Sustainable Banking with the Poor - Case Studies in Microfinance : Peru - The Cajas Municipales de Ahorro y Credito

SECTORAL & IT RESOURCE CENTER FEB 2 9 2000 - WORLD BANK SUSTAINABLE BANKING with the POOR Case Studies in Micro finance PERU- ---------- THE CAJAS MUNICIPALES DE AHORRO Y CREDITO Jill Burnett Carlos Cuevas Julia Paxton May 1999 The World Bank  '. , ,, Sustainable Banking with the Poor (SBP) is a collaborative effort of the Gender and Poverty Team and Rural Finance at the World Bank, funded by the World Bank, the Royal Ministry of Foreign Affairs of Norway, the Swiss Agency for Development and Cooperation* (SOC), and the Ford Foundation. The study aims at improving the ability of donors, governments and practitioners to design and implement policies and programs to build sustainable financial institutions that effectively reach the poor. The SBP task managers are Lynn Bennett and Jacob Varon; the technical manager is Carlos Cuevas, and the associate manager is Korotoumou Ouattara. The administrative assistant is Sangae Sherman. The World Bank SBP, J4-241, 1818 H Street, N.W. Washington, D.C. 20433 Phone (202) 458-0277 Fax (202).477-2978 Internet: CCuevas@WORLDBANK.ORG or KOuattara@WORLDBANK.ORG WEB PAGE: http://www-esd.worldbank.org/html/esd/agr/sbp *Note: In the back cover box, Swiss Development Corporation, should read Swiss Agency for Development and Cooperation (SDC). • by Jill Burnett, Consultant, Carlos Cuevas, SBP Manager, and Julia Paxton, This report was prepared _ consultant. This case study benefited tremendously from the input of Felipe Portocarrero Maisch of LusoFinancial Systems. Valuable insights were also contributed by Gabriel Gallo Olmos of the Federaci6n de Cajas Municipales del Peru.  TABLE OF CONTENTS ACRONYMS AND ABBREVIATIONS ................................. .- ........................ .ii INTRODUCTION ................................... : ................................................ .1 COUNTRY OVERVIEW ............................................................................ 1 ORGANIZATIONAL BACKGROUND AND ORIENTATION .............................. 7 INSTITUTIONAL STRUCTURE AND MANAGEMENT ................................... 14 FINANCIAL PRODUCTS ......................................................................... .18 PERFORMANCE .................................................................................... 30 DISCUSSION: ACHIEVEMENTS, CHALLENGES AND THE FUTURE ................ 37 CONCLUSIONS ................................. : .......................................... ~ ......... 41 BIBLIOGRAPHY .......................................... : ......................................... 43 APPENDIX 1 - LIST OF CASE STUDIES ...................................................... .44 APPENDIX 2-LIST OF DISCUSSION/TECHNICAL PAPERS ............................ .45 SUSTAINABLE BANKING WITH THE POOR  ACRONYMS AND ABBREVIATIONS BancoAgro Banco Agrario BW Banco Wiese CMAC Cajas Municipales de Ahorro y Credito (Municipal Savings and Credit Banks) COFIDE Corporacion Financiera De Desarrollo S.A. (Financial Corporation for Development) CTS Compensation for Time in Service EDPYME Entidades de Desarrollo de las Pequefzas y Micro Empresas (Small and Microenterprise Development Agency) FEPCMAC Federaci6n Peruana de Cajas Municipales de Ahorro y Credito FOCMAC Fondo Peruano CMAC (The Peruvian Fund CMAC) GDP Gross Domestic Product GTZ German Technical Cooperation agency IDB Inter-American Development Bank IPC Interdisciplinaire Projekt Consult (Interdisciplinary Project Consulting) MITINCI Ministerio de Industria, Turismo, Jntegracion y Negociaciones Comerciales lnternacionales (Ministry oflndustry, Tourism, Integration, and International Business Negotiations) NGO Non-governmental Organization ROSCAs Rotating Savings and Credit Associations SDI Subsidy Dependence Index WWB Women's World Banking EXCHANGE RATE 1995 US$ I = 2.26 Nuevos Soles 1996 US$1 = 2.45 Nuevos Soles 1997 US$ I = 2.66 Nuevos Soles 1998 US$!= 2.93 Nuevos Soles SUSTAINABLE BANKING WITH THE POOR II  IBRD 30293 COLOMBIA A~DOR / 0 ~0 '~OtJ).S~),;;k-.., ... ~ o fSELECTE01TOWNS ® '~ATIQNAL:CAPiTAL -- i1NiERNATIONAU10 ~a{ - •· 0 7ttJ?)J:}:,· 0•Fj:}~ i ,) BRAZIL ;l llpa BOLIVIA JUNE 1999 Country Profile Economic and Social Context Inflation GDP per capita (1997) us $2,587 1995 10.2% 1996 11.8% Population (1997) 24. 7 million 1997 6.5% Population density 35 inhab./km 2 1998 6.0% SUSTAINABLE BANKING WITH THE POOR iii  CASE STUDIES IN MICROFINANCE PERU THE CAJAS MUNICIPALES DE AHORRO Y CREDITO INTRODUCTION The network of Cajas Municipales de Ahorro y Credito (CMAC) of Peru represents one of the most significant and interesting cases of micro finance in Latin America. With 13 branches located throughout Peru, the CMACs offer savings and credit services to areas that are underserved by commercial lending. Since their creation in 1980, the CMACs have attained an impressive scale. In 1998, the system granted some 400,000 loans, averaging US$440. In addition, the CMACs offer voluntary savings instruments that have grown in popularity. While performance indicators have been favorable, the CMACs face several obstacles as they enter into the millenium. One of their primary concerns is determining an appropriate legal and ownership structure. In addition, they face the challenge of managing growth while curbing increasing arrears, lessening dependence on subsidized funds, and promoting savings. The CMAC experience and the way in which they are able to confront these obstacles provide invaluable lessons for other microfinance institutions worldwide. COUNTRY OVERVIEW Soc10-EcoNOMIC ENVIRONMENT Peru is a country of stark contrasts. The beautiful Andes mountains are juxtaposed against lush tropical rain forests and one of the driest deserts in the world. Also, many contrasts exist within the l>eruvian population with its mix of Spanish, latino, and indigenous cultures, and the disparity between the wealthy and very poor. Endemic poverty plagues Peru as is exemplified by the thriving "pueblos jovenes" (urban slums) and extensive informal microeconomic sector. From 1981 to 1995, nearly 50 percent of the population lived on less than US$1 per day on average 1. The 1995 life expectancy was 66 years and adult illiteracy ran 11 percent (approximately 18 percent for women and 7 percent for men). After years of economic instability resulting from conditions of hyper-inflation and terrorism, Peru has been in a period of significant economic growth since 1993. Structural reforms to liberalize the economy introduced by President Fujimori by 1990 have been successful in promoting expansion of the economy and reduction of inflation. Major privatizations of mining, electricity and telecommunications have been completed. In 1993, political stability was restored with the capture of the Shining Path Leader, further facilitating the in- flow of foreign investments. During this period, inflation was reduced from 139 percent in 1 Using Purchasing Power Parity (PPP) from World Development Report 1997, The World Bank, Washington D.C. 1997. SUSTAINABLE BANKING WITH THE POOR 1991 to 6 percent in 1998 and exchange rates stabilized (Table 1). The financial sector also achieved stabilization and expanded its credit activities substantially responding to increased demand for investments (up by 35 percent in 1994 from the previous year). Peru's growth domestic product (GDP) grew from -5.5 percent in 1990 to 6.4 percent in 1993 and to almost 13 percent in 1994, making it the fastest growing economy in Latin America during that period. Growth occurred in all sectors, especially in construction resulting from increased spending in both the public and private sectors. More recently, these trends have been maintained, though at substantially more modest levels. Concerns of hyper- inflation have led to the introduction of measures by the government to curb expansion and ensure that growth is stabilized at manageable levels. By 1998; GDP growth rates were a modest 0.6 percent. Table 1. Overview of Country Indicators Indicators 1992 1993 1994 1995 1996 1997 1998 GDP Per Capita (US$) 1,866 1,810 2,164 2,353 2,469 2,587 - GDP Growth Rate -1.4% 6.4% 13.1% 7.3% 2.5% 7.2% 0.6% Inflation Rate 56.7% 39.5% 15.4% 10.2% 11.8% 6.5% 6.0% Exchange Rates 1.25 1.90 2.19 2.26 2.45 2.66 2.93 (Nuevos Soles per US$) Despite the transformation of Peru's macro-economy that has occurred in the last 5 years, the effects of this growth were not shared among the lower income sectors of the economy. Unemployment has remained high (estimated in 1994 at 11 percent of the economically active), resulting from the down-sizing that has occurred in both the public and private sectors. Recent unemployment figures are more promising with unemployment falling to 8.2 percent in 1996. The expansion of credit to the private sector was limited to large and medium scale businesses, with little change for small and micro-businesses which form the principal source of income for the majority of Peruvians. SUPPORT FOR THE MICROENTERPRISE SECTOR In recognition of the isolation of these sectors, the government introduced policies to promote the integration and development of informal micro-businesses as an important component of the private sector. It identified the expansion of credit to small and microenterprises and their normalization into the official economy as central components of its policies. In 1994, the Micro and Small Enterprise Program was established within the Ministerio de Industria, Turismo, Integraci6n y Negociaciones Comerciales Intemacionales (MITINCI) as a mechanism to implement-these objectives. Financial resources to support these initiatives have come from the central government, with the establishment of rotating credit lines from FONCODES in 1992. In addition, US$50 million funds were designated for microfinance institutions in 1994 by Corporaci6n Financiera De Desarrollo S.A. (COFIDE), Peru's second tier development finance institution. Funds from SUSTAINABLE BANKING WITH THE POOR 2 COFIDE are a mix of government funds and loans from multilateral institutions.2 Since 1993, Peru has also experienced significant in-flow of funds from external. sources dedicated to the financing of microenterprises. Most important has been the Inter-American Development Bank (IDB) through its Micro-Global loan of US$25 million in 1995/96. However, although substantial financial resources have been allocated for the development of microenterprise finance, structural barriers are inhibiting the delivery of these funds to prospective clients. Outlets for these funds incluqe the banking sector, through IDB sponsored Micro-Global projects3; municipal savings and credit banks including the Cajas Municipales; and Entidades de Desarollo de las Pequefias y Micro Empresas or EDPYMEs, non- bank financial institutions4. With banks only recently launching their Micro-Global activities and few NGOs prepared to meet the necessary pre-conditions of EDPYME status, there are relatively few retail agents for this credit. NGOs in particular stressed their frustration at the lack of available resources for the institutional development activities they required to transform to EDPYME status and subsequently to gain access to these funds. RURAL SECTOR With regard to small and micro-agricultural producers in the rural sector, the effect of increased liberalization of Peru's economy in the 1990's has been less than positive. With most of Peru's agricultural production deriv~d from independent small-scale producers (majority smaller than 5 hectares following from land reforms undertaken into the 1980s), the downward pressure on prices resulting from increased imports and price fluctuations based on international markets has undermined the viability of their activities. Without the support of government subsidies and price controls, many smallholders have become more impoverished during recent years. Their economic isolation has . been deepened by the recent failure of the Banco Agrario and Cooperativas (partially due to the inability of farmers to meet loan contracts, combined with poor management and operational efficiency). Non-performance on loans has also induced the government to close its farm input credit program: Fondo Agro. The closure of these institutions has left a significant gap in rural financial markets which is currently being pursued by Cajas Municipales. LOCAL ECONOMIC CONTEXT OF THE CMACS The CMAC network covers numerous provinces in the country, each with distinct levels of economic development, industries, .and client profiles. Most of the CMACs are located in the coastal region of Peru, including Piura, Sullana, Paita, and Trujill<;> in the north, and Santa, Pisco, 2 COFIDE's total inflow of foreign funds for 1994 was l.JS$ 265 milli~n: US$ 115 million from the Andean Development Corporation; US$ 100 million from the IDB; US$ 47 million from foreign trade credit lines and US$ 3 million from USAID. 3 The IDB Micro-Global projects are coordinated through local Central Banks, which channel guaranteed finance and technical assistance to commercial banks and other financial institutions for micro-lending activities. Micro-Globals have been introduced by the IDB throughout Latin America with varying levels of success. 