Microfinance Case Studies Indonesia INDONESIA'S RURAL FINANCIAL SYSTEM: THE ROLE OF THE STATE AND PRIVATE INSTITUTIONS TABLE OF CONTENTS INTRODUCTION | METHODOLOGY THE HISTORY OF INDONESIA'S RURAL FINANCIAL SYSTEM THE EFFICIENCY AND DYNAMIC CHARACTER OF THE RURAL FINANCIAL SYSTEM AND THE ROLE OF THE STATE IN WEST JAVA THE CLIENTELE OF THE RURAL FINANCIAL SYSTEM CONCLUSION | BIBLIOGRAPHY | ENDNOTES ACRONYMS BIMAS: Bimbingan MASsal: "mass supervision" of the Green Revolution Program BPS: Biro Pusat Statistik: National Bureau of Statistics BULOG Biro Urusan LOGistik: National Food Stock Marketing and Management Office CASER: Center for Agro-Socioeconomic Research CIRAD: Center for International Cooperation on Agronomical Research for Development GTZ: Deutsche Gesellschaft für Technische Zusammenarbeit IBI: Institut Bankir Indonesia: central bank research and training unit IPC: Interdisziplinäre Projekt Consult KSM: Kelompok Swadaya Masyarakat: associations, self-help groups PLN: Perusahaan Listrik Nasional: National Electricity Company REPELITA: REncana PEmbangunan LIma TAhun: five-year development plan SDI: Subsidy Dependence Index RFI: Rural Financial Institution FINANCIAL INSTITUTIONS Bank Pasar: Market bank BI: Bank Indonesia: central bank BKD: Badan Kredit Desa: village bank and granary BKK: Badan Kredit Kecamatan: subdistrict credit organization (Central Java) BKPD: Bank Karya Produksi Desa: producers' bank BPD: (West Java) Bank Pembangunan Daerah: regional development bank 1 BPR: Bank Perkreditan Rakyat: people's credit bank BRI: Bank Rakyat Indonesia: People's Bank of Indonesia KUD: Koperasi Unit Desa: village cooperative KUK: Kredit Usaha Kecil: credit for small enterprises KUM: Karya Usaha Mandiri: "establishment of independent activity," a Grameen Bank project KURK: Kredit Usaha Rakyat Kecil: credit for activities of the poor (East Java) KUT: Kredit Usaha Tani: credit for agricultural activities LPK: Lembaga Perkreditan Kecamatan: subdistrict credit institution P4K: Pembinaan Peningkatan Pendapatan Petani-Nelayan Kecil: "Helping small farmers and fisherman earn higher incomes" PHBK: Projekt Hubungan Bank dengan KSM: project to link banks with ROSCA: self-help groups Rotating Savings and Credit Association UD: Unit Desa: village unit Country Profile Economic and Social Context GNP per capita (1994) $880 Population (1994) 190.4 million Population density 100 inhab/km2 Inflation 1994 8.5% 1995 9.4% 1996 8.1% 1997 10.1% INTRODUCTION Following the acknowledged failure of state-owned development banks or credit components of agricultural development projects established during the 1970s, new alternative financial services were introduced in rural areas during the 1980s. These are based on various forms of banking arrangements, cooperatives, mutual benefit societies and solidarity groups, the overall objective of which has been to accommodate the needs and capacities of population groups which until then had been denied access to traditional types of financing. An analysis of these structures indicates that they have generally attracted a sizable clientele and that they have had a good loan recovery record. However, they have had to contend with two issues: first, how to manage their sustainability and their expansion to an extent sufficient to have an impact on development and, second, how to become integrated into the broader financial system. 2 The constraints involved managing risk in an often imperfect market, meeting the challenges of competition, controlling costs and managing resources to ensure financial equilibrium, and adapting to a formal institutional environment. In analyzing these institutions, Governments, donors and researchers face many questions for which few answers have yet been provided. What will the future place and role of these new alternatives be within the formal and informal financial sectors? Will they be sufficiently developed, stable, and efficient to meet the needs of rural populations? What role should Governments play? To what extent is government intervention justified? What regulatory structure should be established to protect macroeconomic stability and clients while simultaneously allowing efficient operations? The case of Indonesia offers very useful insight for answers to these questions in three respects. In political and economic terms, since President Suharto came to power in 1967, development has followed two paths. The first, which lasted until 1983, was characterized by state intervention, while the second, which began in 1984 and continues today, has been more liberal. In terms of financial policy, Indonesia has set out an original path for the liberalization of its financial system by profoundly transforming the network of state entities and defining a legal framework that specifically targets the development of small rural banks. Finally, in institutional terms, the rural financial system now includes many highly diversified institutions serving rural households. These in turn offer a multitude of case studies on the provision of alternative financial services. Answers to some of the questions posed above may be found in an analysis of the structure of Indonesia's rural financial system. This study will explore: 1) the rural financial institutions which have emerged, prospered, or declined during the various periods of state intervention and liberalization and the factors which determined their behavior; 2) how the rural financial system is organized and how it strives for stability and efficiency; and 3) the Government's role in this evolution. The study will analyze how effectively these organizations have performed their role as agents of Indonesia's economic growth (availability of resources, volume and quality of investments, resource allocation and appropriation). Within the context of an economy incorporating a multitude of new institutions, the study will take into account the imperfect nature of financial markets (characterized by such factors as asymmetric information, moral hazard, and adverse selection) which leads institutions to adopt certain procedures (search for information and incentive mechanisms) (Bardhan, 1989; Hoff et al., 1993; Stiglitz, 1986) and may justify some government intervention in order to ensure the efficient operation of the financial system (Besley, 1994; Stiglitz, 1992). 3 METHODOLOGY This analysis is based on secondary data on Indonesia's economy and financial system and on information gathered in interviews with key individuals in the institutions' main offices in Jakarta.1 In order to clarify these data, two fields of study were selected in West Java, a province where the development of the financial system has been very dynamic. The target areas were the districts of Bogor and Tasikmalaya. Although their economies are very similar, their financial institutions have developed along sharply different lines over time. Suffice it to say that, since the 1970s, Tasikmalaya's rural financial institutions have grown rapidly2, although very few new private rural banks have been established. In Bogor, however, very few small financial institutions were established until the liberalization of the financial system in 1988, and many new rural banks are now being established. The analysis of the reasons for these differences will enable the reader to understand what factors have spurred the creation of local institutions. One institution of each type3, for a total of 12, was surveyed in each field of study; demand was assessed by conducting two rapid series of household surveys (approximately 40 surveys in each region). The study begins with the history of Indonesia's rural financial system; next, based on case studies in West Java, it evaluates the factors underlying the system's efficiency and dynamism and then turns to the role of the State; the third part analyzes the demand for financial services in rural areas. THE HISTORY OF INDONESIA'S RURAL FINANCIAL SYSTEM Indonesia is a vast archipelago comprising 13,000 islands on both sides of the equator and spanning an area of 2,000 km from north to south and 5,000 km from east to west. Although this vastness endows Indonesia with a rich variety of land- and sea-based resources, it also makes the country - a divided territory scattered across the Indian Ocean - difficult to manage. In addition to being geographically vast, Indonesia has a population of 190 million, a heterogeneous mix comprising 366 ethnic groups with different languages, religions, and customs. Most of the population (100 million) lives on Java, Bali and Madura (which account for less than 10 percent of Indonesia's territory), resulting in population densities - some 800 inhabitants/km2 - which are among the highest in the world. Indonesia is among the East Asian countries which have been experiencing sustained economic growth for many years. It is able to rely on its resources as well as on the steady growth of its infrastructure and human resources, within a stable and highly structured political and social environment. If rapid growth is to continue, Indonesia must come to grips with the serious challenges facing it in terms of job creation and economic diversification . Moreover, Indonesia has not been spared by the current financial crisis in Asia and must take into account political and social dangers which may worsen in the coming years. 4 From the colonial Period to the Interventionist Era of the Suharto Government During the colonial period, until 1945, the financial system consisted of seven large commercial banks, all of them foreign. The three largest were Dutch institutions. There were only a few Indonesian institutions: BKDs and BRI. BKDs, or village banks and granaries (Badan Kredit Desa), sought to free small farmers from the grip of usurers, who had been condemned by the colonial powers. There were many such institutions in Java (in the rice-producing great plains) and, by the beginning of the twentieth century, there were more than 20,000 (Suharto, 1985). They still exist today, but in more limited numbers (5,300 in 1993). BRI, a national state bank, was established during the Dutch colonial period in 1895 and operates at the district level. Until 1968 it primarily served civil servants and, on occasion, farmers or entrepreneurs. During Sukarno's tenure (1945-1966), the banking system was almost completely nationalized and centralized, and it did not function smoothly. An economic crisis, coupled with a grave political crisis, paved the way for the violent transition which brought General Suharto to power in 1966. The Suharto Government faced a critical economic situation from the outset. During the initial intervention phase beginning in 1967, the Government's efforts to restore Indonesia to financial and monetary health focused primarily on combating inflation. Following the stifling of the financial system during the Sukarno era, efforts were made to foster the growth of financial services, particularly through the establishment of regional development banks. Established during the 1960s, these banks4 were operated by the provincial governments (which provided them with funds) and helped to finance regional economic policies. Located in each provincial capital, there were 27 of them, one for every Indonesian province, and by 1970 there were nearly 300 "people's credit banks"5 (BPR, Bank Perkreditan Rakyat) at the district level. Although Indonesia was a major rice-importing country, in 1969 the Government announced the ambitious objective of making Indonesia self-sufficient in rice production. The "green revolution" was launched primarily on Java and was intended to spur Indonesia's growth by heavily favoring agricultural development. At the same time, however, the financial system was again subject to severe restrictions, particularly on entry into the sector. Beginning in 1969 no new licenses were issued to foreign banks. A 1970 decree prevented all new private banks from opening and imposed