RIPPLE, Circuit Judge.

While these two cases, Billish and Chicago Fire Fighters, are distinct lawsuits and were heard by different judges in the district court, the parties, counsel, legal issues, and facts in both cases significantly overlap. Accordingly, we consolidate them for purposes of decision on appeal. In both cases, nonminority members of the Chicago Fire Department (CFD) challenge as discriminatory the CFD's decisions to promote various individuals pursuant to an affirmative action program implemented by the City of Chicago (the City). They claim that those promotion decisions violated the Equal Protection Clause and, in one case, the Due Process Clause, of the Fourteenth Amendment. In both cases, the district court granted the City's motions for summary judgment. For the following reasons, we affirm the judgment of the district court in Chicago Fire Fighters (No. 90-2182), and we affirm in part and reverse and remand in part the judgment of the district court in Billish (No. 90-1650).

I [2] BACKGROUND

The uniformed, non-exempt fire suppression ranks of the CFD, in ascending hierarchical order, consist of firefighter, engineer, lieutenant, captain, and battalion chief. No special training or academic achievement is required for the entry-level position of firefighter. Vacancies at the upper ranks, however, are filled from eligibility lists compiled after job-related examinations are conducted. In general, only candidates who hold the rank immediately below the rank being tested are eligible to take the examination. A firefighter, however, may sit for the lieutenant exam — and may be directly promoted to lieutenant — without spending time as an engineer. After an exam is completed, the candidates are listed on an eligibility list in rank order, i.e., the highest scoring candidate is ranked first and the other candidates are numerically ranked according to their respective scores.

A. Antecedent Litigation and Settlement Order

These two cases present challenges to the promotion procedures employed by the CFD. These cases are not the first, however, to challenge the promotion (or hiring) procedures of the CFD. In 1973, the United States Department of Justice (DOJ) brought a civil rights action against the City claiming that the hiring and promotion practices of the CFD unlawfully discriminated against blacks and Hispanics. At the time of the suit, blacks and Hispanics comprised less than 5% of the uniformed personnel in the CFD. The DOJ moved for a preliminary injunction of the challenged practices, but the district court denied the motion. This court entered an interlocutory injunction against the City on the DOJ's discriminatory hiring claims. See Chicago Fire Fighters Union Local No. 2 v. Washington, 736 F. Supp. 923, 927 (N.D.Ill. 1990) (describing case history). Thereafter, the City determined that it could not defend successfully against these claims. Accordingly, the City entered into a consent decree in 1974 (the 1974 consent decree), which established an interim 50% minority (black and Hispanic) hiring ratio and a long-range goal requiring the City to increase significantly the minority composition of the CFD so as to reflect ultimately the minority composition of the City.

Little hiring occurred between 1974 and 1978 in the CFD. Consequently, the interim hiring ratio established in the 1974 consent decree did little to improve the minority proportion in the CFD. Indeed, by September of 1978, minorities comprised only about 9% of the CFD's uniformed personnel. The DOJ returned to the district court seeking an order to extend through 1980 the hiring goals set forth in the 1974 consent decree. On September 12, 1978, the district court granted the motion in part; the court ordered the City to fill 120 of the next 280 firefighter vacancies with minority candidates and maintain the long-range hiring goal articulated in the 1974 consent decree. R. 71 (Chicago Fire Fighters) Ex.A. In June 1979, the district court granted another DOJ motion and ordered that, if the new 1979 hiring eligibility list were used for more than two years or 500 names, 50% of all further hiring must be of minority candidates. R. 71 (Chicago Fire Fighters) Cole Aff. ¶ 8.

As for promotions, the DOJ challenged as discriminatory the City's promotion examinations for engineer, lieutenant, and captain administered in the late 1960s and early 1970s. While the district court upheld the validity of these exams under Title VII, this court granted the DOJ's motion for an injunction pending appeal enjoining the City from making any permanent promotions based on the engineer and lieutenant examinations. See United States v. City of Chicago, 573 F.2d 416, 420 (7th Cir. 1978). Once this court reached the merits on appeal, it determined that the district court applied the wrong legal standard in evaluating one aspect of the promotion examinations and that further findings of fact were necessary; therefore, this court reversed the decision of the district court and remanded for further proceedings. See id. at 416, 424-29. Furthermore, this court granted interim relief by modifying its prior injunction. We enjoined any permanent promotions based on the engineer, lieutenant, and captain promotion examinations unless the promotions were "first offered to minority candidates on the respective lists in the order of their rank thereon." Id. at 429. In 1978, prior to the completion of the remand proceedings, the City administered new promotion examinations for each rank in the CFD. The DOJ, in early 1980, informed the City that it intended to file a new suit challenging those promotion examinations because they had a severe adverse impact against minority candidates.

In response, representatives of the City and the CFD met together to assess the strengths and weaknesses of the DOJ's case. They concluded that the examinations indeed had an adverse impact on minority candidates, that the examinations had not been properly validated, particularly for use in ranking candidates, and that the efficiency ratings (one part of the candidate's final score) were racially discriminatory. In light of the strength of the DOJ's case, the litigation costs, and the recognition that the promotion scheme had treated unfairly black and Hispanic members of the CFD, the City decided to settle the suit. During the settlement agreement negotiations, the DOJ proposed a 1-to-4 minority/nonminority promotion ratio as an interim requirement for promotion to all ranks until new, validated promotion examinations could be administered and new promotion lists posted. The City agreed to the proposal, except to the extent that the 1-to-4 ratio applied to the ranks of captain and battalion chief. The City maintained that the ratio was not feasible considering the few minorities whose current rank made them eligible for promotion to captain or battalion chief. The City proposed that the 1-to-4 ratio apply to the ranks of engineer and lieutenant and that a longterm promotion goal be instituted, which would require the City to increase substantially minority participation at each rank (including captain and battalion chief), so that the minority composition of each rank would mirror the racial composition of the preceding rank.

The DOJ agreed with the City's counter-proposal, and it was drafted into a settlement agreement. The district court approved the settlement agreement and incorporated it into an Order of Settlement entered March 31, 1980 (1980 settlement order). It states that

defendants, as a long range goal, shall seek to promote black and Hispanic persons in sufficient numbers so as to increase substantially the minority composition in each of the promotional ranks [of the CFD] with the objective that each such rank become more representative of the racial and ethnic composition of the rank or ranks from which promotion to it are made.

R. 25 (Billish) Ex.I at 2-3. The 1980 settlement order further required the City to employ a 1-to-4 minority/nonminority promotion ratio for any promotions beyond number 213 on the 1979 engineer eligibility list and/or number 113 on the 1979 lieutenants list. If necessary to meet the 1-to-4 ratio, the order instructed that "the 1979 Engineer and Lieutenant Eligibility List shall be expanded by reducing the cut off score from 70." Id. at 3. The City also was required to compile new promotion eligibility lists consistent with the long-range goals mentioned above. The new lists had to be designed to reduce or eliminate any adverse impact on black or Hispanic candidates for promotions and had to ensure against unlawful discrimination against those candidates. Finally, the City was required to provide appropriate training to those seeking promotion.

B. Collective Bargaining Agreement

In 1980 the City and the Chicago Fire Fighters Union (Union) entered into their first collective bargaining agreement (CBA). R. 25 (Billish) Ex.D. Subsequently, the CBA was codified as a city ordinance. It provides that, in filling promotion vacancies, promotions will "customarily" be made in rank order. The CBA also contains an affirmative action provision. R. 25 (Billish) Ex.D at 64-67. Under the CBA, the City is required to comply with agreements reached and decrees issued in court actions affecting hiring and promotion, to create hiring programs with the goal of achieving "in the shortest possible time a total force in which approximately thirty percent shall be Black and Fifteen percent Hispanic," and to provide promotion policies with the goal of including "Black and Hispanic personnel in all categories and all ranks in order to reach as quickly as is reasonably possible a level as close to 45% as is reasonably achievable." R. 25 (Billish) Ex.A at 77. This same language has been in every CBA since 1980. R. 71 (Chicago Fire Fighters) Stensland Aff. ¶ 12.

C. Summary of Remaining Relevant Facts

The facts underlying the Billish case occurred before those in Chicago Fire Fighters. On the other hand, from an analytical perspective, it is more useful to deal with the merits of Chicago Fire Fighters before turning to Billish. To assist the reader, we shall set forth chronologically, at this point, a summary of the factual circumstances underlying each case. A more detailed rendition of the facts pertinent to each case will be presented with our analysis.

Following the 1980 consent decree, promotions were made from the 1979 eligibility lists for each rank. In August 1983, the new Fire Commissioner, Louis G. Galante, realized that these lists were getting stale, so he ordered the preparation of a new set of promotion examinations for each rank. These exams were prepared and given in ascending order: engineer first, lieutenant second, captain third, and battalion chief last. The first examination, the engineer exam, was not administered until 1985. The engineer eligibility list was posted in 1986. Next, a lieutenant exam was administered, and a lieutenant eligibility list was posted. In the fall of 1986, the first part of a two-part captain examination was given. In December 1986, Commissioner Galante determined, according to the City, that he had an operations need for additional fire captains. Because the captain examination was not complete, and thus a new captain eligibility list was far from ready, Commissioner Galante looked to the 1979 list to make these promotions. On December 25, 1986, he ordered the promotion of eighteen lieutenants to captain (and, at the same time, ordered promotions in the other ranks), effective January 16, 1987. The eighteen promotions were made in rank order from the 1979 eligibility list, and each of the promotees was white. Soon thereafter, Commissioner Galante inquired of the City Personnel Department whether there were any black or Hispanic lieutenants remaining on the 1979 list. He was told there were two minority lieutenants on the list but, because both had scored below the cut-off score of 70, that cut-off score would have to be lowered before they could be eligible for promotion to captain. Commissioner Galante asked the Personnel Department to lower the passing score. The Personnel Department determined that the scores and additional experience of these two lieutenants were sufficient to qualify them for promotion and, on this basis, certified them for promotion on January 12. On January 13, Commissioner Galante ordered their promotion to captain, out of rank order, effective the same date the December promotions were to become effective, January 16. By May 1987, the second half of the captain examination had been administered but the new captain eligibility list had not yet been drawn up. Commissioner Galante considered promoting five additional captains, but he decided to wait until the new captain eligibility list was ready.

