REGIONAL AND SECTORAL STUDIES I GO%9 5se. jqq6 The Informal Sector and Microfinance Institutions in West Africa Edited by Leila Webster Peter Fidler The Informal Sector and Microfinance Institutions in West Africa WORLD BANK REGIONAL AND SECTORAL STUDIES The Informal Sector and Microfinance Institutions in West Africa EDITED BY LEILA WEBSTER PETER FIDLER The World Bank Washington, D.C. I C 1996 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W., Washington, D.C. 20433 All rights reserved Manufactured in the United States of America First printing September 1996 The World Bank Regional and Sectoral Studies series provides an outlet for work that is relatively limited in its subject matter or geographical coverage but that contributes to the intellectual foundations of development operations and policy formulation. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this publication are those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to the members of its Board of Executive Directors or the countries they represent. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Dr., Danvers, Massachusetts 01923, U.S.A. The complete backlist ofpublications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from Distribution Unit, Office of the Publisher, 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. Leila Webster is a private-sector development specialist in the Private Sector Development Department of the World Bank. Peter Fidler is at the Kennedy School of Government and was formerly a consultant at the World Bank. Cover photo by Doug Salloum shows the inauguration of a new village bank that is part of the Village Banks Nganda network in Senegal. Library of Congress Cataleging-in-Publication Data The informal sector and micro-fiance institutions in West Africa / edited by Leila Webster and Peter Fidler. p. cm. - Includes bibliographical references. ISBN 0-8213-3597-9 1. Informal sector (Economics)-Africa, West. 2. Informal sector (Economics)-Africa, French-speaking West. 3. World Bank-Africa, West. 4. Banks and banking, Cooperative-Africa, West. 5. Banks and banking, Cooperative-Africa, French-speaking West. I. Webster, Leila, 1950- . II. Fidler, Peter, 1972- HD2346.A58154 1996 330-dc20 96-22495 CIP Contents Acknowledgments ..................... xi Foreword ..................... xii Acronyms and Abbreviations ..................... xiv 1 Introduction ...............................................1 Part 1. Overviews ..................... 3 2 The Informal Sectors of West Africa. 5 Peter Fidler and Leila Webster 3 Microenterprise Support Programs in West Africa .21 Leila Webster and Peter Fidler 4 A Profile of Selected Microfinance Institutions .43 Leila Webster 5 Implications ............... 71 Leila Webster Part 2. Country Studies ............... 79 6 Burkina Faso ............... 81 Ousa Sananikone 7 Cape Verde ............... 91 Ousa Sananikone 8 Chad ............... 105 Ousa Sananikone 9 The Gambia ............... 115 Simel Esim 10 Guinea ............... 127 Angela Walker 11 Guinea-Bissau ............... 139 Simel Esim v vi Contents 12 Mali .................................................. 153 Pierre-Olivier Colleye 13 Mauritania . ............................................. 167 Anne-Marie Chidzero 14 Niger .................................................. 177 Pierre-Olivier Colleye 15 Sao Tome and Principe ............ ........................ 185 Peter Fidler and Sarah Sipkins 16 Senegal ................................................. 195 Anne-Marie Chidzero 17 Sierra Leone . ............................................ 203 Bita Hadjimichael Part 3. Appraisals of Selected Microfinance Institutions ..... ........... 215 18 Reseau des Caisses Populaires ......... ..................... 217 Pierre-Olivier Colleye and Ousa Sananikone 19 Sahel Action: Projet de Promotion du Petit Cr6dit Rural .... .... 231 Pierre-Olivier Colleye and Ousa Sananikone 20 Programme Int6gre pour le Developpement de l'Entreprise ..... 243 Pierre-Olivier Colleye and Ousa Sananikone 21 Credit Rural de Guinee ............ ........................ 255 Pierre-Olivier Colleye and Ousa Sananikone 22 Credit Mutuel de Guinee ........... ....................... 269 Pierre-Olivier Colleye and Ousa Sananikone 23 Kafo Jiginew ................. ........................... 281 Doug Salloum 24 Caisses Villegeoises d'Epargne et de Credit Autogerees du Pays Dogon ........................................... 293 Doug Salloum 25 Credit Mutuel du S6negal ............. .................... 307 Doug Salloum Contents vii 26 Village Banks Nganda ............ ......................... 319 Doug Salloum References ................................................... 333 Annexes 1 Aggregate Data for Microfinance Institutions in West Africa .... 339 Peter Fidler 2 The New Legal Framework for Mutualist Savings and Credit Organizations ........... ....................... 347 Doug Salloum 3 Financial Notes .351 Pierre-Olivier Colleye Maps . ...................................................... 353ff. Boxes 2.1 Main reasons banks do not serve microentrepreneurs ..... ..... 19 3.1 Key principles in the provision of technical assistance .... ...... 27 3.2 Ten principles for sustainable lending to microentrepreneurs .... 30 4.1 Key features of the new law governing mutualist-type institutions .......... ....................... 65 8.1 Main administrative procedures necessary to start a business in Chad ................................................. 109 9.1 Case study of the handicrafts sector in The Gambia ..... ...... 118 11.1 Informal financial services in Guinea-Bissau ..... ............ 142 12.1 Gender constraints in Mali ......... ....................... 156 13.1 Social structures in Mauritania ........ ..................... 169 13.2 Investissement-Developpement en Mauritanie ..... .......... 172 Figures 2.1 Location of the informal sector ......... ...................... 7 2.2 Sources of finance for the informal sector ...... ............... 16 18.1 Model 1 ................................................. 223 18.2 Model 2 ................................................. 224 22.1 Structure of Credit Mutuel de Guinee ...... ................. 271 Tables 2.1 Characteristics of formal and informal sector enterprises. 6 2.2 GNP per capita and growth rates of GNP per capita in West Africa, 1981-91 .10 2.3 Social indicators of development in West Africa, 1992 .11 3.1 Levels of sustainability among microfinance institutions .31 viii Contents 3.2 Indicators of sustainability among selected microfinance institutions .......... ......................... 34 3.3 Services offered by microenterprise development programs in West Africa .................. .......................... 36 4.1 Basic characteristics of nine microfinance institutions in West Africa ................. ........................... 44 4.2 Indicators of the size and growth of nine microfinance institutions in West Africa ......... ........................ 46 4.3 Organizational structure of nine microfinance institutions in West Africa .................. .......................... 48 4.4 Savings services in nine microfinance institutions in West Africa ................. ........................... 51 4.5 Credit services in nine microfinance institutions in West Africa ................. ........................... 52 4.6 Indicators of institutional efficiency in nine microfinance institutions in West Africa ......... ........................ 57 4.7 Staff productivity in nine microfinance institutions in West Africa ................ ............................ 58 4.8 Indicators of operational and financial sustainability in nine microfinance institutions in West Africa ...... ............... 62 6.1 Informal sector activities in Burkina Faso, by gender ..... ...... 84 6.2 Selected microenterprise credit and savings programs in Burkina Faso, mid-1994 .......... .......................... 88 7.1 Selected microenterprise credit programs in Cape Verde .96 9.1 Matrix of -selected nongovernmental organizations providing financial services in The Gambia .. 121 10.1 Sectoral contributions to GDP in Guinea, 1988 .130 11.1 Microcredit programs in Guinea-Bissau, 1995 .145 14.1 Primary activities of women in the informal sector in Niger .... 179 14.2 Constraints on microenterprises at various stages in Niger ..... 180 15.1 Characteristics of microenterprises in Sao Tom6 and Principe, 1991 ............ ........................... 188 17.1 Formal and informal sector employment in urban areas of Sierra Leone, by gender and occupation, 1988-89 ..... ..... 206 17.2 Formal and informal sector employment in urban areas of Sierra Leone, by gender and employment status, 1988-89 ... 207 18.1 Growth of the Reseau des Caisses Populaires in Burkina Faso, 1975-94 ......... ......................... 218 18.2 Savings and credit activities of the Reseau des Caisses Populaires in Burkina Faso, by geographic area, as of September 20, 1994 ......... ......................... 220 18.3 Types of loans of t:he Reseau des Caisses Populaires in Burkina Faso ............... ........................... 221 Contents ix 18.4 Adjusted financial statements of the Reseau des Caisses Populaires in Burkina Faso: Income statement for the caisses, 1994 ....... ................ 225 18.5 Adjusted financial statements of the Reseau des Caisses Populaires in Burkina Faso: Rough estimate of the income statement for the entire institution, 1994 ..... .... 227 18.6 Adjusted financial statements of the Reseau de Caisses Populaires in Burkina Faso: Balance sheet, 1994 ..... ......... 229 19.1 Decisionmaking structure of the PPPCR in Burkina Faso ........ 236 19.2 Growth of the PPP('R in Burkina Faso, 1991-94 ..... ........... 237 19.3 Adjusted financial statements of the PPPCR in Burkina Faso: Income statement, 1993 ............ ....................... 238 19.4 Adjusted financial statements of the PPPCR in Burkina Faso: Balance sheet, 1993 ................. ...................... 240 20.1 Decisionmaking structure of PRIDE in Guinea ..... ............ 248 20.2 Adjusted financial statements of PRIDE in Guinea: Income statement, 1993-94 .......... ...................... 251 20.3 Adjusted financial statements of PRIDE in Guinea: Balance sheet, 1992-94 ............. ....................... 252 21.1 Decisionmaking structure of Credit Rural de Guinee ..... ..... 261 21.2 Basic performance indicators of Credit Rural de Guin6e, 1989-94 ................................................. 262 21.3 Break-even analysis of Credit Rural de Guin6e, 1993 ..... ..... 263 21.4 Adjusted financial statements of Credit Rural de Guinee: Balance sheet, 1992-93 ............. ....................... 264 21.5 Adjusted financial statements of Cr6dit Rural de Guinee: Income statement, 1992-93 .......... ...................... 266 22.1 Growth of Credit Mutuel de Guinee, 1989-94 ..... ........... 274 22.2 Adjusted financial statements of Credit Mutuel de Guinee: Balance sheet, 1993 ................. ...................... 276 22.3 Adjusted financial statements of Credit Mutuel de Guinee: Income statement, 1993 ............ ....................... 278 23.1 Characteristics of loans offered by Kafo Jiginew in Mali, 1992-93 ................................................. 285 23.2 Growth of Kafo Jiginew in Mali, 1992-94 ...... .............. 287 23.3 Adjusted financial statements of Kafo Jiginew in Mali: Balance sheet, 1992-93 ............ ........................ 289 23.4 Adjusted financial statements of Kafo Jiginew in Mali: Income statement, 1992-93 .......... ...................... 291 24.1 Growth of Caisses Villageoises du Pays Dogon, 1991-93 ................................................ 293 24.2 Adjusted financial statements of Caisses Villageoises du pays Dogon: Income statement, 1991-93 ...... ............ 301 x Contents 24.3 Adjusted financial statements of Caisses Villegeoises du pays Dogon: Balance sheet, 1991-93 ...... ............... 302 25.1 Composition of caisses of Cr6dit Mutuel du S6n6gal, 1993-98 . ............................................... 308 25.2 Growth of Credit Mutuel du Sen6gal, 1989-94 ..... .......... 312 25.3 Adjusted financial statements of Cr6dit Mutuel du Sen6gal: Balance sheet, 1991-93 ........... ......................... 313 25.4 Adjusted financial statements of Cr6dit Mutuel du S6negal: Income statement, 1991-93 ......... ....................... 314 26.1 Composition of loans in Village Banks Nganda in Senegal, 1994 ................ ......................... 324 26.2 Potential membership in Village Banks Nganda in Senegal .... 326 26.3 Adjusted financial statements of Village Banks Nganda in Senegal: Balance sheet, 1994 ........ ..................... 328 26.4 Adjusted financial statements of Village Banks Nganda in Senegal: Income statement, 1994 ....... .................. 329 A1.1 Basic characteristics of nine microfinance institutions in West Africa, 1994 .............. .......................... 340 A1.2 Basic credit characteristics of nine microfinance institutions in West Africa, 1994 ............. ......................... 342 A1.3 Savings characteristics of nine microenterprise institutions in West Africa, 1994 ............. ......................... 344 A1.4 Financial performance of nine microfinance institutions in West Africa, 1993-94 .......... ......................... 345 A1.5 Indicators of staff productivity in nine microfinance institutions in West Africa, 1993-94 ........................ 346 Acknowledgments This compendium is very much the product of team effort. Silvia Sagari, former division chief of the Industry and Energy Division of the World Bank's Western Africa Department, initiated this work and supported it intellectually over the ten months it took to complete. Peter Watson, division chief of the Trade and Finance Division in the West Central Africa Depart- ment, supported the institutional case studies in Burkina Faso. Pierre-Olivier Colleye and Ousa Sananikone together wrote ten of the twenty-six chapters in this volume and carried out much of the fieldwork. As members of the team, Anne-Marie Chidzero, Simel Esim, Peter Fidler, Bita Hadjrimichael, Doug Salloum, Sarah Sipkins, Angela Walker, and Leila Webster wrote chapters, verified facts in the field, and carried out institutional appraisals. Mary Dutch made production of this work possible by proofreading and formatting each and every piece. The first draft was reviewed by many people. We are particularly grateful to Silvia Sagari, Ira Lieberman, William Steel, Carlos Cuevas, Lynn Bennett, Elizabeth Rhyne, and Arvind Gupta for their detailed and very helpful comments. Other useful comments were received from Adriana De Leva, Victoria Kwakwa, Gerald Caprio, Tonia Marek, Roza Makonnen, Borje Tallroth, Miguel Saponara, Jacob Yaron, and Hilda Yumiseva. We are very appreciative of the time and energy donated by the many managers of microfinance institutions who were interviewed by field teams, particularly those from the nine programs that were reviewed in depth. We are also grateful to AF51E staff for welcoming us on missions and for long chats in the field and at headquarters about this subject. xi Foreword There are many myths about the informal sector: that microentrepreneurs are poor credit risks, that they cannot pay for services, and that they cannot mobilize savings. However, experience has proven otherwise. Poor people pay and repay, and the financial institutions serving them can be commer- cially viable, especially in the long term. Microfinance is an opportunity to make national investment more effective dollar for dollar. Savings mobiliza- tion in microfinance institutions answers a real need of the "unbanked" poor who want secure places to put their savings. As a result of the International Conference on Actions to Reduce Global Hunger, donors created the Consultative Group to Assist the Poor (CGAP) to help strengthen microfinance institutions, identify best practices, and channel funds through sound microfinance institutions to the very poor. In line with this approach, this study focuses on the informal sectors of selected West African countries and identifies some examples of microfinance best practice. We have long been grasping for ways to reach this vital and vibrant sector, this "silent majority" that accounts for a third to half of gross domestic product and half to three-quarters of total employment in the West African countries reviewed here. Having learned some valuable lessons through experience, we decided a year ago that the time had come to engage in a more aggressive pursuit of solutions. A team was then set up to study the best approaches to helping the informal sector and to make full use of the experience of individuals already engaged in this enterprise at the grassroots level. The writings contained in this volume are the first fruits of this team's work, consisting of country profiles, case studies of some of the most active initiatives already underway, and an overall synthesis of the work to date. They are the work of many people with good minds and good hearts who have taken up humbly but passionately the precious task of development that has been entrusted to them by the international community. It was my privilege to accompany this team in their work. We still do not have all the information we need on the sector, nor do we have answers to all the questions, but we need to start moving ahead with what we do know. The challenge now is to design with care, hope, and courage, in partnership with West African governments, with other donors, and with the grassroots organizations that are working on the ground, practical, effective programs to reach the underserved poor. xii The faces of the people of West Africa, those we have seen on our trips and the thousands and millions we will never see, are guiding our path. Even in silence, they remind us of their dreams for themselves and for their country, our dreams, and the true purpose of our work. Silvia B. Sagari Chief, Industry and Energy Division Western Africa Department The World Bank March 1996 xiii Acronyms and Abbreviations ACEP Alliance du Credit et de l'Epargne pour la Production ADOPEM La Asociaci6n Dominicana para el Desarrollo de la Mujer AGETIP Revue de l'Agence d'Execution des Travaux d'Interet Public contre le Sous-emploi AIDS Acquired immunodeficiency syndrome APEK Association pour la Promotion Economique de Kindia BCEAO Banque Centrale des Etats de l'Afrique de l'Ouest BKD Badan Kredit Desa BNDA Banque Nationale de Developpement Agricole BRI Bank Rakyat Indonesia BRK Bankin Raya Karkara CARE Cooperative for American Relief to Everywhere CFA Communaute financiere africaine CFD Caisse Francaise de Developpement CGAP Consultative Group to Assist the Poor CICM Centre International du Credit Mutuel CIDA Canadian International Development Agency CIDAC Information and Documentation Center, Amilcar Cabral CIDR Centre International de Developpement et de Recherche CIRAD Centre International de Recherche Agronomique pour le Developpement CLUSA Cooperative League of the United States of America CNCA Caisse Nationale de Credit Agricole CRS Catholic Relief Services CSI Contribution du secteur informel FAARF Fonds d'Appui aux Activites Remuneratrices des Femmes FINCA Foundation for International Community Assistance FUNDEI Fundac,o para o Empresarial Industrial GDP Gross domestic product GEMINI Growth and Equity through Microfinance Investments and Institutions GNP Gross national product HIV Human immunodeficiency virus IDA International Development Association IDM Investissement-Developpement en Mauritanie IFAD International Fund for Agricultural Development ILO International Labor Organization xiv Acronyms and Abbreviations xv IMF Intemational Monetary Fund INC-FAC Instituto Nacional de Cooperativas-Fondo de Apoio as Cooperativas IRAM Institut de Recherches et d'Application des Methodologies de Developpement KfW Kreditanstalt fur Wiederaufbau K-REP Kenya Rural Enterprise Program LPD Lembaga Perkreditan Desa NGO Nongovernmental organization OECD Organization for Economic Cooperation and Develop- ment oici Opportunities Industrialization Center Intemational OMCV Organizac,o das Mulheres de Cabo Verde PARMEC Projet d'Appui a la Reglementation sur les Mutuelles d'Epargne et de Cr6dit PNDA Politique Nationale de D6veloppement de l'Artisanat et de Promotion de la Micro et Petite Entreprise PRODIA Promotion de Developpement Industriel, Artisanal et Agricole PPPCR Projet Rural Integre pour le Developpement des Entreprises SAL Structural adjustment loan SDID Societe de Developpement Intemational Desjardins SDR Standard Drawing Rights SLOIC Sierra Leone Opportunities Industrial Center SOLIDAMI Solidariedade e Amizade TANGO Association of Nongovemmental Organizations UNCACEM Union Nationale des Cooperatives Agricoles de Credit et d'Epargne en Mauritanie UNDP United Nations Development Programme UNICEF United Nations Children's Fund UNIDO United Nations Industrial Development Office USAID U.S. Agency for International Development VITA-PEP Volunteers in Technical Assistance-Private Enterprise Project 1 Introduction This study was initiated by Silvia Sagari, who felt that the World Bank needed to know more about informal sectors in the countries of West Africa so that it could do more to help low-income populations in this region. The work was carried out in phases. In phase 1, staff studied the informal sectors of twelve countries in West Africa. A seminar for departmental staff was held in September 1994 to discuss research findings. Bank staff wanted to know more about the types of microfinance institutions that are in this region, and a team from the Private Sector Development Department was dispatched to identify the most effective programs and to evaluate their performance. In phase 2, nine microfinance institutions with a reputation for excellence were selected, and appraisal teams visited each of them in late 1994 and early 1995. As this phase comes to a close, phase 3 is being launched with plans to take action on the recommendations contained herein. The countries included in this review are Burkina Faso, Cape Verde, Chad, The Gambia, Guinea, Guinea-Bissau, Mali, Mauritania, Niger, Sao Tome and Principe, Senegal, and Sierra Leone. As is obvious, this is a heterogeneous group that mixes coastal West African and Sahel countries. It corresponds to regional groupings within the World Bank. Throughout this volume, these countries are referred to as West Africa, even though they represent only a subset of countries in this region. Several important caveats are in order. First, the desk research that underlies the country studies revealed that relatively little has been written about the informal sectors of West Africa, especially compared with countries in other regions. This research was heavily dependent on sector studies carried out by GEMINI (Growth and Equity through Microenterprise Investments and Institutions), which was financed by the U.S. Agency for Intemational Development, and on studies completed by the Intemational Labor Organization. Second, all aggregate statistics in the country studies and all financial data presented in the nine appraisals are estimates. Exact figures on the informal sectors in this region do not exist, and, in the case of the institutional appraisals, researchers had to construct income statements and balance sheets based on available data and their best judgments. Third, there are probably many more programs on the ground than are described in the country papers; desk studies and a few country visits cannot produce fully comprehensive pictures. Fourth, nine microfinance institutions were chosen for an in-depth evaluation, but this does not mean that other good programs do not exist as well. I 2 Introduction This volume is organized as follows. Part 1 contains four overview chapters. Chapter 2 summarizes information from the country studies to produce a profile of informal sectors in this region. Chapter 3 outlines what is known about effective delivery of credit and savings services to the poor and places West African microfinance institutions in this best-practice context. Chapter 4 presents an overview of the nine case studies of selected microfinance institutions in the region, and chapter 5 lays out some implications for the World Bank. Part 2 contains the twelve country studies, and part 3 contains appraisals of the nine microfinance institutions that were evaluated in depth. Part 1 Overviews 2 The Informal Sectors of West Africa Peter Fidler and Leila Webster This chapter first defines West Africa's informal sector and places it in context, and then outlines past and current debates about its role in developing country economies. The focus then shifts to West Africa's regional macroeconomic and social indicators, followed by the main characteristics of West African informal sectors and the primary constraints on enterprise and sectoral growth. Finally, the chapter looks specifically at the financial services available to the informal sector, leading into chapter 3, which focuses on microfinance institutions in West Africa. Defining the informal sector The International Labor Organization (ILO) first used the term "informal sector" in the early 1970s to refer to informal economic activities (ILO 1972, p. 5). In current parlance, this term is typically used in one of two ways. The first refers to illicit or illegal activities by individuals operating outside the formal sphere for the purpose of evading taxation or regulatory burden. This definition has been used most often in the Latin American context, notably by de Soto (1989). The second is simply shorthand for very small enterprises that use low-technology modes of production and management and does not refer to legal status. Throughout this document, the informal sector refers to the smallest enterprises, typically those with ten or fewer employees, the vast majority of which are one-person businesses with few wage workers.' As used, the term also implies a traditional mode of production and therefore excludes small high-technology businesses.2 The degree of legal or regulatory control over the enterprise is not considered. In fact, employing a legal definition would not be useful in West Africa, because the extent to which these very small businesses are registered and pay taxes varies from country to country. Informal sector enterprises share similar characteristics. Two main features are their small scale of operation and labor-intensive mode of production. Other common characteristics include low fixed costs, reliance on family labor, use of personal or informal sources of credit, recycling of goods, and lack of wage employment. Formal and informal businesses are compared in table 2.1. 5 6 Overviews Table 2.1 Characteristics of formal and informal sector enterprises Characteristics Formal sector Informal sector Entry barriers High Low Technology Capital-intensive Labor-intensive Management Bureaucratic Family-based Capital Abundant Scarce Work hours Regular Irregular Wage labor Normal Limited Inventories Large Small Prices Often fixed Often negotiable Financial services Banks Personal, informal Customer relations Impersonal Personal Fixed costs Large Negligible Advertising Necessary Little to none Government subsidy Often large None Markets Often export Rarely export Source: Taken, in part, from Santos 1979, p. 38. Figure 2.1 describes the informal sector in terms of employment and number of enterprises. Medium and large firms typically attract the most attention from governments and donors, but most workers and enterprises are located in the informal sector. Diversity is its trademark, and it includes survivalists (very poor people who work part-time in various nonfarm, income-generating activities); self-employed people who produce goods for sale, purchase goods for resale, or offer services; and very small businesses (microenterprises) that usually operate from a fixed location with more or less regular hours. Participants include very poor, marginal people as well as members of the working class. This sector houses an especially large share of people in West Africa. Importance of the informal sector An early (and, for some, still current) view is that the existence of an informal sector is a symptom of economic dysfunction. From this perspec- tive, the presence of street vendors and home workshop producers signals that the structure and growth rate of the formal economy are inadequate to absorb the national labor force. Accordingly, the formal economy must be managed better so that it can better integrate low-income and unemployed people and, in essence, absorb the informal sector. Macroeconomic management has improved in many developing countries, but the informal sector has continued to grow. Development specialists have emphasized the informal sector's importance as a creator of jobs and an incubator of small and medium enterprises. In this view, the informal sector is a giant sponge, absorbing much of the shock of periodic economic contraction by soaking up excess labor and by providing second incomes to individuals whose real incomes have been eroded by inflation The Informal Sectors of West Africa 7 Figure 2.1: Location of the informal sector / firms Medium-scale firms / Teinformal sector \ / r \~~~~~~~~~co Survivalists employed enterprses and public spending cutbacks. In addition, very small enterprises often are viewed as training grounds for prospective formal business owners (who are in notably short supply in many African countries). Claims of job creation and enterprise incubation have been challenged in recent years. Some observers argue that, in reality, net job creation by informal sectors has actually increased very little (that is, large numbers of new enterprises are offset by equally large numbers of closed enterprises). In addition, a growing body of research shows that few microenterprises become small and mediunm businesses, discrediting somewhat the notion of the informal sector as an incubator of the formal sector (although there may be intergenerational effects). Recent empirical evidence concerning these questions in the African context comes from work by Donald Mead of Michigan State University. Based on surveys of large, representative samples of micro and small enterprises in Botswana, Kenya, Malawi, Swaziland, and Zimbabwe, Mead concludes that: * Net employment in small enterprises in Southern Africa grew by roughly 7 percent a year in the 1980s and early 1990s. * Small enterprises absorbed more than 40 percent of total new jobs created in this region during the 1980s. 8 Overviews * 75-80 percent of all new jobs in small enterprises came from new businesses, with the remainder coming from growth in existing microenterprises. * 20 percent of sample enterprises added a worker or two, but only 1 percent reached ten or more workers. * Because the total number of microenterprises was so large, even small growth increments created a large number of jobs.3 Current thinking looks to microenterprise development as a major tool with which to reduce poverty. In this paradigm, graduating informal micro- enterprises to small businesses is less critical than generating income by creating jobs for the poor. Very small enterprises are seen as a vehicle with which the poor can increase their income; impact analysis of borrowers from several microfinance programs confirms that recipients of microloans have been able to increase and stabilize their incomes.4 In the current debate regarding the importance of the informal sector, two counterarguments often arise. The first asserts that informal sectors create a large number of jobs, but that these jobs are of such poor quality that many microenterprise workers would be better off as employees of medium and large companies. Firm-level surveys carried out by the World Bank's Regional Program onl Enterprise Development, which is managed by the Africa Technical Department, found that manual and nonmanual wage employees of large firms with more than 250 workers earn significantly more than manual and nonmanual wage employees of microenterprises with fewer than ten workers (the countries analyzed were Cameroon, Kenya, Zambia, and Zimbabwe; see Mazumdar 1994, p. 19). Casual observation shows that working conditions in many microenterprises are physically uncomfortable and sometimes dangerous; the work is tedious; and reliable wages, pensions, and job security usually are lacking. A second counterargument says that although microenterprise support programs do raise incomes of the poor, only individuals who are at or near the poverty line are positioned to benefit from them. A recent report, based on a large study of six microfinance institutions carried out by the Universi- ties of Manchester and Reading, argues that microenterprise promotion programs are unlikely to benefit the very poor and that the poorest of the poor might actually be ill-served by credit programs that create debt among persons who can least afford it. The study argues that services that aim to protect rather than to promote income are better suited for the poorest. Such services would avoid enterprise-related loans and include savings services and consumption and emergency loans (Hulme 1995). Debates about the role of the informal sector continue, but, in the meantime, many donors have chosen to support informal sector and microenterprise development for highly pragmatic reasons. Urban migration and high population growth rates have made unemployment a The Informal Sectors of West Africa 9 critical problem for developing country governments, and donors have responded by financing the programs that develop microenterprises, which are meant to absorb a portion of this excess labor. More recently, the failure of structural adjustment to sift economic benefits down fully to the poor has alarmed governments and development experts. Again, the remedy has been to create enterprises with an emphasis on raising and stabilizing incomes, particularly among the self-employed poor. The small but growing number of microfinance institutions that have proved capable of both reaching large numbers of poor people and attaining a high level of financial sustainability has fed hopes that microenterprise development can make a substantial difference in the fight against poverty. Macroeconomic and social context of informal sectors in West Africa The past decade has witnessed deteriorating economic performance in much of West Africa. Since the late 1980s, per capita gross domestic product (GDP) declined in four of the nine countries for which comparable data are available and grew only slightly in the remaining five (see table 2.2). Only Cape Verde and Chad averaged GDP per capita growth rates greater than 2 percent between 1987 and 1991 (World Bank 1994a, p. 241). Seven of the nine countries for which comparable data are available registered a higher per capita growth rate between 1981 and 1986 than they did between 1987 and 1991. Across the region, living standards dropped, population growth rates averaged 2.7 percent, massive urban migration altered demographics, and exogenous shocks including civil strife and declining world commodity prices undermined economic gains. Social indicators describe some of the poorest countries in the world (see table 2.3). For the region as a whole, per capita gross national product (GNP) averaged $440 in 1992 (all dollars are U.S. dollars, unless otherwise noted). With the exception of Sao Tome, half to three-quarters of the countries' populations are illiterate, and secondary school enrollment is 20 percent or less. Infant mortality and child malnutrition rates are among the highest in the world. In addition to such adverse structural forces, high population growth rates, shrinking public budgets, and large-scale urban migration have placed great demand on jobs in the region, demand that cannot be met by the formal enterprise sector or the public sector. Instead, the informal sector provided employment for most of the millions who sought jobs in West Africa over the past decade. Characteristics of West African informal sectors Most informal sector enterprises in West Africa are family-based with somewhat sporadic operations. Rural enterprises commonly are home-based 10 Overviews Table 2.2 GNP per capita and growth rates of GDP per capita in West Africa, 1981-91 (percent, unless otherwise noted) GNP per capita, Difference 1993 GDP per capita growth rate between Country (U.S. dollars) 1981-86 1987-91 the two periods Burkina Faso 300 2.2 0.4 -1.8 Chad 220 4.5 2.6 -1.9 Gambia, The 370 1.2 0.3 -0.9 Guinea-Bissau 220 2.9 1.5 -1.4 Mali 290 0.4 -1.2 -1.6 Mauritania 530 -0.9 -1.0 -0.1 Niger 290 -4.9 -2.4 2.5 Senegal 780 0.4 -0.2 -0.6 Sierra Leone 170 -2.1 0.8 2.9 Note: Cape Verde, Guinea, and Sao Tome and Principe were omitted due to insufficient data for the comparable time period. From 1980 to 1991, the GDP per capita grew at an average rate of 2.1 percent a year in Cape Verde and declined at an average rate of -3.3 percent a year in Sao Tome. Source: World Bank 19931,1994a. and involve agroprocessing or petty trade; urban enterprises are likely to engage in trade or services. Revenues are often able to support only the owner, and use of paid employees is rare. Beyond family members, apprentices who work f or little if any wage income account for the bulk of workers in informal sector businesses. For example, a national survey of informal enterprises in Burkina Faso found that apprentices constituted 67-70 percent of all microenterprise workers (World Bank 1989, p. 34). Competition in most informal sector markets is fierce, due in part to ease of entry and large number of participants, and in part to the small size of many markets served by informal sector enterprises. Lack of specialized skills leads microentrepreneurs to become concentrated in similar activities, which hastens market saturation. Consumers benefit from an excess supply of low-cost foodstuffs and basic goods, but producers and retailers typically realize low profits. Informal sectors in West African countries tend to have few vertical links with larger domestic firms and weak horizontal links among microenterprises due to lack of specialization. Similarly, microenterprises have little or no contact with foreign markets (except in the case of border operations like those carried out by the djilla traders on the border between Guinea-Bissau and Senegal). For all their scale and dynamism, most West African informal sectors operate as parallel rather than as integrated economies. Informal sectors in West Africa have absorbed many unemployed and underemployed people. However, conceiving of these informal sectors as interim way stations for the unemployed and for individuals who need 1-3 Table 2.3 Social indicators of development in West Africa, 1992 (percent, unless otherwise noted) GNP per capita Infant Child Population Illiteracy Secondary Urban Country (U.S. dollars) mortality mortality growth rate schooling population Burkina Faso 300 14.4 46 2.8 82 7 15 Cape Verde 850 6.6 15 2.4 53 20 30 Chad 220 12.4 35 2.5 70 7 34 Gambia, The 370 13.2 20 3.4 73 16 24 m Guinea 500 13.3 23 2.8 76 10 27 Guinea-Bissau 190 14.0 - 2.1 64 4 21 Mali 290 10.2 31 2.7 68 7 25 Mauritania 530 11.9 31 2.8 66 - 50 Niger 290 12.6 49 3.3 72 7 21 Sao Tome 360 6.7 17 2.6 33 66 44 Senegal 780 8.1 22 2.9 62 16 40 Sierra Leone 170 14.3 23 2.6 79 16 34 - Not available. Source: Taken in part from World Bank 1994h. 12 Overviews second jobs underestimates their importance. In Mali and Senegal, for example, the informal sector accounts for the vast majority of all employ- ment (World Bank 1993d, p. 10). Similarly, in Chad and Guinea, the informal sector contributes an estimated 66 and 62 percent of CDP, respectively (World Bank 1994f; ILO 1990, p. 59). In addition to absorbing labor on a temporary basis, the informal sector clearly is home to a majority of the poor in West Africa. From the vendor of millet beer in N'Djamena to the tailor in Banjul, and from the female mat merchant in Niamey to her thread-spinning counterpart in rural Mali, the informal sector comprises a wealth of activities. Sectoral concentration differs by country. In Mauritania and Senegal, trade dominates: 85 percent of all microenterprises in Mauritania and more than 70 percent in Senegal work as vendors in makeshift shops, small kiosks, and marketplaces (this is partially because Mauritania and Senegal have higher urban concentrations than other countries in the Sahel; Charmes 1992, pp. 60-66; World Bank 1994g). In contrast, survey data from Sao Tome indicate that cottage-industry production accounts for approximately 60 percent of all informal enterprises, the majority of which are engaged in sewing, carpentry, handiwork, and woodworking (ILO 1991). In Guinea-Bissau, informal sector employment is distributed more evenly across sectors, with 25 percent of informal firms in trade, 43 percent in services, and 30 percent in artisanry and light manufacturing. Women are active in most urban and rural informal sectors. They account for roughly 50 percent of informal sector employment in Sierra Leone, more than 66 percent in Burkina Faso, and around 80 percent in Cape Verde (Sierra Leone, Central Statistics Office 1991; Burkina Faso, Ministcre de l'Emploi, du Travail, et de la Securite Sociale 1993). In Niger, women run roughly 40 percent of all microenterprises, and in rural areas of Guinea and Guinea-Bissau, many women, particularly women whose husbands work in neighboring countries during the dry season, support themselves fully from informal sector activities. The informal sector is less egalitarian than one might assume. Internal divisions of labor typically channel women into the least profitable activities, and, in some countries, women work only in home-based enterprises. In Sao Tome, for example, women occupy more than 50 percent of all apprentice- ships but constitute only 6 percent of all salaried positions in the informal sector (ILO 1991). Men tend to monopolize the more profitable informal sector jobs, perhaps because they enjoy easier access to education, capital, trading contacts, and transportation. Women tend to be far less mobile than men, because societal norms across West Africa dictate that a woman's primary responsibility is to care for her family, and this means staying close to home. Survey work carried out in many West African countries points to practical, cultural, and even legal constraints that block women's entry into The Informal Sectors of West Africa 13 the formal labor market and, in essence, confine them to the informal sector. At the most general level, women's opportunities are limited by their lack of education, large numbers of pregnancies and children, and substantial responsibilities for agricultural production and home maintenance. In most countries, poor access to information, markets, and financial services combines with cultural mores to limit women's economic participation. Legal discrimination is also evident in some countries. For example, regulations in Mali prohibit married women from opening businesses without their husband's authorization. The labor code in Niger maintains regulations from the colonial period that hinder women's access to employment in a range of activities. Constraints affecting West African informal sectors Low entry barriers facilitate relatively easy enterprise start-up, but managing and expanding small businesses usually call for additional resources and skills that are commonly in short supply. Informal sector surveys and field visits point to the critical nonfinancial and financial constraints that constrict the operations and growth of enterprises. Nonfinancial constraints When surveys ask microentrepreneurs in West Africa to identify their most critical nonfinancial constraints, their most common response is lack of demand for their products and services. This reply typically signals problems with both markets and marketing. The problem with markets is usually saturation. Informal sector markets in many of the countries reviewed are small because customers have low incomes and because rural areas, where many informal activities take place, are sparsely populated. As the number of new entrants grows, the supply of goods and services surpasses the demand, thereby depressing prices and incomes. Informal sector markets often resemble a small (and sometimes shrinking) pie that is being cut into smaller and smaller pieces. Problems with marketing are obvious if one visits a rural or urban market in one of these countries. Clusters of market vendors hawk identical products daily from the same location, and few make more than marginal profits. When asked why they do not identify niche products for local markets or expand to larger markets beyond their immediate area, vendors typically answer that their goods and services are always sold at this location, that everyone in the local area knows where they are, and that they would not know where else to sell.5 Lack of product differentiation (a root cause of most marketing problems) is related to the low skill base of most informal sector entrepreneurs. Enterprise owners typically start their businesses with a specific skill, such as dressmaking, or with specialized knowledge, such as knowledge of 14 Overviews markets for a particular product. But such skills and knowledge tend to be easily accessible to others as well, and new activities are often initiated in already saturated or near-saturated markets where profits are marginal at best. Lack of technical training (and often illiteracy) become binding constraints to upgrading product quality and diversifying production, both of which are needed to survive the fierce competition encountered in most of these markets. Problems with product differentiation also reflect the limited skills of the pool from which West African microentrepreneurs draw labor. In most cases, profits are too small to pay salaried workers, and entrepreneurs rely heavily on unpaid family workers and apprentices. Entrepreneurs who can afford to hire a paid employee or two typically must choose from among individuals who have learned their skills via classic apprenticeships. Because most apprenticeship training simply passes on known practices, the skills of its graduates tend to be based on emulation rather than innovation, a sure recipe for noncompetitive production. Another common, nonfinancial constraint is the poor state of infrastruc- ture in the region. Many roads in West Africa are unusable (except by foot) for much of the year. Few microenterprises have access to running water or electricity. In Chad, the unit cost of electricity is far too expensive for most microentrepreneurs (about CFAF 100 per kilowatt compared with CFAF 16 in neighboring Nigeria; the currency is the franc of the communautefinanciere africaine, or CFA). In Guinea-Bissau, for example, it is easier to make a telephone call from Bissau to New York than within the country. The impact of poorly developed infrastructure on informal enterprises is mixed, but mainly negative. In rural areas, poor roads and nonexistent telecommunications block the flow of information about markets and technologies. Poor roads increase the cost of transporting inputs and finished products to and from markets.6 Limited availability and the high price of electricity often are incentives for continued use of traditional, low- technology methods of production that typically turn out noncompetitive products. West African microentrepreneurs rarely identify legal and regulatory problems such as licensing, registration, location regulations, and other government regulations. It could be that informal sector enterprises are too small to become ensnared in the kinds of regulatory bottlenecks that affect larger businesses, or regulatory problems may be second-order problems that only become significant once more immediate barriers have been surmounted. Financial constraints When West African entrepreneurs are asked to name their most serious financial constraint, most will answer lack of money. Further questioning usually narrows the problem to the dual dilemma of insufficient working The Informal Sectors of West Africa 15 capital and no source from which to borrow it. Personal savings and loans or gifts from family provide start-up capital, but microentrepreneurs typically experience shortages of working capital within months of start-up. This, combined with an irregular supply of raw materials, quickly under- mines their ability to respond to market demand.7 Small enterprises commonly work on a stop-and-go basis, only purchasing raw materials and taking on workers when they have an order from a customer and sometimes an advance payment. Inventories range from minimal to nonexistent. Start- up funds usually include the purchase of one or several used machines needed to launch operations, but further investment is rare. Of perhaps equal importance, few microentrepreneurs in West Africa have secure places to deposit their savings, particularly if they are in need of liquidity or would like to earn a return. Rotating savings and credit clubs are popular means of accumulating savings, but members are bound to their turn in the rotation, and their deposits do not earn interest. Savings collectors also play an important role in mobilizing deposits throughout West Africa, but they charge a monthly fee in return for the security and convenience of their services. In sum, interest-bearing, liquid savings accounts are rarely, if ever, available to informal sector entrepreneurs in most West African countries. Sources of finance for West African informal sectors Informal sector entrepreneurs rely on multiple sources of financial services, from the least formal and most available (family) to the most formal and least available (banks; see figure 2.2). Almost all informal sector financial transactions take place with family, friends, moneylenders, and esusus and tontines (informal savings and credit clubs). In recent years, these traditional sources have been supplemented in some locations by credit unions and international and domestic nongovernmental organizations (NGos) that offer microloans and, in some cases, savings services. Some NGOS and credit unions have grown quickly, and a few have demonstrated relatively high levels of efficiency, but in most countries they meet only a fraction of the demand for financial services among low-income people. Family andfriends The most common sources of credit for informal enterprises in West Africa are family and friends. In urban areas of the Gambia, for example, friends and families frequently pool their incomes to create an effective social safety net for one another. A survey of microenterprises in Guinea found that 95 percent of all firms were either self-financed or financed with the aid of family members during their initial start-up period (ILO 1990). In the medium term, the larger problem for microentrepreneurs is obtaining access to working capital after resources from family and friends run dry. 16 Overviews Figure 2.2 Sources of finance for the informal sector8 Family, Savings Supplier Esusus Nongov- Credit Banks friends collectors, credits, and ernmental unions money- customer tontines organiza- lenders advances tions Least formal, Mostformal, most available least available Moneylenders and savings collectors Moneylenders are a primary source of credit for individuals and microenterprises, especially where no other options exist (as is the case in much of West Africa). Some moneylenders work full-time (and may even be registered as such), and some work part-time, using surpluses from trading to lend short term at high rates of interest (Aryeetey and Steel 1994, pp. 3-4). In both cases, moneylenders use their own funds as working capital and do not intermediate between savers and borrowers. Savings collectors throughout West Africa mobilize savings from households and microentrepreneurs. They collect a fixed amount from each client daily, and, at the end of the month, they return the sum of the deposits minus a small fee. Aryeetey and Steel identify three principal reasons why depositors are willing to pay to save, even in inflationary environments (Aryeetey and Steel 1994, pp. 34). First, contractual savings are a form of financial management that ensures the availability of working capital needed to restock supplies. Second, savings collectors provide security. The contractual nature of the system not only protects funds from theft but also, which is more important, shields funds from other demands and thus satisfies an "illiquidity preference" (Shipton 1990). Third, the transaction costs of a collector who comes daily to the home or business are much lower than those of more formal savings and credit institutions, and depositors are therefore willing to pay a "convenience fee." Suppliers' credit, customer advances, and trade credits Use of suppliers' credit and customer advances is rare during enterprise start-up but becomes more common during the life of a business. For example, up to two-thirds of traders in the Gambia use trade credit to finance their operations. When available in the start-up phase, suppliers' credit tends to be positively linked to the size of the enterprise (Aryeetey and others 1994, p. 84). As microenterprises mature, suppliers' credit and advances from customers are much more likely to serve as sources of working capital. The Informal Sectors of West Africa 17 Esusus and tontines Informal savings and credit clubs (usually referred to as tontines) have a long tradition in most West African countries. With some local variation, members make fixed contributions at regular intervals, with one member taking the entire pot at each interval. Because all recipients but the last receive the accumulated sum sooner than if they had acted alone, esusu and tontine funds can be attractive (van den Brink and Chavas 1991).9 In principle, tontines disband after each rotation, but some continue over long periods of time, and many develop supplementary social welfare, insurance, and loan funds. Esusus and tontines are a popular source of finance among informal sector operators in West Africa. In Niger, roughly 75 percent of urban women and 50 percent of rural women participate in tontines. In Senegal, tontines are major financial institutions with an estimated CFAF 65 billion ($125 million) transferred through them annually (a billion is 1,000 million; World Bank 1993g). Nongovernmental organizations In some countries, international and domestic NGOS have become an important provider of financial services to low-income people. Many NGOS came to West Africa to provide relief services in response to droughts, and some have since expanded their programming to include the provision of credit and savings services. Some countries like Burkina Faso, the Gambia, and Mali have a wealth of NGO programs; others such as Mauritania have very few. Desk reviews and field visits indicate that most NGO microfinance programs in West Africa have been successful in reaching the poor with their services, but most have small scale of outreach and weak financial sustainability. A typical program reaches several hundred low-income people (often mainly women) with small loans and sometimes training courses, but the cost per person served is very high. Such programs are fully dependent on the flow of donor funds for their survival. Some NGOS in the region are aware of the need to achieve greater scale and to move toward financial sustainability. For example, Sahel Action PPPCR in Burkina Faso offers small loans to the rural poor on a relatively efficient basis. As of February 1995, the program had 13,960 loans outstand- ing and an average loan size of $48. These 13,960 loans were disbursed by only thirty-two agents, for an average of 436 loans per agent. The real effective interest rate is 26 percent a year, and repayment rates are just under 100 percent. Sahel Action PPPCR has a long way to go before achieving full financial sustainability, but it appears to be on the right track (see part 3 for a case study). 18 Overviews Credit unions Credit unions are the most formal financial institutions that serve informal sector clients. Most West African credit unions are modeled after the French "mutualist" system and have a membership structure and a network of branches linked by a federated union. Mobilization of savings is emphasized more than provision of loans. Indeed, the maximum loan size is a multiple of the borrower's savings, with repayment motivated by the potential loss of savings and by social pressure from the larger membership. Their reliance on local savings has made credit unions more conservative in their lending than credit-based NGO programs that have access to donor funds. Credit unions tend to take a minimalist approach to credit delivery and rarely provide training, technical assistance, or other services to their members. Several credit unions visited in West Africa operate on sound financial principles."0 Reseau des Caisses Populaires in Burkina Faso maintains a fairly conservative lending philosophy and yet has been successful in extending its reach throughout the country. The Reseau has 64,000 members but dispersed only 3,710 loans in 1994. Each of the 64,000 members has deposited savings in the credit union (the average savings account is $76) even though deposits do not yield interest. This kind of performance once again demonstrates that low-income entrepreneurs often need a secure place to deposit their savings even more than they need access to credit. Savings deposits are 160 percent of the Reseau's loan portfolio. Cr6dit Mutuel de Guin6e operates on similar principles. Of its 45,000 members, only 2,800 are borrowers. Deposits total 248 percent of Cr6dit Mutuel de Guin6e's loan portfolio. Banks As a rule, banks do not serve informal (or small) enterprises (see box 2.1). They have little or no presence in many rural areas in West Africa. In addition, even when they are physically present, few are either structured or inclined to respond to the financial needs of very small savers and borrowers. Conclusions Conceptual debates and empirical research continue to refine our under- standing of the role of the informal sector in developing country economies. In the meantime, govemments and donors struggle with spiraling unem- ployment rates in cities and deepening poverty in the countryside. As employment opportunities in the public sector have decreased (particularly relative to the number of job seekers), informal sectors have become sources of employment for more and more people. In hopes of enhancing the informal sector's capacity for job creation, donors (and increasingly The Informal Sectors of West Africa 19 governments) have supported microfinance and training programs for informal sector entrepreneurs. Box 2.1 Main reasons banks do not serve microentrepreneurs * Small size of the transactions * Lack of collateral * Inexperience, illiteracy, and lack of numeracy of the borrower * Physical remoteness of many informal sector enterprises * Mobility of many informal sector enterprises * Lack of bookkeeping or an appropriate compliance framework * General lack of information about the borrower and of predictabil- ity surrounding the transaction. West Africa is no exception to these trends. Given the prevalence of widespread, severe poverty in the region, the large size of the informal sector's contribution to GDP and employment, and the promise of microfinance as an effective poverty-reduction strategy, it makes sense that microenterprise promotion should occupy center stage in development strategies. This chapter has gone into some detail about the financial constraints facing microentrepreneurs and the most common sources of financial services. Chapter 3 outlines global best-practice principles in the delivery of financial and nonfinancial services to microentrepreneurs and places West African microenterprise assistance programs within this context. Notes 1. Throughout the text, the terms informal sector, informal sector activities, very small enterprises, microenterprises, and microentrepreneurs are used more or less interchangeably. 2. As typically employed, the term informal sector does not include small farmers, and they are not included here in general discussions about the informal sector. However, unlike in most other regions, some of the microfinance institutions referred to in the text do include small farmers among their clients. 3. Mead 1994, pp. 5-8. These surveys included some large firms, but 73 percent of sample firms had fewer than ten employees at start-up and 60 percent were one-person enterprises. 4. Studies by Chen (1992), Wahid (1993), Khandker, Khalily, and Khan (1994), Khandker and Khalily (1994), and Khandker, Khan, and Khalily 20 Overviews (1994) have documented the role of the Grameen Bank in Bangladesh in increasing household incomes, household expenditures for basic needs, employment opportunities, and nutritional intake. These studies also identify social benefits for women who participate in microfinance programs: participants feel less marginalized, have higher aspirations for their children's education and future, use more reliable sources of drinking water, are more likely to use latrines and contraceptives, and are less likely to marry at an early age. 5. Most microentrepreneurs have little if any financial margin with which to invest in product development and new markets. 6. Notwithstanding these disadvantages, this isolation also protects microentrepreneurs from competing goods and services. 7. Microentrepreneurs everywhere use personal savings as start-up capital. In West Africa, however, initial capital requirements tend to be very low. In a review of fourteen developing countries, the lowest initial capital requirement for microenterprise formation was in Sierra Leone ($49). See Liedholm and Mead 1987, p. 35. 8. As one proceeds to the right-hand side of the continuum, financial transactions occur with less frequency. 9. The costs for early recipients are low, but esusu and tontine partici- pants whose turn arises late in the rotation are rarely compensated for use of their savings or for the risk that early recipients will drop out. 10. The two credit unions mentioned here were appraised by field teams, and case studies can be found in part 3 of this volume. 3 Microenterprise Support Programs in West Africa Leila Webster and Peter Fidler This chapter begins with a discussion of the rationale for providing low- income people with credit and savings. It then turns to what is known about the provision of training services to informal sector entrepreneurs. It next outlines the conceptual transformation that has occurred in the field of microfinance over the past decade and summarizes additional lessons that have been learned in this area. This is followed by a section that lays out a set of criteria for assessing the performance of microfinance institutions, describes some model programs in South Asia and Latin America, and lays out the principles of best practice in microfinance delivery that are used in part 3 to evaluate nine West African microfinance institutions. Turning to West Africa, the next section aggregates information about the sixty-two microenterprise support programs documented in this book to produce a general profile of microenterprise assistance programs in this region. The final section places these programs within a best-practice context and discusses their strengths and weaknesses. Rationale Before outlining best practices in the delivery of microfinance services, it is important to clarify the rationale for selecting financial services as a means of intervening in the informal sector. What benefits are expected to result from delivering financial services to low-income people? Most arguments emphasize the role of financial services in promoting equity and economic growth. Individuals who focus on equity argue that providing financial services to low-income people has the potential to reduce poverty, mainly by smoothing income but also by increasing income. Specifically, savings services enable the poor to accumulate even very small surpluses that can be used at a later date to ride out dips in income or respond to emergencies. Lending services enable the poor to invest in opportunities that will increase their incomes, such as buying instead of renting tools, increasing their livestock, or getting access to bulk prices for raw materials.' Individuals who focus on growth emphasize the efficiency gains of extending financial services to low-income populations. Financial sectors play a key role in the efficient allocation of resources by shifting funds from 21 22 Overviews surplus to deficit locations and from less profitable to more profitable activities. To the extent that resources remain outside the financial sector and thus are not available for use by the larger economy, the efficiency of resource allocation is undermined. In most developing countries, the formal financial sector does not capture the savings of low-income groups, nor does it provide poor people with loans for investment. Because low-income people represent a large share of the total population in most developing countries, large amounts of savings go uncaptured and numerous invest- ment opportunities go unrealized. It follows that efficiency gains can be realized to the extent that these underserved groups are brought into the system: their savings can be channeled into loan funds for use elsewhere, and local lending facilities can allow them to exploit high-return investment opportunities. The question that follows asks why financial institutions in most developing countries do not reach low-income clients: what constrains banks from mobilizing savings and making loans to low-income people? The shallowness and fragmentation of financial sectors in most developing countries, particularly in Africa, is one reason. Until recently, most banks in Africa provided financial services primarily for import and export trade and for public enterprises. Moreover, government controls on deposit and lending rates resulted in the rationing of services, and low-income people were not given high priority in times of rationing. Many such controls have been lifted in recent years, but there is little indication that banks have significantly deepened their reach. Rather, they remain constrained by the lack of regulations that enforce contracts, by poorly functioning information systems, and by the high costs of small transactions. Moreover, few are interested in entering new markets, perhaps due to lack of serious competi- tion. A second set of constraints is that most low-income people in developing countries have characteristics that make them particularly difficult for banks to serve. Typically, they have no history with the banking sectors, they have little in the way of assets other than their personal reputation (which is not known to loan officers), and many live in rural or otherwise hard-to-reach locations. Many are illiterate, and few have any knowledge of banking operations. Most need very small loans for short periods and on an almost immediate basis, because investment opportunities arise and fade very quickly for this group. Most prefer to make very small deposits and to withdraw savings frequently. In sum, low-income people are among the hardest clients for banks to serve, particularly banks in developing countries. Over the past decade, microfinance institutions scattered across all parts of the developing world have led the way in overcoming these barriers and delivering credit and savings services to low-income people. They have done so by employing many of the financial technologies described in this Microenterprise Support Programs in West Africa 23 chapter. In addition to achieving high levels of outreach, a small but growing number of microfinance institutions have moved beyond depend- ence on donor subsidies to full financial sustainability. The emphasis on achieving financial sustainability has increased in recent years-in part because experience has shown that it is possible and in part because donor funds are becoming increasingly scarce. Many development specialists question the impact of microfinance programs. To what extent have anticipated benefits been realized? That is, do participants have more stable and higher incomes, and has financial intermediation become more efficient as microfinance institutions have joined the financial sector? These questions have proved difficult to answer largely because of the technical difficulties associated with measuring the impact of microfinance programs on poverty reduction. It is possible to compare the income of control groups of nonparticipants to that of groups of participants. The difficulty comes with controlling for other variables, such as determining whether the entrepreneurs who participate in microfinance programs are better equipped for success than those who do not. The implication is that the selection criteria predict success better than does access to financial services. Several donors have taken up the question of impact as a major research effort, and their findings should be forthcom- ing in the next several years. Impact studies to date are essentially before-and-after studies of participants. Virtually all show that clients of microfinance programs have higher and more stable incomes than they did prior to their participation. Studies of participants in the Grameen Bank loan programs have docu- mented the program's impact on increasing participants' household incomes, household expenditures for basic needs, employment opportuni- ties, and nutritional intake. These studies also identify social benefits that women gain from participating in microfinance programs: they feel less marginalized, have higher aspirations for their children's education and future, use more reliable sources of drinking water, are more likely to use latrines and contraceptives, and are less likely to marry at an early age (see Chen 1992; Wahid 1993; Khandker, Khan, and Khalily 1994; Khandker, Khalily, and Khan 1994; and Khandker and Khalily 1994). Another impact study from Bank Rakyat Indonesia (BRI), the largest microfinance institution to date, shows that its borrowers have increased their net household incomes by 75 percent and net enterprise income by 93 percent. A related question concerns the opportunity costs of supporting microfinance institutions. The positive impact of having access to credit and savings services is clear, although the long-term impact has not yet been established. The larger question is whether the impact of microfinance institutions is greater than the impact of other possible interventions: what are the costs and benefits of supporting a microfinance institution in a remote area compared with those of building a road to that area? More 24 Overviews research is needed to determine the kinds of interventions that have the largest long-term impact on household income. The extent to which microfinance institutions have improved the efficiency of financial intermediation has not been thoroughly studied, but several observations can be made. First, these institutions have mobilized large amounts of savings, and a large share of these funds have been rechanneled as loans. In this sense, the extent of financial intermediation clearly has grown. Links between microfinance institutions and formal financial sectors have been weak, and only now are being forged in a few locations. When these links are strengthened such that savings mobilized by microfinance institutions flow into the pool of investment capital available to the entire economy and such that microfinance institutions are able to draw on the formal banking system for their loan funds, national systems of financial intermediation will have made the kinds of efficiency gains envisioned. Microfinance institutions and the donors that support them have placed a strong emphasis on gender issues. They recognize that the constraints that limit the poor's access to financial services are particularly harsh for women. The country studies contained in part 2 document numerous examples of legal and cultural restrictions on women's activities. In addition, women in many parts of the developing world are disadvantaged in their access to education, skills, and capital that improve their ability to succeed in small businesses. Microfinance institutions have, in many cases, come to prefer women clients: their services appear to have more impact on women than on men, and women are, for the most part, excellent clients. Some impact studies point out that women invest the extra income that flows from investment of their loan funds in improving the welfare of their families; comparative research about male borrowers does not show the same positive impact. Similarly, women in many programs have paid back their loans at higher rates than have men. Training services Some participants in the debate surrounding the delivery of technical assistance to informal sector entrepreneurs contend that technical assistance for microentrepreneurs should not be a high priority, and some feel that training is (or should be) a critical input to microenterprise assistance programs. Those in the first camp-the "minimalists"-judge the cost- benefit ratio of training programs to be unacceptably high, suspect that most trainers have little of real value to offer microentrepreneurs, and point out that entrepreneurs have demonstrated their aversion to training courses in many settings, particularly where training is a prerequisite for a loan. Minimalists also argue that the heterogeneity of microenterprise sectors makes them ill-suited for the kinds of generic training programs that are Microenterprise Support Programs in West Africa 25 capable of reaching large numbers of people at a reasonable cost (Rhyne and Holt point out that microfinance is a generic intervention that can be used successfully by entrepreneurs in all sorts of businesses; Rhyne and Holt 1994, p. 31). Training courses that are narrowly focused to meet specialized needs typically reach only a small number of users and at a high cost per beneficiary. Observers who argue that technical assistance is critical feel that poorly developed business skills are a binding constraint to enterprise growth, even more than lack of access to credit in many cases. In this view, entrepreneurs' complaints about lack of access to credit mask technical and managerial inadequacies that, if remedied, would have a far greater return than would access to loans. Many practitioners also believe that training must accom- pany the provision of microloans if entrepreneurs are to optimize their use of loan funds, particularly if they are not experienced business managers. Debates about the wisdom of financing training programs for microentrepreneurs lead to general questions about the acquisition of skills and the need for training in this sector. In the early 1990s, the Education and Training Division of the Africa Technical Department of the World Bank, the OECD (Organization for Economic Cooperation and Development), and the ILO jointly sponsored a large research project that interviewed 1,570 micro- entrepreneurs working in eleven common activities in four West African towns (Birks and others 1992, p. xi). They sought to answer three main questions: Which skills do microentrepreneurs use? How do they acquire them? and What possibilities exist for promoting skill acquisition in microenterprises? This research produced many interesting findings, three of which are of direct relevance here. First, and perhaps of greatest importance, is the finding that completion of primary school education-the ability to read and write-is a key predictor of the ability of microentrepreneurs to succeed. Specifically, better educated entrepreneurs enter into apprenticeships more frequently than entrepreneurs without formal education, thereby putting themselves on the quickest path to self-employment; they select the most attractive activities within the informal sector (those that use relatively advanced, high-technology modes of production and are less vulnerable to fluctuations in demand); and they benefit more than their uneducated counterparts from subsequent training programs (Birks and others 1992, p. 93). Second, very large systems of apprenticeship training are the primary route through which informal sector entrepreneurs learn their trades, and, as such, apprenticeships are highly valued by both master and student. Apprenticeships generally are closely tied to market demand and suffi- ciently flexible to incorporate shifts in market trends. In fact, the authors of that study believe that traditional apprenticeship training programs are akin to national treasures that could never be replaced by "formal" systems. 26 Overviews Third, formal training institutions seldom target microentrepreneurs, and few have attempted to train informal sector entrepreneurs. To do so, they would need far more local autonomy and flexibility with which to design appropriate training courses. An effective approach might include evening or weekend classes, flexible entry and exit requirements for courses, short, modular courses, demand-driven extension services, and training center facilities that are available to local entrepreneurs (Birks and others 1992, p. 100). Reflecting these findings, the study recommends that governments focus on providing universal primary education, paying particular attention to the inclusion of girls, rural children, and other underserved groups. By educating the previously uneducated, governments significantly enhance the capacity of these groups to earn incomes as informal sector entrepre- neurs. The study also recommends that formal training institutions reorient their curricula to include microentrepreneurs and address their needs. The authors warn that interventions in existing apprenticeship systems should be undertaken with great care to avoid disrupting internal incentive systems that have functioned well for decades. Government- and donor-financed training for microentrepreneurs generally has taken one of two forms. The first focuses on technical institutions, operated and financed by governments and donors, that train artisans as part of their activities. Visits to several such government- operated training centers and a search of secondary documentation concerning these institutions indicate that this approach has been less than effective in most cases. Problems include lack of contact with the market- place, use of capital-intensive technologies that cannot be replicated by microentrepreneurs, sporadic operations due to fluctuations in donor and government funding, lengthy programs of several years with no clear-cut route to graduation so that relatively few people are trained, and high unit costs of training. The second form is typified by training programs delivered by NGOS and other nonbank intermediaries concurrent with microcredit programs. These training courses range from fairly minimal instruction in group formation, basic analysis of cash flow, and use and repayment of loan funds to full- scale skills training programs. Many practitioners believe that at least minimal training is important for first-time borrowers, particularly those who are illiterate or have no experience. They would argue that, despite the additional cost, training in the basics of business enhances entrepreneurs' chances of success and increases their loan repayment rates. Comprehensive training programs that involve extended series of classes have proved exceedingly costly when measured on a per client basis, and their effective- ness is not clear.2 Many NGOS continue to fund such programs, however, particularly for rural women. Microenterprise Support Programs in West Africa 27 A model approach to training microentrepreneurs remains elusive, but some lessons have been learned (see box 3.1). Current thinking supports a simplified, focused approach that addresses a specific training need, such as cash flow analysis. The greatest success has come with training courses that are optional rather than a condition for credit, are demand-driven in approach and content, are designed to minimize time away from a business, and are financed in part by course fees. Box 3.1 Key principles in the provision of technical assistance 1. Treat microenterprises as clients not as beneficiaries * Aim for cost recovery * Charge fees for service * Find at least quasi-sustainable ways to provide the services 2. Aim for scale * Adopt indirect rather than one-on-one mechanisms to increase leverage and to reach more people * Intervene systematically to affect many firms with a single pro- gram 3. Aim for cost-effectiveness and lower per unit costs 4. Use the market and its signals * Let the market determine fee for service 5. Tailor the product to market demand * Identify opportunities for higher payoffs * Adapt training to the needs of illiterate clientele * Use sector-specific analysis to identify effective systematic inter- ventions 6. Use partners that understand all of the above. Much could be learned by experimenting systematically with cost recovery. Microenterprises have shown a willingness to pay for training that meets their needs, and a market appears to exist for flexible, modular, demand-driven training programs. A fee-based approach has the added benefit of providing a more cost-effective product for the training institu- tion. Improving the financial sustainability of technical assistance for microenterprises (and small and medium businesses) is a frontier in this field. A final point relates to the costing of such services. Training and credit services frequently are lumped together in microfinance institutions' 28 Overviews accounts, and this practice distorts proper costing. Technical assistance programs vary in scope and in cost, but few are able to cover their operating expenses through client fees and other sources of income. In contrast, financial services can cover a substantial portion or all of their costs, and training expenses should be separated out so that the real costs of both financial and training services are clarified. Financial services The delivery of financial services to poor and low-income people changed significantly during the past decade. First, and perhaps of most importance, long-standing and fundamental assumptions about the bankability of the poor have been overturned based on well-documented experience in banking with the poor in a selection of developing countries. A small number of highly effective microfinance programs have demonstrated that low-income clients can use small loans productively, will pay them back fully and on time when given reason to do so, can and are willing to pay high real interest rates for their loans, and do save and often need savings services as much as or more than credit services. These findings correct earlier notions that the poor cannot use credit effectively, do not have the capacity to repay loans, cannot afford to pay high interest rates that reflect the real cost of funds, and do not generate sufficient surplus funds to enable them to save. Second, a shift in thinking is under way within which delivering credit and savings services to the poor is being conceptualized as extending the reach of national financial systems to include low-income customers. For example, financial sectors are being redefined to include nonbank financial institutions as well as banks. This framework differs radically from the long- standing practice of separating microfinance programming from financial sector development and including it within a panoply of social welfare services for the poor in a "project" context. In this view, microfinance institutions fit within a continuum of financial institutions and, in fact, represent an essential component of an integrated financial system that serves the majority of citizens. This new paradigm necessarily has far- reaching implications: links between all financial institutions on the continuum become far more important, the need to revise banking regulations to include nonbank financial institutions arises, and a new legitimacy is granted to financial institutions that serve the poor and to their clients who come to represent an enormous new market for financial services rather than a hefty welfare burden. A third breakthrough has been the development of new lending technologies that are effective both in reaching large numbers of low-income borrowers and in moving microfinance institutions toward financial sustainability. These technologies are designed to deliver small loans with terms and conditions that meet the needs of poor clients, to lower lenders' Microenterprise Support Programs in West Africa 29 transaction costs by distributing them over large numbers of borrowers, by increasing operational efficiency, and by reducing delivery costs, and to increase revenues by using full-cost interest rates and high loan repayment rates. Box 3.2 outlines a list of principles that have worked well in designing financial services for microentrepreneurs. A fourth important change is that microfinance institutions are now expected to attain high levels of (if not full) financial sustainability within a reasonable period of time. First-generation credit programs were fueled by a constant supply of donor funds and paid little attention to operating costs, loan losses, and the expense of auxiliary services. Efficiency and cost recovery were not priorities for service providers or for international donors. As a result, credit programs had minimal outreach because credit funds were limited to donor funds, repayment rates were poor, financial interme- diation was stunted because cheap credit discouraged the mobilization of local deposits, and credit was concentrated in the hands of privileged and less-than-creditworthy borrowers who successfully pursued these subsi- dized loans. Over the past decade, a handful of pioneer microfinance institutions have demonstrated not only the bankability of the poor but also the potential for sustainability of financial institutions that serve the poor. Full financial sustainability is reached when administrative, loan loss, inflation, and financial costs are covered entirely by revenues. Few institutions have reached this stage, and the difficulty of doing so should not be understated. In fact, most microfinance institutions fall somewhere between levels 2 and 3 shown in table 3.1. At level 1, programs are highly subsidy-dependent, and they would cease to exist without a constant injection of fresh funds. Almost all microcredit programs during the 1970s and early 1980s would be classified as level 1 institutions. At levels 2 and 3, microfinance programs depend on donor financing but increasingly are able to cover operating expenses, such as salaries, administrative costs, depreciation of fixed assets, and loan losses. By level 3, operations are generally efficient with high ratios of clients to staff and low rates of delinquency and default (Rhyne and Otero 1991). Institutions at level 4 are fully self-financing, with revenues covering both operating and financial costs. Perhaps fewer than a dozen microfinance institutions have reached level 4, but the fact that these programs have done so sets an example for the rest.3 The fifth development of significance is the new focus on mobilizing savings among the poor. Microfinance institutions increasingly are under pressure to mobilize savings, and there is some contention among practitio- ners about the wisdom of this trend. It is clear that savings services are needed urgently by large numbers of poor people around the world to protect their incomes and to serve as an alternative to the assumption of debt. Savings deposits also offer microfinance institutions a valuable source of sustainable local funds.4 30 Overviews Box 3.2 Ten principles for sustainable lending to microentrepreneurs 1. Offer short-term loans. Offer initial loans of three to six months with frequent repayment periods. 2. Offer small initial loans. Start with very small loans appropriate for meeting day-to-day financial requirements of microenterprises and motivate repayment by offering larger loans as incentives for repeat customers. 3. Concentrate on working capital tofirms with a proven track record. New businesses have high failure rates, and initial financing should be left to friends and families. Microfinance institutions should focus on providing working capital to existing enterprises. 4. Specialize services, diversify portfolio, ignore the temptation to target. Institutions should specialize in the provision of either technical assistance or financial services, but not both. It is the loan portfolio, not the range of services, that should be diverse. Portfolios should be diversified to counter supply-driven, targeted credit programs and to reduce the risks associated with a homogeneous portfolio. 5. Simplif services. Make the credit program customer-friendly. Use a simple application process (often less than one page) appropriate to low levels of literacy and numeracy and streamline operations to minimize staff time per loan. 6. Localize services, focus on scale. Locate close to entrepreneurs. Select staff from local communities, including people with lower levels of education (and salaries), rather than from staff in formal banks. Locate where there is a critical mass of clients in order to reduce transaction costs. Microfinance programs reach sustainability in part by making large volumes of loans. 7. Shorten turnaround time. Limit the time between loan application and disbursement. Since the majority of microloans are for working capital, speed is ideal for borrowers and saves administrative costs for lending institutions. Turnaround time can be lowered by relying on solidarity groups to screen clients and by decentralizing loan approval. 8. Motivate repayment. Motivate repayment via group solidarity and joint liability. Group lending is efficient because it externalizes costs. Character-based lending to individuals (as opposed to groups) can be effective where the social structure is cohesive and there is little potential for political abuse. 9. Recognize that the poor do save. Credit programs are more sustainable when they are financed with personal savings. Savings are often the forgotten half of informal finance. However, accepting deposits is labor-intensive and therefore quite costly. Moreover, unstable programs can rob the poor of their savings. Great care must be taken when an institution accepts fiduciary responsibilities. 10. ChargefIll-cost interest rates (factoring in inflation, loan losses and delinquency, and the cost of loanfunds). The administrative costs of lending to the poor require interest rates that are substantially higher than rates charged by commercial banks. Microentrepreneurs have shown a willingness to pay high rates for services that meet their needs. Source: Taken in part from Malhotra 1994; Rhyne and Holt 1994. Those who argue most vehemently for the inclusion of savings services are of the view that true financial intermediation is only possible when deposits are recycled as loans. In this view, savings are key to microfinance institutions' success in attaining full financial self-sufficiency because they Table 3.1 Levels of sustainability among microfinance institutions Characteristic Level 1 Level 2 Level 3 Level 4 Operating costs High operating Lower costs by adapt- Lower costs by achieving Increased efficiency costs ing lending technolo- greater efficiency gies Repayment Mixed repayment Low loan losses Low loan losses Very low loan losses Cost of funds No cost of funds Cost of funds still Transition to near- Commercial sources of funds cheap but higher commercial cost of funds Interest rates Negative interest Commercial interest High interest rate Full-cost interest rates rates rates compared to commercial ratesa Coverage of prices Far from covering Operating expenses Almost full-cost pricing Full financial viability; costs covered (operating plus partial profitable without subsidy financial costs) a. Although rates may be higher than commercial rates offered to the formal sector, they are generally substantially lower than rates on funds offered by moneylenders and traders. Source: Rhyne and Otero 1991. 32 Overviews provide a sustainable, local source of loan funds and because using savings deposits as loan funds (as opposed to grants from donors) imposes greater responsibility and therefore greater discipline on lending institutions. The Unit Desa Program in the Bank Rakyat Indonesia, for example, has six depositors for each borrower (Christen, Rhyne, and Vogel 1994, p. 27). Those who argue the middle ground agree that poor people need and should have access to savings services but argue that not every microfinance institution should be compelled to take deposits, especially if such services are available locally from an alternative source. Those who are most cautious about savings mobilization argue that deposit mobilization is far more labor- intensive and costly than is commonly understood and that it should not be undertaken unless the institution is clearly able to bear these costs (Schmidt and Zeitinger 1994). And finally, some worry about the potential danger posed by the growing number of unregulated microfinance institutions that have begun to mobilize large amounts of savings (Adams 1994, p. 6). Evaluating the performance of microfinance institutions If the objective of microfinance institutions is to deliver financial services to large numbers of previously underserved low-income people, an evaluation of how well this objective is met would measure levels of outreach (the number of poor people reached).5 It would also look at the quality of services provided (the degree to which services are geared to the financial needs of low-income people). An institutional evaluation would include measures of financial sustainability, under the dual assumption that donor funds are limited in amount and duration and that microfinance institutions' long-run capacity to deliver services will depend on their ability to generate revenues sufficient to cover their operating and financial costs. Outreach Over the past decade, a small number of microfinance institutions have achieved impressive scale and depth of outreach. As regards scale (the number of people reached), the largest microfinance institutions are BRI, which provides financial services to 2 million microborrowers and 12 million savers in Indonesia, and the Grameen Bank, which offers financial (and other) services to low-income clients in almost half of the villages in Bangladesh. As regards depth of outreach (the extent to which clients are poor or underserved), most successful microfinance institutions do reach the poor (if not the poorest) with loans that average $200 to $500. Some microfinance institutions have successfully reached the very poor with loans below $100 (Badan Kredit Desa (BKD) in Indonesia has average loans of roughly $40; see Christen, Rhyne, and Vogel 1994, p. 24). Reaching the underserved has meant that many microfinance programs have concen- Microenterprise Support Programs in West Africa 33 trated their operations in rural areas and have focused on serving women. This is particularly true of microfinance programs in West Africa. Experience has shown that low-income people have fairly specialized financial needs, and high-quality microfinance services meet these specialized needs. As regards credit, most microentrepreneurs need working capital loans with terms of twelve months or shorter, simple application procedures and quick turnaround time, few if any restrictions concerning the use of loans, loan guarantee requirements that match their capacities (personal guarantees), repayments on a frequent basis, and ongoing access to loans, based on credit records. One way of measuring the quality of lending services within a given institution is to use repayment rates as a proxy. The assumption here is that entrepreneurs will protect their access to credit to the extent that they value it, particularly if there is any significant competition among sources of financial services. The most successful programs have default rates of a few percentage points or less, and this is usually interpreted to mean that these lending programs are well-suited to their target groups. As regards savings, optimal services offer security, liquidity, and a financial return. Here again, clients' behavior is perhaps the best measure of the quality of service. If voluntary savings deposits are significant in number, a good match has probably been achieved between the target group's needs and the service provided (again, contingent somewhat on the level of competition). The quality of savings services becomes difficult to measure in cases where deposits are mandatory, either as a membership requirement or as a precondition for receiving a loan. Sustainability Financial sustainability refers to an institution's ability to generate sufficient revenues to cover the full cost of service delivery, including operational and financial costs (the cost of funds). Efficient operations lower costs, and several indicators are commonly used to measure operational efficiency in microfinance institutions. Scale of operations clearly is important because costs drop when they are distributed across large numbers of clients. On the revenue side, setting interest rates to reflect costs is crucial, as are maintain- ing high repayment rates and avoiding loan losses. The final indicator of a microfinance institution's sustainability examined here is the durability of its organizational structure. Operational efficiency is commonly evaluated by looking at cost and productivity variables. Several factors are key (see table 3.2). Making very small loans to hard-to-reach clients is far more expensive than commercial banking, which typically makes large loans to easily accessible clients. The most efficient microfinance institutions have administrative costs of 10-30 percent. Salaries usually account for the largest share of administrative costs, and the ratio of salary costs to average outstanding loan portfolio is a 34 Overviews Table 3.2 Indicators of sustainability among selected microfinance institutions (percent, unless otherwise noted) Salaries Average Nominal as a salary as a Financial (real) percentage multiple of Number of Institution and self- effective of loan GNP per loans per country sufficiency interest rate portfolio capita staff member LPD 137 36 (27) 6.6 0.7 30 (Indonesia) BKD 118 55 (48) 11.5 1.8 57 (Indonesia) BRI (Indonesia) 110 34 (25) 4.5 4.2 118 Actuar 104 71 (52) 16.2 5.5 90 (Colombia) BancoSol 103 55 (46) 12.5 5.1 139 (Bolivia) ADOPEM 89 72 (67) 16.8 12.7 74 (Dominican Republic) FINCA (Costa 75 32 (23) 8.7 3.3 270 Rica) Source: Christen, Rhyne, and Vogel 1994. function of salary levels and of scale of lending. Determining average salaries as a percentage of GNP per capita controls for the effect of small portfolios and measures the extent to which salaries are in line with national norms. Models to emulate in this regard would be BRI (Indonesia), LPD (Lembaga Perkreditan Desa, Indonesia), and Grameen Bank (Bangladesh) where salary costs constitute only 5, 7, and 9 percent of respective loan portfolios (in these three cases, salary costs represented 54 to 65 percent of total administrative expenses). Efficiency indicators also include measure- ments of productivity, usually the average number of clients per staff member. In this regard, institutions such as FINCA (Foundation for Interna- tional Community Assistance, Costa Rica) and BRK (Bankin Raya Karkara, Indonesia) have been outstanding, with average ratios of staff to clients of more than 200.6 Three factors in a microfinance institution's ability to earn the revenues needed to cover operational and financial costs are significant: the loan repayment rates, the scale of lending, and the interest rates charged on loans. On the first count, there is now strong consensus that repayment rates of less than 95 percent are not acceptable; large loan losses are an avoidable barrier to cost recovery. Second, there are substantial economies of scale in microenterprise lending; it is exceedingly difficult for institutions with a Microenterprise Support Programs in West Africa 35 small number of borrowers to earn sufficient revenues to cover their costs even when they charge high interest rates. Third, interest rates should be based on the cost of loan delivery, not on prevailing rates in commercial banks. Considerations that should be taken into account when setting interest rates include operational and financial costs, loan delinquency rates, credit demand, and inflation (Christen, Rhyne, and Vogel 1994, p. 35). Nominal rates can be quite high, especially if calculated from the borrowers' points of view in cases where savings that do not bear interest are required to receive a loan. For example, Actuar (Colombia), BKD (Indonesia), and BancoSol (Bolivia) charge nominal effective interest rates of 71, 55, and 55 percent, respectively. All three of these institutions registered positive rates of return on assets for 1993 (Christen, Rhyne, and Vogel 1994, p. 36). Long-term sustainability also depends on the durability of the institu- tional structure (that is, how well it will be able to handle the demands of change and of expansion). Relevant indicators might include the balance struck between social and financial objectives, the extent to which an institution is decentralized and representative of its clients, the effectiveness of staff training programs in transferring leadership and knowledge, and the ability of the institution to withstand negative external factors such as macroeconomic instability and an imperfect regulatory environment. Although difficult to quantify, each of these variables is critical to determin- ing a microfinance institution's likelihood of survival in the long run. Characteristics of West African microfinance programs This section turns to the West African context and looks at what is known about microenterprise support programs operating in this region. This overview is based on aggregate information about the sixty-two microenterprise assistance programs documented in the twelve country studies in part 2. Two caveats are in order: first, these programs are among the largest and best-documented in the region, but they are far from constituting a complete census. As such, conclusions are based on only a sample of programs. Second, much of the information about these programs is taken from secondary sources and may be somewhat dated, particularly in countries that were not visited by the authors.7 A first observation about microenterprise programs in West Africa concerns their distribution. Some countries have a good number of microfinance and training programs-Burkina Faso, the Gambia, Guinea, Guinea-Bissau, and Mali-while others have very few-Chad, Mauritania, Sao Tome, and Sierra Leone. The largest microfinance programs are located in Burkina Faso, Guinea, Mali, and Senegal, and the nine programs that were selected for in-depth study are located in these four countries (see part 3). Programs identified in Chad and Niger are mid-size, with 2,000 and 5,000 clients each. Guinea-Bissau has a fair number of programs, but most serve 36 Overviews fewer than 100 clients a year. Not surprising, the smallest programs are in Cape Verde and Sao Tome. Microfinance institutions in West Africa come in a wide variety of shapes and sizes, although most are either NGOS or credit unions. NGOS tend to be oriented toward poverty alleviation, and most offer credit or training services to low-income, rural people. In many cases, NGOS first came to this region in response to problems associated with drought, and microfinance programming was subsequently added to other programs addressing the basic needs of poor people. With this addition, there is a notable trend toward enlarging urban operations-in part in recognition that larger scale- in-lending can be achieved in urban than in rural areas. The credit unions (most of which are actually mutualist institutions) are membership institutions with a strong orientation toward savings mobilization and a somewhat lesser commitment to lending. Microfinance institutions range in size from Cr6dit Rural de Guinee, with an outstanding portfolio of more than $1.3 million and more than 13,000 loans, to Tiniguena in Guinea-Bissau, with fifteen loans a year and an outstanding portfolio of around $1,000. In Burkina Faso, 64,000 people benefit from membership in the Reseau de Caisses Populaires compared with 164 people who have received loans from all seven microfinance projects established in Guinea-Bissau since 1992.8 Programs also vary in the types of services they provide to their informal sector clients. Most of the sixty-two microenterprise development programs reviewed here have a microcredit component. A quarter also provide savings services. Almost half offer some form of business training, usually in combination with a loan program (see table 3.3). Table 3.3 Services offered by microenterprise development programs in West Africa Component Percentage of programs Projects containing a microcredit component 82 Projects containing a savings component 25 Projects containing a training component 48 Projects exclusively offering microfinance 52 Projects exclusively offering training 18 West African microfinance programs within a best-practice context Best-practice principles in the field of microfinance have gradually been established over the past decade, based mainly on experience in Asia and Latin America. One objective of this review is to place microfinance institutions in West Africa in this larger, global context-both to bring Microenterprise Support Programs in West Africa 37 information about the West African programs to the global table and to get some feel for how well these programs perform relative to the best-known programs. The in-depth case studies in part 3 bring a great deal of detail to this discussion, but an overview of the sixty-two programs documented in the twelve country studies in part 2 points to general strengths and weaknesses. As regards outreach, few West African microfinance programs have been able to achieve significant scale in lending, especially compared with some of the large programs in Asia and Latin America. Specifically, only two institutions in these twelve countries have more than 10,000 outstanding borrowers (Credit Rural in Guinea and Sahel Action in Burkina Faso). Most programs have several hundred borrowers; the largest ones have several thousand. Generating revenues sufficient to cover operational costs at such a small scale is difficult. Several microfinance institutions reviewed have had much greater success in mobilizing savings, and at least a handful have well over 20,000 savers. This strong performance is due mainly to the presence of credit unions that emphasize savings mobilization. The reasons for the small size of most microfinance programs in this region are fairly obvious. Most are located in provincial capitals, but the rural areas on which they focus are very sparsely populated. Per-client costs are high in this environment, in part because it is expensive to reach groups of clients physically (villages are small and isolated) and in part because poorly developed infrastructure increases the expense of delivering even basic services. People tend to be very poor, which means that average loans are very small and that loan portfolios therefore remain small. In addition, most programs offering microfinance services have a small capital base and do not have access to the level of financing that would be required for significant scaling up. Most are also heavily involved in the delivery of other kinds of services without the type of singular focus that is commonly associated with building large microfinance systems. In contrast, many of these programs appear quite strong in their depth of outreach. Figures on average loan sizes, the share of women among clients, the percentage of clients who are illiterate, and the rural location of many programs indicate that most are indeed reaching the very poor who likely have little or no access to financial services other than tontines and moneylenders. Visits to many of these programs and interviews with their clients confirmed an impressive performance along this dimension. Without in-depth analysis, it is difficult to assess the quality of services that are being provided. Lending and savings technologies appear to be fairly current in that there is widespread use of proven lending techniques such as character-based lending, including solidarity groups and comrnittees of village elders, small starter loans with access to repeat loans based on repayment records, and simplified and fast procedures. Substantial information from microfinance programs elsewhere clearly has been 38 Overviews transferred to West Africa, presumably by donors whose staff bring experience and literature from other regions. It is equally clear that a better flow of information would hasten the speed with which cross-fertilization could take place. For example, many program managers who were interviewed expressed an ambition to move their programs toward financial sustainability, but most seemed to have a fairly weak understanding of what this would actually entail. The fact that very little current literature in the highly dynamic field of microfinance has been written in or translated into French is no small contributor to this problem. In terms of sustainability, these programs show a mix of strengths and weaknesses. Three factors are of particular importance in structuring sound microfinance programs: cost-based interest rates on loans, high repayment rates, and scale in operations. Normally, interest rates for microloans should be substantially higher than rates offered by commercial banks. Credit programs in West Africa have been slow to scrap subsidized interest rates despite growing evidence that microentrepreneurs are willing to pay high rates for financial services that meet their needs. Only twelve of the forty- eight microcredit programs reviewed (for which reliable data are available) charge nominal annual interest rates of 24 percent or higher, and most charge much lower rates.9 As described in the next chapter and in annex 2, a new law in the CFAF countries sets out the provisions for licensing microfinance institutions. Once licensed, these institutions will become subject to existing usury laws that cap interest rates at twice the discount rate. Enforcement of these usury laws would have profoundly negative implications for West African microfinance programs, because virtually none would be able to deliver loans at this cost, and all would be forced into loss-making positions. An additional key to financial sustainability is establishing large-scale operations that serve substantial numbers of people and thereby lower the unit costs through volume lending. Scaling up is especially difficult in the predominantly rural and sparsely populated countries of western Africa. Some programs have concentrated on growth: Credit Rural and Credit Mutuel de Guinee have 76,000 members and savers between them and caisses in 131 locations. These institutions are the exception, however, and most West African microenterprise programs have a small number of clients. In Cape Verde, no microfinance institution makes more than 125 interest-bearing loans a year. For obvious reasons, maintaining high repayment rates is important, and West African programs appear fairly strong in this area. Thirty of the forty- four programs for which repayment rates are available reported rates of 90 percent or higher.'" When the repayment rates for the forty-four programs are aggregated, the mean is 89 percent. The impression is that many programs have made effective use of incentives for repayment such as Microenterprise Support Programs in West Africa 39 mobile loan agents, solidarity group and joint liability lending, and short- loan terms with access to larger loans on repayment. Conclusions Looking ahead to the future of microfinance programming in this region, several tentative conclusions can be drawn. First, there is significant room for improvement within current programs using principles and practices developed in other regions of the world. The vision and objectives of individual programs become important here. If considerations such as scale of outreach and financial viability are not of paramount importance to program managers (and if they have access to resources that allow them to assign low priority to these aspects of their operations), much of what has been learned over the past decade is somewhat irrelevant. Many practitio- ners believe that microcredit programming is part and parcel of social service packages that enjoy high levels of subsidy. If managers want and have incentives to build large, cost-efficient programs, however, a great deal of mileage can be gained simply by applying the principles and practices described above. Second, the unique conditions in West Africa may call for some variation in approach. The role of microfinance institutions is somewhat different in West Africa than in other parts of the world. Elsewhere, microfinance institutions operate alongside a formal banking sector even if there are few links between them. Fieldwork in West Africa showed that microfinance institutions in many rural and quasi-rural areas are the only banking institutions operating in the region. Commercial and state-owned banks in many of these countries have pulled their branches out of rural areas in an effort to maximize their revenues, and microfinance institutions have become substitutes for banks, most notably in Guinea. The question then becomes the extent to which this substitution is feasible and desirable. On the one hand, it is favorable for the microfinance institutions that enjoy a large client base that includes middle-class people. On the other hand, serious issues of supervision also arise. Issues concerning training arise sharply in the West African context. As shown, poverty is deep and education is minimal with very high rates of illiteracy, particularly among rural women who are the target group for many programs. Many of the programs documented here provide a mix of credit and training services-a practice that is not in favor among leading practitioners who feel that institutions perform better when they specialize either in delivering credit or in providing training. Two questions are relevant here. First, how important is training when clients are illiterate and entirely inexperienced in forming enterprises and handling loan funds? The research pertaining to skills acquisition is particularly relevant, because it was carried out in West Africa. Primary education and apprenticeships were seen as major predictors of success in 40 Overviews the informal sector. Clients served by many of the programs reviewed here, mainly the rural poor, commonly have access neither to primary education nor to apprenticeship programs. A fundamental question that goes to the heart of issues concerning the impact of microenterprise assistance programs asks where the return is greatest. For example, what maximizes a poor, rural person's (particularly a woman's) opportunities to earn a decent living: a primary education, access to apprenticeship training, or classes in business basics from a microenterprise assistance organization? Although these means should not necessarily be viewed as mutually exclusive, given limited resources, impact and cost-benefit analysis is sorely needed to determine which means have the highest social and economic rates of return. A second question related to training concerns the common view that credit services should not be delivered alongside noncredit services such as nutrition and housing. There are sound reasons for this view in light of the experiences of multipurpose programs, many of which have simply tacked credit components onto other social programs with little appreciation for the specialized skills that are required to build sound microcredit programs. It is not surprising that these programs fail to reach many people, repayment rates are poor, and the programs soon collapse. However, the physical environment in West Africa may call for some qualifications to this argument. Field visits to rural programs underscored the difficulty and expense of building infrastructure through which services can be delivered to hard-to-reach populations. In the case of West Africa, arguments for discrete channels for microcredit services become more difficult to defend, and the need for a better understanding of how to deliver microfinance services alongside other services without sacrificing effectiveness becomes more imperative. A third issue that arises concerns the potential for microfinance institu- tions operating in this region to achieve full financial sustainability. The bottom line is whether it can be done. On the one hand, operating conditions for microfinance institutions in West Africa are clearly difficult. Low per capita GDP and stagnant economies, combined with poor rural infrastructure and sparsely populated regions, might make West Africa the most challenging environment in the world for microfinance institutions. These factors raise the cost of operations and make it very difficult to achieve the needed scale of operations. In addition, the regulatory environment is threatening to take a significant turn for the worse if governments choose to enforce usury laws. On the other hand, as is documented in the next chapter, a small number of microfinance institutions have made good strides in the direction of financial sustainability by increasing their number of clients, raising their interest rates, and cutting their costs. Whether these front-runners will be able to reach full financial sustainability is yet to be seen, but it bears careful watching. In the meantime, many other programs in the region could Microenterprise Support Programs in West Africa 41 improve the cost-effectiveness of their programs by taking similar steps. In West Africa, it is important that the large flow of donor funds (and a fairly donor-dependent culture among NGOs) not weaken the incentives for microfinance institutions to adopt the stringent budgetary measures required to put them on the path to financial sustainability. Notes 1. At an aggregate level, microfinance services are a means of broadening economic participation to include marginal groups that have been left out previously. A main benefit is the heightened political stability that flows from having more stakeholders. 