Nos. 90-1650, 90-2182.
Argued December 7, 1990.
Decided May 4, 1992.
Reargued En Banc December 15, 1992.
Decided March 29, 1993.
John L. Gubbins, Monfort, WI, Kimberly A. Sutherland, Chicago, IL (argued), for plaintiffs-appellants in No. 90-1650.
Sarah Vanderwicken, Darka Papushkewych, Jay M. Kertez, Ruth M. Moscovitch, Asst. Corp. Counsel, Lawrence Rosenthal, Deputy Corp. Counsel (argued), Frederick S. Rhine, Asst. Corp. Counsel, Mardell Nereim, Kelly R. Welsh, Asst. Corp. Counsel, Office of Corp. Counsel, Appeals Div., Chicago, IL, for defendants-appellees, in No. 90-1650.
Irving Gornstein, David K. Flynn, Asst. Atty. Gen., Dept. of Justice, Civil Rights Div., Appellate Section, Washington, DC, for U.S. amicus curiae.
Robert S. Sugarman, Stephen B. Horwitz (argued), Jacobs, Burns, Sugarman Orlove, Scott F. Turow, Karen H. Flax, Sonnenschein, Nath Rosenthal, Chicago, IL, for plaintiffs-appellants in No. 90-2182.
Sarah Vanderwicken, Mandell Nereim, Office of Corp. Counsel, Judson H. Miner, Davis, Miner, Barnhill Galland, Darka Papushkewych, Jay M. Kertez, Ruth M. Moscovitch, Asst. Corp. Counsel, Lawrence Rosenthal, Deputy Corp. Counsel (argued), Kelly R. Welsh, Asst. Corp. Counsel, Office of Corp. Counsel, Appeals Div., Chicago, IL, for defendants-appellees in No. 90-2182.
Irving Gornstein, John R. Dunne, and David K. Flynn, Asst. Attys. Gen., Department of Justice, Civil Rights Div., Appellate Section, Washington, DC, for U.S. amicus curiae.
Appeal from the United States District Court for the Northern District of Illinois.
Before BAUER, Chief Judge, and CUMMINGS, CUDAHY, POSNER, COFFEY, EASTERBROOK, RIPPLE, MANION, and KANNE, Circuit Judges.
We granted rehearing en banc to examine the following difficult and important question: In what circumstances can admitted racial discrimination, challenged as a violation of the equal protection clause because committed by an organ of state government, be upheld as a proper remedial measure when there has been no trial to determine whether the discrimination can survive "strict scrutiny"? For that is the constitutional standard applicable to discrimination against as well as in favor of whites. City of Richmond v. J.A. Croson Co., 488 U.S. 469, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989). A divided panel of this court upheld the district court's grant of summary judgment for the defendants in one of two cases that had been consolidated for appeal, Chicago Fire Fighters Union v. Daley, No. 90-2182, but reversed it in the other, Billish v. City of Chicago, No. 90-1650, on the ground that the district judge (a different one) had failed to apply the standard of strict scrutiny to the defendants' conduct. 962 F.2d 1269 (7th Cir. 1992). Because the district court decided the cases on summary judgment, we can affirm only if, had the record before the district court in the two cases been the record of a complete trial, the defendants would have been entitled to a directed verdict. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248, 251, 106 S.Ct. 2505, 2510, 2511, 91 L.Ed.2d 202 (1986); Cygnar v. City of Chicago, 865 F.2d 827, 834 (7th Cir. 1989).
The cases are so similar that we can treat them as if they were one. Both involve challenges to racial favoritism in promotions within the uniformed ranks of the Chicago Fire Department. The ranks are firefighter, engineer, lieutenant, captain, and battalion chief, but the last is not in issue. Promotions are made on the basis of competitive examination. Eligibility to sit for an examination is limited to persons in the rank below, except that a firefighter may, if he wants, sit for the lieutenant's examination, and be promoted to lieutenant, without spending any time as an engineer. The record is silent on whether there is some minimum amount of time that one must spend in one's present rank before being eligible for promotion; but as examinations for promotion are not given every year, there would be delays in promotion even if there were no required waiting periods. After an examination is administered and graded, the candidates are ranked from first to last on the basis of their grades and are placed in rank order on an eligibility list from which promotions to the next rank are made. All listed persons whose grades are above a cut-off point deemed to mark minimum eligibility — all, in short, who have a passing grade on the examination — are eligible for promotion to the next rank.
