Pursuing Diversity: From Education to Employment

Amy L. Wax1Amy L. Wax is the Robert Mundheim Professor of Law at the University of Pennsylvania Law School.

* * *

A part of the series, Affirmative Action at a Crossroads.

A core ideal of Anglo-American law is that legal wrongs should be remedied by restoring the injured victim to the “rightful position.”  That position is defined as the one the victim would have occupied had the legal injury never been inflicted. This aspiration has exerted a powerful influence on American legal practice across the board.  Although, curiously, the phrase “rightful position” barely figures in cases or executive orders applying federal civil-rights laws, the restorative imperative embodied in that concept has exerted an important influence on their administrative and judicial application.

Following the enactment of the Civil Rights Act of 1964, the question loomed large of how to define the “rightful position” for the targets of unlawful discrimination, exclusion, and ill treatment, and especially for Blacks. What should be the touchstone for full correction, and thus racial justice?  The answer that soon took hold was that Blacks would occupy the same social, educational, and occupational positions as Whites.  Perhaps the closest to a formal expression of this concept was President Lyndon B. Johnson’s famous speech at Howard University articulating as the goal of the federal civil-rights laws not just “equality as a right and a theory,” but “equality as a fact and equality as a result.” 

 The “equality as a fact” benchmark necessarily raised the question of how to achieve that result for victims of racial bias and discrimination.  A consensus grew that the desired outcome demanded more than the legal command to stop discriminating.  Reversing and undoing the lingering vestiges of past wrongs would be required.  This in turn would call for the adoption of race-conscious measures across multiple domains.  As Justice Harry Blackmun stated in Bakke v. Regents of the University of California in 1978, “In order to get beyond racism, we must first take account of race. There is no other way.”  The practice of affirmative action, or reverse discrimination, was born of this idea.

Although specific, limited, and targeted race-conscious remedies were not hard to reconcile with established legal-equitable principles, formidable obstacles existed—political, doctrinal, and practical—to the aggressive pursuit of “equality of result” through the systematic use of race preferences.  Much of the public remained wedded to individualist, impartial ideals of meritocratic selection that tolerated expanded opportunities but resisted “hard” measures like racial quotas or racial proportionality.  Retrofitting remedial doctrines designed to redress well-defined, individual injuries to more nebulous group harms proved controversial and perplexing.  Courts grappled with whether specific proof of discrimination, as opposed to vague allegations of societal racism, should be required.  They also considered whether race-conscious orders could properly be imposed on entities never demonstrated to violate the law or in favor of persons never shown to have suffered actual discrimination.  On the practical level, the “rightful position” project was stymied by entrenched customs, outright resistance, and Blacks’ lack of readiness to step into a full range of social and economic roles due to poor education and skills.  Attempts to grapple with these realities produced policies of mixed efficacy and a tangle of conflicting and confusing court decisions.

Confounding this project in the important employment realm was a Civil Rights Act provision, Section 703(j) of Title VII, that disavowed any requirement that an employer “grant preferential treatment to any individual or to any group” due to racial imbalances in the workplace.  The executive branch, in enforcing the law, effectively ignored and repeatedly flouted the clause’s limitations by demanding that businesses and other employing entities address lopsided racial representation regardless of whether discrimination was shown to be the cause.  In The Affirmative Action Puzzle, a recent history of affirmative action, Melvin Urofsky quotes an Equal Employment Opportunity Commission (EEOC) staff member’s statement from the 1970s that his agency treated the antipreference bar in the statute as “a big zero a nothing, a nullity.  [It doesn’t] mean anything to us.”  Accordingly, the EEOC and an expanding network of federal offices and agencies proceeded to impose on businesses, corporations, and educational institutions ever more intrusive and onerous requirements, including hiring quotas, targets, and timetables as well as massive paperwork and documentation.   

Judicial practice was more equivocal.  Through a welter of fractured rulings in employment, business set-asides, and education, the courts and the Supreme Court established a set of rules permitting some types of race-conscious remedial orders to rectify proven statutory violations, and at times allowing voluntary affirmative-action programs in the absence of adjudicated legal infractions.  Judicially crafted limitations on such programs included the requirements of showing some evidence of past or present discrimination beyond statistical imbalances, avoiding inflexible quotas and numerical targets, tailoring race-conscious measures as narrowly as possible, and applying them only temporarily.  These strictures were not always consistently applied, which created uncertainty for economic actors subject to their mandates.

