CONTENTS

INTRODUCTION

1. FINANCE FOR SMALL-SCALE ENTERPRISE DEVELOPMENT

1.1 Credit schemes for small-scale enterprise development

1.2 The informal sector

1.3 Policies and regulations

2. THE FINANCIAL MARKETS

2.1 Informal financial markets

2.2 Linkages between SHOs and formal financial institutions

2.3 Savings and credit

3. SELECTED PROGRAMMES IN ASIAN COUNTRIES

3.1 Bangladesh

3.2 Indonesia

3.3 Philippines

3.4 Sri Lanka

4. CREDIT PROGRAMMES FOR MICROENTERPRISES IN LATIN AMERICA

4.1 Bolivia

4.2 Colombia

4.3 Mexico

4.4 Peru

5. FINANCIAL INTERMEDIATION SYSTEMS IN AFRICA

5.1 Ghana

5.2 Ivory Coast

5.3 Kenya

5.4 Nigeria

5.5 Togo

6. REVIEW AND CONCLUSION ON MAJOR ISSUES

6.1 Delivery systems

6.2 Transactional costs

6.3 Interest rates

6.4 Credit guarantees

6.5 Loan recovery

6.6 Graduation to formal lending institutions

6.7 Impact of programmes

BIBLIOGRAPHY

CHAPTER 1

FINANCE FOR SMALL-SCALE ENTERPRISE DEVELOPMENT

1.1 Credit schemes for small-scale enterprise development

Over the past quarter of a century or so, governments of developing countries have laid great stress on the development of small-scale enterprises (SSEs) in their programmes for economic development.

Advisors to developing countries have also emphasised that SSEs can serve vital development functions, such as:

1. to help generate employment by using more labour in relation to capital invested. This is specially important in developing countries where labour is abundant and capital relatively scarce;

2. to act as "seed-beds" for entrepreneurial talent, another element considered in short supply in the third world;

3. to operate in less populated rural areas with limited markets and poor infrastructure - a feature common in developing countries;

4. to be able to start up with very limited resources; "ease of entry" as this is called is a significant feature of small businesses especially in the services;

5. to provide "hands on" training facilities for people of varying levels of education in both management and technical skills;

6. to supply both low cost items for the poor and in certain circumstances high cost quality products for the rich and for export.

Some would add as an asset the flexibility of small enterprises that enables them better to weather recessions, material shortages and market changes.

Not surprisingly in the light of the above significant positive attributes, governments of developing countries have encouraged and supported programmes to help SSE development.

Programmes of financial assistance have been one of the principal ways in which such governments have tried to help SSE development. Until the mid-1980s most of this assistance was directed at - or at least ended up with - the larger or medium-scale formal SSE units.

As an example one might cite the small enterprise development projects supported by the World Bank that channelled during the 15 years 1973-88 more than US$3 billion in 70 projects to help small and medium enterprises in 36 developing countries. Probably less than 10% of this sum actually reached the very smallest of the SSE, that is those with less than 10 workers and assets of less than US$10,000. To reach these borrowers one would have to lend sums of the order of US$1,000-US$2,000 or at least less than US$5,000. As the average loan size was US$24,287 in Latin America, US$31,811 in Asia and a staggering US$83,408 in Africa, it is patently clear that these projects reached few of the very smallest firms or microenterprises.

In the same period, most other programmes of financial assistance to SSE, whether using government funds or external donor funds - both multilateral and bilateral - suffered from this same bias towards the larger, more formal units of the SME sector. This was principally due to the channelling of these funds through development or commercial banks in one form or another. These institutions were only willing to participate in such programmes or projects if they were able to target their lending to the largest group of borrowers that conformed to the category of formal small- and medium-scale enterprises as defined.

To explore further the relationship of the banks to the small-scale enterprises it is worth repeating again the well known reasons why small enterprises have difficulty in obtaining access to credit from the formal banking system. These may be categorised as follows:

1. Lending to small enterprises is perceived as being risky. The uncertainties that face a small industry, the high mortality rate, the susceptibility to market changes, and to economic fluctuations, make banks reluctant to deal with these clients. Non-payment, or even delayed payment, by a few major clients can cause the collapse of a small business.

The mortality rate of small businesses is indeed high. This was estimated in a survey in the Philippines to be over 4% a year. In the developed countries the figures are even higher: somewhere between 20% to 50% of new small enterprises in the United States fail within the first one or two years of operation. Various studies have shown that small businesses are no less creditworthy than the larger enterprises, but even though small-scale businessmen may take great pains to repay their loans in order to maintain their credit worthiness, in expectation of increased borrowing, these enterprises usually depend on a single-owner manager. This makes them more vulnerable to what might happen to the manager, which would obviously not be the case in a larger industry.

2. Parallel to the reluctance of banks to lend to small enterprises is the reluctance of these enterprises to borrow from banks. The administrative and costly formalities of obtaining bank finance, particularly the time and paper work involved, are a formidable deterrent to most smaller businesses. Some of them lack the formal education to cope with the bureaucracy and others, compounded by problems of location and time pressures have difficulty in complying with what the institutions require before they grant a credit. In many cases, potential borrowers have to pay for preparation of accounts or special studies on top of the cost involved in the numerous visits to the lending institutions. The transaction costs on the part of a borrower in obtaining a loan from a bank may be proportionally as high as are these costs for the lender.

3. There is a distinct institutional bias on the part of banks towards lending to the larger corporate sector. In many cases there are links in directorships, joint ownerships and various other common financial dealings between banks and the large enterprises and automatically this induces preference for directing finance to these borrowers.

4. The administrative costs of lending to small enterprises are high, which cuts deep into the profitability of such transactions for lending institutions. This is undoubtedly the case and has been borne out in studies.

A World Bank study in the Philippines in 1978 showed that whereas the administrative costs of handling large loans was in the order of 0.3% to 0.5% of the cost of the loan, such costs rose to a range of 2.6% to 2.7% when lending to small enterprises. Most subsequent studies have shown that most lending, even to formal organised small businesses, requires administrative costs of closer to 4-5%. One institution in Latin America - the Corporación Financiera Popular of Colombia - which lent exclusively to small and medium enterprises - had administrative costs of over 7% and this institution hardly lent at all to the very smallest microenterprises. As the transactional costs must take account also of the risk element and potential losses, these overall costs are even higher. Despite what has been said about the credit-worthiness of small entrepreneurs, there is no doubt that several small enterprise lending programmes in the past have suffered from a high level of arrears and defaults.

5. Small enterprises are unable, or unwilling, to present full accounting records and other documentation called for by banks. In most cases such records just do not exist, making appraisal of loan applications difficult. One may quote the example of the Development Bank in the Philippines (DBP), when a study in 1980 showed that it usually took four to six weeks for a professional staff worker to collect the necessary data to make a recommendation and to come to a decision on a loan to small or medium enterprises, largely because of the poor records kept by the borrower.

6. Usually such small borrowers are unable to provide the collateral and security demanded by lending institutions before approving loans.

These are all real problems and it would serve little purpose to ignore them and hope to overcome these difficulties simply by exhorting the banks to adopt a more understanding attitude to the needs of the SSE Sector.

Most approaches that try to expand access to finance to the small microenterprises, address these important issues along two lines: first, is a governmental imposition on the banks mandating them to lend a minimum percentage of their portfolios to small-scale borrowers, and second, to develop new institutions, mainly non-governmental organisations (NGO's), to act as further intermediaries either as short-term lenders or to act between the smaller borrowers and the banks in an effort to reduce the transactional costs of such lending. A few countries - Colombia, Korea, Turkey - have created special government banks to lend to SME, but while this has increased lending to the larger SMEs, these institutions have understandably suffered from the high cost of such operations bearing in mind the high transaction costs of the small-scale borrowings.

The mandating of quotas of lending to SMEs also has its limitations as it has proved costly and difficult to monitor the banks' compliance. Many economists also claim that mandating quotas for lending to specific groups distorts the whole financial system and can have severe repercussions on the profitability of the whole financial sector. Such mandating also does little to open up access to finance for the very small microenterprises.

In general, through the provision of low-cost government funding and refinancing arrangements, the larger formal SMEs have managed to obtain some increased access to finance from commercial and development banks, but these special credit schemes, many at subsidised interest rates, have benefited only a small proportion of the intended target groups. It is claimed that in a number of cases, a high proportion of those who benefit are politically well connected.

Latterly, more of such credit schemes for SMEs have lent at close to commercial rates but here, too, access has been limited to those with sufficient collateral.

"Microenterprises", the term given to the smallest of the SME sector, have generally benefited little from all these schemes. To understand the special problems and needs of this vast underclass of tiny units one has to look closer at the special characteristics of this subsector and of what has been called in many countries the "informal sector".

1.2 The informal sector

The term "informal sector" was first used in 1973 in the report of an ILO employment mission to Kenya. Prior to this, little or no account had been taken by development economists of the very considerable part of the societies - some times over 50% - of most developing countries that operated at the margin of the economy in the form of large numbers of microenterprises, artisans, cottage industries, household businesses and self-employed persons.

The "informal sector" is difficult to define but, as has been commented by many economists and sociologists, there is little difficulty in knowing what it constitutes. It is immediately recognised in all countries.

In the early 1970s many development economists dismissed this "informal sector" as "largely parasitic" engaged in "unskilled and obsolete activities" which made little contribution to the national economy.

Since the 1973 ILO Kenya Report most development economists have taken a more benign and positive view of this "informal sector". Some begin to see it as a "political" source of strength, in whose workshops and tiny units "practical skills and entrepreneurs' talents were being developed at low cost" (ILO report, 1973).

Legality, claims Hernando de Soto (the Peruvian analyst of his country's large "informal sector", author of "The other path"), is at the heart of the concept of "informal and formal" sectors. The "informal sector", says de Soto and those who follow his views, is simply a euphemism for the "illegal sector" - basically illegal enterprises which have legal objectives. They are forced into illegality by a complex web of laws or regulations with which they are totally unable to comply. Compliance, or each attempt to comply, would, it is claimed, cripple any likelihood of their operating at all. According to this argument, the legal system with its imposed restrictions has caused the informality (or the illegality) of these businesses.

There is, however, another explanation of the "informal sector" by other analysts. It represents the failure, they argue, of the more "modern market economy" as it develops to provide jobs and livelihoods for large sections of the population. These individuals and groups left out of the development of the marketing economy must, in an effort to survive, develop entrepreneurial initiatives in traditional forms but on a tiny, very small scale.

This latter view assumes a dual economy - a formal and an informal economy - developing side by side until the formal economy gradually absorbs the informal as it develops.

The concept of the dual economy undoubtedly has some validity, but in most parts of the developing world the "informal sector", far from disappearing, has expanded and the formal sector is relatively stagnant, if not actually contracting. Growth of the formal sector where it has taken place is on a relatively small scale.

As an example, the so called "informal sector" composed almost entirely of microenterprises (the subsector of the smallest units within the SME sector) has grown in most Latin American countries. It is reported that in 1950 informal activities in Latin America occupied 30% of the economically active population and by 1980 this had grown to at least 50%. At the beginning of the 1990s over 50% of the population live off these "informal sector" activities in countries such as Bolivia, Colombia, Ecuador and Peru.

A number of economists have explored the relationship between the two sectors and in particular the role of the state. All types of structural benefits and favours flow from the state government to capital intensive, large modern firms. Highly regulated markets with low competition, subsidies and privileged access to credit and foreign exchange are a few examples of such favours. The privileged access to resources combined with limited competition are usually denied to informal sector firms. Because of these barriers, informal sector firms necessarily tend to use a higher proportion of indigenous factor inputs, labour intensive technologies and, ultimately, they generate lower incomes.

A fallacy believed by some economists has been the assumption of a free market economy in the Third World. Overly regulated, monopolistic or oligopolistic state-nurtured markets, bureaucracies, state corporations and numerous parastatals render unrealistic any economic reasoning based on a free market model in most of the developing world.

Both the formal sector and the state stand in a mutually protected and exclusive relationship to one another. The formal sector depends on the state, as the state depends on the formal sector. This mutual dependency has in the past taken the form of subsidies, tax breaks, special incentives and support, import and export aid, all under a cloak of political legitimacy. For the formal sector, economic success appears less a function of entrepreneurial talent, but rather one of gaining access to, and obtaining, the preferential treatment of the state. With the advent of structural adjustment and some reforms in the late 1980s and early 1990s some of this is undergoing change, but at a slow pace.

In contrast, the informal sector has been excluded from the distribution of state benefits. Located outside the area of law enforcement, it is claimed by many that the informal sector often presents virtually the only real independent sector in developing urban economies.

The informal sector has no access to goods and services distributed by the political system. Small-scale entrepreneurs in the informal sector have to rely on themselves to a far greater extent than those in the formal sector. These small-scale operators are entrepreneurs in the classical Schumpeterian innovative sense in an economy characterised by extreme scarcity. They are innovative and flexible operators not so much because of any personality disposition, but simply because the business environment in which they operate requires such skills for economic survival.

The first steps towards "self-employment" or starting up a microenterprise in the "informal sector" rarely, if ever, call for borrowing from an institution. Overwhelmingly these microenterprises rely on their own, their friends', or their family's savings, which overall add up to the major part of total small investments in any developing country. Informal credit sources (including money lenders, pawnbrokers, rotating savings and credit associations - ROSCAs - wholesalers and shopkeepers, etc.) play an important role in financing microenterprises both when they start up and later in the early stages, when they need more working capital.

As shown by the cases presented later in this publication and taken as examples from the experience of developing countries in different regions, there is now a great variety of credit programmes for microenterprises in the "informal sector" run by poverty focused banks, special NGO's, or banking type operations (Grameen, KUPEDES-Unit Desa, BRAC, BKK in Asia and the various foundations in Latin America). Some have reached impressive numbers of borrowers representing a significant percentage of the eligible target groups. Others have been limited to microenterprise borrowers in their immediate localities. The economic and social impact of their programmes is dealt with in greater detail in a later chapter, so is the all important question of what success these financing programmes have had in graduating the microenterprise borrowers into the mainstream of formal commercial banking.

In general most of the funds in these lending programmes were requested as working capital for microenterprise operations rather than for new "start-ups" or for fixed asset investment. Although there are some exceptions most of the bigger loan programmes have - or are approaching - acceptable loan recovery rates. As can been seen from the cases this has largely been achieved through the use of group pressure to enforce repayment and by using the carrot of larger loans to those who repay on time.

Most of the more successful lending programmes aim at poverty alleviation and enhancement of income rather than business development as such. Few programmes have developed links with formal financial institutions although many have expressed this as an aim. There is also evidence that some credit programmes have been reluctant to transfer good clients to the banks for fear, no doubt, that this would adversely affect the financial viability of the programme in the course of time.

1.3 Policies and regulations

Financial policies of governments have in the past mainly relied on trying to cheapen credit for SSEs by lowering interest rates on SME loan programmes. Development banks were set up by governments to provide long-term finance for investment but these institutions generally adopted procedures suitable only for large projects. Development banks were discouraged or even prohibited from mobilising deposits so that they became totally dependent on government funds and donor institutions. Usually, the development banks soon faced financial problems and thus influenced these institutions to follow conservative lending policies. This meant lending to larger investment projects supported by the government rather than extending credit to small enterprises.

By setting a ceiling to the interest rate at a level well below the market rate, formal financial institutions were unable to cover costs when they tried to lend to small enterprises. Such lending became completely unprofitable. As the example below on lending programmes in Bangladesh shows, interest rates well above those of commercial banks in no way deterred small enterprises from applying for loans.

Cheap subsidised credit led to political influence in the choice of borrowers and loans frequently came to be regarded as gifts which did not need to be repaid. Recovery rates of less than 50% were not unusual in such schemes and this served to confirm the contentions of those who argued that SSEs were too poor, or were unable or unwilling to repay loans. As loan recovery rates dropped and debt collection was not pursued, those who were repaying their loans were demotivated from continuing their repayments.

As several of the cases described in Chapters 3 to 5 indicate, credit programmes without a savings component tend to lead to poor payment morale. This problem is compounded further by a policy of lower subsidised interest rates on loans to SSEs.

Over regulation by governments, as graphically described by Hernando de Soto, leads to a perpetuation and expansion of the "informal sector". Enterprises in the "informal sector" lacking any legal status face even greater problems in obtaining finance.

In the short term, deregulation could help legitimise a large number of enterprises and this would remove one barrier to institutional finance. The problems referred to earlier of high transactional costs, perceived risks and lack of records and collateral, already cause SSEs enough worries in trying to obtain credit from banks. Lack of legal status makes such access virtually impossible.

Liberalisation of the financial system would help some banks (e.g., cooperative banks, people's banks, savings banks, development banks, etc.) and other non-bank financial agencies to mobilise more deposits in the form of savings, by borrowers and others, and thus increase resources available for lending. Such a policy would also free interest rates so that lending institutions would have a better chance to recover costs and might even make lending to SSEs modestly profitable (as in the BRI/Unit Desa example in Indonesia).

CHAPTER 2

THE FINANCIAL MARKETS

2.1 Informal financial markets

Informal financial markets comprise two major groups:

- self-help organisations (SHO) (such as credit unions and cooperatives and associations, mutual support groups, etc.);

- individual financial brokers (money lenders, pawnbrokers, etc.).

Self-help groups may be formed specifically for mutual financial help or they may have a wider role in mutual support, financial assistance being only one of several different forms of help given.

For non-farm beneficiaries individual financial brokers are of two main types:

- specialised financial brokers where providing credit against security is the primary business; or

- trade-related financial brokers in which finance is related to other commercial activities such as materials purchasing, sales of products, etc.

Moneylenders are individuals who provide credit without collateral and usually with only a minimum of administrative procedures, if any. Interest rates are high (rising sometimes to 100% a month or higher depending on commercial bank rates), varying with the client, loan purpose, size of loan, and availability of funds. Small loans from money lenders may cost as much as 10% per day in some parts of the world.

There are also some trade middlemen who act also as moneylenders. Some of them are traders selling products, such as materials, machines or implements or other inputs, on credit. In agriculture, this is sometimes combined with the advance purchase of outputs, where sales to farmers on credit obliges the latter to deliver his crops at harvest time. Some moneylenders also help to market the products of craftsmen to whom they make loans. Interest rates are usually between 10 and 15% per month, with extreme rates up to 50% or even 300%. In other cases middlemen act as moneylenders to finance purchases made by small shopkeepers from wholesale traders or importers or for the purchase by artisans of materials.

From a study of formal and informal financial institutions in Indonesia, the World Bank concluded in 1983 that:

"the traditional assessment of the informal sector (money lender) has been that it is exploitative, usurious, and disfunctional, and newly established institutional credit is thus often aimed at replacing the informal sector. Such an approach depreciates the functional role that informal lenders play during the incipient development of rural financial markets and may result in policies which displace informal lenders without offering alternate sources of institutional credit... In these circumstances, rather than substitute for these informal arrangements, formal arrangements can, and should, complement these arrangements, supplement and strengthen informal credit sources, and compete with informal lenders. There is greater economy in this approach as it will avoid the costly duplication of intermediation resources. Moreover, newly introduced systems could also learn from the informal sector by addressing those aspects of credit previously ignored by informal arrangements."

The most widespread and important type of SHO and informal financial institution (IFI), is the savings and credit association which exist in most Asian and African countries and in most ethnic groups within these countries. Originally savings were kept in cash or in kind to help each other in case of emergency or to distribute the accumulated items or funds in a rotating order. In recent decades, there has been a general shift towards purely financial institutions involved in savings and credit rather than in insurance, which is now secondary.

Savings and credit associations exist in developing countries at all levels of the population, in urban and rural areas, among traders and market-women, farmers, fishermen, small industrialists, wage and salary earners, even among bankers. The widespread existence in all countries - developed and developing - of credit unions among groups of salaried employees in large organisations is also an example of such SHO. In recent decades, many savings and credit associations have expanded their activities to cover various financial services, mainly to members, but sometimes also to non-members, for the purpose of house building, social insurance services for medical treatment, and for a variety of other expenditures (school fees, marriage costs, purchase of large consumer items, etc.).

Some SHO set up funds for specific activities from regular contributions, membership and registration fees, interest from loans, fines for late payments, and proceeds from communal production or hired labour.

Savings and credit associations which have been operating successfully for decades can generally be divided into the following types:

Rotating savings associations

The procedure is simple. Each member pays in a fixed amount at regular intervals and then, one after the other in turn, each member receives the whole amount at a time. The cycle ends when all members have received the total at least once; then next cycle begins.

Rotating savings and credit associations

Each member pays a fixed amount at regular intervals. Part of the sum collected is allocated to each member in turn in rotating order; the rest is put into a general fund for loans, emergency insurance, etc.

Non-rotating savings associations

Each member pays a fixed or variable amount at regular intervals. The contributions are deposited and paid back to the individual members at the end of the stipulated period.

Non-rotating savings and credit associations

Each member pays a fixed or variable amount at regular intervals. Income from sources such as contributions, fees, fines, joint labour or business, etc., is put into a fund, which may be used for loans, insurance and social services. The fund may be established for a specified or unspecified period. Contributions may or may not be paid back at the end of a stipulated period.

SHO are usually defined by the following criteria:

- they are voluntary associations,

- they have an economic objective,

- frequently they have social objectives as well (e.g. helping weaker members of the community),

- they are permanent bodies beyond just organisations for ad-hoc actions,

- they have a registered membership,

- members have equal standing, elected leadership, and democratic control by the members. They are governed by rules and regulations, by-laws or a constitution which is usually in writing.

Most SHOs are informal organisations. They are organised and subject to rules but are not formally or legally registered, so that they do not have full legal status. They may, however, be considered to be semi-formal organisations which fulfil some but not all formal sector requirements.

SHOs may develop from local indigenous origins, or be initiated by governmental or non-governmental organisations, or possibly a mixture of both.

While SHOs may be considered to belong to the informal financial sector, and may often be informal in the sense of lacking official legal status, they usually do have an organisational structure. Typically they are headed by an executive committee of at least three: the chairperson, a secretary and a treasurer. Officers in SHOs are usually of local origin, elected by the members, and remain in office until the association decides otherwise. The fact that elections are held does not always prevent influential members of the community from dominating in practice and having themselves voted into office. In SHOs initiated by outside agencies, officers may be either elected or designated, or there may be a combination of both, sometimes with a difference between what is set out in the rules and what happens in practice.

These SHOs vary widely in size. The minimum number of members is rarely below 10, and mostly below 100. Groups of over 100 members may be subdivided into sections, each with its own administrative structure. Membership is usually voluntary, though in a few community based associations it may be compulsory.

Almost all SHOs are governed by rules and regulations, by-laws, or a constitution. Such rules may include regulations concerning membership, organisational structure, voting procedures, accounts, control, savings contributions, lending procedures, interest payments, penalties, etc.

There are usually three primary financial functions of SHOs: savings, loans and insurance (emergency) services. Less frequently there may be additional economic functions, such as reciprocal labour, joint production or other forms of common investment (common purchasing, marketing). These SHOs may have a specific social role in providing help to members facing special problems.

Loans are usually given to individual members or to the group as a corporate body or to both. In addition, they may on occasion be extended to non-members, either for social or for financial reasons, or to the community as a whole. In fact, some SHOs also act as community development organisations.

Disbursement is either immediate, as in the case of rotating savings associations, or deferred, as in the case of savings and credit associations. In the former, the risk involved is minimal; in the latter, where a fund is accumulated, special care must be taken to prevent embezzlement. Funds may be kept by elders or recognised and trusted leaders, or, more frequently, deposited in a bank account to earn some interest. In many cases basic rudimentary controls are maintained such as the treasurer keeping the money and the secretary keeping the books, both possibly under the supervision of the president. Many SHO transactions take place in public at regular meetings so that they are widely observed.

The mode of payment is either immediate, at regular meetings attended by all members, or intermediate, i.e. to a collector, a treasurer or the president.

The more members there are within a SHO, the longer the cycle and the lower the regular contribution; fewer members means higher regular payments. Rotating savings associations may therefore restrict the number of members, while savings and credit associations will try to increase the number of members as much as possible. Multiple membership in several associations is common, thus increasing the total amount of savings per individual per time period and at the same time spreading the risk.

Loans from SHOs are generally short-term, with maturities between one week and several months. Virtually all SHOs charge interest on loans. Interest may either be calculated as a percentage of the capital per time unit, usually per month, or as an amount of money per amount of loan and loan period. Interest rates charged by SHO are usually market rates, with the result that they are higher than subsidised rates charged in the formal financial sector when initiated and controlled by the government. Small and very small short-term loans usually carry a much higher interest charge than larger and longer-term loans reflecting the different proportions of transaction costs in terms of total costs.

Loans are frequently secured through one or several guarantors, who stand in with their savings if loans are overdue, or through depositing some specific collateral. Confiscation of savings and personal property and subsequent compulsory auction of a debtor's property after due warning may be the ultimate recourse of a SHO. However, this problem is relatively unknown to most SHOs, where such defaulting rarely occurs. Where it does, a SHO may evaluate the circumstances and then decide whether to reschedule the debt or to take more drastic action against the debtor.

2.2 Linkages between SHOs and formal financial institutions

As can be seen in some of the cases, an objective of most SHOs is to reach a stage where they can link their activities in some way with the formal banking system. This linkage may take place either directly between the SHO and a specified bank or number of banks or indirectly through the intermediation of private voluntary NGOs or, in some cases through governmental or public agency intermediaries as well. Financial linkage between a bank and a SHO may be between the savings activities and the credit operations of the organisation.

Before a link can be established between a SHO and a bank one would expect the former to have been engaged in regular savings activities over a period, usually at least a year, and to have built up a loan fund as well as some direct lending from resources of its own generated through a savings or a deposit programme. One would expected the SHO to have established a suitable organisation, whether formal or informal, and displayed social cohesion with responsible leadership and effective management. On the other side, the NGO or government-based small industry promotion institution could act as the intermediary between the SHO and the bank. Ideally, these organisations should have had some previous experience in working with SHOs. Also, the bank concerned should have a suitable branch network, well established savings and credit activities and the capacity, experience and willingness to work with small enterprises.

Generally, it is the professed long-range objective of SHOs to establish linkages with a bank so as to make the individual small enterprise borrowers bankable clients and obtain direct access for them to the services of a formal financial institution. There is of course the possibility that the formal financial institution may not be a bank of long standing commercial experience, but rather a non-banking institution that has been established in part through the activities of a SHO. Formal financial institutions established by SHOs may be either through the upgrading of an SHO to formal status so that it becomes a formal savings and credit cooperative, or through the setting up of a financial apex organisation by a group of SHOs. Such formal institutions established by SHOs may, in turn, set up linkages with primary banks.

The lack of legal status is one of the core problems facing SHOs attempting to establish links with formal financial institutions. A few are officially recognised but the majority are unregistered and unrecognised. Registration may only be a formality and the NGO or government institution concerned may in practice recognise the SHO unofficially as having a legal status.

The legal status of SHOs can however sometimes be a more complex problem. Clearly only prevailing legislation on registration and official legality can afford an institutions' legal status and the protection of the law, but in many countries, particularly the less developed of the developing countries, social pressures can be stronger and more effective than the law enforcement mechanism based on the existing formal legal framework. As shown in some of the cases presented later, there are many circumstances where groups of people would not violate a mutual obligation incurred by an informal organisation but might default as a member of a formal cooperative society or as a borrower from a formal bank. Despite infringement of the law, in many cases the defaulter may go unpunished.

Despite these diverse attitudes to obligations within developing societies, the fact remains that banks often find it difficult, if not impossible, to deal directly with groups that do not have legal status. Banks are bound by their management to rely heavily on legal contracts. Informal groups without legal status are generally precluded from opening saving accounts and cannot obtain group loans. There have been successful attempts in some countries to find suitable legal forms to give a cloak of legal status to SHOs. These can take the form of cooperative societies, foundations, or voluntary associations that can be registered under the Society Act, or as non-profit-making associations, or even in some cases taking the legal form of a business enterprise.

Lack of legal status has not always been an insuperable barrier in establishing linkage with a bank. Sometimes the senior representative or chairman of an SHO may act on its behalf and open a savings account or deposit money in a bank at his own risk, or the members may form a joint mutual liability group and all sign on joint financial transactions such as group deposits or group loans. And of course, in a number of cases, the SHO may link itself directly to a private voluntary organisation or NGO or even to a government organisation which will take full legal responsibility.

Of course these solutions are not appropriate for more comprehensive linkage arrangements. A representative of the SHO acting on its behalf takes an unreasonable risk; the solution whereby members form a joint and mutual liability group tends to be cumbersome and transaction costs are heavy. It is feasible for the SHO to attach itself to a private NGO, but it is not so straightforward for most of these organisations to find a suitable organisation with which to develop such a relationship.

There are, of course, alternative solutions. SHOs can be given legal status if the national government enacts specific laws, nationally, regionally or provincially to enable SHOs to act, for example, as credit guarantee groups, where the loan contract is established directly between a bank and a small enterprise and the SHO guarantees the transaction.

It is important to make maximum use of the vast reservoir of SHOs with their experience and local and regional acceptance in developing financing programmes for micro-businesses. Some legal form should be worked out whereby they may be registered, possibly under a provincial or regional law and be given the legal status they need to carry on transactions with formal financial institutions such as banks. Ways should be sought whereby banks and SHOs can contract legally binding transactions of the type being considered here. This may require appropriate banking legislation. Certainly as a start, the legal status problem can be overcome by linking banks and SHOs through the mediation of a legally registered NGO or governmental institution.

The process of linking informal financial institutions and NGOs with formal banking institutions can take various forms. Mention has already been made of the arrangement whereby NGOs and SHOs use commercial banks or other formal financial institutions to disburse loans, collect repayments and maintain data and records on borrowers, even though the loan risk and repayment responsibility remain with the lending organisation. This inevitably leads to closer links between borrowers and the bank and in the course of time may bring about some direct lending from the financial institutions to the most advanced and reliable borrowers.

Some links with commercial banks are established by NGOs and foundations when the latter borrow directly from the banks and then on-lend smaller loans to individuals or group borrowers. In this case the lending organisation usually maintains a deposit of some size with the bank. The deposit can also be used as a guarantee in place of collateral for a number of small borrowers. With a suitable relationship developing over a time, a commercial bank may agree to a certain leverage in the guarantee provided by the deposit, namely that the deposit is used to cover three to five times the sum in small loans. Usually the bank agrees to this arrangement for small borrowers who have a good banking record. The borrowers most likely to benefit from such an arrangement would probably have received and repaid on time the maximum loan allowable within a credit programme and are thus ready to graduate to the formal financial system.

2.3 Savings and credit

As has been shown in several of the examples of different financial support systems for small enterprises in developing countries, many of them successfully link savings and credit. Such links can take place in two ways: (1) a joint credit and savings scheme in which a loan is given only when an individual meets specific savings requirements; (2) separate savings and credit schemes, each with its own set of clients, where the bank mobilises its own resources through a savings or deposit scheme, but does not require borrowers to meet specific savings requirements. Most banks in fact operate in the second way and so the most specific form of financial innovation lies in the first approach, namely that of directly linking the granting of a loan to specific savings requirements.

Banking regulations and financial legislation differ from country to country and in some developing countries (e.g. in some parts of Latin America) organisations other than registered banks are severely limited as to the type and amount of savings deposits they can take in. In these circumstances, most savings are mobilised as a requirement to obtaining a loan as is usually the case in credit unions or cooperatives.