2. There is increasing experimentation in the area of technical assistance with the aim of improving effectiveness and lowering costs. An example of such pilot work is a program operated by Women's World Banking that involves firm-to-firm training. 3. As discussed in part 3, which evaluates nine West African microfinance institutions, the assumption that all microfinance institutions can eventually reach full financial sustainability can be questioned. Factors such as income level and population density of client groups, the condition of the local infrastructure, and national financial policies are critical determinants of a microfinance institution's ability to generate sufficient revenues to cover its costs fully. In an environment such as that of West Africa, all of these factors negatively affect the financial sustainability of even the most efficient microfinance institutions. 4. Despite their desire to enter the savings side of financial intermedia- tion, many microfinance institutions have been unable to do so because of national laws and regulations that make it illegal for nonbanks to take deposits. 5. See Yaron 1992; Christen, Rhyne, and Vogel 1994. Most of the parameters of outreach and sustainability used here are drawn directly from these two seminal reports. 6. Christen, Rhyne, and Vogel (1994, pp. 33-35) tested a series of cost and productivity variables as predictors of profitability, including number and amount of loans per staff member, salaries as a percentage of assets, and administrative costs as a percentage of assets. Only three variables were of statistical significance: the country's inflation rate, the microfinance institution's real effective interest rate charged on loans, and the program's average employee salary relative to the country's GNP per capita. 7. World Bank missions visited Burkina Faso, Cape Verde, Guinea, Guinea-Bissau, Mali, Mauritania, and Senegal, but not Chad, the Gambia, Niger, Sao Tome, and Sierra Leone. 8. Programs include Tiniguena, SOLIDAMI (Solidariedade e Amizade), Action pour le Developpement, Service Quaker, Africare's Rural Initiatives 42 Overviews Project, FUNDEI (Fundac,o para o Empresarial Industrial), Caisse Francaise, and United Nations Development Programme (UNDP)/Banco Totta. 9. Three programs are in Guinea, two each are in Burkina Faso, the Gambia, Mali, and Senegal, and one is in Guinea-Bissau. A thirteenth program in Sierra Leone has a credit component that charges 40 percent interest for loans, but it has suffered greatly with the deteriorating security situation, and in 1993 one program center reported a loan recovery rate of only 22 percent. 10. Definitions of repayment rates differ across programs, which makes standardization difficult. Generally, repayment rates are calculated on the basis of loans paid in full within thirty days of the due date, but some are calculated on a ninety-day past-due basis. 4 A Profile of Selected Microfinance Institutions Leila Webster There were several objectives in carrying out these case studies.Staff from the World Bank's Western Africa Department, Industry and Energy Division, were presented with country profiles of West African informal sectors in mid-September 1994 (see part 2).l They grasped clearly the importance of this sector to the Bank's work but were unsure how to proceed, particularly as regards selecting local partner institutions. The first objective, therefore, was to identify and evaluate microfinance institutions that might be well suited for collaboration with the Bank. A second objective was to test current approaches to evaluating microfinance institutions.2 Microfinance institutions in West Africa are much less well known than are their counterparts in Asia and Latin America, and a third objective was to bring these institutions into the mainstream, both to inform practitioners in the microfinance field about them and to place them in the global arena where their experience can be compared and contrasted with that of others. The approach was the following. From the country profiles, microfinance institutions that appeared to reach a large number of poor people and to operate particularly effectively were identified. Other donors active in microfinance work in West Africa were contacted for their recommenda- tions. This process identified nine institutions to be evaluated. After each institution agreed to participate, two-person field teams of Bank staff and consultants visited each one for extensive discussions with managers, accountants, local staff, and clients. Outreach and financial data were examined carefully, and income statements and balance sheets were constructed by field teams, usually as a collaborative endeavor with the institution's accountant or manager.3 Drafts of the case studies were sent to the respective institutions for comments, which were incorporated into the final studies presented in this volume and summarized in this overview. This chapter is structured as follows.4 The following section summarizes the basic characteristics of the nine institutions reviewed, including their origins, objectives, and scale. This is followed by an outline of the organiza- tional structures and systems of governance, a description of credit and savings services, including the types, terms, conditions, and scales of services provided, and, using the parameters laid out in the previous 43 44 Overviews chapter, an analysis of performance in terms of outreach and financial sustainability. The final section projects ahead to future issues that are likely to challenge these institutions. General characteristics The nine microfinance institutions analyzed in this section are of three types: five are credit union-type institutions, two are village banking systems, and two are credit programs (see table 4.1).5 The oldest institution, Reseau des Caisses Populaires, is twenty years old, and the youngest, Village Banks Nganda, has been operating for only eighteen months. They range in size from $90,000 to $5.1 million in total assets and from 600 to 64,000 members. All nine offer credit services to low-income clients, and seven offer savings services (the two credit programs do not take formal deposits). Only one program (PRIDE in Guinea) provides substantial training for borrowers. The remainder offer only minimal training for clients and none of the auxiliary social services that often accompany microfinance services. All operate mainly in rural areas. Table 4.1 Basic characteristics of nine microfinance institutions in West Africa Type of Institution Country Date founded institution PRIDE Guinea 1991 Credit program Credit Rural Guinea 1989 Credit union Credit Mutuel de Guinee Guinea 1988 Credit union Credit Mutuel du Sen6gal Senegal 1988 Credit union Village Banks Nganda Senegal 1993 Village banks Reseau des Caisses Burkina 1972 Credit union Populaires Faso Sahel Action PPPcR Burkina 1988 Credit program Faso Caisses Villageoises du Pays Mali 1988 Village banks Dogon Kafo Jiginew Mali 1988 Credit union The origins of the nine institutions are similar. All were initially started with donor financing: one by the Caisse Francaise de Developpement, one by the Canadian Intemational Development Agency, one by the French Ministry of Cooperation and Development, one by KfW (Kreditanstalt fuir Wiederaufbau) in Germany, two by USAID, and three by a consortia of donors within which the Caisse Francaise de Developpement, the European A Profile of Selected Microfinance Institutions 45 Development Fund, and the European Union have been major contributors. All nine have received significant assistance with implementation from nongovernmental organizations: six French NGOS (including two charged specifically with the development of credit unions), two U.S. NGOS, and one Canadian NGO. All nine institutions have the overarching goal of reducing poverty, and all provide financial services to very poor people. Approaches differ by type of institution. The two credit programs provide microloans to poor clients with the aim of raising and stabilizing their incomes. The remaining seven institutions seek to serve low-income, rural people: five through credit union structures and two through village banks. Most of the institutions reviewed have experienced explosive growth over the past few years, and many have become fairly large institutions (see table 4.2). Six of the seven that have membership structures have more than 10,000 members; five have more than 20,000 (PRIDE and Sahel Action are not membership organizations). Five of the membership organizations have total assets of more than $1 million, and five have year-end loan portfolios of more than $500,000. In several cases, membership has doubled and tripled in the past few years. Growing memberships have brought parallel growth in the number of borrowers in some institutions, but in others, particularly credit unions, the growth in membership has translated into a large number of savers, largely because credit unions emphasize savings rather than loans and have fairly conservative lending policies. Organizational structure and governance In general, organizational structures reflect the type of institution (see table 4.3). The five credit union systems have a relatively large number of local branches (caisses), a small number of regional offices and unions, and a central office or headquarters. At the caisse level, locally elected committees and councils together with local loan agents promote membership and approve and follow-up loans. Regional offices control and consolidate financial data from caisses in their regions, move surplus funds to caisses that need them, ensure that systemwide policies are implemented, and provide some training for caisse committees and staff. Central offices set policies for the institution as a whole, handle institutional finances including investments of surpluses and funds for deficits, provide some training services, and negotiate with donor agencies. The two village banking systems have central offices and village banks without regional offices. Village Banks Nganda operates as a financially linked system under the umbrella of a regional community development agency that has designated one department as the central office that manages the network's operation. The primary donor, Catholic Relief Services, oversees the entire system. The Caisses Villageoises du Pays Dogon Table 4.2 Indicators of the size and growth of nine microfinance institutions in West Africa Amount in U.S. dollars Number Year-end Total Loans at year- Institution Data as of loan portfolio assets Data as of Members end Savers Caisses PRIDEa 7/92 31,000 66,000 1994 n.a. 3,500 n.a. 6b 7/94 447,000 907,000 Cr6dit Rural 7/92 760,000 2,900,000 1989 110 110 0 2 12/93 1,300,000 3,700,000 1994 31,000 13,000 b 31,000 41 Credit 12/93 1,600,000 5,000,000 1990 12,000 176 12,000 29 Mutuel de 1994 45,000 2,800 b 45,000 70 Guin6e Credit 12/92 260,000 1,200,000 1990 5,543 4,500 +'c 5,563 27 Mutuel du 12/93 733,000 2,800,000 1994 22,228 22,226 55 Senegal Village 12/94 65,000 90,000 1994 600 600 600 12 Banks Ngandad Reseau des 1994 1,200,000 5,100,000 1990 25,000 3,710b 25,000 25 Caisses 1994 64,000 64,000 35 Populaires 0 Table 4.2 (continued) Amount in U.S. dollars Number Year-end Total Loans s: Institution Data as of loan porfolio assets Data as of Members at year-end Savers Caisses Sahel Action 12/93 316,000 760,000 1991 n.a. 2,051 n.a. 36.f PPPCRa 1994 13,174 191 Caisses 12/91 49,300 107,000 1991 4,630 455 25 Villageoises 12/93 227,000 360,000 1994 15,330 9,767 1,752 54 Kafo Jiginew 10/92 587,000 1,140,000 1993 13,313 9,320 22,770 42 S 10/93 813,000 1,560,000 1994 20,670 15,105 51 n.a. Not applicable. a. PRIDE and Sahel Action are NGOS without a formal membership base. Neither accepts savings. b. The number of loans outstanding, not the total number of loans made in 1993-94. c. The number of loans made in 1994 as of August 31, 1994. It consists only of rural credit, by far the largest category of loans. d. Village Banks Nganda was eighteen months old when the appraisal was performed, so no trends are available. e. The number of loans from the external account. It does not include first or second intemal account loans. It is also based on a six-month lending cycle. f. Sahel Action does not have caisses; these figures represent the number of villages served. 48 Overviews Table 4.3 Organizational structure of nine microfinance institutions in West Africa Number of Number of staff Number Number of Institution total staff in headquarters of caisses regional offices PRIDE 67 10 6 0 Credit Rural 120 23 41 6 Credit Mutuel 151 11 70 8 a de Guin6e Credit Mutuel 83 7 54 4 du Senegal Village Banks 3.5 1 12 0 Nganda Reseau des 110 15 5 3 Caisses Populaires Sahel Action 57 7 4 0 Caisses 14 b 11 54 0 Villageoises Kafo Jiginew 40 12 51 0 a. To reduce costs, Credit Mutuel de Guinee plans to reduce its regional offices to four. b. Caisses Villageoises du Pays Dogon has fourteen paid staff and more than 130 volunteers who work in the village banks 1.5 days a week. has created fifty-four administratively and financially independent caisses that elect representatives to unions of caisses (that have no physical location), which borrow from the local agricultural bank and pass funds through to creditworthy caisses. The small central unit has taken a major role in establishing new caisses, training local boards and cashiers, and controlling caisse accounts. In the summer of 1995, these central functions were reduced to the provision of basic services by an even smaller unit called the service commun. The two credit programs have central and local caisses, but no regional offices. In the case of Sahel Action, mobile loan agents travel from village to village disbursing loans to groups of borrowers and collecting repayments. In the case of PRIDE, loan agents work out of six large caisses located regionally, and borrowers (also organized in groups) must come to the office for disbursements and repayments. Based on the assumption that effective governance is critical to institu- tional functioning and durability, two aspects of governance were examined: the extent to which authority and responsibility have been decentralized, and the extent to which decisionmakers represent members and clients (Rhyne and Rotblatt 1994, pp. 24-32). In the first case, it seems logical that loan decisions should be as decentralized as possible because doing so is efficient and because strong local involvement in selecting borrowers and A Profile of Selected Micro,finance Institutions 49 handling repayments has been shown to be key to sound credit operations. It also makes sense to handle some functions, such as organizational policymaking, central accounting, links with government and banks, and external relations, at a central level. As for the issue of representation, the assumption is that participatory institutions are more likely than nonparticipatory institutions to have deep roots and thus be more durable. As regards decentralization, the nine institutions are more alike than different. All delegate individual credit and savings decisions to local committees and caisse agents who approve loan applications and follow up on repayment. Caisse staff are selected locally in about half of the institu- tions and centrally in the other half. Central administrative functions are carried out in head offices. All organizations but one have assigned policymaking functions to central offices that set operating policies, including interest rates, loans-to-savings ratios, use of collateral, loan sizes, and repayment policies. The important exception to this is the Caisses Villageoises du Pays Dogon, which differs from the rest in the near-complete autonomy of each village bank and in the very flat administrative structure. Most policy decisions are made by caisse-level committees, including the setting of interest rates, the determination of maximum loan sizes, and the disposition of earnings. The underlying premise here is that local authorities not only are fully capable of making all policy decisions but, indeed, must be empowered to make these decisions if the caisse in their village is to endure. All institutions are highly democratic at local levels in that the main decisionmakers are almost always elected by members or clients. Local assemblies of entire villages or of all members or borrowers are convened at least once a year to receive a status report and to elect members of the committees that make most of the credit decisions. This approach presum- ably leads to a high degree of local accountability for governance of the local unit and for lending decisions. Above local levels, institutions vary in client participation. Reseau des Caisses Populaires, Kafo Jiginew, and Caisses Villageoises are perhaps strongest in this area. Local caisses elect representatives to unions that, in turn, elect representatives to governing boards of the institutions. In the other institutions, there appears to be little in the way of client or member representation above the local level. Regional offices are managed by staff employed from the center, and most policy decisions are made by head staff who typically are managed by an expatriate from the implementing NGO. Some also have governing boards. Savings services As noted in chapter 3, savings mobilization among the poor has been emphasized because savings services are valued highly by low-income people, many of whom need a secure place to store surplus funds and a 50 Overviews mechanism by which they can accumulate wealth; are generally preferable to debt, particularly in the case of poor people; offer microfinance institu- tions a local, and presumably sustainable, source of loan capital; and when used as loan funds, should engender highly responsible use of funds. From the clients' points of view, the highest quality savings services offer convenience, security, liquidity, and financial retum. All institutions reviewed offer savings services to clients, except for the two credit programs. Within the membership-based organizations, the number of savers generally matches the number of members, which reflects a common approach of requiring at least minimal savings to become a member (see table 4.4). Total savings are large among the credit unions, equaling or exceeding their loan portfolios. Average savings accounts are small in keeping with the low income levels of most clients. Small average deposits mask considerable concentration of savings in Credit Mutuel and Credit Rural in Guinea, both of which accept deposits from civil servants and other middle-income people. With the exception of Village Banks Nganda, demand deposits can be withdrawn freely by savers. Two institutions do not pay interest on demand deposits, two pay interest rates of 4-5 percent a year, and the two credit unions in Guinea pay annual rates of 13-15 percent. Only three institutions offer term deposits, two at annual rates of 5.0-6.5 percent and one (Caisses Villageoises du Pays Dogon) at an annual interest rate of 22 percent. Not surprisingly, 94 percent of deposits in the Caisses Villageoises are term deposits, offering an interesting bit of counter evidence to the argument that poor entrepreneurs are not particularly sensitive to interest rates. As noted in chapter 3, the wisdom and effectiveness of forcing members or borrowers to save are questionable. This can be viewed both from the side of the institution (which is trying to increase its pool of funds, secure its loans, and maximize its revenues) and from the side of savers (who want to secure their savings, maintain liquidity, and maximize return). Several of the nine institutions rely heavily on mandatory savings. Village Banks Nganda offers the strongest example of this practice. Borrowers are required to save 35 percent of their loan amount, deposited in equal monthly shares for the duration of the loan. These savings accounts earn no interest, they cannot be withdrawn until the loan is fully repaid, and the borrower ceases to be a member of the organization when savings are withdrawn. Most of the credit unions require minimal opening deposits to establish membership, several months of saving, and a percentage of loan amounts in deposits for the life of the loan. Requiring deposits to secure loans probably is a reasonable practice, but experience with other mechanisms such as character-based lending has shown that the proportion of required savings might not need to be as high as currently set by some of these credit unions.6 A Profile of Selected Microfinance Institutions 51 Table 4.4 Savings services in nine microfinance institutions in West Africa Average savings Total Nominal Savings as a account savings interest percentage of Data as Number (U.S. (U.S. rate Percentage of outstanding Institution of' of savers dollars) dollars) (percent) term deposits loan portfolio PRIDE b n.a. n.a. n.a. n.a. n.a. n.a. n.a. Credit 12/94 31,000 42 1.3 15 0.' 79 Rural million Credit 12/94 45,000 112 5.0 13 0 248' Mutuel million Guin6e Credit 8/94 22,000 70 1.55 4.5-6 37' 146' Mutuel du million Senegal Village 12/94 600 45 27,000 0 100 41 Banks Nganda Reseau des 9/94 64,000 76 4.9 0 0 387 Caisses million Populaires Sahel n.a n.a. n.a. n.a. n.a. n.a. n.a. Action b Caisses 12/94 2,450 76 108,000d 0-22' 94' 59' Villageoises Kafo Jiginew 10/94 20,670 43 888,000 4-6.5 35' 10.1' n.a. Not applicable. a. Except where otherwise noted. b. PRIDE and Sahel Acion do not take savings. c. As of December 1993. d. Average savings for 1994. e. On term deposits, 22 percent interest is paid. No interest is paid on demand deposits. Credit services Reflecting their greater emphasis on mobilization of savings than on lending, the credit unions have far fewer borrowers than savers (table 4.5). The two village banking systems focus more on the provision of credit, and the number of borrowers equals or exceeds the number of savers. The two credit programs are only lenders with no savings mobilization. Average loan sizes are small for the most part: less than $100 for three programs; $100-$350 for five; and $1,000 for Credit Mutuel de Guinee. Loan repayment rates are very high in all institutions, except for Credit Mutuel de Guinee. Nominal interest rates range from a low of 16 percent in Reseau des Caisses Populaires to a high of 54 percent in the Village Banks Nganda. Effective interest rates tend to be somewhat lower, with five programs charging between 19 and 26 percent a year, three between 32 and 39 percent, and one charging at 54 percent.7 52 Overviews Table 4.5 Credit services in nine microfinance institutions in West Africa Average Average loan Maximum Effective Number of number of size loan size interest Quality of the Data loans at loans out- (U.S. (U.S. rate loan portfolio Institution as of year-end standing dollars) dollars) (percent) (percent)' PRIDE 7/94 - 3,500 200 500 37 100 Credit Rural 12/94 - 13,180b 100 1,000 32 97 (12/93) (12/93) (12/93) Cr6dit 12/94 - 2,800 1,000 10,000 22 80 Mutuel de (12/93) Guinee Credit 8/94 5,625 + 2,377 132 300 20 98 (< 90 days) Mutuel du (12/93) to 1,000 (12/93) S6negal' Village Banks 12/94 600 + 871 68 300 54 100 Ngandad Reseau des 9/94 - 3,710 336 6,000 19 95 (<90 days) Caisses Sahel Action 12/94 13,174 9,526 b 48 120 26 100 (12/93) (12/93) Caisses 12/94 9,767 4,109 b 43 - 39 94 Villageoises (12/93) KafoJiginew 9/93 9,320 5,048b 105 116 19 98 - Not available. a. Quality is determined by the percentage of loans paid in full within thirty days of the due date. All figures are calculated on this basis unless otherwise noted. b. Estimate based on average outstanding loan amount and average loan size. c. Calculated on an eight-month basis and based solely on rural credit loans. d. Calculated on a six-month basis and based solely on loans from the external account. Terms and conditions of loans differ across programs. Four programs lend only to clients organized into solidarity groups, four lend mainly to individuals, and one has separate loan products for individuals and for groups.8 Loan terms typically are six to twelve months, with an overall average of eight months for the nine programs. With only a few exceptions, payments of principal and interest are due at the end of the loan terms instead of on a weekly or monthly basis. Programs tend not to limit borrowers' use of funds, but rather to offer several types of loans geared to different uses and needs of rural clients. Loans for agriculture and for agroprocessing predominate, followed by trading activities. Several programs are beginning to offer loans for medium-term equipment. Loans are approved when applicants meet one, and usually several, of the following requirements: they are members who have been prescreened, they are guaranteed by fellow members of a solidarity group, they offer collateral as security for their loan, and they have the requisite savings on A Profile of Selected Microfinance Institutions 53 deposit. Collateral requirements are liberally interpreted to include virtually any personal belonging, but group guarantees and savings requirements appear to be strictly enforced. Only two programs construct balance sheets and analyze the cash flow of applicants as part of credit decisions. Performance Performance of the nine microfinance institutions analyzed is evaluated on the basis of two commonly used parameters: outreach and sustainability.9 Outreach is judged by the number of people served, the extent to which they are poor, women, or otherwise traditionally underserved, and the quality of services provided. Sustainability is assessed in terms of the efficiency of services, the degree of financial self-sufficiency, and the durability of organizational structure. Outreach The objective of microfinance programs is to reach large numbers of poor people with financial services; the scale of operations indicates the extent to which this has been achieved. As pointed out in chapter 3, achieving scale in both lending and savings operations is also a key factor in achieving financial sustainability: if loans are very small, it takes a large number of them to generate substantial revenues; if average savings accounts are very small, it takes many savers to mobilize significant funds. Operational scale is a mixed picture for these nine microfinance institutions. From one point of view, relatively large organizations have been built. All but two have membership or client rolls that exceed 10,000, and five count more than 20,000 people as members. Large-scale member- ship has created more than 20,000 savers among four of the seven institu- tions that take deposits. But from another point of view, large organizations have produced 10,000 or more borrowers in only two institutions (Sahel Action and Cr6dit Rural). The remainder lend to several thousand clients. In sum, most of the microfinance institutions reviewed have been successful in building large institutions, but only one (Kafo Jiginew) has been able to translate this growth into large numbers of borrowers and savers. Several points are relevant to the issue of scale. First, some institutions, notably the credit unions, have deliberately kept their loans-to-savings ratios fairly low, at roughly 0.3-0.5. This can be seen as prudent, because all savings are demand deposits and these institutions have no access to long- term sources of funds that would support higher levels of exposure. As regards covering risks, some observers feel that these institutions could safely lend a larger proportion of their savings, in part by relying more heavily on some of the proven nontraditional means of securing loans, such as character-based lending. 54 Overviews Second, achieving large scale in remote, sparsely populated, rural areas with poor infrastructure (without incurring unacceptably high costs) might be close to impossible in some cases. For example, the Caisses Villageoises du Pays Dogon in Mali estimates that 44 percent of adults living in villages with banks are members, as are 19 percent of adults in adjacent villages. Kafo Jiginew estimates that total membership represents 43 percent of all potential members (which includes all adults). These cases illustrate that capturing large shares of the market will not necessarily result in significant scale of operations, especially where populations are small. A third issue that arises in connection with operational scale is the original design and vision of the institution, that is, whether it was designed to be local, regional, or national in scope. Organizational structures that are designed to be regional can run into problems if they attempt to become national. To expand without incurring significant costs and without increasing in bureaucracy, institutions have to develop replicable, fairly standardized village-level units. Initial governance structures may be poorly suited to larger systems. Large networks need much more sophisticated management information systems than do small networks. In sum, the growing pains associated with achieving scale are significant and should not be underestimated. Helping microfinance institutions manage growth is a frontier in this field, and the institutions reviewed here are excellent examples of the need for this assistance. All institutions selected here are serving hard-to-reach clients, including the very poor, the uneducated, women, and people in remote locations. Average loan sizes are less than $100 for three institutions and between $100 and $350 for another five.'" The average savings account is less than $100 in all cases but Cr6dit Mutuel de Guinee. Another measure of clients' income levels is the ratio of average loan size to per capita GDP. Here, loan sizes in all programs but two average less than 50 percent of per capita GDP, suggesting that most are reaching the very poor."' Precise figures concerning clients' educational levels are not available, but field visits suggested that the great majority are illiterate (and all programs have designed procedures and forms to conform with this reality). A third to half of clients are women in most programs, with the highest share being 95 percent in Sahel Action and the lowest being 5 percent in Kafo Jiginew. Most institutions, such as Credit Mutuel du S6negal, have located regional offices in provincial capitals with their systems of caisses spread throughout more remote areas. In other cases, such as Caisses Villageoises du Pays Dogon, the entire operation is located in a remote area. In most cases, institutions serve clients who live in areas where roads, electricity, and telecommunications are minimal or nonexistent. For all of these reasons, these institutions get high marks for serving the hard-to-reach. A third measure of outreach concerns the quality of services provided, that is, the extent to which services meet the needs of the target group. A Profile of Selected Microfinance Institutions 55 Answering this question involves evaluating the accessibility of services, the liquidity and financial return associated with savings, and a series of characteristics of credit programs, including ongoing availability, restric- tions concerning the use of loans, the level of security required, and the processing time. Members or clients seem to have quite good access to services, particu- larly in light of the remote, rural locations of many caisses. Strong consider- ation has been given to locating caisses where they can serve a large number of surrounding villages. Caisses in more populated areas tend to be open daily; those in very remote areas usually are open one or two days a week, usually coinciding with market days. In the case of Sahel Action, mobile agents visit villages in their assigned areas on a scheduled basis. In sum, clients do not have to travel exorbitant distances to reach local units, nor are they asked to wait a long time to access services. Among the seven institutions that mobilize savings, the quality of service is quite high in terms of liquidity and adequate in terms of financial return. As noted, six of the seven accept demand deposits that clients can access at any time, and four of the six pay interest on savings. Three offer term savings, and all accept very small deposits (including the two programs with minimum balances and transaction amounts). Several institutions rely fairly heavily on mandatory savings, mainly from borrowers. Involuntary savings have been criticized by practitioners who feel that they fail to meet the need of low-income savers for liquidity and that they impose an unnecessarily heavy burden on clients. In this case, some institutions may be requiring higher levels of deposits than are really needed to secure loans, given the additional use of collateral and character- based guarantees. In addition, forced savings probably are not a good source of funds, because borrowers who are required to deposit large percentages of loan amounts will soon cease to need loans and will simply withdraw their savings in the medium term, as could happen in Village Banks Nganda. Credit services also can be rated quite highly for the most part. Most institutions offer several types of loans to accommodate different needs and to compensate for the fact that certain loans-such as agricultural loans-are only available at certain times of the year. In most cases, clients can access at least one type of loan on an ongoing basis. For example, Kafo Jiginew offers agricultural loans only once a year, but it also offers short-term loans for trade and medium-term loans for equipment purchases. A common practice seems to be to take agricultural loans in the month that they are offered and to deposit them in savings accounts to be drawn down when needed. Most lending is character-based. Security for loans varies from personal cross-guarantees among group members to physical assets that are used as collateral. In general, a good balance seems to have been struck between the 56 Overviews need of institutions for security and the ability of individuals to offer something of value to guarantee their loans. Processing time between loan application and disbursement ranges from a week to forty days. More than two weeks would seem overly long when compared to the performance of effective programs elsewhere, and a good number of the institutions reviewed exceed this limit-in part because of poor infrastructure and telecommunications and in part because of seemingly unnecessary extra steps, such as requiring that loans be approved at the regional level and that funds be requested from central offices. Clients' behavior is also a good indicator of the quality of services. Strong growth in the number of members or clients over the past several years suggests that many people have judged these services to be worthwhile. The fact that people are willing to pay fairly high interest rates for loans and, in some cases, to receive no interest on savings is another indication that these services are valued. Very high repayment rates across all programs but one demonstrate a respect for the process and for the institutions, as well as the effectiveness of offering repeat loans after repayment of initial loans. Sustainability The three indicators of sustainability considered here are operational efficiency, financial viability, and institutional durability. The ability of an institution to achieve full (or nearly full) financial sustainability depends critically on how efficient it is. There is no single measure of efficiency for microcredit institutions but rather a series of indicators that, taken together, are highly indicative. In this case, two sets of indicators are employed. The first set looks at overall institutional efficiency by calculating operating costs as a percentage of the loan portfolio, operating costs minus expatriate costs as a percentage of the loan portfolio, salaries as a proportion of outstanding loan portfolios, and salaries as a multiple of GDP (see table 4.6). The second set looks at the productivity of staff members in service delivery and includes the number of savers, the value of savings, the number of loans, and the value of the average portfolio-each on a per staff member basis (table 4.7). Ratios of operating costs to average loan portfolio offer insights into operational efficiency. High ratios suggest high administrative costs, small loan portfolios, or some combination of the two. Lowering this ratio obviously depends on decreasing the numerator (lowering costs) and on increasing the denominator (expanding the size of the loan portfolio). Starting on the cost side, an in-depth look at disaggregate costs suggested some explanations for inexplicably high operating costs in some institutions. Three main sources of apparently high costs were identified: operating expenses in head offices, staff salaries, and expenses associated with expatriate staff. A Profile of Selected Microfinance Institutions 57 Table 4.6. Indicators of institutional efficiency in nine microfinance institutions in West Africa Salary Operating costs as a costs minus Fee plus percent- Operating expatriate interest age of Average costs as a expenses as a income as a average salary as a percentage percentage of percentage of annual multiple of of average average average Data loan GDP per annual loan annual loan annual loan Institution as of portfolio capita portfolio portfolio portfolio PRIDE 7/94 90 9.4 221 186 39 Credit Rural 12/93 30 6.5 166 109 16 Credit 12/93 14 2.1 122 49 23 Mutuel de Guin6e Cr6dit 12/93 22 1.8 208 63 20 Mutuel du Senegal Village 12/94 5 _a 33 17 27 Banks Nganda Reseau des 9/94 - _ a _ _ _ Caisses Populaires Sahel Action 12/93 48 6.4 200 164 26 Caisses 12/93 24 10.6b 202 100 39 Villageoises Kafo 9/93 19 8.9 86 61 19 Jiginew - Not available. a. Only 1994 (after devaluation) data are available for average salaries. Because the only official figures for GDP per capita are for before devaluation, these figures could not be calculated. b. The average salary as a multiple of GNP per capita is extremely high in the case of Caisses Villageoises du Pays Dogon because this institution employs a small number of full-time staff and relies on thirty-seven volunteers to carry out most credit and savings operations. Figuring in volunteer labor drops this ratio to 2.9. Field teams were impressed with the relatively large number of local caisses and regional offices or unions that were breaking even or making profits. Sources of high operating costs could be traced to headquarters and central offices. Administrative costs that seemed inordinately high in some central offices included expenditures on travel, vehicle maintenance, communications, and office supplies. In some situations, the high proportion of operating costs accounted for by these items perhaps cannot be reduced, 58 Overviews Table 4.7 Staff productivity in nine microfinance institutions in West Africa Average out- Number of Value of sav- Number of standing loan savers per ings per staff loans out- portfolio per staff Data staff member standing per member (U.S. Institution as of member (U.S. dollars) staff member dollars) PRIDE 7/94 - 52 5,179 Credit Rural 12/94 258 10,800 110 a (12/93) 10,833 a (12/93) Credit 12/93 298 33,113 19 7,285 a (12/93) Mutuel de Guin6e Credit 8/94 265 18,627 29 (12/93) 6,385 a (12/93) Mutuel du Senegal Village 12/94 171 7,657 249 16,914 Banks Nganda Reseau des 9/94 582 44,545 34 8,773 Caisses Populaires Sahel Action 2/95 - - 167 4,035 a (12/93) Caisses 12/94 1.75 7,714 b 294 12,690 Villageoises Kafo 10/94 517 22,000 126 a (9/93) 13,250 a (9/93) Jiginew - Not available. a. 1994 data are not available, so this ratio was calculated by dividing 1993 data by the total number of staff members in 1994, producing a reasonable indication of staff productivity. b. Based on the average outstanding savings for 1994, rather than outstanding savings as of December 1994. because operating costs clearly are higher in some of these countries than elsewhere. Others probably could be reduced. These kinds of distinctions can only be made internally by persons who understand the individual context. One area of potential savings that stands out is the notable lack of dependence on volunteers and on donations of goods and services-two hallmarks of nonprofit institutions in industrial countries. Caisses Villageoises du Pays Dogon and Kafo Jiginew in Mali are exceptions to this in that they make extensive and highly efficient use of locally donated physical space and of local volunteers (villagers construct their own caisses as their contribution). The other institutions generally limit their use of volunteers to credit committees and pay for everything else, including rent A Profile of Selected Microfinance Institutions 59 or construction costs of local and central offices and all staff. One wonders if costs could not be cut through aggressive campaigns to increase donations and volunteerism. Salary costs have been closely associated with the ability of an institution to become financially sustainable. Views are split regarding the most effective employment strategies in microfinance institutions. The majority view seems to be that employees should be sought whose level of formal training closely matches the job requirements; for example, local agents should be local people who are literate and have basic arithmetic skills. A minority view is that higher levels of skills should be sought, particularly because these institutions handle larger volumes of funds. In these institutions, salary costs (minus all expatriate costs) as a percentage of the average outstanding loan portfolio vary from a high of 90 percent in PRIDE followed by 48 percent in Sahel Action to a low of 5 percent in Village Banks Nganda (table 4.6). In addition to the total number of staff, the major difference between institutions with higher salary costs and those with lower salary costs is whether or not they emphasize university graduation among their staff. Those who hire more university graduates have much higher costs than those who do not. Average salaries as a multiple of GNP per capita indicate the extent to which staff salaries in these institutions are in line with country norms. According to this indicator, Caisses Villageoises and PRIDE pay the highest salaries. Salaries in Caisses Villageoises are only for the small number of central staff, because all field staff are volunteers. The impression that Sahel Action pays high salaries is partially negated because, as a multiple of per capita GNP, they are about average. This is a good example of a case where the size of the loan portfolio is the source of the apparent inefficiency (although productivity measures also need to be factored in). A third source of high costs in some institutions is expatriate expenses. The fourth column in table 4.6 shows operating costs as a percentage of per capita GDP when expatriate expenses are excluded."2 Institutions with proportionally the highest expenses for expatriates include Credit Mutuel du S6negal, Caisses Villageoises du Pays Dogon, and Credit Rural and Credit Mutuel de Guinee. This issue is moot in the case of Caisses Villageoises, because the expatriate director left in the summer of 1995, but figures for the two Credit Mutuels, and to a somewhat lesser extent for Cr6dit Rural, indicate that the expenses of expatriates as a share of the average outstanding loan portfolio are very high and likely will be significant barriers to these institutions' attaining true financial sustainability. The denominator of this ratio is the size of the outstanding loan portfolio, and relative efficiency is improved if this portfolio increases in size. This can be achieved in two ways: the number of borrowers can increase, or the average loan size can increase. The first option takes us back to issues of 60 Overviews scale and to the conclusion that some institutions clearly could extend more loans than they currently do; institutions in sparsely populated areas would find this more difficult. Expanding the number of borrowers is a question, first, of deepening the reach in locations where caisses have already been established and costs are mainly fixed and, second, of investing in new caisses in additional locations. Significant expansion is not an easy proposition. Another strategy to increase the loan portfolio is to increase the average loan size; this too raises important issues. The ability to raise the average loan size depends to a large degree on the institution's commitment to serving only the very poor and, to a lesser degree, on the availability of creditworthy clients who can effectively use larger loans. Average loan sizes can increase by choosing to include a higher-income strata of the population, by including the less poor among the very poor as clients, or simply by gradually increasing the loan size of current clients as their businesses mature. When the issue of raising average loan size arises, the common fear is that the very poor wilL be abandoned. Evidence indicates that this is not a valid fear in most cases (see Christen, Rhyne, and Vogel 1994). Programs that serve the very poor with very small loans can raise the average loan size without exorbitantly high per loan costs if they can achieve large scale (assuming that other factors are in order). Increasing the size of the loan portfolio is difficult for the institutions reviewed here for a series of reasons. None has a sufficient source of long- term capital. Savings are small due to the low incomes of the clientele, almost all savings are demand deposits that cannot be re-lent in significant amounts, and none has been able to obtain the credit lines from the commercial banking sector that are needed to ensure a steady flow of long- term capital."3 As a result, almost all are forced to ration their credit funds, and demand for loans among their members is far from satisfied. Some have begun to increase their loan portfolio by including better-off people among their savers. Without exception, management in these institutions expressed their commitment to not abandoning the poor in their quest for more loan funds. In some cases where institutions are located in extremely poor areas, significantly increasing savings is not a viable option. At a general level, increased competition probably would force institu- tions to become more efficient. Microfinance institutions are proliferating in West African countries, but the trend is for donors to carve out territory to avoid competition. As long as programs avoid using subsidies to undercut one another with below-market interest rates, a more competitive environ- ment could go a long way toward lowering costs. The productivity of staff members varies greatly (table 4.7). Caution must be taken here to assess staff productivity by examining both the number of savers and the number of borrowers per staff member, because some institutions devote a much larger share of their staff time to savings A Profile of Selected Microfinance Institutions 61 transactions than to loan transactions; this is true of Reseau des Caisses Populaires and Credit Mutuel du Senegal. In the case of PRIDE, staff spend half their time on training, which skews staff per loan ratios. Several facts about staff productivity become clear from these figures: * Staff in Reseau des Caisses Populaires, Kafo Jiginew, and Credit Mutuel du Senegal manage large amounts of savings on a per person basis. * Loans and savings per staff member are very high in Caisse Villageoises du Pays Dogon and in Kafo Jiginew, in large part because they rely heavily on volunteer and part-time workers. Staff in Sahel Action also handle a large number of borrowers, in part because all borrowers are in groups. Program managers recognize the importance of staff productivity, and most have put in place mechanisms to improve staff performance. Several programs offer incentive packages to staff that involve significant bonuses for strong performance, defined in part by the number of clients that are served and in part by repayment rates. Others have sought to delegate more responsibility to volunteer credit and management committees so that paid staff can serve more clients. A second measure of sustainability is financial viability, which is determined by the ratio of total revenues to total expenses. Revenues include net interest income, fees, and interest on investments. Expenses are categorized here as operating expenses (all administrative costs, deprecia- tion of fixed assets, and losses from loan defaults) and as total expenses (all of the above plus the cost of loan funds and allowance for inflation). It follows that strategies to improve financial viability would focus on increasing revenues, decreasing costs, or both. All but two of the nine institutions cover roughly a third to half of their operating costs with their revenues (see table 4.8). This means that the dependence on subsidies remains high in most cases. Subsidies come overwhelmingly from donors, and it is important to separate out the "good" subsidies, such as investments that build institutional capacity, from the "bad" subsidies, such as investments that are simply spent down with no lasting effect. Because interest income from lending is by far the main source of income, the search for ways to increase revenues begins with interest rate policies. The experience of some of the most successful microfinance institutions has been that interest rates must be set at roughly 20 percent above the cost of funds (including adequate adjustment for inflation) for an efficient microfinance institution to break even. Less efficient (and smaller) programs have to charge even higher rates if they are to cover their costs. Nominal interest rates in these institutions vary from a low of 16 percent to a high of 54 percent. Effective annual interest rates vary from 19 to 54 percent, with 62 Overviews Table 4.8 Indicators of operational and financial sustainability in nine microfinance institutions in West Africa Fee plus Fee plus interest Fee plus interest interest income as a Fee plus income as a per- income as a percentage of interest centage of total percentage operating costs income as a costs minus of operating minus expatriate percentage expatriate Institution Data as of costs expenses of total costs expenses PRIDE 7/94 26 30 26 30 Credit Rural 12/93 24 36 22 32 Credit Mutuel 12/93 34 87 26 47 de Guin6e Credit Mutuel 12/93 18 57 17 48 du Senegal Village Banks 12/94 160 307 123 229 Nganda Reseau des 9/94 - - 53 - Caisses Populaires Sahel Action 12/93 13 16 12 14 Caisses 12/93 19 39 18 33 Villageoises Kafo Jiginew 9/93 29 40 29 38 - Not available. most charging between 20 and 40 percent. Since only two of these institu- tions are near or beyond break-even, the question arises as to whether interest rates should be raised. If pricing is considered "correct" when interest rates are set to cover lending costs fully, then all but two of these programs have not set their interest rates high enough. The question then becomes how high rates can and should be raised in the interest of cost recovery. Little analysis has been done on the issue of how high interest rates have to be before they exclude the poor, although the common assumption is that they can be very high without doing so. National political realities militate strongly against high interest rates for very poor people. As described in annex 2, the new law governing mutualist credit and savings institutions will cap interest rates at twice the discount rate. If current usury laws are implemented as part of this new law, interest rates will be limited in most countries to around 20 percent, which clearly is far too low to allow the institutions reviewed here to break even. In fact, implementation of this cap would virtually preclude the possibility of their breaking even. A Profile of Selected Microfinance Institutions 63 There are reasons to believe that most of these institutions are well on their way to covering a larger share, if not all, of their operating costs. First, operating costs have been very much inflated by the tremendous expansion that has taken place over the course of the last several years, particularly the large increase in the number of new caisses. Time is needed for the payoff from this investment to be realized. Second, where trends in cost recovery are evident, they are quite positive in all cases; these institutions are covering a far larger proportion of their costs than they were previously. Regardless of their efficiency, their scale, and their pricing policies, the ability of these institutions to cover costs is undermined by the extreme poverty of their clients and by the costliness of doing business in their environments; for example, roads are bad, and four-wheel drive vehicles are a must, even though they depreciate very quickly in the climate of the Sahel. A third indicator of an institution's sustainability is the durability of its organizational structure. An assessment of this indicator would look at the balance that has been achieved between social and financial objectives; issues of governance, such as levels of decentralization and participation; staff incentive and training systems and plans for local people to assume leadership; and factors in the environment that increase the institution's chances of survival in the future. Striking a sustainable balance between social and financial goals is both critical and difficult. As long as donors are willing to make up the shortfall in revenues, institutions are free to emphasize social objectives (alleviating poverty) over financial objectives (breaking even). If donors should withdraw their funds, these institutions would be forced to give higher priority to financial objectives and perhaps to sacrifice some of their social goals. Striking a sustainable balance might mean making every possible effort to maximize financial sustainability without abandoning the hard-to- reach clients, with donors making up small shortfalls on a sustained basis. Governance issues could affect the ability of institutions to endure if the balance of centralization and decentralization is not sustainable. Lending decisions are handled locally in all cases reviewed, but key policy decisions are made centrally with the exception of Caisses Villageoises. The extent to which this arrangement will prove durable probably depends, in part, on the degree to which those who manage the institution are representative of the members or clients. Organizations that have structured in high levels of participation such as Reseau des Caisses Populaires, Kafo Jiginew, and Caisses Villageoises could prove stronger in the long run than those that have not. Staff training is an investment that should pay off in the medium and long terms, and most of these institutions invest heavily in training their staff, especially agents who work at the local level. A related issue is the speed with which local people take over management of the institution once they are properly trained. On this point, the institutions reviewed vary 64 Overviews widely. Those that are fully managed locally include the Caisses Villageoises, where the expatriate manager was training local Malians to take over central functions in the summer of 1995, and Kafo Jiginew, where the board of directors asked the expatriates to leave. Those that continue to rely partially on expatriates include Reseau des Caisses Populaires, PRIDE, and the Village Banks Nganda. Those where expatriates appear to manage (directly or indirectly) include Sahel Action, Credit Mutuel du Sen6gal, and Cr6dit Mutuel and Credit Rural in Guinea. Long-term sustainability probably depends in some part on expatriates giving up the helm to local people. A final determinant of the ability to endure in the long term concerns the environment within which institutions operate. The degree of macroeco- nomic stability, as reflected in the level of inflation and the stability of exchange rates, has a major impact on the survival of microfinance institutions. To this can be added the extent to which financial regulations, especially interest rate caps, foster or handicap nonbank financial institu- tions. Clearly, higher ratings on each of these counts are more conducive to the maintenance and growth of microfinance institutions. The external environment in these West African countries is difficult. As noted in chapter 2, per capita GDP is low, and growth rates have been either low or negative in recent years. Long-standing macroeconomic stability was shattered by the devaluation on January 1, 1994, and the impact on these programs was unclear. At the time of the appraisals, high inflation, which erodes the capital base of microlenders, had not materialized following a sharp spike after the devaluation but neither had it fully dissipated. The new law for mutualist savings and credit institutions contains several provisions that, if applied to the institutions reviewed here, would seriously-perhaps critically-undermine their efforts to achieve financial sustainability, first by placing caps on interest rates, and second by limiting the amount that institutions can lend to twice the level of their savings (see box 4.1). In sum, achieving sustainability in this region is a complex and difficult proposition that depends on all of the factors discussed in this section. Remote, underdeveloped, and sparsely populated rural areas are perhaps the most challenging environment of all for microfinance to work well. In addition, the economic and regulatory environment is not an easy one, especially as regards the new law. In this context, the nine institutions here might have performed as well as possible. Future challenges Four issues emerge when these nine case studies are summed up and their futures are anticipated. First, it is interesting to compare the institutions reviewed here with the best microfinance institutions globally to see how and where they differ. Second, the largest current and future challenge may be effective management of the tremendous growth that is under way in A Profile of Selected Microfinance Institutions 65 Box 4.1 Key features of the new law governing mutualist-type institutions The law subjects institutions to existing usury laws, which cap the effective rate of interest at no more than twice the discount rate. It also includes the following ratios designed to ensure prudential operation and to reduce risk: * Loan portfolio must not exceed twice the deposits of the members * 15 percent of annual operating surpluses must be allocated to a general reserve * Loans to one member cannot exceed 10 percent of the value of the members' deposits * Short-term assets must at all times equal at least 80 percent of short-term liabilities. these institutions. Third, the importance of the new law governing savings and credit institutions that has been passed and is coming under implemen- tation should not be underestimated. And fourth, it is important to ask what would happen if even the best microfinance institutions in a large region cannot become fully financially sustainable. Are there situations where sustainability cannot be achieved? If so, should the institutions be supported anyway? The global context At a general level, all nine of these institutions are very much in the mainstream of best practice in the field of microfinance. Returning to the ten principles for sustainable lending to microentrepreneurs presented in chapter 3, we see that these programs receive high marks on almost all counts. They are located near their clients but in the largest catchment areas possible; they use lending technologies that are simple, well-tailored to the cultural environment, and inexpensive for both lender and client; they have employed very effective techniques for obtaining high repayment rates; most include savings, which meets a critical need of many poor people; they have avoided weighing down their financial services with additional social services; and they price their loans far above commercial lending rates, though not at full cost recovery. They have problems reducing their administrative costs, especially in central offices, as do many microfinance institutions. These West African institutions have an added cost that most programs elsewhere have managed to minimize over time, and that is the cost of expensive expatri- ates. Programs in other regions also employ expatriates, but for shorter periods of time and rarely in direct management positions. Where the loan portfolio is large, the proportion accounted for by expatriate expenses can 66 Overviews be relatively small, but in the case of institutions with relatively small loan portfolios, expatriates account for a very large share of the loan portfolio. In some cases, it was obvious from field visits that expatriates either are on their way out or have a very light presence. But in others, their presence seemed to be permanent, and this does not bode well for attaining self- sufficiency, either managerial or financial. Comparing these programs with those considered among the best in the world, it is easy to see where they are fully comparable and where they fall short."4 On outreach, the West African institutions fall short in terms of total assets and number of clients: they do not reach a large number of borrowers. On depth of outreach, these programs excel in serving the very poor and other hard-to-reach groups. On quality of service, as defined in this report, the West African institutions are very much on par with the institutions analyzed by the USAID study. On measures of sustainability, the West African programs fall below the best institutions, both in efficiency and financial viability. Operating costs as a percentage of loan portfolios are relatively higher, and revenues cover a substantially smaller share of operating costs. Field staff are less produc- tive than the norm in some programs, in part due to the physical difficulty of reaching clients. These weaknesses can be traced to many factors, including small scale of operations and small average loan sizes. Managing growth The largest challenge confronting these institutions could be management of their extremely rapid growth. Most have evolved from small NGO- managed projects to much larger financial institutions over the past three to five years, and the institutional demands are tremendous. The following are among the most important demands: - Financial management. As they grow larger, issues of liquidity, cash flow, and classifying the risks of borrowers become more important. - Management information systems. With only few exceptions, these institutions have extensive outreach and financial statistics. But too often, they are unable to consolidate them properly for lack of computers and appropriate software. They are also unable to use their data as sound -planning indicators, because in-house financial analytical skills are inadequate to meet the needs of the now much larger institution. * Financial planning. The construction of realistic business plans that reflect the rapid expansion under way may not be sufficient. * Staff training. Although training is a standing priority for most institu- tions, growth is so explosive that even large investments in training A Profile of Selected Microfinance Institutions 67 cannot keep up with the need, and insufficient training can lead easily to problems with the quality of the portfolio and governance. * Governance issues. Good governance becomes much more critical as institutions grow. Boards of directors and management need to be strong, the democratic process becomes more important, and issues associated with decentralization become critical. The new legal framework The new law governing the operations of credit union-type institutions could pose serious problems for some of these institutions as well as for many others in this region. This law establishes the legal basis of mutualist credit and savings institutions and spells out how they should operate. Many provisions appear quite reasonable for credit unions, but two are very worrisome: a maximum loans-to-savings ratio of two to one (for the institution as a whole) and a cap on interest rates at twice the discount rate (once institutions are registered and hence subject to usury laws). Implemen- tation of these provisions will block virtually all of the institutions reviewed here from attaining financial viability. This law seems to assume that all microfinance institutions should be credit unions, because no provisions are made for other types of microfinance institutions that take deposits and make loans. This approach could inadvertently narrow the range of institutions that are recognized as effective sources of financial services for the poor. It leaves noncredit union institutions with the unpalatable choice of registering under this new law and becoming subject to its provisions or of not registering and forgoing the advantages that derive from inclusion. In addition, the same or a similar law is reportedly close to being passed in Central Africa and in several individual African countries. There may still be time to join the dialogue in these countries. Attainingfinancial sustainability The widespread collapse of first-generation microcredit programs when donors directed their funds elsewhere has led to a very strong emphasis on financial sustainability of microfinance institutions. The thinking is that these institutions will eventually be able to function independently of all subsidies, and their operations will then not be jeopardized by the with- drawal of donor funds. The usual assumption is that all microfinance institutions can eventually become fully financially sustainable if they scale up sufficiently to generate adequate revenues, price their products and services properly, and employ proven lending and savings technologies. A widely shared belief is that microfinance institutions can and should reach full financial sustainability. 68 Overviews The performance of the nine microfinance institutions reviewed here raises questions about these commonly held beliefs and assumptions. The first asks whether the possibility exists that some programs will be unable to reach financial sustainability because of the unavoidably high costs associated with reaching the very poor in difficult locations. In West Africa, these costs would include the expense of delivering services to clients who live in remote, sparsely populated areas with almost no basic infrastructure; the costs of serving very poor clients who are only able to make very small savings deposits and absorb very small loans; and the costs of starting from scratch with a poorly developed financial sector. Suppose that a microfinance institution attains a very high level of efficiency but that revenues are still insufficient to cover these high costs. This possibility raises the issue of the "purposeful" subsidy, which is one that is highly targeted to fill shortfalls in revenues that have been judged as essentially inescapable. An excellent example of a purposeful subsidy comes from Caisses Villageoises du Pays Dogon. This village banking program serves very poor people with very small loans in a highly cost-efficient manner. Local village banks cover their costs, and some are profitable, but the costs of the soon-to-be minimalist central unit cannot be covered by revenues from the interest on loans (which is set at 40 percent a year) because of the small size of both the loans and the total loan portfolio. For example, 40 percent of the operating expenses of the central unit are the costs of one four-wheel-drive vehicle that is critical to reach banks in remote locations across almost nonexistent roads. It will be next to impossible for this small system to cover the $20,000 cost of this one vehicle. Is there not a strong case to be made for an ongoing subsidy to cover the cost of this vehicle so that the entire network can continue to deliver its services? One further point related to sustainability concerns the incentives that are being created by donors who are increasingly pressing microfinance institutions to achieve or at least move strongly toward full financial sustainability. On the positive side, this emphasis forces microfinance institutions to behave like real financial institutions and to pay close attention to revenues and costs. In addition, it is clear that many microfinance institutions are highly inefficient, very costly, and in need of this discipline. On the less positive side, placing a strong emphasis on covering costs may tip the delicate balance that exists between social and financial objectives, forcing institutions such as the ones reviewed here to move away from serving the very poor and rural areas where the delivery of financial services is very costly and where there typically are no other providers. Most of the microfinance institutions studied here are moving into urban areas, but this does not necessarily mean that they are moving away from rural activities. Nevertheless, if requiring all microfinance institutions to reach financial sustainability at any cost encourages them to abandon rural activities, the costs may be too high. A Profile of Selected Microfinance Institutions 69 Notes 1. Peter Fidler, of the World Bank's Private Sector Development Department, calculated most of the figures presented in this chapter. 2. The methodology was taken directly from other sources, notably from a recent study by USAID (U.S. Agency for International Development) and from a technical guide published by the Inter-American Development Bank. 3. As noted in the case studies, audited financial information was rarely available, but most institutions had fairly detailed raw data that could be used with reasonable confidence. 4. Aggregate tables that summarize the data presented in this chapter are found in annex 1. 5. The five institutions labeled here as credit unions are not legally registered credit unions. Rather, their legal status is still that of an NGO project, with the exception of Reseau des Caisses Populaires, whose local branches are registered individually as small savings and loan banks. This is a transitional period, because the new law regulating credit unions creates a legal category for these institutions. The five institutions classified as credit unions share most of the basic characteristics of credit unions: they are membership institutions set up to serve only their members, they generally govern themselves through representative structures, they are federated at a union level, and they give precedence to savings mobilization over lending. Village banking programs aim to establish independent or quasi- independent, community-managed credit and savings associations that provide financial intermediation services to very poor residents in villages. 6. For example, Cr6dit Mutuel de Guinee requires a guarantee of 50 percent of the loan principal in savings, and Reseau des Caisses Populaires requires 100 percent in assets and an additional 25-30 percent in savings. 7. The effective interest rate is calculated by dividing the credit income by the average loan portfolio. Real interest rates were not calculated because inflation prior to the devaluation of January 1994 was negligible followed by a one-time spike that has now leveled out to a large extent. 8. The literature on microfinance touts the efficacy of group lending in producing high repayment rates, but the nine institutions reviewed here demonstrate that group lending is not a precondition for high performance but rather one option that may be appropriate in some circumstances and not in others. 9. See Yaron 1992; Christen, Rhyne, and Vogel 1994. Most of the parameters of outreach and sustainability used here are drawn directly from these two seminal reports. 10. Data were not available on the distribution of loan sizes, which is a more accurate measure of outreach to the poor. 11. By way of comparison, many of the best programs have difficulty reaching the "very poor." Grameen Bank's average loan size as a percentage of GDP per capita is 48 percent, K-REP'S is 64 percent, ADOPEM'S is 68 percent, 70 Overviews BRI'S is 81 percent, and BancoSol's is 82 percent. (ADOPEM is the Asociacin Dominicano para el Desarrollo de la Mujer, and K-REP is the Kenya Rural Enterprise Program.) See Christen, Rhyne, and Vogel 1994. 12. Expatriate expenses account for the largest share of the loan portfolio in institutions whose ratios improve the most from column three to column four in table 4.6. Here again, the size of the loan portfolio must be consid- ered when evaluating these figures. 13. Several, such as Caisses Villageoises, have links with commercial banks, but these links are almost exclusively with banks that accept institutions' deposits and disburse donors' funds, not their own. 14. In very general terms, we compare the nine West African programs with the eleven programs selected by USAID as among the best in the world to get some sense of how well the West African institutions compare. A full discussion of best-practice principles is contained in Christen, Rhyne, and Vogel 1994. 5 Implications Leila Webster The objective of this study is to deepen our knowledge of informal sectors in West Africa so that the World Bank can be of more help to their low- income populations. Phase 1 of this work was devoted to learning as much as possible about the characteristics and constraints of the informal sectors in twelve West African countries. In phase 2, staff of the Private Sector Development Department selected and appraised nine microfinance institutions considered to be among the most effective in the region. Here, the objectives were to identify and evaluate institutions that might be suitable partners for the Bank and to learn more about the dynamics of microfinance institutions in this region, especially compared with programs from other regions. The first section briefly sums up major findings of the country profiles and of the institutional case studies, while the second lays out a series of options for World Bank involvement. Summing up Studies of the informal sectors in the West African countries indicate that the following facts might be most significant: * Informal sectors in West Africa are very large, and appear to be growing rapidly. They account for roughly one-third to half of GDP and anywhere from one-third to three-quarters of total employment in most countries. * The majority of informal activity is rural and associated with agriculture. Large-scale urban migration is changing this, and informal sectors are growing more rapidly in urban than in rural areas. * Women, particularly very poor women, are major participants in all informal sectors. * Key constraints include saturated and stagnant markets, lack of access to credit and savings services, weak technical skills, inadequate information, and poorly developed infrastructure. * Low-income entrepreneurs rely mainly on family and friends, money- lenders and trade creditors, and esusus and tontines for savings and 71 72 Overviews credit. A few access financial services from donor-supported NGOS and credit unions. Virtually none interact with banks. * Microenterprise assistance programs are scattered unevenly across the region: there are many in some countries and few in others. They range from very small, high-cost programs to fairly large, efficient institutions. The case studies of nine microfinance institutions produced a wealth of information about common strengths and weaknesses of West African microfinance programs. Major strengths are related to outreach. The number of members or clients is large, and most of the seven programs that offer savings services reach a relatively large number of savers. The number of borrowers has grown rapidly, but, in absolute terms, the scale of credit services remains very small in most cases, with fewer than 10,000 borrowers. As regards depth of outreach, these programs are serving some of the hardest-to-reach populations in the world. Clients are very poor, and most live in remote, rural areas where basic infrastructure is minimal. Most count large numbers of women among their clientele. The quality of both savings and credit services is generally high. With a few exceptions, savers are able to access their accounts freely, and most programs offer at least minimal interest. Borrowers can obtain small loans through the use of simple procedures. Requirements for character-based guarantees and collateral are well-tailored to local conditions, and the terms and conditions of loans are in line with the needs and abilities of clients. Indicators of the financial viability of these institutions, such as the ratio of revenues to expenses, point to a wide range in performance. On average, revenues cover 30-40 percent of expenses; only one institution is operating at a profit. Revenues can be increased by making more loans or by charging higher interest rates. In most cases, institutions are blocked from making more loans by lack of capital (savings are insufficient in size, and most are demand deposits). Interest rates are not likely to be raised significantly, because most institutions are already charging more than 30 percent a year and, if the new law is implemented as written, a much lower cap on interest rates will be imposed. Costs can be lowered by reducing inefficiency (particularly administrative costs), and transaction costs per loan can be lowered by scaling up (which returns to the issue of lack of capital). In sum, the microfinance institutions reviewed here are delivering much- needed and valued financial services to some of the poorest people in the world. Their depth and quality of outreach are on par with those of microfinance institutions considered to be among the most effective in the world. Their financial sustainability is below that of the best programs globally, but available data on trends indicate that most will improve on both counts in the next several years as growth slows and operations become more streamlined. Implications 73 Several important issues arise from these case studies. First, the built-in costs of reaching very poor people in remote, rural areas of this region may always exceed possible revenues, and subsidies may always be needed to fill the shortfall if the institution is to remain in operation. This possibility challenges the common assumption that all microfinance institutions can eventually reach break-even, but in no way does it imply that microfinance institutions such as those reviewed here should not make every effort to recover costs fully. This possibility also challenges governments and donors to think carefully about the use of subsidies. In this context, impact becomes key. A few studies have been carried out concerning the impact on poor households of participation in microfinance programs, and they have found a significant impact on family welfare. That access to savings and credit services raises and stabilizes incomes among the poor is of little doubt. Of more signifi- cance is the question of how these benefits compare with those flowing from other types of interventions, especially in situations where ongoing subsidies are at issue. For example, which has a greater long-term impact on family welfare: cash incentives for sending girls to school or a four-wheel-drive vehicle that will enable a village banking system to stay in business? Where should governments and donors invest to realize the largest return? Little empirical work has been done on this specific question, but current thought says that raising incomes is the most direct route to improving the welfare of the poor and that providing credit and savings services is a fairly direct, cost-effective route to raising incomes. The underlying assumption is that poor people are fully capable of making welfare-enhancing choices once they have income with which to purchase services. Impact studies indicate that this is a valid assumption. The nutritional, educational, and housing status of participants' families improves with the increase in income associated with the use of savings and credit services. This approach empowers participants in that it affords them the choice of how they will improve their lives; it respects them in that it involves a contract between two parties rather than a handout; and it is efficient in that welfare choices are driven by demand rather than by supply. These case studies raise the issue of the role of donors in supporting the growth of microfinance institutions. Two questions arise. First, where is the line between start-up grants and operating subsidies? Field teams were surprised to find that expatriate costs were not included in the cost accounts of the institutions appraised, and there were long discussions with program managers and with donor agencies about how expatriate costs should be understood. Program managers did not think of these costs as part of their operating costs, perhaps because they generally were paid directly by donors and did not come out of the manager's budget. Donors did not think of expatriate costs as operating costs, but rather as separate investment costs. Field teams from the Bank argued that expatriate costs should be included 74 Overviews in operating costs but placed them under a budget category labeled "institution building." The argument on the Bank's side was that these costs are not one-time, start-up costs but ongoing costs that must be factored into assessments of financial sustainability. This rather practical debate has significant conceptual content. A second, related question asks in what capacity and for how long expatriates should manage microfinance institutions in Africa. Expatriates have left or are leaving two of the nine institutions, have a visibly soft touch in three institutions, and are clearly in charge in four institutions. Those classified as having a soft touch have delegated a great deal of the decisionmaking authority to Africans, but those who have management positions appear to hold them permanently. Even where expatriates work in the capacity of advisers, this use of expensive expatriate assistance does not appear to be clearly time-bound and focused on transferring skills to locals. The relatively heavy, ongoing reliance on expatriates appears to be counter to the objectives of attaining high levels of institutional and financial sustainability. Implications for the World Bank A good grasp of the dynamics and constraints faced by West African informal sectors and a deeper understanding of the issues confronted by the microfinance institutions that serve them are critical to formulating strategies for Bank assistance. Based on the information and insights that have emerged from this research, it appears that the Bank could take an active role in four arenas: within the Bank itself, with microfinance institutions, with client country governments, and with other donors. Within the Bank First, a conceptual adjustment is needed. Delivering financial services to very small borrowers has proven effective in raising and stabilizing the incomes of the poor. This experience has been translated within the Bank to mean that the delivery of credit and savings services to the poor should be incorporated into poverty loans, usually accompanied by social welfare services that are far afield from finance. A preferable approach would categorize microfinance delivery as a financial sector operation. The underlying premise here is that financial systems should be capable of serving low-income as well as middle- and high-income people. Microfinance institutions have proven themselves able to reach these lower strata, and as such, they are part and parcel of the financial system. This conceptual shift means that microfinance program- ming would fit best within financial sector work, regardless of the category of the larger Bank project within which it would be included (see Webster, Riopelle, and Chidzero 1994 for a full discussion of this issue and for an Implications 75 overview of the Bank's current work in the area of microenterprise development). Second, more Bank staff should acquire the skills needed to work effectively in the area of microfinance. Background knowledge of the dynamics and constraints of informal sectors, familiarity with the history of assistance efforts, understanding of current thinking in the field, and accounting skills that can be used to appraise microfinance institutions would be a good start. Third, the same minimum operating standards for intermediaries should be applied by all Bank teams that are active in microenterprise assistance in any given country. An agricultural loan for small farmers that offers subsidized interest rates undercuts the task manager's insistence in an urban project that rates be set above commercial rates.' Fourth, better coordination should be achieved across the many functional divisions of the Bank that are active in the area of microenterprise assistance, such as the human resources, finance and industry, agriculture, urban, and infrastructure divisions. Assisting microfinance institutions The Bank's work with microfinance institutions should focus on helping them to increase their outreach and move more quickly toward sustainability. The following options emerge when possibilities are narrowed down to those that meet both the needs of most microfinance institutions and the capabilities of the Bank. Provision of capital. As was the case with the institutions reviewed here, loan portfolios that are too small critically undermine efficiency and financial viability. Most of the institutions reviewed here constantly ration loan funds, and demand is far from met. Where the problem is lack of capital, the Bank's objective should be to devise ways of leveraging financial resources. In the long run, most microfinance institutions that achieve large scale and high levels of financial viability must be linked to the formal financial sector, and the Bank's approach should work to strengthen the links between microfinance institutions and banks. This might involve one- time capitalization of institutions so that they can leverage funds from local banks, or it could involve credit lines to microfinance institutions through banks with an increasing percentage of participating bank funds, for example, 60 percent in the first period, 80 percent in the second, and so forth. Such credit lines would be re-lent at commercial rates to microfinance institutions that have been evaluated as creditworthy. Training. Most of the programs reviewed here have invested substantial resources in staff training, but approaches, curricula, and quality are highly disparate. In place of typical technical assistance programs where expatriate experts are assigned to given institutions or where staff are taken abroad for 76 Overviews training, the Bank might consider underwriting the establishment of a regional training center for the West Africa region. Such a center could serve the training needs of all levels of staff in microfinance institutions from field agents to managers. Many benefits would flow from a regional approach, most notably the delivery of standardized, high-quality training services that would raise regional standards over time and the establishment of a channel for the flow of information from outside the region. If such a center could be set up such that fees are reasonable, demand for services probably would be enormous. Development of effective management information systems. The institutions reviewed here are highly adept at gathering micro-level data, but some have difficulty consolidating local data and, more important, interpreting their significance and using the results for remedial action and planning. Managers need help in working out practical and useful accounting systems. Hardware, software, and technical advice also are needed. Several software packages have been developed for use by microfinance institutions, but their relative strengths are not clearly understood. Assisting governments Education should be the first step. Field visits confirmed that few govern- ment officials understand clearly the effectiveness of microenterprise development as a poverty alleviation strategy. Even fewer have up-to-date information about best practice in the delivery of financial services. A relatively inexpensive and easy approach to this problem would be for the Bank to run a series of seminars in selected countries taught by experts in the field. Participants would include relevant government officials as well as managers of local microcredit programs. There is now a sufficiently long track record and a large enough body of information on this subject so that developing such seminars would not be difficult. The Bank can also help governments to fashion their banking regulations such that they include and are appropriate for a range of financial institu- tions, including those that serve the poor. The new law governing credit union-type institutions in most West African countries contains several provisions that are not conducive to the growth of microfinance institutions. The Bank should have encouraged more dialogue about this law. Further investigation is needed to uncover other regulations that hamper efficient development and functioning of microfinance institutions. Donor coordination Working more effectively with other donors would focus mainly on improving coordination, particularly at country levels. Even the limited fieldwork undertaken in this research revealed many instances where Implications 77 donors seemed to be duplicating services and undercutting each others' programs, due less to conscious decisionmaking than to poor coordination. The existence of two donor committees, the Committee of Donor Agencies for Small Enterprise Development and the Working Group on Financial Sector Development, offers good opportunities for colleagues in donor agencies to know one another and to exchange information. The Bank serves as the secretariat for both committees, and a more aggressive effort could be made to effect better donor coordination.2 A strong beginning in this direction is the process of establishing common standards for evaluating microfinance institutions. If these standards are accepted by all donors and fully implemented in country programs, enormous strides will have been made. Notes 1. In an effort to raise and bring about greater uniformity in the area of standards, much work is under way to establish standards for microfinance institutions that can be used by the Bank and by other donors. These are now available for circulation. 