In June and July of 1987, in two separate letters, Commissioner of Personnel Jesse E. Hoskins informed Fire Commissioner Galante that promotions to engineer and captain "should be made on an affirmative action basis, with the goal that 20% of the persons promoted to Engineer would be Black and an additional 5% of those promoted would be Hispanic." R. 71 (Chicago Fire Fighters) Ex.E at 1, Ex.F at 1. In August 1987, Commissioner Galante issued an order to promote fifty-six firefighters to engineer from the new engineer eligibility list. The first forty-eight, including seven black firefighters and two Hispanic firefighters, were promoted in rank order. However, the last eight were black firefighters who were promoted out of rank order. R. 71 (Chicago Fire Fighters) Ex.L, Ex.N. In October 1987, the new captain eligibility list was posted and Commissioner Galante promoted twenty-four lieutenants to captain. The first twenty-three were promoted in rank order, including five black lieutenants, but the twenty-fourth lieutenant — a Hispanic — was promoted out of rank order. R. 71 (Chicago Fire Fighters) Ex.M, Ex.N.

The Billish suit was brought by nine white fire lieutenants who were next on the 1979 captain eligibility list when Commissioner Galante lowered the cut-off score to allow the two minority lieutenants to be promoted in January 1987. The Billish plaintiffs challenge the lowering of the cutoff score and the subsequent promotions, as well as Commissioner Galante's refusal to draw further from the 1979 captain eligibility list. The Chicago Fire Fighters suit was brought by the Union and twenty-two white firefighters and lieutenants who were passed over in the affirmative action promotions to engineer and captain in August and October 1987. In both cases, plaintiffs sued under 42 U.S.C. § 1983 for violations of their rights under the Equal Protection Clause. In addition, the Billish plaintiffs asserted a violation of their due process rights.

II [18] ANALYSIS

A. Applicable Standards

Both district courts granted the City's motions for summary judgment on the equal protection claims. The Billish court also granted the City's motion for summary judgment on the plaintiffs' due process claim. We review de novo a district court's grant of summary judgment. Doe v. Allied-Signal Inc., 925 F.2d 1007, 1008 (7th Cir. 1991). Our task is to determine whether the record reveals that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law. Fed.R.Civ.P. 56(c); see Adickes v. S.H. Kress Co., 398 U.S. 144, 159, 90 S.Ct. 1598, 1609, 26 L.Ed.2d 142 (1970). "A motion for summary judgment is not an appropriate occasion for weighing the evidence; rather, the inquiry is limited to determining if there is a genuine issue for trial." Lohorn v. Michal, 913 F.2d 327, 331 (7th Cir. 1990); see Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249, 106 S.Ct. 2505, 2511, 91 L.Ed.2d 202 (1986). We "`must view the record and all inference drawn from it in the light most favorable to the party opposing the motion.'" Lohorn, 913 F.2d at 331 (quoting Holland v. Jefferson Nat'l Life Ins. Co., 883 F.2d 1307, 1312 (7th Cir. 1989)).

Both the Billish and Chicago Fire Fighters plaintiffs charge that the City's affirmative action promotions violated the Equal Protection Clause of the Fourteenth Amendment. The Equal Protection Clause provides that "[n]o State shall . . . deny to any person within its jurisdiction the equal protection of the laws." U.S. Const. amend. XIV, § 1. In an equal protection challenge, affirmative action programs are subject to strict scrutiny analysis. SeeCity of Richmond v. J.A. Croson Co., 488 U.S. 469, 493-94, 520, 109 S.Ct. 706, 720-22, 735, 102 L.Ed.2d 854 (1989). Strict scrutiny is used to determine "what classifications are `benign' or `remedial' and what classifications are in fact motivated by illegitimate notions of racial inferiority or simple racial politics." Id. at 493, 109 S.Ct. at 720-21. Under strict scrutiny, affirmative action must be justified by a compelling governmental interest and must be narrowly tailored to serve that interest. Id. at 505-08, 109 S.Ct. at 727-29; see Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 274, 106 S.Ct. 1842, 1847, 90 L.Ed.2d 260 (1986) (plurality). We address both of these requirements in relation to the issues raised by each case.

B. Chicago Fire Fighters Union Local No. 2 v. Daley (No. 90-2182) [23] 1. Facts

The Chicago Fire Fighters case involves promotions from the 1986 engineer eligibility list and the 1987 captain eligibility list.

During 1985, the City, in cooperation with the CFD and the Personnel Department, administered a two-part engineer examination consisting of a written test and a simulator (or hands-on) test. All firefighters were eligible to take the written exam. When the written part of the exam was scored, the City realized that the exam had a severe adverse impact on minorities. Therefore, to minimize the effect of the adverse impact, the Personnel Department standardized the scores. All firefighters who passed the written test qualified to take the simulator portion of the examination, in which they operated specially designed pumps similar to those used by the CFD. The simulator portion of the exam accounted for 90% of the total score; seniority comprised the other 10%. In February 1986, the City posted the new engineer eligibility list.

In June and December of 1986, the new engineer eligibility list was used to promote two sets of firefighters to engineer. In June 1987, Commissioner of Personnel Hoskins informed Fire Commissioner Galante that the third set of engineer promotions from the new list "should be made on an affirmative action basis, with the goal that 20% of the persons promoted to Engineer would be Black and an additional 5% of those promoted would be Hispanic." R. 71 (Chicago Fire Fighters) Ex.E at 1. Commissioner Hoskins stated that the affirmative action was necessary to comply with the affirmative action directives of the 1980 settlement order and the current CBA. Moreover, the affirmative action was necessary to "remedy the effects of past discrimination." Id. at 2. Commissioner Hoskins also indicated that failure to promote on an affirmative action basis would violate Title VII because the engineer promotion procedure, when used as a ranking device, had an adverse impact on minorities. Accordingly, in August 1987, Commissioner Galante issued an order that eventually promoted fifty-six firefighters to the rank of engineer. The last eight firefighters — all black — were promoted out of rank order.

The City administered a two-part captain examination in the fall of 1986 and the spring of 1987. This examination consisted of a written test and oral interview. The written portion accounted for 65% of a candidate's score, the oral interview 25%, and seniority 10% of the total score. Because, in the oral interview, some administrators gave lower average scores or narrower ranges of scores than others, the Personnel Department standardized these scores to neutralize the differences. The passing score was set at 70. On July 13, 1987, the City canceled the 1979 captain eligibility list and made the new captain eligibility list effective.

On July 9, 1987, Commissioner Hoskins notified Commissioner Galante that promotions to captain should also be made with the goal of promoting 20% black and 5% Hispanic. Thus, in October 1987, Commissioner Galante issued on order promoting twenty-four lieutenants to the rank of captain. The first twenty-three were promoted in rank order, the final lieutenant — a Hispanic — was promoted out of rank order.

The Union and certain nonminority members of the CFD holding the rank of firefighter or lieutenant (collectively " Chicago Fire Fighters plaintiffs" or "plaintiffs") brought a two-count complaint against the City under 42 U.S.C. § 1983 alleging that the out-of-rank promotions subjected them to discrimination in violation of the Equal Protection Clause. In Count I, the firefighter plaintiffs maintained that the promotions prevented them from being promoted to engineer. In Count II, the lieutenant plaintiffs contended that the promotions prevented them from being promoted to captain. The City moved for summary judgment.

2. Holding of the district court

The court established the framework for its analysis by recognizing that, under City of Richmond v. J.A. Croson Co., 488 U.S. 469, 109 S.Ct. 706, 102 L.Ed.2d 854 (1989), a court must apply strict scrutiny to an affirmative action policy to determine whether it violates the Equal Protection Clause. "To survive this heightened scrutiny," the court noted, "the policy must be justified by a compelling governmental interest and must be narrowly tailored to serve that interest." Chicago Fire Fighters Union Local No. 2 v. Washington, 736 F. Supp. 923, 926 (N.D.Ill. 1990).

The court first addressed the compelling governmental interest component of the strict scrutiny analysis. The court agreed with the City that there would be a compelling interest "in remedying past (and present) racial discrimination in making promotional decisions." Id. While a "`contemporaneous or antecedent finding of past discrimination by a court or other competent body'" is not a prerequisite to a public employer's voluntary affirmative action program, id. (quoting Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 289, 106 S.Ct. 1842, 1855, 90 L.Ed.2d 260 (1986) (O'Connor, J., concurring)), the district court stated that there must be some showing of prior discrimination by the governmental entity involved. Moreover, the court observed that, as a factual matter, a governmental body engaging in affirmative action must show a "`strong basis in evidence for its conclusion that remedial action was necessary.'" 736 F. Supp. at 926 (quoting Croson, 488 U.S. at 500, 109 S.Ct. at 724). Relying on Wygant, 476 U.S. at 277-78, 293, 106 S.Ct. at 1848-49, 1857, the court noted that, while the City had to establish a strong basis for the propriety of its remedial action, the Chicago Fire Fighters plaintiffs retained the ultimate burden of demonstrating that the out-of-rank promotions were unconstitutional. 736 F. Supp. at 926. The court then recounted the history of the DOJ's litigation against the City and the resulting affirmative action directives contained in the 1980 settlement order, and the affirmative action goals embodied in the CBA between the City and the Union. See id. at 927-29. The court concluded that evidence of prior discrimination was provided by our rulings in United States v. City of Chicago, 573 F.2d 416 (7th Cir. 1978). See 736 F. Supp. at 928. The court also discussed the disparities between the minority representation in the engineer and captain ranks and the representation in the ranks immediately below them, and the discriminatory adverse impact of the engineer examination on minorities. See id. at 929-31. Based on this evidence, the court concluded that "there is no genuine issue of material fact that the City had a strong basis in evidence for its conclusion that its affirmative actions promotions were necessary." Id. at 931-32.

The court next turned to the issue of whether the affirmative action program was narrowly tailored to accomplish its purpose. Relying on United States v. Paradise, 480 U.S. 149, 107 S.Ct. 1053, 94 L.Ed.2d 203 (1987), the district court identified several factors relevant to this inquiry: the necessity of the relief and availability of alternative remedies; the flexibility and duration of the remedy; the relationship of the numerical goals to the relevant labor market; and the impact of the relief on third parties. Id. at 932. The court concluded that each of the factors favored the affirmative action remedy implemented by the City. Therefore, the court held that "there is no genuine issue of material fact that the City's affirmative action goals at issue here were narrowly tailored to their remedial purpose." 736 F. Supp. at 933. Having found that the affirmative action program served a compelling governmental interest and that the program was narrowly tailored to serve that interest, the court granted the City's motion for summary judgment.

