Sustainable Banking with the Poor - Case Studies in Microfinance : Colombia Cupocredito Credit Union

SUSTAINABLE BANKING with the POOR Case Studies in Microfinance COLOMBIA- CUPOCREDITO CREDIT UNION Gloria Almeyda March 1999 SECTORAL & iT RESOI 1or.r:: r,FI\ITER The World Bank FEB 2 3 2000 WORLD BAr~K  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* (SDC), 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 Yaron; the technical manager is Carlos Cuevas, and the associate manager is Cecile Fruman. The administrative assistant is Stephanie Joseph-Goes. The World Bank SBP, J4-241, 1818 H Street, N.W. Washington, D.C. 20433 Phone (202) 473-3153 Fax (202) 4TT-2978 Internet: CCuevas@WORLDBANK.ORG or Sjosephgoes@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). This report was prepared by Gloria Almeyda (SBP Consultant). Cecile Fruman and Korotoumou . Ouattara (SBP) contributed useful guidance c1nd support. The author wishes to thank Beth Chung for • editing. ••••  TABLE OF CONTENTS MICROFINANCE IN COLOI\IBIA ................................................................................................ 1 POVERTY AND MICROENTERPRISE IN COLOI\IBIA ........................................................... .3 INSTITUTIONAL PROFILE .......................................................................................................... 4 LENDING SERVICES .................................................................................................................... 9 SHARES AND SAVINGS SERVICES ........................................................................................ 11 PERFORMANCE: MEASURING OUTREACH AND FINANCIAL SUSTAINABILITY ....... 14 LOAN PERFORMANCE: OUTREACH ........... :................................... ;...................................... 14 SHARES AND SAVINGS OUTREACH ..................................................................................... 16 MEASURING SUSTAINABILITY: FINANCIAL AND OPERATIONAL PERFORMANCE ........................................................................................................ 17 CONCLUSIONS AND RECOMMENDATIONS: CHALLENGES FOR CREDIT UNIONS IN MICROFINANCE ....................................................................................................................22 ISSUES OF CONCERN ................................................................................................................23 BIBLIOGRAPHY .......................................................................................... 28 APPENDIX A. PEARLS INDICATORS OF PERFORMANCE .................................... 29 APPENDIX B. LIST OF CASE STUDIES ............................................................ 31 APPENDIX C. LIST OF DISCUSSIONS/TECHNICAL PAPERS ............................... 32 SUSINABLE BANKING WITH THE POOR 1  LIST OF ACRONYMS AND ABBREVIATIONS CAVs Corporaciones de Ahorro y Vivienda (Savings and Housing Corporations) CUPOCREDITO Cooperativa Union popular de Credito DANCOOP Departamento Administrativo Nacional de Cooperativas (National · Administrative Cooperative Department) IFI Instituto the F omento Industrial .MMS Minimum Monthly Salary WWB Women's World Banking EXCHANGE RATES As of December 31 Colombian Pesos per 1 US$ 1994 US$1= P831 1995 US$1=P988 1996 US$=Pl,000 SUSTAINABLE BANKING WITH THE POOR 11  IBRD 30005 gue '= "---.,_,.....,..--.. '<~-,.; Tl,;, map""' p,oduc,,d l,y'"" Map Do,;go u,,;, of Tl,o Wodd &ml The boundariM, color,,denominotions and anyod,er inlonnation sbownon this map do nor imp~ on tl,e port of~ World Banl Group, any jvdgment on the legclstotus of any territory, or any 1t11don.ement or occeplo~ of such boundaries. 70° DECEMBER 1998 Country Profile Economic and Social Context Inflation GNP per capita (1995) US $1,910 1993 22.6% Population (1995) 36.8 million 1994 22.6% Population density 32 inhab./km2 1995 19.5% Adult literacy (1995) 9 percent 1996 21.6% SUSTAINABLE BANKING WITH THE POOR 111  CASE STUDIES IN MICROFINANCE COLOMBIA- CUPOCREDITO CREDIT UNION MICROFINANCE IN COLOMBIA Among Latin American countries, Colombia has been a pioneer in policies and programs for the informal and microenterprise sectors. 1 This support is the result of the efforts of a group ofNon-governmental organizations (NGOs) that initiated specialized training and microenterprise ·credit programs.2 In addition to the initiatives of the NGOs, institutions such as savings and credit cooperatives or credit unions have also provided financial services (credit and savings) to low-income groups for the past three decades. These groups include wage workers, _ self-employed workers, microentrepreneurs, female domestic servants, and the underemployed. International research on credit unions started in the early 1970s as part ofUSAID's efforts to examine financial markets in developing·countries. 3 The development finance literature has discussed credit unions in association with informal finance, rural finance, small farmers, savings mobilization, and housing finance. However, most of the available data is from internationally funded projects. Less is known about credit union development in other non- USAID recipient countries, such as Colombia. This case study was developed after a five-day visit to Cupocredito in Colombia in November 1995, and updated during a second visit in June 1997. Beginning in 1997-98, some Colombian credit unions were caught up in a crisis resulting both from their own institutional problems and the country's deteriorating political and economic conditions. Other factors are contributing to changes in credit union operations. A reform of the cooperative law was approved by the Colombian legislature in mid 1998 (as of July 1988, the regulations for implementation are still pending). In June 1998, the Superintendency of Banks assumed control of a group of the largest 52 credit unions, including Cupocredito. Other credit unions will be overseen by a new Superintendency of Economia Solidaria (institutions that are formed by association of individuals or non-profit institutions). Cupocredito continues its normal operations despite the changes in its regulatory enviro~ent. More recently, the literature on financial services for the poor has focused on institutions dedicated entirely to microenterprise credit, which excludes low-income worker groups, i.e., women workers in maquiladoras as well as other financial services, i.e. savings, housing credit (Otero and Rhyne, 1994). The major issues in microenterprise credit revolve around the financial viability of these institutions and the outreach and depth of the credit programs. Table 1 presents the most advanced Colombian NGOs specializing in microenterprise credit. The 1 For a summary of governmental policies from 1968 to 1994, see Ubemel Arboleda, 1997, Una misi6n posible: politicas de apoyo a Ia microempresa en Colombia, Bogota: Departamento Nacional de Planeaci6n, p. 63-85. 2 Among the leading NGOs are Carvajal Fundaci6n, Banco Mundial de Ia Mujer and CORFAS (Corporaci6n Fond de Apoyo a Empresas Asociativas). 3 Research has been led by the rural finance group of Ohio State University, Agricultural Economics Department and projects such as GEMINI financed by USAID. • SUSTAINABLE BANKING WITH THE POOR 1 NGOs with the largest equity and assets have been capitalized by some of the strongest industrial groups in the country (e.g., Grupo Santo Domingo, Carvajal). The affiliates of Women's World's Banking (WWB) have been capitalized by internationally and locally donated funds. Table 1. Microfinance in Colombia: Comparison ofCupocredito and NGOs: Equity and Assets - 1996 (US$ 000) Organizations Outstanding Total Equity Total Assets Loans Actuar:.Antioquia 3,348 7,768 14,232 Actuar-Caldas 522 139 804 Actuar-Quindio 539 299 776 Actuar-Tolima 3,966 923 · 4,525 Corfimujer 1,060 200 1,450 Fundesan 303 479 1,831 Micros Antioquia 2,063 595 2,647 Fundaci6n Santo 9,398 7,445 15,148 Domingo Corfas 2,673 519 • 4,476 Palau 90 33,258 34,221 Subtotal 11,294 51,625 80,110 Affiliates of Women's World Bank (WWB) WWB-Cali 4,049 2,342 4,810 WWB-Bogota 1,665 247 2,014 WWB-Medellin 1,3.97 616 1,814 WWB-Bucaramanga 1,594 414 1,879 WWB-Popayan 2.589 1,358 2,827 Subtotal 11,294 4,977 13,344 TOTAL NGOs & WWB 35,256 56,602 93,454 Cupocredito Credit Union 345,200 145,574 447,790 Note: Data for NGOs as of June 30, 1996. Data for Ct.ipocredito as of December 31, 1996. 4 Source: Data for NGOs: Marulanda (1997, Table 2) 4 Beatriz Marulanda, "Servicios financieros a la micro y pequefia empresa-Diagn6sticos, analisis y fonnulaci6n de estrategias." Washington, D.C.: The Inter-American Development Bank. March 1997. SUSTAINABLE BANKING WITH THE POOR 2 In contrast, Cupocredito Credit Union serves a broad low-income population with a range of financial services (including depository services and credit). Microenterprise financing, however, constitutes one of its major lines of lending. This case study examines Cupocredito, an unsubsidized institution that provides a broad array of financial services to low-income populations. It follows the progress of a credit union or a financial cooperative, as they are commonly called, over the past five years in Colombia. 5 POVERTY AND MICROENTERPRISE IN COLOMBIA Although Colombia compares well with other Latin American countries, its macroeconomic indicators show that the number of absolute poor continues to grow. Major differences in income, wealth, and living standards between regions and socioeconomic groups continue to be significant. Income distribution is highly skewed--IO percent of the population receives 38.5 percent of the total wealth. Within a population of 35.6 million, 18.8 percent of Colombians earn incomes that are below subsistence level. 6 The majority of these (70 percent) live in rural areas. 7 Furthermore, the rural poor are poorer than the urban poor. In rural areas the incomes of the poor are 43.3 percent below the subsistence level; the urban poor fall 31.2 percent below subsistence. 8 During the 1980s, the Colombian labor market was characterized by increasing informalization of the economy. From 1984 to 1988, the informal sector provided 63 percent of the new jobs created in the 10 largest metropolitan areas of Colombia. In 1994, the informal sector employed 53 percent of the economically active population (EAP). 