The early affirmative-action cases were primarily about jobs and businesses. By the 1990s, the action had shifted decisively to education and a new justification for race-conscious measures entered the picture: diversity.   Initially developed within academia itself, that idea was put forward in Justice Lewis Powell’s famous concurring opinion in the 1978 case of Bakke, which struck down a quota-based affirmative-action plan at the University of California at Davis Medical School.  Justice Powell suggested that, even if rigid quotas were legally verboten, race-based educational selection could be justified as a device to create a diverse student body.  Being exposed to students from various backgrounds carried pedagogical value and enhanced the educational experience.  It followed that creating student diversity, which delivered those benefits, was a goal that was “compelling” enough to overcome the law’s colorblind imperatives.  Powell cited as exemplary the flexible, individualized admission protocol at Harvard College, which used race as one factor among many to craft a class representing a range of backgrounds, talents, and experiences.

Diversity quickly became the central pillar of the Supreme Court’s educational affirmative-action jurisprudence, with a majority of the Court officially recognizing educational diversity as justifying some degree of race-conscious student selection.  At the same time, the Court imported the prior touchstones of a bar on rigid quotas and the expectation of narrow tailoring of racial methods.  Although appearing to place genuine limits on the use of affirmative action in the educational setting, several aspects of the Supreme Court’s key rulings on the issue, including Grutter v. Bollinger (2003) and the Fisher line of cases (Fisher v. University of Texas (Fisher I (2013) and Fisher II (2016)) gave universities wide discretion to structure their admissions criteria pretty much as they wished.  These included the open-ended and ill-defined nature of diversity, its elevation to a constitutional “compelling interest,” Justice Sandra Day O’Connor’s language in Grutter deferring to educational “expert” judgment, and the Court’s repeated failure to demand specific evidence of diversity’s actual benefits and efficacy.  This situation earned scathing contempt from Justice Antonin Scalia, who repeatedly questioned university officials and other supposed educational “experts” averments of the value of diversity and regarded the elevation of that “compelling interest” as a pretext for the Court’s imposition of its political preferences.2See Amy L. Wax, Trust Me, I’m an Expert: Scientific and Legal Expertise in Scalia’s Jurisprudence, in Scalia’s Constitution (Paul E. Peterson & Michael W. McConnell, eds. 2018).

Justice O’Connor famously stated in her plurality opinion in Grutter that she expected that affirmative action would no longer be needed after twenty-five years.  Albeit well-meaning, Justice O’Connor’s statement represents a misunderstanding of the implications and import of the Court’s adoption of diversity rather than remediation as the centerpiece of educational affirmative-action jurisprudence.  O’Connor’s prediction is grounded in the logic of remediation and thus unavoidably informed by the “rightful position” idea.  The logic of rightful position remediation dictates that proper remedies—and especially those that deviate from and go beyond the law’s mandates—can be justified only if they are designed eventually to achieve the promised result, which is to undo the injuries inflicted by a wrong.  The objective of race-conscious measures to correct the wrongs of racial discrimination is to enable Blacks to catch up with other groups to the point where they can compete by dint of their own efforts and take their rightful place in society under their own steam.  Once that point is reached and the damage fully undone, affirmative action will no longer be needed.  But diversity as the main rationale for race-consciousness provides no reason to believe that affirmative action will reach that end point.  Diversity neither promises nor requires group uplift or equalization.  It rests on the educational value of Blacks’ institutional presence, not on the promise of undoing past harms or making Blacks as capable, academically or otherwise, as other groups.  Diversity, unlike remediation, thus entails no prediction that affirmative action will someday be unnecessary and will be phased out.  Although the point seems not to have been fully appreciated by Justice O’Connor or by those who repeat Justice O’Connor’s famous mantra, identifying diversity as the main rationale for educational affirmative action is fully compatible with affirmative action in perpetuity.