3. Standing of plaintiffs who were not passed over

Initially, we consider the Chicago Fire Fighters plaintiffs' contention that the district court erred in granting the City's motion to dismiss for lack of standing thirteen of the twenty firefighter plaintiffs and one of the two lieutenant plaintiffs. In reviewing a motion to dismiss for want of standing, we "must accept as true all material allegations of the complaint, and must construe the complaint in favor of the complaining party." Warth v. Seldin, 422 U.S. 490, 501, 95 S.Ct. 2197, 2206, 45 L.Ed.2d 343 (1975). "In essence the question of standing is whether the litigant is entitled to have the court decide the merits of the dispute or of particular issues." Id. at 498, 95 S.Ct. at 2205. In Valley Forge Christian College v. Americans United for Separation of Church and State, 454 U.S. 464, 102 S.Ct. 752, 70 L.Ed.2d 700 (1982), the Supreme Court articulated the minimum constitutional requirements for standing:

[A]t an irreducible minimum, Art. III requires the party who invokes the court's authority to "show that he personally has suffered some actual or threatened injury as a result of the putatively illegal conduct of the defendant," and that the injury "fairly can be traced to the challenged action" and "is likely to be redressed by a favorable decision."

Id. at 472, 102 S.Ct. at 758 (quoting Gladestone Realtors v. Village of Bellwood, 441 U.S. 91, 99, 99 S.Ct. 1601, 1607-08, 60 L.Ed.2d 66 (1979) and Simon v. Eastern Kentucky Welfare Rights Org., 426 U.S. 26, 38, 41, 96 S.Ct. 1917, 1924, 1926, 48 L.Ed.2d 450 (1976)).

In the City's view, the affirmative action did not cause the dismissed plaintiffs any injury. The City argued before the district court that, even had the promotions been made in strict rank order, those plaintiffs would not have been offered promotions. See Memorandum Opinion and Order of January 20, 1989, (R. 55 (Chicago Fire Fighters)) at 5. The district court began by noting that the City did not bypass any of the thirteen firefighter plaintiffs or the lieutenant plaintiff when the City certified eight black firefighters and one Hispanic lieutenant for promotion on the basis of race. Id. at 7-8. Furthermore, the court declined to accept the plaintiffs' argument that they sufficiently alleged standing because they were currently threatened by a recurrence of the out-of-rank-order promotions. The court reasoned that, before the challenged promotions, promotions were based on strict rank order, and the City had not conceded nor had the plaintiffs alleged that the City would make future promotions out of rank order. Id. at 9. Therefore, the court granted the City's motion.

The district court's ruling was correct. The plaintiffs in question suffered no direct injury from the non-rank-order promotions. As the court noted, even had the City made the promotions in strict rank order, those fourteen plaintiffs would not have been promoted. Cf. Donaghy v. City of Omaha, 933 F.2d 1448, 1455 (8th Cir. 1991) (plaintiff would not have had a "distinct and palpable injury" if he would not have been promoted had the promotions been made in strict rank order), cert. denied, ___ U.S. ___, 112 S.Ct. 938, 117 L.Ed.2d 109 (1992). The plaintiffs contend, however, that, because the City has expressed its willingness to continue to pursue affirmative action promotions, they have standing based on a threat of future injury. To satisfy the standing requirement under this theory:

The plaintiff must show that he . . . "is immediately in danger of sustaining some direct injury" as the result of the challenged official conduct and the . . . threat of injury must be both "real and immediate," not "conjectural" or "hypothetical."

City of Los Angeles v. Lyons, 461 U.S. 95, 101-02, 103 S.Ct. 1660, 1665, 75 L.Ed.2d 675 (1983) (quoting Golden v. Zwickler, 394 U.S. 103, 109-10, 89 S.Ct. 956, 960-61, 22 L.Ed.2d 113 (1969)); see O'Shea v. Littleton, 414 U.S. 488, 497, 94 S.Ct. 669, 676-77, 38 L.Ed.2d 674 (1974) (plaintiff must base standing on more than speculation and conjecture that harm will occur in the future). The plaintiffs' allegation that they will be injured in the future is speculative. Even if the City pursues affirmative action, the vast majority of the promotions to engineer and captain will be in rank order. Given the plaintiffs' positions on the eligibility lists, they will most likely be promoted in the course of the next promotion. Therefore, we affirm the district court's dismissal of certain plaintiffs for lack of standing.

4. Equal protection analysis

a. compelling governmental interest

(i) prior litigation and the 1980 settlement order

The City entered into the 1974 consent decree concerning hiring and the 1980 settlement order concerning promotion only after extensive litigation with the DOJ. The City argues that the litigation history and the 1980 settlement order provide a strong basis for concluding that affirmative action promotions to the rank of engineer and captain were necessary to remedy prior discrimination. We believe that the entire litigation history of the CFD on the issue of racial discrimination in its ranks provides strong support for the City's position.

We shall not repeat here the factual basis of the initial suit. See supra, pp. 1280-81. A brief summary will make it clear that the City of Chicago's assessment that its policies violated the Constitution was a realistic one. The DOJ brought suit against the City in 1973 to challenge as discriminatory the hiring and promotion procedures in the CFD. On interlocutory appeal of the district court's denial of the DOJ's motion for preliminary injunction, this court temporarily enjoined the City from further hiring pending resolution of the appeal. Soon thereafter, in 1974, the City determined that it could not defend successfully its hiring practices and thus agreed to abide by a consent decree which established a temporary minority hiring ratio and a long range goal of substantially increasing minority representation in the CFD. This action suggests that the DOJ had established a discrimination case likely to succeed on the merits. Thus, the 1974 consent decree provides strong evidence of prior discriminatory hiring practices in the CFD.

In 1978, reviewing a later DOJ challenge to the CFD's promotion practices, this court enjoined promotions stemming from the 1969 engineer, 1970 lieutenant, and 1973 captain examinations unless the promotions were first offered to minorities on the eligibility lists. See United States v. City of Chicago, 573 F.2d 416, 429 (7th Cir. 1978). During remand proceedings in 1978, the City administered a new set of promotion examinations for each rank in the CFD. After the DOJ announced its intention to challenge the adverse impact of these new examinations, the City agreed to enter into the 1980 settlement order. This order required the City to create and implement long-range promotion goals to increase substantially the minority composition of each rank of the CFD so that each rank would represent more closely the composition of the preceding rank. R. 71 (Chicago Fire Fighters) Ex.B at 2-3. More specifically, the settlement order required the City to promote one black or Hispanic for every four whites promoted from the 1979 engineer and lieutenant eligibility lists until new, validated promotion examinations for these ranks were administered. R. 71 (Chicago Fire Fighters) Ex.B at 3. While the captain rank was not subject to the 1-to-4 minority/nonminority promotion ratio, the negotiation history of the settlement order reveals that this difference was due to the absence of enough blacks and Hispanics on the 1979 captain eligibility list to support such a ratio. R. 71 (Chicago Fire Fighters) Cole.Aff. ¶ 16. The 1980 settlement order, like a consent decree, was subject to judicial review and approval. "While the entry of an affirmative action consent decree does not guarantee that the decree serves a remedial purpose or is narrowly tailored, the heightened judicial oversight inherent in a properly entered decree helps attain that end." Donaghy, 933 F.2d at 1459; see also Stuart v. Roache, 951 F.2d 446 (1st Cir. 1991) (consent decree requiring affirmative action promotions in police force upheld under strict scrutiny), petition for cert. filed, No. 91-1516 (Mar. 20, 1992).

Despite the specific hiring ratios and long-term goals, however, by 1987, the percentage of minority captains (3.6%) had actually decreased from the percentage in 1979 (6.0%), and the percentage of minority engineers (11.0%) was still less than half the percentage of minority firefighters (29.2%). R. 71 (Chicago Fire Fighters) Cole Aff. ¶ 12, Stensland Aff. ¶ 19. Faced with this situation in 1987, the City concluded that affirmative action was necessary to comply with the 1980 settlement order. The City's compliance with the 1980 settlement order was "`supported not only by the governmental interest in eradicating [the City's] discriminatory practices, it [was] also supported by the societal interest in compliance with the judgment of federal courts.'" United States v. Paradise, 480 U.S. 149, 170, 107 S.Ct. 1053, 1066, 94 L.Ed.2d 203 (1986) (quoting Local 28 of the Sheet Metal Workers' Int'l Ass'n v. EEOC, 478 U.S. 421, 480, 106 S.Ct. 3019, 3052, 92 L.Ed.2d 344 (1986)); see also Ustrak v. Fairman, 781 F.2d 573, 577 (7th Cir.) (consent decree establishing preference for minorities alone not enough to "carry the day" in affirmative action case), cert. denied, 479 U.S. 824, 107 S.Ct. 95, 93 L.Ed.2d 47 (1986). On this basis, we agree with the district court that "although there exist no formal findings of past discrimination in the CFD, the litigation history of the CFD's past hiring and promotion practices offers at least some evidence of past discrimination." Washington, 736 F. Supp. at 929.

The Chicago Fire Fighters plaintiffs contend that the litigation history provides no evidence of past discrimination because no formal finding of discrimination was made during the litigation. No such finding was necessary. As noted previously, see supra p. 1278, in Wygant, six justices held that a government body may engage in affirmative action without a court or other government body formally finding that the government body previously discriminated. See 476 U.S. at 277-78, 106 S.Ct. at 1848-49 (plurality, including O'Connor, J.); id. at 287, 289-92, 106 S.Ct. at 1853-54, 1854-56 (O'Connor, J., concurring in part and concurring in the judgment); id. at 304-05, 312 n. 7, 106 S.Ct. at 1862-63, 1867 n. 7 (Marshall, J., dissenting); id. at 313, 106 S.Ct. at 1867-68 (Stevens, J., dissenting); see also Donaghy, 933 F.2d at 1459 ("Formal findings of past discrimination need not precede a public employer's adoption of a remedial affirmative action plan."). Indeed, as Justice O'Connor has warned:

The imposition of a requirement that public employers make findings that they have engaged in illegal discrimination before they engage in affirmative action programs would severely undermine public employers' incentive to meet voluntarily their civil rights obligations. This result would clearly be at odds with this Court's and Congress' consistent emphasis on "the value of voluntary efforts to further the objectives of the law."

