By: Melissa Hart*
When Congress passed the 1991 Civil Rights Act (“1991 Act”), the new disparate impact provisions of the law were heralded as a victory for civil rights plaintiffs. After all, the statute was enacted in response to the Supreme Court’s cramped, “near-death” interpretation of disparate impact law in Wards Cove Packing Co. v. Atonio. The new law was a legislative sanctioning of the judicially created doctrine that facially neutral policies may still violate Title VII if their impact falls too heavily on a protected class and they cannot be justified as “business necessity.” This aspect of antidiscrimination law was viewed by many as the best chance for challenging the “built-in headwinds” that continue to keep equal employment opportunity out of reach.
Twenty years later, it is not at all clear that the disparate impact provisions of the 1991 Act have delivered their promised victory. Disparate impact claims are very rarely successful. Moreover, the Supreme Court’s 2009 decision in Ricci v. DeStefano, while technically a disparate treatment case, may well have done as much to eviscerate disparate impact’s potential as Wards Cove did twenty years earlier. The decisions share many common themes: both have particularly unusual facts, both reveal the Court’s willingness to eschew procedural limitations to reach substantive questions not properly before the Court, and both show sharp divisions among the Justices. Perhaps most importantly, both reveal deep skepticism on the part of many Justices about the underlying premise of disparate impact law: that racial inequalities persist because of continued systemic and institutional biases that can and should be addressed.
But while Wards Cove spoke directly to standards of proof for litigating disparate impact claims, Ricci’s consequences will be felt on the compliance side of the law. These consequences may be especially dire because disparate impact was always most useful for its deterrence and compliance effects. Even though plaintiffs have only rarely succeeded in bringing disparate impact claims, the powerful statement of equality inherent in such claims—embodied in the principle that employers should not use facially neutral practices that create a disparate impact unless there is a true business necessity to do so—is an essential message of antidiscrimination law. And the possibility of disparate impact litigation prompts companies to evaluate their own practices and to make internal adjustments that make employment policies more fair.
This Article begins, in Part I, by considering the early potential of disparate impact law, and the Supreme Court’s response in Wards Cove. Part II evaluates how much the Civil Rights Act of 1991 actually promised discrimination plaintiffs and examines how disparate impact litigation developed in subsequent years. Part III considers the Court’s decision in Ricci and its consequences for the voluntary compliance efforts that disparate impact law has encouraged.
When the Supreme Court in 1971 first recognized disparate impact as a legal theory under Title VII, the Court explained that the “absence of discriminatory intent does not redeem employment procedures or testing mechanisms that operate as ‘built-in headwinds’ for minority groups and are unrelated to measuring job capability.” Forty years later, it is the built-in headwinds of a Supreme Court skeptical of—perhaps even hostile to—the goals of disparate impact theory that pose the greatest challenge to continued movement toward workplace equality.
I. Giving Disparate Impact Life and Taking It Away
The disparate impact cause of action was first recognized by the Supreme Court as a necessary element of Title VII in order for that statute to truly reach all employment practices that operated to deny equal opportunity. In Griggs v. Duke Power Co., the Supreme Court explained that Title VII “proscribes not only overt discrimination but also practices that are fair in form, but discriminatory in operation. The touchstone is business necessity. If an employment practice that operates to exclude Negroes cannot be shown to be related to job performance, the practice is prohibited.” The Griggs Court understood that intentional discrimination was not only hard to prove but was also only part of the problem in workplaces that had for so long unthinkingly imposed rules that disadvantaged women and people of color.
During the 1970s and 1980s, disparate impact theory was used to challenge the kinds of “objective” employment criteria—primarily standardized test requirements—that had been disputed in Griggs. Importantly though, it also encouraged employer compliance efforts and even voluntary affirmative action programs. Lawyers and human resource professionals advised companies to carefully evaluate their job requirements and to “initiate and implement more creative selection and training procedures.” And many civil rights advocates viewed disparate impact theory as a driving force behind Title VII’s success as a “major instrument of social progress.”
