Files in this item



application/pdf9211001.pdf (2MB)Restricted to U of Illinois
(no description provided)PDF


Title:Understanding coordination failures
Author(s):Straub, Paul Gregory
Doctoral Committee Chair(s):Villamil, Anne P.
Department / Program:Economics
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Economics, General
Economics, Theory
Abstract:Many recent macroeconomic models exhibit multiple Nash equilibria that can be Pareto ranked. Many (if not most) economists expect subjects to obtain a Pareto optimal equilibrium in these models. However, results of recent laboratory experiments suggest that subjects regularly fail to coordinate their actions to achieve a Pareto optimal equilibrium. These results are puzzling and motivate the study of the following questions. Why would a rational player select an action that supports an equilibrium that is Pareto dominated? More fundamentally, how do rational players form expectations of the play of other rational players in noncooperative games with multiple Nash equilibria?
In single-period, simultaneous-move, noncooperative games with no communication, players have no explicit mechanism to select an equilibrium as a group. The players must individually form their expectations of the play of other players and select their actions as individuals rather than a group. For this reason, we argue that rational play in these games must be defined in terms of individual rationality rather than collective rationality.
In chapter 1, individual rationality and collective rationality are defined in terms of Harsanyi and Selten's risk dominance and Pareto dominance. This chapter shows that a model which modifies Harsanyi and Selten's (1988) A General Theory of Equilibrium Selection in Games by applying individual rationality as a constraint on players' choices can answer the puzzling questions motivated by the results of these recent laboratory experiments. Furthermore, the robustness of this model is documented by a series of laboratory experiments involving coordination games and battle of the sexes games. The conclusion of this chapter is that coordination failures in single-period games are due to the conflict between individual rationality and collective rationality. Subjects act as if they abandon individual rationality in favor of group rationality only in special circumstances. The finitely repeated versions of these games are analyzed in chapter 2.
Issue Date:1991
Rights Information:Copyright 1991 Straub, Paul Gregory
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9211001
OCLC Identifier:(UMI)AAI9211001

This item appears in the following Collection(s)

Item Statistics