Just as a wage gap can be found in earnings of men and
women, a wage gap also exists among some racial and ethnic
groups in America. The controversial question is why the
wage gap exists - to what factors can it be attributed?
Research suggests various answers - skill disparity, differences
in work patterns, differences in choice of industry/occupation,
economic changes, and discrimination. Each of these possibilities
has different policy implications. Before any progress can
be made in eliminating wage disparity between racial and
ethnic groups, it must be determined which of the possibilities
is responsible for the wage gap.
One's level of education plays a big role in how much one
earns and will earn in the future. The combination of data
on level of enrollment and level of completion give a clear
picture of how different groups measure up to one another.
U.S. Census data on enrollment in primary, kindergarten,
elementary, high school, college, and college as a full
time student, reveals that while enrollment is very similar
among racial and ethnic groups for kindergarten through
high school, it varies substantially for college and college
full-time enrollment. While whites' college enrollment is
at 23%, blacks' is at 20%, Hispanics' is at 16%, and Asians'
is at 35%. For full time college enrollment, whites' is
at 16%, blacks' at 13%, Hispanics' at 10%, and Asians' at
26% (U.S. Census - 2).
However, rates of enrollment do not tell the whole story.
While rates of enrollment are very similar among all groups
for high school, Hispanics' and blacks' rates of high school
completion are lower than those of whites and Asians. According
to the U.S. Census Bureau, of all eighteen through twenty-four
year olds who were included in the census in 2000, 91.8%
of whites, 83.7% of blacks, 64.1% of Hispanics, and 94.6%
of Asians completed high school (NCES - 2). A similar trend
can be found for college completion. According to the Integrated
Postsecondary Education Data System (IPEDS) Graduation Rate
Survey published in 2003, blacks and Hispanics complete
college at lower rates also. Of all people who began college
in 1997, 59% of whites completed college within six years
or less, while only 40% of blacks and 42% of all Hispanics
that began college in 1997 completed it within the same
time period. A huge gap exists also in advanced degrees.
According to the U.S. Census Survey of Income and Program
Participation of 2001, out of the total 16,180,000 advanced
degrees held by people in America, 82.4% were held by whites,
6% were held by blacks, 3.6% were held by Hispanics, and
the rest by other minorities (U.S. Census - 1). As the data
reveals, at practically all levels of education, blacks
and Hispanics have a lower level of participation and completion.
Table 1: Group Completion Rates
* 18- through 24-year-olds
who had completed high school, by race/ethnicity: October 2000
** First-Time-In-College, Bachelor-Degree-Seeking
Students Enrolled fall 1997 Who Graduated from the Same College
or University by August 2003, IPEDS GRS.
Why is education so important? It has been
proven in various research that level of education and earnings
have a positive correlation. A study conducted by the U.S.
Census Bureau and published in "The Big Payoff: Educational
Attainment and Synthetic Estimates of Work-Life Earnings"
displayed this correlation. In estimating the work-life earnings
for full-time workers of different education levels, the article
revealed that while a white non-high-school graduate would
earn 1.1 million over a life time, the same individual with
an advanced degree would earn almost three times the amount
at 3.1 million dollars. For a black individual a similar trend
of earning growth exists with experience, however, non-high
school graduates would start out at .8 million dollars, while
a person with an advanced degree would earn 2.5 million. The
data for Hispanics and Asians is very similar to that of blacks,
except at the advanced degree level, Asians' earnings mirror
those of whites at 3.1 million (Cheeseman Day, 7). Thus, while
ultimately, blacks and Hispanics earn less than whites of
roughly the same level of education, there is a great return
on education for all racial and ethnic groups. In fact, the
return on education is greater for blacks and Hispanics because
in calculating the increase in earnings of a person who starts
out without even a high school degree and then works his way
up to an advanced degree, the increase in earnings for whites
is 280%, while for blacks and Hispanics it is 315%. The fact
that the return on education is actually greater for black
men than for white men is also confirmed by the National Center
for Education Statistics. Their study showed that in 2003,
black college graduates earned 60% more than black high school
completers, while black high school completers earned 30%
more than black workers who dropped out. On the other hand,
whites with a bachelor's degree or higher earned just 20%
more than whites who finished high school (NCES - 1).