But disparate impact faced vocal criticism from the beginning. Courts and commentators worried that
acceptance of the idea that discrepancies between racial composition of the community and the plant or department alone make out a prima facie case of discrimination leads inevitably toward a narrowing of the Court’s options in fashioning a remedy. If the problem is to be demonstrated by the mere fact of a discrepancy, then the solution logically must amount to an order to bring the employment statistics into line with the population statistics . . . .
This fear, that employers would simply engage in quota hiring to avoid disparate impact liability, was a constant threat to disparate impact law’s development.
Five years after deciding Griggs, the Court concluded that the disparate impact theory was not available to plaintiffs bringing constitutional claims; instead, the Equal Protection Clause is violated only by intentionally discriminatory conduct. Indeed, the Washington v. Davis majority revealed considerable skepticism about disparate impact as a theory of discrimination, announcing that, “[a]s an initial matter, we have difficulty understanding how a law establishing a racially neutral qualification for employment is nevertheless racially discriminatory and denies ‘any person . . . equal protection of the laws’ simply because a greater proportion of Negroes fails to qualify than members of other racial or ethnic groups.” This rejection of disparate impact theory in constitutional analysis put disparate impact claims on shaky ground by creating a distinction between “true” discrimination and claims of disparate impact.
The question of whether disparate impact effectively required employers to implement quotas to avoid liability was presented to the Supreme Court as early as 1977. The concern expressed by critics of impact theory was that, if plaintiffs can make out a prima facie case of disparate impact discrimination merely by showing that an employer’s hiring or promotion policies lead to statistical underrepresentation of a protected class, then defendants will have an incentive to avoid liability by simply ensuring that their workforce does not show that statistical underrepresentation. This is troubling, critics argue, because Title VII specifically provides that the statute shall not be interpreted to require any kind of proportional representation.
In International Brotherhood of Teamsters v. United States, the Supreme Court dismissed the concern that reliance on statistical proof will lead to race-based quota hiring. In a disparate treatment case, statistics are probative because they are “often a telltale sign of purposeful discrimination.” In disparate impact litigation, statistical disparities push the employer to justify its business practices—to explain why the practice that is creating the disparity is actually necessary for the workplace. Liability will not flow from statistical disparities alone, but from reliance on business practices that are unnecessary and that impose a disproportionate disadvantage on women or people of color.
The tension between those who viewed disparate impact as the best hope for challenging continued workplace inequality and those who viewed impact theory as an illegal directive to implement hiring quotas came to a head in Wards Cove Packing Co. v. Atonio. In Wards Cove, the Supreme Court confronted a disparate impact challenge to the racially segregated world of salmon canneries in Alaska. At the two canneries that were the subject of the litigation, jobs were classified as “cannery” (unskilled) and “noncannery” (skilled). The cannery jobs were filled almost entirely by Filipinos and Alaska Natives who were either hired through one union or resided in villages near the canneries. The noncannery jobs, which paid more than the cannery positions, were filled predominantly by whites who were recruited in Washington and Oregon. Cannery employees lived in separate dormitories and ate in separate dining halls from the noncannery employees. Justice Blackmun described these working conditions in his dissenting opinion:
The salmon industry as described by this record takes us back to a kind of overt and institutionalized discrimination we have not dealt with in years: a total residential and work environment organized on principles of racial stratification and segregation . . . . This industry long has been characterized by a taste for discrimination of the old-fashioned sort: a preference for hiring nonwhites to fill its lowest level positions, on the condition that they stay there.
In 1974, fifteen years before the case would reach the Supreme Court, a class of nonwhite cannery workers brought suit challenging a broad range of the companies’ employment policies: nepotism, separate hiring channels for cannery and noncannery positions, a rehire preference, a practice of not promoting from within, an English language requirement, no posting for noncannery positions, and a lack of objective hiring criteria. The plaintiffs contended that these practices “were responsible for the racial stratification of the work force and had denied them and other nonwhites employment as noncannery workers on the basis of race.” They claimed both disparate impact and disparate treatment violations of Title VII. The Wards Cove litigation had a tortuous procedural history during which the lower courts rejected the plaintiffs’ disparate treatment claims but permitted the impact claims. The dispute arrived at the Supreme Court on an interlocutory appeal, and the Court took the case as an opportunity to make a number of pronouncements about Title VII’s disparate impact standards.
