A NEW METHOD FOR DETECTING INDIVIDUAL AND GROUP LABOR-MARKET DISCRIMINATION

In this paper we examine the pay and performance of several hundred major league baseball players and adjust for productivity differences to see if there are any unexplained wage differentials that are correlated with race. The technique used is new to this literature. Frontier estimation, most often used in connection with firm level efficiency measurement, can be adapted to measure wage differentials traced to nonproductivity characteristics, such as race. The distinct advantage to this technique is that each observation (in this case, players) receives a measure of wage discrimination. Thus, the information content of our analysis is greater than previous discrimination analyses in that we are able to say by how much each player's salary deviates from this productivity. Our study provides disaggregate information and allows us to compare the wage deviations for specific players as well as a more aggregate comparison, for example, race.