heavy restrictions on the opening of new branches. The Government used the credit policy of national state banks to manage Indonesia's economic development process through complex mechanisms relating to targeted credit and ceilings. BRI thus became the predominant institution in rural areas at the expense of the provincial government banks. With the advent of the green revolution, the financial system's structure was dictated primarily by the extension of two major networks: BRI's village units (Unit Desa), which were mandated to distribute subsidized credit for rice 5 production, and the KUD agriculture cooperatives, which provided technical training. As from 1970, only non-bank financial institutions could be established in rural areas; however, their maneuverability with respect to savings mobilization and loans on the interbank market was severely limited, a factor which meant that they generally were restricted to operating on a very small scale. The Government's aggressive intervention in the financial system beginning during the late 1960s led to the predominance of the public sector in banking6. The Government's involvement in every aspect of economic development reflected its response to the economic crisis. Interventions were conducted to stabilize the macroeconomic situation and then to revitalize growth, focusing in particular on the agricultural sector. The central Government in Jakarta managed the entire state budget, and even regional governments were rather limited in their capacity to intervene. During the early 1980s, a number of converging factors challenged Indonesia's interventionist approach to development. To begin with, the country had to contend with the drop in oil prices and with the global economic crisis, which precipitated a decline in international aid flows. Furthermore, the economic environment had evolved substantially over a period of 15 years, and the objectives established for the first three five-year plans had largely been met, particularly self-sufficiency in rice production, which was achieved in 1984. Indonesia then had to define new priorities in the face of increasingly urgent problems which threatened to undermine the country's stability over the long term: on the domestic front, these were unemployment, the exodus from rural areas, and the crisis affecting certain institutions, including BRI; on the international front, Indonesia faced mounting foreign debt and oil was its major export in a market characterized by plummeting oil prices. The Gradual Liberalization of the Financial System (1983-1996) Indonesia began to liberalize its financial system as part of its broader effort to liberalize its economy, a step necessitated by the direction its economy had taken and recommended by the World Bank and the International Monetary Fund (IMF). Indonesia charted its own course towards a more open economy (Vittas, 1992), and the liberalization of the financial system in particular saw a gradual evolution of the regulatory structure and an effort to strike a balance between liberalization and economic stability. The evolution of laws and decrees7 During the period of government intervention, regulations virtually froze the growth of the private banking sector and created a market structure almost entirely controlled by the State. The liberalization process passed through the following stages: - June 1983: The first stage began with a 1983 decree which reduced the volume and number of institutions eligible for subsidized loans from the central bank, and raised ceilings on rates, loan volume, and the volume of most deposits. 6 - October 1988 (Pakto): The market was not truly opened up to competition until 1988, when the 1970 decree was superseded. The Government thus allowed established institutions more than five years (June 83 - October 88) to prepare themselves for the liberalization of the sector. In addition, new prudential regulations were introduced and a distinction was drawn between "general banks" and "people's credit banks", or BPR8. Banks in the first category, which operated like classical commercial banks, were the only ones authorized to accept sight deposits. Their minimum capital requirement was 10 billion rupiahs ($5 million). The BPRs, which were much smaller, could mobilize savings only in the form of time deposits or passbook accounts, and some of their activities were subject to restrictions (foreign exchange, insurance, and foreign ownership), as was their geographical scope (they were permitted to operate only at the subdistrict level, with each bank serving approximately 15 villages). However, the minimum capital required was only 50 million rupiahs ($25,000). Rural banks already in place were given two years to comply with the regulations introduced for the people's credit banks. The Pakto would profoundly transform the financial system. - March 1989 (Pakmar): In order to alleviate the difficulties which rural banks experienced in adapting to the new regulatory structure, the time limits established in the Pakto were eliminated. - January 1990 (Pakjan): The central bank's subsidized credits still targeted four priority sectors (primarily food production and small farmers). However, under the KUK program (Kredit Usaha Kecil, credit for small businesses), all national banks, whether state-owned or private, were required to earmark 20 percent of their portfolios for small credit9; similarly, state-owned enterprises were required to allocate 5 percent of their annual profits to the KUK program. - March 1992: all these changes were summarized and expanded upon in the 1992 Banking Act, which synthesized the regulations introduced since the liberalization process had begun in 1983. Its provisions may be summarized as follows: Ceilings on loan rates and amounts were eliminated; Savings mobilization and financial intermediation were emphasized (through incentives for savings mobilization and deposit protection); State banks were placed on an equal footing with the private sector (they were permitted to issue shares and the Government's responsibility was limited to the amount of subscribed capital, although this was not borne out in practice); International prudential regulations based on the CAMEL rating system were introduced, the central bank was required to conduct routine monitoring and supervision operations, and penalties were introduced; The Government was empowered to call on all the banks (not only state institutions) to become involved in development programs; The distinction between the "general banks" and the "people's 7 credit banks" (BPR) was confirmed. Reversing the action taken pursuant to the 1989 decree, the Banking Act again set a five-year time limit for financial institutions to adjust to the new regulatory framework. This applied essentially to rural banks established between 1967 and 1970, which, despite the fact that they were described as "banks," had continued to be classified as non-bank financial institutions. However, any new institution which had joined the financial system since 1988 was required from the outset to comply with regulations governing banks. The Act explicitly stipulated what status banks were permitted to adopt: the "general" banks could be set up as limited liability companies owned by the central or provincial governments or by private entities, or as cooperatives. People's banks could be established as limited liability companies owned by the provincial governments or private entities; they could take the form of cooperatives, or they could adopt some other status established by government regulation. The impact of financial liberalization on institutional growth The liberalization process as summarized in the Banking Act resulted in simpler financial market operating procedures, since there were only two categories of banks, the classical commercial banks and the "people's banks" (BPR) serving the rural population.10 The procedure for securing operating licenses was clarified. The rules allowed some flexibility, given the adjustment periods envisaged and the options offered with respect to institutional status. The liberalization process therefore led to the "deepening" of the financial system, as the figures below clearly demonstrate. Table 1. Increase in the number of banks (and branches) of the Indonesian financial system between 1983 and 1993 Type of bank 1983 1993 Commercial banks - state-owned 5 (727) 5 (981) - private 81 (337) 183 (2830) Development banks - state-owned 1 (22) 1 (42) - provincial 27 (185) 27 (425) 8 - private 1 (1) 1 (19) Savings banks - state-owned 1 (12) 1 (43) - private 2 (2) 2 (89) Rural banks(1) 5642 5343 - Village banks(2) - Other banks 391 1594 Source: Bank Indonesia, 1994. (1)In the case of rural banks, the number of banks equals the number of branches. (2)The "village banks" are village banks and granaries (BKDs); the "other banks" comprise the older BPRs established before 1970 which obtained bank status in 1983 (BKPD and Bank Pasar); these were joined in 1993 by new BPRs (930) and non-bank institutions which had achieved institutional status (273). In terms of the development of the number of banks, there were essentially two categories: on the one hand, private commercial banks and, on the other hand, "other rural banks," the number of which increased almost four-fold. These new rural banks were the 273 non-bank provincial government institutions which had obtained institutional status and 930 new BPRs, which were 95 percent privately owned. The new BPRs were established primarily on Java and Bali and in urban perimeters (see the second part of this study for a discussion of their operations). The number of banks in the national and provincial state network remained stable; none closed, no new ones were established and none were privatized; only the agency network expanded. The year 1993 brought further progress towards greater liberalization, as rules which had been considered too strict were amended. Bank credit expansion was encouraged by the relaxation of some prudential regulations and the elimination of certain limits imposed on the KUK program. The geographical restrictions on BPRs were removed, which subsequently enabled them to expand beyond the subdistrict level (May 1993 decree). The current diversity of Indonesia's rural financial system evolved over a long time and is a product of sustained government intervention, both direct (through the influence of state institutions) and indirect (through the gradual establishment of a regulatory framework). Involvement by the Government initially was justified in order to achieve macroeconomic stability and to enable it to lead the country out of the crisis of the late 1960s. The Government subsequently invested directly in the development of the financial sector: it established innovative institutions such as the BRI village units; it made it possible to establish at least a small measure of equality among the different islands; and it compensated for the lack of complementary markets such as an 9 insurance market. Lastly, the Government gradually established a regulatory framework applicable to all financial institutions, from the very smallest to the largest. The liberalization of Indonesia's financial system may be considered to have changed the rules of the game (Vittas, 1992), since financial institutions enjoyed the utmost flexibility with respect to their operations at the same time that they were subject to increased surveillance. The liberalization process reflected the capacity of the financial authorities to react to the evolution of the financial system and constraints on its development, as well as their desire to adapt to changes and signals in the economic environment. A balance was needed in order to ensure a more stable and efficient system. The Government continued to play a major role as both direct and indirect actor, and even retained the right to require banks to participate in certain development programs (such as KUK). At this point, the "Indonesian model" seems remarkable for the flexibility with which regulations were introduced over a 10-year period of reflection (1983-1992), during which reforms were gradually introduced, realigned, and finally synthesized in the Banking Act. This study will now explore the strategies implemented by the various types of institutions, the information-gathering mechanisms and incentives they employed during the screening, monitoring, and enforcement process, and the results they achieved in terms of economic and financial efficiency. THE EFFICIENCY AND DYNAMIC CHARACTER OF THE RURAL FINANCIAL SYSTEM AND THE ROLE OF THE STATE IN WEST JAVA This chapter will analyze the financial system in West Java by looking at four types of institutions notable for their flexibility and effectiveness. A distinction will be drawn between the older institutions which were established prior to liberalization and new institutions which came into being as a result of the liberalization process. Lastly, consideration will be given to relations between institutions and to the Government's role in regulating the financial system. How Older Institutions Adapted The older institutions were those established before or during the period of government intervention, i.e., those created up until the early 1980s. These were entirely or partly state-owned and generally received government support, or at least benefited from a flexible regulatory and legal structure which enabled them to grow. However, these institutions did have to adapt to the changing environment once the liberalization process began. 