Cooperation Dynamics in Repeated Games of Adverse Selection∗

We study dynamic games with private information. After any history, signaling reveals information that helps players coordinate their future actions, but also makes the problem of predicting the informed player’s actions harder for the uninformed player. In equilibrium, the informed player may play aggressive or uncooperative actions, but his partner only tolerates a certain number of such actions. We discuss several applications of our results: We explain the cycles of cooperation and conflict observed in World War I, show that price leadership and unilateral price cuts can be part of an optimal signaling equilibrium in a repeated Bertrand game with incomplete information, show that communication between cartel members may be socially efficient in a repeated Cournot game, and study the optimal organizational design when direct communication leads to delays in decision making. Finally, we explore a model with frequent interactions and show that when types are more persistent, informational frictions are smaller.