1. Background and relevance of the study:

In Vietnam, alleviating poverty is one of the main targets of the country during its process of development. Data from the Vietnam Living Standard Survey (1993) show that 51% of Vietnam’s population live under the poverty line (Quang, 1996: 12). A report produced by UNDP/ UNFPA/ UNICEF in 1995 indicated that 20%- 40% of Vietnam’s population are poor and 90% of the poor live in rural areas ( Ibid., 12). The report also pointed out some reasons for poverty in Vietnam, such as isolation in terms of geography, society, and knowledge; high risk; loss of crops, diseases, and lack of resources, especially capital.

One way to help the poor escape from the poverty is to provide them with capital. Currently, in Vietnam, there are two main sectors which provide capital to rural areas: the informal and the formal ones.

The providers of credit in the informal sector include relatives, friends, traders and moneylenders. Of the informal credit relatives and friends provide 60% and the remaining 40% are met by others. However, it is easy but too expensive for the poor to borrow from traders or moneylenders. It was estimated that if the poor have to sell their crop before harvest to the traders, the actual monthly interest rate charged can be 10% or more (United Nations, 1996: 71). Moneylenders also charge high interest rates. A survey carried out by Rural Development Service Center (RDSC) in 10 communes of 4 provinces in the North and the Center of Vietnam showed that the monthly lending rate charged by moneylenders in those communes ranges from 4% to 10% (RDSC, 1996).

The formal sector includes the Vietnam Bank for Agriculture and Rural Development (VBARD), the Vietnam Bank for the Poor (VBP), rural shareholding banks (in the South) and People Credit Funds (PCFs- in the North). In 1998, the VBARD provided only 10% of its total loans portfolio to the poor (Thuc, 1998). This small portion of its loans portfolio implies that the VBARD is a non- poor targeting financial institution. In response to the non- poor focus of the VBARD, the VBP was established in 1995 with the purpose of providing the poor with subsidized credit. However, the subsidized credit provided by the VBP is not sufficient to meet credit demand of the poor. The rural shareholding banks and PCFs also operate in rural areas but are limited by sources of fund. Furthermore, their nature also prevents the poor from accessing their credit. As a result, they do not reach a significant number of rural poor.

In summary, in rural areas of Vietnam the poor have limited or little access to credit from the formal sector while the informal sector charges high interest rates. The situation has created room for foreign Non- Governmental Organizations’ (NGOs) operation in microfinance field with the poor as the target group.

In Vietnam, operation of foreign NGOs, bi- and multi- lateral agencies in rural microfinance has satisfied about 10% of credit demand of the poor (Quang, 1996: 16). Though small in comparison with great demand of the poor, the operation of the NGOs has been drawing much attention of rural microfinance experts and specialists. The NGOs lie somewhere between informal and formal sectors. They combine aspects of informal credit with the management system of the formal sector to reach increasing numbers of the poor. However, questions are raised relating the efficiency and sustainability of microfinance programmes carried out by foreign NGOs as the programmes cannot receive continual donations necessary for their long- term operation and existence. Therefore, for further development of microfinance in rural Vietnam it would be relevant to have a study on efficiency and sustainability of foreign NGO microfinance programmes currently carried out in Vietnam.

2. Focus of the study:

The study will focus on efficiency and sustainability of three foreign NGO microfinance programmes carried out in the North and Center of Vietnam in lending the rural poor. These are AAV Commune Bank programme, GRET Village Bank

programme and SCF/ UK S&C programme.

3. Research questions:

The study attempts to answer the following questions:

a/ How efficient are the NGO microfinance programmes in lending the rural poor?

b/ What factors influence their efficiency?

c/ Can they cover their lending cost?

d/ How does savings component of the programmes affect their sustainability?

e/ What policy recommendations can be withdrawn for sustainable development of banking services for the poor by the formal sector, especially by the VBP and by the NGO microfinance programmes?

4. Methodology:

Secondary data are collected to analyse efficiency and sustainability of the selected NGO microfinance programmes by using qualitative and quantitative methods. The efficiency of the programmes will be addressed in terms of their operational cost effectiveness. Their sustainability will be analysed through their ability to cover their lending cost out of their interest income.

