Empirical Results

Tables 9.2 [ PDF 103.3KB | 1 page ], 9.3 [ PDF 145.6KB | 1 page ] and 9.4 [ PDF 146KB | 1 page ] present the results of estimation of equation (1). Since there are many variables included in the regression to control for individual or village characteristics, the tables report only the main variables of interest: the coefficient estimates on the three variables indicating participation in the microfinance program offered by Khushhali Bank. Each coefficient estimate represents a separate regression – dependent variables are reported as column headers and the independent variables of interest in the five rows. (Note that the independent variables indicating access or participation were included in five separate regressions, but they are reported in one row in the tables for economy of space).

The first set of regressions reported in Table 9.2 [ PDF 103.3KB | 1 page ] look at conventional monetary indicators of poverty. The first outcome variable, monthly consumption per capita, looks at the impact of the program on caloric consumption as measured by expenditure of food items. The items used in calculating this variable correspond as closely as possible to the items used by the government of Pakistan in calculating the official poverty line, although we could only conduct the survey once and so had to rely on respondents recall and could not control perfectly for seasonal variations in consumption. The other items included here are monthly per capita consumption of non-food items, monthly per capita expenditures on health care and annual educational expenditure per child in the household.

Naturally, the core poor have lower overall levels of expenditure on almost all items, as indicated by the statistically significant parameter estimates on the dummy variable for core poor. The dummy variable for Khushhali Bank clients, however, is statistically insignificantly different from zero in most cases, indicating that upon joining the microfinance program, Khushhali Bank clients do not differ significantly from the overall population. Turning to the parameter estimates of interest, the regression results indicate the program does not impact most consumption expenditure measures – almost all coefficient estimates in Table 9.2 [ PDF 103.3KB | 1 page ] are insignificantly different from zero. There is some evidence that participation in the program has a positive impact on educational expenditures for the very poor, as indicated by the statistically significant positive coefficient estimate in column 4. The more loan cycles very poor clients have taken, the higher the household’s annual educational expenditures per child.

Poverty Indicators – Social Indicators

The next set of regressions look at social indicators of poverty: non-expenditure indicators of education and health. The results of these regressions are reported in table Table 9.3 [ PDF 145.6KB | 1 page ].

The coefficient estimates on the dummy variable for the core poor are statistically significantly negative in some cases, indicating this group of core poor are poorer in nonfinancial terms as well: their children are less likely to be enrolled in school and less likely to be vaccinated. However, participation in the microfinance program reverses these trends.

For example, the program is found to have special impacts on children’s education for the poorest borrowers. Although there is some evidence that the probability of their children being enrolled in school may be lower for client households than for nonparticipants, as indicated by the statistically significant negative coefficients reported in column 1 of Table 9.3 [ PDF 145.6KB | 1 page ], for the poorest borrowers these effects are reversed. The longer their participation in the program the more likely children in their household are to be enrolled in school.

Children in the poorest households also reap health benefits. The program positively impacts indicators of children’s health for all borrowers – children in participating households are more likely to get medical treatment for their illnesses and that treatment is more likely to be provided by a trained professional – but the poorest borrowers also benefit from higher likelihood of vaccination, as indicated by the statistically significantly positive coefficient in column 6 of Table 9.3 [ PDF 145.6KB | 1 page ]. Since the microfinance program analyzed here does nothing explicit to promote awareness of health issues, these findings most likely reflect the preference or need of poor households to increase the quality of their health care, especially for their children.

Income Generating Activities

The next set of regressions analyse the impact of the program on income-generating activities run by the poor households: animal-raising, microenterprises and agricultural activities. Since microenterprises are mainly in urban areas, in the statistical analysis a dummy for urban households was included and interacted that with the treatment variables as well as the dummy for the poorest households. The regression results are reported in Table 9.4 [ PDF 146KB | 1 page ].

Participation in the microfinance program yields the most impact for urban households running microenterprises and for very poor borrowers engaged in agriculture. Although there is no evidence of higher sales or profits in animal-raising, urban households reported statistically significantly higher sales and profits for their microenterprises, and strong positive impacts are found for sales of agricultural products, especially for the poorest clients.

Comment(s)

There are [1] comment(s) for this entry. Post a comment.

1. Khan Hidayat Ullah
   (posted 23 November 2006 / 10:14:28 PM)
   
   Econometric Model Selection:
   
   I went through your paper and really liked the approach adopted by you while found your results interesting as well.
   
   I have one comment to make. When it comes the components of microfinance pertaining to the term of loan, e.g., amount, duration and repayment mode, in practice these factors are determined by the both the borrower and lender through negotiation between line staff of MFI and Managers of CBOs. This results into adoption of resolution by members in joint meeting held by field staff MIF and CBO.
   
   Keeping these factors in view, the claim that an explanatory variable relating to terms of micro-credit is exogenous becomes doubtful. This creates endogenity problems in the model and the results which are estimated by this model can not be claimed as unbiased.
   
   In you model you have used variables like amounts of loan and times of loans borrowed in past, which brings in endogenity problem in the model and leads to biased estimates, if I am correct? I would like to know how you addressed this problem in your econometric model. If not what kind of instrumental variable you intend to use in order to deal with situation will.