Contrary to general belief, savings do take place among the poor and in low income strata of developing countries, but most is through SHOs. In fact, many of these have their own credit delivery systems, using the resources mobilised through savings. Of course the capacity of low income, poor, small-scale businessmen or women to save is limited and, as a result, the funds that are generated by such savings schemes are inadequate for the credit needs of the SHO members, but savings are needed and not only for the creation of internal resources.

Savings is a means of motivating borrowers to consider more seriously repaying their loans since they have in fact a vested interest, however small, in the financial sustainability of the organisation from which they have borrowed. There is also considerable evidence to show that where credit is given with a savings requirement, there is greater commitment to using the loan for income generation. There is also less inclination to invest the resources borrowed in risky business ventures. There is an established direct relationship between savings and higher repayments of loans. The accumulation of savings by a group of SHO members provides a fund which can also be used as collateral to guarantee credit for one of the group.

The credit-savings link determines a distinct relationship between a member and the SHO to which he belongs. The SHO may in fact use the resources saved in different ways. Some may be set aside in an internal loan fund for immediate re-lending to members, or part of the internal fund may be deposited with a bank, NGO or promotional agency to serve as blanket collateral for loans to the SHO as a whole or to individual members thus using the savings as a mutual guarantee fund.

The credit-savings schemes of SHOs can be described as a form of saving to which an SHO member commits him or herself with the objective of obtaining a loan from the organisation. This loan may be less or equal to the amount of savings, but in most cases there is an agreed ratio which allows individuals to borrow up to an agreed multiple of the savings made depending on the resources available to the organisation. The organisation's loan fund may be made up of the savings of members and borrowers, interest earned, other income, outside grants (if any) and the organisation's own borrowings.

Most state run credit programmes do not make any stipulations as regards savings requirements. In fact, those credit schemes that are run by development banks make savings deposits impossible since the institution concerned generally does not have the right to mobilise deposits. There are some exceptions. There have been moves by some commercial banks to require a new small entrepreneur to open a deposit account before applying for a credit but these have usually been efforts to develop a form of collateral. It should be recognised however that all compulsory saving as a requirement for obtaining a loan has an element of providing collateral even if the amount does not cover the loan. It should also be understood that such a savings requirement inevitably makes the loan more costly since the interest earned on the deposit is usually less than the real value of the money deposited or the opportunity cost of the funds to the borrower.

The right of a member to obtain a loan many times greater than the amount he has saved in the lending organisation (SHO or NGO) gives him or her direct encouragement to save as much possible. If the organisation only allows a loan equal to the amount of the savings there is less incentive because theoretically the member may withdraw his own money and use it and thus get no extra benefit from the savings, although it has been shown that, even in this situation, the borrower is more likely to repay the loan over a period and is encouraged to accumulate some limited capital through the savings mechanism. The agreed multiple relating the loan size to savings deposits is usually set so that it may increase as members successfully repay a loan. The actual loan size may increase up to three (or even more) times the amount saved. Because of the limited resources usually available to an SHO, the multiple can never be very high and there will always be a constraint on the amount of lending that can take place within an SHO/NGO. SHOs seek linkages with NGOs or formal banking institutions so that the resources available for loans can be much higher than the limited amount generated internally. Most SHOs in direct lending schemes set a maximum value not only of the multiplier between savings and credit but also on loan ceilings to limit the potential drain on resources and increase opportunities for new borrowers. Resources available will also depend on the onlending interest rate, and more and more SHOs are learning that setting an acceptably high interest rate with relatively easy access to loans, works to the benefit of all concerned, even though in most cases the interest set is not much higher than the full commercial rate prevailing in the country. Because of the short repayment period, the manner of calculation and sometimes of payment and collection, the annual interest rate may be considerably higher than that set in the loan transaction.

The most desirable form of linkage between an SHO and a bank or with an NGO operating as a credit institution, is that of a dynamic financial linkage model, in which cycles of increasing savings lead to multiple cycles of increasing loans, with corresponding cycles of timely repayment.

The model suggested is one of gradual growth based on the experience that the loan repayment capacity of small enterprises is directly related to their investment capacity, both being linked to the ability to save. Investment capacity is not only a function of internal capital formation and access to credit, but also a function of business experience which in most cases is with small enterprises. This is slow to develop.

The incentive to save and so accumulate capital is more likely to be effective if increased saving leads to larger loans. The approach of maintaining loans at a low level in relation to the volume of savings may appear to be a prudent policy but the result will, in the long run, be counter productive. The incentive to save will in the long run be closely linked to the possibility of obtaining bigger loans and the ability to repay the loans. Low upper loan limits more or less preempt the likelihood of a totally client being able to graduate to direct borrowing from a formal banking institution.

CHAPTER 3

SELECTED PROGRAMMES IN ASIAN COUNTRIES

3.1 Bangladesh

Bangladesh with its 109 million people, 90% of whom live in rural areas, is one of the poorest countries in the world. Land distribution is highly uneven and surveys have shown that one-third of rural households own no land at all, while another third own less than 2 acres. Eighty per cent of the population is regarded as living below the poverty line.

It is not surprising that a number of programmes have developed in Bangladesh to try to provide some form of finance to alleviate the inexorable poverty that pervades the country. Among the most prominent and innovative of Bangladeshi schemes to provide credit for the poor, is the now famous Grameen Bank, the programmes of the Bangladesh Rural Advance Committee (BRAC), and the Micro-Industries Development Assistance Society (MIDAS). There are several other NGO programmes of which we only mention one here: the Rangpur-Dinajpur Rural Service (RDRS), an arm of the Lutheran World Service.

The Grameen Bank

The Grameen Bank was started in 1976 by Dr. Muhammad Yunus, Professor of Economics at the University of Chittagong to lend money to the landless poor who had no assets and who had no clear way of repaying their debts. In 1976 the Grameen made its first loan through a Bangladeshi commercial bank. Dr. Yunus structured his activities in a village called Jobra as part of a Chittagong University programme. He secured a commitment from a commercial bank to provide the initial finance and the scheme was explained in detail to the villagers.

Until 1983 Grameen was a project, not a Bank, and operated in the villages through branches of commercial banks. In October 1983, dissatisfied with the slow development and with the relationship with the commercial banks, Yunus succeeded in persuading the Government to help him set up a separate bank specifically for the landless poor. The new Grameen Bank succeeded in obtaining an interest-free loan of US$3.4 million from the International Fund for Agricultural Development (IFAD). The Bangladesh Bank matched the IFAD loan and the Grameen Bank started operating. By end October 1983, it had 82 branches in five districts.

Today the Grameen Bank's structure resembles a pyramid. At its base are groups of five selected villagers who carry out the Bank's activities in that village. Ten or so of these village groups form larger groups called "centres". Grameen Bank branch offices supervise several centres.

Grameen loans are granted up to a limit of Taka 5,000⁽¹⁾ (about US$230) and repaid in weekly instalments over 50 weeks, each instalment being 2% of the loan amount. The interest rate (16% in 1989) is all paid at the end. In practice the borrower pays interest and deductions amounting to 24.5% of the amount loaned: 13% as interest, 5% in a form of savings to a group fund, and 6.5% to an emergency fund. When compared to some NGO programmes this interest rate may appear to be rather high; however it is substantially lower than the rate charged by money lenders who were the alternative source of finance before the Grameen Bank brought finance to this target group and whose annual rates were anything from 120% to 3,000%. Grameen Bank operations have expanded at a phenomenal rate - at more than 50% each year for most of the 1980s - from 362 villages involved in the programme in 1982 to 22,784 in July 1991. The Grameen Bank's success has been attributed primarily to the fact that decision-making powers are delegated to the lower levels of the structure, thus creating a process of horizontal expansion.

By July 1991 the Bank had 894 branches, and had given out 953,734 loans to the rural poor in 22,784 villages, which represented over 20% of all the villages in Bangladesh. Average loan size was around US$67 and over this period the Bank lent the cumulative total equivalent of US$354.20 million. It is estimated that Grameen Bank disbursed well over the equivalent of US$100 million during 1991. One of the most significant features of Grameen Bank's lending is that 876,244 of its borrowers, or 92%, are women (1991). The table below gives a summary of the growth of the Grameen Bank, its members and the amounts lent and saved in the equivalent of United States dollars.

Table 1: Growth of the Grameen Bank (US$ millions)

	1982	1985	1986	1987	1989 July	1991 July
Members	30 000	172 000	235 000	291 000	580 000	953 734
No. of villages	363	n.i.	n.i.	6 570	13 047	22 784
Percent women	45	66	74	n.i.	n.i.	92
Amount disbursed	23.1	32.5	49.0	n.i.	162.48	354.20
Amount saved	0.444	2.94	4.56	5.6	11.90	24.43

Sources: Hossain 1988; Rippley 1988; Yunus 1990: Grameen Dialogue (Newsletter of the Grameen Trust) No. 8, October 1991.

An important element of the Grameen Bank operation is the use of savings to prepare borrowers to manage their credit and encourage them to accumulate small capital reserves of their own. Grameen has established a rule that all borrowers must deposit weekly savings and that credit eligibility is based, at least in part, on their ability to maintain the savings discipline. Each borrower through his or her group must save 1 taka, or approximately US$0.3, per week, and in addition 5% of every loan is set aside at the time of disbursement for a group fund managed by the group, to provide emergency loans to its members. The group has full authority to determine the allocation of this fund. By March 1990 the group fund had resources of US$15.4 million. Capitalised by saving the equivalent of 25% of the interest on all loans used for emergencies, by March 1990 the fund had resources of US$6.03 million. As has been pointed out, the savings operations of the Grameen bank have become an important source of capital for the Bank's lending operations. This has helped considerably to expand the bank's lending operations, even though until now it still relies heavily on IFAD funds, through successive low interest credits which have been supplemented over the years by further soft loans or grants from the Ford Foundation, NORAD (the Norwegian Agency for International Development) and SIDA (the Swedish International Development Agency). Savings accumulated, amounted in late 1986 to 45.4% of loans outstanding (Hossain 1988).

Despite considerable success in loan recovery and the wide coverage of its loans, Grameen had not yet reached profitability by the end of 1991. Most ascribe this to the Government imposed interest rate limits. The current interest rate is inadequate to cover the cost of funds available and the high overheads in transactional costs. The administrative structure of Grameen is quite costly as the bank employs over 10,000 people. The figure of 80-90 borrowers per staff member seems quite an efficient operation but when one considers that this implies loans totaling only 5,500-6,000, the heavy overheads become clear.

However, as donors provide subsidised funding either as grants or loans at very low interest rates (2-3% per annum), Grameen in quite viable without actually making a profit on its transactions.

The repayment level on Grameen loans has been high - in the order of 98%. It is generally recognised that the high repayment rate as well as the rapid expansion has been due to the effective use by Grameen Bank of the group process. These groups have demonstrated a remarkable degree of cohesion, a commitment to mutual support and a significant capacity to appraise each others' projects. They have become important, viable, solidarity groups that are also able to exert substantial "peer pressure" on members to accept the discipline of borrowing and repaying their loans on time.

Probably the most significant success of Dr. Yunus and the Grameen Bank has been to demonstrate that the poor represent a bankable clientele who repay their loans if treated with a degree of trust and a if suitable lending organisation is established. Although the Grameen Bank can claim considerable success in its objective of providing finance to the landless poor, figures indicate that it has only reached a relatively modest percentage of the 8 million landless families in Bangladesh and has catered even less for the large numbers of agricultural labourers who today represent only a small proportion of Grameen borrowers. It is also claimed by some that Grameen Bank's transaction costs are relatively high because of the hierarchical structure of its group system. Whilst some view positively the conversion of Grameen to a bank mainly for women (92% of borrowers in 1991), others believe that this may represent a weakness in that the credit needs of male clientele do not appear to have been serviced to an equivalent extent. Dr. Yunus makes no excuses for the heavy emphasis on women borrowers. He believes them to be more reliable borrowers and most of all is of the opinion that lending to women is the most effective way of ensuring the money goes towards alleviating poverty.

The intention of the Grameen Bank was to lend to the poor and thus the low limit of 5,000 taka (originally 200) was set, but in time this has become a limiting factor in the ability of the bank to assist the more enterprising of its entrepreneurial borrowers. To some extent Grameen has attempted to overcome this by developing a group lending programme where larger sums are loaned for group projects, but this has also produced some mild deterioration in the repayment records and has led to Grameen become involved in some larger schemes that are perhaps less appropriate for a bank of the poor. Grameen Bank has now set up a foundation partly funded by the United Nations Capital Development Fund to finance projects to generate employment.

Unlike other large-scale non-governmental efforts to assist the poor, Grameen is basically an institution to provide a forum through which credit can be extended to the landless without collateral and without offering other forms of assistance. Other institutions in Bangladesh, such as BRAC seek to provide a comprehensive programme for the poor. Apart from credit, the BRAC programmes and those of some other NGOs offer training and technical support which are not part of Grameen activities. Grameen's policy is to leave the decision on how the loans are used to the borrower, where in other organisations the credit is more strictly supervised and efforts are made to influence the utilisation of the money for specific purposes. Grameen bank was in fact the first major institution to follow what is now called in the United States a "minimalist" programme of credit without additional non-financial support.

Studies of the impact of Grameen lending have been few. In 1984 Hossain attempted to study 62 borrowers comparing them with a control group. He found a 30.8% increase in per capita income of borrowers. As regards employment generation the difference between Grameen borrowers and others was very small. There is no real basis for asserting that Grameen lending has had any considerable impact so far on the creation of new employment, rather its success has been in enhancing the incomes of borrowers.

Apart from the Grameen Bank which is the best known of such institutions in Bangladesh, there are a number of other programmes operated by NGOs (called private development organisations (PDOs) in Bangladesh) that attempt to provide credit for very poor and disadvantaged groups in the country. Significantly, in 1987, 2.6% of loans (according to the Bangladesh Bureau statistics survey of 1989) were from Grameen Bank, but this has to be compared with loans from professional money-lenders who provided 29.21% of loans to poor households.

Bangladesh Rural Advancement Committee (BRAC)

BRAC is perhaps one of the largest of the NGOs/PDOs involved in poverty alleviation in Bangladesh. It has a long history of helping the landless, providing them with training and latterly with loan facilities. Loans have been given mainly through BRAC's Rural Development Programmes (RDPs). Started in 1979, the RDPs are now one of the BRAC's major programmes. To understand better how RDPs operate it is necessary also to understand some of the BRAC's philosophy.

In early 1972, after the emergence of Bangladesh, the BRAC became a comprehensive social assistance programme in the fields of health, family planning and education. Its earliest experience with credit goes back to 1974 in a limited programme to some villages in the Sylhed District. From 1976 these programmes were expanded to provide credit on an organised basis to landless groups.

The major objectives of RDPs were:

(a) to build up organisations for the poor capable of instituting changes in their socio-economic circumstances;

(b) to improve the social and economic status of the rural poor through providing credits to generate income and employment; and

(c) to develop the managerial and entrepreneurial capabilities of the poor.

As can be seen, the objectives are wider than those of the Grameen Bank. Apart from the landless, BRAC's target group also includes men and women employed as manual labourers and usually landless.

RDPs are administered through branches headed by a manager, assisted by four or five programme organisers. Since 1983, RDPs have endeavoured to recruit local men and women to administer the credit activities. Approximately 45-50 villages with a total population of about 60,000 are covered by a branch. A regional manager then supervises about 10 branches and reports to the programme coordinator based in BRAC's head office in Dacca. RDPs have a monitoring unit and an independent research and evaluation division. As of December 1989, RDPs were working from 81 centres in 45 sub-districts scattered throughout Bangladesh. This means that the programme was operating in close to 3,400 villages where 65% of households belonged to the RDP defined target groups.

Three types of credit are advanced by RDPs to group members: short-term credit (12 months or less); medium credit (one to three years); and long-term credit (more than three years). The interest rate on all loans is 16%. Care is taken to ensure that the borrowers use the loan for the requested and approved income generating purpose. Loan repayment is on a weekly basis. To obtain a loan, members have to be recommended by their respective village organisations which means that all members have to have a significant stake in the venture, equivalent to at least 5% of the loan. No collateral is requested, but BRAC does look into the borrower's ability to repay the debt. There is also intensive and continuous monitoring by BRAC's staff throughout the loan period. Also, when a loan is granted to procure an income generating asset, the asset remains mortgaged until the loan has been fully recovered. Loan amounts vary significantly. The smallest to date was for 500 taka (approximately US$20) and the largest for 1 million taka (US$40,000). Large loans are given for collective enterprises such a brick fields or development of wells. Most individual loans vary from 500 to 8,000 taka ($320), so that the upper limit of BRAC is somewhat higher than that of Grameen.

From 1979 to 1989 over 470 million taka was disbursed, equivalent approximately to US$15 million. Overall repayment rate was 96.5%, although there were variations between the branches. Those branches which operated credit programmes earlier, generally had worse repayment records than those started later. Half the loans are short-term and very few were for long-term, that is for more than three years. Most of the loans went to small-scale trading and agriculture. As in the case of the Grameen Bank repayment records were marginally better for individual borrowers than for collective schemes. The latter was estimated at around 80%.

A major survey was undertaken in February 1988 to evaluate the impact of BRAC's RDP lending programme. A comparison was made between 50 male and female members from four randomly chosen branches who had received loans and a selected control group who had not received a loan as yet from BRAC. Efforts were made to ensure that both groups came from a similar socio-economic background. The survey showed that the per capita annual income of the households that borrowed from RDPs was 26% higher than that of the control households. In fact, it was found that of the borrowing households 50% had an annual income of more than 19,000 taka annually, whereas this was true for only 29% for the control group. To identify the extent to which the improved situation of the borrowers (as against the control group) was due to RDP credit, an examination was made of the source of income of the respective groups. Since RDPs provided 53% of total credit to the trading sector, the fact that a higher proportion of borrowing households earned their income in this sector (34.5% as against 29% in the control group) seemed to indicate that RDP credit had helped these borrowers to generate income other than from direct employment. A rather higher proportion (59.5% of the RDP households) were actually deriving their main income from trading, cottage industries or transport as against less than 50% in the control households.

An attempt was also made to estimate the impact of RDPs on employment. The average number of workers per household in the sample group of borrowers was 1.61 compared to 1.38 in the control group and the ratio of female to male employment was more favourable in the RDP group. Another study undertaken in 1988 compared the income of RDP group members with a baseline and found 160% in real income increase, 84% in employment, and 153% in possession of assets.

In comparing BRAC and Grameen credit operations it is clear that the latter services a far higher number of beneficiaries but concentrates almost exclusively on credit, whereas BRAC has wider objectives of which credit is only one of the services it offers. BRAC also engages in widespread training activities and aims to improve the status of the rural poor. In considering loan applications both BRAC and Grameen emphatically recommend the group for credit worthiness and for ensuring repayment. Furthermore, both programmes emphasise not only borrowing but savings as well. BRAC also insists on compulsory individual savings and group members had saved almost 70 million taka up to 1989 and approximately 192 taka per group member. This is less than the corresponding savings accumulated by Grameen Bank through its borrowers.

In 1989, a consortium of seven donor agencies, including Scandinavian and West European bilateral donors, developed a project to provide a fund of US$50 million to convert BRAC's RDP into a fully-fledged rural bank. However, the BRAC management preferred to use the funds to expand their operations without creating an independent bank. This was mainly to avoid possible government interference and regulation of activities. During 1990-91, BRAC accordingly considerably expanded the RDP and its Rural Credit Programme (RCP).

A report by Calmeadow Foundation of Canada (November 1991) on BRAC's credit programmes shows a very substantial expansion in the institution's lending volume.

In the two years from October 1989 to the end of September 1991, the number of villages where BRAC was operating grew from 2,667 to 4,537 and the number of borrowers went from 140,000 to 415,692. Outstanding loans rose from TK337 million (approximately US$9 million) to TK514 million (approximately US$14 million) and savings from TK66 millions (US$1.8 million) to TK181 millions (US$4.9 million).

The report stated that on an operational basis BRAC's RCP breaks even with operating costs at 14% per annum in relation to loans outstanding. The report estimated that it takes 5000 borrowers for a BRAC RCP office to break even with average loans of TK2750 ($75). On that basis it reported that most RCP offices are now breaking even. The report goes on to say that on a more rigorous basis RCP as a whole is not breaking even and that BRAC would have to charge between 20% and 30% on its loans to break even on a market basis. At the end of 1991 the BRAC RCP lent at 16% so that with 14% operating costs it must pay only 2% for its funds available for lending to break even. All donor funds are given at present at 0% but this is an artificial situation. BRAC pays 9% on savings deposited which account at present for about one-third of the loan funds. On the savings converted to lending, BRAC incurs a loss even though it is paying a deposit rate which is negative in real terms.

The analysis seems to show that BRAC (and Grameen) type lending which offers very small loans requires low-cost funding to cover the high cost of the transactions.

The Micro Industries Development Assistance Society (MIDAS)

Another institution expanding its credit programmes in support of small enterprises in Bangladesh is MIDAS which started operations in November 1983 with a grant from USAID, still a major contributor. MIDAS has subsequently received assistance also from the Norwegian Agency for International Development (NORAD), the Swiss Development Corporation and the Ford Foundation as well as some UNIDO technical assistance. In the five-year period 1986-91, USAID assistance to MIDAS was planned to exceed US$8 million. From 1982 to 1990 (8 years) MIDAS actually disbursed TK140 million (approximately US$4 million). In 1989-90, MIDAS financed 30 projects for a total of TK48.65 million ($1,3 million), 17 of which were new enterprises. MIDAS also provided equity participation of TK5.66 million in four new projects.

Although its name refers to micro industries, MIDAS has adopted a far wider target group than Grameen and BRAC. Some of MIDAS' loans are equivalent to US$100,000 or more, so that it can be said to target both micro-enterprises and formal small businesses.

USAID has supported MIDAS mainly as an organisation working in the private sector. Even though its donor assistance is now quite considerable, MIDAS has managed to keep its staff relatively small (around 80 with 49 professionals) and of reasonably high quality. Despite the wide target group and average loan size of TK2.5 million (around US$80,000), MIDAS has concentrated on helping labour intensive microenterprises and claims that its projects generally support employment creation at an investment cost per job of around US$1,000. MIDAS has mainly aimed at promoting and financing new small industrial projects rather than expanding of existing enterprises. Under the influence of USAID, MIDAS put its interest rates at around 18%, which was close to the commercial rate for lending in Bangladesh. MIDAS's overall loan recovery rate was 84% in 1990.

In contrast to the other organisations mentioned, MIDAS aims to identify opportunities for new small enterprises development and has worked with the Danish International Development Agency (DANIDA) to identify possibilities and arrange entrepreneurship training for selected entrepreneurs. In 1989 MIDAS developed 71 small projects with DANIDA's financial assistance; 45 of these received loans to a combined total of TK7 million (US$218,750).

MIDAS does not regard itself exclusively as an NGO providing credit. It operates a fee-based consultancy service, carries out small-scale industrial sector studies and engages in activities to help expand the volume of sub-contracting, franchising and entrepreneurship development. Sponsored by donors, it has signed agreements with foreign consultants who have undertaken to develop a number of new innovative industries around Dakha and Chittagong which would create about a thousand job opportunities. Women entrepreneurs will be given priority.

Rangspur - Dinajpur Rural Service (RDRS)

As a final example of non-formal lending for income generation one might look at the RDRS model, an arm of the Lutheran World Service, which has experimented in providing credit over a number of years. The RDRS Credit Programme has many features similar to those of Grameen and BRAC. It also lends through groups and places emphasis on the responsibility of the group for repayment of the loan. The RDRS generally works through larger groups than Grameen and BRAC, which concentrate on groups of five to ten. RDRS favours groups of about 15 members, believing this size to be more stable yet small enough to function on the basis of trust and to operate cooperatively together.

RDRS is now not only forming groups for credit but is moving away from its original mandate of providing relief to the poor, to promoting development. Like other NGO/PDOs operating in Bangladesh (PROSHIKA, Care and others), RDRS now seeks to maintain its own revolving loan fund.

RDRS credit is only provided to groups that have been in existence for at least six months. The groups must have a bank account and show that there has been some prior group activity. Evidence of group savings is also required before credit can be obtained.

Groups

RDRS only gives credit to groups. The loans are for use either by the group or any individual within the group, but in all cases decided by the group, which also guarantees repayment.

RDRS comments on group size is of interest. It sets out the different categories to which loans may be given in a credit programme as follows:

1. Individuals.

2. Groups of five (Grameen Bank, BRAC, Swanirbhar⁽²⁾, and several government agencies) formed to generate social pressure for loan repayment. A group size of five was first used by the Grameen Bank. Such tiny groups are mostly artificial in terms of banding together for an investment project, and now after some years of experimentation other agencies are finding that the model does not always suit them.

3. Medium-size groups of 12 to 15, which are, in principle, optimum for group trust and project involvement.

4. Larger groups or cooperatives of 25, 40 or more. PDOs have found that such large groups often have a boss who then may exploit the others, or they need to follow some formal system involving elections, like the rules of the Department of Cooperatives, which brings its own problems.

5. Village organisation, which uses a loan to make small loans to individuals in the community.

RDRS' experience on group size is useful. RDRS used to sponsor groups of about 8 or 10 small farmers. It has now concluded that groups of 15 work the best for all its target population in the comprehensive programme.

RDRS groups have many of the qualities which make informal savings and credit groups successful in Bangladesh. They meet regularly and there are new instances of trust within the group. Group officers are appointed by democratic consensus without the divisiveness of formal elections or the power of a few vested in a management committee.

Interest rates in credit programmes

Originally a more philanthropic type of organisation, RDRS credits used to be interest-free, but by 1988, in an effort to retain at least some of the value of its capital fund, RDRS started charging 14% interest. According to RDRS reports the agency believes that small bank loans, even at 16% (the rate charged by Grameen and BRAC), cannot cover the full cost of operating a lending programme. It believes that the 8% margin between the level charged by the Bangladesh Bank and that charged to the borrower is not enough to cover overheads and defaults. Some commercial banks in Bangladesh that have been privatised (the Rupali Bank for instance) have stopped giving rural credit because they found it was not profitable at prevailing interest rates. After carrying out a study, RDRS concluded that small loans could not fully cover costs at less than 22-30%, and even this did not make full allowance for inflation (RDRS suffers because it does not receive savings and so loses income from investments available to others who do take deposits). Findings also showed the borrowers repaid loans equally whether they were at 30% or at rates below 15%. However, RDRS stated that it did have a repayment problem when interest rates were raised above 36%. It found borrower viability best at 18% interest, but up to 30% there was hardly any decline in loan repayment or in demand for loans.

The studies made by the RDRS programme show that while a loan from a bank may actually cost only 16% in interest, to a poor borrower the real cost is much higher due to the considerable time needed to make multiple trips to the bank, delays and possible bribes or extra fees to pay for formalities. It is because there are so many extra costs involved in obtaining institutional credit even when this is available, that the non-institutional credit (i.e., loans from friends, relatives or moneylenders) is attractive to poor borrowers because it is available without delay and formalities and because the actual very high interest rates (possibly over 50% to more than 100%) may not be all that much higher than the real cost of borrowing from banks with all the bureaucracy involved.

According to one report of the Bangladesh Bureau of Statistics (1985), because of the high demand for credit and the reluctance of banks to lend, anybody in Bangladesh with some capital can engage in money-lending. The report estimates that in 1987 that there were 1.59 million moneylending households as against 5.05 million borrowers. A RDRS report asserts that Grameen borrowers pay in all about 27%, of which 16% is interest, and the remainder being compulsory savings, insurance, special fund, membership fees, and so on.

3.2 Indonesia

The 1980 census in Indonesia gave the number of enterprises in the country as 27 million and the total number of inhabitants as 148.04 million. By 1990 the population was estimated at 190 million. Most business enterprises are very small, with an average number of employees at less than two. Such data as exist on size (1980) indicate that 92.1% of the enterprises which account for 62.2% of the total employment in the country are in the very small category, i.e. with less than five employees; a further 7.3% of enterprises and 18.4% of employment falls into the next category of small enterprises with 5 to 19 employees. In agriculture and trade the proportion of small and very small enterprises is even higher, probably more than 95%.

Over the past 20 years Indonesia has had a varied experience with financial schemes for small enterprise development and has experimented with a wide range of different programmes on a national and a regional scale with and without a savings component, with subsidised and with commercial interest rates, with broad and narrowly defined target groups and loan purposes. Four of these programmes which are thought to represent a sample of the different types are described below.

The KIK/KMKP Programme

Kredit Investasi Kecil/Kredit Modal Kerja Permanent (KIK/KMKP) (small investment credit/permanent working capital credit) was a government-sponsored credit programme targeted at Pri-bumi (indigenous Indonesian) small-scale entrepreneurs in all sectors of the economy, in both urban and rural areas. Financed and administered from 1973 by Bank Indonesia, the central bank, and implemented through a number of handling banks, it became the most important credit programme for small enterprises throughout Indonesia, with a total of 2.5 million loans from its inception to the end of 1987 after which it was phased out.

The objective of KIK was to provide medium to long-term financing for the acquisition of fixed assets through small investment credits. Short to medium-term financing for working capital was provided through KMKP. The programme aimed at employment generation, entrepreneurship development, expansion into local markets and a broader geographic spread of productive economic activities.

The programme was administered by Bank Indonesia, through state commercial banks, national and regional development banks and a few private banks with over 1,200 branch offices. Of these, Bank Rakyat Indonesia, the major state rural commercial bank, had around 50% of the portfolio. To administer the programme, Bank Indonesia set up a Central Project Management Unit (CPMU) to monitor and evaluate the handling banks, who were required to submit monthly reports. Monitoring and evaluation was further decentralised through 13 Regional Project Management Units (RPMU), which provided advice and assistance on project identification, and in the improvement of lending procedures. They also helped in training staff.

There were two types of loans: loans to individual borrowers and loans for group projects, the latter amounting to about 16% of the KIK portfolio and 11% in the case of KMKP. "Pri-bumi" small enterprises were eligible for credit if their net worth, excluding land and buildings, did not exceed Rp100 million⁽³⁾ (US$61,000) in 1987 for industry and construction and Rp40 million (US$24,000) for other sectors. Although Bank Indonesia declared that "the project financed is the main collateral", additional collateral of up to 50% of the loan amount was usually requested by the handling bank in addition to the credit insurance referred to below.

The maximum maturity for KIK loans was ten years, including a grace period of up to four years. KMKP loans have a maximum maturity of five years, including a grace period of one year. The credit ceiling was Rp15 million (about US$9,000) under each programme. Loans under both programmes may be combined to a total of Rp30 million (US$18,000).

In 1987 the interest rate was 12% per annum effectively, i.e. about half the commercial lending rate at that time. The chief source of finance for KIK/KMKP has been a government rediscount facility at a rate of 3% for 80% of the loan volume, supplemented since 1981 by World Bank funds for KIK. In 1984 the third World Bank loan of US$200 million included a component for KMKP as well.

ASKRINDO, the government financed Indonesian Credit Insurance Corporation, automatically insured up to 75% of a KIK/KMKP loan. The ASKRINDO premium was a 3% one-time front-end fee for loans up to five years, shared equally by Bank Indonesia and the handling bank. For loans for a longer maturity, the premium was 5%, shared at a 2:1 ratio. Claims settlement was lengthy until 1982 but was greatly expedited thereafter. Claims made by handling banks on ASKRINDO's guarantees rose dramatically after 1980 and became a major drain on the financial resources of the government.

Backed by the government, the substantial ASKRINDO coverage for a low premium, reduced the handling banks' incentive to take care in selecting projects and improving loan supervision and debt collection. Attempts were made after 1987 to review the credit guarantee terms to raise the incentive for banks to collect loans but all the changes proposed were vigorously opposed by the participating banks who threatened to withdraw from the programme without the automatic ASKRINDO 75% guarantee.