10 Resistance to change on the part of non-bank financial institutions The performance and contrasting characteristics of non- bank financial institutions Non-bank financial institutions are government-owned in whole or in part. Among the state-owned institutions, those owned by provincial governments were established after 1970, when bank status no longer could be obtained; in West Java, these institutions included the LPK (Lembaga Perkreditan Kecamatan, a subdistrict-level credit institution). Partly government-owned institutions included the long-standing village banks and granaries (BKDs) and the village cooperatives (KUDs) established during the green revolution. Non-bank financial institutions assumed various forms and followed divergent paths within the rural financial system; however, despite these differences, they all appeared to be rather lethargic and they did not adapt well to the rural financial system reforms. They were, in effect, micro-institutions which provided small loans. There were many of them, particularly in the KUD and BKD categories, but they were scattered, and coordination problems prevented them from working in a collaborative fashion. Their operating procedures were rather vague, as indicated by a provincial government statement that "wages and management systems are to be set according to the capacity of each LPK". The system undoubtedly resulted in a vicious circle of low salaries, lack of motivation and poor institutional performance. These organizations also were lacking in professionalism. Most employees had received no training in bank operations and were simply selected from among village populations to staff the BKDs. The KUDs had a "captive" market for savings accounts11, which put a damper on employee involvement and motivation. The major difficulty for the BKDs was that they were open only once a week, which restricted their clients' access to financial services and reduced contact between the institutions and their clientele. The manner in which these micro-institutions functioned did not qualify as financial intermediation. They collected very little in the way of savings and extended very few loans. The relative absence of monitoring by their owners and their staffs' lack of motivation and competence generally relegated them to a very secondary role within the rural financial system. This in turn resulted in a very poor record in terms of labor productivity, capital leverage, diversity and flexibility of services, and credibility. Mechanisms introduced to overcome the constraints of the financial market Against this background, a number of innovations recently were introduced to revitalize these institutions. The principal remedy introduced in response to the risk of loan default generally was to make village authorities share responsibility for compliance with the terms of loan 11 contracts. For example, LPKs require that 10 percent of loan amounts be pledged in the form of compulsory savings, and this requirement is backed by a written recommendation from the village chief (Pak Lurah) and a photocopy of the borrower's identity card is required. Because village political organizations were highly structured as a result of government intervention since independence, village chiefs, at least on Java, often are respected authorities. Some institutions were aware of their overall inefficiency and have attempted to reorganize themselves. A number of LPKs, worried about their excessive reliance on state financing, have introduced a locked "piggy bank" (Kotak Tabungan) which is placed in the houses of the borrowers, into which they can deposit excess cash at any time. Each week an employee collects the accumulated savings. This savings mobilization effort initially targeted only borrower members and the amounts collected were likely to be earmarked specifically for loan repayments. Even so, such initiatives enabled LPKs in Bogor to double their disposable savings. The Government also issued regulations to revitalize the cooperatives and provincial government institutions. The KUDs were granted autonomy (KUD Mandiri) so that - provided they met 90 percent of the requirements - it would be easier for them to borrow from banks and mobilize household savings. The requirements for autonomous status were related, inter alia, to financial statements and compliance with program monitoring rules. However, these requirements were not very selective and did not require any substantial change in operations on the part of cooperatives in order for them to obtain autonomous status. In particular, they were not asked to do much to improve service to clients, which was a weak point of the cooperatives. Their "autonomy" thus was likely to be rather limited. These efforts notwithstanding, it is surprising to note that the West Java institutions studied here appeared to have been bypassed by the transformation process of the rural financial system. The long history and stability of BKDs, the lack of interest in LPKs - and, by contrast, the systematic support provided to KUDs - help to explain these institutions' lack of vitality and their isolation in the midst of the transformation. Without withdrawing or trying to modernize these institutions, the Government remained directly involved in activities of limited scope and continued to provide operating and guarantee funds. These institutions apparently played a political role which justified their perpetuation in the eyes of the Government: village cooperatives were instrumental in organizing, structuring and monitoring the rural world. Institutions such as LPKs or BKDs, coming as they did under the umbrella of the provincial governments or village authorities, gave those governments and authorities a certain degree of freedom to direct the growth of the local economy. However, the State did not systematically adopt this approach, since some older government institutions experienced profound changes in recent years which have enabled them to face the future with greater confidence. This is true for KURKs (Kredit Usaha Rakyat Kecil, credit for activities of the poor) as described in Box 1, as well as for BRI, which will be dealt with in the next part of this study. 12 Box 1. The Institutionalization of the KURKs in East Java The institutionalization of provincial government entities proceeded in accordance with the 1988 decree (Pakto) and the 1992 Banking Act. The movement had begun in East Java in 1993. The institutionalization of East Java's KURKs was based on a regional regulation developed by the provincial government and East Java's regional development bank (BPD Bank Jatim) adapting features of the Pakto and the Banking Act to the specific circumstances of East Java and its KURKs. In particular, of the three legal entities whose status rural banks were authorized by the Banking Act to adopt - namely, regional enterprise, cooperative or limited liability company - the KURKs, which belonged to the provincial government, were urged to become regional enterprises. Capital subscriptions were available to the provincial government (approximately 50 percent), the district government (approximately 40 percent), and Bank Jatim (10 percent). This arrangement was intended to: (1) enable Bank Jatim to supervise the KURKs and receive a counterpart payment in the form of dividends; (2) ease credit and funds transfer procedures; and (3) in terms of development costs in particular, facilitate the diversification of KURK products along the lines of those offered by Bank Jatim. The transformation of the KURKs12 into institutions with the status of regional enterprises did not fundamentally alter their operations; rather, the central bank was able to monitor those operations more closely and the new status offered increased avenues for growth (such as savings mobilization and larger loans from other banking institutions). The shift from administrative functions to commercial functions by state banking institutions: BRI, a noteworthy example BRI is one of five state commercial banks (see Table 1). A national bank, it meets the criteria for "general bank"13 set out in the Banking Act and primarily serves rural communities, mainly through its network of village units. This network (BRI Unit Desa - UD) has been widely praised recently for its success in terms of the size of its clientele, the volume of loans disbursed, and the amount of savings mobilized (Robinson, 1992; Yaron, Benjamin, 1997). In the early 1980s, however, the future of these village units was in doubt owing to the serious difficulties they faced. This study will now examine how BRI transformed itself as the liberalization of the financial system progressed and what the determining factors were in its success. Changes made by village units in order to recover from the economic crisis During the green revolution, BRI was a government development agency. It managed a policy to promote technical innovations in order to make Indonesia self-sufficient in rice production. Subsidies, support and targeted credit were justified by the economic situation and the policy objective. However, the crisis of the 1980s imposed a strategy change. BRI's village units experienced an alarming decline in loan repayments, interest 13 in the technical assistance components which they financed dwindled, and loans tended to be disbursed to wealthier farmers. Moreover, the village units had no financial autonomy nor were they responsible for their own operations: they were merely the village representatives of local banks (Cabang) and therefore lacked the means and incentives to successfully encourage loan repayments (Patten, Rosengard, 1991). In 1984, BRI moved to transform the existing network byestablishing new objectives for it. BRI's UDs operated a tremendous shift away from their past practices. In addition to setting interest rates at a level which covered costs, BRI had to alter its methods of operation and engineer a change in attitude on the part of its clientele. Many units were closed down when their problems grew excessive, or they were transformed into "secondary units"14 which were open one day a week to serve areas with limited potentialities. Although BRI units underwent significant change in institutional terms (internal structure, allocation of responsibilities, redefinition of