The Affirmative Action Debate
For more than a quarter century, controversy about affirmative action has tormented American politics. It still arouses intense passions, spawns litigation, and frustrates public dialogue. Some commentators explain the fierceness of the debate by noting the conflicting interests of groups competing for a limited number of coveted jobs and educational opportunities. Others point to the symbolic importance that affirmative action has assumed for each side in the debate.
"Proponents regard the continuation of affirmative action as a litmus test of our nation's commitment to racial justice," Glenn Loury has written. "Opponents see it as an unacceptable violation of the ideal of equality of opportunity, and the principle that government should treat its citizens in a color-blind fashion." Both these ways of framing the issue, Loury suggests, are "mired in confusion." And both have at times preempted a broader discussion of strategies to address the problem of racial inequality.
A variety of rationales has been put forward to justify affirmative action programs. The single most important, writes Robert K. Fullinwider in an essay for this special Report, grew out of a recognition that abolishing facially discriminatory policies would leave in place a complex of long-established attitudes and informal practices that were likely to thwart rapid progress toward equal opportunity. In the years following the passage of the Civil Rights Act of 1964, this recognition gave rise to a new legal concept of discrimination, and provided a limited justification for racial preferences in hiring and admissions.
William A. Galston, in his "status report" on affirmative action, looks at how the resulting policies have affected a range of institutions, from the military to public universities. He also analyzes the judicial and political constraints that have recently been placed on preferential programs, and suggests alternative strategies for promoting equal opportunity as certain forms of affirmative action are dismantled.
Opponents of affirmative action have long argued that color-conscious selection procedures can only undermine meritocratic standards. In their essay for this issue, Judith Lichtenberg and David Luban consider the role that merit plays in modern economic life, and the role that it ought to play. Without accepting the view that judgments of merit are merely subjective, they identify imperfections in the usual processes by which institutions define and detect it. They explore the fate of the merit principle in a "winner-take-all" society. And they argue that a just society, in its allocation of posts and rewards, would make considered trade-offs between merit and other important values.
In recent years, many employers and university officials have placed the idea of diversity, rather than the moral imperative of antidiscrimination, at the center of their arguments for affirmative action. For some, achieving greater diversity is a way of enabling organizations such as police departments, colleges, and corporations to perform their missions better. For others, diversity is a good to be promoted in the service of ideals such as community solidarity and integration. In his second essay, Robert Fullinwider examines a standard argument for diversity in higher education, and assesses its adequacy as a justification for affirmation action in admissions and faculty recruitment. David Wasserman, in his essay on diversity and stereotyping, addresses some of the more stringent objections to the pursuit of institutional diversity. Finally, Owen M. Fiss argues that the only sustainable basis for affirmative action is a commitment to dismantling the "caste structure" of American society; the diversity rationale, he writes, seems "shallow" when compared with appeals to distributive justice.
Some of the contributors to this special Report engage their colleagues' arguments directly, while in other cases areas of agreement or dissension are implicit. Against Owen Fiss's call for a "transcendent" commitment to social justice, one might set William Galston's reminder that some means to an acknowledged good may be "effective but nonetheless unacceptable." Robert Fullinwider's analysis of discrimination as a "de-moralized concept" converges with Judith Lichtenberg and David Luban's account of institutional racism: "the view that discrimination does not always disappear when personal bias does." Like the two papers on diversity, the essay on merit addresses important questions of stereotyping and group identity. By exploring basic issues of law, social policy, and institutional practice, this Report, we hope, will bring some clarity to what has often been a misguided and bitter debate.
An Affirmative Action Status Report:
Evidence and Options
An ideal introduction to the affirmative action debate would supply nothing less than a crash course in the history, law, sociology, economics, and politics of the issue. My ambitions in this essay are somewhat more modest. Without masquerading as an expert, I have tried to summarize the basic facts as I understand them, put forward some illustrative case studies, and identify contentious areas of principle and policy. I also offer an inventory of options that have emerged for changing affirmative action programs. My purpose is to stimulate productive discussion by sketching a map of the terrain, not to announce and defend a particular position.
