By 1983, years of heavy losses and fiscal cuts had made the vast network of village banks belonging to the state-owned Bank Rakyat Indonesia unsustainable. Shutting the banks down risked serious unemployment and the loss of an important credit delivery mechanism. The government responded by taking a novel approach. It established a new, market-based rural small credit program, the Kredit Umum Pedesan (KUPEDES), aimed at transforming village banks into self-sustaining full-service financial units. The program's principal goals were to provide credit to small borrowers at market rates and to mobilize rural savings. By the time the World Bank's $102 million loan to KUPEDES was approved in 1987, the program was already working with a profit. By loan completion in 1992, the program's assets had grown 21 times over the 1984 level. The village banks were reaching 5 percent of the country's households with loans and 15 percent with deposit services.
A recent OED audit 1/ draws two major lessons from the KUPEDES experience. First, it is possible to create a microlending program that serves the poor and is profitable and self-sustaining. But to succeed the bank units will need to lend at market rates, use their income to finance their operations, and devise appropriate savings instruments to attract depositors. Second, developing a savings instrument for the poor is at least as important as providing them with loans. In Indonesia, SIMPEDES, the village savings program, attracted thousands of depositors. And by mobilizing rural savings, it not only provided the banks with a stable source of funds, it also kept financial savings in rural areas, thus helping continue development in the countryside.
Finally, the audit highlights the main reasons for village banks' success: simplicity of loan design, which enabled the banks to keep costs down; effective management at the unit level backed by close supervision and monitoring by the center; and appropriate staff training and performance incentives. By themselves, however, these factors would not have been enough for the program to succeed. The presence of an existing institutional network of village banks and a thriving economy that had spread to many rural areas were also critical.
The KUPEDES program
Background and program design
Since the 1970s, Bimbingan Massal, or BIMAS, an agricultural diversification program with a rural credit component, had been providing small loans to farmers at below market rates. BIMAS channeled the loans through unit desas (village banks) belonging to Bank Rakyat Indonesia (BRI), a state commercial bank. Through the years the banks evolved into a vast network of 3,300 units employing 14,300 people. But by 1983, heavy losses and fiscal cuts (the latter in response to Indonesia's structural adjustment program) had made the banks unsustainable. When the BIMAS program ended, the government decided to transform the bank units into self-sustaining full service centers to provide financial services to small borrowers at market rates. As a first step, the banks were restructured. Many of the units were moved to market centers, and about a third were downgraded to village service posts to reallocate credit resources. An accounting system was installed in each unit to ensure accountability and performance.
The program, Kredit Umum Pedesan (KUPEDES), began in 1984, on the heels of the financial liberalization reforms introduced in mid-1983. The government (including the central bank) provided about 210 billion rupees ($20 million in 1984 dollars) in seed capital and start-up loans and two years' administrative subsidy to help cover projected early losses.
Simplicity was the critical element in the program's design. Its only restriction was loan size, which automatically limited the loan's appeal to small borrowers. (In 1984, the minimum loan was equivalent to $24 and the maximum to $930.) Otherwise, the banks were free to offer loans to any creditworthy borrower in any sector and for whatever purpose. Instead of charging a yearly rate, the banks followed rural lending traditions, charging a flat lending rate of 1.5 percent a month. This method translated into a 33 percent yearly interest rate, as opposed to 12 percent under the old subsidized program. In addition, a refundable 0.5 percent monthly penalty was added as incentive for early repayment. In 1984, the program also introduced the Village Savings Program, or SIMPEDES, to encourage rural savings and to meet the deposit needs of clients.
By 1986, the bank units had fully used the seed capital and by year's end were turning a profit. KUPEDES loans had grown three times and deposits had expanded fourfold. The initial fear that the high interest rates would make the program unacceptable to small borrowers never materialized. A major reason was the much higher credit rates demanded by informal lenders, ranging in 1989 from 30 percent a month for suppliers of raw material to 6 percent a month at neighborhood stores. Private banks offered credit at about 3 percent a month.