Continuing social and educational trends reaffirm “affirmative action forever” as the most realistic future scenario.  Despite the initial exuberance of the Civil Rights era, it has become increasingly clear that the holy grail of steady racial progress towards “equality in fact” has remained elusive across multiple domains.  More broadly, the high hopes surrounding race preferences in education and other arenas to effect enough social and economic progress in the Black community to close existing gaps have not been realized.  Blacks as a group continue to suffer disproportionately from poverty, high crime rates, family breakdown, and low male workforce participation.  In the education sphere specifically, Blacks on average still lag behind Whites and Asian Americans in standard measures of learning, ability, and academic proficiency, such as national K–12 achievement tests, SATs, LSATs, MCATS, and GREs.  Observed differences are pervasive:  even controlling for socioeconomic status and factors such as family income, school quality, and neighborhood residence, Blacks underperform other groups academically, often by a significant margin, in virtually every public-school district in the United States.  Average differences translate into a stark undercount of Blacks in the higher academic ranks nationwide.  This means that, without affirmative action, Blacks will be severely underrepresented at competitive and selective institutions that rely heavily on standard measures of academic ability and achievement. 

Whatever the etiology of existing disparities—and factors such as racism, poverty, culture, and innate group differences have been cited—they have proved highly resistant to elimination or even enduring narrowing through instruments of law and policy.  Little progress has been made in the past few decades towards closing observed racial gaps in multiple dimensions despite the widespread adoption of affirmative action in educational programs as well as in other sectors.  These are the realities on the ground that continue to make affirmative action necessary for generating a so-called “critical mass” of Blacks in selective universities. In sum, current developments belie Justice O’Connor’s prediction that affirmative action will eventually be phased out some time soon. 

I.  Challenges with Prioritizing Diversity

In the meanwhile, the shift from remediation to diversity as the central doctrinal rationale for educational affirmative action has had several practical and legal consequences in light of important social changes that have occurred.  In recent decades, a surge in immigration has produced unprecedented levels of demographic diversity.  The growing numbers of potential students from a variety of ethnicities, backgrounds, and identity groups are in a position to contribute to the diversity of educational institutions.  Although the situation is complex, the great majority of these groups have more or less surpassed American-born Blacks in achievement, income, employment participation, and occupational status.  By enhancing the salience of some groups’ relative success in light of difficult circumstances and hardships, the diversity goal has threatened to draw attention away from the civil-rights focus on Blacks’ plight and to dilute Blacks’ primacy as uniquely deserving.  And the increasingly multicultural landscape has also thrown into renewed relief the ongoing and seemingly recalcitrant educational deficits in the Black community and the failure of decades of public and private initiatives to close existing racial gaps. 

The pressure to maintain what is considered a desirable student mix, including a significant presence of relatively underperforming Blacks and Hispanics in the face of growing competition, especially from Asian Americans, has induced universities to become both more aggressive and less forthright about their emphasis on race relative to other conventional academic factors in their admissions practices.  This has elicited backlash in the form of proposed referenda and legislative initiatives at the state level to curtail the use of race by public universities in admissions and other programs.  Some of these have succeeded.  More recently, efforts to reduce race preferences have taken the form of legal action, including a high-profile lawsuit by an Asian-American advocacy group claiming that Harvard University discriminates against Asian Americans in its admissions process in violation of Title VI of the Civil Rights Act.  The Justice Department Civil Rights Division also just commenced an investigation of Yale University.  Although the Division’s August 2020 letter questioning Yale’s admissions practices stopped short of declaring that race could never be taken into account, it argued that Yale violated Title VI of the Civil Rights Act by routinely making race a “determinative factor” in its admissions decisions and by applying a hard, quota-like limit on the number of Asian-American students at the school.

Both legal challenges face uphill battles.  The Supreme Court, in its main affirmative-action precedents, has given universities permission to make use of race as a factor in pursuit of the compelling goal of student demographic diversity.  Put another way, the Court has effectively issued a license to discriminate based on race in favor of some groups—most notably Blacks and Hispanics—that would otherwise be underrepresented at their institutions based on conventional academic criteria.  But it is an undeniable fact that the places available at the most competitive and desirable institutions are limited relative to the number of applicants seeking them.  That means that freeing up places for Blacks and Hispanics (to achieve a “critical mass” of those student categories) necessarily entails admitting fewer students from higher-achieving groups—including Whites and especially Asian Americans.  This can only be done by holding such groups to higher academic standards and reducing their likelihood of gaining admission relative to other applicants.  In Glenn Loury’s words, this is “simple logic.”  In sum, the affirmative-action practices the Supreme Court permits will inexorably and necessarily result in discrimination against students from some groups at selective schools. 

