Home Home | About Us | Sitemap | Contact  
  • Info For
  • Professionals
  • Students
  • Educators
  • Media
  • Search
    Powered By Google

VI. ECONOMIC EFFECTS OF AFFIRMATIVE ACTION ON TARGET GROUPS

It is also important to know whether affirmative action has improved the employment status of women and minorities. Research on the issue of target group attainment has revolved around three categories of outcomes: (a) employment rates, (b) income attainment, and (c) promotion rates and occupational attainment.

A. Employment of Women and Minorities in Organizations

Several studies have examined how employment has varied as a function of affirmative action policies. Leonard (1990), in a review of the economic literature, concluded that affirmative action increased the proportion of Black males in federal contractor firms in the early 1970s, although this was frequently limited to nonskilled occupations (e.g., Beller, 1978; Burstein, 1978; Goldstein & Smith, 1976; Heckman & Wolpin, 1976). Results for Black females and White females are less consistent in showing gains (e.g., Goldstein & Smith, 1976; Heckman & Wolpin, 1976). However, between 1974 and 1980, Black male and female employment increased significantly faster in organizations that were growing and in federal contractors than in other establishments (see further Leonard, 1984a, 1984b, 1984c).

In an analysis of EEO1 compliance reports, Smith and Welch (1984) demonstrated a shift in Black employment from noncontractor to contractor firms during the years from 19661970. They also concluded that Black males were three times as likely to report they were managers or professionals in 1980 than in 1966; women and Black females were twice as likely to do so. Likewise, Osterman (1982) concluded that women employed in industries that received greater attention from the Office of Federal Contract Compliance and had higher rates of federal purchases had lower quit rates than women not employed in these industries. Warner and Steel (1989) reviewed data from over 280 municipal police departments and concluded that departments with a stronger commitment to hiring women had greater utilization rates of women in policing jobs, although this difference was sometimes dampened in times of budgetary stress.

More recently, Uri and Mixon (1991), in an analysis of male and female employment data from 1947 to 1988, demonstrated that women aged 2054 had more stable employment from 19651980, although they lost some of these gains during 19811988. Men showed the opposite effect. Uri and Mixon also suggested that affirmative action programs increased women's share of projected employment while decreasing that of men. Other researchers suggest that the magnitude of these latter changes may be quite limited (for reviews see Clayton & Crosby, 1992; Johnson, 1990; and Leonard, 1989).

Research using more focused samples also suggests some effects of affirmative action policies on employment outcomes. Konrad and Linnehan (1995a), using a sample of 138 firms in a major metropolitan area, demonstrated that human resource management policies that took protected category status into account were positively associated with the percentage of people of color in the firm's management and the rank of the highest woman in the firm. However, these policies were not significantly associated with the percentage of women employed by the firm, the percentage of women in management, the percentage of people of color employed by the firm, or the rank of people of color. Konrad and Linnehan also found that human resource management policies that did not take protected category status into account were not associated with any of their measures. In contrast, government contractor status was positively associated with the percentage of women employed and the percentage of women in management. Further, firms which were subjected to compliance reviews had lower percentages of female employees and female managers than did firms which were not subjected to compliance reviews.

B. Income Attainment

Examinations of income attainment by women and minorities before and after the enactment of civil rights legislation has yielded somewhat inconsistent results. Burstein (1978) suggested that salaries of non-Whites were positively affected by affirmative action legislation. Smith and Welch (1984) reported substantial wage gains for Black men and women during 19671981 but suggested that most wages gains came prior to 1974. Son, Model, and Fisher (1989) found that the income gap between Blacks and Whites increased during the years between 1974 and 1981. This finding was moderated by educational attainment. Blacks with a high school education or less earned less than their White counterparts. However, young Black college graduates narrowed the income gap such that their income was quite similar to that of college educated Whites. In contrast, Crosby, Allen, and Opotow (1992) concluded that affirmative action legislation has not contributed to widening the income gap between middle and lower class Blacks. Leonard (1990) suggested that affirmative action increased the occupational attainment of non-White males and therefore has narrowed the wage gap between Whites and non-Whites.

C. Promotional and Occupational Status

Research has also examined rates of promotion and occupational status of women and minorities before and after the implementation of federally mandated equal employment opportunity programs begun in the early 1970's. This is relevant to the present review because the first principle of affirmative action is that the organization must abide by a policy of equal opportunity. EEO programs implemented in elementary and secondary public education systems of Oregon and New York reduced discriminatory hiring practices by about half (Eberts & Stone, 1985). DiPrete (1987) and DiPrete and Soule (1986) also found that EEO programs had positive effects on the promotion of nonWhite and female lower level employees to higher level positions in federal agencies. DiPrete and Soule point out that female and nonWhite lowerlevel employees did not have a greater probability for promotion than did male and White lowerlevel employees. Leonard (1990) reported that Black males increased their representation in skilled trades from 1974 to 1980 in federal contractor firms, although the employment rates of all Blacks did not increase over these years. Black females also increased employment share in all occupations except technical, craft, and whitecollar trainee (see Leonard, 1989 for a review of the impact of affirmative action on women's employment). Son et al. (1989) found that Blacks' occupational status was moderated by level of education. College-educated Blacks fared as well or even better than their White counterparts; Blacks with high school education or less fared more poorly than their White counterparts.

D. Summary

Research on the impact of affirmative action on demographic group attainment indicates that, in general, gains have been made, although there is some evidence of losses for those with low educational levels. All three categories of group level outcomes (employment rates, income attainment, and promotion rates and occupational attainment) show some improvement over the time period in which affirmative action has been implemented. These field studies do have several limitations, however. First, they have often have used proxy measures of affirmative action such as the percentage of minorities employed or the date of the implementation of a federal program. These measures have several drawbacks for examining specific responses of recipients. They lack precision and provide little information on what seems to be a crucial factor--the specific implementation strategy adopted by each firm. Second, such studies may be insensitive to certain variables. For example, individuals who are hurt by a policy may leave the organization and not be included in the study. The resultant sample may present an inaccurately rosy picture. Third, performance measured at the firm or aggregate level provides no information about individual responses to selection processes or the potential distribution of those responses. Finally, singlewave time series and crosssectional designs, such as those studies measuring promotion rates before and after the implementation of EEO programs, are subject to multiple interpretations that make it difficult to infer causal relations and to rule out alternative explanations for the findings.