Hasta las 5:15 pm.
Abstract: This paper proposes and implements a mixture model to account for the unobserved group heterogeneity when modeling repayment behavior in group lending. We discuss the model properties and identification. We estimate the model using a rich dataset from a group lending program in India. The estimation results support the existence of two different group types: “responsible” and “irresponsible” groups. We find that the effects of the factors driving the repayment behavior differ across types. The model also shows a higher predictive performance than standard probabilistic models, particularly in identifying potential defaulters. We provide evidence supporting the robustness of the estimations.