9 Most of those who belong to the informal sector are independent workers (48 percent) or workers in establishments of either two to five employees (40 percent) or 6 to 10 employees (12 percent). More than 1,571,000 microenterprises operated in the ten largest metropolitan areas in 1994. The demand for microentrepreneurial and other productive activities is high in rural areas as well. Tapping this market niche adequately is a continuous challenge for Colombian microfinance mstitutions. How is Cupocredito doing it? The organization is succeeding through an extraordinary mobilization of community savings--without any external subsidies. 5 Article 99 of the Colombian Cooperative Law (Ley 79 of 1988), states that specialized financial services are to be provided by first tier savings and loans cooperatives [cooperativas de ahorro y credito, known in English as credit unions] and by cooperative organiz.ations, second-tier groups, and other finance-related entities. The term "financial cooperative" is also used for credit unions that provide services to non-members. 6 World Bank. 1994 Poverty in Colombia. Washington, D.C. p.2. The subsistence level is based on the "indigence line" calculated by Colombia's DANE (Departrnento Administrativo Nacional de Estadistica). in addition to the "indigence line," DANE estimates the "poverty line" which is twice the indigence line in rural areas and 2.3 times in urban areas. 7 Inter-American Development Banlc, "Colombia" in Basic Socio-Economic Data, April 21, 1997. 8 Ibid. 9 In Colombia, according to the DANE workers in the informal sectors are: unpaid family workers, domestic servants, non- professional independent workers, and workers in non-formal manufacturing enterprises that employ up to 10 people or other non-manufacturing sectors that employ up to five people. Commonly, the term microenterprise refers to enterprises of less than ten employees. See Arboleda (op.cit 1997, p. 41) and Hugo L6pez, 1996, Ensayos sobre economia laboral colombiana. Bogota: Editorial Fonade/Carlos Valencia. p. 173. SUSTAINABLE BANKING WITH THE POOR 3 INSTITUTIONAL PROFILE Cupocredito is Colombia's largest credit union--a cooperative enterprise specializing in credit and savings--built on the investment of low-income people. 10 As of December 1996, Cupocredito's total assets reached US$447 million, and its outstanding loans topped US$352 million (table 2). Savings plus deposits exceeded US$203 million, and members' capital shares were worth US$113 million. Cupocredito's assets are equivalent to 16 percent of the consolidated assets of the 84 largest Colombian credit unions. As a cooperative, it is owned by its 486,272 members, 59 percent of whom are men and 41 percent women. The average outstanding loan is US$1,444, while savings and shares per member are US$462 and US$233 respectively. Lo_ans are primarily for housing (down payment and improvement), business- related activities (manufacturing, commerce, services, and agriculture) and household consumption and investments (consumer durables, and education). Through its 128 branches, Cupocredifo serves middle- and lower-income neighborhoods and rural communities nationwide. Table 2. Cupocredito Compared to Top 84 Credit Unions(CUs) as of December 31, 1996 Group of Cupocredito Cupocredito's 84CUs Participation Members 2,111,172 486,272 .23% Assets (Millions US$) $2,863 $447 16% Deposit Savings Volume (Millions US$) $1,444 $203 14% Avg Savings Account Size $684 $462 Shares Volume (Millions US$) $499 $113 22% Avg. Share Account Size per Member $236 $233 Loans Outstanding Volume (Millions US$) $1,640 $352 21% Avg Loan Outstanding Size <US$5,000 n.a. $1,444 Source: Colombian Confederation of Cooperatives ORIGINS Cupocredito was created in 1960 by 28 working class leaders with 3,338 Colombian pesos. They sought to improve the economic situations of workers and their families. The main impetus for credit unions in Colombia came from Francisco Mejia, a Jesuit priest who promoted them among peasants and urban factory workers in the 1960s. At the time, credit for low-income groups was almost non-existent, and expensive if available at all. Cupocredito used "enterprise committees" to promote the credit union among workers. 10 No data are available for the exact number of credit unions in Colombia According to the National Administrative Department cif Cooperatives, there are over 1,000 registered credit unions in Colombia Many of them are closed-bond. Data for the largest 84, which are all open-bond credit unions, were available from the Colombian Confederation of Cooperatives. SUSTAINABLE BANKING WITH THE POOR 4  The mission and corporate objectives of Cupocredito make specific reference to financial services (savings and credit) for its members and the community: Cupocredito, in agreement with the cooperative contract, n seeks to contribute to the solution of its members' needs through the promotion ofsavings from which it will grant loans as well as other financial services, thereby strengthening an entrepreneurial culture based on solidarity and the well being ofthe members and the community. Although Cupocredito's mission has not changed, the process by which it is :fulfilled has evolved over the years. Initially, savings meant shares which received dividends based on surplus (total revenues minus total expenses). In 1978, a special savings line was created that paid interest every quarter at a minimum market rate (18 percent at the time, to keep up with inflation). Later, quarterly payments became monthly payments. In 1984, the credit union requested authorization from DANCOOP to pay interest on members' shares and to add a percentage interest at least equal to the inflation rate to each member's account. This constituted a major advancement among credit unions in Latin America and the Caribbean. Current by-laws identify a broad range of financial services for individuals as well as for communities (e.g., infrastructure projects within municipalities). LENDING AND SAVINGS ACTIVITIES Cupocredito exhibits an impressive scale of operations: 486,272 members (or clients), in 1996, whose small savings and shares of US$203 million and US$113 million respectively finance an outstanding loan portfolio of US$352 million via 159,729 loans (table 3.). It offers a variety of savings schemes, from daily interest-bearing accounts to fixed-term deposits. Lines of credit include a variety of terms, loan repayment schedules, collateral requirements and interest rates. Credit is available for business and household investments. US$302 million in loans disbursed during 1996 financed 121,816 loans of which 28.8 percent (US$87 million) were for business investments and 27.4 percent (US$82.9) for housing. Loans to members are financed with members' shares (in 1996, 25 percent of total assets) and savings-(45.5 percent of total assets). The ratio ofloans to savings is 1.3 and Cupocredito has a ratio of borrowing to assets of 16 percent. Six percent of the borrowing is in the form of loans from commercial banks, and 10 percent is long-term debt raised through issuing 35-year bonds. 11 The cooperative contract is defined by Law 79 of 1988 as the contract among a group of persons who agree to form a legal entity called cooperative whose activities are to be fulfilled with social objectives and not for profit. Therefore, the "profits" of a cooperative are called "surplus" or "excedentes" in Spanish. SUSTAINABLE BANKING WITH THE POOR 5 Table 3. Outreach: Cupocredito Credit Union Savings and Lending Activities SHAREHOLDERS Dec. 31, 1994 Dec. 31, 1995 Dec. 31, 1996 (also Members and Clients) (1US$=P831) (1 US$=P988) (1 US$=Pl,000) Number of members 353,166 407,363 486,272 Women members: (data for January1994) Savers 30% Borrowers 44% SAVINGS AND FIXED TERM DEPOSITS Volume of savings & deposit (Million US$) $115 $148 $203 1 Number of savings accounts 351,494 375,406 447,370 Average savings & deposits $328 $395 $455 balance/account Avg. Savings & Dep. to GDP/per capita 20% 23% 26% SHARES Volume of shares (Million US$) $84.8 $88.4 $113.3 Number of share accounts 353,166 407,363 486,272 Average share account size $240 $217 $233 Average share account to GDP/per capita 15% 13% 13% LOANS Volume of loans disbursed (Million US$) $167.7 $195.2 $302.5 Number of loans disbursed 92,554 99,063 121,806 Average disbursed loan size $1,812 $1,971 $2,483 Volume of loans outstanding (Million US$) $174.6 $219.1 $352.6 Number of outstanding loans 159,729 Average outstanding loan under US$5,000 2 $1,444 Average outstanding loan to GDP/capita 82% Notes: 'The number of savings accounts for 1996 was estimated using as a proxy the same percentage of accounts to total members as in 1995, equivalent to 92 percent (in 1994 it was 99 percent). 2 As of December 1996, 91 percent of the outstanding loans were under US$5,000. Source: Data from Credit Union Files COMMUNITIES SERVED Through its first 20 years, Cupocredito's management was cautious about opening branches outside the main office in Bogota because it lacked the necessary technology to provide secure transactions and communication. Starting in 1985, the credit union began a period of expansion, mainly through the rural communities in the central departments close to Bogota SUSTAINABLE BANKING WITH THE POOR 6  (Cundinamarca, Boyaca, Meta), and then opened branches in the rest of the country. As of 1995, Cupocredito had 111 branches distributed in five regions nationwide. Colombia uses a system of six strata for classifying household income distribution at the municipal level. These strata are used to establish utilities tariffs and for other purposes. Level 1 is the lowest income level ("low-low''), and level 6 is the highest. Cupocredito's br_anches are located in strata 1, 2, and 3, delivering services to a membership of medium-low, low and low- low income levels. Members include both wage and non-wage workers from rural and urban communities. A large percentage are microentrepreneurs and self-employed individuals. Fifty-four percent of the membership and 51 percent of the branches are located in urban areas. Rural branches are found in some 89 municipalities. In areas such as the Llanos (Eastern part of Colombia), Cupocredito is often the only financial institution. 