1 | P a g e
Oscar Uyamadu
Dr. Zafar Nazarov
Senior Seminar Introduction
21st February 2017
Do you earn what you deserve?
Motivation
In 1964 the civil rights act mad it illegal for an employer to discriminate against
individual based on their race, gender, color and national origin (Coleman, 2003). This was an
attempt to help bring equality to men and women no matter what race or gender they may be;
however, a gender and race wage disparity still exists. Up until the 1970s, it found that women
only made 60% compared to their male counterparts. Since then, the wage difference has
continued to decrease, in part, due to a high number of women entering the labor force, coupled
with the outlawing of gender discrimination and an increase in the number of women attending
colleges and professional schools (Stone, 2004). In a most recent study, it found that women now
earn 76% of men’s wages (Blau 2006). An even more interesting aspect of the gender wage
disparity is how it differs between races. For instance, in 1975 white women earned 43% less
than men, while today white women earn 21% less than they earn men (Green 2005). Similarly,
black women also earn less than black men do, but not as much of a margin as their white
counterparts. In fact, in 1975, black women also earned 25% less than black men did and in
2003, they only earned 11% less (Green 2005).
During the 1960s and the early 1970s, the racial wage disparity has narrowed
significantly but since then it has remained constant over the years. A large body of studies has
emerged in labor economics that tries to explain the reasons for early convergence and the recent
stagnation in wage disparity. The wage convergence of the 1960s and the 1970s was because of
2 | P a g e
improvements in the quality of schools attended by blacks (Smith and welch 1989) and the
enactment of anti-discrimination legislation (Heckman and Payner 1989). The slowdown in
convergence during the 1980s and 1990s is because of racial differences in the levels and return
to observed and unobserved skill (Neal and Johnson 1996). Neal and Johnson demonstrated that
nonwhites and whites differences in “pre market” skills as measured by the AFQT scores
account for about 3/4 of the disparity in wages and this is further emphasized by Bollinger in
2003 as he found the same result. Neal and Johnson focused on the AFQT scores, as they
believed it to be an appropriate measure of skill due to schooling levels are believed to be
“noisy” proxies of actual skill.
Using this improved measure, they find that discrimination explains less of the disparity
between races than previously found. From past studies the general rule of thought was that
discrimination had contributed anywhere from 1/3 to 1/2 of the wage disparity between races but
it may not be as simple as that and other factors may play an even more vital role. The role of
labor market experience is another explanation economists feel could explain the wage disparity
between races. Numerous studies have shown that blacks receive lower returns to experience
than whites do. This therefore has implications for the observed wage disparity between races. If
wages grow at a slower rate for blacks than they do for whites then this could in fact explain the
differences in wages observed between races during their careers as well as differences in there
labor force participation.
With there being a wage disparity between gender and race it would then be plausible to
think that nonwhite women would be the most effected group as they are hit with both being a
woman and being nonwhite. In the United States, two ascribed characteristics stand out for being
strongly associated with the labor force outcomes: gender and race/ethnicity. A large body of
3 | P a g e
work in both sociology and economics has been focused in recording the wage disparity by
gender and by race. In general, these works have found that women earn less than men do,
nonwhites earn less than whites do and such disparity cannot be just attributed to human capital
factors (Corcoran and Duncan 1979). It is therefore plausible to think that nonwhite women are
likely to face considerable, perhaps even sever, disadvantages in the labor market. Despite the
significance of this topic and the increasing attention given to gender and racial disparity in past
works, our empirical knowledge of the differences in the labor outcomes by gender and race is
surprisingly disappointing. Previous works on this topic have focused on either race or gender
disparity but in doing so have overlooked the interplay of both.
In studies, assessing the labor force outcomes of nonwhite women it is first necessary to
choose an appropriate group in which to compare them. One possibility would be to compare
female nonwhite workers to male workers of the same group, in better words the “gendered
centered approach”. Another possibility would be to compare female nonwhite workers to
female white workers, in better words the “race/ethnicity– centered approach” (Bound and
Dresser 1999). Even though the two approaches have the appeal in not confounding gender and
race, they preclude a direct comparison between nonwhite women and white men or indeed
between any two groups that differ from one another in both race and gender. This shortcoming
has given rise to alternative approaches. The first being to compare all gender-race combinations
simultaneously to one reference group, most likely being white men (Corcoran and Duncan
1979). The second being, to understand the gender effects by race and then sequentially to
understand the race effects by gender (Beron 1994). These approaches are useful because it
allows a degree of flexibility in empirical description of wage disparity by race and gender. This
so-called flexibility is preferable to the approaches mention above because it negates the strong
assumption of additivity. Under this assumption, nonwhite women presumed to incur two wage
disadvantages: one being associated with being female and another with being nonwhite.
Consequently, the total disadvantage faced by nonwhite women compared to that of white men
would simply be the sum of a gender penalty and a race penalty. Appropriately called, “double
jeopardy” (king 1988).
Research question and hypothesis
This study will examine the question of why there is a wage disparity between gander
and races, and the explanations for it. The two main hypothesis will be that men earn higher
wages than women controlling for certain factors and that whites earn more than non-whites also
controlling certain factors. If these two theories turn out to be true then it would then be plausible
to think that nonwhite female will be the most disadvantaged group and they are impacted by