Definitions
"Affirmative action" means many different things. Among them: outreach to broaden the pool of eligible individuals to include more members of specific groups; targeted or compensatory training to upgrade the qualifications of individuals in these groups; goals and timetables to measure progress; preferences; set-asides; and actual quotas. Affirmative action programs have arisen as a result of executive orders, legislation, consent decrees stemming from government investigations, court-ordered remedies, and voluntary action by corporations and other non-public institutions.
The distinction between government-mandated and voluntary programs is important: for the most part, court decisions restricting public programs on constitutional grounds do not directly affect voluntary programs in the private sector. Some scholars argue that retrenchment in public programs could nonetheless lead to private-sector retreat. "Without government enforcement," writes sociologist Alan Wolfe, "some private companies may indeed drop their enthusiasm for diversity and retreat to birds of a feather' hiring policies." On the other hand, Wolfe notes that support for affirmative action is unexpectedly strong among leading American corporations, and he expects them to go on practicing it for the same reasons they do now: "out of pragmatism, trying to meet particular corporate objectives."
Measuring Results
Evidence on the effects of affirmative action efforts is frequently imperfect and ambiguous. For example, while programs addressing employment and government contracting have had modestly positive effects in the aggregate, their role is frequently difficult to disentangle from other antidiscrimination or opportunity-creating efforts. The Clinton administration, in its review of governmental affirmative action policies, found that active federal enforcement during the 1970s "caused government contractors to increase moderately their hiring of minority workers. According to one study, for example, the employment share of black males in contractor firms increased from 5.8 percent in 1974 to 6.7 percent in 1980. In non-contractor firms, the black male share increased more modestly, from 5.3 percent to 5.9 percent."
A significant number of the 1.3 million black government employees owe their jobs (or promotions to managerial rank) to affirmative action. Corporate affirmative action programs (some voluntary, others reflecting consent decrees in response to government pressure) have opened up managerial ranks, though not yet the very top echelons, to minorities and women.
Any assessment of the expansion of employment opportunities over the past three decades must take into account, in Jerome Karabel's words, "how exclusionary many labor markets were in 1965 not only in high-status professions such as law, medicine, and academe but also in strategic working-class domains such as construction unions and police and fire departments." Karabel notes that while increased minority representation in the professions is widely recognized, "the record in certain blue-collar jobs is just as impressive. For example, between 1970 and 1990, the number of black electricians more than tripled (from 14,145 to 43,276) and the number of black police officers increased almost as rapidly (from 23,796 to 63,855)." A significant portion of these gains is attributable to affirmative action plans applicable to unions and local governments. Aggregate gains in employment and promotion for women and minorities have continued during the past decade, considered by some to have been a period of diminished attention to equity.
Effects on wages appear to have been modest. According to a benchmark report by James Smith and Finis Welch, "the racial wage gap narrowed as rapidly in the 20 years prior to 1960 (and before affirmative action) as during the 20 years afterward." The report argues that "slowly evolving historical forces," such as education and migration, "were the primary determinants of the long-term black economic improvement. At best, affirmative action has marginally altered black wage gains about this long-term trend."
Aggregate effects on higher education enrollments appear to have been larger. In 1955, only 4.9 percent of college students ages 18-24 were black. This figure rose to 6.5 percent during the next five years, but by 1965 had slumped back to 4.9 percent. Only in the wake of affirmative action measures in the late 1960s and early 1970s did the percentage of black college students begin to climb steadily: to 7.8 percent in 1970, 9.1 percent in 1980, and 11.3 percent in 1990. Educational gains for Hispanics have been less impressive. According to a recent report based on Census Bureau statistics, Hispanics holding bachelor's or advanced degrees rose from 5 percent of the Hispanic population in 1970 to 9 percent in 1994. (These figures exclude recent immigrants.) By contrast, blacks holding bachelor's or advanced degrees rose from 4.5 percent of the black population to 12.9 percent during the same period.
