Estimating Customer Discrimination in Baseball Using Panel Data: 1972-1991
In contrast to owner or co-worker discrimination, customer discrimination can exist indefinitely in a competitive environment such as baseball. As a result, persistent wage differentials across racial groups could arise because team owners will pay the discriminated players less to compensate themselves for employing these players. Therefore, an empirical study of customer discrimination would provide some insight into the persistence of wage discrimination in baseball. Furthermore, does the presence of customer discrimination in baseball force owners to choose between playing black players or alienating fans? In this paper, I investigate whether this trade-off exists for team owners and whether customer discrimination has affected the salaries of black baseball players.
The majority of work on this topic examines customer discrimination in basketball, but there have been several empirical investigations using baseball data. For example, using 1968-1969 baseball data, Scully ( 1974b) finds that a one percent increase in the fraction of a team's players who are black reduces a team's revenue by $59,000. Sommers and Quinton ( 1982), however, find no evidence of customer discrimination using 1976-1977 baseball data. Gwartney and Haworth ( 1974) find that in the initial period of baseball integration, increasing the percentage of black players on a team boosted attendance. Andersen and La Croix ( 1991) and Nardinelli and Simon ( 1990) find evidence of customer discrimination in baseball using baseball cards. Finally, Andersen ( 1992) finds that black baseball players faced customer discrimination in All-Star voting.
Overall, the evidence of customer discrimination in baseball is not conclusive. One reason for the ambiguity is data limitations. Most of the prior work relies