Does racial discrimination continue to limit the employment opportunities available to minorities? This has been a controversial issue in economics. Statistical evidence of racial disparities in employment outcomes across groups does not necessarily imply that some groups face discrimination, since there are many other personal characteristics and preferences of workers that are correlated with race. Recent evidence suggests that, when we net out racial differences in educational attainment and/or cognitive ability (such as test scores), there remains little difference in wages between whites and blacks(11) or between whites and Mexican Americans.(12) Of course, discrimination in housing markets or unequal funding of school districts could help to generate differences between groups in average educational attainment or quality.
Wage vs. Employment Effects
Furthermore, it would be incorrect to conclude that labor market discrimination is no longer a factor for minorities, especially blacks. For one thing, the evidence cited above is based on hourly or weekly wages rather than employment rates; major racial differences in the latter still can be found even after netting out these differences in skills. For instance, Neal and Johnson report that significant racial differences in annual earnings, which reflect employment rates over the year as well as wages, remain even after controlling for test score differences.
These findings are consistent with the recent results of "audit" studies in the labor market, in which matched pairs of white and black job applicants with equal credentials are sent out to apply for advertised jobs. These studies generally show that white applicants are more likely than equally qualified blacks to receive job offers.(13) This evidence is also supported by ethnographic studies of employers,(14) which reveal that very negative perceptions of African American workers are held by many employers.(15)
The fact that discrimination may persist in employment but not in wages could also result partially from how Equal Employment Opportunity (or EEO) laws are administered in the U.S. Most EEO cases currently involve charges relating to discharges or promotions, rather than hiring activity.(16) Employers might therefore face a higher probability of lawsuits when they do hire minorities than when they do not, which might then lower their willingness to hire from these groups.(17)
Differences in Discrimination by Sector and Minority Group
Recent evidence from studies of employers also suggests that hiring discrimination against blacks is much more severe at some kinds of firms than others. For instance, black applicants are much more likely to be hired at large establishments than at smaller ones, and they are less likely to be hired in jobs involving contact with white customers.(18) The latter may be part of a larger pattern of greater discrimination against blacks at suburban than central-city establishments. The evidence also suggests that hiring discrimination is more severe against black males than females and against blacks than Hispanics.(19) These inferences are based on comparisons of hiring or employment rates of specific demographic groups with the rates at which they apply for jobs at various kinds of establishments. While the tendency of any group to be hired relative to its share of the applicant pool might reflect relative skills or other factors, these seem unlikely to account for the particular patterns that we observe in the data. The relatively greater preference for Hispanics likely reflects a broader preference among employers for immigrants over native-born blacks, in jobs that do not require cognitive or verbal skills, because of a stronger perceived work ethic among the former.(20)
Summary: Does Discrimination Matter?
Of course, if there are sufficient numbers of non-discriminating employers relative to the size of the minority labor force, it might well be possible for minorities to avoid the adverse effects of discrimination on their employment or earnings by applying for work primarily at non-discriminating firms. Indeed, there is some evidence that Hispanics may successfully be doing so, while blacks are not. Hispanics are hired in rough proportion to their share of the applicant pool, while blacks are hired much less proportionately.(21) But even if there are sufficient numbers of nondiscriminators in the market, and if they could be clearly identified, the employment opportunities of minorities are likely to be limited by other barriers when they seek employment in these firms (such as those associated with skills, etc.). Indeed, skill demands facing noncollege graduates seem to be relatively higher in larger firms and in those located in the central cities, precisely those where discrimination against blacks seems least severe.