Behavioral Experiments in Networked Trade

We report on an extensive series of highly controlled human subject experiments in networked trade. Our point of departure is a simple and well-studied bipartite network exchange model, for which previous work has established a detailed equilibrium theory relating wealth to network topology. A notable feature of this theory is its prediction that there may be significant local variation in equilibrium wealths and prices purely as a result of structural asymmetries in the network. Our experiments mix recent lines of thought from algorithmic game theory, behavioral economics and social network theory, and are among the first; and largest behavioral economics experiments on network effects conducted to date. They continue a line of human subject experiments on networked games and optimization allowing only local interactions.