Overall, the past three decades have witnessed huge growth in the black middle class, educational attainment, incomes of black married couples relative to white married couples, and suburbanization, along with a substantial decline in overall black poverty rates. These gains are attributable to a complex of causes: national economic growth, a decline in discriminatory attitudes and practices by whites, and programs -- including affirmative action -- targeted to African-Americans. While we can sometimes trace the quantitative impact of single causes within particular economic or educational organizations, it is far more difficult to ascertain the relative contribution of each factor to the aggregate gains of recent decades.
In conjunction with other broad economic and social trends, these positive changes have led to what Seymour Martin Lipset has characterized as "growing differentiation" within the black community. On one side, there is now a burgeoning black middle class; on the other, there are the ghetto poor, economically and socially isolated from the rest of society. These poor people belong primarily to two groups: single mothers and their minor children, and young men who have dropped out of both high school and the labor force.
As numerous scholars have argued, affirmative action programs have not had significant positive consequences for the bulk of the ghetto poor; nor are they likely to. The reason is straightforward: these programs work most effectively when they remove barriers to opportunity for those who possess the credentials to succeed or who are strongly motivated to acquire them within established norms and institutions.
Two Case Studies
To understand how affirmative action works in practice (and some of the reactions, positive and negative, that it evokes), it may prove useful to look briefly at two important institutions that have undergone fundamental changes during the past generation.
The U.S. Military. The armed forces were almost completely segregated until President Truman's famous 1948 order. While the number of uniformed minorities and women rose during the next quarter century, racial tensions intensified sharply during the Vietnam era.
In response, the services implemented affirmative action plans in the 1970s. The President's review confirmed what many observers had concluded: these plans have succeeded in expanding representation of minorities and women, especially as officers, while improving race relations, promoting integration, and enhancing overall combat readiness. The noted military sociologist Charles Moskos has characterized the Army as "the only institution in America in which whites are routinely bossed around by blacks." How has the military achieved these results? Many analysts have emphasized its special institutional characteristics as a highly closed, controlled, hierarchical, disciplined system with the ability to establish, and attain compliance with, organization goals. In addition, some specific features of military affirmative action efforts have contributed to their success.
- The pool of applicants accepted for military service is highly selective. A very high-ranking military officer told me that "it's harder to get into the All-Volunteer Forces than into most colleges." Further contributing to quality is the fact that the military has become the career of choice for many African-Americans.
- Military affirmative action plans do employ goals: promotion of minorities and women within the eligible pool is to occur in the same percentages as overall promotions from that pool. But in many cases the goals are not linked to timetables. In addition, the goals serve as presumptions, not mandates; promotion boards that fail to meet them are deemed to have done their job correctly if they can demonstrate due diligence.
- All candidates for promotion are placed in a common pool and are subject to the same standards. Race can serve as a factor, but only when other differences are very small. As one officer put it, "Only fully qualified people are promoted, but not necessarily the best qualified. But don't forget, we are talking micro-millimeter differences in these cases."
- The armed forces engage in constant training, including compensatory training, before as well as after admission to the All-Volunteer Forces, to enable the highest possible percentage of individuals to meet high standards. While outreach efforts are not racially exclusive, some are "race-conscious." New recruits who are diagnosed as having particular weak spots are given numerous opportunities to remedy them.
The University of California System. At the time of the famous Free Speech Movement in the early 1960s, the University of California System was virtually all white; this was particularly true of elite institutions such as Berkeley and UCLA. As late as 1984, whites constituted 70 percent of all students in the System. Today, that figure is 49 percent.
Since 1964, admission to the University of California has been governed by a legislatively crafted "Master Plan." To be eligible for admission through the regular process, students must be in the top one-eighth of their high school graduating classes, take challenging courses, and do reasonably well on their SATs. From this pool, about half of each entering class is selected strictly on the basis of an "academic eligibility score" -- a weighted sum of grades and SATs. Roughly 45 percent are selected from the eligible pool on the basis of additional factors such as race and ethnicity. Five percent enter through a "special admit" category outside the normal eligible pool. Eighteen percent of African-American first-year students are in this last category; the other 82 percent are in one of the first two.
