Relative Performance Contracts versus Group Contracts with Hidden Savings

This paper studies the effects of hidden savings on the relative benefits of two optimal incentive contracts, namely, relative performance contracts and group contracts. As an analysis framework, this paper develops a dynamic moral hazard model in which agents can secretly save. The results from the model suggest that hidden savings affect relative performance contracts more than they affect group contracts. In addition, under group contracts, agents rely more on risk-sharing networks and less on own savings than they do under relative performance contracts. To test the model's predictions, this paper uses a unique data set with detailed information on households' characteristics, their choices of loans, and their responses to liquidity shocks. The empirical results confirm that, in the areas where hidden savings problem is likely to be more severe, households are more likely to choose group loans. In addition, the results also show that households with group loans rely more on networks to prevent themselves from future liquidity shocks.