RISK-SHARING MATCHING AND MORAL HAZARD

In this study we model the endogenous relationship formation between risk-averse principals and agents in a CARA-normal framework. Agents exert unobservable efforts to increase mean outputs and reduce risks. We show that risk-reduction efforts are more important than mean-increasing efforts in determining the matching patterns. Compared to cases without moral hazard, the agency problem in risk reduction induces more positive-assortative matchings.