Intentions and Ex-Post Implementation∗

Intention-based reciprocity is an important motivation for human behavior, and it can be exploited in the design of economic allocation mechanisms. In this paper, we address questions of robustness that arise in the context of asymmetric information about intentions. We propose allocation mechanisms that eliminate uncertainty about the players’ intentions, by making all types of each player equally kind, and we investigate a first notion of ex-post fairness implementation, based on the property that learning about a player’s type does not change the perception of that player’s intention in such mechanisms. We show that efficient social choice functions which provide payoff insurance to the agents can be implemented in this way, with or without voluntary participation constraints.