An Empirical Analysis of Microfinance: Who are the Clients?

This paper aims to:

• Test for the effect of self-selection into a borrowing relationship and program placement on estimates of impact;
• Determine how important it is to control for these attributes;
• Use panel data sets of households with access to micro-enterprise over time to test techniques that might better control for selection problems;
• Test for the possible endogenity of loan amounts to prove that there are no statistically significant benefits of microfinance on enterprise profits.

The paper:

• Reviews recent literature on the study of impact;
• Describes the Assessing the Impact of Micro-enterprises Services (AIMS) project, on which its data is based, discussing:
  • The survey respondents;
  • Bias in the data;
  • Empirical tests;
  • Test results and their policy implications;
  • The random effects model.

The paper concludes that:

• Estimates that ignore selection bias will overstate the impact of credit;
• Whether the poor benefit at all from these services is debatable;
• Self-selection into the lending program is a significant problem in the data;
• Benefits to microfinance loans are large and significant;
• Ignoring the endogenity of credit while testing for impact would inject bias into the results;
• There are no significant benefits of microfinance on enterprise profits.