4 The legal status ofEDPYME (translates as ''.Small and Microenterprise Development Agency") was created in 1994, and sets institutional, performance and financial prerequisites for non-bank microfinance lenders. Requires that all financial NG.Os gain this institutional status before being eligible to gain access to loans from COFIDE. SUSTAINABLE BANKING WITH THE POOR 3 and lea in the central region. The CMACs of Huancayo; Cusco and Arequipa are located in highland regions and the Maynas CMAC is in the northern jungle region. Table 2 shows the various provinces where the CMACs operate and their relative importance to the CMAC net~ork in terms of outstanding loan portfolio. Table 2. Ranking of CMACS by Size - Outstanding Loan Portfolio, April_ 1998 CAJAS MUNICIP ALES TOTAL % No AREQUIPA 24.49 1 PIURA 21.86 2 TRUJILLO 9:97 · 3 cusco 9.92 4 SULLANA 8.15 5 ICA 5.29 7 TACNA 3.14 8 HUANCAYO 3.09 9 MAYNAS 2.55 10 -. PAITA 2.16 11 SANTA 1.77 12 PISCO 0.98 13 CHINCHA 0.03 14 TOTAL 100.00 Source: Supermtendency of Banks and Insurance of Peru, 1999 The CMAC with the greatest outstanding loan portfolio is in Arequipa, Peru's second largest city. Located in the southern Andes, Arequipa is an important market center for the region. Economic activities include a wide range of commerce and services, as well as varied micro-industries and manufacturing. As one of the most dynamic CMACs, ,Arequipa is currently the site of much product development and testing for the CMAC system: It has experienced substantial growth, particularly in its microenterprise lending operations. Piura is the capital of the northern desert region of Peru and is an important business and administrative center for the area. The Piura CMAC is responsible for 22 percent of the total outstanding loan portfolio of the CMAC network. Piura CMAC is the oldest and most institutionally developed of the CMACs. In recent years, the Piura CMAC has.moved away from the target "micro" market and appears to operate increasingly like a bank (very conservative eligibility and screening mechanisms; inflexible terms; preference for classic forms of collateral ovei: substitutes). As a result, Piura has specialized in the areas of savings and Pawn loans and Personal Credit. Besides the two largest CMACs, three others stand out for the scale that they have obtained. Trujillo and Cusco each have loan portfolios that comprise approximately 10 percent of the overall CMAC loan balance. The Sullana CMAC is responsible for 8 percent of the 1998 loan portfolio. Trujillo is a fishing and agricultural center along the northern coast of Peru. In contrast, Cusco is one of Peru's principal tourist attractions, as the ancient capital of the Inca SUSTAINABLE BANKING WITH THE POOR 4  civilization. Market research performed by the Cusco CMAC shows that a high percentage of savings remain outside the financial system and that the demand for local credit outstretches is significant. As a northern desert secondary city 40 km. from Piura, Sullana serves as a local market center for surrounding agricultural areas. Sullana's principal economic activities are commerce and agriculture: fruits, vegetables, and rice are among the most important crops. Sullana has two seasons around which its lending activities are organized. Having successfully saturated its urban and peri-urban markets, Sullana has provided the principal testing ground for the CMAC' agricultural credit product development. Natural disasters and environmental conditions make lending more volatile in this region. COMPETITION FROM OTHER SOURCES Banking·institutions operate in cities where the CMACs are located and, without exception, provide stiff competition in the area of savings. The CMACs are attempting to compete with the banks by offering slightly higher rates of interest (usually 0.1 to 0.2 percentage points); and by offering a rapge of products to suit different components of client markets. Client . treattrtent and rapid delivery of services are other areas where the CMACs have attempted to gain the edge over the banks. Conversely, the banks are able to attract savers on the basis of their strong public image as stable and secure institutions. Some business owners (especially larger), also cited preference for a bank account which provided checking services and the ability to conduct inter-branch business when purc~asing business inputs from Lima. Other aspects of deposit services offered by banks and CMA.Cs are similar. In the area of credit, most of the CMAC management indicated that they regarded the banks as their leading competitors. Some banks express interest in expanding outreach to micro- business sectors. One example of this is Banco Wiese (BW) which is working with non- governmental organizations (NGOs) to extend lending to micro-businesses throughout the . country. Other institutions are approaching micro-lending through participation in the IDB's "Micro-Global" project which provides technical support and guaranteed loan capital for microentetprise lending. As one CMAC manager commented, the fact that the banks only are willing to expand into these market areas with direct financial incentives provided by the IDB project demonstrates that most banks continue to perceive microenterprise lending as a separate activity from their other "standard" operations. Nonetheless, as with other Latin American countries undertaking micro-global projects, if successful, this initiative should have significant impact in promoting competition in micro credit markets. The CMACs believe their main advantages over bank credit lie in ~ ,C? following: • Quick turnaround time of loan applications: CMACs deliver credit within 2-4 days for a first loan and more quickly for renewals; bank loan applicants can wait up to three • months for an answer. • • Guarantees: CMACs are flexible, look to informal financial practices on collateral substitutes; bank require formal, official guarantees that can be time-consuming and expensive to obtain. SUSTAINABLE BANKING WITH THE POOR 5 ·L • Client service: the CMACs tend to be more service oriented and work closely with clients. • Transparency: the borrower costs of CMAC loans are easily calculated; banks charge service fees and commissions on loans which make their effective rates much higher than those advertised. In the agricultural sector, the Banco Agrario (Banco Agro) provided an important source of finance to rural producers until its failure in the early 1990s. Reasons for its collapse were attributed to classical explanations of development finance institutions (inadequate enforcement, public perception of loans as soft government funds; etc.), along with changing market conditions resulting from opening of markets to international trade. Combined with the collapse of the cooperatives in early 1990s and the elimination of a government sponsored agricultural credit program, Fondo Agro (which provided farmers with inputs for their production activities), the agriculture sector has suffered shortages in credit access during the last years. The CMACs, recognizing this market opportunity, have begun to expand their services in this area. lnformaljinancial services are widely available in the form of moneylenders (prestamistas), supplier credit, rotating savings and credit associations (ROSCAs), and friends and family. Most clients interviewed indicated that they had used informal lenders in the past. Money-lender credit is used especially for emergencies; is very expensive and is paid back over the short term. Supplier credit was either partially or fully replaced by CMAC loans. (When clients can gain access to the full amount they require, they use CMAC credit to pay for supplies. Until this is achieved, clients tend to use loans for a variety of uses, including offsetting some debt to suppliers.) One credit officer interviewed suggested that the informal lenders provided stiff competition in the rural areas because client confidence had been shaken by the collapse of rural financial institutions. He noted that they preferred the personal and reliable services of the . moneylenders who visited them and suggested that CMAC services had to compete with informal lenders on these fronts. NGOs based in Lima and other centers of the country provide credit to lower income groups. ACP (an affiliate of ACCION), CARE, CEPES, !DESI and Credinpet were identified as the most significant. CRS, Socodevi and other international NGOs are also present. MIBANCO, a recently established commercial bank owned by shareholders including ACP, ACCION, PROFUND, Banco de Credito and Banco Wiese, is one of the most important .microfinance institutions in Peru. Typically, NGO loans are more expensive than those of the CMACs. Interest rates vary at around 5-6 percent per month, often with additional service charges of up tp 2 percent. Almost all NGOs have a smaller outreach of clientele. Some are specialized in reaching poorer clients and provide subsidized rates of interest. Lack of technical support for NGOs and moreover, funds to support their institution building processes appear to be important themes. SUSTAINABLE BANKING WITH THE POOR 6 ORGANIZATIONAL BACKGROUND AND ORIENTATION HISTORY In 1980, the Peruvian government enacted a law to create publicly owned municipal savings and credit banks in areas outside of the greater metropolitan region of Lima. 5 This law was created with two main goals: to promote the decentralization of financial services into outlying areas by creating institutions specifically designed to intermediate funds between local savers and borrowers, slowing the drain of rural resources to financial institutions in the capital of Lima; and to encourage the democratization of credit by serving low income clients unable to gain access to formal financial institutions. -- The CMACs were created as public financial institutions, with both social and economic missions. Each CMAC was created as an economically and politically independent institution and included as part of the formal financial system and thereby subject to regulation of the Banking Superintendency and the requirements of the Central Reserve Bank of Peru. The principal owner of the CMACs was established as the Provincial Council which is required to place a minimum amount of capital at the disposal of the Banking Superintendency each year. In 1982, the first CMAC was founded in Piura,. Following a request for support from the German Technical Co-operation agency (GTZ), financial experts were brought in from the German consulting firm, Interdisciplinaire Projekt Consult (IPC). In addition to working with Piura, IPC assisted in the development of a· strategy to create a nation-wide network of CMACs based on the model of the German savings banks, Sparktisse. This initiative resulted in the creation of a longer term project signed in 1985 involving a three-way partnership between GTZ, the Municipal Savings Banks of Germany, and the Banking Superintendency of Peru. Under this agreement, IPC was contracted to serve as the principal implementing agent for the project, providing on-going technical support to the project. The IPC assigned high-level personnel to this project, including the director of IPC, based in the CMAC at Piura for two years. German and local IPC consultants were directly involved in the project until 1996. The-second CMAC was established in Trujillo in 1984. Between 1985 and 1990, eight new cajas were established in the provinces of Santa, Arequipa, Sullana, Maynas, Cusco, Huancayo, lea and Paita. Two more were added in 1992 with the creation of Pisco and Tacna. Since 1995, a CMAC was started in Chincha, bringing the total number of CMACs to thirteen located throughout the country. In 1987, the Federacion Peruana de Cajas Municipales de Ahorro y Credito (FEPCMAC) was founded to serve as the national and international representative of the CMAC system and to promote the creation of new Cajas. The mandate of the FEPCMAC is both to audit and control operations of the CMACs and to provide each agency with specific support, training and assistance as required. IPC consultants work in close conjunction with the Federation to monitor the needs of individual CMACs and to review the results of regularly conducted audits. In 1990, a new decree was introduced by the government which allowed Cajas to create branch agencies in other provinces as deemed necessary. It also opened the way for external 5 The exclusion of the Lima region from the law was specifically designed to protect against the focusing of development resources on the capital, and where separate similar government initiatives were established. SUSTAINABLE BANKING WITH THE POOR 7 loans to the CMACs from national and international capital markets and provided for the creation of the Fondo Peruana CMAC or FOCMAC as a central channel through which these funds could flow. In 1993, the CMACs aggressively pursued a new strategy to expand microenterprise lending by mobilizing international funds. This strategy was motivated by new opportunities created by increased economic stability and the subsequent expansion of external funds; and by the recognition of the inadequacy of local savings as the sole source of finance for microenterprise lending. External financing to the CMACs was provided by FONCODES, COFIDE and IDB: partially through FOCMAC and partially through contracts negotiated directly with individual CMACs. Although this new focus on microenterprise lending was more heartily adopted in some CMACs, the results across the system were impressive. The microenterprise portfolio grew by US$11 million in 1994 and by another US$9 million in the first nine months of 1995. The active loan portfolio for microenterprise lending reached approximately US$25, 750,000 at the end of September, 1995. From 1995 to June of 1998, the loan portfolio has more than doubled, reaching US$65,000,000. While this record of expansion has been noteworthy, not all of the CMACs have achieved equal success. In June 1998, the five largest CMACs represented 75 percent of total assets for the system and the levels of institutional development between these and the smaller CMACs is significant. The implications of this are that the Federation has an increasingly challenging role in providing the technical support required by these diverse organizations. Moreover, as individual CMACs have matured, some have opted for policies that are increasingly independent to that advocated by the federation, as the central body of the CMACs. The issues raised by these circumstances form the challenges currently being addressed by the CMACs. The year 1996 was pivotal for the CMAC system. Several Federation staff members and the manager resigned or were released after several years of internal strife centering around the autonomy of the individual CMACs and the role of the Federation. The long standing relationship with the external consultants from IPC was strained as a conflict of visions came to the forefront. Despite the controversies, the role ofFEPCMAC with support from GTZ still remains critical to the operation of the CMACs. FEPCMAC coordinates CMAC activities, supervises training, management and accounting, and provides auditing. Also in 1996, the General Law for the Financial System was passed in Peru giving the CMACs a legal status of public institutions. The CMACs have provincial ownership which leaves them in a somewhat precarious economic and financial position. Different municipal governments have differing views on the role and management of the CMACs, thus limiting the role of the Federation. In addition, all activities and budgets must be passed by the Ministry of Economy. One of the goals of obstacles facing the CMACs is to find ways of clarifying their ownership structure and increasing their equity base in order to solidify their financial position. SUSTAINABLE BANKING WITH THE POOR 8  VISION AND GOALS Purpose The CMACs are non-profit organizations which seek to stimulate economic development in the provincial regions through providing access to reliable and affordable sources of financial services to support the productive activities of small and micro-busin~sses. A strong ethic of social development pervades their operations. This was apparent from discussions with all levels of staff during· our visits to different CMACs. The principal goal of the CMAC system is to establish strong and effective provincial fina.ncial institutions to meet these objectives. Each CMAC operates according to the following Management Principles: 1) to achieve full cost-recovery through interest charges to clients to avoid decapitalization of loan equity; • • 2) to operate according to clearly established institutional strategy, defining range of activities and services provided and the target market to be served; and 3) to use effective financial technologies to ensure competitiveness within local financial institutions. Conservative growth process. In ·addition, all new CMACs are required to undertake a gradual institutional development process. In the first year of operation, services are limited to gold "pawn" loans. In the second year, savings instruments are offered. Once established, progressively more complicated services are added. Microenterprise loans, which represent higher risks and require the greatest amount of credit assessment, are typically initiated only in the third year of operation once performance standards have been attained. A further goal