Abstract

Many people contribute to public goods but stop doing so once they experience free riding. We test the hypothesis that groups whose members know that they are composed only of “like-minded” cooperators are able to maintain a higher cooperation level than the most cooperative, randomly composed groups. Our experiments confirm this hypothesis. We also predict that groups of “like-minded” free riders do not cooperate. Yet, we find a high level of strategic cooperation that eventually collapses. Our results underscore the importance of group composition and social learning by heterogeneously motivated agents to understand the dynamics of cooperation and free riding.

This content is only available as a PDF.
You do not currently have access to this article.