World Bank loan
The World Bank provided the program with a major boost in 1987 through a $102 million loan, of which $97 million was for onlending and the rest for technical assistance and training. The loan had three objectives: (1) to support BRI's efforts to develop the unit desas (village banks) into a financially viable network of full-service subdistrict branches; (2) to encourage BRI to improve its resource mobilization to fund KUPEDES, thus reducing its reliance on central bank credits; and (3) to improve the institutional capacity of both BRI and KUPEDES. The onlending portion of the loan was disbursed by September 1989, 31 months ahead of schedule.
KUPEDES became a model credit program in Indonesia. Between 1984 and 1994, it grew sixfold, much of which occurred during the first three years. The increase in the number of new loans and in the maximum loan size (to $3,150) led to another growth spurt between 1989 and 1990.
On two occasions (in 1988 and again in 1991-92), the program suffered setbacks. Drought, mild in 1988 and long and severe in 1990-92, caused a rural recession, pushed up arrears, and reduced demand for credit. The BRI's initial response was to restrict lending, which only increased the portfolio's deterioration because it kept many creditworthy clients from getting loans. But in 1991, BRI lifted the restrictions and in 1992, in keeping with the Project Agreement with the World Bank, issued a plan of action to reduce delinquencies. These steps produced immediate results. Arrears started to decline steadily, from 8.6 percent of the portfolio in 1991 to 5.6 percent in 1994.
Who was served
KUPEDES made credit avail-able to small borrowers in villages across the country. In spite of the market cost, small borrowers were able to obtain reliable credit more cheaply than from informal channels. The majority of borrowers were from landless and near landless families, who used their house plot or house to guarantee their loans. Women participated actively. A 1989 survey found 25 percent of the sample borrowers to be women. In reality the number of women benefiting from KUPEDES is probably much higher since small family enterprises in Indonesia involve both husband and wife.
The lending program also became fairly diversified. In 1994 more than half of the loans financed microenterprises and small-scale trading activities. Agricultural lending accounted for about one-fifth of the lending program. The biggest increase was in loans for working capital, used mostly for home construction and repairs. Started in 1989, working capital loans made up 29 percent of the portfolio by 1994.
The village banks developed into an effective channel for banking services, playing an important role in financial intermediation (see box). Moreover, BRI now has a cadre of 16,000 trained employees. Banking technology has expanded to village units, enabling the 3,300 units and 622 village service posts to reach close to two million borrowers or about 5 percent of Indonesia's estimated 36 million Indonesian households.
Savings mobilization was the most successful element of KUPEDES, supplying the bank units with a stable source of funds. By the end of 1993, savings accounts (excluding current accounts and time deposits) made up almost 28 percent of the accounts and 11 percent of the rupiah value in the entire Indonesian banking system. By 1994, village banks were servicing more than 12 million savings accounts. With many households having several savings accounts, the program reached an estimated 15 percent of the country's households.
The unit banks offered a variety of savings schemes, but their major program was SIMPEDES. Developed by BRI in 1984, SIMPEDES offered a safe, convenient, and liquid savings instrument that paid savers market-based interest rates (positive in real terms, that is, net of inflation) and allowed them unlimited withdrawals. The savings plan evolved from market surveys of villagers and its design was refined through pilot studies. Its purpose was to meet the needs of rural organizations and households. The program allowed deposits to be made in the name of an organization under joint savings held by the group leader or treasurer.
To mobilize deposits, the village banks used creative approaches. They conducted strategic mapping of potential savers, advertised their services, visited potential customers, and established links with village chiefs and community leaders. Many organizations responded, ranging from village treasuries and government offices to religious institutions, town organizations, and the like, helping to create a large market for rural deposits. As of 1994, deposits had grown 27-fold from their 1986 levels when SIMPEDES was adopted as a national program for the unit desas. And savers substantially outnumbered borrowers, thus helping the program achieve self-sufficiency (see figure). Moreover, by helping to mobilize rural savings within the banking system, the program provided a stable, credible channel for keeping financial savings in rural areas.
