by Kevin Stainback and Donald Tomaskovic-Devey. New York: Russell Sage Foundation, 2012. 412pp. Paper $45.00. ISBN:978-0-87154-834-4.

Reviewed by Leslie F. Goldstein, Department of Political Science and IR, University of Delaware, LESL [at] UDEL.EDU


editor's note: A coauthor notified the reviewer that the definition of "segregation" used in the review published on 6.19.13 was incorrect. The first two lines of paragraph 2 of this review have been revised to reflect the correct definition

This is a thick book by two sociologists, replete with charts, tables and sophisticated statistical analyses; the sheer listing of figures and tables in the front matter runs to nine pages of small print. The book purports to document the degree to which the United States has succeeded (or not) in desegregating its workplaces by race and gender since 1966, the first year that the EEOC issued reports on the racial and gender composition of the workforce. The book traces these reports up through 2005, and makes an argument about why we have seen this change. Law and society scholars will recognize an interplay of political institutions and private action, but scholars of courts and law have reason to question some of the story told here.

The book’s definition of “segregation” is the degree to which two racial or gender groups work in separate occupations within a workplace as measured by proportion of the labor pool; perfect desegregation would mean that if blacks are fifteen percent of the labor pool, every occupation within a workplace would be fifteen percent black. The book also examines progress toward what one might think of as race and gender equity in the workplace by assessing the degree to which blacks and women are represented proportionately in the higher status job categories of “Highly skilled craft worker,” manager, or professional. “Proportionately” in this sentence refers to proportion of the available labor pool represented by the group. White males were 60% of the labor force in 1966 but only 38% of the labor force in 2005 (p.30). In 1966 white males held 93% of managerial jobs (p.246); this percentage decreased significantly by 2005 but the authors still consider white males’ advantage to have been INCREASING during this period (and especially during the years 1966-1980) because their share of the labor pool decreased even more sharply.

The basic picture set forth by this book is that racial segregation in employment decreased substantially in the years 1966 to 1972 when employers were maximally uncertain how the federal government was going to enforce the antidiscrimination provisions of the 1964 Civil Rights Act, and perhaps decreased to a similar degree beginning in 1961, when President John Kennedy issued an executive order calling for “affirmative action” by all federal contractors in hiring and promoting racial minorities. (These authors’ discussion of executive orders from presidents Franklin [*297] Roosevelt through Eisenhower is misleading, because from time to time they blur together prohibitions on discrimination with requirements for affirmative action, at e.g., pp.51, 108, 113.) That this desegregating trend began in 1961 is suggested by Frank Dobbin (2009), whose work is frequently cited by Stainback and Tomaskovic-Devey, but the EEOC reports do not cover the period of 1961-1965, so it remains in this book a surmise. Dobbin reported that in 1960 five percent of employers had antidiscrimination policies in place for racial minorities (whereas many other firms openly discriminated), and by 1966, this percentage of antidiscrimination firms had doubled. Among the largest firms, percentages were far higher; by 1967, 71% of “medium and large firms” had such policies (Dobbin 2009, pp.53-54). Stainback and Tomaskovic-Devey also document that state-level fair employment laws had been adopted in 18 (non-Southern) states before 1964 and these appear to have significantly reduced anti-black discrimination by employers prior to 1966 (pp.74-78).

Documenting Desegregation examines three separate stages of desegregation over four decades. After the uncertainty period of 1966-1972, when blacks made their most rapid advances, the period of 1972-1980 is termed by Stainback and Tomaskovic-Devey the “regulatory decade,” when the federal government actually took clear steps to enforce the Civil Right Act and made clear that gender segregation too would be treated as unlawful. In 1967 President Lyndon Johnson issued an executive order that added women employees to the group for whom contractors had to take “affirmative action.” In 1973, the Supreme Court declared unlawful gender-segregated want ads for jobs (“Help Wanted: Female”; “Help Wanted: Male”) (Pittsburgh Press v. Human Relations Commission; see Baer and Goldstein 2006, pp.134-135). During this period Congress granted the EEOC the power to file lawsuits against discriminating employers and in 1968 the Office of Federal Contract Compliance Programs notified contractors and subcontractors that they needed to maintain annual affirmative action reports with goals, plans, timetables, and progress reports. These were to be kept in order to be provided should the agency decide to do a compliance review of that firm (p.86). Richard Nixon played a strangely pivotal role in this period. Even as his administration and his 1972 electoral campaign took positions (such as opposition to desegregative busing) to attract Southern racists to the Republican fold, Nixon in 1969 endorsed the Philadelphia Plan, which called for the use of affirmative hiring quotas by federal contractors (pp.123-125), and authors tell us that the “regulatory apparatus organized through the EEOC and the OFCCP were strengthening across the decade” – i.e. from 1972, when Nixon was still President, through the administrations of Ford and Carter (p.126).

