1 Introduction and background 1.1 Key demographic and economic data Sri Lanka is an island of some 66,000 square kilometres, located south-east of the Indian subcontinent. The population was 18.1 million in 1995. This equates to a population density of 274 people per square kilometre, below Bangladesh (832) and India (283), but higher than for the other six countries included in this study. Between 1990 and 1995 the population grew by a modest 1.3 per cent per annum, with a total fertility rate of 2.3 births per woman in 1995. These figures suggest that Sri Lanka has proceeded further along the demographic transition than any other country in this study apart from Thailand. With a per capita income of $700 in 1995, average incomes in Sri Lanka are much higher than those of any other country in South Asia, but below those of the four Southeast Asian countries included in the study. GNP per capita increased by a reasonably healthy 2.6 per cent per year between 1985 and 1995, above the growth rate for most countries in South Asia, but slightly below India and well below the dynamic Southeast Asian economies of Thailand, Indonesia and Malaysia. The sectoral allocation of GDP also suggests that, in terms of economic development, Sri Lanka is between the other countries of South Asia on the one hand, and the Southeast Asian countries on the other. In 1995 agriculture contributed 23 per cent of GDP, the lowest in South Asia but higher than for any of the Southeast Asian countries included in the study. Industry contributed 25 per cent, while the remaining 52 per cent came from services. On the other hand only 22 per cent of the population lived in urban areas, lower than for all countries in the study except Bangladesh and Nepal. The picture regarding human development is noteworthy. Sri Lanka has long emphasised human development, and it is widely recognised that social indicators here are generally much better than for other countries with similar income levels. For instance, Sri Lanka achieved a score of 0.711 on the UNDP human development index for 1994, higher than for all countries in this study except Thailand and Malaysia. While GDP per capita was lower than in Indonesia and the Philippines, human development was higher. In 1995 Sri Lanka had the highest life expectancy at birth of any of the nine countries included in the study (72 years), the second lowest infant mortality rate (16 per 1,000 live births), and the third highest adult literacy rate (90 per cent). 1.2 Poverty Estimates of poverty Official poverty statistics in Sri Lanka have been derived using broad rules of thumb, rather than rigorously derived estimates of the poverty line, and do not necessarily provide a clear indication of the incidence of poverty. World Bank (1995) estimates suggest that the incidence of poverty in Sri Lanka is considerably lower than in any other country in South Asia but slightly higher than in three of the four countries in Southeast Asia included in this study. The official poverty line in Sri Lanka is based on monthly household income. In 1977 it was set at Rs300 ($5.40) per month per household. It remained at this level until 1989, when it was increased to Rs700 ($12.70) per month per household. Based on the official poverty line, the government estimates that the percentage of households below the poverty line has fallen from 62 per cent in 1979 to 37 per cent in 1991 and 39 per cent in 1994. The poverty line has been adjusted infrequently, and has therefore tended to become out of date between adjustments. Partly for this reason, the official poverty line has not always been used in targeting households under the government’s anti-poverty programs. For instance, in the second round of the Janasaviya program beginning in 1991, poor households were selected on the basis of a set of qualitative criteria determined at the sub-district level. The National Development Trust Fund (NDTF) defines the poor as those with a household income of less than Rs1,500 ($27.20) per month, more than double the official poverty line. The World Bank (1995) has conducted a more rigorous study of poverty in Sri Lanka. It notes that in the post-independence period and especially since 1965, Sri Lanka has made good progress in reducing poverty. Between 1965 and 1985 there was a rapid decline in poverty, propelled by rapid growth in per capita income and consumption, and a stable and egalitarian distribution of household income. Between 1985 and 1990, poverty continued to decline as a result of an improvement in the distribution of consumption. The World Bank study derived estimates of the incidence of poverty for 1985–86 and 1990–91 from socioeconomic surveys undertaken by the Department of Census and Statistics. Based on a poverty line of Rs471.20 ($8.50) per person per month, the incidence of poverty declined from 27.3 per cent of the population in 1985–86 to 22.4 per cent in 1990–91. Using 1995 population data, this equates to around 4.1 million people. The incidence of poverty was 24.4 per cent in the rural areas, 18.3 per cent in the urban areas, and 12.6 per cent in the estates (around 7 per cent of the population, predominantly Sri Lankan Tamils, live on estates in the plantation sector). These estimates are significantly lower than the official estimates. Using a higher poverty line of Rs565.44 ($10.20) per person per month, the incidence of poverty declined from 40.6 per cent in 1985–86 to 35.3 per cent in 1990–91. The data also enabled some conclusions about the distribution of poverty. Poverty was lower in the western region than in other parts of the country. Surprisingly, female-headed households did not have a significantly higher incidence of poverty than male-headed households. The incidence of poverty among Sinhalese and Sri Lankan Tamils was also about the same. However, it should be stressed that because of lack of data as a result of the civil war, the calculations exclude the north and east provinces. Given the civil war it is likely that the incidence of poverty is higher in these regions than in the rest of the country, and poverty in these regions is likely to have increased between the two surveys. Policies for poverty reduction Clearly, the level of poverty is affected by the full range of the government’s policies, including macroeconomic, investment and human development policies, as well as policies specifically directed at reducing poverty. As noted above, the government has placed a great deal of emphasis on education and health policies in particular, and there is considerable evidence that these have been very effective in reducing poverty. This discussion, however, focuses on those government policies that are specifically targeted to the poor. The central component of the government’s anti-poverty policies is the Samurdhi program, with a budget of Rs10 billion ($180 million) in 1997. The program commenced in 1995, but incorporated elements from a series of earlier government programs such as the Janasaviya program, which operated from 1989 to 1994. One component of the Samurdhi program provides direct welfare payments to an estimated 1.2 million families at the bottom of the income scale. Direct welfare payments to poor households, whether in the form of food stamps or cash, have existed since 1979 and are far more prevalent in Sri Lanka than in any other country included in this study. The poorest of the poor, around 45,000 families with incomes of around Rs500 ($9.10) per month, receive a direct income transfer of Rs1,000 ($18.10) per month. Around 760,000 poor families, with incomes of around Rs1,000 ($18.10) per month, receive an income transfer of Rs500 ($9.10) per month. Another 220,000 families who are slightly better off receive a transfer of Rs200 ($3.60) per month, while another 180,000 receive a transfer of Rs100 ($1.80) per month. In addition, government-appointed ‘animators’ establish a Samurdhi Task Force in each village. These task forces undertake five activities. First, they conduct a family profile survey to determine the number of poor families for the provision of welfare. Second, based on the information gathered, they identify community works to benefit the village and provide employment. Third, they encourage small savings and credit activities within the village (this Samurdhi microfinance program is described in more detail below). Fourth, they encourage and assist persons to undertake self-employment projects. And fifth, they implement nutrition programs for vulnerable groups. 