is the continual refinement of credit and savings methodologies and the addition of new services. Part of this aim is to increase the efficiency of services to improve the profitability of CMACs with an objective of expanding lending capacity. The other aspect is to improve the range and quality of services available to clients. Regular market analysis and monitoring is undertaken to support these goals. Operational Goals Current objectives identified for each CMAC are the following: (a) to consolidate the growth of the CMAC system in order to ensure that it is able to compete effectively within local financial markets; (b) to strengthen the liquidity position of individual CMACs by improving access to external sources of funds (IDB, COFIDE, other) as a complement to local savings; (c) to reduce real interest rates charged to clients and financial margins used; and SUSTAINABLE BANKING WITH THE POOR 9  (d) to make more efficient use of existing capacity within the CMACs to rationalize costs and increase operational efficiency. TARGET MARKET Client Characteristics The principal target market for CMAC services is unbanked, low income sector of the population. ·Microenterprise clients vary widely. Box 1 presents a client profile of a typical client. A 1996 study6 found that the average client of the Huancayo CMAC had a loan of less than US$500 and used it for working capital for commercial activities. Thirty-eight percent of the borrowers were female. The educational background of clients was quite diverse, with 22 percent having incomplete primary education and 35 percent having some higher education. Some CMACs (including Arequipa and Cusco) are exploring the possibility of attracting larger, medium sized enterprises as borrowers:.. including those who could access bank credit. Piura, by contrast, is already focusing on a higher income client market, but obviously at the expense of smaller/microenterprise clients. The objective of extending their client base in this way is to generate income from these clients that could help to offset the higher costs of lending • to the target micro borrower. This activity is also viewed as strengthening the institutional status of CMACs. However, it is only viable insofar the CMACs have adequate access to liquidity to ensure that local demand for microenterprise credit is being satisfied. Outreach The CMAC system, driven by the Federation of all CMACs, aims to expand the outreach of the organizations in two ways. First, to extend the services of existing CMACs into other provinces of the district by providing mobile services and opening new branch agencies; and second, by launching new products to penetrate more deeply into existing local markets. Individual CMACs and the Federation regularly face pressure from Municipal; Provincial or District Councils wanting to launch either new CMACs or agency offices in their region. While their support is vital, expansion of the CMAC system through the addition of new cajas and branches is strictly regulated by the Federation based on their assessment of market viability. Federation approval for new offices is determined by their projected ability to become a profitable cost center within a one year period. One CMAC visited suggested that the criteria used by the Federation analysts to assess the market viability of new sites was too conservative (Cusco). However, the Federation points to the danger of over expansion of the system based on political pressures versus sound economic oppo~ities. 6 Benavides, Marisela "Cajas Municipales de Ahorro y Credito del Peru, Lima, 1996. SUSTAINABLE BANKING WITH THE POOR 10  Box 1. Typical Client Profile Female Food Merchant in Central Market Profile: IO years in business; three loans from CMAC, each for six months. Payments made weekly. Loans for working capital and for purchase of small equipment for display of goods. Loan amounts: 1st loan - 800 Soles (US$365); 2nd loan - 1200 Soles (US$545); 3rd loan - 1500 Soles (US$681) Monthly Gross Income: 5,000 Soles (US$2272) Net Monthly Income: 800 Soles ( US$365) Source oflnputs: Almost all products come from Peru. Some from Brazil and Chile. Products are purchased from local mobile sellers. Other Expenses: She pays a rent of 30 Soles per month for her stall at the market and a fee of 7 Soles per month for secure storage place. Other Sources of Credit: Capital structure of business has changed since use of CMAC credit, In the past purchased inputs on credit from suppliers; now uses CMAC loan. Supplier credit offered at between 20-50 percent interest per month. When consumption needs are higher, may revert to partial use of supplier credit and use CMAC loan to meet other household needs. Had used moneylender finance in the past for emergencies only. Very expensive (sometimes over I 00 percent per month- too expensive for business capital). Has never had a bank account or used the services of the bank even though it is located just outside the market. Savings: Member in a rotating savings club. Ten women save 5 Soles per day. Every 15 days, one member gets 750 Soles. The first to receive the money is the collector of the funds. She receives no other compensation. The second disbursement is given to the woman who was last in the previous round. Others go in tum according to a lottery. Money is typically used for consumption needs. "This system works well; no problems so far!" Socio-economic Information: Has four children; all have at least primary school education. (Education is a priority in Peru). Most help with business or do other jobs to earn income. Household/Business Assets: Husband has transport business. Used his truck to guarantee loan. Observations: Poor but stable economic conditions. No shoes: sandals. Lives at edge of city in barrio in rented house. Has aspirations to build their own house with accumulated family savings. Family is well nourished and has access to state medical services. Children required to work as soon as possible. She works long hours, but her business provides reliable source of income. SUSTAINABLE BANKING WITH THE POOR 11  SOURCES OF FUNDING AND ASSISTANCE Technical Assistance While the CMACs are undeniably Peruvian institutions, the effects of continued direct outside involvement cannot be underestimated. Perhaps most important is the vision and energy that GTZ/IPC and IDB bring to both the individual CMACs and to the Federation. While IPC's involvement ended in 1996, the IDB has continued to support the six weakest CMACs in 1997 and 1998 through Luso Financial Systems, a German firm specializing in microcredit. The impact of external technical assistance has not only helped to shape the evolution of the CMACs, but it also has connected the CMAC system with other international micro-lending projects, integrating the experience and innovations accumulated in other parts of the world. GTZ support has also been essential to the CMACs in assisting their access to international resources from international sources. At the local level, many clients indicated their knowledge of German support to the CMACs and suggested that this connection brought credibility to the CMACs. Donations It is estimated that the GTZ has provided approximately US$7.5 million between 1985 - 1995. These funds were applied toward technical assistance and institutional development purposes for both the individual CMACs and the Federation. None of these funds have been used for loan capital, although GTZ has made donations to Huancayo and Paita to support their net worth when these institutions failed to achieve banking law regulations requiring that each CMAC maintain minimum capital requirements. In addition, GTZ also contributed to a separate ~ trust with FOCMAC to finance CMAC lending activities. In 1993 and 1995, the IDB contributed grants ofUS$500,000 to the Federation for institutional development purposes. Loans and Investments The opening of financial markets during economic reforms has facilitated access to international funds. Loan contracts for finance of CMAC lending activities have been negotiated both through FOCMAC and directly with individual CMACs. The capital structure of each CMAC is distinct. Table 3 summarizes the structure of funds in December 1994 for five · CMACs. The CMACs have only limited financing from Peru's commercial banking system. These loans are secured against mortgages and other fixed assets and have been for relatively short periods (1 year). In no cases have loans been guaranteed against CMAC loan portfolios. As the performance of the CMACs has strengthened, there is strong potential to gain further access to private commercial markets. SUSTAINABLE BANKING WITH THE POOR 12  Table 3. Capital Structure in 1994 (Amounts in Soles) ~ohxc_ - e:·• • ·•.·.: .,• :.. it~ ·~~':·,_:· :\"~(~;~:i:;ri};-iAI~t~i~ - J~~ }s:i t{:}t,l;ll~C~/{:<:fi 1; :/ ~~~}t/ Equity 1 966.61 3 401.72 1 997.94 1 532.32 898.64 29.82% 16.30% 14.15% 20.10% 48.90% Deposits 2 564.37 15 443.21 9 096.37 4 247.80 511.45 38.88% 74.02% 64.43% 55.73% 27.83 % External Funds 2 064.88 2018.35 3 023.90 1 842.05 427.57 31.31% 9.67% 21.42% 24.17% 23.27% Bank Loans 1 587.32 2 018.35 2 872.58 1 842.05 0.00 24.07% 9.67% 20.35% 24.17% 0% IDB 65.5% 0% 15% 32% 0% National Banks 0% 51% 8% 32% 0% COFIDE 34.5% 27% 77% 36% 0% FONDEMI 0% 22% 0% 0% 0% FOCMAC 477.55 0.00 151 33 0.00 427.57 Loans 7.24% 0% 1.07% 0% 23.27% FONCODES 83% 0% 100% 0% 22% IDB 0% 0% 0% 0% 51% FOCMAC 17% 0% 0% 0% 27% Funds TOTAL 6 595.85 20 863.29 14118.21 7 622.17 1 837.66 100% 100% 100% 100% 100% SUSTAINABLE BANKING WITH THE POOR 13 INSTITUTIONAL STRUCTURE AND MANAGEMENT INSTITUTIONAL LINKAGES The Structure of the overall Caja Municipales system is set out in Figure 1 below. The CMACs are joined together as a system by the Federation (FEPCMAC) which, as the principal advocate for the CMAC system, plays an active role in lobbying government to ensure that guidelines and pre-conditions governing the operations of the Cajas are appropriate (as distinctive from those used to govern the commercial banks). Figure 1. Organigram of the CMAC System· Ministry of Econ. & IDB, Finance COFIDE,GTZ, GTZ:PERU Project Agreemen~.-+--I -P-C- ---+-,1 FEDERATION FOCMAC I I '--------t:::=----t-.J (FEPCMAC) ($$) □□□□□□□□□□□□ The Federation was established in 1987 as a public institution which has full economic, financial and administrative independence. It is financed by monthly contributions from the CMACs, with additional contributions from GTZ and the IDB. The Federation was created as a mechanism to co-ordinate the development of the CMAC system and with the mandate to serve as their central representative to national and international bodies. An important function of the Federation in this regard is assisting the CMACs in soliciting and negotiating credit lines with lenders; and in obtaining necessary authorization for these from the Banking Superintendency. The operational activities of the Federation include internal auditing, monitoring performance, and training staff. The Federation provides on-going staff training for all levels of CMAC personnel through regional and national workshops, courses and coordination of study visits to other microfinance institutions. The Federation is guided by two bodies: The General Assembly and the Board of Directors. The General Assembly has participation of the Mayors of CMAC municipalities, the Presidents of the Boards of Directors of the CMACs; and the Managers. The Board of Directors for the Federation consists of the following representation: three CMAC Board Presidents, three Managers and the President of the Federation (as elected by the General Assembly). Both of these bodies are designed to represent both technical and community development interests. SUSTAINABLE BANKING WITH THE POOR 14  FOCMAC was created as an independent entity with the objective of providing a central channel through which funds allocated to the CMACs from national and international sources could flow. The FOCMAC is supervised and regulated by the Banking Superintendency and is administered by the Federation. Capital managed through FOCMAC is derived from deposits placed by individual CMACs and external lines of credit, predominantly from CO FIDE and the IDB. The FOCMAC is directed by a Board of Directors which includes six members: • one Manager of the Federation (President); • three Managers from the CMACs (the top three financial contributors); and • two repres~ntatives from external contributors. The Board of Directors is responsible for the management, evaluation and control of FOCMAC operations. These include the procurement of lines of credit and loans for external sources; and the allocation of these as loans or investments to specified CMACs. Budgets are prepared on an annual basis for each CMAC through participatory methods involving dialogue with credit evaluators and based on strategic planning and forecasting sessions. Spending allocations are also set according to strict guidelines imposed for public institutions by the Ministry of Economy and Finance. Budgets are subject to approval by the Board of Directors and the Ministry. The Federation also reviews and comments on CMAC Budgets, but does not have binding authority on financial decisions made by each CMAC. Monthly variance reports are prepared for by each CMAC for review by management, Boards and the Federation. These are used by the Federation to guide it in designing its technical assistance for each of the CMACs and for comparing standards set by different institutions. INSTITUTIONAL FRAMEWORK OF THE CMACS The institutional structure of the CMACs has been rigidly defined by the framework of the legislation upon which they were founded. Each CMAC is an economically and politically independent organization. All CMACs are governed by separate Boards made up of local representatives. Each CMAC is subject to regulation by the Banking Superintendency and by the regulations set out by the Reserve Bank of Peru. Figure 2 illustrates the intraorganizational relationships within the CMAC Federation. Each Board cifDirectors must comprise seven individuals nominated by the Provincial Municipal Council and include the following representation: • three members designated for the Municipal Council (two from the majority and one from the minority); • one representative of the church; • one representative from either the Banco de la Nacion or COFIDE; • one representative from the local Chamber of Commerce; and • one representative of the small or micro-business community. SUSTAIN.ABLE BANKING WITH THE POOR 15  The Municipal Council has ultimate responsibility for the nomination of Board members. However, in practice, nominations are submitted through local organizations. Board members who are council representatives are nominated for a one year term, with all others appointed for two years. All positions are renewable for unlimited periods. Each representative receives an honorarium at each meeting. Approximately two meetings are held per month, unless otherwise deemed necessary. The primary function of the Board is to develop general policies for the CMACs and to hire. and supervise the managers. The Board does not undertake Executive functions and has no direct involvement in the operations of the CMACs, which are fully the responsibility of the management. Figure 2. .CMAC Internal Structure ·~unicipal Council I Board of Directors I External Auditor (Bank Superintendency/ Internal Audit Dept. I FEPCMAC) Managers I Credit Manager Savings Manager j J Administration Manager I ' I I I I I ! Information Human I Logistics and Dept. I I Resour~es I I l l Planning Branch I Program I Manag~.