services provided), the system continued to benefit from the fact that it had a network of offices and employees stretching across the entire country, a network that was particularly dense on Java. Those units which remained open were required to operate in the black and assume responsibility for all financial intermediation services. Starting in 1984 each unit established its own accounts, which were independent of the local bank (Cabang). They had greater autonomy in managing loans, savings and their clientele.15 Local bank approval was required only for loans above a certain amount. In this manner, units which up to that point had served a purely administrative function went on to become profit centers. Services were offered primarily under two programs: KUPEDES for credit, and SIMPEDES for savings. These were simple programs which had very flexible requirements and met a wide range of needs. In particular, they were open to all rural activities, not just agricultural undertakings, and small loans were available (the minimum loan was 25,000 rupiahs, or $12; some went as high as 500,000 to 1 million rupiahs, or $250 to $500). Commercial strategies accompanying savings and credit services, such as incentive mechanisms, gradually facilitated a shift in attitude and bridged the information gap between lenders and clients. Prompt payment incentives (Insentif Pembayaran Tepat Waktu) were offered giving clients a partial interest rebate when loan installments were paid on time; good borrowers were encouraged to sponsor work colleagues to become new clients; deposits were guaranteed by the Government; and savings lotteries were held. Staff incentives were linked to employees' share of the unit's portfolio performance, as well as to their level of training and responsibility. On balance, BRI's transformation was fueled primarily by the "privatization" of its internal operations (decentralized decision-making, making employees responsible for their work, a profit orientation, giving employees a stake in performance, and incentives), the improved professionalism of its staff, increased flexibility, and the targeting of its services to rural activities in general rather than to purely agricultural undertakings. The gradual liberalization of Indonesia's financial system facilitated the 14 transformation of BRI. Although rates were allowed to rise beginning in 1983 when the institution's transformation began, the market was not opened to competition until 1988. This allowed BRI (and other existing institutions) to prepare for the new competitive environment. The Government served as a valuable catalyst in this transformation. Current performance of BRI village units The fact that BRI village units began operating profitably was a major achievement of the reorganization. The network's transformation was largely completed within only four years (1984-1987), and the cost to the Government in terms of subsidies was relatively low.16 By 1985, most of the units were turning a profit. In 1991 the unit network accounted for 25 percent of BRI's deposits, 10 percent of outstanding loans and over 70 percent of its profits (Robinson, 1992). The factors contributing to this success were the wide mobilization of savings with government guarantees, increased staff productivity (the result of motivation and training), and the support which the village unit network received from BRI's entire structure (whereby Cabang provided technical follow-up and monitoring services and ensured a balance between profitable and unprofitable units). In December 1996, the village units had 16.2 million depositors, 2.5 million borrowers, and a surplus of $1,260,000 in savings over outstanding loans ($1,705,000). BRI's outreach places it among the world's largest institutions offering services to rural areas.17 A comparison of unit performance of BRI and KUD ­ two networks which were created to support the green revolution ­ suggests that the Government has established two very different entities. One is a commercial enterprise free of cumbersome operations and constraints which might undermine its efficiency (although it is still vulnerable to the pressures of corruption) and possessing the financial and human resources enabling it to operate profitably. The other is a policy-oriented channel for government support for specific agricultural development programs and a player in defining the structure of rural areas. Older institutions thus were affected differently by the liberalization process. The study will now examine its impact on new institutions seeking a place in the financial landscape. The Development of New Types of Institutions Following liberalization The study will look first at the institutions established by the Government in the context of poverty alleviation programs. It will then turn to the highly dynamic system introduced by the private sector, which created a dense network of rural banks following the liberalization effort. New institutional channels for government intervention: poverty alleviation programs The Government increasingly withdrew from playing a direct role in the operations of Indonesia's financial sector. However, it continued to be an active participant in poverty 15 alleviation programs and established new institutional channels for intervention which allowed it to retain the initiative at the same time that it increasingly brought other actors onto the stage. The state-sponsored system joins the informal sector The informal financial system had been denigrated for a long time as an operation that was tolerated only because of the inadequacies of the formal system. However, not all informal practices disappeared with the establishment of formal institutions, and the advantages of the informal sector were in fact deemed worthy of interest. The informal financial system relied on close ties between lenders and borrowers, a situation that fostered more flexible and rapid service delivery without compromising contract compliance over time. Transactions proceeded on the basis of largely perfect information, requiring no special acquisition costs. The proximity of the two sides also meant lower transaction and coordination costs. As the informal system regained its respectability, formal institutions began turning to familiar community practices to renew their organizational structure. The Ministry of Agriculture, either directly or through the Center for Agro-Socioeconomic Research (CASER) under its supervision, launched projects patterned on the Grameen Bank18 model. Groups based on joint liability guarantee arrangements relied on the relationships among members to apply pressure to repay loans, since all loans were a shared responsibility. The central bank, with the support of GTZ (the German technical cooperation agency), worked alongside local village associations. These included rotating savings and credit associations, farmers' groups, religious cooperatives, family planning centers, and associations of microenterprise owners. In such instances, the projects built upon pre- existing associations which were already providing savings and credit services. The new institutional innovations implemented by the state thus relied on the support of, or were inspired by, established social and economic relationships, in particular, traditional self-help groups. The establishment of contractual relationships between institutions The Government also established links between development projects and financial institutions, a step which enabled it to take advantage of the specific features and capabilities of each partner. Project P4K was launched by the Education and Training Department of the Ministry of Agriculture, which did not have the capabilities of a financial institution. BRI therefore was brought in to manage project loans (by assisting with implementation, helping those involved to agree on proposed activities, disbursing funds, and collecting payments). In 16 addition to disbursing loans, BRI became the depository for the compulsory savings of members. Project P4K established a four-loan ceiling on the number of successive loans a client could receive, the objective being to kick-start the client's operation so that it would qualify later for one of BRI's standard loans. Staff at BRI and the ministry were given the task of monitoring the project. What made the operations of the PHBK project developed by the central bank and GTZ unique was the fact that links were established between banks, state and private institutions in the formal sector, and village associations, including self-help groups and volunteer cooperative networks (Kropp et al., 1989; Koch, 1993). There were three scenarios for this institutional linkage. In the first model, a bank disbursed loans directly to the group and received compulsory savings; an intermediary such as a nongovernmental organization facilitated the transaction by providing training, advice and assistance. In the second model, indirect links were forged between the bank and the self-help group. Here, the third-party institution served as a financial intermediary by receiving funds from the bank which it then disbursed to the group. Lastly, in the third model, the linkage was direct, in that the bank itself made the loans and provided advice and training. The project favored the third model, in order to encourage all the partners to establish direct contacts and assume direct responsibility, and in order to cut interest rates as far as possible by reducing the number of intermediaries involved. All these projects, government-sponsored or otherwise, were based largely on practices which are now acknowledged to foster sound institutional operations, particularly with respect to loan recovery: preliminary surveys to gather data, frequent meetings or work sessions with local staff, social pressure, incentives, and the creation of client expectations of future loans. Savings mobilization, however, continued to be a marginal activity. As the Government withdrew from direct involvement, it delegated the tasks of screening, disbursement, and monitoring to agents more qualified to conduct such activities given their proximity to clients and their information networks. Nevertheless, some programs remained heavily dependent on the State or on subsidies and risked engaging in unfair competition with established financial institutions which were striving to become independent and profitable. The development of the private sector The number of private rural banks grew substantially following the liberalization of the financial system, and particularly on the heels of the Pakto. This study will now examine the reasons for this growth. 