12 Yet Bogota, including its surrounding rural areas, continues to be the major area of operations for Cupocredito: in 1995, 26 percent of the total number of branches and 41 percent of the membership were located in Bogota. This share has decreased over the past two years as Cupocredito expands to other areas. Cupocredito's financial services contribute to enterprise at two levels: at the community level, where there are no other financial institutions; and at a second level, where financial institutions exist but Cupocredito spurs competition and improved services. In many rural areas, the sole financial institution had been the Caja Agraria, a relatively inefficient, government-run agricultural development bank. Cupocredito's share of rural branches increased from 49 percent in 1995 to 52 percent in 1996. And now, as towns develop on the urban fringe, the credit union is providing new semi-rural populations with loans for commercial and construction activities. GOVERNANCE, SUPERVISION AND CONTROL According to the Colombian Cooperative Law, the administration of a credit union is based on three main internal bodies: the general assembly, the Board of Directors (a.k.a. the administrative council) and the manager. The external supervision of the credit union is performed by a supervisory committee and an external auditor (who has to be a certified public accountant). Until 1998, annual reports approved by the general assembly were presented to DANCOOP; but as of June 1998 Cupocredito is under the control of the Superintendency of Banks. One of the major challenges for Cupocredito is maintaining accountability to its membership. Members are elected to a nine-member administrative council. Annual meetings and elections to the administrative council are organized through delegates who are elected at the • branch offices. A member needs to be in good standing in order to serve as a delegate or to vote. Only about 20 percent of the eligible members vote, for a variety of cultural and logistical reasons. 12 Until mid-1997, Cupocredito branches had never been attacked by the guerrillas in some rural areas. SUSTAINABLE BANKING WITH THE POOR 7  Eighty-one percent of delegates at the General Assembly in 1995 were men, and 19 percent women. The Central region (Bogota) had the largest proportion of women nominated as delegates, almost 40 percent, compared to the 10 percent nominated in other regions. This is because in provincial areas women play more of a traditional role and participate less in cooperative meetings which are often held at night. Of the eight members elected to the Administrative Council, currently only one is a woman. The lack of elected women leaders reflects their historical absence from public affairs.13 Despite the limited leadership role of women as elected officers, they occupy key administrative positions in Cupocredito (discussed below). This is the result of Colombian women's increasing levels of education and participation in the labor force. • Members also elect a Supervisory Committee of three people responsible for monitoring the fulfillment of the obligations of the Administrative Council and the management. This "watchdog" committee is responsible for receiving complaints from the members and taking action when needed. ADMINISTRATION AND STAFF STRUCTURE The credit union is headed by a president and an executive vice president.. The main structure is made up of vicepresidencias: three regional divisions and five central departments in Finance, Marketing and Development, Administration, Technology, and Credit & Loan Collections. There are also a general secretary, three comptrollers, a director of human resources, and a legal director. Under these managers there are 11 division chiefs in the areas of services (Credit, Loan Collections, General Services, Investments), technical support (Operations, Organizational Methods, Telematic, On-line Operations, Technical Support), and other services (Insurance and Social Protection). The regional divisions report directly to the vice president. In 1995, 56 percent ofCupocredito's employees were women. At the executive level there were 21 female managers and division chiefs (representing 43 percent of executive management) across all areas of major responsibility. By early 1997, women were well- represented in key positions, such as the vice president for administration, the comptroller of computer systems, the legal director, the chief of organizational methods, and the chief of loan collections. A typical branch office (between 2,000 and 6,000 members) has the following staff: a manager, administrative secretary (second to the manager), one to two credit assistants, one collection officer, one membership assistant (help with members' new accounts and savings), one to two cashiers, an assistant secretary, and a maintenance person. Branch managers are paid 13 Colombian women did not vote until 1948. During the 1950s the countiy was immersed in a civil war. Women's traditional place was at home. According to Yolanda Gonz.alez, during the 1960s the right to vote did not have much importance for Colombian men nor women since the two ruling political parties shared power. Women's issues were mainly discussed at universities. For a discussion of Colombian women's political participation, see, for example, Magdalena Velasquez Toro, Las Mujeres en laHistoria de Colombia, Torno I, editorial Nonna, 1995. SUSTAINABLE BANKING WITH THE POOR 8 an average ofUS$1,250 per month, which is similar to the salary paid by other financial institutions. A credit assistant is paid about US$480 per month, and a cashier about US$380 per month, which are above average wages for financial institutions. • Since 1988 employees have received a share (bonus) of the profits of the cooperative, and salaries of employees are comparable to those of banks. Cupocredito has a quarterly performance review for branch managers. Factors they evaluate include level of savings mobilized, loans, delinquency rates, membership growth, and net income. In addition, branch managers are responsible for making Cupocredito' s services known in their neighborhoods and communities. Although Cupocredito advertises, word of mouth remains the most common form of institutional promotion. If the branch manager is not performing well, her/his approval privileges are suspended until performance improves LENDING SERVICES LOAN PROCESSING Only individual members can apply for loans from Cupocredito. Membership requires a minimum ofUS$30 in shares, to be increased, within a year, to one official minimum monthly wage (in 1997, the equivalent of US$! 70). Loan applications can now be submitted immediately upon joining the credit union. Branch managers have the primary responsibility for credit operations. Approval of small loans is decentralized, with branch managers approving loans of up to US$2,550. Ninety-one percent ofloans are for less than US$5,000. One unique feature of Cupocredito among large Colombian credit unions is its use of a credit committee at the branch level for loans in the range ofUS$4,250 to US$11,050. The •traditional credit union model usually has a credit committee made up of volunteer members for loan approval. As credit unions grow, credit committees are being replaced by technical credit staff. The credit committee comprises the branch's manager, the administrative assistant, credit assistant, loan collections assistant, and two members. The branch manager selects people to serve on the committee based on their knowledge of the community and the members. This committee evaluates the borrowers' loan application, and helps reduce the cost of information. The credit committee meets every week and loans are processed within five working days. A typical medium-sized branch processes about 20 loans per week. TERMS AND INTEREST RATES There are two criteria for determining the amount to be borrowed. First, the maximum request is determined based on a multiple of shares and savings. Second, the maximum is subject to debt capacity. For example, a person can deposit $1,000 and his/her maximum amount would be US$5,000. But the person's debt capacity might be as low as US$3,000. A summary of loan characteristics is presented in table 4. SUSTAINABLE BANKING WITH THE POOR 9 Table 4. Loan Terms: Credit Line Characteristics - December 1996 Type of Credit Terms Annual Nominal Adm. Reciprocity Guarantees Repayment Interest Rate1 Fee Transitory Upto30days 0.12% (day) None The average Shares Monthly savings or share balance of the past 90 days. Regular loans 12 months 31.5% 3% Up to 5 times based on multiple 24months 32.52% 2.75% average share of shares. New balances loan available after current loan is 70%paid. Loans based on 12months 32.98% None Up to 5 times Co-signers Monthly, or Fixed Term 24months 34.00% shares and average and/or collateral every 2, 3, 6 or Deposits 36 months 35.02% savings balances determined 12 months. 48months 35.98% according to 60months 38.02% debt capacity and risk. Real Estate: for loans >US$10,000 Loans based on 12 months 33.58% None Up to 5 times savings average. 