1 of 56

Chapter 10

Discrimination: Theory, Measurement, and Consequences

FIFTH EDITION

The Economics of Sports

MICHAEL A. LEEDS | PETER VON ALLMEN

2 of 56

Introduction

Jackie Robinson became the first black player in MLB’s modern era (NL; hired by manager Branch Rickie)

1946—Montreal Royals
1947—Brooklyn Dodgers

Larry Doby became the first black in the AL (hired by Bill Veeck)

1947—Cleveland Indians

It is remarkable that Robinson and Doby flourished despite the tense atmosphere in which they played

10-2

3 of 56

Learning Objectives

Understand the Becker model of discrimination
Describe how various forms of discrimination can occur in professional sports
How economist measure discrimination
Statistical discrimination

10-3

4 of 56

Empirical Studies and Definitions

A major study finds that all players face the same opportunities in MLB, NFL and NBA
Some economic studies find that there are differences in pay in the NBA as well as European soccer
They also find evidence of discrimination in the NFL draft
We need to distinguish between

Prejudice -- a feeling or emotion
Discrimination -- an action

Economists separate two issues

Equal access to work (draft order)
Equal pay for equal work (equal salaries to equally talented players in same positions)

10-4

5 of 56

Becker’s Theory of Discrimination

Gary Becker (1957) uses neoclassical theory to analyze and evaluate discrimination like any other “good”

Neoclassical theory assumes that people maximize utility subject to constraint
Economists focus much more on the outcome of discrimination than on its origin

Becker considers prejudice a taste

Discrimination is indulging that taste
This indulgence comes at a cost

We can reduce discrimination by raising its cost

10-5

6 of 56

Different Forms of Discrimination

This section introduces

Employer prejudice discrimination
Employee prejudice discrimination
Consumer prejudice discrimination

Although each model uses money to measure the intensity of the discrimination, they are all based on utility maximization
As people with a taste for discrimination maximize their utility, their willingness to pay to indulge their tastes has a variety of effects

10-6

7 of 56

Employer Discrimination

In Becker ’s model, employers have preferences regarding employees with whom they do and do not want to associate
There have been many studies of discrimination against French-speaking hockey players (Francophones)

Their findings depend in many cases on the players’ positions

To simplify matters, assume that there are only two groups of players

English-speaking (E) and French-speaking (F)

We assume for now that, although the players’ styles may differ, they are equally productive

They have the same MP_L

10-7

8 of 56

Employer Discrimination

If a hockey team owner dislikes Francophones, employing them brings a psychic cost

He feels that he pays them more than others
The perceived wage is

w_F= (1+d_F)w_E

d_F = discrimination coefficient (%)

If the employer pays both F and E the same, he feels as if he is paying F more
As a result, the demand for F falls

10-8

9 of 56

Employer Discrimination

Assume for simplicity that the supply of both players is the same
Panel b in Figure 10.1 indicates that the wages of both kinds of players will be the same
Panel a in Figure 10.1 indicates that with discrimination, the demand for Francophones falls

At any given wage, the employers’ distaste for them reduces their willingness to hire them

As a result, F’s wage falls relative to E’s

W_E > W_F

10-9

10 of 56

Figure 10.1

10-10

11 of 56

Who Wins and Who Loses: Summary

Francophone players lose

Pay is lower and fewer are employed (Figure10.1)

Anglophone players win

Pay is higher and more are employed
Less qualified workers are hired

Unequal pay for equal work results
Prejudiced employers lose

Francophones’ lower pay offsets psychic cost
Employers pay more for worse players

10-11

12 of 56

Does Anyone Win with Employer Discrimination?

Consider the case of racial discrimination in MLB
Blacks were effectively barred from organized baseball from 1888 to 1947 by a “gentlemen’s agreement”
One group that benefited from discrimination was white players of that era
More white players played in the major leagues than would have been possible otherwise
Figure 10.2 reflects the result of assuming that there is a large supply of white players (deep pool of talent) willing to play at market wages—no blacks are hired

10-12

13 of 56

Statistical Discrimination

What if productivity differs across the two groups?
Group statistics may indicate nothing about individual performance
The use of group averages to judge individual productivity is called statistical discrimination

Employers may use it if obtaining individual information is costly

Small differences in group statistics can have a strong impact on a team’s hiring practices and may result in decisions that look like taste discrimination

Statistical discrimination may generate a self-fulfilling prophesy

10-13

14 of 56

Statistical Discrimination

Before Jackie Robinson proved his talent in MLB most players and owners felt that Negro League players did not really have what it took to play in the white major leagues.
This may also apply to foreign players in the NBA before U.S. pro players could compete in the Olympics.
What about Japanese baseball players?

