D+C Development and Cooperation (No. 1, January/February 2001,
p. 8-11)

The wisdom in development finance has long been that lending to and saving by poor micro-entrepreneurs and farmers is doomed to failure: costs are too high, the poor are not creditworthy and they are not able to save and insure, and so on. Bold experimentation with new institutions in microfinance, supported by public action, have resulted in a number of success stories and changed this pessimistic assessment during the past fifteen years or so. In this paper, a number of arguments are put forward that call for a continuation of the support towards institutional innovation and bottom-up adaptation of Microfinance Institutions (MFIs) in developing countries.

The large and successful MFIs reaching the poor in developing countries have all relied on the support by donors and governments, at least during their formation stage. Because of widespread market imperfections concerning financial services to the poor, institutional innovation and expansion in microfinance is rarely market-driven, but a process that has been nurtured by the public sector or by altruistic leaders. Private R&D has little contributed to the microfinance revolution that we have witnessed in the past fifteen years. Some examples may substantiate this claim.

• The Grameen Bank. In Bangladesh, Professor Muhammad Yunus addressed the banking problem faced by the poor through a programme of action-research. With his graduate students in Chittagong University, he designed in 1976 an experimental credit programme to serve them. It spread rapidly to hundreds of villages. Through a special relationship with rural banks, he disbursed and recovered thousands of loans, but the bankers refused to take over the project at the end of the pilot phase. They feared it was too expensive and risky in spite of his success. Eventually, through the support of donors, the Grameen Bank was founded in 1983 and now serves more than 2 millions of borrowers.
• State-owned microbanks in Indonesia. Another flagship of the microfinance movement is the village banking unit system of the Bank Rakyat Indonesia (BRI), the largest microfinance institution in developing countries. This state-owned bank serves about 22 million microsavers with autonomously managed microbanks. They are highly profitable. The microbanks of BRI are the product of a successful transformation by the state of an state-owned agricultural bank during the mid-1980s.
• Credit union movement in 19th century Germany. Similar innovations took place in Europe in the past century. The concept of the credit union was developed by Friedrich Wilhelm Raiffeissen and his supporters over a period of 18 years, from 1846 until 1864. Their altruistic action was motivated out of the concern to assist the rural population to break out of their dependence from moneylenders and to improve their welfare. Since about 1870, the unions expanded more and more rapidly over a large sector of the Rhine Province, and also in other regions of the German States. The cooperative movement quickly spread to other countries in Europe and North America, and eventually, supported by the cooperative movement in developed countries and by donors, also to developing countries.

The above MFIs have something in common. They were not created by a market-induced process, but rather are the product of public action by the state, donors, and altruistic leaders, that facilitated social experimentation and institutional innovation. Recent policy reviews of poverty-oriented banking approaches rightly emphasize the important role that government, donors, and civic institutions together can play in forming MFIs. However, it is important to note that innovation and expansion of the microfinance sector could be improved in many cases in the future through more explicit partnership with the private banking sector.

Major types of
microfinance institutions

Institutional innovation in microfinance does not necessarily mean to create a new institutional type at the international level (as Friedrich Raiffeissen did), but includes the adaptation of an existing institutional type to the constraints and potentials of a certain client group in a specific local environment. In the following, the five major types of MFIs are presented

1.The cooperative model. The cooperative model was the first to introduce microfinance in developing countries, inspired by the successes in Europe and North America at the end of the nineteenth century. The main principles of the cooperative approach include:

• Participation: The cooperative members are the owners, contributing to the equity capital through shares. Loans are only granted to the members.
• Minimalist approach: The MFI solely focuses on the provision of financial services.
• Responsibilities: Members participate in major decisions and democratically elect officers from among themselves to orient and monitor the administration of the cooperative.
• Profit sharing: Surplus earnings remain in the cooperative in form of equity capital or are distributed to the members.
• Structure: Each cooperative is geographically limited. However, a cooperative can join a second financial level of organization (e.g., a regional union or federation), which can ensure supervision, refinancing and technical support among the federated cooperatives. A third financial level, the central union, may exist at the national level.