24months 34.54% shares and average Special line for 36months 35.5% savings balances of members with previous 90 days business activities Special line for 36months Market rate on 90 None Up to Stimes Business day fixed term savings Investments deposit (DTF) + 8% Special DTF+7% None A share balance Real guarantee Monthly Microenterprise 12months DTF+7.40% equivalent to 1 Line with IFI* 24months DTF+7.5% MMS** (or 36months DTF+7.75% US$170). Share 48 months DTF+7.9% and or savings 60months average balances helps to leverage loan and to adjust interest rates Housing 120months DTF+9 None Up to 10 times Real guarantee, Monthly, or <US$25,000 DTF+lO level of shares and mortgage every 2, 3, 6 or >US$25,000 average savings 12 months balance D D Notes. . As on June 1997, average interest rates on 90-day loans were at 33.231/o m banks and 36.761/o tn financial companies. As of June 1997, the interest rate on a 90-day fixed term deposit (DTF) was 20.41%. *IFI = Institute de Fomento Industrial (government second-tier financial institution) ..MMS=Minimum Monthly Salary; Source: Data from Credit Union files SUSTAINABLE BANKING WITH TIIE POOR Loan Terms: Cupocredito offers several credit schemes for a variety of investments. Loan terms vary from one month (working capital) to 10 years (housing), and are determined based on the activity to be financed and debt repayment capacity. The monthly payment--or other repayment--schedu.le may not be more than 30 percent of income. Business investment loans can have terms of 12 to 60 months. Seventy-four percent of the outstanding loan portfolio as of December 31, 1996 was less than one year. Interest Rates: Interest rates are determined based on market rates and are adjusted to terms and risks. Nominal interest rates vary according to the terms of the loan, the purpose ofhe loan, the guarantees, and the source offunds{special lines are available from the second-tier financial institution IFI). Administrative fees are charged according to the level of savings and shares a borrower uses to leverage a loan and the loan term (the longer the term, the smaller the fee). As of June 1997, Cupocredito's nominal rate was 33.58 percent, a rate that compared •well with the banks' rate of 41.20 percent and the commercial financial companies' rate of39.87 percent. The financial corporations, with a 31.3 7 percent rate, target larger borrowers. The real a effective interest rate in Cupocredito for a 12-month loan with nominal rate of33.58 percent was about 16.33 percent, lower than a comparable rate from a commercial financial company of 17.23 percent. According to a recent study conducted by the IDB (Marulanda, 1997), the effective interest rates of a group ofNGOs specializing in microenterprise credit was 59.6 percent, which resulted in a real effective interest rate of 34.57 percent (assuming an inflation rate of 18.60 percent for the last 12 months). SHARES AND SAVINGS SERVICES One ofCupocredito's major contributions to microfinance has been its strong record of savings mobilization among lower-income groups in urban and rural Colombia. As of December 31, 1996, the level of shares was US$113 million and savings US$203 million, a total of US$316 million. Average balances per member were US$233 in shares and US$455 in savings. The minimum share investment required to become a member is one official minimum monthly wage, which in 1997 was equivalent to US$170. The ratio of number of accounts to total members is 99 percent (351,494 to 353,166). Most shareowners are also depositors. ' SAVINGS OPERATIONS The main difference between Cupocredito and other Colombian microfinance institutions is its savings mobilization capability. Therefore, the a~strative structure and operations of the credit union need to respond to the delivery of both savings and credit services. Branch managers are responsible for savings mobilization, investment, and income generation. SUSTAINABLE BANKING WITH THE POOR 11  TYPES OF SAVINGS AND DEPOSIT ACCOUNTS Cupocredito offers both highly liquid, interest-paying savings accounts and fixed-term deposits (table 5). Most of Cupocredito's shares and savings accounts are under US$500. In April 1997, the average savings passbook balance was US$186, and shares were US$259. Savings accounts are opened with a minimum of US$30; the minimum balances required are flexible as well. There are four types of savings: regular passbooks, daily interest-paying passbooks, special savings for children and youth, and sharedraft or checking accounts. Fixed-term deposits are also available for a variety of terms (7 to 30, 60, 90, 180 and 360 days). At the end of 1996, savings passbooks amounted to US$112 million and fixed-term deposits to US$91 million. Savings constituted 35 percent of the total of savings plus deposits and shares; fixed- term deposits were about 29 percent and shares were 36 percent. The liquidity of the savings accounts and access to checking accounts makes savings flexible enough for withdrawals and liquidity management, particularly for microentrepreneurs and independent workers. The checking account was created at the end of 1996 and was launched initially in Bogota. To develop the checking account, Cupocredito worked with various banks to provide a debit card with access to a nationwide network of ATMs (Automated Teller Machines). INTEREST RATES Cupocredito pays competitive market rates on savings and deposits. As of August 1997, the effective interest rate on savings was 21 percent (nominal 19.52 percent) (table 6). Assuming an inflation rate of 18.60 percent (for the last 12 months), the real effective interest rate is 2.02 percent for savings passbooks and 1.18 percent and 2.2 percent for Cupodiario accounts (less than US$100 and_ between US$100 and US$1,000, respectively). Cupocredito pays 20 to 21 percent effective interest on smaller minimum balances than those required by other financial institutions. In general, Colombian commercial banks do not pay interest on savings accounts under US$500. According to Cupocredito representatives, the higher interest rates paid on savings by the credit union and the ability to borrow based on average savings continue to be advantageous factors for credit unions. And the savings balances maintained by credit union members can be withdrawn or leveraged when necessary. The interest rate paid on fixed-term certificates of deposits (CDs) varies with the amount, the term and the calculation modality. For example,a short-term CD(90 days) ofless than US$5,000, pays an annual nominal rate of20.80 percent, i.e., (effective rate of23.58 percent and real rate 4.20 percent). SUSTAINABLE BANKING wrm THE POOR 12 Table 5. Types of Shares, Savings and Liquidity Management Services at Cupocredito Shares Minimum required: US$30 to become a member. A share balance ofUS$170 (equivalent to a monthly minimum wage) required within a year after joining the credit union. Withdrawals: A balance of US$! 70 should be maintained. Other amounts can be withdrawn/ transferred to savings. Leverage: Up to 10 times (for housing loans) and up to 5 to 8 times for other types of credit (e.g., microenteprise). Return Paid: Equivalent to inflation plus a dividend determined every year based on net profit. Savings Savings Passbook: Account Minimun to open account: US$30. Minimum Deposits:None Minimum Withdrawls: $5, any time Leverage: Up to 5 average Interest Paid: Daily (market rates) on average monthly balance and compounded quarterly. Minimum US$30 average quarterly balance required to receive interest. Minimum US$10 daily balance to maintain account active. Other benefits: Life insurance. Cupodiario: Minimum to open account: US$50 Minimum· Deposits:None Minimum Withdrawls: $5, anytime Interest paid: Daily ( market rates) and capitalized on daily balance. Minimum US$50 daily balance to receive interest. Minimum US$10 daily to maintain account active. Other benefits: Life insurance. Cupoahorrito and Cupoahorro Juvenil: Minimum to open account: US$20 Minimum Deposits:None Minimum Withdrawls: $5, any time Interest paid: Market rate. Minimum US$20 quarterly balance to receive interest. Other benefits: Life insurance. Special accounts for kids under 14 years of age (cupoahorrito) and for 14 to 18 years old (cupoahorro juvenil). Account holders participate in periodic raffles. Accourit provides a life insurance in case of death of children's parents (or adult account holder) to cover education expenses. Cupocheque: Savings with a checkbook and a special transitory credit to cover overdrafts. Minimum US$700 to open the account, however, no minimum balance required. Pays 12% interest on average monthly balance. Savings can be withdrawn with a d~bit card that can be used in nationwide ATM networks Fixed Fixed Term Deposit Certificates: Certificates are available from a minimum of 7 days to Term 90 days or more. Certificates can be used to leverage loans and as guarantee. The Deposits minimum amount required to open an account is US$500. Source: Cupocredito files SUSTAINABLE BANKING WITH THE POOR 13 Cupocredito's rate is higher than the· average market annual effective interest rates of22.95 percent for a similar CD (table 6). The market for CDs requires a higher income group since the minimum balance to open a certificate is US$500. Table 6. Effective Annual Interest Rates Paid on Savings and Deposits (August 1997) Shares Passbook Cupodiario Fixed Short-term Savings Deposits Cupocredito 18% 21% 20%(<US$100) 23.58% 21% ($100- (<US$5,000) $1,000) Avg. mkt rate n.a. 3%-9% 8%-14% 22.95% Source: Data from the credit union files PERFORMANCE: MEASURING OUTREACH AND FINANCIAL SUSTAINABILITY The performance of Cupocredito affects the savings and capital shares of 486,272 Colombians. These Cupocredito members' savings and shares average less than US$500 per person, amounting to US$203.6 million in savings and US$113.3 million in capital shares In 1996, compared to 31 commercial banks in the system, Cupocredito ranked 23rd in asset size, 20th in loans outstanding, 23rd in deposits, and 11th in capital. It is striking that Cupocredito has been able to reach this level of operations by providing financial services to populations which are not traditionally served by formal financial institutions. LOAN PERFORMANCE: OUTREACH · SCALE During 1996, Cupocredito disbursed 121,816 individual loans totaling US$302 million (table 7). One-fourth of the savers are also borrowers. Cupocredito compares well with other providers ofmicrofinance in Colombia. Recent data for a group of 10 NGOs in micro-credit shows that they provided services to 31,500 microentrepreneurs in June 1996 (Marulanda, 1997). In addition, the Colombian group of five Women's World Banking affiliates disbursed 10,779 loans. The total number ofloans made by these 15 NGOs was about 42,279 for 1996. Cupocredito alone disbursed some 41,034 loans related to business, averaging US$3,788 (table 7). SUSTAINABLE BANKING WITH THE POOR 14  Table 7. Scope of Outreach: Activities Financed- Loans Disbursed in 1996 . Loan Purpose Number Volume Distribution Average US$ Million Loan US$ Number Volume Business 41,034 $155.4 34% 51% $3,788 Housing 27,249 $82.9 22% 28% $3,045 Consumption 42,631 $48.9 35% 16% $1,147 Education & Health 6,934 $6.8 6% 2% $878 Other 3,968 $9.3 3% 3% $2,348 Total 121,816 $302.7 100% 100% $2,485 Based on the estimated number ofmicroenterprises in Colombia (1,247,000), Cupocredito's business related loans would have covered about 3 percent of the market. This share is equivalent to that reached by the group of 15 NGOs. Sixty percent of these NGOs lend primarily short-term (6 to 12 months). Cupocredito has longer-term loans (one to five years). Moreover, a percentage