Between 1974 and 1981 the lending volume increased rapidly with an annual growth rate of about 57%. After 1982, the growth rate declined, after growing arrears made banks more hesitant to grant loans. Increased government losses on the programme including the heavy ASKRINDO guarantee payments, led the government to reduce the financial resources it made available. In 1986 the World Bank also decided against a new loan for the programme, thus further cutting funds. By 1986, a total of 2.4 million loans had been made, amounting to a total nominal amount of Rp 4,376 billion (US$2.7 billion at 1986-87 exchange rates). After 1987 the programme was run down and phased out.

Of all KIK/KMKP loans the proportion of working capital (KMKP) loans up to December 1986 was 88.4% in terms of number of loans and 74.1% in terms of loan value. The biggest share of loans went to the trade sector (wholesale and retail trade, hotels and restaurants) - 62.5% of loans outstanding up to end 1986; agriculture's share was 11.7% and that of industry, including crafts, was 10.5%. The need for investment and working capital loans is unevenly distributed by sector: the transport sectoris need was for investment loans while the trade sector predominantly needed working capital loans. During the lifetime of the programme, the shares of agriculture and transport have declined while that of trade has increased.

There has been a move towards larger loans - that is, to the maximum permitted under the programme - which reflects both the impact of inflation and the handling banks' preference to lend as large amounts as possible. Yet, average loans remained small - Rp 3.5 million for loans in 1986 or Rp 1.8 million for all loans since 1974 .

Performance

Although the KIK/KMKP programme has been defended by the Indonesian authorities as helping in the consolidation, if not growth, of a large number of small and very small enterprises in both rural and urban areas in Indonesia, generally most outside consultants and donor agencies have been very critical of the performance of the programmes in financial terms. Performance on loan recovery has been difficult to evaluate due to poor data on arrears and defaults. There was also a considerable amount of loan rescheduling which distorted the real figures on arrears. Most external reports put the loan recovery rate at between 70 and 80% at the most, but others claim that the real figure was much lower - even below 50% - taking into account the high number of claims settled by ASKRINDO and the loan rescheduling that took place.

Impact

By July 1987, 2.5 million loans had been disbursed through the KIK/KMKP programme but this has to be seen in relation to a universe of 30 million small enterprises, and one has also to take into account that the loan figure include multiple lendings. The limited estimates of "presumed impact" are based on a sample of 470 small enterprises which received their first loan during the first quarter of 1984. Of the loan recipients 76.6% of the sample were male, 46.8% were below the age of 40 and 70.9% had more than five years business experience. The average number of employees in the enterprises was 14 in agriculture, 15 in manufacturing, 5.6 in trade and 5.4 in transport. Fixed assets per unit were Rp 9.1 million in agriculture, Rp 17.8 million in manufacturing, Rp 14.0 million in trade and Rp 15.5 million in transport. Of the loans 68.1% were used for the expansion of enterprises, 14.9% for the repair or replacement of equipment, 12.1% for starting up a new enterprise and 5.0% for other purposes. This seems to confirm what had previously been stated by many that the contribution to small enterprise creation was small and that most loans went to those already in business with good connections, especially with the lending banks. Assuming that credit was the major determinant of changes, in various performance indicators during the two-year period of 1983-85, it was found that the enterprises studied hired on average between 2.1 and 4.8 additional workers (full-time job equivalents), according to sector and that they acquired additional assets of between Rp 6.9 million and Rp 16.2 million.

Extrapolating to the total number of small enterprises in the scheme, it was estimated that by 1985 employed an additional 66,326 persons with a "full-time job equivalent" of 43,412, that they increased their assets by Rp 233.4 billion and their value added by Rp 344.4 billion. Of course KIM/KMKP loans were but one of several factors contributing to this increase, but it can be said they contributed to this growth. Most independent observers believe these extrapolations exaggerate the real economic impact of the programme.

Despite the widespread criticism of the KIK/KMKP programmes, it still remains in its coverage one of the largest small enterprise financing schemes ever undertaken, lending over the equivalent of US$ 2 billion over 15 years. The poor loan recovery combined with the heavily subsidised interest rates and the large number of ASKRINDO claims made it a very high cost programme for the Indonesian Government. Moreover, the 75% automatic ASKRINDO cover was certainly a disincentive to banks to make any great effort to recover loans or for the borrower to make repayments.

The programme was poorly managed, up-to-date data on the state of the portfolios and the level of loan recovery was inadequate or unavailable. Furthermore, banks gave out new loans after rescheduling old loans which had fallen into arrears.

The handling banks were given margins around 3% - too low to justify careful screening of loan applicants and supervision of borrowers. The banks preferred to lend to known customers, mainly those operating in trade rather than in manufacturing or agriculture and to rely on the credit insurance, rather than to lend to first time entrepreneurs engaged on more risky projects.

Some have criticised the KIK/KMKP programme in that it made no attempt to mobilise savings, but clearly the banks involved did not believe that such efforts were justified in the light of the small sums involved.

Notwithstanding all the justified criticism of the KIK/KMKP programme it did result in a very high number of loans being given to small enterprises. It was also one of the first examples of a lending programme to small enterpreneurs in which central bank overall coordination and control was combined with a highly decentralised implementation, covering a vast geographic area and involving 30 development and commercial banks, both public and private, with 1,200 branches in all parts of a widespread country.

The lessons learned and the mistakes made during the KIK/KMKP experience may have contributed to the development of more successful programmes in future.

BKK: A Regional Program in Central Java

Badan Kredit Kecamatan, BKK ("subdistrict credit institution") is a regional government programme in Central Java supervised and refinanced by Bank Pembangunan Daerah (BPD), the regional development bank which started operations in 1970. In October 1990, the programme comprised 502 branches and 3,928 BKK subdistrict units, which covered almost one-third of all villages in Central Java. The programme provides "easy, fast", very small, short-term loans at commercial interest rates, predominantly to microenterprises engaged in off-farm income-generating activities. For the first ten years of its existence, the programme had the status of a project, when in 1981, by a provincial law, it was turned into a BUMD (Badan Usaha Milik Daerah), a profit-making and taxable corporation owned by the provincial government. It received assistance from USAID. The programme was designed to avoid the shortcomings of other credit programmes (such as KIK/KMKP): subsidised interest rates, poor loan screening, an over-centralised administration, with high transaction costs and high default rates.

The BKK programme was capitalised in 1972 through the BPD, the regional development bank, with a Rp 200 million (US$476,190) loan from the Central Java Provincial Government. The three-year disbursement of the funds began with 200 BKK (subdistricts) and grew to 486 BKK in 1975, and then to 1,445 branches and 4,085 village posts in 1990 supervised by Provincial Development Banks in Central Java and by other Indonesian provinces where the programme has been replicated. Each BKK received a three-year Rp 1 million ($2,381 at 1975 exchange rate) loan at 1% per month with a one-year grace period on the repayment of the principal.

Each BKK made loans from the amount it received to individual clients for periods of 22 days to 6 months. Interest due and compulsory savings had to be paid up before anything was credited against repayment of the principal. All payments were made in equal amounts, according to one of the six repayment plans chosen by the client.

Each BKK is an independent subdistrict credit body that is locally administered and financially autonomous. Operational responsibility is delegated through district commissioners in rural areas to mayors of municipalities or subdistrict chiefs depending on where the administrative authority is situated. Financial supervision and management is performed by the provincial development bank, BPD, which has also introduced a unified system of bookkeeping and procedures and is in charge of BKK staff training. Most lending is done through the 4,085 village posts (in 1990), thus "the bank is taken to the people", usually on market days. This keeps borrower transaction costs at a minimum.

Each BKK operates with a staff of at least two: a bookkeeper and a cashier. A total of 1,740 staff members gave out an average of 302 loans totalling Rp 15.9 million in 1986. BKK staff are paid out of income received from interest on loans. Low salaries are supplemented by an incentive scheme based on a BKK's performance. This helps to motivate the staff. About 10% of quarterly profits are given as a bonus to employees on the basis of their grade, the better paid civil servants receiving less. The bonus permits some staff members to more than double their fixed salary.

Unprofitable BKKs are penalised by a lowering of the credit ceiling. Classification is based on six factors: the equity ratio of villages to village posts; the number of new borrowers; the portfolio quality based on collection of payments due; total savings and capital circulation. The proportion of BKKs in the two best-performing classes (out of a total of five) has increased from 13.8% in 1981 to 21.0% in 1982, 32.9% in 1983-84, 41.7% in 1984-85, and 48.8% in 1985-86; and to 55.1% by the end of 1986.

Lending procedures for individual borrowers, while minimising risks, are simple and rapid and are characterised by:

- emphasis is on granting loans for business purposes rather than for consumption;

- decentralised approval authority: BKK approval for loans up to Rp 25,000; subdistrict chief approval for bigger loans;

- loans being increased gradually depending on satisfactory repayment performance and savings. On average, 90% of loans are for sums ranging from the equivalent of US$22 to less than the equivalent of US$60;

- lending being carried out from village post or subdistrict office, whichever is more convenient to the borrower;

- no material collateral being required.

BKKs aim at a "graduation policy", that is encouraging clients eventually to move on from the BKK and to borrow amounts over Rp 200,000 from the BPD.

Non-subsidised commercial rates are charged and the average effective rate works out at 30% per annum which, in 1990, was higher than the commercial bank lending rate. Penalties for late repayment are not officially part of the programme but many BKK units have started to impose a charge.

There is a modest compulsory savings component - between 6% and 20% of the loan - which is for educational rather than for fund-raising purposes. By December 1986, compulsory savings stood at Rp 2.48 billion, equal to 14.9% of all liabilities. Since 1988 in some provinces BKKs have also offered deposit savings on a voluntary basis which has been popular and given the system an important source of capital. Approximately 56% of loan capital in 1990 came from domestic savings.

In 1990 there were 800,000 borrowers and 1,033,000 savers in the BKK system.

The BKK programme is one of the very few publicly funded credit programmes for microenterprises in the world that is financially viable and even making money from providing loans to very small rural enterprises. Interest rates are high compared to most institutional finance (low by informal sector standards), but are sufficient to cover the cost of funds, administrative expenses, default losses as well as to allow a build-up of modest reserves from retained earnings.

The quality of the BKK portfolio is good and loan recovery is high. By December 1985, 113,739 loans amounting to Rp 1.23 billion were in arrears on payments for more than six months amounting to 10.2% of the loan portfolio at that time. However, when expressed in terms of accumulated loans granted since 1972, the figure would be 1.2% of the total value lent and 2.7% of the total number of loans.

Loans in arrears for up to six months numbered 36,897 (or 8.2% of all loans given in 1985) and were worth Rp 0.69 billion (or 5.8% of loans outstanding at December 1985), or 2.7% of the value of all loans given during that year (taking into consideration a portfolio turnover ratio of 2.2).

The combined balance sheet of the BKK shows that the programme is profitable and is essentially self-financing its own expansion, a situation which should continue into the 1990s. From 1984 to 1986, annual profits rose from Rp 1.71 to Rp 2.25 billion, contributing to a gradual build-up of reserves from Rp 4.58 billion in 1984 to Rp 8.02 billion in 1986.

Based on a study of 662 BKK clients in November-December 1982 (Goldmark & Rosengard (1983)), it was found that 60% of clients in the sample were women, who dominate the very small trading sector. This was reputed still to be characteristic of the programme in 1990. As their primary economic activity, clients were in trading (53%), in agriculture (15%), in services or repair crafts (12%), in handicrafts (11%) and in agricultural processing (7%); 58% had little or no primary education. Additional loans were made to 34% of clients from other sources, mainly from indigenous savings and credit associations. Other loans came from informal financial institutions such as moneylenders (5%).

The average BKK loan in the sample was for Rp 59,000 ($88 at 1982-83 exchange rates) which was higher than the overall BKK average. Borrowers in the sample had received an average of 13 consecutive loans from the BKK. In addition to the BKK compulsory savings scheme, 40% of clients had additional savings accounts 83% of which were held in informal and only 17% in formal financial institutions.

The impact of BKK loans is difficult to estimate. However, the majority of respondents stated that the BKK loans had a positive effect on their business activities: on the volume of materials purchased (82%), the variety of goods sold (37%), the number of customers (75%), sales volume (81%) and, last but not least, net profits (92%); 50% said that BKK loans led to additional employment.

In rural Central Java BKKs have considerably improved access to credit. However, in spite of the increased number of village posts and the broader coverage, the programme can satisfy only a small proportion of credit needs. Its 400,000 clients in 1987 represented only 8% of Central Java's 5 million households.

BKK can be described as a "minimalist" programme in that it offers only credit and virtually no other support services such as training or advice. Its rapid expansion in volume and its high loan repayment rate are often given as indications that "minimalist" programmes are more effective.

The BKK programme has been highly praised and is quoted as an early example of lending to microenterprises which has achieved high loan recovery at unsubsidised commercial interest rates, where all costs have been covered and where even a surplus of income over expenditure has been earned enabling it to build up modest reserves. As BKK predates the Grameen bank in Bangladesh it can claim a pioneer role in showing that the poor, particularly women, are bankable, and will repay loans made at commercial or even higher rates.

BKK succeeded in developing a fast, simple, highly decentralised system of processing loans delegating a fair amount of authority approving loans to the village and subdistrict levels. It was among the first schemes to give incentives to staff with monetary rewards for good loan performance (disbursement and recovery) and BKK profitability. By gradually increasing loans to clients based on their past loan repayment record, borrowers were given positive encouragement to repay their loans on time.

Some have criticised BKK for its low credit ceilings which undoubtedly reduces the economic impact of its loan programme. The average loan size in 1986 was less than the equivalent of US$60. The original loan ceiling was Rp 200,000 (then US$122) but this was raised in 1987 to Rp 500,000 (in 1991 the equivalent of US$250 approximately). Also the low loan size makes the average cost of the transaction a high percentage of the amount lent. The lack of links with any training or technical, economic or management advice inevitably enforces limits on the business expansion that might result from the lending. The lack of such help and the low loan limits more or less preempts the likelihood of graduation to direct lending from BPD.

In summing up it may be said that BKK is a publicly funded and administered credit programme with a moderate savings component that appears to have achieved financial self-sufficiency and is making profits from lending to small rural enterprises.

KUPEDES Unit Desa System: A commercial credit

programme for small rural enterprises

KUPEDES (Kredit Umum Pedesaan - "general village credit") is a credit programme for small rural enterprises which was introduced in February 1984 and run along commercial lines by Bank Rakyat Indonesia (BRI) through village units called "Unit Desa".

As already mentioned, BRI was the major participant in KIK/KMKP. It is the largest state commercial bank in Indonesia and the most important financial institution in rural areas. In addition to providing financial services to the rural population, it guides and supervises all village and market banks. By 1986, at the subdistrict level, BRI had 297 branch offices, 2,383 village units (Unit Desa) - with an average staff size of four - and 1,116 temporary "Unit Desa", totalling about 3,500 permanent and temporary "Unit Desa". In 1990, staff totalled 31,760, 38.8% or 13,000 of them in the Unit Desa.

In 1983, bank deregulation enabled BRI to further develop its commercial banking operations in rural areas by setting up a commercial small enterprise loan programme (KUPEDES) and increasing mobilisation of domestic savings. The Unit Desa loans ranged from Rp 25,000 to Rp 25 million (US$15 to US$1300), but the upper limit was generally only available to repeat customers with good repayment records. Table 2 shows the growth in average loan size in the KUPEDES programme.

Table 2: Average KUPEDES loan size - loan balance outstanding

(rupiahs thousands)

	1986	1987	1988	1989	Jan.-June 1990
Average loan size	199	294	377	391	468

Source: BRI Head Office.

There are two types of loans:

- working capital loans of a maximum maturity of two years with a nominal interest rate of 1.5% per month for timely repayment and 2.0% when in arrears. Working capital loans also have options of single balloon payments for three to 12-month maturities but these options are not encouraged due to poor repayment records.

- investment loans of a maximum maturity of three years at a nominal interest rate of 1% per month for timely repayment and 1.5% when in arrears. Grace periods are of eight to nine months.

Interest rates on working capital and investment loans are the same. On loans of Rp 3 million or less, the interest rate is 1.5% per month, calculated at a flat rate on the original loan principal. For loans of more than Rp 3 million the interest rate structure is two-tiered - 1.5% per month, calculated on a flat rate basis on the first Rp 3 million of the original loan principal and 1.0% per month on the amount of the original loan principal above Rp 3 million.

The effective annual interest rate is 31.7% for loans of Rp 3 million and less; for loans of more than Rp 3 million, the effective annual interest rate ranges from slightly less than 31.7% declining to 22.7% for a maximum loan of Rp. 25 million. The high loan interest rate permits a profit to be made, even though funds are borrowed at market interest rates and intermediation costs are high due to the large number of small loans.

While these rates are higher than market rates for larger loans (which currently range between 18% and 24%), they are much lower than rates charged by informal lenders, which may exceed 10% flat rate per month.

During six years of operations, the Unit Desa have disbursed about Rp 3.4 trillion in KUPEDES loans (about US$1.9 million equivalent at the end of 1989). The growth of the KUPEDES loan portfolio is shown in Table 3. The total number of loans made during this time was 6.4 million. Average loan size has increased almost threefold since 1984 from Rp 287,000 (US$176) to Rp 777,000 (US$450) in 1989. (The consumer price index rose approximately 50% during that period.) The number of annual loans made during this same period has increased twofold from 0.64 million to 1.38 million. After stabilizing at around 1.1 million loans annually in 1986-88, loans increased by 21% in 1989.

Figures show that the greatest need of rural small enterprises is for working capital loans. Over 95% of the loan volume of outstanding Unit Desa loans was for working capital loans and less than 5% for investment. There are some sectoral differences, with a higher proportion of investment loans in industry/handicrafts (12%) and a lower one in trade (2.1%). Sectorally, 68% of the loans go to trade, followed by agriculture (27.6%); crafts and industry receive a negligible share (2%).

Table 3: Summary of KUPEDES lending operations (rupiahs billion)

	1984	1985	1986	1987	1988	1989	1990 June
Loans outstanding; start of period	-	111	229	334	429	539	846
Loans disbursed	172	340	483	598	708	1073	937
Loan repayments	61	222	378	498	586	646	639
Loan write-offs	-	-	-	5.5	12.4	20.3	3.2
Loans outstanding; end of period	111	229	334	429	539	846	1140
Number of loans (000)	639	991	1150	1136	1137	1379	1808

Source: Bank Rakyat Indonesia, Head Office.

BRI tries to raise its own funds for financing its commercial loans, including KUPEDES. Although still relying on BI (central bank) funds, BRI has gradually stepped up its domestic resource mobilisation. Total savings at BRI village units amounted to the equivalent of US$460 million (at December 1989) representing 6.6 million savers with average deposits of US$70.

SIMPEDES, Simpanan Pedesaan (Village Savings Programme) was introduced by BRI as a Unit Desa savings instrument in 1985. SIMPEDES interest rates, calculated on the basis of minimum monthly balances and compounded, are 0% on balances of less than Rp 25,000; 9% on balances from Rp 25,000 to Rp 200,000; 14.4% on balances between Rp 200,000 and Rp 750,000; and 15% on balances above Rp 750,000. The saver is permitted unlimited withdrawals, which is considered a key factor in the success of the programme. As an additional incentive for the SIMPEDES programme, savers receive coupons for participation in a lottery held every six months for prizes.

With the expansion of the Unit Desa system to urban locations, a new savings instrument, SIMASKOT, Simpanan Kota (City Savings Programme) was developed for the cities and introduced in December 1989. This instrument is similar to SIMPEDES but in addition provides an interest rate of 15.6% on accounts larger than Rp 2 million. The lottery for SIMASKOT awards cash prizes which are not offered in the SIMPEDES programme.

In addition to the savings instruments, Unit Desa offer time deposits and demand or checking accounts. The time deposits have maturities ranging from one month to one year. It is interesting to note the growth of the savers and borrowers in the Unit Desa system. As the following table shows, the number of savers has sharply outpaced the number of borrowers each year (except 1989 when the growth of savers and borrowers was about even). This is not surprising as more people in rural areas tend to be savers than borrowers at any one time. At the end of 1989, the Unit Desa had almost four times as many savers as borrowers.

Table 4: Savers and borrowers at Unit Desa

	1984	1985	1986	1987	1988	1989
Borrowers unit	1 060	1 232	1 315	1 386	1 634	1 997
Borrower growth	-	16.2%	6.7%	5.4%	17.9%	22.2%
Savers	2 500	3 544	4 184	4 998	6 245	7 500
Saver growth	-	41.8%	18.1%	19.5%	24.9%	20.1%

Source: Bank Rakyat Indonesia, Head Office.

KUPEDES is a commercially run credit programme. Arrears - defined as loans overdue for any period of time - stood at 2.0% in 1985 rising to 7.5% in 1988, but declining in 1989 to 5.4%. In 1990 they dropped below 5%. Given the high growth rate and the short history of the programme, at the time these figures were calculated, not surprisingly, arrears rose between 1985 and 1988, but only moderately. The long-term loss rate to the ratio of the total amount of the loan portfolio overdue to the total loan amount outstanding was 3.3% in June 1990.

KUPEDES charges commercial interest rates calculated so that BRI can cover the cost of funds, administrative expenses, a risk premium, and leave a surplus. KUPEDES has proved that a credit programme for small rural enterprises on commercial terms can be financially viable and can attract borrowers. Access to credit at reasonable transaction costs has proved more important to small enterprises than the level of interest rates.

In most rural parts of Indonesia, the Unit Desa is the only commercial credit institution to which very small enterprises have access. Together with the savings deposit facilities these BRI village units offer a range of financial services to the rural population.

KUPEDES and the Unit Desa System have proved profitable. Before the introduction of KUPEDES the village unit system of BRI made huge losses (up to 50% of the expenses). KUPEDES reached breakeven point 18 months after it started in 1984 and since has generated increasing profits.

The profits of the Unit Desa System rose from Rp 9.8 billion in 1986 to Rp 36.9 billion (US$20.3 million) in 1989. Pre-tax profits in 1990 rose to US$25 million equivalent despite a big increase in administrative expenses.

Impact data on the KUPEDES programme depend mainly on a study in 1990 by Ann Sutors for BRI which found that at the time of their first loan 15.1% of KUPEDES borrowers fell below the poverty line as defined by the World Bank. Just under 25% were women. Their loans were mainly used for working capital for non-farming enterprises. The enterprises supported employed on average 3.4 employees.

The average KUPEDES borrower interviewed in the study received three loans averaging Rp 466,000 or US$250. Income of the beneficiaries grew at an annual rate of over 20%. Total employment in borrower enterprises increased at an annual rate of 18.2%. Workers per enterprise increased from 3.4 to 5.6 but the increase was mainly in enterprises that borrowed more than Rp 1 million (US$520).

The study's limited sample should be treated with reservation especially as it was "biased" towards successful borrowers, i.e. those that had received at least three loans. There are indications however that the KUPEDES programmes helped enhance incomes and made a modest contribution to employment creation.

The Unit Desa and KUPEDES programmes can claim to run their programmes along good banking principles, lending at commercial interest rates. The programme has shown rapid volume growthe, is financially viable and is profitable, a rare feature for a programme of this sort. Credit delivery is decentralised and transaction costs are low for borrowers. The programme has also high levels of loan recovery.

However, amounts lent are small which again reduces economic impact on employment and enterprise growth. Lending for investment is low and most lending is for working capital, mainly for trade rather than manufacturing.

Lumbung Pitih Nagari (LPN): A regional programme of financial self-help groups in West Sumatra⁽⁴⁾

LPN are village-based savings and credit associations in West Sumatra. They afford an interesting example of upgrading indigenous "self-help" groups and linking them to banks.

LPN originate from two indigenous institutions - Lumbung Padi, the communal rice storage go-down which every village had for emergencies and to help poor people in need, and "Julo-julo", a rotating savings association, of which there were several in every village. These institutional origins have the important elements of self-help in small groups, mutual aid at the village level, and concern for poverty.

LPN are non-rotating savings and credit associations with up to about 200 members, comprising one or more villages within a subdistrict. They have two major functions:

- they provide a deposit facility, based on a system of self-imposed savings;

- they provide a credit delivery system based on internally generated funds and are partially refinanced by the Provincial Development Bank.

In the early 1970s, the provincial government decided to reorganise these group activities and to link them to the banking system. It initiated the establishment of one LPN per Nagari (subdistrict), which comprises several villages. Small indigenous savings associations and government-initiated LPN continued to exist side by side. While the LPN built up its loan fund, subgroups within the LPN continued their rotating savings practices.

LPN were provided with a one-off capital grant by the provincial government of Rp 500,000 for each LPN, followed by allocations of Rp 500 million and Rp 1 billion to the regional development bank, BPD (Bank Pembangunan Daerah), by the Ministry of Finance. The LPN were also given access to the banking system to deposit their savings and obtain loans for onlending to members. In addition, BPD provided training, monitoring and auditing services.

Two types of LPN developed: those which were genuine self-help groups formed from existing associations, and those that were newly created for the purpose of obtaining a loan. Of the 584 LPN listed by July 1987, only about 230, or 39%, were active, and even fewer - 133, or only 23% - qualified for refinancing by the BPD. About 60% only existed on paper. LPN established solely in response to the government's initiative and credit offer were generally slow in generating their own funds, delayed payments and in some cases defaulted or even misappropriated funds. Eventually, most of these LPN became inactive.

Around 98% of the LPN were registered with the provincial government and enjoyed a recognised legal status in West Sumatra.

The LPN loan portfolio for members in rural West Sumatra grew from Rp 11.9 million in 1974 to Rp 2.5 billion in 1985. The maturity of loans to members was mostly between six and eight months, up to a maximum of 12 months. The interest rate was 10% flat for six months or 20% per annum, on the nominal amount of the loan. Instalments were mostly paid fortnightly, sometimes monthly. The effective interest rate was thus approximately 40% per annum. Most loans varied from a minimum of Rp 10,000 (less than US$87) to Rp 1 million (approximately US$800 at 1987 exchange rates).

Loans are approved by the LPN management which is autonomous in this respect. The BPN which refinances loans from some LPN claims that the group is better able to judge a borrower's credit-worthiness and the project to be financed than a bank's credit officer. It is claimed that this improves repayment, as the group is able to apply social pressure on defaulters and in this way bank transaction costs are also reduced.

In reviewing a loan application, the LPN management usually takes into consideration the following criteria:

- the applicant's length of membership;

- applicant's savings;

- the purpose of the loan and the viability of the project to be financed;

- financial strength of LPN.

Ten per cent of the loan volume is retained as compulsory savings.

In principle, loan purpose is limited to income-generating activities but in practice, as money is "fungible", loans are sometimes diverted for consumption purposes. Internal group control results in relatively close adherence to the regulations of the LPN. According to 1983 data, 60% of the loans went into trading, 25% into crafts and small industries, and 15% into agriculture. Group projects and community enterprises have not been financed under the LPN programme.

The annual profits of LPN are allocated as follows: 45% are distributed as dividends among members according to their savings; 20% each are allocated to LPN reserves and to a village development fund, respectively; and 5% each are paid to the manager, to LPN staff, and to an LPN social fund. According to data submitted by 543 LPN in March 1987, profits for the fiscal year 1985-86 were Rp 278.5 million. Accumulated reserves amounted to Rp 284.1 million, which was 95.3% of total capital.

LPN have legal status and are linked to the Development Bank BPD where they deposit their capital and the savings. BPD in turn is refinanced by the Ministry of Finance, from which it received Rp 1.5 million per LPN for onlending to these associations.

BPD loans to LPN are heavily subsidised and the effective interest rate is 7% per annum, or well below market rates. An LPN may obtain up to 3.5 times its deposits in loans but has to meet eligibility criteria in bookkeeping, regular savings, management development, and auditing by BPD. Only 133 LPN have qualified for BPD loans and, by July 1987, Rp 1.38 billion had been approved, and Rp 1.18 disbursed.

The performance of LPN is mixed. The experience with "self-help LPN" is good, but around 60% of those initiated by the government and set up by members to obtain cheap government loans have performed poorly. The 133 LPN which are being refinanced by BPD and which are viable self-help groups gave Rp 2.1 billion in loans to members by March 1987. Arrears at that date were 8.9% of amounts lent, and repayment was estimated to be around 98%. Of the amount lent, 80% was financed from internally generated funds.

Positive features of operations of active LPN include:

- an effective financial intermediation system through self-help groups at the grassroots level integrated into the local culture,

- local deposit and credit facilities very accessible to small borrowers at minimal transaction costs,

- sympathetic use mainly of internally generated funds for loans to members,

- local credit supervision by LPN management,

- use of social controls and pressures to influence repayment,

- linkages to banks as deposit and refinancing institutions,

- provision of technical services by formal banking institutions.

Some negative features include:

- offering loans to some LPN before groups had proven their capability to save and their financial viability,

- providing subsidised refinancing, which results in limiting the scope of the programme,

- large number of inactive LPN, not really based on the principles of "self-help",

- bad debts not written off.

These represent weaknesses in the way the LPN programme was carried out rather than its actual experience.

In West Sumatra around 40% of the LPN were successful. They created "self-help" groups, providing a financial intermediation system with commercial rates with both savings and credit activities at the grassroots level. About half of them established viable linkages between "self-help" groups and banks. On the other hand, the 60% of inactive LPNs show that government pressure to organise groups through subsidised credit does not really lead to a viable financial intermediation system.

3.3 Philippines

More special credit programmes for small enterprises have been introduced in the Philippines than in most other developing countries.

A report in April 1989 stated that at least 15 existing national financing programmes specifically addressed the needs of small and medium-sized enterprises. Not includes in this figure were local or regional efforts catering to microenterprises or cottage industries. The 15 programmes referred only to those that lent more than Ps. 100,000.⁽⁵⁾ Several of these programmes were developed after 1987 when it became policy for credit to be onlent to SMEs in all programmes at commercial interest rates. The credit programmes included lending specifically for project development or acquisition of technologies, financing of fixed assets, acquisition of factory space, working capital, and extra finance for financial restructuring.

These programmes have focused especially on the difficulties encountered by small enterprises in obtaining access to formal credit due to collateral requirements. A number of guarantee schemes have therefore been drawn up. Primary among these was the Industrial Guarantee and Loan Fund (IGLF) which was started in 1952, became active after 1975, but which has not been very successful as a guarantee fund. Later other guarantee funds, such as the Guarantee Fund for Small and Medium Enterprises (GFSME) and the Philippine Export and Foreign Loan Guarantee were set up. Recently, efforts are being made to help set up Mutual Guarantee Associations and a new National Small Business Finance and Guarantee Corporation is planned in the Magna Carta Law for Small Enterprises approved in 1991.

Government banks and commercial financial institutions are actively engaged in some of these special credit programmes for SMEs, but in recent years a feature of SME financing has been the activities of non-governmental organisations or private voluntary organisations (both local and national) addressing the specific financing needs of micro and small enterprises. The best known of these are the Philippine Business for Social Progress (PBSP) and Tulay Sa Pag-unlad (TSPI).

Industrial Guarantee and Loan Fund (IGLF)

The IGLF was first set up with the assistance of USAID in 1952 to provide credit to small and medium industries by re-financing their loans from other financial institutions. The concept of the IGLF was that a part of the loan (60%) would be guaranteed as an incentive to financial intermediaries to use the loan fund. In the first years of operation, the IGLF was little used by banks.

