Labor Market Discrimination

1
Labor Market Discrimination
2
Outline

• 1. Basic Facts Race and Gender Inequality
• 2. Models of Labor Market Discrimination
• Personal Prejudice Models (Taste Discrimination,
  Preference-Based Discrimination)
• Statistical Discrimination
• 3. Evidence of Discrimination
• Oaxaca Decompositions
• Audit Studies
• Other Studies
• 4. A Brief History of Federal Anti-Discrimination
  Policy
• 5. Trend in Black-White Wage Inequality Over Time

3
Basic Facts About Race and Gender Inequality
4
Time Trends in Race and Gender Inequality
5
Theories of Labor Market Discrimination

• Differences in labor market outcomes do not
  necessarily imply discrimination.
• There are two main theories of labor market
  discrimination
• Preference-based discrimination
• Statistical discrimination

6
Preference-Based DiscriminationEmployer
Discrimination

• Start from the supposition that men and women are
  equally productive. Labor is the only factor of
  production.
• However, employers do not like hiring women and
  act as if there is an extra cost to hiring women.
  
• VMPEWF(1d)
• Where d is the
• d may vary across firms

7
A Firm that Does Not Discriminate(d0)

• Suppose that WMgtWF.
• As always, they will hire women up to the point
  where the value of the marginal product of labor
  equals WF.

Dollars
WF
VMPE
Employment
EF
8
Firms that Discriminate(dgt0)

• Hire only men if WF (1d)gtWM
• Hire only women if WF (1d)ltWM

Dollars
WM
VMPE
E
EM
Male-only firm WF (1d)gtWM
9
Labor Market Equilibrium
Distribution of d in the Economy
10
Labor Market Equilibrium
WF / WM
S
1.00
0.75
D
L
Nd0
11
Predictions
WF / WM
Decrease in number of women in the labor force
S2
S1
1.00
0.75
D
E
N1
N2
Nd0
12
Predictions
WF / WM
Increase in number of non-discriminating firms
S
1.00
0.80
0.65
D2
D1
E
N2
N1
13
Discrimination and Profits

Profits in the absence of discrimination
WF
VMPE
N
L
14
Preference-Based DiscriminationCustomer
Discrimination

• Imagine that consumers dont like buying goods
  made or sold by blacks.
• Purchasing decisions are based on the utility
  adjusted price, pd.
• Consumer discrimination reduces the demand for
  goods and services made and sold by minorities.
• Segregated workforce. Firms have an incentive to
  place minority workers in jobs where they have as
  little customer contact as possible.
• No wage differential. No wage differential
  between blacks and whites as long as firms can
  hide black workers.
• If firms cannot hide black workers, wages of
  blacks will fall.

15
Preference-Based DiscriminationEmployee
Discrimination

• Suppose that whites do not like working with
  blacks and that blacks are indifferent between
  working with whites and members of other racial
  groups.
• This implies that employers would have to give
  white workers a compensating wage differential.
• Segregated Workforce. Firms will want to avoid
  these higher labor costs wherever possible and
  the result will be that employers will hire
  either only blacks or only whites.
• No Wage Differential. Employers will simply
  hire whichever group is the cheapest. Thus,
  competition will eliminate any racial wage
  differential.
• Firms may not be able to avoid putting groups
  together in the workplace.

16
Statistical Discrimination

• Basic Logic
• Two people, identical resumes.
• One of the applicants is a man, and the other is
  a woman.
• Statistical evidence indicates that women are
  more likely to quit their jobs than men.
• Quits hurt employer because they disrupt
  teamwork, etc..
• Thus, the (profit maximizing) employer will
  logically hire the man.
• Same logic as racial profiling.

17
The Impact of Statistical Discrimination on Wages

• Combine all of the information that an employer
  has into T.
• If T is not a perfect measure of productivity,
  then firms have an incentive to use average
  productivity.

• a measures the correlation between the test score
  and the workers productivity.
• T is the average test score for the group.

18
Two Ways in Which Statistical Discrimination
Leads to Wage Inequality
CASE 1 This corresponds to the case in
which blacks are less productive than whites on
average.
Even if blacks and whites have the same test
score, blacks will be offered lower wages.
19
Two Ways in Which Statistical Discrimination
Leads to Wage Inequality
CASE 2 This corresponds to the case in which
firms have less accurate information about
blacks than about whites.
High scoring blacks earn less than high score
whites. Low scoring blacks earn more than low
scoring whites.
20
Should Employers Use Group Averages?

