1 Introduction and background 1.1 Key demographic and economic data Thailand is located in mainland Southeast Asia, opening on to the South China Sea and Indian Ocean, and sharing land borders with Malaysia, Myanmar, Laos and Cambodia. In 1995 it had a population of 58.2 million people occupying an area of some 513,000 square kilometres, with a density of 113 people per square kilometre. Population growth over the period 1990 to 1995 averaged only 0.9 per cent per year, the lowest for any of the nine countries in the study. In 1995 the total fertility rate stood below replacement level at 1.8 births per woman. Thailand has proceeded much further along the demographic transition than the other countries. Thailand is at a very different stage of development to most of the other countries in the region, and is now classified as a middle-income country by the World Bank. In 1995, income per capita was $2,740, less than Malaysia but well over double that of any other country included in this study. This reflects rapid economic growth over an extended period. Over the period 1985 to 1995 per capita income grew at an exceptional rate of 8.4 per cent per year, much higher than for any other country included in this study. However, since mid-1997 Thailand has experienced a financial crisis, requiring an emergency loan from the International Monetary Fund. Little economic growth is expected over the next two to three years. Very rapid economic growth has been accompanied by significant changes in the structure of output. Between 1980 and 1995, the share of agriculture in GDP declined from 23 to 11 per cent, the lowest for any country in this study. This decline was matched by an increase in the share of industry, from 29 per cent of GDP to 40 per cent, while the share of services remained broadly unchanged at just under 50 per cent. These changes have had significant effects on settlement patterns, with the proportion of the population living in urban areas more than doubling, from 17 per cent in 1980 to 36 per cent in 1995. Fifty (50) per cent of the urban population live in Bangkok, a higher concentration in the largest city than for any other country in the study. Human development is also relatively high compared with most other countries in the region. Thailand scores highest of the nine countries in this study in the UNDP human development index for 1994, and scores in the top three in terms of life expectancy at birth, the infant mortality rate, and adult literacy. Nevertheless, on some indicators Thailand’s performance is disappointing given its per capita income. For instance, life expectancy at birth at 69 years is less than Malaysia (71), and three years less than Sri Lanka (72), despite the fact that per capita income is nearly four times that of Sri Lanka. Similarly, the infant mortality rate (35 per 1,000 live births) is more than double Malaysia (12) and Sri Lanka (16), and only slightly lower than the Philippines (39). 1.2 Poverty Estimates of poverty Reflecting the high per capita income, the incidence of poverty is lower than in most other countries in the region. The official poverty line in Thailand is based on an estimate of nutritional adequacy. In 1976, the recommended daily per capita consumption was set at 1,978 calories but in 1981 it was revised to 2,016 calories, based on changes in the population structure. This recommended daily consumption is then converted into typical ‘food consumption baskets’, consisting of representative samples of food items consumed in rural and urban environments. These baskets are then costed, based on average retail prices in rural and urban areas. Finally, a component is allowed for non-food items, based on the ratios of food to total expenditure for the lowest quintiles of the rural and urban populations in the socioeconomic survey of 1975–76. Based on official estimates, the incidence of poverty stood at 30.0 per cent in 1975–76. While the incidence of poverty fell to 23.0 per cent in 1981, it rose sharply to 29.5 per cent in 1986, reflecting large drops in commodity prices in that year. Since then it has fallen significantly, to 22.2 per cent in 1988, 18.0 per cent in 1990 and 13.1 per cent in 1992. Estimates of the depth and severity of poverty have also fallen. The incidence of poverty differs significantly between regions, ranging from 1.1 per cent in Bangkok to 22.3 per cent in the north-east. In 1992, around 58 per cent of the poor lived in the north-east, almost 85 per cent lived in rural areas, and 98 per cent had household heads with primary school education or less. Much of the data on which the official poverty line is based are now somewhat dated. The estimates of nutritional adequacy are based on daily dietary allowances obtained from the Ministry of Public Health in the mid-1970s. While the minimum calorie intake was revised in 1981 to take account of changes in the population structure, it has not been revised since then. The allowance for non-food items is based on expenditure patterns from 1975 to 1976. In recent years Dr Medhi Krongkaew of the National Economic and Social Development Board (NESDB) has undertaken considerable research on the poverty line (see, for example, Medhi 1997), and has proposed a number of changes to revise and update it. His estimates suggest a much higher poverty line than official estimates. Using Medhi’s methodology, the poverty line in 1990 was equivalent to annual per capita expenditure of B12,764 ($505) for urban areas and B8,366 ($331) for rural areas, compared with the official poverty lines of B7,632 ($302) and B4,968 ($196) respectively. Not surprisingly, his estimates of the incidence of poverty are much higher than the official estimates. Medhi estimates the incidence of poverty in 1992 at 36.5 per cent, more than double the official estimate, with poverty incidence ranging from 5.0 per cent in Bangkok to 55.0 per cent in the north-east. Deloitte and Touche Management Consultants (1996) have also derived estimates of poverty, based on daily calorie intake. Their estimates of the incidence of poverty in 1992 are slightly higher than the official estimates, but significantly lower than the Medhi estimates. In summary, poverty in Thailand has reduced considerably in recent years, and the official estimates suggest that the incidence of poverty has now fallen to around 10 to 15 per cent of the population. On the other hand there is considerable regional variation, and analysis by Medhi suggests that poverty remains a pervasive problem. Moreover, it is clear that there has been a long-term trend, dating from at least the mid-1970s, towards greater income inequality. Between 1988 and 1992 alone, the income share of the top 20 per cent of households rose from 54.9 to 59.4 per cent of GDP, while the share of the bottom 20 per cent of households fell from 4.5 to 3.8 per cent. In this respect, Thailand’s experience is quite different from that of other countries in East Asia, which have experienced reduced income inequality over the last twenty years. While income inequality in Thailand was substantially lower than in the Philippines and Malaysia in the mid-1970s, it is now substantially higher, with Thailand now having the most unequal distribution of income in East Asia. The World Bank (1996) suggests that reasons for the increase in inequality include structural change in the economy, the fact that formal sector jobs have not been distributed equitably across households, and the lack of access by the poor to secondary education. Policies for poverty reduction Poverty reduction has been a central objective of development planning in Thailand since the Third Plan (1972–1976). However, it has not consistently been a top priority. Thailand has adopted policies which have emphasised high economic growth and led to very rapid growth in GDP and per capita incomes. This has contributed to significant reductions in the proportion of people in poverty. However, as noted above, the trickle down to the poorest segments