Wygant, 476 U.S. at 290, 106 S.Ct. at 1855 (citations omitted) (quoting Regents of the University of California v. Bakke, 438 U.S. 265, 364, 98 S.Ct. 2733, 2785-86, 57 L.Ed.2d 750 (1978)). Instead, both Wygant and Croson teach that the touchstone for affirmative action is whether the government entity had a "strong basis" to conclude that remedial action was necessary. See Croson, 488 U.S. at 500, 109 S.Ct. at 724; Wygant, 476 U.S. at 277-78, 106 S.Ct. at 1848-49 (plurality), at 286, 106 S.Ct. at 1853 (O'Connor, J., concurring in part and concurring in the judgment); see also Davis v. City County of San Francisco, 890 F.2d 1438, 1446 (9th Cir. 1989), cert. denied, ___ U.S. ___, 111 S.Ct. 248, 112 L.Ed.2d 206 (1990); Howard v. McLucas, 871 F.2d 1000, 1007-08 (11th Cir.), cert. denied, 493 U.S. 1002, 110 S.Ct. 560, 107 L.Ed.2d 555 (1989).

The Chicago Fire Fighters plaintiffs further contend that the City's reliance on the litigation history as a basis for its compelling interest is a "mere afterthought." The plaintiffs concede that the City took the 1980 settlement order into account in 1987 when it decided to engage in affirmative action, but submit that the City never mentioned the 1970's litigation until it became embroiled in this lawsuit. The record belies this argument. In Personnel Commissioner Hoskins' June 16 and July 9, 1987, letters to Commissioner Galante recommending affirmative action in the ranks of captain and engineer, Commissioner Hoskins cited the litigation history among the reasons compelling affirmative action. Thus, the litigation history is not "mere afterthought," but contributed to the City's basis for concluding that affirmative action was necessary.

(ii) the collective bargaining agreement

The City also maintains that the CBA between the City and the Union necessitated the remedial action taken in this case. As we discussed, the CBA requires the City to comply with affirmative action decrees and agreements reached in court actions and to promote with the goal of including minorities at all ranks to "reach as quickly as is reasonably possible a level as close to 45% as is reasonably achievable." R. 82 (Chicago Fire Fighters) Ex.3 at 87. The district court acknowledged that "[t]he existence of an affirmative action goal in a [CBA] alone may not be sufficient" to provide the City with a strong basis for concluding that non-rank-order promotions were necessary. 736 F. Supp. at 929. Nevertheless, as the district court concluded, while the affirmative action provisions in the CBA may not prove the existence of past discrimination, they certainly corroborate it.

(iii) minority underrepresentation statistics

Next, the City maintains that statistics showing minority underrepresentation in the CFD ranks provided strong evidence that past discrimination still had a strong effect in the CFD. "[S]tatistics can be an important source of proof in employment discrimination cases. . . ." Hazelwood School Dist. v. United States, 433 U.S. 299, 307, 97 S.Ct. 2736, 2741, 53 L.Ed.2d 768 (1977). The district court noted that "a statistical imbalance sufficient for a Title VII prima facie case would satisfy the constitutional requirement that the City have a `strong basis' for believing that remedial action was required." Chicago Fire Fighters, 736 F. Supp. at 929. Indeed, the district court cited the City's statistics as "the strongest evidence of past discrimination in the CFD." Chicago Fire Fighters, 736 F. Supp. at 929.

As noted, a prerequisite for promotion to engineer or captain is a candidate's position as firefighter or lieutenant. Thus, for purposes of evaluating underrepresentation in the engineer and captain ranks, the appropriate statistical comparison is between the percentage of blacks and Hispanics in those ranks and the percentage of blacks and Hispanics in the firefighter and lieutenant ranks. "The `proper comparison [is] between the racial composition of [the at-issue jobs] and the racial composition of the qualified . . . population in the relevant labor market.'" Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 650, 109 S.Ct. 2115, 2121, 104 L.Ed.2d 733 (1989) (quoting Hazelwood, 433 U.S. at 308 n. 13, 97 S.Ct. at 2742 n. 13); EEOC v.Chicago Miniature Lamp Works, 947 F.2d 292, 302 (7th Cir. 1991); see Croson, 488 U.S. at 501-02, 109 S.Ct. at 725-26; Janowiak v. Corporate City of South Bend, 836 F.2d 1034, 1039-40 (7th Cir. 1987), cert. denied, 489 U.S. 1051, 109 S.Ct. 1310, 103 L.Ed.2d 579 (1989); see also Mayor of Philadelphia v. Educ. Equality League, 415 U.S. 605, 620, 94 S.Ct. 1323, 1333, 39 L.Ed.2d 630 (1974). A significant statistical disparity between the ranks would support an inference of past racial discrimination. See Wards Cove, 490 U.S. at 650, 109 S.Ct. at 2121; Croson, 488 U.S. at 509, 109 S.Ct. at 729 (plurality); Bazemore v. Friday, 478 U.S. 385, 398, 106 S.Ct. 3000, 3007-08, 92 L.Ed.2d 315 (1986); International Bhd. of Teamsters v. United States, 431 U.S. 324, 337-39, 97 S.Ct. 1843, 1855-56, 52 L.Ed.2d 396 (1977). In June 1987, when Commissioner Hoskins requested the affirmative action promotions, 11% of the engineers were minorities, compared to 29.2% of the firefighters, and 3.6% of the captains were minorities, compared to 13.6% of the lieutenants. See Chicago Fire Fighters, 736 F. Supp. at 930. As the district court observed, if the difference between the expected percentage of minorities and the observed percentage of minorities in the relevant labor pool is more than two or three standard deviations, then the hypothesis that promotions were made without regard to race would be suspect. The City presented uncontradicted evidence that the difference between the expected percentage of minority engineers and the actual percentage of minority engineers was 8.7 standard deviations, and the difference between the expected percentage of minority captains and the actual percentage of minority captains was 3.96 standard deviations. R. 80 (Chicago Fire Fighters) Supplemental Aff. ¶¶ 2, 4. "[A] range of four to five standard deviations corresponds to a probability range of 1 chance in 15,786 to 1 chance in 1,742,160." Ottaviani v. State Univ., 875 F.2d 365, 372 n. 7 (2d Cir. 1989), cert. denied, 493 U.S. 1021, 110 S.Ct. 721, 107 L.Ed.2d 740 (1990). Thus, the odds that the disparity among the ranks occurred by chance are in the neighborhood of 1 in 10,000. These numbers reveal disparities "well beyond the possible effect of chance," and therefore support an inference of past discrimination. R. 80 (Chicago Fire Fighters) Supplemental Aff. ¶ 2.

The Chicago Fire Fighters plaintiffs present several challenges to the use of statistical comparison as evidence of past discrimination. They first argue that the City cannot rely on the statistical comparisons between the labor pools because the City did not consider the statistics when it decided to engage in affirmative action. Because the City did not fashion the remedial relief based on such statistics, in the plaintiffs' view, the City cannot now proffer those statistics as part of its "strong basis" for employing remedial relief. This argument fares no better than the argument that the City's reliance on the litigation history was a "mere afterthought." See supra, p. 1286. Commissioner Hoskins in fact relied on the minority underrepresentation statistics as one justification for the affirmative action in his letters to Commissioner Galante. R. 71 (Chicago Fire Fighters) Ex.E at 2, Ex.F at 2.

The plaintiffs also suggest that in Cox v. City of Chicago, 868 F.2d 217, 222 (7th Cir. 1989), this court concluded that underrepresentation statistics are no longer relevant to disparate impact analysis. They characterize the inquiry here as a disparate impact case and argue that underrepresentation statistics are thus irrelevant to this case. This is not a disparate impact case; the suit brought by the plaintiffs is a section 1983 action based on the constitutional right to equal protection. And, at this point in the case, we are dealing with the City's defense that it had a "strong basis" for its determination that affirmative action was necessary to remedy prior discrimination. This inquiry — the City's self-assessment of the relevant situation before it — could be analogized to disparate treatment analysis just as well as to disparate impact analysis; the statistics suggest a pattern or practice of past intentional discrimination as much as they suggest that the past promotion procedures had an adverse impact on minorities. In reality, the City considered the statistics for neither purpose but, rather, as evidence that the hiring and promotion practices that the DOJ challenged in the 1970s and early 1980s had not yet been remedied. Thus, the City's reliance on underrepresentation statistics as part of its "strong basis" for concluding that remedial action was necessary is not inconsistent with Cox.

The plaintiffs next argue that the district court should have narrowed the comparative labor pools by considering only the persons in each preceding rank who are eligible for promotion. Particularly, plaintiffs suggest that the relevant labor pools are those firefighters who actually took the 1985 engineer examination and those lieutenants who actually took the 1986-87 captain examination, rather than the complete firefighter and lieutenant ranks. In support of this argument, plaintiffs again draw an analogy to Cox, in which statistics were used to illustrate the discriminatory effect of the City's promotion mechanism, specifically the "captain rule," which states that only candidates who have passed the captain examination are qualified to take the battalion chief exam. 868 F.2d at 219-20. In Cox, this court held that the proper comparative labor pool was not the pool of all lieutenants but, rather, the pool of lieutenants who had sufficient experience to be eligible for promotion to battalion chief. Id. at 221-22. However, as the City notes, Cox is not analogous on this point because the comparisons there were made to illustrate the disparate impact of the promotion hurdle which separated the two groups. Here, as just discussed, the comparisons are made to show that the current upper ranks in the CFD remain affected by prior discrimination. More fundamentally, evidence in this case suggests that the minority representation among those eligible for promotion to engineer and captain was diluted by prior discrimination; therefore, the comparison between complete ranks is a more appropriate measure of past discrimination.

(iv) adverse impact of the 1985 engineer examination

Finally, the City contends that the adverse impact of the 1985 engineer examination contributed to its compelling governmental interest in affirmative action. After comparing the passing rates for white, black, and Hispanic firefighters, the City concluded that the examination taken as a whole — including the written exam as well as the practical exam — had an adverse impact. In drawing this conclusion, the City relied upon the Uniform Guidelines on Employee Selection Procedures, which contain a brightline "four-fifths" test for determining whether a scored examination has an adverse impact when used as a selection procedure:

A selection rate for any race . . . which is less than four-fifths (4/5) (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate will generally not be regarded by Federal enforcement agencies as evidence of adverse impact.

29 C.F.R. § 1607.4(D) (1978).

The engineer examination consisted of two parts. The first part was a written exam, which was administered on January 19, 1985. The raw scores of this test revealed that it had a severe adverse impact under the four-fifths rule. To counteract this, the City "standardized" the scores by converting the mean scores and standard deviations of the white, black, and Hispanic groups of test-takers to a single mean and standard deviation. The City then set a passing score which allowed the top 617 test-takers to pass. This procedure lessened the adverse impact of the written test.

