Analysis of Wage Discrimination in Major League Baseball

Analysis of Wage Discrimination in Major League Baseball

Introduction

Baseball has become one of the most popular sports in the United States. It became a professional sport since 1960s when the increasing victory led to recruit better players despite their ethnicities. However, since then the issue of unequal salaries among players arose. Many people believe that racial inequality is present in the baseball league. Even in this highly integrated labor market, the minority (non-white) might receive lower salaries than white players. In response to the argument of discrimination in payroll, many researchers have tried to examine the hypothesis of unequal treatment toward black baseball players. However, the result of those studies did not come to the same conclusions. In the early 1970's Anthony H. Pascal and Leonard Rapping and Gerald W. Scully, using the 1968-69 major league baseball players, arrived at the conclusion that salaries are independent of races. In contrast, Scully found racial salary discrimination existed in baseball. The difference arose because of the various methods used to estimate the effect of races on salaries. One model uses the wage differential from a single linear regression. The other model is based on the average salary of two separate salary model of white and nonwhite players.

In this paper, we will present a single linear regression to estimate whether or not there is discrimination among baseball players. The estimates that will be reported from this paper will provide crucial information for the continuing debate on the effect of race on the inequality of labor services in professional sports markets. The result of this research is important because the sample data refer to players in a more recent year, 1997, almost thirty years after Scully and Rapping estimated their models. In addition to races, the model used in this study includes other variables that may affect salaries, such as city-size, performances, national league to see what other factors might significantly influence salary.

Data-Model-variables

The study sample consists of 60 players (non pitchers) in 1997 Major League Baseball season. The reason of excluding pitchers is the uniqueness of pitchers' role, and the differences in the measures used to assess their performances and those used for other players. The players' salaries are collected through the website http://www.baseball-almanac.com/players/player-a.shtml and The Lahman Baseball Database. First of all, 30 white players and 30 non-white players are randomly chosen from The Lahman Baseball Database. Then, players' information about slugging, fielding percentage and races are collected through their profiles. Variables that are too highly related with each other are not included because this will increase the estimated errors on the individual parameter estimators. Hence, slugging percentage and average batting which are linearly correlated are not included in the model.

Findings

Below is the summary of data. Sixty players are chosen. Of which, 30 are white players and 30 are nonwhite.

Table 1

Summary of information about slugging, years, fielding, white, nl and marketsize.

Variables

Mean

Std. Dev.

Total

Salary

Nl

Years

Slugging

Fielding

White

Marketsize

Lnsalary

3229311

.5333333

7.216667

397.9322

97.75733

.5

.4833333

14.21035

3022061

.5030977

4.566242

129.6057

1.987797

.5042195

.5039393

1.472707

White Players

_cons

nl

marketsize

fielding

years

slugging

241306.2

1002170

564616.1

-94591.31

8346.857

6759379

1.88e+07

866638.7

986284

199563.1

3418.068

88598.37

Non-White Players

_cons

nl

marketsize

fielding

years

slugging

164740.2

-505254.4

229935.8

-44570.87

13005.91

324884.1