Chapter II

Analytical framework

1. Basic concepts used in the study

In this study, NGOs refer to foreign NGOs established in foreign countries other than Vietnam, coming to Vietnam in the aim to conduct development and umanitarian activities without profit- making purposes. These organizations operate outside Vietnam’s regulatory system and target the disadvantaged population of the country.

In Vietnam, many NGOs operating in microfinance field work in partnership with mass organizations, especially the Vietnam's Women Union. In the field, they act as financial intermediations for the poor by giving loans and receiving deposits. However, the most apparent difference between them and financial intermediations is that they operate outside the regulatory framework and supervisory system that establishes the guidelines for0, and monitors formal financial institutions in the country. They structure their financial services on the basis of either established guidelines followed by financial intermediations or guidelines that they define themselves. As a result, they remain accountable to their donors rather than to the country's supervisory system.

Microfinance is the activity of lending poor people what they need to invest in their self- employment and it is for them an escape route from poverty (Getubig et al., eds., 1997).

Initially, microfinance programmes were credit- based, that is to mainly provide the poor with credit. Today they go beyond credit and typically also offer savings services as well as other non- financial services such as insurance and family budgeting services.

In general, efficiency represents the relationship between input factors of scarce resources and output factors of goods and services. There may be technical efficiency and economic efficiency. Technical efficiency represents the relationship in physical terms, which is the ability of productive equipment in making a high quality product regardless of production costs. Economic efficiency represents the relationship in cost terms, which shows the ability to produce a unit of a good or service at the lowest production costs.

In rural microfinance, NGOs provide the poor with financial services, mainly credit. Their operational efficiency can be measured in terms of total operational cost per average total assets. The lower the ratio, the more operationally efficient the microfinance programme is. In the context of this study, efficiency of lending activity of the programmes is taken into account. Therefore, lending operational cost per unit of average loans outstanding will be used.

Sustainability covers many aspects, such as management capacity of staff, clarity of ownership, legal status. However, the present study is only concerned with financial aspect of sustainability.

Financial sustainability is defined as a stage of financial operations where all costs of the lender are fully met from the interest charges, and where such charges are not subsidized, partly or fully, from outside sources (Seibel, 1993: 49). Financial sustainability also comes with the ability to earn a reasonable profit on equity so that the lender can assure itself of loanable funds mobilised from its members/ depositors.

Lending cost of a financial institution in giving loans to its clients comprises cost of funds for lending (financial cost), operational cost and risk- related cost.

NGOs normally lend their borrowers with three sources of funds: grants from donors (donated funds or donations), soft loans and members/ borrowers savings. Donations are usually the greatest initial source of funds for lending of the NGOs. As they are granted, no payment has to be made by the NGOs for the use of the funds. However, these NGOs have to keep the value of the donated funds against devaluation due to inflation. As a result, the cost of the donated funds for lending (CD) is determined by the inflation rate (INF) and the amount of donated funds (AD), or in formulating it we have:

In exchange for the use of the soft loans for onlending the NGOs have to pay interest. The cost of this source of fund for lending (CL) is the amount of soft loans (AL) multiplied by the interest rate (IL) they pay or

The NGOs often combine credit and savings activities, therefore, savings is one of their source of funds for lending, for which they pay savings interest rate (IS). As a result, the cost of funds for lending incurred by the NGOs also include the interest (CS) they pay their savers for the amount of savings mobilized (AS) or

The total cost of funds for lending (CF) incurred by the NGOs is, therefore,

Operational cost (CO) refers to all costs incurred in initiating, recording, handling, and collecting loans. With an increase in the loan size the average operational cost (AO) will decrease. Therefore, a negative relationship between operational cost per loan unit and loan size is evidenced.

Risk- related cost (CR) is the amount necessary to cover possible loss from default. It is determined by the amount of loans outstanding (L) and the default rate (DR) or

The total lending cost(LC) is then expressed as:

However, in order to achieve financial sustainability the NGOs should also build up equity or institutional capital. The cost of doing this is to be covered by interest income. Therefore, not only the lending cost but also the cost of building up equity should be taken into consideration if the NGOs aim for full financial sustainability. Once built up and lent out, the value of the equity should be preserved against inflation, which is considered financial cost.