Those who passed the written examination were eligible to take the practical examination. A few chose not to. The results of the practical examination, when considered separately from the results of the written test, did not reveal an adverse impact. The City drew up a rank-order eligibility list, calculating rank scores from the practical examination scores (90%) and seniority (10%). On June 5, 1986, the CFD promoted the fifty top-ranked firefighters to engineer. Breaking down these fifty promotions into racial groups and comparing the numbers with the numbers of persons who took the practical exam, the City noted that the success ratio for blacks and Hispanics was more than four-fifths that of the success ratio for whites, and thus the rank-order promotion did not have an adverse impact. On December 25, 1986, the CFD promoted thirty-three more firefighters to the rank of engineer. This time, however, when the numbers of persons of each race promoted were compared with the numbers who took the practical, an adverse impact was evident. But, if the two groups of promoted firefighters were pooled together for comparison with the group that took the practical, there was no adverse impact. On this basis, the rank-order promotion was approved.

Nevertheless, the City realized that, beginning with the third set of promotions, the use of the eligibility list for rank-order promotions would have an adverse impact: when either the third group of promotions or the pool of all three groups of promotions were compared with the group of firefighters who took the practical examination, the numbers revealed that the practical examination had an adverse impact when used as a rank-order device. Furthermore, when the written and practical portions of the 1985 examination were considered together, the calculations revealed that the overall examination had an adverse impact. This adverse impact of the overall examination means that even if the eligibility list were used as a pass/fail promotion device — for example, if names from the list were drawn randomly for promotion orders — the use of the list would have an adverse impact. The City thus argued that, with this information available, it had reason to believe that promoting from the 1986 engineer eligibility list in strict rank order would be discriminatory and could subject it to Title VII liability. The district court agreed. See Chicago Fire Fighters, 736 F. Supp. at 931.

The plaintiffs concede that the statistics reveal evidence of adverse impact. Nevertheless, they argue that such evidence only begins the inquiry. The examination cannot be declared discriminatory unless it is not job-related, and the plaintiffs maintain that the simulator (or hands on) portion of the exam was clearly job-related. In Cox we noted that, once a device is found to have had an adverse impact, the focus shifts to whether the challenged promotion device is job-related. 868 F.2d at 222 n. 3. As one court commented, "a prima facie case is established by a showing that an examination has an adverse racial impact on minority candidates. Thereafter, legitimate, job-related explanations for differences in score distributions become relevant to rebut the prima facie showing of adverse racial impact." Bushey v. New York State Civil Serv. Comm'n, 733 F.2d 220, 224 (2d Cir. 1984), cert. denied, 469 U.S. 1117, 105 S.Ct. 803, 83 L.Ed.2d 795 (1985). In response, the City submits that neither the practical nor written parts of the examination were properly validated as job-related. The City also points to expert testimony in the record to support its contention that the examination could not have been validated. R. 71 (Chicago Fire Fighters) Barrett Aff. ¶¶ 16, 17; Chicago Fire Fighters, 736 F. Supp. at 931. Thus, the City had a basis for concluding that the 1985 engineer examination had an adverse impact on black and Hispanic firefighters, especially when used as a rank-order device. This evidence contributes to the "strong basis" which the City seeks to establish for its remedial action in this case. see Croson, 488 U.S. at 500, 109 S.Ct. at 724 (a "strong basis in evidence" supporting "remedial action" may be shown by something "approaching a prima facie case of constitutional or statutory violation").

The City presents the four grounds we have just discussed to establish that it had a "strong basis" for its conclusion that racial discrimination pervaded hiring and promotion practices at the CFD and that, consequently, it has a compelling interest in taking remedial action. The City's evidence with respect to some of the bases may be stronger, from a quantitative or qualitative point of view, than it is with respect to others. The district court was on solid ground, however, in holding that, when evaluated in the aggregate, the evidence establishes that there is no genuine issue of material fact as to whether the City had a strong basis for concluding that the personnel structure of the CFD was a product of racial discrimination.

b. narrowly tailored

In addition to having a compelling governmental interest for remedial action, the remedy chosen must be narrowly tailored to accomplish its purpose. Croson, 488 U.S. at 507, 109 S.Ct. at 728; Paradise, 480 U.S. at 171, 107 S.Ct. at 1066-67 (plurality); Wygant, 476 U.S. at 280, 106 S.Ct. at 1850 (plurality); Cygnar v. City of Chicago, 865 F.2d 827, 838 (7th Cir. 1989). The purpose of the affirmative action promotions was "to address the problem caused by the city's past discrimination in hiring and promotion: the low percentage of Black and Hispanic captains and engineers" and "to comply with the long-term goal of the consent decree." Appellee's Br. at 37. As we assess the propriety of the City's affirmative action, we are mindful that "the `tailoring' requirement is to be interpreted both carefully and flexibly, for there is no universal answer to the problem of remedying racial discrimination." Donaghy v. City of Omaha, 933 F.2d 1448, 1461 (8th Cir. 1991), cert. denied, ___ U.S. ___, 112 S.Ct. 938, 117 L.Ed.2d 109 (1992). As the Supreme Court has recently confirmed in Croson, 488 U.S. at 507, 109 S.Ct. at 728, in determining whether race-conscious remedies are narrowly tailored, we utilize several factors, including: (1) "the necessity for the relief and the efficacy of alternative remedies"; (2) "the flexibility and duration of the relief, including the availability of waiver provisions"; (3) "the relationship of the numerical goals to the relevant labor market"; and (4) "the impact of the relief on the rights of third parties." Paradise, 480 U.S. at 171, 107 S.Ct. at 1066-67 (plurality); see Stuart v. Roache, 951 F.2d 446, 454-55 (1st Cir. 1991), petition for cert. filed, No. 91-1516 (Mar. 20, 1992); Davis v. City and County of San Francisco, 890 F.2d 1438, 1447 (9th Cir. 1989), cert. denied, ___ U.S. ___, 111 S.Ct. 248, 112 L.Ed.2d 206 (1990); Howard v. McLucas, 871 F.2d 1000, 1008 (11th Cir.), cert. denied, 493 U.S. 1002, 110 S.Ct. 560, 107 L.Ed.2d 555 (1989). After reviewing the record, we believe that the district court correctly determined that "the City's policy of affirmative action promotions was narrowly tailored to serve its remedial ends." Chicago Fire Fighters, 736 F. Supp. at 932.

First, with regard to the necessity for the relief and the efficacy of alternative remedies, Personnel Commissioner Hoskins noted in his 1987 letters to Fire Commissioner Galante that the decision to pursue affirmative action promotions was arrived at "only after concluding that no other approach would permit meaningful affirmative action progress." R. 71 (Chicago Fire Fighters) Ex.E at 3, Ex.F at 3. Because promotions to the rank of engineer and captain are made from the ranks immediately beneath those ranks, some race-neutral procedures for increasing minority representation, such as aggressive recruiting, would not correct discrimination in the upper ranks. Thus, contrary to the plaintiffs' suggestion, this is not a case like Croson, in which the government entity failed to consider any alternatives to race-based relief. Cf. Croson, 488 U.S. at 507, 109 S.Ct. at 728 ("there does not appear to have been any consideration of the use of race-neutral means to increase minority business participation in city contracting").

Second, the affirmative action was flexible and limited in duration. The promotions were to be made "with the goal that 20% of the persons promoted to Engineer would be Black and an additional 5% of those promoted would be Hispanic." R. 71 (Chicago Fire Fighters) Ex.E at 1, Ex.F at 1 (emphasis added). A goal is by definition "more flexible than an unyielding quota." Chicago Fire Fighters, 736 F. Supp. at 932. Furthermore, as the district court noted, under the affirmative action promotions, no unqualified minority had to be offered a promotion. See Paradise, 480 U.S. at 177-78, 107 S.Ct. at 1069-70 (plurality) (remedy flexible because it required qualified applicants and did not require "gratuitous promotions"). This case demonstrates adherence to that principle. The plaintiffs do not challenge the qualifications of those minority candidates promoted out of rank order. The record also reflects that the affirmative action program was limited in duration. It was designed to last three years or until 350 promotions were made to engineer or 150 promotions to captain. Cf. Donaghy, 933 F.2d at 1461 (consent decree narrowly tailored in part because limited to seven years). Indeed, Commissioner Hoskins indicated that the "goals are temporary" and stated that "to insure flexibility and to guarantee that the ratios are used only so long as they are necessary and appropriate, the affirmative action promotion goals will be reevaluated on an annual basis." R. 71 (Chicago Fire Fighters) Ex.E at 4-5, Ex.F at 4-5. In short, the policies were of limited duration and were annually reviewed.

Third, with regard to the relationship of the numerical goals to the relevant labor market, at issue are the goals set forth in Commissioner Hoskins' letters to Commissioner Galante on June 16, 1987 (engineers) and July 9, 1987 (captains): "the goal that 20% of the persons promoted . . . would be Black and an additional 5% of those promoted would be Hispanic." R. 71 (Chicago Fire Fighters) Ex.E at 1, Ex.F at 1. The district court considered the relevant labor market to be the members of the ranks from which promotions are drawn: the pools of firefighters and lieutenants in 1987. However, even if we narrow the relevant labor market to the pools of engineers and lieutenants who had passed the recent eligibility tests and made the 1986 (engineer) and 1987 (captain) eligibility lists, the goals of the City were sufficiently related to the labor market. Using this more demanding methodology, of the 557 candidates ranked on the 1987 engineer eligibility list, 112 were black (20.1%), 35 were Hispanic (6.3%), and the remaining 410 were white (73.6%). The engineer promotion goals are clearly realistic in light of the racial composition of this labor market. Of the 341 candidates on the 1987 captain eligibility list, 42 were black (12.3%), 15 were Hispanic (4.4%), and the remaining 284 were white (83.3%). There is a disparity between the 12.3 percent of black lieutenants available for promotion and Commissioner Hoskins' announced goal of 20 percent black promotions. While we cannot say that this disparity constitutes the sort of "outright racial balancing" condemned in Croson, 488 U.S. at 507, 109 S.Ct. at 728, we believe that, in other circumstances, it would deserve more careful scrutiny than that given it by the district court in this case. Here, however, it is clear that the disparity made no difference in the outcome of this litigation. At issue is the October 1, 1987, promotion of twenty-four lieutenants to captain. The first twenty-three of these promotions were made in rank order, including the rank-order promotions of five black lieutenants. Only the twenty-fourth promotion, that of a Hispanic lieutenant who was ranked fortieth, was made out of rank order.