In 1986 the fire commissioner, defendant Galante, promoted eighteen lieutenants to captain on the basis of their rank order in a captains' eligibility list that had been compiled in 1979, the last time the captains' exam had been given. All the promoted lieutenants were white. Only two members of recognized minorities (blacks and Hispanics, so far as relevant to this case) were left on the list, and their scores on the captain's exam were well below the cut-off point. The commissioner asked the personnel department to lower the cut-off point, and it did. He then, early in 1987, promoted the two minority candidates to captain out of rank order. The plaintiffs in the Billish case are white lieutenants who had scores above the (old) cut-off point on the 1979 list and were passed over when the two minority lieutenants were promoted over them. They also complain because, the list having become all-white after the minority lieutenants were promoted to captain, the commissioner refused to make further promotions from the list, deciding instead to await the compilation of a new list based on a new exam.
In 1987 new eligibility lists were compiled to guide promotion from firefighter to engineer and from lieutenant to captain. These lists were based on examinations administered between 1985 and 1987; part of the engineer's exam was "race normed" to improve the scores of the blacks and Hispanics who took it. From these lists Commissioner Galante in 1987 ordered the promotion of 56 firefighters to engineer and 24 lieutenants to captain. The first 48 promotions to engineer were made in rank order and included 9 minority candidates; the last 8 promotions were of black firefighters taken out of rank order. The first 23 promotions to captain were made in rank order and included 5 blacks (no Hispanics); for the twenty-fourth, the commissioner reached down for a Hispanic and promoted him out of rank order. These departures from rank order were in conformity with a 25 percent goal, established by the Washington administration, for minority promotions in the fire department. The plaintiffs in the Chicago Fire Fighters case are the whites passed over for promotion to engineer or captain as a result of the promotion of nonwhites out of rank order in the 1987 eligibility lists.
The blacks and Hispanics promoted out of rank order to the detriment of the promotional opportunities of the plaintiffs in these two cases were promoted because of their race or ethnicity. The equal protection clause forbids states to give preference to persons on the basis of their race (or ethnicity, but this needn't be discussed separately), even if theirs is not the white race. City of Richmond v. J.A. Croson Co., supra, 488 U.S. at 493, 109 S.Ct. at 720; United States v. City of Chicago, 870 F.2d 1256, 1261 (7th Cir. 1989). To that extent the Constitution is truly colorblind. The Supreme Court has, however, carved an exception for the case in which discrimination against whites is necessary to rectify previous discrimination in their favor committed by the state agency that is seeking to practice remedial discrimination. City of Richmond v. J.A. Croson Co., supra, 488 U.S. at 509, 109 S.Ct. at 729; Wygant v. Jackson Board of Education, 476 U.S. 267, 280, 106 S.Ct. 1842, 1849, 90 L.Ed. 2d 260 (1986) (plurality opinion); Associated General Contractors of California, Inc. v. Coalition for Economic Equity, 950 F.2d 1401, 1412 (9th Cir. 1991). We must decide whether the defendants have so far succeeded in bringing themselves within the exception that no reasonable trier of fact could doubt that the discriminatory promotions described above were appropriate remedial measures; only then would the defendants be entitled to summary judgment, foreclosing the plaintiffs' right to a trial. Since a discriminatory remedy, to pass constitutional muster, must discriminate no more than is necessary to rectify the discrimination for which it is the remedy, City of Richmond v. J.A. Croson Co., supra, 488 U.S. at 505-08, 109 S.Ct. at 726-28; see also Wygant v. Jackson Board of Education, supra, 476 U.S. at 274, 106 S.Ct. at 1847 (plurality opinion), we must consider not only whether the Chicago Fire Department ever discriminated unlawfully in favor of whites but also whether, if so, the curative measures taken in these two cases have been carefully designed to avoid unnecessary injury to white persons.