In a sharply divided opinion, the Court first criticized the lower court’s comparison of the percentage of cannery positions held by nonwhites with the percentage of noncannery positions held by nonwhites. The relevant comparison, the majority explained, is between the percentage of job holders and the percentage of qualified applicants for those jobs. In telling its story about what qualifications were relevant to that comparison, the Wards Cove majority focused exclusively on the noncannery jobs that required special skills, such as accountants, doctors, and other professionals. To compare those jobs to the unskilled positions held by cannery workers was to hold the employer responsible for differences between the two labor pools that had nothing to do with the employers’ policies and practices: “If the absence of minorities holding such skilled positions is due to a dearth of qualified nonwhite applicants (for reasons that are not the petitioners’ fault), petitioners’ selection methods or employment practices cannot be said to have had a ‘disparate impact’ on nonwhites.”
The Court went on to hold that a plaintiff bringing a disparate impact challenge must identify with specificity what particular employment practice caused the complained-of disparate impact. Plaintiffs cannot make out a prima facie case of disparate impact simply by pointing to significant racial disparities in workforce composition. The Court concluded that “[t]o hold otherwise would result in employers being potentially liable for ‘the myriad of innocent causes that may lead to statistical imbalances in the composition of their work forces.’”
Finally, and most controversially, the Court reversed twenty years of disparate impact law and concluded that an employer seeking to explain racial disparity with a “business necessity” will not have to demonstrate that the practice in question is “essential” or “indispensible.” Forcing the employer to meet this burden, the majority explained, imposes too onerous a standard, and “would result in a host of evils.” This “host of evils” is the possibility that employers will engage in quotas or hiring goals in order to avoid disparate impact liability. Instead, the Court held that an employer facing a charge of disparate impact discrimination would not have to “demonstrate” anything, in the sense of meeting a burden of proof. Instead of being an affirmative defense—which “business necessity” had been since Griggs—the majority concluded that the employer’s burden should be merely a burden of production. The disparate impact plaintiff would be required to demonstrate that the challenged practice was not a business necessity. Moreover, the Wards Cove majority significantly weakened the “business necessity” threshold, concluding that an employer’s challenged policy need only serve “the legitimate employment goals of the employer.”
Wards Cove produced two impassioned dissents, one penned by Justice Blackmun and the other by Justice Stevens. Blackmun’s dissent observed that the legal changes wrought by the decision “essentially immunize[d] . . . from attack” the range of practices that entrenched “racial stratification and segregation” in the salmon industry. Justice Stevens’s dissent accused the majority of “[t]urning a blind eye to the meaning and purpose of Title VII,” when it “perfunctorily reject[ed] a longstanding rule of law and underestimate[d] the probative value of evidence of a racially stratified work force.” One of the most striking things about the three opinions—the majority and the two dissents—is what radically different meaning the dissenting Justices took from the facts of the case than did the members of the five-Justice majority. As Justice Blackmun concluded, “One wonders whether the majority still believes that race discrimination—or, more accurately, race discrimination against nonwhites—is a problem in our society, or even remembers that it ever was.”
Justice Stevens’s dissent began by observing that this case had very unusual and complicated facts and should not have been used to rewrite the law. He went on to detail the ways in which the Wards Cove majority broke from the settled law in disparate impact cases. A substantial part of the dissent was occupied with challenging the majority’s view of how to think about the statistical evidence offered to the lower courts. Where the majority disregarded the segregation of the noncannery and cannery workforces as being irrelevant comparisons, Justice Stevens argued that in the “unique industry” of Alaskan salmon canneries, there are key elements that make the comparison of these two groups particularly appropriate. He presented a very different picture of the “skilled” noncannery positions filled almost entirely by white employees; instead of focusing on the doctors and accountants that occupy the majority, he pointed out that the “skills” required for many of those positions included only things like English literacy, typing, good health, and possession of a driver’s license. Moreover, Justice Stevens pointed out that one of the most important job qualifications for both cannery and noncannery employees in this industry was a willingness to be available for and to accept seasonal employment. That important variable makes the comparison between these two groups of employees arguably more relevant than any other comparison and certainly as relevant as a comparison of noncannery workers with the general labor force.