ofloans in non-business categories (e.g. housing improvement) are also made to microentrepreneurs and self-employed clients. DEPTH Cupocredito serves medium- and low-income populations. Cupocredito does not target exclusively the very poor. Neighborhoods in urban areas and rural communities contain both poor and very po.o r populations; Cupocredito provides services to both. The loan portfolio is . diversified in terms of loan size. Ninety-one percent of the total number of loans outstanding are under US$5,000. Fifty-four percent of the number ofloans are under US$1,500. In 1996, • Cupocredito disbursed 21,593 loans under US$500 and a total of almost 86,000 loans under US$1,500 (54 percent). Despite its growth, Cupocredito has maintained low average loan sizes. The average loan size for the portfolio under US$5,000 is US$ l ,444, equivalent to 83 percent of the GDP per capita. • GENDER COVERAGE An IDB study of Cupocredito (Bonilla et al. 1994; Almeyda 1996, p.32), estimated that about 44 percent of the number of outstanding loans under US$5,000 (1992 data) were made to women. The participation of female borrowers was higher than their overall membership of 30 percent. Assuming a similar level of participation of women in the loan portfolio, for loans under US$5,000, would yield approximately 37,800 female borrowers for 1996. The average of women borrowers as a percentage of total borrowers is similar to that of specialized microenterprise NGOs. SUSTAINABLE BANKING WITH THE POOR 15 SHARES AND SAVINGS OUTREACH SCALE Cupocredito fuels local economic development by mobilizing small savings, which in tum are invested in the fonn ofloans in the area. At the end of 1996, there were 447,370 savings passbooks, for a total ofUS$112 million, and fixed-tenn deposits totaling US$91 million. Members' shares amounted to US$113 million. At the current level of average balances, Cupocredito is the largest of the top 84 Colombian credit unions with 23 percent of the total combined membership, 16 percent of the total savings and deposits, and 22 percent of the total capital shares. Savings volume increased 29 percent in 1996, with a 7 percent real growth. Fixed-term deposits grew 54 percent (28 percent real growth). The average annual membership growth for the past three years was 15 percent. Overall, the coverage and annual growth performance of Cupocredito is good, given that Colombia's economic growth has slowed down in the past two years, and inflation continues to be around 20 percent. • DEPTH Cupocredito illustrates its commitment to lower-income groups in the minimum balances required of members to open savings accounts. Only US$30 is required to become a member. In early 1997, the average savings passbook balance was US$ I 86. The average savings balance as a percentage of GDP/capita has remained the same for the past three years, about 13 percent. The annual real growth of the average savings account increased 17 percent in 1996, however it was actually negative in 1996 with a -9 percent real annual change. Cupocredito has maintained its growth with members' savings. For the past few years, the average balance for shares as a percentage of GDP/capita has also continued to be about 13 percent. OUTREACH: GENDER COVERAGE The most recent data (1993) indicate that about 30 percent of Cupocredito members are women. This percentage may have increased over the past two years as a result of Cupocredito' s expansion. Differences in savings behavior among male and female members of Colombian credit unions have not been studied; Cupocredito could benefit from such research. In focus group research on Bogota's microentrepreneurs and their savings, women tended to be more concerned than men about proximity to the savings institution and its hours of operation (Bonlla et al, 1994). Some microentrepreneurs said that they did not have much information about savings programs because either they were not interested or--if they had tried to get information from formal financial institutions--the attention they received was very poor, and they felt discouraged. Cooperatives such as Cupocredito provide a non-threatening environment for both women and men in low-income groups. SUSTAINABLE BANKING WITH THE POOR · 16  MEASURING SUSTAINABILITY: FINANCIAL AND OPERATIONAL PERFORMANCE This section evaluates the performance of Cupocredito using the PEARLS system which was developed by WOCCU, and measures Protection, Effective financial structure, Asset . quality, Rates of return and cost, Liquidity and Signs of growth. PEARLS facilitates the evaluation of micro credit providers among countries. PEARLS is designed to identify key issues in strengthening credit union performance (see Appendix A). Cupocredito PEARLS measures (PERLAS in Spanish) are shown in table 8. REPAYMENT PERFORMANCE AND PROTECTION As of December 31, 1996, Cupocredito' s total portfolio at risk was 21.8 percent (including the total outstanding balance ofloans over 30 days past due). Cupocredito's delinquency levels can be explained by both internal and external factors. The internal factors relate to Cupocredito's monitoring ofloan delinquency. In the past two years Cupocredito has embarked on a completely new strategy to improve loan payment collection. A new vice presidency of "Credit and Collections" was created in 1996, reflecting a new institutional commitment to maintain a well-performing portfolio. Cupocredito has considered portfolio performance an important benchmark for branch managers' evaluations; however, there has been no system geared toward improving their monitoring. For example, in the past managers would only begin to follow up on delinquent loan after 90 days. Now follow- up starts on day one of delinquency, Management information systems have also improved. Loan analysis includes checks with Datacredito, a service which provides information about prospective borrowers' credit histories. External factors affecting loan delinquency include the general economic health of Colombia in the past two years. National production grew by only 3.2 percent, inflation increased by two points compared to 1995, and unemployment rates reached a national level of 11.5 percent . In some cities it was as high as 17 percent (Cartera de Credito, 1997) The loan delinquency rate for financial corporations increased by 96 percent during 1996. Cupocredito's membership includes some of the populations most affected by this economic slowdown. SUSTAINABLE BANKING WITH THE POOR 17  Table 8. Cupocredito Credit Union PEARLS (PERLAS) INDICATOR 12/31/94 12/31/95 12/31/96 P=PROTECCION (PROTECTION) Provisions / Loans Delinquent> 12 months 75% -- I 80% Net Provisions / Loans Delinquent< 12 Months 77% 14% E=ESTRUCTURA FINANCIERA (FINANCIAL STRUCTURE) Loans / Total Assets 76% 75% 78% Liquid Instruments / Total Assets 10% 10% 7% Financial.Investments/ Total Assets 2.5% 1.6% 1.4% Deposits / Total Assets 49% 50% 46% External Credit/ Total Assets 4% 7% ' 16% Shares / Total Assets 36% 30% 25% Institutional Capital / Total Assets 4.5% R=RENDIMIENTOS Y COSTOS (EARNINGS AND COSTS) Return on Loans /Average Loans 31% 31.5% 32% Returns on Liquidity / Average Liquid Instruments 29% 34% 41% Return on Financial Investments / Avg. Financial 13% 17% 4.5% Investments Cost of Deposits/ Average Deposits 14% 14% 15% Cost of Shares / Average Shares 15% 12% 16% Cost of External Credit / Average External Credit 23% 26% 22.5% Gross Margin /Average Assets 11% 14.5% 14% Operating and Administrative Expenses / Average 7% 8% 7.6% Assets Provisions for Losses / Average Assets 0.7% o.~% 0.9% Net Income /Average Assets 0.6% 0.6% 0.6% Net Income / Average Equity 2.4% 2.5% 2.5% Interest Paid/Interest Earned 46% 48% 45% Financial Expenses/Financial Revenues 61% 49% 52% L=LIQUIDEZ (LIQUIDITY) (Liquid Instruments - lmmedjate Obligations) / Deposits 17% 18% 9% Liquidity Reserves I Deposits 10% 10% 10% Cash & Non-earning Liquidity/ Total Assets 1% 1% 1% A= ACTIVOS IMPRODUCTIVOS (NON-PRODUCTIVE ASSETS) Delinquency/ Total Loans (Includes unpaid balance) 2.4% 2.4% Non-Productive Assets/ Total Assets 12% 15% 15% S=SENALES EXPANSIVAS (GROWTH) (based on average outstanding balances) Total Assets - Annual Growth 47% 51% 52% Real Annual Growth 19% 25% 27% Loans - Annual Growth 42% 49% 60% Real Annual Growth 15% 23.5% 33% Savings - Annual Growth 56% 53% 39% Real Annual Growth 26% 26% 16% Shares - Annual Growth 40% 24% 30% Real Annual Growth 13% 2.5% 8% Institutional Capital 2 5% Members- Annual Growth 28% 15% 19% 1 Notes: In order to issue bonds in 1995, debts over 181 days were written off. 2 The General Assembly of 1996 required an "irreducible share capital," currently at a level of 5% of total assets. SUSTAINABLE BANKING WITH THE POOR 18  At the end of 1996, Cupocredito's provisions covered 80 percent of the loans delinquent over 12 months and covered 14 percent of loans delinquent less than 12 months (table 8). Provisions plus shares (held as guarantee) covered 28.5 percent of the total delinquent loan portfolio. These loan provisions are above DANCOOP's requirements. However, the level of provisions is lower than PERLAS' recommended level of 100 percent provisions for loans delinquent over 12 months and 35 percent for loans less than 12 months overdue. Cupocredito continues to rely mainly on personal guarantees (e.g., credit and savings history of borrower, co-signers). Eighty percent of the outstanding loan portfolio of Cupocredito is personally guaranteed. The remaining 28 percent has real guarantees. A large percentage of the real guarantee portfolio uses the goods to be financed as the guarantee (e.g., housing, vehicles, and machinery). FINANCIAL STRUCTURE Savings deposits and share capital constitute Cupocredito' s primary sources of lending funds. The ratio of savings deposits plus share capital to total assets has been maintained at an average of 81 percent over .the past four years. For the past five years, loans to total assets continue to be maintained at the level of 78 percent. The loan portfolio is 80 percent short term (under five years) and 20 percent long term (over five years). The recent trend has been an increase in long-term loans; five years ago the portfolio was all short-term. Over the past few years, Cupocredito has changed its basic financial structure. While in 1992 shares to total assets was 41 percent, this ratio