A Brief Description of the Khushhali Bank Survey

In Pakistan, it is estimated that of 6.5 million poor people who need microfinance services, only about 5% are being served by microfinance institutions. To expand microfinance services to the poor, the Khushhali Bank (KB) was established in August 2000 by 14 private and 2 state-owned commercial banks as a flagship microfinance bank. It became an integral part of the Islamic Republic of Pakistan's Poverty Reduction Strategy and its Microfinance Sector Development Program (MSDP), developed with the assistance of the Asian Development Bank (ADB). Khushhali Bank's primary objective is to provide sustainable microfinance services to the poor in order to reduce poverty and promote economic development through community building and social mobilization.

Through consolidation of several NGOs' microfinance operations, Khushhali Bank shortly became one of the largest microfinance institutes in the country with 31 branches covering 33 districts and US$12 million in disbursements. But real expansion occurred after the government of Pakistan signed a loan agreement of US$150 million with the Asian Development Bank in 2002 to support the operations of Khushhali Bank and to promote the microfinance sector in Pakistan. Khushhali Bank obtained a US$70 million component of this loan for microloans to the poor—particularly to women of the country's rural and urban areas—and a US$10 million component allocated toward institutional capacity building. Another US$70 million component has been allocated to support policy reforms of the microfinance sector in Pakistan.

By the end of 2005, KB had 63 branches and employed 1,576 people. KB reported that in 2003, it provided loans to 100,000 households. It aimed to reach 700,000 households by 2007. The bank's line of products includes short-tenure microloans, up to US$500 for working capital and asset purchase, as well as training and consulting. It does not offer deposit services. Its lending is based on the Grameen model, i.e., it loans to community groups without collateral. To ensure that the loan reaches the target segment (e.g., the poor), KB has limited the loan size to about US$150, an equivalent of 36% of the per capita GDP, an amount in which wealthy people would not have interest. In practice, however, different members of the same household often borrow in tandem, rendering the small loan limit an ineffective deterrent. Khushhali Bank also uses another method of targeting: ranking poor prospective borrowers by tracking the economic status of their beneficiaries. This approach, however, is not adopted rigorously and suffers from subjectivity since participating households generally self-select themselves as poor. Moreover, the very aims of the Khushhali Bank—obtaining group guarantees for repayment and maintaining a close scrutiny of monthly cash flows to determine the repayment capacity of potential clients—can potentially exclude the poor, a priori.

A Khushhali Bank survey was conducted by the Asian Development Bank Institute in May- June 2005. The survey covered 2,881 households, of which 1,416 are KB borrowers and 1,465 nonborrowers. Additional data on loan characteristics such as purpose, duration, installments, and interest rate were provided by Khushhali Bank. When matching survey data and KB data on borrowers, 18 observations were lost due to unknown reasons and thus data from 1,398 client households were used in the final analysis. The survey covered KB borrowers from 11 regions: 8 rural and 3 urban. Table 1 [ PDF 42.7KB | 1 page ] shows that out of 2,881 households, 2,126 (74%) were rural households and 726 (26%) were urban households.

This sampling of rural versus urban households is consistent with Khushhali Bank's portfolio, in which around 75% of Khushhali Bank's lending goes to the rural population. To be representative, the survey design covers the same distribution with 74% of rural households.

Table 2: Distribution of KB Borrowers and Nonborrowers by Region [ PDF 46.8KB | 1 page ]

In Pakistan, due to cultural reasons, extended families are very common. Graph 1 [ PDF 46.8KB | 1 page ] depicts the distribution of household size among survey participants.¹ While households with 6 family members are most common, 16% of all households have 10 or more members. The average number of household members in the sample is 6.5 people.

Table 3 [ PDF 46.8KB | 1 page ] shows that female borrowers consist of 44% of all borrowers in the survey, which is significantly higher than the 30% of women in the overall pool of KB borrowers. This fact is mainly explained by the survey design to give greater weight to women so as to infer the program's impact on gender empowerment. For comparison, the labor force participation rate of Pakistani women in 2005 was only 11%, compared with 49% of men.

From the time of its establishment in 2000 until the survey in 2005, the average loan repayment rate was more than 97%. High repayment rates occur either by good portfolio management or by rapid expansion into new areas. In the latter case, high repayment rates among the first-time borrowers bolsters the average. Indeed, the Khushhali Bank has been expanding very fast. This growth is illustrated by the number of first-time borrowers. Table 4 [ PDF 51.4KB | 1 page ] shows the number of loan cycles broken down by gender. In particular, the survey shows that around 70% of KB borrowers have borrowed only once or are first-time borrowers.

The standard loan size of Khushhali Bank is 10,000 rupees or less. To discern whether multiple family members borrow from the Khushhali Bank at the same time, we turn to the distribution of currently active loans. Table 5 [ PDF 51.4KB | 1 page ] shows that 11% of the households appear to have outstanding loan amounts exceeding 10,000 rupees, indicating that some households can effectively circumvent KB's loan size restriction and borrow more at the same time. Table 6 [ PDF 51.4KB | 1 page ] shows the cumulative loan size by sample households since the establishment of Khushhali Bank in 2000. By mid-2005, 60 households (more that 4% of all surveyed households) have borrowed more that 41,000 rupees.

To investigate if KB achieved its goal of targeting the poor, Table 7 [ PDF 51.4KB | 1 page ] shows that 62% of sample households can be classified as poor using the criterion of the national poverty line in 2005, less than 878.6 rupees of consumption per capita, which is necessary to provide 2,350 calories per day. This ratio of the poor among KB clients is slightly lower than the comparable ratio of the poor among nonborrowers in the sample (64%).

To further investigate if KB reaches the core poor, defined as people spending less than 500 rupees on consumption, which is a little bit more than half of the national poverty line, Table 8 [ PDF 42.6KB | 1 page ] shows that around 21% of all borrowers can be classified as the core poor. This ratio is consistent with the ratio of the core poor in the whole sample. Thus, it could be concluded that KB was effective in reaching the poor.

Download this Discussion Paper [ PDF 198.6KB| 31 pages ].

Post a Comment

We welcome your feedback on this publication. Post a comment. ADBI is not obliged to acknowledge or publish comments and may abridge or edit them before web posting.

Comment(s)

There are [0] comment(s) for this entry. Post a comment.