Finally, the impact of the relief on nonminorities is minimal. The affirmative action program did not bar nonminority advancement, nor did it burden unduly nonminority candidates. See Vogel v. City of Cincinnati, 959 F.2d 594, 598 (6th Cir. 1992). As will be discussed in the context of the Billish plaintiffs' due process claim, the nonminority candidates were not entitled to rank-order promotion. The CBA provides that promotions will "customarily" be made in rank order, but also requires the City to pursue affirmative action hiring and promotion goals. Arbitrators have interpreted the CBA to permit out-of-rank-order promotions. Furthermore, the plaintiffs do not refute the City's determination that, as of May 1989 (two years after the implementation of the affirmative action promotions), 73% of those promoted from the 1987 captain eligibility list were nonminorities, and 77% of those promoted from the 1986 engineer eligibility list were nonminorities. Thus, the plaintiffs' opportunities for promotion have not been denied, but postponed. Postponing a promotion opportunity for qualified nonminorities is less intrusive than denying them any future promotion opportunity. In fact, by September 1988, approximately a year after the promotions in dispute, each of the Chicago Fire Fighters plaintiffs not dismissed for lack of standing had been promoted or offered the opportunity for promotion.

Relying on Croson, the plaintiffs also contend that the City's affirmative action program could not be narrowly tailored because the beneficiaries under the program had not been identified as victims of past discrimination. In the plaintiffs' view, an individual beneficiary of an affirmative action program must have personally been a victim of the past discrimination that is being remedied. Croson faulted the City of Richmond's affirmative action plan because "there [was] no inquiry into whether or not the particular [minority business] seeking a racial preference has suffered from the effects of past discrimination by the city or prime contractors." 488 U.S. at 508, 109 S.Ct. at 729. The plaintiffs contend that the district court erred by not inquiring "whether any minority group member who would be awarded a preferential promotion to Engineer or Captain had been the victim of past discrimination." Appellants' Br. at 44-45.

After a careful review of the Croson decision, as well as opinions of this and other circuits, we cannot accept the plaintiffs' contention that Croson requires affirmative action plans to be limited to identifiable victims of the precise past discrimination that is being remedied. The Supreme Court's demand in Croson for an "inquiry into whether or not the particular [minority business] seeking a racial preference has suffered from the effects of past discrimination by the city" cannot be read without reference to the particular factual setting of the case. 488 U.S. at 508. The City of Richmond adopted an affirmative action plan that required contractors awarded city construction contracts to subcontract 30% of the dollar amount of each contract to Minority Business Enterprises (MBEs). An MBE was defined as a business from anywhere in the United States that was at least 51% owned by black, Spanish-speaking, Oriental, Indian, Eskimo, or Aleut citizens. See 488 U.S. at 469-70, 109 S.Ct. at 706-08. The Court concluded that the plan was not narrowly tailored because, in part, Richmond had never identified the prior discrimination it was attempting to remedy. MBEs were defined so broadly that the potential beneficiaries included ethnic groups that may never have been represented in the City of Richmond. Therefore, the plan was simply too broad to be remedial. See id. at 506-08, 109 S.Ct. at 727-29. The factual context of Croson thus suggests that the Court was concerned that affirmative action programs be narrowly tailored enough to remedy only proven local discrimination, rather than attempting to remedy "societal discrimination."

Similarly, the jurisprudential context in which Croson was decided confirms that the Court did not intend to overturn its prior approval of "race-conscious" relief programs that are not necessarily individual-victim conscious. It must be remembered that the Court has said specifically that affirmative action plans need not benefit only identifiable victims. In Local 93, International Association of Firefighters v. City of Cleveland, 478 U.S. 501, 516, 106 S.Ct. 3063, 3072, 92 L.Ed.2d 405 (1986), a Title VII case, the Court held that "voluntary action available to employers and unions seeking to eradicate race discrimination may include reasonable race-conscious relief that benefits individuals who were not actual victims of discrimination." This principle, expressed many times by the Supreme Court, was reiterated in Wygant v. Jackson Board of Education, 476 U.S. 267, 106 S.Ct. 1842, 90 L.Ed.2d 260 (1986), by Justice O'Connor — the author of the Croson majority opinion — when she wrote that, with respect to the tightness of the means-end fit, "the Court has forged a degree of unanimity; it is agreed that a plan need not be limited to the remedying of specific instances of identified discrimination for it to be deemed sufficiently `narrowly tailored,' or `substantially related,' to the correction of prior discrimination by the state actor." Id. at 287, 106 S.Ct. at 1854 (concurring opinion). Justice O'Connor also stated that "[t]he Court is in agreement that . . . remedying past or present racial discrimination by a state actor is a sufficiently weighty state interest to warrant the remedial use of a carefully constructed affirmative action program." Id. at 286, 106 S.Ct. at 1853. Furthermore, Justice Powell, writing for Chief Justice Burger and Justices Rehnquist and O'Connor, noted that "[public schools] are under a clear command from this Court . . . to eliminate every vestige of racial segregation and discrimination in the schools. Pursuant to that goal, race-conscious remedial action may be necessary." Id. at 277, 106 S.Ct. at 1848. Justice Powell also noted that, "in order to remedy the effects of prior discrimination, it may be necessary to take race into account." Id. at 280, 106 S.Ct. at 1850.

Croson did not question the continuing validity of these cases. Croson focused on the fact that the Richmond plan potentially benefitted minority businesses from all across the country, including businesses owned by minority groups against whom there was no evidence of discrimination in Richmond. Thus, it is not surprising that the majority of circuits that have addressed this issue, including this circuit, have read Croson to require that, for an affirmative action program to be considered narrowly tailored, there must be a showing of prior discrimination against the racial or ethnic group that benefits under the program. In Associated General Contractors v. Coalition for Economic Equity, 950 F.2d 1401, 1417 (9th Cir. 1991), the Ninth Circuit upheld a plan which benefitted groups that had suffered discrimination:

[T]he plan remedies only specifically identified discrimination: the City provides preferences only to those minority groups found to have previously received a lower percentage of specific types of contracts than their availability to perform such work would suggest.

The court specifically addressed the argument at issue:

We find [the plaintiff] unlikely to prevail on its argument that tailoring the remedy to specific minority groups is insufficient to meet constitutional requirements, and that to pass constitutional muster any remedy adopted must provide redress only to specific individuals who have been identified as victims of discrimination. To reach any other decision would thwart the Supreme Court's directive in Croson that race conscious remedies may be permitted in some circumstances. As pointed out by the district court, "an iron clad requirement limiting any remedy to individuals personally proven to have suffered prior discrimination would render any race conscious remedy superfluous."

Id. at 1417 n. 12 (citations omitted). In Peightal v. Metropolitan Dade County, 940 F.2d 1394 (11th Cir. 1991), cert. denied, ___ U.S. ___, 112 S.Ct. 969, 117 L.Ed.2d 134 (1992), the Eleventh Circuit contrasted the affirmative action plan under scrutiny with the Richmond plan struck down in Croson: "However, the over-inclusiveness at issue in Croson is not present in our case, since Metro Dade is not offering hiring preference to any groups, e.g., Aleuts or Eskimos, against which no discrimination has been found." Id. at 1409 n. 37 (separate opinion of Judge Brown). In Donaghy v. City of Omaha, 933 F.2d 1448, 1460 (8th Cir. 1991), cert. denied, ___ U.S. ___, 112 S.Ct. 938, 117 L.Ed.2d 109 (1992), the Eighth Circuit found a remedial consent decree not over-inclusive because "the consent decree's interim and long-term goals applied only to blacks, the minority group identified as being underutilized." In Cunico v. Pueblo School Dist. No. 60, 917 F.2d 431, 437 (10th Cir. 1990), the Tenth Circuit held that "[t]he purpose of race-conscious affirmative action must be to remedy the effects of past discrimination against a disadvantaged group that itself has been the victim of discrimination." And the Eleventh Circuit observed that

the Richmond plan was not `narrowly tailored' to remedy past discrimination in Richmond's construction industry because the set-aside requirement applied equally to African-Americans, Hispanics, Indians, Eskimos, and Asians, in spite of the fact that some of those groups were not represented in Richmond. The Richmond plan thus did more than remedy past discrimination, it benefitted groups against whom there may have been no discrimination.

Cone Corp. v. Hillsborough County, 908 F.2d 908, 917 (11th Cir.), cert. denied, ___ U.S. ___, 111 S.Ct. 516, 112 L.Ed.2d 528 (1990). Similarly, this circuit, speaking through Judge Posner, has held that " Croson holds that the Equal Protection Clause forbids states and municipalities to discriminate in favor of blacks and other minorities unless the discrimination is necessary to rectify discrimination against the favored groups." Milwaukee County Pavers Ass'n v. Fiedler, 922 F.2d 419, 421 (7th Cir.), cert. denied, ___ U.S. ___, 111 S.Ct. 2261, 114 L.Ed.2d 714 (1991); see id. at 421-22 ("And a majority of the Justices of the Supreme Court believe that racial discrimination in any form, including reverse discrimination, is unconstitutional when done by states or municipalities, unless the purpose is to provide a remedy for discrimination against the favored group."). Thus, we conclude that Croson demands only that there be a showing of prior discrimination against the racial or ethnic group that benefits under the program, not the individual beneficiaries. As previously determined, in this case the City has justified its affirmative action promotions by demonstrating that the groups that benefitted — blacks and Hispanics within the CFD — were the victims of past discrimination.

Even if, as it is argued, Croson did require that an affirmative action program benefit only identifiable victims of the past discrimination it seeks to remedy, we would uphold the program under review in this case. We have already concluded that the City has provided substantial evidence that there was prior discrimination against blacks and Hispanics in the CFD's hiring and promotion procedures; from 1973 through 1980, the Justice Department challenged the CFD's hiring and promotion practices as discriminatory against blacks and Hispanics. Each beneficiary of the affirmative action promotions is employed within the ranks of the CFD, a specific, identified body infected by the ills of past discrimination. In 1987 there remained a significant statistical disparity between the ranks. The minority membership of the lower ranks was still underrepresented in the upper ranks. Minority lieutenants still suffered the isolating effects of past discrimination, and, even if there was no barrier to entry at the firefighter level, minority firefighters still encountered the frustration of a discriminatory promotion system and of working under leadership that was the product of that system. Therefore, every one of the beneficiaries of the affirmative action practices challenged in Billish and Chicago Fire Fighters suffered at least the general effects of the past discrimination that the CFD was seeking to remedy.