Interest represents the cost of the use of money. It is the amount paid by a borrower to a lender in exchange for the use of the money for a certain period of time. Interest paid by borrowers enables lenders to cover the cost of lending and generate a reasonable profit. For microfinance institutions, interest income partly determines whether the institutions are dependent on donor funds or able to maintain itself with its own earned income. Therefore, setting appropriate interest rates is crucial for the sustainability of microfinance institutions.

In microfinance, savings mobilization is generally an integral part of NGO microfinance programmes. It plays an important role in the process for NGO microfinance programmes to achieve their long- term sustainability for three reasons. First, savings provides funds for onlending, reducing the dependence on external sources of funds. Second, savings serves as a source of collateral in the event of loan default by the members. Third, savings creates the members’ sense of ownership of and commitment to the programmes, thereby improving the overall performance of the programmes.

Padmanabhan (1988) has pointed out three main groups of factors affecting savings mobilization. These are ability to save, opportunity to save and incentive to save. The ability to save refers to the availability of surplus income generated by the rural poor. The opportunity to save addresses the presence of secure and accessible institutions. The incentive to save means a condition where the sacrifice involved in saving will bring a better future reward. This is a very important factor determining saving behavior of rural people.

"PERLAS" or "PEARLS" system is a set of financial ratios, which has been developed and used by the World Council of Credit Union, Inc. since 1990. Each letter of the word "PEARLS" measures key areas of Credit Union's operations: Protection, Effective financial structure, Asset quality, Rates of return and cost, Liquidity and Sign of growth. The system comprises 38 separate ratios and measures, which are applicable to all types of credit unions.

In Vietnam, the system was considered to be too complex in view of the much simpler financial structure of the typical community financial institutions. To improve the quality of the operations of such institutions and to assist them to achieve sustainability, AAV, MRDP and WOCCU have been working together to develop a set of "prudential standards" for these institutions. "Prudential standards" are a defined set of targets covering key areas for the financial management of community financial institutions (CFIs), including loan loss provision, financial structure, asset quality, rates of return, liquidity and growth rates. An initial set of 14 ratios has been selected from the "PEARLS" categories.

Chapter III

Profiles of the selected programmes

1. Profile of AAV Commune Bank programme:

Figure 1: AAV Commune Bank programme- Scheme structure.

District Women Union Banking Service

(DWUBS)

(CB) (AAV)

- Lending service: The main source of funds to lend group members is grant from AAV to DWUBS which is then on- lent to CBs and group members. CBs charge interest rates of 1.5%- 1.8% per month on loans to group members. The amount of loans is 25 times the member’s smallest monthly deposit amount of 3 successive months, but not larger than VND 1,000,000.

Prior to May 1999, loan term was extended to 2 years. Now it is reduced to 1 year. Repayments of both principal and interest are made in regular monthly installments for 5% of the principal and total monthly interest. No collateral is required for when loans are disbursed. However, the guarantee for loan repayments is member’s savings and peer pressure of group membership. Loan purpose is determined by borrowers themselves.

- Savings service: Savings mobilisation is another service provided by AAV. Savings is a compulsory precondition for loan disbursement. Regular voluntary savings is encouraged. No minimum savings amount is stipulated. This is to encourage participation of very poor people in the community. AAV offers members full and easy access to savings, which aims to consolidate confidence in savings services provided. Interest rate paid for savings is 1% per month, which is similar to that of formal financial sector.

GRET’s Village Bank scheme is organised as in the diagram below:

Figure 2. GRET Village Bank programme- Scheme structure

Project Local Authorities/

VBA/ VBP- GRET Women Union

Village Bank

Credit Committee

Credit Agent President Cashier

- Lending service: Each Village Bank borrows from VBA up to VND 33,000,000 for 3 years at monthly interest rates equal to 50% of interest rates it charges its borrowers. Funds from the VBA/ VBP is Village Bank’s main source of fund for lending. Another source is savings from borrowers.

The Village Bank lends group members at interest rates equal to short-term interest rates on loans for farmer- households, which are set by the banking system.

Loan duration is determined by group members themselves but not longer than 12 months. However, as loans are disbursed in cycles, all members have to repay at the end of the cycle though their loans have not matured yet.

Loan amounts range from VND 200,000 - VND 500,000, which is 20 times the member’s initial compulsory savings amount. The maximum lending amount is VND 500,000.