Wages are not only affected by the level education of the
individual, but also correlate to the level of education
of the individual's parents. For whites and blacks whose
parents had less than a college education, whites consistently
earn more than blacks. However, in a situation where the
parents had some college education or more, blacks earn
more than their white counterparts (Black, 19).
While various data demonstrate that blacks and Hispanics
are less educated than whites and Asians when measuring
by degrees earned, the question that remains is why an earnings
gap remains for people of roughly the same level of education
but of different racial or ethnic groups. One explanation
is that the data available often does not control for both
level of education and years of experience. Just as in comparing
wages of men and women, women of all ages tended to have
less work experience than men, differing work patterns of
different racial and ethnic groups may have an affect on
wages and earnings.
Various resources show that a greater percentage of black
and Hispanic men than white and Asian men do not participate
in the labor force; of those people who are in the labor
force, there are twice as many blacks unemployed as whites.
Moreover, blacks and Hispanics tend to work fewer weeks
per year and fewer hours per week, are overrepresented in
temporary and on-call work, and tend to be unemployed for
longer periods of time than whites.
Rates of participation in the labor market, as well as
rates of employment and unemployment are one way to compare
work experience among racial and ethnic groups, which could
explain some of the gap in wages and earnings. The U.S.
Census Bureau report showed that in 2000 white people had
a higher rate of participation in the labor force, than
blacks, Asians, and Hispanics, with 64.6% of the total white
population, 60.2% of the black population, 63.3% of the
Asian population, and 61.4% of the Hispanic population,
participating. The same report showed that among all people
in the labor force in 2000, blacks had a higher rate of
unemployment than whites; the unemployment rate for whites
was 3%, for blacks 6.9%, for Hispanics 5.7%, and for Asians
3.2%. A review of the U.S. Census data for different years
shows that the gaps in the rates of unemployment among different
groups have proportionally persisted over the years. Whether
it is by choice or due to other factors, statistically,
black, Hispanic, and even Asian people overall are employed
less than whites (Spalter-Roth, 2).
Table 2: Labor Force Participation, Employment,
Unemployment in 2000*
||In Labor Force
*Original Source: U.S.
Census Bureau, 2000. "Profile of Selected Economic Characteristics."
Census 2000, Summary File 4, DP-3.
The differences in number of weeks worked
per year and number of hours worked per week by the different
racial and ethnic groups may also reveal information about
the gap in wages and earnings. According to the California
labor market data, among all working men compared in 2000,
blacks worked 46 weeks per year on average, while whites worked
48. In terms of hours worked per week, blacks and Hispanics
worked about 41 hours per week, while whites worked 44 hours
per week (Reed). This is also reflected when hourly wages
are compared to annual earnings. According to "Basic Skills
and the Black-White Earning Gap" by Neal and Johnson, black
men in America earn 48% less per year than whites of the same
age, even though their wages are only 24% lower (Johnson,
12). This statistic suggests that black men may be working
less time overall.
The type of jobs people hold can greatly affect their wages
also. According to "The Big Payoff" the earnings of workers
who work full time year round tend to be significantly higher
than the earnings of workers who work part time or just
part of the year (Cheeseman Day, 2). When compared to whites,
blacks' and Hispanics' participation in non-standard work
(regular part-time, temporary help agency, on-call/day labor,
self employed, independent contractor) is proportional to
the size of its population, and maybe even slightly low.
However, in two worst areas of non-standard jobs - temporary
and on-call labor, both of which tend to pay little and
offer few benefits, if any, blacks and Hispanics are over
represented. While blacks made up only 12% of the U.S. population
in 1997, they made up 20% of all temp workers in the U.S.
In the same year, Hispanics represented 13% of the temp
workers and 15% of all on-call/day laborers (Hudson, 12).
Moreover, whether people work full-time or non-standard
jobs is often closely tied to their level of education.
