Cooperation and Incentives in Models with Asymmetric Information

Brown University, Providence RI

Investigators

Abstract

The analysis of incomplete information has had a profound impact on economic theory, especially in the context of non-cooperative models. However, cooperative theory in the presence of asymmetric information is much less settled. This project will study important open problems of cooperation and incentives under asymmetric information, with an eye on applications such as auctions and bargaining. Due to the importance of individual incentives, the research will rely heavily on insights from non-cooperative game theory and mechanism design. When economic agents can cooperate with each other, it is important to consider economic allocations that are immune to changes induced by self-interested groups or coalitions. The idea of coalitional stability is typically formalized through the set of allocations known as the `core,' also connected with the theory of competitive markets. This project is concerned with the theory of the core in an environment with asymmetric information. When a coalition of asymmetrically informed agents forms, it is not clear how much information will be shared among them because it may be in their individual interest to hide some of it. The research will analyze how to formalize coalitional stability in this context. It will provide a way of making endogenous the amount of information pooling among agents, and investigate conditions under which the set of "core" allocations is non-empty. The research will also explore how non-cooperative decentralized mechanisms may be designed to implement socially desirable outcomes. The results of the project will lead to a better understanding of important social, political and economic phenomena, including collusion in auctions and coalition formation in problems of bargaining.