~~j • • • • • • • Administration I Accounting Dept. I General Cashier Dept. J IBranch ! I Credit Coordinators Staff: Savings Officers, Credit Microenterprise Disbursement and Payment Credit Assessors Collection NOTE: indicates basic structure of CMACs ------- indicates structures used in larger, more developed CMACs SUSTAINABLE BANKING WITH THE POOR 16  HUMAN RESOURCES Management The executive body of each CMAC is formed by the management team made up of two to three managers (depending on the size of the Caja). Although each manager has a specified area of jurisdiction (either credit, savings and/or administration), the management team also has joint overall responsibility for the performance of the institution. Managers typically share offices at the principal branch of the CMAC (where branches exist) and fulfill duties as the branch managers. Under this organizational framework, thei:e is no single executive officer with _ultimate authority over others; instead, all major management decisions are taken jointly. This system is used as a "check and balance" in decision-making and to promote shared responsibility in management of the cajas. Program Administration . The Program Administration Department incorporates all staff involved in program operations: Branch Officers (and Branch Managers where separate branches are established), credit evaluators (and in larger CMACs, Credit Coordinators) and Program Promoters. Each Branch also has a general cashier who is responsible for monitoring and controlling cash flow. Branch Office Sta.ff include tellers who are responsible for receiving savings deposits, administering term deposits, receiving loan payments. Specific tellers have designated responsibility for the administration of Pawn and Personal Loans. New staff to the Program Administration Department typically begin their work in the branch offices. Branch Managers are typically veteran Credit Assessors who have a broad range of exposure to CMAC activities. They carry overall responsibility for the operations and administration of branches and are directly responsible to managers at head office. Branch managers submit daily performance reports to head office management. Credit Coordinators are appointed in larger CMACs as middle management between Credit Assessors and the Credit Manager. They are essentially senior Credit Officers and are required to maintain a full client portfolio, but have additional administrative responsibilities of monitoring and supporting more junior Credit Officers. They assist in developing strategies for collecting on delinquent loans and for extending the client portfolios of Credit Assessors. Microenterprise Credit Evaluators are typically located in separate offices from the branches. Their time is divided between three main activities: assessing and administering new and renewed loans; monitoring and following-up delinquent loans through client visits; and recruiting new clients. In addition, Credit Assessors are responsible for holding daily training sessions on microenterprise credit in their offices. In these sessions, potential applicants are given details on eligibility requirements, loan terms and conditions and application procedures. Each is individually responsible for developing and managing their personal client portfolio. Credit evaluators are able to approve loans up to specified amounts on their own (this SUSTAINABLE BANKING WITH THE POOR 17 amount differs between CMACs), above which they must present the case to the Credit Committee for approval. Most credit evaluators manage 200 to 600 clients. A credit evaluator receives a daily loan delinquency report around which the day is planned. In some CMACs, Evaluators are also responsible for regular reporting of delinquent clients in a centralized "Risk Center" information system. Credit evaluators are always recruited from local areas. They are required to have university backgrounds in economics, business, or another related field. Agricultural economists and industrial engineers have also been hired to work in related sectors. All Credit ,Officers undergo an intensive six-month period of training. SALARIES AND STAFF INCENTIVE SCHEMES Salaries of credit evaluators are 60-70 percent base salaries, with 30-40 percent determined by an incentive scheme. This structure is determined by each CMAC individually, although is based on a standard equation developed by the Federation. Salary bonuses are calculated according to four principal factors: number of loans; number of new clients; total portfolio amount and delinquency rates. Variations on this equation are determined based on the local lending environment (agricultural lending for example in Sullana; and saturation of urban . markets in Paita). All staff members of each CMAC are also entitled to a percentage of the profits generated (as determined by each CMAC). The average earnings of credit evaluators with incentives included are comparable or superior to employees of local public institutions. In some instances, such as Piura, credit evaluator salaries are also competitive with private sector institutions. However, in our visit, most CMACs indicated that competition from the banks had been problematic and that many had lost well trained staff to the banks. This problem, linked to the "public" status of the CMACs and restrictions placed by the Ministry of Economy and Finance on pay levels, was being addressed during the period of our visit. CMAC staff were to be classified under new categories to ease salary ceilings. 7 Regional and National Meetings for different levels of staff provide a source of training and inspiration for staff members. Training missions in other IPC programs include one year apprenticeships in Germany with Sparkasse, visits to Calpia in El Salvador and Caja de los Andes in Bolivia. In addition, university and other professional enhancement courses are used to •upgrade skills of staff. FINANCIAL PRODUCTS LOAN PRODUCTS The CMAC system granted approximately 400,000 loans in 1998 with an average loan size of US$440. At the end of 1998, 222, 329 loans were outstanding, totaling US$98,250,000. The terms, conditions and credit assessment procedures for each of the credit products offered by the 7 These are to be implemented retro-actively, providing pay increases from the beginning of 1995. SUSTAINABLE BANKING WITH THE POOR 18 CMACs vary considerably. All CMACs loans are subject to the legal requiremenfthat they cannot exceed 5 percent of the net worth of each organization. There are three principal types of loans: Prendario or "Pawn" Loans; Personal Loans; and Microenterprise Credit. Pawn loans are a popular short term loan instrument. In June of 1998, nearly 80,000 clients had pawn loans, representing 7 percent of the outstanding loan volume. The average size of pawn loans in June of 1998 was US$77. Pawn credit is the first product offered by Cajas as they launch operations and is currently offered by all agencies. CMACs begin with this credit product not only because it is low risk (especially important since client savings are used to finance loans from the start of operations), but also \because it is simple, requiring minimal assessment and low cost to administer. Only 18 karat gold jewelry (sometimes silver) is accepted for pawn. Gold is effective as a pawn commodity since most Peruvians saved in the form of jewelry during the past years of hyperinflation. Borrowers are also required to provide identification (usually in the form of electoral cards) upon application. Credit delivery is instant. Clients receive up to 50 or 60 percent of the value of the pawned object, based on the assessment of the administering officer. The term of the loan typically ranges between 2 weeks to 3 months. Installments of loan capital and interest are paid on a weekly basis. Once repaid, loans can be renewed up to three times consecutively using the same pawned item. In cases of delinquency, borrowers face late payment charges. Pawn loans are used for emergency situations and are designed to provide a low cost alternative to the usurious rates charged by moneylenders. Interest rates charged are higher than microenterprise credit, set at between 6 and 9.5 percent monthly (calculated on a declining balance). Few borrowers use pawn credit for investment purposes. Personal loans comprised 23 percent of the June. 1998 loan portfolio for the CMAC system. At this time, some 48,500 clients used personal loans, averaging US$440 per borrower. In 1991, the CMAC system introduced Personal Credit as a consumer loan product. This form of credit is targeted at a higher income client market than the Pawn Loan clientele. Eligible applicants· must be able to guarantee loans either against their salaries or from savings held in the CMAC from CTS payments. CTS translates as Compensation for Time in Service and are essentially unemployment or pension funds paid by both employers (13 percent monthly salary) and employees (9 percent monthly salary). These funds are held in personal savings accounts at financial institutions and are restricted by law from withdrawals (unless in cases of personal emergencies: illness, funerals and marriage, when equivalent of 50 percent of one month's salary can be withdrawn twice per year). In cases where loans are guaranJ~_ed against salaries, applicants must be employed for a minimum of one year. The maximum size of the loan is determined by the monthly repayment capacity of the borrower, with loan payments not exceeding one third of the borrowers monthly salary. Credit assessment includes assessment of client's cash flow (income and expenses) and evidence of their credit rating (bill payment receipts and credit checks). SUSTAINABLE BANKING WITH THE POOR 19  Three procedures are used to administer personal loans: 1) Applicants have their employers sign a contract agreeing to have a portion of their salaries paid directly into the CMAC in flat rate installments. Most loans are structured this way and lower rates of interest are often charged because of reduced risks to the CMAC. 2) Employer signs as a guarantor for the loan and borrowers make payments independently. Interest rates for this arrangement are highest for personal credit (e.g. Sullana 6 percent) 3) Applicants borrow up to 50 percent of CTS savings in deposit. Credit guaranteed against savings is provided by most CMACs at extremely low rates reflecting low administration costs (e.g. Sullana 5.5 percent). The term of Personal Credit loans ranges between 1 month and 2 years, with an average term of 10 months. Interest rates for this product range between 5 percent and 7 percent monthly. Similar loans from banks - if available- are offered at 7 percent plus service charges and costs of providing official documentation, bringing effective rates to as much as 10 percent. Total time for turnaround of personal loan applications by CMACs is approximately 2-3 days. Microenterprise lending has been the institutional priority of the CMAC operations since 1993. In June 1998, microenterprise represented 70 percent of the total.outstanding loan balance. Nearly 80,000 clients had microenterprise loans, averaging US$817 per borrower. Table 4 shows that 37 percent microenterprise loans in 1995 were for sums less th,an US$500 while less than I percent of the loans were for quantities greater than US$ I 0,000. Prior to the full launching of this product line, extensive testing and refining of the microenterprise lending methodology was carried out in selected CMACs for a period of nearly 5 years. ,Experience from other IPC project~ in different Latin American countries was amassed and specific credit technologies (particularly credit assessment and enforcement strategies) were developed to overcome the high risks of business lending. Studies were made of informal lending practices, focusing on their flexibility and their agility in responding rapidly to client demand for credit and aspects of their methods were incorporated into the microenterprise lending technology. Table 4. Microenterprise Credit by Loan Size in 1995 No. of Loans % Volume % $0- $500 10,332 37.34% $3,135,473 10.65% $500-$1000 9,362 33.83% $6,683,578 22.70% $1000-$2000 5,168 18.68% $6,898,197 23.43% $2000-$5000 1,972 7.13% $5,973,079 20.29% $5000-$10,000 679 2.45% $4,552,498 15.46% $10,000-$20,000 131 0.47% $1,592,607 5.41% $20,000 + 27 0.10% $608,710 2.07% Total 27671 100.00% $29,444,142 100.00% . . Source: FEPCMAC, "Infonne Econom1co Fmanc1ero del Sistema CMAC al 30.06.95", 1995 . susrAINABLE BANKING WITH THE POOR 20  During this period, systematic studies of this segment of the client market were also ci;irried out to ensure that products introduced were appropriate to the needs of borrowers. Continuous attention has been centered on the introduction of new products and loan terms in response to perceived market opportunities. Examples· of this are specialized agricultural and fishing credit products which have been introduced for working capital and fixed assets, and the development of Automatic Credit and Parallel Loans. (Each of these is described in further detail in the Methodology section below.) Terms ofloans vary according to activity: working capital loans are usually for between 1 and 12 months; fixed asset loans range between 6 months and 2 years. Payment rates are also set individually by loan officers for each client. Some working in markets make payments on a daily basis; other agriculturists may repay their loans in as little as one or two installments over the period of a year. Most pay either weekly or monthly. • Rates of interest are different for each CMAC, based on market rates and requirement for each·CMAC to recover costs (average between.4.5 and 5.5 percent per month). Loan sizes also ·· range widely between different CMACs according to local ml;ll"ket conditions and economic activities (in lea: US$924, compared with US$264 in the Mayo.as CMAC), with an average of approximately US$823 in September 1995. Guarantees used to secure loans vary according to borrower capacity to pledge security. Both real assets, such as houses, vehicles, equipment and household goods and personal guarantees (usually through a co-signer) are accepted. Most important is the judgment of the credit officer in assessing both the ability and willingness of the applicant to repay. In principle, no client is turned away because of their inability to provide collateral. Initial loan sizes are typically small, increasing with regular increments as the credit record of the client is established. An important element of the success of the CMAC activities in the area of microenterprise lending is the training of loan officers. Credit assessors frequently have branch administration experience as part ~ftheir six month training to become a loan officer. By the· a time they are fully initiated into the field, they have good understanding of the operations of the CMAC and extensive field experience with trained lending officers. Training and supervision of personnel involves a substantial investment of GTZ funds (via the Federation). In 1997 and 1998, Luso