There has been a vigorous empirical debate about the academic effects of these policies. On one side, Karen Paget emphasizes that most of the students admitted to Berkeley under the Master Plan continue to be drawn from the top 12.5 percent of high school graduates. This fact, she writes, contradicts "the common but erroneous assumption that minority students have been admitted from below the line' or outside the eligibility pool." On the other side, there is evidence of substantial disparities in academic preparedness among students from different racial and ethnic groups. For the entering class at Berkeley in 1994, the mean grade-point averages were 3.43 for blacks, 3.65 for Hispanics, 3.86 for whites, and 3.95 for Asians. Among these same freshmen, the mean SAT scores were 994 for blacks, 1032 for Hispanics, 1256 for whites, and 1293 for Asians. (Let it be noted that the gap in average SAT scores for blacks and whites at Berkeley -- 262 points -- is wider than the nationwide gap between the two groups.)
The picture becomes more complex when we look at admission and graduation rates for different categories of students. Let me turn again to statistics provided by Jerome Karabel, one of the authors and staunchest defenders of the Berkeley affirmative action program.
Under the Berkeley plan, the odds of a black or Hispanic being admitted from the eligible pool are roughly three times as high as for whites or Asians, but their chances of graduating are about one-third lower. In 1988, for example, 75 percent of black applicants, and 85 percent of Hispanic applicants, were admitted, as opposed to 28 percent of white applicants and 25 percent of Asians. After five years, graduation rates for this class were 37.5 percent for blacks, 43.5 percent for Hispanics, 71.5 percent for whites, and 67.3 percent for Asians. After six years, the graduation rates rose to 51 percent for blacks, 53 percent for Hispanics, 77 percent for whites, and 75 percent for Asians.
Critics of affirmative action often assume a direct link between the statistical disparities among average GPAs and SATs and the lower graduation rates of black and Hispanic students. Abigail Thernstrom, speaking last fall at a conference at Yale University, cited "the racial gap in cognitive skills" as the cause of low minority retention rates, and argued that "we did these students no favor" by granting them preferences in the admissions process. In contrast, Paget cites studies showing that the reasons for lower graduation rates among minority students are "many and varied," including "personal or family financial problems." Karabel has documented black and Hispanic students' "comparative lack of financial resources" through detailed income comparisons; more recent studies, indicating striking differences in financial assets between whites and blacks, could be adduced as well. In Karabel's view, the fact that graduation rates for blacks and Hispanics "look significantly better after six rather than five years" lends support to economic factors as a "part" of the explanation.
However the statistics are interpreted, they cannot answer the policy question of whether the lower graduation rates for black and Hispanic students represent a reasonable trade-off against the gains in educational opportunity for large numbers of minority group members. Nor do the admissions figures answer the question of whether the costs imposed on non-minorities by the Berkeley plan are diffuse enough (see the Wygant case below) to pass moral muster. Clearly, however, affirmative action in the University of California System has failed to pass political muster. On July 21, 1995, by a vote of 14 to 10, the University of California's Board of Regents rejected the use of race, sex, religion, color, or national origin as factors influencing admission. In place of these factors, the System may adopt supplemental criteria based on evidence of social disadvantage.
Public Attitudes
Popular support for the full range of current affirmative action programs has slipped significantly in recent years and is now very limited. For example, a July 1995 CNN/USA Today poll gave respondents three options on affirmative action: "basically fine the way it is"; "good in principle but needs to be reformed"; and "fundamentally flawed and needs to be eliminated." Sixty-one percent chose the reform option; 22 percent opted for elimination; only 8 percent favored the status quo.
Not surprisingly, there is a significant racial split, with only 11 percent of African-Americans in favor of outright abolition (versus 24 percent of whites). Nonetheless, only 15 percent of blacks support the status quo, and 23 percent believe that on balance affirmative action has been bad for the country. Many other recent polls point in the same direction.