of the population has been slow, and there has been a significant increase in income inequality. Clearly, a wide range of government policies, including macroeconomic, investment and human development policies affect poverty, both positively and negatively. However, there are several programs in Thailand that are specifically designed to reduce poverty. These are described by the World Bank (1996). First, there are a number of programs that provide cash transfers and in-kind transfers to the poor. The Department of Public Welfare provides a monthly allowance to the elderly without other means of support. It has also established village community funds, administered by village welfare committees, to be used in assisting poor residents. The Ministry of Public Health provides low-income cards to around 20 per cent of the population, entitling them to free medical services. Another in-kind transfer program is the school lunch program operated by the Ministry of Education, which provides a free meal each school day to children from poor families at the preschool and primary levels. Second, there are some programs directed at income generation by the poor. The most significant of these is the poverty alleviation project, a microfinance program operated by the Community Development Department, and discussed in more detail below. Another major program is the Tambon Development Program (the tambon is the administrative unit at the sub-district level) administered by the Office of the Prime Minister, which includes a public works component aimed at providing rural infrastructure and employment generation. The World Bank notes that expenditure on these anti-poverty programs is modest, accounting for only 1.6 per cent of government expenditures in 1995. Moreover, the share has fallen in recent years, although it was budgeted to rise to 2 per cent in 1996. The World Bank argues that better targeted policies are needed to reach the hard-core rural and urban poor. It argues that the government should identify the major schemes by which the poor are to be assisted, and define clear poverty-oriented goals for these. The Eighth Plan, to cover the period 1997–2001, signals some changes in development policy. It advocates a shift to ‘people-centred development’, intended to support human development, improve quality of life and promote increasing community participation. Among the targets are to:
Nevertheless, this plan was developed on the assumption of continued rapid economic growth, prior to the current financial crisis. Clearly, much slower economic growth and the need for significant reductions in government spending will make it harder to reduce poverty. It is likely that a major reorientation of government spending will be necessary if the poor are to be protected from the effects of the downturn and inequality is to be reduced. 1.3 Overview of the financial system The formal financial sector in Thailand consists primarily of commercial banks, non-bank financial institutions, specialised financial institutions established by the government, and cooperatives. As at the end of 1996, there were 15 domestic commercial banks and 14 foreign commercial banks, with a total of 3,138 branches throughout the country. There were also a number of credit foncier companies, finance companies, and finance and securities companies. The above institutions are supervised by the Bank of Thailand. In addition, there were seven specialised financial institutions supervised by the Ministry of Finance. Most notable among these are the Bank for Agriculture and Agricultural Cooperatives (BAAC), 628 branches and sub-branches; the Government Savings Bank (GSB), 543 branches; and the Government Housing Bank with 169 branches. Finally, there were 1,127 thrift and savings cooperatives and 2,832 agricultural cooperatives, supervised by the Ministry of Agriculture and Cooperatives. Some, but not all, of the agricultural cooperatives provide financial services to their members. The financial system in Thailand is reasonably well developed. In 1995 bank interest rates on both loans and deposits were significantly positive in real terms (that is, well above the inflation rate), encouraging financial development. There is a general ceiling on interest rates applying to loans of 15 per cent, but since 1992 all financial institutions regulated by the Bank of Thailand, including commercial banks, have been exempted from this provision, and are free to charge interest rates based on market conditions. The ratio of money and quasi-money (currency outside banks, demand deposits, time and savings deposits, and similar bank accounts) to GDP stood at 73.8 per cent in 1995, slightly lower than for Malaysia but much higher than for any other country included in this study. This ratio tends to increase with the development of the economy and the financial sector, because deposits tend to increase as monetisation accelerates and banking facilities increase. On the other hand, it can fluctuate widely in the short term depending on the macroeconomic situation. The Bank of Thailand has carried out two financial development plans since 1990, designed to identify obstacles to financial development and implement measures to overcome them. The measures taken relate to regulation reform and the establishment of financial infrastructure. Important measures include relaxing exchange controls, abolishing interest rate ceilings for financial institutions, introducing greater competition, and modernising the payments system. Nevertheless, the current crisis suggests that there has been insufficient attention to prudential regulation and supervision. The crisis at least partly reflects financial factors, with banks incurring significant losses after lending large amounts on the basis of unsustainable property values. While it remains to be seen what longer term effects this will have on the financial system, it is clear that there has been a considerable erosion of confidence in the short term. The crisis has led the Bank of Thailand to increase its focus on prudential regulation and supervision, and it is unlikely that the bank will issue any new banking licences in the short term. Bank for Agriculture and Agricultural Cooperatives The Bank for Agriculture and Agricultural Cooperatives is by far the largest lender in rural areas. In 1995–96, BAAC provided credit of B112 billion ($4.4 billion) to farmers and farmer institutions. It reached 4.65 million farm families, or 82 per cent of all farm families in Thailand. This consisted of 3.33 million individual farmers, 855 agricultural cooperatives with a total membership of 1.28 million, and 323 farmers’ associations with a total membership of 43,000. BAAC also provides extensive deposit services, with the value of deposits totalling B91 billion ($3.6 billion) at end-1995. Interest rates for small loans are maintained at below market rates, with interest rates for loans to individual farmers set at 9 per cent for loans below B60,000 ($2,370) and 11.25 to 12.25 per cent for loans between B60,000 ($2,370) and B1 million ($39,530). These loans are cross-subsidised by the bank’s other operations. Overall repayment rates are satisfactory and BAAC operates profitably, although this in part reflects its access to soft loans from the Bank of Thailand, the Asian Development Bank and other sources. As noted above, BAAC also lends to agricultural cooperatives, and has a small program for lending to farmers’ associations. Farmers are organised in joint liability groups of five or more members who guarantee each other’s loans. Loans to cooperatives and associations attract interest rates of 6 per cent where the amounts on-lent to individual farmers are below B60,000 ($2,370), and 9.25 per cent where the amounts on-lent are between B60,000 ($2,370) and B1 million ($39,530). For loans up to B50,000 ($1,980), no collateral is required. Hence, BAAC uses some lending techniques similar to those used by microfinance institutions (MFIs). However, BAAC