2. Another vehicle for coordination among donors is the Consultative Group to Assist the Poor (CGAP). As a multidonor effort to address poverty through microfinance, the CGAP Secretariat is housed within the World Bank. Part 2 Country Studies 6 Burkina Faso Ousa Sananikone Situated in the heart of western Africa, Burkina Faso is a landlocked country covering 275,000 square kilometers. With a population of 9 million people, it is one of the most densely populated countries in Africa. More than 85 percent of its population live in rural areas, with the majority concentrated in the center and east of the country. In 1992 the population of the two main urban centers, Ouagadougou and Bobo Dioulasso, grew at a rate of 10 percent, compared with less than 3 percent for the population as a whole. With a per capita income of less than $300, Burkina Faso is one of the world's poorest countries. Basic social indicators are among the worst in all of Sub-Saharan Africa: life expectancy is forty-eight years, infant mortality is 133 per 1,000 births, 46 percent of all children are malnourished, and 82 percent of the total population and 91 percent of women are illiterate. Furthermore, primary school enrollments (36 percent of school-age children) and secondary school enrollments (7 percent of school-age children) are far behind those in most Sub-Saharan countries (World Bank 1994h, pp. 52-53). In 1992 agriculture and services were the leading sectors in the economy, accounting for 44 and 36 percent of GDP, respectively. The formal industrial sector contributed 20 percent to GDP but provided only about 5 percent of total employment. Burkina Faso's export earnings came primarily from agricultural and mineral products, with raw cotton, live animals, hides, skins, and gold accounting for 88 percent of the total value of exports. Burkina Faso's economic performance is limited by a number of constraints: high population growth and related environmental and socioeconomic problems, dependence on rain-fed agriculture and vulnera- bility to drought, inadequate physical infrastructure, and, until 1991, a highly regulated and distorted economic environment that discouraged private investment and exports. Since the adoption of a structural adjustment program in 1991, the government has been implementing policies aimed at restoring efficiency in public resource management, promoting private sector growth, and reallocating public expenditures in favor of the priority sectors of health and primary education. The government has reduced prices and import restrictions, eliminated export taxes, and reformed the banking sector. Despite some delays in the implementation of structural reforms, the adjustment program is broadly on track, and GDP has been growing at 4 81 82 Country Studies percent a year since 1992. The devaluation of the CFA franc in January 1994 provided renewed impetus to press on with economic restructuring and policy reforms, as improved competitiveness has provided new opportuni- ties for growth. Main characteristics of the informal sector According to official statistics, the informal sector accounted for 22 percent of the labor force and 32 percent of GDP in 1992 (Burkina Faso, Ministcre de l'Emploi, du Travail, et de la Securite Sociale 1993, p. 32). In terms of value of output, it accounts for an estimated 65 percent of the secondary sector and 58 percent of the tertiary sector. Between 1986 and 1992, the informal sector grew at an average annual rate of 2.5 percent in the secondary sector and 5.0 percent in the tertiary sector (Ouedraogo and Ou6draogo 1994). Women constitute roughly two-thirds of the informal sector's labor force. The informal sector in Burkina Faso plays an important role in unemploy- ment reduction and poverty alleviation by absorbing surplus labor at minimal wages. Its growth is most visible in large urban areas; in Ouagadougou, it accounts for more than 86 percent of all paid employment (Burkina Faso, Ministcre de l'Emploi, du Travail, et de la Securite Sociale 1993). The growth of the urban informal sector can be attributed in large part to urban migration, which brings approximately 11,000 people to Ouagadougou every year. In addition, thousands of secondary school and university students graduate annually, further adding to the already high level of unemployment. Because the formal sector creates only 1,000 jobs each year, the informal sector provides an important means of livelihood to new entrants in the job market. Most informal sector enterprises produce simple, low-priced consumer goods and services that cater to the needs of low-income groups in rural and urban areas. Basic foodstuffs, for example, are distributed mainly through a network of small traders in the informal sector. Unlike certain countries in Asia, where subcontracting arrangements between formal, larger firms and smaller, informal enterprises are common, such relationships are rare in Burkina Faso. In this respect, the informal sector in Burkina Faso is rather typical of the informal sector in many Sub-Saharan African countries. The majority of informal sector microenterprises in Burkina Faso employ fewer than five workers, and most entrepreneurs are self-employed. The average microenterprise lasts about four years. With the exception of one or two subsectors, links between enterprises do not appear to be an important feature of the informal sector. Certain activities, however, tend to cluster together in a kind of informal collaboration. For example, it is quite common for a motorcycle repair business to settle in front of a shop selling spare parts or for a traditional cafe to attract related food activities (Grant and others 1991, p. 9). Burkina Paso 83 Skills for informal sector activities are invariably acquired within the sector itself. According to a 1990 survey by the National Employment Office, 82 percent of entrepreneurs in the informal sector learned their trade through apprenticeships, usually with family members. Apprentices are common in male-dominated occupations such as tailoring, upholstery, woodworking, and brass or bronze working. Interviews with trainees in the bronze workshop at the Artisans Training Center and with owners of individual bronze craft stores confirmed that most bronze workers first learned the techniques at home in apprenticeships with family members, usually with fathers, uncles, or older brothers. A 1986 survey of entrepre- neurs (male and female) in Ouagadougou revealed that only about 8 percent had received professional training other than an apprenticeship; 30 percent had attended primary school, most without obtaining a degree; and only 9 percent had learned to read and write (van Dijk 1986, p. 67). National surveys conducted by the National Employment Office indicate that informal sector enterprises generally employ three categories of workers: salaried workers, apprentices, and family members. Apprentices generally are not paid but do receive food and occasionally a small stipend for their work. A recent study of the informal sector revealed that the majority of apprentices are family members and that apprentices who work for free constitute nearly 70 percent of all microenterprise workers. The significance of this must be seen in a larger social and economic context: although family members are not paid, employing them is a means of taking care of one's family. The minority of owners who pay their workers typically pay low wages, sometimes only when they have orders from customers (World Bank 1989, p. 35). Temporary and daily workers dominate the construction, iron melting, and service subsectors, where workers are paid on either a daily or a weekly basis. The most common informal sector activities in Burkina Faso center around production, services, preparation and sale of food, and petty trade. In recent years, the number of operators in activities such as aluminum smelting and metallic carpentry has increased. Service activities such as food catering and bicycle repair have also expanded. Women constitute two-thirds of the labor force in the informal sector. Yet, because most subsectors are highly differentiated by gender, women are restricted to operating in only a few activities. Because it is easiest for women to engage in economic activities that require low capital investment and use existing skills, women dominate informal sector activities such as food preparation, sorghum beer (dolo) brewing, basket weaving, hairdress- ing, cosmetics, and tailoring (see table 6.1 on gender-differentiated activities). These activities are flexible enough that they can be generated from the home and combined with child care responsibilities and other related tasks. Rarely do women's activities in the informal sector evolve into real employment in the formal sector. In general, women entrepreneurs in 84 Country Studies the informal sector have the lowest incomes, and they seldom have the capital needed to scale up. Table 6.1 Informal sector activities in Burkina Faso, by gender Gender Activity Female Brewing and selling of sorghum beer, production and sale of doughnuts, preparation and sale of other edible products Male Car and motorcycle repair, construction, electrical repair, upholstery, woodworking, transport, selling of coffee and tea, blacksmithing (including metallic furniture), plumbing, repair of watches, photography, milling, trading of meat, shoe shin- ing Mixed, but pre- Tailoring, domestic employment, laundry, shoe mending and dominantly tannery, trade of cloth and second-hand clothes, production male and sale of artistic handicrafts Mixed, but pre- Hairdressing, trading of fruits and vegetables, work in hotels dominantly and restaurants, spinning, dyeing, basket making, weaving, female trade in groundnuts Source: Grant and others 1991. Main constraints on the informal sector Because of their lack of collateral and small size, microenterprises generally lack access to credit from formal financial institutions. In a recent survey of informal sector enterprises, 97 percent of entrepreneurs reported that they had never received a bank loan. The vast majority of enterprises self-finance their start-up using personal savings or loans from family and friends-all of which necessarily means very small investments. The key constraint for microenterprises is not a lack of start-up capital, but inadequate access to working capital. Surveys of the informal sector in Ouagadougou indicate that two-thirds of all artisans have had to stop working at some point because they lack working capital. ln this respect, lack of access to working capital in effect condemns microenterprises to a cycle of low productivity, low revenues, and unstable incomes. Contrary to their image as tax evaders, most informal sector enterprises do pay some form of taxes. In 1993 the government established a new tax regime for the informal sector that was considerably less punitive than previous regulations. The contribution du secteur informel (csi) combines all taxes the informal sector must pay, except for municipal and wage taxes. Prior to the fiscal reform of 1993, informal sector enterprises mainly paid the patente, a tax on sedentary enterprises, and an additional market tax on nonsedentary enterprises. The patente consists of two separate taxes: a fixed tax based on the enterprise's profits and a flexible tax proportional to the value of the location. Since the patente was partly tied to the value of the Burkina Faso 85 enterprise's location, entrepreneurs often hesitated to make new, visible improvements to the working place that could lead to a higher tax (Oudraogo and Oudraogo 1994, p. 22). Under the csi, the main tax on informal enterprises is no longer tied to the value of the location, but rather to the enterprise's profits. As of January 1994, enterprises registering profits in excess of CFAF 15 million ($30,000) are subject to taxes. Because few microenterprises achieve this level of profits, the csi affects only a minority of them, and the average tax rate imposed on the informal sector is less than 3 percent (Oudraogo and Oudraogo 1994, p. 22). The greatest threat to entrepreneurs in the informal sector comes from competition within the sector. The inability to upgrade product quality and to move into more profitable niches leads many microenterprises to "bunch up" in similar activities, often in the same geographical areas, resulting in market saturation. Insufficient technical skills and poor access to appropriate technology prevent entrepreneurs from producing finished goods of high quality, keeping them locked into poor-quality, noncompetitive products. Although advanced management skills are not really needed to run efficient microenterprises, the lack of marketing skills-selecting which products and services to provide and how to market them-often limits enterprises' profits. When asked how they market their products and services (for example, whether they go around the neighborhood to "advertise" their products), many entrepreneurs respond that word of mouth will eventually bring them customers. Lack of information about markets prevents many entrepreneurs from becoming more competitive. For example, the owner of a small dressmaking business in Ouagadougou indicated that her enterprise would be more competitive if she had access to up-to-date foreign fashion catalogues that would enable her to produce women's clothes in new styles and fashions. Most training programs for entrepreneurs in Burkina Faso are delivered by public or semipublic institutions and tend to be weak and poorly developed. They often feature outdated curricula, use old and poorly maintained equipment, and have weak links with the private sector. Few charge fees, and some have "flexible" training periods such that training can go on indefinitely, depending on whether or not the participant finds employment. For example, the curriculum of the Centre National de Perfectionnement des Artisans Ruraux, the leading state agency for training artisans, has not been updated over the past ten years, even though domestic demand and technological innovations have changed consider- ably. Finally, a key constraint that restricts the growth of informal sector enterprises in Burkina Faso may simply be the high costs of becoming a formal enterprise. Cumbersome bureaucratic procedures, the lack of clear- cut regulatory norms, and the high taxes imposed on formal enterprises may 86 Country Studies well deter many microenterprises from moving up into the formal sector. Depending on the type of enterprise, the entire process of registering and operating a private business may take anywhere from six months to three years and costs an average of CFAF 100,000-CFAF 300,000 in various fees (including "facilitation" fees). An entrepreneur who wants to open a private company has to have a minimum capital of CFAF 2 million for a limited liability company and CFAF 10 million for a corporation, to pay legal fees equivalent to approximately 10 percent of the capital, and to pay registration fees and taxes equal to about 3 percent of the capital. Policies and programs for microenterprise development Burkina Faso is blessed with a large and diverse community of governmen- tal and nongovernmental organizations, many of which are committed to promoting microenterprise development. As of February 1990, 150 NGOS were registered with the Bureau de Suivi des Organisations Non- Gouvernementales, the government's NGO monitoring agency. Because of the droughts that affected the Sahel region during the 1970s and 1980s, relief activities such as food aid and medical assistance have dominated NGOS' agendas until recently. Over the last several years, however, many NGOS have focused their attention on promoting microenterprise development. NGO assistance in this area has included supporting income-generating activities for women (growing vegetables, raising goats and cows, making crafts, processing food, and conducting small conunercial activities), managing small savings and loan programs, providing training and education, and managing health and water resources. This section provides an overview of some agencies and programs that offer financial and technical assistance to microenterprises. Institutional infrastructure In 1986 in recognition of the growing importance of the informal sector in Burkina Faso's economy, the government formed an interministerial commission for microenterprise support-the Commission Inter- Minist6rielle sur la Politique de Promotion de l'Artisanat. The commission is chaired by the Direction Generale de l'Artisanat of the Ministry of Economic Development. There are no fewer than a dozen government institutions with a mandate to support or regulate informal sector activities. Key institutions include the ministries of planning, labor, and social affairs and agencies such as the National Employment Office. In 1992 the govern- ment appointed the Bureau de la Promotion du Secteur Informel within the Ministry of Labor as the coordinating agency for all government initiatives affecting the informal sector. Burkina Faso 87 Assistance programs As noted earlier, microenterprises lack access to formal financial services. However, they are able to obtain credit from a number of informal savings and credit programs, most of which are financed by local and international NGOS, donors, government agencies, and a combination thereof. These institutions usually take the form of savings and credit unions (caisses d'epargne et de credit) that provide basic banking services in villages and urban neighborhoods (see table 6.2). There are three major savings and credit unions in Burkina Faso: R6seau des Caisses Populaires du Burkina Faso, Six "S," and the Union de Cooperatives d'Epargne et de Credit Burkinabe. A World Bank mission in May 1994 visited the first two institutions, and in January 1995, another mission conducted an in-depth financial evaluation of Reseau des Caisses Populaires. Reseau des Caisses Populaires is a savings and credit cooperative established in 1972 with the assistance of a Canadian NGO (Societe de Developpement International Desjardins) and financing from the Canadian International Development Agency. The project was initiated to help village associations mobilize savings in the Bougouriba region, and credit activities were subsequently added as a means of recycling funds from savings. Reseau des Caisses Populaires operates in both rural and urban areas, and it has become one of the largest savings and credit cooperatives in West Africa. As of February 1995, it consisted of thirty-five regional branches with 64,000 members, CFAF 2.3 billion in savings ($4.9 million), and CFAF 600 million ($1.2 million) in outstanding loans. Between 40 and 80 percent of loans are nonagricultural loans, with agricultural loans making up the remainder. Interest rates are fixed at 16 percent a year for all types of loans. The average loan size is CFAF 168,000 ($336), and the repayment rate is reportedly 95 percent. Reseau des Caisses Populaires makes individual as well as group loans. Six "S" is a large, pan-African NGO that provides support to the Naam, a cooperative whose structure is based on traditional rural associations in Burkina Faso. Its objectives are to promote rural development by providing credit for income-generating activities and training in agriculture, health, and nutrition. A savings program was added in 1992. As of 1994, Six "S" had fourteen branches and 3,000 savers across the country, with CFAF 50 million ($100,000) in savings. The minimum loan size is CFAF 5,000 ($10), and the maximum loan size is CFAF 15,000 ($30). The interest rate is 32 percent a year. To date, Six "S" has lent a total of CFAF 4 million ($8,000) to about 400 borrowers, primarily women. The repayment rate is reportedly almost 100 percent. Many NGOS also provide savings and credit services to microenterprises. Loans are typically small, averaging CFAF 10,000-CFAF 50,000 ($20-$100), and terms usually do not exceed one year. Interest rates range from 15 to 25 Table 6.2 Selected microenterprise credit and savings programs in Burkina Faso, mid-1994 Credit and Subloan (in cFA francs) Annual interest savings rate Number of Repayment rates programs Minimum Maximum Average (percent) Term of loan Number of loans savers (percent) FAARF, credit - 50,000 for indi- 500,000 for 19 One year 2,314 individual - 90 program, 1991 vidual loans group loans loans plus 44 (first loan); collective loans 150,000 reaching 1,000 (second loan) individuals PPPCR, credit 5,000 50,000 24,000 20-25 One year 30,000 - 98 program, 1988 PRODIA, credit 50,000 1 million 250,000 17 One year 3,000 - 97 program, 1991 Reseau des - 3 million 168,000 16 One year - 50,000 95 Caisses Populaires, savings and credit union, 1972 Six "S," 5,000 15,000 - 32 One year 400 3,000 99 savings and credit program, 1992 - Not available. cJ.L Burkina Faso 89 percent a year, and repayment rates are generally high, about 95 percent. A solidarity group lending approach is often used. Three NGO programs notable for their outreach are reviewed in this section. They are Fonds d'Appui aux Activit6s Remuneratrices des Femmes (FAARF), Promotion de Developpement Industriel, Artisanal, et Agricole (PRODIA), and Projet de Promotion du Petit Credit Rural (PPPCR). FAARF is an autonomous, government-supported program established in 1991 with financial and technical assistance from the UNDP. It provides loans and business training to women in urban and rural areas. Loans are provided for one year at 19 percent interest a year.' The majority of loans finance agroprocessing, artisanal industries, and service activities such as small shops and restaurants. As of March 1994, FAARF had lent a total of CFAF 129 million ($258,000) representing 2,134 individual loans and forty- four collective loans. The average loan for individual borrowers is about CFAF 500,000 ($1,000). PRODIA is an NGO established in 1991 and financed primarily by a German charity. Its objective is to provide small loans to individuals, particularly women. Loan sizes range from CFAF 50,000 ($100) to CFAF 1 million ($2,000). The average loan size is CFAF 250,000 ($500), and the interest rate is 17 percent a year. As of 1994, PRODIA had made approximately 3,000 loans, mostly to women. The average repayment rate is 97 percent. There appears to be a strong demand for credit, as evidenced by the backlog of credit applications and the long lines of women waiting to see credit officers. PPPCR was designed in 1988 by a French research institution, the Centre International de Recherche Agronomique pour le D6veloppement, to provide credit to villagers who had been impoverished during the drought. It is funded by the Caisse Francaise de Developpement and implemented by Sahel Action, a local NGO aimed at promoting rural women's development. The project consists of four regional units located in the provinces of Ganzourgou, Soum, Tapoa, and Yatenga and serves nearly 200 villages. Lending activities take place at the village level, with mobile agents visiting villages once a week to make new loans and collect payments. As of November 1994, the project had made nearly 30,000 loans, with 13,000 loans outstanding. It had an outstanding portfolio of CFAF 232.6 million ($465,200) and an average loan size of CFAF 24,000 ($48). Most loans are extended for one year at 20 to 25 percent interest a year. The average repayment rate reportedly is 98 percent, and 95 percent of borrowers are women. Training programs in Burkina Faso-at least those visited in Ouagadougou-appeared to be somewhat less effective than many of the microfinance programs. One of the key institutions with responsibility for training entrepreneurs is the Chamber of Commerce and Industry, which offers general business and specialized technical training. However, few microentrepreneurs have access to these services or are able to take advantage of them. Another leading institution is the Centre National pour le Perfectionnement des Artisans Ruraux, which has focused on training 90 Country Studies rural and urban artisans in traditional crafts such as blacksmithing, carpentry, welding, and bronze and brass working. Because most of these are male-dominated activities, few women have access to these training programs. Conclusions The outstanding fact emerging from this brief overview of microenterprise credit programs is the strength of existing savings and credit programs in Burkina Faso. A number of programs, such as R6seau des Caisses Populaires and PPPCR, are able to reach a relatively large number of individuals rarely served by formal financial institutions. These microcredit programs use methods for appraising loan applicants, delivering credit, and collecting repayments that are simple, effective, and well-tailored to the local environ- ment. At this point, none is approaching full financial sustainability, but over the past few years many have become keenly aware of the need to increase their operational and financial efficiency. Some have taken steps in that direction by charging higher interest rates, increasing the number of loans per agent, and establishing a solid information system that enables them to monitor loan repayments more closely. Given the relative success of microcredit programs such as Reseau de Caisses Populaires, Six "S," and Sahel Action, the World Bank could find ways to support these programs rather than starting new programs to assist microenterprises. With many donor agencies vying to assist the working poor in Africa, caution should be taken not to undermine the significant advances that these institutions have made toward attaining self-sustainability, or at least greater efficiency, by injecting funds into programs that are less cost- effective and that take a welfare rather than a business approach to service delivery. Despite the large number of training programs for entrepreneurs, none of the existing programs appears to be adequate in meeting the needs of microentrepreneurs. Training programs should be short-term, focused, and updated to reflect market and technological trends; above all, they should be demand-driven. Greater efforts could be made to integrate new and more appropriate production technologies into training programs and to encourage links among enterprises, both within and outside the informal sector. Notes This chapter was written in May 1994 and revised in February 1995 based on field visits. 1. An additional 10 percent of the loan value is factored into the repayment schedules. This 10 percent is paid in the first four weekly installments and then held as a deposit that serves as a guarantee fund, to be retumed to the borrower after the loan is repaid in full. 7 Cape Verde Ousa Sananikone Cape Verde is a small volcanic archipelago of ten islands located 650 kilometers off the coast of Senegal. It has a resident population of nearly 400,000, half of whom live on the island of Santiago. With only a tenth of the country's surface arable and a dry climate exacerbated by long cycles of drought, Cape Verde faces major environmental and developmental challenges, including shortages of fresh water and soil erosion. Confronted with such inhospitable living conditions and limited employment opportu- nities, Cape Verdeans have traditionally resorted to emigration. Almost twice as many Cape Verdeans (750,000) live abroad as live in Cape Verde itself, and their remittances contribute significantly to the economy.' In 1992 per capita income was $850. Social indicators in Cape Verde are quite impressive compared with those of other Sub-Saharan countries. In 1990, illiteracy was 53 percent, primary school enrollment was universal, infant mortality was low (56 per 1,000), and life expectancy was sixty-seven years. These achievements are largely the result of concerted efforts and substantial allocation of resources to social services on the part of the government, women's associations, and donor agencies (in 1992 social services accounted for 30 percent of the govern- ment's budget). Safe water remains in critically short supply. According to the UNDP, 70 percent of the population lacked access to safe water in 1990. Scarce natural resources, prolonged periods of devastating drought, and a scarcity of arable land have limited the opportunities for economic development. The disadvantages of a small domestic market are compounded by Cape Verde's geographic isolation and inadequate internal communications system. Unemployment poses a major challenge for the government. In 1990, more than a quarter of the economically active population was unemployed, and only 28 percent of the workforce held permanent jobs. To achieve full employment, the ILO estimates that 8,000 to 10,000 jobs would need to be created annually until the year 2000 (Woilet 1992, p. 167). Largely because of the paucity of natural resources, the service sector rather than agriculture has been the mainstay of the Cape Verdean economy. In 1991 the service sector accounted for 69 percent of GDP and employed nearly 40 percent of the population. Industry, consisting primarily of light manufacturing, fish processing, and artisanal production, accounted for 18 91 92 Country Studies percent of GDP. Due to low agricultural output and inadequate exploitation of fisheries resources, the primary sector contributed only 13 percent of GDP. Export earnings traditionally have come from nonfactor services rendered to international maritime transport and, more recently, from international air traffic services (World Bank 1993a, p. 10). The economy is highly dependent on imports of food and of intermediate and capital goods. Following Cape Verde's independence in 1975, development of the private sector was not a major government priority. Solid macroeconomic management, reinforced by heavy inflows of emigre remittances and foreign aid, enabled GDP to grow at a healthy pace of 6 percent a year between 1980 and 1988.2 However, the excessive involvement of the public sector, combined with inward-looking economic policies, resulted in an undevel- oped private sector, limited competition, an inadequate financial system, and a growing informal sector. Since 1991 the government has been steadily reducing its central role in the economy and has set as major priorities the eradication of poverty and major reductions in unemployment. New legislation governing foreign investment, trade policies, and privatization was adopted with the aim of orienting the economy toward potentially competitive activities such as export-processing zones, tourism, fisheries, offshore services for banking, and teleporting (World Bank 1994d, p. 3).3 The government's current economic development strategy calls for continued promotion of the private sector through improvements in the nation's economic and social infrastruc- ture. Given the state's long dominance of the economy, it is not surprising to find that the formal private sector in Cape Verde remains small and underdeveloped, consisting primarily of local entrepreneurs engaged in small- and medium-scale enterprises in commerce, construction, and, to a lesser extent, shipping. Fishing companies operate small processing plants, and extraction enterprises produce salt for export. Carpenters, bakeries, and repair shops supply local markets (World Bank 1985a, p. 3). Skilled labor, particularly at the managerial and marketing levels, is in short supply, and experience in export marketing is virtually nonexistent (World Bank 1993a, p. 10). Main characteristics of the informal sector The informal sector has mushroomed since the late 1980s, largely in response to limited opportunities in the formal sector. Many informal sector entrepreneurs are engaged in multiple, part-time activities, thus making the size of the sector difficult to determine. A 1994 survey conducted by the Austrian Cooperation in the town of Santa Cruz, near Praia, found approximately 2,000 microenterprises employing a total of 3,000 persons. A 1992 study of the artisanal sector by the ILO estimated that approximately 23,000 persons were engaged in artisanal activities (Woilet 1992, p. 116). Cape Verde 93 Government statistics currently estimate that 30-40 percent of the active population are employed by microenterprises. The informal sector is the dominant force in trading goods such as textiles, footwear, and farm and fish products. A wide variety of interna- tional products ranging from clothing to audiovisual equipment is traded in the large informal markets of Sukopira in Praia and Asomada in Santa Catarina. Most goods traded in informal markets are purchased in nearby countries such as Senegal and the Gambia, thanks largely to relatively inexpensive air transportation to and from continental Africa. Because many Senegalese and Gambian entrepreneurs import their goods from Western Europe, Cape Verdeans raise their retail prices to cover second- and third- hand mark-ups. Given the scarcity of raw materials and the resulting absence of a tradition of craftsmanship, the informal sector remains negligible in productive activities. Commonly found service activities include restaurants, laundries, and hair salons-usually the domain of women-and shoe repair, transportation, and mechanical and electrical repairs-activities dominated by men. Microenterprises in Cape Verde are typically one-person units, but can employ up to five assistants. Some are organized as cooperatives, employing up to ten workers. Microenterprises are characterized by low initial capital investments, relatively easy entry, seasonal or part-time activities, use of rudimentary tools, low product quality, and weak marketing of products and services. Contrary to some countries in continental Africa, many microenterprises in Cape Verde are registered and do pay some form of taxes, primarily to the municipality. This is mostly the case of urban microbusinesses such as bakeries, taxis, rum distilleries, and hair salons. Taxes are 20 percent of declared annual sales (World Bank 1994d, p. 11). A 1988 survey of selected microentrepreneurs revealed that most had not completed primary school and that many were illiterate. However, the majority had more than ten years of practical experience. Women comprise nearly 40 percent of the labor force in Cape Verde, but they are believed to account for 70-90 percent of informal sector establish- ments. Women dominate the trade sector, where they work mostly as street vendors. Unlike some other countries in Sub-Saharan Africa, there are few overt barriers to women's participation in the economy. Legally, women are accorded equal status in the ownership of land, access to credit, and ownership and operation of businesses. In practice, however, a number of constraints militate against their full participation in the economy. Female entrepreneurs generally have lower levels of education than men and thus are more likely to have a lower income level, which means reduced savings from which they can finance small business activities. In interviews, women microentrepreneurs cite as primary constraints the lack of market informa- 94 Country Studies tion, inadequate working and investment capital, and weak managerial skills, particularly related to planning, accounting, and marketing. Main constraints on the informal sector A review of the literature on microenterprises in Cape Verde suggests that the major constraints to informal sector development arise from the country's physical environment and are technical and financial. These constraints include the following: • Small internal markets. A key obstacle to microenterprise development in Cape Verde is the small size of the domestic market, a problem that is exacerbated by inadequate transport and communications infrastructure among the islands. This situation in effect confines the domestic market to the island where goods and services are being produced (with the exception of food). The results are highly saturated markets with excess capacity in key sectors such as garments and footwear. * Lack of raw materials and prohibitive import policies. Because of Cape Verde's poor natural resource base, most microenterprise activities are highly dependent on imported raw materials and intermediate goods. However, a number of regulations are biased against import-intensive, small-scale activities. For example, to import materials valued at more than Cape Verde escudos (C.V. Esc) 100,000 ($1,260), an entrepreneur must have a capital stock of at least C.V. Esc 5 million ($63,300), an investment level far higher than most microentrepreneurs reach.4 i Weak product development and marketing strategies. Entrepreneurs in the informal sector typically have weak product development and marketing skills. Many products are developed based purely on tradition rather than on real market preferences. Furthermore, many microenterprises are started simply by imitating a neighboring activity, further saturating an already small market. • Lack of access to credit. Like their counterparts in other Sub-Saharan countries, microenterprises in Cape Verde lack access to formal financial institutions. Unlike them, however, Cape Verdeans have little recourse to informal savings and credit institutions such as tontines and coopera- tives, which are far less common in Cape Verde. As a result, many microenterprises remain self-financed. Remittances from overseas emigres play an important role in financing small business ventures at home. The formal financial system in Cape Verde is very small and predomi- nantly serves the public sector. It is comprised of four state-owned institutions: Banco de Cabo Verde, Caixa Econ6mica de Cabo Verde, Banco Comercial do Atlantico, and the new Instituto de Seguros e Providencia Cape Verde 95 Social.' Given the high administrative costs and financial risks involved in making very small loans, none of these institutions serves microenterprises. As of February 1995, legislation was in place for the creation of a Caixa de Cr6dito Rural that would provide credit to microenterprises, lending mainly to cooperatives for fishing and agriculture-related activities. Although microenterprises in Cape Verde clearly lack access to financial services, demand for microcredit is difficult to determine. The fact that foreign remittances play an important role in financing microenterprises and that existing microcredit programs have made very few loans over the past three to four years may suggest that there is little demand for small loans. However, staff of NGOS and parastatals currently providing credit to microenterprises indicate that demand for credit far exceeds the current supply. Policies and programs for microenterprise development Compared to other countries in Sub-Saharan Africa, the few microenterprise assistance programs operating in Cape Verde show a striking lack of outreach, defined as the number of clients reached and loans made, even when considering the small size of the population. This section reviews financial and technical assistance programs as well as government policy for microenterprise development. Financial services A mission to Cape Verde in September 1994 identified five institutions providing credit services to microenterprises: Instituto Nacional de Cooperativas-Fondo de Apoio as Cooperativas (INC-FAC), Caixa de Credito Rural, Organizaqao das Mulheres de Cabo Verde (oMcv), Caritas Caboverdiana, and CITI-Habitat. Of these, one has ceased providing credit due to lack of funds (INC-FAC), three maintain active credit programs (Caritas, CITI-Habitat, and oMcv), and one is not yet operational (Caixa de CrGdito Rural). Table 7.1 outlines the lending terms and conditions of these programs. INC is a government institution designed to establish and assist coopera- tives across the country. Within the INC, the Fondo de Apoio as Cooperativas (FAC) was created as a special unit to channel funds from donor agencies to cooperatives. FAC was capitalized with funding from several foreign sources, including the British, U.S., French, Dutch, and Danish governments and NGOS. In 1989 FAC'S lending portfolio totaled $425,000, most of which was lent to cooperatives for purchasing equipment and constructing buildings. A third of the outstanding portfolio was used to finance agricultural and small business activities (Clement, Souto, and Magill 1989, p. 31). Interest rates were 12.5 percent a year but could be '-0 0-, Table 7.1 Selected microenterprise credit programs in Cape Verde Down- Interest Average Repayment Number payment Loan Term of Credit program rate a year loan size rate of loans required fee loan Status or comments INC, Caixa de - - _ _ _ _ _ Legislation in place; however, Credito Rural, institution not yet operational 1994 Caritas, Credit de 10% $2,700 80-90% 125 0% $0 Two Program provides follow-up Developpement years training Agricole, 1992-94 Caritas, Nutrition 0% $26 Less than 600 0% $0 Flexible Program in progress; charity- Program, 1992-94 80% based (given to families in the Caritas emergency program); many grants offered; no strict repayment period OMCv, 1990-94 6% $266 80-90% 28 0% $0 Sixteen Philanthropic program, with no -$2,000 months strict repayment period or en- forcement mechanism CITI-Habitat, 5% $650- 100% 3 0% $0 One year Lending resources exhausted; 1993-94 $2,000 repayment of first loan cycle in progress - Not available. r Cape Verde 97 lowered depending on the cooperative's needs. In addition, interest-free loans were extended to newly established cooperatives. The average loan size was C.V. Esc 100,000 ($1,265). The FAC was dissolved recently due to lack of funds. Following the dissolution of FAC, a law decree was adopted in April 1994 to establish the Caixa de Credito Rural to take over the management of INC'S cooperative lending program. Due to slow administrative processes, Caixa's opening has been delayed, and funding has not yet been secured. The government is currently negotiating with the African Development Bank and other donors for possible funding. Once the program is operational, loans are expected to be given to cooperatives to finance agricultural and fishing-related activities. Interest rates are expected to be 7-9 percent a year, significantly lower than commercial rates. The average loan size is expected to range from C.V. Esc 300,000 to C.V. Esc 500,000 ($3,800-$6,330). Caixa de Credito Rural will be headquartered in Santa Catarina, with branches on other islands. The oMcv was created by the government in 1988 to mobilize women to undertake small, income-generating projects. In 1990 oMcv changed its status to that of an NGO and was restructured, reducing its staff from nearly 100 to 9. It is funded entirely by foreign donors, including the Swedish and Canadian governments, UNICEF (United Nations Children's Fund), and Novib, a Dutch NGO. oMcv focuses on women's issues, such as adult illiteracy, family planning, and children's education, and has created approximately eighty kindergartens in Cape Verde. It has more than 12,000 members who, in principle, are supposed to pay a membership fee of C.V. Esc 5 to C.V. Esc 500 ($0.06 to $6) per month, depending on their financial means. However, few members pay this fee, and no enforcement mechanism is in place. In 1990 OMCV started a credit program that, as of December 1994, had generated a total of twenty-eight loans ranging from C.V. Esc 20,000 to C.V. Esc 150,000 ($253-$1,900). Loans were extended for a maximum term of two years at 6 percent annual interest, well below prevailing market rates. The total amount lent was $9,084 from an available credit line of $10,000. Loans financed livestock, meat processing, sewing, and other commercial activities. As of December 1994, 10 percent of the loans were in arrears, and 4 percent had defaulted. oMcv continues to make small loans using the funds from reimbursed loans but has not received additional funds or credit lines from donors. Caritas Caboverdiana is a humanitarian organization created in 1976. Its objective is to improve living conditions among the poor through support for agricultural and livestock activities and training in women's and family issues. Caritas currently operates two credit programs: Credit de Developpement Agricole and a credit component within its nutrition program. Credit de Developpement Agricole provides loans for productive activities at an interest rate of 10 percent a year. This lending program was 98 Country Studies started in 1985 on an experimental basis, and it was only in 1992 that a concrete effort was made to follow up on loans and ensure repayment. Between 1992 and 1994, about C.V. Esc 4 million ($50,633) was disbursed to approximately twenty-five groups (125 individuals). The average loan is about C.V. Esc 200,000 ($2,532), and the average term for a loan is two years. Repayment rates range between 80 and 90 percent. The credit component within the nutrition program was begun in 1992 to alleviate the impact of the drought on families. Some 500 families received loans for amounts as low as C.V. Esc 2,000 ($25), and the total amount lent was C.V. Esc 9 million ($114,000). Caritas expects to disburse an additional C.V. Esc 3 million ($3,800) in 1995. CITI-Habitat is an NGO that was created in 1988. It is funded by a Belgian NGO called Solidarite Socialiste, the UNDP, the Austrian Cooperation, and several private individuals. Its main objectives are to provide training in construction, carpentry, and solar energy and to support youth activities. CITI-Habitat also sells the research and consulting services of its employees to private enterprises as well as to national and international organizations such as the UNDP and ILO. In 1993 CITi-Habitat began a lending program in the city of San Francisco on Santiago Island. A local association manages the program. As of February 1995, cITi-Habitat had received ten loan applica- tions but had made only three loans due to lack of funds. The loans ranged in size from C.V. Esc 50,000 to C.V. Esc 150,000 ($633-$1,900) and carried an interest rate of 5 percent a year for a term of one year. To date, only one loan has been fully reimbursed, and, according to CITI-Habitat staff, the remaining two loans are being repaid on time. A common characteristic of these programs is their philanthropic, almost charity-based approach. All existing programs are dependent on donor funds, and none has built-in mechanisms to reach financial self-sufficiency. Loans are provided at highly subsidized rates of 5-6 percent a year, compared to market rates ranging between 12-14 percent a year, and some programs even offer "flexible" rates depending on the need of the borrower. In many cases, timely repayment of loans is not required. Except for the Caritas programs, which reached approximately 700 people, these programs average only three to ten loans a year, a minuscule number compared to other microenterprise programs-even very small ones-in Africa. Several other NGOS are involved in small enterprise credit programs that target women. For example, Welt Freundienst, a German NGO, finances artisanal production in Maio, specifically in clay, ceramics, and limestone. Oxfam/America has been involved, through its Boston office, in a project that assists the oMcv in marketing lace and embroidery handicrafts in the New England area. Cape Verde has a rather large number of donors relative to its small size, but few of them target microenterprise development. Some donor agencies such as the UNDP and ILO have channeled funds through existing NGOS, like Cape Verde 99 CITI-Habitat and OMCV, to support microenterprises. The World Bank's Industrial Finance and Promotion Project has provided assistance to small enterprises through formal financial institutions. Although technically not a microenterprise project, the Industrial Finance and Promotion Project deserves mention because it was relatively successful in providing technical assistance to small-scale enterprises through existing local institutions. The objectives of the $4.0 million project (1985-92) were to support the expansion of Cape Verde's economic base through the creation of employment and the establishment of an institutional infrastructure for industrial development. Two components were aimed at the development of small and medium- sized industries: a $3.3 million line of credit to Banco de Cabo Verde to finance enterprise investments and a $0.3 million technical assistance component designed to set up and operate a new investment department within the bank. The project was relatively successful in stimulating industrial develop- ment and establishing the institutional infrastructure for it. The two units within the Banco de Cabo Verde-the Investment Department and the Directorate of Industry-have become effective institutions for industrial development and financial intermediation. A total of forty-nine subprojects were financed at an estimated cost of Special Drawing Rights (SDR) 3.2 million, creating 337 jobs. At the request of borrowers, the original ceiling of $300,000 for small-scale enterprises was raised to $650,000. Similarly, the ceiling for fixed assets of medium-scale enterprises was raised from $1.5 million to $2.0 million. All of the projects financed were new ventures covering a broad spectrum of industrial and service activities, including fishing, bakeries, hotels and tourism, textiles, and carpentry. USAID is currently working on a proposal to establish a microcredit program. As of February 1995, this proposal had not yet been submitted to the government. The proposed program would make available small loans averaging C.V. Esc 22,500 ($284) at about 3 percent interest a month. Technical assistance Few organizations in Cape Verde offer business skills, marketing, manage- ment, or vocational training for entrepreneurs in the informal sector. oMcv and Caritas provide most of their training in health-related fields. oMcv has offered training for women in general business and secretarial skills. CITi-Habitat provides technical training in construction, carpentry, and pottery making, and INC offers vocational training as well as business skills training. In 1993 the government created the Instituto de Apoio ao Desenvolvimento Empresarial to provide free business training to entrepre- neurs. However, its services are geared more toward small and medium-size enterprises than toward microenterprises. Two other government institutions-Centro de Formaq5o e Aperfeico- amento Administrativo and Instituto de Formacao e Aperfeicoamente Pro- 100 Country Studies fessional-provide practical training in areas such as carpentry, stone working, and typing. Both institutions receive funding from USAID, and both have expressed an interest in introducing business-related courses targeted to the private sector. Government policy Cape Verde's industrial development policy calls for the development and promotion of artisanal and informal activities. The 1989 law on industrial development states that "a special regime will be established to support and promote the effective development of artisanal and informal production and encourage its transition towards more advanced forms of organization, notably through the creation of industrial cooperatives or other forms of associations of industrial production" (Woilet 1992, p. 192). Since the creation of this law, a number of government agencies, including the ministries of education, tourism, industry and commerce, labor, youth, and social promotion, and fishing, agriculture, and rural activities, have each been struggling to take the lead in promoting the development of the informal sector. As a result, coordination among the various ministries has been sorely lacking. The Ministry of Labor, Youth, and Social Promotion, recently named as the coordinating agency for all matters affecting the informal sector, is drafting a policy framework for the development of the informal sector that will provide the context for government action and donor assistance. Prospects Cape Verde appears to offer better prospects for the development of small and medium enterprises than of microenterprises. Given its limited resource base and small domestic market, the country seems better suited to developing modern, efficient small and medium enterprises that can provide export-oriented, relatively inexpensive products and services. Cape Verde has many elements that would allow it to move in this direction. It has a strong base of human resources, as refleeted in a hard-working, disciplined, and relatively well-educated population. Furthermore, it has the advantage of a large overseas emigrant community that maintains close ties to the country and contributes significantly to the economy through remittances. In this respect, Cape Verde presents many interesting similarities with Taiwan, another small island economy whose long-term development seemed initially handicapped by poor natural resources as well as by other factors. Yet, despite these great odds-or perhaps because of them-the Taiwanese economy recorded growth rates averaging 10 percent during the 1970s and 1980s. Most of this economic growth was fueled by small and medium enterprises that responded to government incentives designed to Cape Verde 101 promote export-oriented small and medium enterprises. Like Cape Verde, Taiwan had the advantage of a large network of overseas Chinese who not only provided capital but also facilitated trade links with other countries. Most important, Taiwan made huge investments in education and training, which, in the short span of three decades, transformed its comparative advantage of cheap, unskilled labor in the 1960s to inexpensive computer engineers in the 1990s. In the long run, a similar outward-oriented strategy might be the most effective way to promote private sector development and reduce poverty in Cape Verde. Given the small, fragmented domestic market, it would be difficult for Cape Verde to promote industrial development strictly within a national context. The largely unexploited tourist and fishing industries and maritime-related services might offer significant opportunities for develop- ing small businesses. Joint ventures with foreign investors or with overseas Cape Verdeans in these sectors could stimulate private sector growth through the transfer of capital and knowledge. Similarly, microenterprise activities serving the internal market, such as construction, food processing, and electrical and mechanical repairs, could be strengthened to become more productive and efficient. Investments in higher education and professional training would enable Cape Verde's already educated workforce to compete in specialized sectors such as offshore banking and teleporting. Financial services A major conclusion of this review is that financial services for microenterprises are very limited in Cape Verde. The absence of credit mechanisms for low-income entrepreneurs, in both formal and informal financial sectors, constrains business development. The assumption here is that although private transfers from overseas emigr6s play an important role in financing microenterprise start-up, they are not sufficient to finance operating costs or improvement and expansion activities that would lead to more efficient, profitable, and stable businesses. Available information suggests that microenterprise lending is not an established tradition in Cape Verde and therefore should be developed slowly. Unlike other parts of Sub-Saharan Africa, it is not clear how strong the demand for microcredit is. Thus, care should be taken not to overwhelm beneficiaries (institutions as well as borrowers) with more credit than they need or are able to manage. Support to microenterprise development in Cape Verde should be developed on a pilot basis and undertaken as a small- scale activity combining financial support with appropriate training and technical assistance. 102 Country Studies Technical assistance Because there is little tradition of microenterprise support programs in Cape Verde, technical assistance should be an important component of any microenterprise assistance program. Technical assistance to entrepreneurs would be helpful in areas such as identifying and understanding market opportunities, designing products for existing markets, improving product quality and presentation, and managing the business effectively (Woilet 1992, p. 38). To this end, technical assistance should focus on providing enterprises with marketing assistance and business management to help them develop skills and practices for sound business. It should be channeled through one or more of the existing local support institutions (preferably through private institutions) and should be carried out in such a way as to develop a sustainable capability in those institutions. The staff of existing support programs, such as INC and OMCV, lack the resources and experience they need to carry out their activities on a large scale. These institutions would require assistance if they were to launch a large-scale training program for entrepreneurs. Government coordination The government of Cape Verde has recognized the need to take a coordi- nated approach to microenterprise development. One possible way in which the World Bank could assist the government to define its approach to microenterprise development would be to bring together the various actors in the microenterprise sector: the government, donors, NGOS and other specialized institutions, and, most important, representatives of microenterprises themselves. This could take the format of a one- or two-day seminar that would address the specific needs of and constraints on microenterprises and the means through which assistance could be provided more effectively. Such a seminar would help to establish an open dialogue between key players, to develop a framework for government, donor, and NGO interventions in the sector, and to disseminate best practices for delivering financial and technical services to the poor on a sustainable basis. Notes This chapter was written in June 1994 and was revised based on new information gathered from missions to Cape Verde in September 1994 and February 1995. 1. Cape Verdeans traditionally have emigrated to the United States, Europe, and continental Africa. More than 500,000 Cape Verdeans reside in the United States. Remittances from emigrants averaged nearly 12 percent of GDP between 1989 and 1992, with 60-70 percent of Cape Verdean families receiving money from relatives abroad. About 50 percent of transfers are Cape Verde 103 investments in durable goods, with 80 percent going to build homes. See World Bank 1994d, p. 2. 2. Both public and private transfers accounted for more than 36 percent of GDP during the second half of the 1980s. During the 1970s, when Cape Verde suffered a long drought, international aid-particularly food aid-represented as much as 60 percent of GDP. See World Bank 1993a, p. 15. 3. Teleporting is a type of information technology infrastructure that supplies remote access to computers around the world and allows services to be outsourced to remote locations. 4. Based on an exchange rate of 79 Cape Verde escudos equal 1 U.S. dollar. 5. Caixa Econ6mica de Cabo Verde was converted from a postal credit union to a commercial bank in 1993. It provides consumer and housing loans as well as small business loans. Banco de Cabo Verde is the central bank. Established in 1993, it serves as a commercial and development bank. Instituto de Seguros e Providencia Social is an insurance and social security mortgage company that was created as part of the financial sector reform and is the first joint venture with a foreign partner. 8 Chad Ousa Sananikone Chad is the fifth largest country in Africa, covering 1.28 million square kilometers. With 6.2 million inhabitants and a per capita income of $220 in 1992, it is also one of the poorest countries in the world. Economic develop- ment efforts have been hampered by recurrent civil strife, severe droughts, and precipitous falls in the world price of cotton, the country's main commodity. Chad presents a major development challenge. It juxtaposes an uncertain political situation and extreme poverty with good long-term prospects for growth (World Bank 1993k, p. 59). Compared with other Sub-Saharan countries, Chad is relatively well endowed with natural resources, including significant oil reserves, which, if properly exploited, could stimulate economic growth in the medium term. Food production is generally sufficient to meet domestic demand, and agriculture-particularly livestock and cotton production-is a key source of future growth. Yet, despite major postwar reconstruction efforts, Chad's physical infrastructure, human resources, and institutional capacity to support development remain at the same level as those of most low-income Sub- Saharan countries in the 1960s. Chad ranks 150th out of 160 countries on the UNDP'S 1992 Human Development Index, and it has some of the lowest development indicators of all Sahel countries. The crude death rate of 19 per 1,000 is twice the average of low-income countries, life expectancy at birth is forty-seven years, and the under-five mortality rate is 21 percent. In 1990, 70 percent of the adult population (including 82 percent of women) were illiterate, and the secondary school enrollment ratio was only 7 percent (World Bank 1994h, pp. 66-67). The Chadian economy is based almost entirely on agriculture, which accounts for 43 percent of GDP and provides a livelihood for more than 78 percent of the population. Subsistence production of food crops is the principal economic activity, but output fluctuates widely depending on rainfall. Livestock and cotton provide just over 50 percent of export earnings, down from 77 percent in 1983, and are the major sources of cash income in rural areas. The formal industrial sector contributes about 18 percent to GDP and is dominated by Cotontchad, a state-owned enterprise that processes and exports cotton. The rest of the industrial sector includes several medium-scale enterprises and a large number of microenterprises 105 106 Country Studies that produce consumer goods for the domestic market. The service sector makes up the remaining 39 percent of GDP and includes a large trading sector. In 1987 the government launched a structural adjustment program aimed at improving public resource management. By 1990 the fiscal deficit had been reduced to 2.2 percent of GDP, fiscal revenues were increasing steadily as a result of tax reforms, and progress was being made in revitalizing key sectors such as cotton and transportation. Real GDP growth, though erratic due to fluctuations in agricultural output, averaged 4.5 percent a year between 1987 and 1990. Inflation varied widely with food crop prices, but averaged less than 3 percent a year during the adjustment period. The recurring military and civil conflicts that have beset the country since 1979 have severely hampered the government's capacity to manage projects and absorb donor investments, with some 40 percent of government expenditures going to the military, and civil service salaries paid irregularly (World Bank 1994b, p. vii). Since 1992 the IDA (International Development Association) has been implementing a minimal assistance program to alleviate poverty by financing activities that address basic poverty issues, place minimal demands on government administration, and improve the prospects for long-term growth. The program has emphasized operations in health, education, agriculture, infrastructure, and energy. In response to the opportunity presented by the realignment of the CFA franc in 1994 and the government's stated commitment to implement a new program of economic reforms, the Bank and IMF (International Monetary Fund) are now deepening their policy dialogue with the government toward the adoption of a medium-term policy framework (World Bank 1995, p. 1). Main characteristics of the informal sector The development of the informal sector in Chad can be best understood in the context of the adverse economic and political climate that prevailed in the country during much of the past twenty years. Intermittent civil war destroyed regional economies and uprooted hundreds of thousands of people. In addition, the economy was severely affected by drought in 1984 and 1990, which dramatically reduced the production of fish, livestock, and cotton-all mainstays of the economy. This crisis situation was exacerbated by the precipitous drop of nearly 50 percent in world cotton prices between 1981 and 1986. The combination of continuous political instability, natural disasters, and external shocks impoverished the population even further. In response to the loss of economic activities, many peasants migrated to urban areas in search of income-generating activities and thus contributed to the growth of a large informal sector. Another significant factor account- ing for the rise in informal sector activities, particularly in the capital city of N'Djamena, was the collapse of the formal commercial and service sectors in the late 1970s, when most of the Greek, Lebanese, and Syrian traders who Chad 107 handled retail and wholesale trade left Chad and did not return due to the civil war. Consequently, Chadians moved to fill these gaps by expanding a wide range of informal activities. Chad has a large and vibrant informal sector that is estimated to account for two-thirds of GDP, including the bulk of agricultural, trade, and services output. The informal sector accounts for roughly 75 percent of employment in N'Djamena (World Bank 1994f). Within N'Djamena, 50 percent of informal sector activities are concentrated in trade, approximately 35 percent in services, and the remaining 15 percent in artisanal activities (World Bank 1989). Informal sector activities in Chad include the manufacturing of basic agricultural tools and household utensils, the preparation and sale of food, and the sale of basic consumer goods. In the service sector, informal sector workers earn income as porters, water carriers, messengers, launderers, and vendors of millet beer. Most artisans do not produce new goods but instead provide repair services for radios, bicycles, sewing machines, cars, and tires. Monthly wages in the informal sector were around CFAF 20,000 ($66) in 1991, substantially less than the average monthly income of CFAF 65,000 ($216) in the public sector (World Bank 19940. The informal sector supplies goods and services primarily to low-income groups, including low-paid and in many cases unpaid civil servants who cannot afford to buy commodities in the formal urban market. N'Djamena is the hub for this clientele, because the majority of the civil administration is located in the capital. Main constraints on the informal sector Many of the key constraints confronting informal sector enterprises in Chad are the same regulatory, financial, and technical problems that plague the informal sector in other Sub-Saharan African countries. These include the following. * Limited size of the domestic market. Low population density and inadequate transportation and communication systems have resulted in relatively small, isolated domestic markets that cannot sustain the growing informal sector. Even urban markets are not sufficiently large to support the increasing numbers of people who depend on informal activities for their livelihoods. * Low purchasing power of households. The fiscal crisis besetting the govern- ment has dramatically reduced the disposable income of many Chadian households and thereby limited demand for goods and services supplied by the informal sector. In N'Djamena, where low-paid public workers constitute the main clientele of the informal sector, thousands of civil servants have not been paid for as long as ten months. 108 Country Studies * Insufficiency and high costs of public services. The shortage and high cost of public services and physical infrastructure in Chad are major impedi- ments to private enterprise development in general. These barriers exact even greater costs for microenterprises that need basic services such as water, electricity, and adequate roads for their day-to-day operations but have limited financial resources. Electricity in Chad is particularly expensive and is not available in adequate quantities to many microentrepreneurs. * Lack of access to credit. Microentrepreneurs apparently can obtain sufficient funds to start a business, but most experience chronic shortages of working capital. The formal financial sector reportedly does serve some small and medium enterprises, but it does not lend at all to microenterprises.' The bulk of available credit in the formal financial sector is allocated to Cotontchad. In addition, many Chadians stay away from formal financial institutions because they do not trust them. A key factor behind this distrust is the lack of a secrecy code in Chad's banking laws, which has resulted, during certain periods, in "unofficial" taxes being levied directly on bank accounts. Many Chadians do not use formal banks because they are afraid their savings will be fined or confiscated, and formal banks have only been able to mobilize a very low level of savings. As a result, microenterprises typically avoid interaction with formal financial institutions and must turn either to informal sources of credit such as tontines or to credit programs provided by a small number of NGOS. By virtue of their limited capital, microenterprises operate under significant regulatory disadvantages vis-a-vis larger firms. For example, under the current investment code, small-scale enterprises must have an investment of at least CFAF 15 million in equipment to benefit from certain tax exemptions. Those that qualify pay a tariff of 5 percent on imported equipment and are exempted from paying income and profit taxes for five years. These regulations clearly are biased against small businesses, which rarely have capital exceeding CFAF 300,000 (World Bank 1989, p. 5). The growth of the informal sector may also be due to the high costs of entering the formal sector. For example, it reportedly takes more than a year to process all the requisite papers to start a business in Chad. Box 8.1 contains a list of administrative procedures that must be completed before permission to start a business is granted. Once registered, entrepreneurs in the formal sector are subject to roughly eighty different taxes. In addition to these official taxes, licensed firms are often asked to make "voluntary" contributions to the war efforts (taxes de guerre). Moreover, formal sector firms are subject to a host of government regulations that few informal sector entrepreneurs can afford, such as minimum wage requirements, restrictions on imports and exports, and some Chad 109 obligations to trade with state-owned enterprises. Given these numerous constraints in the regulatory and policy environment, it is not surprising that many microenterprises choose to remain informal. Box 8.1 Main administrative procedures necessary to start a business in Chad To obtain a permit from the Department of Commerce and Industry, the following steps are required. 1. The entrepreneur must present a request with a health certificate and documentation showing no prior criminal record. 2. If approved, the entrepreneur must complete an application form and submit it with the following: a draft of the by-laws, the nature of investment and sources of financing, proposed employment and training programs, and proposed field of activity. 3. The application must then be approved by the following departments: the Industrial Promotion Office, the Division Chief of the Planning and Industrial Promotion Office, the Director of Industry and Cooperatives, and the General Director of Commerce and Industry. 4. After authorization has been received from all four departments, the entrepreneur must provide two photos, a receipt from the tax office for cFAF 5,000, and a stamp for CFAF 3,000. 5. Authorization to set up a business is then granted. Once authorization to set up a business is granted, entrepreneurs must complete the following: 1. The by-laws must be registered at the Direction des Domaines, de l'Enregistrement, et du Timbre. 2. The entrepreneur must open a bank account. 3. The firm must be entered into the official registry (Registre du Commerce). 4. A legal notice must be published. 5. The business must be registered at the Tax Office (Direction de l'Imp6t et Taxes). 6. The business must be entered in the employer registry and registered with the Social Security Office. 7. The business must be registered with the statistical office (Direction des Statistiques). Source: Hennart 1992. 110 Country Studies Policies and programs for microenterprise development A review of donor agency and NGO programs in Chad identified three that are relatively successful: the USAID-funded VITA Private Enterprise Project, the World Bank's Social Development Action Project, and the N'Djamena Market Women Project of CARE (the Cooperative for American Relief to Everywhere). The Private Enterprise Project was started in 1984 with funding from USAID and was implemented by Volunteers in Technical Assistance, a U.S. NGO based in the region of N'Djamena. The objectives of the project, known as VITA-PEP, were to assist postwar reconstruction by promoting small business development, particularly in manufacturing and services, within the areas of N'Djamena and Moundou. The success of the first phase of the program provided the basis for a second phase in 1987 and a third phase in mid-1990, which called for additional microlending activities. The World Bank's Social Development Action Project was started in 1991 and finances the third phase of VITA-PEP. The objective of the project is to mitigate the adverse impact of structural adjustment on the poor and the unemployed. To that end, the project includes a microenterprise component ($2.6 million out of a total project amount of $13.4 million) that aims to stimulate employment creation by providing credit to small and microenterprises and to improve the productivity of microenterprises by providing technical assistance. The credit scheme offers three credit windows, each in conjunction with technical assistance, to address the investment and management needs of micro, small, and large enterprises. The maximum loan sizes are $1,000 for microenterprises, between $1,000 and $25,000 for small-scale enterprises, and between $25,000 and $50,000 for larger enterprises. Interest rates range from 15 to 24 percent a year. The project also includes a technical assistance component implemented by a small unit within the Ministry of Labor. Technical assistance in product design and quality control, facilitation of joint projects among small enterprises and technology transfer among microenterprises, and training in record-keeping and selection of appropriate machinery are provided. VITA was selected to implement the microenterprise component because of its proven effectiveness in small-scale lending. In effect, the USAID- financed Private Enterprise Program was expanded under the Bank's Social Development Action Project. To date, results have far exceeded project objectives. The initial aim was to commit $1.6 million (out of the $2.6 million microenterprise component) in credit to 700 micro and small enterprises and to create about 1,200 jobs. From July 1990 to December 1993, the project disbursed $3.6 million. A total of 3,513 loans were made, of which 91 percent went to microenterprises and 9 percent to medium- and small-scale enterprises. The average loan size was $284 for microenterprises and $9,286 for medium- and small-scale enterprises. During this period, 2,568 jobs were created at an average cost Chad 111 of $1,405 per job. The average cost per job among microenterprises was $525 (World Bank 1990a, p. 3). Three-quarters of total loans were made to women. As of March 1994, agriculture-related loans comprised 62 percent of the value of all medium- and small-scale loans and 92 percent of all microloans. Of these loans, 6 percent were used for production-related activities, 16 percent for the processing of agricultural products, and 78 percent for the marketing and transport of agricultural products. Clients' balance sheets for microloans indicated that gross returns increased 32 percent after the first loan, sales increased 24 percent, profits increased 35 percent, and net cash flow increased 32 percent. During the first quarter of 1994, loan repayment increased from 92 to 100 percent of the total amount due. This improvement was primarily a result of better loan recovery procedures, a shift toward more microclients, and a greater number of repeat and female borrowers (VITA/Chad Private Enterprise Development Project, First Quarter CY 1994b, Quarterly Progress Report). The project's experience is that women borrowers repay at higher rates than men. In 1991 VITA-PEP opened a Business Resource Center that provides training, technical assistance, and reference publications to clients. Group and individual training sessions provide simple accounting skills and administrative and contractual procedures. VITA-PEP trained fifty-one clients from September 1991 through March 1992. Most trainees were women, and nearly all were microclients (VITA/Chad Private Enterprise Development Project 1992). With financing from USAID and later from the World Bank, VITA-PEP has demonstrated that it can serve the financial needs of micro and small enterprises. Its strategy is to become an autonomous local NGO capable of generating its own financial resources. With this in mind, measures to achieve greater financial self-sufficiency have been taken over the past few years, including streamlining lending procedures and reducing operating costs. Loan demand continues to be very high, and there is no evidence that it would change significantly if interest rates were higher (VITA/Chad Private Enterprise Development Project 1992). VITA-PEP'S financing from the Bank expired in June 1994, and the project is seeking additional funding from USAID (which will fund operations through 1997) and the World Bank. CARE'S N'Djamena Market Women's Project is a collaborative effort between CARE and VITA-PEP. Its objectives are to facilitate market women's access to credit and to help them mobilize savings. CARE'S staff assist women to form solidarity groups and to prepare loan requests for VITA-PEP. Most women are engaged in the processing and sale of agricultural products. An evaluation conducted in September 1993 indicated that the project had achieved satisfactory results, given that it was the first to work with urban market women and the first collaboration between the two NGOS. The 112 Country Studies evaluation found that the original project goals were overly ambitious and that too much emphasis had been placed on group organization. However, valuable experience was gained from working with market women in an urban setting. Between 1991 and 1993, the project helped 180 women (far below the goal of 800) to form twenty-eight solidarity groups and to apply for loans. It made sixty group loans (corresponding to 359 individual loans) for a total amount of CFAF 20 million ($67,000). The average loan size was CFAF 30,000 ($100), and the repayment rate was 100 percent. Project participants reported that their incomes and profits had increased. Many women were able to negotiate lower prices by paying cash for their purchases instead of buying on credit. Some women diversified their range of products, and some graduated to become medium-scale retailers and wholesalers. Finally, the World Bank's Public Works and Capacity Building Project aims to reduce poverty by generating employment through labor-intensive public works. Work is carried out mainly by local contractors, including t&cherons (tradesmen) from the informal sector who are among the poorest workers. The project is expected to provide employment and to assist people in making the transition from the informal to the formal sector.2 Conclusions With the continued deterioration of the Chadian economy and the spiraling fiscal crisis, there is a high demand for very small loans, particularly on the part of women. Moreover, the need for microcredit is being met almost exclusively by informal sources of credit and by a relatively small number of NGO credit programs. VITA currently meets only about 11 percent of the demand of microentrepreneurs in N'Djamena. More and larger microfinance programs will need to be developed if the current minimal scale of outreach is to be expanded. Evidence also suggests that the current legal and regulatory framework may be biased against the development of small businesses. Areas where further examination and possible reform might be in order include the number and complexity of administrative procedures required to register a formal business, the number and rates of taxes imposed on informal businesses, the schedule of import taxes and any biases it contains against small enterprises, the effect of minimum wage regulations on small firms, mandatory caps on interest rates, and the security and confidentiality of bank accounts. Notes 1. The formal financial sector consists of the Banque des Etats de l'Afrique Centrale, the Central Bank, and five commercial banks: Banque Meridien Chad 113 BIAO du Tchad, Banque Tchadienne de Cr6dit et de D6p6ts, Banque Tchadienne de D6veloppement, which was recently restructured with the help of the African Development Bank, Banque Tchadienne Arabo- Lybienne, and Financial Bank of Chad. 2. World Bank 1994f. The project is based on the positive experience of previous IDA public works and service projects using labor-intensive jobs in countries such as Mali, Niger, and Senegal. Similar projects are being implemented or being proposed in Benin, Burkina Faso, the Gambia, and Mauritania. 9 The Gambia Simel Esim The Gambia is a coastal, riverine West African country bordered by Senegal on all three sides, and by the Atlantic Ocean to the west. Per capita GNP is $370. Population densities vary throughout the country between the urban areas of Banjul and Kanifing, which have more than 3,000 persons per square kilometer, and the rural areas of Kuntaur and Mansakonko, which have less than 50 persons per square kilometer. Approximately 76 percent of the country's I million inhabitants depend on farming the flat, arable soil along the Gambia River. Groundnuts are the main agricultural commodity, accounting for 55 percent of GDP and more than 80 percent of exports. Rice is a staple food crop. Other resources include fisheries and sites with the potential for tourism. In 1991-92 agriculture contributed about 20 percent of value added, and industry contributed about 12 percent; the remainder came from services, including trade (16 percent), tourism (4 percent), and public administration (10 percent). Manufacturing made up only 7 percent of GDP (World Bank 1993c, pp. 187, 191). The economy is vulnerable to fluctuations in the world prices of agricultural products such as groundnuts and cereals, adverse weather conditions, a changing foreign aid climate, economic conditions in neighboring countries, and shifts in the pattern of Western European tourism. The economy is highly dependent on imports. Because of limited natural resources, most food, all fuels, many intermediary and capital goods, and most manufactured goods must be imported (World Bank 1993c, pp. 188-89). The Gambia's poverty is reflected in its low level of human resource development. The Gambia trails most of West Africa in literacy (73 percent illiteracy) and has high infant mortality (132 per 1,000 live births) and low life expectancy (forty-five years).' The population growth rate is quite high, reaching 3.4 percent in 1992.2 On average, the per capita income of urban households is one and a half times the average per capita income of rural households.3 These income differences reflect the low level of education and the limited employment opportunities of most rural households. The Gambia's Economic Recovery Program, a comprehensive macroeco- nomic stabilization and adjustment program, was adopted in 1985. Facing mounting external debt, a depletion of official reserves, an excessively expansionary fiscal policy, high levels of inflation, and an overvalued 115 116 Country Studies currency, the government initiated a series of macroeconomic reforms that included liberalizing trade regulations and releasing exchange and interest rates. Restrictive credit and fiscal policies were adopted to control the growing budget deficit and to bring down inflation. Early implementation of reforms was generally satisfactory, and most macroeconomic targets were met. The Program for Sustained Development, which started in 1990, experienced some delays in implementation; however, reform efforts have been resumed, and the program is back on track and supported by a large influx of external aid. Tangible effects of structural adjustment reforms measure quite positively (Hadjimichael and others 1992). Some of the most notable include: * A reduction of the budget deficit from 17 percent of CDP in 1988 to -2.7 percent in 1993. * Divestiture of 60 percent of public enterprises. * Growth in real GDP by an average of 4 percent a year in 1991, 2.1 percent in 1992, 1.3 percent in 1993, and -4 percent in 1994. * Reduction in annual inflation, as measured by changes in the consumer price index, from 70 percent in 1985 to 5.9 percent in 1992. * Maintenance of interest rates, through open market operations, at positive levels in real terms (measured in relation to inflation during the previous twelve months) and with appropriate differentials with respect to interest rates abroad. The Gambia's adjustment efforts since 1985 have stimulated employment opportunities in the urban and formal sectors of the economy, but not everyone has benefited. The leveling off of agricultural output in response to a substantial decline in groundnut prices has brought deterioration in living standards among much of the rural population. One result of this has been massive urban migration and the development of a large urban informal sector. With an urban population growth rate of more than 6.4 percent a year and insufficient job opportunities in the formal sector, the informal sector has grown in both numbers and importance (World Bank 1994h). Main characteristics of the informal sector A 1990 study by the International Labor Office in Geneva estimated that approximately 60 percent of the urban labor force is engaged in informal employment, and 30 percent of the rural labor force is employed in the informal sector.4 Figures from The Gambia's June 1993 budget indicate that the informal sector has probably grown even larger since 1990. The June 1993 budget listed total employment at 106,000 and formal sector employ- The Gambia 117 ment at 36,000, indicating that roughly 66 percent of employed Gambians work in the informal sector.5 Many small business activities have always taken place in villages. Women constitute more than half of the agricultural workforce and contribute to approximately 40 percent of agricultural production.6 Although they are becoming more involved in cash crop production, they are primarily engaged in the production of food crops (growing and processing vegetables and groundnuts). Rural women also trade fresh and processed foods at local markets, along with basic household goods, homemade cosmetics, and processed roots and leaves. Men are involved in agriculture, the drying and smoking of fish, hulling of grains, bee keeping, and rope making. At the Yellinda ferry, a main trading center, men run open-air food stalls that sell meat, T-shirts, cold water, and fruits. They also offer their loading services to traveling passengers. The Gambian economy is based primarily on agriculture, but thousands of microenterprises operate in urban areas. In the Greater Banjul area, these enterprises absorb a large portion of the urban and semiurban unemployed. They seldom have more than three workers per unit, but unpaid apprentices typically assist in metal and woodworking shops. According to official national accounts, small-scale manufacturing (formal and informal) accounts for less than 10 percent of GNP. Informal manufactur- ing includes tailoring, handicrafts, blacksmithing, welding, automobile repair, baking and processing of peanuts, processing of fish, and preparation of hides (see box 9.1 for a description of the handicrafts sector). Most masons and carpenters work in the urban informal sector. Most are self-employed or employed for wages by enterprises with fewer than five people. Women have been particularly vulnerable in the years following adjustment. Gambian women comprise roughly 50 percent of the population between the ages of fifteen and fifty, but they account for only 15 percent of formal employment. They do, however, play a large role in the informal sector, supplementing household income through a range of activities. An estimated 30-50 percent of women living in urban areas earn income in the informal sector, typically as self-employed traders, fish processors, grain millers, and seamstresses.7 Jewelry is a repository of women's capital, especially in the form of large gold earrings. If a female entrepreneur cannot raise working capital through her own household savings or through her husband, brothers, or other close relatives, she will often try local pawnbrokers or the local osusu networks (see discussion in this chapter under the heading of Osusus).8 These sources of financing can be sufficient for small amounts of working capital. They typically are inadequate for enterprise formation and growth, which require larger investments. 118 Country Studies Box 9.1 Case study of the handicrafts sector in The Gambia The handicrafts sector in The Gambia provides income to large numbers of people, especially urban dwellers. Artisans produce wood carvings, batiks, tie-dye fabric, dolls, dresses and shirts, leather goods, and woven cloth and rugs. Most are small-scale operators who derive most of their income from the tourist industry. Many women produce tie-dyed fabric for clothing and home accessories. A few operate their own stalls at tourist markets, but the majority produce and sell to other vendors. There are two basic markets for handicrafts produced in The Gambia: one for local consumption and the other for tourist consumption and for export. In order of declining importance, the main buyers are tourists, expatriates, Gambians, and foreign buyers. The market for domestic consumption is already saturated, and the tourist and expatriate market is limited by its small size. Excess supply predictably depresses prices. The export market has the most potential for expansion, but it is relatively untouched by Gambian producers. An important constraint on entering export markets is the limited number of persons who produce handicrafts of sufficiently high quality. Product quality is negatively affected by lack of access to precision equipment. Another constraint is the enterprises' lack of experience with and knowledge of export markets. Most fledgling handicraft exporters are producers who travel overseas with suitcases of samples to sell to retailers. Any attempt to increase the exports of Gambian handicrafts will require time to develop links with overseas markets, improve and adapt products, train local artisans in quality control, and establish trust and confidence among artisans and buyers. Training and technical assistance are needed in three areas: product design and adaptation, quality control tied to market requirements, and assistance with development of long-term links with overseas buyers. Such assistance could be provided by foreign experts, but it is important to build local capacity in these areas. Source: Silcox 1993. Main constraints on the informal sector A key constraint for microenterprises in The Gambia is lack of access to credit. Formal financial institutions are reluctant to lend to most informal sector entrepreneurs without a substantial guarantee. In addition, many rural micro and small-scale entrepreneurs have insufficient access to larger markets, due to transportation difficulties. Existing markets for some informal sector products are saturated, but export markets are rarely explored. The regulatory framework also appears to contain some biases against small enterprises, particularly in obtaining access to import licenses. A further constraint is lack of marketing skills, poor market information, and inability to differentiate products. This frequently results in a concentration The Gambia 119 of vendors selling identical products at the same location and targeting the same customers (market saturation). Policies and programs for microenterprise development Social safety nets are based on social and religious traditions. Primary among these is a Gambian phenomenon known as the F-connection (family, friends, and firms), whereby households and small businesses operate as an extended family. Friends and family are entitled to live and work with other friends and relatives, especially in urban areas (World Bank 1993c). Sharing incomes and work is an effective safety net in times of economic crisis; this practice is a major factor explaining the low levels of income inequality found in The Gambia. Virtually everything is lent among friends and family, including land, labor, livestock, seeds, fertilizer, pesticides, and farm tools as well as craft tools, vehicles, and household goods. A family's savings are held in nonmonetary forms such as livestock, grain, machinery, and jewelry, but further wealth lies in obligations owed by kin or neighbors. Such a concept of savings is not clearly distinguishable as investment or consumption, especially when it consists of animals or consumer durables. Kafos Most prominent among local informal finance practices are kafos, which are informal but cohesive village groups that are multifunctional in nature.9 Kafo groups act as contribution clubs for money, and they resemble rotating savings and credit associations.'° Kafo groups are commonly unnamed, or their names shift with activity. Members usually are of the same age, ethnicity, gender, and occupation. Women's groups are not always fully autonomous. Commonly a group has a trusted male patron in its village who may hold the group's money or keep an account. While the head of a men's kafo usually plays a subordinate role to the village headman, the head of a women's kafo is likely to be the most powerful woman in her village. Some agricultural kafos cultivate a common plot of groundnuts and vegetables that are sold in collective crop stores, and the proceeds of sales are contributed to a village fund. Some kafos operate small restaurants, hotels, and milling machines. Kafo groups accumulate funds from members' dues and fines, contract labor, and hold dances, concerts, wrestling matches, and market stall sales. A major purpose of kafos is to help members in the event of emergencies. Such groups can mobilize contributions quickly to pay hospital bills and bail and to rebuild houses after a storm. A kafo may lend or give money, food, seeds, and tools to members and their families or to other villagers (although members have priority). In 1988 cash loans from kafo groups to individual villagers typically ranged from D 10 to D 100 (10 120 Country Studies dalasis equal 1 U.S. dollar) with nominal interest rates of 10 to 60 percent over six months. Osusus Osusus, in contrast, are informal groups that provide financial services. Osusus members make fixed contributions of money at regular intervals. At each interval, one member takes the entire pot. Every member takes a turn until the cycle is completed, and then it starts again. For people who take their turn late, osusus function as a savings mechanism. The Gambian osusus are especially important among market women in towns, but they also play a role in farming villages. In villages, nearly all osusus are run by women. In towns and cities, they are more likely to include both genders. Ties of neighborhood, gender, age, kinship, and ethnicity create peer group pressure that encourages regular participation. Each group has a recognized leader. Osusus have seven to thirty members in villages and seven to fifty members in towns. In most groups, each member pays between D 1 and D 5 weekly. The Gambian osusus are often used as a model by NGOS trying to establish microfinance programs in urban settings. Financial NGOS Since The Gambia's independence, NGOS have been active in community development efforts including health, education, and agricultural extension activities. As the informal sector has expanded, so too has the number of local NGOS, especially among groups that have been hurt by economic adjustment. In the early 1990s, the USAID mission in Banjul, along with an Ohio State University research team, undertook a study to identify the nature of NGOS' microfinance programs in The Gambia (see table 9.1). Of approximately 100 international and national NGOS identified, fourteen targeted rural areas. Little information exists as to the volume, coverage, costs of operation, and terms and conditions of microloans offered by these NGOS. Apparently, most operate under the assumption that support for income-generating activities of the rural poor requires subsidized loans. The survey also found that most loan programs focus on specific geographic areas. As elsewhere, financial NGOS in The Gambia tend to be embryonic institutions that offer fairly rudimentary financial services to targeted populations in limited geographic areas. The separate markets in which they operate lack broad national penetration. Their operational philosophies typically either view microcredit as charity or view credit and savings services as business. The large number of NGOS, their different approaches, and a lack of coordination have created some confusion in rural financial markets. The Gambia 121 Table 9.1 Matrix of selected nongovemmental organizations provid- ing financial services in The Gambia Repayment NGo and year Annual rate started Scope interest rate Average loan size (percent) AATG, 1983 177,300 mem- 0% to kafos D 2,875 ($288) 42% in late bers; over 300 until 1991; 1991 kafos; 582 currently villages 12% Save the - 15% D 500 to D 28,500 75% Children, 1985 ($50 to $2,850) Wisdom, 1989 9,031 members; 11% D 1,000-D 3,000 High 215 rural and ($100-$300) for urban kafos; short-term. D 300 villages 20,000 ($2,000) for long-term AFET, 1990 9,300 members; 24% - High 1,343 loans (82% women) Caisses 1,384 members; 40-60% D 200-D 400 92-97% Villageoises, 56 kafos; 827 ($20 to $40) 1988 loans; 6 villages Gambia 300 members; 13%, revolv- D 5,000-D 20,000 - Women's Fi- 60 villages ing; 24%, ($500 to $2,000) nance Associa- standard tion, 1987 - Not available. Source: Graham, Meyer, and Cuevas 1992, chap. 2, pp. 30-39; GEMINI 1993. Some of the most prominent local NGOS provide services to informal sector entrepreneurs in The Gambia. The Caisses Villageoises d'Epargne et de Credit Program, started in 1988 under the tutelage of the Centre International de D6veloppement et de Recherche with support from the Kreditanstalt fur Wiederaufbau (KfW), seeks to create village banks (caisses) that can mobilize local capital as savings and reinject it as credits to finance productive activities. Most village banks rely initially on savings, and then, after a couple of years, they begin lending, sometimes obtaining refinancing from a commercial bank, a development bank, or the Central Bank to leverage their funds. In part as a result of the program's decision to stress indigenously mobilized deposits over external funding, annual interest rates tend to be high. Of the six individual caisses currently offering financial services in the southern part of The Gambia, none charges annual interest rates lower than 40 percent, and some charge as much as 60 percent. The caisses also reward depositors at above-market rates, because savings earn 20-30 percent 122 Country Studies interest a year. These high interest rates are not set by foreign practitioners concerned with sustainability. Village assemblies meet and decide on the composition and responsibilities of credit committees, the interest rates, and the types of savings instruments. Moneylenders are the alternative source of credit for microentrepreneurs, and villagers reportedly are willing to borrow at annual interest rates as high as 40-60 percent. Repayment rates are high. In 1991 the lowest repayment rate reported among the six caisses was 92 percent, and some registered rates of 97 percent (GEMINI 1993). Loans are quite small. The maximum loan size is set at $300, and the average loan size is between $20 and $40. In 1991 the six caisses made 621 loans for D 240,000 ($24,000) with an average maturity of 5.6 months. They mobilized CFAF 3.2 million ($10,700) in savings (Graham, Meyer, and Cuevas 1992, p. 36). Although the caisses program operates on a moderately small scale, the quality of service appears to be high. Unlike in many other Gambian microcredit programs, there is no targeting of loans in terms of their use, repayment rates are high, interest rates are above the commercial level, and collateral substitutes, such as animals, farm tools, and jewelry, are used efficiently and effectively to enforce contracts. This program illustrates clearly that the fundamental constraint in financial services for microenterprises is not the high price of credit, but rather the access to loans. The largest NGO offering microfinance services in The Gambia is Action Aid The Gambia, a U.K-financed program that started in 1986. Action Aid lends primarily to kafos, and in 1992, it provided credit to 250 groundnut kafos and more than 53 rice kafos (Graham, Meyer, and Cuevas 1992, p. 35). The strength of the program is its scale. Its loans accounted for nearly 12 percent of the volume of all agricultural loans made in 1992, reaching more than 375 villages. Until 1991 it did not charge any interest on loans made to kafos, and only in the past couple of years did it raise interest rates to 12 percent a year. The repayment rate was only 42 percent in 1991. The Gambia Women's Finance Association, an affiliate of Women's World Banking, was set up in 1987 to facilitate the access of female entrepreneurs to bank loans. It serves roughly 300 members concentrated in the Banjul area, and its loans range between $500 and $2,000 (Graham, Meyer, and Cuevas 1992, p. 36). Annual interest rates are set at 12.5 percent for revolving funds and 23.5 percent for standard funds. In 1991 only a very small number of loans were provided, because the association concentrated on providing technical assistance to businesswomen and on working with collateral issues. Similar to the principles of the Caisses Villageoises d'Epargne et Credit Program, the Association of Farmers, Educators, and Traders emphasizes the mobilization of local deposits. Its operational philosophy is to use kafos as solidarity groups and screening committees to improve the distribution The Gambia 123 of credit, while reducing the risks associated with rural finance. The savings program links savings to members' access to credit. The interest rate on loans is 24 percent, and savings earn 15 percent for deposits under $25 and 18 percent for deposits over $25 (Graham, Meyer, and Cuevas 1992, p. 36). Collateral requirements for loans are rather stringent, requiring three guarantors and 75 percent of the loan amount in savings. In 1991 the association had 1,343 borrowers (83 percent of whom were women) and operated in thirteen villages. The Gambian Cooperative Union is a government union of cooperatives that provides quality inputs such as transport and commercial credits to small farmers. Although their records have not been impressive, these cooperatives have dominated the rural credit picture with their inefficient yet rather indispensable operations.'" Catholic Relief Services has been supporting the Sesame Seed Growers Association since early 1987. The program originally had a strong subsectoral focus on oil production and processing but has recently adopted a more broad-based development agenda in response to the needs of its members. The program has moved out of direct technical support and now provides more general training in business management and marketing. Recently, the association has begun providing savings and lending services to economic activities besides sesame seed growing. Currently 56,000 women belong to about 800 village kafos associations. One final NGO that does not offer financial services but nevertheless plays an important role in the lives of many Gambian microentrepreneurs is the Association of Nongovernmental Organizations (TANGO). TANGO was formed in 1983 to provide an umbrella framework under which NGOS can consult and cooperate with one another to enhance developmental activities. To facilitate coordination, TANGO publishes a local directory of NGOS and a quarterly magazine: TANGO Talks. It coordinates the flow of funds from international donors such as UNDP and USAID and from national NGOS. It also organizes short-term training sessions for NGOS involved in rural finance. Formal financial institutions Formal financial markets have been shrinking in The Gambia over the past decade, due, in part, to the rapid decline of the country's largest bank: the Gambian Commercial and Development Bank. Further contraction resulted from the collapse of agricultural lending, which fell from 35 percent of total lending in the early 1980s to 7-8 percent in the early 1990s, in part due to the withdrawal of government guarantees. The Central Bank of The Gambia regulates and monitors financial institutions and has initiated regular dialogue with NGOS, with the objective of implementing its policies on interest rates, linking loans to savings, and encouraging coordination among the various NGOS. With the recently 124 Country Studies created Agricultural Credit Unit funded by the UNDP, it is promoting village-level financial institutions (Graham, Meyer, and Cuevas 1992). The Indigenous Business Advisory Services promotes microentrepreneurs in The Gambia by providing loans to laid-off civil servants. It is perhaps perceived by borrowers as an implicit severance grant rather than as a loan, given a default ratio of 77 percent. Such a program clearly can continue only through a substantial flow of subsidies from donors or the government, since loan recoveries are not sufficient to cover transaction costs or to maintain the capital base. Internationally coordinated programs and projects Numerous projects have been initiated for women in The Gambia. For example, Save the Children-U.S.A., through an OXFAM grant, established a loan program in 1985 for income-generating activities among rural women. Also The Gambia's Women's Training Program provides training in income- generating activities such as tie-dyeing, soap making, and food preservation. The World Bank's Women in Development Project is a five-year, multisectoral, free-standing project (the first of its kind in the World Bank) with total project costs of $15 million and a World Bank loan amount of $7 million. It has passed its mid-term review, and its overall status is positive. It was designed to improve women's productivity and potential to earn income, to strengthen government institutions to enable them to deal effectively with women's issues, and to contribute to changing The Gambian society's perception of the role of women. Three of its six compo- nents-agriculture, skills development, and a small project fund in support of NGOs-are aimed at enhancing the productivity of women operating in the informal sector. The agriculture component has achieved a notable degree of success in increasing the participation of female farmers in extension activities. T'he skills development component has two subcomponents specifically targeted toward rural women: the Skills Development Program, which provides literacy, numeracy, and skills training to approximately 10,000 rural women, and an Improved Access to Credit Program, which provides credit and related services to rural women who have obtained the basic literacy and skills training. The Skills Development Program has been relatively successful, because classes have been initiated in most of the target villages and participation and enthusiasm are high. The Improved Access to Credit Program has experienced delays due to difficulties in securing the technical assistance. Local tutors have been identified and trained, and classes have been initiated. However, requiring that technical assistance be in place before proceeding to other substantive activities continues to delay the credit component of the project. Still, the credit program is operational, and a repayment rate of 90 percent has been reached. The Gambia 125 GVD Trust Fund/uNDP is a trust fund of $325,000, established by the UNDP in eleven villages to support village development groups and national NGOS that promote participatory income-generating projects, especially for women and youth. Initially, the program was designed to support technical assistance and training for local NGO savings and credit associations, but the strategy has changed to make direct loans to local NGOs. Recommendations Credit is not the only answer to microenterprise development."2 Osusus have several sound principles, such as the reliance on savings and the use of members to screen applicants for character, that have been the mainstays of these informal financial networks. International organizations would be well served to incorporate such principles into their programs. The large numbers of NGOS in The Gambia appear to be poorly coordi- nated. In order to overcome this, TANGO could be strengthened to carry out its ability to serve all NGOS throughout The Gambia. Currently, its staff is too small to provide more extensive training of NGOS or to coordinate NGO activities. However, with sufficient resources, it could provide an effective body for NGO interests, a forum for dialogue, and a centralized training center to offer a range of services for local NGOS. The scale of existing programs could also be expanded. Although many NGOS are in operation, microenterprise development programs only reach a very small subset of microentrepreneurs. Scaling up should be a concern for the World Bank's Small Enterprise, Women in Development, and Program for Sustained Development projects. There has been a high volume of misallocation and theft charges in The Gambia, within both formal and informal financial institutions. This raises the question of how programs could be monitored more effectively. For example, a loan tracking information system could be introduced to the activities of the Indigenous Business Advisory Services to reduce the risks of default. Some women in development programs are operating in The Gambia, but their scope is limited. These programs should be expanded by strengthening existing women's associations. Notes 1. World Bank 1994h. In recent years, the Gambian government has increased budgetary allocations for social services. The combined share of recurrent outlays on education and health in total current spending (excluding interest payments) after declining to 13 percent in 1987 increased to 32 percent by 1992. Seven health centers have been upgraded to improve services in rural areas, maternal and child health services have been expanded, and national immunization campaigns against childhood 126 Country Studi!s diseases have raised the national immunization average from 55 percent in the mid-1980s to more than 70 percent. Hadjimichael and others 1992, pp. 24-25, 37. 