Although there has never been a formal judicial determination that the Chicago Fire Department ever discriminated in favor of whites, we do not understand the plaintiffs to deny that, for at least a brief period of time after the date in 1972 on which Title VII of the Civil Rights Act of 1964 first became applicable to states and municipalities, the department was in violation of that statute under a "disparate impact" theory of discrimination. The fact that when the city first came under the Act the percentage of blacks and Hispanics in the uniformed ranks of the fire department — 5 percent — was far lower than their percentage in the relevant labor market would not of course have established a violation, Hazelwood School District v. United States, 433 U.S. 299, 309, 97 S.Ct. 2736, 2742, 53 L.Ed.2d 768 (1977), but the city appears to have violated the Act by not changing, when the Act became applicable to the city, the hiring practices responsible for this disparity, or showing that they were essential to the department's efficient operation. Id. at 309 n. 15, 97 S.Ct. at 2742 n. 15. The objectionable hiring practices were, however, eliminated by a consent decree entered in 1974 in a suit against the city brought by the Department of Justice. Promotions were the subject of a separate consent decree, entered in 1980 in another suit by the Justice Department against the city. This decree required the fire department to compile new eligibility lists on the basis of new tests properly validated as fair to minorities, and in promoting from the old lists to engineer or lieutenant to promote one black or Hispanic for every four whites who were promoted.
Assuming, then, that there was actionable discrimination in favor of whites in hiring from 1972 to 1974 and in promotions from 1972 to 1980, we come to the question of the appropriateness of the fire commissioner's actions in 1987 as measures for rectifying that discrimination. Even if the measures had been required by the consent decree, which they were not — the decree contained no captain's quota at all, and the quotas for engineers and lieutenants were only 20 percent and are thus exceeded by the measures challenged here — this would not be a defense. The plaintiffs were not parties to the decree and there was no judicial determination that the decree was necessary to rectify previous discrimination. Martin v. Wilks, 490 U.S. 755, 109 S.Ct. 2180, 104 L.Ed.2d 835 (1989); United States v. City of Chicago, 978 F.2d 325, 330 (7th Cir. 1992); People Who Care v. Rockford Board of Education, 961 F.2d 1335, 1337 (7th Cir. 1992); Donaghy v. City of Omaha, 933 F.2d 1448, 1456-58 (8th Cir. 1991). The decree moreover was entered before the Croson decision made clear that reverse discrimination is subject to strict scrutiny. The civil rights amendments enacted in 1991 forbid challenges to civil rights consent decrees by persons having "actual notice" of the decree and a "reasonable opportunity to present objections" to it, 42 U.S.C. § 2000e-2(n)(1)(B)(i), but the city does not argue either that the provision applies to this case or that, if it did, it would bar these suits.
We have three sets of promotions to consider. The first, of 20 lieutenants to captain, was from an old eligibility list — that is, an eligibility list based on an examination given before the 1980 consent decree: the 1979 list of lieutenants who had taken the then most recent captain's test. The second set of promotions was of the 56 firefighters promoted to the rank of engineer on the basis of an examination given in 1985. The third set consisted of the 24 lieutenants promoted to captain on the basis of a two-part examination given in 1986 and 1987.
The city's strongest case for legitimate remedial discrimination involves the promotion of the two minority lieutenants to captain in January 1987. At the time, the only eligibility list for captain had been compiled in 1979 on the basis of an examination that had not been validated for racial neutrality. Neither of the two minority lieutenants on that list had scores above the cut-off point. The commissioner had just ordered the promotion of eighteen lieutenants from the list — all white. Only 4 percent of the captains in the Chicago Fire Department at the time were nonwhite, while 29 percent of the firefighters (the lowest rank) were. By reaching down to the two minority lieutenants who were below the cut-off point, the commissioner caused 10 percent of the promotions to captain to be of nonwhites, and this was a percentage must lower than that of nonwhites in any of the lower ranks.