Loan purpose is freely chosen by the borrowers provided that it is productive and income- generating. The loan purpose must be informed to the Credit Committee and all members within the group.

Repayment of principal is made at the end of loan terms while interest is paid monthly. Repayment of principal prior to maturity is not encouraged though acceptable. Borrowers who wish to repay loans before due date have to pay 100% interest for that current month even though just one day of that month has passed. Furthermore, withdrawal of savings can only be possible at the end of loan term when all group members have fulfilled their repayment obligations. Members are tied to commitment in their joint- liability groups until the end of the loan term. Such discouragement exists as Village Banks lend in cycles and it would be difficult for the banks to find new borrowers to revolve loans paid back prior to maturity. As a result, losses may occur to the banks, damaging bank’s performance.

- Savings service: In the programme, no regular savings is required but initial savings is compulsory for all members who apply for loans. The amount of savings is 5% of registered loans. Savings earns interest rate of 1% a month. Repayment of savings and interest is made at the end of the lending cycle when all group members have paid all their loans principal and interest. Therefore, savings requirement is also a condition to guarantee repayment by group members. Without full repayment of any group member, other members in the group are not allowed to withdraw savings or receive savings interest.

a/ Scheme structure: The scheme is organised as in the chart below:

Figure 4. SCF/ UK S&C programme- Scheme structure

Hamlet 1 Hamlet 2 Hamlet3

S&C Group S&C Group S&C Group S&C Group

- Lending service: The programme is based on the principle of small, short- term, repeated loans provided through a group mechanism for productive uses.

The first loans are small because they suit needs and capacity of the poor. When they have gained some investment experience and repaid the first loans, loans in the next cycles will be expanded to a larger amount. The maximum loan size for each borrower in the first cycle is VND 500,000 and for all subsequent cycles is VND 1,000,000. Loan duration depends on production cycles, which are normally short (less than 12 months). Repeated loans are both an incentive for full and timely repayment and a way to help the poor to gradually get out of poverty. Group mechanism provides a social guarantee instead of collateral, which the poor normally do not have but usually is required by banking system when giving loans. Productive purposes ensure that the borrowers are able not only to repay the loans and interest, but also to earn profits for further investment. No property collateral is required from borrowers.

The CMC lends borrowers at the interest rate of 2% per month. Repayment of loans and interest is made in monthly installments. Prior- to- maturity repayment is encouraged.

- Savings service: All members must save regularly in their groups, beginning 3 months before any loan is disbursed and continuing throughout their participation, regardless of whether they have a loan. Minimum monthly amount of savings is VND 4,000- VND 5,000. Additional voluntary savings is encouraged. Savings earn interest rate of 1.4% per month and the interest is added to the initial savings when withdrawal is made. No interest is paid if savings is withdrawn prior to the agreed time.

Savings mobilised then can be relent to either members or non- members at interest rate of 2% per month or deposited at a bank. Loans from savings have the duration of up to 6 months with the maximum amount of VND 300,000. Loans can be used not only for productive purposes, but also for consumption, often in the cases of emergencies, family events. Repayment of loans from savings is made once at the end of the loan term.

Chapter IV

Efficiency of the selected programmes

Lending operational cost per unit of loans outstanding shows how efficient the rural microfinance programmes are in lending the poor. The lower the ratio, the more cost- effective the programme is. The ratio is calculated on year- to- year basis to see how it changes over time with changes in average loans outstanding during a year. Operational cost includes money paid to staff who are responsible for the operation of the organisations in form of salary, allowance or remuneration and money paid for stationery necessary for their work. Depreciation of fixed assets or rent is excluded from operational cost. Lending operational cost is operational cost allocated for lending activity of the organisations, which is determined by portions of staff working time consumed in the activity. The average loans outstanding during a year is the average of total loans outstanding at the beginning and loans outstanding at the end of that year.

Graph 1 shows lending operational cost per unit of loans outstanding of the programmes.

Without subsidies from donors GRET Lung Hoa I Village Bank is the least cost- effective one in lending the poor, incurring the highest lending operational cost per unit of average loans outstanding. AAV Phuong Dong Commune Bank appears to be the most efficient one in lending the poor. Though, a similar trend of development is evidenced in lending activity of AAV Phuong Dong Commune Bank and SCF/UK Cam Hoa Commune Management Committee. They both experienced increasing lending operational cost per unit of average loans outstanding in their first 3 years of operation. After that the ratio decreased with increase in average loans outstanding.