2. World Bank 1993c, 1994h. Population growth rates for the past decade were approximately 4.1 percent a year for the country as a whole, but with significant variations between urban and rural areas due to extensive rural- urban migration. 3. The Cornell Food and Nutrition Policy Program undertook a household income, employment, and consumption survey in The Gambia during 1989 and 1990. Analysis of the household data set has identified important socioeconomic characteristics of urban and rural households in The Gambia. 4. Previously, the informal sector was thought to account for only 40 percent of total employment in urban areas. The ILO survey of employment in The Gambia found similar results, although it primarily measured informal sector activities such as manufacturing, construction, and repair services and neglected to include the large number of self-employed people in transport and trade at marketplaces and along roadsides. Ahmed and others 1992, p. 25. 5. The fact that the June 1993 budget placed the number of people available for employment at about 450,000 reflects the extent of the country's unemployment problem. 6. Agricultural production is not considered a part of the informal sector in this chapter, although the growing and processing of food for commercial purposes are. 7. Women in the urban informal sector are also heavily involved in tie- dyeing, handicrafts, and soap making. Many women in rural areas are involved in informal employment, but it is hard to differentiate between family labor in agriculture and informal sector activity. 8. When it comes to cash, there is much less free sharing of wealth in Gambian marriages than in most countries outside of Africa. It is quite common for married entrepreneurs to keep separate accounts. 9. In Fula, kafo groups are more commonly called compin, from the word company. 10. Some NGOS in The Gambia recently have experimented with using kafo groups as conduits for microloans. Some kafos seem to provide a good link for connecting poorer microentrepreneurs with informal financial institutions. 11. An investigation has revealed that the Gambian Cooperative Union is effectively bankrupt. The disappearance of funds from a World Bank revolving credit scheme and an agricultural development scheme in the Jahally Pacharr Project is still under review. Shipton 1992, p. 25. 12. All of these recommendations are based on the results of a desk study and need to be tested in the field. 10 Guinea Angela Walker Guinea is a coastal country bordered by Guinea-Bissau, Mali, and Senegal to the north and C6te d'Ivoire, Liberia, and Sierra Leone to the south. It spans 245,852 square kilometers and has a population of between 5.0 million and 7.2 million (current population estimates vary considerably due to an influx of refugees from Liberia and Sierra Leone). The population is projected to grow at 2.8 percent a year between 1992 and 2000 (World Bank 1994j, p. 210). Since its independence from France in 1958, Guinea has endured much economic and political hardship, spurring approximately 2 million Guineans to flee the country by the mid-1980s. Many of the emigr6s (called diaspos) were among Guinea's most educated citizens, and few have returned. Guinea is divided into four regions. Guinee Maritime (coastal Guinea) constitutes 16 percent of the total territory and is home to about 20 percent of the population, 950,000 of whom live in the capital city of Conakry. Guinee Maritime is inhabited primarily by the Soussou ethnic group. Moyenne Guinee or Fouta-Djallon (east of the shore) constitutes 20 percent of Guinean terrain and houses 24 percent of the population, predominantly Peulhs. The savanna of Haute Guinee (northeast) has the lowest population density, with 21 percent of the population on 40 percent of the country's land. This area is inhabited mostly by Malinkes. Guinee Foresticre (south) constitutes roughly 25 percent of the country and is home to 19 percent of the population. Several small ethnic groups, such as the Guerzee, Kissi, Manon, and Toma, inhabit this region (Economist Intelligence Unit 1994b, p. 11; the remaining 16 percent of the population is not recorded due to sporadic migration and lack of accurate census records). Guinea is rich in bauxite and other mineral resources, and climatic and soil conditions are conducive to a wide variety of agricultural activities. In 1992 the primary sector (agriculture, livestock, fishing, forestry, and mining) contributed half of GDP and employed more than two-thirds of the country's 2.9 million working population. Agriculture and mining, each estimated at 21 percent, were the main contributors to GDP in the primary sector. The secondary (manufacturing, energy, and construction) and tertiary (com- merce, transportation, and administration) sectors constituted roughly 11 and 39 percent of GDP, respectively. Construction was the leading contribu- tor in the secondary sector (6 percent); commerce was the leading contribu- 127 128 Country Studies tor in the tertiary sector (24 percent; Economist Intelligence Unit 1994b, p. 15). Despite Guinea's rich natural resource base, its per capita GNP is estimated at just over $500 (World Bank 1994j, p. 162). From 1988 to 1992 GDP grew at an average yearly rate of 3.9 percent. This is a marked improvement over the 1980 to 1987 average annual growth rate of 0.9 percent. In addition to poor economic indicators, Guinea's social indicators rank unfavorably with those of other Sub-Saharan African countries. In 1992 life expectancy for both men and women was forty-four years. Although this is an improvement relative to the 1965 estimate of thirty-five years, it is still well below the average Sub-Saharan African life expectancy of fifty-one years (Economist Intelligence Unit 1994b, p. 11). Approximately one in eight Guineans has access to health care, and only a handful of countries have a higher infant mortality rate than Guinea's, which is estimated at more than 13 percent. In 1990 the average adult illiteracy rate was 76 percent. Illiteracy estimates for women are even higher, at around 87 percent (Economist Intelligence Unit 1994b, p. 162). In conjunction with the govermment's aggressive 1985 economic and financial reform program, the World Bank approved the first structural adjustment credit for Guinea (SAL i) in 1986. SAL i was designed to help remove state controls and interventions, to set the policy and institutional framework for private sector development, and to mitigate the impact of reforms on public works and civil service employees in urban areas. This loan was preceded and complemented by a 1985 technical assistance project called Pagen I. Approval of both these programs was a long and difficult process. The government had requested Bank support as far back as 1981, but several political groups opposed foreign technical assistance and blocked the initiation of structural adjustment programs until 1985. Under SAL I, the government made substantial strides toward liberalizing the economy by removing most price distortions, privatizing or liquidating many public enterprises, and reducing the size of the civil service. The encouraging results from SAL I gave rise to a second structural adjustment loan (SAL 11), which was approved in May of 1988 to support the govern- ment's reform program. SAL II was designed to liberalize the economy further by improving the banking system, trade policies, the legal frame- work, and the government's economic management capability (as a follow- up to Pagen i, Pagen ii was implemented to provide additional technical assistance). The success of this second adjustment loan was limited due to the difficulty in achieving political consensus for reform measures and weak administrative capacity. SAL II remained technically open until mid-year 1994 due to a small remaining balance from cofinancing sources. SAL II's performance was rated unsatisfactory in fiscal 1992 but was raised to satisfactory by the World Bank country team in fiscal 1994. Guinea 129 Main characteristics of the informal sector From 1958 to 1984, the Sekou Toure government pursued a centrally planned model of economic development. This legacy of central planning has given birth to a large and dynamic parallel economy. At least two-thirds of the economically active population (2 million-3 million people) are employed in the informal sector. According to some estimates, the informal sector employs as much as 80 percent of the working population in Guinea (ILO 1990, pp. 7, 66). In 1988 the informal sector's contribution to GDP was estimated at close to 62 percent. Ironically, SAL I and ii may have tempo- rarily decreased the informal sector's contribution to GDP by raising the cost of imported inputs and temporarily decreasing consumer spending power (ILO 1990, pp. 59, 61). Nevertheless, from 1980 to 1992, informal sector employment grew at an annual rate of 2.6 percent (the same as average annual population growth; World Bank 1994j, p. 210; ILO 1990, p. 67). Average annual growth in the informal sector's contribution to GDP was about 3 percent from 1987 to 1993 (ILO 1990, p. 62). The activities of urban informal enterprises fall into three broad categories: production and artisanry (crafts and jewelry, mosquito nets, construction, and traditional clothing), commerce (sale of imported and local goods), and services (small restaurants; appliance, automobile, and home repairs; domestic help; grain hulling; and blacksmithing). Table 10.1 contains a 1988 estimate of formal and informal sector contributions to GDP. The reform program of late 1985 brought extensive public sector layoffs. Registered unemployment in Conakry increased from approximately 5,000 in 1986 to 15,000 in 1990 (Economist Intelligence Unit 1994b, pp. 16-17). Even though a 1986 monetary reform program resulted in a 100 percent salary increase for civil servants, IMF statistics show that Guinean civil servants still earn 40 percent less than their colleagues in Togo and 20 percent less than their colleagues in C6te d'Ivoire. In fact, a 1986 survey in Conakry carried out by the Direction Generale de la Statistique et de l'Informatique showed that higher-level civil servants earned an average net monthly salary of GF 23,675 ($65), a salary six times smaller than a taxi owner, tailor, or mechanic with three employees.' The official workday in Guinea ends at 3 p.m., and it is likely that many civil servants also operate informal sector enterprises to supplement their income.2 A 1987 survey of informal sector enterprises in five major Guinean cities determined that the average microenterprise employs 2.5 persons.3 Informal sector businesses in manufacturing and services average 3.2 employees, while enterprises in commerce average fewer, with 1.7 persons. Informal enterprises in Conakry average slightly higher employment in manufactur- ing (4.0 persons) and services (3.9 persons) than in the other four cities surveyed. Approximately two-thirds of informal sector entrepreneurs use only hand tools, roughly 17 percent use nonelectric machines, and between 17 and 27 percent use electric machines (ILO 1990, pp. 95, 106). An estimated 130 Country Studies Table 10.1 Sectoral contributions to GDP in Guinea, 1988 (percent) Total contribu- Sector Informal sector Formal sector tion for sector Primary sector 49.6 60.1 53.5 Agriculture 35.1 0.3 21.8 Livestock and hunting 6.4 - 3.9 Fishing 0.3 0.1 0.2 Forestry 4.1 - 2.5 Mininga 3.7 59.7 25.1 Secondary sector 8.2 9.8 8.8 Manufacturing and 4.2 2.0 3.4 communications Energy - 2.1 0.8 Construction services 4.0 5.7 4.6 Tertiary sector 42.3 26.4 36.2 Commerce 30.4 6.9 21.4 Transportation 3.1 4.5 3.6 Administration 2.2 12.5 6.2 Other 6.6 2.4 5.0 Indirect tax - 3.7 1.4 GDP 100.0 100.0 100.0 - Not available. a. The ILO includes mining as a secondary sector activity. In accordance with IMF sectoral breakdowns, mining is listed as a primary sector activity for the purposes of this chapter. Source: Government of Guinea, Direction Generale de la Statistique et de l'lnformatique, as noted by ILO 1990 and Economist Intelligence Unit 1994b. 50 percent of entrepreneurs have not completed primary school, and roughly 97 percent obtain informal sector employment via family or friends. Specialization is rare. Most entrepreneurs participate in the entire product cycle, including sales. Main constraints on the informal sector The following constraints limit growth in the informal sector. Small markets and poor infrastructure. Local markets are small, and demand is insufficient to absorb the informal sector's increasing supply of goods and services. Inefficient transportation and communication networks constrain access to new markets. Basic inputs such as electricity, water, and fuel are expensive, and supply is unreliable. Guinea 131 * Limited access to resources. There exist both a human and a natural resource constraint to growth in the informal sector. Skilled employees are difficult to find. Local and imported material resources are scarce and costly. * Limited interaction with theformalfinancial sector. As a result of the 1985 liquidation of the country's six state banks, Guineans tend to distrust the formal financial sector. Most households hoard savings, which limits the mobilization of savings by the formal banking sector. It is estimated that twice as much money circulates in the informal sector as in the formal sector.4 Microentrepreneurs have no access to formal financial services and, thus, no access to medium- or long-term credit. Policies and programs for microenterprise development Under the Sekou Toure regime, the government created and operated a multitude of mutualist-type groups called cooperatives artisanales to support microenterprises. These organizations never prospered, primarily due to lack of private sector ownership and autonomy. Little information is available on private production groups that may have formed since economic liberalization. Informal financial markets in Guinea are fairly well developed and involve transactions in Guinean francs, French francs, CFA francs, and U.S. dollars. Lack of collateral is mitigated by the use of informal social mechanisms to encourage repayment. Informal financial markets reportedly provide short-term loans, but little in the way of medium- and long-term credit. Projet Rural Integre pour le Developpement des Enterprises (PRIDE) develops microenterprises in rural areas of Guinea by providing training and microcredit. This program began in 1991 and is supported by USAID ($6 million) and implemented by the U.S. NGO Volunteers in Technical Assistance. PRIDE focuses on solidarity-based lending with ongoing training for borrowers. The credit portion of the program is based on the Grameen Bank in Bangladesh. What distinguishes PRIDE'S microenterprise program is its emphasis on training. The training is designed for a clientele that is 90 percent illiterate and is offered free of charge. Credit is available to members of five-person solidarity groups who have taken a basic credit training program and have had their loan application approved by a neighborhood council of elders. Once a loan is granted, the borrower must participate in one business training session a month. The lending program started in late 1992, and by November 1994 PRIDE had granted more than 10,000 loans (70 percent of which went to women) totaling GF 2.2 billion ($2.2 million) with 3,505 loans outstanding and repayment near 100 percent. Six branches were open by end-1994, and PRIDE plans to add three more sites in early 1995.5 132 Country Studies Projet de Credit Agricole et Rural is a rural banking program based on the Grameen Bank model. This project is funded by the Caisse Francaise de D6veloppement, the European Development Fund, the African Develop- ment Bank, and USAID and is implemented by the Institut de Recherche et d'Application des Methodologies de Developpement, a French NGO. The maximum loan amount is GF 300,000 ($300) per person, and the average loan size is GF 71,000 ($71). A savings component established in 1991 pays 15 percent interest a year. Forty-one caisses have been created, and the majority of borrowers come from the trading sector. The program began in Moyenne Guinee, and during the first year, repayment was 100 percent. As of 1994, the project had expanded to three other regions, and repayment was around 98 percent. The number of members grew from 250 in 1989 to 31,000 in 1994. As of November 1994, the project had a lending portfolio of GF 2.0 billion ($2.0 million) and had accumulated GF 1.3 billion ($1.3 million) in savings. Credit Mutuel de Guinee was designed in 1988 to mobilize local savings and to extend loans for agriculture and livestock, housing, commerce, artisanry, and consumption. It is managed by the Centre International du Credit Mutuel and financed by the Caisse Francaise de Developpement, the Fonds d'Aide et de Coop6ration, the Fonds International de Developpement Agricole, the World Bank, the Office pour la Promotion de l'Investissement Prive (a local organization, formerly the Centre de Creation et de Developpement des Entreprises), and the government, and it is linked to the Credit Mutuel of France. The program operates both in urban and rural areas of Maritime Guinea and around the towns of Fria, Gu6ckedou, Kankan, Kinia, Macenta, Mamou, N'Zerekore, and Suiguiri. Once estab- lished, a branch collects savings for six to twelve months before lending. Interest on demand deposits is 13 percent a year. There are no term deposits. Once a savings base is established, borrowers can obtain up to two times their savings at a 30 percent annual interest rate. Loans average GF 1.0 million ($1,000), and applicants come mainly from urban areas. No one can receive more than 10 percent of the available money supply at once. The project now has roughly seventy branches and 45,000 members, 30 percent of whom are women. Membership grew at a rate of about 85 percent a year from 1990 to 1993. At the end of 1994, the program had roughly GF 5.0 billion ($5 million) in savings, with GF 2.5 billion ($2.5 million) extended in credit. Maison de l'Entrepreneur was established in 1988 by the Centre International de Developpement et Recherche with French francs (FF) 1.5 million a year in funding from Fonds d'Aide et de Cooperation and other NGOS. This program, now in its second phase (1993-95), is under the Secretary of State for Decentralization, but it functions independently. Maison de l'Entrepreneur provides existing and would-be entrepreneurs with training, advisory services, business contacts, and assistance in Guinea 133 obtaining credit. The group is for-profit, and it operates in Macetan and Nzerekore. A technical assistance program offers basic training for a fee of GF 20,000 ($20) or a complete cycle of business training for GF 40,000 ($40).6 The Nzerekore office also manages a small United Nations High Commis- sion for Refugees credit fund with a solidarity group approach that lends to groups of twenty to thirty persons. In 1991 the two offices combined trained 100 individuals, eight village groups, and thirteen informal groups with two to three members each. As of 1993, Maison de l'Entrepreneur had eight contracts with individuals for follow-up training (GF 10,000 or $10 a month), thirteen contracts with rural groups (also GF 10,000 a month), and two government village contracts (GF 40,000 or $40 a year). Under the credit program, the Nzerekore office made 112 loans for a total of GF 20 million ($20,000). According to a 1993 GEMINI evaluation report, all loans were repaid. In 1991, Maison de l'Entrepreneur covered 39 percent of its operating costs in Macetan and 57 percent in Nzerekore (GEMINI 1993, p. B23). Integrated Assistance to Small and Medium Enterprises in Guinea provides business training and loans to entrepreneurs starting or expanding small and medium enterprises. Training is free of charge, and any entrepre- neur is eligible to participate. The project is funded by UNDP and executed by the United Nations Industrial Development Office (UNIDO), the Centre de Cr6ation et Developpement de l'Entreprise, and the Office de Developpement Industriel du Maroc. Phase one of this project ran from 1987 to 1992, and total funding was $900,000. Phase two is ongoing (1993-95), with an estimated cost of $2.23 million. The unit presents an entrepreneur's dossier to a comife de gestion composed of staff from UNDP, UNIDO, the government, and Banque Intemationale pour le Commerce et l'Industrie de Guinee for loan approval. Loans range from GF 5 million-GF 12 million ($5,000 to $12,000) at a nominal interest rate of 12-15 percent a year.7 A main weakness of the program is that all services are free, negating the possibility of cost recovery and jeopardizing the sustainability of the project. Loans were made to forty enterprises during the first phase of the project. The majority were disbursed in late 1990 and early 1991. The average loan amount was GF 8.0 million ($8,000). A February 1992 review of portfolios showed that 92 percent of loans were in the process of being repaid, although none had yet been fully reimbursed (GEMINI 1993, pp. B24-B25; UNDP 1991, pp. 3-7). The Africa Small Business Assistance Program Revolving Credit Fund in Guinea was designed to create additional employment in ten target cities outside Conakry by creating rotating credit funds for small enterprises. The program began in 1988 with funding from a philanthropic U.S. businessman and is implemented by Peace Corps volunteers, who serve as managers and technical assistants for local loan review committees. 134 Country Studies The project has closed all but one office, Kamsar, and has put the recoverable funds into self-help projects. Disbursement of loans was stopped in most cities before the branch was closed so that staff could focus on recovering bad loans. Between May 1989 and September 1990, the project made 158 loans averaging GF 431,000 ($431), and total assets increased from GF 34.8 million to GF 45.6 million ($34,800 to $45,600). Repayments hover around 80 percent, but the program apparently has not been evaluated since 1990 (GEMINI 1993, p. B27). The Association pour la Promotion Economique de Kindia (APEK) was created in 1989 and is supported by Loire Atlantique, the European Economic Community., Fonds d'Aide et de Cooperation, and the chamber of commerce in Kindia's sister city, Nantes. It provides multiple services to local enterprises in Kindia and surrounding areas. Originally targeting female entrepreneurs, APEK now assists anyone in the Kindia area interested in its services. The association has six units: * APEK Agriculture provides technical and financial assistance to small agriculture-related enterprises. * APEK Recherche focuses on product development by exploring the possibilities for transforming and subsequently marketing existing agricultural products (transformation includes the drying and processing of vegetables or grains). * APEK Service manages the APEK incubator center and provides services such as accounting, building and equipment rental, and quality control to clients, who are charged the full costs of these services. * SOKIDIS markets APEK products in Kindia and Conakry through coopera- tive warehouses. * APEK Amenagement is a contractor that oversees a team of local construc- tion workers assigned to APEK and other construction projects. * APEK Finances manages a guarantee fund from the Fonds Europeen de Developpement. APEK had an operating budget of GF 60 million ($60,000) in 1991 and six branches. As of 1993, seven enterprises were benefiting from its incubator services, and twenty groups were being supported by financing from APEK Agriculture. Standard loans from APEK Agriculture range from GF 50,000-GF 100,000 ($50 to $100). APEK Agriculture classifies clients with large loan requests as "modem entrepreneurs" and has disbursed five loans (minimum GF 500,000, or $500) totaling GF 8.0 million ($8,000) to them for commercial plantations. In 1993 twenty requests for modem entrepreneur loans were in the pipeline (GEMINI 1993, pp. B23-B29). The Opportunities Industrialization Center provides vocational training to youth and retired civil servants. The program began in 1986 and is Guinea 135 supported by USAID (GF 42.8 million, or $42,800), UNDP ($200,000), the Canadian International Development Agency (GF 12.3 million, or $12,300), and the Canadian Center for International Studies and Cooperation (two administrators). In 1991 this project trained thirty female entrepreneurs and managers of small firms in business management, marketing, and bookkeeping. It also trained twenty-four people in masonry, twenty-one in carpentry, and sixteen in secretarial services (GEMINI 1993, p. B25). The author of this chapter was unable to locate a technical assessment of this project. On October 18, 1994, the World Bank approved a financial sector adjustment credit in the amount of $23 million. This credit has a component designed to broaden the reach of the financial sector by developing and bridging informal and formal financial systems. To accomplish this task, the formal financial system will begin working closely with Credit Mutuel and with Credit Rural. Both financial intermediaries plan to open new branch offices. The main branch of each organization will act as a central bank and regulator to the local offices. A unit within the formal Banque Centrale de la Republique de Guinee will be established to supervise Credit Mutuel, Credit Rural, and other mutualist savings and loan institutions. Measures will be taken to channel some of the foreign exchange in the informal sector to the formal banking system, and insurance intermediaries will be developed to service the formal and informal sectors (World Bank 1994e). Because Credit Mutuel and Credit Rural primarily serve rural areas, a new financial intermediary with four branch offices will be established in Conakry (Caisses d'Epargne et de Credit de Conakry). The government of Guinea has agreed to provide a $2 million grant for start-up costs. Institut de Recherche et d'Application des Methodologies de Developpement, Centre International du Cr6dit Mutuel, and Societe de D6veloppement International Desjardins of Canada have already expressed an interest in supporting Caisses d'Epargne et de Cr6dit de Conakry (World Bank 1994e). In 1989 the World Bank launched the Socio-Economic Development Support Project ($9 million) including a pilot program ($6 million) designed to improve living conditions among the poor and to generate employment for recent school graduates and laid-off civil servants. In the beginning, the pilot program focused only on nonprofit projects such as schools and health programs. Credit for the private sector has been available since November 1992, and as of March 1995, eight of nine approved projects were under way. Issues and recommendations A field visit in November 1994 confirrned that few, if any, small and microenterprises in Guinea are served by banks. Donor-funded NGOS have moved to fill the gap by offering credit and training programs for micro and 136 Country Studies small enterprises in rural and semirural areas. These programs range from very small credit-only programs to large savings and credit networks. Considered on a purely financial basis, these programs are a far cry from achieving full sustainability. However, they fulfill a critical function from an economic and social perspective in that they provide financial services to a significant portion of the Guinean working population that is not reached by formal financial institutions; they create income-earning opportunities for a large number of low-income individuals; and some of them offer significant training services to entrepreneurs regarding credit, management, and marketing. Some programs, such as the USAID-funded PRIDE and the French-financed Credit Mutuel and Credit Rural, have established a strong national presence and recognition. To the extent possible, a World Bank strategy for microenterprise assistance in Guinea should seek to work with these and other institutions that have a track record of reaching a large number of the poor, have high repayment rates, and are making an effort to attain financial and operational sustainability. Bank assistance to microenterprises could be envisaged at three levels: * Policy. The Bank could play a useful role by assisting the Central Bank and the Ministry of Finance in providing a legal and regulatory environ- ment that facilitates the growth of microfinance institutions. Such a legal framework would address specific issues such as interest rate caps and liquidity ratios. - Education and dissemination. Another level of assistance would be to galvanize the government, donors, and NGOS to coordinate their approach to the informal sector and to microenterprise development. This may best take the form of a seminar that would highlight the importance of the small-scale informal sector in the Guinean economy, examine the policy, regulatory, and institutional constraints facing informal enterprises, and disseminate best-practice cases of microfinance institutions around the world. a Project. To the extent that some microfinance institutions in Guinea are already reaching large numbers of the poor, the Bank could help them to increase their outreach and improve their sustainability by linking them up with the formal financial sector, by including microenterprise credit components in financial sector projects, and by providing technical assistance in specific areas of the institution's needs, such as managing growth and establishing sound management information and accounting systems. Guinea 137 Notes This chapter was written in June 1994 and was updated based on a field visit in December 1994. 1. ILO 1990, p. 89. The exchange rate in 1986 was 365 Guinean francs to I U.S. dollar. Today, it is roughly GF 1,000 to $1.00. 2. A government agency, Le Bureau d'Aide r la Reconversion des Agents de la Fonction Publique, was created to facilitate the transfer of workers from the public to the private sector. However, 85 percent of the loans advanced are nonperforming, and few resources remain to help new applicants. Economist Intelligence Unit 1994b, p. 16. 3. The cities surveyed were Conakry, Kankan, Kindia, Labe, and Mamou. The survey was conducted by the Programme des Emplois et des Competences Techniques pour l'Afrique and Direction Generale de la Statistique et de l'Informatique. 4. Interview with Sherri Archondo, financial analyst, Western Africa Department, Industry and Energy Division, World Bank, December 1994. 5. Interview with Paul Rippey, Volunteers in Technical Assistance country representative, Guinea Rural Enterprise Development Project, September 1994. 6. If the government pays, the fee charged for a group training event is GF 120,000 ($120). If participants pay themselves, the fee charged is GF 60,000 ($60). 7. Interest rates are rising slowly. Interview with Paul Rippey, Volunteers in Technical Assistance country representative, Guinea Rural Enterprise Development Project, September 1994. 11 Guinea-Bissau Simel Esim Guinea-Bissau is a small country on the West African coast with approxi- mately 1 million inhabitants. It is bordered to the north by Senegal, to the east and south by Guinea, and to the west by the Atlantic Ocean. The Republic of Guinea-Bissau was established in 1974 after a long war of national liberation, which was launched with Portugal in 1963.' Victory by the Partido Africano da Independencia de Guinea e Cabo Verde led to the adoption of an economic model of central planning, based on near-complete state control of economic activities with the exception of the traditional rural sector (Swedish International Development Agency 1989, p. 14). The large rural sector relies primarily on subsistence activities such as farming, fishing, and forestry, which account for about 90 percent of employment and an estimated 44 percent of GNP. Rice is the main food crop and is cultivated mostly in reclaimed coastal swamps, but peanuts grown in the drier interior are Guinea-Bissau's principal export. Marketed agricultural output is largely confined to export crops, primarily cashews, groundnuts, and palm kernels. Cashew nuts account for about 80 percent of exports. Exports are sold mainly to Portugal. With a per capita income of $220, Guinea-Bissau is one of the poorest countries in the world. Basic social indicators including life expectancy (thirty-nine years), infant mortality (140 per 1,000 live births), adult illiteracy (64 percent), and secondary school enrollments (4 percent of school-age children) also place Guinea-Bissau among the poorest countries. In April 1992 the director of an AIDS prevention program in Bissau announced that 10 percent of the population was HIv-positive. Malaria and cholera pose serious health problems as well. The poor health situation is compounded by the fact that only 5 percent of the population has access to safe drinking water (Economist Intelligence Unit 1994a). Economic development has been hampered in Guinea-Bissau by a number of crucial factors, including dependence on agriculture and vulnerability to drought, inadequate transportation infrastructure, an undeveloped human resource base, and a decade-long experiment with central planning. Large-scale emigration further constrains economic performance. The frustrations of daily life that began soon after independ- ence (including the inability of many to find suitable work) motivated large numbers of professionals to leave the country for France, Portugal, and 139 140 Country Studies Senegal.2 The period of political repression under the regime of Luiz Cabral, the tensions generated by the 1980 coup d'etat, and ongoing political uncertainties have all contributed to a slow but steady exodus of well- educated Guineans.3 This migration has been particularly damaging to the government's development plans, which depend on the presence of skilled professionals. In 1983 the government broke with the policies of the decade following independence and adopted an economic recovery program supported by the IDA and IMF. The first phase of structural adjustment began in 1987 and lasted through 1988. This phase focused on implementing broad macroeco- nomic policies, restructuring the public investment program, decontrolling agricultural prices, simplifying tariff and customs procedures, and liberalizing trade policies. A second phase began in 1989 and progressed more slowly than expected. Since late 1992, the authorities have renewed their commitment to reform. In 1993 the government implemented a demanding IMF-monitored program, and progress continued in 1994.4 This has led to a pronounced deceleration in inflation, which dropped from about 88 percent annually at the end of 1992 to approximately 12 percent a year in mid-1994. In addition, exchange rates were unified in 1994. Reform has been slower than originally envisaged. Nevertheless, the structural adjustment process has transformed the economy of Guinea- Bissau. Real GDP grew steadily at annual rates of 2-6 percent from 1986 to 1993. Since 1987, when the adjustment process began, Guinea-Bissau has evolved from a rigidly controlled command economy into one guided mainly by market forces. Prices (with the exception of petroleum) and the exchange rate have been liberalized, and the swollen public enterprise sector has been partially privatized and restructured. There are no firm statistics, but it is estimated that private sector investment has grown at about 20 percent a year from a very small base. Main characteristics of the informal sector The informal sector in Guinea-Bissau emerged during the colonial period. At that time, about 70 percent of all economic transactions were based on barter, enabling rural Guineans to engage in economic activities beyond the watchful eyes of Portuguese officials. Guineans relied on informal trade in two ways during the colonial era: as consumers acquiring goods not available or not affordable in the formal sector and as producers obtaining higher prices for crops than those offered by the Portuguese-controlled stores. Informal activities dominated the rice trade. Deliveries of rice to state-controlled outlets never totaled more than 7,305 tons in colonial times, even though rice production is estimated to have ranged between 43,800 tons in 1946 and 182,251 in 1963. Private commerce, transport, and other services have been the fastest growing sectors since the beginning of the adjustment process. Officially, Guinea-Bissau 141 5,000 registered private firms are now operating in Guinea-Bissau. Most registered firms are microenterprises, but many additional microenterprises reportedly operate without registration. Economic reform has brought a considerable increase in retail and service activity. Informal sector products, including cloth, processed food, wooden tools, and furniture, are appearing in local markets in increasing numbers. According to a survey of 605 enterprises carried out in 1990 by the Instituto Nacional de Estudos e Pesquisa, most informal sector activities are concentrated in Bissau. Among the enterprises surveyed, 30 percent were in industry, 25 percent in commerce, 43 percent in services, and 2 percent in agriculture. An overwhelming proportion were owned by individuals (92 percent in services and 71 percent in agriculture). The others were owned by cooperatives and community or family groups. Only 8 percent of surveyed enterprises kept records of their operations, and about 40 percent were registered. Almost all surveyed firms identified infrastructure as a major problem. Less than 1 percent had access to piped water, only 1.5 percent had telephone connections, and less than one-third had access to electricity. More than 65 percent started with an investment of $750 or less. Approxi- mately 75 percent of initial capital came from personal savings, and less than 2 percent was obtained from the banking system. Three-quarters of surveyed enterprises had financed their working capital entirely from business profits, with less than 3 percent ever having had access to a bank loan. Similarly, three-quarters of those surveyed did not have bank accounts, and more than half of business operations were interrupted due to lack of working capital. A glimpse into the informal sector of Guinea-Bissau reveals an intricate network of barter. Iron workers barter their tools, and tailors exchange finished clothing for rice or millet. Women barter handmade cooking pots and salt for rice, and farmers barter their rice surpluses for sugar or tobacco. Beafada women commonly barter fish for paddy rice, usually at equal volumes.' Cattle raisers barter their animals, with young bulls valued at 360 kilograms of paddy rice. The barter system occupies a much more central role in Guinea-Bissau than in most countries, and thousands of barter exchanges occur daily. Cash trading also takes place with djila traders.6 The djila traders bring finished goods from Senegal to the north and from Guinea-Conakry to the south and sell them for cash, commercial crops, and food grains. Djila traders have flourished due to shortages that accompany state-controlled markets. Since retail trade was privatized in the late 1980s, legal private stores have been able to offer some imported items, but illegal imports still ensure the prosperity of the djila traders. Since independence, peasant women have had legal access to land, but remaining gender inequities have denied most women access to capital and 142 Country Studies training. In semiurban areas where land has been increasingly sold or leased, some women have lost access to their plots of arable land. Most rural women have primary responsibility for growing and harvesting food crops, fabricating artisanal products, transporting produce to markets, and selling their output. In recent years, many female cultivators in Guinea-Bissau have had to take on additional tasks when their husbands have migrated to earn cash incomes. Rural Manjaco women fabricate oil filters, fishing nets, and baskets and make lemon vinegar, cashew wine, and ceramics.7 They dry palm fibers and press them into balls called candjirba for use as food supplements. These women travel as far as 60 kilometers on foot to find water from which to extract salt. Gender roles and heavy family responsibilities are major impediments to the educational advancement of women in Guinea-Bissau. Only 38 percent of elementary school children and 36 percent of secondary school students are girls, reflecting the reluctance of families to release female children to pursue a formal education. Main constraints on the informal sector The primary constraint faced by microentrepreneurs in Guinea-Bissau is lack of access to financial services. Low levels of literacy and numeracy, and high collateral requirements constrain access to bank loans. Nonbank, microfinance institutions like those that have developed in neighboring Senegal and Guinea-Conakry are largely absent in Guinea-Bissau. Therefore, virtually all informal sector operators rely on personal savings and loans from family to finance the start-up and growth of their enterprises (see box 11.1) Box 11.1 Informal financial services in Guinea-Bissau Most informal sector financial services are provided by informal savings and credit institutions. These take two general forms: bilateral relationships between two unequal partners and informal guilds involving two or more relatively equal partners. Bilateral relationships involve an informal written or oral contract between two unequal parties in which one party agrees to transfer capital to another for a fixed time period. * Kin-based credit. The majority of lending takes place among relatives, neighbors, friends, and acquaintances. Terms vary greatly depending on whether the loan assumes the guise of benefaction, reciprocity, or moneylending. These loans tend to involve no or low interest, no guarantees or collateral, and indefinite or flexible repayment schedules. Among some Islamic ethnic groups, loans are made without interest. However, interest can be disguised by calculating high barter rates for in-kind loans. Guinea-Bissau 143 Box 11.1 (continued) * Client-oriented credit. Most informal credit markets involve some advance of cash, goods, and services from a moneylender to a borrower. Contracts are usually oral. Loans of consumer goods have terms of two to six weeks, while perishables are lent for one day to one week. Three types of collateral are used to secure loans. The most common is a guarantor who vouches for the borrower. A second form is a pledge or penhor in which jewelry, or some other article of value greater than or equal to the goods lent, is given to the creditor as security against the loan. A third form was once common among entrep6t actors: the possession of a license or alvara. Informal guilds are of three general types: blocs, buyer and vendor associations, and savings associations. * Blocs. A cartel is organized into several distinct blocs that promote the interests of members and are guided by sanctionable codes of behavior. Blocs procure consumer goods at wholesale rates from importers with the objective of comering the market for a particular commodity. Each member contributes a fixed sum to a common pool used by one member to purchase all or most of the available supply of one selected good. * Buyer and vendor associations. Other associations based on kinship or friendship organize into small trusts composed of two to five small missions. An envoy purchases cloth and distributes it to each member firm according to the funds contributed. Firms then combine merchandise at one vending table, but they know which cloth is whose. * Savings associations. Savings associations or abotas are a common feature of small-scale marketing systems. In an abota, several marketing agents, usually of similar social or professional status or occupying adjacent stalls in a marketplace, contribute a fixed sum of daily, weekly, or monthly earnings to a common fund. This fund is then granted in its entirety to each member of the association in succession. Abotas usually group together five to thirty people of similar age, gender, ethnicity, and occupational status. Many abotas form only seasonally, in keeping with different crop and labor cycles. Some also function as emergency funds. A third variety is oriented specifically toward the purchase of expensive, durable equipment. * Religious affiliations. Some communities have set up emergency funds through religious affiliations. * Professional security associations. Employees of a few high-risk informal professions have developed professional security associations. The most notable examples are locataire associations in which roughly fifteen members contribute to a common fund kept by one of the dues collectors. Source: Crowley 1993. Poor roads, infrequent and costly transport services, unreliable ferry crossings, fuel shortages, inadequate electricity, water and telecomrmunica- tion services, and lack of or inappropriate storage, shelter, and sanitation facilities confine most informal sector activities to small, local markets. 144 Country Studies Transporting people and goods in and out of many villages is almost impossible, especially during the lengthy rainy season. Procedures for registering and licensing a business, registering land, and authorizing imports and exports are very complicated and time-consuming. Official fees apparently are not excessive, but the transaction costs of intricate and prolonged administrative procedures encourage businesses to remain outside the formal sector. The legal and regulatory structure excludes most informal sector participants and employees from a legal framework within which they could mobilize collateral for acquiring credit, protect innovations and inventions, and gain access to or expand business premises. Existing laws and regulations are poorly known, and the disparities between laws and regulations and actual trade and investment practices are immense. For example, there are few legal supports for contracts involving parties from different ethnic, religious, and national backgrounds, and the judiciary provides little guarantee of enforcement to litigants in civil disputes. The low literacy rate also contributes to the population's ignorance of how the formal business environment operates. Although in many villages male migration requires women to become the head of household for months at a time, government policy has paid little attention to the specific needs of rural Guinean women. "Integrated" rural development projects to improve and diversify agricultural production demonstrate little awareness that women peasants need independent access to capital, tools, and training. Women have rarely been targeted for specific development programs, and few have received financial or technical assistance. Policies and programs for microenterprise development Foreign donors and forty-two national and sixty-four international NGOS operate in Guinea-Bissau. The following are some of the most prominent informal sector programs recently completed or under implementation (see table 11.1, which contains information about programs visited in February 1995). International initiatives The World Bank Social and Infrastructure Relief Project, a $5.0 million loan to the government, seeks to increase employment among low-income people by underwriting activities with high economic and social rates of return. The project focuses on rehabilitation of infrastructure, provision of urban housing, improvement of health care services, job placement, development of data bases, and strengthening of government support agencies. Table 11.1 Microcredit programs in Guinea-Bissau, 1995 Number Annual of loans Life of the Program interest rate Temn Loan size a year Repayment rate lending project Special conditions Existing Tiniguena 0% One year $120 35 100% 1994-present Next year hopes to imple- ment a charge of 5% inter- est a year Solidami 10% Two years, $0-$10,000 6-7 In progress 1992-December The governing body of all flexible 1995 NGOS; lends to microenterprises directly as we-ll as via other NGOS; has a flexible repayment schedule; visits each pro- ject once every six months Action pour le 10%, Six months to $4,000-$30,000 21 98% 1993-present Provides subsidized Developpement negotiable two years, to associations money and materials; negotiable of fifteen or supported by Solidami, more people among others Service Quaker} 0% Six months - 7 99% Scheduled to begin in 1995 Fundei 12%, long One to ten $100,000 maxi- - - Beginning in Has no lower limit on loan term; 24%, years mum May 1995 size; will require no collat- medium term; eral; has no training com- 36%, short ponent term Table 11.1 (continued) Number Annual of loans Life of the Program interest rate Term Loan size a year Repayment rate lending project Special conditions Africareb Inflation plus - $2,000 80 90-95% Beginning in sum- Offers solidarity lend- 1% maximum (estimate) mer 1995 ing; provides training Reseau Africain de 10% (which - Very small - - Working toward Takes a bottom-up ap- Poupance et Credit will increase) self-sustainability proach; needs training and capital Caisse Francaise de Subsidized - $200 average - - Three-year life Lends for agricultural Developpement beginning in 1995 activities Banco Totta Rural Subsidized Medium term - - - - Much confusion sur- Credit to long term rounds the project; un- certain if money or ma- terials will be lent Recently completed UNDP and Banco Inflation rate One to five $7,000-$8,000 15 93% June 1993 to In the repayment phase, Totta years to groups of December 1995 with only one late pay- five persons or ment to date; UNDP indi- more cates some trouble in monitoring - Not available. a. The mission was unable to meet with representatives of Service Quaker. Information on Service Quaker was provided by Africare. b. The statistics for Africare are taken from interviews and projections made in feasibility study done by Nicolas, Lewis, and Associates 1994. Source: World Bank mission in Guinea-Bissau, February 8-10, 1995. Guinea-Bissau 147 The loan has many components that focus on poverty alleviation and job creation. One component, the Reorientation Program ($100,000), provides unemployed people with job training and support for microenterprise start- up. This component contains two programs for the Reorientation Unit in the Ministry of Public Function and Labor. The first establishes a data base of unemployed people (with special emphasis on women and youth) to be matched with a roster of available jobs. The second involves a training program for former government employees. The training provided by the Technical Institute for Vocational Training provides basic accounting and management information. The World Bank also finances a Social Sector Project for SDR 6,400,000, of which about 10 percent is allocated for a social action fund, which finances community-based microprojects. Local associations (composed mainly of women) are eligible for a grant of up to $15,000 for activities that generate income or that improve health or nutrition. These microprojects include activities such as horticulture, rice production during the dry season, and housing construction for women. Since late 1993 the Social Action Fund has financed eleven microprojects for a total of $125,000, benefiting at least 2,500 small and microentrepreneurs. USAID's Trade and Investment Promotion Support Project is a six-year effort financed by a grant from the Development Fund for Africa and implemented by Labat-Anderson Incorporated, a consulting firm (USAID 1992). This project focuses on removing legal and regulatory constraints to increased trade and investment in Guinea-Bissau's main activities, including the following: * The production, processing, and marketing of rice, cashews, fruits and vegetables, and forest and fisheries products. * The commercial and light manufacturing assembly sectors of informal and semiformal enterprises. * The service sector. The project offers free training to small businesses in basic business skills and on how to access formal credit; provides training and advisory services to the legislative, judicial, and executive branches of the government; and increases public awareness of the judicial system in a democracy via weekly television and radio programs. Africare's Rural Initiatives Project in Tombali is also financed by USAID. Africare began work in the Tombali region in 1991, offering training courses in basic business skills, group dynamics, and group problem-solving. The group dynamics portion of the training program was implemented in anticipation of a pilot lending program to associations or grassroots groups to begin in 1993. In this pilot, two lines of credit were extended, one for $22,000 to an association of eighteen fishermen and a second for roughly 148 Country Studies $3,000 to an association of twenty women who manufacture soap. Both were one-year loans with annual interest rates of 20 percent. Both loans are delinquent and are not anticipated to be repaid. Interviews with project coordinators indicate that failure was due to the excessive size of the loans, their disbursement in one lump sum rather than in tranches, poor follow-up, and lack of training about loan programs and the importance of repayment. With these lessons in mind, Africare will launch a new microenterprise lending program in Tombali in June 1995. A network of roughly eighty associations (approximately 1,000 people) has received the free, preparatory training for this solidarity lending program. Before receiving a loan, Africare is requiring associations to become formal entities and register with the government. Project forecasts made by a U.S. consulting firm estimate that the project will provide loans to 320 associations over the next four years. The initial capital cost is an estimated $598,400. The target borrower is "engaged in informal sector activities rooted in agricultural production and processing" (Nicolas, Lewis, and Associates 1994) Average loan size will be $2,000 at an interest rate of inflation plus 1 percent, and the term will be three to twenty-four months (when this chapter was written, inflation was approximately 15 percent). Africare will cover all administrative costs associated with the program. In the fourth year of the program, it is envisaged that an additional $50,000 will be placed in a bank branch in Tombali to serve as a guarantee fund allowing borrowers with a good credit history to access commercial bank loans. The Fundac,o para o Desenvolvimento Empresarial Industrial is a pilot medium- to long-term lending program initiated by Swede-Corps, UNDP, and the Ministry of Industry. It is due to be operational in May 1995, but it has seen several delays; a start date later than May would not surprise coordinators. Swede-Corps has donated start-up capital, training for employees, and investments of Swedish kronas (SKr) 18,000,000 ($2.4 million) in international markets (the investment covers administrative costs). The government of Guinea-Bissau has donated office space in Bissau. Interest rates on loans are expected to be 12, 24, and 36 percent a year for long-, medium-, and short-term loans, respectively. Representatives from the Ministry of Industry indicate that some degree of flexibility in repay- ment periods and interest rates may be granted. Program coordinators want the program to have a private sector image and, therefore, have established it as a legal foundation. Although the foundation will lend to microenterprises, and no lower limit will be placed on loans, the program will be tailored more toward small- to medium-scale financing. The loan ceiling is $100,000, and collateral will be required. Coordinators hope that this foundation will evolve into a development bank. UNDP and Banco Totta credit pilot is funded by UNDP and administered by Banco Totta e Acores, a Portuguese commercial bank with a branch in Bissau. Follow-up with borrowers has been provided by a celluile of the Guinea-Bissau 149 Ministry of Industry created specifically for this project. From July 1993 to December 1994, fifteen loans were granted to associations of five persons or more. Average loan size was approximately $7,000, the interest rate is linked to inflation, and loan terms range from one to five years. Borrowers were not offered training in connection with this program, and collateral in the form of one's possessions or a guarantee by someone else is required. Although still in the repayment phase, UNDP and Banco Totta report only one late payment and a timely repayment rate of roughly 93 percent. Domestic initiatives Reseau Africain de Poupance et Credit is headed by five philanthropic, well- educated Guineans who are interested in community development in the Gabu region. It is not an NGO or a formal institution, although members are considering formalizing their efforts in 1995. They have assisted four women's associations in mobilizing revenues from flour and honey processing. As a result, the four associations have deposited Guinea- Bissauan pesos (PG) 120 million ($9,500) in the Gabu branch of the Central Bank. With the women's revenues, Reseau Africain assists the associations in developing a lending program for residents of their respective villages. Interest rates average 10 percent a year, and repayment is in the form of money or products. Reseau Africain is working to help these women's associations evolve into four self-sustaining credit and savings banks. This involves offering training in literacy, numeracy, and business skills; having local coordinators from other grassroots savings and credit programs like Grameen Bank and BancoSol visit the coordinators; having association members visit neighboring financial intermediaries like PRIDE in Guinea- Conakry and Catholic Relief Services in Senegal; and making an initial capital investment in huts, bookkeeping supplies, and so forth. Roughly $100,000 in financing is needed over the next five years. SOLIDAMI (Solidariedade e Amizade) is an umbrella NGO for the country's 105 national and international NGOS. It is housed in the Ministry of Cooperation and receives funding from several donors, including Oxfam England, Interpares Canada, the European Community, and Centre Canadien d'Etude et de Cooperation Intemationale. SOLIDAMI provides training and financial support to NGOS, associations, and groupements de bases. In 1991 it launched Fundo de Pequenas Iniciativas to provide credit for the start-up and growth of microenterprises. As of February 1995, twenty-four loans had been provided at an interest rate of 10 percent for one-year loans and 15 to 20 percent for two-year loans. Representatives of the program indicate that these interest rates are somewhat flexible, as is the repayment schedule. The fund is due to end in December 1995. Tiniguena is a national NGO that provides small lines of credit for agricultural activities. It began its lending program in 1994, and its major contributors are the Belgian, Dutch, and Swiss governments. Tiniguena has 150 Countra Studies provided fifteen loans of PG 2.0 million ($120) each in 1994 and expects to provide thirty-five in 1995. Loans are disbursed in tranches over a six-month period, and project coordinators visit borrowers once a month to follow up. The interest rate is 0 percent, but staff is considering establishing a 5 percent interest rate in 1995. Of the fifteen loans issued in 1994, all were repaid in full in a timely manner. Acqao para o Desenvolvimento is a national NGO with three offices: one in the south, one in the north, and one in the east. Among its supporters are CIDAC (Information and Documentation Center, Amilcar Cabral) and CECI (Canadian Center for Studies and International Cooperation). Action pour le Developpement facilitates the formation of professional associations, provides some vocational training, and lends or grants funds and materials to newly formed associations. Loans and grants range from $4,000 to $30,000. The loan term is generally two years, but repayment schedules and interest rates are negotiable throughout the life of the loan. Because of the flexibility built into the program, representatives were unable to cite firm repayment rates. Issues and recommendations Unlike its neighboring countries, Guinea-Bissau does not have well- developed microfinance institutions. Several NGOS and donors have piloted lending programs for microenterprises, but none appears to be using best- practice principles and technologies or to be working toward financial sustainability. Education in the realm of microfinance is sorely needed at local, national, and international levels for organizations lending to the informal sector. Existing programs should be remodeled applying microfinance best practices. Investments in transport services and intermediate storage facilities in key market towns would enhance the ability of producers to market their products. These enhancements would also facilitate regional development. Greater use of appropriate technology could be emphasized for the informal sector, perhaps by providing technical support for the introduction of improved technology for processing and manufacturing. Simple methods, such as sun drying, green harvesting, and canning or bottling using sterilized recycled containers could be encouraged. Interviews indicate that the legal and regulatory system is not necessarily "prohibitive," but regulations reportedly could be revised to provide greater opportunities, protection, and incentives to informal sector entrepreneurs. For example, contracts conceivably could be prepared and written by a trained local agent or scribe in a legally binding manner so that commercial procedures could be widely used. Procedures for the acquisition, transfer, and registration of land could also be simplified. Guinea-Bissau 151 Notes This chapter was updated by Angela Walker of the World Bank's Private Sector Development Department based on a field visit carried out in February 1995. 1. The history of Guinea-Bissau is closely connected to that of Cape Verde because the Portuguese settled in both areas beginning in the fifteen century. 2. In 1981 an estimated 80,000 Guineans (nearly one-tenth of the population at that time) had taken up residence in Senegal. This emigration has affected mainly the northern regions of Guinea-Bissau, which have become increasingly separate from the economy of the south. 3. Luiz Cabral is the brother of Amilcar Cabral, the national liberation war hero of Guinea-Bissau who was killed in 1973. Luiz Cabral cofounded the Partido Africano da lndependencia de Guinea e Cabo Verde and was the first president of Guinea-Bissau. He was overthrown and jailed in 1980. He is now in exile. 4. The first multiparty elections were held in July 1994 and were generally considered to be free and fair. The ruling party-Partido Africano da Independencia de Guinea e Cabo Verde-won a majority of the seats in the legislature. The incumbent president was reelected, by a narrow margin, in a second ballot in August 1994. 5. The Beafada are a non-Islamic Senegambian group that is concentrated in the region north of Bambadinca. 6. Because the peso has little purchasing power, many traders rely on the CFA franc for trade. In numerous areas of rural Guinea-Bissau, the franc is the dominant currency of market exchanges. In fact, the peso is disregarded in some rural areas to such an extent that items of necessity like rice, sugar, and soap can only be obtained with francs. Even in the capital city of Bissau, the franc plays a major role in informal trade. 7. The Manjaco are an ethnic group of the Senegambian family. They represent 14 percent of the total population of Guinea-Bissau and are concentrated south of the Cacheu River and north of the Mansoa River. 12 Mali Pierre-Olivier Colleye Mali is a landlocked country of 1.2 million square kilometers in West Africa. Its geography is characterized by largely barren deserts in the north, Sahelian scrublands reaching across much of the middle of the country, and wet, forested subtropical savanna in the south. The economy is greatly affected by two opposing natural phenomena: the long traverse of the Niger River, whose interior delta of small tributaries and canals enriches the country's agricultural potential, and the extensive drought and desertifica- tion that have occurred over the last two decades. Unstable agricultural yields, the loss of herds as a result of diminished acreage for grazing, and lack of water have been chronic problems for Malians. Mali's population is around 8.7 million, with a 2.7 percent annual growth rate. In spite of a steady 5 percent growth in urban population, close to 80 percent of Mali's population live in rural areas and are largely dependent on agriculture, which accounts for more than 44 percent of GDP. Population growth exerts severe pressure on the limited natural resource base and on the financial resources available for basic public services. Per capita income was estimated at $288 in 1991, and real GDP growth has averaged about 3.3 percent a year since 1987. Much of the recent GDP growth can be attributed to increases in cereal and cotton output. The industrial sector remains small, is dominated by relatively low value added agroprocessing, and grew 8.1 percent annually in the late 1980s. Inflation averaged 3.3 percent in 1993.1 As measured by a number of social indicators, the quality of life in Mali is among the lowest in the world. Mali's infant mortality rate of 102 per 1,000 births and combined child and infant mortality rate of 200 per 1,000 are among the highest in Sub-Saharan Africa. Primary school enrollment is about 32 percent generally and is only 26 percent for girls. Adult illiteracy is estimated at 68 percent, less than one in five Malians has access to safe drinking water, basic health services are not widely available, and life expectancy at birth is only forty-eight years (World Bank 1994h). Mali's economic performance is dictated by a narrow resource base, highly variable weather, and a legacy of ineffective development policies that remained intact until the early 1980s. An overly complex fiscal system severely distorted the allocation of resources, while nonmarket monetary 153 154 Country Studies policies led to rapid growth of credit. Price controls, a restrictive trade regime, and excessive economic regulations were prevalent. The government launched a series of adjustment programs in 1982 that eventually reduced fiscal imbalances. However, reform slowed in 1986, and economic crisis ensued in 1987. A second, medium-term adjustment program began in 1988, targeting private sector growth, savings and investment promotion, improvements in international competitiveness, and better management of public resources. Measures have included restructur- ing the Banque de Developpement du Mali as a joint stock company, eliminating import and export licensing and price controls, and devising a trade reform package that abolished export taxation, simplified import taxes, and introduced a value added tax. Macroeconomic indicators began to improve in 1991. For example, the fiscal budget moved from a deficit of 0.3 percent of GDP in 1987 to a surplus of 2.1 percent of GDP in 1991. The current account deficit fell to 13.7 percent of GDP in 1991, from a peak of 28.5 percent in 1985. Deviations from the reform program occurred during the 1992 political transition to a democrati- cally elected government, but quick and decisive actions were taken to recover some of the losses. Still, budgetary revenues were lower in 1992, partially due to tax evasion facilitated by the transitional government and partially due to a drop in world cotton prices. The current government-a second democratic coalition-has pledged to take additional structural adjustment measures in the coming years. One area of structural adjustment with direct and indirect impact on the urban informal sector is public sector reform. Cutbacks in public employ- ment under the adjustment program have increased the number of people working in the informal sector while simultaneously decreasing real incomes and the demand for consumer goods (World Bank 1993d, p. 10). Main characteristics of the informal sector Microenterprise activities in Mali contribute significantly to supplying products and services to both urban and rural populations. The informal sector accounts for 20-25 percent of GDP, and its share of GDP should continue to grow (Grant 1991, p. 4). It is difficult to get an accurate estimate of employment in the informal sector. The Direction Nationale de Statistiques et d'Information estimates that there are approximately 80,000 artisans and 35,000 apprentices in this sector, but the real figure is probably much higher. In Burkina Faso, which has roughly the same population and many other similarities, 560,000 artisans were identified in a national census. Even this number, according to experts, seems to be underestimated (Grant 1991, p. 5). According to the World Bank (1993d), the informal sector employs 80-90 percent of the economically active population; roughly 70 percent of women in Bamako Mali 155 work in the informal sector. Most households, including those with salaried workers, engage in informal activities to supplement their income. The main nonagricultural activities are trade, artisanal production, and other services. Yet a large number of microenterprises, located mainly in Bamako, S6gou, and Sikasso, concentrate on wood and metal production, mechanical and electrical repair, and construction activities. In rural areas, farmers take informal sector jobs as a secondary source of income after the annual harvest. Many farmers become blacksmiths, cut firewood, harvest animal fodder, and go to urban areas to weave and enter the construction trades. The National Accounts indicate that 1.2 million women are employed in Mali, representing 37.6 percent of the employed population. Many are involved in commerce and services. Common activities include sewing, hairdressing, and food preparation. In rural areas, women produce vegetables, make pottery, and spin thread from November to March. These activities are important in maintaining the cultural and social traditions of villages, and they are often passed down from generation to generation (Grant 1991, p. 5). Women are relative newcomers to the business world in Mali. Sociocultural norms have encouraged women to remain in the home to serve the male head of the household. However, decreasing purchasing power has meant that one source of income is no longer sufficient to satisfy family needs, and more women are earning incomes outside their homes. Because there are few opportunities for employment in the formal sector, these women turn to self-employment and the creation of enterprises. Although women are key actors both in economic development and in safeguarding the welfare of the family, development agencies and govern- ment officials traditionally have overlooked them when designing economic and sectoral policies and programs. This is in large part due to the cultural milieu. With regard to education, employment opportunities, access to credit, and extension of services, there seems to be a general bias against women. The government has publicly called for integrating women into economic life, and Mali's recently revised constitution assures the protection of all Malians without regard to race, language, sex, or religion. Established laws and practices, however, often function as barriers prohibiting women from economic participation. For example, Malian women need written authorization from their husband to get a visa to travel, and Malian law prohibits married women from opening businesses without explicit authorization from their husband (U.S. Department of State Dispatch 1994a). Main constraints on the informal sector Microenterprises in Mali have limited access to formal channels of credit. For small borrowers, the family is the primary source of credit, followed by village associations and wealthier community members. The most wide- 156 Country Studies spread form of savings is the tontine, but tontine credit is used primarily for consumption rather than for investment. There are more than 10,000 tontines in Bamako alone, each with an average of ten members. See Box 12.1 for a discussion of gender constraints facing entrepreneurs in Mali. Box 12.1 Gender constraints in Mali Entrepreneurs in Mali face a number of common constraints regardless of gender: a limited domestic market, onerous and irrational taxes, lack of credit for investment, lack of access to market information, and lack of technically skilled labor. A number of constraints, however, particularly affect women. Perhaps the most important are limitations on women's legal rights. Although the Malian constitution assures gender equality, the marriage code and the commercial code both limit the right of the Malian women, especially married women, to conduct business. Under the marriage code, a woman may in principle undertake legal acts and separate activities from her husband. However, certain articles of the code contradict these rights. Article 34 stipulates that the husband is the head of household, article 35 mandates that married women who have a separate profession from their husbands are required to contribute to the payment of household bills, and article 38 states that married women cannot deal in commerce without their husband's authorization. The commercial code adds to these restrictions on women's economic activities, requiring married women to obtain their husband's authorization before participating in commercial activities. Overall monetary policy, particularly credit and interest rate ceilings, discourages banks from lending to micro and small enterprises. Women's access to credit is additionally limited by the stringent collateral requirements of local banks. Women generally own few assets acceptable to banks as collateral which puts them at a particular disadvantage. Women also have little access to well-connected bank customers willing to co-sign loans. Although women have the legal right to access banking services independently, many banks prefer to obtain the husband's authorization or another man's signature before granting credit. Women in business are constrained by the heavy tax burden and complicated tax system which are common to all small businesses. Complex government regulations pervade many critical aspects of business. These procedures are difficult for anyone to comprehend and to follow, but they are especially so for women, who lack education, experience and self-confidence necessary to operate in the business and bureaucratic world. Faced with these incomprehensible rules and regulations, most women opt to stay in the less regulated informal sector. Source: Grant 1991, p. 6. Mali 157 Programs and projects for microenterprise development Apart from informal credit structures (families, tontines), nonprofit organizations have initiated a wealth of programs to provide more organized forms of financial services in rural areas. These schemes typically rely on characteristics most valued by the poor: flexibility, timeliness, and little or no collateral. Some initiatives link small savers and borrowers to the formal financial network. Moreover, some of these programs appear to avoid the trap of providing subsidized credit by making financial viability a key priority. The remainder of this chapter will examine fourteen programs aimed at promoting informal sector growth in Mali. Four provide credit and technical assistance in an attempt to graduate informal sector enterprises into the formal sector. These programs serve a more economically advanced target group (some of whom might not fit under the conventional definition of a microenterprise). The remaining ten are microfinance and technical assistance programs that promote microenterprise growth. Some of the savings and credit programs stress indigenously mobilized deposits, while other NGO programs are completely donor-driven. Some programs also provide (anC subsidize) consulting and accounting services. The following section discusses how these different approaches to microfinance delivery have led to varying degrees of success. Programs to graduate informal sector enterprises The ILO has been involved in the Malian informal sector for more than a decade. The ILO project was nearly shut down in 1989, but it continues today under a new structure financed by the Caisse Francaise de Developpement. It aims to help microenterprises move from the informal to the formal sector by registering enterprises with the chamber of commerce, encouraging proper bookkeeping, and opening bank accounts. The ILO has helped entrepreneurs to form groupements d'intetret economique or professional associations. Organizing the target group of entrepreneurs and building capacity among their associations are the main focus of the project. Self-management, self-financing (through membership fees), and training are the three fundamental inputs. With these advantages, associa- tion members should be able to improve the performance of their business and to protect themselves from unfair govemment policies. The ILO has also funded technical assistance to bureaux d'etudes (business support centers) that helps microentrepreneurs to prepare feasibility studies and action plans to present to banks when applying for a loan. The ILO assists with bank applications and monitors the performance of microenterprises for the duration of their loans. On occasion, it also mobilizes funds from organizations such as the Caisse Francaise de D6veloppement when bank loans are not available. 158 Country Studies The problems experienced by the project are instructive, and efforts have been made to overcome them since 1990. Still, some problems persist, including the following: * The program lacks a participatory approach that would engender local commitment. * The associations are financed by grants and therefore have little interest in raising local funds. * The heads of the original associations are experienced artisans; some have used the associations largely for their own benefit and have not encour- aged democratic processes or allowed new initiatives. * There has been little transparency within these associations. * Although the program has achieved some success, the project has only reached a limited number of artisans (U.S. Department of State Dispatch 1994a, pp. 137-42). The Fonds Europeen de D6veloppement program was started in 1988 and has since assisted 1,000 enterprises. Similar to the ILO program, it assists entrepreneurs with the preparation of business plans, with bookkeeping, and with efforts to gain access to formal financial services. Once proper business plans have been prepared, the Fonds Europ6en de Developpement approaches different banks, such as the Banque de Developpement du Mali or the Banque Nationale de Developpement Agricole (BNDA), on behalf of the enterprise. The program is highly dependent on external funding. Loans vary from CFAF 2.0 million ($4,000) to CFAF 50.0 million ($100,000) with a subsidized annual interest rate of 12-13 percent.2 The program is committed to maintaining the interest rate at no more than twice the Central Bank lending rate and therefore makes no effort to cover the costs of operations. Repay- ment rates are around 75 percent. Project d'Appui aux Iletites et Microenterprises (PAPME), financed by the Canadian International Development Agency, has a budget of $3.6 million. It assists residents of Bamako to develop productive enterprises in agroindustry, construction, and light manufacturing. Its goal is also to move enterprises from the informal sector into the formal one. PAPME serves as a link between enterprises and commercial banks (the Bank of Africa and the Banque Malienne de Credit) by assisting entrepre- neurs with business plans and loan applications. The program is designed to assist 200 small enterprises and 700 microenterprises. The financing of these enterprises is equally divided among the entrepreneurs, the banks, and PAPME itself. Initially, the program enjoyed a good deal of success. Enterprises' performance improved tremendously, repayment rates were high, and the Mali 159 program was run efficiently. However, as the banks were not able to provide many services to borrowers and the quality of the business plans prepared by the business support centers diminished, PAPME'S role changed. The organization became more involved in the everyday management of the enterprises and began offering monthly consulting and accounting services. Not surprisingly, the efficiency of the program plummeted. In the last year, the nineteen employees were assigned to work with only fifty-five enterprises. With such high operating expenses, 13 percent a year interest rates did not come close to covering costs. Repayment rates were below 60 percent, and collateral, when it existed, was not claimed. The Canadian International Development Agency has announced that it will stop funding the project in 1994. The Caisse Francaise de Developpement offers subsidized loans to help small Malian businesses transfer from the informal to the formal sector. In order to access credit, an enterprise must register with the chamber of commerce, have adequate bookkeeping, and be part of a professional association. Entrepreneurs can borrow up to CFAF 40 million ($80,000), usually for five to six years at annual interest rates of 8-12 percent. Many applicants are referred from the Fonds Europeen de Developpement and the ILO. The Caisse Francaise de Developpement has only made fifty-five loans over a two-year period; its operations are labor-intensive, and their cost efficiency is questionable. Programs to promote microenterprises Caisses Villageoises d'Epargne et de Credit Autoger&es in Dogon Country (a village savings and credit association) is managed by the Centre International de Developpement et de Recherche, with the assistance of the Direction Nationale de l'Action Cooperative au Mali. It is funded by KfW. Its goal is to create a series of autonomous village banks that can mobilize local savings and provide small loans. Each caisse is indigenously and independently managed. An elected board and the bank managers set interest rates and approve loan requests. Caisse members can open one of two types of savings accounts: one that does not pay interest but can be withdrawn at any time or one that keeps their money with the caisse for more than three months and earns annual interest of 20 percent. Loans carry an annual interest rate of 40 percent. Three associations or unions represent the village banks and negotiate between the BNDA and the caisses. These unions use the caisses as collateral and have enough credibility to quell the BNDA'S concerns about the size and riskiness of informal sector lending. Refinancing from the BNDA amounted to CFAF 22 million ($73,000) at the end of 1993. The unions are flexible, small organizations that are financed by the differential between interest rates paid to BNDA for credit lines and rates charged the caisses as the funds are passed down. 160 Country Studies By the end of 1994, this project had created fifty-five village caisses with a total of 15,300 members (33 percent of whom were women). The caisses had collected CFAF 143 million ($286,000) in savings, and they had made 9,767 loans totaling more than CFAF 211 million ($422,000) to village members. The average loan size was CFAF 21,600 ($43) for an average term of 4.5 months. In 1994, repayment rates to the village associations were 99.5 percent, and association repayment to BNDA was 100 percent. The Centre International de Developpement et de Recherche will be removing its expatriate staff by mid-1995 and is working actively to make the project self-sufficient. Central to its strategy is the creation of a private organization called service commun that will provide the services that villagers cannot perform themselves, such as audits and negotiations with banks. The Societe de Developpement International Desjardins (SDID) manages a caisses villageoises project in Segou. The project, financed by the Canadian International Development Agency, started slowly, but by May of 1994 the network included fifteen caisses with 2,019 members. Caisses charge an entrance fee of CFAF 2,000 ($4), and transactions average 2.5 per member a month.3 The average number of caisse members is ninety-seven, with women representing approximately 30 percent of membership. Caisse members are usually part of a solidarity group that relies on peer pressure to motivate repayment. These groups can be formed according to profession, sex, ethnicity, neighborhood, or even political agenda. In order for a member to borrow funds he or she has to have been a member of the caisse for at least six months, to have 30 percent of the loan amount in a savings account, and to guarantee the remaining 70 percent with personal items. The lending interest rate is 10 percent semiannually, and most loans are for three to six months. In the case of delinquency, a delay of forty-five days is given to the borrower (with a penalty of 5 percent a month), after which the guarantee is used. As of May 1994, repayment rates were at 100 percent, and delinquency had yet to occur. The SDID program illustrates that given the proper financial infrastruc- ture, the poor do save. In one village where a caisse has been operating for three years, not a single loan has been made. When asked for an explanation for the lack of credit, the village chief responded simply by stating, "Borrowing money is the first step to bankruptcy." For most members, the caisse functions as a useful disciplinary tool, serving as a means to resist the temptation to spend money in the present. The credit to savings ratio in all fifteen caisses is only 11 percent. Currently, savings do not yield interest, but it is hoped that as soon as the caisses cover their operating costs, they will start paying about 3 percent semiannually.4 SDID finances 100 percent of the operating costs during the initial stages, but this subsidy is targeted to go down to 75 percent in 1995, 50 percent the Mali 161 following year, and 25 percent the third year. The caisses are expected to cover all their expenses by 1998. Kafo Jiginew has helped to create rural savings and credit cooperatives linked to cotton production since 1987. With support from a consortium of European NGOS and the European Development Fund (including technical assistance from the French Ministry of Cooperation), Kafo Jiginew has helped to set up fifty-five savings and credit cooperatives with roughly 20,000 members (98 percent of whom are men). Each member-managed savings and credit cooperative serves five to eight villages within a 10-kilometer radius. The cooperatives pay 4 percent interest for checking accounts and 6.5 percent for savings deposited for a minimum of one year. Interest rates on loans vary from 2 percent a month for equipment loans of CFAF 20,000-CFAF 100,000 ($40-$200), with terms up to three years, to 4 percent a month for loans of CFAF 5,000-CFAF 20,000 ($10-$40), with terms of one to three months. After six years, the project is recovering roughly a third of its operating expenses and expects this share to increase steadily. The outstanding loan portfolio at the end of 1993 was approximately $800,000, with an average loan of about $105. Total savings were $820,000. The repayment rate was nearly 100 percent. The project has enjoyed a good deal of success, in part because it works closely with various commercial banks. Regional agricul- tural production is dominated by the cash crops of tobacco and cotton, giving banks a certain guarantee of repayment. The Cr6dit Mutuel de Segou was developed with the assistance of the European Union to provide business advisory services and access to bank credit for young entrepreneurs. The Credit Mutuel is a savings and credit organization, structured to attract member savings and to convert the funds into monetized guarantees for loans to participants. Monthly deposits are collected from each member until the group has reached the percentage of total credit needed. The deposits are then placed in a group savings account that is used to guarantee bank loans for members. Credit Mutuel's strategy acknowledges some of the key constraints encountered by the poor. Collateral is based on group savings, processing procedures are simple because members know each other, and repayment rates are high due to social pressure. Under this scheme, banks save on transaction costs, because distribution, monitoring, and collection of loans are handled by the borrower group. In 1988 Save the Children-U.S.A. set up pilot village-level credit and savings banks on the southern border of Mali to lend predominantly for agricultural activities, including food storage. The scheme works on a revolving fund basis under which Save the Children gives advances to village banks that, in turn, lend to individuals. New funds are available only when loan repayments are up to date. Although Save the Children handles the daily administration of the scheme, village committees have formed a 162 Country Studies union of banks that has begun to make policy decisions. Save the Children provided a capital base of $75,000, and the BNDA provided an additional $10,000. Once members pay an initial membership fee of CFAF 1,500 ($3) and open a savings account of CFAF 1,000 ($2), they can borrow up to ten times the amount of their savings. Women dominate among savers, and men dominate among borrowers. Annual interest rates are 10 percent, the same rate charged by commercial banks. However, it is expected that the interest rate will increase to 25 percent by the end of 1999. The annual savings interest rate is around 4 percent. Repayment rates for the program are roughly 96 percent. Freedom from Hunger (a California-based NGO) currently administers a microcredit program using the village bank methodology developed by the Foundation for International Community Assistance. Village banking was chosen over other models, because it delegates maximum responsibility to village associations and stresses the importance of self-management. Village banking involves the following approach: * The distribution of small loans to meet the credit needs of clients with the promise of additional loans in the future. * The provision of credit on a short-term basis with repayment through regular installments. * The restriction of loans to productive activities that enhance incomes. * The limiting of loans to familiar activities that do not require costly training and technologies. i The provision of credit for the production of goods and services that rely on local inputs and have readily accessible markets. D The promotion of personal responsibility via individual loans adminis- tered through borrower groups made up of self-selected peers who are jointly responsible for repayment. - The financing of activities deemed feasible by peer groups and not by expensive feasibility studies. Freedom from Hunger made one significant modification of the Foundation for International Community Assistance model by breaking down the associations into borrower groups of approximately five members. By banding small borrower groups into larger credit associations, Freedom from Hunger has been able to bring together groups of women who know one another, thereby minimizing the risk of delinquency; to reinforce the association by shifting accountability to smaller groups; and to streamline the loan review process (loan applications are discussed in the smaller groups before they are approved by the association as a whole). Mali 163 This approach has been quite successful in Mali. The first loans were made in January 1989, shortly after Freedom from Hunger created the Centre d'Appui Nutritionel et Economique aux Femmes, which was to provide the credit and savings services. The program began to expand significantly six months later. Personnel and systems are now in place for a significant expansion of the program, both in the Dogon region and elsewhere. The center is providing services for more than 2,000 rural Malians organized into credit associations of twenty to thirty people, 90 percent of whom are women. It has a portfolio of $47,047 (57 percent of which is provided by the BNDA, which has promised to double its line of credit this year). By the end of 1996, Freedom from Hunger projects are expected to reach 7,500 group members organized into 300 associations in three regions of the country (Ashe and others 1992). USAID has sponsored two programs that represent the range of success and failure in Malian microfinance programs.5 The first program was designed to assist civil servants leaving the public sector to start their own businesses. USAID placed guarantees in Malian banks so that civil servants would be able to borrow without collateral. Unfortunately, the riskiness inherent in providing start-up capital rather than working capital became apparent in this program. Most of the loans were spent on consumption or family trips rather than on enterprise creation. The few that were used to create microenterprises failed miserably. None of the loans was fully repaid to the bank, and USAID quickly canceled the program. The second program, financed by USAID in 1988, commissioned the Office de la Haute Vallee du Niger, a parastatal agency, to develop village cooperatives and to assist them in their dealings with commercial banks. This program, implemented with the assistance of the Cooperative League of the United States of America, has enjoyed a great deal of success, creating more than 308 cooperatives in the cotton and tobacco regions south and west of Bamako. To become a cooperative member, a village must have a minimum literacy rate of 15 percent. If this condition is not met, the Office de la Haute Vall6e du Niger organizes intensive literacy courses. Once this target is reached, the Cooperative League trains cooperative managers in bookkeep- ing, feasibility studies, general management, cooperative organization, and banking operations. On completing the training program, the cooperatives can go directly to any one of four commercial banks. USAID offers a guarantee (from a $710,000 fund) of up to 45 percent for the first loan, 30 percent for the second, and 15 percent for the third. By the fourth loan, it is hoped that the borrowers will have built a credit history and have accumu- lated enough wealth to serve as collateral for the loan. Loans vary in size from CFAF 3 million ($6,000) to CFAF 10 million ($20,000), and terms are one to two years. Banks charge 14 percent for these loans. Repayment rates have been higher than 98 percent, and interest earned on the fund has more than 164 Country Studies covered losses on payment defaults. The activities financed range from the production of cotton and tobacco to the building of nurseries. Responsibility for managing the loans is left to the cooperatives, and the subsidy compo- nent of this program is very small. The replicability of this program is not clear. Participation by commercial banks is key, and they have been active in this program because of the proximity of the cooperatives (the farthest is only 100 miles from Bamako), a guaranteed cash crop of cotton and tobacco, an ample guarantee fund, and an organization (Office de la Haute Vall&e du Niger) that provides technical assistance when village projects are in trouble. It is questionable whether this program design could achieve the same degree of success in more remote regions. Programs to offer technical assistance and support The Programme d'Animation Economique, a ten-year program managed by the Centre International de D6veloppement et de Recherche, is linked to the Caisses Villageoises d'Epargne et de Credit Autogerees in Dogon Country described above. The program helps to prepare feasibility studies, identifies promising subsectors where microentrepreneurs could concentrate, prepares technical training and assistance on product design in these subsectors, and disseminates the information to the caisses of the Dogon project. The objective of Acccs des Femmes au Credit (financed by SDID) is to improve the living standards of women by providing technical assistance with their economic activities. Its functions are manifold, including teaching women in Segou about the importance of savings, helping local caisses to design credit products that are tailored to women's needs, producing feasibility studies for women borrowers and assisting them with the loan application process, providing market information to microentrepreneurs, organizing women into solidarity groups, and providing technical assistance when businesses are in trouble. Issues and recommendations The range of financial and technical assistance services available to Malian microentrepreneurs is a great deal more extensive than elsewhere in the Sahel.6 A review of these programs raises a number of issues about how and where programs to assist the informal sector should be structured. Market saturation is a big problem for microenterprises in Mali. Entrepreneurs badly need to acquire new skills for new kinds of businesses. However, the USAID experience in providing start-up capital to the enterprises of former civil servants illustrates the disastrous results that can come from targeting credit to start-up microenterprises. There is an abundance of programs in rural areas, but few credit programs for microenterprises in urban areas. It might be useful to help Mali 165 successful programs with a proven track record to expand their activities to urban areas. The large numbers of savings and credit programs for microenterprises in Mali have, in effect, created the beginnings of a parallel financial system. The implications of this are unclear. An increasing number of low-income people are receiving financial services that they otherwise would not receive. The efficiency with which these services are provided varies by program, but some programs clearly are moving toward higher levels of cost recovery. As this system of informal financial institutions grows, policymakers will be faced with a series of issues. For example, to what extent should informal finance institutions be allowed to substitute for banks? How should these kinds of institutions be regulated, and how can regulations be crafted such that they improve standards in this field rather than introduce new barriers to efficient service delivery? Because microfinance institutions are relatively more developed in Mali than in other Sahelian countries, these kinds of issues are likely to be on the table sooner in Mali than elsewhere. Opportunities for linking informal financial institutions with banks might be greater in Mali than elsewhere because of the large number of microfinance institutions, their relative strength, and the fact that the BNDA has already extended lines of credit to some programs. These kinds of links between formal and informal financial sectors are rare, and they merit further examination regarding the terms and conditions of these lines of credit, the experience thus far of the involved parties, and the possibilities for replication, both with commercial banks in Mali and in other countries. Finally, many Malian microfinance institutions seem to operate with little substantive contact with similar institutions in Mali and other countries. Rather than seeing the common problems facing them, managers in these institutions seem to concentrate on the differences in their approach and the irrelevance of others' solutions to their own situation. Donors could assist in disseminating best practices at the national and regional level and in facilitating contacts among Malian and foreign institutions.7 Notes This chapter incorporates the findings of field visits in July and November 1994. 1. Even though inflation surged in 1994 because of the devaluation of the CFA franc, real interest rates in the programs described here are based on the low inflation rate of 3.3 percent. 2. The exchange rate used in this chapter is 300 CFA francs equals 1 U.S. dollar prior to devaluation and 500 CFA francs equals 1 U.S. dollar after January 1, 1994. 3. This fee includes a CFAF 500 membership fee, CFAF 500 for capital (equity), CFAF 500 for identification pictures, and CFAF 500 for stationery. 166 Country Studies 4. Of the deposits, 10 percent are kept in a safe, 20 percent are used for making loans, and 70 percent are placed in a local bank account that yields 7 percent annually. 5. USAID also finances a third microfinance program that is implemented by CARE. Unfortunately, information on this program was not available. 6. Additional Malian microfinance programs that were not visited include Projet d'Appui f la Creation d'Entreprises et f l'hnsertion des Jeunes Dipl6mes, l'Office Riz S6gou, les Associations Villageoises de l'Office du Niger, le Projet FIDA-Fonds de Developpement Villageois, Kondo Jigima, and Appui f la D6centralisation 7. An association called Pivot Petite et Moyenne Entreprise/Petite et Moyenne Industrie located in Bamako disseminates information on a small scale through a magazine called NAAFA. This effort could be expanded to include more information and more institutions. 