For example, according to "The Big Payoff," high school
dropouts are less likely to work full time and year round
than people with bachelor's degrees. While only 65% of high
school dropouts worked full time and year round in 2000,
77% of people with bachelor's degrees worked the same amount
(Cheeseman Day, 2).
Another important factor that must be considered is whether
there are differences between how long people of different
racial and ethnic groups are unemployed. Hispanics and blacks
are more likely than whites to be unemployed for longer
periods of time. In 2000, 29% of all long-term unemployed
Americans were black, 16.9%, were Hispanic, and 48.3% were
white. When compared to the percentage each racial and ethnic
group makes up in the total population (whites - 69%, blacks
- 16%, and Hispanics - 12%), it is clear that blacks and
Hispanics are disproportionately represented among the long-term
unemployed group. Moreover, when compared to the 20% that
blacks made up of the total unemployed in 2000, the 29%
is very high. Of all people long-term unemployed, blacks
had the highest percentage of people that were unemployed
for over six months at 22.7%, while whites had 17.6%, and
Hispanics had 14.2% (Stettner, 2).
Table 3: Long-Term Unemployment
||Long Term Unemployed
Over 6 Months*
* % rate of the Long Term Unemployed
Choice of Industry/Occupation
Besides the differences between racial and ethnic groups
in work patterns, differences can also be found in their
choices of industry and occupation. According to the U.S.
Census Survey of 2000, 35.6% of white men, and 44.6% of
Asian men were employed in managerial, professional and
related occupations, compared with 25.2% of black men and
just 18% of Hispanic men. On the other hand, about 40% of
black and Hispanic men held jobs in service, production,
transportation, and material moving occupations, compared
to 27% of white men and Asian men. A disproportionately
high percentage of black and Hispanic women compared with
white and Asian women held jobs with poor pay, few benefits,
and little career mobility such as food preparation, cleaning,
and personal care (Spalter-Roth, 4).
Table 4: Occupational Data for Employed
Population 16 and over*
*Original Source: U.S. Census
Bureau, 2000. "Profile of Selected Economic Characteristics."
Census 2000, Summary File 4, QT-P28.
These statistics beg the question why people
of different races end up in different occupations. One answer
is obvious - differences in education.; because a great percentage
of blacks and Hispanics do not acquire a high school or a
college degree, they work jobs in service, production, transportation,
and material moving. Another reason may be the existence of
so called "ethnic niches". New York city provides a broad
example of ethnic niches; there, Hispanics predominantly work
in construction, Asians run laundry mats and dry cleaning
businesses, white men work as fire fighters, etc. While such
niches can help members of the prevalent racial or ethnic
group at that job obtain a job by providing training and shelter
from discrimination, such jobs pay less, and can often constrain
job mobility. Once an ethnic niche is created in a certain
occupation or industry the desirability and availability of
the job becomes limited (Spalter-Roth, 5).
Another difference could be simply the variation in choices
made by people of different racial and ethnic groups in
college. According to "Why Do Minorities Earn Less? A Study
of Wage Differentials among the Highly Educated", the index
of dissimilarity indicates that 14% of Hispanic men, 20%
of black men, and 31% of Asian men would need to change
their major to match the distribution of majors among whites.
Asians, for example, are more likely to major in engineering
than any other group, while black men tend to be underrepresented
in engineering and over represented in education. Black
men also choose majors that on average have a higher fraction
of women, while Asian men choose majors that have a lower
fraction of women (Haviland, 12).
One other possibility that could explain why people of
different racial and ethnic groups end up in different occupations,
is discrimination. Rather than looking at each person's
credentials like education and experience, employers look
at skin color, and base their hiring decisions on racial
and ethnic identities of past employees. For example, if
in all the years of a company's existence the position of
vice-president has been filled by a white male, it may take
a long time before a woman or a minority will be hired for
that position, simply because the hiring personnel may feel
more comfortable giving the position to someone who is similar
to other people who have held that position in the past.
Thus, blacks continue to be hired for certain types of jobs
in certain occupations, reinforcing existing ethnic niches.