This section attempts to identify some factors affecting operational efficiency of the selected programmes. Among many factors the average loans outstanding and the length of time in operation are expected to influence the ratio of operational cost per unit of loans outstanding.

The average loans outstanding directly affects the ratio of operational cost per unit of loans outstanding. An increase in the average loans outstanding implies an increase in total operational cost that the programme has to incur because larger loans require more time and more staff to monitor, handle and collect the loans. However, it is expected that the rate of the increase in the loans outstanding would be larger than the rate of the increase in the total operational cost. Therefore, the larger the loans outstanding, the lower the ratio of operational cost per unit of loans outstanding.

The length of time in operation also influences the operational efficiency of the organization. It is expected that the longer the organization is in operation, the more cost- effective it is.

In the light of these considerations, the ratio of lending operational cost per unit of loans outstanding is specified as follows:

where:

LOCL denotes lending operational cost per unit of loans outstanding

L denotes average loans outstanding (VND)

T denotes the length of time in operation (year). It is a dummy variable.

T = 0 if the programme is in operation for less than 3 years, and

T = 1 if otherwise.

The Ordinary Least Square (OLS) method is applied to quantify the influence of the two factors on the ratio of lending operational cost per unit of loans outstanding, using data on lending operational cost, average loans outstanding and the length of time in operation of the selected programmes. The regression results follow.

*******************************************************************************

Dependent variable is LNLOCL

13 observations used for estimation from 1 to 13

*******************************************************************************

Regressor Coefficient Standard Error T-Ratio[Prob]

CONST 10.2425 1.3824 7.4094[.000]

LNL -.47888 .074421 -6.4347[.000]

T .48480 .16876 2.8727[.017]

*******************************************************************************

R-Squared .80797 F-statistic F( 2, 10) 21.0379[.000]

R-Bar-Squared .76957 S.E. of Regression .25283

Residual Sum of Squares .63923 Mean of Dependent Variable 1.3102

S.D. of Dependent Variable .52669 Maximum of Log-likelihood 1.1346

DW-statistic 1.5478

*******************************************************************************

Diagnostic Tests

*******************************************************************************

* Test Statistics * LM Version * F Version *

*******************************************************************************

* * * *

* A:Serial Correlation *CHI-SQ( 1)= .17891[.672]*F( 1, 9)= .12559[.731]*

* * * *

* B:Functional Form *CHI-SQ( 1)= .54126[.462]*F( 1, 9)= .39100[.547]*

* * * *

* C:Normality *CHI-SQ( 2)= 1.1685[.558]* Not applicable *

* * * *

* D:Heteroscedasticity *CHI-SQ( 1)= .86676[.352]*F( 1, 11)= .78580[.394]*

*******************************************************************************

A:Lagrange multiplier test of residual serial correlation

B:Ramsey's RESET test using the square of the fitted values

C:Based on a test of skewness and kurtosis of residuals

D:Based on the regression of squared residuals on squared fitted values

The model passes all diagnostic tests. The F- tests indicate reasonable explanatory power of the model. The coefficient of determination (R²) is high (0.81).

In the equation, the determinants of operational cost per unit of loans outstanding are found to be statistically significant.

The negative sign for the variable L, as expected, shows that the larger loans outstanding is, the lower the ratio of operational cost per unit of loans outstanding. This means that the microfinance programmes can increase their operational efficiency with increases in their loans outstanding, therefore, gain economies of scale in lending the poor. With maximum loans amount specified by the programmes, the increases in the total loans outstanding imply greater outreach of the programmes. As a result, the operational efficiency of the programmes is positively related to their outreach and the total loans outstanding. Since the dependent and independent variables (OCL and L) are measured in logarithms, the regression coefficient is interpreted as elasticity. Ceteris paribus, 1 percentage increase in the loans outstanding reduces operational cost per unit of loans outstanding by 0.48%.

The positive sign for the variable T indicates that the length of time in operation contributes to the increase in operational efficiency of the programmes. After three years of operation, the programmes incur decreasing operational cost per unit of loans outstanding with increases in the number of years they are in operation.