The fundamental difference between the stories told by the dissents and the story told by the majority is a crucial element of Wards Cove. The majority saw the facts through a lens of skepticism about—even perhaps hostility to—the reach of disparate impact theory. The absolute segregation of the salmon industry did not worry the Justices in the majority because they viewed that segregation as occurring naturally, unrelated to policy choices being made by the employer. For the dissenting Justices, the “unsettling resemblance to aspects of a plantation economy” was the major concern, and the lens through which the applicable legal standards were considered. Wards Cove revealed how completely divergent views about disparate impact law mirrored similar debates about affirmative action. In both contexts, one sees the substantial divide between those who view workplace discrimination against people of color as a continuing serious problem and those who believe that antidiscrimination laws have themselves become a source of unfair treatment of white workers.
II. The Rebirth of Disparate Impact
The Civil Rights Act of 1991 was an emphatic and hard-fought rejection of several 1989 Supreme Court decisions—most especially of Wards Cove’s changes to disparate impact law. The bill that passed and that was signed by President George H. W. Bush was heralded as a victory for plaintiffs in part because of the process that led to its passage. The bill was first vetoed, and the subsequent year-long negotiations ended with what many called a “capitulation” by a Republican White House to the demands of civil rights leaders that disparate impact law remain a viable litigation theory. The core of the debate that shaped the relevant provisions of the legislation was about the relationship between disparate impact and quotas.
The 1991 codification of disparate impact explicitly returned the law, in certain respects, to its pre-Wards Cove status. In particular, section 703(k) of the Civil Rights Act of 1964, amended by section 105(a) of the 1991 Act, now specifies that “business necessity” is an affirmative defense, which the defendant carries the burden of demonstrating after the plaintiff has made out a prima facie case that an employer practice disproportionately impacts protected employees. “Business necessity,” which the Wards Cove majority had described as anything consistent with “legitimate employment goals,” is defined in the new section 703(k) as “job related for the position in question and consistent with business necessity.” The 1991 Act also specifically returned the meaning of “alternative employment practice” to that which it had been under “the law as it existed on June 4, 1989.” As to the prima facie case, which the Supreme Court had said required identification of a specific employment practice, Congress provided that a plaintiff typically does have to demonstrate a particular practice that causes a disparate impact, but the legislature offered an exception for circumstances in which the plaintiff can demonstrate “that the elements of a [defendant’s] decisionmaking process are not capable of separation for analysis.” In that circumstance, “the decisionmaking process may be analyzed as one employment practice.”
Given the battle over disparate impact that led to the 1991 Civil Rights Act, it would be reasonable to imagine an increase in the number of disparate impact cases following the statute’s enactment. In fact, however, there was no surge in the number of disparate impact suits filed after 1991. And, as Michael Selmi’s 2006 empirical evaluation of disparate impact cases demonstrated, plaintiffs had significantly more success with disparate impact claims before 1991 than after.
There are a number of possible explanations for the relatively small number of disparate impact claims in the federal courts. Perhaps most significantly, the 1991 Act added compensatory and punitive damages to Title VII’s remedial arsenal, but only for claims of intentional discrimination. This change created substantial incentives for plaintiffs to frame their suits as disparate treatment rather than disparate impact claims. Further, although the 1991 Act was quite explicit in rejecting Wards Cove, the statute still left considerable uncertainty about core interpretive questions—including what constitutes an “employment practice” subject to challenge and precisely what “business necessity” means—in disparate impact litigation. And importantly, the number of disparate impact claims was lower by the 1990s because disparate impact theory was doing what it was in large part intended to do: encourage employers to develop internal practices that did not have a disparate impact on protected classes. Indeed “[t]he disparate impact standard . . . triggered reconsideration of a wide range of promotion practices and other devices that failed to accurately measure and predict candidates’ job performance.” By 1991, twenty years after Griggs, employer practices that caused obvious disparate impact without any business justification had been eliminated in many workplaces through employers’ own internal compliance efforts.
Just as the promise of the 1991 Act might have been more rhetorical than substantive for potential disparate impact litigation, the perils that opponents saw lurking behind disparate impact theory did not emerge in the wake of the new law. There is absolutely no evidence to suggest that the newly codified disparate impact theory led employers to adopt quotas or to lower their employment standards. But the fear that potential disparate impact liability might lead employers to adopt hiring quotas—and more generally the anxiety that antidiscrimination laws were themselves prompting discrimination against white employees—has not diminished.