1.3 Overview of the financial system The formal financial system in Sri Lanka includes eight domestic commercial banks and 18 foreign commercial banks, with a total of 1,060 branches between them as at December 1996. The two state commercial banks, the Bank of Ceylon and the People’s Bank, have by far the largest branch networks. There are also a number of specialised financial institutions, including the National Savings Bank, two development banking institutions and a number of non-bank financial institutions. In addition, there are 17 regional rural development banks (RRDBs). The above institutions are all supervised by the Central Bank of Sri Lanka. In the past, the two state commercial banks, the Bank of Ceylon and the People’s Bank, have lent heavily in the rural areas. While they still have extensive outreach in the rural areas, their lending to agriculture has declined in real terms in recent years as they have been forced to become more competitive. They operate a number of loan schemes designed for small entrepreneurs, and act as conduits for some of the government’s microfinance programs. Except for these government microfinance programs, however, they have not tended to reach the poor. Moreover, repayment rates on small loans have tended to be low. A more successful program is the small microfinance program of a private commercial bank, Hatton National Bank, discussed in more detail below. The RRDBs were established by the central bank in 1985 as specialist institutions for lending for rural development. With some exceptions, they have tended to lend primarily to borrowers in the rural areas who are better off. Some commentators suggested that they have been subject to political interference, and their repayment performance has been poor. The government has recently passed legislation to amalgamate the existing RRDBs into larger rural development banks at the provincial level. It is understood that it has also given some consideration to floating shares in some or all of the new rural development banks, to expand their capital base and enable them to operate more independently. Additional to the banks discussed above are the cooperative rural banks (CRBs), which are supervised by the Department of Cooperative Development. They have extensive branch networks, with a total of some 1,250 branches and 4 million depositors at end-1995. These are operated by the state-dominated multi-purpose cooperative societies, and are quite separate from the thrift and credit cooperative movement. Their main activity is to mobilise savings, which are deposited primarily with the People’s Bank and used in its lending operations. Their own lending accounts for only around 40 per cent of their deposits, and around half of this is related to their pawnbroking operations. Moreover, it is understood that they lend mainly to people in rural areas who are better off. Hence, CRBs provide only very limited microfinance services. Since 1990, there has been considerable deregulation of the banking system under a series of reforms sponsored by the World Bank and the International Monetary Fund. While in the past the central bank was heavily involved in refinancing loans by commercial banks to priority sectors, this has been withdrawn. There has also been a series of reforms aimed at strengthening regulation and supervision of banks, and bringing banking supervision into line with international practices. Areas that have been strengthened include classification and provisioning of non-performing loans and advances, capital adequacy requirements, and accounting and auditing standards. Most exchange controls have also been removed. Surprisingly, macroeconomic data suggest a lack of depth in the banking system, with a ratio of money and quasi money to GDP of only 31.8 per cent in 1995. This is lower than for any other country in the study for which data are available, including much poorer countries such as Bangladesh and Nepal. While the reasons for this are not clear, it may reflect a reliance on thrift and credit cooperative societies instead of banks. Unlike bank deposits, deposits with thrift and credit cooperative societies are not included in the definition of quasi money. There are no restrictions applying to interest rates for banks or other financial institutions, either for deposits or loans. According to the World Bank’s World Development Report for 1997, in 1995 bank interest rates for both deposits (16.1 per cent) and loans (14.7 per cent) were considerably higher than the inflation rate (7.7 per cent). While these data suggest that interest rates on deposits were higher than those applying to loans, data from the central bank suggest that the prime lending rate charged by commercial banks was actually 20.1 per cent. There are no longer any requirements on banks to lend to particular priority sectors of the economy. However, there is an incentive for them to lend for agriculture, with the government paying an interest subsidy of 7.5 per cent on short-term crop lending. To be eligible for the subsidy, banks must lend at an interest rate of 16 per cent. This subsidy was to be reviewed at the end of 1997. 1.4 Overview of microfinance Since 1989, microfinance has been a central element of the government’s poverty reduction programs. The Presidential Commission on Finance and Banking (1992) considered the appropriate policy for credit to the poor. It concluded that: . . . a pluralistic approach which provides scope for the involvement of a wide array of financial institutions, cooperatives, NGOs and a range of governmental and informal agencies, each employing different techniques and strategies based on their different credit cultures, together with market based interest rates and adequate support services, would be the appropriate policy to adopt in order to improve the access of the rural population to formal credit institutions (p.122). The Janasaviya program, which operated from 1989 to 1994, included a microcredit scheme where loans were disbursed to poor beneficiaries for income-generating projects. Loans were made on the referral of village-level support teams, and were disbursed through the state commercial banks, RRDBs and CRBs. In addition, the Janasaviya Trust Fund (JTF) was established in 1991, one of its main functions being to manage a credit fund for disbursing credit to partner organisations, for on-lending to beneficiaries for income-generating self-employment projects. The JTF, now known as the National Development Trust Fund (NDTF), is still in operation. However, recent attention has shifted to the Samurdhi program, which replaced the Janasaviya program in 1995. Under the Samurdhi program, village task forces select poor borrowers to receive loans from a fund managed by the two state banks. As noted above and with some exceptions, the commercial banks and RRDBs do not engage in microfinance to any significant extent except as a conduit for government programs. While the CRBs are active in mobilising savings in the rural areas, they do not provide microfinance loans to any significant extent. Other than the direct government programs, the major institutions engaged in microfinance in Sri Lanka are the thrift and credit cooperative societies (TCCSs), and a small number of non-governmental organisations (NGOs). TCCSs operate throughout Sri Lanka, although their coverage and strength varies between regions. As at end-1996 there were 8,340 primary societies with a total membership of 772,000. Given that the total population is only around 18.1 million, this clearly represents a significant proportion of households in Sri Lanka. Around 55 per cent of members are women. The societies collect savings from their members, and provide loans for income-generating activities and consumption. The credit cooperative movement in Sri Lanka has been far more successful than in most other countries in the region. For instance, Hulme, Montgomery and Bhattacharya (1996, p.177) argue that: Cooperatives in developing countries have a weak record for helping poor people to improve their living standards and for assisting microentrepreneurs. The experience of thrift and credit cooperative societies in Sri Lanka stands in marked contrast to this general situation and supports the case for a reconsideration of the role of credit cooperatives in rural finance. Since 1978 the thrift and credit movement has pursued a strategy of rapidly expanding its membership and activities, and of shifting from an orientation towards middle income households to a greater focus on poorer rural people. Membership of the TCCSs is open to anyone and is not confined to the poor, and it is understood that most members are not poor. Nevertheless, the movement has served as a vehicle for various targeted poverty lending programs operated by the government and donor agencies. Moreover, it has also had its own projects for lending a portion of its funds to low-income groups. There are also a number of NGOs involved in microfinance. In 1986 Sarvodaya, the largest NGO in Sri Lanka, established a microfinance division called Sarvodaya Economic Enterprises Development Services (SEEDS). It provides microfinance services through some 2,000 independent primary societies, with a combined membership of around 150,000 households. Other well-regarded NGOs involved in microfinance include Women’s Development Foundation, People’s Rural Development Association and Janashakti Banku Sangam. It should be noted that a large number of credit NGOs have sprung up in recent years as a result of government programs such as NDTF, but it is understood that most of these are not operating on a very sound basis. One innovative microfinance program is the Gami Pubuduwa Upadeshaka (GPU) scheme operated by Hatton National Bank, a private commercial bank. The program was introduced in 1989, and is aimed at rural and semi-urban unemployed youth. It is based on the concept of ‘barefoot banking’. Essentially, the GPU officers function as a linkage between the bank and the rural community, with most banking transactions carried out in the village, rather than at a bank branch. At present there are around 7,000 borrowers under the program, and the program has