In 1975 the World Bank made a loan to the IGLF in an agreement that it would finance 60% on condition that the Philippines Government contributed an equal share. The World Bank followed up with two further loans in 1979 and 1982.

The refinancing part of the scheme was relatively successful, but the guarantee scheme proved ineffective and the banks continued to insist on collateral lending before making loans. The main reason for this was that the banks were required to exhaust all possible means of collecting from borrowers before claiming from the IGLF guarantee and the banks found this too bureaucratic. After 1982 the financial institutions and borrowers were given the option of contracting for the guarantee. Since they were reluctant to pay the 2% guarantee fee while not availing themselves of the guarantee provisions, only very few took the guarantees up after this.

Originally the loans from the IGLF were intended to finance fixed assets but later this was revised to include permanent working capital. In March 1987 it was decided also to include the financing of short-term working capital for cottage and small and medium industries.

To facilitate the screening of loan applications an accreditation system of banks and non banks was established. Each financial institution was assessed as to financial resources, management and administrative competence and the performance of its loan portfolio. After accreditation loan applications were evaluated. Approved by the accredited financial institution, subject to review by the special unit in the central bank which administered the programme. Non accredited financial institutions were allowed to continue use of the IGLF but on a case by case basis, and with a comparatively low maximum loan of Ps 150,000.

Commercial banks were the most important channel for use of IGLF loans, followed by some private development banks and a few quasi-financial institutions. Average loan size was smallest for the rural banks, but these failed to obtain accreditation and later, because of poor loan recovery, were barred completely from the programme. The rural banks suffered from the highest arrears affecting nearly 60% of the total IGLF loans extended through them.

Until 1986 arrears were high and the IGLF only managed to collect 57% of the amounts due in that year. However, by 1988 this had been substantially improved and the loan recovery rate rose to over 90%.

Depending on the size of the borrowing firm, there were ceilings on IGLG loans. The ceiling for cottage industries with assets of less than Ps. 500,000 (around US$20,000) was Ps. 400,000 (US$15,000); for small industries with assets up to Ps. 5 million (US$250,000) it was Ps. 4 million; and for medium-sized industries with maximum assets of Ps. 20 million (US$ 800,000) the ceiling was Ps. 16 million. In 1988 the IGLF refinanced a total of 556 loans valued at Ps. 1,499 million.

Before 1983 about three-quarters of the loans were to finance fixed assets, but, starting in 1984, most loans were for working capital. By 1987, 63% of all approved loans by IGLF were for working capital and as a result there was a shortening of loan maturity. While in 1981 the majority of loans were for seven or ten years, by 1987 76% were for less than seven years. There has also been a shift in loan beneficiaries from small to medium-sized industries. Medium industries accounted for only one-third of the approved loans in 1983, but by 1987 they constituted three-quarters of the total.

The tendency to cater to larger enterprises is indicative of the conservative lending policy of Philippine commercial banks who were also reluctant to rely on the guarantee part of the programme. There was, in some ways, a conflict in the IGLF regulations with regard to portfolio performance for accreditation. Loan managers claimed that if they relaxed their stringent requirements, arrears would increase and possibly lead to the central bank withdrawing their accredited status. They continued therefore to require collateral in full before approving a loan.

Microfinancing

In 1991, faced with the difficulty of getting its accredited institutions to unbend to very small microenterprises and cottage industries, the IGLF launched a special CB-IGLF Financing Programme for Microenterprises using affiliates of the Philippines Chamber of Commerce and Industry (PCCI) and member companies and private voluntary organisations. The IGFL allocated Ps 50 million to be channelled through PCCI affiliate chambers and member companies or associations to microentrepreneurs. The Land Bank of the Philippines (LBP) was designated a trustee and given responsibility for distributing the funds and collecting repayments from borrowers. The Government Department of Trade and Industry (DTI) was asked to coordinate the programme. On-lending would be done by the PCCI Chamber affiliates who would be screened by the PCCI to determine their eligibility to take part in the programme. The DTI might also take part in the screening process.

PCCI affiliate chambers and the member companies are the actual borrowers of the funds from the IGLF and are therefore also responsible for the repayment of the loan. The loan, however, is not for the chamber or member company's own use but for relending to existing and would-be microentrepreneurs.

IGLF loans made through LBP to PCCI's chamber affiliates and member companies (who would be responsible for repayment to LBP) are for sums from a minimum of Ps. 50,000 (US$2,000 approx.) to Ps. 1,000,000 (approximately US$ 40,000). Loans on-lent to microentrepreneurs to be repaid to an affiliate chamber/member company would have no minimum but a maximum of Ps. 25,000 (approximately US$1,000).

The loan beneficiary would pay an interest rate of 10% to 13% (substantially below commercial levels) which would be apportioned as follows: IGLF 7%; LBP 2%; PCCI 1%; affiliate chambers/member companies 0-3%. The affiliate chamber/member company would therefore be required to pay 10% towards the combined cost of the loan to IGLF, LBP and PCCI.

In order to qualify for participation as an intermediary the affiliate chamber would be required to show that it had the capacity, staff, experience and financial and social position to manage the programme and repayment of loans,and to furnish the reports required. To achieve accreditation as an intermediary the affiliate chamber must also provide proof of the need for the programme in the community.

Until March 1992 there was little movement of funds in this IGLF microentrepreneur programme partially because of the weakness of the PCCI and its chamber affiliates and its lack of experience and capacity in this field.

Guarantee Fund for Small and Medium Enterprise (GFSME)

The GFSME was established in April 1983 and began operating in 1984. It is a special programme attached to the Kilusian Kabuhayan Kaunlaran (KKK) Processing Centre Authority, now renamed The Livelihood Corporation (LifeCor) a government owned corporation. The Board that manages the fund is composed of three professional bankers representating the private sector and four government representatives.

An initial Ps. 302 million was allocated as the GFSME seed fund, Ps. 300 million of which came from the government's Economic Support Fund. By the end of 1989 GFSME resources had grown to Ps. 500 million. GFSME supports export or domestic orientated projects which are engaged in the direct production and processing of food. Export orientated projects are given priority.

GFSME operates through a network of accredited commercial, development, and rural finance institutions and savings banks. To encourage banks to lend to small and medium enterprises, GFSME has been operating the following sub-programmes:

Guarantees. The provision of an 85% guarantee to banks lending to the approved projects within the specified groups to cover all types of credit risks.

Interest rate subsidies. GFSME absorbs interest rate variability and so ensures the profitability of the loan portfolio of participating banks. Recently this subsidy was being phased out in favour of all the special credit programmes lending at market interest rates.

Re-discount facilities. This sub-programme which was phased out in 1990 provided participating banks with a re-discount facility.

There are plans to develop a secondary market for GSME guaranteed loans, but this has been slow to develop.

For projects to be eligible for guarantees under GFSME they must be small with assets of not less than Ps. 82,500 (US$3,500) and not more than Ps. 2.5 million after financing; or medium-sized with assets between Ps. 2.5 million and Ps. 10 million after re-financing. The projects must have a projected rate of return of at least 20% per annum, have a clear market for the output, and be capably managed.

As part of the idea of providing guarantees to encourage commercial banks to lend more to the SMI sector, a number of new guarantee facilities have been developed by GFSME. These include special facilities for farm machinery dealers and for non-farming projects.

In February 1989 the number of accredited financial institutions using GFSME stood at 23 which was a drop from 39 in 1985. By February 1989, 346 projects had been approved, with an aggregate loan value of Ps. 620 million. The average loan size was Ps. 1.8 million (US$90,000). About 42% of the projects approved were to borrowers defined as small-scale enterprises. At the end of February 1989, the guarantee was claimed on 40 accounts due for a total of Ps. 38.7 million, or 10.25% of all projects. Unlike the IGLF, GFSME paid banks promptly for their claims. A default is recognised to have occurred when the borrower has failed to pay three consecutive monthly payments. The lending institution may call in the guarantees ten days after sending a letter demanding payment to the defaultee. The GFSME undertakes to act on the claim within 30 days of receiving the bank's notification. Once the claim has been settled, the GFSME can pursue the defaultee to collect the debt. By 1989 GFSME had managed to recover five accounts on which it had paid guarantees.

Apart from interest the borrower pays an origination fee of 3% of the loan amount and a guarantee fee of 2% per annum of the guarantee portion of the loan still outstanding. Banks may also charge other minor fees for registration and documentation. However, the accounts of GFSME between 1984-87 indicated that the income derived from these guarantee and loan origination fees had not been enough to cover general administrative expenses.

Apart from the guarantee programmes there are also a number of special credit programmes concerned with direct lending.

Tulong Sa Tao Financing (TST)

TST, established in 1987, is among the most important credit programmes in the Philippines. It is a rural based financial assistance programme for small-scale entrepreneurs. It was set up with a funding of Ps. 100 million from the Philippines government and has also benefited from a US$8.8 million loan from the Asian Development Bank (ADB). Out of this ADB loan, $8 million is used for lending, while $0.8m is used for institution strengthening of NGOs and entrepreneurship development programmes for clients. The TST fund is administered by the Department of Trade and Industry (DTI) through its regional offices. TST financing has two programmes: (1) self-employment loan assistance, and (2) subcontracting financing.

Until early 1990 a total of 135 NGOs nationwide had availed themselves of TST-SELA (Self-Employment Loan Assistance loans) with a total value of Ps. 38.7 million to benefit 7,287 borrowers. Under the TST-SELA scheme, the fund lends to DTI accredited NGOs who in turn re-lend to individuals or groups. To be accredited an NGO must be registered, have a legal identity and have operating experience of at least one year in community development income generating projects for low income groups, have a board of managers with standing in the local community, a minimum staff/beneficiary group ratio of 1:20, a staff training programme and at least 20 square metres of office space. On the financial side it must have external auditors, a minimum net worth of Ps. 100,000 and a net worth to risk asset ratio of not less than 1:5. In its lending operations for at least one year, the NGO must be able to show a debt collection rate of over 80%. The projects implemented must have a record of investment cost per job generated of not more than Ps. 15,000 and be capable of providing 15% of the financing requirements of projects.

The TST-SELA requires NGOs to re-lend at least 80% of the total loan received from the programme in the field of manufacturing and processing. NGOs may obtain a second loan after 80% of the original loan has been successfully lent to intended beneficiaries.

An accredited NGO can borrow at least Ps. 50,000 to a maximum of Ps. 2 million at a rate of 7% per annum which remains fixed during the term of the loan. The repayment period is for five years inclusive of a grace period. There is a penalty charge of 1% per month for arrears. Despite the subsidised rate at which the NGO receives TST-SELA funding, the lending rate to the final beneficiary should be at an interest rate close to the local commercial bank rate. In general, recipients are microenterprises who may use the loan for working capital purposes or for capital investments. The sub-loans to microentrepreneur and "self-help" groups onlent by the NGO may be for a maximum of Ps. 25,000 for individuals and to Ps. 200,000 for self-help groups and smaller NGOs.

The TST has a second programme, TST-SCF (subcontracting and financing), which provides financing to contractors and subcontractors engaged in business with cottage, small and medium-scale enterprises. The sectors that are given priority in this programme are the garment, gift, household appliance, furniture, metal working, footwear and leather goods sectors. To benefit, an enterprise in any of these sectors must have been in operation for at least a year, be registered with the DTI and with the Department's subcontracting programme. Loans can be used either for the acquisition of machinery, costs of production, and/or as cash advances to subcontractors. The maximum amount lent varies according to the debt servicing capacity of the borrower and the collateral offered, but the amount shouls not to exceed Ps. 1 million per borrower. The interest is fixed for the term of the loan and is determined by the TST Fund Management Committee. A 2% fee based on the loan value is charged for each loan, of which the maximum term is five years. An important feature of this programme is that the loan is flexible, depending on the project's cash flow and the interest is calculated on the principal outstanding.

Each loan is secured by the signature of the owner, or owners in the case of a company or partnership, plus a first or second mortgage on lands, buildings or machinery. Borrowers can use savings or time deposits as collateral.

The Tulay Sa Pag Unlad Inc (TSPI) (The Bridge to Progress) and the loan programme of the PBSP (The Philippine Business for Social Progress) are special credit programmes for small and microenterprises of particular interest.

Tulay Sa Pag Unlad Inc (TSPI)

The TSPI, which began operating in 1982, is the product of collaboration between a group of Filipino and American Christian business people. Its board of local businessmen is very active, not only in managing the programme, formulating the policy and raising funds, but also in selecting clients and monitoring lending procedures. Board members also try to help programme beneficiaries become clients of commercial banks.

TSPI specialises in loans for the expansion of existing enterprises rather than to support new entrepreneurs or start ups. However, two of the six provincial subsidiaries of this programme, the KMBI and TKSI, do try to help aspiring entrepreneurs because of the extreme poverty in the areas in which they operate.

TSPI and its provincial subsidiaries make loans to individuals normally without a savings requirement. Repayment is tailored to the the agreement of the members of the board. The Board and programme staff take active responsibility for assessing project proposals, monitoring repayments and helping clients.

Because of the transaction costs involved minimum loan size is Ps. 10,000 (US$500). Repayment is made through a system of post-dated cheques signed at the time the loan is approved and which are presented to the bank by TSPI on a monthly basis. The borrower undertakes to make regular deposits of not less than the amount necessary to meet the agreed repayment schedule. The purpose of the TSPI programme is to "graduate" clients towards commercial banks. In 1988 TSPI started negotiating with the Bank of Philippines Foundation to find ways with minimal administrative costs to service the credit needs of small borrowers. TSPI has been replicated in six provincial subsidiaries. Its interest rate policy is to lend at the same rate as the formal banking sector.

Philippines Business for Social Progress (PBSP)

PBSP was set up in December 1970 in an attempt by the Philippine Council for Economic Development and the Philippine Business Council to deal with problems of social unrest. Through PBSP the Philippine business community raises funds and provides technical support to socio-economic projects across the country.

PBSP sees itself as a Foundation dedicated to the establishment of a community-based credit and extension programme to enable the poor with no access to formal credit sources, to become involved in business to generate income and so to improve their livelihood. Apart from credit, PBSP also includes in its programme training and consulting activities and other support services for very small enterprises. From 1988 to 1991 it set itself the goal of helping projects that would create 1,800 new full-time jobs.

PBSP is an independent, indigenous NGO, whose major sources of income until recently were contributions from member companies (67%) and interest earned on funds loaned to PBSP's property programme (28%). Each of its 118 member companies contributes 0.6% of pre-tax income to PBSP, which in 1987-88 amounted to Ps. 74.8 million (US$3.6 million). In the last year USAID has channelled funds through PSBP but some believe that donor funding has simply served to dilute domestic contributions.

PBSP has implemented its programmes by making loans and grants to local NGOs, community organisations, cooperatives and marketing groups. PBSP acts as an intermediary between microentrepreneurs and domestic and foreign corporations in the formal modern economy. In this brokering role, and as a direct lender to community-based groups for on-lending to microenterprises, PBSP channelled an estimated US$2 million for income generating purposes between 1971 and 1987.

PBSP has also set up an Enterprise Fund Facility (EFF) for NGOs, people's organisations, cooperatives and individual entrepreneurs. The conditions for NGO participation in the EFF are similar to those for the TST. For the people's organisations and cooperatives there too, as in the case of NGOs, there must be willingness and capacity to put up as equity 20% of the total project cost. Individual entrepreneurs must be ready to share up to 30% of the equity with the employees and to finance 30% of project costs from their own resources.

Proposed EFF projects must have capital of Ps. 250,000 to Ps. 5 million, a minimum projected investment return of 20% and generate one employment place for every Ps. 20,000 invested. Requests to the EFF are processed through an Enterprise Resource Centre (ERC) set up with local Chambers of Commerce. The ERC screens the proposal and after confirming its viability will recommend financing from the EFF but only after formal financial institutions have refused to fund the project.

Rural Banks

A feature of the Philippines which some consider could be used more for SME lending is the large network of over one thousand rural banks. Interviews with small and medium entrepreneurs show that they consider it a great advantage to deal with a bank close to their place of operations and that they are reluctant to deal with banks far away in the capital. Rural banks are a part of the community and are a possible conduit for lending to SMEs because they are part of Filipino social culture. On the negative side, however, the collection record of some rural banks has been poor and when the IGLF tried to use them, arrears were very high. Furthermore, a many rural banks have closed because of poor operations. If a way could be found to screen the rural banks and monitor their operations, it should be possible to allow the more successful and more competent lf them to participate in special credit programmes for SMEs.

Cottage Enterprise Finance Project (CEFP)

The CEFP is a programme which started in 1991 to facilitate cottage enterprises' access to credit through the Mutual Guarantee Association (MGA) to overcome the problem of lack of collateral.

The programme is open to individual enterprise owners, partnerships, and small companies with gross assets of between Ps. 50,000 and Ps. 1 million and less than 50 employees.

The CEFP offers short-term loans during the first year of operation of an enterprise to finance inventories or receivables. The project also offers medium to long-term loans (five years maximum) with a maximum of three years grace to finance fixed asset/capital equipment, to build new plant, or to expand existing facilities.

Finance for the CEFP is provided by the World Bank (WR) through a US$15 million loan to the Development Bank of the Philippines (DBP) for rediscounting loans made by participating financial institutions (PFI). The Kreditanstalt for Weideraufbau is cofinancing the CEFP with a DM20 million loan to DBP for similar purposes as the WB loan. KFW is providing a small Technical Assistance Grant for the project.

The Department of Trade and Industry (DTI) acts as liaison for producer associations throughout the country. Members of these associations form the bulk of MGA membership; PFI and DGP accredited banks act as the direct lenders to MGA members.

To satisfy bank collateral requirements the CEFP establishes a Guarantee Fund to guarantee up to 80% of individual loans to members of the MGA. Borrowers provide direct collateral for the remaining 20%. MGAs are formed by 40 to 100 cottage industry entrepreneurs. MGA members take equity in the common stock of the association which registers as a corporation with the Securities and Exchange Commission (CEC). Initial equity contributions (EIC) form the Reserve Liquid Fund (RLF). The DBP matches the RLF on a 2:1 basis during the first year of the MGA's operations. The RLF and the DBP's Matching Loan Fund (MLF) constitute the Guarantee Fund (GF).

The programme allows a leverage coefficient of 3 x GF, thus increasing the amount available for borrowing by MGA members. The maximum loan that may be made to a MGA member is GF x 3 divided by the total number of members. This equals 9 times the EIC, but may not to exceed Ps. 1 million.

As the RLF increases either due to new members, increased membership contributions or investment earnings of original EICs, the maximum loan available to each MGA member also increases. However, the programme also imposes a minimum loan size of Ps. 50,000 (approximately US$2,000). A guarantee fee of 2.5% of the guaranteed amount is charged for each loan. The income from the fee goes into the RLF.

The CEFP programme is open to enterprises engaged in industry and services including agri-businesses. New enterprises are eligible if they can show their businesses are viable but in general membership requires one year proven track record in business.

At the time of writing the CEFP is in its very early stages. The MGA concept is borrowed from successful experiences in European countries. The international development finance institutions have been interested in promoting MGA in developing countries and this effort in the Philippines is one of the first trials in this direction.

Conclusion

As can be seen, over the last five years special credit programmes for SMIs and microenterprises have proliferated in the Philippines. This is due in part to the fact that the financial liberalisation introduced by the government in 1987, when ceilings on lending and deposit rates were removed and the Central Bank abandoned resource allocations. This did not result in improved access for SMEs to financial services offered by the banking sector. It highlighted the over conservatism of the banks in the face of excessively high transaction costs and administrative and default expenses of lending to SMEs in the Philippines. Banks continue to be collateral orientated and there is little likelihood of them using their own deposit funds for SMEs who do not have adequate collateral.

An interesting study undertaken in the Philippines in 1987-88 of 600 sample firms showed that, as in many other countries, small firms do not depend on institutional finance to start their enterprises. They usually tap informal sources of credit, and the resources of family and friends, as well as their own savings, to start and expand their enterprises. It is clear therefore, that in the Philippines all credits, whether from banks or other institutions, only complement the other resources of the small entrepreneur.

The survey also found that NGOs in the Philippines, were successfully introducing credit programmes for microenterprises as an additional package of services to the management, training and counselling which they were offering. Because of this close relationship with their clients and the considerable information they have on them, they are able to lend with lower transaction costs.

3.4 Sri Lanka

Sri Lanka, an island economy with 16 million inhabitants, torn over the past few years by ethnic strife, needs small and medium enterprise development to diversify its economy and achieve balanced growth. Since the 1970s the government has looked to SMEs as the centrepiece of its policies for industrial development after a long flirtation with large state-owned enterprises in the 1960s.

Credit is recognised as a vital element to improve the capacity of the poor and unemployed and to stimulate the well-educated Sri Lankan entrepreneurial talent to contribute to industrial and economic development. There are considered to be three possible forms of SME credit in Sri Lanka.

1. Credit provided by the formal banking system or through non-bank financial institutions which function as banks.

2. Credit obtained through an informal system combining traditional credit structures and new thrift structures that embody the main features of the traditional informal system.

3. Credit obtained from a system that combines features of both the above and that attempts to link the formal and informal systems.

Small enterprise credit

Credit institutions addressing self-employment and microenterprises in Sri Lanka can be classified into three broad types: formal, semi-formal and informal sector institutions.

Formal institutions include the state commercial banks (Bank of Ceylon and the Peoples' Bank), the private banks, and the National Development Bank, the National Savings Bank, the Development Finance Company, and the foreign banks.

Semi-formal institutions include the Regional Rural Development Banks (RRDB), the Co-operative Rural Banks (CRB), the three levels of the Thrift and Credit Societies (TCCS) - and those NGOs that are delivering credit as part of their assistance programmes. The largest of the local NGOs is Sarvodaya, where credit and income productivity is managed by Sarvodaya Economic Enterprise Development Services (SEEDS).

The informal financial sector includes moneylenders, merchants, family and traditional savings groups.

Formal institutions operate through a branch network of over 600 state and private commercial bank branches, who have been constrained in their outreach to small borrowers by low profitability and high defaults. The informal style is characterised by a familiarity between lender and borrower, a long-term relationship, and usually involves much higher interest rates than those of banks.

Some semi-formal institutions have branch networks but rely primarily on group-savings and mutual guarantee approaches to reach groups of borrowers. These groups operate through "peer pressure" to mobilise savings and collect loans from members. Hence, the semi formal branch network lenders combine aspects of both formal and informal styles. The RRDBs are the biggest example of a semi-formal institution created to fill the credit gap unserved by banks. They experiment with "group-lending" schemes and use NGOs as intermediaries.

The semi-formal CRBs serve mainly as savings mechanisms to absorb rural deposits for the People's Bank. CRB lending is concentrated in collateralised credits but they do not engage in lending to groups.

The NGOs endorse group savings and approaches where the creditworthiness of individual borrowers is backed by solidarity groups and guaranteed by mutual savings. Generally the group approach lowers the transactional costs for both the borrower and the lender, and results in a much improved level of loan recovery. In certain cases, NGOs and government agencies act as brokers, introducing borrowers to banks and performing services such as project analysis and loan appraisal that banks would normally undertake with larger customers.

Informal sources of credit include non-bank pawnbrokers, established moneylenders, supplier credits, voluntary credit groups (cheetus) and family transfers.

Links between the informal and the formal credit systems have been institutionalised in the Praja Naya Niyamakas (PNN) schemes of the state banks, and the Sampath Seva Sanyojaka (SSS) scheme of Sampath Bank (a new private bank started in 1986), whereby credit is extended by commercial bank branches to approved individuals and merchants for on-lending in turn to their clients.

Of special interest at this time in Sri Lanka is whether the Formal Sector can innovate credit practices to serve profitably a large number of viable small borrowers excluded until now from the formal credit markets. Another possibility is more competition in the informal credit markets which over time may bring down the high informal interest rates.

Formal systems in Sri Lanka suffer from high administrative costs, based on banking practice geared to large enterprises, and on standards of credit analysis which are inappropriate and costly for micro enterprise and SSE landing. These tend to preclude handling small transactions. Low cost solutions to small transactions in the urban sector include cash machines and credit cards, but these are inappropriate for the poor and unemployed clients. Using intermediaries who are known to the bank is a way Formal Lenders can maintain credit standards and at the same time reach out to smaller customers.

Linking formal and informal credit systems

There are definite advantages to linking the informal and the formal credit systems, so incorporating the positive features of informal moneylending into a broader system to serve micro enterprises and small entrepreneurs. The benefits are lower transaction costs, higher loan recovery, and provision of services (credit and management assistance) to poor borrowers. Transaction costs can be kept low by using agents, by shifting responsibility to the clients themselves to perform credit analysis, and by using "peer pressure" to ensure debt collection. Besides controlling costs, these "grassroots techniques" improve loan recovery and make sure that the money is going to the targeted client for the purpose for which it was requested.

The advantages to the formal lender of rural agency schemes extend beyond cost control to improved credit repayment. With loan agent schemes, the funds are not perceived as government funds (quasi grants) since the loan agent is at full risk. Using loan agents helps prevent a situation where large numbers of clients would default due to expectations of "government forgiveness". Also, the agent has better knowledge of the individual borrowers and their circumstances than do the formal lenders. Relationship with and information on clients are the most important factors in successful credit management and in keeping defaults low.

Credit approaches which bring together formal and informal credit are similar to the traditional "shroff" system (as in India), where a bank uses a local person as an intermediary. As intermediary the "shroff" might recommend a customer to the bank (broker role), share risk on a loan (guarantee role), or borrow funds from the bank to on-lend at full risk (full on-lending role). The value of the "shroff" is that he knows the small borrower and this allows him to take a risk that would be unacceptable to a bank. In the hybrid approaches, NGOs and loan agents of banks (PNNs and SSSs) serve the same, traditional role as the "shroff". Criticism of the informal credit market has focused on abuses rather than on the less recognised advantages of brokerage intermediation.

The difficulty in the agency system lies in the selection and professionalism of the agents. A weak or poorly motivated agent limits his risk by lending only to family and close personal contacts. Most of the PNN agents of People's Bank and Bank of Ceylon are political choices. State banks find it most difficult to control agent quality and yet this can be the most important factor in the success or failure of the agency programme to reach poor and unemployed clients.

Targeted programmes

Most programmes in Sri Lanka providing credit for self-employment and microenterprise "target" the unemployed and the chronic poor. Targeted programmes screen borrowers for eligibility based on social considerations which usually leads to the poor being given heavy subsidies and non-targeted borrowers being excluded. Consequently, targeted programmes dictate to lenders non-financial eligibility criteria of clients that is difficult and costly for the lender to administer, and time consuming for the borrower. To keep borrowing costs low, targeted programmes lay down interest rate ceilings virtually making them unprofitable which saps the motivation of lenders to carry on with the programmes. Lenders are saddled with additional costs but cannot pass these on to the borrower, so the result is low lending volume.

Targeted programmes it is claimed are typically costly for the lender to administer and troublesome even for the borrower who qualifies. Excluding borrowers who do not fit the eligibility profile may deny credit to those entrepreneurs who have the best potential for creating jobs.

A non-targeted approach

A non-targeted approach could influence the supply side of credit through better management and training of loan agents and bank staff to accommodate the special nature of small enterprise and group-based lending.

The problem with demand-side programmes is uncertainty about reaching client groups. This is a criticism of the PNN loan agent programmes. The best results are obtained when groups of poor and unemployed manage their own savings and credit programmes. They know best how to select borrowers and recover loans. Groups usually offer their members good rates for deposits, charge higher rates for loans, and distribute profits periodically. The TCCS and Sarvodaya programmes partly fit this group approach. It takes years to build up trust and savings, to train and alter behaviour, and so the best programmes have grown slowly.

TCCS in particular, fear that donor money will drown their traditional savings and credit culture and so are cautious about the larger bilateral and multilateral donors such as USAID and the World Bank. The history of "Government forgiveness" of previous USAID loan schemes hangs over all discussions and is ever present as an implicit, political risk.

Supply of credit for microenterprises and self-employment

The People's Bank has allocated Rs 224.0 million for PNN with about half of the Rs 80 million already advanced having gone for SMEs. In addition, People's Bank has about Rs 30 million in various other schemes for self-employment and microenterprises.

The Bank of Ceylon has allocated about Rs 700 million⁽⁶⁾ for its PNN, NYSCO (National Youth Services Council), Self-Employment and Poverty Alleviation Scheme. Sampath Bank has on-lent about Rs 4.5 million through its SSS loan agent scheme. Hatton National Bank, another private bank, has lent Rs 7 million during the past six months through its village lending experiment. The RRDBs lent Rs 17.7 million for SMEs among their 11 regions and 80 branches. The National Savings Bank has an SME programme of Rs 17.8 million. Among the semi-formals, the CRBs are lending about Rs 100 million while their pawnbroking amounted to almost Rs.400 million. Aside from its bank related programmes, NYSCO has raised as much as Rs 20 million annually in shares, deposits, and a lottery to contribute to youth enterprise schemes. Sarvodaya SEEDS has advanced Rs 22.0MM in two years of operations to September 1989. TCCS is reported to be lending more than Rs 300 million per year to its members.

While the total estimated credit available to small enterprise is impressive, the incentive factor for mass lending is lacking. The terms and conditions of lending to the poor are either unprofitable - due to high transaction costs and high defaults - in targeted and intensely monitored programmes, or in the case of PNN, the funds do not always reach the poor and unemployed categories for productive investments.

These large-scale, supply-driven, short-term credit programmes for the poor and unemployed are generally unsustainable. The supply side programmes are usually cautious of expanding too quickly. In future more will be gained by building up the demand capacity from the targeted borrowers. Response to a market-led approach will involve financial institutions not only as lenders, but as networkers and brokers. A market-led approach also requires promotion, training, and technical assistance, through NGOs and extension services. Where rural credit markets respond to an organised credit demand - at the right prices - sustainability will be more possibly in the long run.

A microenterprise approach to self-employment through the informal sector, complemented by credit responding to organised demand, can help the poor by providing know-how and finance to individuals and groups with feasible projects and markets.

People's Bank (PB)

As of September 1989, PB had 26 various loan schemes on its books. The volume of loans extended through these schemes was generally very low, partly due to the security problem in the whole country. As of March 1990, schemes to assist self-employment and microenterprise were picking up. Examples of such schemes are:

Praja Naya Niyamaka (PNN). As of 15 March 1990, 2,603 agents were in operation in 25 districts, about half of the total number appointed. The goal is 7,000 agents. The amount allocated by the bank for the scheme is Rs.224 million, of which 34% has been advanced in about 17,500 loans. This scheme is widely criticised on grounds that the PNN agents are political appointees, and that there is no monitoring system for the on-lending.

Athamaru Loan Scheme. Small loans, for two months, at 5% per month to members of producer groups, with forced savings. Group leadership is compensated through commission incentives. As a pilot project, activities are spread island-wide, averaging 50 loans per branch, Rs 17.6 million allocated, 5.5% disbursed to 979 borrowers by February 1990.

Startup Enterprises Loan Scheme for Technically Qualified Youth. eligibility for the scheme are: 18-40 years old w/ vocational training; It offers loans of Rs.30,000, for 5-7 years, at 18% per annum, for equipment and working capital. Security requirements are lax relative to loan size. The scheme uses religious organisations and NGOs to market the idea. Rs.37 million has been advanced to 2,793 borrowers by end 1989.

Self-Employment Loan Scheme for Members of NYSCO. This scheme is for people 18-35 years old, identified and recommended by NYSCO which provides supervision. The scheme offers small business loans up to Rs 25 million at 16% per annum. By the end of 1989, 1,163 loans had been disbursed totalling Rs 12.3 million.