III. Ricci: Is Disparate Impact Dead Again?
Twenty years and twenty days after announcing its ruling in Wards Cove, the Supreme Court issued another sharply divided set of opinions in Ricci v. DeStefano. Ricci was a disparate treatment case, but the allegation of disparate treatment stemmed from the City of New Haven’s effort to avoid disparate impact liability. A five-Justice majority concluded that the City had engaged in intentional discrimination against white firefighters when it declined to certify the results of a promotion test that had a disparate impact on minority firefighters.
Ricci shared a number of similarities with the Wards Cove decision. One of the most immediately notable is that in both cases the Court’s majority ignored basic procedural norms that are supposed to constrain the Supreme Court in order to reach its preferred outcome. In Wards Cove, the Court significantly altered disparate impact law in a case that came to it on interlocutory review, and the dissent was sharply critical of what it saw as procedural impropriety. Similarly in Ricci, the dissenting Justices observed that the majority was departing from the Court’s usual procedural rules by not simply reversing the summary judgment granted and upheld below, but actually reviewing the record and granting summary judgment for the other side. The willingness to ignore procedural norms gives both opinions an aura of “judicial activism” that heightens the sense that both are part of a political debate in which statutory interpretation is just one argument.
Wards Cove and Ricci are also notable for their complex facts, and for the widely different view of the facts offered by the majority and the dissent in each case. The highly contested facts in Ricci made especially surprising the majority’s decision to grant summary judgment based on the record as it stood at the Supreme Court.
In 2003, the City of New Haven administered a written test as part of the process for selecting promotion-eligible employees for officer positions in the fire department. The test was developed to account for sixty percent of the promotion process because the City’s decades-old contract with the firefighter’s union provided that promotion would be based sixty percent on a written exam and forty percent on an oral exam. The City charter provided that, after the exam was administered, the Civil Service Board would rank applicants, creating a list from which vacancies would be filled. Candidates had to be chosen from among the top three scorers on the list, and the list would remain valid for two years. Seventy-seven candidates completed the 2003 lieutenant examination and forty-one candidates completed the examination for promotion to captain. The results on both examinations showed significant racial disparities for both African-American and Latino test takers sufficient to make out a prima facie case of disparate impact under Title VII.
As soon as the exam results were made publicly available, “[s]ome firefighters argued the tests should be discarded because the results showed the test to be discriminatory. They threatened a discrimination lawsuit if the City made promotions based on the tests. Other firefighters said the exams were neutral and fair. And they, in turn, threatened a discrimination lawsuit” if the City did not certify the results. At this point, the City found itself between the proverbial rock and a hard place.
In January 2004, the Civil Service Board met to decide whether to certify the results of the exam. At the beginning of the meeting, the City’s director of Human Resources informed the board that she believed the exam created a “significant disparate impact” on test takers. Over the course of five meetings, the Civil Service Board heard testimony from the person who had developed the test for the City, additional firefighters, New Haven community members, other professional test developers, individuals employed in fire departments in other cities, the City’s legal counsel, and a psychologist from Boston College, among others. At the close of these meetings, the Civil Service Board voted on whether to certify the results. With one member recused, the remaining four board members were deadlocked, two to two, on whether to certify; consequently, the list was not certified.
Following the decision not to certify the results, seventeen white firefighters and one Hispanic firefighter filed suit, alleging, among other claims, that the decision not to certify was an act of intentional race discrimination. In district court, the City successfully argued that the Civil Service Board’s good-faith belief that certifying the exam would expose it to liability for disparate impact discrimination shielded it from liability for disparate treatment, and was granted summary judgment. The Supreme Court rejected this argument, concluding that “there is no genuine dispute that the examinations were job-related and consistent with business necessity,” and granted summary judgment for the firefighters. For the majority, the story—the undisputed and indisputable story—of what happened in New Haven was this:
The record in this litigation documents a process that, at the outset, had the potential to produce a testing procedure that was true to the promise of Title VII: No individual should face workplace discrimination based on race. Respondents thought about promotion qualifications and relevant experience in neutral ways. They were careful to ensure broad racial participation in the design of the test itself and its administration. As we have discussed at length, the process was open and fair. The problem, of course, is that after the tests were completed, the raw racial results became the predominant rationale for the City’s refusal to certify the results.