This unavoidable equation means that Harvard University’s official litigation position that it does not practice bias against Asian-American applicants is not only unnecessary under existing jurisprudence, but also illogical and transparently implausible.  These points were not completely lost on Allison Borroughs, the district judge on the Harvard case.  In ruling in the university’s favor, she effectively acknowledged that the law allows Harvard to impose more exacting academic requirements on Asian-American applicants, and to reject a higher percentage of Asian Americans than Blacks, to achieve the all-important goal of undergraduate racial diversity. In other words, Harvard can discriminate against Asian Americans to the degree necessary to achieve what the school regards as a sufficiently diverse student body. The clear implication is that, absent more restrictive rules (such as might exist under state law) than the Supreme Court has imposed, similar complaints against other competitive universities are also unlikely to succeed. Unless universities dramatically change their admissions practices—for example by downgrading or abandoning conventional merit-based metrics such as the SAT, which is already happening in some places—the juggernaut of affirmative action and its double standards will continue apace.

Nonetheless, there are scenarios under which Harvard could still lose its case on appeal or in the Supreme Court (should it arrive there).  Although taking account of race is precisely what Supreme Court precedent allows, the Court has articulated limits on the methods that can be used and the weight that can be assigned to race to achieve demographic balance. These limits furnish a potential basis for finding that Harvard has gone too far (as the Justice Department alleges Yale has done) under existing precedent. But even if the Supreme Court demands that Harvard curtail or modify its practices, the most likely outcome is that the university will still be allowed to limit the number of Asian-American (and/or White) undergraduates as a necessary step to freeing up spaces for Black and Hispanic students.  Race preferences will continue to be part of the admissions equation. 

Alternatively, albeit improbably, the Supreme Court could elect to use the Harvard case to tighten up significantly on its existing doctrine. One promising avenue would be to follow the suggestion, in concurring and dissenting opinions in past affirmative-action cases, that the Court abandon its deference to education “experts” by scrutinizing claims made on behalf of diversity more carefully and by adopting more exacting standards for evaluating its supposed educational benefits.  The Court could demand that universities precisely identify, measure, and demonstrate superior outcomes from diverse educational settings.  A useful natural experiment is presented by secondary schools such as the Bronx High School of Science and the Stuyvesant High School in New York City, and the Thomas Jefferson High School of Science and Technology in Fairfax County, Virginia.  All of these schools are required by law to admit students solely on the basis of a competitive exam, with race-conscious selection verboten.  This process has recently produced a student population that is overwhelmingly Asian, with a miniscule number of Blacks and Hispanics. Is there any indication that the students at these schools learn less or otherwise suffer academically in palpable and demonstrable ways compared to those who attend schools with a greater demographic range of students? That is the type of question that the courts should be asking.  In general, the burden should be on the universities seeking to defend their affirmative-action practices to show with specificity whether and how a more varied demographic profile advances pedagogical effectiveness.

II. Affirmative Action in the Employment Context

The discussion so far has focused chiefly on affirmative action in the educational context, which is an area that has commanded outsized attention from the courts and legal commentators. What about affirmative action in the workplace?  Recent events may prompt renewed interest in that topic.  National soul-searching in the wake of the death of George Floyd in police hands has spawned a raft of pledges by companies, firms, corporations, foundations, and other “woke” organizations to increase the numbers of underrepresented minorities in their staff ranks, with some even promising to achieve percentage targets and goals reminiscent of job quotas.  The centerpiece of the “antiracist” initiatives undertaken by this growing list is a commitment to creating a more diverse workplace.  Although not expressly disavowing remediation, public pronouncements have repeatedly emphasized the prime importance of enhancing workplace “diversity, equity, and inclusion” to reflect the profile of groups in society as a whole, and especially to increase Blacks’ presence in the full array of jobs from top to bottom. 