In November 1996 California voters approved Proposition 209, a measure outlawing racial preferences in the operation of public employment, education, and contracting, by 55 to 45 percent. Sixty percent of white voters -- including 57 percent of white women -- supported the initiative, while 74 percent of blacks and 70 percent of Latinos opposed it. A Field Poll found that even among those who voted against Proposition 209, "50 percent said they believe affirmative action should be relaxed or eliminated." At the same time, the poll found that 8 in 10 voters "recognize discrimination is still common in society today."
At least a dozen other states are considering measures similar to Proposition 209, either through legislation or ballot initiatives.
Points of Contention
Means and Ends. Among the means typically employed in affirmative action programs, enhanced outreach and targeted training enjoy widespread support, while quotas and set-asides are widely regarded as unacceptable. The ongoing debate is about preferences, goals, and timetables. One portion of this debate is empirical: do certain policies in fact work effectively to fight discrimination and promote equal opportunity? Another portion is moral: regardless of their efficacy, are the policies in question acceptable as means to the end in view?
Unless we are prepared to endorse the proposition that means are fully justified by ends, we must be prepared for the possibility that some means are effective but nonetheless unacceptable. Consider Justice Powell's opinion in the Wygant case, distinguishing between race-conscious hiring goals and race-based layoffs. "In cases involving valid hiring goals," Powell observed,
the burden to be borne by innocent individuals is diffused to a considerable extent among society generally. Though hiring goals may burden some innocent individuals, they simply do not impose the same kind of injury that layoffs impose. Denial of a future employment opportunity is not as intrusive as loss of an existing job.
Layoffs, Powell continued, "impose the entire burden of achieving racial equality on particular individuals, often resulting in serious disruption of their lives. That burden is too intrusive." Thus, he concluded, the "selection of layoffs as the means to accomplish even a valid purpose cannot satisfy the demands of the Equal Protection Clause."
I would draw an analogy from the world of criminal justice: it might well be the case that we could enhance arrest and conviction rates by significantly expanding the scope of warrantless searches, but our willingness to move in this direction is constrained by our commitment to individual liberty.
From transitional to permanent. The understanding at the outset was that affirmative action programs would be temporary and transitional, pointing toward a future in which only the relevant talents and accomplishments of individuals would be taken into account. (When the Bakke case was decided in 1978, most of the justices who favored affirmative action in higher education believed that it would be necessary and justified for a single generation at most.)
One element of this understanding was that the rapid expansion of the black middle class would reduce or eliminate the need for affirmative action for their children. For reasons that are poorly understood, this has not come to pass. As Glenn Loury notes, "In 1990 black high school seniors from families with annual incomes of $70,000 or more scored an average of 854 on the SAT, compared with average scores of 855 and 879 respectively for Asian-American and white seniors whose families had incomes between $10,000 and $20,000 per year."
Democratic authorization. The majority of federal affirmative action programs were put in place through executive order, regulation, or the courts rather than legislative action. Many people feel that these programs were somehow imposed without their advice and review, let alone consent. In a democracy, no course of action can be indefinitely sustained unless it rests on a solid foundation of public understanding and acceptance.
The Legal Status of Affirmative Action
Recent Supreme Court decisions (especially Adarand v. Pena) will have the effect of significantly reducing the scope of acceptable federal government affirmative action programs. In that case, the Court applied to federal actions the standard already binding on states and localities: programs must serve a "compelling" interest and must be "narrowly tailored." An analysis by discrimination law expert Paul Gewirtz reaches the following conclusions:
- Objectives such as enhancing "diversity" and "inclusion" or addressing general "societal discrimination" do not qualify as compelling.
- A specific showing of particular discrimination, going beyond simple statistical disparities among racial and ethnic groups, must be made.
- Even when a compelling interest is found, race-based methods may be used only after race-neutral methods are considered and found wanting, only to the extent needed to remedy the identified discrimination, only when the plaintiffs seeking a racial preference have themselves suffered from past discrimination, and only if undue burdens on non- beneficiaries (such as layoffs) are avoided.