17 Characteristics of the new institutions The new rural banks took advantage of the opportunities created between 1970 and 1988 by sustained economic growth. These institutions, which were 95 percent privately owned, were found primarily in the urban peripheries of Java and Bali. In geographical terms they complemented the regional rural banks, but they competed directly with the BRI village units, which enjoyed government support and benefited from years of experience since the 1970s. Seeking to distinguish themselves, the new institutions succeeded initially because they cultivated the image of quality service providers: their offices were spacious and often new, their staffs were large, and their clients felt welcome. Their strategy generally was to provide clients with a wide range of services. They were unique by virtue of their business volume and interest rates, the incentives and gifts they supplied, and their attentiveness to important household milestones, as illustrated by the fact that they offered savings plans to finance pilgrimages to Mecca and school loans for children. Emulating BRI to some extent, they screened their clientele first by requiring borrowers to be able to present physical collateral such as land titles, regular salaries or, more often, productive capital. The BPRs also counted on a sufficiently large and professional staff to evaluate and monitor their clients' projects. Some of these banks adopted an approach distinctly tailored along the lines of Islamic banking. In so doing, they adhered to the Islamic prohibition on fixed, preset interest rates. The banks and their clients shared profits and losses. By adopting Islamic principles, rural banks were quickly able to win over large numbers of clients and thus mobilize substantial savings. In addition to BMI, Bank Muamalat Indonesia, based in Jakarta and Java's provincial capitals, in 1994 there were some 40 rural Islamic banks scattered throughout Indonesia's Muslim regions. Characteristics of the new rural bank networks The 1992 Banking Act limited the geographical reach of BPRs (restricting them to subdistricts) and the rural banks which came into being with the liberalization of the financial system frequently organized themselves into networks of banks dispersed over a wide geographical area. The purpose of these spontaneous organizations was to reduce risks that all the individual rural banks faced due to their limited geographical diversification (inadequately diversified portfolios and exposure to local risks such as droughts, epidemics, and declining prices in local markets). Nearly 60 percent of all new rural banks were organized into networks, and 70 percent of the networks had at least five banks (BPR Bagian Pengawasan, July 1994). These networks were varied in nature, and none had a status imposed or controlled by the central bank. Some included a central organization which monitored and coordinated the activities of the various BPRs; others were structured horizontally, with no umbrella institution, and member banks organized their own funds transfers and training programs. Some networks adopted an original approach, such as the network of rural 18 Islamic banks and regional networks financed and organized at the provincial level by regional development banks (BPDs) and individuals living in these provinces. Some rural banks were grouped into networks because they belonged to the same owner. These might be Indonesian nationals owning several banks, or private national banks which preferred to serve the rural market through rural bank networks instead of setting up local branches. The network structure made it possible to optimize supervision and monitoring efforts and achieve economies of scale with respect to training and technical support services. General banks which created networks with BPRs were able to open up a new market in rural areas and disburse KUK program loans pursuant to government directives. The growth of this type of network occurred because risk was limited, since the activities of the BPRs were independent of the operations of the general banks and other BPRs. If problems arose, each BPR served as a "fuse" within the rural bank/general bank network structure. The networks thus constituted a spontaneous innovation which was widely developed, and the central bank adapted to their presence, particularly by relaxing earlier geographical constraints (see the May 1993 decree). The performance of the new rural banks From the very outset, the structure of the new rural banks was impressive in terms of staff size and premises. They were known for offering convenient services, some institutions even having mobile loan officers to visit the clients. Their productivity consequently remained relatively low in terms of business volume and numbers of clients per employee.19 Nevertheless, the extremely rapid increase in business volume with no increase in staff suggests that these banks can improve employee productivity very quickly to a level matching that of their competitors, such as the BRI village units. A review of the rural banks' financial statements reveals that they have different liability management strategies. While leverage is important to all the banks, their funds are derived from different sources. For example, the Islamic banks, which focus on savings mobilization, devote considerable effort to this activity, and a substantial share of the savings they attract is deposited in long-term accounts. Some BPRs rely on long-term debt through private banks, often as part of the KUK program (20 percent of the national banks' portfolio is in the form of small loans disbursed through BPRs). However, the rural banks often adopt similar strategies with respect to assets, i.e., they systematically disburse approximately 80 percent of their assets in the form of loans. They are seeking to expand and to capture the financial market by offering credit. Their balance sheets indicate that they have assumed considerable risk, given their low liquidity ratios and reserves. Over the medium term, this may mean liquidity crises and perhaps failure for some of these BPRs. The private sector thus has expanded very aggressively into rural markets and has seized the opportunities presented by the financial liberalization reforms while maintaining a highly innovative approach that has enabled it to adapt very flexibly to 19 regulatory constraints. The constraints persist, owing either to the Government's insistence on retaining excessive control or to the inflexibility of the administrative machinery, which had to wait until 1993 for some of the rules that had been in place since 1988 to be amended. Given this situation, the central bank has faced heavy supervisory responsibilities, particularly since competition has spurred established institutions to adopt reckless lending and savings mobilization policies. By opening up the financial sector, the Government of Indonesia performed two functions justified by economic theory: first, by developing original mechanisms for state intervention as a way to reduce poverty in rural areas it acted in the interests of a more equitable distribution of resources; second, it created an environment hospitable to competition and to more efficient institutional operations. However, the Government appears to have exceeded its prerogatives in some respects (as illustrated by the geographical restrictions imposed on BPRs) or to have intervened in a manner that was not economically sound (subsidy programs and deposit guarantees on BRI savings for example). Given the existing regulatory structure and the incentives offered by the Government, some institutions took advantage of the more open market and devised endogenous innovations to facilitate efficient operations. This study will now consider whether the Government has retained control of the financial market and will examine how the rural financial institutions, taken as a system, relate to one another. Regulation of the System and relationships among Institutions: Towards a Contestable Market? The many types of financial institutions now in existence differ considerably with respect to their operations, scale, size and efficiency. It would appear useful in this context to introduce the concept of the "contestable market." In general terms, this concept addresses circumstances pertaining to the efficiency of an industry's organization or the manner in which freedom of entry into a market can enable the majority of established businesses to operate efficiently (Baumol et al., 1988; Baumol and Lee, 1991). 20 Box 2 . Characteristics of a contestable market 1. Definition By definition, a contestable market has no entry barriers, which means that there is complete freedom of entry and absolutely no sunk costs are incurred.20 The advantage of a contestable market is that ever-present concern regarding the arrival of new entrants prompts established firms to adopt behaviors that are beneficial to their clientele. Accordingly, even if an oligopoly or a monopoly exists, established firms cannot set prices too high or reap excessive profits; new arrivals can always adopt "hit-and-run" tactics and seize opportunities to make a profit and then exit once the environment turns less hospitable. For the same reason, it is incumbent on firms to operate efficiently in terms of resource allocation, management capacity, innovation, and know-how. Finally, they are compelled to act in good faith. They cannot engage in unfair competition by setting prices below costs ("predatory pricing"), because they can not make up for such losses later with excessive profits. Discriminatory pricing ("cross-subsidies") also is eliminated. Originally used to analyze the structure of industrial markets, the theory of contestable markets helps in the context of this study to explain the efficiency of the rural financial market. 2. Application of the theory to Indonesia's rural financial system At state and private institutions alike, assets, outstanding debt, and collected savings are scattered and, although liberalization reforms have been implemented, the Government continues to play a major role. The system's structure is essentially similar to that of a contestable market, in that it includes oligopolies as well as small institutions often new to the market. a. Financial market entry and exit criteria Entry into and exit from the financial market are governed by precise rules which are clearly set out in the Banking Act, and the central bank has already issued many licenses. The procedures are therefore well-known. Although investments (offices and equipment) are largely "recoverable," the characteristics of the banking business are such that loans must be managed with the utmost efficiency in order not to jeopardize capital invested at the outset. Nevertheless, through the Banking Act, the Government apparently has established criteria allowing complete freedom of entry given the inevitable constraints of the banking business. b. Economies of scale and scope A central tenet of the theory of contestable markets relates to economies of scale and scope. Baumol and Lee (1991) note that "the role of economies of scale and scope is crucial ­ without the availability of such economies, no one would seriously advocate 21 promoting or perhaps even tolerating bigness." However, this assessment does not appear to take into account the fact that economies of scale and scope challenge the concept of the complete freedom of market entry, since firms practicing such economies that wish to enter a particular market must make substantial investments in order to reach an efficient size. It would therefore be useful to examine economies of scale and scope in detail, in order to understand the advantages of contestable structures in terms of efficiency of production, as well as the fairly substantial constraints on new firms. The advantages and disadvantages with respect to the size and organization of rural financial institutions (RFIs) may be represented schematically as follows: Advantages and disadvantages of institutions according to size Advantages Disadvantages Small institution - Proximity, familiarity with - Limited access to the local conditions (lower ex ante interbank market; contract costs); - Risk management. - Small structure (lower management costs). Large institution - Better protection against - Cost of managing and covariant risks; monitoring a geographically dispersed - Flexible balance sheets; network; - Impact of experience; "Too big - "Decreasing return on to fail". supervision". Although entry into the financial market was completely blocked from 1970 to 1988, the current structure of Indonesia's rural financial system is more characteristic of a contestable market. Since 1988, the Government's role has been to protect state banks, which, in addition to providing classical financial intermediation services, also serve the general public throughout the country, even on the more remote islands (although they are fewer in number there). They also provide the Government with a mechanism for developing certain elements of its economic policy, as well as for financing innovations such as the BRI network of village units. However, because the state system was likely to prove inefficient, the Government pressed for arrangements which gave the private and public sectors economic incentives to manage these institutions efficiently. Lastly, the cost structure appears to allow small institutions to enter the market freely and to compete with established institutions, whether public or private. 