Development Nepal

Development and Stability

Research abstract

May 6, 2006

Title: Do Microfinance Programs Really Have as High Repayment Rates as they Claim?

Author: Ani Rudra Silwal, PhD student at the University of Maryland, College Park

Take away:

1) Repayment rate – share of a microfinance institution’s (MFI’s) loans that are repaid on time — in Nepal’s microfinance programs is much lower than claimed by MFIs.

2) In order to be effective, MFIs should not expect perfect repayment performance, as some participants miss payments on their loan because of systemic and idiosyncratic risks associated with their investments.

3) Having said that, MFIs should not be too lax on repayment because members may lose trust on the discipline of the MFI and disharmony may arise among the members. In order to maintain a balance between allowing delinquencies but not letting them get out of control, VBs should establish ground rules about repayment and follow them strictly.

Background:

The 1970s and the 1980s saw a surge in the number of MFIs in development countries in response to the weaknesses – such as low repayment rates and high costs of subsidy – of government-run agricultural-credit programs popular at the time. The success of MFIs in extending and recovering millions of loans spurred their replication globally.

The success of a MFI is often measured by the repayment rate, as a high repayment rate is crucial for an MFI to survive without subsidies. However, no consensus exists on how repayment rate should be calculated. MFIs often report repayment rates without outlining the methods used in calculating them. One major consequence of the lack of a clear standard for measuring repayment rates is that there is often a significant discrepancy between the repayment rates reported by different entities for the same MFI.

PACT, for example, has received substantial acclaim for its version of microfinance program – Village Banking (VB) – summarizes the successes of VBs in its promotional brochure by claiming that, “no more than 4% of the groups [had] experienced default”. The final report claims that 95.5% of the groups had no defaulted loans since joining becoming a VB and only 88.1% of the groups had no late payments on current loans. Such reports, however, can be misleading.

In Nepal’s case, it is not uncommon to read performance reports of MFIs with repayment rates above 95%, which are significantly higher than those of government-sponsored credit programs of recent past. But is the repayment rate really that high? This study attempts to answer this question by examining the repayment performance of Village Banking, a popular microfinance program in Nepal introduced in a three-year program in 1999 by PACT, with funding from the USAID. At its height, the program had 125,000 women participants.

Data:

Data used in this paper comes from a field study conducted by the author in summer 2002 in Chitwan, Nawalparasi, Jhapa, and Banke along the East-West highway, around which most VBs are located. Repayment data from the photocopied records of VBs was coded into Excel spreadsheets along with loan size, time taken for repayment, and age of the VB when the loan was taken out. The study examines data for 9 VBs which had 308 members and 1,937 loans. The VBs in the study were divided into two groups according to the frequency of their loan repayment schedule - weekly or lump-sum repayment at the end of the loan period. Except for one group, four of the nine groups in this study that saves every week also require weekly repayment. Repayment data was calculated separately for the two categories because of the difficulty of finding a common definition for repayment rate.

Major findings:

1) Although VBs have very few loans that are uncollectible, a significant proportion of their loans are paid behind schedule. Only 27.2% of loans of weekly VBs and 45.6% of loans of monthly VBs were repaid in complete installments by the end of the loan cycle.

2) These results both corroborate and contradict previous studies on delinquency rates of VBs and other MFIs. Like other studies have found, none of the nine groups in this study had any irrecoverable loans; they had defaults rates of 0%. On the other hand, VBs have very high delinquency rates—the percentage of loans whose (installment or complete) payments were not made on time—unlike what most other studies have found. Recovery rates of 100% can, therefore, mask a completely different reality.

3) Borrowers often have to unintentionally miss loan payments because their investment project failed. Decisions to use financial resources are often made in consultation with other members of the household. For example, women often borrow loans for other members of their household. They often get betrayed by the household members and have to delay loan payments.

4) Lack of a credible threat of penalties also makes it easier for borrowers to default, especially in cases where a borrower has multiple sources of credit from other MFIs—a common scenario in light of mushrooming MFIs in Nepal’s rural areas.

5) Delinquency rates are also high because of a lack of a proper monitoring and record-keeping mechanism. Monitoring every single payment of every single loan in a VB of 25 people is time-consuming. Therefore, VBs end up allowing some members to get away with delinquencies. A VB can monitor its loans only if its officials have an accurate picture of the VB’s loan portfolio, which requires that all the records of the VB are recorded and neatly stored. However, most VBs that the author visited neither keep good records nor store previous records safely. The consequence has been that VB officials cannot track all loans well to identify and follow up on delinquencies.

6) The VBs are not concerned the high delinquency rates. This is because VBs are confident that they will eventually recover their loans. Further, members of VBs consider that the cost of delinquencies are worth bearing for the benefits that they get from the VBs’ existence. VB officials feel that it is not right to be strict on members who miss their loan payments since they see tolerating delinquencies as helping other women in need.

This paper is also recommended for:

1) Analysis of what makes (and what does not make) an MFI effective and sustainable.

2) An example of using both qualitative and quantitative techniques to examine institutions in developing countries.

3) A detailed description of data collection, which might be useful to someone planning to do a field-based research in Nepal.