Contracting in dynamic games

We study the bargaining problem in the dynamic framework. The classical way of solving a dynamic bargaining problem is to transform the extensive form game into the normal form and then apply the theory of bargaining well developed for normal form games. This means that the parties sign a binding contract in the beginning of the game which defines their actions for the full duration of the game. In this article, we consider the setting where the players monitor the contract as the game evolves. The main purpose of the article is to study conditions under which the players do not have a rationale to renegotiate a new contract at any intermediate time period; i.e., the contract is time consistent. Time consistency restricts the set of bargaining solutions in dynamic games. We will show that time consistency of the contract is guaranteed if the bargaining solution satisfies the controversial independence-of-irrelevant-alternatives property.