Bank of Ceylon (BC)

As a state bank, BC has many of the same microenterprise and self-employment lending programmes as the People's Bank. It participates in the World Bank funded SMI III scheme and is also a lender in the NYSCO loan scheme.

Sampath Bank

Sampath Bank, a private bank begun in 1986, had 10 branches within three years, with deposits of Rs 2.1 billion covering total loans of Rs 1.5 billion. Sampath prides itself on being at the forefront of banking innovation in Sri Lanka. It has connected its branches in Kurunegala and Kandy to the central computer system in Colombo for on-line processing of transactions with updating of records. Sampath maintains a loan loss reserve by charging provisions for loss against net interest income. Loan losses are deducted from the reserve and subsequent recoveries are added back.

Sampath has begun to innovate profitable programmes to serve self-employment and microenterprise. It also participates in the World Bank funded SMI III scheme. Examples of some of its small business assistance schemes are:

The Sampath Seva Sanyojaka Scheme (SSS). Sampath has also engaged in agency banking experiments to serve in these self-employment schemes in Biyagama. Individuals are chosen by Sampath from the community on the basis of capability and suitability to serve as intermediaries between Sampath and small entrepreneurs. Some of these agents are retired civil servants. These agents can serve not only as moneylenders (receiving funds at 21% pa, lending at 42%), but can also provide technical assistance and marketing advice for their clients.

The SSS network resembles a group scheme. The average number of borrowers per SSS agent is about 40, a generous number given the intensity of the operations. With a modest failure rate on lending (many loans are startups), an agent can earn in excess of Rs.5,000 per month. Some mushroom growing and ornamental fish raising projects have been financed. Sampath studies markets and, where appropriate, sets up subcontracting arrangements. It contacts suitable local training institutions to deliver training to agents who then can impart the skills to their client producers/borrowers. Sampath also finances exporters of the finished products.

After some "startup" losses Sampath was showing profits on the scheme in 1990. As of 31 March 1990, 1,979 loans had been funded to the cumulative amount of Rs.4.4 million with no loan arrears exceeding three months. Savings have been mobilised to the amount of Rs.1.4 million or 32% of loans. The administration of the programme has required only two persons in the bank.

Mini-Banking. Sampath management plans to expand the SSS experiment and establish a Mini-Banking Division linked to the branch network, to help microenterprises, primarily in the informal sector. Sampath plans to set up seven new branches.

The Sampath SSS scheme is similar to the state bank PNN scheme, but Sampath has chosen its agents carefully. The other features are: greater spread on loans, better loan supervision, and the bank offering technical assistance. The loan spread is 1.75% per month versus 1.0% under the PNN. Rather than leave the programme to pure market forces, Sampath keeps a close watch on lending activities to see that they are applied correctly. The bank also assists its agents by arranging training for them and their borrowers and by helping them to identify investments which have markets and buyers. The bank even helps to make important market connections.

The programme will have higher administrative costs but this may be covered by higher on-lending rates which could offer greater incentives to agents. Borrowers, while paying more than to PNNs, may find the extra services of technical assistance and market connections worth the price.

Hatton National Bank

Hatton Bank, another private bank, has an experimental Gami Pubuduwa "Village Awakening" programme. It uses bank employees as mobile bankers in villages to bring services to the very poor. The average loan size of Rs.20,000 is relatively large compared with other "social banking" programmes, and this is due to Hatton's aim to stimulate small industry in rural areas. At the end of 1989, 13 Gami Pubudu Units had been established and financing extended to 802 projects. Recovery rate was 99% which is excellent for any programme.

For project selection, credit analysis, and loan supervision, Hatton depends on the active involvement of its Gama Pubuduwa mobile bankers to familiarise themselves with the management talent and commitment of the entrepreneur, and to assess the productive capacity of the enterprise.

National Development Bank (NDB)

The NDB, set up in 1979, provides funds and facilities to commercial banks to encourage them to embark on long-term lending for small and medium-sized projects. It acts mainly as an APEX refinancing institution (IDA) for participating credit institutions such as in the World Bank SMI loan programme.

NDB earned Rs 103.9 million in 1988 of which 41% was attributed to SMI fund operations. These operations consisted of net interest income on funds on-lent to participating credit institutions, and interest earned on reflows invested in liquid instruments.

In 1988, 28% of NDB's total assets of Rs 3.7 billion, Rs 819 million were SMI loans. The NDB finances its portfolio with Government of Sri Lanka/World Bank/IDA and Asian Development Bank credit lines and a small amount of Central Bank refinance. NDB approved 528 SMI loans in 1988, for an average size of Rs 493,000 ($12,325) compared to the average loan size for equipment in non-SMI credit lines in 1988 which was about 13 times greater, at Rs 6.5 million ($164,325).

The smallest loans under SMI refinancing are about Rs 25,000 (US$625). Of the industries financed through direct facilities, metals and chemicals (20%), service industries (16%), and food, beverages, and tobacco (13%) predominate. The textile industry, which creates 7 out of every 10 new jobs among their enterprises, takes 10% of NDB's direct financing. Some private commercial banks have taken the opportunity to set up business links between producer groups and these SMEs.

National Savings Bank (NSB)

NSB attracts savings by paying attractive rates of interest and by maintaining convenient deposit facilities at more than 3,000 post offices. In 1988, operating through a network of 64 of its own branches and 3,828 post offices and sub-post offices, NSB mobilised Rs.1,663.6 million in savings net of withdrawals. NSB's market share of total savings mobilised by all banks amounted to 32.5%. A Pilot Rural Credit Project in the Matara District was funded through NSB by the German Savings Bank Association, lending Rs 17.8 million to 3,700 borrowers.

Non-Governmental Organisations (NGOs)

The Sarvodaya Economic Enterprises Development Services (SEEDS)

The Sarvodaya Shramadana Movement, started in 1958, has grown from a small informal movement into one that is active in 5,600 villages of Sri Lanka. Sarvodaya operates through village societies which are established and run by the villagers themselves.

SEEDS was set up in September 1986 as a separate division within the Movement, with a mandate to focus on economic development of the village societies, their members, and of the Movement itself. SEEDS runs three complementary programmes which target rural producers (1) Rural Enterprise Programme (REP) which organises savings and credit units in villages and trains its members in financial management - savings are re-lent by the villages to members and also used to leverage loans from the SEEDS district level; (2) Rural Enterprises Development Service (REDS), a technical assistance programme designed to help producer groups improve their enterprises through market development and technology improvement. REDS is a potential profit centre through sales of literature and through consultancy fees; and (3) Sarvodaya Management Training Institute (SMTI) which trains Sarvodaya and SEEDS staff in management and programme administration. SMTI charges fees for its courses to external students. Other features of SEED's programmes include the following:

Borrower eligibility. SEEDS do not dictate conditions of eligibility for group membership. Group members are responsible for selection. Members must have been saving for six months prior to receiving a loan, must have a feasible proposal, and the proposal must be approved by the group.

Programme feasibility. SEEDS feel that the goodwill gained through the grass-roots organisation of Sarvodaya, can fuel enterprise development. Linkages to market advice and training improve productivity and profitability. High repayment is achieved through close field control by the members and by REP staff.

Loan security. Before it releases a loan the REP district office takes a non-refundable 2.5% cash collateral as contribution to a Loan Risk Fund from the member's village society. The member is also required to have deposited 10% of the requested loan amount to the village society which is refundable with interest at maturity of the loan. The village society guarantees all loans to its members.

Programme expansion. During the first two years of operations, SEEDS extended 11,602 loans in the amount of Rs 21.9 million. Approximately 36% of loans were approved by SEEDS proper, 28% funded by district savings, and 35% by village savings. Banks funded 105 loans or 1%.

SEEDS projects the volume of lending to expand from 6,000 loans totalling Rs 30 million in 1990, to over 30,000 loans by 1994. Average loan size should double over the same time period. In addition, SEEDS is negotiating with state banks to obtain better interest rates on deposits for its savings. It has also demonstrated that it can work effectively as a brokerage unit for the banks by doing the preparatory work normally done by the bank before approving a loan.

CHAPTER 4

CREDIT PROGRAMMES FOR MICROENTERPRISES IN LATIN AMERICA

For Latin America the 1980s were characterised by economic crisis and regression. Throughout the decade per capita production fell and in most countries was lower in 1990 than it had been in 1980. It is estimated that fixed investment productive capital fell by over 30% between 1980 and 1985.

Not surprisingly, poverty has increased in all countries of the region and population growth has exacerbated the problem. The informal sector has grown at a faster rate reflecting the inability of the formal economy to absorb the new entrants to the labour force and governments have begun to recognise the informal sector as an alternative source of jobs and are beginning to respond to its needs in an effort to create employment opportunities for the growing population and to alleviate the increasing poverty.

In most Latin American countries today there are programmes providing credit and other forms of assistance to the microenterprises that comprise the large informal sectors. In countries such as Colombia, Ecuador, Mexico, Peru, Bolivia, and the Dominican Republic the informal sector provides 50% or more of employment.

Most programmes of assistance to microenterprises and the informal sector in Latin America originate with private development organisations (PDOs), in the form of foundations established in the various major cities by individual businesses or community business groups motivated to try to help the informal sector enhance incomes and create more employment, and possibly ultimately to graduate units from the informal to the formal economy. It is estimated that such organisations now extend credit and training in over 65 cities throughout Latin America to very small microproducers and retailers. The scope of these NGO/PDOs is considerable and their impact significant. Today it is estimated that these organisations disburse over US$2 million each month in very small loans. Examples of assistance developed by NGOs/PDOs in countries such as Colombia, Bolivia, Peru and Mexico are described below to illustrate how these organisations have sought to provide finance and other forms of assistance to microenterprises in the informal sector. Similar programmes operate in Costa Rica, Honduras, Ecuador, Paraguay, Dominican Republic and in some Caribbean countries.

4.1 Bolivia

Bolivia's economic problems developed over the period 1950-85 when governments favoured strong state involvement in the economy. Government policies in 1982-85 resulted in an economic collapse that was characterised by hyperinflation and decapitalisation of the formal financial sector. During this period Bolivia's financial sector recovered only a fraction of the loan repayments due to them and as a result very few loans were made to businesses or individuals outside a relatively small group of corporations that themselves owned the banks. A strong austerity economic recovery programme introduced by a new government in 1985 put greater reliance on the private sector's role in the economy. Taxes and tariffs were reformed and as a result of new policies on wages and a liberalisation of labour laws, inflation was reduced from over 22,000% in 1985 to 12% in 1988.

This policy helped to rehabilitate the Bolivian economy but its short-term effects on employment and the income of the poor was serious. More than 25% of the labour force was unemployed and another 25% considered significantly underemployed.

The microenterprise sector, which includes both formal and informal businesses, is reckoned to provide employment for well over half of the non-farming population. As might be expected the sector expanded rapidly during the economic depression and hyper-inflation. Large numbers of urban poor and unemployed established tiny service and manufacturing enterprises as a means of survival. A 1986 study showed that 52% of the economically active population was employed in the informal sector. In La Paz the study showed that the informal sector accounted for 74% of jobs in manufacturing and 86% in commerce. At least 90% of all businesses employed fewer than five persons.

As the informal sector grew and poverty and unemployment increased, external donors, especially USAID, came in with programmes to help the microenterprise sector. In particular they helped set up a private non-profit organisation (PRODEM) which started a programme to support micro-enterprises in La Paz in 1987. PRODEM also received support from Accion International, a United States based non-profit NGO with widespread programmes to help microenterprises throughout Latin America. PRODEM followed the methodology that had been employed in Colombia and other Latin American countries. Its clients were organised into self-formed solidarity groups whose members were collectively responsible for loan repayment and for guaranteeing the loans of group members. Generally, these clients of PRODEM had one or two employees when they first became involved in the programme. PRODEM provided short-term working capital loans of US$123 on average and some minimal technical assistance mainly related to business analysis and loan supervision.

While PRODEM was regarded as a suitable vehicle for USAID support to microenterprises, the loan portfolio for larger small enterprises was given over to the National Federation of Credit and Savings Cooperatives (FENACRE) which has nation-wide coverage through 100 affiliated credit unions providing credit as well as marketing and management assistance through its member cooperatives. It also has a savings programme.

The USAID designed programme for Bolivia intended that PRODEM should target informal family-based businesses with assets of less than the equivalent of US$5,000, annual sales of a similar volume and employing fewer than five employees. PRODEM would lend the equivalent of $US500 to $US5,000 mainly for working capital and beneficiaries would be entitled to higher loans once they had successfully repaid off earlier loans. FENACRE was to target the next size of small enterprises, those with assets of US$5,000-10,000 and with 5-10 employees and would offer a loan for investment capital up to a maximum of US$10,000. The programme also encouraged the development of credit unions, which would act as support groups for the financial needs of small enterprises. Efforts were also made to involve the regional associations of small industrialists and artisans (ADEPI) which were part of FEBOPI, an organisation formed in 1986 as a national association for small-scale entrepreneurs.

A review of PRODEM's activities in the period 1987-91 showed that it provided microcredits and business training to 25,000 informal micro-enterprise owners with assets of under US$2,500. Women represented 82% of the clients and average client income was US$30 per month; 64% of the loans were granted for commercial activities - mainly selling food; and 36% of loans were for manufacturing enterprises. PRODEM also developed a loan programme for solidarity groups of 4-7 members who were responsible for loan repayment and for guaranteeing the loan.

In October 1990, PRODEM became the first of Accion's affiliated programmes in Latin America to lend over US$1 million in a month, a considerable achievement for a staff of only 150 persons. Its loan recovery rate was over 99%. In the four years 1987-91 PRODEM disbursed US$13 million (US$7.7 million in 1990) and mobilised savings of over US$500,000.

Potential borrowers were given assistance with their initial loan application. Loans were repaid directly to PRODEM's account at the Commercial Bank in weekly or monthly instalments. Although the official loan ceiling was US$3,000, PRODEM's average loan size was US$227. The objective was that after receiving a few loans from PRODEM over four or five years, they would graduate to the formal banking system.

After four years PRODEM came to the conclusion that in its present limited legal structure it could not grow sufficiently to meet the increasing needs of the Bolivian informal sector. It therefore decided in conjunction with USAID, Accion International and the Calmeadow Foundation of Canada to set up the Banco Solidario SA (or Banco Sol as it is called) which started operation in January 1992 and is the first microenterprise bank in Latin America. The International Finance Corporation and private the sector fund of the Inter-American Development Bank also have equity shares in the US$5.5 million capital of the new bank.

The decision to create a new bank has been questioned. Banco Sol explains that its main advantage over PRODEM is that it has the legal and financial structure to increase its client base many times over. Banco Sol will be able to mobilise savings (PRODEM suffered from serious limitations in this respect) and make new loans against these deposits. It will also be able to use its profits (assuming it is be profitable) to expand its operations.

PRODEM will continue but mainly in orienting new clients and training the credit officers of the new bank. Banco Sol's projections are ambitious: it anticipates a client base of 250,000. It has sent missions to Grameen bank in Bangladesh and BRI-Unit Dea in Indonesia and has learnt a great deal from these institutions on how to structure and operate the new bank.

4.2 Colombia

Colombia has a very large microenterprise informal sector. There are no exact statistics on its size. However, a study by SENA, the national training organisation, stated that in 1980 of the of one million employed in the manufacturing industry, only 400,000 (or less than 40%) were working in formally registered enterprises. The majority, over 600,000, were in the informal sector.

The first organised programmes to assist microenterprises in Colombia were developed in Cali in 1976 by the Fundacion Carvajal, a foundation set up by a family of local industrialists well-known for many years for their social programmes to help the poor of Cali in health care, education and housing. Some early training programmes were organised in 1978-79, but it was only in late 1979 when the Inter-American Development Bank (IDB) gave a US$500,000 loan to the Foundation to provide credits to microenterprises that the programme began to take off. The Carvajal Foundation, under its dynamic director, Jaime Carvajal, laid great stress on business training and education. In fact, the Carvajal methodology saw credit as the final stage in a programme of assistance, after the microentrepreneur has passed through a training programme over a number of years. It became part of the Carvajal Foundation's procedure that only those who had passed basic training courses - in accounting, investment analysis and marketing - were eligible to obtain credit. They were eligible also to receive advice in their own workshops in which these subjects were clarified. The first loans were of about US$1,000 and could be followed by two further loans of up to a maximum in the final loan of the equivalent of US$3,000, after which, on repayment, the microentrepreneur was expected to seek further financial assistance from a formal financial institution. The moneys loaned by the Carvajal Foundation, mainly from the IDB and some from the local business community, were offered at an interest rate that was slightly lower than the prevailing commercial rate. The Foundation received the IDB funds virtually as a grant and so was able to use part of the interest for training and technical assistance.

Support from private institutions to the informal sector in Colombia has generally followed one of two different approaches. There are private foundations that use the Carvajal methodology in which credit is given to individuals after they have passed a process of training and education in business management. There are other private foundations that use the solidarity group method whereby they lend only to solidarity groups of three to six microentrepreneurs who guarantee repayment of the loans that each receive. The solidarity groups not only guarantee the loan, but also help each other solve their business problems.

Generally it is the Carvajal approach that has been followed by the 30 or so foundations established in cities all over Colombia. The earlier foundations looked for microenterprises with potential for growth and saw training as the most important input. The foundations that followed the Carvajal approach believed that credit given after training was used better.

Table 5: Results of programmes that use Carvajal methodology

Source: Miguel Fadul: Country studies - Colombia, in "The critical connection", by Stearns & Otero, ACCION International, 1990.

Table 6: Results of programme with solidarity group methodology

Source: Miguel Fadul: Country studies - Columbia, in "The critical connection", by Stearns and Otero, ACCION International, 1990).

The programmes of these private sector based foundations throughout Colombia showed considerable growth from 1980-85. Then several of them went through managerial crises and tried to find a more effective methodology. As the programmes expanded, some constraints in the availability of resources also developed and this curbed the ability of the foundations to increase their loans to the microenterprises. Growth of the foundation's individual lending throughout Colombia is shown in Table 5.

Since 1983 programmes based on the solidarity group approach have also been operating throughout Colombia, and the growth of this type of programme is shown in Table 6. In 1983-85 there was fast growth in the number of such groups and again in 1987-88, but the number of groups stabilised at around 1,000 in 1988. The total number of solidarity groups receiving loans during the period 1983-89 was 4,281 covering 15,775 enterprises and 29,548 loans. A total of US$7.758 million was loaned at an average loan amount of $256 for each group.

The figures show that the cost of lending through solidarity groups was considerably lower than using the Carvajal approach of lending to individuals, primarily because of lower transaction costs but also because the training involved was less expensive. Also the average loan size for the individual borrowers was considerably higher (US$1,150 in 1988) than the loans for solidarity groups ($287 in 1988). Another significant figure brought out by the review was that average number of employees per microenterprise helped in the Carvajal type programmes was 3-4.5 per enterprise which was appreciably higher than the national average of 1.7 for the microenterprises in the country as a whole indicating that the programmes were reaching the larger than average microenterprises.

Even when engaged in lending to individuals some of the other foundations in Colombia did not rigidly follow the Carvajal methodology, insisting on training before granting credits to individual microentrepreneurs. Some foundations engaged in both Carvajal type lending to individuals and lending also to solidarity groups as well. Some foundations approved loans without insisting that the borrowers take prior training in business management. Virtually all the foundations screened the applicants and accepted the risk of default, but relied on the bank or financial institution to disburse the credit and collect the debt. For this these institutions were usually given a management fee of around 3%.

There have been attempts to assess the impact of these programmes. Most of the evidence seems to show that there was an average employment increase as a result of the assistance given of approximately one employee per microenterprise. Those that had entered the programme with three employees were usually able to raise their employment to four. Family income also tended to go up and business income increased by 45% in commerce and 57% in production.

A government review of the microenterprise support programmes of the private sector recognises that the government became interested in the need to help the informal sector largely as a result of earlier initiatives by the private foundations. What began as a local effort, to help with problems of employment, low income and poverty, grew into a nationwide development alternative to help the underprivileged sectors and gave the opportunity for assistance and help to thousands of microentrepreneurs.

Critics of the foundations have generally claimed that their cost of operation is too high in relation to the training and credits offered. Furthermore, since the Carvajal approach aims at training all microentrepreneurs, but financing only those with economic potential, it is claimed that this is not really helpful in alleviating poverty, since the owners of the enterprises that have the greatest potential for growth are those whose incomes are already above the poverty line. The foundations would argue that in the long run these are the businesses that make the greatest contribution to economic development and employment creation. The poorer elements are generally those that abandon self-employment when they obtain employment in another enterprise.

Despite these weaknesses, some overlapping and even conflicts at times between the different foundations and between the government and the foundations, the Colombian example has revealed interesting positive features and has been followed with interest in other parts of the world, particularly in other countries of Latin America. Not least of interest has been the support given to the programmes by the local business community and the efforts made to raise the levels of management and entrepreneurship before providing basic finance to microenterprises. The Carvajal approach is based on the idea that after receiving the basic training courses and up to three guaranteed loans the microenterprise would be ready to graduate to the formal sector and negotiate its own financial needs from a formal banking institution.

In reality, cases of microenterprises graduating to the formal sector in Colombia are limited, a situation experienced by this type of programme in most countries. Inevitably both the banks and the foundations have a vested interest in maintaining the situation as it is. The banks are more comfortable managing programmes where the risk of loan defaults is carried by the private foundations; the latter, for their part, are reluctant to lose good clients. There is evidence that by bringing in banks to administer the credit programmes, handle the disbursements and collect the debts without taking the loan risk, in fact puts the banks into contact with the entrepreneurs and gradually, in the course of time, this can result in some direct lending from the formal institutions when it is seen that most of these clients are creditworthy. However, nowhere does this happen quickly or on a large scale.

The involvement of at least one important financial institution in Colombia -the National Savings Bank (Caja Social y de Ahorros (CSA)) - in managing some foundations' lending programmes did result in the bank lending to microenterprises from its own resources and at its own risk. The CSA has managed foundations' credit programmes in no less than nine cities in Colombia, and more recently has taken an active part, as a direct participant, in the IDB loan to the Government for the microenterprise sector.

4.3 Mexico

Mexico has long carried out widespread programmes to help develop formal small and medium industries, but it was not until 1988 that the government began to address its programmes at the microenterprise sector as well. In that year a Federal Law was enacted to promote investments in new microindustries and to simplify the regulations governing their operation so as to make it possible for them to operate formally and legally, by granting them tax, financial and marketing considerations. In Mexico a microenterprise is defined rather liberally as an establishment with less than 15 employees and with annual sales of less than US$114,000 (1989 rates).

Mexico differs somewhat from other Latin American countries in that the Government and provincial states have played a major role in implementing support programmes for microindustry along with a number of private development organisations. Each state within the Federation has a public development agency with public funds channelled through the Nacional Financiera, the Federal Government's major development bank. These funds are used to provide credits to microindustries.

In 1987 the Nacional Financera started a programme (PROMICRO) for the support of informal microindustry by providing credit to microenterprises through both public and private sector development agencies that met certain requirements. In addition to providing credit, PROMICRO supports advisory services and training for microenterprises.

Since 1985, the Nacional Financiera has also channelled credit to microenterprises at preferential rates through its Artisan Workshop Programme operated by the development fund supporting small and medium enterprises (FOGAIN) that has been operating in Mexico since 1954.

Mexico was one of the first Latin American states to provide major incentives to microenterprises to register and formalise their activities. These incentives include a 20% income tax exemption for three years from the date of registration, a certificate for a 40% tax deduction on all investments in machinery and equipment, and help with standardising and simplifying bookkeeping.

Despite the easing of restrictions and licensing procedures, relatively few microindustries have taken advantage of these special incentives to register officially. In a country where there are hundreds of thousands of microenterprises only a negligible percentage have registered officially. Clearly the simplification of administrative procedures has not worked and apparently the incentives offered are not enough to motivate enterprises to move from the informal to the formal sector.

Of Mexico's 31 states, 29 participate in a programme to guarantee bank financing for microenterprises. Since late 1988 the state development and guarantee funds of the 29 participating states have guaranteee 3,538 loans for the equivalent of US$7.8 million. In 1989, the most active of these state funds, FOMICRO in the state of Nueva Leon, guaranteed 845 loans for a total of US$1.45 million, and, it is reported that between 1985-89 it granted a total of US$3.456 million in credits. This generated almost 2,000 jobs at an average an investment cost of less than US$1,800 loaned for each job created. However, over 80% of the guarantees given in the state of Nueva Leon were in the metropolitan area of Monterrey, which means that one of the objectives of the programme, the decentralisation of industry, was not being achieved. However, some smaller municipalities in less populated regions did receive a significant number of loans through the FOMICRO guarantee programme.

The purpose of PROMICRO is to help integrate informal sector enterprises into the formal sector. PROMICRO provides training courses and technical assistance as well as credit. It also supports microindustries through development agencies that are authorised to receive a line of credit specifically to make loans to microindustries.

From the start of the programme in 1987 until September 1989, 27 agencies had received authorization for a line of credit from PROMICRO and 21 of them were already disbursing funds from the credit line by early 1990. Nacional Financiera has other resources to cover the expenses that the agencies incur in training and technical assistance for programme beneficiaries. There were complaints, however, that there was considerable bureaucracy to go through for an agency to obtain access to a credit line.

FOGAIN, which disburses resources made available to Mexico by the World Bank and the IDB, channelled more than US$4 million in 1988 through its Artisan Workshop Programme. These were loans of up to US$30,000 for 3-6 years. From the size of these loans, it is evident that this programme is serving enterprises that are larger than the usual microenterprise.

A private institution that has considerably enlarged its operations in the last few years to help microenterprises is the Asesoría Dinamica a Microempresas (Dynamic Advice for Microenterprises (ADMIC)) in Monterrey. This is the largest private sponsored programme in Mexico for this sector, although there are lesser programmes in Chihuahua and also a privately sponsored agency for Mexican universities (DESEM) to help students set up microcompanies.

ADMIC is a private, non-profit foundation set up in 1980 by a group of large-scale businessmen from Monterrey. Until 1988 ADMIC operated with its own funds and some limited loans obtained from international institutions such as USAID and IDB. In 1989 it began also to disburse resources from FOGAIN and this substantially increased the volume of credit funds in its programmes. From January to July 1989 ADMIC disbursed credits to microenterprises for US$700,000 from FOGAIN's Artisan Workshop Programme. This represented 44% of that programme's total nationwide disbursements for that period.

ADMIC operates in six states in northern Mexico and in the central districts around Queretero and Puebla and in Mexico City as well. It has focused its efforts on family businesses that have never had access to credit. Its loans are small - the first loan to an entrepreneur is generally for less than the equivalent of US$600, but its maximum loans can be around US$7,000.

In 1989 ADMIC gave out more than 4,700 credits for a total of US$2.5 million. From its founding in 1980 to the end of 1991 ADMIC claims to have given 8,500 loans and the subsequent business growth has crated 10,400 jobs. Of these 4,600 new jobs were created in 1989. Operations in that year also helped sustain 10,000 existing jobs.

ADMIC is not a "minimalist" operation and is closer to the Carvajal approach in that it believes that training and technical assistance are as important for microenterprises as credit. A good deal of its training efforts are focused on the management of the small businesses to which it provides finance. One of its special achievement has been that it has obtained support from some regional banks for the microenterprise sector. In 1989, after considerable effort, ADMIC was able to use a line of credit from the BANORTE-FOGAIN for $1.2 million for microenterprise loans with a $100,000, or less than 10%, letter of credit used as guarantee.

As in other Latin American countries, private sector programmes, while recognised as having made significant contributions, are regarded by many as inadequate for the needs of the microenterprise sector. These programmes, it is claimed, have not expanded enough for a significant percentage of existing entrepreneurs to take advantage of their services. It is also argued that there is insufficient support given to the development of self-help organisations by the microenterprises themselves. A challenge that remains is for ADMIC to use its links with the business community to generate more commercial and business assistance directly from larger businesses to microenterprises. These private development organisations have so far not succeeded in persuading large businesses to buy from the microenterprise sector and when they do, it is alleged, they pay very low prices, thus making it difficult for the sector to find profitable outlets for its products. The private sector has so far done little to integrate microenterprises into the formal structures, nor is there evidence of trying to develop supplier credit for microentrepreneurs.

Experience in Mexico with assistance programmes for the microenterprise sector specifically raises the question of the relationship between public and private sectors in fostering the development of this sector. In Mexico both are trying to help but the public sector has not designed an overall strategy and has earmarked very limited resources in view of the considerable needs. There are indications too that the Government and public organisations have deliberately prevented NGOs from obtaining access to any significant extent to the resources of international institutions and foundations for their microenterprise support programme.

It is clear that there is insufficient cooperation between the public and private sectors in support of microenterprises and this has reflected negatively on their development. There are indications that Nacional Financiera has overcome its reluctance to work with the private sector and is now willing to distribute an increasing amount of its own and donor resources through private institutions such as ADMIC. It is argued strongly by some in the private sector that organisations such as ADMIC should be given more encouragement and that efforts should be made to establish such private development agencies in all leading urban centres in Mexico. The situation could be improved by closer contacts through regular meetings between organisations of the two sectors.

In December 1991 the President of Mexico expressed strong support for ADMIC's activities and promised a government grant of US$1.7 million to help capitalise a new loan fund of 10 billion pesos (US$3.3 million).

4.4 Peru

Peru's must be the most widely described informal sector. This is due principally to the activities and writings of Hernando de Soto, President of the Institute for Liberty and Democracy, and particularly his widely acclaimed book "The other path" (El Otro Sendero). Peru's informal sector is estimated to comprise 50% to 60% of the man hours worked in the country, and to include close to 50% of the working population. One estimate calculates that the output of the informal sector in Peru is 38% of gross domestic product.

Unlike previous governments, the Government that came to power in 1985 recognised the importance of the informal sector and the need to take it into account in overall economic strategy. The Government recognised the dual origin of the urban informal sector in Peru: firstly, it was structural, in that the formal urban economy was unable to create sufficient jobs to absorb the growing labour force forcing them into informal activities; and secondly, the many regulations and controls encouraged informal operations in order to avoid these burdens.

After 1985 the National Development Bank set up a multi-sectoral decentralised bank for each financial district; a series of loan funds were established to help the informal sector. In 1986 the Industrial Bank of Peru - the most important industrial development financing bank in the country - created a special fund (FOFISI) to finance the informal sector with a grant from the Ministry of Finance equivalent of to nearly US$3 million. This fund gave loans to individuals and groups engaged in production, services and commerce and, for the first time, did not require the recipients to be officially registered nor to have legal status. Eligibility was limited to beneficiaries with annual sales of less than the equivalent of US$35,000 or 20% of the amount set as the maximum sales volume of businesses defined as small. Despite efforts only a very small number of loans were actually disbursed through this fund in the first two years of operations.

The National Fund for credit to the urban informal sector (FONCRESI) was established in 1986 with the equivalent of US$10 million from the National Treasury. The Agrarian Bank used these resources after 1986 to make loans to more than 10,000 clients, particularly in the Andean region, where most of the indigenous population live. In this way the Agrarian Bank became a much more dynamic provider of finance for microenterprises than the Industrial Bank. This national fund was available also for other institutions and, by the end of 1986, the Central Credit Cooperative and the Banco Continental also helped in disbursing these funds.

As a result of the change in government policy, the urban informal sector became a priority target for assistance programmes. In addition to the activities of the Agrarian Bank and the Credit Cooperative Central, two major private, non-profit institutions were started. The guarantee fund for the Informal Sector (FOGASI) was set up in 1985, followed by the Development Institute for the Informal Sector (IDESI), which was established in 1986.