13 Mauritania Anne-Marie Chidzero Mauritania is a vast country of 1.1 million square kilometers, bordering western Sahara and Algeria to the north, Mali to the east and southeast, Senegal to the south, and the Atlantic Ocean to the west. The majority of its 2.1 million inhabitants have extremely low standards of living as evidenced by a per capita GNP of $530 in 1992, an average life expectancy at birth of forty-eight years, an infant mortality rate of 117 per 1,000 births, and a child malnutrition rate of 31 percent (World Bank 1994h, pp. 226-27). According to a recent poverty assessment, 57 percent of the population spend less than $370 a year and are classified as poor. The very poor represent 44 percent of the population and spend less than $275 a year.' Additionally, approxi- mately 66 percent of the population are illiterate, and barely half of school- age children attend primary school.2 Mainly infertile, arid desert, Mauritania has a very limited resource base. Its main economic activities are iron ore mining, offshore fishing, agricul- ture, and livestock. During the 1960s, mining and animal husbandry provided the main sources of income. This situation has changed dramati- cally over the last two decades. Periodic droughts and difficult economic conditions in rural areas have spawned large-scale migration to urban centers, where about half of Mauritania's population now live.3 Addition- ally, a fall in iron ore prices brought decline to the mining sector, and its contribution to GDP dropped from 17 percent in 1973 to 11 percent in 1991. Similarly, livestock's share of GDP declined from 27 percent in 1973 to 17 percent in 1991. The fishing sector's contribution to GDP increased margin- ally from 7 percent in 1973 to 8 percent in 1983 but declined during the 1980s to reach 5 percent in 1991 (the sector is facing severe shortages as a result of overfishing). Tertiary sector activities have increased from 37 percent of GDP in 1973 to 43 percent in 1991 (World Bank 1985b, annex table 2.2, p. 87; World Bank 1993k, pp. 317-19). In 1985 the country faced the most severe economic crisis in its history. In response, the government, under the new leadership of Colonel Ould Taya, undertook fundamental reforms to alter the structure of the economy. The structural adjustment program was designed to redress internal and external imbalances that had emerged in the early 1980s as a result of persistent drought, the decline of export revenue from iron ore, poor economic management, and political instability. For the first four years of 167 168 Country Studies the adjustment program, between 1985 and 1989, GDP grew by an average of 3.5 percent a year. Between 1989 and 1992, the economy again suffered a series of internal and external shocks that slowed the adjustment effort. In 1989 the reform program was disrupted by land tenure problems along the Mauritanian- Senegalese border that brought domestic turbulence.4 This situation was aggravated by Mauritania's decision to side with Iraq during Iraq's invasion of Kuwait in 1991, which precipitated a dramatic drop in international aid to Mauritania. With the addition of chronic drought, productivity declined, financial imbalances reemerged, the country's external position deteriorated, and reform efforts slowed. Annual GDP growth fell to 1.7 percent in 1992. The economic outlook improved in 1992, in part due to constitutional reforms, elections, and an expanded public dialogue. The government embarked on a new adjustment program in late 1992 with broad reforms in areas including trade, taxation, pricing, public enterprise restructuring, public expenditures, civil service reform, and social policy. In the Fourth Policy Framework Program, the government renewed its commitment to economic liberalization and the development of a free market economy. In 1993 GDP grew by 4.9 percent. Despite this progress, the economy remains fragile. Mid-term prospects continue to be limited by a poor resource base of natural and human capital, a high degree of vulnerability to exogenous factors (climatic conditions and commodity prices), an inelastic revenue base, and an extremely heavy debt burden. Mauritania has a complex social structure of ethnic groups, socioeco- nomic classes, and clans (see box 13.1). The major ethnic division is between Moors and non-Moors. A rigid hierarchy within both groups separates the nobles (rulers and owners of land and wealth) from the non-nobles and the poor (workers, artisans, and servants). According to tradition, the nobles support the poorer members of their group, who, in turn, repay that support in the form of labor, gifts, and political support. Over the past three decades, Mauritania has experienced enormous socioeconomic change. Largely a rural nomadic society in 1965, it is now 50 percent urban (World Bank 1994c, pp. 11, 17). In 1965, 85 percent of Mauritanians were nomads; today this figure is less than 10 percent. This massive movement of people out of the desert and into urban areas has brought urban slums known as kUbb&s to the outskirts of Nouakchott, and it has provided the impetus for growth of the informal sector.5 It has also contributed to the erosion of Mauritania's traditional social structure, which determined the distribution of wealth, defined the social divisions within economic activities, and provided the solidarity networks for poorer groups. Mutual support networks have weakened as ruling classes have been unable to keep pace with the needs of a growing population of the poor and found less use for their services. Poverty is no longer restricted to groups at the bottom of the traditional hierarchy, because economic hardship has affected Mauritania 169 Box 13.1 Social structures in Mauritania Ethnically, Mauritanians can be divided between Moors and non-Moors. The dominant Moors are divided between the Arab/Berber group (white Moors) and the Haratins (freed slaves). The non-Moors include the Peuls, Soninkes, Toucouleur, and Wolofs. The two ethnic groups share similar hierarchical social structures consisting of a noble elite (warriors and marabouts), a non-noble group of free persons, artisanal caste groups, and serfs or freed slaves. Moorish society is organized among extended family lines. Every Moor belongs to a clan that consists of a large group of people descended from a common ancestor. Within the clan are factions or fahds that function as organized interest groups. Each fahd consists of family units called aial. Non-noble elements of Moorish society align themselves with a particular noble fahd or aial. Most decisions are made at the aial level, and the interests of the aial are represented by the fahd. In non-Moorish society, individual households form part of a compound headed by a senior male who represents the interests of the compound in village discussions. Source: World Bank 1994c. almost all Mauritanians. Regardless of their clan and traditional social status, many Mauritanians have been forced to learn new skills and to find new means of survival. Main characteristics of the informal sector Informal sector enterprises employ about a third (150,000) of the active workforce (458,000), and they contribute roughly 10 percent to GDP (World Bank 1994i, pp. 3-4). The informal sector is growing rapidly at a rate of 7 percent a year. The lack of available jobs in the formal and public sectors, desertification, and generally poor economic conditions have pushed many Mauritanians to self-employment. Microenterprise activities fall into three broad categories: trade and services (85 percent of informal sector employment), artisanal and light manufacturing (4 percent), and construction (10 percent). In recent years, there has been a marked movement toward trade and services. In 1980 trade and services generated only 71 percent of informal sector employment, while manufacturing accounted for 15.6 percent and construction for 11 percent (Charmes 1992, pp. 60-66). The trend toward trade and services can be explained in part by changes in the traditional structure of Mauritanian society. Manufacturing activities (artisanry) were traditionally perforrned by specific castes. The younger and better educated members of these castes are moving out of traditional 170 Country Studies artisanal activities and into trade and services where economic opportunities are more favorable. Trade liberalization measures have made it easier to import consumer goods and have encouraged many entrepreneurs to enter into trade (Charmes 1992, pp. 60-66). Operating mainly in groups, women are very active in the informal sector. Twenty-four women's associations and more than 600 women's cooperatives have registered with the government's Secretariat d'Etat pour la Condition Feminine. Group members typically pool their financial resources, lend moral support to one another, maintain intemal savings accounts, and extend informal loans on a very small scale. Men dominate in agriculture, jewelry manufacturing, and construction, and women are more active in trading and clothes manufacturing. Under Islamic law, men do not have rights over the assets of their wives, and women manage their businesses independently. There are a large number of cooperatives and associations in Mauritania. In addition to the 600 women's cooperatives, a 1989 study suggests the existence of more than 3,000 cooperatives in the agricultural sector, twelve in fisheries, and an undetermined number in artisanry (Societe de Developpement International Desjardins 1990, pp. 13-16). Cooperatives mainly provide a solidarity network for members. No cooperative mobilizes savings and extends credit on any significant scale, as is seen in other Sahelian countries. Attempts to organize the informal sector at a national level have not met with much success. The Union Nationale de Groupements Artisanaux Mauritaniens, created with donor assistance, fell apart when it was clear that only a few members were benefiting from its services. Three separate chambers were created early in 1994 to replace it: La Chambre Feminine; La Chambre d'Artisanats Traditionels, a group of 700 artisans producing crafted jewelry, trinkets, farm tools, and traditional arts and crafts; and La Chambre des Metiers. The impact of these associations on the development of the informal sector is unclear, because they are fairly new. Main constraints on the informal sector Entrepreneurs in the informal sector face a number of constraints that limit their ability to operate effectively. Some of these follow: * Limited access tofinancial services. Financial resources for start-up, working capital, and long-term investment are very limited. The family network and informal financial systems provide the small sums needed for start- up costs, but they reportedly are inadequate to meet the need for working capital and for long-term investment capital. These informal networks also are limited, because a small number of people have access to them. Other sources of financial services are practically nonexistent. Mauritania 171 * Lack of skills. Mauritanians generally are entrepreneurial, but most lack technical and productive skills. The ability to expand, diversify, innovate, improve product quality, and develop effective marketing strategies is constrained by weak technical and business skills. * Lack of demand. Demand for informal sector products is constrained by the small size of the population, very low levels of income, and fierce competition stemming from the existence of many producers of most product lines. * Poor infrastructure. Basic infrastructure in the cities and urban slums is very poor. Businesses are managed from shacks, from tents, or "under the sun" where there is no running water or sanitation. It must be assumed that informal businesses could operate more efficiently with better infrastructure. Policies and programs for microenterprise development Very few NGOS and other nonbank financial institutions provide financial services or technical assistance to microenterprises in Mauritania. The World Bank poverty assessment for Mauritania identified only twelve international and seven local NGOS. These NGOs are involved exclusively in humanitarian and health activities. These few include the government-sponsored Union Nationale des Cooperatives Agricoles de Credit et d'Epargne en Mauritanie (UNCACEM), an agricultural bank registered under the cooperatives law;6 recent efforts by donors to organize informal credit networks, as illustrated by the Bureau d'Appui aux Artisans; and failed efforts to channel funds to microenterprises through a development bank. Only one other nonbanking institution, Investissement-Developpement en Mauritanie, which is described in box 13.2, can be identified, and it caters to small- and medium- size enterprises, not microenterprises. UNCACEM was created in 1992 and is financed by the government of Mauritania in conjunction with several donor organizations to provide credit for agricultural projects. It has been active for fifteen months, and 550 clients have received at least one loan; 70 percent of all loans issued are medium-term loans with terms of four years. The annual interest rate is 12 percent. Repayment rates average 91 percent on short-term loans and 95 percent on medium-term loans. A ceiling of ouguiyas (UM) 30,000 ($240) is placed on medium-term loans.7 The 1993-94 loan portfolio stood at UM 750 million ($6 million). UNCACEM takes deposits from borrowers only, and no interest is accrued on these savings accounts. In early 1994 the Bureau d'Appui aux Artisans was set up by the Chambre Feminine, Artisanats, et Metiers with the assistance of the German Agency for Technical Cooperation to provide technical and financial assistance to its members. To be eligible for loans, entrepreneurs must have 172 Country Studies Box 13.2 Investissement-Developpement en Mauritanie The Investissement-Developpement en Mauritanie (IDM), an affiliate of the Societe d'Investissement et de Developpement International, was created in 1991 with funds from the Fonds Europeen de Developpement and the Caisse Franqaise de Developpement. It was established with a capital base of UM 200,000 ($1,600) under the supervision of the Ministry of Planning, the Central Bank, and the Delegation G6n&rale des Mauritaniens a l'Etranger et a l'Insertion. IDM was originally set up to assist repatriated Mauritanians after the conflict with Senegal in 1989. It has since broadened its mission to serve all small entrepreneurs in all sectors except agriculture and trade. It provides credit and takes equity positions. However, because banking regulations prohibit institutions outside the banking sector from undertaking any lending activity, IDM acts as a manager of the credit lines from the Fonds Europeen de Developpement and the Caisse Francaise de Developpement. IDM's role is to identify projects and provide technical assistance to entrepreneurs. Loan and equity agreements are signed between the donors and beneficiaries, not with IDM. The credit lines from the Fonds Europ&en de Developpement are used for equity investments starting at UM 200 mnillion ($1.6 million), and they have no upper limit. The credit lines from the Caisse Franqaise de D6veloppement are granted at an interest rate of 13 percent (inflation is 34 percent), and they range from UM 400,000 to UM 7.5 million (approximately $3,200 to $60,000). Source: World Bank mission, July 1994. savings deposited with the Chambre equal to at least 30 percent of the loan amount requested. Loans average about UM 50,000 ($400). Repayment rates are 100 percent among women and about 85 percent among men. Loans are used mainly for trade transactions. Offices have been established in Atar, Kaedi, and Nouakchott. To fill the gap in financial services to small borrowers and respond to the crisis of 1989, the French Ministry of Cooperation created the Cellule d'Appui au Cr6dit f la Reinsertion. This program directed a line of credit through a development bank, Union des Banques de Developpement, for onlending to repatriated Mauritanians seeking to start small businesses. The program closed at the end of 1992 when the development bank was liquidated as part of World Bank-sponsored financial sector reforms. Loan repayment rates were generally quite low. There are several hypotheses for the scarcity of financial NGOS in Mauritania. One is the absence of regulations for nonbanking intermediaries and restrictions under the existing banking laws. Banking regulations governing the operations of financial institutions stipulate that to register as a financial institution, an initial capitalization of no less than UM 500 million Mauritania 173 ($4 million), not including deposits, is required. This high level of capitaliza- tion discourages NGOS from implementing credit programs. Until 1988 there was no framework for holding a dialogue between the government and NGOS. Since 1988 the government has been trying to increase the participation of NGOS. In 1989 the Federation des ONG (the Federation of NGOS) was created. In 1992 the Cellule de Coordination Gouvernement-ONG was created to serve as a permanent contact between NGOs and the government. The lack of credit and savings programs in Mauritania also can be traced to the absence of donor funds. NGOS tend to be present where there is a strong donor community. Donors are not particularly active in Mauritania, in part because of its human rights record and, in part, because of the position it took during the Iraq-Kuwait conflict. Kuwait, Saudi Arabia, and the United States have closed their operations in Mauritania. In addition to these programs, the UNDP and the World Bank have supported organizations that promote microenterprise development. In general, these programs have focused on strengthening support organiza- tions. The following two projects were executed by the UNDP in support of microenterprise development. No further assistance is envisaged for this sector under UNDP'S Fifth Country Program covering 1992-97. The Creation of Artisan Enterprises is a twenty-six-month project that was launched in November of 1991 and executed by the ILO. The objectives were to train repatriated Mauritanians in trade, artisanal, and vocational skills, to support the creation of artisanal enterprises, and to develop the institutional capacity of the implementing agency, the Direction de l'Artisan (the government body in charge of artisanal activities). The project closed in December of 1993. According to the final evaluation, the project achieved its stated objectives. UNDP has also financed two private sector development projects, both executed by UNIDO. The first ($500,000) was designed to strengthen the capacity of La Confederation Generale des Employeurs de Mauritanie to provide services for start-up microenterprises. Services for entrepreneurs included assistance in developing business plans and in preparing project proposals for funding. The second was launched in 1991 as a follow-up to the first and involved $903,000. No evaluation has been done, but at least 300 projects received technical assistance under the second project. Funding under this project expired at the end of 1994. Two World Bank projects are designed to promote urban microenterprise sectors: an education and a construction project. Previously, the Bank financed two industrial development projects that included components for artisans. As of March 1994, the Technical Education and Vocational Training Support Project supports development of a coherent system of technical education and vocational training. Activities under the vocational training 174 Country Studies component ($1.6 million) include a survey of the training needs of small- scale enterprises in the Nouakchott area, the streamlining and training of staff of the Centre de Formation et de Perfectionnement Professionnel, and the development of new training courses. The project also intends to establish a training support fund in a commercial bank to finance training activities initiated by and for microenterprises, small-scale enterprise employees, and apprentices. This facility will finance project proposals developed by informal sector businesses in cooperation with the Centre de Formation et de Perfectionnement Professionnel. Proposals are to be approved by a technical committee before they are awarded to local training agencies for execution. The fund will finance 100 percent of the cost of the training courses, and grants can range from $1,000 to $10,000. The Construction Capacity and Employment Project, established along the lines of the Revue de l'Agence d'Execution des Travaux d'Interet Public contre le Sous-emploi (AGETIP) project in Senegal, seeks to create new employment activities in urban areas by subcontracting public works projects to local small enterprises and to develop a construction capacity within the country. The Agence Mauritanienne d'Execution des Travaux dl'nt&et Publics et pour l'Emploi has been created to develop construction capacity in the country. The Industrial and Artisan Development Project was approved in 1985 and closed in December 1991. As a follow-up project to the first Urban and Rural Development Project, its objectives were to support diversified economic growth through the development of the industrial and artisanal sectors. One component provided technical assistance, training, and equipment for carpet weavers through the Office Mauritanien du Tapis. Project implementation was adversely affected by poor macroeconomic conditions and institutional weaknesses. The line of credit financed only eleven subprojects with a total of $1.6 million-clearly not reaching the small artisans for which it was partially intended. There is little local institutional support for the informal sector. The following were identified as providing limited support to microenterprises: • La Confederation Generale des Employeurs de Mauritanie was created in 1960 by private businesses as a support organization for formal businesses. Recognizing the need to provide support for microenterprise development, it created a special small enterprise unit under the UNIDO projects described above. * Le Secretariat d'Etat pour la Condition Feminine operates twenty women's promotion centers throughout the country. These centers provide literacy training, courses in handicrafts, and basic business training such as bookkeeping and management. They reportedly could benefit from staff training and from assistance in developing suitable courses and material. Mauritania 175 * Le Centre de Formation et de Perfectionnement Professionnel was established in 1978 by the government, with the assistance of the World Bank under the First Education Project, as a vocational skills training and upgrading center for workers in the industrial sector. Equipped to provide training in skills such as automobile mechanics, carpentry, plumbing, and sewing, the center has suffered from poor management and lack of funds, and is to be restructured under the World Bank Technical Education and Vocational Training Support Project (approved in 1993 and effective in 1994). Prospects Urbanization and the consequent growth of the informal sector have changed Mauritanian society. In the urban centers, rapidly growing numbers of poor people find ways to support themselves under difficult conditions with limited support. If the growing needs of Mauritania's urban poor are to be addressed, it is vital to address the needs of the informal sector. An important issue is the dearth of microfinance institutions that serve the needs of the informal sector. In this context, it is recommended that the Bank conduct a survey of informal financial systems to determine how financial services for microenterprises can be expanded. A second recom- mendation is for the Bank to disseminate state-of-the-art information regarding the principles and characteristics of effective delivery of microfinance services. Notes This chapter incorporates the findings of a field visit in July 1994. 1. These two poverty lines refer to the expenditure-based poverty lines adopted by World Bank 1990b. See also World Bank 1994c, p. 7. 2. However, Mauritania has made great strides in this area in the past twenty-five to thirty years. In 1965 only 13 percent of school-age children in Mauritania attended primary school. 3. According to the World Bank poverty assessment, the urban popula- tion growth rate is 7.8 percent (among the highest in the world), while the population growth rate is 2.9 percent. World Bank 1994c, p. 2. 4. In 1989 tensions between Mauritania and Senegal over land on the riverbanks between the two countries erupted into violence. Mauritanians residing in Senegal returned home, as did Senegalese residing in Mauritania. The departure of Senegalese from Mauritania left a large skills gap in the Mauritanian labor force. The Senegalese were plumbers, electricians, mechanics, and carpenters. The repatriated Mauritanians were mainly traders. The influx of Mauritanians increased the population by 10 percent 176 Country Studies in a one-month period, critically affecting health and employment in urban centers. World Bank 1994c, p. 17. 5. The population of Nouakchott increased from 4,800 inhabitants in 1959 to 378,000 in 1992, according to World Bank 1994c. The influx of Maurita- nians as a result of the 1989 disturbances between Mauritania and Senegal also exacerbated the urban population explosion and the growth of slums. 6. In 1960 the Direction de la Cooperative et de la Mutualit6 (the ministry in charge of cooperatives) was created to provide the legal framework for promoting the development of cooperatives and regional structures. The first law on cooperatives was created in 1963 and amended in 1967. 7. The exchange rate used in this chapter is 125 ouguiya to 1 U.S. dollar. 14 Niger Pierre-Olivier Colleye Niger is a landlocked country of 1.27 million square kilometers in West Africa. Rainfall is limited and often irregular, and soil fertility is declining due to intensive use. The population numbers roughly 7.9 million, almost half of whom are less than fifteen years old. Rural to urban migration is increasing and placing mounting pressure on limited urban resources. The urban population has more than doubled in the past two decades to reach its current level of 15.4 percent. Nearly 90 percent of Niger's population is concentrated along the southern border in the 12 percent of land considered arable. Niger's social indicators reflect a poor quality of life. The life expectancy of forty-five years (forty-eight for women) is one of the lowest on the continent. The infant mortality rate is high (126 per 1,000 live births) and accounts for 60 percent of all deaths. Niger also ranks toward the bottom of the list in the provision of education for its citizens. The adult illiteracy rate is 72 percent, less than 30 percent of school-age children are enrolled in primary school, and less than 7 percent are in secondary school. Subsistence agriculture dominates the economy, with millet and sorghum accounting for 80 percent of the cultivated area. Livestock also provides an important source of income and export receipts. Industrial development is handicapped by the limited size of the domestic market, the landlocked position of the country, the lack of raw material resources, and eroding competitiveness in the trade of goods, especially compared to Nigeria. With a per capita GDP of approximately $290, Niger is one of the least developed countries in the world. Despite almost nonexistent inflation rates in recent years (0-1 percent), economic growth has been lackluster due partly to a collapse in uranium markets. After a 1.9 percent contraction in 1990, GDP growth reached 1.4 percent in 1991 and 3 percent in 1992 due to small booms in agriculture. Niger's economy is relatively open, with imports and exports approach- ing 35 percent of GDP. It has strong trade links with Europe, mainly France, and neighboring African countries, particularly Nigeria. Once an exporter of processed raw materials like meat and tanned leather and hides, Niger increasingly exports unprocessed raw materials. The uranium export boom of the late 1970s and early 1980s financed government expenditures. The ensuing uranium bust led the government 177 178 Country/ Studies to launch stabilization efforts in 1983 and a structural adjustment program in 1985. Various debt reduction initiatives throughout the late 1980s decreased Niger's debt service ratio significantly. Still, a continuing fiscal imbalance in the early 1990s (equivalent to more than 11 percent of GDP in 1992) reduced the imports of capital goods and continued to threaten investments, which were compressed to below the replacement rate of existing assets. Tax revenues declined 20 percent in 1991 and 6.5 percent in 1992. Although the government has sought to commit itself to a meaningful structural adjustment program by privatizing parastatal enterprises and improving the efficiency of public infrastructure investments, political turmoil has slowed implementation of all of these reforms. The adjustment program was halted in 1991 when a national political conference prohibited downsizing the wage bill and the civil service, and outlined a set of irreconcilable policies. Main characteristics of the informal sector The size of the informal sector in Niger is in dispute. Official estimates have placed its contribution to GDP at approximately 20 percent, with an annual growth rate of 2.5 percent. The 1988 census includes 130,000 microenterprises, which employ some 240,000 people, 38 percent of them located in urban areas. According to a usAID-commissioned survey of the informal sector in Niger conducted by Michigan State University and the GEMINI Project, the informal sector is significantly larger (GEMINI 1990). The survey collected basic information from small nonfarm enterprises in selected locations in the Maradi and Dosso departments (political subdivi- sions of the Nigerien territory). According to these data, the two depart- ments alone accounted for 237,875 nonfarm microenterprises, with a labor force of 358,814 individuals (proprietors, working family members, and hired workers). Other reports estimate that the informal sector contributes 66 percent of GDP (World Bank 1989, p. 55). All estimates indicate that the informal sector plays an integral role in Niger's economy. It is also apparent that the sector is growing rapidly, primarily because people continue to leave rural areas due to drought, and adjustment programs and layoffs in the cities have forced many Nigeriens to take up informal activities as a means of employment. Most informal activities take place within the manufacturing, service, and trade sectors. Among manufacturers, the most common activities are tailoring, wood and metalworking, printing, and jewelry and shoemaking. Most informal services revolve around restaurants, bars, radio and electronic repair, automobile repair, and milling. Inforrnal sector trade consists primarily of textiles, garments, jewelry, furniture, food, and other consumer goods. The geographic division of informal sector activity is highly specialized. For example, in rural Dosso, mat weaving accounts for Niger 179 50 percent of total informal sector employment. In other locations, manufac- turing provides 30 to 40 percent of all microenterprise employment. As is the case in most Sahel countries, women play an important role in the informal sector. Table 14.1 describes the most common activities undertaken by Nigerien women. A 1987 study of women-owned micro and small enterprises identified 37,890 such enterprises, the vast majority of which-32,330-were in rural areas. These microenterprises created employment for between 45,000 and 50,000 women, and they provided a variety of services to local residents, including the house-to-house sale of small quantities of goods, the manufacture and sale of mats, and the preparation and sale of spices. Women also provide short-term credit to their customers (Bayard-Gamatie 1993). Table 14.1 Primary activities of women in the informal sector in Niger Sector Activity Processing Oil, fritters, prepared dishes, spices, soap, dried tomatoes and onions, rice husking, cheese, butter, sour milk, and spun cotton Marketing All of the transformed products above, gathered products (palm leaves, karite, fruits, and leaves), kola nuts, tobacco, cereals, vegetables, blankets, cloth, and perfume Handicrafts Pottery, mats, and leather work Services Village midwives, hairdressers (braid making), women praise- singers, healers, water sellers, seasonal workers for harvesting cotton and tomatoes in the Agadez region Source: IRAM 1976, pp. 47-48. Seasonal migration among men is a common occurrence in Niger, and, as a result, the number of temporary female-headed households is quite high. In some villages, virtually all adult men leave to work in neighboring countries during the dry season. Women's income then becomes the sole income for feeding the family (World Bank 1993e, p. 9). Main constraints on the informal sector The Michigan State University/GEMINI survey of the Dosso and Maradi departments identifies some of the problems microentrepreneurs face at the time of start-up and subsequently as their firms grow and mature. A lack of access to credit and a shortage of funds appear to be most pressing during a microenterprise's start-up phase. Another common constraint is the lack of product innovation. Outdated technology and the poor quality of machinery and tools tend to affect microenterprises most severely while their firms are growing. Table 14.2 summarizes the most severe impedi- ments stifling the growth of Niger's informal sector. 180 Country Studies Table 14.2 Constraints on microenterprises at various stages in Niger (percentage of the firms interviewed) Problem area At start-up During growth Currently Shortage of funds 31.1 23.3 22.3 Regulations 15.5 19.4 10.7 Availability of 12.6 25.2 11.7 machines and tools Technical 12.6 2.9 2.9 knowledge Markets 11.7 11.7 35.9 All others 16.5 17.5 16.5 Total 100.0 100.0 100.0 Source: GfEMINI 1990. Female microentrepreneurs face additional constraints in Niger. Social pressures, illiteracy, and inadequate access to credit hinder women's advancement in the informal sector. Women's situations vary according to ethnicity, class, religious affiliation, and production systems, but few have access to productive resources (secure land tenure, capital, technology, and credit), education, information, and training (U.S. Department of State Dispatch 1994b). Women's advancement is perhaps more severely constrained in the formal than in the informal sector, although since the last election, five women have been named to positions at the level of a minister or state secretary. No legal barriers are created by the civil code, but the overall regulatory environment for small- and medium-scale enterprises is particularly inhospitable for women. The labor code carries over from the colonial period a number of provisions governing the conditions of work, and protective regulations tend to limit women's access to the full range of employment. In part as a result, most women turn to the informal sector as their primary provider of employment. Policies and programs for microenterprise development Niger has six banks. The formal financial sector has adopted a very conservative lending posture. In general, banks refuse to finance the informal sector for the following reasons: * Microenterprises seldom have bank accounts. * Microenterprises are frequently located in rural areas far from a bank branch. Niger 181 * Banks are afraid that loans will not be used for their stated purpose, and the lack of bookkeeping does not allow banks to follow a borrower's performance. * Information about the microenterprise borrower is sparse. * The borrower has little, if any, collateral. * Many microenterprises do not operate from fixed locations. In rural areas, tontines or rotational credit and savings schemes are popular. Reports vary as regards the extent to which individuals, and particularly women, make use of tontines. One reliable report estimates that in urban areas, 75 percent of all women participate in tontines, and in rural areas, about 50 percent of all women participate. A number of programs provide savings and credit to the informal sector in Niger. The most successful programs are described in the following paragraphs. In October 1988, the Ministry of Commerce, Industry, and Artisanry set up the Direction de la Promotion de l'Artisanat. Its first step was to define, with the help of the ILO, a series of policies favorable to micro and small enterprises. Subsequently, in 1991, the ILO initiated a project called Politique Nationale de D6veloppement de l'Artisanat et de Promotion de la Micro et Petite Entreprise (PNDA) aimed at providing an appropriate legal framework and a mechanism for promoting the informal sector. The PNDA is now entering a two-year implementation stage with two main programs: * A program to improve the informal sector environment (financial and tax) and to provide services to entrepreneurs (information and statistics, training and apprenticeships, and market development). - A program to create, through assistance with feasibility studies, close to 7,000 informal sector enterprises: approximately 3,500 in food processing, 450 in mining, 230 in metalworking, 430 in woodworking, 80 in leather, 360 in textile and garment making, 520 in chemical products, 210 in construction and public works, 560 in services, and 10 in transportation. These enterprises are expected to generate more than 30,000 jobs. The PNDA also has planned an expanded five-year program to create an additional 4,000 informal sector enterprises and 23,000 new jobs. The project finances feasibility studies for each type of business activity. These studies are scheduled to take place between 1992 and 1997, but no results have been published thus far. The project aims to facilitate the access of microenterprises to financial services, but it does not address the actual financing of these enterprises. The Maradi Microenterprise Development Project forms part of USAID'S Small Enterprise Activities Development Project. Sponsored by CARE/Niger, 182 Country Studies it began with a trial phase in which loans were made and training was offered to 813 microenterprises between July 1988 and June 1991 (Grant 1993). No savings are involved in this project. It has since expanded its credit program in phase 2, while narrowing its training program to focus on specific vocational needs that will increase productive capacity. Phase 2 will run through 1996.1 Phase 1 had a budget of $2.4 million, supplied entirely by USAID. Phase 2's current budget stands at $7.6 million, $2.9 million of which is capital for loans. This program is distinguished by its delegation of responsibility, focus on profitability, minimizing of bureaucratic procedures, centralized financial controls, and systematic collection of repayments. The top priority is to transform the project into a viable, self-sustaining independent institution. The project makes individual and group loans for any legal purpose (20 and 80 percent, respectively). Traditional loan analysis is minimal, with the agent's personal assessment of the client and the client's contractual pledge to repay the loan taking the place of titled, physical collateral. The agents, with budgets of CFAF 5 million a month, determine the amount and term of loans in consultation with their clients. The 18 percent interest rate charged to borrowers covers the credit program's operating costs plus a provision for losses. The cost of capital itself is entirely subsidized by the donors. By March 1992, 1,716 new enterprises and 2,344 new jobs had been created, and 4,548 loans had been disbursed. The repayment rate reportedly is around 95 percent. The technical training component consists of technical (vocational) seminars lasting four to eight months. These courses, which are open to everyone regardless of educational background and experience, are given on subjects such as automobile mechanics, tailoring, radio repair, mill repair, wood and metalworking, and electrical and construction wiring. The choice of subject usually depends on the needs in each town, and a fee of CFAF 8,000 to CFAF 12,000 is charged for the courses. After six years of operation, fees recover 35 percent of the costs of the training program. With greater efficiency and simplification, this number should increase. CARE/Niger noted that when it raised the price of the courses, interest and enrollment rose. It also noted that paying for courses induced students to participate more actively and to give better feedback for enhancing the curriculum. Completion of the training program does not automatically entitle partici- pants to receive loans. The credit component of this project currently depends on external funding for 100 percent of its pool of credit. Without a local source of funding (savings deposits, government of Niger subscription of capital, or sale of shares), the project will never become an independent, self-sustaining institution. As William Grant (1993) concludes, "Unlike many other countries, Niger currently has a zero percent or even negative inflation rate. Niger 183 This rate helps preserve the capital in the loan fund but could be reversed should the CFA [franc] depreciate." Caisses d'Epargne et de Cr6dit, an $8 million project financed by USAID, is executed by the World Council of Credit Unions. It targets poor, rural Nigeriens living in the region around Zinder with a locally managed and funded rural savings and credit system. This is a classic savings and credit union program adapted to the local environment in Niger. The project was started in 1989, and it took roughly a year to get the first credit unions up and running. By June 1991, five credit unions were established. After sixteen months of operation, these five unions had about 550 members and about CFAF 2.5 million in savings and share capital, plus another CFAF 0.5 million in earnings. Initial activities included training managers in the basics of credit, setting up adequate record keeping systems, and conducting literacy training. The project has now entered a second stage (1992-97) in which its operations have spread to urban areas, including Niamey. As of March 1994, the number of credit unions had increased to twenty- four, representing a membership of 2,844 Nigeriens. In the first quarter of 1994, 229 loans were made for a total of CFAF 13 million, or 168 percent of the total amount of loans made for all of 1993. Repayment rates are encourag- ing. Loan delinquency reports show that 8.14 percent of the loans are late by one to two months, 0.4 percent by two to six months, and less than 0.5 percent by six to twelve months. No loans are delinquent beyond twelve months (World Council of Credit Unions 1994). The project is now trying to increase its loans-to-assets ratio without increasing its delinquency rate and, thereby, to increase the share of operational costs covered by loan fees and interest. Prospects The government has been receptive to promotion of the informal sector. The Nigerien Ministry of Commerce expects the informal sector to expand and reach at least 25 percent of GDP by the end of the century. The number of enterprises is growing relative to GDP across all sectors, but performance varies from sector to sector. The number of persons involved in mat weaving continues to grow, despite extremely stagnant markets. Various types of metal and leather products are enjoying growing markets and increasing returns. Evidence suggests that informal sector development projects should include basic information on markets and should particularly communicate to microentrepreneurs the importance of taking consumer preferences into account. Further recommendations must be withheld until an adequate field assessment of the current characteristics, constraints, and needs of microenterprises in Niger can be carried out. 184 Country Studies Note 1. Initially, the project also intended to provide management training and to introduce new products and technologies, but it met only a weak response and was eventually discontinued (because participants were not willing to pay for management training, and customers for appropriate technology products never materialized). 15 Sao Tome and Principe Peter Fidler and Sarah Sipkins Situated in the Gulf of Guinea, about 300 kilometers off the coast of Gabon, the Democratic Republic of Sdo Tome and Principe consists of two islands: Sdo Tome with an area of 859 square kilometers and Principe with some 109 square kilometers. According to a 1981 census, the population totals approximately 125,000 inhabitants, 120,000 residing on the island of Sao Tome and 5,000 residing on Principe. The population is growing at an estimated rate of 2.4 percent a year. During its first decade of independence, Sao Tom6 and Principe converted its colonial-based economy into a centrally planned command system, with disastrous results.' Immediately after gaining independence from Portugal in 1975, Sdo Tome and Principe moved to central planning associated primarily with three commodities: cocoa, coconut, and palm oil. Poor planning, lack of technical expertise, significant Portuguese divestiture, and nationalization of the cocoa estates brought a substantial decline in export earnings and precipitated an economic crisis with large fiscal deficits, a huge external debt, and monetary instability.2 In part as a result of these events, Sao Tome and Principe has experienced a significant decline in per capita GNP over the last two decades. Twenty years ago, per capita GNP was $440, higher than in any country in the Sahel region. By 1992, it had dropped to $360, an average annual decline of 3.3 percent. The average growth in real GNP between 1980 and 1990 was 0.1 percent a year. The performance of the economy is even more disappointing because the island nation received some of the most development assistance per capita in the 1980s. In recent years, Sdo Tome and Principe has sought to reverse this tide of negative growth, in part by implementing a structural adjustment program designed to liberalize the economy and to decrease the role of the government in economic activity. In spite of structural gains, Sao Tome and Principe has found it difficult to maintain macroeconomic discipline, due partly to the low capacity of its public administration. The formal economy is largely dependent on agriculture, which accounts for 95 percent of total export earnings. Cocoa exports account for nearly 80 percent of this amount.3 Agricultural earnings comprise the bulk of income derived from exports, but the sector occupies a much smaller role in the country's GDP. In 1992 agriculture comprised only 30 percent of GDP, while 185 186 Country Studies services accounted for nearly 60 percent, reflecting the vital role that informal service providers play in the economy. In contrast to its macroeconomic performance, Sao Tome and Principe receives relatively high marks for its basic social indicators (World Bank 1994h). Life expectancy is among the highest in Africa (sixty-eight years), the child malnutrition rate (17 percent) is much lower than in most Sahelian countries, the infant mortality rate is relatively low (65 per 1,000 births), and the adult illiteracy rate (43 percent) is far lower than in neighboring countries. Sao Tome and Principe serves as a model for the rest of Africa in the provision of education for its citizens. It has attained very high rates of enrollment in primary education, and around two-thirds of all eligible students attend secondary schools.4 Despite these comparatively high standards of living, social indicators have declined moderately over the past decade. Recently, there has been a surge in malaria and endemic diseases, and a shortage of trained teachers and health workers has developed. Today, more than half of the population do not have access to safe water (World Bank 1993b, p. 3). In 1987 Sao Tome and Principe initiated a series of economic policy reforms within a structural adjustment program. Assisted by the World Bank and the African Development Bank, the program was aimed primarily at liberalizing price and trade systems, privatizing the economy, diversify- ing agricultural output, and improving the efficiency of cocoa production.5 Currency devaluation, price liberalization, privatization of state-owned farms, and reduction of budget deficits became key policy elements of a second structural adjustment program. In addition to focusing on macroeco- nomic stability, the government pursued an import-substitution and export- led growth strategy that emphasized market forces. Modest increases in cocoa production were evident, although a continuing decline in global prices reduced net revenues. By 1993 annual inflation had fallen to 20 percent from 36 percent in 1991. In December of 1994, the government unified the official and free market exchange rates. As a result, inflation dropped to nearly zero during the first two months of 1995. One possible spin-off that has accompanied economic adjustment has been a slight rise in unemployment. Total formal sector employment fell from 15,140 in 1986 to 13,970 in 1990 (in the context of an annual population growth rate of 2.6 percent). A similar shift took place in the agricultural sector, where the number of workers fell from 15,072 in 1981 to only 13,000 in 1990. This drop reflects both a transition from publicly funded agricul- tural production to subsistence farming and migration from rural to urban centers. It is possible that, in the coming years, the adjustment process will cause the informal sector to become more important as a provider of employment. In Sub-Saharan Africa, the informal sector has been growing quite rapidly over the past decade to the point where informal sector activities now Sao Tome and Principe 187 account for more than 70 percent of nonagricultural employment. Sao Tome and Principe is no exception. The balance of this chapter describes the role played by the informal sector in the economy, the kinds of constraints that impede its development, and the programs that are addressing its needs. Main characteristics of the informal sector With 19,000 persons working in the public sector and only a few dozen private enterprises in operation, the informal sector serves as a major source of employment. The degree of its importance, however, is a point of some contention. According to the Ministry of Health, Labor, and Social Security, 4,000 persons or 12.5 percent of the economically active population are employed in the informal sector (Economist Intelligence Unit 1994a, p. 43). Unofficial estimates, however, have placed the share of informal sector employment much higher. In a 1993 study, a team of World Bank observers estimated that the formal sector employed only 23 percent of the labor force (World Bank 1993b, p. 10). Given an unemployment rate in the neighbor- hood of 27 percent, the informal sector employs close to 50 percent of the labor force (Peck 1994, p. 1; precise data on unemployment are almost impossible to obtain). This figure is expected to increase, because structural adjustment could require the government to lay off 2,000 civil servants and an additional 2,000 workers employed in the public agricultural sector. A 1991 survey by the ILO of 564 nonagricultural micro and small enterprises found that 59 percent of the businesses were located in the homes of the owners, 38 percent had fixed locations outside their homes, and 2 percent were mobile (ILO 1991). The average enterprise had 3.5 workers per establishment, with 51 percent of the enterprises having one employee or fewer. The survey also discovered that 35 percent of the establishments did not hire any outside labor and that manufacturing enterprises were more likely than service enterprises to hire additional workers. Of the workers employed in these enterprises, only 47 percent were permanent employees. Of the remaining workers, 11 percent were temporary workers, 15 percent were family members, and 28 percent were unpaid apprentices. The survey also revealed a very low degree of integration between the informal sector and the rest of the economy. Roughly 95 percent of enterprises sell their products directly to individuals rather than to larger businesses. Microenterprises also tend to be labor-intensive, with very little investment in machinery or equipment. Characteristics of some of the most common forms of microenterprises are given in table 15.1. Because the informal sector employs such a large percentage of the working population, the range of activities is broad. In urban areas, a snapshot of the informal sector reveals carpenters working with hand tools, welders on side streets fabricating stoves with ovens made from diesel drums, and tailors sewing shirts out of scant material. In rural villages, the 188 Country Studies Table 15.1 Characteristics of microenterprises in Sao Tome and Principe, 1991 Percentage of Average nuimber Percentage of Type of enterprise female-ownedfirms of workers in-housefirms Restaurants and bars 40.0 5.5 58 Hairdressers 30.0 2.3 100 Butchers 26.5 3.1 60 Clothesmakers 22.5 4.9 57 Shoemakers 15.0 4.3 70 Carpentry and cabi- 14.0 3.1 54 netmakers Basket makers 10.0 3.2 68 Distilleries 5.0 6.0 48 Source: ILO 1991. portrait is very different: 90 percent of Sao Tome and Principe's total land area was once under the control of state-owned agricultural estates, but privatization is slowly producing small-scale farmers and farming coopera- tives. Here one observes fish processing, handicraft making, goat raising, food processing, wood carving, and bamboo working. Almost two-thirds of the 564 nonagricultural, informal sector enterprises included in the ILO survey were cottage production industries. Within this group, clothing (21 percent) and carpentry and cabinetmaking (17.5 percent) accounted for the largest percentage of enterprises; woodworking and wood engraving (5.1 percent), distilling (3.5 percent), and basketmaking (3.4 percent) were the next most common activities. An additional 18 percent of enterprises surveyed were involved in various services, of which shoemak- ing (6.6 percent), automobile and bicycle repair and general mechanic work (5.3 percent), and hairdressing (2.4 percent) were the most prevalent. Another 17 percent of enterprises were engaged in mini-restaurants, bars, and hotel services, and the balance contracted themselves out to the construction and public works sector. No regulations in Sao Tome and Principe restrict female economic activity, and the informal sector is open to many forms of female entrepre- neurship. The ILO survey found that 20 percent of business owners were women. Although women comprise a significant part of informal sector employment and outnumber men in a number of sectors, women generally are relegated to the smallest and least profitable activities. The ILO survey found that women accounted for more than 50 percent of all unpaid apprenticeship positions, but they comprised only 6.5 percent of permanent salaried employees. Sao Tome and Principe 189 Main constraints on microenterprise development A number of factors inhibit the growth of microenterprises in Sao Tome and Principe. Lack of access to adequate financial services, a high dependence on imports, lack of indigenous natural resources, the physical remoteness of the islands, and an insufficient supply of energy all negatively affect informal sector production. The most severe impediments are described below. Lack of access to credit is a major constraint to informal sector develop- ment in Sao Tome and Principe. A commonly cited reason for the general credit shortage is IMF-imposed credit ceilings on bank lending. More realistically, banks typically are not interested in lending to informal sector enterprises because of the small size of the transactions, lack of available collateral, inexperience of potential borrowers in dealing with lending institutions,6 and the general lack of predictability surrounding the transaction. As a result, informal sector enterprises are forced to turn to alternative sources of credit. Savings and loan cooperatives have been one such source of credit for microentrepreneurs. Article 3 of the Sao Tome and Principe financial institutions law includes cooperatives in its definition of "financial institutions" and leaves the door open for savings and loan institutions to participate in the system (World Bank 1993b, pp. 27-28). In the last couple of years, cooperatives have entered into the microfinance arena. More than fifty cooperatives-several of which are savings and loan associations-have been established. Informal savings and credit clubs are relatively undeveloped in Sao Tome and Principe, especially compared with the elaborate networks that exist in some Sahelian countries. As a result, most informal sector enterprises are financed through personal and family savings. The 1991 ILO survey found that of the 564 enterprises surveyed approximately 94 percent were self- financed or were financed with the aid of family members during their initial start-up stage. Although personal and family savings are frequently sufficient to get firms up and running, working capital tends to run dry rather quickly. The dearth of sources of short-term loans means that many informal sector enterprises have little prospect of growth beyond their start- up. The legacy of a twenty-year command economy, combined with geographic isolation, has limited the flow of information about domestic and foreign (European and African) markets. Microentrepreneurs could benefit from basic business information about market supply and demand, but processed information of this kind is generally lacking. While informal operators have saturated certain markets, other activities such as the production of bamboo furniture, handicrafts, wood carving, flower exports, and fruit exports might expand if the right information were available. 190 Country Studies The geographic isolation and tiny domestic market of Sao Tome and Principe mean that the economy must rely on the efficient transport of goods abroad. However, neglect of road rehabilitation in recent years has negatively affected the mobility of goods and people. In many cases, erosion has damaged roads to such an extent that vehicles can no longer pass (World Bank 1993b, pp. 57-58). Further exacerbating the transport difficulties, the main port at Ana Chaves has no usable berthing facilities for oceangoing vessels. Vessels must anchor in deep waters outside the harbor to load and unload their cargo onto intermediary ships for transport to the port. Goods are frequently lost or damaged, which increases the cost of transport. Many shipping lines no longer call at Sdo Tome and Principe, thus isolating the island country even further and raising the cost of inputs for microentrepreneurs and others. With a population of 125,000 people, Sao Tome and Principe is one of the smallest countries in the world. The tiny size of the domestic market forces producers to search for buyers abroad. This need is compounded by the weak purchasing power of the population. It is somewhat surprising, therefore, to observe that the economy has not been adequately integrated into the global trading system. For the period 1980-90, on average, exports amounted to 25 percent of GDP, and imports amounted to approximately 60 percent (World Bank 1993b, p. 10). For comparable island economies, such as Fiji, Kiribati, the Solomon Islands, Tonga, Vanuatu, and Western Samoa, exports averaged 55 percent of GDP, and imports averaged 67 percent. Furthermore, Sao Tome and Principe's average annual rate of export growth was 6 percent in this period. At first glance, it might appear as though the underdevelopment of export trade would have little impact on the welfare of informal sector operators who are solely concerned with local, domestic markets. However, this underdevelopment intensifies market saturation, as both formal and informal enterprises are forced to battle for control of a domestic market whose purchasing power is already quite thin. Programs for microenterprise development Given the physical remoteness and small size of the population in Sao Tome and Principe, a surprising number of international organizations have programs to promote informal sector development on the islands.7 The most effective program is the Cooperative League of the United States of America (CLUSA). The following paragraphs describe that program in detail and highlight a few other programs that either are at the planning stage or are just getting under way to promote microenterprises in Sao Tome and Principe. In 1989 USAID authorized $1.5 million for the CLUSA program to increase rural, private sector productivity and to reduce credit constraints encoun- tered by microentrepreneurs. The organization has sought to accomplish Sao Tome and Principe 191 these goals by promoting the establishment of a national cooperative movement. The program started by assisting six pilot cooperatives formed to facilitate the attainment of land titles. By the end of 1993, the program had expanded its network to assist fifty cooperatives in the areas of small business management and training. Thirty of these cooperatives are agricultural associations, fifteen are fishing village cooperatives, and four are women's trader groups. Recently, the success of the rural cooperative program (1,764 members or roughly 1.5 percent of the population) has led CLUSA to assist in the creation of urban cooperatives as well. CLUSA operates a loan guarantee fund, primarily through Caixa Nacional de Paupance e Credito, but also through a second financial institution that specializes in the delivery of agricultural credit, Mesquita Center. The fund is earmarked for cooperatives wishing to borrow from the formal financial sector, and it provides a guarantee (for a fee) up to 20 percent of the amount borrowed by the group. As of October 1993, CLUSA had guaranteed twenty- two loans and distributed more than $53,000 to fifteen cooperatives (CLUSA 1993). According to reports, the lack of village tradition and social structure in most of the farm associations has made lending activities a bit risky. Only 72 percent of all Caixa Nacional loans and 87 percent of Mesquita Center loans have been repaid.8 CLUSA'S technical assistance to microentrepreneurs is an equally vital component of the project. Cooperative members receive training in literacy and numeracy as well as assistance in management, finance, accounting, and property law. CLUSA has assisted with implementation of an agricultural cooperative export strategy, designed a rural development public education campaign, and been instrumental in drafting new national cooperative legislation. CLUSA will take on an even greater role in the future as the government implements its land reform program. The risks associated with managing one's own rural microenterprise without technical services, appropriate information, and adequate financing are great. The government recognizes this fact and has joined the International Fund for Agricultural Development (IFAD) in calling on CLUSA to create sixty-four new cooperative associations over the next six years. In 1985 IFAD provided a loan of Standard Drawing Rights (SDR) 2.1 million to supply fishing equipment, updated technology, repair facilities, and technical assistance to small-scale fishermen. Various NGOS have served as distributors for outboard engines, equipment, and gear. The program doubled the income of more than 1,500 fishermen (1.4 percent of the population). Phase 2 began in 1991 with a $1.7 million loan from IFAD. The objective is to draw from current success and create a series of viable fishing associations. Men dominate fishing, but women also benefit from the program, because they frequently are engaged as fish traders and processors and as operators of transport vehicles. 192 Country Studies Lutte Contre la Pauvrete par le D6veloppement Rural was established in January 1993 by the UNDP in an attempt to improve the rural business environment. Recognizing the key role of microenterprise development, this program supports micro and small enterprises in agriculture, agroprocessing, and transportation. The program focuses exclusively on the provision of technical assistance and currently does not have a credit or savings component. Central features of the program include analysis of market supply and demand in rural villages to identify new markets, assistance in product diversification, introduction of training programs in promotion institutions, and formula- tion of future projects that could help in promoting micro and small businesses. In one year, the project has assisted in the creation of twenty micro and small enterprises. Most of the businesses are located in the Agua Ize Zone, and more than half (60 percent) are managed by women. In conjunction with the World Bank Land Distribution Project and with the assistance of IFAD and UNDP, the Caisse Francaise de Developpement has begun a preliminary appraisal of a small rural credit scheme. No project has been implemented to date, but a modest plan may be implemented in 1995. The Women's Organization Fund has a program to promote microenterprises in the assessment stage by identifying cottage industries and commercial activities operated by viable clients of small and micro loans. Initial plans indicate that primary recipients would be engaged in agriculture and food processing. A United Nations Capital Development Fund project soon will be implemented to provide credit to a small enterprise project in Angolares. It also plans to implement a $2.3 million project in 1995 for infrastructure rehabilitation and microenterprise promotion. Approximately $190,000 is expected to be allocated for a small rural credit scheme. Prospects To the extent that firms in Sao Tom6 and Principe rely on domestic markets, most will be forced to remain small. In such an environment, one would expect informal operators to possess a comparative advantage: microentrepreneurs can use their flexibility to their advantage, while firms in the formal sector fail to benefit from economies of scale. But the reality of the informal sector looks quite different. Lack of access to working capital and, perhaps more important, to updated technology and market informa- tion has led to saturated markets. The informal sector occupies a central role in the economy of Sao Tome and Principe. With the impending restructuring of the public sector and the agricultural sector, this role will increase. To ignore the importance of microenterprises is to ignore the reality. A study of microenterprises and, specifically, of demand for credit should precede programming. The supply of credit that will come with programs now under preparation will need to Sdo Tome and Principe 193 be specified and compared with estimated demand. Further recommenda- tions must be withheld until an adequate field assessment of the characteris- tics, constraints, and needs of microentrepreneurs in Sao Tome and Principe can be carried out. Notes 1. In recent years radical measures have been taken to open up the economy with an overall market orientation. 2. Once independence was obtained, virtually all Europeans departed, and the country found itself starved for administrative know-how. Prior to 1975, senior administrative positions and technical posts of authority were occupied entirely by colonial officials. 3. Economist Intelligence Unit 1994a, p. 4. Export trade is relatively underdeveloped. As a result of the failure to diversify and cultivate export markets, Sao Tome and Principe ran a $21 million current account deficit (including grants) in 1992. 4. This figure is even more impressive because more than 47 percent of the population is under the age of fifteen. 5. Diversification away from the cocoa monoculture is a goal that has continuously eluded Sao Tome and Principe. Despite efforts to diversify, the economy still remains 80 percent reliant on cocoa production. Such dependence has led former President Pinto da Costa to conclude that, "We managed to free ourselves from Portuguese colonialism, but not from cocoa colonialism." 6. The one formal sector institution that could possibly play a functional role for microenterprises is Caixa Nacional de Paupance e Credito. Formerly called Caixa Popular, this savings and loan was established in 1981 to mobilize small-scale savings and to channel them into loans for housing construction and small business development. Recently there have been signs that it is planning to emphasize lending to small farmers and microenterprises, two groups previously ignored. 7. Most are engaged in technical assistance, and few focus on the delivery of microfinance services. 8. Of the five loans in arrears, almost all are agricultural input loans. Although various reasons have been presented for the arrears, Caixa Nacional and the program recognize that a better repayment rate could have been achieved if the former Caixa Popular had monitored loans more aggressively. With the strengthening of the Caixa Nacional, it is expected that better bank procedures and more adequate monitoring will be used to follow up the loans. Caixa Nacional is also planning to open three new branches to become more accessible. 16 Senegal Anne-Marie Chidzero Senegal is a small, semiarid, coastal country, which surrounds The Gambia, and is surrounded by The Gambia, Guinea, Guinea-Bissau, Mali, and Mauritania. In 1991 per capita GNP was $720, and the population totaled 7.6 million, with the majority living in rural areas (World Bank 1993k, p. 238). The average annual population growth rate between 1986 and 1992 was 3 percent, slightly lower than the average annual real GDP growth rate of 3.1 percent (World Bank 1993k, p. 434). Senegal's economic indicators resemble those of a middle-income country, but its social indicators reflect poor economic conditions. The average life expectancy is forty-eight years, the child malnutrition rate is 17.5 percent, the infant mortality rate is 68 per 1,000 live births, the illiteracy rate is 62 percent (75 percent among females), only 59 percent of school-age children attend primary school, and only 16 percent attend secondary school (World Bank 1994g, pp. 302-03). Agriculture employs roughly 80 percent of the workforce and accounts for 21 percent of GDP. Primary sector activities include crop production (10 percent of GDP), livestock (9 percent), fishing (2 percent), and forestry (1 percent). The secondary sector contributes 19 percent to GDP, with industry accounting for 13 percent, construction for 3 percent, energy for 2 percent, and oil mills for 1 percent. The tertiary sector accounts for 60 percent of GDP, consisting mainly of transport (10 percent), commerce (22 percent), and government and other services (28 percent).' Since its independence in 1960, Senegal's overall economic performance has been disappointing (World Bank 1993h, 1993k, 1994g). In 1966 it lost the preferential treatment accorded to its agricultural exports by the European Community. Between 1967 and 1974, real GDP growth slowed to an average rate of 1.3 percent a year compared with 3.5 percent between 1960 and 1966. In the mid-1970s, economic growth was led by strong international demand for phosphate and groundnuts. Severe droughts in 1979 and 1980, a fall in world prices for groundnuts, and a doubling of world prices for iron ore had a severe impact on the performance of the economy. During this period, real GDP growth rates dropped to 0.8 percent a year, and agricultural output fell by 40 percent. Since 1980 Senegal has implemented a series of adjustment programs with the assistance of the World Bank and the International Monetary Fund. The main goals have been to reduce the state's role in the economy, to 195 196 Country Studies stabilize internal and external financial balances and public savings, to liberalize trade, and to increase investment in productive sectors. Reforms were designed to tighten monetary and fiscal policies, to reduce import tariffs, to liberalize markets and prices, to reduce the size of the public sector, and to encourage private sector development. In 1986 a series of reforms were launched under the Nouvelle Politique Industrielle, which aimed to improve the competitiveness of Senegalese firms and to encourage the development of high value added activities by abolishing quantitative restrictions on imports, rationalizing the system of protection, removing price controls on manufactured goods, revising the investment code, and implementing a program to divest parastatal enterprises. Important sectoral reforms were also implemented in agricul- ture, banking, transport, and human resources. Results have been mostly positive. The budget deficit, which amounted to 8.8 percent of GDP in 1983, was eliminated by 1991. The current account deficit was reduced from 18.6 to 7.9 percent of GDP during the same period. The annual average real GDP grew by 3.2 percent a year between 1986 and 1991, up from 2.1 percent during the first half of the 1980s. However, 1993 estimates show a slump in economic performance, with average real GDP growth rates dropping to 1 percent a year between 1991 and 1993. Senegal is better off than it was in 1981, but it still faces the challenge of creating an environment for long-term growth. Main characteristics of the informal sector Informal enterprises dominate activities in the private sector. The private sector contributes around 85 percent of GDP, roughly 60 percent from the informal sector and 25 percent from the formal sector. Moreover, the informal sector employs more than 90 percent of the labor force (because most of the informal economy is not documented, these statistics are only estimates; Labat-Anderson 1990, p. 4). A 1988 census conducted by USAID of the urban informal sector in Dakar showed that about 30 percent of Dakar residents earn their living from informal sector activities (Lubell and Zarour 1990). This might indicate that the majority of informal sector activity is rural. The informal sector in Dakar is currently estimated to be growing at a rate of 3 percent a year, absorbing the majority of new entrants to the job market as well as those who have lost jobs in the formal sector. Unofficial sources estimate that at least 50 percent of employment in urban areas is now generated by the informal sector (Catholic Relief Services 1994, p. 2). Informal sector enterprises are traditionally owned by Africans, while most formal enterprises are run by foreign nationals. Owners of informal sector enterprises were between twenty-five and forty-four years old in 1992 (World Bank 1993i, p. 7). Informal sector enterprises are closely linked to the family: 54 percent of enterprises use business revenues for family expenses Senegal 197 (Muteta and Niang 1992, whose findings are based on interviews conducted in September 1992 with forty-two entrepreneurs operating in Dakar's informal sector). Similarly, family savings frequently provide start-up and working capital, and family members almost always constitute a significant proportion of an enterprise's labor. Most businesses included in the USAID census of Dakar employed from one to eight workers (Lubell and Zarour 1990). The study judged the maturity of an enterprise by the percentage of its revenues that is used for household expenses. Qualitative analysis of enterprises found evidence to support three stages of microenterprise development: * Stage one. Of the enterprises surveyed, 10 percent use 90 percent of their revenues to cover daily expenses of the household. They are barely able to sustain themselves in the absence of any investment. * Stage two. An enterprise graduates to stage two when net revenues exceed basic household expenditures. At this stage, owners have resources to invest in a house or a car or to reinvest in the enterprise; 78 percent of the microenterprises surveyed were in stage two. * Stage three. When enterprises reach stage three (12 percent of the microenterprises surveyed), revenues exceed household expenses by a large margin. Roughly half of revenues are used for the household, and the remainder can be invested in the enterprise for expansion and diversification. The informal sector dominates activities in commerce. In Dakar roughly 70 percent of all informal businesses are engaged in trade (vendors in the markets, in small shops and kiosks, and on the streets). The remaining 30 percent participate in light manufacturing (textiles, leather, woodworking, metal working, fumniture, and agroprocessing), building trades, services, and transport. Informal enterprises in Senegal are divided along gender lines. Men tend to be involved in manufacturing, such as woodworking and metal working, and some service activities. Women dominate hairdressing, dressmaking, food preparation, and trade. Women represent the majority of operators in the urban informal sector, but men monopolize the most stable informal sector jobs, perhaps because they have easier access to capital, trading contacts, and transportation (World Bank 1992, pp. 4-5). Men also tend to be more mobile, because societal norms dictate that a woman's primary responsibility is to care for her family. An estimated 3,600 women's groups are operating in Senegal. These self- help groups provide their members with financial and social support and training; they also perform a number of functions traditionally performed by the rural family. Membership ranges from 50 to 500 persons (World Bank 198 Country Studies 1992, p. 6). In 1987 the National Federation of Women's Groups was established to coordinate women's groups. Unlike in most other countries, there are substantial links between the informal and formal sectors in Senegal, at least in urban areas. The USAID survey found that 47 percent of all urban informal sector customers were private individuals, 20 percent were other small enterprises, 18 percent were government officials, 9 percent were small artisans, 3 percent were department stores and other large enterprises, and 3 percent were wholesal- ers. Surveyed enterprises purchased their inputs from both informal and formal sector enterprises, including waste products recovered from industrial sites. Main constraints on the informal sector Numerous informal sector surveys point to the following constraints as the most significant blocks to informal sector growth: * Lack of access to credit. Of the entrepreneurs interviewed by the USAID survey, 86 percent indicated that lack of access to credit was a major constraint. In the 1992 World Bank study, 90 percent of the enterprises surveyed found it difficult to obtain needed working capital. * Sociocultural pressures. Senegalese entrepreneurs have numerous family obligations that place constraints on the financing of businesses. Family and community, ethnic, and religious groups pressure entrepreneurs for financial contributions. These assets would otherwise be invested in growth of the business. * Lack of physical space. Finding physical space to start a business is a problem for microentrepreneurs. Most work in their homes, but many would prefer to move to more convenient sites. Policies and programs for microenterprise development Microenterprises in Senegal use a wide array of informal financial institu- tions, including associations, tontines, and moneylenders. Of the entrepre- neurs interviewed in a survey conducted by the World Bank and the Canadian International Development Agency of informal financial institutions, 40 percent were members of associations, 52 percent were members of tontines, 43 percent saved their money with other individuals, and 53 percent borrowed from other individuals.2 The key factor underlying the success of these systems is trust. Associations and tontines have low overhead costs and few formal rules or regulations. Guarantees are rarely used, and repayment rates are generally higher than 95 percent. Most associations in Senegal have economic as well as social functions. About 85 percent of all associations collect savings, and about 60 percent Senegal 199 provide credit. Savings per member average CFAF 3,515 ($7), and loans average CFAF 12,900 ($26).3 Extrapolating from a review of 540 associations to a national level would place aggregate savings held by associations at CFAF 5 billion-CFAF 8 billion ($10 million to $16 million). Two-thirds of this amount are kept as cash. The remainder is deposited with banks or microfinance programs. These associations have lent approximately CFAF 9 billion-cFAF 11 billion ($18 million to $22 million; World Bank 1993g, pp. 18-21). Tontines (which are called natt or, depending on the region, teck, piye, or pari) have a purely financial function. About half of survey respondents belong to a tontine (80 percent in urban areas and 35 percent in rural areas). The majority are women. Tontines provide a mechanism whereby the savings of one member can be leveraged by the savings of other members for individual use on a rotating basis. About 40 percent of tontine funds are used for production purposes, and the remainder is used for consumption. Individual deposits range from CFAF 500 ($1) to CFAF 50,000 ($100). Deposits are made daily, monthly, or bimonthly, depending on the tontine and the number of members. Extrapolating from these figures to the national level, the net amount transferred through tontines is estimated to be CFAF 65 billion annually ($125 million; World Bank 1993g, pp. 8-12, 22-23). Some tontines also operate loan funds. A fraction of the savings are deposited in a credit fund that can be used to make small and very short- term loans to members and nonmembers. Interest rates are generally set at about 25 percent a year. This practice expands the fund and provides additional revenue to members. In addition to savings and credit from associations and tontines, 18 percent of those interviewed had deposits with a guardian, 54 percent had lent money to other individuals, and another 53 percent had borrowed from a moneylender (World Bank 1993g, pp. 13-17). A recent survey conducted by ILO and Banque Centrale des Etats de l'Afrique de l'Ouest found thirty programs that engaged in savings and credit activities, but only three that reached a large number of borrowers and had made significant steps toward achieving financial sustainability. These include Credit Mutuel du Sen6gal, Catholic Relief Services-Small Enterprise Development Pilot Project, and Alliance du Credit et de l'Epargne pour la Production. Sponsored by the Caisse Francaise de Developpement, the Centre International du Credit Mutuel, and the Fonds d'Aide et de Cooperation, the Credit Mutuel du S6negal seeks to create a financially viable savings and credit institution for agriculture, livestock, housing, commerce, artisanry, and social needs based on the French mutualist system. Credit Mutuel was created in 1988 and has since developed an extensive network of branches. It is headquartered in Thies and has regional offices in Koalack, Tambacounda, and Ziguinchor. By the end of 1994, it had fifty-four branches 200 Country Studies and 22,228 members. This is a marked increase from the forty-one caisses and 13,114 members reported at the end of December 1993. The program's loan portfolio stood at CFAF 220 million ($440,000), and savings totaled CFAF 332 million ($664,000) in December 1993. The average loan size is CFAF 66,900 ($134). The interest rate charged on loans is 2 percent a month, and savings earn between 4.5 and 5.5 percent, depending on whether they are forced or voluntary. Although not yet financially sustainable, the program could cover the cost of operations by the year 2000, depending on the number of branches created and on the size of the membership. Credit Mutuel plans to expand the network to eighty-seven branches by 1998 (Caisse Francaise de Developpement 1991, p. 14). The Catholic Relief Services-Small Enterprise Development Pilot Program in Senegal is a rural microfinance program that uses the village banking methodology developed by the Foundation of International Community Assistance in Latin America to promote community-managed credit and savings associations established to improve members' access to credit, to build a community self-help group, and to help members accumulate savings. Catholic Relief Services works through local partners-the Ententes des Groupements Associes-who set up the network of village banks. The village banks provide small working capital loans in direct proportion to savings deposits. Catholic Relief Services has developed village banks with the Entente des Groupements Associes de Birkilane and the Entente de Groupement Associes Nganda. The first project was started in 1989, and by 1993 it had twenty-six banks and 1,300 members. Unfortunately, repayment problems arose in September 1994, and Catholic Relief Services has temporarily suspended its support. The second project was started in 1992, and by December 1994 it had twelve village banks and about 600 members. The average interest earned on loans is 54 percent a year, with an average loan size of CFAF 33,800 ($65). As of December 1994, the program had a loan portfolio of CFAF 32.5 million ($62,000). Alliance du Cr6dit et de l'Epargne pour la Production, formerly the Agence de Credit pour l'Entreprise Privee, is one of the most successful projects in Senegal in terms of credit delivery, repayment rates, number of borrowers reached, and potential for long-term sustainability (Grant 1993). The program was launched by USAID in 1986 with the objective of providing microcredit to enterprises in the Koalack region. It has since expanded to Dakar, Diourbel, Kolda, Louga, Tambacounda, Thies, and Ziguinchor, with headquarters in Dakar. By December 1993, the program had served more than 4,500 borrowers, had a loan repayment rate of 98 percent, and an average loan size of $1,060.4 The total portfolio in 1993 was CFAF 1,212 million ($2.4 million). On average, enterprises financed by the agency have assets of $5,000 and fewer than ten employees (Grant 1993). Operational costs are in the moderate to low range, accounting for roughly 20 percent of Senegal 201 the total loan portfolio (Christen, Rhyne, and Vogel 1994, p. 31). The agency uses a minimalist approach that provides very little training to entrepre- neurs and emphasizes establishing viable credit operations. Each branch operates as a profit center, and branch managers and credit officers receive an annual bonus based on the branch's income. The yearly interest rate on loans is 20 percent (equivalent to a 14 percent real rate; Christen, Rhyne, and Vogel 1994, p. 31). Collateral is required and can consist of a property deed, a pledge of contract, equipment collateral, or a pledge of working capital. A life insurance policy is also required for the amount of the loan, and the agency is named as the beneficiary. Other programs providing microfinance services in Senegal are the Conseil National pour la Promotion des Caisses d'Epargne et de Cr6dit, which has twelve credit unions in the Louga region and two in Dakar; La Caisse des Femmes du Grand Yoff, which is centered in Dakar and has 103 members; the Federation of Nongovernmental Organizations in Senegal, which provides financial services and technical assistance to rural enter- prises; and the Caisse Nationale de Credit Agricole du Senegal, an agricultural savings and credit bank that serves individuals, enterprises, cooperatives, and solidarity groups in agriculture and agroprocessing. These programs are small in their scope (Caisse des Femmes du Grand Yoff), too new to assess adequately (Conseil National Pour la Promotion des Caisses d'Epargne et de Credit), or facing financial difficulty (Federation of Nongovernmental Organizations in Senegal and Caisse Nationale de Credit Agricole du Senegal). One of the main policy issues concerning support to microenterprises is the absence of a regulatory framework for operating microfinance programs. Recognizing this need, the government, with the assistance of the World Bank, the Canadian International Development Agency, and the Societe de D6veloppement International Desjardins, created the Cellule d'Appui aux Operations Bancaires Mutualistes within the Ministry of Finance and Planning in 1990 (see annex 2). This unit developed an administrative, legal, and operational framework for the activities of savings and credit unions, village banks, and other NGO microfinance programs in Senegal. The result was an administrative structure within the Ministry of Finance set up to supervise and regulate specialized credit and savings institutions and to develop appropriate regulatory guidelines. In December 1993, as a result of this work, the Conseil des Ministres of the Union Economique et Monetaire de l'Ouest de l'Afrique adopted a regional law governing savings and credit unions: Le Projet Loi Portant Reglementation des Institutions Mutualistes ou Cooperatives d'Epargne et de Credit. Modalities for the application of the law are left to the individual member countries. This law was adopted by Senegal on December 17, 1994, and the decrees of application are still to be adopted. 202 Country Studies The law was created to govern the activities of institutions that have the profile of savings and credit unions. Other institutions can ask to be recognized under the law; otherwise they fall under the banking law. The law provides operational guidelines for savings and credit unions and defines eligibility for tax exemptions. What the law does not do, however, is to define the interest rate an institution can charge on loans. This rate is defined by the current usury law mandating that interest rates cannot exceed twice the Central Bank discount rate. To the extent that this usury law is applied, it could pose a problem for the long-term financial sustainability of many programs, because it would not allow them to charge interest rates that are sufficiently high to cover operational costs. Recommendations The World Bank could support Senegal's effort to develop its informal sector on two fronts: * By analyzing and discussing the Projet Loi Portant R6glementation des Institutions Mutualistes ou Cooperatives d'Epargne et de Cr6dit to ensure that microfinance institutions are constrained in pricing their loans as low as possible. * By expanding existing microfinance programs, possibly through an apex fund that could capitalize institutions and expand the geographic coverage of programs with a credible commitment to reaching opera- tional efficiency. Notes This chapter was updated following a field visit in January 1995. 1. These estimates are in 1987 constant prices and are based on govern- ment statistics. World Bank 1993h, annex table 7; 1994h, p. 303. 2. World Bank 1993g presents the findings of a study conducted on the practice of informal savings and credit in Senegal in fourteen subregions (arrondissements). The report analyzed 128 associations and interviewed 256 individuals. 3. World Bank 1993g, pp. 3-8. The exchange used is 500 CFA francs equal 1 dollar. Savings amounts are underestimated because they were calculated as amounts in the pool at the time of the census and exclude the amounts loaned out. Savings are also often in kind. 4. This figure is obtained by dividing the amount of the outstanding credit portfolio in CFA francs by the number of borrowers. Data were provided by the Programme d'Appui aux Structures Mutualistes ou Cooperatives d'Epargne et de Credit dans l'Union Economique et Monetaire de l'Ouest de l'Afrique sponsored by ILO and the Central Bank of West African States. 17 Sierra Leone Bita Hadjimichael Located on the west coast of Africa, Sierra Leone covers 72,000 square kilometers. It is bordered by Guinea to the north and east, Liberia to the southeast, and a 570-kilometer coastline along the Atlantic Ocean. About one-third of Sierra Leone's 4.2 million inhabitants are urban dwellers. Although approximately 75 percent of the land in Sierra Leone is arable, only 10 percent is cultivated for agriculture. Deforestation has depleted much of Sierra Leone's interior, and today forests account for only 6 percent of the total land area (UNDP 1993, p. 1). Despite an impressive endowment of mineral wealth (gold, diamonds, rutile, and bauxite), ample arable land, and rich fisheries, Sierra Leone is one of the poorest countries in the world. In 1992 Sierra Leone registered a per capita income of only $170 (World Bank 1994j, p. 162). During the period 1987-93, per capita real GDP in Sierra Leone declined 0.6 percent a year. The agricultural sector dominates the economy, accounting for 38 percent of GDP and 9 percent of exports in 1991. Its principal subsectors are crop production, livestock, forestry, and fisheries. About 60 percent of the workforce are absorbed by the agricultural sector, most producing at subsistence levels (World Bank 1993j, p. 11). Manufacturing accounts for approximately 5 percent of GDP and 11 percent of the labor force. This sector includes both modem factories that produce consumer goods with a high import content and small-scale and handicraft establishments, which contribute about 50 percent of the value added and the bulk of employment in the sector (UNDP 1993, p. 6). The mining sector comprises a capital- intensive enclave and substantial small-scale alluvial operations. Mineral production accounts for 10 percent of GDP and 60 percent of export earnings. Services constitute the remaining 47 percent of GDP (World Bank 1993j, p. 16). The distribution of income is markedly uneven in Sierra Leone. About 2.8 million people, or 68 percent of the total population, live in absolute poverty. In the last few years, poverty has deepened, in large part due to a population growth rate of 2.6 percent a year and a massive influx of refugees leaving the volatile Liberian border. For many years, human resource development has been neglected in Sierra Leone. The average life expectancy is quite low (forty-three years), and infant mortality is one of the highest on the continent (143 per 1,000 live 203 204 Country Studies births). Access to basic health services, safe drinking water, and sanitation is beyond the reach of the majority of the population, especially in rural areas. Average daily caloric intake is only about 1,800, less than 80 percent of the required amount. Primary school enrollment (48 percent) is among the lowest in the region, and the illiteracy rate (79 percent) is among the highest in the world (World Bank 1994h, pp. 306-7). Urban living conditions are extremely difficult, and rural villagers struggle to remain at subsistence levels. In 1994 the UNDP ranked Sierra Leone 170th out of 173 countries on its Human Development Index (calculated using a weighted composite of GNP per head, life expectancy at birth, and adult literacy). Despite its rich natural resource base, Sierra Leone suffered a steady and prolonged economic decline from 1975 to 1989. The main reasons for Sierra Leone's poor economic performance were the adoption of poor macroeco- nomic policies and the deterioration of the mining sector. Reduction in the tax base due to the decline in imports and official diamond sales, combined with an inefficient tax administration, led to a sharp fall in budgetary revenues from 16 percent of GDP in 1980-81 to 5 percent in 1985-86. Major cuts in capital expenditure followed. Investments in new social and physical infrastructure were halted, causing rapid decay in existing facilities with chronic shortages of electricity, petroleum, and water. The civil service became demoralized in the face of declining real wages and shortages of materials and supplies. Inflation accelerated, peaking at close to 170 percent a year in 1986-87 (World Bank 1993j, pp. 1-2). State control over external and internal trade became pervasive, fostering widespread economic distortions. Inappropriate pricing policies and an overvalued exchange rate discouraged agricultural production and encouraged smuggling. External payments arrears accumulated, and international reserves were depleted. Moreover, Sierra Leone's rich natural resources either were grossly underused or were used to generate income that continued to escape official channels (IMF 1992, p. 123). Faced with this deteriorating economic environment, the government began to implement a series of reforms in 1989 that were aimed at stabilizing the economy. Government efforts have since been sustained in the face of large external shocks to the economy. The influx of nearly 200,000 Liberian refugees (equivalent to 5 percent of Sierra Leone's population) and Sierra Leone's participation in the West African states' peacekeeping initiative exerted significant pressure on government finances. Deep incursions into the eastern and southern provinces by Liberian rebels in March 1991 and subsequent guerrilla activities displaced 15 to 20 percent of the population. As a result, alluvial diamond mining was curtailed, and rice production as well as cocoa and coffee exports were reduced, thus imposing additional strains on the economy (World Bank 1993f, p. 2). Sierra Leone's structural adjustment program, supported since 1992 by the World Bank-financed reconstruction import credit and an IMF Rights Sierra Leone 205 Accumulation Program, has involved liberalization of the exchange rate, trade, and prices, strengthening of the government tax collection system, and reform of the civil service and public enterprises.' The Bank-financed structural adjustment credit, approved in 1993, was designed to extend and deepen the reforms supported by the reconstruction import credit in the above areas. The structural adjustment credit also initiated Bank support for private sector development through reform of the financial sector, the legal and regulatory framework, and the indirect tax system. Despite continued security problems in the eastern and southern provinces and ongoing strains created by the inflow of Liberian refugees, progress under the structural adjustment program has been satisfactory: * The overall budget deficit was reduced from 10 percent of GDP in 1990-91 to 7 percent in 1992-93. * The growth of money supply has slowed from 140 to 35 percent a year. * The rate of inflation has decelerated sharply, from 115 percent a year in 1991 to 35 percent in 1992; by the end of August 1993, inflation had reached 20 percent a year. * Interest rates have been positive in real terms since mid-1992, and the margin between official and parallel market exchange rates has virtually disappeared. e Prices and trade regulations have been liberalized, and the state marketing company's monopoly on coffee and cocoa exports has been abolished. e Fiscal management has been strengthened. i The civil service has been downsized by about 30 percent, and salary differentials between the private and parastatal sectors have been reduced. * Exports in 1991-92 increased 15 percent in real terms but declined 9 percent in 1992-93, as rebel activities interrupted mining. Overall, between 1990-91 and 1992-93, the external current account deficit narrowed slightly from 18 percent of GDP to about 16 percent (World Bank 1993f, pp. 2-3). Main characteristics of the informal sector The proliferation of informal sector enterprises reflects, in large part, the failed macroeconomic policies of the 1980s. The informal sector has continued to grow as the formal economy has progressively failed to deliver annual wages capable of supporting an individual-let alone a family-for more than a month. The informal sector in Sierra Leone encompasses both 206 Country Studies registered and unregistered private and cooperative establishments. These enterprises usually operate at the owner's domicile, and most employ fewer than five persons. Today, the informal sector is a major source of employ- ment and the primary source of sustenance for many Sierra Leoneans. Employees in the formal sector often take informal sector jobs to supplement their incomes. Estimates based on the results of a 1989 survey indicate that the informal sector employs about 387,000 people or roughly three-quarters of the urban labor force, about half of whom are women (Sierra Leone, Central Statistics Office 1991). The occupational distribution of persons employed in the formal and informal sectors is presented in table 17.1. The formal sector is dominated by professional, technical, administrative, managerial, clerical, and service workers, and the informal sector is comprised primarily of sales, agricul- tural, production, and transport workers.2 While men tend to dominate all occupations in the formal sector, the number of women surpasses the number of men in both informal sales and agriculture. Table 17.1 Formal and informal sector employment in urban areas of Sierra Leone, by gender and occupation, 1988-89 Formal sector Informal sector Type of employment Total Male Female Total Male Female Professional and 27,700 17,600 10,100 4,600 3,300 1,300 technical Administrative 1,600 1,400 200 0 0 0 and managerial Clerical 34,800 20,100 14,700 600 600 0 Sales 7,800 5,400 2,400 190,000 59,100 130,900 Service 22,500 18,500 4,000 7,100 6,100 1,000 Agriculture and 2,700 2,200 500 99,600 42,400 57,200 fisheries Production and 45,200 41,800 3,400 78,900 68,700 10,200 transport Not classified 3,000 3,000 0 6,100 6,000 100 Total 145,300 110,000 35,300 386,900 186,200 200,700 Source: Sierra Leone, Central Statistics Office 1991. As is typical, the informal sector is dominated by self-employed workers and unpaid family workers, while the formal sector is composed predomi- nantly of wage earners. Men outnumber women in most employment categories except for unpaid family workers. The distribution of employed persons by employment status in table 17.2 illustrates the contrasts between the two sectors. Sierra Leone 207 Table 17.2 Formal and informal sector employment in urban areas of Sierra Leone, by gender and employment status, 1988-89 Formal sector Informal sector Position Total Male Female Total Male Female Employer 1,300 1,100 200 1,200 700 500 Self-employed 21,000 14,300 6,700 176,100 93,500 82,600 worker Paid employee 111,100 87,400 23,700 22,600 16,200 6,400 Unpaid family 7,900 3,900 4,000 172,700 62,300 110,400 worker Apprentice 2,200 1,700 500 11,000 10,400 600 Other 1,800 1,600 200 3,300 3,100 200 Total 145,300 110,000 35,300 386,900 186,200 200,700 Source: Sierra Leone, Central Statistics Office 1991. About 90 percent of microenterprises in the informal sector of large towns in Sierra Leone are sole proprietorships, 6 percent are partnerships, and 4 percent are family or cooperative ownerships. These enterprises employ an average of three persons, with unpaid family members accounting for 90 percent of employment. Start-up capital is quite low (on average Leone (Le) 3,000 or $19.80), and it is usually either self-financed or raised through loans from relatives and friends (World Bank 1993j, p. 39). Women in Sierra Leone cannot be considered a homogeneous group. Their living standards are heavily influenced by religion, region, and age. Non-Christians (80 percent) and women located in rural areas (two-thirds of all women) are least likely to have access to services and opportunities to ensure their health, education, and economic participation. More than 80 percent of rural women depend on subsistence farming. Most women, particularly in the northern regions, are impoverished. Generally, women in Sierra Leone do not own land. The land they farm is usually the property of their husbands, and few have opportunities to earn cash and to build savings. Most of the farm labor performed by women is viewed as part of their customary obligations to the household, and it therefore goes unpaid. Participation in nonfarm economic activities such as petty trading and the production of soap, palm oil, and tie-dye fabric is common among many rural women, especially those who are unmarried and without the necessary resources to farm independently. In urban areas, women's lower level of literacy (13 percent) compared to that of men (31 percent) limits their entry into the formal labor market and forces most of them into low-skilled, low-paying informal sector jobs. In this sector, they are predominantly engaged in wholesale and retail trade and in petty trading, accounting for 83 percent of total employment. Although this type of commerce generally produces little income, a small number of 208 Country Studies women living in western Sierra Leone control a large share of the trade and distribution markets, and many enjoy reasonable incomes (World Bank 1993j, p. 68). Policies and programs for microenterprise development The formal financial institutions in Sierra Leone are not very active in the informal sector.3 Owners of informal sector enterprises generally face great difficulties obtaining credit from the formal financial system because they lack collateral and the costs of processing their loans are high relative to the small size of the transaction. This forces them to seek credit from alternate sources within the informal financial system. Informal financial institutions abound in Sierra Leone, and nearly all adults belong to at least one savings group, commercial organization, or family financial network. These groups lend money to their members to finance commercial transactions, to cover farmers' needs for seasonal credit, to finance consumer purchases, and to support members in times of emergency. The following four informal sources of credit and savings dominate: Esusus. Esusus are savings groups usually composed of people who know one another through family, ethnic, or professional relationships. Each esusu is headed by a manager who acts both as a money collector and as a payment enforcer. Every esusu member is required to deposit a fixed amount in a common pool on a regular basis. The leader usually collects and disburses the money on the same day in order to minimize misappro- priation of funds. Since deposits are withdrawn as soon as they are collected, there is no need for esusus to maintain bank accounts. Members take turns using the money, with new members joining at the bottom of the waiting list. Although late payments do not carry any interest, defaulting on a payment has negative social and professional implications for the member. * Rotating savings and credit associations. Members of rotating savings and credit associations are required to deposit funds on a contractual basis.4 A portion of the funds are made available on a rotating basis, and the remainder is loaned to members at a predetermined interest rate (normally higher than the market rate). Some savings and credit associations also lend to members on an ad hoc basis. The size of the loan depends on the size of the potential applicant's savings. At the end of each lending period, members receive rebates on part of their interest payments, and part of the remaining profits are distributed as dividends. Defaulting members can lose their deposits as well as the right to apply for loans. Sierra Leone 209 * Traders. Traders often act as intermediaries for credit available through major exporters of agricultural produce. They offer farmers any number of items including fertilizer, pesticides, seeds, processed food and groceries, household items, agricultural tools, building materials, and money in exchange for the farmers' produce. This type of credit does not bear any explicit interest, but there are hidden costs in the sense that the produce is apparently traded below market prices. * Moneylenders. Moneylenders are also abundant in the informal sector. Despite the high interest rates (well in excess of the interest on formal sector credits), many informal sector operators use moneylenders as a primary form of finance, presumably because other forms of credit are not available. Most NGOS and international organizations have been forced to suspend operations because of the security situation in Sierra Leone. Those that provide grants and loans to small businesses also provide training and supervision to small business owners. Most operating NGOS that lend to small-scale businesses require a loan guarantor who owns land, require a contribution from the borrower, accept production equipment and purchased goods as collateral, and require their borrowers to have bank accounts. As a result, the scope of NGO lending is very limited, and few microenterprises have profited from it. The following programs are among the most important institutions supporting microenterprise development. The Sierra Leone Opportunities Industrial Center (SLOIC) was founded in 1976 with financial support from USAID and technical support from the Opportunities Industrialization Centers International (oICI, which is a nonprofit, private, voluntary organization based in Philadelphia, Pennsylva- nia). In 1989 SLOIC and oici collaborated on a six-year Small Enterprise Development and Training Program. The program was projected to cost $5.1 million, of which $3.58 million were granted by USAID, $1.4 million were provided through the government of Sierra Leone and other local contribu- tions, and $97,500 were contributed by oici (Flaming, Nguyen, and Sellu 1993, p. 2). The Small Enterprise Development and Training Project was initiated to increase the level of income, productivity, and employment in Sierra Leone by assisting in the development of micro and small enterprises. SLOIC offers a comprehensive package of services tailored to the needs and capacities of participants in its programs. These services include prevocational training, technical skill development, training in management and entrepreneurship, and loans for informal sector entrepreneurs. SLOIC operates six centers in Sierra Leone, which include two vocational education and business management training centers in Bo and Makeni, two small and microenterprise development foundations, one management and business 210 Country Studies development center in Freetown, and one agricultural skills training center. A new vocational center has been planned for Mattru Jong. The vocational training centers seek to provide hands-on training that eventually produces graduates with valuable job skills. Given the limited employment opportunities in Sierra Leone, the introduction of trainees to the concept of self-employment and basic business operations has been an appropriate addition to the curriculum. The center in Bo conducts training in automobile mechanics, carpentry, masonry, electricity, secretarial science, and agricultural surveying. Vocational courses in the Makeni center include carpentry, masonry, blacksmithing, and tailoring. From 1989 to 1992, the two centers graduated 708 trainees, about 77 percent of whom have subsequently found jobs (Flaming, Nguyen, and Sellu 1993, pp. 54-55). The credit component of the program is handled by the small and microenterprise development foundations located in Bo and Makeni. The Bo center was started in 1989, and the Makeni center was started in 1991. Loan applicants receive a package of services that consists of business manage- ment training and credit.5 The training session is required in order to qualify for a loan.6 Borrowers are also required to obtain two guarantors who must be able to submit the title to their homes as collateral on the loan. Following approval, the borrower is required to make an advance payment of 10 percent of the principal amount of the loan. This amount is subtracted from the final payment. Credit assistance is in the range of $200 to $1,000 for microenterprises in furniture production, electrical appliance repair, shoe making, tailoring, automotive repair, and other production or service activities. Most loans are specifically for working capital (Stern 1994, p. 22). Until September 1992, loans were disbursed on a twelve-month basis; however, the program has since amended its guidelines and now provides credit on six-month terms only. The annual average nominal interest rate in 1993 was 40 percent (Flaming, Nguyen, and Sellu 1993, pp. 68-70). The staff of the foundation report that loan applicants are eager and attentive during the training session up to the disbursement of the loan but that their interest wanes after they receive the loan. The findings of a mid- term evaluation indicate that most of the borrowers have not seriously applied the techniques taught in the training courses. The centers have spent a good deal of time monitoring and encouraging entrepreneurs to imple- ment specific management practices and to follow through with their business plans. The overall repayment performance has not been impres- sive. Although the average repayment rate of loans obtained from the Makeni center was 89 percent in 1991-92, the performance in Bo has steadily declined since 1989 to reach a recovery rate of only 22 percent in 1993. The drop in the repayment rate can be attributed to an increase in the volume of loans, inadequate screening criteria, and the deteriorating security situation.! Credit policies are currently being amended to reduce the loan amounts, Sierra Leone 211 shorten the loan terms, and lower the maximum debt burden to 25 percent of gross profit (Flaming, Nguyen, and Sellu 1993, pp. 71-75). In December 1993 the Sierra Leone Women's Development Movement was set up under the direction of Mrs. Gloria Strasser. The goal of this organization is to become an umbrella body that will liaise with other organizations, both local and international. It will be responsible for providing finance to women for small-scale and cottage industries ("Blueprint for Development" 1994, p. 12). Further analysis of this program is not available at this time, because it is still in the implementation phase. The ILO Secondary Roads Project concentrates on improving rural feeder roads. The construction, rehabilitation, and maintenance projects supported by ILO are labor-intensive, and they create direct opportunities for the poor to earn income. The program is task-based, so laborers are able to combine the work with their regular employment activities. The project includes a training component designed to teach local contractors how to run their own firms as well as the fundamentals of the construction industry. Women are actively recruited to participate in training programs. Contractors are able to obtain necessary equipment through a hire-purchase scheme financed through a local bank. In addition to fostering indigenous entrepreneurship, the ILO promotes development by working with chiefdom committees (Stem 1994, p. 23). Recommendations Very little of the large sums of money that have flowed into Sierra Leone to alleviate poverty have actually reached the people for whom they were intended. The government estimates that as much as 50 percent of assistance offered by donors in the form of grants has not been used. At present, few mechanisms are in place to ensure that funds go where they are most needed. A number of reasons including allocation of funds to other uses, corruption, and high administrative costs could explain the diversion of funds. It is, therefore, important to improve the supervision and accountabil- ity of assistance programs (Stern 1994, pp. 15-16). The record of donor coordination in Sierra Leone has not been very good. Since 1982 several unsuccessful attempts have been made to establish a central agency for the coordination of international assistance to Sierra Leone. Part of this failure can be attributed to opposition by individual officials within the government who resist the loss of their direct relation- ship with donors. Better donor coordination would be helpful to ensure the continuity of programs, minimize the fragmentation of efforts, avoid duplication, and facilitate more effective use of international assistance (Stern 1994, pp. 15-16). Another issue of concern is the coordination of NGOS. At present, NGOS must register with six different departments (health and social services, agriculture, labor, energy and power, trade and industry, and foreign 212 Country Studies affairs). Their activities are not coordinated, and the government is not always aware of their activities, project areas, and resources. This has resulted in duplication of development efforts among NGOS and between NGOS and the government. The National Council for Social Services in the Ministry of Social Welfare and Rural Development was intended to be an umbrella organization that would coordinate all NGOS operating in Sierra Leone, but, in practice, it only registers them.8 Women in Sierra Leone are often constrained by negative sociocultural attitudes, a lack of knowledge about their civil rights, weak links with regional and national women's groups, and restricted access to income- earning opportunities. A strong national organization to coordinate the activities of women's organizations could perhaps lobby for measures to improve the status of women. The need to optimize the use of women's productive capacity and to increase their income-earning options is a critical area where institutional support programs must be undertaken. Necessary measures are required to provide women farmers with more land for farming. Women should also be included as recipients of support services to facilitate their work in farming. Training programs in efficient techniques for weeding, harvesting, processing, and preservation would be a fine place to start. Finally, steps could be taken to prepare more women for work in wage employment in both rural and urban areas (Beoku-Betts 1990, pp. 32-35). Notes 1. In 1992 the IMF agreed that Sierra Leone could accumulate "rights" to use fund resources through February 1994 up to a maximum of Standard Drawing Rights (SDR) 87.3 million (about $119 million), close to the level of its arrears to the IMF at the time. By implementing an IMF-monitored macroeconomic and structural reform program and clearing its IMF arrears, Sierra Leone was in fact able to obtain assistance from the IMF under the enhanced structural adjustment facility. 