This understanding of what happened in New Haven rests on a number of much contested assumptions about the neutrality and fairness of the City’s test and the process used to design it. The majority simply disregarded the catalog of contested factual questions. With these blinders on, it could perceive the statistically significant disparate impact of the test as legally irrelevant.
The Ricci dissent told a very different story. The dissent described a long history of race discrimination in the New Haven Fire Department and pointed to portions of the record that suggested that the challenged test was significantly more problematic than the majority’s recitation of the facts suggested. While the majority lauded the test-development process, the dissent pointed out that there was no determination before hiring the test writer of what kind of test would best evaluate candidates for promotion. In fact, the City didn’t consider any other testing mechanism; didn’t question its use of a decades-old decision to weight the written exam sixty percent and the oral exam forty percent; and didn’t vet the written exam with any experienced local firefighters.
Indeed, only after the test was administered, and the significant adverse impact became apparent, did the City seem to realize the range of flaws in the test and refer the question to the Civil Service Board. At this point, too, the dissenting opinion demonstrates that a very different story can be read in the record than the majority’s view that only statistical racial disparities mattered in the Civil Service Board’s process; the record included evidence that Civil Service Board members understood that “their principal task was to decide whether they were confident about the reliability of the exams: Had the exams fairly measured the qualities of a successful fire officer despite their disparate results? Might an alternative examination process have identified the most qualified candidates without creating such significant racial imbalances?”
The dramatically different readings of what actually happened in New Haven presented in the Ricci opinions are a result of the widely divergent views held by the majority and the dissenting Justices about the problem of discrimination. Why did the original test end up with such disparate results? The Supreme Court’s majority believed that it was because white people do better on objective tests that evaluate merit. The Ricci majority’s description of the facts was replete with quotes accounting for this discrepancy: “usually whites outperform some of the minorities on testing”; “[n]ormally, whites outperform ethnic minorities on the majority of standardized testing procedures”; and “regardless of what kind of written test we give in this country . . . we can just about predict how many people will pass who are members of under-represented groups. And your data are not that inconsistent with what predictions would say were the case.” Of course, this was all testimony that was in fact presented to the Civil Service Board. But it is just a very small sample of the testimony offered during the course of the five meetings the Civil Service Board held about these tests. There was also a great deal of evidence—the evidence credited by the dissenting Justices—that showed New Haven’s test was not developed with care and other tests would more accurately measure qualifications and would do so with much less racial disparity.
The conviction that whites just do better is central to the majority’s conclusion that the decision not to certify the test results constituted “race-based” discrimination. As Girardeau Spann has observed,
The reason that the Ricci Court displayed such unquestioning deference to the standardized promotion exam is precisely because whites outperform minorities on standardized tests. I am not suggesting that the Court conspiratorially chose to utilize an invalid selection criterion in order to favor white firefighters over minority firefighters. I am suggesting something much more troubling. I am suggesting that—despite a mass of contrary evidence—the Court actually believed the standardized test to be valid because the results of that test corresponded to the racially-correlated expectations that the culture had taught the Justices equate with merit. Because whites outperformed minorities on the exam, the exam must have been measuring qualities that were relevant to merit-based promotions. Therefore, any decision not to certify the results of that exam must have been rooted in a desire to abandon merit in favor of unwarranted racial affirmative action.
This is the point at which Ricci becomes a case about disparate impact’s increasingly uncertain future. While the majority specifically declined the opportunity to hold that Title VII’s disparate impact provisions are unconstitutional, it began its analysis “with this premise: The City’s actions would violate the disparate-treatment prohibition of Title VII absent some valid defense.” This statement could be read—and is being treated by many employment lawyers—as suggesting that efforts to avoid disparate impact on minority employees will always present white employees with a cause of action for discriminatory disparate treatment and that employers will only be able to avoid liability in those cases in which they can satisfy Ricci’s new “strong basis in evidence” defense.