Yet at the time the commissioner acted, it was plain that the compilation of a new eligibility list, one based on examinations designed under the aegis of the consent decree and therefore presumptively unbiased, was only months away, for the department had already administered the first half of a new captains' test. And in fact by October 1987 the new eligibility list had been compiled and new promotions made from it. It is not argued that the department had a desperate need for two more captains in January, that it couldn't wait till October. It seems that, distressed by the all-white composition of the group of captains appointed in the latest round of promotions, Commissioner Galante decided to appoint possibly unqualified blacks without waiting to see how blacks would perform on a race-neutral examination. If that is what happened — a question for trial — it would not bespeak the kind of sensitivity to the importance of avoiding racial criteria in making employment decisions, whenever it is possible to do so, that Croson requires. It should be unnecessary to add that a public employer cannot be allowed to justify reverse discrimination by the bootstrap method of an alternating sequence of racial promotions (or hires). That is, the city cannot get points for first using a presumptively biased eligibility list to make a string of white promotions and then turning around and trying to do some rough racial justice by promoting two blacks from the bottom of the list.
Of course if the city had a desperate need for twenty captains just then, and the only eligibility list it had was based on a quite possibly biased examination that had been given in 1979, a modest departure in favor of remedial racial balance might well be justified. When the new captains' examination was finally administered, 21 percent of the lieutenants promoted to captain on a strict rank-order basis with no racial favoritism were black, which is more than twice the percentage under the old list even after Commissioner Galante promoted the two black lieutenants out of rank order. But the record does not reveal whether the city had an urgent need for twenty captains and thus could not await the administering and scoring of an unbiased exam.
If on remand the promotion of the two black lieutenants out of rank order does not survive strict scrutiny, it does not follow that the Billish plaintiffs will win. They have no vested rights in their position on an eligibility list compiled on the basis of an examination that may have been biased in favor of whites. They will have to show that if promotions had been delayed until the 1987 list was available, and made from that list, they would have been promoted sooner than they were.
The engineer examination administered in 1985 was devised by the city, of which Harold Washington, a black, was now the mayor, with the aid of an academic expert; in consultation with the Justice Department, which had brought the suit concluded by the consent decree entered in 1980; and in apparent or at least intended conformity with the EEOC's guidelines for "content validation" of employment tests. 29 C.F.R. § 1607.5. Moreover, the examination was approved by the district court. The first part of the examination was a written test. The ratio of the minority pass rate to the white pass rate was less than the 80 percent benchmark that the EEOC and other federal agencies use to decide whether an employment test has a racially adverse impact. So the city raised the minority candidates' scores sufficiently to increase the ratio to 84 percent. Candidates could sit for the second half of the examination, a "hands on" practical test (operating a fireengine simulator), only if they passed the written test, the scores on which, as we just noted, had been adjusted to minimize the adverse impact on blacks and Hispanics. The practical test did not require a racial adjustment, because on it the ratio of the minority to the white pass rate was 87 percent. Rank order on the eligibility list for promotion to engineer was based on performance on the practical test, weighted 10 percent by seniority.