One important factor that may shine some light on the
cause of the wage gap between racial and ethnic groups is
skill. While looking at the level of education has been
the traditional and common way to determine one's ability
level and predict future wages, recent researchers have
contended that this information can be misleading because
the quality of schools and intensity of education in different
schools vary greatly in America. Just as age is not a valid
predictor of one's level of education, the amount of schooling
one has doesn't truly reveal that person's ability. In "The
Role of Premarket Factors in Black-White Wage Differences"
Derek Neal and William Johnson discuss a different measure
of education - skill. For their research, Neal and Johnson
used the scores from the Armed Forces Qualification Test
(AFQT) found in the National Longitudinal Survey of Youth,
to examine the black-white wage gap among workers in their
late twenties (age 26-29). The AFQT is known to be a racially
unbiased measure of basic skills that helps predict job
performance, and is often used in military testing. The
data set included a sample of individuals who were tested
at ages 16-18, just before they entered the labor force
full time or began their secondary education. Testing for
math and reading skills, the results of the test revealed
that three-fourth of the racial wage gap for men is due
to a skill disparity. For women, the test scores explained
all of the wage disparity. In fact, when the AFQT scores
were held constant for white, black, and Hispanic women,
black and Hispanic women earned more than white women.
Carneiro, Heckman, and Masterov, the authors of "Labor
Market Discrimination and Racial Differences in Premarket
Factors," sampled the children of the mothers in the 1979
NLSY to see if ability disparity can be found in children
before they enter school. Their data from the Children of
the National Longitudinal Survey of Youth of 1979 (CNLSY79),
showed that minorities do in fact enter school with lower
measured ability than whites, and the gap in ability widens
as the children obtain more schooling. However, the increase
in gap with schooling is much less significant than the
original gap. According to the CNLSY79, 5-6 year old black
boys scored 18 percentile points below white boys of the
same age, while Hispanic boys scored 16 percentile points
below white boys. These findings are consistent for the
different tests and in various data sets. Schooling, rather
than closing the gap, substantially widens it. By ages 13
to 14, the gap in scores widens to 22 percentile points
for blacks, and remains the same for Hispanic boys
at 16%. Therefore, when they enter the market, they have
a much poorer set of skills than whites.
Besides the disparity that exists in cognitive skills,
disparity is apparent also with non-cognitive skills such
as motivation, self control, time preference, and social
skills. In the CNLSY, mothers were asked age-specific questions
about the anti-social behavior of their children, including
aggressiveness, violent behavior, cheating, lying, disobedience,
peer conflicts, and social withdrawal. The results showed
that by age 5 and 6, the average black is roughly 10 percentile
points above the average white (the higher the score, the
worse the behavior). This gap is important because non-cognitive
skills are directly related to what the labor market calls
"soft-skills". These skills involve ease of interaction
with colleagues and customers, enthusiasm and a positive
work attitude - all skills essential in a service driven
economy. Thus, if such disparities in social ability exist
at such a young age, they can have very negative effects
in the future, unless some sort of intervention occurs (Carneiro,
19-20). In fact, it has been documented that black men are
at a particular disadvantage during job interviews, because
their body language and communication skills often do not
meet employer expectations regarding politeness, indications
of motivation, or enthusiasm (Spalter-Roth, 7).
All of this information on skill disparity begs for some
explanation for the cause of the skill disparity between
racial and ethnic groups. According to Neal and Johnson,
the ability disparity can be explained by varying school
and home environments. In fact, the authors found that children's
scores on the AFQT correlated with the level of education
and the professional status of their parents, the number
of children in the family, measures of family reading material,
and school characteristics of the children (including student/teacher
ratio, disadvantaged student ratio, dropout rate, teacher
turnover rate) (Neal, 887). According to Carneiro, Heckman,
and Masterov, however, most of the important factors would
be those related to the family environment, since ability
gaps are substantial before children even enter school.
Among the factors they mention are measures of family background,
family income, mother's level of education, home environment,
and mother's cognitive ability. More specifically, black
and Hispanic children tend to come from much poorer and
less educated families than white children. They are more
likely to grow up in broken or single parent homes. The
home score, which is based factors such as the number of
books, magazines, toys and musical recordings available
to the child, family activities, methods of discipline and
parenting, learning at home, TV watching habits, home
cleanliness and safety, etc, is always higher for whites
than for blacks and Hispanics (Carneiro, 8-11). All of these
factors may explain the cause of the skill disparity between
racial and ethnic groups.
