September 1, 2014
Although discrimination can be costly for agents who practice it, it was not until 72 years of being founded Major League Baseball (MLB) that a team had a black player put his feet on the ground. (For an interesting analysis of the costs of discrimination, see “Is it profitable to discriminate? A baseball lesson”). In the case of the National Basketball Association (NBA), the waiting time was much smaller since by 1950, only four years after being founded, Nat ”Sweetwater” Clifton debuted as the first black player in the professional basketball.
Sports provide an excellent scenario to test the existence of racial discrimination because of the abundance of statistics on the productivity of the athletes. That is, when you take control of a wide range of statistics measuring both performance and quality of athletes, it is possible to determine whether differences in the treatment of athletes are due to their racial characteristics or their performance.
Early research on the issue of discrimination in sports attempt to prove the existence of wage gaps in athletes on the ground of race origin. In the case of the NBA, some authors have shown that black NBA players faced wage discrimination until late 1980s. However, these differences seem to have diminished in recent decades. In the case of the MLB, evidence also points in the same direction (see Kahn, 1991, 2000; Bellmore, 2001; Jiobu 1988). Moreover, in professional baseball is possible to observe how athletes from ethnic minority groups have received some of the best wage contracts in recent years. For example, in a sample of 511 players who were free agents and signed contracts in the period 1998-2006, Holmes (2011) shows that, on average, white players signed contracts valued at US$2 million annually, while Hispanics did in the amount of US$2.4 million annually.
The most recent studies on discrimination in professional sports try to prove the existence of different treatments of athletes by their evaluators. Specifically, based on the idea that individuals tend to favor those with whom they feel they have some affinity, these studies attempt to demonstrate whether the umpires or referees tend to favor in the plays those athletes belonging to their ethnic group.
The results are simply astonishing. In the case of the NBA, Price and Wolfers (2007) find that black players are signaled from 2.5% to 4.0% more fouls than white players for every 48 minutes per game, when the number of white umpires arbitrating the game increases from zero to three.
In the case of the MLB, Parson et al. (2011) used an amazing database of over 3.5 million records of pitches to show whether the umpires favors pitchers belonging to the same ethnic group. The results show that when the pitcher and home plate umpire belong to the same ethnic group, the pitcher is favored on average with one strike more per game. Despite its small size, this effect is amplified (or disappears) when the plate umpires see they are poorly (or highly) monitored. For example, when the ballparks are little frequented by fans and have no monitoring camera systems for pitches, about 30% fewer pitches falls into the strike zone when the pitcher and the umpire are of different ethnicities. That is, if in a game, where usually 150 pitches are thrown, the pitcher and the umpire are both Latinos, then about 49 pitches are considered in the strike zone. However, when the ethnicity of these individuals differs, everything else been equal, then only about 34 pitches are considered strike pitches.
In general, wage discrimination is a phenomenon that has been vanishing in sports. This is due to the level of competitiveness leading to team owners to hire and pay well the best athletes no matter their ethnicity. But there is another kind of discrimination, far more subtle and difficult to detect, which is given by the affinity that individuals may have with people they feel they share something in common (like race). Fortunately, putting the right incentives (e.g., the use of cameras for pitching monitoring) can significantly mitigate this phenomenon.