Financial Systems has been instrumental in training loan officers from the six weakest CMACs with the financial support of the IDB. There is no forced savings component for loans. Interestingly, one CMAC did require borrowers to deposit funds as security against loans. However, this was quickly abolished when Federation consultants criticized this as a prohibitive and high cost pre-condition for clients (also in direct opposition to the IPC credit methodology used across the CMAC system). Due to the different client.markets for each of the CMAC credit and savings products, there is frequently no crossover between microenterprise borrowers and users of other products. In case a microenterprise client establishes credit history through use of another loan product, this is considered in the credit analysis process. However, there are no requirements that an applicant "graduate" from one product to the next. SUSTAINABLE BANKING WITH THE POOR 21  Loan Delivery System Screening Procedures Clients are visited by the credit evaluator in their homes and businesses (if different). Copies of national identity cards are required, or some equivalent. Applicants are requested to show evide11ce of bill payments, bank and other financial statements. Total household income is assessed with both the applicant and their partner since both are required to sign for the loan. The credit evaluator completes the application form with the applicant which includes an extensive cash flow analysis from all activities. Repayment capacity of the client is determined by the repayment capacity of the applicant before use of the loan. In the case of first application, detailed information regarding the status of the household is also gathered. Major assets (house, vehicle, equipment, inventory and other household items) are verified. Goods to be pledged as collateral are identified and original document of ownership is provided to the credit evaluator to be held during the period of the loan. If co-guarantors are used, they are assessed by credit evaluator and s_ecurity is pledged. During the visits to home and business, the credit evaluator verifies information given by the applicant with friends, neighbors and employees. Character information is sought informally. In cases where minimal or no collateral is provided, this portion of the analysis is more extensive. (In these cases a co-signer is always used, even if the real value of the security they pledge is substantially less than the value of the loan.) \ Once all information is gathered, a loan amount and payment system is established by the credit evaluator and discussed with the client. (This is sometimes completed during the first visit with the applicant.) The average time required for all aspects of the screening and application procedure is between 2-4 hours: one to two hours in the field; and one to two hours of administrative work verifying and processing disbursement for the loan. This time frame varies according to the location of the borrower (rural clients require additional travel time, although they are typically visited in batches), whether the client has been referred by an existing client, the nature of their business and the availability of collateral to secure the loan. Credit assessment for larger~ more established businesses that are able to fully collateralize loans is significantly more efficient. This encourages credit evaluators to develop a mixed portfolio of borrowers to assist in · achieving targets required by their incentive scheme. Loan Approval Once the application is completed, it is sorted into two categories: those within the loan size limit of the credit evaluator that can be disbursed directly; and those for loan sizes beyond the approval capacity of the credit evaluator which require signatures of one or more managers through the Credit Committee. Each CMAC determines its own approval ceilings. In most instances, credit evaluators can lend up to approximately US$ l ,OOO on their own; amounts SUSTAINABLE BANKING WITH THE POOR 22 between US$1,000 and 3,500 require the signatures of one manager and above US$3,500, the signatures of all managers are required. Loan approval usually takes place within one to three days of the request. Once approved, the file information is input by the Information Department into the portfolio management system and the disbursement process is initiated. (This process varies between the CMACs) Disbursement System As the disbursement order is issued through the system, a report is generated listing loans to be collected at the branch. The credit evaluator indicates to clients the date when their loans will be available at the specified office of the CMAC. This means that clients can use whatever branch is most convenient to collect loans and make payments. However, most CMACs are not yet able to provide this service. Designated tellers at each branch use the report listing as a reference. Files for all new loans are pulled in anticipatio~ of clients' visits to cross check that the disbursement order is correct. Once clients collect their loans, the teller instantly updates the system which issues a receipt for signature by the borrower and a payment schedule. The payment schedule provides a transparent breakdown of the capital and interest portions of each payment. All loans are issued in cash (Soles or US Dollars). Repayment is always in the same currency of the loan issued. Cash at each branch is managed by the General Cashier who reconciles at the end of each day. Only tellers and the Cashier's Department have direct access to cash in the system. The system is highly automated and is standardized across the CMAC system. Statistics are posted for the client portfolios of each credit evaluator at the end of each day: number of • clients, amount outstanding, delinquency rates and total branch performance. These are monitored by the mangers and credit coordinators on a daily basis. Loan Payment Procedures Payment schedules are determined jointly by the credit evaluator and borrower. Clients are required to go to the CMAC branches to make payments on their loans. Reports of payments due are issued each day and are used by (other) designated tellers who are also receiving savings and conducting other services. Once the payment is received, the teller updates the system which generates a receipt for the borrower which documents all payments made and indicates the amount still outstanding. Clients are able to repay loans before due dates without penalty. Clients who fail to make payments on the dates specified automatically are charged a late payment fee. This is typically 0.9 percent of the capital due on payment and is charged on a daily basis until the loan is repaid. Late payments are flagged in the system and are recorded on the receipt provided to the client. . "\';"" SusrAINABLE BANKING WITH THE POOR 23  Loan Renewals Clients are able to renew loans immediately upon repayment of their credit. Loan amounts are usually incrementally increased with ea~h new loan. Terms and size ofloans are mutually determined by borrower and credit evaluator. Credit evaluators undertake a reduced credit assessment building on existing data available in client files. Visits to the business and home may or may not be undertaken with each new application (to the credit evaluator's discretion). Cash flow analysis, household income sources and profit levels are updated; loan activities are considered and repayment capacity is calculated. If required, additional documents are gathered and signatures of co-signers obtained. Completed applications are then forwarded through the approval system and disbursement is arranged. The turnaround time for loan renewals is between 1-3 days. For clients with strong records, this process is fairly standard. Clients with late payments are subject to a more rigorous analysis. Reasons for late payments are considered and verified. Credit evaluators determine whether their access to larger loan sizes should be affected and may wish to discuss cases with loan coordinators or managers. Loan Monitoring System • Credit evaluators do not visit clients on a regular basis beyond the contact that their work in the markets and communities brings them. No verification of loan use is undertaken and no form of technical assistance or support is provided to clients. Unless a meeting is either specifically solicited by the client or the credit evaluator is flagged by a non-repayment on the part of a borrower, contact between client and credit evaluators is usually limited. \ Performance of clients is stringently monitored through daily portfolio reports prepared for each Credit Officer. These, supplemented by reports on overdue payments and delinquent loans, provide credit evaluators and managers with vital up-to-date information on portfolio quality. Immediate response to problem cases is regarded as a fundamental element of the CMACs' low delinquency status. Credit evaluators follow up late payments with a visit or telephone call to the client as soon as this information is available (sometimes in the evening of the first day). Most instances of late payments are for logistical reasons and are resolved within five days. Although credit managers monitor overall performance rates and review the frequency of late payment cases per credit evaluator, they do not become directly involved in these cases unless requested by the credit evaluator, or after a specified period of time. SAVINGS .PRODUCTS Based on the driving philosophy of the CMACs to provide intermediation of local resources, savings activities form a vital pillar of their operations. With its experience in the area of voluntary savings, the example of the CMACs is significant in Latin America as a microfinance institution. The institutional design and emphasis on savings is similar to many credit unions operating in Latin America. SUSTAINABLE BANKING WITH THE POOR 24  The initial focus on savings has required CMACs to develop a strong and professional public image to build confidence among prospective clients of their ability to manage the security of their deposits. In order to compete with the services of the banks that they operate alongside, CMAC savings are guided by the following principles: • attention to building a professional and solid institutional image; • guaranteeing security by providing deposit insurance for all savings held in the CMACs, up to a limit of US$20,000; • · rapid and friendly service; 8 • appropriate business hours; and • attractive interest rates. The terms and specifications for each loan product vary among the CMACs according to local market conditions. In the 1990's, the CMACs have attracted an increasing number of savers by offering interest rates slightly above those available in the commercial banks. Economic growth, a liberalized financial system, a reduction in inflation, improved consumer confidence, and the availability of attractive savings instruments have allowed for savings to increase throughout the country. The combination of these factors has allowed savings in the CMACs to increase from US$5.8 million in 1991 to US$69.8 million by September of 1998. Table 5 shows a breakdown of deposits by loan size for the CMAC Federation. Over 3/4 of the volume of savings (and only 10 percent of the number of accounts) come from medium and larger savings accounts of over US$1000. These accounts provide the CMAC system with a significant volume of funds (nearly US$55 million in September 1998) while offering a valued financial service to middle income clients. Interestingly, 64.1 percent of all depositors have accounts of less than US$100. These accounts contribute only 4.9 percent of the volume of savings and therefore are not a significant source of funds for the CMAC system. The high number of very small accounts is not surprising given the income disparity in Peru9 . The transaction costs associated with these accounts are quite high. However, these small accounts provide the very poor with a highly demanded financial service. There are several well documented 10 justifications for incorporating savings into the financial intermediation process. First, the use of internal funds leads to conservative lending strategies. When funds are external, a more careless attitude towards lending can prevail, leading to higher default rates. In addition, those with savings accounts are interested in seeing that the bank is sustainable, with sound banking practices and attention to profitability. Most importantly, savings facilities provide members with an opportunity to smooth consumption, speculate, and hedge against inflation, rather than using informal sources that may have higher risk. 8 Arequipa reported that it could conduct deposit transactions in 8-12 seconds! 9 The poorest IO percent of Peru's population represents only 1. 9 percent of the national income. The World Bank, World Development Report 1997, Oxford University Press, 1997. • IO For example, USAID, "Mobilizing Savings and Rural Finance: The AID Experience" USAID Science and Technology in Development Series, 1991. SUSTAINABLE BANKING WITH THE POOR 25  Table 5. Deposit Accounts by Loan Size as of September 30, 1998 Loan size in No. of deposits % Volume % Average US$ deposit ::;100 75,907 64.1 $3,458,000 4.9 $46 101-1,000 30,876 26.1 $12,298,000 17.3 $398 1,001-5,000 8,925 7.5 $17,745,000 24.9 $1,988 5,001-25,000 2,346 1.9 $18,637,000 26.2 $7,944 25,001 + 507 0.4 $19,049,000 26.7 $37,573 TOTAL 118,561 100.0 $71,190,000 100.0 $600 Source: Portocarrero Maisch, F. "Mecanismos para Promover el Ahorro Popular: La Experiencia de las Cajas Municipales de Ahorro y de Creditos (CMACs) del Peru" Luso Financial Systems, 1998. Deposit Accounts Two types of deposit accounts are available: a savings account that allows for payment orders to be issued from it (essentially a current account, without checking facilities); and a passbook savings account. Payment orders may be used among local businesses. There is no minimum balance requirement for these accounts. Interest rates paid on deposit accounts vary between I and 1.2 percent and are typically between 0.1 and 0.2 percent higher than local banks. In June 1995, 92 percent of active savings accounts held deposits of less than US$500. These accounts made up 18 percent of total savings held by the CMACs. Lack of checking facilities is a major disadvantage of saving with the CMACs. Many microenterprise loan clients interviewed indicated that they held savings in banking institutions because of their need to pay for inputs for their business purchased in Lima with checks. Term Deposits Term Deposits are offered for periods of 1, 3, 6, 9, 12, 18, 24, and 36 months. Term Deposit clients include private individuals, public institutions and enterprises (provincial and municipal organizations, including local municipality authorities and district councils, colleges, clubs and others). In June 1995, 79 percent of term deposits were for amounts less than US$1000. These deposits contributed a total of 17 percent of total savings held in the CMAC system. An example of the structure of 1995 interest rates for term deposits from the Sullana CMAC is as follows: • TERM 31 days 90 days 180 days 360days 720days RATE 1.7% 1.8% 1.9% 2.0% 2.2% SUSTAINABLE BANKING WITH THE POOR • 26  CTS Deposits CTS deposits (described in Personal Loans