What drives the repayment of agricultural micro loans? Evidence from Nicaragua

We analyze the drivers of the probability of default (PD) and the loss given default (LGD) of a unique dataset consisting of individual agricultural micro loans granted by a Nicaraguan microfinance institution which does not charge interest rates from borrowers according to their creditworthiness. The research method accounts for the clustered structure of the data (loans are clustered by borrowers, loan agents and regions) with crossed random effects. The results show that being married or being of a lower age increases the PD significantly, as do high interest rates. Moreover, the micro loans' economic objective influences the PD. Building an irrigation system appears to be more risky than livestock farming or planting vegetables. The age of the borrower and the duration of the relation with the MFI have a negative influence on the LGD. Furthermore, we do not find any significant gender effect.