The shift from remediation to diversity in the rhetoric of workplace affirmative action, which recapitulates what has occurred in education, can be understood as proceeding from similar real-world conditions, disappointments, and failures.  The “rightful position” measure of racial justice proceeds from the expectation that the temporary use of race preferences will eventually enable Blacks to compete effectively and to qualify, without a race-conscious boost, for the range of jobs and occupational positions in proportion to their numbers.  As in the education sphere, that employment goal has so far proved elusive.  Decades after the enactment the Civil Rights Act, and in spite of a plethora of policies, programs, and initiatives across an array of domains, race-based preferences are still necessary in order to maintain a significant Black presence in many workplace positions, and especially in the most demanding, remunerative, and skill-intensive.  Given the profile of qualifications of Blacks as compared to other groups in society, there is no indication that “equity,” as it has now come to be designated (which means equal outcomes), can be achieved without continuing race-conscious interventions and every reason to believe these will be needed indefinitely.

The emphasis on diversity can be seen as an adjustment to this reality.  As in the case of education, embracing diversity as the principal rationale for race preferences in employment does not entail the expectation that race-conscious personnel practices will be or can be phased out in the foreseeable future.  The workplace is facing “affirmative action forever.”

III. The Court Should Not Allow the Goal of Increasing “Diversity” to Justify Affirmative Action in the Workplace

Renewed efforts to increase racial diversity in employment to the point of achieving proportional representation and “racial equity” raise important legal and practical questions. As Melvin Urofsky notes, the Supreme Court has declined to forbid voluntary, private race-conscious affirmative action programs under Title VII, despite explicit language in the statute that would seem to place them off-limits in many cases, and despite the Court’s own repeated, albeit erratic, endorsement of the need for narrow tailoring and proof of specific infractions under longstanding remedial conventions and doctrines.  In United Steelworkers of America v. Weber (1979), as Urofsky explains, Justice William Brennan’s majority opinion “somehow managed to find that the explicit wording of Title VII prohibiting racial discrimination did not foreclose [a] private race-conscious affirmative action plan[ ]” so long as the plan was “transitional in nature . . .  designed to correct statistical imbalances . . . and allowed flexibility in hiring non-minorities.”  The Court did not justify its permission in that case on specific findings of past discrimination by the employer or union that implemented the plan. Rather, it relied on the generalized remedial goal of “eliminat[ing] present and future discrimination” as well as “wip[ing] out the burden of past discrimination.” Because that objective was within the “spirit of the authors of the Civil Rights Act” even if contrary to its precise terms, the Court permitted the race-conscious initiative at issue in that case to stand.

The Supreme Court has never expressly repudiated or overruled Weber, but it is unclear how the Court would apply it today, especially in light of its recent embrace of a strict textualist reading of Title VII in Bostock v. Clayton County (2020). Based on this development, the Court could choose to invoke the colorblind language of Title VII or the statute’s disavowal of a numerical balancing requirement to severely curtail Weber, perhaps by limiting race-based hiring to a corrective for specific instances of proven past or present discrimination. But even if the Supreme Court declines to take that course, it is unclear whether the degree of race-consciousness that employers would need to deploy in many cases to achieve racial “equity” would satisfy the caveats and limitations in Weber, including the rejection of quotas and rigid numerical goals, as well as the requirement, still alive and well, that affirmative-action programs be transitory in duration.

One important and underappreciated fact that is central to the fate of the aggressive “racial equity” measures many employers have promised is that the Supreme Court has never expressly recognized diversity as a legitimate, let alone compelling, justification for race-conscious decision-making in the employment context. In fact, education is the only sphere in which the Supreme Court has expressly acknowledged any legitimate interest in maintaining diversity.    

A key question is thus whether the courts would, and should, accept the goal of enhancing workplace diversity as valid basis for race-conscious personnel decisions. For many reasons, the blanket permission to advance diversity that the Supreme Court has granted to higher education, whether defensible or not, should not be extended to the workplace. Specifically, the courts should refuse to assume that diversity is desirable enough as a general matter to justify race preferences in hiring and promotion. Rather, race-conscious decisions should be permitted only in particular instances where an employer can specifically demonstrate that creating a more diversified workforce than would otherwise result without the deliberate use of race preferences generates net concrete benefits for the enterprise overall in light of its central mission and purpose. It is far from clear a priori how commonly such net benefits would result or could be shown. An honest assessment may sometimes, and maybe often, reveal no identifiable, tangible, or measurably positive payoffs.  Or the downsides of pursuing something like proportional group representation in the workplace will outweigh the upsides.  