In 1989, after three years of operations, IDESI had given out 267,000 loans for a total value of $25.7 million benefiting 65,000 participants, approximately 64,000 of whom participated in training programmes. Arrears were reported to be 3% by 1989. FOGASI, which operated differently as a guarantee fund for the informal sector, actually provided over 10,000 guaranteed loans to the informal sector for a total of US$930,000 to 2,900 beneficiaries.

Among other private institutions that developed in Peru to help the informal sector, most significant was the Acción Communtaria (AC), which was established in 1969, but which developed its credit programme in 1982. In the six years 1982-85, Acción Communtaria gave out over 22,000 loans for a total value of US$18.78m to 6,785 participants, with an arrears rate on repayments of 4.5%. As can be seen, AC's average loan size was more in the region of US$2,500, considerably higher than those of IDESI.

As the data shows, the most comprehensive programme for providing credit to informal sector microenterprises in Peru was that of IDESI. This private sector NGO came into being after unsuccessful attempts by governments and financial services (i.e., the Industrial Bank) to offer financial support to the informal sector. Despite its focus on the informal sector, IDESI established links with formal institutions and its board of directors included representatives from universities, banks, research centres and governments. Apart from providing credit, IDESI worked to improve the status of the informal sector by securing social security and health provisions for workers in these enterprises and their families.

Over the first two years of operations IDESI reached and supported over 40,000 beneficiaries, 60% of whom were women. As with private sector NGO's in other Latin American countries, IDESI's experience showed that the creation of solidarity groups was the most effective way to help overcome difficulties, not merely by obtaining access to credit, but also by providing mutual support and "self help" within groups.

IDESI's mode of operation allows it to deposit the funds it receives from international organisations as a guarantee deposit fund to enable poor people to become regular bank customers. IDESI has managed to obtain help from the ILO, UNDP, the Netherlands, the CIDA programme of Canada and from some international NGOs. Because of the manner of its operations, IDESI claims to have reduced its costs of intermediation down to less than 2%.

As in the case of Colombia, there are those who claim that the private programmes have not met the expectations as regards increased income for the informal sector beneficiaries nor in the generation of employment.

While the financing programmes for the informal sector have not yet been fully assessed as to their economic impact, there is some empirical data that would indicate that the credits permit beneficiary informal sector enterprises to increase their working capital and, as such, their total assets and income.

As in other countries, it is generally regarded that the greatest impact in Peru has been on the financial system, by demonstrating that these microenterprises are able to repay loans, are creditworthy and, that in these private programmes, they have very low rates of arrears.

There is no hard evidence as yet in Peru as in other countries that formal sector credit is significantly more available to microenterprises as a result of the activities of these NGO credit programmes. The sector is still regarded as one of high risk for lenders involving high transaction costs in the processing of small loans.

IDESI, Acción Communtaria, and others have tried to combine credits with training and counselling and other types of non-financial assistance, but some claim that the training offered does not correspond to the real needs of the beneficiaries. Also, there has been additional criticism that the programme was not monitored adequately so that it is not clear to what extent the training and the credits reinforce each other in helping the clients. There have been efforts to organise the microenterprises into "self-help" type of associations, but so far these have had limited results. Associations that have been created have been relatively weak. It is recognised that in the long term it is important to raise the status of these groups so that more can be achieved through "self help" sectoral organisations based on membership.

CHAPTER 5

FINANCIAL INTERMEDIATION SYSTEMS IN AFRICA

5.1 Ghana

Ghana, a country with 14.6 million inhabitants (1990), had a per capita income of US$390 in 1986 which represented a significant decline since independence in 1957.

The banking system in Ghana is dominated by the central bank (the Bank of Ghana (BOG)) and three commercial banks: the state-owned Ghana Commercial Bank (GCB), Barclays Bank and the Standard Chartered Bank (both 40% state owned). There are eight other national non-commercial banking institutions - referred to as the secondary banks - mainly state-owned development banks including the Agricultural Development Bank (ADB) and the Social Security Bank, the second most important bank, after the GCB, in terms of volume of business.

Although each of the secondary banks, ADB, NIB (Natural Investment Bank), BHC (Bank of Housing and Construction) operated primarily to channel government and foreign donor contributions to their respective sectors, they have in the course of time offered normal commercial banking services to try to attain better financial returns on their operations.

The ADB was started in 1967 to extend the range of rural credits and financial services available to the agricultural populations who were considered to be inadequately covered by the existing network of commercial and development banks. Since its inception, the ADB has opened 50 branches operating all over the country (1989).

ADB's results fell short of expectations. Its services were used by only a limited number in Ghana's rural areas. Estimates indicated that the ADB reached only 16% of small farmers in the country in any year. At its peak it enjoyed a 85% repayment rate but, as it was forced to lend at subsidised regulated rates (negative real rates in relation to the high inflation) and its administrative costs were high because of the need to cover a widely dispersed clientele, ADB was unable to recover its expenses and became unprofitable. ADB has also been criticised for having ineffective management especially at the branch level.

Although ADB operated a lending programme for small-scale farmers, fishermen and cottage industries, it was considered by BOG to have had limited success in meeting the demand for small-scale credit among the rural population leaving the latter dependent on informal sources or moneylenders. The BOG and its Rural Credit Unit, impressed by the system of rural banks in the Philippines, decided in 1976 to set up a similar system in Ghana. The first rural bank opened in July 1976 and by 1991 there were 122 rural banks in operation.

The rural banking system in Ghana represented an innovative initiative by BOG in that it broke with traditional banking in the country. The new rural banks were independent, locally owned and managed, and catered to a local clientele unlike other banks which were all controlled from central head offices. The development of the rural banks (RB) was fostered by the BOG which continued to be heavily involved in their internal affairs. The RBs of Ghana, unlike their counterparts in the Philippines, are owned by large numbers of persons in the local community. The RBs operate as "unit banks" which means they are not allowed to operate more than one branch or office. Yet they differ from "village banks" in other African countries (e.g. Mali) in that each bank has a direct relationship with the BOG which formally regulates and supervises the RBs' operations. All RBs, however, belong to the Association of Rural Banks (ARB), an umbrella organisation founded in 1981 to foster the exchange of ideas among the banks and to organise common training activities. Some overlapping has developed between the Rural Bank Department (RBD) of the BOG and ARB especially in the matter of arbitration of disputes between Boards and Management.

Although the ARB has taken on some internal legal issues, the RBD of the BOG maintains continued interest in the supervision and regulation of RBs and plays a major role in the authorisation of new ones. This process is quite lengthy requiring formal submission from an individual or group in the village where the RB will be located, including information on the proposed area where the new bank will operate. The BOG appraises the area and makes a recommendation to the RBD. If the outcome is positive, the local group collects the share capital from the community and submits a list of candidates for the positions of bank manager, accountant and for members of the board of directors, all to be approved by the BOG. The staff concerned undergoes a period of training organised by the BOG.

In practice there is contradiction between the BOG's role as the promoter of new RBs and its efforts to supervise and regulate them. There is also contradiction between BOG's general desire to have the ARB play a greater role on behalf of the RBs and its apparent reluctance as evidenced until mid-1991 to relinquish any of its powers. There is a parallel situation with regard to ARB which has not shown great readiness to accept new responsibilities as regards the RBs.

RBs have a low level of capital assets. As of June 1988 the total assets of the existing 120 RBs represented only 3% of the total assets of the whole banking system. As a comparison it should be noted that the GCB (with 150 branches) has about 30% of total assets and Barclays (30 branches) and Standard (24 branches) have 8% and 7% of assets respectively.

RBs were limited by the BOG in the first years to a maximum capital base of 1 million Cedis (1979). The minimum was set at 100,000 Cedis (pre-1979)⁽⁷⁾. BOG then provided up to 50% of the paid up capital requirements. The RBs were given a five year tax holiday and an interest free establishment loan of 40,000 Cedis (pre-1979). Capital requirements were adjusted in 1979 and 1986 to keep up with inflation. In 1986 new RBs required a minimum capital of 1.5 million Cedis with the BOG reducing its share to one-third. Participation by any individual in bank ownership was limited to 1.7% (2.5% for a family) and 12.5% for an institution.

The general functions of an RB as set out by the BOG were:

- to collect savings, deposits and securities for safe custody;

- to act as an agent for other financial institutions;

- to serve as executors and trustees for clients;

- to provide loans and credit for local enterprises; and

- to engage in investment activities that promote social and economic development in its area of operation.

RBs are precluded from engaging in foreign operations of any kind without the approval of the BOG.

The Rural Banking Department of Ghana also formulated a number of specific RB operational guidelines including the following policy areas:

- Delay in start-up of lending operations: No loans or advances were made in the first six months of business to allow management and clients to familiarise themselves with each other. Borrowing was limited to customers of the bank, which encouraged people to deposit their savings. The delay also helped the banks to establish contact with clients and become aware of their needs, and allowed customers to develop a credit rating.

- Lending policies: sectoral guidelines were prescribed as to the proportions of lending. Initially, these were: 50% for agriculture; 30% for rural industries; and 20% for all other sectors.

To ensure that lending was biased toward small producers, guidelines were also prescribed regarding the size of loans. Suggested distribution was: 50% for small loans; 30% for medium loans; and 20% for large loans. The nominal amounts of each of these categories were adjusted regularly to keep up with inflation.

In anticipation of a predominance of short-term seasonal loans, the maturity structure of loans was also set out: short-term loans (due within 1 year) - 70%, medium-term loans (due within 1 to 5 years) - 20%, and long-term loans (due in not less than 5 years) - 10%.

Additional recommendations regarding credit disbursements were made regarding the lending limits for each level of management. Each Manager, the Local Board and the full Board were given approval limits, with the Board ratifying all disbursements made by the Manager and the Local Board, and the RBD authorising loans above the full Board's limits. All loans to Board members were to be referred to the BOG to avoid abuse.

- Issues of security and collateral: alternative strategies of lending were proposed for groups with varying amounts of collateral. Group lending and guarantee schemes were proposed to encourage loan repayment.

- Establishment of interest rates: until 1987, interest rates had been set by the government and applied to banks of all types. Under the Structural Adjustment Programme, they were later liberalised at this time although deposit rates continued to be fixed as of January 1991.

- Targeted population: income and occupational groups from which clients should be drawn were defined and included small and medium-scale producers, industrialists, traders, fishermen, retailers and artisans among others.

It was further suggested that clients be drawn from within a defined geographical area (a 48 kilometre radius).

These guidelines have been modified by the Boards of individual RBs and have been adapted or even totally relaxed according to the needs of each institution.

As of June 1988 the average size of loans was 20,000 Cedis for agriculture, 25,000 for processing, 50,000 for trade, 140,000 for transport.

There were 122 RBs in operation in June 1991 with an estimated 60 or so RBs approved but not yet operational. A further nine had conditional approval and there were 229 pending applications.

As with the whole banking system, RBs suffered from a serious problem of non-performing loans. Studies undertaken by the World Bank in 1989 on 20 RBs showed that as many as 70% of them had 50% of their loans overdue for three years or more. The World Bank estimated that 25-30% of the loans in the sampled banks could be treated as bad debts.

As part of the World Bank's Rural Finance Project, 1989 the BOG and Bank began assessing the status of the RBs in an effort to restructure them as part of a programme of comprehensive financial reforms (GEMINI Tech No. 33 Report). This assessment is still going on. The restructuring being undertaken has been trying to grade the RBs into three groups based on their level of profits: those making a profit of more than 5 million Cedi; those with less than 5 million Cedi; and those operating at a loss.

To achieve Category A status a RB must also meet a capital adequacy requirement ratio of 6% to be calculated on the basis of secured assets to those that are unsecured or risk oriented. Those that failed to meet this capital adequacy ratio would have to collect outstanding loans, increase share equity and/or reduce operating overhead costs.

The World Bank/BOG assessment looked in depth at each bank and at the system as a whole and set requirements for basic performance levels and operating standards that each bank should meet to be regarded as solvent and functional.

By mid-1991, 50 RBs had been reviewed and a total of 17 banks had been classified as Category A profitable institutions. This seemed to indicate that about 40 of the 122 banks (around 30% of the RBs) would be able to meet performance standards and capital adequacy requirements. Another 40 banks would be in a questionable situation while the remainder could be considered as having failed.

In general it was found that the RBs maintained excessive liquidity mainly through the purchase of Government treasury bills earning 35%. Whilst helping to shield the banks financially this rather undermined the function of those RBs to provide access to bank credits for the rural population.

Early on the RBs enjoyed some success especially mobilising savings but the surge in the number of new RBs (from 54 in 1982 to 107 in 1984) quickly created problems. The BOG could not give close attention to all these banks as it had done with the first group. After 1985 the BOG considerably reduced the number of new RB approvals, as evidenced by the number of pending applications.

RB's declining performance is mainly due to poor management and inexperienced staff. The training helped but was inadequate to improve decision making by senior staff. As the number of RBs grew the supervision and advice of the BOG became less effective and the RBD's resources were taxed beyond its capacity. RBs also suffered from a degree of isolation, poor communications and serious transportation and logistics problems. The BOG representation on the board of RBs was important as usually the persons involved were the most knowledgeable in banking matters.

Many RBs suffered from unpaid loans after the 1983 draught. A number also have been undermined by a series of bad loans and embezzlements by bank staff and board members. To meet capital adequacy requirements the banks have been emphasising over the last few years calling in loans and raising deposit levels.

The general policy environment also helped to undermine the ability of the RBs to achieve financial sustainability. As in the case of the ADB, the government fixed interests rates were set too low in the belief that the rural clients needed cheap credit, but administrative costs were high thus generating losses. The guidelines and restrictions imposed by the BOG meant the RBs had to operate under highly regulated and artificial conditions.

As the RBs tried to recuperate losses by charging supplementary "fees" for extra services and began to demand more collateral and guarantees to reduce defaults, clients were discouraged from using RB facilities. To achieve some degree of profitability - or at least to reduce losses - RBs began to lend more to larger, wealthier borrowers thus destroying the whole purpose for which they were created.

RBs achieved greater success in mobilising savings. This is remarkable in view of the fact that rates on savings accounts were only 3% to 9% when inflation was over 36%. In 1988 the savings mobilised by the RBs accounted for nearly 70% of the total funds available. In 1988 RBs actually controlled 7% of total savings deposited in banks throughout the country. However a closer look reveals that although the nominal amounts deposited increased substantially in real terms, allowing for inflation, the amounts saved have been declining since 1983.

RBs have also been criticised for maintaining higher than necessary primary and secondary reserves, placing a higher proportion of their capital in government securities, thus reducing the amount of loanable funds available. A GTZ study in 1989 estimated that of the 7% of average total savings deposited with RBs, less than 2% of savings was being channelled from savers to borrowers. Thus the contraction of funds for lending, due to loan defaults, high reserves, etc., forced the continued dependence of the rural population on informal financial markets.

The Rural Banks of Ghana have been important in providing a network of financial institutions in the outlying rural regions of the country. However malfunctioning, they have introduced banking to the rural population which, in the case of the remoter areas of the north, would only have had access to informal intermediaries or money lenders. RBs have thus helped to highlight the financial needs of the rural economy. Unfortunately, due to high transaction costs and losses through poor loan recovery, the capital available for loans is continually decreasing and credit opportunities for small scale rural borrowers from RBs is becoming even more limited.

The situation as reported in June 1992 indicates that the number likely to succeed has dropped and may be no more than 20, leaving the future uncertain for the other Rural Banks.

Although it is generally regarded that Ghana is the only African country to have developed a rural financial system through the establishment of unit rural banks, some efforts to establish such banks have also been made in Nigeria and Sierra Leon. The significance of the Ghanaian effort can be understood better when one considers the limited nature of the rural financial system in most African countries. Richard Meyer and Douglas Graham have produced an interesting table categorising the components of the rural financial system in sub-Saharan African countries, dividing them into formal, semi-formal and informal sectors. The division they produce is given in Table 7.

As can be seen from this interesting table the unit rural banks of Ghana are significant because they belong to the formal financial sector under the direct supervision and regulation of the central bank. This is not so in the case of the village banks in Mali and Burkina Faso which are community based and have been set up in most cases with donor help from Switzerland, France and Germany. The main purpose of these is to mobilise savings and provide loans in the village. They are not subjected to the regulation of the Central Bank as in Ghana. In most cases village savings mobilised in the countries concerned have been augmented by the donors from the countries referred to above.

Table 7: Important components of the rural financial system

Formal	Semi-formal	Informal
Regional and national central banks, treasury, and other regulatory bodies	Agricultural cooperatives	Communal clubs
Financial intermediaries:	Credit unions	Mutual aid associations
Commercial banks	Banques populaires	ROSCAs
Development banks	Integrated development projects	Moneylenders
Savings banks	Village banks	Moneykeepers/ Mobile bankers
Postal savings	Self-help groups	Input suppliers
Cooperative banks	Savings Clubs	Storeowners/ Merchants
Unit rural banks		Trader-lenders
Finance corporation		Farmer-lenders
		Friends
		Neighbours
		Relatives/family

Source: Richard Meyer and Douglas Graham: Developing financial markets in Sub-Saharan Africa, Ohio State University, Colombus, 1991.

5.2 Ivory Coast

The Ivory Coast is one of the more prosperous countries of Africa with 9.2 million inhabitants, a large proportion of whom are immigrants from neighbouring countries. The government's policy is to foster agricultural development for both internal consumption and export. It is based on an internal resource mobilisation working together with external capital.

The group which needs finance comprises small rural enterprises, including on-farm and off-farm activities in villages and rural towns.

Informal financial institutions (IFI)

There are some 50-60 indigenous ethnic groups in Ivory Coast, which fall into four major ethno-linguistic divisions: the Kru in the south-west (17.5% of the population), the Gur in the North (13.0%), the Akan in the south-east and centre (42.7%), and the Mande in the north-west (26.8%). In the rural areas, the predominant IFIs are the hired-work association with primary financial functions, the rotating savings association and the savings and credit association. Of these, the hired-work association is the most important, unlike in many other West African countries. This is probably due to the shortage of manual labour in agriculture. In urban areas, the predominant IFIs are the rotating savings association, the savings and credit association and the system of daily deposits to a money collector.

On the whole, IFIs are important among the tribes in the south of the country but not in other areas. The southern tribes have also organised strong financial and non-financial self-help organisations, well integrated into household activities.

There have been some changes over the past 30 years: from work associations to financial self-help organisations; from savings associations to savings and credit associations with their own revolving fund; and from an insurance or mutual assistance fund to a loan fund. These changes emanate from increasing monetarisation of the economy. IFIs are likely to continue to spread in the Ivory Coast.

Also, in geographical terms, IFIs have spread from rural to urban areas. In sectoral terms, IFIs first expanded among small enterprises in two sectors - agriculture and trade - but later spread to small crafts and transport enterprises. They have also become widespread among white collar employees.

The core problem of the Ivorian IFI is the lack of access to refinancing institutions and to financial consultancy services. A breakdown of the problems is presented in Table 8.

Formal institutions

The Ivory Coast has an array of formal financial institutions. Extension and, to some extent, supply and marketing services are provided by state development organisations such as CIDT (Compagnie Ivoirienne pour le Developpement du Textile) in the north, SATMACI (Societé d'Assistance Technique pour la Modernisation de l'Agriculture en C.I.) in the centre and SODEPALM (Societé de Developement pour le Palmier) in the south.

The above linkage model can be applied, with due modification, by any of these agencies. However, it has been specifically designed for the agricultural development bank BNDA, a decentralised structure of six regional offices, 31 branch offices and 57 temporary offices. Within the next five years, the BNDA will step up its decentralisation to reach all 750 to 900 central villages in the country, providing both saving and loan facilities. The BNDA is to link up with existing IFI to offer more effective financial services to the rural population.

Linking informal financial institutions with formal institutions

To develop links between IFIs and formal institutions inthe Ivory Coast, the following procedures may be followed:

1. Local development initiatives should be linked to indigenous self-help organisations, such as IFIs.

2. Local self-help organisations should be promoted within the framework of the relevant social group, such as the village among the Senufo, the quarter and the village among the Dan and Malinke, or the individual household among the Kru and Guere.

Table 8: Problems of IFI in Ivory Coast

		Unproductive utilisation of loans
	Undeveloped loan business
Lacking examination of loan applications
Insufficient interest calculations
Rudimentary book-keeping
Lacking financial consultancy services		Unsafe deposits	Lacking agricultural consultancy services
CORE PROBLEM Lack of access to refinancing institutions			Lacking collaboration with development services
		Lacking linkages with formal financial institutions

3. Interventions should be adjusted to local decision-making which should lead to:

- broad "grassroots" participation in a decentralised manner;

- primary involvement of local authorities and control agents in centralised societies with limited social mobility;

- a mixed strategy based on development from above and below with elements of both decentralised decision-making and centralised. control.

The linkage procedure

Given the variety of IFIs in the Ivory Coast, the diversity of ethnic groups and the differences in degree of exposure to the monetary economy, the linking of informal and formal financial institutions must follow a flexible procedure. One such approach, worked out for the BNDA, is as follows:

1. collaboration between the BNDA and a research institute to study IFIs and local social systems, followed by organised communication between farmers, IFI committee members, local authorities, BNDA agents and representatives of development organisations, aimed at initiating links between IFIs and the BNDA.

2. Meetings of village authorities and social control agents with IFI leaders; in decentralised societies meetings of IFI members with their committee members. The results will be presented to meetings of IFI and village authorities in the presence of BNDA agents and representatives of development organisations. The meetings will discuss projects to benefit from the linkage arrangements and group projects; and a discussion of the elements such as self-help, savings mobilisation, BNDA-lending to IFI and IFI-lending to members, as well as extension services to be offered by BNDA, development organisations and cooperative agencies.

3. Linking IFIs to the BNDA and, if appropriate, to cooperatives as intermediaries and to other development organisations.

4. Establishing a credit and savings scheme for IFI, to be administered by BNDA.

5. Evaluating and monitoring the linkages.

5.3 Kenya

Kenya has a long history of promoting SMEs. Since 1967 the country has implemented a widespread programme of setting up industrial estates through Kenya Industrial Estates (KIE) in the major cities and population areas and rural industrial development centres in the rural towns. In general, beneficiaries have been the larger formal SME rather than the more numerous small informal sector enterprises. In 1987 KIE extended its activities and started the Informal Sector Loans Programme (ISP) to provide credit to microenterprises in the informal sector. The ISP has mainly targeted microenterprises in rural areas.

Informal Sector Loans Programme (ISP)

The ISP was supposed to assist microenterprises with a recognised potential for business growth so as to attempt to maximise economic impact through employment creation.

An evaluation of the ISP in November-December 1991⁽⁸⁾ found the programme had an average loan size of Ksh 23,550 (around US$800).⁽⁹⁾ Loans range from Ksh 5,000 to 50,000 ($1700), increasing to a maximum of Ksh 100,000 for second-time borrowers. Loans are repayable within 2-3 years at 18% interest - the prevailing commercial bank rate.

To be eligible borrowers must have been in business for at least six months. ISP is not intended to provide capital to start a business. Borrowers must be licensed (apparently this is less than formal registration) and must run their own business which should have a capital investment of no more than Ksh 150,000.

At the end of 1991, one year after disbursement, an evaluation made of 82 businesses assisted by ISP reported that loan repayment was around 81%, but that there was a marked repayment difference between women (92%) and men (70%) borrowers. The study found 48% employment growth among those beneficiaries evaluated.

The process of applying for, and obtaining an ISP loan can be time consuming for the borrower, but managers of the programme say that the steps have been designed to test the level of maturity of loan applicants as entrepreneurs. Before the loan is approved, borrowers must supply the sales records of their business and prepare a business plan. A training programme is arranged for loan applicants to help them prepare the business plan.

In some cases, after preparing the business plan and making the necessary calculations, entrepreneurs conclude that they do not really need the loan to expand their business, or in other cases, that they will be unable to serve the debt to be incurred through the loan. It is claimed that the exercise of preparing a business plan helps all loan applicants determine the real size of the loans they need.

The ISP is only one of many credit programmes for microenterprises in Keyna. As stated earlier, the informal sector was first mentioned in 1972 in a report by an ILO employment mission to Kenya. Despite this early recognition of its importance little was done at the time to develop access to credit for informal sector enterprises. In the late 1980s one of the first organised efforts to provide credit to informal microenterprises was made by the National Christian Council of Kenya (NCCK).

National Christian Council of Kenya (NCCK)

NCCK's loan fund was capitalised in 1981 at $275,000. By 1988, 148 loans involving US$19,605 were outstanding. First and second loans are available, each with a maximum payback period of three years. Loan applications take three months to be screened by extension workers and regional and national committees. Repayments improved after 1987; NCCK did not write off a single loan during 1988, an improvement over the 1981-86 average of 52%. Loans are guaranteed either by fixed collateral, or individual, group, or third party guarantors.

NCCK charged 15% interest on its loans at a time when other financial institutions in Kenya were charging 20%-30% and traditional moneylenders 50%. Inflation was at 10%.

An evaluation showed that after one year NCCK clients increased their incomes by an average of 375%, although only to a modest level of $75.49. Increased savings, assets, and consumption levels were also reported. NCCK helped create 100 new jobs in 20 new enterprises, and 258 in 86 existing enterprises. There were no business failures.

In order to expand their outreach capacity, NCCK relied on church networks to identify borrowers. NCCK used the infrastructure of the church to select and reach out to clients, integrating community development with the selection and promotion of entrepreneurs. To reach more clients, promote skill sharing, and specialise production functions, group work was emphasised. Selection was based equally on social and economic factors.

Banks were used as disbursing agencies to improve collection rates. However, a need to improve collection rates changed NCCK's emphasis toward more affluent enterprises.

NCCK favours direct and participatory outreach with decentralised decision making. However, NCCK reports that staff tended to relax on client selection rules with decentralisation.

NCCK found it difficult to convince banks to handle the credit programme. After two years of negotiations, banks still did not permit the use of their own funds for NCCK loans. Although the bank disburses each loan, NCCK project staff does the promotion, extension work, follow-up, and loan supervision. However, NCCK reported in 1988 that collaboration with the banks has helped; recipients no longer consider loans to be grants from the churches. Repayments are more timely and efficiently processed. Unfortunately, high transaction costs remain a problem even though staff remuneration is low, despite an increased workload.

NCCK's effectiveness appears to lie in its "grass roots" character; it is able to transfer some transaction costs to beneficiaries especially as its portfolio of clients shifts to comparatively larger microenterprises. NCCK does not believe that its programmes can attain self-sufficiency but plans a savings programme to create more resources and to enlarge the guarantee base.

Kenya Rural Enterprise Programme (KREP)

In 1984 USAID started funding the Kenya Rural Enterprise Programme (KREP) which aims to develop credit programmes for informal sector manufacturers through NGOs as intermediaries. KREP was a project of World Education Inc., an international NGO.

KREP's original mandate was to provide financial and technical assistance to NGOs in Kenya for the development of their small enterprise programmes.

During its first operational phase, KREP funded 10 NGOs to establish credit programmes. The grants were typically split 60:40 between the NGO operational costs, and the establishment of a revolving loan fund for entrepreneurs. In addition to assisting the NGOs in the design of their projects, KREP provided training throughout the life of the project to enhance NGO capabilities in accounting, marketing, pricing, credit administration, monitoring and overall financial management. In line with the KREP methodology, most of the NGOs pursued an integrated approach to their own credit programmes, offering training and technical assistance as part of the credit package.

The evaluation in 1988 of the first three-year phase indicated that NGO management capacity was still weak, lending costs high, the volume of loans small, and repayment rates averaging 78% - not as good as KREP felt they should be. These findings stimulated a rigorous reappraisal of KREP's programme, resulting in the following three major changes for the subsequent phase:

1. KREP decided to introduce a different lending methodology based on the Grameen Bank model (Bangladesh);

2. KREP undertook its own direct lending programme;

3. With NGOs, KREP shifted for straight grant-making to providing a financial package of which 70% is a loan repayable to KREP and 30% is a grant.

Initially, the decision to lend funds was met with resistance from NGOs and led to a reduction in the number of NGOs KREP was willing to assist. However, because the four new NGO projects now lend without insisting on training and technical assistance, the number of ultimate beneficiaries has actually increased. After only one year (in July 1990) there were dramatic results for all participating NGOs, including:

1. Increased efficiency in the costs of lending. Costs per dollar loaned dropped from an average of 2:1 (e.g. US$2 costs for each dollar loaned) to a 0.25:1 (25 ct. per dollar loan) costs per loan dropped from approximately $2000 to $86.

2. Higher loan volumes with higher repayment rates which produce more income for the programme. Average repayment rates have risen from 78% to 97%. The number of loans disbursed per NGO per year increased from 120 to over 1,100. The four participating NGOs borrow from KREP at 7% and on-lend at 28% (on a declining balance). Loan reflows are designed to cover direct operating costs and by the end of Year 3 the field programmes should be fully sustainable.

3. Significant savings mobilisation among borrowers. A total of approximately $280,400 has been saved by 6,787 participants for all programmes. Savings per person increased from zero to close to $50 during the first year.

Clearly, KREP's assistance strategy has helped its NGO partners to build both management and financial resources. NGOs formerly dependent on grants are now able to finance their credit programmes with loan funds.

Important challenges lie ahead (KREP organisers report). Graduation to commercial credit is one of them. Another lies in the KREP-NGO relationship itself. While KREP is mobilising substantial resources for these programmes, the bulk of them do not stay with the NGO since the loan principal and the 7% interest is repaid to KREP. NGO's may opt to access donor grant funds directly to capitalise their loan programme and keep a greater portion of loan reflows. A third question is, how can these programmes most effectively use member savings? KREP is now considering whether groups of borrowers could link up and form their own savings and credit societies.

5.4 Nigeria

Transforming IFI into Cooperative Societies

Nigeria is the most populous country in Africa with over 100 million inhabitants of whom about 50% live in towns. Exports now consist almost totally of petroleum even though Nigeria was a food exporting country in the 1960s. Agricultural production has declined continuously since independence while food imports have risen steadily.

Nigeria has a developed modern industry and a formal financial structure. Since independence it has tried to modernise rapidly but the informal sector has continued to thrive.

Informal financial institutions (IFI)

IFIs are well developed and widespread in Nigeria. Neither the oil boom nor the extensive, decentralised banking network have impeded the development of IFI.

While work associations have existed for a time, the origins of savings and credit associations are more recent. The total membership of IFIs in Nigeria is estimated at about 20 million, of an eligible adult population of approximately 50 million. In peasant societies, such as exist in rural Nigeria, between 30-90% of craftsmen and market-women and between 20% and 50% of industrial workers in urban centres use informal financial markets.

Linkages between IFIs and cooperatives:

Organising market-women in Lagos into cooperatives

There have been several attempts to forge a connection between banks and women traders in markets in Lagos. The Yoruba tribe has developed a system of daily deposit collection called "ajo". A money collector visits the clients, who may number up to several hundreds, during or after the market day and collects fixed and variable amounts of savings from every person who has signed up. After a stipulated period, generally one month, the total savings of each individual are paid back, minus a deduction of one, occasionally two, daily contributions as a fee. Confidence and trust in the "ajo-man" are essential in this system and there are many stories of dishonest persons absconding with the deposits. Despite its apparent shortcomings, the ajo system has certain advantages which induce its participants to adhere to it.