More specifically, the evidence suggests that the beneficiaries of these programs suffered particular effects of past discrimination in CFD hiring and promotion. The beneficiaries of the practices challenged in Chicago Fire Fighters were a group of black firefighters who were promoted to engineer before the plaintiff firefighters, and one Hispanic lieutenant who was promoted to captain before the plaintiff lieutenants. All of the beneficiaries were promoted out of rank order.

With respect to the black firefighters, they had ranked lower than the white firefighters on the 1986 engineer eligibility list because of their performance on the 1985 engineer examination. The district court found that this examination was discriminatory insofar as it "had an adverse impact on minority candidates when used as a rank order device." 736 F. Supp. at 931. As discussed above, see supra pp. 1289-92, there was substantial evidence to support the district court's finding. Thus, the black firefighters who were beneficiaries of the affirmative action plan were not only general victims of past discrimination, they were specific victims of the adverse impact of the 1985 engineer examination.

With regard to the Hispanic lieutenant, he was promoted from the 1987 captain eligibility list out of rank order. This list was compiled from results of the 1986-87 captain examination which, unlike the 1985 engineer examination, was adjusted to dissipate any adverse impact. Thus, the Hispanic lieutenant cannot claim to be a victim of any discriminatory impact of the 1986-87 captain examination. Nevertheless, the evidence suggests that the Hispanic lieutenant was a victim of past discrimination in the 1978 promotion examinations. From the October 1, 1987, promotion order and the 1987 captain eligibility list, one can deduce the name of the Hispanic lieutenant and the fact that he has served in the CFD since July, 1977. R. 71 (Chicago Fire Fighters) Ex.N at 7, Ex.M at 1. In order to be promoted from firefighter to lieutenant, he had to take and pass a promotion exam for lieutenant at some point between July 1977, when he was hired, and September 1987, when he sat for the first portion of the captain exam. His first opportunity would have been the 1978 set of promotion exams, which were discriminatory. His next opportunity would not have been until after the engineer exam in 1985. The first promotions from the 1985 engineer eligibility list, which was produced by the 1985 engineer test, were not made until June 5, 1986 and the first promotions from the new lieutenant eligibility list could not have been any earlier. The promotion orders from June 5, 1986 through September 1987 include only one set of promotions to lieutenant — the December 25, 1986 Personnel Order — and the Hispanic lieutenant at issue is not named on that order. R. 71 (Chicago Fire Fighters) Ex.N at 2-3. Thus, from the evidence available in the record, it is reasonable to conclude that the Hispanic lieutenant must have taken the 1978 lieutenant exam and suffered its specific discriminatory effects: he was ranked lower on the 1979 lieutenant eligibility list than he would have been on a fair test. A lower rank meant a delayed promotion to lieutenant and less experience as a lieutenant before he took the 1986-87 captain exam. To the extent that the lack of experience translated into a poorer performance on the 1986-87 captain exam and thus a lower ranking on the 1987 captain eligibility list, the challenged action of promoting him to captain out of rank order provided a direct remedy of the effects of prior discrimination.

Given this record, we agree with the district court that "there is no genuine issue of material fact that the City's affirmative action goals at issue here were narrowly tailored to their remedial purposes." Chicago Fire Fighters, 736 F. Supp. at 933. In sum, the City had a compelling governmental interest justifying its affirmative action program, and the program was narrowly tailored to accomplish its purpose.

C. Billish v. City of Chicago 1. Facts

The Billish case involves promotions from the 1979 captain eligibility list. As we already have noted, the City administered an examination for the rank of captain in the CFD during 1978-79. The examination consisted of a written test, an oral test, and an efficiency grade. Each of the three parts was graded separately, weighed, and averaged for a final score. In 1979, the City posted the results of the exam and listed the lieutenants in rank order on an eligibility list. Each of the plaintiffs passed the examination and was ranked between numbers 245 and 281. The City began making promotions from the eligibility list in rank order. Of the 298 persons who scored above the passing score of 70, 13 were minorities. Shortly after the posting of the eligibility list, the district court entered the 1980 settlement order, and the City and Union entered the CBA described previously.

In August 1983, Fire Commissioner Galante requested that a new examination be developed and administered for each of the ranks from firefighter to battalion chief because the 1979 eligibility lists were still in place. In March 1985, the last minority lieutenant with a score above 70 on the 1979 captain eligibility list was promoted. In the fall of 1986, the first part of a two-part captain examination was given. On December 25, 1986, before the City had completed administering the second part of the captain examination, Commissioner Galante issued a promotion order for several ranks of the CFD. The promotions were ordered to take effect January 16, 1987. Among those promoted were the eighteen remaining top-ranked persons on the 1979 captain eligibility list, each of whom was white. Commissioner Galante asked the Department of Personnel to identify any minorities that remained on the 1979 list. Deputy Commissioner of Personnel Robert Joyce reported that there were two minority lieutenants remaining on the list, but, because they had scored below the cut-off score of 70, the Personnel Department would have to lower the cut-off score to make them eligible for promotion. Precedent existed for such an action; passing scores for other ranks had been lowered. R. 44 (Billish) Ex.A ¶ 6. Deputy Commissioner Joyce determined that the scores (64.17 and 56.69) and experience of the two minority lieutenants were sufficient to consider them qualified for promotion to captain. He testified that, because passing scores involved administrative factors such as the number of anticipated promotions, "failure to achieve the cut-off score does not mean that a candidate is necessarily unqualified for promotion." Id. ¶ 4. Accordingly, on January 12, 1987, Commissioner of Personnel Hoskins authorized lowering the cut-off score to make the two minority lieutenants eligible for promotion. On January 13, Commissioner Galante ordered the promotion of the two lieutenants effective January 16, the same date the December 25 promotions were effective.

By May 1987, the second half of the captain examination had been administered but the captain eligibility list had not yet been drawn up. Commissioner Galante needed five additional captains and inquired of Personnel Commissioner Hoskins when the new captain eligibility list would be ready. Commissioner Hoskins reported that the new list would be ready in approximately thirty to sixty days. Based on this report, Commissioner Galante decided to wait for the new list rather than draw from the 1979 list. On July 13, 1987, Commissioner Hoskins canceled the 1979 eligibility list and made the 1987 eligibility list effective. Each of the nine plaintiff lieutenants had passed the exam and were among those ranked on the list. On October 1, 1987, the first promotions from the 1987 list were made when twenty-four lieutenants were promoted to captain. These promotions included eighteen whites, five blacks, and one Hispanic. None of the plaintiffs — the highest-ranking of whom was forty-first on the new list — was among the twenty-four promoted. The plaintiffs note, however, that, had the promotions come from the 1979 eligibility list, each of them would have been promoted.

On October 7, 1987, the plaintiffs filed a two-count complaint, under 42 U.S.C. § 1983, against the City and Commissioner Galante. The plaintiffs challenged the lowering of those scores and the subsequent promotions, as well as Commissioner Galante's May 1987 refusal to draw further from the 1979 captain eligibility list. Count I alleged that the promotions of the two minorities deprived the plaintiffs of their right to promotion to captain without due process of law. Count II alleged that the defendants refused to promote the plaintiffs solely because of race, in violation of the Equal Protection Clause. The City moved for summary judgment on both counts. In separate opinions, the district court granted both motions.

2. Due process claim

As the district court noted, a due process claim has two components: (1) the plaintiff had a protectible property interest, and (2) he was deprived of that interest without due process of law. Memorandum Opinion and Order of November 28, 1988 (R. 46 (Billish)) at 6-7;see Perry v. Sindermann, 408 U.S. 593, 599, 92 S.Ct. 2694, 2698-99, 33 L.Ed.2d 570 (1972). To have a constitutionally protected property interest the plaintiffs "must have more than a unilateral expectation." Board of Regents v. Roth, 408 U.S. 564, 577, 92 S.Ct. 2701, 2709, 33 L.Ed.2d 548 (1972). Instead, they must "have a legitimate claim of entitlement." Id. The Constitution does not create property interests; the plaintiffs must find their property interests in sources such as state law, existing rules, or mutual understandings. Id. "A person's interest in a benefit is a `property' interest for due process purposes if there are such rules or mutually explicit understandings that support his claim of entitlement to the benefit and that he may invoke at a hearing." Sindermann, 408 U.S. at 601, 92 S.Ct. at 2699. As the district court also noted, the crux of the plaintiffs' argument is that they had a property interest in being promoted from the 1979 captain eligibility list based on the City's custom and practice of filling all vacancies before retiring an eligibility list and promoting individuals in rank order as vacancies occur. The court considered statutes, rules, ordinances, the City's Personnel Rules, and the CBA between the parties, but found nothing to serve as a sufficient basis for a protected property interest.

On appeal, the plaintiffs draw our attention to the text of the CBA as the source of their property interest in rank-order promotion. Specifically, the plaintiffs point to the language in Article IX, paragraph 9.3B(2)b, which provides: "No employee shall be promoted from a new promotion list until all employees who may have been passed over on a previous promotion order have been promoted." R. 25 (Billish) Ex.D. According to the plaintiffs, this agreement formalized a custom and practice of the City to promote in rank order. The district court reasoned that paragraph 9.3B(2)b did not apply because the plaintiffs were not "passed over" within the meaning of the paragraph. Relying on an arbitration decision that construed the paragraph, the court concluded that the plaintiffs' failure to be promoted from the 1979 eligibility list was not the equivalent of being "passed over." R. 46 (Billish) at 9-10. The court also reasoned that an individual is passed over within the meaning of the paragraph when that individual "comes up for promotion but [was] not promoted for a particular reason, valid or otherwise." Id. at 10. In this case, the court continued, none of the plaintiffs came up for promotion, the entire 1979 eligibility list was canceled and everyone who wished to compete for subsequent promotions was required to take the new captain examination. Furthermore, again relying on an arbitrator's interpretation, the court concluded that the plaintiffs were not "on an order" as that term is used in section 9.3B. Id. at 10-11. Our review of the arbitrator's decision confirms the district court's reasoning. Paragraph 9.3B(2)b applies only to individuals first certified for promotion and placed on a promotion list but later passed over. The plaintiffs were never certified for promotion or placed on a promotion list, therefore they were not "passed over" within the meaning of paragraph 9.3B(2)b.

