| Summary: | The equal split is a widely observed outcome in experimental studies of two-person bargaining. We report on an experiment that controls for the preferences of subjects and therefore sheds light on the social orientation of those proposers who offer the equal split. The data suggest that when the bargaining environment is personal and the cost of conflict is high, both egoists and fairmen (i.e. bargainers that demonstrate other-regarding preferences) make 50–50 proposals. However, as the interaction becomes more anonymous and as the cost of conflict falls, egoists experiment with more demanding proposals while fairmen stick to the equal split. When responding to offers, the data suggest that egoistic and competitive individuals are responsible for most rejections. Chapter three develops a model of the egalitarian conventions seen regularly in field studies of sharecropping contracts, in more general ethnographic reports on small-scale political structures, and in the experimental lab. We develop an evolutionary bargaining model incorporating systemic perturbations to derive stable equilibria in an environment where agents are allowed to experiment with different demands. The first major result is that using a minimal set of assumptions, the only systemically stable bargaining convention is the equal split. The second result demonstrates that other unequal conventions are also robust to systemic perturbations if the asymmetry of equilibrium demands is great enough. Monitoring by peers in work teams, local commons situations, and residential neighborhoods is often an effective means of attenuating incentive problems. Most explanations of the incentives to engage in mutual monitoring rely either on small group size or on a version of the Folk theorem. In chapter four we provide an explanation of mutual monitoring in single shot interactions among members of large teams. A key element of our approach is that when team members are residual claimants, some members are motivated by reciprocity to punish fellow members that shirk. To test the model's predictions about the role of reciprocity and the informational effects of team size we conduct an experiment that varies the residual claim of teams and information about other members of the team.
|
|---|