also been very successful in terms of savings mobilisation. In addition, the bank has a linkage program for lending to the poor, involving around fifteen NGOs. The bank’s microfinance program has recently been considered in detail by the World Bank (Gallardo, Randhawa & Sacay 1997). One concern expressed about microfinance in Sri Lanka is the lack of targeting. The TCCSs and most of the NGOs discussed above do not specifically target the poor, and no data on the number of borrowers who are poor are available. The major government programs, such as NDTF, the Small Farmers and Landless Credit (SFLC) project and the Samurdhi program are, in principle, targeted to the poor. Where TCCSs and NGOs have received funds under NDTF and the SFLC project, they have been required to establish targeted cells within their overall lending programs. However, it is understood that the targeting guidelines in the government programs have often not been followed by either NGOs or government agencies, and there has been little if any monitoring of the extent to which these projects have reached the poor. The government microfinance programs, most particularly the Samurdhi program, do not operate on a sustainable basis. Government has provided the loan capital, and this is lent through the state commercial banks to final borrowers at highly concessional interest rates (for example, 10 per cent under the Samurdhi program). Moreover, administrative costs associated with the village task forces are borne by the budget. On the other hand, the TCCSs and some of the major NGOs, such as SEEDS and the People’s Rural Development Association, are more concerned about sustainability. While they still rely on donor funds to a considerable extent, these programs meet most of their operational costs from interest earnings. Few data on the smaller NGOs are available, but it is understood that they rely much more heavily on donor funds and cheap loan capital, and are not operating on a sustainable basis. 2 Arrangements for direct support 2.1 Support for specialised microfinance institutions National Development Trust Fund In the past, the government and World Bank have provided extensive support for microfinance through the National Development Trust Fund (NDTF), previously known as the Janasaviya Trust Fund (JTF). In addition, many donor agencies have provided direct support to particular microfinance institutions (MFIs). NDTF was established in 1991 as a second tier microfinance institution, similar in functions to the Palli Karma Sahayak Foundation (PKSF) in Bangladesh and second tier institutions in other countries. It was incorporated under the Trusts Ordinance, and is governed by a 20-member board of trustees, selected by the government and comprising around half public-sector and half private-sector representatives. Its main function is to manage a credit fund for disbursing credit to partner organisations, whereby funds are channelled through various microfinance institutions — including NGOs, RRDBs, commercial banks and cooperative rural banks — to final borrowers for income-generating self-employment projects. In addition, it manages a human resources development fund for providing skills training to beneficiaries, a rural works fund, and a nutrition fund. A credit fund of $35 million was established, with support from the World Bank ($20 million), the government of the Federal Republic of Germany ($10 million) and the government of Sri Lanka ($5 million), and NDTF commenced operations in 1992. By March 1997, it had provided funds totalling Rs1.188 billion ($21.5 million) to 157 partner organisations, including commercial banks, RRDBs, thrift and credit cooperative societies (TCCSs) and NGOs. Of these, around 130 are currently active. NDTF resources are available to partner organisations at an interest rate of 7 per cent, and are on-lent to final borrowers at a minimum interest rate of 19 per cent. Interest rates on loans to final borrowers range between 19 and 36 per cent. NDTF also provides assistance to partner organisations for institutional development. For instance, it provides funds to small partner organisations for the purchase of equipment. It also employs a number of accounting technicians who are available for secondment to partner organisations to develop their financial management capacity. Nevertheless, a World Bank mission in November 1996 found that NDTF had not focused sufficiently on capacity building among its partner organisations. NDTF has achieved a solid repayment rate of 90 to 95 per cent on its loans to partner organisations. However, while in some respects the operation of the fund has been reasonable there have been a number of problems, to the extent that the mid-term project review by the World Bank in May 1995 classified the project as a ‘problem’ project. Difficulties included the following:
These failures appear to reflect a number of factors. Some commentators stated that NDTF was under pressure to disburse funds too quickly, and that there were not enough sound MFIs available to use as partner organisations. It would also appear that it did not have sufficient staff with the technical skills to properly implement the program. Finally, some commentators pointed to politicisation of the organisation. It was reported that the problems experienced under NDTF have done considerable damage to the NGO movement as a whole, and have led to doubts within the government about the capacity of NGOs to engage in microfinance. More recently, some measures have been implemented to try to overcome these problems, and there appears to have been a marked improvement in the performance of NDTF. It has now adopted more detailed guidelines for the selection of partner organisations, and is taking steps to ensure that these are implemented more rigorously than in the past. A World Bank mission in November 1996 found that the main recommendations of the mid-term project review had been satisfactorily implemented, and advised that the project’s problem status should be lifted with immediate effect. Nevertheless, uncertainty as to the future role of NDTF continues to affect its performance. For instance, the World Bank mission found that staff were demotivated because the government has not given clear signals about NDTF’s future. Hence, staff turnover was high and it was difficult to recruit good professionals. The project was originally scheduled to terminate at the end of 1996, but the closing date was extended to the end of 1997. It is intended that there will be a second World Bank project after the first project terminates, and the board of NDTF has recently prepared a concept paper with some ideas for a project design. However, the final shape of any second project is not clear at this stage. Second tier microfinance institutions such as NDTF offer a number of advantages over the traditional arrangements, whereby donor agencies and governments have provided largely ad hoc support to individual MFIs. They can ensure that different forms of support to a particular MFI are complementary and directed to the same objectives. They can ensure that all MFIs face a level playing field, with support available to all MFIs meeting predetermined standards. And they can require that MFIs face one set of performance and reporting standards, rather than a number of different standards or no standards at all. By lending to MFIs, such institutions also harness the strengths of non-government microfinance programs, which have generally been much more successful than programs involving direct lending by government agencies or commercial banks. It would be appropriate for the government to renew its commitment to NDTF as the primary vehicle for channelling direct support to microfinance in Sri Lanka. This in itself would improve the operating environment for the fund and make it easier to recruit and retain high-quality staff. NDTF also needs to continue to improve its performance. One critical issue is to ensure that it is not politicised. In practice, this may be difficult to achieve. However, it may be possible to secure greater autonomy by altering the trust deed to:
