Modeling collusion as an informed principal problem∗

In this paper we address the question of collusion in mechanisms under asymmetric information by assuming that one of the colluding parties offers a side contract to the other one. We develop a methodology to analyze collusion as an informed principal problem. We show that if collusion occurs after the agents accept or reject the principal’s offer, the dominant-strategy implementation of the optimal contract without collusion is collusion proof. In the second part of the paper, we look at a different timing, assuming that the agents’ decision to accept or reject the principal’s offer is taken after collusion, so the agents share their private information before accepting the principal’s offer. On the other hand, we assume that the collusion offer includes a punishment strategy, to be used whenever the other agent rejects the side contract. We establish the conditions that have to be satisfied for a contract to be collusion proof and we show that the optimal contract without collusion is no longer collusion proof. The optimal collusion proof contract is asymmetric, both in transfers and in quantities. JEL Codes: C72, D23, D82, L23