was 25 percent in 1996 (table 8). Cupocredito has shifted to savings and fixed-term deposit instruments from a traditional credit union approach that emphasizes shares as the main savings mobilization and source of funds instrument. The current ratio of savings deposits to total assets is 45.5 percent. One of the major changes in the financial structure is the increase in external debt from 2.8 percent in 1992 to 16 percent in 1996. Twenty-five percent of the long-term external debt is with commercial financial institutions and 75 percent is in bonds. Cupocredito currently has an outstanding long-term debt ofUS$16 million with commercial banks and with IFI (Instituto de Fomento Industrial). Eighty-one percent of this debt (US$13 million) is with IFI's discount line for project financing, including small and microenterprise development. Cupocredito uses this line mainly for small enterprises (average loan size: US$11,000), which represents a higher risk for the credit union. Cupocredito issued bonds in 1995 for US$16 million. This first emission was .sold to three banks and two financial corporations. During 1996, Cupocredito had a second bond issues for US$30 million. One of the _main reasons for using the local capital market was to access longer-term debt for housing finance (Cupocredito now offers 10 year housing loans). Fixed-term deposits over 360 days constitute less than 1 percent of the total fixed-term deposits. Bonds are to be paid in six years (starting with SUSTAINABLE BANKING wrrn THE POOR • 19 the fourth year at 25 percent, the fifth year at another 25 percent, and the sixth year at the remaining 50 percent). The institutional capital is 4.5 percent of assets (table 8). According to the PERLAS systems, the minimum required is 10 percent. At the end of 1996, Cupocredito's ratio of total equity capital was US$145 million. Members' share capital constituted 78_ percent. EARNINGS AND COSTS Cupocredito is a self-sustaining institution. Income from operations covers financial and operational expenses. The ratio of net income to average total assets was 0.54 percent in real terms, and to average total equity, 2.52 percent in real terms (table 8). The credit union grew at a higher average than other financial institutions and did not show a loss for the period. For other financial institutions, the average return on total assets for banks was 2.5 percent; for housing and savings corporations 1.52 percent; for financial corporation 1.79; and 0.27 percent for financial companies (table 9). • Cupocredito showed almost twice the return of financial companies (which are also more comparable to credit unions than other financial institutions). With respect to returns on equity, results for 1996 were mixed. For example, banks had a return on equity of 10.63 percent; Housing and Savings Corporations (CAVs) had the highest, 19.36 percent; financial corporation, 7.27 percent; and commercial financial companies, 1.89 percent. Thus, Cupocredito's return on equity was higher only to that of commercial financial companies (table 9). Table 9. Returns-Comparison of Cupocredito with other Financial Institutions (as of 12/31/96) Indicator Cupocredito 84 CUs Banks CAVs inancial Financial orporations Ccial Cies Return on 0.54% 0.64% 1.66% 1.52% 1.79% 0.27% Assets Return on 2.52% 2.40% 10.60% 19.30% 7.27% 1.89% Equity Source: Estimated based on data from "El mundo de las Cooperativas," Podery Dinero, Bogota, July 1997, p. 96-104. Cupocredito is a profitable financial intermediary. The yield of the loan portfolio (interest from loans to average outstanding loans net of provisions) was 32 percent, and has been at this level for the past three years (table 8). The ratio of interest paid on loans to interest earned for the period was 45 percent, as it has been for the past four years. The ratio of total financial expenses to total financial revenue was 51.6 percent for 1996; the average for the past four years has been 56 percent. The ratio of total administrative expenses to total assets was 7.6 percent (table 8). According to the PERLAS system, the ratio of total administrative expenses to total assets should be between 3 to 5 percent, but Cupocredito' s ratio was 16 percent. • The average number of loans disbursed during 1996 per loan officer (including branch managers) was 287, and the average volume was US$713,515. The outstanding loan portfolio per loan officer for the same period was 377 and the volume was SUSTAINABLE BANKING Willi THE POOR 20 US$83 l,782. Given that Cupocredito uses individual loan methodology, the average level of operations per loan officer is high (compared to NGOs, for instance). The credit union clearly needs to improve the quality of its loan analysis, in order to reduce delinquency. LIQUIDITY Cupocredito maintains a liquidity fund of 10 percent of savings and deposits, which was a DANCOOP requirement. The liquidity fund is maintained with the cooperative banks and is paid market-rate interest. The ratio of liquid instruments minus immediate :financial obligations to total deposits is 9 percent. The average for the past •four years has been 14 percent. ASSET QUALITY The delinquency ratio is the key measurement in determining the institutional strength of a :financial intermediary. Cupocredito's ratio should have been at least close to the average market delinquency rates of7 percent in 1996. The current ratio of21 percent diminishes all the strong factors that support the institutional long-term viability of Cupocredito. The new preventive measures that apply to loan collections should help to improve the portfolio's performance. The ideal delinquency rate recommended by PERLAS is under 5 percent. Another key measure for asset quality is the ratio ofnon-productive assets to average assets. This ratio has increased to 15 percent. According to the PERLAS monitoring system, the recommended level is 5 percent. The high ratio of 15 percent is the result in part of a growth (adjusted for inflation) in fixed assets by 50 percent between 1995 and 1996. The accounting reporting obligation to adjust fixed assets to inflation explains 37 percent of the change in fixed assets and the remaining 63 percent change is as a result of increases in property. Table 10. Measuring Operational Self-Sufficiency (Million US$} 12/31/94 12/31/95 12/31/96 Total Financial Income $47 $70 $109 Financial Expense 1 $28 $34 $56 Administrative Expenses2 $18 $31 $45 Total Expenses $46 $65 $101 Net Operational Income $1 $5 $8 Operational Self-Sufficiency 102% 107% 108% (Admin. Expense +Prov.)/ Avg Loans Outstanding 17% 16% · Notes: Includes dividends on shares 2 Includes bad debt provision SUSTAINABLE BANKING WITH THE POOR 21  GROWTH Cupocredito' s •operations have grown at very satisfactory levels given the Colombian financial environment of the past two years. Membership increased 20 percent during 1996. The credit union has maintained a nominal growth in total assets over the past three years of SO percent with an average real growth of23 percent (table 8). Outstanding loans increased 60 percent (nominal) and 33 percent (real) during 1996. The average real growth of the loan portfolio has been 24 percent for the past three years. Savings showed an increase of39 percent (nominal) and 16 percent (real) during 1996. Savings deposits' annual average real growth has been 22.5 percent for the past three years. Shares, which continue to have a lower participation as a percentage of total assets, had a nominal growth of 30 percent and 8 percent real growth in 1996. CONCLUSIONS AND RECOMMENDATIONS: CHALLENGES FOR CREDIT UNIONS IN MICROFINANCE Cupocredito' s main strengths as a microfinance provider are reflected in the average size of its loans, its flexible savings instruments, the type of membership it serves, its level of self-sustainability, and spreads sufficient to cover all the financial and operational expenses. Cupocredito's growth over the past four decades demonstrates that: • Self-sustainable financial intermediaries can emerge from local capital and savings by lower-income clienteles: Cupocredito's equity amounts to US$145 million and savings deposits to US$203 million. These local resources are currently financing a loan portfolio ofUS$352 million. Revenues from operations cover all financial and operational costs. Cupocredito's branches, located in urban and rural areas, are serving 486,272 members. • The nature of financial market provides a framework for the development and strengthening of credit unions: Colombia's reforms of the financial system have contributed to a reduction of reserve requirements, interest rate controls, targeted- • credit programs as well as entry of foreign banks and investors. Cupocredito has benefited from the dynamism and competitiveness that these reforms have caused. • A regulatory framework that accomodates the nature of the credit unions, is key to their evolution: Colombian credit unions have promoted legislation that responds to their evolving needs. Revisions of the cooperative law over the years have allowed Cupocredito to change and expand its financial products. Regulation has also contributed to leveling the playing field shared by credit unions and other financial intermediaries. Adequate control, supervision, and regulation are crucial to credit unions' ability to offer fmancial services on a viable basis SUSTAINABLE BANKING WITH THE POOR 22 • Self-discipline has been a necessary condition (although not sufficient) in developing performance standards. DANCOOP, the agency responsible for Cupocredito's supervision, has lacked the technical capability and resources to do its work effectively. Cupocredito had to develop its own standards of govei:nance, managerial and operational practices in order to grow without any outside external supervision. • Access to appropriate technology has accelerated credit union growth. The availability of local expertise and technology, such as on-line database systems, was a key factor in Cupocredito' s expansion. Access to technology has resulted from the . dynamics of the financial sector in Colombia. Technology is crucial to the efficiency of the operations and the delivery of services to members. This, in turn, contributes to keeping costs low and maintaining a competitive position in the market. • Focus on community outreach contributes to