It will also be necessary for the fund to set and, more importantly, to enforce appropriate performance and reporting standards for the MFIs that it funds. This will mean striking an appropriate balance between the need to increase outreach and the need to strengthen the capabilities of existing MFIs. On the one hand, given that there is currently only a relatively small number of strong MFIs, it may not be possible to impose standards as rigorous, for instance, as PKSF is able to impose in Bangladesh. On the other hand, it would clearly be counter-productive to provide large loans to small and inexperienced MFIs that do not have the institutional capacity to handle them. It would be appropriate to require all partner organisations to meet a set of minimum standards, and to increase these gradually over time. NDTF is already moving in this direction. 2.2 Support through the banking system Directed credit schemes As noted above, there are no longer any requirements on banks to lend to particular priority sectors of the economy. Sri Lanka is slightly unusual in this regard, with Sri Lanka and Bangladesh the only two countries in this study to have abolished all such directed credit schemes. Small Farmers and Landless Credit Project There are, however, two major microfinance programs which are channelled through the banking system. One is the Small Farmers and Landless Credit (SFLC) project. This project is executed by the central bank, with the funds channelled through RRDBs to NGOs and other partner organisations. It is funded by the International Fund for Agricultural Development (IFAD), $6.7 million; the Canadian International Development Agency (CIDA), $6.6 million; and the government of Sri Lanka ($4.1 million). The project has operated in four districts since 1990. A key objective of the project is to establish a cost-effective and sustainable microcredit delivery system for the poorest of the poor. The project has a tighter poverty focus than other programs in Sri Lanka, with borrowers required to have an annual per capita income of less than Rs5600 ($101), the income necessary to obtain at least 2240 calories per day. This requirement has meant that participating agencies which do not normally means-test their clients, such as the TCCSs and Sarvodaya, have had to establish means-tested components within their broader operations. Another objective is to inculcate savings habits and thrift among the rural poor. Loans are granted only after clients have demonstrated regularity in savings for a period of three to six months. The central bank lends to the RRDBs at an interest rate of 3 per cent, and they in turn on-lend to partner organisations at 9 to 11 per cent. Partner organisations on-lend to final borrowers at interest rates of between 16 and 20 per cent. A number of NGOs, including the district TCCSs and Sarvodaya, have been involved in the project. They are required to lend using a Grameen-style methodology, although some flexibility is allowed. By end-December 1996, 35,480 loans had been disbursed under the project, with an average loan size of Rs8,560 ($155) and a recovery rate of 91 per cent. Some 65 per cent of borrowers were women. The project appears to have been reasonably successful, and more successful than most government programs in other countries channelled through the banking system. This may reflect the involvement in many cases of NGOs as conduits between the RRDBs and the final borrowers. However, relatively few data are available, and the project does not appear to have been evaluated rigorously. The project was scheduled to be completed in December 1997, at which time the central bank would hand over the revolving funds to the participating RRDBs for them to continue the project. The government has applied to IFAD for a second loan, and initial indications are reported to be favourable. Until now, the project has been executed by the central bank. As with other projects where funds are channelled through the central bank, the central bank lends only to banks under its supervision, in this case RRDBs. It does not lend directly to NGOs or other institutions. Indeed, this is one factor motivating the federation of TCCSs to apply for a banking licence, to enable it to borrow directly from the central bank under such projects. Where RRDBs on-lend to NGOs, this restriction appears to impose an unnecessary layer between the central bank and the final borrower. To the extent that the central bank continues to be involved in this and other microfinance projects, it would be appropriate for it to lend directly to NGOs rather than through RRDBs. In any case, the central bank was scheduled to withdraw from the project at the end of 1997 and hand it over to individual RRDBs. This is perhaps symptomatic of a change in the attitude of the central bank towards microfinance. In the past, the central bank has played an important role in rural credit and microfinance by providing refinancing, managing funds such as under the SFLC project, and providing technical support to RRDBs and other MFIs. However, some commentators suggested that the central bank was withdrawing from this kind of development work and concentrating on more traditional central banking functions. It is not clear that the RRDBs are well placed to take over the project. As noted above, their repayment performance has generally been poor, and some commentators suggested that they have been subject to political interference. Moreover, they are currently being restructured, with the government recently passing legislation to amalgamate the existing RRDBs into larger rural development banks at the provincial level. As such, it is not clear that they have the capacity to undertake effective supervision and monitoring of either loans to final borrowers or loans to NGOs for on-lending. With regard to loans to NGOs, the RRDBs are in any case acting as quasi-second tier institutions and duplicating the role of NDTF. One option would be to consolidate these functions with NDTF, which could continue to lend to both RRDBs and specialist MFIs for on-lending to final borrowers. It may perhaps be appropriate for the government to consider transferring the central bank’s loan portfolio under the project to NDTF at the end of the project period. Samurdhi Development Credit Scheme The government has introduced two major microfinance programs in recent years:
Both schemes commenced operations in 1996. In the case of the Samurdhi program, the government provided Rs500 million ($9.1 million) to the two participating banks for 1997. The program is intended to operate in 12,141 villages. Each village has a Samurdhi task force, and will be allocated credit of Rs41,500 ($750). Additional funds come from the compulsory savings from welfare payments provided to poor households under another arm of the Samurdhi program. The task force selects people to receive loans of between Rs2,500 ($45.30) and Rs10,000 ($181.20). The loans are then released by the relevant bank. The interest rate is 10 per cent, repayable over 6 to 24 months. The program is targeted to Samurdhi beneficiaries, although it was suggested that in some cases people who are better off have benefited. Village task forces are required to prepare monthly reports on the progress of the schemes. These are provided to the divisions, districts, and ultimately to the head office of the ministry. Because of the newness of the scheme, no data are yet available on repayment rates or other aspects of performance. Nevertheless, there are some causes for concern in the design of the scheme. The scheme involves state commercial banks lending directly to individual borrowers on the recommendation of village task forces. Commercial banks are generally not geared up to lend directly to poor borrowers in a cost-effective manner. Moreover, the funds for the scheme have been provided to the banks by government, reducing the incentive for the banks to ensure high repayment rates. The village task forces are effectively government-sponsored agencies, albeit at the local level, also with a limited incentive to ensure high repayment rates. The scheme does not envisage any explicit role for NGOs or other specialist MFIs, either in identifying and motivating borrowers or as direct financial intermediaries. Community-based NGOs generally have a comparative advantage over commercial banks and government agencies in reaching the poor, reflecting factors such as proximity, trust, commitment, flexibility and responsiveness. While the reasons for this lack of involvement by NGOs are not entirely clear, it was suggested that not many NGOs operate in the poorer and more remote villages, and that some of the problems experienced under NDTF have contributed to a suspicion of NGOs in official circles. A number of commentators suggested that, given its design, it is unlikely that the scheme will achieve high repayment rates. Moreover, the scheme does not appear to use the most cost-effective lending technologies for reaching poor borrowers. It would appear appropriate to provide a role for NGOs and other specialist MFIs, either in identifying and motivating borrowers or as direct financial intermediaries. One alternative would simply be to merge the scheme into NDTF, subject to measures being taken to ensure that NDTF continues to improve its performance. Moreover, the scheme is not designed to be sustainable. Loanable funds have been provided to the participating banks from the budget. The banks are charging an interest rate of 10 per cent, which would clearly not be sufficient for them to operate the scheme on a commercial basis if they were using their own funds. The cost of administering the village task forces is also being met by the government. Clearly, outreach could be increased significantly, for the same resource cost, if the scheme was redesigned to operate on a more self-sufficient basis. There is also a danger that the availability of cheap credit will crowd out existing NGOs and other specialist MFIs, which do not have access to large subsidies and generally lend at interest rates between 19 and 36 per cent. Regardless of any other changes that might be desirable, it would be appropriate to redesign the scheme to require it to be more self-sufficient. Linkages between banks and specialised MFIs As noted above, one highly innovative program in Sri Lanka is the microfinance program operated by Hatton National Bank. The bank involves around fifteen NGOs in the program as a means of reducing costs. In some cases the NGOs operate as non-financial intermediaries, identifying and training borrowers, but the loan contract remains between the bank and the final borrower. In other cases, NGOs deposit members’ savings with the bank, and the bank extends a loan to the NGO based on a multiple of the savings. The bank is careful in selecting NGO partners, looking in particular at their management, field structure, the sustainability of their operations and their legal status, but there are no written guidelines about the selection of NGOs. While there do not appear to be any regulatory impediments to other commercial banks establishing such programs, they have been very reluctant to do so. It would therefore be appropriate for the central bank to take a more active role in encouraging linkages, by documenting and publicising the experience of Hatton National Bank, and issuing clear guidelines that other banks could follow. In this regard, the guidelines issued by the National Bank for Agriculture and Rural Development (NABARD) in India may be a useful starting point (see the India country study), although they would need to be adapted to suit the institutional framework in Sri Lanka. 