The role of the World Bank
The World Bank supported KUPEDES at a time when the program was still fragile and prone to outside pressure, particularly because of its continuing dependency on central bank loans. Moreover, Bank staff, aware of the poor results of many targeted and subsidized credit programs, strongly supported the program's simple, nontargeted design and working principles. As a result, the Bank lent credibility to the program's innovative approach and protected it from pressure to move to a targeted approach.
KUPEDES evolved in response to the failure of the subsidized BIMAS program and on the strength of an already existing delivery network. It took a major challenge to the survival of the delivery system to make the KUPEDES's market-based approach an acceptable option. As a result, some of the program's innovations may be unique to Indonesia. Nevertheless, the program's reliance on simplicity of design and application of sound banking principles is applicable to most small credit efforts. The use of a variety of institutional innovations (including those that helped motivate staff), attention to savings instruments, and close supervision and monitoring are also of general applicability. But in the end, institutional development takes time, and much nurturing. The audit highlights some of the important lessons from Indonesia's experience.
- Success of rural credit programs is facilitated by sound macroeconomic and sector performance. KUPEDES was initiated at a time of rapid growth spurred by Indonesia's structural adjustment and sound sector development program. The economic expansion fed into the growth of incomes in rural areas, and the strong rural infrastructure and agricultural supporting services programs helped to increase demand for credit, and as important, create the basis for growth of domestic savings in rural and small households.
- Government commitment and ownership are at the heart of a sound market-based credit program. KUPEDES was designed and launched by the government, with help from long-term foreign consultants from the Ministry of Finance. Thus, the program was implemented with a high level of government commitment and guided by carefully chosen technical experts who not only understood the issues but appreciated the need to adapt solutions to Indonesia's unique setting.
- A focus on objectives and simplicity of design are critical in credit programs geared to self-sufficiency. KUPEDES's objective of turning money-losing units into selfsustaining profit centers formed the framework for the program's approach to the types of financial services offered, the policies and procedures selected, the organizational characteristics adopted to run the program, and the type of support and technical assistance requested from the government, donors, and the Bank. The success of the BRI unit desas system, and thus the BRI/KUPEDES project, rested on its simple design, which in turn, was guided by good policies: (1) nontargeted credit, except for loan size; (2) nonsubsidized lending; and (3) self-sustainability through savings mobilization and profitable operations.
- Simplicity in design must be backed by careful attention to organizational detail and staff incentives. The program's management instituted a system of quality control that emphasized monitoring, supervision, and accountability backed by significant attention to staff training. Management also devised performance incentives, rewarding employees and units for good performance.
Box: The unit desas' structure and operation
In 1993, the Bank Rakyat Indonesia ran a vast network of almost 3,300 unit desas (village banks) and 620 village service outposts. Together, the units employed 16,000 people, representing the biggest single rural banking network in the country. The transition from nonmarket to commercial banking required a major change in company culture and work ethics. This transformation took place during the first four years of KUPEDES, and produced clear operational procedures.
Every subdistrict town had at least one unit desa. Staffing of units was determined by a ratio--for example, one credit officer per 400 borrowers, one teller per 200 daily transactions in automated units, and one teller for 150 daily transactions in nonautomated units. Village service outposts were manned by two people. The operational structure of each unit was kept simple. Units were staffed by at least four persons and at most 11. If the business of a unit expanded beyond the maximum staff limit, the unit was split, thus keeping the operation small and focused.
Institutional incentives were used to motivate staff and instill competition among them. For example, performance bonuses allowed 10 percent of the unit desas' profits (up to one-and-a-half months of an employee's salary in a year) to be paid to staff. To promote competition among the units, the central office devised a system to recognize the top 20 performers of each region, who then competed for becoming the top 20 performers at the national level. This competition was based on several criteria, including savings mobilized, outstanding portfolio, short- and long-term loss ratios, and management.
1/ Performance audit report, "Indonesia: BRI/KUPEDES Small Credit Project," Report No. 14511, May 1995. Performance audit reports are available to Bank executive directors and staff from the Internal documents Unit and from Regional Information Services Centers.