Coalitional Rationality

This paper investigates the implications of groups or coalitions of players acting in their collective interest in non-cooperative normal form games. It is assumed that players are unable to make binding agreements, and pre-play communication is neither precluded nor assumed. The main idea is that each member of a coalition will con ̄ne play to a subset of their strategies if it is in their mutual interest to do so. This leads to an iterative procedure of restricting players' beliefs and action choices in the game. The procedure de ̄nes a noncooperative solution concept, the set of coalitionally rationalizable strategies. The solution set is a re ̄nement of the the set of rationalizable strategies. In contrast to equilibrium based solution concepts, it is always nonempty, despite the fact that every coalition is simultaneously allowed to \make agreements". It is also robust to the order in which agreements are made. Relations to other non-cooperative solution concepts, re ̄nements of the basic concept and various applications are also o®ered. * Princeton University. I thank Dilip Abreu for his continuous support and encouragement. I am also indebted to Faruk Gul, Wolfgang Pesendorfer, Ariel Rubinstein and Marciano Siniscalchi for many useful comments and discussions in the process of writing this paper. Finally I would like to thank Erica Field, Heluk Ergin, Wojciech Oszlewski, Hugo Sonnenschein, Andrea Wilson and seminar participants of the Princeton Micro Theory Workshop for useful comments and suggestions.