We quantify the importance of endogenous peer effects in group lending programs by estimating a static game of incomplete information. Endogenous peer effects describe how one's behavior is affected by the behavior of her peers. Using a rich dataset from a group lending program in India, our empirical analysis presents a robust finding of large peer effects. The preferred model suggests that the probability of a member making a full repayment would be 12 percentage points higher if all the fellow members were to make full repayment compared with a scenario in which none of the other members repay in full. We find that peer effects would be overestimated without controlling for unobserved group heterogeneity and that inconsistencies exist in the estimated effects of other variables without modeling peer effects and unobserved heterogeneity.