• From the standpoint of profit maximization yes.
• From the standpoint of efficiency yes.
• From a moral/legal standpoint no.
• The very process of using the group average to
  infer productivity can lead to self-fulfilling
  prophecies.
• Example women and the decision to leave the
  labor force to have children.

21
Measuring Discrimination

• How do you measure or quantify discrimination?
• One possible definition of discrimination is to
  look at the difference in the mean wage between
  two groups, say men and women.

• A more appropriate measure would compare the
  wages of equally skilled workers.

22
Oaxaca Decomposition

• To simplify the exposition, lets suppose that
  only one variable, schooling, affects earnings.

Male earnings function
Female earnings function

• Can estimate these parameters by running a
  regression.
• ß tells how much an individuals wage increases
  as the result of an additional year of schooling.
• a tells the level of the earnings profile for
  each group.
• So that the difference in the average wage of men
  and women becomes

23
Oaxaca Decomposition, Continued
Wage differences can arise because
And because
or
24
Diagram of the Oaxaca Decomposition
SlopeßM
Wage
Men
wM
Women
wFM
aM
wF
SlopeßF
aF
sM
sF
Skill
Average skill of men
Average skill of women
25
Critique of the Oaxaca Decomposition

• The Oaxaca decomposition may overstate the extent
  of discrimination.
• If we cannot control for all of the relevant
  observable differences between men and women,
  then we will erroneously interpret differences in
  the wages of men and women as the result of
  discrimination.
• The Oaxaca decomposition may understate the
  extent of discrimination.
• Differences in observable characteristics may
  themselves be the result of discrimination.
• Many discrimination law suits in the U.S. court
  system use the Oaxaca decomposition.

26
Evidence of Discrimination Oaxaca Decomposition
27
Audit Studies

• Train auditors who are selected to be similar
  on as many dimensions as possible, but who have
  different race or gender.
• Send them out to interview for jobs.

28
Kahn and Sherer, 1988

• Look for evidence of racial discrimination in the
  NBA.
• Conceptually, study has the advantage of having
  very good measures of productivity.

29
Measured at the start of the 1985-1986 season.
30
(No Transcript)
31
Means, 2001-2002 Season
Here I have repeated the analysis for the
2001-2002 season. Note that blacks earn
significantly more than whites.
32
Results Suggests that whites earn 5.5 more
than blacks, BUT this difference is not
statistically significant. Little evidence of
salary discrimination in the NBA in the 2001-2002
Season. One problem A large proportion of
compensation comes in the form of signing
bonuses, etc. We have no measure of this.
33
Bertrand and Mullainathan, 2003

• Respond to job ads listed in the newspaper in
  Chicago and Boston.
• Send out resumes that are identical apart from
  the names at the top of the page.
• Tries to avoid the problem with audit studies
  that auditors may not be identical.

34
One potential problem is that these names may
connote more than race. They also may connote
socioeconomic status.
35
A Brief History of Federal Anti-Discrimination
Policy

• Brown vs. the Board of Education, 1954
• Invalidated separate but equal schooling
• The Equal Pay Act of 1963
• Prior to 1960 sex discrimination was officially
  sanctioned by protective labor laws that
  limited womens total hours of work and
  prohibited them from working at night, lifting
  heavy objects and working while pregnant.
• These laws were overturned by the Equal Pay Act
  of 1963, which also outlawed separate pay scales
  for men and women with the same skills and
  performing work under the same conditions.
• The act said nothing about equal opportunity in
  hiring and promotions. Thus, an unintended
  consequence may have been that women were less
  likely to be hired.

36
History, Continued

• Title VII of the Civil Rights Act of 1964
• Title VII of the Civil Rights Act of 1964 made it
  unlawful for any employer to refuse to hire or
  to discharge any individual, other otherwise to
  discriminate against any individual with respect
  to his compensation, terms, condition, or
  privileges of employment, because of such
  individuals race, color, religion, sex or
  national origin.
• The law was not retroactive.
• In order to enforce the provisions of the law,
  Title VII also created the Equal Employment
  Opportunity Commission (EEOC).
• The EEOC has the authority to mediate complaints,
  encourage lawsuits, and (since 1972) bring suits
  itself against employers.

37
History, Continued

• The Federal Contract Compliance Program (OFCC)
• Established in 1965 by Executive Order 11246 to
  monitor the hiring and promotion practices of
  federal contractors (and firm supplying goods and
  services to the federal government that are above
  a certain size (above 50,000 or 50 employees).
• Requires contractors above a certain size to
  analyze the extent of their underutilization of
  women and minorities and develop plans to remedy
  that underutilization (starts in 1968).
• Defining the percentage of women and minorities
  the firm should hire is not straightforward.

38
What might explain this picture?