2.3 Other issues Some commentators noted that the government had waived loan repayments for small loans by state commercial banks and under some of its own microfinance programs from time to time, and that this had negatively affected the environment for microfinance. The most recent debt forgiveness took place in 1995. The government should avoid debt forgiveness for small loans in the future. 3 Regulation of non-bank microfinance institutions 3.1 The broad regulatory framework Thrift and credit cooperative societies The operations of thrift and credit cooperative societies (TCCSs), along with other cooperatives, are governed by the Cooperative Societies Law of 1972. Cooperatives are required to be registered with the Registrar of Cooperative Societies. To be registered, a society must have as its object the promotion of the economic, social or cultural interests of its members in accordance with cooperative principles. It must have at least ten members owning shares with a value of at least Rs100 ($1.80) each, and must have its own by-laws (to a large extent, these are based on a standard model). Moreover, officers from the Department of Cooperative Development supervise the operations of a pre-cooperative for around six months before it is registered, to ensure its economic and organisational viability. Every society is required to have a monthly meeting of the membership. It is also required to have an annual general meeting, and to table an annual report and financial statements. The executive committee is elected from the general body at the annual meeting. Societies are exempt from stamp duties but are liable for all other taxes, including company income tax. The federation of TCCSs commented that it found the provisions of the Cooperative Societies Law of 1972 and associated regulations restrictive in a number of ways. In particular, the law gives considerable discretion to the departmental minister and to the Registrar of Cooperative Societies, and requires their approval for numerous activities undertaken by the societies. For instance, various staffing decisions, relatively small capital purchases, and even large loans to borrowers require approval from the registrar. The federation stated that it found these requirements cumbersome and inefficient. It would be appropriate to review the regulatory requirements facing cooperatives, with a view to streamlining them and removing unnecessary requirements. The law specifically permits TCCSs to receive deposits from persons who are not members. Many societies are now actively seeking deposits from non-members by offering competitive interest rates, and deposits by non-members account for around 15 per cent of total deposits. Given the lack of prudential requirements, it is perhaps surprising that societies should be permitted to receive deposits from non-members. Indeed, the federation itself commented that the lack of prudential requirements reduces the credibility of societies as deposit-taking institutions, making it difficult for them to attract deposits on a large scale. This is a major reason for the federation’s proposal to establish a bank, which would be subject to more rigorous prudential requirements and would be able to attract a greater volume of deposits. It does not seem appropriate for institutions other than banks, such as TCCSs, to accept deposits from the general public in their own right. When the federation establishes a bank, however, there is no reason why the primary societies should not serve as agents, accepting deposits and offering withdrawals on behalf of the bank. Non-governmental organisations Non-governmental organisations (NGOs) engaged in microfinance are subject to a different regulatory regime to cooperatives. There are three avenues for registration. First, some large NGOs, such as Sarvodaya and Lanka Mahila Samiti, have been incorporated under their own Acts of Parliament. This is probably the most complicated method. The NGO will need to arrange for a bill to be introduced into Parliament (generally as a private member’s bill), and publish notices in the government gazette and in the newspapers in case anyone objects to their establishment. While the requirements may vary from one Act to another, such NGOs generally need to have a constitution, and to produce an annual report and audited financial statements. Second, some NGOs are registered as non-profit companies under the Companies Act of 1982. The registration procedure is essentially the same as for companies, except that there is no minimum capital requirement. Non-profit companies are required to have a memorandum and articles of association covering matters such as their objectives, governance and management structure. They are also required to prepare an annual report, to have their annual accounts audited, and to lodge these with the Registrar of Companies. However, it is understood that not all non-profit companies meet these conditions, and that the conditions are not systematically enforced. Third, some NGOs are registered under the Societies Ordinance of 1891. This is the simplest method, and has been adopted by most of the smaller NGOs. Broadly, the requirements are similar to those for non-profit companies. Registrations are approved by the Registrar of Companies. To be registered, a society must have at least seven members and a subscribed capital of at least Rs10,000 ($180), and must provide two copies of its rules to the registrar. The rules must cover the objects of the society, the operation of the management committee, internal governance and the keeping of accounts. In addition, societies are required to provide an annual report to the registrar, and to have the annual accounts audited. Again, however, these requirements are not strictly enforced. Even though these various provisions for registration are not particularly onerous, it is possible for an NGO to operate without being registered. However, unregistered NGOs may have difficulty in attracting external funds. International NGOs operating in Sri Lanka do not need to be registered. They generally sign a memorandum of understanding with the Ministry of Policy Planning and Implementation (as at May 1997, 53 international NGOs had signed such a memorandum), or at least an agreement with the Director of External Resources, before they commence operations. However, there is no legal requirement for them to do so. NGOs are also allowed to receive funds from foreign donors without any restrictions, control or supervision by the government. Hence, while most NGOs involved in microfinance are registered, it is quite possible to operate as a microfinance institution (MFI) without any legal backing whatsoever. It would appear that this unstructured regime has led to some problems, with some NGOs reportedly engaging in activities of a controversial nature. On the other hand, the environment for NGOs has at times been highly politicised, with NGOs subject to considerable uncertainty. In the early 1990s, a presidential commission was established to examine and report on alleged irregularities in the operations of some NGOs. However, to date, there have not been any changes to the registration procedures. It would be appropriate to tighten up the procedures, to require NGOs to be registered and to monitor their activities to ensure they are properly accountable to their members and third parties. At the same time, NGOs meeting the necessary requirements should be permitted to operate with maximum flexibility and certainty. According to the law, NGOs are not allowed to provide savings facilities. Under the Banking Act of 1988 and the Finance Companies Act of 1988, an institution must be licensed as a bank or finance company in order to collect deposits. This applies to all deposits, whether from members or non-members, and includes compulsory savings that may be a pre-condition for obtaining a microfinance loan. Any such savings must be invested in a bank account, and cannot be used for re-lending by the NGO. (As noted above, these provisions do not apply to cooperatives, which are specifically permitted under the Cooperative Societies Law of 1972 to raise deposits. Moreover, the Samurdhi task forces are specifically