above) are a decreasing source of deposits in the CMACs dropping from US$500,000 in December 1993 to US$446,000 (-20 percent) in 1994. These funds have restricted withdrawals and thus provide a source of long term capital for the CMACs. The nominal interest rate in 1995 on CTS deposits was between 2 and 2.2 percent. Most of the CMAC savings is comprised of savings deposit accounts and fixed term accounts. Table 6 shows the growth of these accounts during the 1990's. Each type of account has demonstrated steady growth throughout the 1990's in terms of total volume. This growth can be explained by both external and internal factors: the effects of an improved economic environment (increased potential for higher profits and lower inflation levels encouraging cash savings); and the higher profile and strengthened public image CMACs have developed as a result of their expanded lending activities and their institutional development (visible through the opening of new offices and branches). • Despite this expansion, internally generated savings have become a less important source of funds to the CMAC system. The ratio of total deposit volume to total liabilities has fallen from 0.83 in 1991 to 0.59 in 1998 (Table 6). The reason for the declining deposit to liability ratio is that the CMACs are increasingly reliant on long term lines of credit from public and multilateral donor organizations 11 . While this trend has provided a low cost line of funds for expansion of the CMACs, it has also made them more dependent on subsidized lines of credit and reduced the need to stimulate savings mobilization. While savings are a vital source of finance for their lending operations, the costs of mobilizing savings are relatively high. This is especially the case when compared with the subsidized credit lines available from multilateral aid institutions and development finance institutions. The Federation advocates that, since CMAC clients are expected to cover the real costs of credit (including operational and financial costs as represented in interest charges), it is preferable to have a mix of lower cost external funds to offset expensive savings. While access to subsidized donor credit allows the CMACs to pass savings on to their borrowers, it also has the effect of discouraging the CMACs to mobilize savings. The common dilemma of using readily available donor funds versus relatively expensive internal savings is one unintentional effect of donor funds that can lead to long term subsidy dependence. Table 6. Growth of Savings 1991-1998 (000 US$) DEC91 DEC92 DEC93 DEC94 DEC95 DEC96 DEC97 JUN98 Savings $3,767 , $3,475 $5,662 $9,807 $14,073 $17,043 $21,869 $23,467 Tenn $2,089 $2,064 • $5,987 $10,498 $16,958 $25,353 $39,867 $45,065 Deposits/L 0.83 0.86 0.81 0.67 0.55 0.54 0.56 0.59 iabil. .. Source: Portocarrero Maisch, F., "Caracterist1cas de! Sistema de CaJas Mumc1pales de Ahorro y Credito" Luso Financial Systems, I 998. 11 Such as COFIDE (US$23.2 million) and IDB (US$3.4 million) SUSTAINABLE BANKING WITH THE POOR 27  Large scale deposits are very volatile. In Paita, for instance, reports of seasonal variations in deposits have fluctuations from US$80 to US$275,000 within the year. Small scale savings average about US$ l 50 and are typically for very short periods. This means that they are not appropriate to the needs of CMACs for lending capital. 12 Attempts to address this have increased the percentage of savings held by individuals to 88 percent, with only 12 percent from institutions (June 1995). Recently, attention has been focused on attracting savings from the middle and upper income sectors of the population whose saving patterns (higher amounts for longer periods) provide a more agile source of funds. While still dedicated to serving the lower income groups, the CMACs perceive that this sector of the pofulation cannot provide adequate resources to finance their credit activities. A recent study 1 by Luso Financial Systems emphasizes the importance of targeting medium and large scale deposits as a stable source of capital. An important finding of the CMAC experience is that savings and credit clients are almost totally distinct from one another. Microenterprise credit clients in particular are unlikely to have savings in the CMAC system. Those that do hold deposit accounts tend to use banks with checking facilities to make financial transfers for purchase of inputs. In this regard, the ·banks provide especially tough competition. OTHER SERVICES Bill Payment Services B.oth clients and non-clients are able to make payments on utilities (water and electricity), school bills, various municipal taxes and charges through the CMACs. These services are costly in administration time to the CMACs, but are an important client service and help to develop public image of the CMACs as part of the financial system. Foreign Currency Services CMACs offer savings and credit in US dollars as well as in national currency. This reflects the highly "dollarized" nature of the Peruvian economy resulting from long periods of high inflation. Currency exchange services of US dollars are also offered by the CMACs. Clients typically demand credit in dollars. Rates of credit offered for foreign currency loans are lower to reflect the costs to client resulting from currency fluctuations {devaluation). However, attention should be drawn to the fact that client bears highe'st risk in this arrangement. CMACs are encouraged to lend in dollars since loans from the IDB and COFIDE are both allocated and repaid in dollars. 12 IPC study on the structure and costs of savings has recently been prepared. 13 Portocarrero Maisch, F. "Mecanismos para Promover el Ahorro Popular: La Experiencia de las Cajas Municipales de Ahorro y de Creditos (CMACs) del Peru" Luso Financial Systems, 1998. SUSTAINABLE BANKING WITH THE POOR 28  Products and Services under Development The CMACs have a continual evolution of financial products available. Several new products recently have been offered or considered including: • More longer term (2 to 4 years) investment capital loans for businesses (and personal loans in Arequipa), providing larger loan sizes to improve the capacity of businesses to capitalize; • An increase in financing for rural enterprises, including long term investment loans for the development of new crops (such as fruits and vegetables) and businesses, ensuring adequate diversification of the loan portfolio to .contain risks. The failure of the Banco Agrario left a void in the agricultural lending _market. The CMACs began offering agricultural credit in 1994. Rural microenterprise lending comprised only 4.4 percent of the loan portfolio in 1994 and it had grown to 11.6 percent by October 1998; • Leasing services are under consideration for agricultural, office and construction equipment; • Insurance contracts, providing security both at the institutional level and the client level. The experience of El Niiio in 1998 highlighted the need for insurance. Insurance covering fire, natural disasters, civil unrest and market fluctuations could be offered to cover both the Cajas and their clients. In addition, health and life insurance could lower individual risk, especially to those clients without access to other forms of insurance; • New credit card technology developed at Calpia in El Salvador was introduced to the CMACs in 1996 for Preferential Clients using parallel credit products. Parallel loans are typically working capital loans used to increase inventory during peak seasons, such as pre-Christmas or at the start of the school year, when sales substantially increase. Loan periods are typically from two weeks to two months. Rates of interest are the same as for microenterprise credit; • Automatic credit products have been introduced in pilot tests but are still under development. Automatic credit products provide long term clients with access to instant credit. Ultimately, automatic credit clients will be able to negotiate these loan contracts through Automated Teller Machines in their branches. This technology has been developed and tested in El Salvador at !PC's Calpia project. Automatic Credit is regarded as an important tool in enabling credit evaluators to extend their number of clients and reduce delivery costs for loans. This is a noteworthy innovation that could be replicated on a widespread basis in programs outside of Peru; • Private housing loans (home completion or improvement loans) and financing for public urban housing and utility development projects serving low income populations are products that are under consideration for development. These loans SUSTAINABLE BANKING WITH THE POOR 29  may be for sums up to US$10,000 and be granted in loan cycles of 2 to 5 years. The home and land could be used as collateral; and • Diversified savings products offer medium and long term savings including CDs, bonds, and long term deposit accounts. In.order to improve deposit services, the CMACs plan on consolidating the savings services of the CMAC network and link to national banks in order to lower costs and provide superior service. As rural lending is expanded, the CMACs will need to refine their delivery methodologies and technologies and explore alternative institutional strategies to improve service access to a diverse clientele. Maintaining security and cash controls are key issues faced in this area. PERFORMANCE PORTFOLIO QUALITY In June 1998, 7 percent of the outstanding microenterprise loan portfolio was in arrears (by 15 days or more). As Figure 3 illustrates, arrears reached a low of 3.58 percent in 1993 and have risen gradually since then. Figure 3. ARREARS 1992-1998 15 10 ~ 0 5 0 Dec. 92 Dec. 93 Dec. 94 Dec. 95 Dec. 96 Dec. 97 Jun. 98 Source: Portocarrero Maisch, F. "Caracteristicas del Sistema de Cajas Muncipales de Ahorro y Credito" Luso Financial Systems, 1998. Various explanations can account for the deteriorating repayment performance. On one hand, the slow increase in arrears corresponds with macroeconomic fluctuations in Peru during the past couple of years. A parallel trend has occurred in commercial banks. The effects of El Nifio have been particularly devastating in Peru in 1997 and 1998. On the other hand, the presence of increasing arrears also may be related to systemic problems that have been documented in other micro finance institutions. Von Pischke, Yaron, and SUSTAINABLE BANKING WITH THE POOR 30 Zander 14 offer several explanations for why repayment performance declines over time. For example, credit officers' attention to borrowers may decline over time. The bank may offer loans to riskier clients as it grows or may be inexperienced in managing risk for larger sized loans. The growth of loan size over time may increase the incentive to default. In addition, poor design, inexperience, and rent-seeking may contribute to default. On the part of the borrower, the clients may realize the inherent weaknesses of the program, including the common reluctance to forcibly collect repayment. Whether any of these explanations are valid for the CMACs will be revealed over time. Certainly, the growth in arrears merits careful monitoring. In order to curb this negative trend, the CMACs have slowed their rate of portfolio expansion by using a more cautious approach towards lending and have increased the provisions for loan loss from 3.65 percent of the average outstanding loan portfolio in 1996 to 6.54 percent in June 1998. CMAC loans are delinquent after one payment is missed (depending on the payment schedule). Delinquent loans remain the full responsibility of the credit evaluator until they are either re-scheduled or written off. As such, they can adversely affect the incentive bonus available to the credit evaluator for months. The following steps are taken to enforce loan collection: • ,,,--- • Credit evaluator calls or visits clients. Assesses whether the case is a can't pay or won 't pay situation. If client has specific difficulties, credit evaluator advises credit manager and loan is re-scheduled. • If client does not appear to be in difficulty and refuses to repay, a letter is issued by a lawyer stating deadline for payment of due amount or foreclosure. (Most cases are resolved at this point. However, clients no longer have access to services at the CMAC once legal services are used.) Where risk center information exists, data on the client is recorded for local/regional use, damaging the credit record of the individual. This is removed once repayment is received. • • • If client still neglects to pay, foreclosure proceedings are launched and assets are removed from the client. Clients are given a final opportunity to repay loan at this point. If not, collection agencies are hired to enforce foreclosure. Clients are charged a late payment fee of 0.9 percent of capital due. After 8 days, overdue clients are suspended from receiving future credit for periods determined by their loan officer. Partial payments are accepted, but outstanding portions are treated as late payments. Clients' continued access and ability to secure larger amounts is determined by repayment performance. No client who has been involved in legal action to recuperate loans is eligible for future credit. No systematic rating system is used to determine client performance levels or access levels. Good performance is reward_ ed with higher loan amounts, immediate renewals and eventually automatic and parallel credits. CMAC system aims to build strong relationships with clients through good service. Knowledge of clients is not necessarily as intense as in other micro- lending programs such as ADEMI where clients are visited at regular intervals and/or training 14 J.D. Von Pischke, J. Yaron, l!nd R.M. Zander, "Why Credit Project Repayment Performance Declines," Savings and Development, 1998. SUSTAINABLE BANKING WITH THE POOR 31 and technical advice are provided. Rather, knowledge of clients is developed through credit performance. • Detailed unaudited financial reports are ·generated by each CMAC and by the Federation for the CMAC system as a whole for monthly, quarterly and annual periods. These are produced by internal auditors and monitored by the Boards of individual CMACs and by the Federation. They include Income Statements, Balance Sheets and Portfolio Information. During the last year, monitoring of costs per product has also been introduced. Auditing is conducted at three levels. Formally, the CMACs are audited.by representatives of the Banking Superintendency as part of the formal financial system. The Federation also undertakes·an arinual audit of each CMAC using its own consultants. In addition, CMACs have their own internal audit department which is involved in a continuous review of CMAC operations. Branch offices, in operation since the adjustment of the law in 1993, have not been operated as profit/cost centers. All revenue and expenses for these offices is reported through CMAC head offices. However, this is gradually under reform with Arequipa having introduced cost profit/center management for its branches in September, 1995. Monthly Aging of Portfolio Reports are produced at three levels: credit evaluator; CMAC and overall CMAC system. Loan loss provisions are estimated from trends in these reports and from historical experience. There is no specific write-off policy. Loans stay on books until recuperated, or approximately