10-14

15 of 56

Statistical Discrimination Problem

Draw a normal distribution of productivity for black and white baseball players such that the mean for white players is greater than that for the black players but such that the upper tail of the black distribution overlaps into the upper tail of the white distribution.
Demonstrate that a star black player can possess productivity greater than that of an average or even above average white player.

10-15

16 of 56

Hypothetical Distribution of Worker Productivity

among Blacks and Whites

10-16

17 of 56

Segregation Results

The Negro League owed its existence in part to such discrimination

The league effectively folded once the major league integrated

Figures 10.3 shows the market for Negro League

Demand (D_S) was much higher before integration (D_I)

Figure 10.4 shows the MLB market

Demand (D_I) rose after integration (D_S)
Tastes of consumers changed, both blacks and whites

10-17

18 of 56

Figures 10.3 shows the market for Negro League�Demand (D_S) was much higher before integration (D_I) �

10-18

19 of 56

Figure 10.4 shows the MLB market�Demand (D_I) rose after integration (D_S) �Tastes of consumers changed, both blacks and whites

10-19

20 of 56

Competition Can Eliminate Discrimination

Unprejudiced employers have an advantage over prejudiced employers

They hire more productive players (and get higher revenue if consumers and employees are not prejudiced)
They do not have to pay as much as teams with Anglophones
They have higher profits

As more unprejudiced employers enter, or as more people worry about profit than taste

Pay of Francophones is driven up
Pay eventually equals those of Anglophones

Perfect competition eliminates discrimination

10-20

21 of 56

Competition and Discrimination

In the 1950s, integrated teams were better

Dodgers, Giants, Indians, Braves, and White Sox were dominant teams

Dodgers and Giants won pennants

Only the Yankees were good and remained white

Owners had recognized black talent but could not act on it
MLB kept Bill Veeck from buying the Phillies in 1943

Veeck had wanted to hire Negro League players
He was the first to integrate AL with 1947 Cleveland Indians

10-21

22 of 56

When Markets Are Not Competitive

In competitive markets, discriminated players work at a discount
We will now assume that the market for labor is not competitive—it is a monopsony
The standard model is presented in Figure 10.5
Without discrimination, the employer pays the same low wage (found on the supply curve) to all players

In this figure, the wage is $800 while the value of the player is $1000

10-22

23 of 56

Figure 10.5

10-23

24 of 56

Figure 10.6

10-24

25 of 56

Monopsony Model with Discrimination

Figure 10.6 shows the wages paid to F players
It assumes that

F players are more productive than E players, so F players are worth $1500 rather than $1000 that E players receive (for example)
Employers discriminate against F players, which shifts the demand curve from D to D’

F players are paid more than E players ($960>$800)
F players are paid less than they are worth because of monopsony power and because of discrimination

10-25

26 of 56

Employee Discrimination

Employees with a taste for discrimination regard the market wage, w, as w(1 - dj)

d_j represents the coefficient of discrimination for employees
They feel like they are being paid less

In 1880s, a few blacks played in the American Association

The first was Moses Fleetwood Walker for Toledo in 1884
Cap Anson, a dominant player for Chicago White Stockings, played in an exhibition game against Toledo team with Moses Fleetwood Walker
Next year he had a contract not to include black players
White players drove blacks out of MLB by 1888

10-26

27 of 56

Employee Discrimination History

Attempts to reintegrate MLB started after WWII even as employee discrimination was rampant
The Dodgers traded 5 players for Al Gionfriddo & $100,000, a marginal player, who was willing to take the locker next to Robinson (& “carry the money”)
In 1947, Dodger players circulated a petition refusing to play with Robinson
Dodgers manager, Leo Durocher, called a late-night meeting and (in an exquisite speech of which there are different versions) said

See the next slide

10-27

28 of 56

Quote

…I hear that some of you don’t want to play with Robinson. Some of you have drawn a petition. Well, you know what you can do with that petition
I hear Dixie Walker is going to write Mr. Rickey a letter asking to be traded. Just hand him the letter, Dixie, and you’re gone! I don’t care if a fellow is yellow or black or if he has stripes like a fuckin’ zebra
I am the manager, and I say he plays. …I’ll play an elephant if he can do the job, and to make room for him I’ll send my own brother home
He’s going to win pennants for us. He’s going to put money in your pockets and money in mine

10-28

29 of 56

Discrimination in Perfect Competition

Assume the labor market is perfectly competitive

The supply curve is horizontal

Prejudiced players require a compensation above the market wage: w/(1-d_j)

Such a wage does not exist

Prejudiced players are replaced by those willing to take the market wage
Alternatively, the employer could ban black players

Players no longer require a premium wage

Segregation thus results in this case from employee discrimination

10-29

30 of 56

Discrimination v. Segregation

The two are not the same – though they usually are related
There could be separate markets for prejudiced and unprejudiced players