exempted from the restriction on mobilising savings by virtue of the Samurdhi Authority Act.) In practice, the prohibition on mobilising savings has generally not been applied strictly. Indeed, major government programs such as the National Development Trust Fund (NDTF) and the Small Farmers and Landless Credit (SFLC) project have required NGOs borrowing from them to establish savings programs, and have encouraged them to use these savings in their credit programs. On the other hand, in 1992 the largest NGO, Sarvodaya, was prosecuted for mobilising savings and required to return all savings to members. It now mobilises savings only through independent primary societies, and these are not remitted to Sarvodaya itself. Amendments to the Finance Companies Act that would give NGOs more flexibility in accessing savings have been mooted, but have not been introduced. Clearly, the current situation leads to uncertainty, and if the law were interpreted strictly, would severely impede if not effectively prohibit NGOs from engaging in microfinance. Savings represent a stable source of funds for MFIs in many Asian countries. Moreover, saving can be important in ensuring borrower discipline, and ‘savings first, credit later’ has been the working guideline of many successful MFIs. Therefore, it is important that MFIs should at least be permitted to accept members’ savings. There is a need to amend the law to give NGOs greater certainty in this area. 3.2 Interest rates There are few regulations governing interest rates. In the case of the TCCSs, societies are free to set their own interest rates for both deposits and loans. However, where societies receive outside funds for on-lending to members, they are not permitted to charge a margin of more than 4 per cent. While this may seem restrictive, the margin has been set in consultation with the federation, and the federation commented that it is sufficient to enable societies to cover their transaction costs. There are no restrictions on the interest rates that NGOs engaged in microfinance can charge. 3.3 Prudential regulation and supervision There is virtually no prudential regulation and supervision of MFIs in Sri Lanka, whether TCCSs or NGOs. In the case of TCCSs, the registrar advises them that they should maintain a minimum of 15 per cent of their deposits in the form of liquid assets, but this does not appear to be binding and is not enforced. Other than this, there are no capital adequacy requirements, reserve requirements, or requirements relating to loan loss provisioning, loan documentation or collateral applying to either TCCSs or NGOs. In terms of reporting and supervision, TCCSs are required to prepare an annual statement of accounts, including a statement of income and expenditures and balance sheet, and to submit this to the registrar within three months of the end of the financial year. They are also required to prepare an annual budget, setting out projections of revenue and expenditure, and provide it to the registrar. In addition, officers from the Department of Cooperative Development visit societies from time to time and carry out inspections. The department is required to conduct an annual audit of every society, although it is understood that the department is constrained by a lack of resources, and that such audits are not always timely nor very thorough. In the case of NGOs, there are no requirements for the maintenance or provision of portfolio or other accounting data, beyond the general requirements concerning annual accounts that apply where NGOs are registered. These do not apply to all NGOs, and are not strictly enforced even where they do apply. There does not appear to be a strong case for requiring a government agency to monitor and supervise TCCSs on a day-to-day basis. The performance of the TCCSs has generally been good. It is understood that several societies become dormant each year as a result of inadequate financial management, but this is only a very small proportion of the more than 8,000 societies in operation. Given the large number of societies, it would require a considerable input of resources for the Department of Cooperative Development or another government agency to monitor and supervise them in a rigorous manner. Moreover, any such effort would largely duplicate the monitoring and supervision already undertaken by the district unions and the federation. It may therefore be more cost-effective for the Department of Cooperative Development to monitor the activities of the federation and perhaps conduct spot checks on individual primary societies, rather than establishing a parallel system for monitoring all primary societies. The situation regarding NGOs involved in microfinance is somewhat different. There are far fewer NGOs than there are TCCSs, with maybe 200 in total. There is also no apex body responsible for monitoring or supervising them. And, as noted above, many of them have not operated on a very sound basis. Hence, it would appear that some more intensive monitoring and supervision is necessary. On the other hand, it is generally accepted that it is unnecessary and inappropriate to subject MFIs which do not accept deposits from the general public to full prudential regulation and supervision. It would therefore be appropriate to consider other approaches to ensuring that MFIs maintain high standards in their operations. 3.4 Performance and reporting standards of the National Development Trust Fund In some countries, de facto standards for self-regulation are being established based on the policies of second tier institutions which lend to MFIs. Where such second tier institutions lay down standards for MFIs wishing to borrow from them, there is a major incentive for MFIs to meet them, and a tendency for them to become industry standards. It would appear that NDTF has the potential to play a similar role. NDTF has laid down operational procedures for partner organisations which borrow from it. According to the guidelines, an organisation must have legal status, and must have at least two years’ experience in lending to the poor, with a recovery rate of at least 90 per cent over the last year and at least 70 per cent over the last two years, to qualify as a partner organisation. In addition, it must meet the following criteria:
In the past, these conditions were not enforced. Many partner organisations with little or no previous experience in microfinance were given loans. An external audit of partner organisations completed in August 1996 found that a significant number had not maintained proper accounting records and did not follow generally accepted accounting principles. More than 60 per cent did not have a recovery rate of 90 per cent or above (although as noted previously, the repayment rate for partner organisations to NDTF has remained solid), and a fair number did not mobilise savings. It is understood, however, that since 1995 the requirements in terms of experience and repayment rates have been enforced more rigorously. NDTF has also introduced a number of more specific prudential requirements. In particular, partner organisations are required to:
There are also a number of criteria for the selection of borrowers. They must have a household income of less than Rs1,500 ($27) a month, with priority given to women and people under 35. They must participate in social mobilisation and training activities for a minimum of three months before obtaining the loan, and be organised in mutual support groups. They must participate in an individual savings program, and should not have defaulted on previous loans. Moreover, sub-loans are only to be given for income-generating activities, with a maximum of Rs10,000 ($181) for the initial loan. NDTF does not generally monitor these conditions. Some commentators suggested that many loans were, in fact, going to borrowers with higher incomes. In terms of financial reporting, NDTF requires partner organisations to have their accounts audited by a qualified auditor, and partners are eligible to have the cost of the audit reimbursed by NDTF. Nevertheless, not all partners have had their accounts audited. In 1993 less than half had their accounts audited, but by 1995 the proportion with audited accounts had risen to around 80 per cent. NDTF has decided that from 1997 only those with audited accounts will be eligible to receive loans. Partner organisations are also required to provide a quarterly report to NDTF. The report covers loan balances and arrears with respect to loans from NDTF to the partner, and statistics on the group contingency fund. In addition, partners are required to provide single line data on the number of members, value of members’ shares and savings, other savings, value of loans from savings, number of loans, average loan size and cumulative recovery rate. While these data are not particularly onerous, it is understood that many partners do not provide the necessary data, and need to be reminded to do so. Hence, while NDTF has laid down detailed operational procedures for partner organisations borrowing from it, in the past most of these procedures have not been enforced and many, if not most, partner organisations have failed to meet them. It may be possible to develop the standards adopted by NDTF, with appropriate modifications and in consultation with MFIs, into more general standards for MFIs operating in Sri Lanka. For this to happen, however, it would be necessary for the standards to be enforced more rigorously than they have been in the past. Moreover, there is currently considerable uncertainty as to the future role of NDTF, and a need for it to improve its performance further. These issues would need to resolved before it could operate as an effective apex institution for MFIs in Sri Lanka. 