The main reason that the Supreme Court’s educational affirmative-action jurisprudence should not control in the workplace is that the two settings are incomparable and serve entirely different purposes.  Schools are sites of teaching, learning, developing human capital, and preparing young people for constructive citizenship.  The “expert” consensus, upon which the courts have heavily relied, is that exposing students to people from a variety of ethnic and racial backgrounds advances these goals, with the benefits sufficiently substantial and important to justify race-conscious selection. Moreover, the relatively narrow and well-defined set of objectives that institutions of higher education share make it reasonable to assume that a diverse student body will have similar positive effects on the educational experience across the board.  (Curiously, though, the Supreme Court has not extended this assumption to K–12 education.)  It therefore makes sense for the courts to adopt a general rule for race-conscious efforts in the educational sphere.

But employment is fundamentally different from education. Quite simply, the purpose of the workplace is not pedagogical.  Rather, employees are hired and paid to do a job, deliver a service, produce a product, and complete specified tasks.  The ordinary expectation is that their activities will contribute to the employing entity’s profitability or at least to preserving its solvency.  This requires the efficient and effective operation of the enterprise at issue, and workers are expected to contribute to that state of affairs.  Additionally, employing entities are much more variable than educational institutions on dimensions that bear on whether a diverse workforce and the steps needed to produce and maintain it create a net positive or negative effect.  Although employing organizations share the goal of economic viability, they vary dramatically in function, purpose, size, type, and mode of operation.  They also present their staff with a spectrum of job demands and employment situations, ranging from complex, intricate teamwork to solitary, self-directed production.  Given these realities, it is far from obvious that a diverse set of workers can be counted on to best serve the objectives of each and every employer across the entire economy, regardless of the organization, firm, business, or activity at issue, or the attributes, availability, and responsibilities of potential workers.  

Accordingly, the courts should not assume that achieving workplace diversity is a “compelling interest” that always justifies a departure from race-neutral principles.  Nor should the judicial acceptance of demographic diversity as a compelling interest for education automatically carry over into the employment sphere.  Rather, hiring entities and managers should be required to justify race-conscious practices for diversity purposes on a case-by-case basis by demonstrating the concrete benefits of diversifying workforce composition in their specific operational settings using accepted, well-defined, quantitative metrics like productivity, profits, quality services, and growth, or other appropriate and precise metrics. Certainly the courts should not simply defer to employers’ assertions that a more diverse workforce actually delivers favorable outcomes without clear evidence of that result. 

The employer’s burden will not be easy to carry.  It is today widely assumed that enhancing workforce diversity is an unalloyed positive.  But despite all the feel-good platitudes extolling the virtues of racial, ethnic, (and gender) mixing in the workplace and repeated invocations of the mantra that “diversity is our strength,” the advantages of creating a more racially and ethnically varied staff, and especially of increasing the number of workers from underrepresented groups than would otherwise result from colorblind hiring practices, are speculative and unproven.  The social science on the question is sparse, spotty, and equivocal, and the results decidedly mixed.  Given the paucity and type of evidence, it is just as plausible to assume that hiring individuals from similar backgrounds is a better practice, or that workplace uniformity operates more effectively in many settings.  Historical examples of decidedly undiverse teams producing outstanding results are not hard to find.  The Apollo 11 moon mission was run mostly by White guys with buzz cuts. They put a man on the moon. Would a team with significantly more women or minorities have done a better job? The staff at Bletchley Park who unscrambled the German Enigma machine and cracked the Nazi secret war code consisted mostly of young, male, well-educated, Caucasian British citizens.3See generally Hugh Sebag-Montefiore, Enigma: The Battle for the Code (2000) (telling the stories of the men who cracked the Enigma code). Their successful efforts proved crucial to an Allied victory in World War II.  The highly lucrative Belgian diamond trade has for centuries been almost entirely run by a small group of Orthodox Jewish merchants based in Antwerp. None of these examples—and there are many more—rules out the possibility that more diverse teams would sometimes do a better job.  But whether, when, and under what circumstances is ultimately an empirical question. A priori generalizations are baseless, and any determinations on the question must proceed piecemeal.  Moreover, any rigorous assessments must take into account that, in our increasingly racially and ethnically mixed country, some degree of diversity will already spontaneously exist in many jobs and workplaces without race preferences, even if not strictly proportional to all population groups. This means that the supposed benefits of the enhanced diversity traceable to such preferences will, in many cases, be merely incremental. 

