| Abstract: | In this paper we examine the allocation of labor under free agency by developing and testing a model to predict the migration of free agents. Data for individual players are used in a logit model that estimates the probability that a free agent will change teams as a function of several independent variables. Along with other authors, we find that free agents tend to migrate to big cities. In addition, we find evidence that the allocation of labor may be different under free agency than under the reserve clause and suggest a reason why Coase's theorem may not be applicable to this labor market. |
