Long-term relationships, group lending, and peer monitoring in microfinance: Experimental evidence

Microfinance is generally associated with high repayment rates. However, it is not clear what drives such high repayment rates? Does the success of microfinance results only from the use of group lending, peer sanctioning or long-term relationships that typically characterize microfinance contracts? In this paper, we aim to contribute to the existing literature by using a laboratory experiment to disentangle the effect of these different potential mechanisms. Our experiment involves both students and real bankers in the role of lenders. We find that group lending alone, in absence of peer sanctioning mechanisms induces free riding and is not sufficient to mitigate the problems of ex ante and ex post moral hazard. In sharp contrast, we find that individualized long-term relationships perform significantly better than either form of group lending with or without peer sanctioning. Consequently, borrowers' final payoffs are lower in the group lending schemes than in the individual lending scheme due to the fact that borrowers receive on average fewer loans from lenders in the group lending schemes. Our study also sheds light on the social cost of peer pressure mechanisms.