Published online by Cambridge University Press: 02 September 2015
We develop a model of adaptive learning with social comparisons. Actors are more likely to choose actions that recently yielded satisfactory payoffs; satisfaction is evaluated relative to an aspiration level that reflects previous payoffs and possibly other players’ payoffs. This captures the phenomenon of social comparison via reference groups. We show that if agents compare themselves to those who are receiving higher payoffs then in stable outcomes all payoffs must be equal. If, however, agents’ aspirations are driven by less ambitious social comparisons then very unequal distributions can be stable. We apply our general results to collective action problems in socio-political hierarchies and derive conditions for stable exploitation. Finally, we develop a computational model, which shows that increases in payoff inequality make outcomes less stable.
Jonathan Bendor, Walter and Elise Haas Professor of Political Economics and Organizations, Graduate School of Business, Stanford University Stanford, CA 94305 ([email protected]). Daniel Diermeier, Dean, Harris School of Public Policy and Emmett, Dedmon Professor of Public Administration, 1155 E. 60th Street, Chicago, IL 60637 ([email protected]). Michael M. Ting, Department of Political Science, Columbia University, New York, NY 10027 ([email protected]). An earlier version of this paper was presented at the 2007 Annual Meeting of the American Political Science Association. The authors thank Michael Harrison and David Kreps for helpful discussions about non-stationary Markov chains, and Greg Martin and Holke Brammer for their research assistance. To view supplementary material for this article, please visit http://dx.doi.org/10.1017/psrm.2015.47