Dixie Walker refused to play alongside Robinson
Walker was not refusing to play
He just did not want to play with Robinson

Dodgers traded Walker to the Pirates in 1948

Teams separate into integrated and non-integrated teams
Pay and treatment on those teams is not affected by race

10-30

31 of 56

Consumer (Fan) Discrimination

Consumers have a taste for discrimination if they prefer not to purchase goods or services from members of a specific group

Customers feel a psychic cost

Some fans feel they pay a higher price p(1+d_k) when watching black players

p is the price of watching white players

Consumer discrimination can be difficult to isolate
A team feels pressure to discriminate

It fears that fans will go elsewhere
Unprejudiced employers are led to discriminate

10-31

32 of 56

Consumer (Fan) Discrimination

Consumer discrimination can affect attendance
Consumers could show their taste for discrimination by supporting teams that have fewer players from the group that they dislike
Consumers could follow teams that are integrated but do so with less intensity—going to fewer games, buying fewer jerseys, watching the team less on television

10-32

33 of 56

Fan Discrimination

For the preferred group, there is a race premium. More of their services are hired at a higher wage than they really are worth relative to their actual performance.
Bigoted fans pay the cost of their discriminatory preferences through less winning for the amount of money paid to preferred players.

10-33

34 of 56

Detecting Consumer Discrimination

Studying consumer behavior can be difficult

It is not obvious how to separate taste for the team from the taste for individual players

Several studies focus on memorabilia rather than game attendance

Do fans prefer the cards of white players, all else equal?
Evidence suggests they do

10-34

35 of 56

A Digression into Boxing

Jack Johnson and “unforgivable blackness”

In some ways, he resembled Muhammad Ali
Times did not tolerate a brash black man who taunted his opponents
Race riots accompanied his title win and defenses (1915)
He spurred the search for a great white hope

No black man fought for the title between 1915 and 1936
Joe Louis was a supremely talented fighter

He was groomed with Jack Johnson in mind
His achievement went beyond boxing 2 bouts v. Max Schmeling
He became symbol of American ideals v. Nazism

Schmeling beat Louis in 193
The rematch in 1938 brought interracial celebration rather than riots

10-35

36 of 56

Measuring Discrimination

In 1988 Kahn & Scherer asked:

Is there racial discrimination in the NBA?
How can there be when the NBA is 80% Black?

Black players earn more than white ones on average

The theory of discrimination says that

Average pay is suggestive but not conclusive
Discrimination means paying equals unequally
Compare black players with otherwise identical whites

Kahn and Scherer found that black players received about 20% less than white players

10-36

37 of 56

The Residual Method

A statistical technique that separates the wage gap between persons of two groups into the parts due to productivity differences and discrimination, respectively.

10-37

38 of 56

Residual Method (cont.)

Assume that only efficiency determines productivity.
Consider two lines that estimate the relationship between wages and efficiency for whites and blacks. Using the residual approach we can determine if there is still pay discrimination in the NBA despite equal average pay between the races.

10-38

39 of 56

The Residual Approach to Measuring �Wage Discrimination

Efficiency

Wage in Dollars

White Players

Black Players

W_W=W_B

W_B*

E_W

E_B

10-39

40 of 56

Residual Method (cont.)

Total Wage Gap: W_W - W_B
Explained: W_W - W_B*
Unexplained: W_B* - W_B
Where W_W is the average wage of White workers, W_B is the average wage of Black workers, W_B* is the wage Black workers would be paid (given their level of efficiency) if they were White.

10-40

41 of 56

Discrimination Problem

The average pay of black and white players was equal at $400,000 in the 1985-86 NBA season. Does this mean there was no pay discrimination?
Economists discover that black players earn, on average, $28,500 for each extra efficiency point of their performance.
White players earn $30,700 for each extra efficiency point.
Black players earned base earnings of $57,000 if they had no efficiency points; white players had base earnings of $61,000.
Reconcile the apparent contradictions in this data using the residual method.

10-41

42 of 56

Solution

Solve: 400,000 = 61,000 + (30,700xE_W)
Solve: 400,000 = 57,000 + (28,500xE_B)
E_W= 11.042
E_B= 12.035
If black players were paid along the white pay line the value of their performance would be approximately $430,475!
So we would conclude that there is still pay discrimination despite equal average pay.
The amount of the pay discrimination would be approximately $430,475-$400,000 = $30,475

10-42

43 of 56

Economic Findings on Pay Discrimination

There is evidence that pay discrimination existed in pro team sports in the past.
But by the mid-1990s, pay discrimination is pretty much gone.
Research from the NBA using 1985-1986 data suggested there was a pay premium for white players and that the source of this premium was fan discrimination.
Some research on the NHL suggested there was pay discrimination against French-speaking players.