2. Solidarity groups. The second type of MFI are solidarity credit groups:

• Participation: Three to ten clients join a group to receive access to financial services (primarily credit). They may have to save before receiving a loan.
• Complementary services: In addition to financial services, the support agency may offer non-financial services, such as training or market information, to the group members.
• Responsibilities: Group members collectively guarantee loan repayment, and access to subsequent loans is given only if previous loans are paid in full.
• Profit sharing: The profits are not shared among the members, but used to build up group funds and emergency reserves. The ownership of these funds is often unclear. Intragroup savings accounts may be opened; they can be used according to the members' wishes.
• Structure: The group (three to ten members on average) can join a center (around five groups). The center allows for economy of scale in disbursement of loans, collection of savings or repayment, and training. The upper level (regional, or national) is in charge of decisionmaking (top-down approach).

3. Village banks. The village bank can be seen as a mix between the cooperative and solidarity group models, seeking to capitalise on the advantages of each. The village bank has usually fewer members than a cooperative, and is less formalised and complex in structure. Some international NGOs, such as FINCA, CARE, CRS, and CIDR (see also the article by Mathias Adler in this issue), promote the establishment of village banks. Their main form of credit guarantee relies on peer pressure among the members, like it is the case in solidarity credit groups. Other characteristics are:

Participation: The members of a village organise themselves to provide community-based savings and credit services. Membership fees may contribute to the equity capital of the bank.

Responsibilities: The village bank is a community credit and saving association managed by a committee elected by, and among, members.

Profit sharing: Profits are either distributed to members or used to increase the equity capital.

Structure: Highly decentralised institutional structure. Ideally, the village bank is independent of any upper level organisation for provision of financial products. The decentralised structure allows to serve villages in more remote areas.

4. The Linkage model. The fourth model builds on existing informal self-help groups (SHGs), such as rotating credit and savings associations described in the article by Sika and Strasser in this issue. The linkage model seeks to combine the strengths of existing informal systems (client proximity, flexibility, social capital, reaching poorer clients) with the strengths of the formal system (e.g. risk pooling, term transformation, provision of long-term investment loans, financial intermediation across regions and sectors). The main principles are:

• Participation: Members of a SHG enter into a group contract with a bank that provides savings and credit services to the group. An intermediary NGO may provide complementary services, such as training or certification of creditworthiness of groups.
• Responsibilities/profit sharing: The bank, sometimes assisted by an NGO, provides the services. Internally, the SHG may organize member-managed savings accounts.
• Structure: The SHG is linked to the bank through a group contract. Individual members of the SHG do not have any links with the bank.

5. Microbanks with individual financial contracts. The above four MFIs are member-based. Members contribute to a varying degree in the management, ownership and control of the MFI. On the other hand, microbanks, such as BancoSol in Bolivia, mainly rely on individual contracts between the institution and its client. This type of MFI is closest to the classical banks. However, the loan collateral may not be conventional, using for example savings of the client, knowledge on his/her creditworthiness, or other persons as guarantees of the loan. It is obvious that clients prefer to have an individual loan if they could get it on the same terms as those provided by member-based institutions, such as the first four types described. This is so because participation in any of the above MFI-types carries additional transaction costs on behalf of the client (for example for meetings). Yet, in order to reduce transaction costs for the MFI itself, member-based institutions can be more efficient in environments with lower population density, higher illiteracy, and poor road and communications infrastructure. It is therefore not surprising that microbanks saw their greatest success so far in urban areas of better-off developing or transformation countries. Because of the relatively high loan sizes of microbanks (see table below), it is unlikely that many poor (as defined by national poverty mappings) are reached. However, the somewhat better-off clients may not have any access to traditional banks, and loans to small enterprises certainly will make an indirect contribution to poverty alleviation. Some of the main principles of microbanks are:

• Participation: Clients are selected by the microbank based on creditworthiness.
• Responsibilities/profit-sharing: The client is individually responsible for loan repayment, and is not involved in management, ownership or profit-sharing.
• Structure: Emphasis is given on a decentralised structure that gives decision flexibility and strong performance incentives to managers of the microbanks.