2. The informal sector in Freetown is dominated by petty traders, tailors, artisans in metal work and automobile body repairs, cooks and bakers, watch and radio repairers, shoe repairers and shoemakers, carpenters, and construction workers. 3. Most of the information in this section is based on discussions with the Country Operations Division of the World Bank's Central Western Africa Department, which formerly included Sierra Leone. The financial sector consists of a Central Bank (Bank of Sierra Leone), four commercial banks (Barclays, Standard Chartered, Meridien Biao, and Sierra Leone Commercial Bank), two development banks (National Development Bank and the National Cooperative Bank), four rural banks, and thirteen insurance companies. Supplementing the activities of these institutions, the post office offers banking services on a small scale, primarily the acceptance of savings Sierra Leone 213 deposits. There is only the nucleus of a money market, which is operated by the Central Bank, mainly through the discounting of treasury bills. A foreign exchange bureau, registered late in 1991, has started operations. There is no stock exchange. 4. An esusu mainly functions as a depository for savings, while a rotating savings and credit association allows members to save and earn profits. 5. The training course introduces participants to concepts and techniques in bookkeeping, marketing, business management, business planning, and credit management. In addition, it teaches the social role of microentrepreneurs, taxation, insurance, and regulations that govern the participants' particular businesses. 6. The training program is implemented every three months. Loan applicants who are accepted into the next training session will wait approximately two months, and if they are placed in the following cycle, they will wait five months, thus increasing the amount of time an applicant has to wait in order to receive a loan. 7. It has been extremely difficult to appraise microentrepreneurs who are illiterate and do not keep records and who are by nature guarded about disclosing information regarding their business and personal activities. The centers have therefore calculated the profitability of the activity without taking into account the money that the owner draws from the business. This approach has not provided an accurate picture of the debt burden of the loans. Consequently, loans have been on average too large for borrowers to manage. The program is therefore seeking to reduce its dependence on the analysis of profitability. 8. Stern 1994, p. 17. The Association of NGOS (TANGO) that exists in The Gambia could serve as a useful model for strengthening the coordination of NGOS in Sierra Leone. To facilitate coordination, TANGO publishes a local directory of NGOS and a quarterly magazine, TANGO Talks. Part 3 Appraisals of Selected Microfinance Institutions 18 Reseau des Caisses Populaires Pierre-Olivier Colleye and Ousa Sananikone The R6seau des Caisses Populaires was established in Burkina Faso in 1972 with the assistance of the Societe de Developpement International Desjardins (SDID), a Canadian NGO, and financing from the Canadian International Development Agency (CIDA). The project began as a small assistance program to help village associations to mobilize savings in the Bourougiba region. Very quickly, management realized that villagers could increase their standard of living if savings were transformed back into credits. The organization first viewed itself as a project but eventually began to think of itself as a financial institution on the road to full sustainability. The Reseau is a credit and savings cooperative, operating in both rural and urban areas. It expanded slowly during its first fifteen years but grew very rapidly between 1990 and 1994 to become one of the largest of such institutions in West Africa (see table 18.1). It now includes thirty-five caisses with 64,000 members, CFAF 2.4 billion ($4.6 million) in savings, and CFAF 620 million ($1.2 million) in outstanding loans.' Program characteristics The Reseau is a three-level credit and savings cooperative with thirty-five caisses, three unions, and one federation. Units at each level function as independent legal entities. The caisses Caisse members are individuals and groups, with an average of 1,500 members per caisse. Urban caisses tend to have more members than rural ones. The decision to open a caisse is driven by demand, because the Reseau is now known throughout the country and no longer has to promote its services. Requests to open a caisse are sent by a group of individuals or by the authorities of a village. After a short socioeconomic study of the area's social cohesion and economic potential, a caisse is opened to serve villages within a 15-kilometer radius of the selected location. The caisse is located either in a building constructed with funds from the federation, from donors with programs nearby, or in a rented building. 217 218 Appraisals of Selected Microfinance Institutions Table 18.1 Growth of the Reseau des Caisses Populaires in Burkina Faso, 1975-94 Characteristic 1975 1980 1985 1990 1994 Number of caisses 7 7 11 25 35 Number of members 2,800 7,200 11,700 25,000 64,000 Outstanding loans (millions of 18 30 80 250 620 CFA francs) Mobilized savings (millions of 35 100 265 725 2,406 CFA francs) To become a member, an individual must pay a subscription fee of CFAF 1,000 ($2): CFAF 500 ($1) as paid-in capital, CFAF 400 ($0.80) for a fund to construct other caisses, and CFAF 100 ($0.20) for processing fees. Additional requirements include an identification card (which is easily obtained) and two pictures. A general assembly of members elects a board consisting typically of seven persons, a management committee consisting of three persons, and a control committee consisting of five persons. Board and committee members are volunteers, and their two-year term can be renewed once SDID hires one agent to manage each caisse and a few cashiers or trainees, depending on the size of the caisse, to conduct daily operations.2 The unions The thirty-five caisses are grouped into three unions, one each in the regions of Bourougiba, Plateau Central, and Yatenga. Union members are the individual caisses, and union boards are elected by two delegates from each caisse. A union typically includes caisses within a 100-kilometer radius. Management staff of a union consist of roughly seven people. Below the director are two departments: training and accounting. The unions assist the caisses with the following activities: * Hiring and training agents. * Transporting funds between the union and the caisses. * Assisting with logistics and supplies. * Consolidating financial documents. * Collecting data on outstanding amounts, number of members, and repayment rates. * Collecting savings from the caisses and depositing them with commercial banks. Reseau des Caisses Populaires 219 * Intervening in case of serious problems (low repayment, theft, and conflicts). * Maintaining the link between the federation and the caisses. The federation The federation is almost fully developed and should become a legal entity in 1995. Its members will be the unions, and its functions will be to consolidate financial information from unions, conduct audits, disseminate best practices, and provide guidance for the development of new caisses. Until the federation is finalized, the institution is managed by a coordina- tion center that is operated by SDID. It provides services similar to the ones planned for the federation, but is more involved in development issues related to growth, new methods of control, and subsidies. SDID used to be actively involved in managing the unions, but this is no longer necessary because SDID has provided extensive training to union employees. Once created, the federation will take over the responsibilities of the coordination center and will continue to be assisted by SDID. The coordination center has a staff of roughly fifteen (plus drivers and guards) and is divided into three departments: accounting, control, and training. It is headed by a Burkinabe and assisted by a representative from SDID, who is a Canadian national and the only expatriate in the institution. In addition, the center cooperates with Freedom from Hunger, an NGO that provides education to several women's groups to whom the Reseau also makes loans. Savings All savings are demand deposits that do not yield interest. Total savings reached CFAF 2.4 billion ($4.8 million) by end-1994. Members reportedly save with the R6seau for one of three reasons: because they wish to protect their money against theft and social pressure, because there are no other local institutions where they can save, and because savings are required to receive loans. There is little competition for small savers in Burkina Faso, particularly in rural areas. Banks do not encourage small savings accounts, especially now that they have excess liquidity. Reportedly, they charge high fees to open an account (up to CFAF 40,000, or $80), impose minimum savings amounts, pay low interest rates, and charge high monthly administration fees. The R6seau experimented with paying interest on savings but quickly gave it up as unnecessary and financially unsustainable. Savings accounts average CFAF 38,180 ($76; see table 18.2). Precise figures for the number of savers are not available, but managers reported that the overwhelming majority of members are active savers with little concentra- 220 Appraisals of Selected Microfinance Institutions tion in savings. Depositing and withdrawing money are simple and fast operations, and the caisses are open during regular business hours. Table 18.2 Savings and credit activities of the Reseau des Caisses Populaires in Burkina Faso, by geographic area, as of September 20, 1994 Characteristic Overall Rural areas Urban areas Number of members 63,907 38,178 25,729 Number of caisses 35 27 8 Number of members per caisse 1,803 1,414 3,216 Percentage of women 28 25 31 Savings Savings (millions of U.S. dollars) 4.8 2.3 2.5 Average savings per member (U.S. 76 60 97 dollars) Average savings per caisse (U.S. 137,490 85,000 312,500 dollars) Loans Outstanding loans (millions of U.S. 1.2 0.6 0.6 dollars) Average loan amount per caisse 35,458 26,000 70,000 (U.S. dollars) Number of loans per caisse 106 100 127 Average loan size (U.S. dollars) 336 242 552 Credit The R6seau has developed an extensive product line of loans, and new products are added frequently. Interest rates are fixed at 16 percent for all loans.3 As of early 1995, there were ten types of loans (see table 18.3). Until the end of 1994, maximum loan sizes were CFAF 100,000-CFAF 300,000 ($200-$600), depending on the type of loan. The average loan size is CFAF 168,000 ($336): CFAF 121,000 ($242) in rural areas and CFAF 276,000 ($552) in urban areas. There are currently 3,700 loans outstanding, for a total of CFAF 620 million ($1.24 million). At the end of 1994, the maximum loan size was increased to CFAF 3 million ($6,000), except for salary advances and group loans, which are limited to 25 percent of the member's salary and CFAF 150,000 ($300) per person, respectively. This change in policy will be discussed below. Reseau des Caisses Populaires 221 Table 18.3 Types of loans of the R6seau des Caisses Populaires in Burkina Faso Type of loan and term Purpose Loansfor consumption Advance on salaries (one Civil servants and other salaried employees year) Social (one year) Baptisms, weddings, education, funerals, and health Household (two years) Furniture, installation of sanitation, and house- hold goods Vehicles (five years) Bicycles, mopeds, cars, and trucks Agricultural loans Input (ten months) Seeds, livestock, pesticides, and fertilizers Equipment (five years) Animals and tools Housing loans Housing (five years) Purchase, construction, and renovation of homes Community loans Group loans (two years) Village associations and women's groups Commercial loans Inventory (one year) Stock Equipment (two years) Tools and "office" furniture Individual loans. Application procedures vary slightly from caisse to caisse, but as a general rule, a borrower must have been a caisse member for at least four months, have made regular deposits and withdrawals during those four months (at least one or two transactions per month), be an adult, and live or work in the area served by the caisse. Loan applicants must have a short interview with the caisse manager who helps them fill out the loan application with information about the size of the loan, the intended use of the loan, and the borrower's income and debt.4 To qualify for a loan, borrowers must put up collateral worth 100 percent of the loan amount and have 25-30 percent of the loan amount in savings (which cannot be withdrawn until the loan has been repaid). The caisse, therefore, requires security equivalent to 125-130 percent of the loan, and effective interest rates for the borrower increase to 21 percent. According to the management, collateral requirements are not a major constraint for members. In urban areas, houses, furniture, and household goods are accepted as collateral. In rural areas, animals, bicycles, mopeds, and ox carts are accepted.5 If a loan applicant has no collateral, she or he can join a solidarity group for which collateral is not required. 222 Appraisals of Selected Microfinance Institutions Once the application has been completed, the collateral has been assessed by the credit committee and the caisse manager, and all other requirements have been met, the credit committee votes on whether or not to extend the loan. The committee works by majority rule with a quorum of three people. Group loans. The procedure for group loans is slightly different. Groups are usually composed of women and have from three to sixty members. Each group selects a president who collects repayments from group members and deals directly with the caisse. To increase group cohesion, the group is considered as one client. It must save and transact as a group for six months before it can apply for a loan. No collateral is required for a loan, but members must have 10 percent of the loan amount on deposit. Group loans show a high repayment rate (more than 98 percent). Of the Reseau's 64,000 members, 5,600 are groups. The Reseau's overall repayment rate reportedly is more than 95 percent.6 If a loan payment is fifteen days late, the member is called to talk to the credit committee. The committee will decide on the penalty, new repayment dates, and whether or not to set up a provision for bad debts.7 Collateral is rarely used to repay the debt. If a caisse's overall repayment rate drops below 90 percent, all credits are suspended until the problem is solved. Commercial loans represent between 40 and 83 percent of the loan portfolio, depending on the union. Up to 30 percent of loans are for agricultural activities. The Reseau is trying to diversify its portfolio and could reach an agreement with the World Bank to finance housing construction. Flow of funds The Reseau is experimenting with two different operating models, which are described below (see figures 18.1 and 18.2). The three unions have thus far used model 1, but for the past year, one of the unions has switched to model 2. It is expected that, within a couple of years, all unions will have adopted the new model. Under model 1, the caisse: : Receives savings from its members on which it does not pay interest. * Lends part of the money back to the members at a 16 percent interest rate. * Deposits the rest of the savings with the union from which it receives 2.5 percent interest. * Receives subsidies from the union during its first few years of operation.8 Moreover, the union: * Receives savings from its members (the caisses) on which it pays 2.5 percent interest. Reseau des Caisses Populaires 223 * Lends part of the money to large borrowers (the loan limits are larger for a few big merchants who deal directly with the union). * Deposits the rest of the savings with a commercial bank from which it receives 4.5 percent in interest. * Receives subsidies from SDID. Figure 18.1 Model 1 |Members \redit (@16%) B1 \ u Subsidies > C Credit (@ 16%) M A Savings (@4.5%) N M N~ ~ ~ ~~~~~ Saig N > 2.5%) S Savings (@ 0%) B Subsidies (grant) |S.D.I.D. | Under model 1, the union is financed mainly by the difference between the interest rate it pays to the caisses (2.5 percent) and the interest it receives from the banks (4.5 percent). In contrast, with model 2 the union passes the full 4.5 percent received from the commercial banks to the caisses. It also transfers the revenue from its own lending operations to the caisse. The union has a budget that is set by the union board (made up of caisse representatives) during the general assembly that takes place at the beginning of the fiscal year. To cover its operating expenses, it then bills each caisse in proportion to its total assets. Model 2 is likely to be expanded throughout the R6seau. It has two main advantages, as follows: * Unions no longer compete directly with the caisses by making direct loans to larger borrowers because the interest from these larger loans is passed down to the caisses. 224 Appraisals of Selected Microfinance Institutions * Unions are more accountable for their budget. Theoretically, under model 1, if rates were to triple (making the interest spread 6 percent instead of 2 percent), the union could spend three times as much money without having more costs to cover. Under model 2, the benefits from the increased interest would be passed down to the caisses and presumably to their members. Figure 18.2 Model 2 |Members \redit (@16%) B \ u ~~~~~~~Subsidies )O C Credit (@ 16%) m A < Savings (@4.6%S) _l Costtinanclng Il j S ) S Savings (@ 0%) K X N Savings (@2.5%) ES Subsidi (grant) Performance The Reseau de Caisses Populaires does not compile consolidated balance sheets or income statements. Each union and each caisse is an independent legal entity, and each publishes its own financial statements. Controlling services rotate from caisse to caisse throughout the year, which means that official statements are issued in different months. Still, the Reseau collects all information made available from the caisses on a monthly basis. This information includes general statistics about outstanding loans, savings, repayment rates, and other income and expenses and balance sheet items. Only sixteen caisses were able to provide complete financial statements for the appraisal at hand, because the other nineteen had not yet been audited in 1994 and therefore could provide only partial data. The income statement presented in table 18.4 is extrapolated from these sixteen caisses, under the assumption that they are representative of the thirty-five caisses.9 This appraisal, therefore, cannot include accurate Reseau des Caisses Populaires 225 consolidated statements, which would include the caisses, the unions, and the coordination center, nor can it calculate overall recovery rates.'" However, we can look at these three levels independently. Table 18.4 Adjusted financial statements of the Reseau des Caisses Populaires in Burkina Faso: Income statement for the caisses, 1994 (millions of CFA francs) Amountfor the As a As a twelve months percentage of the percentage of the Financial category ending 9/31/94 average portfolio' net operating income Income Credit income 82 17 59 Investment income 53 11 38 Interest expense 0 0 0 Fees 0 0 0 Net interest income 135 28 91 Other income 3 1 2 Net operating income 138 29 100 Expenses Salaries 51 11 37 Administrative ex- 70 15 51 penses Depreciation 15 3 11 Other expenses 12 2 9 Total expenses 148 31 107 Net income -10 -2 n.a. n.a. Not applicable. a. The average loan portfolio for 1994 was CFAF 482 million ($965,000). Note: This is an extrapolation based on sixteen caisses. These figures do not include the income and expenses of the three unions nor do they include the expenses of the coordi- nation center. See the financial notes in annex 3. This financial statement was recon- structed from data provided by the institution. It has not been formally audited but was approved by the institution's management. The caisses In 1994 the caisses saw their investment income drop and some expenses increase due to the devaluation of the CFA franc. As of September 30, 1994, the caisses' interest income covered 91 percent of their operating costs. This average figure is high, but financial performance varies from caisse to caisse. Indeed, a few caisses make relatively large profits, while many realize small losses. Because the Reseau is not a single legal entity, the large profits cannot be transferred from one caisse to another to compensate for the small losses. 226 Appraisals of Selected Microfinance Institutions However, the management is considering establishing a solidarity fund that would redistribute some of the profits among caisses. The unions Overall, unions break even and do not require subsidies. All together, they generated approximately CFAF 83 million ($166,000) in revenue, which covered the CFAF 82 million ($164,000) in operating expenses in 1993. Unions under model 1 did not perform as well as in previous years due to a decline in interest rates after the devaluation. Until the devaluation, commercial banks paid 8-10 percent on savings to the unions. The unions passed down half of this income to the caisses. This large spread of 4-5 percent provided a comfortable income for the unions to cover their costs. Today, commercial banks only pay 4.5 percent on savings, half of which is passed down to the caisses. Unions' income therefore has dropped almost 50 percent. The union employing model 2 was less affected by the devaluation, because it is financed by a fixed fee charged to the caisses. The coordination center This coordination center is entirely financed by a CIDA grant and has a budget of approximately $500,000 a year. This grant covers all the expenses of the center, including the SDID expatriate and operating subsidies from the center to new or unprofitable caisses. SDID disbursed only about 20 percent of the operating subsidies budgeted for 1994 because the caisses' results were better than anticipated. Each level of the Reseau appears able to provide complete and accurate financial statements on a monthly basis. Unfortunately, the R6seau does not have the infrastructure or the equipment to consolidate these multiple financial statements into a comprehensive statement for the entire institution (see table 18.5 for a rough estimate). Looking ahead The Reseau de Caisses Populaires is a large organization that has expanded tremendously in the last few years while managing to keep its infrastructure and operating costs to a minimum. Looking ahead, three strategies would appear to enhance the profitability of the organization and the impact on its members: increase the profitability of credit operations, set up a solidarity fund for caisses that are unprofitable during their first few years, and computerize the accounting system so staff can publish complete consoli- dated reports that can be audited. Reseau des Caisses Populaires 227 Table 18.5 Adjusted financial statements of the Reseau des Caisses Populaires in Burkina Faso: Rough estimate of the income statement for the entire institution, 1994 (millions of CFA francs) Amountfor the twelve months As a percentage of As a percentage of net Financial category ending 9/13/94 average portfolio operating income Income Credit income 90 19 42 Investment income 120 25 57 Net interest income 210 44 98 Other income 5 1 2 Net operating 215 45 100 income Expenses Caisses salaries 51 11 24 Union salaries 20 4 10 Administrative 100 21 58 expenses Depreciation 25 5 12 Provisions 12 2 6 Coordination 200 41 95 expenses Total expenses 408 85 194 Net income -198 -41 n.a. n.a. Not applicable. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Credit activities First, the Reseau could try to lend a larger portion of the savings collected. The commercial bank interest rate decreased after devaluation (from 8-10 percent to 4.5 percent), and this has had a serious impact on the profitability of the institution. Previously, it could be profitable by depositing 65-70 percent of savings with commercial banks and lending the remainder to members." With the new interest rates in effect, this ratio is no longer possible. By lending as little as 40 percent of mobilized savings, all caisses probably could cover all of their operating costs and the costs of the unions. Re-lending 40 percent of savings might be the limit. The management is not willing to re-lend more because it does not have any long-term funding. Indeed, members can withdraw all their savings at any time, while loans are 228 Appraisals of Selected Microfinance Institutions for periods of ten months to five years. Two possible sources of long-term funds are term deposits and lines of credit. However, even if the institution could obtain lines of credit from commercial banks, these lines could not be profitable if banks charge approximately 13 percent on their loans and if the Reseau cannot charge more than 18 percent (that is, the usury rate). Second, the Reseau can raise the ceiling on loans. The management believes that collateral is not a problem but that many members do not turn to the R6seau for loans because loan ceilings are too low. This is why these ceilings were raised to $6,000. Nonetheless, management expects most caisses to continue to make small loans and does not expect the R6seau to move away from its original goal of alleviating poverty. Solidarity fund The Reseau is also looking for ways to finance its own growth. Caisses allocate their profits to different accounts: 10 percent of profits usually go to reserves, another 10 percent are used to set up an education fund for women, and 80 percent go to a fund used to cover theft and default. The management is considering setting up a new fund that would be used either at the union level or at the Reseau level. Each caisse would transfer a percentage of its profit to this fund at the end of each year. The fund would be used to help new caisses during their first few years of operation until they can show a profit. Accounting procedures At every level of the organization, the staff seems quite knowledgeable about accounting procedures. Although each caisse is able to issue a complete balance sheet and income statement every month, the institution is not able to consolidate all of these statements (see table 18.6). It could be very useful to provide unions with the proper computer equipment and software so that they could consolidate financial data. Consolidated data would allow the Reseau to have a deeper understanding of its financial situation, to identify and remedy problems quickly, to issue financial statements that can be audited, and to improve record keeping-for example, by converting the financial statements from cash accounting to accrual accounting methods."2 In addition to several computers for the coordination center and the unions, the Reseau would need to determine whether existing programs such as the U.N. Food and Agriculture Organiza- tion's Micro-Banker, a software package for microfinance institutions, could be adapted to its needs or whether a program would need to be designed specifically for it. Reseau des Caisses Populaires 229 Table 18.6 Adjusted financial statements of the Reseau de Caisses Populaires in Burkina Faso: Balance sheet, 1994 (millions of CFA francs) Financial category Amount Assets Cash 240 Bank accounts 1,509 Loan portfolio 621 Provision for loan losses 62 Fixed assets (net) 92 Other assets 145 Total assets 2,545 Liabilities Savings 2,406 Payables 200 Total liabilities 2,606 Equity Paid-in capital 34 Reserves 124 Subsidies plus retained earnings -219 Total equity -61 Total liabilities plus ecuity 2,545 Note: Extrapolated from financial data for the thirty-five caisses, the three unions, and the coordination center. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Future plans Now that the coordination center no longer needs to take part in the daily management of the caisses and the unions, management is working on expanding the network. This includes creating new products, funding new caisses, and setting up new unions. It might also consider providing venture capital for small and medium enterprises. A major step toward abandoning the status of the project also will be to finalize the federation, which will replace the coordination center as a legal entity under the supervision of the Ministry of Finance and in accordance with the laws set by the Central Bank of West African States on credit and savings institutions. 230 Appraisals of Selected Microfinance Institutions Notes The information on Reseau des Caisses Populaires is based primarily on discussions with the management, staff, and members of the institution. 1. By the end of 1995 these numbers had increased to 93,000 members, CFAF 3.8 billion ($7.6 million) in savings, and CFAF 2 billion ($4 million) in outstanding loans. The exchange rate used throughout this chapter is 500 CFA francs to 1 U.S. dollar. 2. Caisse managers are paid CFAF 60,000 ($120) a month, and cashiers earn about CFAF 40,000 ($80) a month. The Reseau also hires trainees, who are paid CFAF 25,000 a month ($50). Trainees work as cashiers for twelve months before being confirmed as permanent staff. 3. Inflation before devaluation was approximately 1 to 2 percent. Devaluation caused inflation to reach 35 percent, but this rate is decreasing rapidly. The consumer price index rose 0.65 percent in December 1994. 4. Debt is calculated as the ratio of household yearly debt payments divided by the household's yearly income. This ratio must remain below 35 percent. 5. If a member has no collateral, he or she can use a procuration, whereby a friend or a relative pledges his or her asset as collateral. 6. The Reseau calculates its repayment rate as the portion of the loan amount that is less than ninety days overdue. 7. All provisions for bad debts are transferred to the income statement at the end of the year. 8. There are two types of subsidies: operations and investments. It takes approximately two years for an urban caisse and five years for a rural caisse to break even. 9. The sixteen caisses appear representative of the entire Reseau: they cover both rural and urban areas, belong to the three unions, and are of all ages and all sizes. 10. In addition to the extrapolated income statement for the thirty-five caisses, table 18.5 contains a rough estimate of the income statement of the entire institution. This estimate assumed that the income statements, despite being for different months, could be consolidated and that the cost of the coordination center was approximately $400,000 (a $500,000 budget minus approximately $100,000 in potential subsidies to the caisses). 11. At one point before devaluation, the Reseau was lending up to 40 percent of mobilized savings. But right after devaluation, the number of loans increased only slightly, while large flows of cash that had been secured abroad came back to Burkina Faso and were deposited with the Reseau. 12. This is quite important, because many loans are between ten months and five years. 19 Sahel Action: Projet de Promotion du Petit Credit Rural Pierre-Olivier Colleye and Ousa Sananikone Projet de Promotion du Petit Credit Rural (PPPCR) provides small loans to groups of women in predominantly rural areas of Burkina Faso. It was designed in 1988 by a French agricultural research institution, the Centre International de Recherche Agronomique pour le Developpement (CIRAD). PPPCR is financed by the Caisse Francaise de Developpement (CFD) and implemented by Sahel Action, a local nonprofit NGO aimed at promoting the development of rural women. PPPCR has an operating budget of FF 20 million (which it receives as a grant) and a FF 2.5 million line of credit. The CFD extends this credit line to the Caisse Nationale de Credit Agricole (CNCA), a state-owned agricultural bank in Burkina Faso, at an annual interest rate of 4.5 percent. CNCA on-lends these funds to PPPCR at a rate of 10-13 percent a year. PPPCR grew out of an initiative by CIRAD to alleviate poverty in the northern province of Yatenga following droughts that struck the Sahel in 1973 and 1984. Modeled after the Grameen Bank, its goal was to revive the rural economy by providing loans to low-income farmers and other poor individuals who had lost their savings and assets during the droughts and lacked access to formal financial services. In particular, it sought to reach women, not only because they are less likely than men to have access to resources, but also because they generally have stronger repayment performance in microcredit programs. Now in its second phase (1992-97), the project operates in four regional units located in the provinces of Ganzourgou, Soum, Tapoa, and Yatenga, and it serves nearly 200 villages. A central unit located in Ouagadougou oversees the project and coordinates activities at the regional level. PPPCR has a staff of fifty-seven employees, including thirty-two field agents and one full-time and one part-time expatriate. The project has recorded impressive growth in recent years. Between January 1, 1988, and November 30, 1994, a total of 29,276 loans were made, 13,960 of which were outstanding in February 1995. The outstanding portfolio is CFAF 231.6 million ($463,200), and the average loan is nearly CFAF 24,000 ($48). Loan performance is strong, with a reported repayment rate of 99.6 percent. Although lending activities have been expanded to urban 231 232 Appraisals of Selected Microfinance Institutions areas, the majority of borrowers (65 percent) remain rural. Typical clients are small traders of agricultural products, farmers, and artisans. Nearly 95 percent of borrowers are women, and at least 95 percent are illiterate. Program characteristics Lending activities at PPPCR take place at the level of the four provincial units, each of which manages lending activities for a group of villages. Mobile village agents each supervise an average of eight villages that they visit once a week to disburse new loans and collect payments. Agents report weekly to the provincial unit but otherwise live in one of the villages under their supervision. Agents use a rented office and a safe, but because they are mobile, there is no physical caisse or bank, as is observed in other rural credit programs. Selection of a village Two basic conditions must be met before lending activities can be estab- lished in a given village or region. First, there must be strong local demand for credit, usually expressed verbally by the villagers themselves. Credit demand can also be estimated by the presence (or absence) of other sources of informal finance in the region. Second, the region must demonstrate both economic potential and a high level of social cohesion. A region typically is selected after the central unit conducts a socioeco- nomic study to determine whether the physical, social, economic, and political environment is conducive to economic opportunities and effective group lending. In such instances, village agents promote the project. Agents hold meetings with local authorities (village chiefs, imams, and heads of village associations) to explain the terms and conditions of the lending program. If participants express an interest, agents will organize similar meetings with villagers. In general, the aim is to establish credit operations in a province where significant economic opportunity and social dynamism already exist, conditions are favorable for agriculture, and villages are located relatively close to one another. For example, the province of Ganzourgou has the advantage of three large regional markets, and a long tradition of strong social cohesion. Eligibility Because PPPCR targets the rural poor, with an emphasis on women, salaried workers (who are mostly civil servants) are not eligible. Loans are always provided to groups rather than to individuals, hence the importance of strong social cohesion among borrowers and within the village community as a whole. Clients are organized into groups ranging from three to six members, all of whom are liable for loans given to any member of the group. Sahel Action: Projet de Promotion du Petit Credit Rural 233 Process After formation, a group requests a loan from the village credit committee, whose members are selected by all the groups. Committees range in size from ten to thirty persons. In keeping with the very small loan amounts and the illiteracy of most borrowers, a loan request consists of a meeting between the group and the credit committee and does not involve documents. In one of the few formalities, a borrower must "declare" the projected use of the loan. Loan approval is always made jointly by the committee and the agent. Once a week, village agents submit loan requests to the provincial unit for formal approval, and a request for funds is then sent to the central office in Ouagadougou. Village agents visit each village once a week to disburse new loans. Loan processing takes approximately one month. Credit activities PPPCR provides four types of loans: weekly credit, livestock credit, cereal credit, and artisanal credit. All loans have a maximum term of thirteen months and annual interest rates of 20-25 percent, depending on the type of loan. Because borrowers do not like to have their interest payments mixed with principal payments, the interest is paid at the end of the loan, separately from the principal. The interest is calculated on a straight line. * Weekly credit is used primarily to finance the processing of agricultural products. Loan amounts range from CFAF 5,000-CFAF 50,000 ($10-$100) and have an annual interest rate of 24 percent for a term of either seven or thirteen months. Disbursed during the rainy months of Octo- ber-November just before harvest, the loan is reimbursed in thirty or fifty-six weekly installments. - Livestock credit is used to finance livestock (cows, sheep, and goats). Loans are available for CFAF 10,000-CFAF 30,000 ($20-$60) for a term of six months at 20 percent annual interest. The loan is repaid in one install- ment. * Cereal credit is used to finance the storage of grain. Clients can borrow CFAF 20,000-CFAF 60,000 ($40-$120) for a term of nine months at an annual interest rate of 25 percent. It is reimbursed in three installments. * Artisanal credit is an experimental credit that finances small artisanal production. It has a term of six months and an interest rate of 20 percent a year. In 1994 weekly credit accounted for 75 percent of the total number of loans; cereal credit for 15 percent; livestock credit for 9 percent; and artisanal credit for 1 percent. The average loan size was CFAF 24,000 ($48), up from CFAF 19,500 ($65) in 1993. 234 Appraisals of Selected Microfinance Institutions Each borrower must deposit 10 percent of the loan amount in a guarantee fund to cover loan losses and emergencies. This fund yields no interest and is returned to the borrower after loan repayment. PPPCR is currently debating whether the fund should pay interest to contributors (that is, borrowers). Loan repayment PPPCR has a very strong loan collection performance, with 99.6 percent of loans reportedly repaid on time in 1994. This is partly the result of a good reporting system. The project initially experienced some repayment problems that were due mainly to the absence of an information system that would allow village agents to know the status of all loans at all times. In 1993 each provincial unit set up a monthly record of basic indicators such as outstanding loans, repayment dates, interest due, and level of guarantee funds accrued. Each village agent maintains a data book to track the flow of all funds, enabling the agent to intervene only when payments are due and thereby keeping costs to a minimum. This book is checked by the provincial unit once a week when village agents deposit the contents of their safes at the provincial unit. In addition, financial records have been computerized at the provincial unit, greatly simplifying the task of provincial agents who consolidate data from village agents. High repayment rates are also due to the effective use of social pressure, underlining the importance of social cohesion in selecting a community for group lending. Group leaders collect loan repayments and pass them on to the village agent in the presence of the credit committee, which serves as a kind of "moral witness." If a repayment is late, the committee attempts to understand the nature of the problem. However, once a loan is more than two payments late, the entire neighborhood is blocked from receiving new loans until all scheduled payments have been received. A third factor accounting for high repayment rates is the fact that most borrowers are women. As demonstrated in many microcredit programs around the world, women tend to repay loans better than men. The reasons behind this may be that women in developing countries generally bear the financial burdens of rearing children and maintaining the household, and they must keep their credit in good standing because they usually lack access to other forms of financing. Training of village agents Agents are hired at the provincial level and trained on-the-job for about a year before being confirmed as agents. To facilitate contacts with the client population, agents typically come from the same province to which they are assigned and thus speak the local language and know many people in the region. To avoid conflicts of interest and pressure to lend exerted by family Sahel Action: Projet de Promotion du Petit Credit Rural 235 and friends, they cannot supervise their home village. PPPCR agents have an average of six years of primary education, which apparently is adequate for the lending activities they need to perform. In 1994 a few classes were added to the agents' training. The training section of the central unit coordinates training activities and plays an advisory role in the hiring of provincial agents. Training of clients Other than initial promotional sessions between village agents and borrowers, there is no training of clients. This keeps costs down, and because both the credit services offered by PPPCR and the economic activities of borrowers are relatively basic, PPPCR management believe that their borrowers do not require additional training. On average, village agents and borrowers have two to three meetings in which agents explain the terms and conditions of the credit program. Governance The decisionmaking structure at PPPCR is relatively decentralized, taking into account the geographical remoteness of borrowers, the very small transactions, and the simple credit services offered (see table 19.1). Decisions directly affecting clients, such as the screening of applicants, approval and repayment of loans, and enforcement of loan collection, are taken at the village level by village agents, the credit committee, and the borrowers themselves. This arrangement keeps administrative costs down by using local sources of information to screen borrowers and group pressure to enforce loan repayment. It also fosters a successful program based on local ownership. The levels of governance at PPPCR are as follows: * The borrowing group is self-selected and thereby provides effective screening of applicants. The willingness of group members to assume responsibility for one another's loans ensures loan repayment. * The credit committee is selected by all members of the borrowing group in the village. It functions as moral adviser, recruiter, and loan officer, enabling village agents to devote more time to promotional activities. To be most effective, the credit committee has to have sufficient moral authority and social standing, and it also has to be relatively small. In the past, PPPCR experienced considerable delays in loan approval in villages whose committees had 100 to 200 members. Since then, the size of the committee has been reduced, and the number of members who have the authority to make loan decisions has been limited to two to five persons, although that varies from one village to another. 236 Appraisals of Selected Microfinance Institutions Table 19.1 Decisionmaking structure of the PPPCR in Burkina Faso Level Function Credit committeea Advise prospective clients; advise village agents about applicants; approve loans (jointly with agents); help with loan repayment (with village agents) Village agents Promote the project; inform clients about the terms and conditions of credit; register new groups; approve loans (jointly with the credit committee); transmit loan re- quests to the provincial unit; disburse loans to villages; visit villages weekly; collect repayments (with the credit committee); maintain books Chief of provincial Formally approve loans; coordinate lending, training, unit and promotion activities at the provincial and village levels Central unit Provide centralized administrative, accounting, and technical support to the provincial units; develop finan- cial products; study the impact of credit; provide intel- lectual leadership Steering committeeb Govern the project a. Selected by group members for an indefinite period and ranging in size from 20 to 200 members; however, only two to five members have the authority to make loan deci- sions. b. Made up of Sahel Action, CNCA, CFD, and CIRAD. * The village agent is responsible for all aspects of loan approval and implementation and works with the credit committee to approve and supervise loans and to collect payments. A key operating principle at PPPCR iS that loan decisions are always made jointly by the credit committee and the village agent. * The provincial unit chief heads the provincial unit and formally approves loan requests submitted by village agents and credit committees. Apart from the unit chief, the provincial unit typically employs three to eight village agents (some of whom may be trainees), an accountant who consolidates the agents' books, and a security guard. * The central unit was established in 1993 to coordinate the activities of the provincial units and to oversee the project as a whole. It is divided into three sections: the service unit, which coordinates the provincial units, the training unit, and the administration and finance unit. The finance unit transfers funds to the provincial units. As of November 1994, the central unit had a staff of seven, including one French expatriate from CIRAD. * The project is currently headed by a Burkinabe woman (who is also the chief of PPPCR's provincial unit in Yatenga) and governed by a steering committee made up of representatives of the implementing and donor Sahel Action: Projet de Promotion du Petit Credit Rural 237 agencies: Sahel Action, CFD, CNCA, and CIRAD. The steering committee decides large financial and institutional issues affecting the project, including interest rates, advances on donor funds, the hiring of manage- ment, and the future legal status of the project. Performance PPPCR'S outreach has grown significantly (see table 19.2). Since 1988 credit activities have expanded rapidly from two to nearly 200 villages. Between 1991 and 1994, the number of loans grew by an average rate of 80 percent a year, and the amount lent grew 104 percent. Although statistics are not available, the project clearly has many repeat borrowers, which points to continued demand for credit (and, presumably, satisfaction with credit services). Table 19.2 Growth of the PPPCR in Burkina Faso, 1991-94 Characteristic 1991 1992 1993 1994 Number of villages 36 62 105 191 Number of loans 2,051 3,890 6,758 13,174 Amount lent (millions of 33.4 95.0 131.6 332.2 cFA francs) Average loan amount 2,675 16,332 19,478 24,458 (CFA francs) Average repayment rate 96.4 99.8 99.9 99.9 (percent) Number of staff 18 32 46 57 For the 6,758 loans disbursed in 1993, the average loan amount was CFAF 19,478 ($65) or approximately 22 percent of the real per capita GDP of Burkina Faso.' This small amount indicates the low income of the borrowing population, which consists of poor rural women. Interest income for the period ending December 31, 1993, covered only 13 percent of operating expenses (including the cost of the expatriate staff) and 12 percent of operating plus financial expenses. Excluding expatriate- related expenses (categorized as institution building in table 19.3), interest income covered 16 percent of operating expenses and 14 percent of operating plus financial expenses. 238 Appraisals of Selected Microfinance Institutions Table 19.3 Adjusted financial statements of the PPPCR in Burkina Faso: Income statement, 1993 (millions of CFA francs, unless otherwise noted) Amountfor the twelve months ending As a percentage of As a percentage of net Financial category 12/31/93 average portfolio operating income Income Interest income 18 26 60 Interest expenses (12) -17 -40 Net interest income 6 9 20 Other income 24 35 80 Net operating in- 30 43 100 come Expenses Salaries Field staff 21 30 70 Headquarter staff 12 17 40 Total salaries 33 47 110 Administration 60 87 200 Depreciation 15 22 50 Provisions 0 0 0 Extraordinary 5 7 17 expenses Institution building 25 36 83 Total expenses 138 199 460 Net income (108) (156) (360) n.a. Not applicable. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. The income statement shows very high costs relative to income: • Total salaries (for field and headquarters staff and excluding expatriate salaries) account for 48 percent of the average outstanding portfolio. Expatriate-related expenses make up 36 percent of the average outstand- ing portfolio. Interest income does not even cover salary costs. * Administrative expenses account for 87 percent of the average outstand- ing portfolio.2 This is quite high considering that the project operates in only four provinces, especially compared with other microcredit projects, which have many more regional branches and yet manage to keep their administrative costs lower. Sahel Action: Projet de Promotion du Petit Credit Rural 239 Looking ahead Like many donor-funded projects, PPPCR was handicapped from the start by a charity-based mind-set that stressed social and economic impact over financial sustainability. However, with CFD funds projected to dry up in 1997, PPPCR staff are keenly aware of the need to move toward financial sustainability.3 With this in mind, tremendous efforts have been made over the past two years to increase operational and financial efficiency. The goal of PPPCR is to move from the status of a donor-dependent project to a financially sustainable institution. Toward that end, its short-term strategy is to deepen its growth and outreach by: * Expanding operations in the central region of Burkina, where group lending appears to be well suited to the hierarchical structure of the predominantly Mossi population. . Developing operations in geographical areas where there are few other donor credit programs so as to avoid duplication of activities as well as competition from other credit programs that may offer loans at subsi- dized interest rates. * Continuing to develop partnerships with other rural development programs to build credit and other rural development projects. For example, in the province of Soum, PPPCR cooperates with the Agence Francaise des Volontaires du Progrcs, which provides infrastructural assistance. Institutional challenges The institutional dilemma facing PPPCR as it moves toward autonomy revolves around the twin problems of legal status and capital. On the one hand, remaining a nonprofit organization would allow PPPCR to retain the operational flexibility and fiscal advantages conferred by its NGO status but would not resolve the basic problem of lack of capital (see the balance sheet in table 19.4). On the other hand, becoming a private, for-profit organization would require PPPCR to pay high import and other taxes (from which it is currently exempt) and to deal with the problematic issue of ownership. One option being considered is to become a private financial institution in which CNCA and a consortium of commercial banks are minority shareholders. The current thinking within PPPCR management, however, is that they are in a weak position to turn to the private sector, because the project has yet to show any profits. Financial challenges As the income statement shows, PPPCR is miles from achieving sustainability for two main reasons. First, it is too small. PPPCR'S lending portfolio is 240 Appraisals of Selected Microfinance Institutions Table 19.4 Adjusted financial statements of the PPPCR in Burkina Faso: Balance sheet, 1993 (millions of CFA francs) Financial category Amount as of 12/31/93 Assets Cash 65 Loan portfolio 95 (provision for loan losses) 0 Fixed assets (net) 33 Other assets 36 Total assets 229 Liabilities Payables 21 Lines of credit 138 Reserves 29 Others 20 Total liabilities 208 Equity Donors' equity For fixed assets 150 For operations 350 Total donor's equity 500 Members' equity Retained earnings (479) Total equity 21 Total liability plus equity 229 Note: The average loan portfolio in the last twelve-month period is CFAF 69 million. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. minuscule-CFAF 231.6 million ($463,200) out of a credit line of CFAF 250 million ($500,000). Moreover, its average loan size is very small, less than $50. Second, its source of funds is expensive. PPPCR pays the CNCA 13 percent interest on the credit line, which alone consumes two-thirds of current revenues. To reach sustainability, PPPCR must work simultaneously to increase revenues and reduce costs. Net income would need to increase by more than five times just to cover salary costs. To increase revenues, PPPCR can consider the following options: Sahel Action: Projet de Promotion du Petit Credit Rural 241 * Raise interest rates. This would be problematic, because the 20-25 percent real interest rates currently charged by PPPCR already exceed the 18 percent usury rate. As an NGO, PPPCR is exempt from many regulations affecting formal financial institutions. However, the new banking law requires it to abide by Central Bank-imposed interest rates, and manage- ment is concerned that it may not be in a position to negotiate an exemption. * Increase the loan portfolio. PPPCR has nearly reached the limits of its credit line (the outstanding portfolio is CFAF 231.6 million out of a CFAF 250 million line of credit). Therefore, the project will need to find additional sources of funds. One possibility is to obtain new lines of credit from either local banks or donors; the project is currently negotiating for an additional line of credit through the CNCA, but at a lower rate of 4-5 percent. Assuming approval, PPPCR would need to ensure that the increase in income generated by an additional line of credit is not offset by a corresponding increase in expenses. * Increase the size of loans. The average loan size is currently CFAF 24,000 ($48), and the largest loan available is CFAF 60,000 ($120). Increasing the size of loans would require PPPCR to move away from poor borrowers and to redistribute loans to better-off clients-the classic dilemma facing many NGOS caught between the need to cover the costs and the desire to reach the poor. PPPCR should also find out whether there is a demand for larger loans among its borrowers, and, more important, whether the income-generating activities of borrowers would enable them to repay such amounts. i Increase efficiency. Over the past three years, there has been a gradual increase in the productivity of agents: each agent managed an average loan portfolio of CFAF 6.5 million ($13,000) in 1994 compared with CFAF 2 million ($6,700) in 1992. This increase is due in part to the fact that many of the agents' functions have been delegated to the credit commit- tee, thus freeing agents to promote the project and recruit more borrow- ers. Management plans to have the credit committee follow up on loans (as it already does, jointly with the agent) and have agents work half of the time on promotion and half of the time on loan follow-up. In 1995, the average portfolio per agent is expected to increase to CFAF 9 million ($18,000). * Start savings activities that provide additionalfundsfor lending. This initiative obviously would only be profitable if interest rates on deposits are lower than the 10-13 percent PPPCR currently pays CNCA for the line of credit. However, the costs of managing these savings are not clear. Providing savings services would require considerably more work on the agents' part (recording deposits and withdrawals) and may require more than 242 Appraisals of Selected Microfinance Institutions the current once-a-week visit. Such costs could easily wipe out the difference between the interest rate that PPPCR pays savers and the interest rate that it pays to CNCA for its funds. * Reduce administrative costs. Increasing revenues alone will not allow PPPCR to cover its costs in the long term. The project must also look for ways to reduce operating costs, particularly administrative costs. Administrative expenses appear to be inordinately high, accounting for one-third of total costs. The problem posed by administrative expenses does not appear to be caused by salaries for field agents, which are relatively low. Further- more, productivity is very high (295 outstanding loans per agent on average as of 1994) and expected to increase as PPPCR delegates many of the agents' loan monitoring functions to the credit committee. However, a close look at audited financial statements indicates that expenses such as rent, communications (telephone and fax), and upkeep of agents' motorcycles (including gasoline) account for the largest part of adminis- trative costs. PPPCR should consider reviewing its expenditures to see where administrative costs could be cut. In conclusion, PPPCR has achieved a great deal in terms of scale and outreach, but its cost structure indicates that it is a very expensive small project. To reach sustainability, it will need to increase revenues and cut costs dramatically and aggressively. Notes PPPCR was evaluated in February 1995. Most of the information for this evaluation was drawn from discussions with project managers in Ouagadougou and field visits to the provincial unit of Yatenga. 1. The average loan size increased to CFAF 24,000 ($48) following devaluation of the CFA franc. 2. Administrative expenses include expenses related to rent, utilities, furniture, equipment, supplies, transportation (motorcycles for agents and office vehicles for other staff), communications (telephones and fax machines), visits to field units, hiring of an external accountant, and diagnostic studies. 3. The termination of funds was set for 1996, but a one-year extension was granted after the devaluation. 20 Programme Integre pour le Developpement de l'Entreprise Pierre-Olivier Colleye and Ousa Sananikone Progamme Integre pour le Developpement de l'Entreprise (PRIDE) is a five- year project aimed at supporting the development of small and microenterprises in Guinea through the provision of credit and training. PRIDE began operations in 1991 and is wholly funded by a $5.5 million grant from USAID. The project is implemented by Volunteers in Technical Assistance (VITA), a U.S. NGO. Within the development field, VITA is known as an effective provider of services to microenterprises in several Sub- Saharan countries, including Chad, where it serves as the financial intermediary in the microcredit component of the World Bank's Social Action Development Project. This project has two components: a microenterprise credit and training program and a separate entrepreneurship development program. Although partly modeled after the Grameen Bank in its solidarity-based lending approach, the microenterprise program is governed by a strong emphasis on training. Training is mandatory not only before credit is allocated but also throughout the duration of the loan. Between project start-up in 1991 and December 1, 1994, a total of 10,000 loans were made to roughly 3,505 borrowers, including 3,500 current loans. The outstanding loan portfolio is Guinean francs (GF) 597 million ($597,000).' The average loan size is $220. Approximately 70 percent of borrowers are women, and 90 percent are illiterate. As of December 1994, the project had developed six regional branches located in Boke, Kankan, Kindia, Kissi, Labe, and Mamou. PRIDE employs a staff of sixty-seven people, which includes thirty-one field agents (twenty-five agents and six branch chiefs) and ten staff (including two U.S. expatriates) at the central office in Conakry. The remaining staff include security guards, drivers, and maintenance personnel. With the first phase of the project ending in 1996, PRIDE is examining ways to move from being a donor-dependent project to being a self- sustaining, autonomous institution. Serious consideration is being given to turning the project into a local NGO capable of functioning over the long term with locally generated revenues under local management. 243 244 Appraisals of Selected Microfinance Institutions Program characteristics PRIDE has two components: a microenterprise credit and training program and an entrepreneurship development workshop. The microenterprise program was established with a grant of $5.5 million by USAID, including a pool of credit of $600,000.2 Initially implemented by the Council for International Development, the program has since changed hands and is being implemented by VITA. The program provides an integrated package of credit and training and an entrepreneurship development workshop, the Atelier sur l'Esprit d'Entreprise, for potential and existing entrepreneurs and managers of private and public enterprises. The credit program provides short-term loans to microenterprises. The lending methodology is based loosely on the solidarity approach used by the Grameen Bank. The target group is small enterprises that do not have access to other forms of institutional financing. Civil servants are not eligible for loans, and farmers are also excluded because the repayment schedule is not tailored to the agricultural cycle. Both existing and prospective entrepre- neurs can apply for loans. The initial step in the process is to establish a branch based on the demand for credit within the district and on the extent to which the region's social, economic, and physical environment is conducive to economic activities. Field agents first visit a neighborhood (known as a quartier) and meet with the neighborhood council of elders. A quartier generally has about 1,000 residents, and an agent usually is responsible for several quartiers. Informal meetings are then held in which agents present the principles, components, and terms and conditions of the program. Individu- als who join the credit program must pay a subscription fee of GF 2,000 ($2). To be eligible for loans, clients must first organize themselves into groups of five. The following conditions are required to form a group: members must not belong to the same immediate family or household, they must not work in the same business, and they must live in the same district. In the majority of cases, groups are formed before borrowers receive basic training. Groups can only be formed by individuals who have subscribed to the program by paying the fee. Group membership must be approved by the neighborhood council of elders who know the neighborhood residents. Each member of the group must pledge to guarantee the other members' loans and to repay his or her own loan on time. Similarly, groups in the quartier pledge jointly not to have late repayments in their district. Each group of borrowers must elect a new president every six or twelve months. After completing basic training, the client fills out a loan application and submits it to the neighborhood council of elders. Loans are formally approved at the central office in Conakry, but the loan decisions are effectively made at the branch level based on the council of elders' stamp of approval and the field agents' appraisal of the viability of the proposed Programme Intigr6 pour le Developpement de l'Entreprise 245 project. Compared with most microenterprise programs, the methods used by PRIDE to assess the viability of an enterprise are relatively elaborate in that they include a cash flow analysis and a balance sheet describing the assets and liabilities of each microenterprise. Loan processing takes about a week. In most cases, a branch disburses loans once a month. PRIDE provides six- and twelve-month loans. For six-month loans, repayments are due every two weeks, and first-time clients can borrow from GF 50,000 up to GF 250,000 ($50-$250). The same terms and conditions apply to twelve-month loans, except that repayments are due every four weeks instead of every two. Repeat borrowers can borrow up to GF 500,000 ($500). As a provision for loan loss, PRIDE takes 5 percent from the principal loan amount for two separate guarantee funds: 1 percent is placed in a solidarity fund that serves as an insurance fund for all borrowers in the same neighborhood and is used in case of death, accident, fire, major theft, and incapacitating illness, and 4 percent is placed in a guarantee fund. This is essentially a forced savings plan that yields 10 percent interest a year. This fund is intended to cover loan losses within the district after all recovery measures have been exhausted, and unlike the solidarity fund, it is returned to the client after the loan has been fully repaid. So far, it has not been necessary to use the guarantee funds because PRIDE is enjoying a 100 percent repayment rate. Until February 28, 1995, loans carried an interest rate of 36 percent a year (calculated on a declining balance). As of March 1, 1995, PRIDE lowered its interest rates to 24 percent and raised its fees by adding GF 650 ($0.65) to every repayment. As of December 1994, PRIDE had 3,500 outstanding loans with an average loan size of $220. Loans are increasing in size. PRIDE iS in the process of introducing medium-term loans for amounts ranging between GF 500,000 and GF 5 million ($500-$5,000), with terms of twelve to twenty-four months. These loans will be available for activities other than commerce and mining. Eligible borrowers will be established clients of PRIDE-either borrowers under the credit program or individuals who have completed the entrepre- neurship development workshop. Medium-term loans will have interest rates of 24 percent a year and processing fees of GF 5,000 ($5). As with short- term loans, 5 percent will be deducted from the loan principal (1 percent for the solidarity fund and 4 percent for the guarantee fund). PRIDE will pay annual interest rates of 10 percent on the guarantee fund, which will be returned once the loan has been repaid. A solid management information system allows PRIDE to recover full repayment of loans. Loans must be repaid before 10 a.m. on the due day. If the group fails to show up by that time, the agent in charge visits the group's president and inquires about the delay. Anyone making payments later than 10 a.m. must explain the delay to the branch chief who puts the individual's name in a red book. If even one loan payment is late, further loans to that 246 Appraisals of Selected Microfinance Institutions entire neighborhood are blocked until repayments are made, thereby placing significant pressure on borrowers to repay promptly. Savings There is no voluntary savings program at this time. Rather, the 4 percent of the loan principal that is deposited in the guarantee fund functions as forced savings. To date, the guarantee fund has accumulated GF 83 million ($83,000). PRIDE is considering creating a voluntary savings program as a means of increasing its capital base. Training PRIDE operates with the philosophy that successful use of credit depends on knowledgeable borrowers who are well trained not only in basic credit management but also in business development. Toward this end, it offers three types of training: basic training, ongoing training, and an entrepre- neurship development workshop. The first and second are mandatory for borrowers, and the third is a separate program.3 Once clients have paid a membership fee of GF 2,000 ($2) and formed solidarity groups, they must attend basic training to be eligible for a loan. Basic training seeks to familiarize potential borrowers with the terms and conditions of the credit program and to assess the credit needs of their proposed activities. Basic training consists of three to four sessions of about two hours each for classes of thirty to forty clients. After completing their training, participants-the majority of whom are illiterate-are tested to determine how much of the training they have absorbed. They must be able to sign their name, be fully familiar with the terms and conditions of the program, have identified an investment project, and have selected a loan package that is appropriate. Those who pass the test are eligible for a loan. Those who do not must take the test again in the following training session. If one group member fails the test, the whole group must wait until that person has passed. After obtaining a loan, new and repeat borrowers are required to attend training sessions once a month throughout the duration of the loan. The objective is to promote successful entrepreneurship by providing clients with basic financial skills such as accounting, budgeting, and marketing, helping them to resolve specific enterprise-level problems, and promoting successful management skills by learning from other entrepreneurs. Training courses are developed by branch agents, in conjunction with the branch support unit of the central office in Conakry, based on the specific concerns of their clients. Classes allow borrowers to share their experience and solve problems; they may also feature presentations by successful Guinean entrepreneurs or by local authorities on laws and regulations Programme Int6gre pour le Developpement de I'Entreprise 247 affecting small businesses. Strong emphasis is placed on adapting training to the changing needs of borrowers and to the environment in which they operate. PRIDE also offers a separate, for-fee entrepreneurship development workshop. This program was developed by the Washington, D.C., consulting firm of Management Systems International and is designed for existing and potential entrepreneurs and managers of public as well as private enterprises. Its objective is to provide an opportunity for participants to evaluate their potential to be entrepreneurs, strengthen their entrepre- neurial skills, and learn from the experience of others. This intensive workshop combines self-evaluation exercises, business management classes and simulations, and individual meetings with bankers. The workshop costs GF 200,000 ($200) per person. It lasts two weeks and is offered several times a year in various locations around the country. Space is limited to between twenty and thirty individuals per workshop, and participants are carefully screened. The fee includes all training materials as well as a mid-day meal, but other costs such as lodging and transportation are not included. This fee is subsidized by USAID and does not cover the full costs of training, which are $400-$500. Many participants are sponsored by their companies or by other donor-financed programs, and a few receive scholarships from PRIDE. The workshop has achieved significant results. It has the distinction of being the first and one of the few training programs in Guinea for which participants have been willing to pay out of their own pockets. To date, 540 individuals have completed the workshop. By most accounts, participants consider it to have been highly useful in imparting entrepreneurial skills, knowledge, and confidence. However, in a country with a per capita income of $500, individuals who can afford a $200 training workshop and take two weeks off from work are necessarily few. In fact, demand for the workshop has started to taper off, because most potential participants in the main geographical areas have been reached. In addition, other training institu- tions have begun to offer competing workshops. To expand revenues, PRIDE is planning to offer the workshop to medium-term borrowers and to repeat borrowers of the microenterprise credit program at a lower fee. It also has started offering its courses in local languages, making the programs accessible to the 90 percent of rural Guineans who do not read and write French. Governance PRIDE is implemented by VITA, a U.S. NGO based in Rosslyn, Virginia. Staff at the central office in Conakry include three managers (the expatriate VITA representative heading the project, the local director, and an expatriate technical assistant), four trainers, and three administrators. The central office 248 Appraisals of Selected Microfinance Institutions has five units: finance, administration, training, branch support, and the Business Excellence Center. PRIDE appears to be well organized and to have an adequate balance of centralized management and local decisionmaking at the branch level (see table 20.1). The management information and reporting system is good, and there is a high level of uniformity in accounting. Expatriate and local staff seem to work well together with a high level of mutual trust and respect. Table 20.1 Decisionmaking structure of PRIDE in Guinea Level Functions Local Council of elders Evaluate membership application (moral char- acter of applicants) Supervisory committee' Evaluate loan applications and monitor loan repayments in conjunction with field agents Field agents, including the Organize promotional meetings, register mem- branch chief bers, provide training (basic and ongoing), ap- prove loans based on the supervisory commit- tee's recommendations, and monitor loan re- payments National Management in the head of- Manage administrative, accounting, and train- fice ing services a. Members are elected by the presidents of borrowing groups in the neighborhood. At the branch level, local community involvement is provided by a council of elders that is formed in each neighborhood where the program is introduced. Council members are nominated by local leaders and authori- ties, and are not remunerated. They evaluate each application for member- ship based on their personal knowledge of the applicant's moral character (not his or her managerial competence or the viability of the enterprise, which is the task of the agent). Council members are not allowed to receive loans, but members of their families are eligible to borrow from PRIDE. The council usually has five to ten members, at least two of whom must be women. With guidance from the council of elders, a supervisory committee approves each loan and ensures that borrowers abide by the project's rules. Members of the supervisory committee are elected by all the solidarity group presidents in the neighborhood, and, like the council of elders, they are not remunerated. The committee generally consists of five members. Branch chiefs, who also function as field agents, have administrative responsibility for the branch. Field agents disseminate information to prospective clients, collect membership fees and loan application forms, Programme Integre pour le Developpement de l'Entreprise 249 monitor and supervise loans, and deliver basic and ongoing training. Field agents are usually recruited by the central office's branch support unit through radio announcements, which are a popular means of disseminating information in Guinea. All field agents are university graduates, but PRIDE does not require them to have financial or accounting backgrounds, and field agents are quite diverse in their skills. Applicants are screened through a general examination (held in French). Finalists are tested in the local languages, undergo two weeks of training, and are sent to the field for on- the-job training. In addition to offering relatively high salaries, PRIDE has established a comprehensive, and expensive, incentive system for its employees. Branches are evaluated every quarter according to the following criteria: repayment rates, quality and quantity of training provided, quality of accounting and reporting, level of productivity and initiative, and general office environ- ment. For example, performance criteria for training include whether at least 98 percent of borrowers at the branch are participating in ongoing training, the degree to which clients' needs and concerns are taken into consideration in planning the courses for basic training, and the extent to which branch staff stimulate client participation during training. Scores are given for each criteria, and the bonus is determined by each branch's total score. From July to October 1994, the average monthly bonus per agent was around GF 48,000 ($48). In addition, branch chiefs receive a "responsibility" bonus that averages $60 a month. Every quarter, a report from the central office showing the results of the evaluation is distributed to the branches, and it usually contains strong kudos for high-performing branches and gentle jibes at laggards, along with suggestions of how to improve perfornance. Performance As of December 1, 1994, PRIDE had made a total of 10,000 loans to 3,505 borrowers; 3,500 loans were outstanding. The number of clients grew 52 percent from 1992 to 1993 and 77 percent from 1993 to 1994. The depth of PRIDE'S outreach is evident in the fairly small size of the average loan ($220) and the fact that 70 percent of borrowers are women and 90 percent are illiterate. Although PRIDE does not specifically target the poorest of the poor, it does aim to reach the poor, in part by excluding civil servants from borrowing. The majority of loans are used to finance commercial activities, with the remainder distributed among small artisanal, transport, and construction activities. The outstanding loan portfolio amounted to more than GF 597 million ($597,000) as of December 1, 1994. Services appear to be adequately tailored to the nature and needs of the clientele. Branch offices are open every weekday from 9 a.m. until 4 p.m. In keeping with the illiteracy of most borrowers, loan documentation is kept very simple, with three or four basic forms that are easy to complete and 250 Appraisals of Selected Microfinance Institutions that neatly lay out the repayment installments (borrowers are already familiar with these forms). Loan processing takes a week at most. As of July 31, 1994, total interest, fee, and investment income covered 26 percent of operating costs, up from 17 percent as of July 31, 1993 (see table 20.2). There are no financial costs, because the pool of credit has been provided by USAID as a grant. However, when analyzing its own perfor- mance, PRIDE adjusts this cost recovery rate by adding a theoretical 20 percent cost of capital on the outstanding loan. If applied, the cost recovery rate (interest, fee, and investment income as a percentage of operating costs plus financial costs) becomes 23 percent in 1994, up from 16 percent in 1993. Another adjustment that can be made is for expatriate-related expenses. Taking out these expenses and not applying the cost of capital adjustment, the recovery rates become 30 percent for 1994, up from 21 percent in 1993. Results for the period ending in December 1994 are expected to be better because the pool of credit is growing, demand for credit and basic training is strong, and cash management has become more efficient. Provisions for loan loss (4 percent of portfolio) seem appropriate. Because of close follow-up by agents and a high level of peer pressure at group and neighborhood levels, repayment rates remain at 100 percent. Adjustment also needs to be made for fixed assets. Under USAID'S guidance, PRIDE listed all its investments in cars and office equipment directly in the income statement. Table 20.3 reflects a straight-line deprecia- tion with an average lifetime of five years. Because the program is only thirty months old, such an adjustment lowers administrative costs. However, these costs remain relatively high at 69 percent of average outstanding portfolio. The cost structure shows that salaries are extremely high. Including bonuses and medical expenses, salary costs represent approximately 90 percent of the average portfolio. In contrast, salary costs are less than 25 percent of average outstanding portfolio in most of the other programs surveyed. PRIDE employs university graduates with an annual average salary cost of $4,700 per person. In addition to performing relatively simple financial transactions (on average, 160 loans per agent), agents spend approximately 50 percent of their time providing the training services described above. An agent with 250 clients will generate GF 325,000 ($325) in training fees every four weeks and earn a salary, social security, and bonuses totaling about GF 425,000 ($425) during the same period.4 Looking ahead PRIDE has laid the groundwork for a sound microfinance institution: it has a well-trained staff, provides effective training and credit services, and features seemingly one of the best accounting controls in the West African region. Despite high costs, all trends remain positive: the number of clients Programme Intggre pour le Developpement de l'Entreprise 251 Table 20.2 Adjusted financial statements of PRIDE in Guinea: Income statement, 1993-94 (thousands of Guinean francs, unless otherwise noted) Twelve months ending 7/31/93 Twelve months ending 7/31/94 Asa Asa As a percentage percentage of percentage of of average loan average net operating Item Amount portfolio' Amount loan portfolio3 income Income Credit income 57,110 32 126,960 37 64 Fee income 8,970 5 8,624 2 4 Investment income n.a. 0 8,248 2 4 Net interest 66,080 37 143,832 41 72 income Other income 32,791 18 55,554 16 28 Net operating 98,871 56 199,386 57 100 income Expenses Salaries Field staff 59,431 33 177,844 51 89 Administrative 117,697 66 135,934 39 68 staff Total salaries 177,128 100 313,778 90 157 Administrative 172,423 97 240,191 69 120 Depreciation 6,631 4 20,399 6 10 Provisions n.a. 0 n.a. 0 0 Extraordinary 33,490 19 350 0 0 expenses Institution 196,168 111 191,033 55 96 building Total expenses 585,840 330 765,751 221 384 Net income (486,969) -274 (566,365) -163 n.a. n.a. Not applicable. a. Average loan portfolio was GF 177.5 million ($177,500) in 1993 and GF 347 million ($347,000) in 1994. These percentages of average portfolio do not give a complete picture of PRIDE'S efficiency, because they do not consider the growing portion of revenues coming from training services. Note: The average loan portfolio was GF 177.5 million ($177,500) in 1993 and GF 347 million ($347,000) in 1994. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. 252 Appraisals of Selected Microfinance Institutions Table 20.3 Adjusted financial statements of PRIDE in Guinea: Balance sheet, 1992-94 (thousands of Guinean francs) Amount as of Amount as of Amount as of Item 7/31/92 7/31/93 7/31/94 Assets Cash 18,101 290,826 435,238 Loan portfolio 30,971 278,713 447,176 (provision for loan - 26,070 53,459 losses) Fixed assets (net) 15,001 22,773 71,216 Other assets 2,072 2,223 7,169 Total assets 66,145 568,465 907,340 Liabilities Payables 1,983 5,691 4,375 Total liabilities 1,983 5,691 4,375 Equity Donors' equity For fixed assets 18,751 33,154 101,996 For operations 261,523 1,207,544 2,031,975 Total donors' equity 280,274 1,240,698 2,133,971 Other equity Reserves - 12,907 20,453 Retained earnings (216,112) (690,831) (1,251,459) Total other equity (216,112) (677,924) (1,231,006) Total equity 64,162 562,774 902,965 Total equity plus liabili- 66,145 568,465 907,340 ties - Not available. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. is growing, repayment rates are high, and the cost structure has been reduced. However, reaching sustainability by the end of the first phase (1996) is probably not realistic, and major improvements will have to be achieved if the project is to become a viable financial institution by the end of the second five-year phase (2001). It is significant that USAID covers 100 percent of operating costs. The entire fee and interest revenues ($307,000 so far) are used to increase the loan fund, not to pay for operating costs. This strategy is likely to continue as long as USAID agrees to it, and it benefits the project, because income increases the loan fund. In this sense, the net income in tables 20.2 and 20.3 is less relevant for PRIDE than it is for other programs. Instead, ways to increase revenues and decrease costs should be explored. PRIDE started with a relatively small loan fund ($600,000) that will need to be increased if the project is to cover its costs. USAID has increased this fund, and the Caisse Francaise de Developpement is expected to finance two Programme Int6gr6 pour le Developpement de l'Entreprise 253 additional caisses in Conakry, which would bring the total loan fund to $1.7 million. A fund of this size could increase total revenue more than 400 percent. Contrary to many microcredit programs that offer training services, PRIDE considers training as a major source of income. It intends to expand these services, because clients seem to consider training to be at least as important as credit and apparently are ready to pay for classes. Each training program is tailored to the requests of clients and is priced according to the budget of target clients. Prices vary from $5 for the two-day seminar on total quality management in Saguaro to $200 for the entrepreneurship workshop in Conakry. Increasing income alone will not allow PRIDE to cover its costs. It must also look for ways to reduce operating costs, and it has begun to do so. On the credit services side, PRIDE has been working to increase agents' efficiency. In the first years of the project, management emphasized quality rather than quantity. Now, it is motivating agents to increase their number of clients and is emphasizing productivity in an agent's evaluation. Between December 1, 1994, and March 10, 1995, the number of outstanding loans jumped from 3,500 to almost 5,000 without hiring new employees. The cost of the Conakry office is relatively fixed. This means that, as new branches open and revenues increase, the cost per dollar lent or per training delivered falls. This is very important for PRIDE'S future. Also, agents have now been trained as trainers. Each branch can therefore train its own clients without relying on the staff at headquarters. Staff salaries are set higher than market levels. If donors cut back their financing, staff salaries will fall. According to management, staff understand this and will accept cuts to safeguard their jobs and the program. PRIDE is in the process of opening a credit window for small businesses, offering loans up to $5,000 for terms of one to two years. These loans would be offered to current clients and would incur lower transaction costs per dollar lent, which would therefore increase profitability. Having clients visit the caisse twice a month for repayments and once a month for training allows agents to keep traveling costs to a minimum, while maintaining a close relationship with their clients (PRIDE bought three cars, but no motorcycles, for agents during the project). Finally, staff is well aware that some institutions have given up training because it is a heavy burden on their profitability. PRIDE argues that, in its case, linking training to credit is good business because it creates a niche that does not compete directly with other microfinance institutions, it provides additional revenue at low costs, and it meets a clear demand from the clients. The medium-term objective is to be a network of branch offices across the country that presents a rich variety of training and financial products, with each branch responsible for covering its costs and a share of headquarters' 254 Appraisals of Selected Microfinance Institutions costs. So far, all trends and strategies seem to point in the right direction, but many challenges lie ahead. The project is still young, and the next few years will tell how successful PRIDE is at managing growth, reducing costs, and graduating from a project into an institution. Notes PRIDE was evaluated in December 1994. The information presented here is based on discussions with project managers and field agents. 1. As of February 22, 1995, the number of loans had increased to 12,855; 4,815 loans are outstanding for a total of GF 759 million ($759,000). Conver- sions into dollars are based on a December 1994 exchange rate of 1,000 Guinean francs to 1 U.S. dollar. 2. This loan fund has increased since 1991: USAID'S contribution is currently $740,000 and is expected to reach approximately $1.2 million. The Caisse Franraise de Developpement will finance an additional F 2.05 million ($400,000) for two other caisses in Conakry. Also, interest and fees collected during the three years of operation, amounting to GF 307 million ($307,000), have been added to the loan fund. 3. Between December 1994 and March 1995, PRIDE introduced a series of one- or two-day workshops on issues such as the balance sheet, the legal rights and responsibilities of small enterprises, marketing concepts, and total quality management. 4. On average, each agent has 160 clients. However, three of the six caisses recently started operating (in February 1994), and they lower the overall average. Many agents actually have 200-350 clients. 