22 Some restrictions remained, however, concerning the pace and terms of the liberalization effort. Between 1983 and 1988, the Government allowed state institutions a delay during which they adapted to the transformation process; they were then in a position to implement "retaliatory measures" to protect themselves against future entrants by advertising their services to clients and winning them over through incentives (e.g., lotteries, loan interest rebates). In addition, new entrants faced restrictions on their activities, particularly regarding the geographical areas they were allowed to serve. However, the 1993 regulations relaxed the constraints somewhat and BPRs were now permitted to operate outside subdistricts. Finally, the support received by certain state institutions could be construed as giving them an unfair advantage over new entrants; the government guarantee on savings deposited with BRI village units has enabled them to mobilize a substantial volume of savings which earn a lower interest rate than that paid by competing institutions. This situation enables the village unit network to earn a considerable profit. With regard to loans, lively competition21 compels new institutions to adopt a firm offensive strategy which might take the form of an extremely high ratio of loans to assets (approximately 80 percent at some BPRs), a situation requiring sound insurance arrangements. Moreover, the number of institutions is growing rapidly and the central bank now has heavy supervisory responsibilities both in terms of granting licenses and in regular monitoring of banking activity. The financial deepening of the Indonesian system in turn has enabled institutions to find ways to complement one another. These complementarities relate primarily to technical support and funds transfers and, through recourse to loans on the interbank market, they enhance the system's global image and no doubt sometimes help to alleviate liquidity crises. However, it would appear that some market segments are not served - small businesses still do not receive adequate financing from the financial sector - and that the financial services provided are primarily small short-term household loans. The Government has steered the rural credit market on Java toward the structure of a contestable market, although it has not necessarily done so elsewhere in Indonesia. However, given the distinct advantage which deposit guarantees provide, state institutions continue to benefit considerably from their savings mobilization efforts, from which they are able to derive extremely high profits. THE CLIENTELE OF THE RURAL FINANCIAL SYSTEM Having reviewed the extensive development of the rural financial sector, this study will now turn to the perceptions of clients, particularly rural households, and will consider the impact which these perceptions may have on the institutions' future strategies. The data for this discussion was gathered during field surveys in West Java. As is generally the case in Indonesia, the rural households of West Java are characterized by their nuclear structure. Household economic activities generally are very varied. Maurer (1986) speaks of the "occupational multiplicity" of Java's rural 23 areas, combining agriculture (rice, soybeans, and cassava in particular), animal husbandry, crops grown in village gardens, salaried labor, activities in the informal microenterprise sector, crafts, trading and temporary migration for work abroad. This diversity may be attributed to a strategy to reduce activity-specific risks as well as to the fact that access to full-time salaried employment is limited (Dury, 1997). Access of Households to Financial Services Household access to financial services is determined by geographical location and the type of collateral households can provide. Household surveys indicate that in Tasikmalaya, a region where traditional formal financing mechanisms are prevalent, approximately 20 percent of the households have outstanding loans with official institutions in the region; in Bogor the formal market serves slightly over 10 percent of the households. Thus, despite the arrival of new institutions, they have made few inroads in the market. As to the informal sector, 16 percent of Tasikmalaya's households have outstanding loans with informal lenders; the corresponding figure for Bogor is 38 percent. Field studies show that the informal sector continues to have a strong presence, not only among the poorest members of the population, but also at all levels of society, even among people who can gain access to formal services. The informal financial services are described in Box 3. The differences between Tasikmalaya and Bogor, however, show that the higher- income segments of the population in income-earning households and households which own real estate assets are abandoning the informal financial system as formal institutions gain ground. There would appear to be an adaptation period during which the formal system and its clients get to know one another. Factors such as habit, mutual understanding, and progressive introduction of clients to formal financial services play a major role in shaping the impact that the sector has on households. The formal sector therefore must develop a reputation for itself and win over local populations. To this end, it must gather sufficient data on the nature of demand and the risks involved in order to offer services which are flexible and which reflect the needs of its clients. However, there is reason to believe that the approximately 15 percent of Tasikmalaya's households which continue to deal with the informal sector represent an irreducible core of clients for this sector, given the intricate social relations woven in villages among self- help and mutually dependent groups, the inadequacy of insurance mechanisms, and the rapid response and flexibility characteristic of the informal sector. Box 3. Principal informal sources of financing The following four sources are the most widespread: a. Rotating Savings and Credit Associations (ROSCAs or Arisan in Indonesian): Indonesian ROSCAs primarily take the form of neighborhood groups of 10 to 40 people (mainly women) organized into cooperative groups (in which each participant receives 24 the amount contributed) or commercial ROSCAs (in which the ROSCA's organizer collects a management fee). The amount of money involved usually is very small (approximately 100,000 rupiahs or $50) and generally is used to purchase consumer goods. The ROSCAs also play a key social role. For example, neighborhood women may organize events in connection with religious ceremonies, such as the feast of Ramadan, at ROSCA meetings. The ROSCAs also fulfill a consensus-building social function which the State encourages in all areas of community life in order to promote the activities of groups of individuals meeting in pursuit of economic, social, or religious objectives. Accordingly, while the ROSCAs belong to the informal sector, they are fully recognized by official institutions. b. Shopkeepers: Retailers, grocers, and owners of very small restaurants (warung) often extend credit to their customers. Such credit is used to cover basic necessities or, occasionally, durable goods, and interest rates may be steep (up to 20 percent monthly). There are also itinerant vendors who sell clothing and dishes on credit. Finally, many dealers in agricultural products buy the harvest ahead of time. This practice is called ijon, from ijau, which means green. These "term contracts" exist for rice and highly perishable fruits such as bananas. c. Pledged property (gadai in Indonesian): The gadai is an exchange between two individuals whereby a sum of money is lent against a pledge of a productive asset (such as land, a rice field, a fruit tree, or a vehicle). Dury et al. (1996) have extensively studied gadai involving trees. The owner of the pledged asset relinquishes his right to use it to the creditor for the duration of the loan. The creditor can then use the property as he sees fit and harvest the crop, which serves as interest on the loan. Since no duration is specified and harvests can vary, particularly in the case of fruit trees, interest can fluctuate between 40 percent and 100 percent annually. The gadai is very flexible and makes it possible to borrow relatively high amounts (between 100,000 and 500,000 rupiahs for a durian tree, or $50-$250, or several million rupiahs for a papaya plantation). d. Uncollateralized personal loans: Loans of this type between members of the same family, neighbors, colleagues at work, or employers and employees are fairly common. Although they are generally interest-free, social or traditional ties linking the households concerned generally mean that it is more or less explicitly understood that work, gifts, material aid or future support will be given in return. There are also "money keepers," trustworthy individuals who agree to safeguard cash for individuals, sometimes in exchange for a small fee. Persons who make such deposits generally try to avoid being in debt to persons who are close to them. The informal sector, which is traditional and of long standing, remains firmly anchored in the practices of rural populations, the development of the formal sector notwithstanding. These systems reflect diverse and at times interwoven elements: economic (shopkeepers, gadai) social (ROSCAs, employer/employee relationships) or cultural such as religious prohibitions against collecting interest (gadai). The informal sector is therefore much more than a substitute to the formal financial sector. 25 Segmentation of Financial Networks Market segmentation is dictated by the requirements of each institution as set out in their credit policy. BRI and the BPRs extend loans against physical collateral (such as a certificate of title to land, a house or a vehicle, or a pay slip). Accordingly, well-off families are BRI's primary borrowers. The surveys conducted in Tasikmalaya indicate that the average household utilizing BRI's services differs from the average sample household in that the income of each member, the loan amount, and the value of productive assets are twice as high. However, these results are widely scattered, indicating that BRI also serves less- well-off households, but in fewer numbers. Because of the physical security requirement, BRI and BPRs are the only formal institutions, along with government employee cooperatives, which offer sizable loans (up to several million rupiahs). Only the gadai system in the informal sector offers loans for more than 1 million rupiahs ($500). The ability to provide physical collateral is therefore a determining factor in securing more substantial loans. Indonesian Governments have promoted the cooperative movement since independence. Many population groups have a registered cooperative. These take many forms and often are organized first and foremost as social and self-help groups. Government employee cooperatives require that members earn a regular salary; microenterprise owners engaged in the same activity often organize cooperatives. Social cooperatives, such as school cooperatives or those organized at mosques, are semi-formal systems whose members' participation is based on their "social capital" and the fact that they share the same values. Finally, village cooperatives (KUD) function as quasi-compulsory intermediaries for the marketing of rice and many subsistence crops. Loans are extended based on the borrower's social capital as well as his "political capital," which is established through work performed in