Repayment Performance in Microcredit Programs: Theory and Evidence

Abstract

This paper revisits the conventional wisdom on the determinants of the success of microcredit programs. The paper first develops a simple moral hazard model of borrowing in a group lending context, and then tests this model using data from a survey conducted in Bangladesh in 1991-92. One of the predictions of the moral hazard model, regarding the effect of group size on repayment performance, is rejected by the empirical evidence. The paper then develops an alternative approach, which is consistent with the observed effect of group size as well as the other empirical results.