This book tells us that desegregation for women of both races improved substantially in this period and that desegregation for black men continued but at not as rapid a pace as in the first period (pp.126-131). To authors’ surprise, while black women made sizable advances with respect to working alongside white men, their segregation [*298] from white women intensified (p.131). This worsening pattern for black women vis-a-vis white women continued up through 2000, when it effectively froze in place through 2005 (pp.168-169). White women in this 1972-1980 period made particularly rapid progress into the higher status jobs of manager and professional. Black men and black women progressed only slightly in this measure.

By the late 1970s public support for affirmative action noticeably cooled. Authors report misleadingly that the Supreme Court “redefined” affirmative action as the “crime” of “reverse discrimination” (p.126). (In fact, the Court in Fullilove v. Klutznick (1980) upheld minority owned set-asides in federal contracting. I return below to their discussion of the Supreme Court). And President Ronald Reagan made clear that he did not favor it. His election introduced what authors term the Neoliberal Era of 1980-2005. In this period, EEOC funding and staffing were cut and ditto for the funding and staffing of the Office of Contract Compliance Review. EEOC funding came back somewhat during the Clinton years but not its staffing (pp.158, 167). The basic picture for this era is that black male segregation stagnated at its 1980 level, while women of both races continued to desegregate the workplace, although black women far more slowly than white women. White women in particular advanced into desirable managerial and professional jobs, and even skilled craft jobs up through 2000. Authors claim that white women’s advances into higher status jobs ceased or even declined after 2000, but this is again, because they are looking at the question of advance RELATIVE to the labor pool. In other words, in a given year, if there are 5% more women in a given labor pool and they increase their percent of professional jobs by only three percent, authors describe this as a decline in professional jobs for women.

At this point (about midway) the book becomes more nuanced and traces which parts of the country and which industries are better and worse for minorities, and also attempts to unravel what forces make a difference. Authors survey the literature on the latter question and are particularly swayed by the arguments of Dobbin that it is within-firm practices more than lawsuits or threats of them or federal compliance reviews that produce the most opening up of opportunities. In particular, those firms that set specific targets for hiring and promotion of members of disadvantaged groups, and hold managers accountable for their progress toward the goals are the ones where progress is made (ch.8). Also, because managers are inclined to hire others like themselves, the more diverse the managerial ranks are, the more diverse will be the opportunities available to a range of social groups (p.291). Also those workplaces that hire based on educational credentials, such as universities, are the least segregated. Authors also offer suggestions for how our society could make even more progress toward desegregated workplaces, were there a commitment to do so. For America’s workplaces to be what these authors consider “desegregated,” approximately one half of all workers would have to exchange places.

This book is thick with arguments and statistics to which a short review cannot do justice, and it does seem correct in [*299] terms of its basic historical trajectories. But it has a couple of problems. One is captured in the comment on its final page of text. Authors accuse our society of a “misplaced faith in legal changes” (p.321). Although authors provide two pages of tables of Supreme Court decisions (pp.120 and 165), they do not treat them as having mattered much. They do not discuss the Weber (1979) decision, beyond mentioning it in one table, and they omit the Bakke decision (1978) from the table altogether, although they do give it literally a few words “deemed affirmative action to potentially be ‘reverse discrimination’ against whites in … college admissions” (p.121). Neither Fullilove v. Klutznick (1980), Richmond v. Croson (1989), FCC v. Metro Broadcasting (1990), nor Adarand Contractor v. Pena (1995) are mentioned in this volume.

This neglect of many Court cases causes them to miss certain things. For instance, they treat the shift by human resource managers from the rhetoric of affirmative action to that of “diversity management” as a rebranding by personnel managers because affirmative action rhetoric was no longer politically legitimate. The reader is never told that the Supreme Court in Bakke in 1978 had singled out “diversity” as of compelling interest (for state universities) and constitutionally legitimate, whereas the rectifying of past or ongoing societal discrimination had been declared by the Supreme Court a not constitutionally justified grounds for race-based affirmative action (by a state university). Nor is one told that to justify affirmative action in distribution of broadcasting licenses the Supreme Court reiterated the importance of having a “diversity” of views presented on the air (FCC v. Metro Broadcasting, 1990).

In other words, their treatment seems to underrate the impact of the Supreme Court in developments beginning in the late seventies. (For some reason, by contrast, they are very attentive to the Supreme Court’s role and its interaction with Congress over the matter of pregnancy discrimination). They see social movements and executive branch enforcement as causal motors (to the degree that these prompt effective in-firm practices) but downplay the role of courts.

The authors brand as “clearly a legal error of gigantic proportions” the courts’ acceptance (for showing companies are making “good faith” efforts to desegregate) of affirmative efforts by firms to offer “diversity training” or the public posting of job openings without a requirement of actual hiring goals and manager accountability therefor (p.290). This could be a legal error only if judges believed such hiring goals to themselves be legal. There is good reason to believe that many judges, including a majority on the current U.S. Supreme Court, do not. See Adarand (1995).

