Team beats collusion

This paper analyzes optimal contracts in a linear hidden-action model with normally distributed returns possessing two moments that are governed jointly by two agents, who can observe each others' effort levels and draft enforceable side-contracts on chosen effort levels and realized returns. After showing that standard constraints, resulting in incentive-contracts, may fail to ensure implementability, we examine (centralized) collusion-proof contracts and (decentralized) team-contracts. We prove that optimal team-contracts provide the highest implementable returns to the principal. In other words, the principal may restrict attention to outsourcing/decentralization without any loss of generality. Moreover, situations in which incentive-contracts are collusion-proof, thus implementable, are fully characterized