This June, in Ricci v. DeStefano, the Supreme Court held that the city of New Haven discriminated against white firefighters when it rejected the results of a promotion exam that eliminated almost all of the minority candidates. Frank Ricci and other white firefighters, as well as one Latino, claimed that the city intentionally discriminated against them on the basis of race, in violation of Title VII of the Civil Rights Act, by rejecting the exam simply because too many of the highest scorers were white and thus denying them promotions. * New Haven's defense was that it rejected the exam because using it would have violated another part of Title VII that prohibits tests that have a disparate impact on minorities—meaning any test that needlessly screens out a disproportionate number of minorities. During oral argument, Justice Souter worried that Ricci's lawsuit put New Haven in a "damned if you do, damned if you don't situation": liable for disparate-impact discrimination against blacks if it used the test and liable for intentional discrimination against whites if it didn't use it.
Sure enough, last week, just as New Haven prepared to promote a group consisting almost entirely of white fire captains and lieutenants based on the exam results, a black New Haven firefighter, Michael Briscoe, filed a disparate-impact lawsuit against the city. Like Frank Ricci, Briscoe is a sympathetic plaintiff. He received the highest score of any candidate on the oral portion of the lieutenant's promotion exam. But he isn't eligible for promotion because the city based 60 percent of each candidate's score on the written exam. On this part of the test, Briscoe—like most black candidates for promotion—did comparatively badly.
It's tempting to dismiss Briscoe as a sore loser. But if he's right, New Haven is using an outdated promotion exam that screens out some of the most qualified candidates and locks in the effects of past discrimination. That's precisely what Title VII's disparate-impact prohibition is designed to prevent. And disparate-impact doctrine may be the most important part of civil rights law today, because although fewer and fewer employers openly discriminate on the basis of race, plenty still inadvertently perpetuate the effects of past discrimination by using flawed tests.
Briscoe's claim is a perfect example. Why didn't black candidates do as well as whites on the written exam? Black firefighters argue that because whites are more likely to come from families where firefighting is a legacy (for instance, one New Haven captain's father and grandfather both served as fire chief in New Haven), they are more likely to get help from a network of friends and relatives in studying for the written exam. Few blacks have such family connections—in large part because blacks were deliberately shut out of firefighting jobs until the 1970s, when black firefighters won discrimination suits in New Haven and in many other cities nationwide. (In fact, Briscoe's lawyer, Yale Law School lecturer David Rosen, brought the first discrimination case against the New Haven fire department back then.) So heavy reliance on a written exam, if it gives an advantage to legacy candidates, could perpetuate the evils of past discrimination.
That violates Title VII, unless the exam is job related and there are no less discriminatory alternatives. New Haven's written exam may be as good as any written exam could have been: The Supreme Court in its ruling in favor of the white firefighters in Ricci pointed out that the city's written test was carefully developed by a professional company to be job-related and to avoid racial disparities. But Briscoe argues that the written exam did not, in fact, test for the skills that fire captains and lieutenants need on the ground; instead, it rewarded rote memorization. As for alternatives, Briscoe says that the city could have relied more heavily on the oral exam, which required candidates to respond to real-life firefighting and training scenarios. Neither the city nor the company that designed the exam defended making it worth 60 percent of the promotion score. Briscoe also points out that New Haven could have used an assessment-center model, which tests candidates through simulations of real-life job challenges. Many other cities use assessment centers successfully.
Although Ricci was often described as a challenge to affirmative action, getting rid of a flawed exam isn't affirmative action and doesn't push diversity at the expense of merit. Instead, it replaces a poor way of measuring merit with a better measurement. Rosen puts it this way, "Taking a hard look at what merit really means for a particular job may increase the diversity of the workforce and looking hard at the barriers to diversity may lead to choosing better people. I live about three blocks from the firehouse where Mike Briscoe works. If there's a fire on my street … everyone on the block will be fortunate if Mike Briscoe is the lieutenant."
Of course, all of this is academic if the Ricci decision forecloses Briscoe's lawsuit. And Justice Anthony Kennedy's majority opinion seems to do just that: "If, after it certifies the test results the City faces a disparate-impact suit," Kennedy writes, "then in light of our holding today it should be clear that the City would avoid disparate-impact liability based on the strong basis in evidence that, had it not certified the results, it would have been subject to disparate treatment liability"—in other words, liability for discriminating against the white test-takers. But this is not clear at all. In Ricci, the Court held that New Haven did not have sufficient evidence that the exam had an unlawful disparate impact to justify dropping it. But that doesn't mean that the evidence doesn't exist. New Haven's failure to gather that evidence can't foreclose Michael Briscoe's attempt to prove that the city discriminated against him.
Given the tense and complicated racial politics of New Haven, the city could be guilty of discrimination against both white and black firefighters. Suppose, as Justice Alito speculates in his concurrence, that New Haven dropped the exam because of political pressure from black community activists. Even so, New Haven may also have discriminated against black firefighters by caving in to pressure from the union, dominated by white firefighters, to use an exam with an unjustified racial impact in the first place.
It's also possible that New Haven could be caught in Justice Souter's "damned if you do, damned if you don't" bind despite the best of intentions. Suppose city officials suspected that making a written exam worth 60 percent of an applicant's score wasn't the best way to evaluate firefighters—and thought that there might be equally good and less discriminatory alternatives. But faced with the need to satisfy the union, and comply with various city charter and civil service rules, they saw no alternative. So they decided to cross their fingers and hoped the test wouldn't actually eliminate most of the minority applicants. When the test instead screened out all but one of them, New Haven refused to approve the exam results, fearing it couldn't defend the process under Title VII. In response, Frank Ricci sued. And now, as New Haven was about to go ahead with the promotions, Mike Briscoe has filed the disparate-impact lawsuit the city was trying to avoid in the first place.
The law should provide clear direction so that willing employers can voluntarily comply and avoid litigation. But despite Justice Kennedy's claim to have clarified the relationship between intentional discrimination and disparate impact under Title VII in Ricci, the state of the law today is more confused that ever: Title VII after Ricci makes contradictory demands and threatens to trap the scrupulous as readily as the scofflaw.
Since disparate-impact doctrine has been controversial since the Supreme Court first articulated it in 1971 in Griggs v. Duke Power, it is not surprising that some people propose to resolve the confusion created by the Ricci decision by essentially scrapping it. But even if you think this is a good idea, it is the job of Congress, not the courts, because disparate impact comes from a statute. In fact, the last time the court tried to gut disparate impact—in 1989 in Wards Cove Packing Co. v. Atonio—Congress responded by explicitly endorsing and strengthening the doctrine in the Civil Rights Act of 1991.
And Briscoe's case suggests why gutting disparate impact law is a bad idea. A promotion process that screens out well-qualified candidates and perpetuates past discrimination is both unjust and inefficient, even if it is formally evenhanded. As overt racism becomes less and less common, practices that reinforce the racial exclusion of the past have become the greatest impediments to equal opportunity. Disparate-impact doctrine allows us to dismantle these impediments without the finger-pointing that typically accompanies claims of intentional discrimination. We needn't hunt today for racists to blame in order to eliminate practices that needlessly perpetuate the racism of yesteryear. That's why the law against disparate-impact discrimination should appeal broadly—to zealous champions of social justice, beady-eyed utilitarians who just want the best person on the job, and jaded curmudgeons who are sick and tired of people playing the race card.
Correction, Oct. 23, 2009: The original sentence omitted the Hispanic plaintiff in Ricci's suit. (Return to the corrected sentence.)