Cooperative Networks: Theory and Experimental Evidence

We consider a modified pure public good game characterized by a pre-play negotiation stage, on which pairs of players can form binding cooperation commitments. As the introduced mechanism only supports pairwise rather than more inclusive commitments, it does not implement the efficient outcome. We theoretically derive the incentive compatible and efficient cooperative networks and evaluate the behavioral efficacy of the suggested mechanism to promote and stabilize cooperation. We present the results of two separate experiments. The first experiment serves to provide necessary methodological prerequisites and establishes that neither repetition with an unknown end nor voluntary costly monitoring are behaviorally sufficient to induce cooperative outcomes. In the second experiment we introduce the pairwise commitment mechanism. We show that the mechanism induces aggregate cooperation rates not only beyond the rates observed under the voluntary contribution mechanism operationalized in the first experiment, but also beyond the rate which is supported by the formation of incentive compatible networks. We observe a large heterogeneity between groups: while some groups converge to full cooperation by managing to coordinate on the formation of efficient networks over time, both networks and cooperation rates unravel in other groups. An extended version of our theoretical setting with inequity averse players in the form suggested by Fehr and Schmidt (1999) captures the stylized facts of both experiments