self-sustainability. Cupocredito serves communities and neighborhoods which encompass a broad membership base of wage workers, self-employed, and microentrepreneurs. A mixed memberspip contributes to risk diversification. • Credit unions increase competition and expand local microfinancial markets: The microenterprise sector is heterogeneous: there are very poor, subsistence entrepreneurs as well as advanced entrepreneurs with growing businesses. When the local supply of microfinancial services is broad and competitive, microentrepreneurs have choices: they may go to a credit union or go to an NGO. ISSUES·OF CONCERN Cupocredito's challenge is to maintain manageable growth within the same clientele that it serves: low-~ncome wage earners, the self-employed, and microentrepreneurs. It should continue playing a role in rural and urban communities. And it must correct weaknesses in the system that contribute to loan delinquency. INSTITUTIONAL lsSUES: • Current levels ofloan delinquency: Cupocredito's total portfolio at risk is 21.8 percent for loans more than 30 days past due. Although the level of debt more than one year old is 2.6 percent, loan delinquency is worrisome. The loan delinquency reflects weaknesses in loan processing and monitoring. Cupocredito has responded to high delinquency rates by developing a new "collections strategy." Although credit unions do evaluate debt-payment capacity before approving a loan, proper analysis of the cash flows of a business is rare. Time will reveal whether Cupocredito is able to improve in this critical area. SUSTAINABLE BANKING WITH THE POOR 23 • External debt vs. savings mobilization: do savings and credit services balance? Cupocredito used the local capital markets to issue bonds to increase sources of funds for longer-term housing loans. Low-income housing fmance is a more developed market in Colombia in comparison to other markets such as microenterprise credit. Cupocredito participates in social housing programs--contributing to the privatization and market orientation of subsidized housing construction. Twenty to 30 percent of the loan portfolio of Cupocredito has been in housing (improvement and acquisitions). Cupocredito can access funds from NGOs (such as Fedevivienda) for housing loans. Should Cupocredito provide long-term fmancing for housing? Are there other institutions that have a better competitive advantage in this market? Is Cupocredito targeting areas where there are no other institutions offering this type of finance (e.g., rural areas)? Maintaining an appropriate balance between external financial obligations (e.g., bonds) and savings is key for Cupocredito to continue responding to its membership. • Relations with government institutions: Can the experience of the past two years be avoided? Cupocredito and other credit unions have leveraged their funds with government to finance low-income housing, microenterprises and infrastructural projects. During the past two years there have been cases of credit unions being used by government officials as channels for special funds (e.g., housing programs) for personal businesses. In one unfortunate case, a rural credit union experienced an intervention by Dancoop because of one large loan that the credit union should have never granted. The management of the credit union jeopardized ~e small savings of hundreds of savers. The lack of governmental supervision became an issue of major concern because it could damage the good work of many other credit unions that are playing a key role in many co-financed programs. Cupocredito has management controls, including internal and external auditors, to maintain a close eye on its operations. The rapid growth of the past few years and the nationwide coverage of credit union operations make closer supervision and prudential regulation even more important for the continued strengthening of credit unions in Colombia. CREDIT UNION SYSTEM • Level of integration -Too much dispersion? For some in the local microfmance sector, Colombian credit unions suffer from "dispersed institutional action." Credit unions and cooperative banks have expanded horizontally without much system integration. This dispersion is costly in terms of promotion, technology, and infrastructure. Credit Unions lack a "system" strategy in terms of market penetration and network building. • Credit Union Industry Standards and Self-Discipline: Colombian credit unions lack a basic system of financial and statistical information to develop good standards of performance. Currently, the Confederation (a national body which includes representatives of credit unions and other cooperatives) publishes some general statistics. These are not sufficient to perform a financial or operational analysis, however, since no delinquency rates are available. People now hope that the SUSTAINABLE BANKING WITH THE POOR 24  Superintendency of Banks will generate more reliable statistics. Credit unions should take the initiative to conduct such analysis. Other countries are considering the use of a standard system such as PERLAS for affiliated credit unions. Industry standards need to be proposed according to levels of assets (given the significant number of small credit unions). • Supervision and Prudential Regulation: Continuous collaboration with the official supervisory institutions is needed to develop adequate credit union supervision and prudential regulation. Maintaining the flexibility to tailor financial instruments (credit, savings and others) to the socioeconomic profile of low-income members is crucial to credit unions' success, and necessary to reach the large population currently outside the frontier of formal finance. • Credit Unions as a Sector: Economic and Social Challenges: The more advanced credit unions and cooperative banks have a unique opportunity to tap into a large market: the poor. The cooperative financial system (counting only the top 84 credit unions) has 22 percent of the branches in the entire Colombian financial system. Credit unions need to face the challenge of competing with other financial institutions. Also, credit unions are working separately (and perhaps duplicating efforts) in different areas of the country. For example, Cupocredito has worked in the Llanos (Oriental Region) and more recently in the Caribbean coastal departments, , while Cooperativa Solidarios has opened branches on the Pacific Coast. These are some of the regions with major socioeconomic problems that could benefit from a joint credit union (cooperative) development strategy. At a minimum, efforts to exchange information and experiences about operations in these regions could be useful in developing regional programs. • Leveraging Social Intermediation Capabilities ofNGOs: Credit unions are active in financial intermediation for lower-income groups--traditionally those not reached by formal financial institutions. Credit unions' programs with NGO have proven · beneficial to both the institutions and the clients. NGOs have a comparative advantage in reaching community groups, and. conducting projects on such themes as housing, business development, and technical assistance to microentrepreneurs. Credit unions do not have these capabilities (nor are they supposed to be involved in these activities). However, their members can benefit from the advisory and social intermediation services ofNGOs (and other institutions) in order to improve their members' (and potential members') projects which, in turn, can be presented to a credit union for financing. Given the present social conditions-many areas have suffered from decades of violence--joint efforts to strengthen social and financial intermediation by credit unions and NGOs need to be developed. SUSTAINABLE BANKING WITH THE POOR ' , 25  INTERNATIONAL DONORS AND MICROFINANCE PRACTITIONERS • Recognize the diversity of microfinance providers: Worldwide, credit unions expand _the supply of microfinancial services to low-income people. Credit union experiences derived from both donor-supported programs and those that have developed without such support (e.g., Colombian credit unions) can provide useful lessons (the good, the bad and "mixed" experiences) about institution-building. Regulation, governance, system integration, savings instruments, credit products, operational viability, delinquency control, internal controls, membership coverage-- these are issues, among others, that have been and continue to be tackled in a variety of ways by credit unions in both rural and urban areas. • Highlight the role of credit unions as mobilizers of savings and local capital: International donors tend to underestimate or fail to acknowledge the capacity or potential of credit unions as mobilizers of small savings. Even more troublesome, some credit union leaders and managers may also see credit unions as a channel for international funds, rather than ~ institutions that should be built on local savings. Instead of using credit unions as a simple conduit for credit, donors can provide technical assistance to improve credit unions' capabilities to mobilize savings and manage solid loan portfolios. • Support technical assistance and institutional development rather than individual subsidies: Credit unions such as Cupocredito show that it is possible to operate without subsidies. Colombian members are willing to pay for financial services if they know they have the possibility of obtaining credit in a long-term relationship that ensures their access to financial services. At the institutional level, the more advanced credit unions can benefit greatly from access to technical assistance (in areas such as technology, governance, prudential regulations, industry standards, microenterprise credit, etc.). Credit unions' own funds for technical assistance can be leveraged with international funds. International exchanges between credit unions and other microfinance institutions are useful in expanding the access of financial services to lower-income groups in Latin America and other regions. • Foster the appropriate legal, regulatory, and supervisory framework for credit unions: Regional seminars, training, and exchange among supervisory officials can enhance the development of a regulatory framework for credit unions, just as the formal financial sector has benefited from international exchanges and