does not specifically target the poor, and some commentators suggested that it may not reach the poorest farm families. Moreover, the poorest people are often without land and therefore are not farmers. These people are not eligible to receive loans from BAAC, by virtue of its mandate as stipulated in its charter (it is understood that the government is considering amending the charter to enable BAAC to be a rural development bank). Therefore, while BAAC has huge outreach in the rural areas, it may not reach the poorest rural households. It should also be noted that repayment rates by members to cooperatives and farmers’ associations, and by the cooperatives and farmers’ associations to BAAC, are relatively low. In 1995–96, the repayment rate by members to cooperatives was 70.2 per cent, and the repayment rate by members to farmers’ associations was 67.9 per cent. BAAC has commented that the farmers’ associations lack knowledge, understanding and true cooperation, and BAAC has found it difficult to expand lending through them. Historically, the BAAC charter only permitted it to provide loans to farmers for on-farm activities. In 1992 the charter was amended to allow a fixed percentage of lending for agriculture-related activities, but only if operated by farmers. In 1995, the Asian Development Bank provided a soft loan of $50 million to facilitate this lending. This amount has already been drawn down, and the Asian Development Bank is considering a further loan of $100 million. There is also some discussion of amending the charter to enable BAAC to lend for non-agricultural projects in rural areas. However, it is unlikely that the bank will increase its poverty lending. While there is some pressure for it to do so, its cost structure and interest rate structure mean that it is unlikely that it could engage in poverty lending on a commercial basis. The German agency for technical cooperation (GTZ) is currently in the process of establishing a small revolving fund to enable BAAC to develop lending procedures and products which cater to the financial needs of non-farm enterprises, and to develop the capacity and skills of credit officers in an experimental stage. The experience from this pilot test will be used for replication if and when BAAC obtains approval from government for lending to the non-farm sector. 1.4 Overview of microfinance The government’s main policy thrusts for reducing poverty appear to be promoting community participation in general, and increasing employment opportunities, especially wage employment, in rural areas. Nevertheless, the Eighth Plan contains some references to the role of microfinance and financial services. It calls for the extension of the scope of the operations of the rural development fund managed by GSB, to encourage savings at the community level and give wide-ranging support to local and community development activities. It also calls for the extension of financial services by formal financial institutions, especially commercial banks, BAAC and GSB, to lower income and less capitalised customers in regional and rural areas. There are a number of government programs which support microfinance. The largest is the poverty alleviation project operated by the Community Development Department (CDD). This project provides for the establishment of a fund at the village level, for on-lending to poor households. By 1995 loans outstanding were nearly B1 billion ($40 million), benefiting up to 200,000 households. The other two major programs are the rural development fund operated by GSB, and the program operated by the Urban Community Development Office (UCDO) within the National Housing Authority. These two programs operate by providing loans to community organisations (sometimes referred to as savings and credit organisations), which then on-lend to individual members. Between them they have provided loans to nearly 1,500 community organisations with a combined membership of around 200,000 households. There are also a number of smaller government programs. Of the three major programs, the CDD program and UCDO program generally reach poor borrowers, whereas the GSB program reaches both poor and non-poor borrowers. The community organisations financed under the GSB and UCDO programs may take a variety of forms. Some are registered as cooperatives, but most are not registered. Such community organisations obtain funds from a variety of sources, including the savings of members. The performance of different community organisations in reaching the poor also differs. There are a number of types of cooperatives, with both agricultural cooperatives and thrift and credit cooperatives providing savings and credit services. It is understood that the agricultural cooperatives generally do not reach the poorest borrowers. Moreover, many of the thrift and credit cooperatives cater for groups of wage and salary earners rather than the poor. However, is understood that the credit union cooperatives, that subset of thrift and credit cooperatives based on locality rather than employment, generally reach poor borrowers. There are 236 credit union cooperatives, with a total membership of around 93,000 members. The Credit Union Federation of Thailand serves as an apex body for these credit union cooperatives. Generally, the community organisations lend on an individual basis, but UCDO in particular has been trying to encourage them to adopt a group lending approach. There is also a small number of private bodies and NGOs involved in microfinance. Of these, the Small Enterprise Development Company, based in Surin in the north-east, is perhaps the only one that is involved exclusively in microfinance. In addition, some multi-purpose NGOs such as the Local Development Institute and the Population and Community Development Association, also operate microfinance programs. Nevertheless, NGOs have a much more limited role in microfinance than in most other countries in the region. The demand for microfinance Before analysing the policy and regulatory environment, it is useful to consider the demand for microfinance. A number of commentators suggested that the demand for microfinance in Thailand is much less than in most other countries in the region. If this is the case, the imperative to increase outreach may not be as great as in some other countries. The size of the potential market is clearly more limited than in many other countries. As noted above, the incidence of poverty in Thailand is much lower than in all other countries included in the study except for Malaysia. Moreover, the population tends to be highly mobile, and growth in wage employment has been very rapid, with very low rates of unemployment. A number of commentators noted that in most cases poor people have access to wage employment. While wages for unskilled workers are low, the returns may still be higher than from self-employment in the informal sector of the economy, especially for people with few business skills. Again, the availability of wage employment sets Thailand and Malaysia apart from the other countries included in the study. Notwithstanding these points, the distribution of resources in Thailand is heavily skewed. Opportunities for wage employment are heavily concentrated in Bangkok and the other major urban centres while the vast majority of the poor live in rural areas. While there has been substantial migration from rural to urban areas, this causes considerable social dislocation and places additional demands on urban infrastructure which is already stretched. Moreover, much migration is ‘circular migration’, where one family member works in Bangkok for part of the year because of the lack of opportunities in rural areas. As well as causing disruption to family life, the income generated may not be sufficient to support family members remaining in the rural areas. There are also increasing pockets of poverty in the main urban centres. The government’s policy is therefore to encourage development in the