3.5 Self-regulation Federation of Thrift and Credit Cooperative Societies The Federation of Thrift and Credit Cooperative Societies is the apex body for the credit cooperative movement. Below the federation are 33 district unions which in turn are responsible for some 8,340 primary societies. The apex body provides a range of services to the movement. It provides training to staff from the district unions and more specialised training to staff from some of the larger primary societies. The district unions in turn provide training and support to the primary societies. The apex body also plays a coordination role, and acts as a focal point for policy dialogue and discussions with donor agencies. It also coordinates guidelines and procedures for the operation of primary societies. For instance, it has introduced standard loan documentation and other forms for all primary societies. Primary societies deposit excess funds with district unions, which are then able to give loans to other primary societies. District unions also deposit funds with the federation, which makes loans to other district unions. The federation also acts as a conduit for funds from other sources, such as government programs and donor agencies. Individual district unions and primary societies are free to set their own interest rates for deposits and loans. The district unions supervise the activities of the primary societies. The primary societies are required to provide monthly financial statements covering portfolio data, income and expenses, and balance sheet data to the district union. Officers from the district union monitor this information and conduct visits to the primary societies. If primary societies do not provide the necessary information, or if their statements reveal problems, they are not able to obtain loans from the district union. Hence, there is a well-developed structure for coordination and self-regulation, with the federation and district unions providing a range of services to the primary societies. These structures have contributed importantly to the overall strength of the movement, which is much more successful than in most other countries in the region. It may be appropriate for other countries to study the operations of the credit cooperative movement in Sri Lanka, to see what lessons there may be for their own operations. Forum of microfinance NGOs Networks of MFIs have an important role in a number of areas, including information exchange, training, research, policy dialogue with government and donor agencies, and establishing standards for self-regulation. An informal forum for microfinance NGOs has recently been established under the UNDP Civil Society Organisations Development project. The forum follows and builds on previous efforts to establish an apex body for microfinance NGOs. An earlier grouping was convened by the federation of TCCSs following the Sri Lanka national workshop on Banking with the Poor, organised by The Foundation for Development Cooperation in January 1994, and met on several occasions during that year. The current forum was established in early 1995, and is coordinated by UNDP. Originally, it met once a month, but it now meets once every two months. There are around 55 members, with about 35 attending meetings regularly. Around ten to twelve of the participants are local NGOs, including the major NGOs such as the federation of TCCSs and Sarvodaya. Other participants include donor agencies, the central bank, the Bank of Ceylon and international NGOs working in Sri Lanka. It is understood that a sub-group consisting solely of NGOs has also been formed under the auspices of the Sri Lanka Business Development Centre, and has been meeting since late 1996. The forum has three main activities. First, it is collecting and disseminating information on training in credit and savings operations. At this stage, it is not planning to conduct its own training courses. Second, it is coordinating research and documentation on microfinance. So far, it has mainly been gathering information, but it has developed draft terms of reference for a study to evaluate the relative strengths and best practice of microfinance programs in Sri Lanka. To date, it has not obtained funding to carry out this study. Third, it provides a focal point for liaison and policy dialogue with government bodies. The forum has held discussions with government about restrictions on MFIs in collecting savings from members, and about problems caused by debt write-offs under government programs. It has not yet looked at issues related to self-regulation of the microfinance sector. Nevertheless, the proposed study of best practice microfinance in Sri Lanka will provide a basis for starting to consider such issues. The forum is a very positive initiative, with the potential to play a significant role in areas such as training, coordination, policy dialogue and establishing standards for self-regulation. It is important that it continues to develop and expand its membership and range of activities. 4 Regulation of banks 4.1 Licensing and minimum capital requirements As noted above, MFIs other than cooperatives and Samurdhi task forces are not permitted to raise deposits from the general public. Indeed, strictly speaking, most MFIs are not even allowed to accept savings from their own members. Generally, only institutions licensed as banks can accept deposits from the general public. The minimum capital requirement for a commercial bank is Rs500 million ($9.1 million). This is likely to be too high for MFIs wishing to establish banks, or bodies wishing to establish small banks operating at the local level. Following amendments to the Banking Act of 1995, it would be easier for such institutions to seek licensing as a specialised bank. The minimum capital requirement for specialised banks is only Rs100 million ($1.8 million). On the other hand, they are not permitted to accept demand deposits (they can accept savings and time deposits only), and cannot offer chequeing facilities. While there is a network of regional rural development banks (RRDBs) operating in the rural areas, these are owned by the government and there would not appear to be any scope for private investors to use this institutional form to establish a new bank. In any case, the government is currently seeking to amalgamate them into larger units. The federation of TCCSs has recently applied for a banking licence under the new provisions. It intends to create a separate bank to mobilise deposits for on-lending to the TCCSs (while the TCCSs already have the right to raise deposits from the public, the federation considers that establishing a bank would increase its credibility with the public and the protection that it can offer depositors). In addition, a new bank, Pramuka Bank, has recently been licensed and intends to include microfinance in its overall operations. Pramuka Bank intends to develop a number of innovative approaches to reaching poor borrowers in the villages, including through the establishment of agencies at the village level and the use of incentive payments for agents. Nevertheless, the scope for other institutions wishing to establish small microfinance banks to make use of the new provisions may be limited, as the minimum capital requirement is still quite high and may be difficult to meet. Experience in other countries suggest that small banks are much more likely to become involved in microfinance than large banks. If banks are to play an active role in microfinance it is necessary to have some mechanism to enable small banks to be licensed. It would be appropriate to establish a framework for licensing small banks, setting realistic minimum capital requirements and removing any other restrictions which may prevent their establishment. In this regard, the experiences of countries such as Indonesia and the Philippines may be instructive (see the Indonesia and Philippines country studies). 4.2 Interest rates As noted above, there are no restrictions on the interest rates that commercial banks can charge. Hence, interest rate regulation is not a factor impeding regulated banks from becoming involved in microfinance. 4.3 Prudential regulation and supervision Prudential standards for commercial banks generally follow the Basle Accord. The required capital adequacy ratio is 8 per cent of risk-weighted assets, there is a liquidity requirement of 20 per cent of deposit liabilities, and loan provisioning requirements are 25 per cent for sub-standard loans, 50 per cent for non-performing loans and 100 per cent for loss loans. Commercial banks are also subject to frequent reporting requirements. Prudential requirements for specialised banks have not yet been developed, but are expected in broad terms to follow those applying to commercial banks. These requirements would not appear to impose any undue restrictions on small banks or banks engaged in microfinance, although it would be appropriate to review them if special provisions for licensing small banks were introduced. 