21 Credit Rural de Guinee Pierre-Olivier Colleye and Ousa Sananikone Credit Rural de Guinee is a rural banking system that extends credit and mobilizes savings for the poor in Guinea. It was created by the Institut de Recherches et d'Application des M6thodologies de Developpement (IRAM), a French NGO, as a pilot project designed to test the methodology of the Grameen Bank in rural areas of Guinea. Credit Rural began operating in 1989 under the auspices of the Ministry of Agriculture and Animal Resources. Credit Rural is financed by the Caisse Francaise de Developpement, the European Development Fund, the African Development Bank, and USAID. Donor funds are received as grants by the head office in Conakry. As a donor-funded project implemented by IRAM, Cr6dit Rural has no legal status. The project currently has forty-one local branches or caisses distributed through each of Guinea's four regions: Guinee Maritime, Haute Guin6e, Moyenne Guinee, and Guinee Foresticre. Although each caisse is managed independently, the head office coordinates the operations of the caisses and provides centralized administration, accounting, training, and inspection. Six regional offices function as administrative and financial liaisons between the head office and the caisses. Credit Rural employs a staff of 120 people, including twenty persons at the head office and five expatriates (three at the head office and two in the field). Between 1989 and 1994, the number of clients grew from 250 to 31,000, 28,000 of whom are borrowers. As of December 30, 1994, the project had a lending portfolio of GF 2.0 billion ($2.0 million) and had accumulated CF 1.3 billion ($1.3 million) in savings, of which 25 percent are forced savings and the remainder are voluntary deposits. The average loan amount is GF 71,000 ($71). Half of the clients are engaged in agricultural activities-primarily the production, processing, and sale of agricultural products-and the other half are engaged in small trade and artisanal production. Client composition, to a large extent, reflects socioeconomic conditions in rural areas: half are women, and 70 percent are illiterate. As the project enters its fourth phase (1995-97), its objectives are to expand operations by increasing the number of caisses from forty-one to fifty-eight and the number of clients from 31,000 to 60,000. For the long term, Credit Rural's strategy is to become an autonomous, private, legal institution capable of operating independently of donor funds. The institutional 255 256 Appraisals of Selected Microfinance Institutions structure envisaged is that of a private bank in which each caisse is a shareholder, management is decentralized, and the government has a minimal role. Toward that end, the intermediary step scheduled for 1996 is to transform Credit Rural into a private institution. In addition, a foundation will be created as a separate entity managed by a board representing various donor agencies. This foundation will finance the private institution until it becomes self-financing. At that point, Credit Rural will operate as a private bank within which each caisse functions as a local branch. As of December 1994, amendments to the banking laws were being drafted to allow informal rural savings and loan institutions such as Credit Rural to become formal banks. Program characteristics Credit services include both short- and medium-term credit. On the savings side, Credit Rural offers both voluntary and forced savings. Training for clients is minimal. Credit services Small farmers, producers, and retailers in rural areas are eligible to apply for loans. Under its statutes, which are designed to limit lending to lower income groups, Credit Rural is prohibited from lending to civil servants and other middle-class groups. To be eligible for credit, borrowers must form groups of five members who guarantee each other's loans. Each of Guinea's four regions is divided into prefectures, which are in turn divided into subprefectures. Each subprefecture is made up of five to six districts (known as districts in rural areas and as quartiers in urban areas). In the case of Credit Rural, a caisse works at the subprefecture level and therefore encompasses several districts. In general, the aim is to open a caisse in relatively high-density areas with at least 10,000 inhabitants within a 13-mile radius. The first step in establishing a new caisse is for agents at the subprefecture level to conduct a general socioeconomic study of potential districts. Once a district has been identified and approved, caisse agents carry out a more detailed study of the region's geographic, economic, cultural, and religious conditions. Because borrowers attach great impor- tance to physical structures, the next step is to rent an office and install a safe. Next, caisse agents organize meetings with local authorities, including the council of elders, to introduce them to the guiding philosophy of Credit Rural and to explain the terms and conditions of the financial services offered. This is followed by a larger meeting for the general population. Additional, more focused meetings are then held to discuss the project in Credit Rural de Guinke 257 detail with interested individuals. These meetings usually take place in public places such as the mosque or market. The following stage is the formation of groups. Prospective borrowers must form groups of five, segregated by gender. Because caisse agents as a rule are outsiders to the region, they rely on the local knowledge of the council of elders, who review and approve all requests to form groups. Members of the council of elders are not remunerated for the duties they perform for Credit Rural. Once groups have been formed, prospective borrowers receive training from caisse agents that conveys basic lending principles such as the terms and conditions of the loan and the obligations of the borrower, familiarizes them with loan documents in the local language, and teaches them how to sign loan documents. On average, prospective borrowers receive about four training sessions, each lasting an hour and a half. A test is given at the end of training, and if one individual fails, the entire group is blocked from borrowing until that member passes the test. Those who pass the test can then submit a loan application. During the first six months of a new caisse's operations, loan requests are approved by the council of elders. Thereafter, a supervisory committee takes responsibility for approving and supervising loans. The supervisory committee is elected by the presidents of all the borrowing groups within the district and usually consists of five nonremunerated members, one of whom must be a woman.' Based on the committee's recommendations, caisse agents then decide whether or not to give the formal approval for a loan. As of December 1994, 63 percent of the money lent to borrowers came from mobilized savings. The remaining 37 percent came from a grant fund provided by the Caisse Francaise de Developpement and the European Development Fund, which the headquarters passed on to the caisses at 10 percent interest. As its name implies, Credit Rural primarily finances agricultural production and related microenterprise activities. During the rainy season (April to July), loans are used to finance agricultural production, including buying seeds, hiring labor to work the fields, and renting ox carts. In the dry season (October to May), loans are generally used to finance the processing and marketing of agricultural products such as kola and palm oil, as well as some artisanal activities such as ironworks. All loans are secured by group guarantees. Four types of short-term loans are available: * Rural credit (66.5 percent of loans) provides loans for agricultural and artisanal production and trade. First-time borrowers can obtain loans ranging from GF 50,000 to GF 150,000 ($50-$150) for a maximum term of twelve months at 36 percent annual interest. Loans are repaid in monthly installments, and because interest is waived for the first month, the loan 258 Appraisals of Selected Microfinance Institutions (principal plus interest) is reimbursable in eleven monthly repayments. If the first loan has been repaid satisfactorily, subsequent loan amounts can increase each year: up to GF 200,000 ($200) the second year, GF 250,000 ($250) the third year, and GF 300,000 ($300) the fourth year. * Rainy season agricultural credit (29.5 percent) primarily finances activities during the rainy months between April and July. The maximum loan amount is GF 80,000 ($80). It is offered under the same terms and conditions as rural credit, with a maximum term of twelve months and 36 percent annual interest. Loans are disbursed at the start of the rainy season, and repayments are due in January, February, and March of the following year, when prices rise during the dry season. Unlike rural credit, which is extended to single-gender groups of five, rainy season agricultural credit is available to mixed-gender groups of five to ten individuals. * Dry season agricultural credit (2.5 percent) provides loans for activities during the dry months between October and December. Repayments are due in two installments the following year, one in April (half of the principal) and one in June (half of the principal plus interest). Interest is 36 percent a year. * Commercial credit (1.5 percent) is a new credit line, with a maximum loan amount of GF 1 million ($1,000) and interest at 2.5 percent a month. Loans are disbursed in November and December, and repayments are due in July and August the following year. Because this credit is still at an experimental phase, the management has limited the number of such loans per caisse. Management believes that forty such loans per caisse would be sufficient to ensure the profitability of a caisse. In addition to short-term loans, Credit Rural also provides medium-term loans to borrowers who have repaid previous loans satisfactorily for two consecutive years. One credit line offers loans ranging from GF 150,000 to GF 350,000 ($150-$350) for a maximum term of twenty-four months at 30 percent interest a year. Repayments are due either every three months or annually. A second credit line provides loans from GF 500,000 to GF 2.5 million ($500-$2,500) to finance nonagricultural activities. Given the low income level of rural clients, the demand for such loans is modest. Of the 13,180 outstanding loans, less than 1 percent are medium-term loans, and they are used primarily to finance the purchase of carts, bicycles, or equipment for artisans. Repayment rates are reported to be around 98 percent, remarkably high considering the large size of the network. Staff attributes high repayment rates largely to the effectiveness of group lending. If overall repayment rates are lower than 90 percent, the district can be barred from additional lending, and its savings are temporarily blocked. The group can recover its savings Credit Rural de Guin6e 259 once the delinquent borrower has repaid his or her loan in full and resigns from the program. Savings services Credit Rural provides three main savings products: voluntary savings, forced savings, and safety deposit boxes for public institutions. Voluntary savings are available to all members. The savings program was established in 1990 partly in response to management's concern about inadequate capital funds. Many voluntary savers-that is, clients who save voluntarily as opposed to borrowers who are required to have a savings account-are civil servants (who are excluded from borrowing) and local institutions as well as low-income individuals. Savers can make deposits and withdrawals at any time; the only requirements are a minimum deposit and a balance of GF 2,500 ($2.50) at all times. Voluntary savings accounts yield 15 percent interest a year. Forced savings are basically a guarantee fund that provides for loan loss. Depending on the type of loan, a percentage of the loan amount is automati- cally placed into a forced savings account: in the case of rural credit, the amount is 5 percent for the first two years and 10 percent thereafter, and in the case of dry and rainy season agricultural credit, the amount is 10 percent. Borrowers cannot withdraw these funds until they have repaid their loans. As of December 1993, forced savings had accumulated a total of GF 213 million ($213,000). A safety deposit box is available for public institutions that wish to keep their cash in the safe of the caisse. Instead of receiving interest on savings, depositors pay an annual fee of 2 percent for the service (the fee includes record keeping). Funds can be withdrawn at any time, so they are not used to make loans. An additional financial service is known as remote withdrawals. Used primarily by rural merchants who travel to Conakry to buy products, this service allows them to make withdrawals at the head office in Conakry rather than at the local branch, thus eliminating the risk of carrying large amounts of cash on the road. The rules are simple: the client must maintain a minimum balance of GF 500,000 ($500) in a savings account, the with- drawal cannot exceed 50 percent of the savings balance, and the maximum withdrawal amount is GF 1 million ($1,000). Training The training offered to staff is more substantive than the training offered to borrowers. Agents are recruited through radio announcements, are screened through a test of general knowledge, and are trained for about a week in Conakry by headquarters staff and outside professionals (including bankers). They are then sent to local caisses for two- to three-month 260 Appraisals of Selected Microfinance Institutions internships, after which they return to the Conakry office for another phase of training (about fifteen days). This process is repeated once more with a different caisse to enable the trainees to work under different conditions and settings. The rule is not to assign agents to their district or region of origin as a means of avoiding conflicts of interest and pressure from family and friends, and agents are transferred every three years. Staff regularly undergo retraining, usually a series of twelve-week sessions covering accounting, management, public relations, evaluation of microenterprise projects, and bookkeeping. A study program in France is sometimes included for more advanced training in banking. Governance The overall structure of Credit Rural is that of a cooperative that belongs to its members. However, at the local level, the structure is not that of a cooperative (one member, one vote) but rather of an organization in which borrowers make management decisions and exercise voting power. The structure of governance shows a high level of ownership at the local level, in that decisions directly affecting clients such as group membership or credit approval are the primary responsibility of community members (the council of elders). As shown in table 21.1, the structure of decisionmaking is fairly centralized. Large issues affecting the project as a whole, such as interest rates and financial products, are handled at the head office under the direction of the steering committee. At the regional level, regional delega- tions provide monthly controls of books and monitor implementation of directives from the head office. This centralized structure is designed to ensure uniformity in accounting and reporting requirements throughout a large network. To motivate staff, Credit Rural provides bonuses based on performance, seniority, and the working environment. Every six months, agents are evaluated and rewarded for their performance using such criteria as the number of clients recruited, repayment rates, level of savings, quality of information provided to clients, control of costs, and the quality and regularity of financial reports. In addition, staff receive relocation expenses if they have to relocate far from home, a seniority bonus, transportation expenses, and a special bonus for "risks linked to the handling of money." Taken together, these bonuses add up to substantially more than the average agent's monthly salary and are highly valued. Performance As of November 1994, Cr6dit Rural counted 31,000 members, 28,000 of whom had borrowed at least once (see table 21.2). As of December 1993, Credit Rural de Guinee 261 Table 21.1 Decisionmaking structure of Cr6dit Rural de Guin6e Level Functions Local Council of elders Make loan decisions for first six months and are often consulted; after that, its members sit in on all the meetings, review application for group member- ship, and make recommendations to agents on loan applications Supervisory committee After its first six months, elect a management com- mittee to take over its functions, approve loans (take over from the council of elders), monitor loan distri- bution, follow up on overdue loans, serve as a per- sonal link between clients and the caisse, and orga- nize promotional meetings to recruit new members Caisse agents Represent clients at the district level and formally approve loans jointly with the supervisory or the management committee Regional Regional delegation Represent headquarters, serves as administrative and financial intermediary between caisses and headquarters, monitors caisses' books monthly, verify the contents of safes, transport funds to and from headquarters to caisses, provide necessary supplies to caisses, collect accounting records from caisses for headquarters, ensure that project policies are implemented at the caisse level, and assist with loan recovery if necessary National Management in head Manage the project, provide central administrative, office and steering com- financial, and training services for the caisses, and mittee govem the project more than 13,000 of these loans were outstanding. All borrowers are forced to save as a result of the automatic deduction for the guarantee fund. At year-end 1994, the project had a lending portfolio of GF 2.0 billion ($2.0 million) and had accumulated GF 1.3 billion ($1.3 million) in savings. Because Credit Rural operates exclusively in rural areas, it reaches many low-income groups that traditionally have been difficult for the formal financial sector to serve. The average loan amount is about $100. Half of the clients are women, and 70 percent are illiterate. Credit Rural does not lend to civil servants or to established merchants. Services seem to be adequately tailored to the needs of the clientele. In keeping with the small loans and the illiterate status of most borrowers, loan 262 Appraisals of Selected Microfinance Institutions Table 21.2 Basic performnance indicators of Credit Rural de Guinee, 1989-94 Indicator 1989 1990 1991 1992 1993 1994 Number of caisses 2 7 18 25 32 41 Number of members 110 2,425 8,280 14,986 23,666 31,000 Number of members 110 2,425 8,280 13,766 21,206 28,000 having borrowed at least once Rural credit Number of loans 110 2,200 7,221 9,642 14,330 n.a. Percent 100 91 87 70 67 n.a. Rainy season agricultural credit Number of loans 0 225 1,059 3,856 6,546 n.a. Percent 0 9 13 28 31 n.a. Dry season agricultural credit Number of loans 0 0 0 268 254 n.a. Percent 0 0 0 2 1.2 Commercial credit Number n.a. n.a. n.a. n.a. n.a. 400 Percent n.a. n.a. n.a. n.a. n.a. 1.5 Mid-term credit Number 0 0 0 0 176 n.a. Percent 0 0 0 0 0.8 n.a. Number of voluntary 0 0 0 1,220 2,460 3,000 savers n.a. Not available. application forms are relatively easy to complete. A major drawback of the project is the length of loan processing, which takes about forty days from the request to disbursement of the loan. This includes the filling out and submitting of loan applications; the review of loan applications by the council of elders, the supervisory committee, and caisse agents; basic training for borrowers; and requests for funds to Conakry. Once a loan has been formally approved, the transfer of funds from the head office in Conakry to the caisses takes approximately a week. As indicated in table 21.5, total interest and investment income for the twelve-month period ending December 1993 covered 24 percent of Credit Rural's operating costs (recalculated to include all costs for the five expatriates) or 22 percent of operating plus financial costs. Excluding the Credit Rural de Guinee 263 Table 21.3 Break-even analysis of Credit Rural de Guinee, 1993 (millions of Guinean francs) Item 1993 Break-even Interest Interest income 517 3,379 Interest expense (184) (1,203) Net interest income 333 2,176 Variable costs Salaries 98 641 Administration 99 657 Provisions 33 216 Total variable costs 230 1,503 Fixed costs Salaries 292 292 Administrative 295 295 Depreciation 86 86 Extraordinary 0 0 Institution building 0 0 Total fixed costs 673 673 Total costs 903 2,176 Net income (570) 0 expatriate-related expenses, the recovery rates become 36 and 32 percent, respectively. These numbers should be higher to reflect accurately the institution's actual situation. Indeed, until 1992, Credit Rural did not include expatriate- related expenses (which totaled GF 1.7 billion or $1.7 million) in its income statement. Instead, it listed these expenses in the balance sheet as an asset called technical assistance and amortized them over the years. However, in 1992 and 1993, with the agreement of the donors, the entire net amount was erased from the balance sheet and transferred back into the income statement (indicated as extraordinary expenses in table 21.4). When these extraordinary expenses are taken out of the income statement (because they relate to costs prior to 1992), the cost recovery rates without expatriate- related expenses become approximately 57 percent of operating costs and 48 percent of operating plus financial costs, as indicated in tables 21.3 and 21.4. Credit Rural has managed to keep salary costs (wages, bonuses, social security, and medical expenses) lower than other Guinean microcredit institutions, despite employing mainly university graduates. Still, salary costs account for 30 percent of the average outstanding portfolio, while net 264 Appraisals of Selected Microfinance Institutions Table 21.4 Adjusted financial statements of Credit Rural de Guinee: Balance sheet, 1992-93 (millions of Guinean francs) Item Amount as of 12/31/92 Amount as of 12/31/93 Assets Cash 1,342 1,420 Loan portfolio 764 1,308 (Provision for loan losses) 0 (31) Fixed assets (net) 668 790 Other assets 126 192 Total assets 2,900 3,679 Liabilities Compulsory savings 70 213 Voluntary savings 344 814 Payables 228 187 Lines of credit 608 648 Total liabilities 1,250 1,862 Equity Donors' equity For fixed assets 1,167 1,770 For operations 2,148 3,540 Total donors' equity 3,315 5,310 Members' equity Reserves 45 61 Paid-in capital 18 27 Retained earnings (1,728) (3,581) Total members' equity (1,665) (3,493) Total equity 1,650 1,817 Total equity plus liabilities 2,900 3,679 Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. interest income is only 25 percent of the portfolio. This means that unless salaries are reduced or agents' productivity is improved, Cr6dit Rural will not be able to cover its costs. Unfortunately, a reduction of salaries is unlikely, because caisse agents have recently requested a wage increase. Increasing the ratio of loans per agent might be possible, however, because agents currently handle an average of 140 loans each. Loan loss provisions seem quite low (none for 1992 and only 2.4 percent of the portfolio for 1993), considering loan defaults of approximately 2 Credit Rural de Guin&e 265 percent and a few isolated cases of fraud and theft. But provisions increased in 1994 and were likely to increase even more in 1995. Looking at the consolidated numbers at the caisse level, as of December 1994, the caisses covered 86 percent of their operating costs and 60 percent of their operating plus financial costs. They do not finance the costs of headquarters (apart from the 10 percent interest paid to headquarters on the line of credit). Financial results for 1994 are expected to be better than for all previous years due to the growth of the portfolio. Positive trends are likely to continue, because Credit Rural aims to establish an additional seventeen caisses by 1997. However, increasing the number of caisses by itself will not make the institution profitable. The number of clients per caisse and the number of clients per agent must increase as well. Additionally, Credit Rural should consider ways to increase the pool of money available for lending. Currently, it uses savings mobilization (on which it pays an effective interest rate of 14 percent a year) as well as an interest-free credit line from the European Development Fund. Credit Rural recently repaid a USAID line of credit that carried an annual interest rate of 17 percent because it was considered too expensive. The extent to which local commercial banks are willing to extend lines of credit to Credit Rural is not clear. Looking ahead As a network, Cr6dit Rural has grown rapidly, as measured by the number of clients and the range of financial products offered. In the process, it has managed to maintain good repayment rates, thanks mainly to the use of peer pressure (which lowers transaction costs). In addition, its savings facilities allow recycling of funds. Overall, Credit Rural seems to have appropriately balanced growth and control. The institution is too large and too complex for a simple yet reliable break-even analysis. However, we can make some assumptions and see how large the portfolio would have to be for the institution to cover its costs. The following assumptions are used only to illustrate what would happen in a best-case scenario: e Interest rates on savings and credit remain unchanged. * 75 percent of salary expenses are fixed (the increase in portfolio should be accompanied mostly by an increase in agents' productivity, not by the hiring of new agents).2 * 75 percent of administrative costs and 100 percent of depreciation expenses are fixed (more loans per caisse mean less traveling expenses per loan; the increase in administrative expenses would be due mostly to an increase in supplies and printing). 266 Appraisals of Selected Microfinance Institutions * The management of Credit Rural no longer needs the intervention of expatriates and can be transferred to Guinean personnel. Table 21.5 Adjusted financial statements of Credit Rural de Guinee: Income statement, 1992-93 (millions of Guinean francs, unless other- wise noted) Twelve months ending 12/31/93 As a As a percent- Amountfor the percentage of age of net twelve months average operating Item ending 12/31/92 Amount portfolio income Income Credit income 203 419 32 125 Investment 135 98 7 29 income (Interest expense) 31 184 14 55 Net interest 307 333 25 99 income Other income 8 2 0.1 1 Net operating 315 335 26 100 income Expenses Salaries Caisse staff 165 233 18 69 Headquarters 156 157 12 47 staff Total salaries 321 390 30 116 Administration 304 394 30 118 Depreciation 129 86 7 26 Provisions 0 33 2 10 Extraordinary 459 535 40 160 expenses Institution 450 750 57 224 building Total expenses 1,663 2,188 166 653 Net income (1,348) (1,853) -140 n.a. n.a. Not applicable. Note: In 1993, the average portfolio was estimated at GF 1.3 billion ($1.3 million). See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Using the 1993 cost structure, as indicated in table 21.3, interest income would have to become GF 3.37 billion ($3.37 million), up from the current GF Credit Rural de Guinee 267 517 million ($517,000), by the end of 1993 (see also table 21.5). To generate such income, the portfolio would have to be multiplied by 6.53. This calculation simply illustrates how difficult it will be for Credit Rural to break even. Management will probably have to review the program's cost structure and reduce salary and administrative costs. Credit Rural must (and intends to) apply new strategies to increase the productivity of agents and therefore the profitability of the caisses. These strategies include streamlining the procedures for requesting a loan, relying more on the supervisory committee to monitor loan performance (thus leaving agents more time to make new loans), and simplifying the savings system. Credit Rural might also start some "mini-caisses" in villages: these would be smaller caisses staffed with temporary and cheaper personnel who would be supervised by regular agents. Such changes could increase the loan-per-agent ratio. Finally, we must also hope for more transparent accounting statements at the headquarters' level if the institution eventually wants to be able to raise funds from the private banking sector (this is not part of the current plan). Although many changes in accounting procedures made in 1992 and 1993 seem appropriate (such as eliminating the $1.7 million technical assistance from the assets of the balance sheet), they are not properly reflected in the financial notes attached to the financial statements. In summary, the main institutional challenge for Credit Rural is to improve productivity both at the caisse and at the headquarters level in order to achieve financial sustainability. This means increasing the number of caisses and diversifying financial products to increase their cash income. It also means reducing the cost of caisses by increasing staff productivity and, more importantly, by decreasing the overhead of the headquarters. Notes Credit Rural de Guinee was evaluated in December 1994. Most of the information presented here is based on discussions with the project director, Jean-Paul Bustin, and on field visits. 1. After a caisse has been in existence for several years, the supervisory committee is replaced by a management committee that monitors loans. Unlike the supervisory committee, the management committee operates at the level of the subprefecture and supervises all the districts under the caisse. 2. To date, salaries per caisse have been mostly fixed. In 1990, with only seven caisses, salaries per caisse were GF 8.4 million. As of 1994, with forty- one caisses, salaries per caisse had increased only very slightly to GF 8.8 million. 22 Credit Mutuel de Guinee Pierre-Olivier Colleye and Ousa Sananikone Credit Mutuel de Guin6e was designed in 1988 and is implemented by the Centre International du Credit Mutuel (cICM) in France. It is based on the traditional credit and savings cooperative model developed in France and Germany in the nineteenth century. This project is financed by various donors, including the Caisse Franqaise de Developpement, the International Fund for Agricultural Development, and the World Bank. As of end-1994, the project had created seventy caisses with 45,000 members, 30 percent of whom are women. Most caisses operate in rural areas, mobilizing demand deposits and providing loans to individuals. The program was started in large part because of the dearth of financial services in rural Guinea. Since they have been privatized, commercial banks have progressively pulled out of the interior and now have only six or seven branches outside Conakry. This lack of access to credit and savings facilities is believed to have negatively affected the development of the rural private sector in general and of rural micro and small enterprises in particular. Credit Mutuel's first caisse was opened in 1988 in Lab6, in the middle of the Peuhl region. This region was selected because of its high population density and the hard-working character of its people. With thirty-three caisses, the Lab6 region is still the main focus of Credit Mutuel. However, the project also has expanded to regions around the towns of Fria, Gueckedou, Kankan, Kindia, Macenta, Mamou, N'zerekore, and Siguiri.' The project is now concentrating on strengthening its control and accounting systems and on increasing repayment rates. Also, it is preparing to become "institutionalized." It currently has the status of a project under the supervision of the Ministry of Agriculture but is starting legal proce- dures to become a credit union owned by its members. This shift in legal status could take place in the next few months. Program characteristics The decision to locate a caisse in a given village depends on the size of the village or town (minimum of 200-300 potential members) and the presence of a village market. The level of social cohesion of the village and its repayment history under former or current NGO credit programs also are 269 270 Appraisals of Selected Microfinance Institutions considered. Requests to open a caisse can come directly from a village or from a donor, if the caisse would complement another program financed by that donor. There have been instances in which a donor has imposed the opening of a caisse in a location that does not meet the program's eligibility criteria. In general, such caisses have remained unprofitable. Through radio advertising, Credit Mutuel calls for a large meeting in the proposed town. During this first meeting, staff explain the rules and benefits of the savings and loan cooperative to potential members and make an initial assessment of the demand for such a project. If enough people are interested, those attending will elect a steering committee with the advice of the local wise men and imams. This committee eventually will become the board. Credit Mutuel does not build a caisse immediately. It usually rents a room, equips it with a safe, and starts operations. Once the caisse has 300 savers and the staff know more about the local demand for credit and savings and the commitment of its members, Credit Mutuel constructs a caisse at an average cost of GF 20 million ($20,000), an expense that typically is financed by donors.2 Caisse members meet yearly in a general assembly to elect or reelect a board of twelve people (most of whom are literate) and to examine the progress of the caisse. Board members are nonpaid (see figure 22.1). They elect a president and approve loan applications up to GF 1 million ($1,000). The president of the board hires one or two caisse agents, who are remuner- ated by the headquarters in Conakry. Although size varies, on average, caisses have 642 members. Caisse employees keep track of transactions, compute trends, collect loan applications, report to the board, and follow up in cases of default. The seventy caisses are grouped into eight regions, each of which has a regional office. The regional offices serve as the link between the caisses and the Conakry headquarters. Regional staff typically include a regional officer, a secretary, two accountants, and one administrator. Their role is to collect monthly data from the caisses on outstanding loans, repayments rates, savings levels, and number of members and to report them to the headquar- ters in Conakry. The regional offices also deliver basic training to the caisse agents, provide caisses with supplies, shift funds from caisse to caisse, and approve loan applications between GF 1 million and GF 10 million ($1,000-$10,000). Directed by an expatriate from CICM (assigned to this post in fall 1994), the headquarters in Conakry has a staff of twelve who supervise the eight regional offices, assist with promotional campaigns, approve loans between GF 10 million and GF 20 million ($10,000-$20,000), centralize accounting, take care of general administration, and manage public relations with the donors that finance the program. Credit Mutuel de Guinee 271 Figure 22.1 Structure of Credit Mutuel de Guin6e Conakry headquarters hires remunerates and // \ ~~~~~~~~~~~~~~~~trains General assembly report to (one per caisse consisting of all members) 8 regional districts |ekcts \ir (+/- 40 staff) him Board and credd committee (1 board of 12 members 100 to agnt contiol ~~~~~per caisse) I ~~~~~report to Approve nmemberhip and loan I applications manage 70 caisses .~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Savings Large caisses are open every day, and rural caisses are open every other day. The procedure for becoming a member is relatively simple, and the initial membership fee is GF 3,000 ($3). Members can save at any time, but minimum transactions are set at GF 5,000 ($5). There are no term deposits, so individuals can withdraw their funds at any time. Savings mobilization started slowly for the first two years, reaching roughly GF 350 million ($350,000) by mid-1991. From then on, savings increased at a very rapid pace, totaling GF 2.7 billion ($2.7 million) by mid-1993 and GF 5 billion ($5 million) by the end of 1994. The average savings is more than GF 100,000 ($100) per member, and savings accounts yield 13 percent annually. All members of Credit Mutuel are active savers, but savings are highly concentrated. The bulk of savings (70 percent) comes from people with fixed income such as civil servants and large traders. According to available statistics, 20 percent of savers account for 80 percent of the total amount saved, with average savings of GF 450,000 ($450) per person, while the remaining 80 percent of members save an average of GF 28,000 ($28) per person. Market penetration is difficult to estimate, but Cr6dit Mutuel seems to have captured some important markets. For example in Labe, one of the larger towns in Guinea, 14 percent of the working population has a savings account with Credit Mutuel. This percentage is high because there are no banks in the area, the caisses protect the savers from theft and from social and family pressures, individuals must save for a minimum of six months before they can take a loan, and individuals must save a minimum of 50 percent of the amount that they borrow. 272 Appraisals of Selected Microfinance Institutions Another interesting service is the remote withdrawal, which allows merchants who have savings on account in their own village to withdraw money from the office in Conakry so as not to have to carry cash on the road. Credit Loans provided by Cr6dit Mutuel are used for agriculture (30 percent), commerce (30 percent), housing and consumption (30 percent), and handicrafts (10 percent). Borrowers must be members for six months before they can take their first loan, and they must have a minimum savings of 50 percent of the borrowed amount (30 percent for agricultural loans). In addition, 1 percent of the loan amount goes into Cr6dit Mutuel as paid-in- capital, which will be reimbursed once the member leaves the institution. Loan terms are up to two years, with an average of roughly twelve months. Annual interest rates are 30 percent, and the loan amount can range between GF 150,000 and GF 10 million ($150-$10,000). No information is available on the distribution of loans by size. Loans of up to GF 20 million ($20,000) can be made, but they are the exception. The average loan size is approximately GF 1 million ($1,000), much larger than for most rural institutions of this type.3 As with savings, credit is highly concentrated: the total outstanding portfolio of GF 2.5 billion ($2.5 million) represents only 2,800 loans. The evolution of these loans and other financial data are described below. Almost all loans are individual loans. Generally, applicants can request a loan anytime during the month. The application form, completed by the caisse manager or the accountant, indicates the profession of the borrower, the use of the loan, and the amount requested. Twice a month (on the fifteenth and thirtieth), the manager submits all applications to members of the credit committee, who decide which loans to grant based mainly on the character of the applicant. In principle, loans requested on the fourteenth of the month could be disbursed as early as the fifteenth. But on average, this process is longer. First, a caisse often reaches its credit limit (half of the savings mobilized), and the credit committee must wait for more repayments.to take place before additional loans can be made. In addition, for some donor lines of credits, each application must go to the donor for approval. This lengthy process handicaps borrowers, because opportunities often require relatively fast action. The average time between loan application and disbursement is roughly two to three weeks. The issue of individual versus group loans is being reviewed. Credit Mutuel would like to encourage more group loans, because it believes that peer pressure would improve the on-time repayment rate (currently 60 percent). Credit Mutuel considers a payment as late when it is more than one installment overdue. Using this definition, the on-time repayment rate Credit Mutuel de Guinee 273 is 80 percent. It also seeks to increase the participation of women (currently 30 percent of membership) who have proven to be better repayers than men. These will be priorities for the new management. Training Training of Credit Mutuel staff is insufficient. Most instruction takes place on the job and through interventions by regional offices. Management believes that, because agents deal with relatively large transactions, proper training is critical to keeping good records and maintaining high repayment rates. Donors, such as the Caisse Franqaise de Developpement, have financed training programs that have sent agents and caisse managers to Lab6 (where the headquarters was located until 1994), Burkina Faso, and France for up to two months. The new director of Credit Mutuel believes that a better option might be to conduct a two- or three-week training program in Guinea with additional short review courses throughout the year. Management also believes that proper training would inspire a stronger sense of commitment on the part of agents and therefore reduce theft and fraud. Training for borrowers is limited to the initial meeting before the caisse is created and to a basic orientation about the operations of the credit and savings program, the obligations of borrowers and savers, and the importance of loan repayment. Occasionally, additional meetings are held for members. Performance The number of members has grown tremendously in the past few years. As indicated in table 22.1, both the number of caisses and the number of members per caisse have increased. Credit Mutuel is planning to have eighty-seven caisses by the end of 1997. The growth in the number of members is impressive but, as previously indicated, their actual level of participation is lower than it appears: 20 percent of savers account for 80 percent of total savings, and only 6 percent of members borrow. The four caisses that are located in larger towns (Gueckedou, Kindia, Kissidougou, and Labe) account for 84 percent of the outstanding portfolio and 50 percent of total savings. The other caisses are located in much smaller villages, where people have less money to save and are less likely to borrow. Credit Mutuel's strategy is to concentrate on fixed-income members for savings and on well-off merchants for loans, because larger transactions have lower costs. Competition is not really an issue, because the only alternatives in the villages are to keep the money at home and to participate in tontines that do not yield interest. By mobilizing more savings, shortening 274 Appraisals of Selected Microfinance Institutions Table 22.1 Growth of Credit Mutuel de Guin6e, 1989-94 Indicator End-1989 End-1990 End- Mid-1992 Mid-1993 End-1994 1991 Number of caisses 16 29 46 52 62 70 Number of members 6,500 12,000 19,800 27,000 34,500 45,000 Number of members 406 414 430 519 556 643 per caisse Number of outstand- 9 176 712 1,375 1,947 2,800 ing loans Outstanding loans 1.2 32.3 244 855 1,277 2,500 (millions of Guinean francs) Outstanding savings 25 175.4 556.8 1,781 2,695 5,000 (millions of Guinean francs) the processing time, and tailoring credit services to a poorer segment of the population by offering shorter-term loans and decreasing the minimum loan amount, currently at GF 150,000 ($150), the program is likely to enlarge its client base and improve its repayment rates. There is little incentive for Credit Mutuel to implement such changes. The program is trying to increase revenues, and it is not inclined to deal with small savers and borrowers, who are the most expensive to serve. Some remote caisses have reached all potential clients and yet are still too small to break even or to share the costs of the regional branches; closing down these unprofitable caisses is being considered. From the large size of the average loan, Credit Mutuel appears to have attracted clients who formerly dealt with commercial banks when these were still serving rural areas. Now that most of these banks have pulled out, Credit Mutuel has moved in as a substitute. However, if the mission of Credit Mutuel goes beyond replacing commercial banks to encompass poverty alleviation, it will have to increase the number of small accounts and decrease the costs of the smaller caisses. Opening remote caisses once a week would probably be sufficient. Caisse agents either could be rotated to different caisses or could work only once a week in their own villages. Credit Mutuel does not want to leave out small rural borrowers, yet it knows that dealing with smaller accounts would make it even more difficult to break even. Therefore, management now has a strategy of focusing on agricultural group loans: "We are a network of caisses geared to agricultural activities. Agricultural group loans would help us on two issues: (i) group loans are likely to show higher repayment rates, and (ii) smaller borrowers can get together to take a larger loan from Credit Mutuel." Credit Mutuel de Guin&e 275 Credit Mutuel might also renegotiate with donors the requirement for certain loan applications to go through the Conakry headquarters and the donor's representative, a practice that does not foster autonomy. Reducing the time needed to disburse loans might prove crucial in increasing the proportion of active members. As of mid-December 1994, no official balance sheet or income statement had been issued since the beginning of Cr6dit Mutuel's operations. The new management realizes that this is a serious problem, and it has been working on financial statements for calendar year 1993. When this project was evaluated by the World Bank team, financial statements had not been completed. The financial analysis included here is based on a draft of those documents and on other financial statistics made available to the authors (see table 22.2). The numbers in the following tables, therefore, are only indicative of Credit Mutuel's performance. No financial trend analysis was possible because only 1993 data were available. Total interest income for Credit Mutuel for the period ending December 31, 1993, covered 34 percent of operating costs (including the cost of five expatriates and all fixed assets) or 26 percent of operating plus financial costs. Deducting the cost of expatriates, total interest income covered 87 percent of operating costs and 47 percent of operating plus financial costs. Cost recovery rates that include financial costs could be higher than they are. They are low because of two conservative policies: * Members must save a minimum of 50 percent of the amount they borrow (in the case of agricultural loans, only 30 percent of the loan amount must be in the savings account; some donors also lower this requirement in specific cases). This means that, for any particular loan, Credit Mutuel has to pay 6.5 cents (the 13 percent savings rate on 50 cents) on every dollar it lends. As of March 15, 1995, Credit Mutuel discontinued paying interest on these compulsory savings. * Credit Mutuel's total outstanding portfolio is never to exceed 50 percent of total savings. Such a policy means that if the interest rates on savings are half the interest rates on credit and the outstanding savings are twice the outstanding credits, the net interest income has to be zero, and there is no other income to cover operating expenses. This approach is not sustainable if Credit Mutuel wants to become profitable. The first policy is a convenient tool to ensure partial collateral and to increase the pool of money available for credit, but it might prevent some members from borrowing. Lowering this 50-percent limit presumably would increase the number of loans and the revenues they generate. But there is little incentive to lower it so long as Credit Mutuel is not willing to lend a higher proportion of savings. Therefore, the two policies must be changed simultaneously in order to stimulate demand and generate a larger 276 Appraisals of Selected Microfinance Institutions Table 22.2 Adjusted financial statements of Credit Mutuel de Guinee: Balance sheet, 1993 (millions of Guinean francs) Item Amount as of 12/31/93 Assets Cash 2,167 Loan portfolio 1,567 (Provision for loan losses) 0 Fixed assets (net) 681 Other assets 537 Total assets 4,961 Liabilities Savings 3,910 Payables 425 Total liabilities 4,335 Equity Donors' equity For fixed assets 1,680 For operations 3,120 Total donors' equity 4,800 Members' equity Reserves 14 Paid-in capital 43 Retained earnings (4,231) Total members' equity (4,174) Total equity 926 Total equiti plus liabilities 4,961 -Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. spread between effective interest income (currently at 22 percent of average portfolio) and interest expense. It also should be noted that operating expenses, excluding the cost of expatriates (which accounts for approximately 60 percent of total operating expenses), represent 48 percent of the average portfolio (see table 22.3). This proportion is too high if the institution is to become sustainable. Indeed, it means that even if Credit Mutuel lends all the funds mobilized from members it would have to charge 44 percent above the interest rate paid on savings to break even. Efforts will have to be made to contain costs, which should be possible because the average portfolio could increase without having to increase the number of caisses or members per caisse. Credit Mutuel de Guinee 277 Fixed assets seem to be depreciated properly. They include all fixed assets, including cars, buildings, computer equipment, and safes. The return on treasury-the return on the money that is not being lent-is low. Only half of savings are lent to borrowers; 20 percent stay in the caisses in the form of cash, yielding no interest. The management would like the caisses to have less cash in their safe and more with the commercial banks. Unfortunately, poor road conditions make the transfer of funds very difficult. This leaves only 30 percent of the portfolio to be deposited with commercial banks at an interest rate of 15 percent. Credit Mutuel therefore must manage its cash flow better so as to leave as little money as possible in the pipeline and to increase the effective interest income. Repayment rates are low. More than 40 percent of loans are not repaid on time, and more than 20 percent are not repaid after thirty days. Overall repayment rates cannot be accurately computed with the data currently available, because management believes that all loans eventually will be repaid and therefore never writes them off. For this reason, no provisions are made for doubtful accounts. Roughly half of borrowers repay their loans without prompt- ing. This means that a lot of time and money are spent going to borrowers' houses to claim repayments. Enforcing repayment is a difficult task, because the legal system does not work effectively, loans are not secured with collateral, and the number of group loans is negligible. There are several reasons for low repayment: most loans are individual loans and therefore cannot be recovered through peer pressure mechanisms, tracking methods are weak (books are controlled only at the end of the month), and membership is open to everyone, including spouses and other relatives of the members of the board and of the employees. Clients with close connections to the caisse reportedly never repay on time. The management commented that most people who do not repay have the ability to repay but simply do not bother. This might indicate little sense of ownership among members. The lack of provisions for loan loss does not seem wise, and the new project manager is likely to set up provisions in 1995-96. Such provisions are desirable, and management might consider writing off loans that are more than ninety days overdue. This would reflect more accurately the actual value of the loan portfolio and the cost of operations. The head office in Conakry has had a policy of minimal interference at the caisse level. At this point, however, the management realizes that it will need to intervene very quickly to solve a series of pressing issues, including the overabundance of individual loans, poor accounting, low repayment rates, theft, and low "active" membership (active members are the members who actually save or borrow). This process could lead to a reallocation of responsibilities within the institution. 278 Appraisals of Selected Microfinance Institutions Table 22.3 Adjusted financial statements of Credit Mutuel de Guinee: Income statement, 1993 (millions of Guinean francs) Item Amountfor the twelve As a percentage of As a percentage of months ending 12/31/93 average portfolio net operating income Income Credit income 244 22 1,626 Investment income 216 20 1,440 (Interest expense) 450 41 3,000 Fees 5 0 33 Net interest income 15 1 100 Other income 0 0 0 Net operating in- 15 1 100 come Expenses Salaries Caisse staff Headquarters staff - Total salaries 159 14 1,060 Administration 219 20 1,460 Depreciation 124 11 827 Provisions 0 0 0 Extraordinary ex- 32 3 213 penses Institution building 807 73 5,380 Total expenses 1,341 122 8,940 Net income (1,326) -121 n.a. n.a. Not applicable. - Not available. Note: The average loan portfolio was estimated at GF 1.1 billion ($1.1 million) for 1993. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Finally, such a large institution with total assets close to $5 million should have proper accounting mechanisms. Cr6dit Mutuel does not at this stage, and such mechanisms are urgently needed, particularly if the project intends to become a registered credit union. Looking ahead Credit Mutuel has experienced tremendous growth during the past four years, and it has built an infrastructure with relatively high potential. Credit Mutuel de Guinee 279 However, it now needs to take advantage of its large membership and good infrastructure to make a definitive movement toward sustainability. Necessary changes might include: * Increasing the number of active members (by reducing the savings requirements and by tailoring products to lower-income members). * Increasing the number of group loans and the number of female members (to increase repayment rates). * Increasing the quality of control methods and cash management (to reduce waste). * Loosening some of its conservative policies (by lending a larger share of savings). * Publishing complete and accurate financial statements (to detect problems early and attract potential donor funds). The new management is aware of these problems and intends to tackle the most urgent ones in the next few months. It would be very useful to monitor the evolution of this institution. If changes are not implemented, Credit Mutuel might run into the same problems experienced by commercial banks in rural areas, such as lack of profitability and exclusion of the poor. If proper policies are put in place, this program could evolve into a sound, rural financial institution. Notes Credit Mutuel de Guinee was evaluated in December 1994. The information presented here is based on discussions with the project director, Jean-Claude Noir, and with donors financing this institution. 1. To reduce costs, the number of regional offices soon will be limited to four, located in the towns of Fria (Guinee Maritime), Gueckedou (Guin6e Foresticre), Kankan (Haute Guin6e), and Lab6 (Moyenne Guinee). 2. The exchange rate used throughout this evaluation is GF 1,000 to 1 U.S. dollar. The World Bank, for example, has committed $44,000 for construc- tion of a caisse in Mamou. Because of slow disbursements, only the foundations were built two years ago, and Credit Mutuel is waiting for further financing. In the meantime, the Mamou caisse is renting a room nearby. This reportedly is problematic, because people seem to attach great importance to physical assets, and potential clients are not willing to deposit their savings with an institution that does not have a building. 3. This is significantly larger than the average loan size of the other eight institutions reviewed. Credit Mutuel still qualifies as a microfinance provider in that its large number of savings accounts average $100 per account. 23 Kafo Jiginew Doug Salloum A jiginew in the Bambara language is a granary. Kafo Jiginew, which means a union or federation of granaries, is the central credit union for a network of small rural savings and credit unions established in the cotton region of southern Mali. The word jiginew was chosen for the credit union system because for the farmers of the region the word connotes security and wealth-two key themes of the kafo system. Kafo Jiginew and the individual credit unions or caisses have the standard structure of a savings and loan cooperative wherein the members and owners are also the clients. The members own the individual caisses, and the caisses in turn own the kafo. Kafo Jiginew plays the traditional role of a central credit union by investing or redistributing surpluses from the network. In addition, it acts as a central service and purchasing agent for the network, providing accounting, inspection, and training services as well as printed materials and office supplies to the local units. Kafo Jiginew began operations in the summer of 1988 with a ten-year funding commitment from a group of donors. The donors supporting the institution include: - The European Community (50 percent). • A consortium of NGO donors (50 percent): Comite Francais Contre la Faim, sos Faim (Belgium), Manitese (Italy), and Deutsche Welt- hungerhilfe (Germany). Until October 1994, technical support was provided by the French Fondation du Credit Cooperatif. Kafo Jiginew has also received important assistance from CMDT-Mali's cotton marketing board-including experi- enced field staff and extensive use of CMDT's existing network of village- based organizations. The economy of the region in which Kafo Jiginew operates is based almost entirely on the production and sale of cotton. As well as being a development institution for the region, CMDT purchases all of the cotton produced and manages the purchase price in a manner that attempts to soften some of the price swings that occur in international markets. Essentially all of Kafo Jiginew's members are either directly or indirectly 281 282 Appraisals of Selected Microfinance Institutions dependent on the cotton crop. The quality of its lending and the stability of its base of deposits are therefore very dependent on cotton sales and prices. In October 1994 Kafo Jiginew had a confrontation with its donors that resulted in a decision by its board of directors to remove the French general manager, the Malian assistant general manager, and the French manager of computer systems. Kafo Jiginew has described this decision as an attempt by the members and the board to assert control over their own institution. The eventual transfer of all operating responsibility to Malian management and directors was always the ultimate goal of all participants-donors, expatriate staff, and local staff alike-but as a consequence of the confronta- tion, this process proceeded somewhat more precipitously than planned. It would appear to be in the best interests of both the donors and Kafo Jiginew, however, to resolve any difficulties and to confirm funding commitments and objectives for the remaining three years of the program. Kafo Jiginew still requires donor funds, because it is not yet financially self-sufficient. To date the donors have committed themselves to fund the program in three phases: September 1987-August 1990, September 1990-August 1994, and September 1994-August 1997. The original budget for each phase of the program was CFAF 450 million ($1.5 million).' Prior to the split with the donors but after the devaluation of the CFA franc in January of 1994, Kafo Jiginew was advised by the donor coordinator that the budget for the last phase would be twice the previous levels. The original scope of the program was to develop eighty caisses serving 550 villages and having a membership of 23,000. It now appears that Kafo Jiginew will halt its expansion at fifty- one caisses serving 450 villages but still having at least 20,000 members (the situation as of September 1994). Membership can be expected to grow in the existing 450 villages (potential membership is estimated at 45,000 persons or approximately twice the current level). Much of this increase in member- ship in existing villages will come from the increased participation by women, who now represent only 5 percent of members (in September 1994, up from less than 1 percent in 1993). Program characteristics An individual jiginew or caisse can only be created where CMDT operates. The development of a caisse is based on an existing grouping of villages called a zaer. As a result of this initial grouping of six to ten villages, the total population potentially served by each jiginew averages 4,000-5,000 persons, 500 of whom are farmers and heads of families. These are the primary target members for the Kafo Jiginew system. Potential members learn about the requirements for establishing a credit union from Kafo Jiginew's training staff and then complete these prerequisites on their own. A basic requirement is construction of a building to house the new caisse, according to a standard 5 meter by 6 meter plan. The cost of this building is borne by the members. Kafo Jiginew provides a steel safe, furnishings, and Kafo Jiginew 283 paper supplies for the caisse once it is constructed. The value of this contribution roughly matches the cost of the building, approximately CFAF 500,000 ($1,000). The caisse is usually built in a centrally located village serving several other villages in the zaer, often in the village that contains the weekly market for the area. Members join their local jiginew by paying a one-time capital subscription of CFAF 5,000 ($10). To date, most members have been men because men dominate the cotton farms that underpin Kafo Jiginew's development. Women are increasingly being targeted as new members, and they are allowed to join for as little as CFAF 1,000 ($2). They are expected to increase their membership subscription over time, as they earn profits from the loans they receive. The complete Kafo Jiginew system consists of four levels of management or governance: * The jiginews. Each caisse has a board of directors consisting of two persons from each participating village. The board selects two persons from among the villagers, with the required level of literacy and numeracy, to be cashiers who will work two days a week for the jiginew. A supervisory committee and a credit committee are also formed from members of the board of directors. Kafo Jiginew trains the board members, the cashiers, and the individual committee members in their various responsibilities. Training for credit activities is the last of the series of courses and only takes place following a successful period of savings activity-generally six months after start-up. * The unions. Five to eight caisses form a union, which has no physical location. Representatives from each caisse in a union meet at selected villages once every three months to discuss common issues. A fieldworker is assigned to each union. These fieldworkers are selected from among the most experienced cashiers, and they work full time supervising and controlling the quality of deposit books and loan records for the caisses within their union. They are expected to visit each caisse at least twice a month. * Regional offices. Seven fieldworkers are stationed at regional offices in Fana (two), Koutiala (two), and Sikasso (three). Each regional office will soon have one more fieldworker. At the three regional offices, four supervisors oversee the fieldworkers. * The kafo level. Head office staff (in Koutiala) consist of the following individuals: a general manager, a manager of training, one female fieldworker who concentrates on increasing the number of women members, two accountants, one cashier/secretary, and five guards and chauffeurs. 284 Appraisals of Selected Microfinance Institutions Savings services Only members can save in the caisses, and savings are completely voluntary, although development of members' savings is a prerequisite to initiating lending activities in an individual caisse. Minimum savings amounts have been set as multiples of CFAF 1,000 to simplify record keeping and calcula- tions. Two savings products are available to members: demand deposits that pay 4 percent annually and one-year term deposits that pay 6.5 percent annually. The success of savings mobilization has been mixed. Virtually all members have demand deposit accounts, but these balances fluctuate significantly over the year, declining to their lowest levels during the dry season. During 1993 the absolute level of these demand deposits varied from approximately CFAF 80 million ($267,000) to CFAF 240 million ($800,000). Demand deposits as a percentage of total deposits are high, but slowly declining. As of September 30, 1993, demand deposits represented nearly 75 percent of all deposits in the system. By September 1994 this ratio had fallen to 65 percent. Meeting the members' withdrawal requirements and, at the same time, funding loan disbursements have proven to be a challenge. In a diversified and healthy economy, demand deposits are a relatively stable source of loan funding, but in this cotton-growing region, this apparently is not the case. As a result, further development of term deposits is critical. Term savings are growing steadily in total volume, increasing from CFAF 64 million ($213,000) in September 1993 to CFAF 188 million ($376,000) in September 1994, but only 10 percent of Kafo Jiginew's members have term deposits. The volume of term deposits in the system depends as much on the success of the cotton harvest as it does on factors that Kafo Jiginew can influence, such as member confidence and commit- ment. Credit services Only members can receive loans from the caisses, but anyone can be a member by paying the capital subscription. Loans are provided by each local caisse without standardized credit approval or evaluation procedures. There is no minimum savings requirement to qualify for a loan, for example, nor is a strict distinction made between social and economic uses of the loan proceeds. The role of the credit committee of each jiginew is to allocate annual credit funds provided by the unions, regions, and Kafo Jiginew central more than to analyze credit applications. Kafo Jiginew offers three basic types of loans (see table 23.1):2 * Rural credit is provided to assist farmers during the dry season. These loans are disbursed in the summer months and are repaid when the farmers receive cash from CMDT for their cotton harvest, usually beginning in March of each year. A typical practice among borrowers is Kafo Jiginew 285 to draw down a loan and deposit the proceeds or a portion of the proceeds in their demand deposit accounts to ensure access to funds during the dry season, when there is very little cash in the local economy. Loan terms average nine months, and the interest rate is 2 percent a month (1.5 percent prior to February 2, 1994). No interest or principal payments are required while the loan is outstanding. Rather, all accrued interest and principal due are repaid at one time at the term of the loan. Borrowers provide personal guarantees and collateral in support of their loans even though neither form of security is legally enforceable. Villages are expected to collect on any unpaid loans, and they reportedly will resort to taking security offered by the borrower. * Short-term credit is provided for profitable commercial activities for periods of one to three months at an interest rate of 4 percent a month. Interest payments are due at the end of each month, and repayment is due in full at the term of the loan. Security features are similar to those of rural credit. * Equipment loans are made available to purchase farm equipment on terms of up to three years. Annual repayments of principal and accrued interest are required. Interest on these loans is only 1.2 percent a month. Security is the equipment plus the same items as for the other credits. Table 23.1 Characteristics of loans offered bv Kafo Jiginew in Mali, 1992-93 Percent- Number of age of Average loan Loans outstand- Percentage of loans dis- loans size (CFA ing (millions of loan Type of loan bursed disbursed francs) cFAfrancs)Y portfolioa Rural credit 7,056 76 31,000 231 95 Short-term 2,231 24 26,000 5 2 credit Equipment 33 - 53,000 2 1 credit - Not available. a. As of September 1993, CFAF 3 million or 2 percent of the loan portfolio was outstand- ing in loans to staff. Governance Kafo Jiginew belongs to its members, the farmers of the CMDT region of Mali. All decisions ultimately devolve to the members through their elected representatives at various levels. All members in each jiginew elect two members from each village to the jiginew board of directors. These board members have three-year terms, and a third of the board is elected each year. Each jiginew also appoints two persons to the union level, which, in 286 Appraisals of Selected Microfinance Institutions turn, names two representatives to the board of directors. Each jiginew subscribes 40 percent of its own capital to the capital of its respective union. The cashiers at the local level are responsible to both the jiginew board of directors and to the kafo central staff through the fieldworkers and the kafo inspector. The kafo staff ensure the quality and uniformity of the cashiers' work. The union-level board meets every three months with the fieldworkers and can request their assistance in particular areas. The union is the level of the structure for training, group development, and local representation. All financial management occurs at the kafo level. Members of all boards are volunteers. The expenses of board meetings are covered by kafo central but are charged back to the caisses. The cashiers are paid by each caisse on a commission basis at the rate of 10 percent of interest income earned and 20 percent of deposit interest paid by the jiginew. Staff costs, from fieldworkers to expatriate staff, as well as administrative costs, are accounted for at the central level and then allocated down to the jiginews. The concept is that all jiginews should be able to cover all of their costs, including financing costs, by the end of the fifth year of operation. The whole project is to be completely self-financing by the end of the donor funding period: the fall of 1997. The structure has been devised to ensure uniformity in accounting, reporting, and pricing, while at the same time giving members an important degree of control over the issues that most concern them, such as credit approval and loan recovery. Without the local members' support and sense of ownership, the credit operations would fail and savings would be seen to be at risk. Performance The number of savers and borrowers is growing rapidly in the kafo system, due both to an increase in the number of caisses (from forty-two to fifty-one in 1993-94) and to an increase in the number of members per caisse (from 317 per caisse at the end of 1993 to 405 per caisse in 1994; see table 23.2). As of September 1994, the total membership of the system represented 43 percent of the potential membership in the villages, assuming that all adults in the villages served are potential members. Of the 9,320 loans disbursed in 1992-93, 75 percent of the loans and 80 percent of the amount disbursed were in loans of CFAF 20,000-CFAF 35,000 ($66-$116). The average loan size of rural credit, the most common type of loan, was CFAF 31,000 ($105) in 1992-93. This average loan equates to approximately 35 percent of the per capita GDP for Mali and, as such, clearly indicates the relative and absolute level of poverty of borrowers. The participation of women is very low (5 percent of members), but recruitment of women is high on the list of priorities, and the number of female members should grow. The low participation rate derives from the original design of the program, which was created to help cotton-growing Kafo Jiginew 287 Table 23.2 Growth of Kafo Jiginew in Mali, 1992-94 Characteristic 1992/93 1993/94 Percentage change Number of caisses 42 51 21 Number of members 13,313 20,670 55 Number of borrowers Rural credit 7,056 11,042 56 Short-term credit 2,231 2,652 19 Equipment credit 33 1,411 4,175 Number of savers Demand deposits - 20,670 - Term deposits - 2,100 - - Not available. heads of household accumulate savings and obtain loans. Women were not originally viewed as participants in the program. Savings services are as convenient as can be expected. To be cost-efficient at low levels of savings and credit activity, the system can only afford to have one caisse per six to ten villages and to be open only two days a week. This relative inconvenience is offset by locating caisses in villages with a weekly market and by being open on market days to the extent possible. Withdrawals are immediate from funds maintained in the safes at each jiginew. Loan application processing is rapid, with disbursement occurring within one week following the presentation of a successful application. All credit decisions are made at the level of the jiginew by local credit commit- tees. Of greater concern is the rationing of credit made necessary by a lack of funds in the system. Each year, Kafo Jiginew allocates a total credit limit for the next twelve months equal to two times the highest level of savings attained in the previous year. In 1993-94 Kafo Jiginew overdisbursed significantly-CFAF 731 million ($1.46 million) compared with a target of approximately CFAF 500 million ($1 million). Savings have grown in the meantime, but Kafo Jiginew clearly needs to have access to an external line of credit to be able to meet both the demand of members for loans and the demand of depositors for their savings over the coming dry months. The demand for credit within the system is not being met. Many members could use some credit, but, due to the need for rationing, not all can receive it in a particular year. Approximately 50 percent of members received rural credit loans in 1993-94. The second most common loan product, short-term loans, is provided when liquidity is strong but is not available at all times of the year. Presumably, more of these short-term loans could be financed to the benefit of members. This kind of small commerce loan has proven to be very popular in other countries and is particularly useful for women. 288 Appraisals of Selected Microfinance Institutions As a matter of policy and consistent with its status as a regulated savings and loan cooperative, Kafo Jiginew cannot have outstanding loans that exceed two times its deposits.3 As long as the organization is growing, it will have capital constraints. Once it reaches a stable size, liquidity will be easier to manage and the potential for a run on the caisses will decline. Additional member training is needed to develop confidence in and support for the jiginews. More member commitment in the form of more savings would increase the availability of loans, which would further develop member confidence and support. Total operating income for the twelve-month period ending September 1993 covered 33 percent of operating costs (operating costs have been recalculated to include all expatriate costs) or 31 percent of operating plus financial costs. No adjustment was necessary to calculate the real cost of capital for Kafo Jiginew-4 percent on the average loan portfolio was considered adequate in the financial environment that existed in Mali before the devaluation in January 1994. A significant portion of Kafo Jiginew's income comes from investment income on donor funds received but not disbursed in loans. Profit will improve as these surplus funds are lent out to members at rates of interest higher than those now being earned. Results for the period ending September 1994 are expected to be better than in any previous year. This improvement is expected to continue as the combined loan portfolio grows and operating costs stabilize or decrease (with the removal of expatriate staff). However, an adjustment for inflation will be required to maintain the value of the shareholders' capital for the 1993-94 period, which will further increase what is still expected to be a net loss. Loan loss provisions are considered adequate in light of what appears to be a correctly reported loan loss experience. Very strong reserves have been built up equal to approximately 10 percent of the loan portfolio as of September 1993. Interest income as a percentage of the average loan portfolio-at 19.3 percent-confirms the quality of the lending activity and relates well to the estimated effective interest rate calculated. The assets shown on the balance sheet are being depreciated appropri- ately (see table 23.3). The value of the individual caisses does not appear on the balance sheet, however, and a correct representation of this would see the fixed assets increase by about CFAF 25 million (fifty-one offices at CFAF 500,000 each), and members equity would be adjusted by an equal amount. The income statement would be reduced by the appropriate depreciation expense, perhaps CFAF 1 million a year (assuming a thirty-year straight-line formula). In terms of the total expense of the kafo system, this was not considered material, particularly in light of the number of adjustments and estimates required to calculate the actual operating losses and the donors' equity. Kafo Jiginew 289 Table 23.3 Adjusted financial statements of Kafo Jiginew in Mali: Balance sheet, 1992-93 (millions of CFA francs) Item Amount as of 9/30/92 Amount as of 9/30/93 Assets Cash 25 35 Investments 101 140 Loan portfolio 176 244 (Provision for loan losses) (0) (2) Fixed assets (deprecia- 17 23 tion) Other assets 25 28 Total assets 344 468 Liabilities Demand deposits 125 182 Term deposits 32 65 Payables 6 8 Loans 0 0 Total liabilities 163 255 Equity Donors' equity For fixed assets - - For operations Total donors' equity 678 799 Members' equity Paid-in capital 52 68 Reserves 61 78 Retained earnings (609) (731) Total members' equity (496) (585) Total equity 182 214 Total equity plus liabilities 344 468 - Not available. Note: The average portfolio in the twelve-month period ending September 30, 1993, was CFAF 159 million. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Looking ahead To date, donors have invested approximately CFAF 800 million ($1.6 million) in establishing the network of fifty-one caisses. How much more donated capital will be required until Kafo Jiginew is financially independent will depend on its strategy over the next three years. There are two basic options: 290 Appraisals of Selected Microfinance Institutions continue to develop more caisses as outlined in the original plan (to a total of eighty) or stop expansion at the current level. There are arguments in favor of each approach, but considering the relatively uncertain economic climate in Mali, it may be best at this point to solidify the existing network rather than to expand further. If the existing network can be shown to be financially self-sufficient in the next three years, a very strong argument could then be made for expanding the network with an additional round of donor funds. Whereas none of the caisses was more than marginally profitable in 1992-93, this performance (before adjustment for inflation) is expected to be significantly better in 1993-94 (see table 23.4). If Kafo Jiginew is able to maintain a 20 percent financial margin on its lending activity and operating costs in the range of $100 million a year (which will be feasible once the expatriate costs have been removed), an average loan portfolio of CFAF 500 million ($1 million) would be required to break even. This would be a loan portfolio approximately twice that of September 30, 1993. This is not considered an unreasonable level of credit activity, and it could be achieved if all current members had an individual average indebtedness of CFAF 25,000 ($50), which is in the range of the average loan now provided. The risks that may threaten this development of the loan portfolio and Kafo Jiginew's financial self-sufficiency include: - A shortage of members' savings due either to a lack of confidence in the jiginews or to a poor cotton harvest. * A lack of funds from external sources to combine with members' savings to finance the expansion of the loan portfolio. * A forced reduction in Kafo Jiginew's lending rates consistent with the Ministry of Finance's new regulation (setting the maximum allowable lending rates at twice the discount rate) as inflation declines in Mali. Member loyalty is key to the survival of the kafo system, and loyalty depends on the quality of service provided. The low level of term deposit activity, for example, must mean that the product being offered is not that attractive. New ideas may be needed to improve the savings service now provided if volumes are expected to increase. Kafo Jiginew may want to reconsider the differential in rates between demand and term deposits, for example. A premium of 2.5 percent a year may not be enough to compel savers to lock in funds for one year. The demand deposit rates could be lowered, or the term deposit rates could be raised. One technique that has generated savings in other countries is a promotion campaign organized along the lines of a lottery. In this case every term deposit of CFAF 1,000 could provide the saver with a lottery ticket in a draw for a bicycle or a motorcycle. Staff will certainly develop their own Kafo Jiginew 291 Table 23.4 Adjusted financial statements of Kafo Jiginew in Mali: Income statement, 1992-93 (millions of CFA francs unless otherwise noted) For the twelve months ending 9/30/93 Asa Asa Amountfor the percentage of percentage of twelve months average loan net operating Item ending 9/30/92 Amount portfolio income Income Credit income 17 31 19 80 Investment income 12 9 6 23 (Interest expense) (5) (6) -4 17 Net interest income 24 33 21 86 Other income 5 5 3 14 Net operating 29 38 24 100 income Expenses Salaries Field staff 2 4 3 11 Administrative staff 24 27 17 69 Total salaries 26 31 19 80 Administration 17 23 15 61 Depreciation 5 7 5 19 Provisions 1 6 4 15 Extraordinary 0 0 0 1 expenses Operating subsidy 28 30 19 78 Institution building 39 39 24 101 Total expenses 116 136 86 355 Net income (87) (98) -62 n.a. n.a. Not applicable. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. ideas for increasing savings activity once the idea of improving service to the client is accepted as the appropriate focus for the institution. The other feature of Kafo Jiginew's quality of service relates to lending activity, and here it may need some external assistance. Members have to believe that they stand to gain not only as savers but also as borrowers. For 292 Appraisals of Selected Microfinance Institutions all members to have access to credit, Kafo Jiginew must increase its external sources of capital. External lenders can assist with both short- and long-term credits. Kafo Jiginew clearly requires more lending capital and also short- term funds to respond to members' demands for savings withdrawals. The risk of lending to Kafo Jiginew would appear to be commercially acceptable, and the institution should be considered by local and foreign lenders. The constraints placed on the operation of the kafo system by the new regulations from the Ministry of Finance are critical issues, particularly the maximum interest rates on loans and, to a lesser extent, the maximum leverage allowed on members' savings. These issues affect all microfinance institutions in the CFA region. Notes Kafo Jiginew was evaluated by Doug Salloum (a consultant) and by Leila Webster and Pierre-Olivier Colleye (Private Sector Development Depart- ment, the World Bank) in November 1994. Almost all of the information presented here is based on conversations with the project director, Alou Sidibe. 1. The exchange rates used throughout this chapter are 300 CFA francs to 1 U.S. dollar for 1993 and 500 CFA francs to 1 U.S. dollar for 1994. 2. The estimated weighted effective interest rate on loans during 1992-93 was approximately 19 percent a year. 3. There will always be some form of credit rationing due to the limitations of leverage imposed by the Ministry of Finance's new regulations on this kind of financial institution. 24 Caisses Villageoises d'Epargne et de Credit Autogerees du Pays Dogon Doug Salloum The Caisses Villageoises d'Epargne et de Cr6dit Autogerees du Pays Dogon (Caisses Villageoises) project was designed and has been implemented by the Centre Intemational de Developpement et de Recherche (CIDR) based in Autreches, France.' The project has received the financial backing of Kreditanstalt fur Wiederaufbau (KfW) of Germany, the sole donor to the project since its inception in 1986. By the end of 1993, KfW had provided roughly CFAF 575 million ($1.9 million) to cover project expenses, including the costs of two French expatriate staff, fixed assets, and operating deficits. As of December 1993, the project had developed fifty-five independent village banks with total membership of approximately 12,500 persons (and a forecast membership of 16,000 in the same number of banks by end-1994) (see table 24.1).2 Table 24.1 Growth of Caisses Villageoises du Pays Dogon, 1991-93 Percentage change, Characteristic 1991 1992 1993 1992-93 Number of village banks 25 31 55 77 Number of membersa 4,630 7,322 12,515 71 Number of borrowers 1,907 3,098 5,507 78 Number of term deposit 455 1,001 1,764 76 savers a. The project director projected membership at 16,000 by end-1994. As the name implies, Caisses Villageoises operates exclusively in the Dogon region of Mali in villages around the towns of Bandiagara, Bankass, and Koro. The economy of the Dogon region is not based on a single cash crop but is diversified to the extent allowed by the relatively dry climate and sandy soil. Villagers are involved in trading activities in addition to their main occupation of farming-raising goats and cattle and growing millet and subsistence vegetables. Most Dogon villages are homogeneous and well established, often having been the home of a few extended families for more 293 294 Appraisals of Selected Microfinance Institutions than a hundred years. This high level of cohesion, combined with the well- known resourcefulness of Dogon villagers, has contributed significantly to the strength and stability of the Caisses Villageoises. The Caisses Villageoises system assists local villagers directly through village banks that deliver credit and savings services, but the program is structured in such a way as to foster community development. Those banks that are able to generate profits from the intermediation between savings and lending retain these funds for investment in community development projects like the construction of schools, well pumps, and pharmacies. In addition to these direct financial benefits, participating villages reap significant social benefits from the community-building process of creating and allocating their own resources on a cooperative, villagewide basis. The design of the Caisses Villageoises system differs from that of the typical rural credit union primarily in the small size of its central technical support system, its lack of common policies on interest rates, and its rudimentary record keeping and accounting systems. The independence and local ownership of each bank and the lack of emphasis on developing a network of village banks also distinguish this program. The Caisses Villageoises system complements the formal banking system in Mali, whereas credit unions often compete with banks. Some of the differences between this program and classic credit unions derive from the fact that villages in the Dogon region often are isolated from one another, and roads are very difficult to travel. CIDR considers the operating costs involved in providing a standardized and frequent monitoring and control system from a centralized location, as is typically done through credit union federations, to be too high. It also assumes that interest income earned on loans issued by village banks would not be sufficient to finance a centralized structure. In consequence, CIDR has designed a highly decentralized system of independent village banks in which villagers themselves will eventually provide all but a small portion of the activity required to manage the banks. After eight years of direct project control and responsibility followed by shared management with local staff, CIDR is committed to leaving the project in the hands of the villagers. CIDR intends to pull back the French expatriate staff by the summer of 1995 and to leave the provision of basic financial and technical support services to a three-person Malian staff in a new, central unit called the service commun. CIDR has been working to prepare the villagers and the future service commun staff for this transition. This withdrawal of the expatriate staff and the lessening of the village banks' dependence on outside assistance is critical to CIDR'S vision for the future of the Caisses Villageoises system. It is expected that, within eight to ten years, the service commun will be financed entirely through billing the Dogon caisses for its services. The feasibility of this plan is one area of review and Caisses Villageoises d'Epargne 295 challenge for CIDR and its donors, as well as for the village banks and their members. Program characteristics There are three levels of operation and management in the Caisses Villageoises system: the village bank, the union of village banks, and the project. The village bank Creation of the village bank begins with selection of the village where it will be established. An individual village must initiate the process by requesting the establishment of a bank, but the Caisses Villageoises will not work with all villages that express an interest. CIDR has developed criteria intended to minimize the potential for failure, defined as a village bank that is not dynamic and does not have strong local support or potential for growth. Criteria for selecting a village include:3 * Economic potential. Does the village have trading links with other villages and with markets in towns of the region? * Literacy levels. Some of the smallest villages cannot produce a minimum of two literate people who can become the administrators for the village bank. In addition, project staff encourage the addition of a controller. a History. CIDR looks at when a village was created and where the villagers came from to ascertain its level of stability and cohesion. The records of previous and current village organizations established for social or production purposes are indicators of unity. * Size. At least 400 adults are needed as potential members. Once advised of its eligibility, the entire village meets as the general assembly for the village bank, and project staff advise them of the initial tasks that must be accomplished, including: * Construction of a building to house the bank. The project provides a safe, paper, and furniture as required. * Selection of a management committee whose members tend to represent family groupings in the village, typically two committee members per family. Members are not paid. * Selection of administrators and controllers who work as volunteers in the bank office the one day a week that the bank is open and the previous evening when accounts due are identified. If the bank proves profitable, these individuals receive compensation for their work, typically one-third of the bank's net interest income. 296 Appraisals of Selected Microfinance Institutions After the village completes these tasks, project staff train the management committee and the administrators and controllers and help to set up the accounts and the savings and credit records. The management committee is responsible for recruiting new members to start saving in the bank. Villages typically are required to develop a savings base of roughly CFAF 100,000 ($200) before lending begins. Once established, the bank operates autonomously but with training and supervision from project staff. The village general assembly sets interest rates. The management committee takes responsibility for delinquent loan follow-up and handles the cash. The administrators record loan payments and deposits and update members' passbooks. They also collect loan requests from members and present them to the management committee for consideration the night before the opening day of the bank. The controller oversees the work of the administrators and, every two weeks, forecasts the pool of capital available for lending. Project staff (and in the future, the service commun) assist the village in planning its loan disbursements if there is no controller. Anyone can become a member of the village bank, and capital subscrip- tion amounts are kept low at CFAF 500-CFAF 1,500 ($1-$3) to encourage the poorest to participate. In part because of the low subscription price and in part because of an emphasis on loans for commerce, membership among women is relatively high at 33 percent. The project estimates that 44 percent of potential members are now members. The union Twice a year, two or three members of each management committee in a region get together for a union meeting. There are three unions, one each for the regions of Bandiagara, Bankass, and Koro. Travel expenses for these meetings are paid directly by the villages. One of the major tasks of the unions is to apply for loans from Banque Nationale de Developpement Agricole (BNDA). Using its own funds as well as funds granted by KfW, BNDA has been supporting the caisses system by providing short-term (less than one year) loans, which the unions lend to village banks in their respective regions. The unions have set their own requirements for villages to obtain this refinancing. Requirements include a lending history of at least one year, less than 10 percent of loans overdue in repayment, and a positive rating from project staff with respect to the bank's management and record keeping. The union will allow a village bank to borrow up to 1.5 times the savings level attained during the previous six months to one year. This can be increased to twice the savings level based on performance. Unions also require that villages provide cash collateral of 10 percent, which is passed through to the BNDA as security for the loans. The unions were able to borrow almost CFAF 40 million ($133,000) from the BNDA during 1993. Net Caisses Villageoises d'Epargne 297 debt outstanding to the BNDA at the end of 1993 rose from CFAF 15 million ($50,000) to CFAF 22 million ($73,000). The unions borrow from BNDA at 9 percent and lend to the village banks at 18-20 percent. The three unions meet together twice a year, primarily to coordinate the activities of the service commun. The project To date, the project staff have included two CIDR expatriates (the project chief and a small business specialist) and nine Malian fieldworkers. There are few support staff (guards and messengers) in the project team. The central accounting function for the project has devolved to the project chief. The project has advised the villages of the coming withdrawal of expatriate staff and the subsequent reduction of the central staff and its services to the more minimalist service commun. In preparation, project staff have set up several six-month tests in which selected banks operate without any project assistance on a trial basis. Fieldworkers train and assist caisse administrators, controllers, and committee members. They also ensure that records are kept correctly and collect information on repayment, loans outstanding and approved, and deposit activity and balances. An audit of each village bank is carried out at least twice a year to reconcile accounts and to produce year-end project results. Project staff also provide technical assistance to a limited number of entrepreneurs who wish to expand their businesses or to obtain somewhat larger business loans from village banks or the BNDA. This activity will continue with the service commun and should earn some income, because entrepreneurs will be expected to pay for the assistance. Savings Interest rates for deposits are set by individual banks, but averages across the system indicate that the trend is toward uniform pricing. Two savings products are available: demand deposits that pay no interest and are not used to fund the lending activity of the banks, and term deposits with interest rates that average 22 percent annually on noncompounded monthly balances. Term deposits are placed for periods of three to twelve months (average five months), and these funds, combined with members' capital and credits from BNDA, are loaned to members. Simple passbooks and central registries record all deposit transactions. Interest on term deposits is calculated at maturity, and interest earned plus principal is returned to the member. The high interest rate paid on term deposits appears to have attracted a significant level of savings from members. The project estimates that 14 percent of members have a term deposit in the system and that the average 298 Appraisals of Selected Microfinance Institutions amount deposited is CFAF 22,000 ($44). Term deposits at the end of 1993 constituted a very impressive 94 percent of all deposits. Credit Only members can receive loans, but membership is open, and the low membership subscription rates are not a major barrier to participation. There are no standard loan products or interest rates, but, as with deposit rates, there is some uniformity across the system. Use of loan proceeds is not strictly controlled, although each applicant states a primary purpose for the loan at the time of application. The following summarizes the purposes identified for credits approved in 1993: commerce (76 percent), agriculture (8 percent), stock raising (8 percent), and artisanal, social, and other activities (8 percent). Borrowers are required to provide collateral for their loans, and almost anything is accepted: goats, bicycles, kitchen utensils. This collateral will be taken in the case of default, but the system relies mainly on the borrower's family to honor any unpaid amounts. Loans are approved based as much on the amount of funds available and the number of requests as on the merits of the application, although borrowers of bad character presumably would not be approved. During 1993, 44 percent of members received at least one loan from the village banks. The average loan amount disbursed was CFAF 14,600 ($48.66). Among members who received loans, 31 percent were women, which is consistent with their proportion of the overall membership. The average interest rate charged on loans during 1993 was an estimated 40 percent annually. No payments are made during the term of the loans (up to nine months, on average four months). Principal plus interest due is paid at the maturity of the loan. Governance The Caisses Villageoises system has a very flat structure. All decisions are made at either the village bank or the union level. There are relatively few technical support staff at the center, and with the coming cutback, there will be even fewer. Responsibility for and ownership of the individual banks clearly are in the hands of the villagers: * The village as a whole acts as the general assembly for each bank and selects the management committee to run the bank. The general assembly meets annually to review the results of the bank and to decide what to do with any operating profit. All decisions regarding the future of a village bank are made at general assembly meetings. * The extended families that make up the village are each represented on the management committee. The committee is very involved in the day- to-day management of the bank, with one or two committee members Caisses Villageoises d'Epargne 299 taking responsibility for each family member who receives a loan. If a borrower is late with a payment, the committee member investigates the situation and, in the case of a default, takes collateral or asks the family to make up the missing amount. Other than the administrators and controllers, there will soon be no technical personnel in the central office other than the very small staff of the service commun. It is a bold initiative on the part of CIDR to attempt to develop what, if successful, will be one of the most decentralized systems of savings and credit imaginable. The risk is that some individual banks will not be able to sustain operations well enough to maintain the confidence of their members. For example, if records are not carefully kept and promptly updated, depositors could lose trust in the bank and deposits could dry up. Performance The number of savers and borrowers continues to grow, although most recent growth has been due mainly to the increase in the number of banks. Membership in banks created prior to 1992 represents a higher portion of the potential membership in the village than does membership in newer ones, which confirms that local interest in participation has increased with time. CIDR estimates that 44 percent of the adult population of villages that have banks and 19 percent of adults in adjacent villages are now members. The greatest opportunity for expansion clearly is in adjacent villages, but the issue could then become how to maintain the social cohesion needed to keep repayments high. Without question, the village banks of the Caisses Villageoises are serving a population that would be extremely difficult for the formal system to reach given their remote locations and small numbers. Loan sizes are very small. For the 8,354 loans disbursed in 1993, the average loan size was CFAF 14,600 ($48.66). Women form a significant portion of the membership and participate proportionately in both savings and credit activities. The quality of service is high. Once-a-week service is probably all that this kind of village banking system realistically can be expected to provide. The fact that most members live in the village where the bank is located means that having to travel to the bank on the days it is open is not a significant obstacle. Deposit transactions appear to be handled quickly, and the level of documentation is very simple. Loan applications are processed within a week, if funds are available for lending. The high proportion of term deposits in the system makes savings a relatively stable source of loan capital, but the total volume of savings is not sufficient to meet the demand for credit. Refinancing from the BNDA has greatly facilitated the growth of the loan portfolio. 300 Appraisals of Selected Microfinance Institutions Both the timing and the amount of credit provided by BNDA directly influence members' satisfaction with the loan service provided by their banks. Managing the relationship with BNDA is critical to maintaining members' confidence in their banks as BNDA increasingly becomes a major source of credit funds. BNDA'S willingness to increase loan amounts to the Caisses Villageoises likely will depend heavily on how this relationship is managed by the service commun. Total interest income for the entire Caisses Villageoises system for the period ending December 31, 1993, covered only 19 percent of operating expenses (including the costs of the expatriate staff and of all fixed assets) or 18 percent of operating plus financial costs. These figures improve to 39 and 33 percent, respectively, if expatriate salaries are excluded (they are included under institution building in table 24.2). It is an interesting feature of the project design that the system on a consolidated basis reported an operating loss of CFAF 95 million ($316,700) but that villages with profitable caisses received a dividend of CFAF 1.2 million ($4,000).4 The villages receive all profits from the savings and loan activities they manage, portions of which they then allocate to the service commun, to the administrators and controllers, to the reserve and guarantee funds, and finally to community projects. The allocation for community projects totaled CFAF 1.2 million in 1993. This prioritization underlines the CIDR concept that direct financial benefits should flow to the individual villages that have earned them rather than to a common fund that under- writes the system as a whole. The trend in cost recovery is improving. Interest income as a percentage of operating plus financial costs went from 4 percent in 1991 to 19 percent in 1993, but this is still a long way from self-sufficiency (see table 24.3). The percentage of operating plus financial costs that can be expected to be covered by revenues should increase fairly dramatically as the expatriate staff withdraws and as the loan portfolio increases. Loan repayment rates have been high, although the project director reports that competition for low-priced credit from other NGOS operating in the area is weakening the repayment discipline of borrowers in some villages. The accounting for loan losses used by the Caisses Villageoises is not consistent with that applied by most financial institutions. Loan loss provisions had not been taken on the financial statements prepared by CIDR. However, significant reserves equal to 13 percent of the loan portfolio outstanding prior to adjustment had been set aside. In the adjusted statements, the amount set aside as reserves was shown as a loan loss provision and deducted from the assets and the equity. In December 31, 1993, 6 percent of the loan portfolio was overdue by at least one month, and 1 percent was overdue by one year or more. The portion that was overdue by one year was spread over the three years shown and was written off Caisses Villageoises d'Epargne 301 Table 24.2 Adjusted financial statements of Caisses Villageoises du pays Dogon: Income statement, 1991-93 (millions of CFA francs) For the twelve months ending 12/31/93 Amountfor Amountfor As a As a the twelve the twelve percentage percentage months months of average of net ending ending loan operating Item 12/31/91 12/31/92 Amount portfolio income Income Credit income 4 9 21 39 167 Investment income 0 0 0 0 1 (Interest expense) (2) (4) (9) -15.9 68 Net interest income 2 5 13 24 100 Other income 0 0 0 0 0 Net operating income 2 5 13 24 100 Expenses Salaries Field staff - - - - - Administrative staff - - - - - Total salaries 9 10 13 24 101 Administrative 19 24 29 54 230 Depreciation 4 8 7 13 54 Provisions 0 1 1 2 10 Extraordinary ex- 0 0 1 1 6 penses Operating subsidy 3 3 3 6 27 Institution building 63 47 55 102 436 Total expenses 98 93 108 202 863 Net income (96) (88) (95) -179 n.a. Dividends 0 0 1 n.a. n.a. Contributions to loan 1 2 3 n.a. n.a. funds Contribution to re- (97) (90) (100) n.a. n.a. tained earnings n.a. Not applicable. - Not available. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. 302 Appraisals of Selected Microfinance Institutions Table 24.3 Adjusted financial statements of Caisses Villageoises du pays Dogon: Balance sheet, 1991-93 (millions of CFA francs) Amount as of Amount as of Amount as of Item 12/31/91 12/31/92 12/31/93 Assets Cash 4 6 9 Investments 0 0 2 Loan portfolio 15 40 68 (Provision for loan losses) (0) (1) (2) Fixed assets 13 23 23 Other assets 1 3 8 Total assets 32 71 107 Liabilities Demand deposits 1 1 2 Term deposits 9 21 38 Payables 2 4 9 Loans 3 15 22 Total liabilities 15 42 72 Equity Donors' equity For fixed assets 44 61 63 For operations 331 415 512 Total donors' equity 375 475 575 Members' equity Paid-in capita 2 4 6 Reserve 2 4 7 Retained earnings (363) (453) (553) Total members' equity (358) (445) (539) Total equity 17 30 36 Total equity plus liabilities 32 71 107 Note: The average loan portfolio in the twelve-month period ending December 31, 1993, was CFAF 53.3 millions. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Caisses Villageoises d'Epargne 303 against the accumulated loan loss provision. The remaining provisions are considered adequate for the quality of the portfolio. Assets appear to be correctly accounted for except that the buildings provided by the villages are not included on the balance sheet. However, the impact on the income statement of the additional depreciation expense is not considered material. The responsibility of replacing or repairing the village bank building would not fall to the system in any case, and an argument can be made for excluding these assets from the balance sheet. Looking ahead Donor funds invested in the Caisse Villageoises system as of the end- December 1993 totaled roughly CFAF 575 million ($1.9 million). The net result has been the establishment of fifty-five independent village banks. This clearly is a very significant accomplishment, especially considering the isolated nature and small size of the villages being served. The question that cannot yet be answered, however, is whether these banks will sur- vive-whether CIDR has done everything that can be done to ensure their survival before it withdraws from the project. As is always the case with member-owned savings and credit institutions, the future viability of the system will ultimately depend on the commitment and confidence of its members. Their participation presumably will remain high as long as they receive quality savings and credit services. If participa- tion is high, profits will be generated and then redistributed for local projects, further tying the community to the bank. Elements of good service that must be maintained at the village level include good record keeping, administrators and controllers who inspire confidence, quick and accurate transactions, and continuing access to credit at competitive rates. Most of these are within villagers' control, but some will become the responsibility of the service commun. To meet the growing demand for credit, the service commun will have to maintain and strengthen the relationship with the BNDA or other commercial lenders. The process of forecasting the demand for loans and of applying for funds from the BNDA can be made more responsive. To increase its exposure, BNDA may need more information about the repayment of loans by villages to realize that the underlying risk of its credits to the unions is acceptable. The service commun also will be critical to maintaining high standards within individual village banks. Will the proposed service commun have enough staff to provide the technical, motivational, and financial services required to maintain the system? Equally important are questions about funding the service commun. If the entire budget of the service commun (CFAF 22 million a year, $44,000) is funded by the 15 percent of the village banks' profits currently allocated, their profits would have to be CFAF 145 million ($290,000) a year. Assuming that the caisses maintain their net interest spread of approximately 24 percent (and this is a significant I 304 Appraisals of Selected Microfinance Institutions regulatory question),5 the average loan portfolio outstanding would have to be CFAF 600 million ($1.2 million), roughly nine times the value outstanding as of December 1993. This level of lending would be very difficult for the members to support. The preceding simple analysis would lead us to conclude that the service commun either will not be fully financed by member banks or is not financially sustainable. Sources of income available to the service commun that have not been included in this analysis include fees from services, 75 percent of the unions' interest income (unions pay BNDA 9 percent a year and lend to the caisses at 18-20 percent a year), and interest income from a pool of guarantee funds invested with BNDA in Mali and with an open-end investment fund in France and from a line of credit funded by KfW to be made available for agriculture loans. It is also possible that the service commun staff may have to be increased. To improve the financial viability of the service commun, either the village banks' profits or the share of profits contributed to the service commun will have to increase.6 Anything that would not increase the risk of lending but that would increase the size of the loan portfolio for each village would improve the profit of each unit. More members, more villages served per bank, and larger loans all would have a positive impact on interest income. It may be necessary to consider the pros and cons of introducing more leverage in the system (a higher ratio of member savings to external debt) and to arrange more external lines of credit to develop a larger, more profitable loan portfolio. With respect to the share of the profit that comes to the service commun, there might be room to renegotiate the arrangement. As it is currently envisaged, village banks will contribute 15 percent of their profits to the service commun and one-third to their administrators and controllers, retaining the remaining funds for reserves, guarantee funds, and community projects. If the banks increase in size, the administrators' commission may be too high and conceivably could be capped at a level comparable to the salary of a civil servant for the same number of days of work. Some of the difference between a third of the profit and the salary cap could go to the service commun without affecting the level of funds flowing to community projects. Anything the CIDR does now to improve the financial viability of the service commun will directly improve the village banks' chances of survival. Notes 1. CIDR operates two other Caisse Villageoises projects in Mali: one in the First region (Kita and Bafoulabe) since 1991, and the second in the Niger region (Niono) since 1994, both financed by KfW. CIDR also operates a credit solidaire project with groups of women in Mopti since 1993, financed by the Caisse Francaise de Developpement. Caisses Villageoises d'Epargne 305 2. The exchange rates used throughout this chapter are 300 CFA francs to 1 U.S. dollar for 1993 and 500 CFA francs to 1 U.S. dollar for 1994. 