the cooperative's office or by helping monitor its agricultural development programs. Programs to assist poorer populations include projects involving mutual benefit societies along the lines of credit unions or "linkage" projects of the PHBK/GTZ variety which require that a certain amount of savings be mobilized before a loan is extended. Future borrowers therefore need to have some savings capacity. In the case of projects based on joint liability, borrowers must be members of a group which in principle relies on "social capital"and solidarity and makes members eligible to share responsibility for repayment. The informal sector also takes considerable advantage of social capital and family networks. Despite this institutional diversity, some members of the population are not served. First among these are the poorest members of society, although this situation raises the substantive question of whether credit markets can really be adapted to meet the needs of the extremely poor. Another unserved group comprises "undesirable" individuals 26 (yang tidak baik) or persons "of bad character" (karacter yang jelek), language which, although rather vague, was actually used in the surveys of some BPRs or cooperatives. The third excluded group comprises small and medium-sized enterprises (SMEs) which are larger than family businesses. The KUK programs (see the January 1990 regulations) were intended to generate support for SMEs, and KUK loan ceilings were raised in 1993 to encourage this and take the focus off lending to the poor. However, the major barrier derived from the risks incurred: financial institutions considered SMEs overly risky investments (inadequate insurance protection, an inadequate legal framework in the event of default, and activity-specific risk). In fact, an evaluation of institutional practices shows that the factors which determine whether households enjoy access to credit are, first and foremost, whether they own assets and have a regular income. Accordingly, landowners or government employees enjoy privileged access to the financial system. Finally, for most institutions providing convenient service, their clients' social capital and level of respectability are key selection criteria. Household Strategies regarding Financial Services Households generally devise strategies to diversify their portfolios, particularly in order to protect their savings. With respect to savings, only seven of the families surveyed in Bogor (18 percent of the households) have deposit accounts in formal institutions. In Tasikmalaya, however, 22 families, or half the sample, have such accounts in formal institutions. BRI in particular has been very successful in this region, and 20 percent of the families save with BRI for a total of 22.8 percent of the volume of savings of the sample. As for informal savings mechanisms, rich households use gold to diversify their holdings.22 The Bogor surveys demonstrate the importance of informal intermediation, since one third of the households make loans to family members, neighbors or friends totaling more than half of the sample's total savings. Villagers make their own very short-term intermediation arrangements which do not entail the direct payment of monetary interest. With respect to loans, while BRI is a major player (86 percent of loan volume), it has very few borrowers (9 percent of the families in the sample). The strength of the informal sector is evident in Bogor (56 percent of loan volume, 50 percent of the families in the sample, and 80 percent of households that are borrowers); in Tasikmalaya, the number of persons that borrow from the formal financial sector equals the number of persons that borrow from the informal sector (18 percent of the sample in each case). However, in contrast to the savings portfolio, there is very little diversification of loans, and the situation in Bogor differs considerably from that in Tasikmalaya. Box 4 describes how loans are used. The formal system often is in direct competition with the informal market because market niches tend to overlap. Formal financial institutions therefore face competition on 27 two fronts: from formal institutions as a result of the liberalization process, and from the informal savings and credit system. The different situations in Tasikmalaya and Bogor reflect the public's attitude towards financial services as defined by how well formal institutions know their clientele and the degree to which traditional family and village ties prevail. Box 4. How loans are used As to their purpose, surveys23 show that loans, used for the most part for productive purposes, generally enable households to increase sales or conduct an ongoing activity on a more regular basis. These productive activities are primarily commercial activities: the sale of foods processed on-site (pastries, salted foods, fruit juices) and sold in a small shop (warung) or door-to-door (pikulan); the purchase of fabrics or clothing in Bogor or Jakarta and their resale, sometimes on credit, in the village. The second purpose for which loans are used is to finance agricultural activities (the purchase of seeds, fertilizer, small ruminants, or poultry). Finally, loans often are used to finance service activities, such as purchases of gasoline and small equipment by motor-taxi operators. Loans are frequently used to finance children's schooling, with women, who take a greater interest in their children's education, accounting for most of such borrowers. With second and third loans, the percentage used for productive activities tends to decline, although in absolute terms the share largely remains constant (reflecting the amount needed to sustain the activity); however, the share of such loans spent on consumer goods tends to rise and is used to improve housing conditions (building materials, electricity, radios, televisions) or to purchase gold in the form of jewelry (a form of liquid savings in the Indonesian context). Household strategies vis-à-vis financial services are flexible, diverse, and, to some extent, complex, which requires that formal institutions know their market well. That being the case, the Government has thus played a major role by bearing the risks of innovation, i.e., the investments needed to develop a stable clientele receptive to the financial services offered by the formal sector. This is illustrated in the Tasikmalaya region, where state financial institutions have been established since the late 1960s. Because support was provided to the earliest rural banks operated by provincial governments, households in the region are more receptive to the financial services offered by the formal sector. CONCLUSION Indonesia's rural financial system offers a particularly interesting case study in the context of current issues relating to rural financial mechanisms, in particular, with respect to the role of the State and private institutions. This report has analyzed the system's structure within the framework of imperfect markets and the liberalization process. The Government has performed three major functions within the financial system: 28 First and foremost, it established a regulatory framework for the financial system through a series of decrees and the 1992 Banking Act. This framework proved flexible and should be able to evolve over time as a function of the changing economic environment and any limitations that may be introduced. This framework also clearly sets out the procedures for market entry and the prudential regulations with which general banks and BPRs must comply. Its specificity also derives from this clear distinction between the two types of banks, which has provided a simple framework for the establishment of nearly 1,000 rural banks since 1988. Second, the Government played an important role by virtue of its impact on its own institutions. The sharp differences in performance by BRI and village cooperatives demonstrates the importance of the State's catalytic role in generating reforms within state institutions. The Government has the option to transform an institution such as BRI by "privatizing" its operations (through decentralization and by increasing the responsibilities of its staff, promoting the profit motive, and providing incentives). In the same economic and regulatory environment, the State can choose to perpetuate the status quo by continuing to provide financial assistance to programs whose employees are not encouraged to develop a strong sense of responsibility or to become involved in decision-making or client selection. Lastly, the Government can play a catalytic role. It supported the formal sector when it was first being established, when households initially were hesitant to turn to the sector, as illustrated by the different situations in Tasikmalaya and Bogor. The Government can also finance the development of innovative financial services; one example was its support for BRI's village unit network, which made it possible to introduce efficient services adapted to the needs of the rural population throughout Indonesia. The State also can promote partnerships between commercial banks, state institutions and largely informal village associations and groups, as a component of poverty alleviation programs. Although such programs generally are launched by the central bank, ministries (cooperatives and agriculture), and state banks, their implementation is largely the work of private partners who are more familiar with local conditions and the clientele, and are often a part of social networks, a factor which bolsters contract performance. Hugon (cited in Castel, 1995) describes the State's "pro" stance in Asian economies as follows: promoter of private initiatives and of the rehabilitation of government institutions; producer of services (such as within BRI) and poverty alleviation programs; and prospector and programmer, as it delineates reforms and continually readjusts them. However, the situation in Indonesia must be placed in its proper perspective. The financial system developed within a context of considerable government stability which left little room for political criticism. Moreover, the country's rural society, particularly in West Java, is highly structured (groups play an important role, as do village leaders) and its population is extremely dense. This has made it easier for financial institutions to operate efficiently with lower travel, client monitoring and data collection costs. There 29 are still questions to be answered. Additional research would be useful in order to delve further into the concept of economies of scale as they relate to financial services, examine financing mechanisms for small and medium-sized enterprises, closely track the evolution of the regulatory structure and the growth of the Indonesian financial system, and analyze the requirements for the development of insurance mechanisms, which are still virtually nonexistent. Lastly, it would be particularly useful to study how rural financial institutions are responding to Indonesia's current economic crisis and to examine the role of the Government during these difficult times. BIBLIOGRAPHY Adams D.W., Graham D.H. , Von Pischke J.D. (Ed.), 1983. Limitations of Cheap Credit in Promoting Rural development. Washington, D.C., World Bank, EDI Training Materials, 135 p. Bank Indonesia, 1984-1993. Reports for the Financial Years 1983/1984 - 1992/1993. Jakarta, Bank Indonesia. Bank Indonesia, 1993. Act of the Republic of Indonesia N°7 of 1992 Concerning Banking with Regulations for Implementation. Jakarta, Bank Indonesia, 231 p. Bardhan P., 1989. The Economic Theory of Agrarian Institutions. Clarendon Paperbacks, 403 p. Baumol W.J. , Lee K.S., 1991. Contestable Markets, Trade and Development. The World Bank Research Observer, 16 (1) : 1-17. Baumol W.J., Panzar J.C. , Willig R.G., 1988. Contestable Markets and the Theory of Industry Structure. Harcourt Brace Jovanovich, Rev.ed. San Diego, California Benoit-Cattin