The role of institutional
diversity and innovation

Three major reasons call for institutional diversity, and further investments in institutional innovation. First, there are multiple development objectives to be fulfilled by MFIs. Donors and governments that support the microfinance sector pursue three main objectives to a varying degree: Financial sustainability, outreach to the poor, and welfare impact.

The five institutional models above clearly differ in their outreach to the poor, or to the poorest, as will be shown in the table below. Our knowledge on impact of different types of institutions is so far negligible, but it is of course conceivable that MFIs which offer additional complementary services in form of business training or agricultural extension may achieve greater welfare impact than MFIs that focus on savings and credit alone. Impact, of course, will also vary with differences in financial products.

Second, the diversity of socio-economic contexts and the different levels of political, social and economic development require that the various institutional types are adapted to the local context. Such local innovation of institutions is presumably more successful if it is based on participatory processes with, by and for the poor. Microbanks and solidarity groups have here less to offer than the other three types as they mirror more of a top-down approach.

Third, the five institutional types shown above all have their justification because each of them is likely to have comparative advantages in reaching a particular socio-economic group, and work better in certain countries or socio-economic contexts than other types. The table shows the loan repayment rate, an indicator of financial sustainability, and a number of indicators of poverty outreach such as loan size, percentage of per capita gross domestic product (GDP) and percentage of female clients.

The data in the table comes from a postal survey that was conducted by the International Food Policy Research Institute (IFPRI) in 1999. The respondents of the survey were international NGOs involved in microfinance as well as national, regional and international microfinance networks. These respondents were asked a number of characteristics of the MFIs they support in developing countries in Asia, Africa and Latin America. Though less than half of the microfinance networks responded, the information provided a broad overview of MFIs by region or country. In total, the data refers to 1468 MFIs in 85 developing countries with an estimated number of 43 million savers and 17 million borrowers.The data suggests that the village bank, linkage model and the solidarity group reach relatively more women and poorer clients than the cooperative and the microbanking models. In terms of repayment rate, there are no significant differences.

Some concluding remarks

In the past 15 years or so, considerable progress in terms of outreach to the poor through a number of different types of MFIs has been made. However, most of the poor still do not have access to savings, credit and insurance services, and less than 2 percent of the population of the developing countries are clients of MFIs. At present, few MFIs are financially sustainable. Future innovation needs to aim at reducing transaction costs for MFIs and clients, and seek to improve services so as to increase outreach to the very poor and related welfare impact. For the poor, access to savings services may be more important than credit.

Institutional innovation in microfinance, and bold experimentation with adapting different institutional types to local environments, has led to a number of large, successful MFIs as well as new promising approaches, like the village banking and the linkage model. It is important to note that these innovations were not borne out of market forces, but relied heavily on financial support from the state and donors. The focus was on building cost-efficient MFIs that are congruent with market principles and that can reach poorer segments of the society as clients. Future innovation could further benefit from a greater involvement of the private banking sector.

The participants of an international workshop in 1998 in Accra organised by the German Foundation for International Development (DSE), the International Food Policy Research Institute (IFPRI), the International Fund for Agricultural Development (IFAD), and the Bank of Ghana summarised the call for institutional innovation in their concluding Declaration. A quote from this declaration reads as follows

"We appeal to governments, financial institutions, and development organisations to support the formation of human and social capital, and contribute to the start-up costs required for strengthening rural MFIs. We recognise that this will require a comprehensive and long-term institution-building effort, including the formation of microfinance networks. This effort has to focus on the organisational capacity of MFIs and a conducive regulatory framework laying out transparent and enforceable ground rules. The regulatory framework should be based on a consultative process that gives due recognition to different categories of MFIs, their potential, and their experience. Private-sector initiative and public support need to come together to achieve this objective; neither one will accomplish the task on its own".