3. In applying these criteria, CIDR has found that for every village selected, four were considered. Villages not selected for the establishment of a bank often have the option of accessing banks that are developed in nearby villages. 4. CIDR points out that a large portion of these costs are traceable to operating costs in the first, promotional stage of the project and, in their view, they should be considered as investment expenses to get the system established, rather than as operating losses to be carried forward. 5. The Caisses Villageoises system is currently outside of regulations that will be applied to credit unions in Mali, but passage of new regulations by the Central Bank of Western African States (BCEAO) will likely bring some form of regulation to Mali and to this program. The impact of this regulatory initiative on microfinance institutions in the Sahel region is dealt with in annex 2. 6. CIDR points out that the following measures are being undertaken to increase revenues: introducing longer-term and larger loans for equipment and infrastructure, negotiating tax exemptions for replacements of fixed assets (cars, mopeds), negotiating with the BNDA for lower interest rates, and providing services, such as feasibility studies, training, and consulting, to NGOs and other projects. I 25 Credit Mutuel du Senegal Doug Salloum Centre International du Credit Mutuel (cIcM) has been the force behind the creation of what is now the Credit Mutuel du Sen6gal. CICM is a Paris-based, nonprofit institution created by the French Federations du Credit Mutuel to design, initiate, and operate savings and credit projects around the world, primarily in French West Africa. cIcM prepared the original feasibility studies for the creation of a network of village savings and credit banks in the rural area of Kaolack in 1983. Between 1983 and 1987, CICM negotiated the funding for the project with the government of France, and in 1988, operations were started. The project was originally called the Caisses Populaires d'Epargne et de Credit du Centre International de Credit Mutuel. In October 1994, it took the name Credit Mutuel du S6negal to distinguish it from the more than seventy other caisses in the country (many of questionable quality and security) and also to establish more clearly the link with the French Credit Mutuel system. All of the funding received by Credit Mutuel has been provided by the French government through the Ministry of Cooperation and Development, either directly to the project or to cicM as salaries and subsidies for CICM project staff and equipment purchases. Total grant funding from the French government to the project is estimated at approximately CFAF 1 billion ($2 million) to the end of December 1993, including an advance of approxi- mately CFAF 200 million ($400,000) to fund expected growth in 1994. As of the end of December 1993, a network of forty-one local caisses with total membership of 13,114 persons was created with these funds. By August 1994, fifty-four caisses had been established, and membership had grown dramatically to 22,228. The national office of Credit Mutuel is located in the town of Thies, approximately 60 kilometers east of Dakar on the main Dakar-Tamba- counda road. Thies also contains one of the four regional offices, with the others located in Kaolack, Tambacounda, and, as of January 1995, Ziguinchor. Each of these towns is the administrative center for the region. The Thies region is the most densely populated of the four, and its economy is strongly influenced by its proximity to Dakar; Kaolack is a region for growing peanuts and raising goats and cattle; the Tambacounda area is very sparsely populated and dry; and Ziguinchor is culturally and economically somewhat separate from the rest of the country. 307 308 Appraisals of Selected Microfinance Institutions Program characteristics Credit Mutuel has four levels of operation and management: the local banks, the regional offices, the central office, and CICM. The local banks The Credit Mutuel network has been designed primarily to serve rural savings and credit needs, but the system has caisses in both urban and rural areas (see table 25.1). Urban caisses have computers to monitor members' accounts and debts; rural caisses have manual accounting and savings and credit monitoring systems. Table 25.1 Composition of caisses of Credit Mutuel du S6negal, 1993-98 December 1998 Type of caisse December 1993 August 1994 (forecast) Urban caisses 6 8 15 Rural caisses 35 46 72 Total 41 54 87 CICM is planning to develop a larger network based on its four regions of operation. By 1998 the network should consist of eighty-seven caisses. With this completed network, CICM expects to be able to serve the entire country with the exception of the regions of Ferlo, Fleuve, and the peninsula of Cap- Vert (Dakar). The future network will consist of seven large urban caisses, eight small urban caisses, seventeen large rural caisses, sixteen medium-size rural caisses, and thirty-nine small rural caisses. The process of establishing a caisse has three phases: * A preliminary attempt is made to assess the social coherence of the community and its willingness to adhere to Credit Mutuel's principles. This is accomplished in open, public meetings held in interested villages. • A cashier window that will accept deposits is established. Only loans that are fully secured by savings are made at this early stage of a caisse. * Training for members is provided. When a caisse reaches about 250 members, a general assembly is called, and the officers of the caisse are elected. Anyone can open a savings account in a caisse by becoming a member (either as an adult individual or as a group of individuals) or a user (as a minor). Members contribute CFAF 6,500 ($13), and users contribute CFAF 3,500 ($7) as follows: CFAF 3,000 ($6) for a share in the caisse, which is Credit Mutuel du Senegal 309 returned if a member leaves the caisse; CFAF 1,500 ($3) for a one-time, nonreimbursable fee; and CFAF 2,000 ($4) for a deposit to open an account. Each caisse has a maximum of three staff, but most employ only one. Staff at this level are all Senegalese. The large urban caisses employ a manager and a cashier; the smallest rural caisses have only a manager who receives a commission for his or her services; and the others have only a manager who may be paid a salary or a commission. In addition to paid (salaried or commissioned) staff, each caisse has two councils: the administrative council, which has ten members, and the supervisory council, which has six members. Each council is elected for six-year terms from among the membership at the annual general meeting. All of the individual caisses are legally constituted and licensed by the Ministry of Finance as cooperative credit societies with variable capital. All features of the recently approved law and the draft decrees to the law governing mutualist financial institutions or savings and credit cooperatives apply to and are respected by the caisses. This legal status exempts them from income and transaction taxes, as well as from taxes on certain purchases for three years from their registration. The regulations require that the caisses have an annual general meeting, an administrative council, and a credit committee. These were already features of the Cr6dit Mutuel system, and they do not represent a change in the way the caisses operate. The regional offices The major functions of the regional offices are to control the consolidated accounts of the local banks for the region and to monitor, train, and supervise individual caisses. Regional offices also transfer funds within the system, such as surpluses to the central office for investment or to caisses for loans or forecast withdrawals. The staff of the regional offices consist of the regional office manager, a secretary or a secretary/accountant, and some monitors. The number of monitors in each regional office varies with the number of caisses under supervision. The goal is to have one for every ten caisses. Currently, there are two monitors in Thies and Tambacounda and five in Kaolack. These staff members are critical, because most are directly involved in fieldwork. Their tasks include training managers and elected boards in caisses, working with managers to attract additional members, supervising and controlling caisses' accounts, and developing new caisses. The regional manager of the Kaolack office is the only Senegalese regional manager in the CICM network. The rest are French expatriates. CICM'S strategy is eventually to have Senegalese nationals in all the regional manager positions, but CIcM believes it is best to promote monitors to these positions after three or four years of field experience. Credit Mutuel has only been operating in areas other than Kaolack since 1993. It is only in the 310 Appraisals of Selected Microfinance Institutions Kaolack region that field staff have sufficient experience, in the opinion of CICM, to take over regional management responsibilities. The central office The central office in Thies acts as the central caisse for the whole network. It invests surplus funds on behalf of the individual caisses, borrows funds from commercial banks, and redistributes funds as necessary. It inspects the caisses and collects and analyzes all financial information and accounts. The central office currently has seven staff members, but this number will increase soon by at least three, as two inspectors and one accountant are hired. The proposed increase underscores an acknowledged shortage of central office staff. The central office has the following positions: one director (an expatriate), two accountants (including one expatriate), two inspectors, one computer systems manager (an expatriate), and one secretary. The central office also employs the regional office staff. The central office is not legally constituted, recognized, or licensed by the Ministry of Finance. However, the Minister of Finance, the commercial banks, and the donors all recognize it as the network's representative. There is no federation of caisses in the Credit Mutuel system and, therefore, no central caisse. Instead, the central office serves as a central caisse, handling the liquidity for the whole network and borrowing from commercial sources as required. CICM CICM operates the network, and its status is somewhat ambiguous. Its structure and role are closest to that of an adviser with strong management responsibilities. Savings Other than the minimum savings deposit required to open an account and to become a member, all savings deposits are voluntary. There are two kinds of deposits: free savings that pay 4.5 percent a year and can be withdrawn at any time, and blocked savings that are available in three-month or six- to twelve-month terms and pay, respectively, 5.0 and 5.5 percent a year. CICM also offers term deposits that are used to secure a member's loan in part or in full. Term deposits pay an interest rate of 6 percent a year. Credit CICM provides five kinds of loans to its members. All loans are approved by the local administrative council based on the recommnendation of the manager. All loans are extended for a period not exceeding one year and Credit Mutuel du SMn6gal 311 have an interest rate of 2 percent a month or 24 percent a year. The following range of credit products is available to members: * Rural credit is the main credit product of the rural caisses (80 percent of all loans). Rural loans finance the purchase of seeds for peanuts, millet, and corn. They are disbursed in June and repaid in full (interest plus principal) in January of the following year. The average loan size was approximately CFAF 66,000 ($132) in 1994. * Regular credit allows members to borrow up to three times their average savings balance over the previous six months. Repayment is made in equal monthly payments. Members must secure these loans by keeping at least one-quarter of the loan amount on deposit during the term of the loan. A 5 percent charge on these loans, due on disbursement, is allocated to a life insurance policy (0.5 percent), a processing fee (2 percent), and a contribution to a guarantee fund (2.5 percent). * Automatic credit consists of loans that are fully secured by a member's savings and are often proposed as an alternative to having a member withdraw his or her funds. * Third-party loans consist of funds lent by Credit Mutuel but provided by NGOS or donors. . Savings are essential to receiving a loan from Cr6dit Mutuel. To obtain a loan, members must save at a certain rate for at least six months, after which they will receive a loan equal to at least three-quarters of the savings amount. The repayment period for the loan is equivalent to the savings period-at least six months. By that time, borrowers have proved that they are able to repay because their contributions to savings were higher than the required monthly loan payments. Credit Mutuel still requires that one-quarter of the loan be secured by the member's blocked savings. An important feature of Credit Mutuel is its strict loans-to-deposits ratio. On average, Credit Mutuel tries to keep the ratio of outstanding loans to deposits to less than 50 percent for the network, although it does allow some variation among caisses and between rural and urban caisses. In general, urban caisses lend less than 50 percent of members' savings (since the major credit product is rural credit), and rural caisses lend more than 50 percent. Many rural caisses reached 100 percent, or in extreme cases even 170 percent, of local members' savings in 1993 and 1994. Credit Mutuel has not condoned these extreme levels of lending and has proposed new regulations to limit rural caisses to a level of lending that would not exceed 100 percent of local members' savings. In 1994 only 4,500 of Credit Mutuel's 22,000 members received loans, due in part to a series of controls intended to limit the total volume of loans in the system. 312 Appraisals of Selected Microfinance Institutions Table 25.2 Growth of Credit Mutuel du Senegal, 1989-94 Indicator 1989 1990 1991 1992 1993 1994 (August) Number of caisses 10 27 28 31 41 55 Number of members 1,587 5,543 7,764 9,557 13,114 22,228 Number of approved - - - - 2,435 4,500 rural credits Value of approved rural - - - 65 163 298 credits (millions of CFA francs) Savings (millions of CFA 0 12.6 93.2 105.0 332.0 773.7 francs) Savings per member 0 2,265 12,001 10,988 25,314 34,809 (CFA francs) - Not available. Credit Mutuel has developed two regulations designed to limit the credit exposure of the caisses based on their size and experience. First, if a caisse has more than 5 percent of the outstanding loan portfolio in arrears (more than three months late), all credit activity is suspended in that location. Second, maximum loan amounts allowed one borrower vary from CFAF 150,000 ($300) for the smallest rural caisses to CFAF 500,000 ($1,000) for large urban caisses. Loans are used primarily by rural members. This is because urban members are more likely to be engaged in small business activities, and the Credit Mutuel system is not oriented toward business credits. In rural areas, 70 percent of loans are used for agriculture, 20 percent for consumption, and 10 percent for commerce. Governance CrGdit Mutuel's governance structure is only representative at the level of the individual caisses, where elections are held for the administrative council and the supervisory council. The elective process ends at this level. There is no regional or national elective structure (in most mutualist systems, this representative is at the union and the federation level). Local caisse presidents apparently have not demanded more representation, and CICM has not promoted a more representative structure. CICM determines policies, procedures, and regulations, and the central and regional offices implement them. CICM does not represent the members in any democratic sense but operates the network to the benefit of the members, as it deems appropriate. A system of regional and national representation will have to be developed as a feature of compliance with the new laws and decrees Credit Mutuel du SMn6gal 313 governing the operation of mutualist savings and credit institutions. Credit Mutuel is preparing for this process. Table 25.3 Adjusted financial statements of Credit Mutuel du Senegal: Balance sheet, 1991-93 (millions of CFA francs) Amount as of Amount as of Amount as of Item December 1991 December 1992 December 1993 Assets Cash 15 19 96 Investments 70 181 358 Loan portfolio 118 78 220 (Provision for loan losses) 0 0 (4) Fixed assets (net) 46 57 102 Other assets 19 17 67 Total assets 268 352 839 Liabilities Demand deposits 98 140 203 Term deposits 0 0 119 Payables 6 2 10 Loans 20 30 95 Total liabilities 124 172 427 Equity Donors' equity For fixed assets 122 128 162 For operations 278 408 870 Total donors' equity 400 536 1,032 Members' equity Paid-in capital 23 32 39 Reserve 0 0 0 Retained earnings (279) (388) (659) Total members' equity (256) (356) (620) Total equity 144 180 412 Total equity plus liabilities 268 352 839 Note: The average loan portfolio in the twelve-month period ending December 1993 is CFAF 159 million. See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. 314 Appraisals of Selected Microfinance Institutions Table 25.4 Adjusted financial statements of Credit Mutuel du Senegal: Income statement, 1991-93 (millions of CFA francs unless otherwise noted) For the twelve months ending December 1993 Amountfor Amountfor As a As a the twelve the twelve percentage percentage months end- months end- of average of net ing Decem- ing Decem- loan operating Item ber 1991 ber 1992 Amount portfolio income Income Credit income 19 13 32 20 71 Investrnent 6 12 26 16 58 income (Interest ex- (6) (6) (18) (11) 40 pense) Net interest 19 19 40 25 89 income Other income 15 6 5 3 11 Net operating 34 25 45 28 100 income Expenses Salaries Field staff - - 23 14 51 Administra - - 12 8 27 tive staff Total salaries 17 21 35 22 78 Administrative 19 23 42 26 93 Depreciation 10 12 20 13 44 Provisions 4 3 9 Extraordinary 0 expenses Operating 0 subsidy Institution n.a. n.a. 230 145 511 building Total expenses 46 56 331 208 736 Net income (12) (31) (286) (180) n.a. Contributions to (12) (31) (286) revenues and expenses Credit Mutuel du Senegal 315 Table 25.4 (continued) For the twelve months ending December 1993 Amountfor Amountfor As a As a the twelve the twelve percentage percentage months end- months end- of average of net ing Decem- ing Decem- loan operating Item ber 1991 ber 1992 Amount portfolio income Change in - (15) revenues to expenses on balance sheet compared with above contribution n.a. Not applicable. -Not available. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Performance Membership in the Credit Mutuel system is open to any adult in the area served by the local caisses. Savings services are freely available and accessible. Members earn interest rates on their deposits comparable to those paid by commercial banks in Senegal. This is a high-quality savings service. The emphasis in the Credit Mutuel system is on developing savings more than on providing credit. Rural credit is authorized only once a year, and lending is constrained by regulations on the ratio of loans to deposits and strict cash collateral requirements. Membership grew slowly for the first few years but has expanded rapidly in recent years as CICM expands the network with a new phase of funding. Table 25.2 shows the recent growth in the system. The average loan size provided under the main credit line, rural credit, was approximately CFAF 66,900 ($223) in 1993 or 30 percent of per capita GNP for Senegal (1993 GNP per capita at the then exchange rate of 300 CFA francs to 1 U.S. dollar). Interest income for the twelve months ending December 31, 1993, covered 17.5 percent of operating expenses and 16.6 percent of operating plus financing costs (see tables 25.3 and 25.4). As with most programs that have high institution-building costs, these ratios increase when expatriate salaries and related costs are excluded (to 57 and 48 percent, respectively). 316 Appraisals of Selected Microfinance Institutions Interest income from investments is relatively high at CFAF 26 million ($87,000) for the period and is almost equal to the interest income from loans (CFAF 32 million, or $64,000)). This underscores Credit Mutuel's approach of restricting the volume of loans to less than 100 percent of the members' savings. Although the level of loans as a percentage of members' deposits increased over the twelve-month period, it is still at a very conservative level. As of December 31, 1993, the ratio of loans to savings was 68 percent, an increase from 55 percent in December 31, 1992. Income on the average level of investments would appear to be very high (9.6 percent) relative to interest rates paid by commercial banks on deposits (in the range of 6 percent a year). Either Credit Mutuel is doing a very good job of investing members' funds in low-risk investments or the actual average level of investments is considerably higher than the simple average. Both explanations are likely to be true. Loan loss provisions, taken for the first time in 1993, are assumed to represent the true value of potential loan losses. Of a total portfolio of CFAF 220 million ($440,000), Credit Mutuel showed a loan loss provision of CFAF 4 million ($8,000) or about 2 percent a year, which would indicate a very high-quality loan portfolio. This is as expected, considering Credit Mutuel's reluctance to risk members' savings and tight credit policies. Looking ahead CICM has prepared a plan to expand the network to eighty-seven caisses by 1998 in the four current regions of operation. One of its objectives is to attain a level of operation that does not require operating subsidies from the French government. The current plan does meet this particular definition of self-sufficiency. What the plan fails to address adequately, however, are the role and salaries of the French expatriate staff. As currently formulated, the plan does not consider the need or the feasibility of either replacing all of the expatriate positions with national staff or having the network generate sufficient income to pay the costs of all of the expatriates directly. CICM assumes that the director's position (which is to be filled by an expatriate) will be required indefinitely-at least until the year 2000-and that the salary and related expenses should not be paid by the network but rather by the French government. As long as cicM assumes that the director must always be an expatriate, no local person will be trained to take this position. This form of dependence may become contentious over the next few years. Other elements of the plan appear to be quite realistic. One area of risk is the issue of interest rates allowable on loans. cicM is well aware of the recent reductions in the discount rate and the related decline in allowable effective interest rates on loans in the country. Presently, the discount rate is 10.5 percent, which means that the maximum interest rate on loans is 21 percent a year. This is less than the rates now being charged borrowers in the Credit Mutuel system. CICM will have to reduce the interest rate charged on loans Credit Mutuel du Senegal 317 in the future but has already forecast this reduction. In addition, CICM has the option of lending a higher proportion of the members' savings in order to improve the interest income from the network. The key to the future self-sufficiency of the Credit Mutuel is the role of CICM and the level of lending in the network. To improve interest income, Credit Mutuel will have to lend more. This may increase the risk to members' savings. This risk can be offset by additional training and improved systems and procedures. CICM will have to concentrate on these institution-building activities. Included in these tasks will be the develop- ment of the unions and the federation structure and all that entails for representative processes and procedures. All of this institutional develop- ment will require a great deal of attention and focus from cicM, but it is apparent that CICM has developed what may yet become a truly Senegalese financial institution for the rural areas. Note Credit Mutuel was evaluated by Doug Salloum (consultant) and Anne- Marie Chidzero (Private Sector Development Department, World Bank) in January 1995. 26 Village Banks Nganda Doug Salloum Catholic Relief Services (CRS), based in Baltimore, Maryland, has been the driving force behind the creation and management of village banks in Senegal since 1989. In that year, CRS created a separate project management group within its operations in Senegal called the Small Enterprise Develop- ment Department to develop Foundation for International Community Assistance- (FINCA-) style village banking in rural areas of the country. Funding for this program has come primarily from USAID, first as one activity within a larger assistance program for small enterprise development (1989-93) and then, beginning in 1993, as a separate program financed by USAID, CRS, and other donors. Funding for the full project, which envisions eighty banks by 1998, has not yet been fully assembled. CRS implements its village banking program through a local partner, one of several professional associations. The associations are nongovernmental organizations, of which there are now twelve throughout the country, created at the departmental level to assist with economic development in rural areas. Through the Entente des Groupements Associes de Nganda (EGAN), CRS has established the network of village banks that are the subject of this review. As of September 1994, twelve village banks with a total membership of 600 individuals had been established in the Nganda area. Donor funding to create this network is estimated at CFAF 7 million ($14,000) as of end-1994. CRS has developed vilage banks in other regions of Senegal as well, most notably in the department of Birkelane. Working since 1989 with the Entente des Groupements Associes de Birkelane (EGABI), CRS established a network of twenty-six village banks in this region. By 1993 this network was well advanced and had approximately 1,300 members. Unfortunately, repayment problems began to surface, and by September 1994, the network was cut from further support. Some of the EGABI village banks reportedly have since accessed credit from Alliance du Credit et de l'Epargne pour la Production (ACEP), a Senegalese small business lending institution based in Dakar, but the fate of the remainder is unclear. An independent analysis of the collapse of the EGABI network would be very valuable. This review focuses on the Village Banks Nganda, which is an EGAN program that was only eighteen months old in December 1994. The department of Nganda is centered some 200 kilometers to the east of Dakar 319 320 Appraisals of Selected Microfinance Institutions in the geographic region known as the Sine-Saloum, so named for two major local rivers. The Nganda area is rural, sparsely populated, and poorly served by roads, electricity, water, and communications once off the main east-west road from Dakar to Tambacounda. The village of Nganda is a major regional center, and has only 1,200 inhabitants. The department's 151 villages have an estimated total population of 62,000 people. Peanut farming is the primary economic activity. Subsistence raising of goats and cattle and growing of millet and other grains and vegetables are common activities among most villagers. The village banking system as originally designed by FINCA, and as modified for Senegal, is intended to provide basic savings and loan services to rural populations. Village banks are capitalized by a combination of outside funds (from cRzs and donors, but through EGAN in this case) and members' savings. The structure is highly decentralized, with most decisionmaking and management carried out by village bank members. The objectives are financial-developing savings to accumulate and secure personal wealth and using credit as an input to increase that wealth-but also social in that community development is assumed to flow from community participation in the management of village banks. A long-term objective is for each village bank to accumulate sufficient savings and experience to be able to continue without outside assistance after three and a half years of operations. This stage of development has not yet occurred in the Nganda region, because the most advanced banks are not yet two years old. The ultimate objective is to establish an alternative financial system that would link formal and informal institutions through use of an apex institution that would funnel funds from the commercial banking sector to village banks. It is central to Catholic Relief Service's vision that EGAN be able to manage the network of village banks once they become financially independent and once CRS withdraws its support. The emphasis is on training the EGAN staff and directors. Program characteristics There are three levels of operation and management in the village bank network of Nganda: the village bank, the EGAN, and CRS/Senegal. The village bank A bank will only be established if the following criteria are met.' The village must be a member of EGAN, which currently limits the potential number of banks to 42 of the 151 villages in the department. Within an eligible village, at least thirty-five people (who must also be members of EGAN) are required to establish a village bank. The size of membership will depend on the managerial capacity of the village bank and available funding. These Village Banks Nganda 321 individuals are allowed to join the proposed bank in order of their seniority as EGAN members. Their membership dues must be current, and they must have saved money with EGAN for a period of two to three years. This self- selection process encourages group solidarity and accountability. A new requirement is that all EGAN members who wish to establish a bank within an eligible village must be women. Based on the lessons of the EGABI experience, women are expected to be better borrowers than men, and so a higher proportion of women borrowers is thought to reduce risk. Women represent approximately 50 percent of current membership. This figure increases to 60 percent with the inclusion of three new, all-women banks opened in January 1995. Once EGAN determines that the minimum membership is available in a particular village, it begins to organize and train prospective members. The steps in this process include the following: * Elections for the management committee are held. The committee consists of three individuals, all members of the new village bank: a president, a treasurer, and a secretary. These individuals hold office from one annual general meeting to the next, but, in practice, they stay on for several terms. All loan decisions amd record keeping are the responsibility of the management committee, whose members receive an honorarium as compensation: CFAF 15,000 ($30) for a first-cycle bank, CFAF 25,000 ($50) for a second-cycle bank, and CFAF 35,000 ($70) for a third-cycle bank (cycles last six months and begin in January and July). * Solidarity groups of six to ten persons are formed within the bank membership. This is a new approach. In the past, a default on a loan by any one member of a village bank would disqualify the whole bank from further credit. Now the consequences of delinquency and default will be borne by the solidarity group of the defaulting member rather than by the bank as a whole. * The basic features of the village banking system are presented by EGAN and discussed by new members until they are familiar with them. These include the necessity of monthly meetings, monthly savings contribu- tions, and complete loan repayment at the end of each six-month lending cycle. * Internal bank regulations and bylaws are discussed by members and written down by the secretary. These regulations describe meeting procedures and fines for late or poor participation, among others. There is some variety within the banks on issues such as the protocol for meetings and the imposition of fines. * Members must make a symbolic savings contribution of CFAF 2,000 ($4) prior to receiving their first loan. 322 Appraisals of Selected Microfinance Institutions Once established, village banks operate fairly autonomously, with some training and supervision from EGAN and occasionally from CRS. All records are kept in the village and maintained by the secretary. At monthly meetings, the management committee receives and records savings from members and disburses loans to members and nonmembers from the funds collected. A monitor from EGAN will usually observe and assist in the monthly meetings. EGAN The EGAN system is not to be confused with the village bank system. EGAN is one of twelve associations that have been created to assist rural develop- ment in given departments-in this case, that of Nganda. In addition to the Village Banking Project, EGAN administers other programs for a range of donors, including well-drilling and improvements, grain banks, literacy programs, health and environmental management education, and reforesta- tion and tree nursery programs. Set up in 1988 primarily by European donors to replace government-run rural development programs, the system was designed to operate autono- mously from the government. Apparently, none of the twelve associations now operating in Senegal is financially self-sufficient, but the plan is for them to earn sufficient revenues from membership and service fees eventually to become self-financing. The Village Bank Program of CRS is valued in part because of its social goals and accomplishments but also because of its contribution to EGAN'S revenues. EGAN receives funds for this program from two sources. As of September 1994, CRS had provided CFAF 20.4 million ($40,800) in loan capital. EGAN received most of this money as a grant (CFAF 17 million) and the rest as a low-interest loan that is to be repaid to the apex institution once it is established. EGAN lends this money to the village banks at an interest rate of 8 percent per six months. The grant also covers a portion of EGAN staff salaries. There are three levels within the system, and each level elects its own president, treasurer, and secretary: the subcommittee (one member village), the committee (a group of five to ten villages), and the administrative council and executive board. There are forty-two subcommittees, one for each member village. There are six committees, and one representative from each is on the executive board. The administrative council has twelve members, one of whom is a woman. EGAN is the intermediary between the village banks and CRS. It is responsible for training and overseeing each bank's management committee, supervising its financial activities, and lending loan funds to it. CRS provides EGAN with the funds it loans to the village banks and with training in village bank management and supervision. Three EGAN staff members are directly involved in the Village Bank Program: a program coordinator (40 percent of Village Banks Nganda 323 his time), an accountant (25 percent of his time), and a bank monitor (full time). CRS/Senegal CRS initiated the project and raised funds from donors to finance it. It now sees its role as being that of technical assistant to its local implementing partners. CRS has established partnerships with three other local organiza- tions in addition to EGAN: the Entente des Groupements Associes de Toubacouta, Caritas/Thies, and EGABI. With these four organizations, CRS hopes to develop a total of eighty village banks in Senegal by 1998. EGAN will oversee fifteen of them, all of which have been established now that the last three opened in January 1995. CRs has been working in Senegal since 1961 to assist other development programs, most notably literacy programs. cRs/Senegal has thirteen staff (including two expatriates), and the office is located in Dakar. Savings services In addition to the symbolic savings of CFAF 2,000 with which a member initiates his (and more commonly now, her) savings account, members commit to save a specific amount over the six-month loan cycle. In Nganda, 35 percent of the loan amount must be deposited as savings. Because all members of the system receive loans, all members also have savings. This commitment to save is met in equal installments at each monthly meeting.2 Savings transactions are entered in the members' savings account book and in the bank's records kept by the management committee. Transactions take place in full view of members at the regular monthly meetings. Members do not receive any interest on their savings. A member who wishes to withdraw savings may do so at the end of a loan cycle but would not be allowed to remain a member after doing so. Credit services Lending is based on cycles of six months that start at the beginning of July and the beginning of January. These periods were chosen because, in the rural economies in which the banks operate, these are neither the best nor the worst of times for members. The theory here is that when cash is available from the harvest (February and March), members are not thinking about investment opportunities. When cash is most scarce, in the dry months of April and May, members are more likely to spend the proceeds from loans on consumption than on productive activities. At the beginning of July 1994, the Nganda banks were entering the following cycles: first cycle (three banks), second cycle (two banks), and third cycle (seven banks). The banks provide three kinds of loans, and each 324 Appraisals of Selected Microfinance Institutions differs by source of funding (see table 26.1). Some loans are available only for members; others are available for both members and nonmembers. * External account loans. At the beginning of July 1994, EGAN made twelve loans, one to each of the village banks. Each loan was for six months, and the total lent was CFAF 20.3 million ($40,600). EGAN charges 8 percent interest for the six-month period on the full principal of the loans, or 16 percent a year. At the beginning of July, the banks on-lend these funds to members only. These loans are allowed to increase in size incrementally from CFAF 15,000 ($30) for the first-cycle loan to CFAF 150,000 ($300) for the seventh-cycle loan. These loans are for six months, and interest rates are 15 percent, or 30 percent a year. Interest and principal are due and payable in one installment at the term of the loan. * First internal account loans. The first internal account is the accumulated "profit" of each village bank plus members' savings contributed over the loan cycle. The banks have a notional "profit" because they borrow from EGAN at 8 percent and lend at 15 percent for the six-month loan cycle. At the beginning of July 1994, the first internal account contained an accumulated "profit" of CFAF 3 million ($6,000). Members' savings increased these funds to CFAF 12.2 million ($24,000) by the end of December. These funds were lent to members and nonmembers for one- month periods at an interest rate of 10 percent a month. * Second internal account loans. The accumulated savings of the members in July 1994 were CFAF 6.3 milion ($12,600). The banks invested these funds in two ways: CFAF 4.3 million ($8,600) were deposited in a commercial bank (earning no interest for most of the period from July to December), and CFAF 2 million ($4,000) were lent to members and nonmembers at 25 percent interest for five months. Table 26.1 Composition of loans in Village Banks Nganda in Senegal, 1994 (millions of CFA francs) Amount outstanding as of Amount outstanding as of Type of loan July 1994 December 1994 External account loans 20.3 20.3 First internal account loans 3.0 12.2 Second internal account 2.0 0.0' loans Total loan portfolio 25.3 32.5 a. Repaid at the beginning of December after five months. The average interest rate earned on all loans made by the twelve village banks from July to December 1994 was 27 percent, or 54 percent a year, but substantially higher if the forgone interest on savings is included. The Village Banks Nganda 325 average loan financed by the external account for the period was CFAF 33,800 ($68). External account loans represent 80 percent by value of all loans provided in July 1994 (no information was provided on the number or size of loans made from the two internal accounts or the number of members and nonmembers who borrowed these funds). Projects presented by members or nonmembers are not evaluated for their viability. All members who apply for a loan receive one, and the vast majority apply for less than the maximum loan size allowable. There are no credit criteria for nonmembers either, although some form of character lending likely occurs so that villagers of questionable reputation are excluded. No collateral is pledged to support the loans. Rather, the system relies on group solidarity and mutual guarantees. If a member defaults on a loan, the bank has the right to seize his or her assets and to sell them to repay the loan from the proceeds. This is apparently not a consistent feature of all of the banks' regulations. It is also unlikely that the banks have any right to seize the assets of a nonmember who is not party to the implied membership or loan agreement. Loans are put to the following uses: commerce (45 percent), livestock raising (44 percent), agriculture (4 percent), relending (5 percent), and other activities (2 percent). Governance All three levels of the Village Banks Nganda program are involved in managing and directing the system. The members of each bank elect their own management committee at the annual general meeting, where EGAN representatives present the bank's progress. Members can present new initiatives, but, in practice, much of the direction comes from EGAN and from CRS. Standard features and any new regulations (such as the requirement that all new members must be women or that solidarity groups must be formed) are unlikely to originate at the member level. To a large degree, ownership of the banks is not actually in the hands of the members. They cannot withdraw savings without losing their member- ship, they earn no interest income from their savings, and they do not receive dividends from the profits of the banks. EGAN'S governing structure has been designed to represent its own members through democratically elected representatives and officers at all levels. However, this is not the same as saying that EGAN is democratically empowered to represent the interests of the banks or its members (even though all bank members are also EGAN members). The almost complete absence of women at the administrative council and executive levels underscores the fact that EGAN does not necessarily represent bank members. 326 Appraisals of Selected Microfinance Institutions There is no real accountability between EGAN and the banks. EGAN was not created by the banks, as has occurred in some other rural credit programs; rather EGAN created the banks and, on some level, probably feels some sense of ownership toward them. Without an effective form of representation from the banks themselves, this sense of ownership may result in conflicts of interest at some point. The staff and management of EGAN seem genuinely committed to the village banks, but if an issue arises that pits the banks' interest against that of EGAN, the first loyalty of the EGAN staff and executive may not be to the banks. Performance The number of banks in the EGAN region has grown steadily but not dramatically. Furthermore, the number of village banks is not scheduled to increase significantly above the current level of fifteen (if funding were available from donors, an additional four banks could be opened). By maintaining membership at roughly 50 members per bank, the system will reach a membership of approximately 750 (950 if four additional banks were opened). The banks directly affect more than their members, because nonmembers are able to receive loans, but the number of these individuals is neither forecast nor followed by CRS or EGAN. To determine market penetration, a clear distinction must be made between potential and eligible membership in the banks (see table 26.2). Potential membership would include all adult residents of the department of Nganda-some 30,000 persons. However, in the village bank system as developed by EGAN, three conditions of eligibility must be met to qualify for membership: the village must be a member of EGAN, the individual in the member village must also be a member of EGAN, and bank members must be women.3 Table 26.2 Potential membership in Village Banks Nganda in Senegal Bank members as a Potential membership percentage of potential Location (above eighteen years of age) membership 15 villages with banks 4,160 18 42 EGAN villages 10,500 7 151 villages in the depart- 31,000 2 ment of Nganda Application of these three criteria reduces the potential eligible member- ship to somewhere in the range of 1,100 to 1,500 persons (the actual ratio of men to women in the EGAN membership was not available). Therefore, the village bank system is serving between half and two-thirds of the eligible Village Banks Nganda 327 population in the department of Nganda, but it could also be said that only 2 percent of the potential membership is being served. It is very obvious from observation and analysis that most members of the village banks are very low-income individuals who live in a relatively remote region of Senegal. Of the 600 loans provided to members during the loan cycle ending December 1994, the average loan size was CFAF 33,800 ($68) or 15 percent of the per capita GNP for Senegal (1993 GNP per capita at the then exchange rate of 300 CFA francs to 1 U.S. dollar). From a purely financial perspective, the quality of services provided is uneven. A portion of the loans provided by the banks is available through- out the year: loans funded by the first internal account. This is a very responsive system that allows members and nonmembers access to credit almost on demand so that borrowers can respond to investment opportuni- ties or financial needs of the family as they occur. Other loans (from the external account or the second internal account) are available only at the beginning of each six-month cycle. Deposit transactions are not voluntary or particularly responsive to members' needs. The fact that members receive no interest and cannot withdraw funds unless they leave the bank makes these contributions appear more like a tax than like savings. CRS proposes to pay interest on deposits in the future and perhaps to provide more access to individual savings, but these details have not yet been worked out. In response to these limitations, members could stop saving, withdraw their money, and cease to borrow, which would result in the collapse of the banks. In sum, if the quality of savings services is not improved, the survival of the banks is at risk. It was not possible to review either a balance sheet or income statement for the Village Banks of Nganda, because these statements are not produced by EGAN or CRS. An effort was made to construct these statements for the six- month period from July to December 1994 based on interviews with CRS and EGAN staff and reviews, where available, of other information (primarily cash balances for the various accounts; see tables 26.3 and 26.4). A number of assumptions and simplifications were necessary but, in the opinion of the evaluators, the analysis and conclusions drawn from these constructed financial statements are valid even if they differ in small ways from the actual situation. Interest income earned during the period July-December 1994 for the twelve banks is estimated to have covered operating expenses by 160 percent and operating plus financial expenses by 123 percent. In other words, before inflation costs are considered, the Village Banks Nganda are profitable. This is a major achievement, and it is due to both the aggressive pricing of loans to members and nonmembers (an average interest rate of 54 percent a year) and a very low operating and administrative cost structure. Also, the fact that the banks leverage members' savings with debt from EGAN 328 Appraisals of Selected Microfinance Institutions Table 26.3 Adjusted financial statements of Village Banks Nganda in Senegal: Balance sheet, 1994 (millions of CFA francs) Item As of July 1994 As of December 1994 Assets Cash n.a. 3.7 Investments 4.3 4.3 Loan portfolio 25.3 32.5 (Provision for loan losses) n.a. n.a. Fixed assets 6.0 6.0 (Depreciation) (1.2) (1.8) Other assets n.a. n.a. Total assets 34.4 44.7 Liabilities Demand deposits n.a. n.a. Term deposits 6.3 13.4 Payables n.a. n.a. Loans 20.3 20.3 Total liabilities 26.6 33.7 Equity Donors' equity n.a. n.a. For fixed assets 6.0 6.0 For operations 4.7 7.0 Total donors' equity 10.7 13.0 Members' equity n.a. n.a. Paid-in capita n.a. n.a. Reserves (interest account) 3.0 8.8 Retained earnings (5.9) (10.8) Total members' equity (2.9) (2.0) Total equity 7.8 11.0 Total equity plus liabilities 34.4 44.7 n.a. Not applicable. Note: The average loan portfolio in the third cycle was CFAF 29.6 million. See the financial notes in annex 3. IThis financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. Village Banks Nganda 329 Table 26.4 Adjusted financial statements of Village Banks Nganda in Senegal: Income statement, 1994 (millions of CFA francs) Six months As a percentage of As a percentage of ending average loan net operating Item December 1994 portfolio income Income Credit income 8.0 54.1 125 Investment income 0.0 0.3 0 (Interest expense) (1.6) (11.0) 25 Net interest income 6.4 43.4 100 Other income 0.0 0 0 Net operating income 6.4 43.4 100 Expenses Salaries 0.8 5.1 13 Field staff 0.6 3.9 9 Administrative staff 0.2 1.2 3 Administrative 1.2 8.1 19 Depreciation 0.6 4.1 9 Provisions 0.0 0.0 0 Extraordinary expenses 0.0 0 Operating subsidy 0.0 0 Institution building 2.3 15.5 36 Total expenses 4.9 32.8 77 Net income 1.6 10.5 n.a. Contributions to internal 5.8 n.a. n.a. account Contribution to the ratio of (4.2) n.a. n.a. revenues to expenses n.a. Not applicable. Note: See the financial notes in annex 3. This financial statement was reconstructed from data provided by the institution. It has not been formally audited but was approved by the institution's management. improves profit significantly (loans as a multiple of savings varied from a high of 4.0 in July to a low of 2.5 in December). The banks will become even more profitable as the loan portfolios grow in size. Loan sizes are expected to increase as members borrow larger amounts and as more savings are made available to be loaned out. The quality of loans appears to be sound, although there was no way to confirm the contention that all loans advanced during the period reviewed were repaid in full with interest. Based on this constantly stated but 330 Appraisals of Selected Microfinance Institutions unconfirmed perfect repayment record, no loan loss provision was included in the constructed financial statements for the banks. Any default or delinquency according to the methodology should result in a bank-or a solidarity group-losing access to further loans, and there was no evidence of this occurrence. Looking ahead Attaining financial viability is not necessarily the primary challenge for village banks. Pricing and costs are currently in line to such an extent that the banks could pay commercial rates for their funds and still break even. If, for example, the banks had paid 5 percent on members' deposits and 25 percent a year on the loan from EGAN, net income would have dropped from CFAF 1.6 million to CFAF 0.4 million for the period, but the banks would still have been profitable. As the loan portfolio increases, the village banks should be able to show true commercial viability (including an inflation cost component to maintain in real terms the value of the banks' equity) if agreement can be reached as to what the banks' equity actually is and who owns it. It is basic to the village banking methodology that banks "graduate" (operate with mobilized funds or establish links with another institution) after they complete the seventh loan cycle. CRS anticipates that an apex organization will be created to on-lend funds to the "graduated" banks. It is unclear how the banks will graduate and make use of an apex structure, but CRS has at least one year in which to work out the details plus an additional year in which to implement and make the new institution operational.4 That this apex institution should be able to access commercial financing is central to the plans. The level of external financing that would be required by the eighty proposed banks is not yet known, but it will likely be at least two or three times the savings of members if a major reduction in the loan portfolio is to be avoided. If CRS can assist EGAN in developing and producing audited financial statements that confirm the levels of profit and loan default, there would appear to be strong potential for accessing commercial capital in the future. This potential cannot be realized, however, with the accounting records currently produced by the banks. CRS will need to address the following issues in the near future: . Real and effective ownership and control of the banks including the disposition of profits . Interest rates and access policies with respect to members' savings * Maintenance of loan quality as the size of loans and the number of nonmember borrowers increase Village Banks Nganda 331 . The future role and responsibility of EGAN in the direction and supervi- sion of the banks * Compliance with or application for exemption from the new laws governing cooperative savings and credit organizations. Notes 1. Applying these criteria and taking into account the current EGAN membership, CRS plans to open only fifteen village banks in the Nganda area by the end of 1998. 2. For example, to receive a loan of CFAF 100,000, members must agree to repay that loan plus interest at the end of six months and to contribute savings of CFAF 35,000 over the same period. These savings contributions are made at a rate of CFAF 5,800 a month paid in at each of the first five monthly meetings and CFAF 6,000 contributed at the sixth. 3. The combination of the first two criteria reduces the eligible member- ship in the forty-two EGAN villages to a maximum of 2,200 persons. 4. Of the fifteen banks now operating, seven are entering their fourth cycle and will "graduate" two years from now. References Adams, Dale. 1994. "Altruistic or Production Finance? A Donor's Dilemma." Occasional Paper 2150. Ohio State University, Columbus. Ahmed, I., and others. 1992. Poverty in the Gambia. Geneva: ILO. Aryeetey, Ernest, and William F. Steel. 1994. "Savings Collectors and Financial Intermediation in Ghana." Paper presented to the thirty-seventh annual meeting of the African Studies Association, Toronto, November 3-6. Aryeetey, Ernest, Amoah Baah-Nuakoh, Tamara Duggleby, Hemamala Hettige, and William F. Steel. 1994. Supply and Demandfor Finance of Small Enterprises in Ghana. World Bank Discussion Paper 251. Washington, D.C.: World Bank. Ashe, Jeffrey, Madeline Hirschland, Jill Burnett, Kathleen Stack, Marcy Eiland, and Mark Gizzi. 1992. "Access to Credit for Poor Women: A Scale-up Study of Projects Carried out by Freedom from Hunger in Mali and Ghana." Technical Report 33. GEMINI, Bethesda, Md., March. Bayard-Gamatie, Mariama. 1993. "Bilan de la politique d'integration de la femme au developpement au Niger." Esquisse d'une Analyse des Documents Produits sur la Question. Niamey. Beoku-Betts, Josephine. 1990. "Agricultural Development in Sierra Leone: Implications for Rural Women in the Aftermath of the Women's Decade." Africa Today 37 (1): 19-35. Birks, Stace, Fred Fluitman, Xavier Oudin, Bernard Salome, and Clive Sinclair. 1992. "Skill Acquisition and Work in Micro-Enterprises: Recent Evidence from West Africa." Technical Paper 4. Africa Technical Department, World Bank, Washington, D.C. "Blueprint for Development." 1994. West Africa 16 (12, January 10-16). Burkina Faso, Ministcre de l'Emploi, du Travail, et de la Securite Sociale, Secretariat Gen6ral, Direction de la Promotion du Secteur Informel. 1993. Journ6es nationales de reflexion sur le secteur informel. Ouagadougou. Caisse Francaise de Developpement. 1991. "Rapport de mission sur la mobilisation de l'epargne et du credit decentralise au Senegal." Catholic Relief Services. 1994. "Innovation and Transfer." SED Newsletter 3 (1, January). Charmes, Jacques. 1992. "La contribution du secteur informel i l'emploi et au produit national en Mauritanie, 1977-1992." Islamic Republic of Mauritania, Ministry of Planning, Nouakchott. Chen, Martha Alter. 1992. "The Impact of Grameen Bank's Credit Opera- tions on Its Members: Past and Present Research." Grameen Bank Mid- term Review Mission. Christen, Robert Peck, Elisabeth Rhyne, and Robert Vogel. 1994. "Maximiz- ing the Outreach of Microenterprise Finance: The Emerging Lessons of Successful Programs." International Management Communications Corp., Arlington, Va. 333 334 References Clement, Peg, Maria Judita Souto, and John Magill. 1989. "Employment and Income Opportunities for Women in Cape Verde: A Review of Options, Constraints, and Program Implications for USAID/Cape Verde." Labat- Anderson Inc., Arlington, Va. CLUSA (Cooperative League of the United States of America). 1993. "Food Crop Production and Marketing Improvement Project: Farmer Associa- tion Development Program in Sao Tome and Principe. Annual Report." Washington, D.C. Crowley, E. 1993. "Guinea-Bissau's Informal Economy and Its Contributions to Economic Growth." Report prepared for usAID/Bissau. [Bissau]. December. de Soto, Hernando. 1989. The Other Path. New York: Harper and Row. Economist Intelligence Unit. 1994a. Country Profile: Congo, Sao Tome and Principe, Guinea-Bissau, Cape Verde. London. . 1994b. Country Profile: Guinea, Sierra Leone, Liberia. London. Flaming, Mark W., Quy Nguyen, and Gabriel M. Sellu. 1993. "External Mid- Term Evaluation: The Small Enterprise Development and Training Project in Sierra Leone." oici and USAID Bureau for Africa. GEMINI (Growth and Equity through Microenterprise Investments and Institutions). 1990. Prospects for Enhancing the Performance of Micro and Small-Scale Nonfarm E.nterprises in Niger. Working Paper 3. Bethesda, Md. February. .1993. "A Review of Donor-Funded Projects in Support of Micro- and Small-Scale Enterprises in West Africa." Technical Report 54a. Bethesda, Md. Graham, D., R. Meyer, and C. Cuevas, ed. 1992. "A Review of Formal and Informal Finance in the Gambia, 1981-91: Markets, Institutions, Pro- grams, and Policies." A draft report to the USAID Mission, Banjul. Grant, William. 1991. "Mali Microenterprise Sector Assessment and Strategy." Technical Report 20. GEMINI, Bethesda, Md. . 1993. "A Reviewv of Donor-Funded Projects in Support of Micro and Small-Scale Enterprises in West Africa." Technical Report 54A. GEMINI, Bethesda, Md. Grant, William, and others. 1991. Burkina Faso Microenterprise Sector Assessment and Strategy. Vol. 1: Main Report. Technical Report 18. GEMINI, Bethesda, Md. Hadjimichael, Michael, and others. 1992. The Gambia: Economic Adjustment in a Small Open Economy. Washington, D.C.: International Monetary Fund. Hennart, Jean-Francois. 1992. "The Informal Sector in Chad, Institutional Reform, and the Informal Sector Project." University of Maryland, College Park. Hulme, David. 1995. "Finance for the Poor, Poorer, or Poorest? Financial Innovation, Poverty, and Vulnerability." Paper presented at the confer- ence on Finance against Poverty at the University of Reading, March 27-28. References 335 ILO (International Labor Organization). 1972. "Employment, Incomes, and Equality: A Strategy for Increasing Productive Employment in Kenya." Geneva. . 1990. "L'economie informelle en Guinee: Analyse et strat6gie de developpement." Geneva. . 1991. "Por uma prornoc,ao das micro e pequenas empresas arteanais em Sao Tome e Principe." Serviqo do Desenvolvimento e da Gesto das Empresas, Geneva. IMF (International Monetary Fund). 1992. "Sierra Leone: Economic Pro- gram." IMF Survey, April 13. Washington, D.C. [RAM (Institut de Recherches et d'Application des Methodologies de Developpement). 1976. The Participation of Rural Women in Development. Paris. Khandker, Shahidur R., Baqui Khalily, and Zahed Khan. 1994. "Sustainability of the Grameen Bank: What Do We Know?" World Bank, Washington, D.C. Khandker, Shahidur R., Zahed Khan, and Baqui Khalily. 1994. "Sustainability of a Government-Targeted Credit Program: The BRDB RD- 12 Project in Bangladesh.' World Bank, Washington, D.C. Khandker, Shahidur R., and Baqui Khalily. 1994. "Designing a Sustainable Poverty Alleviation Program: The BRAC Strategy in Bangladesh." World Bank, Washington, D.C. Labat-Anderson, Inc. 1990. "Manual for Action in the Private Sector (MAPS), UsAID-Senegal. Phase II: Private Sector Description of Senegal." Arlington, Va. Liedholm, Carl, and Donald Mead. 1987. "Small-Scale Industries in Developing Countries: Empirical Evidence and Policy Implications." International Development Paper 9. Michigan State University, East Lansing. Lubell, Harold, and Charbe] Zarour. 1990. "Resilience amidst Crisis: The Informal Sector of Dakar." International Labour Review 129 (3). Malhotra, Mohini. 1994. "Successful MSE and SME Finance Programs." Development Alternatives, Inc., Bethesda, Md. Mazumdar, Dipak. 1994. "The Structure of Wages in African Manufactur- ing." Regional Program on Enterprise Development, Africa Technical Department, World Bank, Washington, D.C. Mead, Donald. 1994. "The Contribution of Small Enterprises to Employment Growth in Southern and Eastern Africa." Michigan State University, East Lansing. Muteta, G. K., and A. Niang. 1992. "Etude sur le secteur informel urbain de Dakar." Synergie Consultants for Development. Report prepared for Poverty and Social Policy Division, Africa Technical Department, World Bank, Washington, D.C. Nicolas, Robert. 1994a. "Draft Field Report: Tombali Sustainable Rural Initiatives Project, Feasibility Study and Program Design." Africare, Guinea-Bissau. 336 References Nicolas, Robert. 1994b. "Executive Summary: Tombali Sustainable Rural Initiatives Project." Africare, Curtis Lewis and Associates, P.C., Washing- ton, D.C. May 4. Ouedraogo, Idrissa, and Jacques Ouedraogo. 1994. "Etude sur le secteur informel au Burkina Faso." World Bank Resident Mission, Ouagadougou. September. Peck, Jerry. 1994. "Sao Tome: Considerations for a Possible Future Opera- tion." Office Memorandum. Washington, D.C.: World Bank. Rhyne, Elisabeth, and Sharon Holt. 1994. "Women in Finance and Enterprise Development." Discussion Paper 40. Education and Social Policy Department, World Bank, Washington, D.C. Rhyne, Elisabeth, and Mara Otero. 1991. "A Financial Systems Approach to Microenterprises." Working Paper 18. GEMINI, Bethesda, Md. Rhyne, Elisabeth, and Linda Rotblatt. 1994. "What Makes Them Tick? Exploring the Anatomy of Major Microenterprise Finance Organizations." Monograph Series 9. ACCION International. Santos, Milton. 1979. The Shared Space: The Two Circuits at the Urban Economy in Underdeveloped Countries. London: Methuen Press. Schmidt, Reinhard, and C. P. Zeitinger. 1994. "Critical Issues in Small and Microbusiness Finance." Interdisziplinare Projekt Consult GmbH, Frankfurt. Shipton, Parker. 1990. "How Gambians Save and What Their Strategies Imply for International Aid." PRE Working Paper 395. Policy Research Department, World Bank, Washington, D.C. Shipton, Parker, ed. 1992. Informal Finance in Low-Income Countries. Boulder, Colo.: Westview Press. Sierra Leone, Central Statistics Office. 1991. "Report on Labor Force Survey 1988-1989." Freetown. Silcox, S. 1993. "The Gambia: Handicraft Sector Assessment with Recom- mendations and Action Plan." Ministry of Trade, Industry, and Employ- ment and USAID/Banjul. September. Societ6 de D6veloppement International Desjardins. 1990. "Programrnme pilote visant l'emergence des coop6ratives d'epargne credit en zone rurale et urbaine en Mauritanie: Etude de faisabilite." SDID 4417-01-05 (6056). Levis, Canada. November. Stern, Joseph B. 1994. "Poverty Reduction in Sierra Leone: A Framework for a National Action Plan." Draft of a working paper series on poverty reduction. Environment and Social Policy Division, Central Projects Department, African Development Bank Group, Abidjan, C6te d'Ivoire. Swedish International Development Agency. 1989. Guinea-Bissau: Economic Difficulties and Prospectsfor Structural Adjustment. Stockholm. UNDP (United Nations Development Programme). 1991. Co-operation au developpement guinee. Conakry. . 1993. "Report on Development Cooperation for Sierra Leone." USAID (U.S. Agency for International Development). 1992. "Trade and Investment Promotion Support (TIPS) Project Paper." References 337 U.S. Department of State Dispatch. 1994a. "Mali Human Rights Practices: 1993." Washington, D.C. . 1994b. "Niger Human Rights Practices: 1993.: Washington, D.C. van den Brink, Rogier, and Jean Paul Chavas. 1991. "The Microeconomics of an Indigenous Africa Institution: The Rotating Savings and Credit Association." Working Paper 15. Cornell Food and Nutrition Policy Program, Washington, D.C. van Dijk, Meine Pieter. 1986. Burkina Faso: Le secteur informel de Ouagadougou. Paris: Editions L'Harmattan. VITA/Chad Private Enterprise Development Project. 1992. "Project Expan- sion Proposal." . 1994. Quarterly Progress Report, First Quarter CY 1994. Wahid, Abu, ed. 1993. The Grameen Bank: Poverty Relief in Bangladesh. Boulder, Colo. Westview Press. Webster, Leila, Randall Riopelle, and Anne-Marie Chidzero. 1994. "World Bank Lending for Small Fnterprises, 1989-1993." Private Sector Develop- ment Department, World Bank, Washington, D.C. Woilet, Jean-Claude. 1992. "Le secteur artisanal et la consommation de biens et de services artisanaux au Cap Vert: Situation et Perspectives." Bureau International du Travail, Service du Developpement et de la Gestion des Entreprises, Geneva. March. World Bank. 1985a. "Cape Verde: Industrial Finance and Promotion Project. Staff Appraisal Report." Washington, D.C. . 1985b. "Islamic Republic of Mauritania. Country Economic Memorandum." Western Africa Regional Office, Washington, D.C. -. 1989. "Working Papers on the Urban Informal Sector in the Sahel." Western Africa Department, Washington, D.C. . 1990a. "Republic of Chad, Social Development Action Project. Staff Appraisal Report." Western Africa Department, Washington, D.C. .1990b. World Development Report 1990. New York: Oxford University Press. -. 1992. "Senegal: Women in Development. Country Assessment and Strategy." Western Africa Department, Washington, D.C. -. 1993a. "Cape Verde: Opening up a Small Economy. An Agenda for the 1990s." Report 10594-C. Country Operations Division, Western Africa Department, Washington, D.C. -. 1993b. "Democratic Republic of Sao Tome and Principe: Country Economic Memorandum and Key Elements of an Environmental Strategy," vol. 1. Report 10383-STP. Washington, D.C. 1993c. "The Gambia. An Assessment of Poverty." Washington, D.C. 1993d. "Mali: An Assessment of Living Conditions." Washington, D.C. 1993e. "Niger WID Assessment." Washington, D.C. 1993f. "President's Report on a Structural Adjustment Credit to the Republic of Sierra Leone." Report P-6157-SL. Washington, D.C. Septem- ber. 338 References - 1993g. "S6negal: Le secteur financier informel et ses implications pour le developpement du secteur financier." Report 11749-SE. Industry and Energy Division, Western Africa Department, World Bank, Washing- ton, D.C. . 1993h. "Senegal: Macroeconomic Update Report." Western Africa Department, Washington, D.C. 1993i. "Senegal: Private Sector Assessment." Washington, D.C. 1993j. "Sierra Leone: Policies for Sustained Economic Growth and Poverty Alleviation." Report 11371-SL. Washington, D.C. May. . 1993k. Trends in Developing Economies 1993. Washington, D.C. 19931. World Development Report 1993. New York: Oxford University Press. -. 1994a. Adjustment in Africa: Reforms, Results, and the Road Ahead. New York: Oxford University Press. -. 1994b "Chad: Country Environmental Strategy Paper." Report 13174-CD. Country Operations Division, Western Africa Department, Washington, D.C. -. 1994c. "Mauritania: Poverty Assessment." Western Africa Depart- ment, Washington, D.C. . 1994d. "Poverty in Cape Verde: A Summary Assessment and a Strategy for Its Alleviation." Report 13126-CV. Country Operations Division, Western Africa Department, Washington, D.C. . 1994e. "President's Report on a Proposed Credit to the Republic of Guinea for a Financial Sector Operation." Washington, D.C. August 31. - ~. 1994f. "Republic of Chad: Public Works and Capacity Building Project. Staff Appraisal Report." Western Africa Department, Washing- ton, D.C. -. 1994g. "Senegal: Private Sector Assessment." Industry and Energy Division, Western Africa Department, Washington, D.C. - . 1994h. Social Inclicators of Development 1994. Baltimore, Md.: Johns Hopkins University Press. -. 1994i. "Staff Appraisal Report: Islamic Republic of Mauritania, Technical Education and Vocational Training Support Project." Popula- tion and Human Resources Operations Division, Western Africa Department, Washington, D.C. . 1994j. World Development Report 1994. New York: Oxford University Press. . 1995. "Republic of Chad: Agricultural and Livestock Services Project." Report 13618-CD. Agriculture and Environment Division, Central Africa and Indian Ocean Department, Washington, D.C. World Council of Credit Unions. 1994. "Development of the Nigerien Credit Union Movement, Phase II, January-March 1994." Progress Report 4. Yaron, Jacob. 1992. Successful Rural Finance Institutions. Discussion Paper 150. Washington, D.C.: World Bank. Annex 1 Aggregate Data for Microfinance Institutions in West Africa Peter Fidler 339 Table A1.1 Basic characteristics of nine microfinance institutions in West Africa, 1994 Average asset Institution, coun- Number of growth try, Type of institu- Total number Number of staff Number of regional Number of Total assets (last twelve and year founded Data as of tion of staff at headquarters caisses offices members (U.S. dollars) months) Guinea PRIDE, 1991 7/94 Credit pro- 67a 10 6 0 n.a.' 907,000 (7/94) 60 (7/94) gram Credit Rural, 12/94 Credit union 120 23 41 6 31,000 3.7 million 27 (12/93) 1989 (12/93) Credit Mutuel de 12/94 Credit union 151 11 70 8' 45,000 5 million - (12/93) Guinee, 1988 (12/93) Senegal Credit Mutuel du 8/94 Credit union 83 7 54 4 22,000 2.8 million 138 (12/93) Senegal, 1988 (12/93) Village Banks 12/94 Village banks 3.5 1 12 0 600 90,000 (12/94) - (12/94) Nganda, 1993 Burkina Faso Reseau des 9/94 Credit union 110 15 35 3 64,000 5.1 million 60 (9/94) Caisses (9/94) Populaires, 1972 Sahel Action 12/94 Credit pro- 57 7 0 d 4 n.a b 760,000 - (12/93) PPPCR, 1988 gram (12/93) Z Table A1.1 (continued) Average asset Number of re- growth Institution, country, Type of institu- Total number of Number of staff Number of gional Number of mem- Total assets (U.S. (last twelve and year founded Data as of tion staff at headquarters caisses offices bers dollars) months) Mali Caisses Villageoises 12/94 Village banks 14 plus volun- 11 54 0 15,330 360,000 (12/93) 50 (12/93) du pays Dogon, teers' 1988 Kafo Jiginew, 1988 12/94 Credit union 40 12 51 0 20,670 1.5 million 36 (9/93) (9/93) n.a. Not applicable. - Not available. Note: Total assets and average growth of assets are as of the date of the last balance sheet for each institution. a. PRIDE has thirty-one field agents and ten staff members at the central office. The remaining twenty-six staff include security guards, drivers, and maintenance personnel. b. PRIDE and Sahel Action do not have formal members. c. To reduce costs, Credit Mutuel is reducing the number of regional offices to four. d. Sahel Action does not have any caisses. Mobile agents operate in 191 villages. e. Caisses Villageoises du Pays Dogon has more than 130 volunteers who work as field agents on a part-time basis. Table A1.2 Basic credit characteristics of nine microfinance institutions in West Africa, 1994 Growth in Growth in number of Average Maximum loan portfolio Number loans (% size of size of Average Outstanding (% during Nominal Effective of loans during last loans loans loan size portfolio last annual annual Quality of Data Number of outstand- twelve (U.S. (U.S. per GNP (U.S. twelve interest interest Term of Guarantees the loan Institution as of members ing months) dollars) dollars) per capita' dollars) months) rate (%) rate (%) loan required portfolio' PRIDE 7/94 - 3,500 77' 200 Soo 44 447,000 (7/94) 60 (7/94) 36 (7/94)d 37 (7/94) 6-12 4 percent of 100 months' loan in guar- antee fund Credit Rural 12/94 31,000 13,180 - 100 1,000 14 1.3 million 71(12/93) 30-36 32 (12/93) Less than 5-10 percent 97 (12/93) (12/93) (12/93) (12/93) 24 months of loan prin- cipal Credit Mutuel 12/94 45,000 2,800 44 1,000 10,000' 200 1.6 million - (12/93) 30(12/93) 22 (12/93) Maximum 50 percent in 80 de Guinee (12/93) 24 savings months; average 12 months Cr6dit Mutuel 8/94 22,000 2,377 85 132 300 29 (12/93) 733,000 182 (12/93) 24 (12/93) 20 (12/93) Maximum 25 percent of 98 (more du S6n6galg (12/93) (12/93) 12 loan than 90 months; principal5 days)' average 6 months Village Banks 12/94 600 871 _ 68 300 Less than 65,000 -(12/94) 54(12/94) 54 (12/94) 6 months 35 percent in 100 Nganda' 15 percent (12/94) savings (12/93) Reseau des 9/94 64,000 3,710 - 336 6,000' More than 1.2 million -(9/94) 16 (9/94) 19 (9/94) 10 months 100 percent 95 (more Caisses 100 per- (9/94) to 5 years in assets plus than 90 Populaires cent 25-30 per- days)> (12/93) cent in sav- ings Sahel Action 12/94 - 9,526 95 43 120 22(12/93) 316,000 -(12/93) 20-25 26 (12/93) 6-13 Up to 10 100 PPPCR (12/93) (12/93) months percentof loan princi- pal Table A1.2 (continued) Growth in number of Growth in loans (per- Average Maximum loan portfolio Nominal Effective Number centage dur- size of size of Average (percentage annual annual of loans ing last loans loans loan size Outstanding during last interest interest Quality of Data Number of outstand- twelve (U.S. (U.S. per GNP portfolio twelve rate (per- rate (per- Termn of Guarantees the loan Institution as of members ing months) dollars) dollars) per capita' (U.S. dollars) months) cent) cent) loan required portfoliot Caisses 12/94 15,330 4,109 17 49 - 17 (12/93) 227,000 72 (12/93) 40 (12/93) 39 (12/93) Maximum Up to 100 94 Villageoises du (12/93) 9 months; percent in' pays Dogon average 4 assets months Kafo jiginew 10/94 20,670 5,048 62 105 116 (9/93) 36 (9/93) 813,000 (9/93) 39 (9/93) 24 (9/93) 19 (9/93) 1-9 Up to 100 98 (9/93) (9/93) months percent in assets -Not available. Note. Total assets and average growth of assets are as of the date of the last balance sheet for each institution. a. Calculations to obtain the average loan size as a percentage of GNP per capita for programs in CFA countries use the 1993 predevaluation average loan size in order to correspond to the predevaluation GDP per capita rate. Figures for the Guinean programs (PRIDE, Credit Rural, and Cr6dit Mutuel de Guinme) use the 1994 average loan sizes. Figures for Village Banks Nganda and the Reseau des Caisses Populaires are estimates. b. Loans are deemed in arrears if the repayment is thirty days past due. All repayment rates are determined on this basis unless otherwise noted. c. The annual growth in the number of borrowers, not in the number of loans. d. PRtDE plans to lower interest rates to 24 percent, while raising its fees by adding GF 650 ($0.65) to every repayment to finance training services. e. PRPDE is introducing mid-term loans for one to two years ($500-$5,000) at 24 percent interest plus processing fees. f. On exception, loans are made for $20,000. g. Figures for average size of Joans, maximum size of loans, average loan per GNP per capita, and repayment rates are based solely on data from rural credit. h. This does not apply to rural credit. The guarantee scheme for rural credit loans is currently under review. i. Credit Mutuel de Guin6e only keeps repayment records on a ninety-day past-due basis. j Extemal account loans carry rates of 30 percent a year, first intemal account loans charge 10 percent a month, and second internal account loans charge 25 percent for five months. k. Figures for average size of loans, maximum size of loans, average loan per GNP per capita, and repayment rates are based solely on data from the extemal account. 1. At the end of 1994, the maximum loan size was increased from $600 to $6,000. m. Reseau des Caisses Populaires only keeps repayment records on a ninety-day past-due basis. n. Based only on loans, which account for 75 percent of all credits granted. Table A1.3 Savings characteristics of nine microenterprise institutions in West Africa, 1994 Average sav- Savings as a percentage Data Number of ings account Total savings Annual Percentage of of outstanding loan Institution as of savers (U.S. dollars) (U.S. dollars) interest rate term deposits portfolio PRIDEa 7/94 n.a. n.a.b n.a.b n.a. n.a. n.a. Credit Rural 12/94 31,000 42 1.3 million 15 0 (12/93)' 79 (12/93) Credit Mutuel de Guinee 12/94 45,000 112 5.0 million 13 0 (12/93) 248 (12/93) Credit Mutuel du Senegal 8/94 22,000 70 1.55 million 4.5-6 45 (12/93) 146 (12/93) Village Banks Nganda 12/94 600 45 27,000 0 100 (12/94) 41 (12/94) Reseau des Caisses 9/94 64,000 76 4.9 million 0 0 (9/94) 387 (9/94) Populaires Sahel Action PPPCRa 12/94 n.a. n.a. n.a. n.a. n.a. n.a. Caisses Villageoises du pays 12/94 2,450 76 108,OOOd 0-22e 94 (12/93) 59 (12/93) Dogon Kafo Jiginew 10/94 20,670 43 888,000 4-6.5 35 (9/93) 101 (9/93) n.a. Not applicable. Note: The percentage of term deposits and savings as a percentage of outstanding loan portfolio are as of the date of the last balance sheet for each institution. a. PRIDE and Sahel Action do not have a savings component. b. PRIDE does not have a savings component but has accumulated more than $83,000 in forced savings from the guarantee fund. c. Term deposits take the form of compulsory savings, because they can only be withdrawn at the end of the loan. d. The average savings for 1994, not the total savings as of December 1994. e. Interest is paid only on term deposits. Demand deposits do not yield any interest. Table At4 Financial performance of nine microfinance institutions in West Africa, 1993-94 Operas sg costsa Totalcopnses aperenage f Feespa ster FeeplIsoteroot Eopatrstec- Totalcpenoes -Vusen attate SaMy-rootsas Aceragsal- outstandng lan Incomeaoap- Income5apercent- pesesasaper- as a percentage eopenoes 000 a percentage ary assa mul portfo is (mmos oestage of tperastg age of total costs centage of oat- Data of not operatmg p-roentag- of not of outstandmg tiple of i -NP espatnate so- costs (..ss espa- lssns epatrate otandmg loan Inst it iii ien as of -nome operat-ngincne loanportfohio percapita penos- tr=tee rpensts) eoenoes Portfolio PompE 7/94 384 288 90 9.4 221 (186) 26 (30) 26 (30)' 35 Cdit Roost 12/93 653 429 30 6.5 166(109) 24(36) 22(32) 57 Credbt MutIel do 12/93 8,940 3,560 14 2.1 122 (49) 34 (87 26 (47) 73 Credt tNlut-el da 12/93 736 224 22 1.8 208 (63) 18 (57) 17 (481 145 Sene gal VPilage Banks 12/94 77 41 S 33 (17) 160 (3071 123 (229) 16 Ng mda RWesu de Ca-ssen 9/94 194' - - ( -( -) 53(-) - Populaues' Sl el Arion rPeev 12/93 460 377 48 6 4 200 (164) 13 (16) 12 (14) 36 Ca-ss VsIlagrower 12/93 863 427 24 101- 202 (100) 19 (39) 18 (33) 102 do psys Dogon Ka.o Igeew 9/93 355 254 19 89 06 (61) 29(39) 28(371 24 -Not avadable a. Net ope-atsg -.oee eqoals mterent income mmursoterent erpensen plus -vnseot income plan inome from fern pln other ncome b. Tbs- figure will vary depending on she degren to whilh the program rehfe on vol- nter and co-snn-sod labor c. ro-e's loan portfolo s rther small, while is -dmm -srative -rpennon are 86gh, m pant becaun it nupplementn tn endit se-rs-es wsth three types of thaag di. pr.io does nos have any financial censb. becaose its m yonep is obsamefd va a danor grant. However, PRIDE adost- wsth a theo-btcal 20 percent cent of rapstal e Bad on ebtapolshon from the encome stateeno of s-nt-rn raisins and a rough esnase of the icome statement for the -uher mshthuon. f Data we- vadahlbe for s slynsosen of the thuity-fiv- caisson. No -sord could be obtained mdicasng the income d eup -nses of the them anso or the chc hd nsbhn canter g The averagesalary ass olpleof GNIP pe capta s Igh e theca of Cais Vooagisen dn pays Dogon, whaehe t htutons salary as apreceotage.of loa portfoo ore moderate to low at 24 percent Thin is doe to the fact that Caws6s Vdl.lage--sen employs only foutenn full-hme staff me..en and reles on more than 130 part-sou vo aoluntfs w to w Table A1.5 Indicators of staff productivity in nine microfinance institutions in West Africa, 1993-94 Value of savings Number of per total staff Average loan portfolio Number of out- Number of out- Data savers per total members (U.S. per total staff members standing loans per standing loans per Institution as of staff members dollars) (U.S. dollars) staff member staff agent PRIDE 7/94 n.a. n.a. 5,179 52 113 Credit Rural 12/94 258 10,800 10,833 (12/93)' 110 b 139 c Credit Mutuel de Guin6e 12/93 198 33,113 7,285 (12/93)" 19 b 28 c Cr6dit Mutuel du Senegal 8/94 265 18,627 6,385 (12/93)a 29 b 38 c Village Banks Nganda 12/94 171 7,657 16,914 249 348 R6seau des Caisses 9/94 582 44,545 8,773 34 49 Populaires Sahel Action PPPCR 12/94 n.a. n.a. 4,035a 167 298 Caisses Villageoises du 12/94 175 7,714 d 12,960 (12/93)a 294 _ Dogon Kafo Jiginew 10/94 517 22,000 13,250 (9/93)' 126 b 180 c n.a. Not applicable. a. Calculated by dividing the 1993 average outstanding portfolio by the number of staff members in 1994. b. Calculated by dividing the 1993 number of loans outstanding by the number of staff members in 1994. c. Calculated by dividing the 1993 number of loans outstanding by the number of agents in 1994. d. Based on the average outstanding savings for 1994, not the outstanding savings as of December 1994. e. Caisses Villageoises uses part-time volunteers to manage its caisses. x4 Annex 2 The New Legal Framework for Mutualist Savings and Credit Organizations Doug Salloum In seven countries of the Union Economique et Monetaire de L'Ouest de L'Afrique (UEMOA)-Benin, Burkina Faso, C6te d'Ivoire, Mali, Niger, Senegal, and Togo-a major initiative is under way to provide a legal framework for mutualist institutions and savings and credit cooperatives. The Projet d'Appui a la Reglementation sur les Mutuelles d'Epargne et de Credit (PARMEC)/UEMOA Project, managed by the Central Bank of West African States, with technical assistance from Developpement International Desjardins of Canada and funding from the Canadian International Development Agency, has the following four tasks to accomplish, proceed- ing from the general to the specific: e Creation and adoption by all participating countries of a general law governing the operation and regulation of mutualist institutions and savings and credit cooperatives. v Drafting and adoption by each participating country of decrees to the laws. - Acceptance and adoption of regulations for individual mutualist and similar institutions. • Creation of statutes for these same institutions. The first of these tasks is almost complete. The law has been drafted and approved by the national assemblies of Mali (August 15, 1994), Senegal (December 17, 1994), and Burkina Faso (December 23, 1994). It is pending approval by the remaining national assemblies. Draft decrees have been developed and will be issued by the respective prime ministers after review by the councils of ministers in each country. To date, this process has been completed only in Mali. The regulations and statutes related to the third task are intended to standardize the internal governing and control structures of individual institutions. These are not yet available to the mutualist and other institutions that may want them, because they are being developed by the PARMEC team. In addition to developing these standardized regulations and 347 348 Annex 2 statutes, PARMEC, along with staff from the Central Bank of West African States, is devoting considerable time and effort to explaining the new law to the institutions that will be affected. The process has been fairly open and consultative. All of the institutions appraised here were aware of the project, the new law, and, to a lesser extent, the decrees. What is not clear to many institutions analyzed in this project is whether or not the new law will apply to them, and if so, how it will affect their operations. These are basic questions that the law does not answer explicitly. Much about this whole process is left to the discretion of individual ministers of finance. Features of the law The law is intended to apply to mutualist or cooperative savings and credit institutions as well as to the components that make up their networks, such as unions and federations. The framework of the law anticipates only three kinds of savings and credit institutions and attempts to place each within the new framework. Every savings and credit operation in the region will be considered to be one of the following: * Mutualist institutions and similar organizations. The structure and operation of these will come under the new law. "Similar organizations" is an issue of interpretation, but it would appear that any savings and credit program that has members and more than one or two branches will be considered a mutualist institution, and the new law will apply to it. • Savings and credit associations. These loosely structured institutions, including tontines and informal savings groups, will not be governed by the law but could apply for recognition from the minister of finance of their country. This recognition would legalize their operation and provide certain privileges, including tax exemption. • Other institutions. Other savings and credit institutions that are neither mutualist institutions nor associations in structure or nature will be regulated by the banking laws of each country. The law is intended to provide the fundamental principles of a mutualist institution as well as clauses that would protect the name of any institution recognized or accepted as belonging in this category. The decrees fix the rules of operation and the management standards for the institutions to which the law applies. To allow adaptation or adjustment of the regulatory structure, the decrees have been designed with a more supple revision and approval process (including councils of ministers and the prime minister rather than the full national assemblies of each country). In addition to defining who is subject to the law, clauses set forth the following objectives: Annex 2 349 • Organization andfunctioning of the institution. Whereas the law recognizes that there is considerable variety in the structures of the mutualist institutions, it expects certain features to be in place, including separation of management and control, credit policies (to be defined by the general assembly or by the management), limits on loans to board members and their families, conditions for dissolving or merging institutions, and provisions for exempting the institution and its members from taxes (on savings and loan transactions). * Structure and operation of the network. The law fixes the rights and obligations of each level of the network as follows. Each network is expected to supervise its own operation. External control will be imposed through the requirement to submit annual consolidated reports to the minister, the Central Bank of West African States, and the Banking Commission as well as to provide other information that these agencies may require to carry out their role. The benefits of being included under the law are somewhat subtle: protecting the name of the institution, exempting the institution from taxes, and legally recognizing its right to operate. Potentially less positive, institutions that are judged to be mutualist will be subject to all other laws that apply to savings and credit operations. This certainly would include the laws against usury and may include labor laws as well. Usury is defined by the rates of interest charged to clients who borrow. The legal definition in most of the countries involved is that an effective rate of interest in excess of two times the discount rate is a usurious rate and that the institution charging this rate is carrying out an illegal activity. If subject to this upper limit on loan interest, many of the institutions reviewed here clearly would be unable to earn enough interest income to cover their operating and financial costs and therefore would never become financially viable. In these cases, savings and loan programs and institutions may decide to apply for exemption from the law but still ask to be recognized, that is, to be tax exempt, to have their name protected, and to be allowed to carry on operations. In doing so, they risk a refusal, in which case the law would likely be applied sooner than if they had kept a lower profile. Features of the decrees The draft decrees reviewed by the Bank team included a number of specific clauses meant to ensure prudent operation, in the governments' view, of a mutualist savings and credit institution. For example, draft decrees include the required content of the statutes of the mutualist, the powers of the general assembly, and the frequency of meetings (at least annually), the roles and functions of councils of administration and credit committees, the conditions of membership, the process by which institutions are recognized 350 Annex 2 or agreed, and various ratios to control risk. For example, credit risk of similar organizations cannot exceed 5 percent of the loan portfolio, 15 percent of annual operating surpluses must be allocated to a general reserve, the loan portfolio must not exceed two times the deposits of the members, loans to directors cannot exceed 20 percent of the value of members' deposits, loans to one member cannot exceed 10 percent of the value of members' deposits, and short-term assets must at all times equal at least 80 percent of short-term liabilities. All of the features of the draft decrees follow from the assumption that the institutions being affected are mutualist institutions or savings and credit cooperatives. The individuals who have prepared the law and the decrees have a great deal of experience with these kinds of institutions. They have less experience with other forms of savings and credit institutions and perhaps httle interest in preserving these other structures. Their assumption seems to be that the mutualist structure is the preferred structure and that all savings and credit programs of significant size should be of this kind. Annex 3 Financial Notes Pierre-Olivier Colleye The quality of financial data varied greatly among the appraised institu- tions. Two institutions (Sahel Action and Credit Rural) had income statements and balance sheets audited by European firms, whereas several others were preparing their first financial statements after five or six years of operations. All financial tables appearing in this book should therefore be taken as estimates and not as official financial statements. Regardless of their quality, all financial statements had to be adjusted for purposes of unifor- mity. This annex details what was usually included under each balance sheet and income statement item and describes the most typical adjustments that were made on these items throughout the nine appraisals. Balance sheet Cash. Unless otherwise specified, includes all the money in the caisses' safe as well as savings deposited in commercial banks. Bank account or investment. Savings deposited in a commercial bank. Loan portfolio. Principal receivable from the borrowers. All institutions function on a cash basis. Therefore, there are no interests receivable. Fixed assets. Includes cars, furniture, and office equipment. Does not include buildings. Adjustments were sometimes made to represent better the book value of these assets. For example, PRIDE'S original statements do not contain any fixed assets, because all purchases are immediately recorded in the income statement. When not enough information was available, a five-year straight-line depreciation was applied. Equity. Broken down into donors' equity (accumulated amount of subsidies coming from the donor agencies for fixed assets and operations) and members' equity (paid-in-capital, reserves, and retained earnings). Most institutions do not distinguish between the two types of equity, and their management considered that donors' equity for fixed assets should be depreciated just as the fixed assets themselves are and that operating subsidies should be considered as income. However, for our analysis, the distinction was crucial. In addition to determining how efficient the institutions are, it shows exactly how much they have cost so far. 351 352 Annex 3 Income statement Credit income. Income generated from loans made to members or clients, calculated on a cash basis. Contrary to most original financial statements, subsidies are not included in income. Investment income. Income generated from savings deposited with a commercial or development bank. Fee income. Membership fees. In the case of PRIDE, it also includes transaction fees. Salary expenses. Salaries of local staff (not including expatriates' salaries), plus bonuses, social security, and medical expenses. Administrative expenses. Typically include office rents, maintenance, cleaning, and utilities; supplies, printing, and photocopying, communications (phone and fax bills, stamps), and local travel (cars, hotels, per diem). Depreciation expenses. Calculated on cars, furniture, and office equipment, using a five-year straight-line depreciation unless otherwise noted. Institution-building expenses. Expatriate-related expenses (salaries, housing, living expenses, and international travel). Could also include some seminar expenses related to the launching of the institutions. Inflation Most institutions in the CFA region expected 1994-95 to be a difficult year. Some experienced a surge in savings, but many were having difficulty generating income by depositing these savings with commercial banks. However, most financial statements in this document are for institutions in the CFA zone before the devaluation, when inflation was minimal. Adjust- ments for inflation were therefore considered unnecessary. Repayment rates Repayment rates. Loan amounts that are thirty days or more past due. IBRD 28326 WEST AFRICA FORMER \ ALGERIAL I B Y A 5PANISH 1A e NATIONAL CAPItALS SAHARA; -- INTERNATIONAL SOUNDARIES ! I '< 20 '7-----' N> N~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 r M A U R I T A N I A !n ',w *Nouak,hor t! CAPE VERfSE f- M N I G E R raro,0 *HE 'E:N EG MALI fvC H A D THE B L.k.~~~'~BUKIA~ GAMBIA- 2 - *Umoko BNiomey ,_.- Chod Boo~~~~~~~~-~~~ P~FA SO 'N- / SU DAN GUIINI" ' 8eoaagadaoagoajf' \i > tN'Djadeno 1 BISSA ' G U I N E A >U-):\ RIA Con<>kr>^, ,~~~~~- , -4 , ) go ~~~~N f G E R I A J';f/ , A T L A N T C SIERRlA *Abulo -- AT LE A N C F,et- LECNEr3) COTE GHANA OZ | LEONE'~~'D'IVOIRE ( rTZ O C EAN 4LBERr' A )O J CENTRAL i Yomoosao&ko ( ei... AFRICAN *~~~~~.* ~~~~~REPUBLIC r) .. Acora / ~CAMEROO N B 0 250 500 750 KILOMETERS ouniA .d6 * ThMs mop not promdoed by the Mop Dolgr Uad of The World BOak EQIJATORIAIL _ - l ,' The hnadnrnesorItf, dnn,a,nioaomeons n omonm,otlXrar inhomoflon hot'n DAI TOBAI a this mop do no- imply, an th. p-rt of Th. Worid B-ni Groop, ony SAO TOAiE AD Z A I R E jrodmento noel, NI * Irbrerille OO ° creptore ar the Inmot e ae S Tome GAB ON CONGO 20- 10 rz , iOo { 1 20' SEPTEMBER 1996 . IBRD 28328 MAURITANIA BURKINA FASO M A L I MA A L I - GE * NATIONAL CAPITAL - > rBURKINA -- INTERNATIONAL BOUNDARIES -5 ( FASO , 50 19O 150 200 | Z i KILOMETERS 5-. D 01Ri0 A ) I V f V ~~~~~~~~~~~~~~~~~~~~~~~~ATjLANTIC QCEAN c~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ fJ a._si j ~~~~~~~~~~N I G E R M A L I , BURKINA FASO *OUAGADOUGOU -l2- -? \, 121 t ,~~~~~~~~~~~~~~~~/- --\,TOGO B E N I N i, ¢ ~~~~~G H A N A ' -- GHANA Ths mop SoT s produced by the Mop Design nht of The WorId ank. _*,> / \ s The boundaries, colors denommna,ons -nd onyother informaon shown -V., ~~~~~~~~~~~~~~~~~~0thi -np do -ot imply, on the port of The World ook Group, -oy judme, on the eol ts f any trriory, or -ny endorse-en; or cOTE D'IVOIRE acpneof -uh b-udaries A' 2- 0 2' SEPTEMBER 1996 IBRD 28325 . . ' ., 25* 24- vSi SANTO ANTAOC.. sAo VICENTE" .SANTADUZIA SAL DRANCO': BASO - . CAPE VERDE * NATIONAL CAPITAL -INTERNATIONAL BOUNDARIES (INSET) ______ A N TIC ~~~~~~~~~~~~~~~~~BOA o 10 20 30 40 soKILOMETERS A T L A N T I C VISTA T6 his ,op ,.'o, p,od'z4 bp - , Mc,p Dsipm Unit of TBo Woid BoA. the wncto6 , coers, delioatioW -4 "oow onno. Fh,-Q.0 C E A N ihh mp d iowb,, Tm $* porl d Th* wid D&k tGnwp, oC Riim e O0 IBOISIoBo dOay w4fm, of ony .B&B,qner,f or, qcTBptqece o such bndoodA2.jS 2S. CAPE VEtDf MAURITANIA ' IMALI SLNEGA.L r~ \MAIO~ GUINFA gss 7- )--I ,;^URKINA" OUINEA 'A C F-ASO SANTIAGO SIERRA LEQNEJ) COTE > 15i IS' ISECOS .S- ATLANflC IA-,OGO i PRAIA OCEAN I -BRAVA 2 SEPTEMBER 1996 356 IBRD 28327 This mop noa produced by the Mop Des,gi Unit of The World BanA. 20- 24 The boundories. colors, denominatons and onyother inOrmotion shown on /his mop do not imply on the pon of The World -on Group, any judgment on the egolStwts ofony teritory. or ny endorement or H A D acceptance of such haundaries. . ~.. *' NATIONAL CAPITAL N. N / - - INTERNATIONAL BOUNDARIES ;. Y- )Z °0 100 200 300 .. 'KILOMETERS - 201 N I G E R , / C H A D A Lake`_, - U D A N Chad SUDAN N'DJAMENA 12- 12, NIGERIAz, ( K -/J '\ _.~~/' ....C2 L 3 RAlE AIGEIA)- OF EGYPT. 8- \7 , _- a '*cNIGER CAMEROON/ / CHAD DAN I CENTRAL AFRICAN REPUBLIC NIGERIAj( j ( ~~~~~~~~~~~~~.,j ) CENTRAL fl 1,6 20- X^gEiCutO APRICAN REPULI SEPTEMBER 1996 IBRD 28330 1 6 1 7 R r THE GAMBIA CE E MAURITANIA I-." ~ ~ AL * NATIONAL CAPITAL ' SENEGAL- INTERNATIONAL BOUNDARIES Y IRAUR FAS IJ - 0 10 20 30 40 50 KILOMETERS 1- 3 L MALI 815/u GUNA-8 * NATIONAL CAPITAL 'O .GUINEA FASO INTERNATIONAL BOUNDARIES 5 SIERAifNEAn COTE 100 200 LB RAIT J D'IVOIRE D'IVOIRE / _______ 200 LIBERIA A r tIA f Ntf T0tii C LIBERlIA> KILOMETERS , - OCEAN -J -1-1-b 8 : ::: IC:::::: :::::: Gvff .^f Gstne 20 o . SEPTEMBER 1996 IBRD 28323 :''',. ~~~~~~16 5-14t ,-' S E N E G A L ;~~~~~~~~~~~~~ -. - ;- - ,- - - - - a- - -_ GUINEA-BISSAU GUINEA- BISSAU -- .BISSAU < * NATIONAL CAPITAL -- INTERNATIONAL BOUNDARIES ;3j ;. - N; -i-.> x; z GUINEA ATLANTIC . .--A N N.: - . '; .s i,;, , .. , .MAURITANIA '. .' , ,, , -. .A'. / T. SENEGAL M MALI o 20 40 60 80 / GUN - . I , I I GU~~~~~~~~~~~~~~~~~~~~~~~~~INEASU KILOMETER$ GUNE ~y( URVJNA 9 ,z A . _ ~~~~~~~~~~~~lRRA c , _ This mop OOs produced by the Mop Design Unit of The World 8.-. -EONE A NC The boundoires, colors, denominotiot ond onyother i,lor,otio shown *L D'IVOIRE on this op do iosply, on the pot of The Word on Group, ony - jodgotent on the legoistoton of ony tnrr.tcny,or ony eadorseinoot to ATLANTIC UBoRIA.- occeptonce of such bo.ndories. OCEAN 16S 116 SEPTEMBER 1996 - IBRD 28322 'SPANS , ALGERIA _ _- 530 5- 5RA . |"\ This mop -os prodoced by the Mop Design Unit of The World 3-nk' . N. ; The boundories, colors. deoominotions ond ooyother inormotion sho-n t I ! \ j > on this mop do not imply, on the port of The World Roof Group, ny MAURITANIA 'r judgment on the logosttus of any territory, or ony endorsement or ~~MALI j oCCe~~~~~~~.pto-r of suh b-udories. SEI~~sEGAL'/~ -T'NTSGER N GUrg'- gJ mm ! \' A L G E R I A MlsA AUNE .I_+,L>t I -~MALI j< - NATIONAL CAPITAL -- INTERNATIONAL BOUNDARIES 0 150 300 KILOMETERS MAU R I TANN I A 15, ,~Q NGR r , ~~~~~N I G E R SENEGAL BAMAKO BURKINA FASO \:\ N G U ~~~~~~~~~~~~~~ ~ ~~~~~~~~~~~I N I G 6 < R,S E R I A 1T0 > X ' r 5. . G H A N A ',,p' 5- SEPTEMBER 1996 ,,r co. t^ 00 r OeO~~~~~~ ~~~~~~~~~~~~~~~ I x * nf°l I z n \ o > M: r ,r>!_+ 0~ ~ ~ ~~~~~~~~~~~~~~~~~~~_ , Al-6 It./t f r~~~~~~~~~~~~~~~~~~~~r m ~ ~ ~ ~ .~ o ..D j s tr t 8 t t P /-* -t1 IBRD 28329 NO~ I G E R 0 >-- /-L I B Y A N I G E R~~~~ ~ LBY f NATIONAL CAPITAL \ /- -- INTERNATIONAL BOUNDARIES -<0/ ! 0 1 0O 200 300 KILOMETERS / - 20' / 20> This nap sos pr-d-ced by tha -r Map D. ign Uint of The W=/dd Rnk. The .oondaHee colors danonntions and any otler informanon shown on this -tp do not imply, on the part of C H A D The W-rld Rank Graup, ny jydgten0 on the legalstatus of any terr-tay, any endorsement ar acceptance of IM I < e-ch ba-nd-ries. M A LI j 15- - 5'2/ 15 ALGERIA t., NIAMEY -o-- A - r/ ~ ~ s.MALI NIE ;\ j t~~~~~~~~~~~~~~~~.s Lar~ cke _ NI G E R Lk. 1. ~/ CHAD/ K / ~~~~~ ~~~~~~~~~~~~~~~Chad 1n BUAKIN NIGERIA BFARKSO A y s N I G E R I A 3 ,~~~~~~~~~~~~~~~~~~~~~~~~~O 15~ OAKAft CbrSENEG AL> DAKA~* C E N E G7 A L MALI Atlantic E~~~~~ Ocean NT H E~:GAMzB1YA,L V-N~~~~~~~~~3 -t~~~~~~~~ .H . . A.M-I.- . . ........ GUINEA - BISSAU r. S GUINEA 12' 0 t*00000 0 7 C ;i ~~~~~~~~~~~~~G U I N E A-BISSAUC lvp9o^~~PvehsXrsWE - Group: ;: .: S G U I Nj E3A 12 er ooy> .do,o..o.ot or ooo.p*ooo. o#§f .od> ~ooodo4... 2S 1 996 SEPTEMBER 1 996 365 IBRD 28320 13* 12- ,,- 10- lo10 G U I N E A -- U, I SIERRA LEONE , FRU1QWN>', ATLANTIC I BERI A Y OCEAN 0r 50 Ig KILOMETERS SIERRA LEONE MI * NATIO .CAPtTAL G U I N E A -- INTERNATIONAL BOUNDARIES SIERRA '"np nop,ndnnadny '.Mby,, rD.ngn i'aTaWadhn,6. 'hEONEp /"' C. l Th 1,onda,-. ,Ins ors, dowm.nne*nnond oenro,aeri n pma) own ,( < o n o nap - - A.n n phpoTAaWnTdB.o kG nnp-n A r LANTIC 7 LRBERIA" o. .iagnenl~ of n * ysllu orrony ............. ncw> vrany annt or.n.............. a .y naplancaof.nf "b ano'd.an O OCEAN 13' I9 SEPTEMBER 1996 I THE WORLD BANK A partner in strengthening economies and expanding markets to improve the quality of life for people everywhere, especially the poorest Other World Bank Regional and Sectoral Studies Financing Government in the Transition: Bulgaria. The Political Economy of Tax Policies, Tax Bases, and Tax Evasion, edited by Zeljko Bogetic and Arye L. Hillman Decentralization of the Socialist State: Intergovernmental Finance in Transition Economies, edited by Richard M. Bird, Robert D. Ebel and Christine I. Wallich Population in Asia, Warren C. Sanderson and Jee-Peng Tan Education in Asia, Jee-Peng Tan and Alain Mingal Indigenous People and Poverty in Latin America: An Empirical Analysis, George Psacharopoulos and Harry Anthony Patrinos The Determinants of Reproductive Change in Bangladesh: Success in a Challenging Environment, John Cleland, James F. Phillips, Sajeda Amin, and G. M. Kamal The Effects of Protectionism on a Small Country: The Case of Uruguay, edited by Michael Connolly and Jaime de Melo Implementing Reforms in the Telecommunications Sector: Lessons from Experience, edited by Bjorn Wellenius and Peter A. Stern Adjustment in Africa: Lessonsfrom Country Case Studies, edited by Ishrat Husain and Rashid Faruqee Russia and the Challenge of Fiscal Federalism, edited by Christine I. Wallich Administering Targeted Social Programs in Latin America: From Platitudes to Practice, Margaret E. Grosh Including the Poor: Proceedings of a Symposium Organized by the World Bank and the International Food Policy Research Institute, edited by Michael Lipton and Jacques van der Gaag 9 702 1 3 3 970 ,G,,EllA3159171 i*