The underlying premise of this book (made explicit with references to a wide array of literature in social psychology) is that the work world continues to be shaped by a great deal of racial and gender discrimination of the non-obvious forms, such as preferences to hire and work with persons similar to oneself and unconscious bias that more highly evaluates the work of white males. The authors acknowledge that many scholars explain the persistent residue of job inequality with the fact of differences in “human capital” across [*300] groups—i.e., there are differences across groups in years of education, quality of education, cognitive skills, and years on the job. Authors maintain that these cannot be the fully explanatory variables because if they were, job segregation would decrease steadily as a given group increased in, say, education years relative to white males. Instead, the pace of desegregation has varied with its socio-political popularity (pp.39-41). This book did drive me to the library where I discovered a book that makes a contrary argument, The Declining Importance of Race and Gender in the Labor Market by June and Dave O’Neill (2012), two labor economists. Using a cognitive skills test that the military administered, they examine wage gaps between white and black workers and between men and women using a study of five thousand people (NLSY79) over thirty years, controlling for number of years in the work force, and level of cognitive skills, and years of formal education. Doing this accounts for the entire wage gap between whites and blacks (O’Neill and O’Neill 2012, pp.208-210). If one were to assign the Stainback and Tomaskovic-Devey to a course, one should accompany it with an assignment of O’Neill and O’Neill. Doing so would certainly make for a lively discussion. Incidentally, the O’Neill’s have an explanation for the finding by Stainback and Tomaskovic-Devey that black women’s job status declined relative to white women’s in recent decades. The O’Neill’s suggest that this is because in the 60s and 70s many white women were relatively new entrants to the job market; as competing employees, black women were attractive as the more experienced. The O’Neills show that although black women’s wages deteriorate compared to white women’s in the 80s and 90s, because now white women have caught up in job experience, black women’s wages relative to those of white MEN (workers who like black women have been in the job market throughout adulthood) continued to improve (slowly) in the 80s and 90s (O’Neill and O’Neill 2012, pp.148-154).

Even though in this review I have faulted Stainback and Tomaskovic-Devey’s utter neglect of the most recent affirmative action decisions by the Supreme Court (such as Adarand, 1995), I have learned something important from their book: Adarand appears to have had virtually no impact on the use of set-asides, preferences for minority-owned firms, and the requirement of annual reports showing affirmative action efforts by federal contractors (Stainback and Tomaskovic-Devey, pp.265-266, 314-316; Civil Rights Commission, “Contracting After Adarand” 2005).This was a shock to me, although I would have known better had I read the article by Neil Devins predicting it in 1996. He pointed to the causal agent as the Supreme Court itself, noting the ambiguous language of the Adarand decision. On the one hand the Court pleased conservatives by declaring race-based affirmative action to be constitutionally suspect, triggering the strict scrutiny test. On the other hand the Court bowed to liberals by warning that strict scrutiny review should not be understood as “strict in theory but fatal in fact” (Devins 1996, p.700, citing Adarand 1995, p.237). The Adarand passage reads in full as follows:

“Finally, we wish to dispel the notion that strict scrutiny is ‘strict in theory, but fatal in fact.’ The unhappy persistence of both the practice and the lingering [*301] effects of racial discrimination against minority groups in this country is an unfortunate reality, and government is not disqualified from acting in response to it. When race based action is necessary to further a compelling interest, such action is within constitutional constraints if it satisfies the ‘narrow tailoring’ test this Court has set out in previous cases.”

Statements like this from the Court have nurtured a reality in which race-based affirmative action (including minority set-asides) in law is now officially “constitutionally suspect,” but still mandated by executive orders and enforced, to a degree, by the EEOC and OFCCP, and, so far, tolerated in the courts (Civil Rights Commission 2005, Devins 1996). I prefer this situation to an outright abolition of affirmative action, but it is a situation that makes understandable the book’s contempt for the influence of courts.


Baer, Judith and Leslie F. Goldstein. 2006. Constitutional and Legal Rights of Women, 3rd ed. NY: Oxford University Press.

Devins, Neil. 1996. “Adarand Constructors, Inc. v. Pena and the Continuing Irrelevance of Supreme Court Affirmative Action Decisions,” William and Mary Law Review 37: 673-721.

Dobbin, Frank. 2009. Inventing Equal Opportunity. Princeton: Princeton University Press.

O’Neill, June and David O’Neill. 2012. Declining Importance of Race and Gender in the Labor Market. Washington, D.C.: AEI Press.

U.S. Commission on Civil Rights. 2005. Federal Procurement After Adarand. Washington D.C. (accessed June 11, 2013).


Adarand Contractors v. Pena 515 U.S. 200 (1995).

FCC v. Metro Broadcasting 497 U.S. 547 (1990).

Fullilove v. Klutznick 448 U.S. 448 (1980).

Pittsburgh Press v. Human Relations Commission 413 U.S. 376 (1973).

Regents v. Bakke 438 U.S. 265 (1978).

United Steelworkers v. Weber 443 U.S. 193 (1979).

Copyright 2013 by the author, Leslie F. Goldstein.