Plaintiffs also point to the more general language in paragraph 9.3B(2), which states: "[Promotion] [v]acancies will customarily be filled by employees in the order of their ranking on the list." R. 25 (Billish) Ex.D. As the district court noted, this general rule is qualified by the language which precedes it, and thus it is properly understood only in context:

The employees certified for promotion shall be the five employees having the highest scores (rank) for each position to be filled, but in no event more than fifteen employees more than the number of vacancies to be filled. Vacancies will customarily be filled by employees in the order of their ranking on the list.

Id. This reflects the practice testified to by Deputy Commissioner Joyce: after the Fire Commissioner tells the Personnel Department that he would like to promote x number of lieutenants, the Personnel Department prepares and sends to the Commissioner a list of x-plus-five-to-fifteen names from which to choose. Plaintiff-Appellant's App. Ex.W at 27. Thus, the rule that "Vacancies will customarily be filled by employees in the order of their ranking on the list" refers to the list of names which the Personnel Department submits to the Fire Commissioner, not the eligibility list. Furthermore, paragraph 9.3B(2)c(ii) states three reasons an employee may be passed over:

c. An employee may be passed over on an order only for the following reasons:

i. At the time the vacancy occurs the employee is on an injury leave of six months or more for a non-duty related injury. . . .

ii. A major disciplinary infraction (i.e. charges resulting in a suspension of 15 days or more) . . . .

iii. In order to comply with affirmative action ordered by a court.

Id. These exceptions provide the Fire Commissioner some non-rank-order flexibility in choosing from the list he receives from the Personnel Department. When the issue of the Commissioner's discretion was examined in arbitration, the arbitrator reviewed the history of section 9.3B. R. 25 (Billish) Ex.B at 13-18. That history reveals the City's long-standing refusal to allow any limits on the Commissioner's discretion and its acceptance of section 9.3B only with the inclusion of the phrase "on an order," which the City understood to limit the Commissioner's discretion only after he drafted a promotion order. The arbitrator concluded that, despite the seemingly restrictive language of section 9.3B, the parties intended that "the Commissioner has at least a degree of discretion to choose from among the candidates listed on the certification list," although the Commissioner could promote out of rank order "only upon reasonable grounds." Id. at 41-43. The district court relied upon the arbitrator's findings when it concluded that "the Fire Commissioner has discretion to determine whether and when promotions should be made." R. 46 (Billish) at 10-11. After reviewing the CBA and the arbitration opinions, we agree with the district court's conclusion that the Commissioner's discretionary latitude in non-rank-order promotion negates any property interest in rank-order promotion. See Stuart v. Roache, 951 F.2d 446, 455 (1st Cir. 1991) ("[W]here an appointing authority may consider factors in addition to the applicants' ranking on an eligibility list, a police officer's expectation of promotion based on that list will not rise to the level of a `property interest' entitled to constitutional protection."), petition for cert. filed, No. 91-1516 (Mar. 20, 1992).

Finally, the district court found no existence of a custom or practice by the City to fill vacancies for captain from the eligibility list in effect at the time a vacancy occurs. Relying in part on the plaintiffs' own depositions, the court concluded that there was no mutually explicit understanding between the City and the plaintiffs. Indeed, several of the plaintiffs were unable to point to a single instance when all vacancies were filled from an existing list. R. 46 (Billish) at 14. Again, we are in agreement with the district court; there is no substantial evidence of such a custom or practice. Thus, we affirm the district court's dismissal of the due process claim; the plaintiffs have presented insufficient evidence to create a genuine issue as to the existence of a property interest in rank-order promotion.

3. Equal protection claim

The district court granted the City's motion for summary judgment on the equal protection claim in a separate opinion and order. Memorandum Opinion and Order of February 16, 1990 (R. 60 (Billish)). The court began its analysis by stating that the plaintiffs had the burden of showing that: (1) "the rationale for affirmative action is inadequate;" or that (2) "the specific actions unnecessarily trammel the legitimate interests of non-minorities." R. 60 (Billish) at 2-3. Addressing the second prong first, the court concluded that the City could lawfully pursue a course of affirmative action, and the course pursued in this case did not unnecessarily trammel the rights of the plaintiffs. The plaintiffs were not denied an opportunity to be considered for promotion, the court reasoned; in fact, each of the plaintiffs was on the 1987 eligibility list. Rather, the opportunity to be considered for promotion was simply postponed. See id. at 3-4. Furthermore, the court concluded that the plaintiffs could not prove that the City's rationale for the affirmative action was inadequate. The court noted the "significant disparities between minority availability and minority employment at all ranks," the "historical fact" that past hiring and promoting procedures had an adverse impact and were inadequately validated, the existence of prior racially discriminatory performance evaluations, and the DOJ's suit against the City and the ensuing 1980 settlement order. Id. at 5. These conditions corroborated the existence of a rational basis for the City's affirmative action. Relying on Wygant v. Jackson Board of Education, 476 U.S. 267, 106 S.Ct. 1842, 90 L.Ed.2d 260 (1986), the district court declined to accept the plaintiffs' position that a formal finding of discrimination was required. R. 60 (Billish) at 5. The court then held that no equal protection ground existed for challenging the City's decision to retire the 1979 eligibility list. "The City could reasonably conclude that the use of a stale list ought to be deferred where its use would perpetuate effects of past discrimination and increase racial disparities." Id. at 6. Finally, the district court distinguished the present case from Croson on the ground that, in this case, "the showing was made that the City discriminated against blacks in developing the 1979 list and the remedy neither involved arbitrary quotas nor excluded whites from advancement." R. 60 (Billish) at 6 n. 7.

The district court applied the wrong standard of scrutiny in its equal protection analysis. The court apparently relied upon an analytical framework which the Supreme Court prescribed for Title VII analysis in United Steelworkers v. Weber, 443 U.S. 193, 208, 99 S.Ct. 2721, 2729-30, 61 L.Ed.2d 480 (1979). As discussed earlier, a majority of justices in Croson held that, under the Equal Protection Clause, race-based remedial programs are subject to strict scrutiny; they must be justified by a compelling governmental interest and be narrowly tailored. See also Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 274, 106 S.Ct. 1842, 1847, 90 L.Ed.2d 260 (1986) (plurality). For this reason, we must remand this case to the district court for reconsideration of the Billish plaintiffs' equal protection claim.

We acknowledge that there is some authority for simply applying the stricter standard of scrutiny to the record on appeal rather than remanding the case to the district court. See Eisenbud v. Suffolk County, 841 F.2d 42, 45-46 (2d Cir. 1988) (even though district court erred in applying rational basis standard, judgment affirmed under intermediate scrutiny); cf. United States v. Paradise, 480 U.S. 149, 171-85, 107 S.Ct. 1053, 1066-84, 94 L.Ed.2d 203 (1987) (applying narrow tailoring requirement despite lower courts' use of less rigorous standard, without deciding whether lower court had committed error). As this court has repeatedly stressed, "an appellate court may affirm on any ground that finds support in the record." Roland v. Langlois, 945 F.2d 956, 962 n. 11 (7th Cir. 1991). We are also aware, however, that appellate courts are careful not to make such a determination if there is need for further analysis or supplementation of the record by the district court. See, e.g., Tennessee Asphalt Co. v. Farris, 942 F.2d 969, 971 (6th Cir. 1991) (reciting history of remand in light of Croson); Evans v. City of Evanston, 881 F.2d 382, 385 (7th Cir. 1989) (remanding for further analysis and/or supplementation of the record in light of Wards Cove). Here, we believe there are especially compelling reasons to defer to the district court for the initial reassessment. As we have already noted, Croson demands that, in cases involving affirmative action programs, courts undertake the narrow tailoring requirement of strict scrutiny analysis with extreme care. 488 U.S. at 493, 109 S.Ct. at 720-21 ("searching judicial inquiry"), at 507-511, 109 S.Ct. at 728-31. Chief Judge Tjoflat acknowledged this directive when the Eleventh Circuit was confronted recently with a similar situation:

As Croson makes clear, equal protection challenges to affirmative action plans turn on detailed factual evaluations. We, as an appellate court, are ill equipped to make initial factual determinations. Therefore I think it best that we remand appellant's equal protection claim to the district court so that it may decide the merits of this claim. On remand, the district court, which previously resolved this case without the benefit of Croson, will be able to analyze the relevant facts extensively. In this way, we ensure that appellant's claim receives full and fair adjudication.

Peightal v. Metropolitan Dade County, 940 F.2d 1394, 1411 (11th Cir. 1991) (Tjoflat, C.J., concurring in part and dissenting in part), cert. denied, ___ U.S. ___, 112 S.Ct. 969, 117 L.Ed.2d 134 (1992). Moreover, upon examination of the record in this case, we have substantial misgivings as to whether it is sufficient to support the exacting degree of thorough analysis required by Croson. For example, we note that, in its presentation to us, the City has placed great emphasis on the immediate operational need for new captains as justifying the promotion of the two black lieutenants who scored below 70 on the 1979 exam and, in that regard, places great reliance on the deposition and affidavit of Commissioner Galante. However, on inspection, we believe that material leaves a significant question as to whether there was an immediate operational need for two more captains in January 1987. Read as a whole, Commissioner Galante's testimony appears to evidence concern more with immediate minority hiring than with meeting immediate operational needs.

On remand, the district court should reappraise the equal protection claim. Because the compelling governmental interest analysis will be, to a large extent, duplicative of our analysis in Chicago Fire Fighters, the principal focus should be on whether the challenged actions were narrowly tailored. Given the exacting standard of Croson, the district court may well determine that augmentation of the record is appropriate.

Conclusion

For the foregoing reasons, we affirm the judgment of the district court in Chicago Fire Fighters (No. 90-2182). In Billish (No. 90-1650), we affirm the district court's grant of summary judgment on the due process claim, but reverse and remand the disposition of the equal protection claim for further consideration in conformity with this opinion. The Defendants-Appellees in No. 90-2182 may recover their costs in this court. The Plaintiffs-Appellants in No. 90-1650 may recover their costs.

No. 90-2182 — AFFIRMED.

No. 90-1650 — AFFIRMED in part, REVERSED and REMANDED in part.

RIPPLE, Circuit Judge.

ORDER [116] July 8, 1992.

A majority of judges in active service have voted to rehear this case en banc. Accordingly,

IT IS ORDERED that rehearing en banc be, and the same is hereby GRANTED.

IT IS FURTHER ORDERED that the opinion entered in this case on May 4, 1992, be, and is hereby VACATED. This case will be reheard en banc at the convenience of the court.