Endogenous Exit from a Stochastic Partnership
Duke University, Durham NC
Investigators
Abstract
This research used the theory of repeated games to deepen our understanding of the endogenous process by which economic relationships survive and fail. It will also help us understand how the prospect of exit creates dynamics within a relationship. Repeated game theory provides some of the most richly applied tools in modern economics. For example, price wars, collusion, and relational contracts have been widely studied as equilibrium outcomes in repeated games. The possibility that a game may end is typically captured in the formal mathematical model through an exogenous probability that the relationship dies. However, in many real world situations people choose when to leave a relationship, such as when a partnership breaks up or when a customer switches its business to a different supplier. This research shows how the prospect of endogenous exit changes incentives and creates a variety of interesting dynamics in a variant of the widely used 'Prisoner's Dilemma' game. Preliminary results demonstrate that partnerships are more profitable when they are less likely to end soon, that older partnerships tend to be more profitable than younger partnerships, and also help us understand why partnerships often 'break down before they break up'. The research results have wide implications for public policy and business practice.
View original record on NSF Award Search →