There are, in addition, legal developments in the application of constitutional law to the states. In 1994, the Fourth U.S. Circuit Court of Appeals struck down a University of Maryland scholarship program restricted to African-American students (Podberesky v. Kirwan). Last year, the Fifth Circuit rejected an admissions procedure at the University of Texas Law School that divided applicants into two groups -- first, blacks and Mexican-Americans, and second, all others -- and then applied different admissions cut-offs to the two groups. The Court held that the law school's interest in diversity did not constitute a "compelling state interest" and that the school could not take race into account in any form in its admissions process (Hopwood. v. Texas). The Supreme Court let both decisions stand without further review. While as a matter of law other states are not absolutely debarred from continuing race-restricted scholarships or preferential admissions policies, the scholarly consensus is that these programs are unlikely to survive all but certain legal challenge.
Ineradicable Tensions
Recent developments in constitutional law, public opinion, and the political arena have made significant changes in existing affirmative action programs all but inevitable. This does not necessarily mean abandoning the goals of antidiscrimination and equal opportunity. A case could be made that we have asked affirmative action programs to bear too much practical and symbolic weight -- that we have neglected other ways of fighting discrimination, and relied excessively on affirmative action at the expense of a broader equal opportunity agenda. (Indeed, some critics have charged that the latter represents a deliberate strategy of "equal opportunity on the cheap.")
Still, we must acknowledge that there are permanent impediments to realizing the dream of equal opportunity -- limitations not just of resources and will, but stemming also from values deeply rooted in the ethos of liberal democracy. Achieving fully equal opportunity would require equalizing all factors that affect the development of talents. As philosophers since Plato have observed, this would imply (among other things) highly intrusive and coercive government action to correct for the differential impact of such variables as family background and culture. The history of the affirmative action debate confirms that we can neither avoid nor fully erase the tension between equal opportunity and personal liberty.
- William A. Galston
Sources: Paul Sniderman and Thomas Piazza, The Scar
of Race (Harvard University Press, 1993); Margery Austin Turner, et al., Opportunities
Denied, Opportunities Diminished: Racial Discrimination in Hiring (Urban Institute,
1991); Alan Wolfe, "Guess Who Likes Affirmative Action?", The New Yorker (Nov.
25, 1996); Christopher Edley, Jr., and George Stephanopoulos, "Affirmative Action
Review: Report to the President" (1995); Jerome Karabel, "Berkeley and
Beyond," American Prospect, no. 12 (Winter 1993); James Smith and Finis
Welch, "Closing the Gap: Forty Years of Economic Progress for Blacks," RAND
Corporation (February 1986); Steven A. Holmes, "Census Report Finds Hispanic
Americans Lagging Further in College Degrees," New York Times (July 27,
1995); Gerald D. Jaynes and Robin M. Williams, Jr., eds., A Common Destiny: Blacks in
American Society (National Academy Press, 1989); William Julius Wilson,
"Race-Neutral Programs and the Democratic Coalition," American Prospect
(Spring 1990); Seymour Martin Lipset, American Exceptionalism: A Double-Edged Sword
(Norton, 1996); Charles Moskos, "The Army Experience," New Democrat
(May/June 1995); Peter Applebome, "Gains in Diversity Face Attack in
California," New York Times (June 4, 1995); Karen Paget, "Controversy:
Affirmative Action at Berkeley," American Prospect, no. 12 (Winter, 1993);
Carol Ness and Annie Nakao, "Voters Back Ban on Affirmative Action by 55 45
Percent," San Francisco Examiner (Nov. 6, 1996); Edward W. Lempinen,
"Affirmative Action Foes Span Spectrum," San Francisco Chronicle (Nov. 22,
1996); Carol Innerst, "Affirmative Action Under Siege in States," Washington
Times (July 26, 1995); Glenn C. Loury, "Getting Involved: An Appeal for Greater
Community Participation in the Public Schools," Washington Post Education Review
(August 6, 1995); Paul Gewirtz, "Affirmative Action: Don't Forget the Courts," Wall
Street Journal (August 2, 1995). 