Ricci did not, in fact, eliminate—or even really change—disparate impact law. Employers are still required under Title VII, if their employment practices have an adverse impact, to ensure that the practices are job related and consistent with business necessity. The majority was quite explicit in stating that an employer may still design job tests and other practices with the goal of avoiding a disparate impact. Importantly, the majority drew a line between voluntary compliance efforts that seek to avoid disparate impact in the creation and administration of employment tests and practices, on the one hand, and the evaluation of test scores after the tests have been taken, on the other. The former are not subject to the Court’s new approach. Only after a test has been taken—when the actual racial makeup of the results is known—will an employer be at risk of disparate treatment liability. At that point, of course, the risk may be significant. The “strong basis in evidence” defense, which the majority imported from case law on affirmative action, may be a hard one to meet. The Court provided no guidance about what kind of information would be sufficient for an employer to demonstrate, after it had administered a test and seen the results, that it had a strong basis in evidence for believing that it would be violating disparate impact law to use the test in making employment decisions.
What Ricci does do is make voluntary diversity efforts less appealing to employers by casting a shadow of potential litigation over these efforts. Will an employer going through a reduction in force, for example, be sued by white employees if it seeks to ensure that the reduction in force will not unduly impact minority employees? Will employers face claims of race discrimination if they participate in minority job fairs or engage in other diversity efforts? Ricci can certainly be read to suggest that any employer action taken to increase opportunities for formerly excluded minority employees constitutes intentional discrimination against white employees. As Justice Ginsburg noted in her dissenting opinion, there is a “sharp conflict” between the Ricci decision and the “voluntary compliance ideal” that has long been central to the Court’s interpretation of Title VII.
Given the important role that voluntary compliance has always played in response to the possibility of disparate impact liability, Ricci’s consequences for the viability of the doctrine as an important tool in antidiscrimination law are as significant as were the doctrinal changes of Wards Cove. Indeed, Ricci may be even more troubling because it is extremely hard to know how to respond to the opinion, not only for employers, as discussed above, but also for those seeking a legislative fix for the Court’s new legal standard. After Wards Cove, the calls for a legislative response were immediate and it was relatively clear what a responsive statute might look like: the Court’s opinion had included a series of specific doctrinal statements, and the 1991 Act contained provisions that tracked those statements. In doing so, Congress made a powerful rhetorical statement rejecting the Supreme Court’s view of the law.
Although there have been calls for a legislative response to Ricci, it really is not clear what that response could look like. Congress could pass a statute providing that the “strong basis in evidence” test is too high a standard for employers to meet when facing a disparate treatment challenge to efforts at compliance with disparate impact obligations. The legislature could instead adopt the standard proposed by Justice Ginsburg’s dissent. But either legislative fix would hardly be responsive to the rhetoric of Ricci. Still standing would be the underlying assumption: when employers seek to avoid tests that unfairly impact minority workers they are engaging in discrimination against white workers. That is the true harm in Ricci.
Many people have pointed out that Ricci, read neutrally, suggests that mere racial consciousness is enough to demonstrate intent to discriminate. This would be a radical change in employment discrimination law if applied to all cases under Title VII. And yet, nobody really believes the import of Ricci was a liberalizing of the standards that all plaintiffs must meet to prove discrimination. Twenty years before Ricci, Justice Blackmun’s Wards Cove dissent expressed the fear that “[o]ne wonders whether the majority still believes that race discrimination—or, more accurately, race discrimination against nonwhites—is a problem in our society, or even remembers that it ever was.” The same could be said of the Ricci majority, which seems to have created and applied a standard for proving discrimination that is applicable only when the plaintiff is attacking an employer’s voluntary effort to avoid disparate impact. The opinion reflects the sad reality that a majority of the Justices today are likely among the fifty-six percent of American Republicans who believe discrimination against whites is the most serious discrimination problem that our country faces. On the twentieth anniversary of the Civil Rights Act of 1991 this is a solemn statement about the true impediments to equality.
* Associate Professor of Law, University of Colorado Law School. Many thanks to Rachel Arnow-Richman, Roberto Corrada, Scott Moss, Helen Norton, and Catherine Smith for their always helpful comments.
/LRColl2009n40Sullivan.pdf (noting the “firestorm of protest” that led to the passage of the 1991 Act).