Conditionally-repeated prisoners dilemma∗

We augment a two-player normal form game by allowing each player the op-tion to wait. Given the other player does not also exercise their option to wait, the player that waited can adopt a best-response to the action chosen by the other player. As the number of potential (waiting) periods is extended the game can switch from a prisoners ’ dilemma (in which both parties have a dominant strategy to wait) to a mixed strategy (in which both parties move immediately with some probability) and again return to a prisoners ’ dilemma. We show that for some parameter values the outcome in the n-period game converges to the outcome of the symmetric equilibrium in the infinite horizon game. This is not always the case; for some parameter values the dynamic model can behave chaotically. Applications of the model include bargaining with deadlines, investment in joint projects, production in teams and busines