Had the department stuck to rank order, 10 of the 56 firefighters promoted to engineer would have been black or Hispanic, which is 18 percent, very near the quota set by the consent decree. Instead, by promoting 8 more nonwhites, taken out of rank order, the department raised this percentage to 30 percent. Yet not only had the examination been approved by the parties to the consent decree — by the Justice Department, which had originally challenged the fire department's promotion procedures as unlawfully discriminatory in favor of whites, and by the City of Chicago, which under the Washington administration could hardly be thought insensitive to the interests of minority groups. Auriemma v. Rice, 910 F.2d 1449, 1451 (7th Cir. 1990) (en banc); Cygnar v. City of Chicago, 865 F.2d 827, 835 (7th Cir. 1989); United States v. City of Chicago, 870 F.2d 1256, 1257-58 (7th Cir. 1989); United States v. City of Chicago, 894 F.2d 943, 949 (7th Cir. 1990) (concurring opinion). In addition, the scores on the examination had been altered to improve the performance of nonwhites — the practice of "race norming" now forbidden by federal law. 42 U.S.C. § 2000e-2( l). In these circumstances, the departures from rank order could not be justified — at least with the certitude required to head off a trial — on the ground that the order had been based on a discriminatory test.
The city has, it is true, presented evidence that the test which it devised was subtly, unconsciously, biased in favor of whites, after all. The evaluation of that evidence is a matter for trial. That the city was determined to achieve or exceed its new goal of 25 percent minority promotions, regardless of what any test showed, is suggested by its insistence that "content validation" is not an adequate method of validating an employment test. To validate a test for content means to establish that it tests for skills or knowledge used on the job: a typing test for secretaries, for example. The city designed the engineers' test as one that would be content validated, as distinct from a "criterion validated" test, which tests qualities predictive of or correlated with job performance (such as IQ). The law is clear that either method of validation is proper. Gillespie v. Wisconsin, 771 F.2d 1035, 1040-41 (7th Cir. 1985); 29 C.F.R. § 1607.5(A). Now the city has backtracked, claiming that content validation is no good, and even suggesting that the fact that whites do better than others on a test proves that the test is biased even if it has been validated. Again, the law is to the contrary. "There is no iron law of human behavior that every racial or ethnic group will perform equally well on nonbiased examinations in all fields of human endeavor." United States v. City of Chicago, supra, 870 F.2d at 1261; see also City of Richmond v. J.A. Croson Co., supra, 488 U.S. at 503, 109 S.Ct. at 726. The city's grounds for rejecting the results of its test sound like pretexts designed to give the city more freedom to practice reverse discrimination by departing from rank order fixed by nonbiased tests. It will require a trial to determine whether this is in fact the city's motive in repudiating its own test.
The city argues that the disparity between the percentage of minority firefighters (29) and the percentage of minority engineers (11) justified heroic efforts, even ones that might require disregarding the results of a perfectly fair test, to advance minority firefighters by way of remedy for previous discrimination in favor of whites. The disparity by itself proves little, and this for two reasons. First — a point explained next — the much larger percentage of entry-level blacks and Hispanics may reflect nonactionable discrimination in years past. Second, the larger minority fraction of firefighters than of engineers may have reflected, to a degree as yet unexplored, the effects of a 50 percent minority hiring quota fixed in the 1974 consent decree. If, to take an example closer to home, law schools through their affirmative-action programs increased the percentage of minority law students to 10 percent, the fact that law school faculties were on average only 2 percent black would not be evidence of discrimination in appointments to faculty positions. The larger fraction of minority students would be an artifact of the law schools' having lowered their admission standards and not a true measure of the pool from which faculty would be recruited.
We come to the last set of promotions. Had the city promoted lieutenants to the rank of captain on the basis of the 1987 captains' eligibility list strictly on the basis of rank order, 5 out of the 24 promotions, or 21 percent, would have been of minority candidates — all as it happened black. Apparently distressed that none of the successful candidates was Hispanic, the fire commissioner promoted a Hispanic lieutenant to captain out of rank order. The result was to increase the minority representation in the promotions from 21 to 25 percent. The city points to the very small percentage of minority captains in 1987 — only 4 percent — and asks us to infer that this must have been a legacy of discrimination. Although the record contains no evidence on the required or average waiting period in one rank before promotion to the next, it is common sense that working one's way up the ladder to the fire department's second-highest rank may well take many years. Hence a consequence of the very low hiring of minority firefighters before 1974 may have been that in the middle 1980s only a tiny fraction of captains consisted of blacks and Hispanics. From this it can be argued that racial discrimination in their favor was necessary to bring the percentage of captains from these groups nearer to where it would have been but for the discrimination that the groups had experienced many years earlier.