rural areas. These problems are likely to increase with the economic downturn. Growth in wage employment is likely to slow considerably, forcing the unskilled and new labour market entrants to rely more heavily on self-employment in the informal sector. There has also been a significant reversal of migration patterns, with large numbers of urban workers who have been laid off returning to rural areas. Hence, it appears that there is a significant demand for microfinance, and this may well increase, at least in the medium term. To what extent has this demand already been satisfied? To date, the CDD program has provided loans to perhaps 200,000 households. The GSB program has provided loans to 1,360 organisations with a total membership of 188,000 households. And the UCDO program has provided loans to 103 organisations with a total membership of maybe 20,000 households. In addition, there are a number of other government and NGO schemes. Hence, on the face of it, outreach is already quite high. Nevertheless, outreach is still relatively low compared to the some 1.5 million households in poverty. While few data on the CDD program are available, it would appear that it essentially provides a one-off loan (or grant) to beneficiary households, rather than any ongoing access to financial services. In the case of GSB and UCDO, the data relate to the total membership of the organisations that have received loans. It is not known how many individual members have received loans. The size of the loans, and whether they were large enough to support income-generating activities, is also not known. Further, it is likely that there has been considerable leakage to the non-poor under some schemes. It would therefore appear likely that there is still a substantial need for microfinance programs to increase their outreach to poor borrowers. 2 Arrangements for direct support 2.1 Support for specialised microfinance institutions There are two major government programs which provide support to cooperatives and community organisations involved in microfinance. These are the rural development fund operated by the Government Savings Bank (GSB), and the program operated by the Urban Community Development Office (UCDO). The credit program for rural development, administered by GSB, was originally administered by the Office of the Prime Minister and later by the National Economic and Social Development Board (NESDB). However, it suffered from low repayment rates. Responsibility for administering the program was assigned to GSB in 1995. The government has provided a total of B546 million ($21.6 million) to the bank as loanable funds for the program. In addition, the bank took over a loan portfolio of around B103 million ($4.0 million) when it took on the program, although much of this was in the form of bad debts. The program provides loans to savings and credit organisations in rural areas at an interest rate of 7 per cent, either for on-lending to their members or for investment projects undertaken by the groups themselves. Currently, 2,600 savings and credit organisations, with a total membership of 367,000, are affiliated with the program. Of these, 1,360 organisations, with a total membership of 188,000, have obtained loans. The program is not means tested, and the savings and credit organisations include members from a range of income levels. Repayment rates are reported to be high since the bank took over the program. While there are some arrears, no loans to savings and credit organisations have been written off. The other major program is the urban poor development program operated by UCDO. This started in 1992 with a budget grant of $50 million. UCDO offers loans to community organisations in low-income urban settlements for income generation, housing and land, and small-scale community activities. Interest rates vary from 3 to 10 per cent, but the interest rate for income generation projects is 8 per cent. While UCDO is a government agency, it is governed by a nine-person board with three representatives from government (Bank of Thailand, Ministry of Finance and NESDB), three elected by the member community organisations, two from NGOs and one from the private sector. There are currently 616 member organisations, with a total membership of 53,000. Of these organisations, 385 have savings activities, and 103 have received loans from UCDO for income generation activities. The recovery rate for loans from UCDO to member organisations, measured as repayments collected as a percentage of repayments due on a monthly basis, is around 90 per cent. As well as providing loans to savings and credit organisations, both GSB and UCDO provide them with technical support. For instance, GSB provides training in leadership and management and some skills training to the organisations. It also assists organisations to develop simple accounting and reporting systems, and visits them from time to time to follow up and keep track of the performance of its investments, exchange information and provide assistance. UCDO plays an even more active role in this regard. It has a number of activities to support the formation of savings groups, including advising people on how to structure their savings groups and financial systems, training in finance and accounting systems for savings groups, training in business skills such as accounting, marketing, planning and investment, and training in community development techniques such as participatory techniques. To date, UCDO has organised 168 training courses on these and similar matters, with a total of 6,200 participants. UCDO also makes frequent visits to member organisations to offer advice and assistance. Both programs rely heavily on funds from the government, and neither of them is close to self-sufficiency. The GSB scheme relies on grant money from the budget for its loanable funds. The bank receives a margin of 4 per cent for managing the scheme, which is not sufficient to cover its costs, and the program is cross-subsidised by its other operations. Similarly, in the case of the UCDO program, loanable funds come from a grant from the government. Interest from loans to community organisations covers only about 30 per cent of administrative expenses. The government is planning to establish the Community Organisation Development Institute (CODI) to organise microfinance on a more systematic basis. The proposal is being developed by the Prosperity Decentralisation Policy Committee, chaired by the Ministry of Finance and with representatives from the Community Development Department (CDD), NESDB, Bank of Thailand, GSB, and NGOs. In effect, the new institute will take over the overall policy responsibility for the programs operated by GSB and UCDO with respect to microfinance. It is expected that, at least initially, GSB and UCDO will continue to administer the schemes. CODI will be overseen by a committee similar to the Prosperity Decentralisation Policy Committee. Moreover, it is envisaged that GSB and UCDO will on-lend funds to individual corporations at the district level. These district corporations will be managed by a board with representatives from government, NGOs and the local community, and will in turn on-lend to cooperatives and community organisations in their districts. The district corporations will also provide training to cooperatives and community organisations. Detailed lending guidelines, such as loan terms and conditions, targeting/means testing, lending methodologies, and accounting and reporting standards, have not yet been developed. The relatively few non-government and private schemes in existence have tended to rely on outside funds, primarily from external agencies. For instance, the Small Enterprise Development Company has been supported by Catholic Relief Services. Comment GSB and UCDO have much better repayment rates than the CDD program and most other government programs, although the available data suggest that repayment rates are well below those achieved by second