5 Summary and recommendations The World Bank estimates that 22.4 per cent of the population were in poverty in 1990–91. This is considerably lower than in other countries in South Asia, but slightly higher than in Malaysia, Thailand and Indonesia. Sri Lanka has long emphasised human development, with social indicators such as life expectancy, infant mortality and literacy generally much better than for other countries with similar income levels. The major institutions engaged in microfinance in Sri Lanka are the thrift and credit cooperative societies (TCCSs) and a small number of non-governmental organisations (NGOs). The credit cooperative movement in Sri Lanka has been far more successful than in most other countries in the region. Whilst most members are not poor, the movement has sought to increase its outreach at the village level and has served as a vehicle for various targeted poverty lending programs. There are also a number of NGOs involved in microfinance. Arrangements for direct support Government and donor agencies have provided support for microfinance through a variety of channels. Recently, there has been some shift away from the use of NGOs and back to schemes more directly involving government agencies. Affirm NDTF as primary vehicle for microfinance The National Development Trust Fund (NDTF) operates essentially as a conduit for donor and government funds, channelling them to microfinance institutions (MFIs) for on-lending to final borrowers. While there have been a number of problems in its operations, this model has worked well in other countries and NDTF has recently improved its operations. The government should renew its commitment to NDTF as the primary vehicle for channelling direct support to microfinance in Sri Lanka. Strengthen independence and clarify objectives of NDTF At the same time, NDTF needs to continue to improve its performance. One critical issue is to ensure that it is not politicised. This may be achieved by strengthening the independence of the board of trustees, and spelling out more clearly the fund’s objectives. The fund should continue to set and, more importantly, enforce minimum standards for partner organisations, and gradually raise these over time. Consider transferring SFLC project to NDTF The Small Farmers and Landless Credit (SFLC) project appears to have been reasonably successful. It is now entering a transition period, with responsibility being handed over from the central bank to the regional rural development banks (RRDBs). To the extent that the central bank continues to be involved in this and other microfinance projects, it would be appropriate for it to lend directly to NGOs rather than lending to NGOs through RRDBs. It is not clear that the RRDBs are well placed to take over the project. The government should perhaps consider transferring the project to NDTF instead. Redesign Samurdhi scheme to include NGO role and achieve more self-sufficiency The Samurdhi Development Credit Scheme is very new, and hence no data are yet available on repayment rates or other indicators of performance. Nevertheless, there are some causes for concern in the design of the scheme. The scheme does not envisage any explicit role for NGOs or other specialist MFIs. It should be redesigned to provide a role for NGOs and other specialist MFIs, either in identifying and motivating borrowers or as direct financial intermediaries. It should also be redesigned to make it more self-sufficient, and to reduce the possibility of its undermining more sustainable microfinance programs. Publicise Hatton National Bank linkage program success One highly innovative program is the program operated by Hatton National Bank for linking to NGOs and poor borrowers. The central bank should take a more active role in encouraging linkages, by documenting and publicising the experience of Hatton National Bank, and issuing clear guidelines that other banks could follow. The government should also avoid debt forgiveness for small loans in the future. Regulation of non-bank microfinance institutions Review and streamline regulations for cooperatives Regulations facing TCCSs are restrictive in a number of ways, with the registrar and relevant departmental minister having considerable discretion. The government should review the regulatory requirements facing cooperatives, with a view to streamlining them and removing unnecessary requirements. Ensure NGO standards and accountability By contrast, it is possible for NGOs involved in microfinance to operate without any legal backing whatsoever. It would appear that this unstructured regime has led to some problems. On the other hand, it is also clear that the environment for NGOs has at times been highly politicised, and that NGOs have been subject to considerable uncertainty. The government should tighten up the procedures, requiring NGOs to be registered and monitoring their activities to ensure they are properly accountable to their members and third parties. At the same time, the government should ensure that NGOs meeting the necessary requirements are permitted to operate with maximum flexibility and certainty. Restrict the acceptance of general deposits to banks only The law specifically permits TCCSs to receive deposits from persons who are not members, and many societies are now actively seeking deposits from non-members. Given the lack of prudential requirements, it is surprising that societies should be permitted to receive deposits from non-members. The government should not permit TCCSs to accept deposits from the general public in their own right. However, when the federation establishes a bank, there is no reason why the primary societies should not act as agents. Allow MFIs to accept members’ savings According to the law, NGOs are not allowed to provide savings facilities, even to members. In practice, the prohibition on mobilising savings has generally not been applied strictly. However, one NGO has been prosecuted for mobilising savings, and the current situation leads to considerable uncertainty. MFIs should at least be permitted to accept members’ savings. The government should amend the law to give NGOs greater certainty in this area. TCCSs are free to set their own interest rates for both deposits and loans. However, where societies receive outside funds for on-lending to members, they are not permitted to charge a margin of more than 4 per cent. While this may seem restrictive, the margin has been set in consultation with the federation. There are no restrictions on the interest rates that NGOs engaged in microfinance can charge. Establish reasonable monitoring of federation activities There is virtually no prudential regulation and supervision of MFIs in Sri Lanka, whether TCCSs or NGOs. Reporting requirements are also minimal. In the case of the TCCSs there is a well-developed structure for coordination and self-regulation, with the federation and district unions monitoring and supervising the primary societies. These structures have contributed importantly to the overall strength of the movement. The Department of Cooperative Development should monitor the activities of the federation and perhaps conduct spot checks on individual primary societies, rather than establishing a parallel system for monitoring all primary societies. Establish appropriate monitoring for NGOs and MFIs In the case of NGOs, more intensive monitoring and supervision is necessary. On the other hand, it is generally accepted that it is unnecessary and inappropriate to subject MFIs which do not accept deposits from the general public to full prudential regulation and supervision. NDTF has laid down operational procedures for partner organisations which borrow from it. It may be possible to develop the standards adopted by NDTF, with appropriate modifications and in consultation with MFIs, into more general standards for MFIs operating in Sri Lanka. For this to happen, however, it would be necessary for the standards to be enforced more rigorously than they have been in the past. Moreover, there is currently considerable uncertainty as to the future role of NDTF, and a need for it to improve its performance further. Encourage MFI networks Networks of MFIs have an important role in a number of areas, including information exchange, training, research, policy dialogue with government and donor agencies, and establishing standards for self-regulation. The new forum of microfinance NGOs is a very positive initiative. Leading MFIs should continue to develop and expand its membership and range of activities, with support from UNDP. Regulation of banks The minimum capital requirement to establish a commercial bank is too high for MFIs wishing to establish banks, or bodies wishing to establish small banks operating at the local level. The minimum capital requirement for specialised banks is lower at Rs100 million ($1.8 million), but is still likely to be too high. If banks are to play an active role in microfinance it is necessary to have some mechanism to enable small banks to be licensed. The government should establish a framework for licensing small banks, with realistic minimum capital requirements. It should also remove any other restrictions which may prevent the establishment of small banks. References
|