This of course is possible, but has not been shown, and in fact is unlikely. Recall that, so far as appears, the city was guilty of no actionable discrimination in hiring for the fire department until 1972, and that the discrimination was rectified in 1974. That the small fraction of minority captains in the mid-1980s was due to any very great extent to the limited hiring of minority firefighters between 1972 and 1974 is implausible. Of course it is possible that the city had been violating the Constitution, and not just the standards of Title VII not yet applicable to it, before 1972, by engaging in intentional discrimination in favor of whites. But of this there is no evidence. And it is difficult to see how discrimination aimed at remedying conduct lawful when engaged in (that is, before 1972) could be thought remedial in the relevant sense and survive strict scrutiny; the Court's dicta suggest it could not. City of Richmond v. J.A. Croson Co., supra, 488 U.S. at 500, 109 S.Ct. at 724; Wygant v. Jackson Board of Education, supra, 476 U.S. at 289, 106 S.Ct. at 1854 (O'Connor, J., concurring); Regents of University of California v. Bakke, 438 U.S. 265, 307, 98 S.Ct. 2733, 2757, 57 L.Ed.2d 750 (1978) (opinion of Powell, J.).
Another possibility is that minority captains were so few in the mid-1980s not because there were so few minority hires before 1974 but because of discriminatory promotional policies not rectified until the 1980 consent decree — indeed not until later, since promotions based on validated captains' tests were not made until 1987. That is another issue requiring further exploration.
Even disregarding these points, we note that, by 1987, 14 percent of the lieutenants in the fire department were black or Hispanic, and with a higher percentage of black and Hispanic lieutenants than this — 21 percent — being promoted to captain without any "race norming" or other discrimination in their favor on the captain's examination, there would in all likelihood soon have been a large percentage of minority captains even if promotions were made on a colorblind basis. The question becomes whether accelerating the process by departing from colorblind procedures was justifiable as a means for rectifying the discrimination in which the department may have engaged in the 1970s. Perhaps so, but an alternative possibility — supported by the testimony of the city's own expert that rank-order promotion from the 1987 captain's eligibility list would have yielded 4.7 percent Hispanic promotions, the only problem being that there were no Hispanics in the first group to be promoted from the list — is that the department was merely trying to curry favor with Hispanics by promoting a Hispanic lieutenant out of rank order. It is impossible to choose between these hypotheses on the basis of the record compiled in the summary judgment proceedings.
There is concern and resentment about the use of policies of affirmative action or (more bluntly) reverse discrimination to practice a form of racial politics that is the mirror image of discrimination against blacks and other minority persons. There is fear that persons who have not been discriminated against will be advanced at the expense of persons who have neither practiced nor benefited from discrimination. The Supreme Court in the Croson decision made clear that cities and states would have to furnish convincing evidence that their affirmative-action programs were not of this character but were instead sensitive, good-faith measures to remedy the unlawful practices in which they had engaged in the past. The defendants have not shown this strongly enough to permit the dismissal of the plaintiffs' claims without a trial.
These cases, filed in 1987, were first argued in this court more than two years ago. We regret the delay in deciding them and would like to bring them to an end but have concluded that it was premature to terminate them by a grant of summary judgment. In the hope of expediting the proceedings that must resume in the district court we direct that the cases be consolidated in that court and assigned to a single judge.
REVERSED AND REMANDED, WITH DIRECTIONS.
These cases were resolved correctly, according to the law and the facts, in the panel opinion. 962 F.2d 1269 (7th Cir. 1992). For the reasons set forth in that opinion, the judgment in Billish v. City of Chicago (No. 90-1650), should be affirmed in part and reversed and remanded in part. The judgment in Chicago Fire Fighters Union v. Daley (No. 90-2182), should be affirmed.