tier microfinance institutions in other countries such as Bangladesh, India and Sri Lanka. It is likely that their reasonably good performance reflects their independent boards and high degree of autonomy, and the fact that they lend to non-government cooperatives and community organisations. In establishing CODI, care should be taken to ensure that it is managed by an independent board and is free from political and bureaucratic interference. It is also important that CODI develop rigorous performance and reporting standards for the organisations that it lends to, to ensure that they operate on a sound basis and that they maintain repayment rates of at least 95 to 98 per cent on their loans to CODI. As noted above, neither the GSB program nor UCDO is close to self-sufficiency. Clearly, if the two programs could increase their self-sufficiency, they could increase their outreach to poor clients for any given outlay of government funds. In setting up CODI, it would be appropriate for the government to establish explicit targets and timetables in terms of outreach and self-sufficiency. This would provide an incentive for CODI to improve its cost-effectiveness over time. It should be possible for CODI to cover all of its operational costs, and a portion of its notional financial costs, through its interest earnings on loans to community organisations. To maximise outreach to the poor, it is also important to minimise leakages to the non-poor. Neither the UCDO program nor the GSB program has an explicit means test. It would be appropriate for CODI to adopt an explicit means test to ensure that it reaches the poor. Given the difficulty of measuring income in the informal and subsistence economy, means tests should be based on a range of indicators in addition to income, including housing quality. 2.2 Support through the banking system The commercial banks are required to lend 20 per cent of their total deposits from the previous year to the rural sector. Fourteen (14) per cent is allocated for lending to agriculture and the remaining 6 per cent for lending to agricultural business. If they do not meet this target they are required to deposit the balance with the Bank for Agriculture and Agricultural Cooperatives (BAAC), but in practice most commercial banks meet the 20 per cent target. However, there are no requirements concerning lending to the poverty sector. Given the extensive outreach of BAAC in the rural areas, the commercial banks tend to serve only large borrowers in rural areas and leave the small and medium borrowers for BAAC. For instance, until the mid-1970s, Bangkok Bank provided a significant number of small loans in the rural areas, but it now concentrates almost exclusively on large borrowers. The only significant microfinance program channelled through the banking system is the rural development fund, administered by GSB and discussed above. GSB is a specialised financial institution supervised by the Ministry of Finance. It operates as a second tier microfinance institution, lending to savings and credit organisations which on-lend to final borrowers. Hence, under this program, GSB does not lend to final borrowers in its own right. The program has been much more successful than most microfinance programs channelled through the banking system in other countries, which have generally involved banks lending direct to final borrowers. 2.3 Government microfinance programs The largest microfinance program in Thailand is the poverty alleviation project operated by CDD, which provides loans in poor sub-districts. The program operates only in villages that are identified as poor, with a revolving fund of B280,000 ($11,070) established in each village. Individual borrowers are required to have household incomes of less than B5,000 ($200) per person per year, although it is not known how rigorously this is enforced. Borrowers are identified by local government authorities using basic minimum needs data collected by the village, and loans are provided for income-generating activities. Loans to individual borrowers are interest free, with administrative costs met through the budget. By 1995, around B2.8 billion ($110 million) had been advanced to village welfare committees in around 10,000 villages, and nearly B1 billion ($40 million) had been on-lent to final borrowers. The World Bank (1996) noted that the repayment record in 1994 and 1995, for loans advanced in 1993, appeared to be extremely low, varying from 0.02 to 3.1 per cent of loans in different villages. There appears to be little hard data available on the CDD program. Nevertheless, the data that are available suggest that repayment rates are very low, and that the program is operating more as a grant scheme than as a genuine microfinance program. The program is consuming a large volume of government funds, and does not appear to provide an ongoing stream of financial services to borrowers. If the government intends to persist with the scheme, it should be comprehensively reviewed as a matter of priority. More fundamentally, a scheme operated by a government department through local government authorities is unlikely to be the most cost-effective way to maximise outreach, and it would appear necessary to review the basic design of the scheme. There are also a number of other government programs for microfinance. For instance, the Cooperative Promotion Department provides soft loans to cooperatives, including for on-lending to their members. Many other departments have also provided funds for microfinance. Most of these programs are quite small. They are generally not means tested, and it is understood that repayment performance has generally been poor. Establishing CODI presents an excellent opportunity for rationalising existing government programs, especially those operated directly by government departments. It would appear desirable to consolidate the CDD program and other government microfinance programs under CODI. 3 Regulation of non-bank microfinance institutions 3.1 The broad regulatory framework Cooperatives, including those involved in savings and credit activities, are registered under the Cooperative Societies Act 1968. While there are a number of types of cooperatives, it is mainly the agricultural cooperatives and thrift and credit cooperatives (including the credit union cooperatives) which are involved in savings and credit activities. To be registered, a cooperative must have at least ten members with a common interest, and regulations covering the name, type and objectives of the cooperative, and details of membership. Cooperatives are required to have a board of directors, elected annually by the members at an annual general meeting. The board is responsible for the overall management of the cooperative. Agricultural and thrift and credit cooperatives are exempt from most forms of taxation, including income tax and dividend tax. While some community organisations engaged in microfinance are registered as cooperatives, most are unregistered and operate informally. For instance, figures from the Urban Community Development Office (UCDO) show that of the 385 members with savings activities, only 66 are legally registered as cooperatives. The other 319 are not registered and operate informally. There is also a small number of private and non-governmental organisation (NGO) microfinance programs operating in Thailand. These institutions are not owned by their members, and hence they cannot register as cooperatives. There is no specific provision for such organisations to register as NGOs and obtain a legal identity. Most of them are registered as associations or foundations, and provide a range of services of which microfinance is just one. Associations and foundations are not liable for taxation, but are not permitted to undertake business activities. Hence, it is not clear if these organisations are strictly entitled to provide savings and credit services. These organisational forms would not appear to be appropriate for specialist microfinance institutions (MFIs) which provide only financial services. The Small Enterprise Development Company is one of very few specialist MFIs. It found that the simplest way to obtain a legal identity was to register as a company under the Civil and Commercial Code. To register as a company, it is necessary to meet a minimum capital requirement of B500,000 ($19,760). This may be prohibitive for a small NGO wishing to engage in microfinance. It is also necessary to have a memorandum and articles of association providing details of the name of the company, objectives, election of board of directors, and similar matters. If an MFI registers as a company, it is liable for company income tax if it makes a profit. Clearly, the company registration procedures are not designed with the needs of NGOs providing microfinance in mind, and some of the requirements may be difficult for NGOs — in particular, smaller NGOs — to comply with. It would be appropriate to establish a simple registration procedure for NGOs wishing to engage in microfinance. To some extent, the lack of such provisions may explain the paucity of microfinance NGOs in Thailand. 