The system is convenient while access to banks is not available, especially not for the illiterate. Transaction costs are low. The "ajo-man" comes to the client instead of the client going to the financial institution. "Ajo" gives the opportunity to accumulate daily earnings, which, under family obligations and other pressures, would otherwise be spent quickly.

The institutional obligation of daily contributions is respected by family members. The system guarantees anonymity vis-a-vis friends, neighbours and relatives, especially one's husband or wife. In emergency cases, participants can get a quick loan.

In the 1960s, various banks worked together with market-women. All women who wanted to participate in a programme had to join a registered cooperative society. The bank received the daily deposits which served as a basis for loans to participants, assisted in the investment of surplus funds in productive ventures by supervising the work of the newly established cooperatives. Between 1968-70, ten cooperatives were established.

By 1980, all these cooperative societies had collapsed. This failure was due to:

- Associations were created of women of different ethnic background who did not know each other and who were not socially integrated.

- Loans were not linked to savings and were issued before any contribution had been made by participants.

- There was no policy to support self-help and self-reliance.

According to the women:

- In big markets, with up to 2,000 traders and more, the question of who gets the money borrowed from the bank proved too complex for a cooperative to solve. If the amount is divided equally, the "sum" for each individual is too small. If bigger loans are given, then the question is who gets the money first.

- Money paid into the bank account was soon regarded as the bank's money and no more as one's own deposit. One of the results of this attitude was that women no longer felt committed to repay and so they defaulted.

- Personal contacts between the members in the informal system were not transferred to the cooperatives. Therefore, social control was weak.

- Transaction costs were too high.

In short, this attempt to link IFIs to banks through cooperative societies did not prove successful.

Transforming indigenous savings and credit associations into cooperatives in Eastern Nigeria

As opposed to Western Nigeria, official policy in Eastern Nigeria was to establish cooperatives on the basis of existing indigenous savings and credit associations in virtually all villages and professions. In the beginning, the "isusu" associations, as they were called, were encouraged to register informally with the local administration and so obtain a degree of legal status. Only a small percentage actually registered, but they soon outnmbered registered cooperative societies. Subsequently, officers of the cooperative department persuaded many of these "isusu" to register as formal cooperatives so as to benefit from the status of a corporate body and to get full official recognition. It is estimated that today in Eastern Nigeria approximately 40% of all cooperatives are based on pre-existing "isusu" associations.

A survey of cooperatives in several states in 1984 showed that cooperatives based on pre-existing "isusu" performed better in all financial matters the others: regular savings were higher enabling these cooperatives to give out more and larger loans. These loans were repaid with a low default rate, and so the amounts in the revolving funds increased being rapidly augmented also by savings. These cooperatives also had much higher returns than the others.

Two further features are relevant to the cooperatives based on indigenous savings associations. First, after registering as a cooperative society most informal associations' regular contributions per member went down by almost half. Second, most members continued to belong to indigenous savings and credit associations that had not been transformed into cooperatives. In the isusu, savings were 80 times higher than in the cooperatives. Possible explanations include:

- there is less confidence in registered, state-controlled cooperatives;

- cooperatives stop members from freely using their savings deposits and the revenues from cooperative ventures; there is therefore little immediate personal benefit;

- members may try to get more financial assistance from the state by showing less savings.

Conclusions

Experience in eastern Nigeria shows that cooperatives fare better if set up on the basis of indigenous associations such as a savings and credit associations. They perform better when following tested practices. Experience in western Nigeria shows that similar associations, such as work associations, social clubs, women's market associations or women's guilds may perform equally well as a basis for cooperatives if they previously saved and gave credits in the manner of the savings and credit associations.

Contributing to the failure of cooperatives included:

- promises of external aid at the founding stage, and

- lack of social coherence among members.

Comparisons between indigenous associations and cooperative societies show that self-help efforts and business success are more pronounced in the former. At least in Nigeria, registration of cooperatives brought few advantages to the members. It may therefore be more appropriate not to force indigenous organisations to accept Western type cooperative legal forms. Indigenous associations represent genuine African cooperatives with the business determined by the members. If members decide voluntarily to register as a formal cooperative society, this should be supported, but it is not advisable to force them to do so.

Linking informal financial institutions and commercial banks: An example

The population of Anambra State in Nigeria as well as of some of the neighbouring states have a long history of self-help activities and organisations. Most important among them is the "isusu", the type of rotating savings association already referred to. These isusu savings associations have written by-laws, an elected committee, a rudimentary system of bookkeeping and they keep membership records. Over the past 20 years or so, most isusu have built up additional funds for loans, common projects or social services.

Apart from savings and cooperatives, there are few formal financial institutions in the rural areas of Anambra State. The Post Office Savings Bank, once important, declined in the 1980s and now offers few attractions to the rural population. farmers prefer saving in isusu as this provides a basis for credit. There are two commercial bank branches in this district, the First Bank of Nigeria (Ltd), keeping current accounts for wage-earners, companies, and cooperatives and savings accounts without restrictions, and the Cooperative and Commerce Bank, Nigeria (Ltd), which started operations in 1984. As is the case in most of Nigeria, small farmers had no access to bank loans, mainly because of their lack of collateral.

In a survey of cooperative societies in Northern Anambra State in 1984 discussions with members both of cooperatives and "isusu" about their experiences and future plans revealed the following:

- savings in isusu are still too small for any larger investment such as in food processing machines;

- credit, especially medium-term credit for larger investments was unavailable;

- consultancy services on the investment of funds were non-existent.

In November 1984, an informal agreement was reached between a union of self-help organisations and the Cooperative and Commerce Bank (CCB), Nigeria (Ltd.), to permit member societies to obtain group loans for on-lending, based on previous savings. Each member of an isusu would be fully liable for the total amount of loans disbursed to the association. A committee was elected, consisting of a president, a vice-president, a secretary and a treasurer representative at that time, of the 40 member societies. Most of the associations involved were of women. From April 1985 on, member societies who wished to save and obtain loans, initiated savings operations, and opened savings and current accounts with the CCB.

Eight months after the establishment of the union, 14 societies had started to save and by December 1985, had saved N 16,160 towards loans. In addition, members continued to save in their respective isusu. At present, isusu savings exceed those in the bank and both sides agree that mutual confidence will have to be built gradually. The first ten associations obtained their first loan, amounting to N 18,000, after a minimal saving period of one year. One of the isusu withdrew the money from its savings account for direct investment.

The bank managers generally have confidence in the performance of the isusu associations. Significantly the employees of the rural bank branch have set up their own isusu savings association in order to participate in the combined savings and credit programme.

5.5 Togo

Linking informal financial institutions and a savings bank

Togo is a small country with 3 million inhabitants. The capital Lome has a population of about 250,000 and the largest provincial town is Sokode with about 55,000 inhabitants. According to 1980 World Bank estimates, employment figures for the formal sector were 48,600 and for the informal urban sector 96,000, of which 57% were in trade and 43% in craft occupations. Of these 38.5% were in Lome and 61.5% in the rest of the country. Agriculture accounted for another 750,000. There are some 40 different ethnic groups in Togo.

After the economic crisis of the early 1980s the Government began to recognise the importance of local resources, indigenous institutions and the need to support small businesses.

Informal financial institutions

Informal financial institutions are widespread in Togo and make contribution significantly to rural village economies. They are to be found in all classes and areas, both rural and urban and are of three major types:

1. Rotating savings associations, most with weekly or monthly contributions.

2. A system of daily deposits by individuals to a money collector who returns the total amount at the end of a month to the contributor after deducting a fee of one day's contribution. This institution generally operates in markets.

3. A combined rotating and daily deposit association, where a cashier collects a fixed amount from each member every day; at the end of the month, one member receives the total amount collected. This procedure is repeated until each member has received the total at least once.

In a study of 36 members of a craftsmen's association in Lome in 1984, all were found to be members of rotating savings associations, belonging to 33 associations with a total of 841 members. The annual amount of savings mobilised is CFA70,000⁽¹⁰⁾ for each individual, CFA1,776,000 for each savings association and CFA58.6 million for all 33 associations.

In 1986, in the central market of Lome, it was found that all market women and traders belonged to product associations of 200 to 300 members each. They had a system of monthly and weekly contributions, both to finance the affairs of the association and for a mutual aid fund. In addition, most were active in one or more of the savings institutions mentioned above.

In Sokode, with less monetary flow, such financial activities are on a smaller scale than in Lome. Of 27 craftsmen interviewed in Sokode in 1984, 55.5% were members of an IFI of which 22.2% were in rotating savings associations and 33.3% in the daily deposit system which operated on the market. In the rotating savings associations annual savings mobilised were CFA9,400 per member and CFA82,000 per association. In the daily deposit system, deposits amounted to CFA182,500 per year and individual. Of 37 market women interviewed on the central market of Sokode all made daily deposits amounting each year to CFA210,000 each.

Formal financial institutions

There are three major and four smaller commercial banks in Togo, three development banks (all in a weak financial situation), a savings bank and the national social security fund. In addition there is a federation of savings and credit cooperatives - CONAUDEC.

So far the formal financial institutions have been unable and unwilling to provide finance to very small businesses and the informal sector. Major obstacles in this respect are a lack of contact with the sector, high transactional costs and lack of collateral. The most significant formal financial institution in this respect is the savings bank of Togo (Caisse d'Epargne du Togo - CET), a public corporation which collects savings through 34 post offices throughout the country and two direct agencies in Lome. In 1978 it started some limited lending operations. In 1984-85 deposits increased by CFA637.3 million (bringing the total to CFA3,195 million), while CFA647.2 million were given out in loans. There were 60 housing loans amounting to CFA630.8 million and 30 investment loans amounting to CFA16.4 million. In 1984 the CET opened a branch office in Lome's central market, but the number of clients remains low and the range of savings schemes limited. Even in the central market branch office, transactions are limited.

Linking IFI to the Caisse d'Epargne du Togo (CET)

In the late 1980s the CET started to make greater use of existing indigenous savings and credit institutions. It began to direct its attention to two target groups: craftsmen organised in craftsmen's associations, and market women and traders organised in market associations.

Members of the craftsmen's associations who belonged to different rotating savings associations agreed to deposit the association's receipts in a group account with the CET. On the basis of these deposits, each group obtained a loan from the CET for on-lending to its members, while the savings deposit remained with the bank as a security. The loan to the group was a multiple of its deposits which could increase after a period of satisfactory repayment. A similar system was introduced for market women and traders on Lome's central market. Market associations would deposit their funds from weekly and monthly contributions with the CET branch office.

Members of rotating savings associations within market associations were able to deposit their funds with the CET. On the basis of the amount paid in, a system of group lending was worked out. In effect, the savings association became a mutual credit guarantee group for the loans made to members.

Market women participating in the daily deposit system could also make their deposits to a CET agent rather than to the informal agent. To arrange this, the bank simplified some of its procedures and started using daily deposit collectors. The bank also examined the possibility of collaborating directly with indigenous deposit collectors either in a limited way by depositing the funds with the CET and leaving the informal system to continue as before, or by employing the collector directly while, at the same time, maintaining the informal practices. This meant that the bank was following the traditional system rather than trying to modernise the informal system.

Increased collaboration of the CET with the market and savings associations and with the daily deposit collection system helped develop a bridge between the market women and the bank and thus motivated more women and men to become their clients. The bank in turn has had to simplify its procedures to speed up transactions.

Both the IFI and the formal institution - the CET - could ultimately benefit from this collaboration which has demonstrated ways in which linkages between the formal and informal financial systems can take place.

CHAPTER 6

REVIEW AND CONCLUSION ON MAJOR ISSUES

6.1 Delivery systems

To overcome the high transactional costs both for lenders and borrowers, credit programmes for microenterprises (or indeed for all small enterprises) must be kept as simple as possible. As has been said, the access to loans for small businesses is directly related to proximity. Banks with a widespread network of branches or a system of unit banks (as in the case of the rural banks of the Philippines and Ghana) reduce borrower transaction costs by making the points of loan application physically close to the entrepreneurs.

Programmes which are highly decentralised in their operations, such as Grameen, BRAC in Bangladesh, and BRI Unit Desa in Indonesia, have achieved much wider coverage and reached a far higher number of borrowers than those operating from a single head office or from a regional office at a distance that may be difficult for some to reach.

Simplified loan processing to bring about rapid approval and disbursement means reducing the amount of information required on application to a minimum and widely delegating authority to approve loans. Branch managers or even loan officers (with certain safeguards) could be authorised to approve small loans. First loans to new borrowers should generally be for small amounts, for example 25-30% of the upper limit of the programme, but in same cases 10% or even less. The principle of increasing loan size as borrowers satisfactorily repay each loan is a good one which has proved successful.

As is noted in several of the cases presented in this paper, when the upper ceiling is low there is less scope for increasing loans as borrowers prove themselves with a good repayment record. The strongest incentive to repay a loan is the anticipation that this will lead to being granted another larger loan. When loan ceilings are low only multiple loans of the same size can be expected. Accordingly the economic growth and impact resulting from the lending programme is limited.

Those who favour low loan ceilings use a social rather than economic argument. A loan average of $200 or less (many of the programmes considered most successful have a loan average even lower) and a loan limit of $500 is intended to ensure that the financial assistance is given to the really poor and needy. As such it becomes "poverty alleviation" rather than business development. A stronger argument is that that with low loan sizes a limited financial resource base goes farther and helps more people whereas larger loans would appreciably limit the number of loan beneficiaries.

In practice, social motives can lead to loans being too small to permit beneficiaries, once they have exhausted the possibilities of the programme (many programmes limit the number of loans to a beneficiary) of bridging the gap to mainstream banks for their financing even when the organisation running the credit programme encourages them to do so.

Faced with this situation some programmes have raised loan ceilings. In Latin America a number of foundations in Colombia give loans of to $3000 or higher. In the Dominican Republic the an upper limit if $5000 in the case of ADEMI (Association for the Development of Microenterprises). The BRI/Unit Desa now lends up to $13,500 when several loans of increasing size have been repaid.

Some NGOs have found that they are ill-equipped for the heavy burden of loan disbursement and collection. Some programmes therefore make arrangements with commercial banks or other financial institutions to disburse and collect loans for a fee - usually about 3%. This is usually not a very profitable business for the banks but it does bring them in contact with potential new clients. This type of arrangement can have a positive "spin-off" when banks decide to lend directly to SSE borrowers with good repayment records. Some object to banks creaming off the best clients in this way.

To operate a successful credit programme for microenterprises and the poorest small businesses, institutions must adopt good accounting and sound portfolio management practices. A major failing of some institutions in developing countries has been a lack of transparency in managing the arrears and writing off uncollectible loans as necessary. Rescheduling of loans is an acceptable practice in some cases, but it should not be used to hide the true state of a loan portfolio. To achieve the desired level of tracking of loan portfolios, the computerisation needed to provide up-to-date reporting on the state of loans should be introduced. This is an additional valid reason for transferring the disbursement and collection of loans to a bank that is better equipped to maintain up-to-date records.

The Food and Agricultural Organisation (FAO) has developed a microcomputer software programme called Microbanker for use in managing small deposits and loan accounts. This programme was successfully evaluated in Asia by APRACA (Asian Regional Agricultural Credit Association). It has reduced some bank operating costs and released staff to improve loan collection. APRACA is part of FAO's system of regional credit associations; a similar organisation, AFRACA, has been created in Africa. AFRACA is reportedly experimenting with the Microbanker programme.

6.2 Transactional costs

It is widely accepted that the relatively high transactional costs in relation to loan amounts disbursed is a major deterrent both to the banks when considering putting out small loans to microenterprises and to these same potential clients from requesting loans from these institutions.

Lending institutions' transactional costs - the costs incurred in making loans - are of two types: administrative costs and losses incurred due to loans that are not repaid. In the best managed portfolio the loss may amount to 2-3% of the value of the portfolio (Grameen, Unit Desa, BRAC, etc.), but usually it is in the order of 5%. While 2% would be a desirable standard to aim at, 5% can be regarded as the upper limit of acceptability.

Clearly there is a relationship between the level of administrative costs and loss through loan defaults. More careful processing of loan requests and screening of applicants together with increased loan supervision could possibly lower loan default loss but the extra administrative costs might make the effort economically not worthwhile. A major increase in administrative costs to reduce loan losses from 3% to 2% might just not be economically justified. However this consideration has to be set against the contagion and demoralisation that might creep into a loan programme when loan delinquency passes a certain level. Over a certain agreed level it is probably better to deal drastically with loan delinquency (exclusion from future loans, vigorous pursuit of penalty payments, etc.) rather than to raise administrative costs unduly to try to prevent potential loan delinquents entering the programme.

Generally it is difficult, if not impossible, to lower administrative costs below a certain level. Unfortunately, total administrative costs will show only minor differences over a wide range of loan disbursements. For a bank the solution might be to set a minimum loan size (US$500-1000 or higher) and then rely on an NGO (or SHO) intermediary to do the selecting and screening of promising small enterprises and provide them with the consultancy help they need to prepare their loan applications so that by the time the request is brought to the bank a good deal of the administrative processing has already been done. Suggestions have been made that groups of microenterprises should present their loan requests together and costs through groups processing, but it is doubtful whether this would be feasible.

The question of how to lower transactional costs raises the issue of the so-called "minimalist" approach. This is the term used (mainly in USAID supported programmes) when credits are given without any other form of non-financial aid (e.g. training or advice) and where little effort is made to probe the use to be made of the money borrowed. Grameen bank does not concern itself with how the loan will be used or whether the project is likely to be profitable. Of course this reduces the number of questions asked of the borrower and the time spent by the loan officer in clarifying the information given. In this way the administrative costs of processing loan requests can be lowered. If, as a matter of policy, initial loans are only for a small amount, it may be that the sum at risk is less than the cost of processing which at best can only reduce - not eliminate - uncertainties.

Credit schemes which are part of wider SSE promotion or development programmes will have to include training and technical consultancy services and in these cases administrative overheads are bound to be high and this will in turn limit expansion of the programme (see the case of Colombia). Administrative costs incurred directly in processing a loan programme (i.e. in selecting clients, in loan appraisal and approval, supervision and collection) should not include expenses incurred in administering training or consultancy. By increasing loan handling efficiency, loan processing and approval can be speeded up, but administrative costs can rarely be brought to less than 10% of the loan disbursed where average loan size is US$300 or less. Even this figure would imply that some preselection of clients and projects has taken place.

For commercial banks to be willing to make loans to very small borrowers a way has to be found to share the risks through some form of loan guarantee (groups, mutual and NGO or special guarantees) and transaction costs have to be reduced through an intermediary (usually an NGO or SHO) assisting in the preselection of clients and in helping clients in loan request preparation.

Hans Seibel has suggested what he calls an "innovative solution to the transaction cost problem" through closer linkage between banks, SHOs and NGOs. This linkage approach was initiated in Indonesia in 1986 by APRACA, the FAO Regional Agricultural Credit Association in Asia. The linkage is based on the following principles:

- Working through existing institutions.

- Respecting the autonomy of participating institutions.

- Recognising SHOs and NGOs as financial intermediaries.

- Promoting savings.

- Linking savings and credit.

- Using group savings as partial collateral.

- Group liability.

- Granting financial incentives for prompt repayment.

- Fast services through simple procedures.

- Financial viability through market risk.

- Encouraging banks to assume the credit risk.

- Ensuring institutional viability of banks, SHOs and NGOs through cost coverage from the interest margin.

SHOs - self help organisations - are groups that may be of indigenous origin or they may have been initiated by governmental or non-governmental organisations (NGOs). As already seen, because they are very close to their members and well informed about borrowers, SHO transaction costs tend to be low whether for savers, borrowers and lenders. In the model proposed by Seibel the functions of the SHO are as follows:

- Collecting savings from the members.

- Building up an internal loan fund.

- Examining the credit worthiness of borrowers.

- Providing financial consultancy services to members.

- Arranging for informal collateral or guarantees.

- Granting loans to members.

- Collecting loan payments.

- Applying social control mechanisms to enforce repayment.

- Acting as a credit guarantee group.

- Acting as a financial intermediary between members and outside agencies (such as government agencies, donors, NGOs, banks).

A prerequisite for effective financial intermediation is that the SGOs must be well organised, maintain good book-keeping and that their members are in a position to absorb and repay additional credits.

In this proposal for reducing transaction costs through direct linkages, Seibel suggests that there could be two arrangements, one where there would be direct linkages between banks and SHOs and a second where the SHO would work through an NGO as an additional intermediary. Even in the direct linkage model SHOs might use NGOs in an advisory capacity.

Where NGOs do work with SHOs as in the intermediation role, their activities may include any of the following:

- Financial advisory services.

- Training in book-keeping and financial management.

- Promoting innovative savings schemes.

- Mediating contacts with a bank.

- Depositing savings of SHOs in a bank.

- Examining the credit worthiness of SHOs.

- Negotiating bank loans for SHOs.

- Entering into a loan contract with a bank as the legal borrower.

- Onlending to SHOs.

- Collecting from SHOs.

- Bearing the credit risk.

- Repaying the loan.

When they are not acting as direct financial intermediaries, NGOs may carry out several of the activities mentioned above. They would not, however, be expected to enter into a loan contract or bear the credit risk.

Seibel reports that in Indonesia during the 19 months prior to the end of 1990, 26 banks or bank branches, 16 NGOs and 417 SHOs actively participated in a pilot programme of linkages. The 417 SHOs deposited the equivalent of $240,000 in banks and obtained 496 bank loans of approximately $1.2 million. Approximately half of this sum was lent directly to SHOs and a little over half to NGOs as intermediaries. In both arrangements arrears were very low - below 0.5%. These arrangements have made it possible for self-help organisations in the informal sector to gain access to bank finance at market rates.

6.3 Interest rates

Over the years there has been a great deal of discussion internationally on interest rates. Financial specialists strongly recommend that, whatever the size of the loan programme, loans should be granted at the prevailing market rate. Governments, on the other hand, have often held the view that subsidised loans helped small enterprises and stimulated development. One should first recognise what is meant by "subsidised". This means not only that levels of interest are below those of commercial banks, but also that they are "negative" in real terms, that is below the annual level of inflation. Thus, after a year, loans are worth less even including the total interest paid. A recent ACCION International Discussion Paper (Castello, Stearns, Christen, July 1991) shows that between 1980 and 1988 the average annual inflation rate in Colombia was 24% (World Bank, 1989), so that for a credit programme portfolio in Colombian pesos worth $100,000 in 1985, if no interest was charged, it would be worth only $43,897 in 1988, i.e. it would have lost 56.1% of its value. The paper further goes on to show that in a situation where the annual rate of inflation was 18%, a portfolio worth $800,000, 63% ($500,000) of which was donated (i.e. at zero interest cost) and 37% ($300,000) was borrowed at 15%, allowing for an annual loan loss rate of 3%, and operating costs of 22% (high but realistic in many cases), an interest rate of 42% of the loan would cover the costs. This would mean a positive real rate of 24% above inflation (18%). It should be noted that if the 37% of borrowed funds in portfolios is made up of savings, the 15% paid would in real terms be a negative interest rate. In the case of voluntary savings, probably a higher rate would have to be offered.

If the effective interest rate offered is less than 42% in the case given above, the value of the portfolio would erode annually. The term "effective interest rate" is used and this may differ from the nominal rate at which the loan is offered. The effective interest rate includes all the costs incurred, including the nominal rate as well as commissions and fees paid to the lending institutions.

Furthermore, as indicated in a case described of a loan programme in Bangladesh (see above) interest payments are often required in ways that make the actual annual effective rate higher. The interest should be charged on the declining balance, that is, on the part of the loan still outstanding, but in many cases (particularly in Latin America) interest is levied on the total value of the loan and is required to be paid in advance or is even deducted from the amounts of the loan actually disbursed. This makes the annual effective rate much higher than the nominal annual rate which is the rate actually paid only when interest is paid on a declining balance basis with every repayment.

Table 9 shows the big difference of effective interest rate when interest payments are made in any way other than on the declining balance. The Table shows how the "real effective rate" goes up as commission and required compensating balances (savings) are taken into account.

For financial transactions between SHOs and banks or between NGOs that act as intermediaries, there will need to be an interest rate that gives a proper margin between the interest rates for savings and those charged on credits. For these financial transactions there must also be proper margins to enable the lending organisation to cover administrative costs and maintain the value of the loan fund in real terms, with consideration being given to the maturity time or repayment period of the loan and its size, or a combination of both. Allowances must also be made for the percentage to cover losses on unpaid loans.

Where an intermediary is involved between lender and borrower, such as an NGO, a premium or additional margin has to be included for the services rendered by the intermediary and possibly some account taken of the correspondingly lower transaction costs of the bank in handling the loan.

Table 9: Effective interest rates (per month)

(Original loan amount $100; loan period three months;

nominal interest rate 2% per month.)

CONDITIONS		PRESENT VALUE (amount borrowers receive)	PAYMENT PER PERIOD	EFFECTIVE INTEREST RATE
A.	No additional financial costs	$100	$34.68	2%
B.	Interest calculated using original loan amount instead of declining balance (flat method)	$100	$35.33 ($33.33 principal + $2 interest)	2.97%
C.	Interest deducted up-front from original loan amount	$95.97 ($100-$4.03 interest)	$33.33	2.08%
D.	5% commission deducted up-front	$95 ($100-$5)	$34.68	4.69%
E.	5% commission added to original loan amount	$100	$36.41 (Borrower must repay $105 because of commission)	4.55%
F.	25% compensating balance required; no interest earned	$75 ($100-$25)	$34.30 ($34.68 - $.38 interest earned on deposit)	18.34%
G.	25% compensating balance required; 1.5% interest per month earned on balance	$75 ($100-$25)		17.65%

Source: "Exposing Interest Rates"; Castello, Stearns, Christen, in Discussion Papers Services, Document No. 6, ACCION International, USA, July 1991.

There is no doubt that availability of money is more important to a small-scale entrepreneur than the cost of the money, and most SHOs/NGOs should set an interest rate that covers the full cost of the loan plus administrative expenses, including some premium for risk and possibly for the borrower's life insurance where this is appropriate. As shown in the case of Bangladesh (see above) there is generally little or no fall in demand for loans as interest rates rise if there is no other source of funds except the informal market. There may be a point at around twice the commercial rate, where there may be a reduced loan demand. An SHO/NGO would try to achieve a small surplus to provide an additional source of funds. Banks of course would charge full market rates to cover their costs and the risk and to afford them a margin of profit. SHOs should be free to set their own interest rates for saving and credit operations which may be higher than the banks' commercial lending rate. In practice in the past many informal SHOs/NGOs paid either no interest or very low interest on members' savings but, in recent years, the rate paid has been increasing in order to generate more internal funds and be able to lend more money, which is the ultimate objective of the organisation.

As regards interest rates for transactions between SHOs/NGOs and banks, it is recommended that these be based on existing national commercial rates. Unfortunately in some developing countries the government controls and restricts interest rates thus distorting the situation.

6.4 Credit guarantees

As has been mentioned, one of the major barriers to small enterprise borrowing from banks has been their inability to provide suitable collateral. As a result schemes have been developed both in industrialised and in developing countries to provide credit guarantees to banks so that they are compensated in the event of a default on the loan. Some would argue that banks should be encouraged to reduce their collateral requirements and base their lending on the viability of the project being financed and the creditworthiness of the borrower. While this may be desirable it is unlikely, in most cases, that the banks'demand for collateral will be eliminated or reduced to accepting the asset that is being financed as security for the loan. This becomes even more complex when the bank is financing working capital.

In practice there is a tendency for some banks to reduce collateral requirement when a client is known and has proved creditworthy by the timely repayment of loans. Usually by this time clients can provide the collateral required.

While it is desirable to educate banks to lower collateral requirements after suitable appraisal of the borrower and the project, there is need to recognise that, in most cases, banking practice will continue to require some degree of security before a loan is approved. To overcome this problem, in a number of countries various credit guarantee schemes have been developed. Most of these are prepared to offer guarantees to lending institutions on anything from 60% to 80% of the loan. When there is an established case of a default that is clearly not due to the negligence of the bank, the credit guarantee will compensate the bank for its loss according to the agreed risk sharing proposal. The fund set up to offer these guarantees usually comes from the government, although in some cases there have been amounts deposited by foreign donors or NGOs. In some cases, guarantee schemes also require a contribution from the banks themselves, or from local associations or representative bodies.

While, to a degree, these schemes have worked in Europe, North America and Japan, they have encountered a number of difficulties in most developing countries. In general, banks find the systems too bureaucratic and fear undue delays before their guarantee claims are settled. There is also grave concern among many - partially substantiated by experience - that such credit guarantee schemes serve to demotivate the borrower and act as a disincentive, both to the person repaying the loan and the bank to collect. In some cases mutual recriminations have arisen after claims begin to increase, with the banks being blamed for inadequate screening of borrowers and projects and a poor collection effort. This is then sometimes used as an excuse by the guarantee organisation not to pay out on the claim.

Nevertheless, credit guarantee schemes can be helpful provided they are embarked upon prudently and there is a process of screening to ensure that guarantees are not given automatically which can lead to a high volume of claims being made (as in ASKRINDO in Indonesia). As has been shown in several of our examples, it is appropriate in the case of very small borrowers (such as the Grameen Bank) for the accumulated group savings to be used as a guarantee or collateral for individual members. The group then acts as a joint liability or mutual guarantee group. Apart from providing guarantees for members who negotiate loans with formal financial institutions, most lending from self-help groups is guaranteed mutually by all the members. Thus there is a certain amount of "peer pressure" from the members of the group which influences individuals to repay the loans. Sometimes a group life insurance scheme is built up to cover exceptional cases where default results from the death of a member.

A credit guarantee scheme that is based only on the savings of an SHO deposited in a bank will of course be limited. However, this situation could be improved by linking up with a broader credit insurance or credit guarantee fund in which the bank itself as well as some NGO or foreign donor also contribute to expanding the fund. In general, for creditworthy customers - which means in general those that have already shown their capacity and willingness to repay loans - these funds could justify a lending of 5 or 10 times the value of the guarantee fund so created, but it is usual prudent practice, in the early stages of such guarantee schemes, to limit the overall guarantee exposure to not more than, say, a maximum of five times the amount in the fund on deposit.

Of course, from a long-term point of view, an NGO acting as an intermediary or operating a credit programme for individual small-scale entrepreneurs might act as guarantor without establishing a distinct fund for this purpose. Although under certain circumstances - particularly where the NGO is supported by important members of the business community with substantial resources behind them - banks may accept such an explicit or implicit overall guarantee, there are still advantages in setting up a guarantee fund that is clearly monitored. This is preferable to a "blanket coverage" by an NGO or other group of persons without clear indication of liability and capacity to meet claims. The group principle is now being applied in some countries (see above the case of the Philippines) in the formation of mutual guarantee associations which are set up by 40-100 small enterprises in the same locality, each contributing a comparatively small sum into a guarantee fund. The contribution of the members is then usually matched by the Government or a foundation to build up a more significant sum. The total guarantee fund is then deposited and the income earned added to the guarantee amount or, together with the premium or fee paid by each borrower, used to help cover operating costs. In every form of guarantee a fee is charged to cover the administrative costs of the scheme. Sometimes the fee is a one time advance payment of 1 to 3% of the total loan or amount guaranteed, but in other cases the fee is paid monthly being related to the amount still under guarantee.

Mention should also be made of a novel approach to credit guarantees practised by FUNDES, a Swiss-based privately-owned company that assists in setting up loan guarantee schemes mainly in Latin America and then participates by accepting to cover with its own resources part of the credit risks of the schemes. FUNDES enters into an international partnership with local business communities by setting up suitably designed local credit guarantee schemes for SMEs.