research to design, implement, and monitor financial reforms. Governmental officials responsible for developing credit union supervision mechanisms could benefit from technical support and educational programs in this area. Activities involving both credit unions and supervisory agencies can help to promote adequate practices.on both sides. SUSTAINABLE BANKING WITH THE POOR 26 • Research on gender issues and their effect on microfinance services: As savers and borrowers, women rank high in their participation in credit unions. Women use these services for household as well as business strategies. Credit unions can be a useful source of information about women's and men's financial practices, particularly among the lower-income groups. However, the high participation of women as savers and borrowers of credit unions is not reflected in their governance structure. Currently, the board of directors of credit unions has no women members. What institutional practices and/or cultural barriers discourage women's participation at the leadership level? What type of actions could help develop female leaders? Could women leaders improve and expand the access of lower-income women to financial services offered by credit unions? These questions need further research. • Tap local capital markets: Cupocredito issued US$46 million in bonds in the local capital market. What can other credit unions and microfinancial institutions learn from this? Research on tapping local capital markets can be :fruitful for other credit unions, as well as microfinancial institutions. Support for research in this area could help expand the current pool of sources for finance. SUSTAINABLE BANKING WITH THE POOR 27  BIBLIOGRAPHY Arboleda, Ubemel, "Una misi6n posible: politicas de apoyo a la microempresa en Colombia", Bogota: Departamento Nacional de Planeaci6n. 1997 Bonilla, Elssy, B. Marulanda, P. Lizarazo, P. Rodriguez. "Servicios financieros para mujeres microempresarias: El caso colombiano." Prepared for the IDB. Bogota: Universidad de los Andes. 1994 Magdalena Velasquez Toro, Las Mujeres en la Historia de Colombia, Torno I, editorial Norma, 1995. Marulanda, Beatriz, "Servicios financieros a la micro y pequefia empresa--Diagn6sticos, aruilisis y formulaci6n de estrategias." Washington, D.C.: The Inter-American Development Bank. March 1997. Otero, Maria and Elizabeth Rhyne, The New World of Microenterprise Finance-- Building Healthy Financial Institutions for the Poor, Kumarian Press, 1994. Richardson, David, "PEARLS--Financial Stabilization and Capitalization," WOCCU, Madison, Wisconsin, August 1997. SUSTAINABLE BANKING WITH THE POOR 28  APPENDIX A. PEARLS INDICATORS OF PERFORMANCE de to adapt The PEARLS system was tailored to the specific needs of credit unions. An early attempt was ma_ the U.S. CAMEL ranking system to credit unions, but too many modifications were needed. In particular, the CAMEL system possessed two major deficiencies which limited its effectiveness. I. The CAMEL system did not evaluate the financial -structure of the balance sheet This was a critical area of concern in the incorporation of a new credit union model required a major restructuring of assets, liabilities and capital. Balance sheet structure had a direct impact on efficiency and profitability. The new model targeted both of these areas as critically important for effective and sustainable credit union operations. 2. CAMEL did not consider growth rates. Growth of total assets was a key strategy used to address the problems which accompanied several monetary devaluations and runaway inflation. In a relatively hostile macro-economic environment, the credit unions had to sustain aggressive growth if they were to preserve the value of their assets. CAMEL was created as a supervisory tool, not a management tool. The main concern of CAMEL ratios was to protect the solvency of the institution and the safety of member deposits. It was not designed as a tool for the analysis of key areas of credit union operations. THE INDICATORS: P= Protection: Protection of assets is measured by comparing the adequacy of the provisions for loan losses against the amount of delinquent loans. Protection is measured by whether a credit union had sufficient provisions to cover 100% of all loans delinquent for more than 12 months, and 3 5% of all loans delinquent for 1-12 months. E= Effective Financial Structµre: The PERLAS system measures the credit union financial structure of loans, investments, deposits, shares and institutional capital. The following ideal targets are promoted: 95% productive assets composed ofloans (70-80%), and liquid investments (10-20%) 5% unproductive assets composed of primarily fixed assets (land, buildings, equipment etc.) 70-80% member deposit savings 10-20% member share capital 10% institutional capital (undivided reserves) Under the new capitalization system, member shares are de-emphasized and replaced with institutional capital. This capital has three purposes: finance non-productive assets (e.g., buildings), improve earnings and absorve losses. R =Rates of Return and Costs: The PEARLS system disaggregates components of net earnings to help management calculate investment yields and evaluate total operating expenses. This assists management in determining which investments were the most profitable. By comparing financial structure with yields, the credit union measures how effectively it places its productive resources into investments which produce the highest yield. By comparing finap.cial structure with yields, it is possible to determine how effectively the credit union is able to place its productive resources into investments which produced the highest yield. Operational costs are broken down into three main areas: (1) Financial intermediation costs paid on deposit savings, share savings, and external loans; unlike commercial banks who try to minimize financial costs, credit unions should try to pay as high a rate as possible without jeopardizing the stability of the institution. (2) Administrative costs target recommended by the PERLAS system is to maintain administrative costs between 3-5% of average total assets. (3) Unrecoverable loan costs evaluated by PERLAS separates the costs SUSTAINABLE BANKING WITH THE POOR 29 (2) Administrative costs target recommended by the PERLAS system is to maintain administrative costs between 3-5% of average total assets. (3) Unrecoverable loan costs evaluated by PERLAS separates the costs of creating provisions for loan losses from other administrative costs and directly links it to credit analysis and effective loan collection techniques. By isolating this expense from the other administrative costs, it is possible to get a much clearer picture of the effect of weak credit administration on the credit union. By segregating income and expenses into the previously mentioned areas, the PERLAS ratios can accurately pinpoint the reasons why a credit union is not producing sufficient net income. L = Liquidity: The PERLAS system analyzes liquidity in terms of the percentage of savings deposits invested as liquid assets. The recommended amount is equivalent to 10% of all savings deposits. A = Assets Quality: The PEARLS system classified all credit union assets as either productive or non- productive. A non-productive asset was one which did not generate income. The delinquency ratio is the most important measurement of institutional weakness. Th,e ideal goal is to maintain the delinquency rate below 5% oftotal loans outstanding. S = Signs of Growth: PEARLS measures growth in assets, loans, deposits, shares, institutional capital and members. By comparing the growth in key areas, it is possible to detect changes in the balance sheet structure which could have a positive or negative impact on earnings. The ideal goal for all credit unions is to achieve real positive growth (i.e., net growth after subtracting for inflation) each year. Source: David Richardson. 1994. PEARLS--Financial Stabilization, Monitoring and Evaluation. Madison, WI: WOCCU. SUSTAINABLE BANKING WITH THE POOR 30  Appendix B. List of Case Studies • Benin, FECECAM. Cecile Fruman. June 1997 (Also in French) • Bolivia: Assessing the Performance ofBanco Solidaro. Peter Fidler. August 1998. • Burkina Faso: Le Projet de promotion du petit credit rural -PPPCR Julia Paxton. August 1997. • Colombia: Banco Caja Social. Julia Paxton March 1999. • Colombia: Womens' WorldBanking. Julia Paxton. December 1998. • Costa Rica: FINCA Village Banking. Julia Paxton, March 1998. (Also in Spanish) • Egypt, Alexandria Business Association. Tom Dichter. December 1997. • Guatemala CARE Village Banks Project. Julia Paxton. October 1997. (Also in Spanish) • Indonesia -Bank Rakyat Indonesia (BRJ) Unit Desa 1970-1996. Stephanie Charitonenko, Richard H. Patten, and Jacob Yaron. iune 1998. • Kenya KREP. Stephanie Charitonenko et al. August 1998. • Mali Self-Managed Village Savings and Loans Banks (CVECA - Pays Dogan). Cecile Fruman. May 1998. (Also in French) • Niger Credit Unions (Caisses Populaires d'Epargne et de Credit). Korotoumou Ouattara, Mayada Baydas and Julia Paxton. April 1998. • Philippines - TSP/. Tom Dichter. August 1998. • South Africa: Get Ahead Foundation. Craig Churchill. January 1998. • Thailand BAAC - The Thai Bankfor Agriculture and Agricultural Cooperatives. Tetsutaro Muraki, Leila Webster, and Jacob Yaron. April 1998. • Zimbabwe, Zambuko Trust. Peter Fidler and-Mohini Malhotra. April 1997. SUSTAINABLE BANKING WITH THE POOR 31  Appendix C. List of Discussion/fecbnical Papers In English: • Microfinance Practical Guide for World Bank Staff • A Worldwide Inventory ofMicrofinance Institutions • An Inventory ofMicrofinance 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 ofMicrofinance Institutions in South Asia • An Inventory ofMicrofinance 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 Institutions • 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 In French: • lnventaire mondial des institutions de microfinance • Inventaire 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 a:fricaines In Spanish: • Inventario mundial de institutiones de microfinanzas • Inventario de instituciones microfinancieras en Latinoamerica y el Caribe • La autosuficiencia financiera: una guia basica para programas de credito en America Latina SUSTAINABLE BANKING WITH THE POOR 32