In summary, the regression results show that operational efficiency of the microfinance programmes examined increases with increases in loans outstanding and after three years of their operation.

Chapter V

Sustainability of the selected programmes

1. Lending cost per unit of loans outstanding and lending interest rates:

The purpose of this section is to calculate lending cost per unit of loans outstanding of the microfinance programmes in comparison with interest rates they charge their borrowers to evaluate their financial sustainability.

Initially, Phuong Dong CB has two sources of funds for lending the poor. The first one is loans from the DWUBS. The second one is savings of members. Therefore, its cost of funds for lending are the sum of interest paid to DWUBS for loans and to members for savings. When the institutional capital of the Commune Bank is accumulated it becomes the third source of fund for lending. The cost of this fund is the amount needed to preserve its value against inflation, which is calculated by multiplying the inflation rate by average institutional capital during a year.

Operational cost of Phuong Dong CB has been calculated in the previous Chapter, including allowance, stationery for GHs and salary, stationery for CB staff.

In regard to its risk- related cost, delinquency does matter. The matter is viewed in light of the PEARLS system. In order to achieve full financial sustainability, the microfinance organizations should be able to cover 100% of loans delinquent for more than 12 months and 35% of that delinquent for less than 12 months.

In general, the bank has performed well in covering cost of funds for lending and operational cost. Most of subsidies was eliminated, except for cost of expatriate advisors. However, delinquency problem remains even though its lending interest income can provide adequate provision for loan loss due to delinquency.

Lending cost of Lung Hoa I Village Bank is now made up by two components: Cost of funds for lending and operational cost. Cost of funds for lending is the sum of interest paid for loans from VBA/VBP and for savings of borrowers. Operational cost of the Village Bank was calculated in Chapter IV.

The graph shows that with no subsidies from the donor for operational cost (salary for the Credit Agent) the Village Bank cannot cover its lending cost with interest income. Therefore, it is not financially sustainable.

SCF/UK S&C programme lends borrowers with two initial sources of funds: the donated fund and savings fund of members. Later, when some surplus is accumulated from lending interest income, one more source of fund for lending is utilized, which is the capital growth fund. Since the second year of operation the Commune Management Committee has been able to use accumulated capital growth fund to lend borrowers. The value of this fund should also be preserved against inflation. Therefore, the cost of funds for lending of the Commune Management Committee is the total of interest paid on savings and inflation expenses on donated fund and capital growth fund.

The operational cost has been calculated in the previous chapter. The loan- loss reserve fund is to cover any loss from loans default. As Cam Hoa Commune Mangement Committee’s default rate has always been zero, therefore, its cost of loan loss is zero. In the first year of operation Cam Hoa Commune Mnagement Committee experienced a negative spread between lending interest rate and lending cost per unit of loans outstanding. It could not cover all types of costs of lending necessary to assure its sustainability. Only operational cost, savings interest, loan-loss reserve and a portion of inflation expenses were covered by interest income. Income generated by the CMC was not sufficient to preserve the value of the loanable fund against inflation. Financial sustainability was not attained due to very high rates of inflation.

In the following years the CMC has attained positive spreads between lending interest rates and lending cost per unit of loans outstanding. It has been able to cover all lending costs out of interest income. Devaluation of loanable fund has been eliminated. The CMC is, therefore, financially sustainable.

This section will examine savings component of the selected programmes toward their sustainability. As a source of funds for lending, for the programmes savings mobilization necessitates consideration of interest rates and amount of savings mobilized. These two issues will be analyzed under the PEARLS system.

Savings interest rates offered by the programmes should closely relate to inflation rates of the economy. Table below gives information on inflation rates and savings interest rates offered by the programmes over their time of operations.

Source: a/ Vietnam’s General Statistics Office

From the table it can be seen that all the programmes have been offering savings interest rates higher than the inflation rates. Therefore, real return on savings of members/ savers has been ensured against inflation. Under the PEARLS system this is up to the standards. However, the rates offered are not flexible enough, which may not be good for programmes from financial perspectives. Keeping high savings interest rates would increase costs of loanable funds, which the programmes have to incur with increasing amount of savings mobilized.

Table 2 shows growth rate of savings of three organizations being examined.