M. , Irawan B., 1995. Adjustment of Rice Supply to the Evolution of the Consumption Pattern in Indonesia (1972-1993); Mid and Long-term Perspective and Implications. In 2nd International Symposium on systems approaches for agricultural development, Philippines, 6-8 december, 1995, 22 p. 30 Besley T., 1994. How do Market Failures Justify Interventions in Rural Credit Markets? The World Bank Research Observer, 9 (1) : 27-47. Bouman F.J.A. , Hall H.A.J., 1992. Informal Finance in Indonesia. In Informal finance in low-income countries, D. W. Adams and D. A. Fitchett Ed., Westview Press, p. 209-223. Braverman A. , Guasch J.L., 1986. Rural Credit Markets and Institutions in Developing Countries. Lessons for Policy Analysis from Practice and Modern Theory. World Development, 14 (10/11) : 1253-1267. Chaves R.A. , Gonzalez-Vega C., 1996. The Design of Successful Rural Financial Intermediaries: Evidence from Indonesia. World Development, 24 (1) : 65-78. Christen R.P., Rhyne E. , Vogel C., 1994. Maximizing the Outreach of Microentreprise Finance. The Emerging Lessons of Successful Programs. IMCC, 60 p. Cole D.C. , Slade B.F., 1996. Building a Modern Financial System, The Indonesian Experience. Cambridge, Cambridge University Press, 452 p. Colliot E., 1996. Complémentarité et concurrence entre banque d'Etat et un projet de crédit décentralisé animé par une ONG dans une petite région du Viet-Nam. In Développement et transition vers l'économie de marché, Hanoi, Viet-Nam, 1-24. Cuevas C. , Benoit-Cattin M. (Ed.), 1991. Finance et développement rural en Afrique de l'Ouest. Montpellier, CIRAD, Actes du XIIème séminaire d'économie rurale, Ouagadougou, Burkina-Faso, 425 p. Danusaputro M., Colter J.M. , Suharto P. (Ed.), 1991. Monetasi Pedesaan: Bunga Rampai Keuangan Pedesaan. Jakarta, Lembaga Pengembangan Perbankan Indonesia, 224 p. Dorléans B., 1992. L'Indonésie, les incertitudes du décollage économique. Paris, Les Etudes de la Documentation Française, 197 p. 31 Dury S., 1997. Les comportements d'épargne des ménages ruraux. Spécification d'un modèle dynamique et estimation sur données d'enquêtes (Java, Indonésie). Montpellier, INRA ESR, Etudes et Recherche, 108, 176 p. Dury S. , Lapenu C., 1995. Evolution du système financier rural à Java. Economie rurale, 227 : 34-43. Dury S., Vilcosqui L. , Mary F., 1996. Durian Trees (Durio zibethinus Murr.) in Javanese Home Gardens: their Importance in Informal Financial Systems. Agroforestry systems, 33 : 215-230. Fischer B., 1994. Financial Deregulation in an Open Developing Economy. Lesson from the Indonesian Experience. Savings and development, XVIII (3) : 257-280. Gentil D. , Fournier Y., 1993. Les paysans peuvent-ils devenir banquiers? Epargne et crédit en Afrique. Paris, Syros, 270 p. Hammad A.E., 1989. Islamic Banking: Theory and Practice. Zakat and Research Foundation, 45 p. Hoff K. , Stiglitz J.E., 1990. Imperfect Information and Rural Credit Markets. Puzzles and Policy Perspectives. The World Bank Economic Review, 4 (3) : 235-250. Holloh D. , Soetjipto S.M., 1991. Self-Help Groups in PHBK. Jakarta, Bank Indonesia, 35 p. Janvry (de) A., Radwan S., Sadoulet E. , Thorbecke E., 1995. State, Market and Civil Organizations: New Theories, New Practices and their Implications for Rural Development. London, Macmillan, 511 p. Krahnen J.P. , Schmidt R.H., 1994. Development Finance as Institution Building. A New Approach to Poverty-Oriented Banking. Westview Press, 154 p. Kropp E., Marx M.T., Pramod B., Quinones B.R. , Seibel H.D., 1989. Linking Self-Help Groups and Banks in Developing Countries. GTZ, APRACA, 151 p. 32 Lapenu C., 1992. Innovations institutionnelles en faveur des populations rurales dans le système financier indonésien. Montpellier, CIRAD-SAR, 49/93, 80 p. Lapenu C., 1996. Vers un nouveau rôle pour l'Etat et les institutions privées dans le développement du système financier rural indonésien. Thèse de doctorat, Economie Rurale, ENSA, Montpellier, 350 p. Lelart M., 1995. Tontines africaines et tontines asiatiques. Orléans, Institut Orléanais de Finance, Document de recherche, 22 p. Mac Leod R.H., 1992. Indonesia's New Banking Law. Bulletinof Indonesian Economic Studies, 28 (3) : 107-129. Martodidjojo Thohari E., 1991. Rôle du crédit agricole dans le développement de l'agriculture en Indonésie. Cas d'étude à Java Est. Thèse Universitaire, Montpellier I, 256 p. Maurer J.L., 1993. L'Ordre Nouveau et le monde rural indonésien. Archipel, 46 : 131-152. Mosley P., 1995. 'Optimal Incentives to Repay' in Institutions Lending to Low-Income Groups: an Indonesian Case Study. Savings and Development, XIX (3) : 257-278. Muldur U., 1991. Echelle et gamme dans les marchés bancaires nationaux et globaux. Revue d'Economie Financière, 17 : 167-196. Patten R.H. , Rosengard J.K., 1991. Progress with Profits. The Development of Rural Banking in Indonesia. San Francisco (USA), ICS Press, Sector Studies, 114 p. Riedinger J.M., 1994. Innovation in Rural Finance: Indonesian's Badan Kredit Kecamatan Program. World Development, 22 (3) : 301-313. Robinson M.S., 1992. Rural Financial Intermediation: Lessons from Indonesia. Part one: the Bank Rakyat Indonesia, 1970-1991. Harvard Institute for International Development, Development Discussion Paper, 434, 121 p. 33 Schmidt L.T., 1991. Rural Credit between Subsidy and Market. Adjustment of the Village Units of BRI in Sociological Perspective. Leiden Development Studies, 343 p. Servet J. (Ed.), 1995. Epargne et liens sociaux. Etudes comparées d'informalités financières. Lyon, Association d'Economie Financière, Cahiers "Finance, éthique, confiance", 295 p. Stern H.N., 1992. Le rôle de l'Etat dans le développement économique. Payot Lausanne, 158 p. Stiglitz J.E., 1986. The New Development Economics. World Development, 14 (2) : 257-265. Stiglitz J.E., 1992. The Role of the State in Financial Markets. Washington D.C., Institute for Policy Reform, 98 p. Syukur M., Suharto P. , Colter Y.M., 1991. KUM: the Grameen Bank Approach to Poverty Alleviation in Indonesia. Bogor, Centre for Agro-Socio-Economic Research, 35 p. Thorbecke E., 1992. Ajustement et équité en Indonésie. Paris, OCDE, 274 p. Traimond P., 1995. Finance et développement en pays d'Islam. EDICEF, Universités francophones, 255 p. Vittas D. (Ed.), 1992. Financial Regulation:Changing the Rules of the Game. Washington, D.C., Economic Development Institute of the World Bank, EDI Development Studies, 449 p. Woo W.T., Glassburner B. , Nasution A., 1994. Macroeconomic Policies, Crises and Long-term Growth in Indonesia, 1965-90. Comparative Macroeconomic Studies. Washington, D.C., World Bank, 216 p. World Bank, 1993. The East Asian Miracle. Economic Growth and Public Policy. Oxford, Oxford University Press, A World Bank Policy Research Report, 389 p. World Bank (Ed.), 1994. Indonesia: Sustaining Development. World Bank, World Bank Country Study, 216 p. 34 Yaron J., 1992. Assessing Development Finance Institution; a Public Interest Analysis. World Bank, World Bank Discussion Papers, 174, 63 p. Yaron J. , Benjamin, M., 1997. Recent Development in Rural Finance. In 23rd International Conference of Agricultural Economists, Sacramento, California, 10-16/08/97, World Bank, 1-20. Zeller M., Schrieder G., Von Braun J. , Heidhues F., 1997. Rural Finance for Food Security for the Poor. Implications for Research and Policy. Food Policy Review 4. Washington, D.C., IFPRI, 139 p. ENDNOTES 1These primarily have been "people's banks" operated by the West Java provincial government and BRI village units, which are heavily concentrated in rice-growing regions. 2In view of the pressures brought to bear on Indonesia's financial system (corruption, embezzlement, and lack of transparency) the reader is urged to view the figures provided with considerable caution. 3The historical analysis of the financial system has made it possible to establish a typology based on the private, public or semi-public nature of the institutions and on their geographical reach (local, provincial, or national). This typology was used to select the institutions surveyed in Bogor and Tasikmalaya. 4BPD, Bank Pembangunan Daerah, regional development bank. 5LPN in West Sumatra: BKK in Central Java, Bengkulu and South Kalimantan; KURK in East Java; LPD on Bali; LPK and BKPD in West Java; BUPK in Yogyakarta; LPK in Nusa Tenggara Barat (eastern islands); LKK in Aceh (North Sumatra). 6Although in terms of the number of institutions the public sector was represented by only seven national state banks and a number of rural banks, it was the largest sector in terms of its volume of business; in the early 1980s state banks accounted for 80 percent of total assets and disbursed 75 percent of loans granted. These institutions also received over 90 percent of the targeted and subsidized credits from the central bank; these refinancing operations thus provided them with their major source of funds (43.8 percent in 1983). 35 7Bank Indonesia, 1993; Cole and Slade, 1996; Lapenu 1996; McLeod, 1992. 8BPR (Bank Perkreditan Rakyat), people's credit banks. 9"Small credit" meant amounts under 200 million rupiahs (USD 100,000) for borrowers whose capital assets did not exceed 600 million rupiahs (USD 300,000). These amounts represented significant sums in rural areas and were to be used for SMEs rather than households. 10The term Bank Perkreditan Rakyat, or people's credit bank, was translated as "rural bank" in the English version of the Banking Act. This designation as a rural service provider primarily reflects the fact that these institutions were permitted to operate only at the subdistrict level. In reality, however, new BPRs were found mainly on the outskirts of urban areas. 11Farmers who sold their crops through the KUDs were automatically members. They therefore were obliged to provide funds in the form of compulsory savings, which explains that each employee served so many savers. In many villages, KUDs also served as channels for payments to the national electricity company. Accordingly, most village inhabitants also were members of the cooperative. 12The provincial government of East Java owned 222 KURKs (Kredit Usaha Rakyat Kecil, credit for activities of the poor). 13See the definition of general banks in the description of Pakto on page 6. 14Pos Pelayanan Desa: "village service post". 15However, village units remained targets for corruption; some employees were known to collect up to 10 percent of the total loan amount before paying borrowers. The financial sector was no stranger to this type of corruption which characterized the Indonesian economy. 16The subsidies were used to cover losses incurred under the BIMAS program, capitalize village units at the rate of 19 million rupiahs/unit, or a total of $33 million, and establish training programs. 17However, because the average loan was approximately US$1,000, or 114 percent of GDP, its services generally did not fall in the "microcredit"category. 18The P4K project (Pembinaan Peningkatan Pendapatan Petani dan Nelayan Kecil, project to raise the income of small farmers and fishermen); the KUM project (Karya Usaha Mandiri, project to generate independent activities). 19The data from the surveys conducted showed that from the beginning the BPRs had 16 or 17 employees, each of whom served only 20 to 50 36 clients. The staff of the BRI units surveyed numbered 5 or 6 employees, each of whom served 160 to 180 borrowers. 20Sunk costs are investments which cannot be recovered upon leaving the industry. 21On Java and Bali; competition currently is less intense on the more remote islands. 22Gold-owning households in both Tasikmalaya and Bogor are wealthier than the sample average and are in the high-income classes for 75 percent of the households. In addition, over half of them (56.2 percent) have deposits in formal institutions (primarily BRI and cooperatives), and the figure increases to 70 percent when all types of informal savings are counted. 23Surveys of clients in a group-based credit program, KUM, West Java, 1992. 37