BibTeX
@MISC{Birchby11communication,coordination,
author = {Jeff Birchby},
title = {Communication, Coordination and Fairness in a Median Game},
year = {2011}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
I explore within a laboratory setting the extent to which communication can increase efficiency within a game where the coordination task cannot be separated from the allocation. Subjects played a median game in which earnings are a multiple of the group’s median minus a cost associated with an individual’s effort. This structure retains the coordination problem of Pareto ranked equilibrium from the standard minimum effort game, but efficiency is now asymmetric and requires a minority to exert the minimum effort. Communication results in large gains to both effort levels and group medians with the gain relative to a no communication benchmark growing across periods. The effect is most dramatic directly following communication, and dissipates with each repetition. Subjects were found to be interested both in coordinating to the highest possible median as well as achieving an equitable distribution of gains. Within group communication they suggested unanimous high effort most commonly, but a robust minority advocated a series of rotating free riders. A small minority announced an intention to free ride all iterations, and depend upon the remainder of the group’s effort for a high median. Individuals were found to be most sensitive to receiving a free riding message, with a corresponding increase in one’s own effort. This effect was not substantial enough to overcompensate for the free riding, with medians negatively impacted by free riding strategies. Efforts were not particularly sensitive to the message type, with two notable exceptions: aggressive, threatening style messages were found to induce higher efforts and trust based messages corresponded with higher efforts across all iterations. 1 1
Keyphrases
median game minimum effort robust minority large gain communication result unanimous high effort high median threatening style message coordination task cannot coordination problem small minority group median free rider median minus free riding message notable exception individual effort within group communication message type effort level standard minimum effort game gain relative possible median corresponding increase equitable distribution free riding communication benchmark