Also essential to evaluating the desirability of diversification efforts is a substantial and solid body of evidence, accumulated over decades, that steps taken to achieve greater racial balancing in the workplace can carry costs that result from the hiring and promotion of less competent employees.  Studies by industrial and organizational psychologists have long revealed that qualifications, skills, experience, interests, abilities, and educational credentials vary significantly across racial and ethnic groups in American society, with Blacks and, to a lesser extent, Hispanics lagging on average behind Whites and Asian Americans on parameters that predict job performance. This means that increasing the presence of Blacks and other underrepresented minorities virtually always requires accepting lower hiring standards and relaxing well-established, meritocratic selection criteria.  Because many common personnel screens employers have adopted have been shown to predict productivity and performance, altering them can result in a less capable workforce that can impose costs on firms and organizations.  Among the industrial experts who study workplace productivity, this pattern is known as the validity-diversity trade-off.  The data show also that negative effects become more pronounced as jobs become more selective, competitive, and demanding of high-level skills, because the paucity of workers from less qualified groups at those levels becomes more pronounced. 

These documented realities are, and should be, pertinent to the legal bona fides of race-conscious employment initiatives.  As a general rule, companies pledged to pursue racial “equity” either downplay or ignore costs to the enterprise associated with diversity efforts and many would bristle at the suggestion, or at least at the open acknowledgment, that any trade-offs exist between hiring greater numbers of underrepresented minorities and an employer’s effective operation and organizational interests.  The trade-offs, however, are real.  Although the balance of costs and benefits will vary somewhat across different occupations and economic sectors, present racial and ethnic differences in educational achievement, qualifications, and skills mean that negative effects will exist, and especially for highly selective and sophisticated positions.  Courts charged with deciding whether diversity in employment serves a “compelling” interest or any legitimate interest at all should not allow employers to sidestep the possibility of these effects.  They should force employers to account for both downsides and upsides, costs as well as benefits, of the race-conscious measures they adopt.  More specifically, any legal assessment should consider how race-conscious organizational practices affect merit-based hiring, job-related qualifications demanded for particular positions, and the changes in customary criteria for staffing and hiring.  A critical and searching look at the impact of these parameters on the employing organization should be an essential part of any inquiry.

To be sure, affirmative action in the educational setting likewise carries costs, which are routinely minimized or simply denied. Competitive universities must ordinarily relax their customary academic standards, at least for some students, to achieve the diversity they seek.  The consequences of that adjustment, however, are much debated.  Whether universities can accomplish their central mission despite the presence of some groups of students who are less academically proficient is a complex question that admits of no straightforward answer.  In contrast, the use of race-conscious personnel methods for the purpose of increasing workplace diversity is a far simpler matter.  Defending that practice would appear difficult, if not impossible, if it fails to advance legitimate business interests or entails a sacrifice of firm productivity, efficiency, or effectiveness.  Absent concrete and demonstrable benefits, a firm’s desire to socially engineer a workplace that “looks like America” for its own sake should not count as an acceptable reason to deviate from the colorblind mandate expressly written into the law.  Likewise, judges should not uncritically accept the oft-heard assertion that increasing workplace diversity enhances an organization’s ability to serve its minority clients more effectively.  For instance, Black and other minority health care workers are claimed to generate superior outcomes for patients from underserved groups.  These statements are easy to make but hard to prove, and their validity should not be assumed.  Yet another reason employers give for pursuing diversity is to cater to customers or clients who prefer to do business with companies that employ a demographically varied workforce.  Courts have traditionally rejected customer preference as grounds for departing from race neutrality, and there is a long history of minority exclusion on that basis.  They should likewise be reluctant to permit employers to justify affirmative action from similar motives.  That such demands might be made for idealistic reasons or based on political convictions should carry no weight.  Pursuing diversity for its own sake or because clients, managers, or other important and influential people regard it as appealing, desirable, or “the right thing to do,” cannot be allowed to overcome the explicit legal protections against discrimination written into Title VII.  In sum, it should be incumbent on organizations that adopt race-conscious methods to show that their hiring practices actually advance their core mission and legitimate business purposes.  Other rationales should not suffice.