3.2 Interest rates There is a general ceiling of 15 per cent on interest rates applying to loans. This ceiling is spelled out in the Civil and Commercial Code, and applies to all bodies and transactions unless they are specifically excluded. Examples of exclusions are as follows:
Agricultural cooperatives, informal organisations, private and NGO microfinance programs and private transactions are still subject to the 15 per cent ceiling. It is understood that registered cooperatives generally stay within the relevant ceilings. However, this is not the case for unregistered community organisations. In theory, unregistered community organisations are subject to the general 15 per cent interest rate ceiling which applies to all bodies and transactions that are not specifically excluded. However, it is understood that many, if not most, charge higher interest rates. It was reported that most community organisations funded by the Government Savings Bank (GSB) and UCDO charge around 2 per cent interest flat per month to their members. These interest rate ceilings affect different institutions differently, and are a constraint to the development of microfinance on a sustainable basis. Both the 15 and 19 per cent ceilings are low compared to the interest rates charged by most successful MFIs in other countries. Given the relatively high unit costs associated with microfinance, it is unlikely that programs operating within these ceilings could achieve self-sufficiency. This is implicitly recognised in that the ceilings are not strictly enforced, and even community organisations funded by government agencies such as GSB and UCDO are in practice not prevented from charging higher interest rates. Nevertheless, the ceilings mean that such organisations are currently operating on an illegal basis, and could be subject to challenge or political pressure. Moreover, the ceilings appear to apply more strictly to some organisations than to others. Given the above, it would seem appropriate to remove all of the interest rate ceilings applying to cooperatives, community organisations and other MFIs. 3.3 Prudential regulation and supervision MFIs in Thailand, whether in the form of credit union cooperatives, unregistered community organisations or NGOs, are subject to few requirements of a prudential nature. Cooperatives are required to have capital divided into shares of equal value. However, there is no specific minimum capital requirement. Nor are there any capital adequacy requirements, liquidity or reserve requirements, or loan loss provisioning requirements on cooperatives engaged in savings and credit activities. Because private and NGO microfinance programs are registered under laws that are not designed with MFIs in mind, they too are not subject to any such prudential requirements. Reporting requirements are also limited. Registered cooperatives are required to provide a monthly trial balance to the provincial office of the Cooperative Promotion Department. In addition, they are required to provide an annual report, audited financial statements and forward program to the departmental headquarters on an annual basis. It was reported that these requirements are not always fulfilled or rigorously enforced, and that some cooperatives may not have their accounts audited for two or three years at a time. They are not required to provide specific portfolio data on their savings and credit operations. In practice, credit unions are much more accountable to their apex body, the Credit Union League of Thailand, than they are to government regulators. Unregistered community organisations, which make up the majority of MFIs, are not subject to any reporting requirements. Private and NGO microfinance programs are also not subject to any reporting requirements of a prudential nature. Programs registering as companies are required to provide an annual report and audited financial statements to the Commercial Registration Department within the Ministry of Commerce. However, they are not required to provide regular portfolio reports as is generally required of financial institutions. The situation for programs registered as associations or foundations is similar. It is generally agreed that it is not appropriate to subject MFIs which do not accept deposits from the public to full prudential regulation and supervision. Moreover, such an approach would be impractical in the case of Thailand. Microfinance in Thailand is dominated by a large number of cooperatives and community organisations owned by their members. There are more than 3,200 such organisations affiliated with GSB and UCDO, and it is not known how many others are in operation. Given the large number of small cooperatives and community organisations, it is unlikely that a government agency could monitor and supervise them in a cost-effective manner. In any case, existing bodies such as the Credit Union League of Thailand, GSB and UCDO are already involved in monitoring many of these organisations. It would therefore be more cost-effective to strengthen the role of existing bodies in monitoring and supervising cooperatives and community organisations, rather than creating a new regulatory structure. This is discussed below. 3.4 Performance and reporting standards of second tier institutions Where community organisations obtain funds from second tier institutions such as GSB and UCDO, they are subject to various requirements. While there is no legal obligation on organisations to meet these requirements, there is a strong incentive for them to do so if they wish to obtain external funding. Moreover, especially in the case of UCDO, these standards are developed largely through a process of consultation with community organisations. To become a member of the GSB program, a community organisation is required to provide detailed information about the activities to be implemented and a list of members. Generally, organisations must have at least 25 members, a management committee selected by group members, and an adequate management system. Organisations are also required to have operated a savings program for at least six months before they can obtain a loan from the bank. When applying for a loan, they are required to provide a financial statement and details of the group’s operating performance during the past six months. Moreover, every month, organisations are requested to provide a trial balance and some basic portfolio data to the bank. It was reported that some community organisations find it difficult to meet this requirement. In these cases, the bank tries to assist them to improve their accounting procedures. The requirements for members of the UCDO program are broadly similar. However, there are some slight differences. Unlike the GSB program, there is no minimum size for groups to become members of the program. In addition