We have addressed why the gap exists among racial and
ethnic groups before school begins. Now we must address
why this gap widens as the children get older and obtain
more education. The positive effect of schooling on test
scores is much larger for whites and Hispanics than it is
for blacks. This could be explained by the fact that whites,
blacks and Hispanics start school at different levels; since
blacks and Hispanics start with much lower abilities than
whites, their subsequent progress and success is less than
that of whites. The quality of schools attended by black
and Hispanic children in comparison to white children could
also explain the lower effect of schooling on the former
groups relative to the latter group. Thus, differential
initial conditions and differential school quality may also
be important determinant of the adult black-white skill
gap (Carneiro, 14-17).
Another important explanation for the widening of the
skill gap with schooling is expectations of the students.
For instance, in a given survey, 22% blacks and Hispanics
reported that they expected to die next year, in comparison
to 16% of whites. Blacks and Hispanics also report higher
expectation of committing a crime and being incarcerated
(Carneiro, 18). Such unfortunate expectations could certainly
reduce how much those two groups invest in their own human
capital - how often they attend school, study, do their
homework, and participate in class. All of these factors
affect their skills and ability, which is subsequently reflected
in future wages. There is the possibility that pessimistic
expectations of black and Hispanic parents lower their investment
in their children, which translates into lower levels of
ability and skill of those children.
Immigration and Language Disparity
Language disparity plays an important role in wage determination,
and according to "Labor Market Costs of Language Disparity:
An Interpretation of Hispanic Earnings Differences" explains
up to one-third of the relative wage difference between
Whites and Hispanics in America. The wage disparity that
is usually attributed to ethnicity, nativity, and time in
the United States, can in fact be explained by differences
associated with English language skills. In the data sample,
all the Hispanics were divided into four groups of English
proficiency: fluent, very well, well, not well. The findings
showed that Hispanic men in the fluent group have earnings
insignificantly different from whites who have the same
school and potential work experience, as well residency
in the same geographic area. Moving a member of the "very
well" group up to full English fluency would raise his wages
by 10%, a "well" member to full fluency by 17%, and a "not
well" member to fluency by 26% (McManus).
Similar results were found in "Why Do Minority Men Earn
Less?" Here, the authors found that the status of immigration
and whether English is spoken at home both affect earnings.
Generally for non-immigrants, if a language other than English
is spoken at home, the people earn less than those who speak
only English at home. When comparing all immigrants, those
who do not speak English at home earn substantially less
than those who do. Moreover, when all people who do not
speak English at home are compared, the immigrants earn
substantially less than non-immigrants. Thus, it can be
concluded that one's immigration status as well as what
language one speaks at home both affect earnings. When non-immigrants
of different racial/ethnic groups who speak English at home
are compared, Hispanics and Asians earn just slightly less
than whites. However, when all non-immigrants who do not
speak English at home are compared, all groups including
whites, blacks, Hispanics, and Asians earn about the same
with blacks earning slightly more than whites, Hispanics
earning slightly less, and Asians earning more. From the
data above, it appears that immigrants who do not speak
English at home are the lowest earning group in America.
Unfortunately, 37% of all Hispanics, and 70% of all Asians
in the U.S fall into this category (Black, 16-17).
Table 4: Wage Gaps by Language Spoken at Home and Immigration Status
|Speaks only English at home
a language other than English at home
||Speaks only English at home
||Speaks a language other than English at home
According to the U.S. Department of Labor, there are other
things that could affect the wage disparity, and in fact
made earnings more unequal during the 1980's and 1990's
- these are technological change, trade liberalization,
increased immigration, value of the minimum wage, and declining
unionization. The economy has transitioned from being driven
by manufacturing to information. Thus, as technology continues
to advance, the demand for skilled workers who are able
to operate the advanced technology and contribute to its
development continues to grow. Moreover, technological advancements
are causing the replacement of lesser-skilled jobs with
automated devices, and thus demand for lesser-skilled workers
is dropping. This situation is aggravated by the increase
in immigration that has been occurring since 1965. Particularly,
less-skilled workers with lower education levels have and
continue to immigrate to the U.S., which increases the competition
for unskilled jobs and drives wages down for unskilled-workers.
