Secret Report and Optimal Collusion-Proof Mechanism Design: the Case of Moral Hazard

This paper bridges the practice of anti-collusion to the theory of collusion-proof mech-anism design. The role of secret reporting as an effective way of fighting collusion is highlighted through a multi-stage Maskin mechanism. Applying this mechanism to two multi-agent settings under moral hazard: mutually observable actions and privately ob-servable actions, we show that the optimal incentive contracts in the absence of collusion can be implemented in spite of collusion, which implies that preventing collusion entails no costs