10-43

44 of 56

Economic Findings on Exit Discrimination

Exit discrimination occurs when white players’ careers last longer than those of equally productive nonwhite players, ceteris paribus.
Some research suggested that black players leave the NBA and MLB earlier than white players of the same ability.
However, my coauthor and I found no evidence of exit discrimination in the NBA in the late 1989-2001 time period.
We also fail to find exit discrimination in MLB for the 1990-2004 period..

10-44

45 of 56

Other Forms of Discrimination

Some evidence that black players are chosen lower in the NFL draft than white players of the same expected ability.
Stacking remains a prevalent phenomenon both on the field and in the front office.
Stacking occurs when nonwhite players or workers are put in less conspicuous positions.

10-45

46 of 56

Positional Discrimination

Positional (or role) discrimination occurs if players have unequal access to specific positions in team sports
Players are stacked in specific positions
Consider positions in the NFL in 2010

83 percent of quarterbacks are white
84 percent of wide receivers are black
In general, white players are more likely to be on offense than on defense

Table 10.1 reports similar inequalities in MLB

10-46

47 of 56

Table 10.1

10-47

48 of 56

Discrimination by National Origin; UEFA

Civil rights legislation outlaws discrimination on the basis of “race, color, religion, sex, or national origin”
Union of European Football Associations (UEFA) is the governing body for European Football/Soccer

Collectively, UEFA feared loss of national identity on club teams if clubs use too many foreigners
Individual teams tried to hire many foreign players to improve
Until 1991 UEFA allowed each team to have only 2 foreign players
This limitation was struck down by European Union

10-48

49 of 56

UFEA

Because the EU (Treaty of Rome…) allows free movement of labor, teams had to adjust
3 + 2 Rule limited each team to three starting players from other countries in 1991

It, too, was struck down—by the “Bosman Ruling”

Jean-Marc Bosman tried to move from Liege (a team in the Belgian League) to Dunkerque (France); he was blocked and appealed to the European Court of Justice
UEFA’s current regulation requires “eight players from every 25-man squad to have been developed in the fielding club’s national association”

10-49

50 of 56

Hockey and Football

Hockey has changed since the fall of the Berlin Wall
Role discrimination is a possible form of statistical discrimination

A member of a group is being judged based on the average performance of that group instead of individual talent
Troubling because if could be “justified” and self-fulfilling

This could be happening in the NFL

Because there are few black quarterback and their low incidence could be discouraging others from trying

10-50

51 of 56

Discrimination in Coaching

Minority coaches and managers have become increasingly common

The NFL lags behind MLB and the NBA
See Table 10.2

Women are represented in all major sports, though they generally have little say in personnel decisions
Division I college coaching resembles professional ranks

Minorities are underrepresented at Division II and III schools
Fewer women are coaching – even in women’s sports

Do minority coaches have to be better?

Manning finds that minority NFL coaches have better records
Kahn finds that minority NBA coaches are no more likely to be fired

10-51

52 of 56

Gender Equity—A Special Case?

Gender equity is harder to measure
Men & women are seldom in the same venue

Babe Didrikson Zaharias, Annika Sorenstam, and Michelle Wie entered the PGA—they are exceptions
Even the same sport may vary: tennis & figure skating

Mixed competition is rare

Mixed tennis doubles is a contrived event
Horse & auto racing are the only real examples
We do not know if there are few women because demand or supply is low

10-52

53 of 56

Fan Demand for Women’s Sports

Demand for women’s professional team sports, such as soccer or basketball, is much lower than for men’s sports
Women fare much better in individual sports

Tennis is a primary example

Wimbledon TV ratings of women’s tennis is high

Women are paid more in gymnastics and figure skating

Women have recently encountered problems in golf

Number of women’s tournaments has fallen

10-53

54 of 56

10.3 Title IX & Discrimination in College Sports

Title IX has been credited with the success of US women in soccer and the Olympics
Titles deals with more than sports

It is part of 1972 Education Amendments to the 1964 Civil Rights Act
It mandates equal access & opportunities for women in federally funded education programs

Secondary schools and college

10-54

55 of 56

Title IX Compliance

Compliance has a long history

It is overseen by the Office of Civil Rights in the Department of Education

There are three ways to comply

Funding and participation must be proportional to enrollment

5% gap is allowed

School must show history of expansion of programs
School must accommodate interests of students

Very few programs are truly in compliance
No school has been penalized/sanctioned

10-55

56 of 56

Title IX Impact

Some results are desirable. Title IX

Spurred rapid growth in women’s sports

Most of the growth was in the 1970s and 1980s

Gave women grounds to seek remediation

Some results are undesirable

Many colleges have cut men’s programs rather than expand women’s programs
As rewards to coaching have risen, women coaches have disappeared

Fell from ~80% of coaches in women’s sports to ~44%

10-56