one year from due date. Each case is reviewed individually. CMAC policy is to pursue recuperation of loans at all expenses to demonstrate the stringency of enforcement to borrowers. This has kept annual loan write-offs at an average of less than 2.5 percent of the portfolio. EFFICIENCY The CMAC system experienced very rapid growth in terms of number and volume of outstanding loans from 1992 to 1995, more than doubling in size each year (Table 7). In addition to granting more loans, the average microenterprise loan size increased from US$633 in 1992 to US$817 in 1998. This phenomenal ·growth was matched by significant increases in efficiency. Both the volume and number ofloans per loan officer increased during that period. However, a parallel increase in arrears led the CMAC management to slow growth in 1997 and 1998. The average microenterprise loan portfolio per loan officer dropped slightly from its 1996 high ofUS$318,000 to US$289,000 in June 1998. SUSTAINABLE BANKING WITH THE POOR 32  Table 7. Efficiency Indicators for Microenterprise Loans No.of Outstanding Average Volume/ loan No. loans/ outstanding loan portfolio loan size officer loan officer loans Dec. 92 4,253 $2,691,000 $633 $67,000 106 Dec. 93 7,116 $5,787,000 $813 $105,000 129 Dec. 94 20,266 $16,899,000 $834 $167,000 201 Dec. 95 -- 44,239 $33,500,000 $757 $238,000 314 Dec. 96 67,187 $47,001,000 $700 $290,000 415 Dec. 97 77,846 $62,583,000 $804 $318,000 395 Jun.98 79,640 $65,036,000 $817 $289;000 354 . . .. Source: Portocarrero Maisch, F., "Caractenst1cas del Sistema de CaJas Mumc1pales de Ahorro y Credito" Luso Financial Systems, 1998. - SUSTAINABILITY While indicators of repayment, efficiency, and profitability shed light on the financial performance of an institution, long run sustainability remains the most essential criteria for a microfinance institution. Several measures of sustainability exist. Table 8 shows the operational and financial self-sufficiency 15 of each of the CMACs. All of the CMACs except for Paita, Pisco, and Chincha have sufficient financial income to cover their operating costs. Half of the CMACs are able to cover both their operating and financial costs with financial income. When "other income" (including sales of adjudicated goods, other fees, etc.) is added into the definition of operational income, all CMACs except Paita and Chincha attain financial self-sufficiency. Huancayo and Cusco stand out as having attained the highest degree of financial self-sufficiency. Otero and Rhyne 16 define four levels of self-sufficiency for microfinance institutions. Level one can be described as nearly wholly subsidized while level four is fully autonomous. Using this classification, some of the CMACs are at the second stage of self-sufficiency, while the stronger ones have passed to the third stage. According to the authors, 15 For this analysis, operational self-sufficiency is defined as financial income / operating expenses. Financial self- sufficiency is defined as interest income/ (operating+ financial expenses). 16 Otero, M. and Rhyne, E. The New World ofMicroenterprise Finance: Building Healthy Financial Institutions for the Poor Kumarian Press, 1994. SUSTAINABLE BANKING WITH THE POOR 33  Table 8. Self Sufficiency Indicators by CMAC (Thousands of new Soles, July 1998) Financial Operating Financial Operat. Self- Financial Self- Income Expenses* Expenses sufficiency Sufficiency Sullana 6641 5501 2231 1.21 0.86 Piura 17,202 8577 8710 2.01 . 1.00 Huancayo 2609 1468 565 1.78 1.28 Arequipa 16,621 9449 5770 1.76 1.09 Cosco 6778 3001 2121 2.26 1.32 Maynas 2544 1641 527 1:55 1.17 lea 3745 2867 1045 1.31 0.96 Paita .1806 2317 410 0:78 0.66 Santa 1846 1393 .425 1.33 1.02 Trujillo 6830 5543 1760 1.23 0.94 Tacna 2845 1545 714 1.84 1.26 Pisco 965 982 262 0.98 0.78 Chincha 89 174 0 0.51 0.51 TOTAL 76,919 60001 26153 1.28 0.89 *includes all administrative and loan loss provision costs Source: "Estado de Ganacias y Perdidas de las Cajas Municipales" Ajustado por lnfacion segun circular S.B.S. 1999. "The lowest level, level one, is associated with traditional, highly subsidized programs. At this level, grants or soft loans cover operating expenses and establish a revolving loan fund. When programs are heavily subsidized and performing poorly, however, the value of the loan fund erodes quickly through delinquency and inflation. Revenues fall short of operating expenses resulting in continuing need for grants. At level two, programs raise funds by borrowing on terms near, but still below, market rates. Interest income covers the cost of funds and a portion of operating expenses, but grants are still required to finance some aspects of operations. At level three most subsidy is eliminated, but programs find it difficult to eradicate a persistent dependence on some element of subsidy. (Note: many programs at this level could move on to the fourth level, but they choose to continue to receive some subsidies in order to expand their operations or experiment with new products.) The final level of self-sufficiency, level four, is reached when the program is fully financed from the savings of its clients and funds raised at commercial rates from formal financial institutions. Fees and interest income cover the real cost of funds, loan loss reserves, . operations and inflation. SUSTAINABLE BANKING WITH THE POOR 34  While most of the CMACs have operating income that is sufficient to cover their operating and financial expenses, they continue to have some reliance on external subsidies in the form of subsidized interest rates from the IDB and free technical assistance. Several of the CMACs could easily elimi_ nate these and pass to the fourth level of self-sufficiency but choose not to given their availability. The Subsidy Dependence Index (SDI) 17 gives a measure of how much interest rates would have to increase hypothetically in order to cover program costs without using subsidized lines of credit. In the case of the CMACs, two sources of subsidization exist. The IDB has granted several loans at concessional interest rates of 1 percent annually. The second subsidy comes in the form of free technical support by GTZ and IPC. Until 1996, GTZ paid for two long term consultants and several short term consultants. After 1996, the role of GTZ / IPC was diminished considerably. Only one short term consultant worked with the CMACs. In addition, GTZ provided $300,000 for training and technical support. The SDI was calculated for the years 1994 to 1998 and the results are shown in Figure 4. The SDI was 0.17 for the year 1998 (a dramatic improvement from the year 1994 level of 0.62). In other words, interest rates would have to be increased by 16 percent in order to cover program costs if no subsidization existed. The downward trend and relatively low SDI measure for the CMAC system is worthy of note. Many microfinance NGOs have SDI measures well above 100 percent. However, other institutions have attained SDI measures reaching zero. For instance, in 1995, BRI (Bank Rakyat Indonesia) had a negative SDI 18 and BancoSol of Bolivia had an SDI of 1.3 percent. 19 The SDI has declined significantly during the 1990's for several reasons. First, inflation has subsided considerably in Peru, allowing the on.:lending real interest rates charged by the CMACs to increase. Secondly, the CMACs have diversified their sources of funds, relying less on donor funds and more on alternative commercial loans and savings. Thirdly, the amount of technical support from the German donors has decreased since 1996. Finally, the SDI is positively affected by the growth of the loan portfolio throughout the 1990's. 17 Yaron, J., "Assessing Development Finance Institutions, A Public Interest Analysis" Policy Research Working Paper No. 174, World Bank, 1992. SDI calculated using the equation: (A(m-c)+[E*m)-p)+k) / (LP*i). 18 Charitonenko, S., Patten, R., and Yaron, J. "Indonesia: Banlc Rakyat Indonesia - Unit Desa 1970 - 1996" The World Bank, Sustainable Banking with the Poor, June 1998. . 19 Fidler, P. "Bolivia: Assessing the Perfonnance of Banco Solidario" The World Bank, Sustainable Banking with the Poor, Aug. 1998. SUSTAINABLE BANKING WITH THE POOR 35  Figure 4. SDI Measures for the CMACs from 1994 to 1998 SDI 1994-1998 0.7 0.6 - 0.5 0.4 c u, 0.3 0.2 0.1 0 1993 1994 1995 1996 1997 1998 1999 CROSS INSTITUTIONAL ANALYSIS The CMACs fare quite well when compared to other well-known microfinance institutions in Latin America. A financial analysis performed by IPC in 1995 compares the performance of the CMAC system with BancoSol, AMPES/Calpia, WWB Cali20 , and PROCREDIT/ Caja Los Andes. According to the study, not only did the CMAC system have the largest number and volume of loans, but its efficiency indicators suggest a strong performance. Table 9 shows that the CMACs have highest number ofloans per staff member of the five institutions. Only WWB Cali has a larger volume of loans per staff member. In addition, the interest income as a percentage of average loan portfolio is the highest for the CMAC system (62 percent). The administrative costs as a percentage of the average loan portfolio are simiiar among the institutions. Interestingly, while the CMAC system has the largest outreach, the depth of its outreach is impressive as well, as is proxied by the average loan size. Of the five institutions, the CMACs have the lowest overall loan size of US$306 21 . 20 It should be noted that WWB Cali is only one of five WWB Colombia affiliates. 21 This average includes each type of loan that the CMACs offer and is much lower than the CMAC microenterprise loan average of$757 in 1995. • SUSTAINABLE BANKING WITH THE POOR 36  Table 9. Cross Institutional Performance Indicators, June 1995 CMACs BancoSol AMPES/ WWBCali PROCREDITO CALPIA /Caja Los Andes No. Loans 112,887 59,590 8357 4253 10,088 Outstanding Outstanding $34,900,000 $26,400,000 $4,200,000 $2,800,000 $4,000,000 Balance .. Average Loan $306 $455 $504 $645 $390 Size No. Loans/No. 295 132 106 250 116 Staff Loan Vol/ No. $91,000 $59,000 • $53,000 $112,000 $46,000 Staff(US$) Interest Inc./ Av. 62% 37% 49% 54% 43% Loan Vol. Admin. Cost /Av. - 25% 25% 33% 23% 27% Loan Vol. Source: IPC, 1995. DISCUSSION: ACHIEVEMENTS, CHALLENGES AND THE FUTURE MAJOR ACCOMPLISHMENTS The achievements of the CMACs in the 1980's and 1990's are impressive. Through their development, they have gained recognition as credible and professional institutions within Peru's financial system. Their strong performance indicates that they have significantly contributed to the expansion of financial services to targeted low income and micro-business clients. At the end of 1998, the CMACs had collectively mobilized US$69 million in savings and held an outstanding loan portfolio of US$98 million. Characteristics of the CMAC system that have contributed to its success include: Vision: • creation of a social development objective for CMACs by improving opportunities in rural regions • insistence on full cost recovery and efficiency: continuous effort to improve client services Operating Context: • new era of stability following economic and political uncertainty • recent closure of rural financial institutions/credit programs has left market gaps • high level of dependency on expensive informal markets • economic boom; new investments and influx of external resources SUSTAINABLE BANKING WITH THE POOR 37  Institutional Features: • continuous and substantial institutional building support has been crucial to the establishment of the CMAC system • • leadership and vision from GTZ/IPC has been central in shaping the institution • affiliation with local council has brought public confidence (sense of permanency); affiliation with German aid has also helped_in this regard • attention building solid public image: secure and professional • staff incentives promote attention to performance • strong institutional identity shared among staff in most CMACs • efficiency and performance standards promoted by the Feder_ ation • integration with GTZ/IPC network has brought access to innovations, ideas and broad micro-lending experience • co~servative development of each new institution • substantial investments ,in staff training and continuous upgrading of the capacity of credit evaluators Emphasis on Savings: • •provision of both savings and term savings accounts • voluntary savings • small savers do not have to borrow, borrowers do not have to save • impressive volume and growth of savings, reaching nearly US$70 million by December 1998 Promoting Repayment: • •emphasis on client service and building long-term relationships • dependable and quick credit access • immediate follow-up on late payments • legal enforcement of all delinquent loans CHALLENGES Ownership and Governance The role of the Federation has been in question throughout the 1990's.22 Some of the smaller CMACs have viewed the Federation as an essential partner in their operations while 1 some of the larger, more established CMACs have favored more autonomy. Several CMAC managers complained of costly, interventionist policies from the Federation that interfered with their operations. From the Federation's perspective (supported by the GTZ/IPC team of advisors), a centralized vision and auditing were critical to the survival of the network. Tensions throughout the CMACs came to a boiling point at the 1996 General Assembly of the CMACs. The Federation manager resigned and several Federation employees were dismissed, leading to a void in the fragile cohesion of the system. As a result, internal auditing requirements were '1 22 Zeitinger, C.P. "La crisis institucional del Sistema de Cajas Municipales de Ahorro y Credito: causas y propuestas para su soluci6n", IPC paper, Peru, 1996. SUSTAINABLE BANKING WITH THE POOR 38 loosened, more autonomy was granted, the role of external consultants was lessened and additional questions related to the legal structure of the CMACs were created. The structure of the CMACs has brought certairi advantages in achieving some of the original goals of the system: most particularly, the development of financial institutions exclusively concerned with local markets. However, the drawbacks of this arrangement, especially in the area of governance, are significant. Important issues relating to ownership and governance structures are as follows: • Efficiency: In most cases, the Boards governing individual CMACs have critically assessed the operating efficiency of the CMACs. Their review of financial indicators has been cursory. As public appointees, Board members have no direct stake in the performance of the institutions. Board access to 50 percent of profits from each CMAC for local development projects has not been adequate incentive to promote efficiency. 23 Moreover, many Boards' are formed without consideration of the technical capacity of members to monitor the financial performance of institutions. Efficiency is monitored by the Federation. However, the Federation is not vested with power to enforce its position. 