Expanded trade also drives down the wages of low-skilled
workers because it displaces the goods they produce. A decline
in unionization in the 1980's has also contributed to increased
wage inequality, because fewer workers are benefiting from
collective bargaining. Finally, the minimum wage fell in
real terms during both the 1970's and 1980's reaching a
level in 1990 significantly below its 1960 level.
What does all of this information mean? It is important
to have a clear understanding of whether the wage disparity
is a result of discrimination in rewarding blacks and Hispanics,
or is a result of the disparity in education, skills, hours
of work, types of work, and types of job, that exist among
different racial and ethnic groups. The distinction is important
because the two different explanations have different policy
implications. "If persons of identical skill are treated
differently on the basis of race or ethnicity, a more vigorous
enforcement of civil rights and affirmative action in the
market place would appear to be warranted. If the gaps are
due to unmeasured abilities and skills that people bring
to the labor market, then a redirection of policy towards
fostering skills should be emphasized" (Carneiro, 3).
Black, Haviland, Sanders and Taylor, Lowell. "Why
Do Minority Men Earn Less: A Study of Wage Differentials
Among Highly Educated". Department of Economics, Syracuse
Carneiro, Heckman and Masterov, Dimitry. "Labor Market
Discrimination and Racial Differences in Premarket Factors".
Institute for the Study of Labor. January 2005.
Cheeseman Day, Jennifer, and Newburger, Eric. "The Big
Payoff: Educational Attainment and Synthetic Estimates of
Work Life Earnings" U.S. Census Bureau. July 2002.
Dunifon, Rachel. "Race and Gender in the Labor Market",
Joint Center for Poverty, 1999, Vol III, No.1.
Hudson, Ken. "No Shortage of Non-Standard
Integrated Postsecondary Education Data System. "Graduation
Rates for Selected Bachelor-Degree-Granting Colleges by
Race/Ethnicity and Gender".
Johnson, William, and Neal, Derek. "Basic Skills
and the Black White Earnings Gap".
McManus, Walter. "Labor Market Costs of Language
Disparity: An Interpretation of Hispanic Earnings Differences",
American Economic Review, September 1985.
National Center for Education Statistics (NCES). "Annual
Earnings of Young Adults by Race/Ethnicity." 2005.
National Center for Education Statistics (NCES). "Dropout
Rates in the United States: 2000".
Neal, Derek and Johnson, William. "The Role of Premarket
Factors in Black-White Wage Difference", Journal of
Political Economy, October, 1996.
Office of the Secretary. "Factors Driving High and
Low Skilled Workers' Wages Farther Apart". Washington,
D.C.: U.S. Department of Labor.
Reed, Deborah, and Cheng, Jennifer. "Racial and Ethnic
Wage Gaps in the California Labor Market." Public Policy
Institute of California. 2003.
Spalter-Roth, Roberta, and Lowenthal, Terri Ann. "Race,
Ethnicity, and the American Labor Market: What's at Work",
American Sociological Association, ASA Series on How Race
and Ethnicity Matter, June 2005.
Stettner, Andrew, and Wenger, Jeffrey. "The Broad Reach
of Long-Term Unemployment", Economic Policy Institute, 15
Thomas, Scott. "Post Baccalaureate Wage Growth within
Four Years of Graduation: The Effects of College Quality
and College Major". School of Industrial and Labor
Relations, Cornell University." January 2000.
U.S. Census Bureau. "Bachelor's Degree Field of Advanced
Degree Recipients by Sex, Race and Hispanic Origin, and
Age." Survey of Income and Program Participation.
U.S. Census Bureau. "School Enrollment of the Population
3 Years Old and Over, by Level and Control of School, Race,
and Hispanic Origin: October 1955 to 2004".