There are at least two potential impediments to the courts taking a harder line on race preferences in the workplace than for education.  First, the prohibition on making employment decisions “because of” race or other protected categories in Title VII, which can ground complaints of employers’ unlawful disparate treatment, has also given rise to the doctrine of disparate impact, which imposes liability for employment practices that produce deviations from proportional group representation without an affirmative demonstration of “business necessity.” An employer faced with making that showing has an incentive to use race-conscious methods (that is, to act “because of” race) to achieve proportional representation, thus escaping the presumption of liability.  Because the disparate-treatment and disparate-impact doctrines potentially subject employers to conflicting demands, allowing employers to cite the goal of diversity to justify race-conscious selection would enable the courts to ease the tensions between the two.  The temptation to grant that permission should be resisted.  A better and more realistic course would be to curtail disparate-impact liability or abolish it altogether.  The latter would require congressional action in light of the 1991 Civil Rights Act, which validates the disparate-impact discrimination doctrine articulated in Griggs v. Duke Power (1971).  Even without any significant change in the positive law, however, the courts have considerable leeway to revise judicially crafted aspects of the doctrine by adopting more flexible and less stringent standards for disparate-impact liability.  As I have argued elsewhere, the courts should abandon the unreasonable expectation that all groups will be represented proportionately throughout the workforce and instead should rely more heavily on data showing the actual profile of ability, talents, interests, and qualifications for different groups.  The disparate-impact doctrine should be modified to negate the presumption of liability for the broad range of numerical imbalances that are to be expected in light of these group differences.

Another possibility is that employers who face challenges to overtly race-conscious efforts to hire more underrepresented workers could revert to defending their practices as a remedy for past societal discrimination, which is a rationale that the Supreme Court has sometimes accepted, albeit implicitly, as a basis for voluntary affirmative-action programs such as the one at issue in Weber.  Although the courts have mostly disfavored vague and open-ended invocations of generalized discrimination as the basis for race-conscious initiatives, that rationale has not been definitively repudiated.  The courts could choose to breathe new life into affirmative action as a corrective for “societal discrimination” by invoking the now popular and pervasive parlance of “systemic” and “structural” racism.  Reviving and broadening the remedial project, which lies closer to the core of the civil-rights laws’ purposes than the pursuit of diversity, is a doctrinally tempting way to sidestep the empirical weaknesses of the diversity justification in the employment context.  But the courts should firmly reject that ploy, however fashionable at the moment, as an unwarranted and ungrounded end run around longstanding principles of remedial fairness.  In contrast to claims of specific and defined instances of discrimination, allegations based on “systemic” or “structural” racism make use of imprecise, unsubstantiated, and protean categories that are subject to ready manipulation for partisan purposes.  Such concepts tend to corrode impartial meritocratic principles and can be deployed indefinitely to the disadvantage of social groups, such as White males, that become disfavored or targeted by politically powerful factions.  If anything, the Supreme Court should use any challenges to aggressive workplace initiatives grounded in the rhetoric of “structural racism” as opportunities to narrow rather than expand the reach of Weber, which not only flies in the face of the express language of Title VII, but flouts longstanding requirements of proof of actual legal violations and specific findings of liability before remedial measures can be implemented under Anglo-American law.

Whether courts will be confronted with any of these issues depends critically on whether and to what extent the newly articulated and widespread promises to pursue workplace “equity” are actually kept, and whether and what kind of legal challenges are brought to these practices.  These issues remain to be resolved in the future.

* * *

Amy L. Wax is the Robert Mundheim Professor of Law at the University of Pennsylvania Law School.

* * *

Click here to return to the Affirmative Action at a Crossroads series main page.

One thought on “Pursuing Diversity: From Education to Employment

Leave a Reply