24 • • Agility: Ministry of Economy and Finance approval is required for spending decisions, including purchasing equipment, hiring new staff; salary levels; rental and property decisions, etc .. This has resulted in delays of months to respond to changing needs and conditions and has inhibited the ability of the CMACs to be competitive. Specific arrangements surrounding these regulations are being addressed, especially in the area of salary and staffing. However, there is evidence that this structure has weighed heavily on the operational agility of CMACs and hence their competitiveness with privately-owned financial institutions. • Financial Management: Partial allocation of profits for local development projects is a positive gesture, but can distort financial management by draining off resources which could be re-invested. This is particularly the case in relatively young institutions: need to reduce costs to clients first and refine/introduce products; or lend resources across CMAC system to assist in liquidity shortages where they exist. Board structure does not promote main concern for health of institution. • Lack of Cohesiveness: As provincially owned public institutions, CMACs are deviating from their original course to meet their own development visions. In some cases, this means movement away from the target market. Some CMACs have tried to become more like commercial banks,.favoring a more upscale clientele (such as Piura and Trujillo) while others continue to consider themselves as institutions serving the poor. The Federation has this role, but again does not have the power to enforce its vision. 23 Only in Piura have significant development projects been undertaken with dividend of profits: US$ I 00,000 in 1994, US$ 400,000 in 1995. Some political interference has occurred in decision-making in this case. Board should decide amount allocated to Council in consideration of business needs of CMAC. Instead, manager beholden to Board and providing funds to maintain their support. 24 Federation conducts regular assessment and financial monitoring, but relies on Banking Superintendency for support and rigorous enforcement. SUSTAINABLE BANKING WITH THE POOR 39  • Cost Intensive Structure: The costs of having an independent Board for each institution are high. Also, development of the independent systems that is emerging as CMACs mature is costly. It is difficult for CMACs to achieve the economies of scale required to compete with banks. This was especially apparent ih visiting Piura, Paita and Sullana, whose close proximity to one another and overlapping with each other's markets suggest that they might be more efficient as branches of a single institution. • Intermediation o/CMAC Funds: As independent institutions, there is no intermediation of excess liquidity within the CMAC system. This results in liquidity shortages in certain CMACs required to rely solely on external sources of funds. • Politicization o/CMAC Management: The CMAC affiliation with the public sector encourages politicization. This may hinder management and efforts to privatize in the future. The Federation is well aware of these issues, however, it can act only if it is given a stronger position of influence. To date, it has been unable to address fully these issues as they have emerged because of its limited powers. With increased power, the Federation could ensure that objectives of efficiency and transparency are maintained through its continuous monitoring and technical assistance activities. Given the history of conflict between some CMACs and the Federation, it is unclear whether the Federation will be poised to implement these goals. Thus, a key challenge facing the CMAC system is to address these issues while continuing to ensure that the CMACs maintain their local focus. By allowing provincial ownership of the CMACs, no clear, systemic vision is being implemented,on a national level.· Furthermore, the role of the government in managing the CMAC may have the effect of diminishing incentives for sustainability and profitability. With the number of protagonists in the CMAC scene (local governments, national government, CMAC management, FEPCMAC, international donors, GTZ, IPC, etc.), it is no surprise that the future path of the CMAC system is in question. The different visions for the future development of the CMACs include i.) a financial branch of the local municipal government, ii.) an autonomous regional bank, iii.) a decentralized network of institutions with partial autonomy, iv.) a completely privatized bank, and v.) a non-profit organization with private and public investors. FEPCMAC is interested in maintaining strong ties to preferential lines of credit. The FEPCMAC leadership has solicited subsidized credit from the World Bank in 1999. Adding another echelon to the institutional structure, FEPCMAC also is soliciting international assistance to create a new non-profit organization FUNDA CMAC (Fundaci6n para el Fomento de la Competitividad Microempresarial a traves del Ahorro 7 el Credito) which would focus on non-financial services to the microenterprise clients of the CMACs, including training, technical assistance and research. Clearly, no unified philosophy of future development exists among the interested parties, a fact that will stunt the CMACs' ability to evolve. Managing Growth Throughout the 1990's, the CMACs have experienced a very rapid growth of the outstanding loan portfolio and of savings. Since 1992, the volume of savings has grown at an average annual rate of27.5 percent and the outstanding loan volume has grown at a phenomenal SUSTAINABLE BANKING WITH THE POOR 40 81.9 percent annually. The extremely fast growth has allowed the CMAC system to generate revenues quickly and to obtain more favorable efficiency indicators (such as portfolio volume per loan officer). While scale economies appear to be present, certain risks are associated with growth of this pace. In a survey of micro finance institutions worldwide, a strong correlation between pace of growth and arrears was found 25 . Indeed, arrears have risen steadily throughout the 1990's for the CMAC system. It is critical that a deterioration of the loan portfolio does not erode public confidence in the CMACs, especially since savings mobilization is present. CMAC management has realized this and is making efforts to slow growth and focus on portfolio quality. One of the issues that the CMAC s'6stem must face as it grows is which source of funds it will use for its operations. A recent study 6 found that more than half of the deposit accounts in Trujillo and Sullana had less than US$10 in_ them, while more than 20 percent were moribund. Given the relatively high cost of maintaining very small accounts, the author indicates a need for the CMACs to mobilize funds from outside sources in addition to savings. If the CMAC system follows the prevalent thinking that mobilizing savings is too costly, the result may lead to some undesirable consequences. 'First, it may discourage small savings. It has been demonstrated in many developing countries that one of the keys to financial development is providing deposit facilities for the poor. 27 Secondly, it may encourage an unhealthy dependence on subsidized donor funds over time, preventing it from becoming fully sustainable. The CMACs focus on savings has allowed them to become one of the most comprehensive microfinance systems in Latin America. For their long term finandal viability, it is important that the CMACs seek ways of encouraging savings through new product innovation, appropriate savings instruments, and competitive rates rather than turn to readily available subsidized funds. CONCLUSIONS The CMAC system of Peru stands out as orie of the leaders in microfinance in Latin America ip terms of scale, outreach, and availability of both lending and savings services. In _ 1998, the CMACs granted nearly 400,000 small loans, averaging just US$440. Not only has the network been innovative in offering an array of loan services that cater to client demand, they have also offered savings products. The financial performance has been worthy of note, wit.Ji increasingly favorable measures of efficiency and profitability throughout the 1990's. As the CMACs enter into the millenium, they face several challenges. One of these is to find an appropriate legal and operational structure that meets the demands of the growing institutions. Currently, a number of conflicting opinions regarding the institutional vision are swirling around the organizations, preventing them from defining a clear path. The CMACs have become politicized with their public affiliations, thereby encumbering efforts of 25 Paxton, J. "A Worldwide Inventory ofMicrofinance Institutions" The World Bank, Sustainable Banking with the Poor, 1997. 26 Bredenbeck, K. "Savings Mobilization: Lessons from the Peruvian Municipal Savings Banks in Trujillo and Sullana," Savings and Development, Vol. 31, No. 1, 1997. 27 a Graham, D. and Von Pischke, J.D. "Factors and Strategies that Lead to Sustainable Supply of Financial Services forthe Rural Poor" IDB Workshop of Financial Services for the Rural Poor, Dec. 1994. SUSTAINABLE BANKING WITH THE POOR 41 privatization. The presence of government, non-profit international donors, local community leaders, private investors, and internal staff each with varying ideas about the future development of the CMACs have led to a degree of uncertainty and immobility. Another challenge facing the CMACs is how they will manage their growth while maintaining favorable performance indicators. The network is currently at a crossroads in deciding whether to behave more as a commercial bank by aggressively mobilizing internal funds and commercial lines of credit in order finance operations, or to act more as a non-profit organization that obtains and grants subsidized lines of credit. Its decision will affect the future viability of the CMACs in Peru. susrAINABLE BANKING WITH THE POOR 42  BIBLIOGRAPHY Benavides, Marisela "Cajas Municipales de Ahorro y Credito del Peru, Lima, Peru, 1996. Bredenbeck, K. "Savings Mobilization: Lessons from the Peruvian Municipal Savings Banks in Trujillo and Sullana," Savings and Development, Vol. 31, No. 1, 1997. Charitonenko, S., R. Patten, and J. Yaron, "Indonesia: Bank Rakyat Indonesia - Unit Desa 1970 - 1996" The World Bank, Sustainable Banking with the Poor,June 1998. Fidler, P. "Bolivia: Assessing the Performance of Banco Solidario" The World Bank, Sustainable Banking with the Poor, August 1998. Graham, D. and J.D. Von Pischke, "Factors and Strategies that Lead to a Sustainable Supply of Financial Services for the Rural Poor" IDB Workshop of Financial Services for the Rural Poor, Dec. 1994. Otero, M. and E. Rhyne, The New World ofMicroenterprise Finance: Building Healthy Financial Institutions for the Poor Kumarian Press, 1994. P~ton, J. "A Worldwide Inventory ofMicrofinance Institutions" The World Bank, Sustainable Banking with the Poor, 1997. Portocarrero Maisch, F., "Caracteristicas del Sistema de Cajas Municipales de Ahorro y Credito" Luso Financial Systems, 1998 . . Portocarrero Maisch, F. "Mecanismos para Promover el Ahorro Popular: La Experiencia de las Cajas Municipales de Ahorro y de Creditos (CMACs) del Peru" Luso Financial Systems, 1998. USAID, "Mobilizing Savings and Rural Finance: The AID Experienceli USAID Science and Technology in Development Series, 1991. Von Pischke, J.D., J. Yaron, and R.M. Zander, "Why Credit Project Repayment Performance Declines," Savings and Development, 1998. World Bank, World Development Report 1997 The World Bank, Washington D.C. 1997. Yaron, J., "Assessing Development Finance Institutions, A Public Interest Analysis" Policy Research Working Paper No. 174, World Bank, 1992. Zeitinger, C.P. "La crisis institucional del Sistema de Cajas Municipales de Ahorro y Credito: causas y propuestas para su soluci6n", IPC paper, Peru, 1996. SUSTAINABLE BANKING WITH THE POOR 43 APPENDIX 1. List of Case Studies SUSTAINABLE BANKING with the POOR List of Case Studies· · • Benin: FECECAM. Cecile Fruman. June 1997. (Also in French) . • Bolivia: Assessing the Performance of Banco Solidaro. Peter Fidler. August 1998. (Also iri Spanish). • Burkina Faso: Le Projet de promotion du petit credit rural - PPPCR. Julia Paxton. August 1997. • Colombia: Banco Caja Social. Julia Paxton. March 1999. • Colombia: Cupocredito Credit Union. Gloria Almeyda. March 1998. • Colombia: $olidarios_Financial Cooperative (Cali). Gloria Almeyda. April 1999. • Colombia: Women's World Banking. Julia Paxton. December 1998. • Costa Rica: FINCA Village Banking .. Julia Paxton, March 1998. (Also in Spanish) • Dominican Republic: ADEMI. McDonald Benjamin and Joanna Ledgerwood . May 1999. • Egypt: Alexandria Business Association. Tom Dichter. December 1997 . • Guatemala CARE Village Banks Project. Julia Paxton. October 1997. (Also in Spanish). • Guatemala: The Case of Uni6n Popular and Uni6n Progresista Amatitlaneca (UPA) Credit Unions. Gloria Almeyda and Brian Branch. April 1999. • Indonesia: Bank Rakyat Indonesia (BR/) Unit Desa 1970-1996. Stephanie Charitonenko, Richard H. Patten, and Jacob Varon. June 1998. • Kenya: KREP. Stephanie Charitonenko, Cecile Fruman and Glen Pederson . March 1999. • Mali: Self-Managed Village Savings and Loans Banks. Cecile Fruman. May 1998. • , . • Niger: Credit Unions (Caisses Populaires d'Epargne et de Credit). Korotoumou Ouattara, Mayada Baydas and Julia Paxton. April 1998. • Pakistan: Aga Khan Rural Support Progr.am 1982-1994. Tom Dichter. August 1998. • Philippines - TSP/. Tom Dichter. August 1998. • South Africa: Get Ahead Foundation. Craig Churchill. January 1998. • Thailand: BAAC - The Thai Bank for Agriculture and Agricultural Cooperatives . Tetsutaro Muraki, Leila Webster, and Jacob Varon. April 1998. • Zimbabwe: Zambuko Trust. Peter Fidler and Mohini Malhotra. April 1997 . SUSTAINABLE BANKING WITH TIIE POOR 44 APPENDIX 2. List of Discussion/Technical Papers SUSTAINABLE BANKING with the POOR List of Discussionrrechnical Papers In English: • Microfinance Practical Guide for World Bank Staff • A Worldwide Inventory of Microfinance Institutions • An Inventory of Microfinance Institutions in Western and West Central Africa • An Inventory of Microfinance Institutions in East, Central and South Africa • An Inventory of Microfinance Institutions in East Asia and the Pacific • An Inventory of Microfinance Institutions in South Asia • An Inventory of Microfinance Institutions in latin America and the Caribbean • Financial Sustainability for Credit Programs: A Travel Survival Guide • Indonesia's Rural Financial System: The Role of the State and Private lnsitutions • Outreach and Sustainability of Savings-First vs. Credit-First Financial Institutions: A Comparative Analysis of Eight Microfinance Institutions in Africa • Outreach and Sustainability of Member-Based Rural Financial Intermediaries in Latin America • Microfinance Handbook: An Institutional and Financial Perspective • Credit Unions in Latin America: Recent Performance and Emerging Challenges In French • lnventaire mondial des institutions de microfinance • lnventaire mondial des institutions de microfinance en Afrique de l'Ouest et du Centre • Le systeme financier rural indonesien : role de l'Etat et des institutions privees • Taux de penetration et viabilite financiere des institutions de micro-finance privilegiant l'epargne ou le credit: une analyse comparee de huit institutions africaines In Spanish • lnventario mundial de institutiones de microfinanzas • lnventario de instituciones microfinancieras en Latinoamerica y el Caribe • La autosuficiencia financiera: una guia basica para programas de credito en America Latina • SUSTAINABLE BANKING wrrn: THE POOR 45