6.5 Loan recovery

Loan repayment delinquency is recognised as the major threat to maintaining the value of a loan fund. A high rate of non-repayment erodes the value of a loan portfolio and reduces income which undermines any hope of achieving sustainability of the programme.

Some years ago, most microenterprise lending programmes were plagued by high arrears and defaults. They came to be regarded in some countries as "give away programmes" where borrowers where not really expected to repay loans. There have been changes and a number of programmes have shown that high loan recovery rates are possible.

A major issue in tackling loan delinquency is how to measure the extent to which a programme is "contaminated" by the non-repayment of loans. As Katherine Stearns shows in a recent ACCION International Discussion Paper (No. 5 January 1991), there are no less than 20 ways of measuring the extent to which a loan portfolio is affected by non-repayment of loans. Twelve of these measures concern the "general quality of the portfolio, five measure the extent of repayment and three measure the loan loss incurred". As Stearns points out the most usual formula is

the outstanding balance of loans with payments overdue

total value of loan portfolio

This is what is sometimes called the "contamination" formula, i.e. the extent to which a loan portfolio is "contaminated" or "at risk" due to non-payment of loans.

There is a problem even with this formula. In the earlier stages of a programme when the loan portfolio is increasing rapidly, this formula will tend to underestimate the real extent of a loan delinquency problem as the portfolio will contain a number of loans on which no payment has yet fallen due. In reality this shortcoming is transitory and simply states that one cannot assess the quality of a loan portfolio until the programme has been running for some time.

Part of the problem of loan delinquency lies in the poor management of portfolios and in the fact that information available on the state of loan arrears is badly out of date. It has been found that several credit programmes cannot give immediate up-to-date information on the amount in arrears or in default. This leads to slow reaction and to an insidious culture setting in where late or even no payment of loans is not taken too seriously. Demoralisation in regard to punctual payment of loans sets in among clients. On the other hand, immediate reaction and swift measures such as imposing penalty payments without delay helps lead to a discipline of payment on time.

Most programme managers believe the biggest incentive to loan repayment on time is the "carrot" or prospect of obtaining another larger loan. This assumes that the borrower has the means to repay but has the choice between deciding whether he can make best use of the money involved by repaying the loan (as an investment in future goodwill) or by using it for some other purpose. This is why low interest subsidised loans often have more loan delinquency: the money is cheap and can be used to better purpose to repay other more costly bills.

Penalty charges for late payments can be effective if followed through but if constantly forgotten or forgiven or not applied and collected they may in fact have an overall negative effect in creating further demoralisation.

Rescheduling of loans can be effective where it is clear that the borrower is facing serious business problems due to some external cause beyond his control or because of a personal crisis. However, rescheduling is frequently abused so that it masks the true state of the portfolio. If most of the loans have had to be rescheduled there is a serious loan delinquency problem, although this is not apparent from the available figures. It has been suggested that loan rescheduling (which means rewriting the loan contract) should require approval at a high management level or even treated as a new loan. However, high loan delinquency can sometimes imply a need to redesign a loan programme. It may mean loan sizes are too high or repayment periods too short for the cash flow situations of the clients.

There have been situations where a culture of loan delinquency has had to be overcome. The credit institution can only do this by denying loans to all who have failed to pay back a loan in the past and rewarding those who do repay their loans.

Above all, an institution must be fully informed at all times on the exact state of loan repayment and much act immediately when loan payments are overdue. A quick investigation is called for to identify the reasons for non-payment and action should be taken accordingly. A high loan delinquency rate will ultimately discredit the credit institution which will be deprived in the end of resources for further lending.

6.6 Graduation to formal lending institutions

The conclusion that might be drawn from the examples given in this paper and from the experience of many other programmes is that the ultimate objective of all the programmes must be to persuade formal financial institutions into providing credit for very small enterprises. In this model, the "self-help" groups and the NGOs would act as intermediaries to build up a group of reliable clients who have successfully received and repaid loans and who can be presented to the financial institutions as suitable borrowers who can be relied upon to service their debts successfully. This would presuppose that these organisations, mainly NGOs and SHOs, see their role as transitional in that they would ultimately help their most successful clients graduate to the mainstream financial sector of the country and meet their financial needs in direct transactions with formal banking institutions.

Unfortunately, in many cases this model breaks down in practice. The NGOs quite often find it difficult to part with their most successful clients and the formal banking institutions for their part are still reluctant to become involved in lending to these small borrowers irrespective of whether it has been demonstrated beyond doubt that they are creditworthy and have a track record of repaying loans.

There have been other examples which seem to indicate that the banks are more concerned by the high transactional costs involved in lending to small enterprises than they are with the risk perceived to be involved in such lending.

The idea of the banks lending sums to intermediaries, be they NGOs or SHOs, who would then use these funds to on-lend to small enterprises possibly larger amounts than their usual loans, may be considered a form of graduation. The concept then would be the graduation not of individual borrowers, but of a whole group of borrowers or loan portfolios at one step. Of course, in such an arrangement, the risk would still remain with the NGO and this would not achieve the objective of the risk and compliance being borne by the bank and the borrower.

The term "graduate" has come to have different connotations among NGOs. Perhaps the minimal definition of the term might be that of an NGO credit scheme in Tunisia where graduation is regarded as moving "from a low level of organisational productivity to a level in which economic opportunities for SMEs are generated". This definition presumes that a "graduate" with such qualifications will ultimately be given access to credit by the formal financial institutions, although there is no way of knowing how long this will take or if it will ever happen. Until the banks are ready to receive these "graduates" there are those who believe that non-banking credit programmes operated by NGOs should grow with their clients. They may start (as many of them do) with a loan ceiling of $500-2,000 but could expand to lending several times this amount by establishing direct links of the sort outlined with the banks. In this model the NGO would undertake true intermediation, that is absorb the risk but use the extra financial muscle to obtain bank funding for the better customers. The negative aspect would be that the NGO intermediary would become increasingly involved in the transactions of its larger clients and its resources would go primarily to them, resulting in the volume of clients and new entrants to the borrowing market being reduced. This is not in keeping with the original aim of most of these NGOs, which was to create a link between the informal and formal financial sectors.

Experience has shown - as the illustrations in this paper indicate - that perhaps the formal financial sector tends to remain conservative in its operations making the objective of cooperating with banks difficult to attain. This explains why some NGOs have decided to establish full banking institutions of their own for their clients (as in Bangladesh and Bolivia).

The role of NGOs in providing large numbers of clients with very small loans inevitably makes their operations somewhat inefficient even though, because of their knowledge of the clients and closer relations with them, they can reduce transactional costs. An illustration might again be given of the NGO, ADEMI in the Dominican Republic. From 1983 to 1990 ADEMI expanded its operations to a nationwide programme with 20 offices dispersed throughout the country, but its staff increased from seven persons in 1983 to 81 by the end of 1989. The total number of loans given out in 1988 was 19,428, but 89% were for manufacturing enterprises. For an NGO this number of full-time workers and the resulting high administrative costs would be considered as taking an excessive share of resources.

It will be recalled that in the examples given earlier of foundations in Colombia, operational costs took more than 50 cents per dollar lent although this included significant costs for training and advisory services. There have been cases where it has taken more than one US dollar equivalent in transactional costs to lend a similar amount. While this level of administrative and transactional costs can be reduced substantially, NGOs will always have higher intermediation costs for the amounts they lend than the banks. This is a strong argument for pressing on to find innovative ways of increasing the involvement of the banks and formal financial institutions in lending to SMEs.

Latterly some doubts have arisen about whether the objective of graduating borrowers to the formal financial system is feasible. Katherine Stearns of ACCION International recently (1991) produced an interesting paper in which she questioned whether "client graduation is a myth". She quotes several credit programme directors in Latin American who regard the graduation objective as unrealistic.

Several reviews of microenterprise lending programmes - in particular the USAID stocktaking exercise (Boomgard, 1989) - have reported that the number of cases of clients graduating from NGO microenterprises credit schemes to borrowing from the formal banking systems is very small. Stearns says that in 10 Latin American credit programmes affiliated to ACCION which together are lending to over 16,000 microenterprises, only 287 cases have been identified of what might be called graduation, giving a rate of 1.8%. Low by any measurement standard.

Why are there so few cases of graduation? Stearns' limited study of 19 clients, while too small to reach wide conclusions, points to some possible explanations. It seems that reluctance of banks to lend is matched at least as strongly by potential clients to borrow. Clients find that application processing by banks is a lengthy and time consuming procedure in most countries and bank guarantee requirements, even in the more advanced cases, are difficult to meet. In some cases the loan amount is too small to be considered by the bank. But, above all, microentrepreneurs feel more comfortable with the personnel in an NGO credit programme than with bank officials. Clients find that the bank requires data and information which it is too time consuming for them to provide.

For these reasons potential graduating clients want to continue borrowing from the NGO credit scheme rather than from the bank. Legality is also a problem in a few cases. The advantage of borrowing from a bank is that larger sums can be obtained and in some cases at lower interest, but this does not offset the features that discourage them from dealing with banks.

It is admitted by many credit programmes that they do not do all they can to encourage graduation largely because of the ambivalence of this move, namely that of losing their best clients. In fact, in some credit programmes, graduation is regarded as a condition imposed by donors rather than a realistic aim of the programme itself. Of course the fact that the banks are less than enthusiastic about taking on these new small clients with high transaction costs and low profitability hardly improves the situation.

One programme which has had rather more success in graduating clients is the ADMIC programme in Monterrey, Mexico which accounts for 180 of the microenterprise clients graduated (out of 287) in the Latin American programmes. Significantly ADMIC has one of the highest loan ceilings - $7,000 equivalent - of all the microenterprise lending programmes. ADMIC reports that it considers that close to 40% of its clients want to borrow amounts which could be considered by banks. Of the other programmes studied, FUCODES in Costa Rica with an average loan size of US$3,000, had a higher degree of graduation - similar to that of ADMIC. It In order to further the process of graduation to formal bank borrowing the microenterprise lender has to be prepared to lend larger amounts closer to the range that would be of serious interest to a bank. In Asia and Africa, most programmes regard themselves primarily as "poverty alleviation" programmes and so loan sizes are kept too low for clients to make the transition to bank borrowing.

6.7 Impact of programmes

The prime purpose of all the credit programmes for small and microenterprises described and reviewed in this paper is without question to raise the living standards of the beneficiaries, their families and their communities. The approach of channelling loans to those at the poorest and lowest economic strata is based on the recognition that private sector and entrepreneurial initiatives are the main vehicle for promoting economic development in a manner which will ultimately spread socio-economic benefits to a broad segment of those living in poverty. Business development is regarded as the most effective way of generating increased income, paving the way to better living standards and alleviating poverty.

Provision of finance for developing enterprises is considered the most direct means of stimulating sustainable long-term growth. The cases reviewed in this paper describe an array of institutional approaches, some quite innovative! All the programmes overcome formidable obstacles to ensure that funds made available are used effectively and in reality benefit the poorest and contribute in the long term to improving living standards.

Evaluation of the impact of programmes is difficult and costly so that most conclusions are based on fragmented and anecdotal impressions. Some evidence is conflicting. There are evaluations that say that credit has little or no impact and fails to raise people out of the poverty in which they are trapped. In general, NGOs moved into financial assistance programmes because they have become frustrated with the lack of impact and results of non-financial programmes such as training and advisory services. Others have moved in the opposite direction - namely expanding from credit into non-credit supplementary activities justifying this by claiming that credit alone was insufficient to make significant impact on the incomes of the poorer beneficiaries.

Some evaluations, notably those of USAID in 1985 (Kilby and Z'mura) have attempted to use techniques of benefit-cost analysis to arrive at the rate of returns (in financial terms) for the funds invested by the donor agency in five different credit programmes. Using some rather imaginative assessments of financial benefits, the evaluations concluded that some of the programmes showed high financial rates of return. Some later comments on these conclusions asserted (Tendler 1987) that these rates of return were at least as high as those obtained on large World Bank projects financing infrastructure and rural development.

Comparative benefit-cost analysis is not easily applied to credit programmes for microenterprises. The benefits are difficult, if not virtually impossible, to assess fully especially if one considers the alternative situation for a microentrepreneur: idleness and destitution for the individual and his/her family. Enhanced income that following a loan may be due to many factors - improved economy, better marketing arrangements, new activities, etc. - in which the credit has been only one element. Time needs to be allowed before evaluating the real results, but then many new elements usually enter the picture. Despite all these "caveats" the evaluations and studies made in Bangladesh, Colombia, Mexico and Kenya, for example, have generally shown a significant percentage increase in income, in value added and some increase - although less marked - in employment. Family income from a microenterprise is difficult to determine. In some cases total value added (sales less purchased input) is close to the family income, if only members of the family are employed and the space used is within the family residence. Increased sales are virtually always achieved through the increased working capital resulting from a loan, assuming the funds are used in the business.

As to employment - perhaps the major social and economic indicator of impact - measurements are more problematic. Most microenterprises benefiting from loan programmes rarely employ anyone from outside the family and so average additional employment has to be measured in fractions of a person (1.38 to 1.61 persons as in Bangladesh) per enterprise. In the Carvajal type programmes in Colombia there is evidence that there was an average employment increase as a result of assistance of approximately one employee per microenterprise (see above). One programme in Mexico - noticeably one that offers larger loans of US$7,000 equivalent - claims the creation of 4,600 new jobs and also, significantly, the sustaining of a much larger number that might have been lost had no financing been available.

As evaluation of programmes is costly and difficult, often anecdotal or less than statistically valid evidence has to be considered. All such reporting indicates a modest rise in the living standards of those who are already heavily engaged in microbusinesses, generally not the poorest and most deprived elements of the population. Some employment increase is attained, but mainly where loans are above a minimal amount. Bearing in mind that even the simplest form of production or service may require a capital of at least $500-1,000 or more to create a new job, it could be assumed that loans substantially below this level are unlikely to create new employment. Small loans contribute primarily by furnishing working capital that will maintain the microentrepreneur and at most some of his immediate family in work.

The difficulty in evaluating impact has led many donors supporting credit programmes to stress such features as efficiency of disbursement, number of beneficiaries and above all extent of loan recovery as the measurement of success. There is a definite correlation between loan recovery and socio-economic or business development arising from the loan. A high level of loan defaults and arrears is invariably indicative of the ineffective use of the loan even if in the short term there may be some cases of "wilful defaulters" whose businesses are successful. Despite this strong correlation generally between high loan recovery and socio-economic benefits to borrowers and communities, it should also be recognised that business growth also must be an important criterion in judging the success of a programme. There must be growth among the individual beneficiaries, and of the programme as a whole and its coverage. It is for this reason that the issues raised in the previous section on the "graduation" of successful beneficiaries are all important.

One of the social indicators being given increased prominence in assessing programmes is the extent to which women entrepreneurs benefit, since one of the serious social problems that all development aid must address, is how to ensure that women are given greater opportunities to play a significant role in business and in the economy as a whole. Over the last few years a number of major credit programmes have been created specifically for women entrepreneurs. As most women entrepreneurs are engaged in microbusinesses most of these women-specific programmes are forms of microenterprise credit programmes developed by NGOs. Other programmes that were not intended to be only for women beneficiaries are increasingly catering for them (e.g. Grameen and BRAC in Bangladesh, IDESI in Peru, KREP and ISP in Kenya). Many explanations have been given. One (perhaps a more cynical observation) is that all the major multilateral and bilateral donor agencies as well as international NGOs have made increased involvement of, and benefits for, women a major objective and have set up "women in development" units to monitor programmes from this aspect. Accordingly, some say, programme managers are motivated towards this aim too. However, there is also the important factor experienced by those engaged in credit programmes that women are better clients - they are more assiduous in their application to their businesses, more prudent in assessing new investments, more ready to learn and apply new skills and techniques and, above all, are more conscientious in repaying loans, all of which make them more attractive clients even in those programmes that are not specifically restricted to women beneficiaries.

The percentage of women borrowers and clients has therefore become an important social indicator in assessing a programme. Unfortunately until now most programmes in which women have participated to a significant extent have until now been programmes with very low loan levels which has led some to conclude that women borrowers because of greater risk awareness and prudence in entering into debt, have tended to borrow only small amounts. There is some evidence to substantiate this view, but some of it may be explained by the fact that the largest programmes providing loans for women have very low loan ceilings (Grameen and BRAC Bangladesh, etc.). It would be an undesirable development if a division developed between programmes with high loan averages ($1,000 and above) where male borrowers predominated while women borrowed small sums from more restrictive programmes. This may indicate a need for wider programmes and to provide women (and men where necessary) with more training and advice to encourage and help them to undertake larger business projects.

Some recent evidence (Eigen, ISP Kenya, 1992) based on a programme that had been going for some time shows that the average loan size made to borrowers is approximately the same for men and women.

Overall, one may conclude that financing programmes in developing countries appear to have a modest positive economic and social impact. Still more innovations may be needed to ensure that these programmes become even more effective in alleviating poverty, redistributing income, promoting economic development and ensuring that the benefits are spread over wider sections of the population.

BIBLIOGRAPHY

Ashe, Jeffrey; Coslett, Christopher E.: Credit for the poor (New York,UNDP, 1989), Policy Discussion Paper.

Asian Development Bank: Formal finance in Asia, in Asia Development Outlook(Manila, 1990).

Bank Indonesia: Report of the financial year 1985-86 (Jakarta, 1986).

Bank Rakyat Indonesia: Principal changes of the BRI credit policy to promote rural economic activities through the BRI Village Unit (Jakarta, 1986).

Bell, C.: "The informal sector and marginal groups", in IDS Bulletin (Brighton,UK, University of Sussex, 1973).

Blayney, Robert G.; Otero, M.: Small and microenterprises: Contributions to development (Washington, DC. United States Agency for International Development, 1985).

Brown, Jason: Small-scale lending in developing countries (Washington, DC, United States Agency for International Development, 1985).

Burnett, Jill: An over-view of Ghana's banking section: Review of financial institutions and rural areas and analysis of rural banks (unpublished thesis), Oxford, 1991.

Carvajal, Jaime: "Microenterprise as a social investment", in J. Levitsky (ed.): Microenterprises for developing countries (London, IT Publications, 1989).

Chandavarkar, A.G.: "The non-institutional financial sector in developing countries: Macroeconomic implications for savings policies", in Savings and Development, 1985, 9: 129-141.

-----: "Informal credit markets in support of microenterprise", in J. Levitsky (ed.): Microenterprises for developing countries (London, IT Publications, 1989).

Christen, R. P.: What microenterprise credit progams can learn from the moneylenders (ACCION International, Cambridge, Mass., 1989), Discussion Papers Series No. 4.

Chuta, E.; Liedholm, C.: Employment and growth in small-scale industry (London, Macmillan, 1985).

Codespa: Modelos de fomento de la microempresa en Colombia (Madrid, Fundación Codespa, 1990).

Castello, C.; Stearns, K.; Christen, R.P.: Exposing interest rates: Their true significance for microentrepreneurs and credit programs (ACCION International, Cambridge, Mass., 1991), Discussion Papers Series No. 6.

Credit Union Coordination of Indonesia (CUCO): Credit Union Movement in Indonesia (Jakarta, CUCO, 1986).

Dasgupta, B.: Calcutta's informal sector, in IDS Bulletin (Brighton, UK, University of Sussex, 1973).

Eigen, J.: Women entrepreneurs and their access to credit: Lessons from a microenterprice credit programme in Kenya (GTZ, Nairobi, 1992).

FUNDES: A new approach to loan guarantees, in Small Enterprise Development Journal, Vol. 1, No. 2 (London, IT Publications, 1990).

Goldmark, S.; Rosengard, J.: Credit to Indonesian entrepreneurs: An assessment of the Badan Kredit Kecamatan Program (Washington, DC, Development Alternatives Inc., 1983).

GTZ: Financing small enterprises: Report on the Conference held at Lusaka, Zambia, 1985 (Eschborn, Deutsche Gesellschaft fur Technische Zusammenarbeit, 1986).

-----: Financing small enterprises: Annex to the Report on the Conference held at Lusaka, Zambia, 1985 (Eschborn, Deutsche Gesellschaft for Technische Zusammenarbeit, 1986).

-----: Rural finance: Guiding principles, by R.H. Schmidt and E. Kropp (eds.) Eschborn, Deutsche Gesellschaft fur Technische Zusammenarbeit, 1987).

Hadad, I.: Development and community: Self-help in Indonesia (Prisma, 1983) 28: 3-20.

Hossain, Mahabub: The Grameen Bank in Bangladesh (Dhaka, Institute of Development Studies, 1984).

Hart, K.: Informal income opportunities and urban unemployment in Ghana, in Journal of Modern African Studies, 1973, No. 11, pp. 61-89.

Inter-American Foundation: Grassroots Development, Vol. 13, No. 1 (Rosslyn, Va., 1989).

International Labour Office: Employment, income and equality: A strategy for increasing productive employment in Kenya (Geneva, ILO, 1972).

Institute for Development Studies: The informal sector: Economic analysis (Jakarta, LPS, 1987).

Jackelen, H.R.; Rhyne, E.: Towards a more market oriented approach to credit and saving, in Small Enterprise Development Journal, Vol. 2, No. 4., Dec. 1991 (London, IT Publications).

de Jong, M.; Kleiterp, M.: Credit for small businesses and microenterprises in developing countries, in Small Enterprise Development Journal, Vol 2., No. 4., Dec. 1991 (London, IT Publications).

Kleiterp, N.: Financiamiento de la micro y pequeña empresa en America Latina (FMO, Netherlands, 1990).

Kessler, D.; Ullmo, P.-A.: Savings and development (Paris, Economica, 1985).

Koll, M.: Crafts and cooperation in Western Nigeria (Dusseldorf, Bertelsmann Universitatsverlag, 1969).

Kusumahade, M.: Basic information on the various perspectives of the urban informal sector in Indonesia (Jakarta, Friedrich-Ebert-Stiftung, 1987).

Lamberte, M.D.: Impact of special credit and guarantee programmes for SMEs on employment and productivity; ILO-JAPAN-DOLE Multibilateral Project on Strategic Approaches Toward Employment Promotion (Discussion paper, National Employment Symposium, Manila 1990) (Manila, ILO Asian Employment Programme, 1990).

Levitsky, Jacob: "Support systems for small-scale enterprises", in M.S. Molenaar,et al. (eds.): Small-scale industries' promotion in developing countries (Delft, Research Institute for Management Science, 1983).

-----: World Bank lending to small enterprise, Industry and Finance Series Vol. 16 (Washington, DC, World Bank, 1986).

-----; Prasad, R.N.: Credit guarantee schemes for small and medium enterprises Washington, DC, World Bank, 1987) Technical Paper No.58.

-----: Microenterprises in developing countries (London, Intermediate Technology Publications, 1989).

-----: Financing of small and medium enterprises (Vienna, UNIDO, 1988).

Liedholm, C.; Chuta, E.: The economics of rural and urban small-scale industries in Sierra Leone (East Lansing, Michigan State University, African Rural Economy Paper No. 14, 1976).

Lizardo, Sonia: Comparative study of micro and small business as poverty alleviation programs in the USA, Latin America and Caribbean (University of Pittsburgh, Thesis, 1990).

Madeley, John: When aid is no help: How projects fail and how they could succeed (London, IT Publications, 1991).

Maloney, Clarence: Rural savings and credit in Bangladesh (Dhaka, University Press Ltd., 1988).

Mann, C.K.; Grindle, M.S.; Shipton, P.: Seeking solutions: Framework and cases for small enterprise development programmes (West Hartford, Connecticut, Kumarian Press, 1989).

Meyer, R.L.; Graham, D.H.: Developing financial markets in Sub-Saharan Africa (Colombus, Ohio State University, 1991).

MIDAS Micro Industries Development Assistance Society: Annual Report 1990 (Thaka, Bangladesh).

Molenaar, N.; El-Namaki, M.S.S.; van Dijk, M.P.: Small-scale industry promotion in developing countries (Delft, Research Institut for Management Science, 1983).

Nihan, G.; Jourdain R.: "The modern informal sector in Nouakchott", in International Labour Review No. 117, pp. 709-719 (Geneva, ILO, 1978).

-----; Demol, E.; Jondoh, C.: "The modern informal sector in Lome", in International Labour Review No. 118, pp. 631-644 (Geneva, ILO, 1979).

Oehring, E.: "FUNDES: A new approach to loan guarantees", in Small Enterprise Development Journal Vol.1, No.2, (London, IT Publications, 1990).

Padmanabhan, K.P: Rural credit: Lessons for rural bankers (London, IT Publications, 1990).

Page, J.: Some economic considerations in small enterprise development (Washington, DC, The World Bank, 1983).

Pinney, A.: Partners in development. The Government and NGOs in Indonesia (Prisma, 1983) 20: 33-45.

Portes, A; Sassen-Koob, S.: "Making it underground: Comparative material on the informal sector in Western market economies", in American Journal of Sociology 1987, No. 93, pp. 30-61.

Quinones, B.R. Jr: Linkages between formal and informal sectors in rural financial markets: The role of self-help groups (Bangkok, APRACA, 1986).

Ramahan, S.G.V.: A study of credit system for small farmers development programme in Indonesia (Jakarta, FAO, 1986).

Remenyi, Joe: Where credit is due: Income generating programmes for the poor in developing countries (London, IT Publications, 1991).

Salloum D., Grace, L.: BRAC donor consortium review: Saving and credit activity (Canada, Calmeadow Foundation, 1991).

Schmidt, J.; Dedorath, G.W.: Small-scale financing and credit intermediaries Eschborn, Deutsche Gesellschaft fur Technische Zusammenarbeit, 1985).

Seibel, H.D.: Rural finance in Africa: The role of informal and formal financial institutions, in Development and Cooperation (Bonn, DSE, 1985) 6/1985: 12-14.

-----; Damachi, U.G.: Self-help organisations: Guidelines and case studies for development planners and field workers (Bonn, Friedrich-Ebert-Stiftung, 1982).

-----; Marx, M.T.: "Mobilisation of personal savings through co-operative societies or indigenous savings and credit associations?", in UN 1986, pp. 107-112.

-----; -----: Dual financial markets in Africa: Case studies of linkages between informal and formal financial institutions (Saarbrucken/Fort Lauderdale, Verlag Breitenbach, 1987).

-----: Linking informal and formal institutions in Africa and Asia, in Microenterprise in Developing Countries (London, IT Publications, 1989).

-----; Parhusip, U.: Linking formal and informal financial institutions, Paper presented at a Seminar on Informal Financial Markets (Washington, DC, World Bank, 1989).

-----: Microfinance for rural microenterprises: Financial innovations in a deregulated policy environment (Padang, Indonesia, 1991).

Sethuraman, S.V.: "The urban informal sector in Africa", in International Labour Review (Geneva, ILO, 1977), No. 116, pp. 343-353.

-----, (ed): The urban informal sector in developing countries: Employment, poverty and environment (Geneva, ILO, 1981).

de Soto, H.: "The other path": Invisible revolutions in the Third World (New York, Harper & Row, 1989).

Stearns, K.; Otero, M.: "The critical connection: Governments, private institutions and the informal sector in Latin America" (Washington, DC, ACCION International, 1990).

-----: The hidden beast: Delinquency in microenterprise credit programs (Cambridge, Mass., ACCION International, 1991) Discussion Papers Series No. 5.

-----: Microenterprise development programs: Is client graduation a myth? Cambridge, Mass., ACCION International, 1989), Discussion Paper Series, Document No. 3.

Steel, W.P.: Small-scale employment and production in developing countries: Evidence from Ghana (New York, Praeger, 1977).

Tecson, G.; Valcarcel, L.; Nunez, C.: Role of small and medium-scale industries in the industrial development of the Philippines (Manila, Asian Development Bank, 1989).

Tendler, J.: "What ever happened to poverty alleviation?", in J. Levitsky (ed.):

Microenterprises in developing countries: Papers and proceedings of an international conference held in Washington, DC, 6-9 June 1988 (London, IT Publications, 1989).

Timberg, T.A.; Aiyar, C.V.: Informal credit markets in India, in Economic Development and Cultural Change (Chicago, 1984).

Tokman, V.E.: "Micro-level support for the informal sector", in J. Levitsky (ed.):Microenterprises in developing countries: Papers and proceedings of an international conference held in Washington, DC, 6-9 June 1988 (London, IT Publications, 1989).

United Nations: Savings for development: Report on the Third International Symposium on the Mobilisation of Personal Savings in Developing Countries (New York, UN, 1986).

UNIDO/UNDP/ILO/Netherlands: Development of rural small industrial enterprise: Lessons from experience (Vienna, UNIDO, 1988).

United States Agency for International Development: Evaluation of MIDAS,Bangladesh (Washington, DC, ARIES, 1986).

-----: ARIES: Small enterprise credit models (Washington, DC, 1987).

-----: ARIES: Key issues in designing microenterprise programs (Washington, DC, 1990).

-----: Access to credit for poor women: A scale-up study of projects carried out by Freedom From Hunger in Mali and Ghana, GEMINI Technical Report No. 33 (Washington, DC, 1992).

-----: Proceedings of Seminar on Frontiers of Financial Assistance to Microenterprise Development (Washinton, DC, 1990).

Von Pischke, J.D.: Finance at the frontier: Debt capacity and the role of credit in the private economy (Washignton, DC, World Bank, 1991).

Webster, L.: World Bank lending for small and medium enterprises: Fifteen years of experience, Industry Series Paper No. 20. (Washington, DC, World Bank, 1990).

Vogel, R.C.; Burkett, P.: Mobilizing small-scale savings (Washington, DC, World Bank, 1986).

World Bank: Indonesia rural credit review (Washington, DC, World Bank, 1988).

-----: UNDP Project SRL/89/013. Employment and Poverty Alleviation Project.

Annex 4: Current approaches and institutions involved in microenterprise credit in Sri Lanka.

World Development Report (Oxford, World Bank, 1989).

Yunus, Muhammad, 1989: Grameen Bank: Organisation and operation, in J. Levitsky (ed.): Microenterprises in developing countries: Papers and proceedings of an international conference held in Washington, DC, 6-9 June 1988 (London, IT Publications, 1989).

1. ⁰ Exchange rate in 1982 US$1 = Taka 21.6; in 1987 $1 = Taka 31; in 1990 $1 = Taka 33.5; in 1991 $1 = 37.1.

2. ⁰ Swanirbhar is a government sponsored multi-pronged development programme for the poor, based on the principle of self-help.

3. ⁰ The exchange rate in 1986-87 was US$1 = Rp 1,627.90; in 1991 $1 = RP 1,973.01.

4. ⁰ This programme from Indonesia in the period 1970-86 is given here as an interesting example illustrating the effectiveness of self-help groups. The material was provided by Prof. Hans D. Seibel of University of Cologne, Germany.

5. ⁰ Exchange rate in 1987 US$1 = 20 pesos; in 1991 $1 = 27.3 pesos.

6. ⁰ In 1991 US$1 = Rs 40.8.

7. ⁰ The exchange rate for US$1 was 1.00 Cedi in 1977; 2.75 Cedi in 1980; 2.28 Cedi in 1987; 3.85 in 1991 ; 4.11 Cedi in 1992.

8. ⁰ Johanna Eigen: Lessons from the micro-enterprise credit programme in Kenya (Eschborn, GTZ, 1992).

9. ⁰ The exchange rate in 1990 was US$1 = Ksh 27; in 1991 it was Ksh 29.

10. ⁰ The exchange rate in 1991 was US$1 = CFA278.15.

International Labour Organization (ILO): Contact us | Site map |

Copyright and Permissions 1996-2009 International Labour Organization (ILO) - Disclaimer