Long-run determinants of moral hazard in microfinance: a study of group lending programs from Malawi using panel data

Abstract

This paper investigates the determinants of moral hazard from a panel data where the individual time series have unequal length (unbalanced panel data). The estimation is done using the XTPROBIT procedure in STATA. The data used is group level panel data from 99 farm and non-farm credit groups of the Malawi Rural Finance Company in Malawi. Results reveal that, contrary to theory, peer selection and peer pressure are not important in mitigating moral hazard in the long run. Results further indicate that, in the long run, social ties, peer monitoring as well as the number of loan cycles have a significant effect on the likelihood of incidence of moral hazard among borrowers.

The implications from these findings are that while focusing on peer monitoring, and social ties, MRFC cannot rely on peer selection and pressure to reduce the incidence of moral hazard. Instead, MRFC has to continuously appraise financial services needs for borrowers to meet their changing and growing demand as they repeatedly borrow.