to having operated a savings program for six months, organisations are encouraged to operate a loans program from their own funds for three months before they obtain a loan from UCDO. Individual members may repay their loans to community organisations daily, weekly, fortnightly or monthly, but not less frequently than monthly. The requirements in terms of management and reporting are similar to the GSB program. UCDO intends to strengthen these guidelines progressively in the future. For instance, UCDO has been giving some thought to developing standards for loan loss provisioning, in consultation with its community organisations. As well, both programs impose limits on the amount that they will lend to a community organisation. GSB tailors the amount of the loan to the particular circumstances of the group, but with a maximum of five times the group’s own funds. Under the UCDO program the maximum loan amount is set at ten times the value of the group’s own savings, reflecting the fact that loans are also given for housing. However, it is preferred that groups build up gradually to this amount. These limits are of a quasi-prudential nature in that, to some extent, they restrict the degree to which organisations can leverage members’ savings to obtain external resources. However, the limits apply only to loans from GSB and UCDO, and do not prevent organisations from obtaining loans from other sources. As noted above, it may be more cost-effective to strengthen the role of these bodies in setting standards for and monitoring community organisations, rather than creating a new regulatory structure. The establishment of the Community Organisation Development Institute (CODI) should provide scope for consolidating these guidelines as well as requiring more rigorous reporting on the part of community organisations. It is also expected that guidelines will be strengthened in the future. For instance, there are currently no guidelines covering loan loss provisioning, but UCDO has been giving some thought to developing standards in this area. Over time, it would be appropriate for CODI to develop standards covering prudential matters such as loan classification and provisioning, savings facilities, accounting and audit standards, etc. These are all critical areas. CODI should consult widely with community organisations in developing such standards, to ensure that they are appropriate to their needs. These standards should also be applied by the Community Development Department (CDD) and other government programs which lend to community organisations, if these are to continue as separate programs. It would also be appropriate to establish more rigorous reporting standards for community organisations, for example, to ensure that monthly balance sheets and portfolio reports are provided as a matter of course, based on standard formats including for loan loss provisioning. It would then be possible to monitor indicators such as the net asset position, the ratio of savings to loans, and the ratio of liquid assets to savings liabilities, and to initiate remedial action where appropriate. 3.5 Self-regulation There are no overall institutions involved in the coordination or self-regulation of microfinance in Thailand. In the case of the credit union cooperatives, the Credit Union League of Thailand serves as the apex body. The league has around 50 staff at its headquarters in Bangkok, and a number of field staff throughout the country. It assists in the establishment of credit unions, and provides training to credit unions. The league also requires affiliated credit unions to provide monthly reports on their loan portfolios and annual financial statements. As such, it requires considerably more information than the Cooperative Promotion Department, and monitors credit unions more closely. There is some liaison between the league and the Cooperative Promotion Department. Representatives from the department attend the monthly meetings of the board of the league as observers. The league also advises the department when new credit unions have been trained to the point where they are ready to be registered as credit union cooperatives. Over time, it would be appropriate to establish national networks for all types of MFIs operating in Thailand, including community organisations and NGO programs. CODI could take the initiative in facilitating networks, although it is important that they operate independently. Such networks could perform a number of important roles, including information exchange, training, research, policy dialogue with government and donor agencies, and establishing standards for self-regulation. 4 Regulation of banks 4.1 Licensing and minimum capital requirements Cooperatives and other microfinance programs are generally permitted to mobilise savings from their members. However, they are not permitted to raise deposits from the general public. Only institutions with a banking licence are permitted to raise deposits from the general public. Banking licences are issued by the Minister of Finance under the Commercial Banks Act, following screening by the Bank of Thailand. There is a minimum capital requirement of B750 million ($29.6 million) to establish a commercial bank and, unlike the situation in many other countries, there is no network of small retail banks operating at the local level. The number of licences is restricted, and even institutions meeting all the regulatory requirements may not be able to obtain one. The Bank of Thailand is likely to be very cautious in issuing any new licences in the wake of the current financial crisis. Hence, there would not appear to be any scope for MFIs to establish banks, or for the establishment of small microfinance banks operating at the local level. While the current cautious approach by the Bank of Thailand is understandable, 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 important that there be some mechanism to enable small banks to be licensed. It would therefore be appropriate to consider establishing a framework for licensing small banks. There would need to be a realistic minimum capital requirement, and other unnecessary restrictions would need to be removed. 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 longer any 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 Banks must maintain a capital adequacy ratio of 8.5 per cent of risk-weighted assets. This is higher than the Basle Accord standard of 8 per cent, although the current financial crisis may indicate some lack of attention to appropriate risk weighting. In addition, they must maintain a liquidity reserve of at least 7 per cent of domestic deposits plus short-term offshore borrowing. This reserve must be in the form of deposits at the Bank of Thailand. In terms of reporting requirements, commercial banks are required to provide detailed monthly reports on their portfolio and balance sheet, and six-monthly reports on income and expenses. These requirements appear to be quite standard, and would not impose any undue burden on small banks or banks engaged in microfinance. The Bank of Thailand also requires that commercial banks classify non-performing assets into three categories:
Hence, while there is no specific requirement for loans by commercial banks to be collateralised, the loan loss provisions treat loans with collateral more favourably than those without. These provisions should not pose any undue difficulties for any commercial banks making unsecured loans, whether to MFIs through linkage arrangements or to final borrowers, so long as they maintain high repayment rates. On the other hand, recent events are likely to make the banks much more cautious in their lending policies and discourage any loans seen as more risky, even if for small amounts. 5 Summary and recommendations Thailand is at a very different stage of development to most of the other countries in the region, with an annual per capita income of $2,740 in 1995. Reflecting this relatively high per capita income, the incidence of poverty is lower than in most other countries in the region, with an estimated 13.1 per cent of the population below the poverty line in 1992. Nevertheless, the distribution of expenditures is the most unequal of any country included in this study. While the demand for microfinance in Thailand may be less than in most other countries in the region, there is still a need to increase outreach considerably. Microfinance in Thailand is a mixture of government and non-government activity. The three main programs, those of the Community Development Department (CDD), the Government Savings Bank (GSB) and the Urban Community Development Office (UCDO), are operated through government agencies. The CDD program is very much government driven, with a government department establishing revolving funds at the village level to be managed by local government authorities. On the other hand, the programs of GSB and UCDO are managed by independent boards and operate by extending loans to cooperatives and community organisations, which are mainly non-government bodies owned by their members. There is also a very small number of non-governmental organisations (NGOs) and other private bodies engaged in microfinance. It is not clear why there is so little involvement by NGOs other than organisations owned by their members, but this is an issue that warrants further investigation. Arrangements for direct support Ensure independence of overall policy body The GSB and UCDO programs provide support to community organisations engaged in microfinance, in the form of loanable funds and technical assistance. The government is currently planning to establish the Community Organisation Development Institute (CODI), which will take over the overall policy responsibility for the programs operated by GSB and UCDO. In establishing CODI, the government should ensure that it is managed by an independent board and is free from political and bureaucratic interference. Establish specific targets and parameters Neither the GSB program nor UCDO is close to self-sufficiency. Clearly, if the two programs could increase their self-sufficiency, they could increase their outreach to poor clients for any given outlay of government funds. In setting up CODI, the government should establish explicit targets and timetables in terms of outreach and self-sufficiency. Moreover, neither the UCDO program nor the GSB program has an explicit means test. The CODI should adopt an explicit means tests to ensure that it reaches the poor. Consolidate microfinance under central organisation The largest microfinance program in Thailand is the poverty alleviation project operated by the CDD, which provides loans in poor sub-districts. There appears to be little hard data available on the CDD program. Nevertheless, the data that are available suggest that repayment rates are very low, and that the program is operating more as a grant scheme than as a genuine microfinance program. There are also a number of other government programs for microfinance, but it is understood that repayment performance has generally been poor. The government should consolidate the CDD program and other government microfinance programs under CODI. Regulation of non-bank microfinance institutions Register NGOs engaging with microfinance Cooperatives, including those involved in savings and credit activities, are registered under the Cooperative Societies Act 1968. While some community organisations engaged in microfinance are registered as cooperatives, most are unregistered and operate informally. There is also a small number of private and NGO microfinance programs operating in Thailand. There is no specific provision for such organisations to register as NGOs and obtain a legal identity. Most of them are registered as associations or foundations, but these organisational forms are not appropriate for specialist microfinance institutions (MFIs) which provide solely financial services. Nor are company registration procedures designed with the needs of NGOs providing microfinance in mind. The government should establish a simple registration procedure for NGOs wishing to engage in microfinance. Remove interest rate ceilings There is a general ceiling on interest rates applying to loans of 15 per cent, which applies to all bodies and transactions unless they are specifically excluded. The ceiling applies to agricultural cooperatives, unregistered community organisations, and private and NGO microfinance programs. Thrift and credit cooperatives are subject to a separate interest rate ceiling of 19 per cent. It is understood that registered cooperatives generally stay within the relevant ceilings. However, many if not most unregistered community organisations charge higher interest rates. Nevertheless, the ceilings mean that such organisations are currently operating on an illegal basis, and could be subject to challenge or political pressure. Moreover, the ceilings appear to be applied more strictly to some organisations than to others. The government should remove all of the interest rate ceilings applying to cooperatives, community organisations and other MFIs. Develop prudential and reporting requirements MFIs in Thailand, whether in the form of credit union cooperatives, unregistered community organisations or NGOs, are subject to few requirements of a prudential nature. Reporting requirements are also limited, and do not cover the regular portfolio reports generally required of financial institutions. Nevertheless, it is generally agreed that it is not appropriate to subject MFIs which do not accept deposits from the public to full prudential regulation and supervision. Moreover, such an approach would be impractical in the case of Thailand. Where community organisations obtain funds from second tier institutions such as GSB and UCDO, they are subject to various requirements. While there is no legal obligation on organisations to meet these requirements, there is a strong incentive for them to do so if they wish to obtain external funding. The establishment of CODI provides scope for consolidating these guidelines, and requiring more rigorous reporting on the part of community organisations. CODI should develop standards covering prudential matters such as loan classification and provisioning, savings facilities, accounting and audit standards through consultation with community organisations. It should also establish more rigorous reporting standards for community organisations. Establish national MFI networks There are no overall institutions involved in the coordination or self-regulation of microfinance. It would be appropriate to establish national networks for all types of MFIs operating in Thailand, including community organisations and NGO programs. Such networks could perform a number of important roles, including information exchange, training, research, policy dialogue with government and donor agencies, and establishing standards for self-regulation. Regulation of banks Establish a framework for licensing small MFI banks Cooperatives and other microfinance programs are not permitted to raise deposits from the general public. Only institutions with a banking licence are permitted to raise deposits from the general public. There is a minimum capital requirement of B750 million ($29.6 million) to establish a commercial bank, and unlike the situation in many other countries, there is no network of small retail banks operating at the local level. The number of licences is also restricted. There is therefore little if any scope for MFIs to establish banks or for the establishment of small microfinance banks operating at the local level. If banks are to play an active role in microfinance, it is important that there be some mechanism to enable small banks to be licensed. The government should consider establishing a framework for licensing small banks. In this regard, the experiences of countries such as Indonesia and the Philippines may be instructive. References
|