Table 2. Savings growth rates by the schemes

Of the three organizations GRET Lung Hoa I Village Bank has the weakest savings mobilization. This weakness results from the programme's policy. A variety of limitations is set up by the programme in each its cycle of loans disbursement. First, limit is imposed upon the number of members. Only about 70 people can take part in the programme for each cycle of loan disbursement. Second, loan size is limited to VND 500,000 for each borrower.

Unlike GRET Village Bank programme, AAV Commune Bank programme and SCF/UK S&C programme conduct different policies toward savings and credit activities. No limitation on number of members is imposed by these two programmes. Two types of savings services are offered: compulsory and voluntary. As a result, they register high growth rates of savings.

In summary, of the three programmes examined GRET Village Bank programme does not conduct policies emphasizing savings side. AAV Commune Bank programmes and SCF/UK S&C programme have paid attention to their savings activities. However, AAV policy pertains to considering savings as a guarantee for loans while SCF/UK policy pertains to building savings habit for the poor with regular small savings amounts.

In the GRET model no additional savings is required or encouraged except for initial compulsory savings, which is 5% of loans amount. As a result, during a cycle of loans disbursement the programme has been able to finance only 5% of its loans portfolio with member's savings. Under the PEARLS system the programme appears to have an extremely weak savings component because the ratio of 5% is far lower than that of 70% - 80% targeted by the system.

Savings components of AAV Commune Bank programme and SCF/UK S&C programme are stronger, which is shown in Table 3.

Though more and more of the loans portfolio of the two programmes has been financed by savings mobilized from members, their savings components are still weak as compared with those targeted by the PEARLS system.

Currently, VBARD lends at 1.15% per month for short-term loans and 1.4% per month for medium-term loans, while AAV Phuong Dong Commune Bank lends at 1.8% per month and SCF/UK Cam Hoa CMC lends at 2% per month. As a result, NGO microfinance programmes are under pressure to reduce lending interest rates. Is the pressure justified if the programmes are to attain financial sustainability? To answer the question reasonable lending interest rates for NGO microfinance programmes in Vietnam have been considered towards their sustainability under the PEARLS system. Conclusions follow:

- Current lending interest rate offered by AAV Phuong Dong Commune Bank still does not assure for the bank to achieve full financial sustainability. Therefore, it is not justified to force the bank to reduce its lending interest rate.

- At present, GRET Lung Hoa I Village Bank lends at interest rate of 1.2% per month, which is far bellow what should be charged (2.84% per month). As a result, it is necessary for the bank to consider increasing lending interest rate if it is to achieve full financial sustainability.

- SCF/ UK Cam Hoa Commune Management Committee could reduce its lending interest rates if its savings interest rate were lowered to 1% per month.

Chapter VI

Conclusions and recommendations

Operational efficiency of NGO microfinance programmes operating in rural Vietnam has been measured in terms of lending operational cost per unit of average loans outstanding. The lower the ratio the more cost- effective the programmes are in dealing with the poor. Of the three programmes examined, GRET Village Bank appeared to be the least cost- effective programme with the highest ratio of about 8.5%- 9% if the Village Bank had to pay for its Credit Agent. AAV Phuong Dong Commune Bank was the most efficient programme.

In the early stage of operation two programmes (AAV and SCF/UK) experienced increasing lending operational cost per unit of average loans outstanding from 1.97% to 3.68% and from 2.69% to 3.83%, respectively. After that they gained economies of scale with increasing in average loans outstanding. It took 3 years for the programmes to achieve this. After three years of operation AAV programme had lending operational cost per unit of loans outstanding of 3.68%, which was similar to that of SCF/UK programme after the same period of operation.

The ability of the programmes to cover their costs of lending is measured in terms of spreads between their interest rates and their lending cost per unit of average loans outstanding during a year. However, subsidies from donors do matter.

There are three main components of lending cost: cost of funds for lending, operational cost and loan- loss reserve. Cost of funds for lending incurred by AAV Commune Bank and GRET Village Bank includes interest paid for borrowed funds and interest paid for members/savers. SCF/UK S&C programme lends the poor with two sources of funds: donated funds and savings of members. Therefore, its cost of funds for lending is a total of inflation expenses on donated funds and interest paid on savings. Later when some surplus is generated and becomes a source of fund for lending, inflation expenses to preserve the value of this fund will be deemed as cost of funds for lending incurred by the programmes. Operational cost of the programmes includes salary for staff, stationery and other operating expenses. Loan- loss reserve is the income of the programmes needed to replace funds lost due to loan defaults.

