No abstract

Using data from the University of California and results from previously published research on the returns to higher education, this article presents a preliminary evaluation of the impact of ending affirmative action in admissions at a large, publicly funded university. At the undergraduate level, eliminating race as a factor in the admissions process will redistribute African Americans, Mexican Americans, and Native Americans away from the most competitive campuses (UC-Berkeley, UCLA, UC-San Diego) towards the less competitive campuses in the California State University system. This redistribution will lower the returns to schooling for those affected groups and could have a negative impact on the educational environment for all students. Affirmative action will, in the short run, reduce the number of African American, Mexican American, and Native American students admitted and, in the long run, will have an adverse effect on the delivery of legal and health care services to those racial and ethnic groups.

No abstract

Antidiscrimination laws are designed to prompt employers to stop excluding black workers from jobs they offer and from treating them unequally with respect to promotion and salaries once on the job. However, a moral hazard effect can arise if the existence of the laws leads black employees to bring unjustified claims of discrimination against employers. It has been argued that employers may become more reluctant to hire black workers for fear of being subjected to frivolous lawsuits.Using the Multi-City Study of Urban Inequality (MCSUI), we find that male and female black workers are far more likely than whites to report racial discrimination at work. This is the case even when a host of human capital and labor market factors are controlled for. Further, nearly all black workers who report they have been discriminated against on the job in the MCSUI Surveys also show statistical evidence of wage discrimination. This is not the case for white males or females. We find little evidence to support a moral hazard effect.