Since the first year of operation AAV Phuong Dong Commune Bank has been able to cover its lending cost with interest income. GRET Village Bank registers negative spreads between its lending interest rates and lending cost per unit of average loans outstanding. It can only pay cost of funds for lending and a part of operational cost. SCF/UK S&C programme has been able to cover its lending cost since the second year of operation. In the first year of operation higher- than- expected inflation rates resulted in a negative spread between lending interest rates and lending cost per unit of average loans outstanding. This means the programme could not pay all its lending cost. In the following years, the programme has been able to cover all costs of lending.

It was also found out that it is not justified to force AAV Commune Bank to reduce its lending interest rate while SCF/UK Cam Hoa CMC could reduce its lending interest rate if its savings interest rate were lowered to 1% per month. To be sustainable GRET Village Bank should charge lending interest rate of 2.84% per month instead of present rate of 1.2% per month.

Savings mobilization is an integral part of S&C programmes. It determines whether the programmes are independent from external sources of funds for lending.

All the programmes examined have been able to offer savings interest rates higher than inflation rates, ensuring real return on savings for members/ savers. This is a big incentive for people to save. However, their weaknesses are that they have not been able to mobilize voluntary savings and their ratios of savings/ average loans outstanding are still low in comparison with that targeted by the PEARLS system.

For sustainable development of the rural sector, creation of self- sustaining rural financial system should be the primary aim of government's rural financial policies. In Vietnam, for the sake of poverty alleviation the VBP has been established. However, the sustainability of this bank in lending the poor is questioned. One of the decisive factors affecting its sustainability is that whether the bank can preserve and expand its funds for lending.

As an intermediation between the formal sector and the poor, operation of GRET Village Bank has proved that the lending rate of 1.2% per month is not sufficient for the bank to become sustainable, let alone 0.8% per month. The lending rate is suggested to be 2.84% per month.

Performance of NGO microfinance programmes examined shows that the poor can borrow and repay at high lending rates up to 2% per month. This suggests that the VBP can lend at interest rates higher than its present ones.

In summary, operation of the three programmes examined suggests that for sustainable development of microfinance for the poor, the VBP should give up supplying subsidized credit and move to market interest rates, which has been proved affordable for the poor.

The above findings show that the programmes are efficient in operation and can achieve full financial sustainability. However, one of the constraints to full sustainability is that the programmes have not been able to have loan funds fully capitalized from savings or raised at commercial interest rates.

Savings component of the programmes is weak under the PEARLS system. Most of their loanable funds is from the donors. They have no access to the formal sector for funds for lending, except for GRET Village Bank programme. However, the Village Bank can only borrow from the VBA/ VBP provided that the donor deposits a certain amount of money with the banks as guarantee.

Overall, the programmes are operating independently from the formal banking system. They operate with lack of legality, which has both positive and negative impacts on their work with the poor. On the positive side, the programmes have independence to set up, experiment, develop and flourish with no interference from the Central Bank or any other regulatory body. They can set their own interest rates, devise their own credit delivery mechanism and accounting practices. In general, they have freedom to work in the way that they feel best which may have positive impacts on poverty. On the negative side, however, the lack of legality has hampered the programmes’ operation in some ways. Most crucially it has limited their ability to mobilize savings deposit, especially from their non- members. The programmes (GRET) feel uneasy about mobilizing savings deposit for on- lending and do not put effort into it. Lack of legal status also makes it less secure for the savers to deposit. Furthermore, the lack of suitable legal identity makes it impossible for the programmes to establish a "track record" with the formal sector, thereby precludes their access to funds from them.

Presently, the S&C programmes in Vietnam mostly work in partnership with mass organizations, especially the Women’s Union. The role of such organizations acts as a legal protection for operation of the programmes, which results in the lack of legal autonomy of the programmes.

In order for the programmes to provide financial services for the poor in a sustainable manner, they must become an integral part of the formal financial sector of the country. In case of Vietnam, it is necessary to institutionalize the NGO S&C programmes, so that they can work with the poor within a specified legal framework.