File Name: 06a0385n.06
                                Filed: June 1, 2006
                  NOT RECOMMENDED FOR FULL-TEXT PUBLICATION

                                        Case No. 04-3904

                            UNITED STATES COURT OF APPEALS
                                 FOR THE SIXTH CIRCUIT

JANELL RUTHERFORD et al.,                     )
                                              )
        Plaintiffs-Appellants,                )
                                              )      ON APPEAL FROM THE UNITED
v.                                            )      STATES DISTRICT COURT FOR THE
                                              )      NORTHERN DISTRICT OF OHIO
CITY OF CLEVELAND et al.,                     )
                                              )      OPINION
        Defendants-Appellees.                 )


        BEFORE: MOORE, ROGERS, and McKEAGUE, Circuit Judges.

        McKEAGUE, Circuit Judge. Appellants, non-minority applicants for the position of police

patrol officer in Cleveland, Ohio (the “City”), filed claims of reverse discrimination and disparate

treatment against the City, the Cleveland Police Department (“CPD”), and several other City

defendants (collectively, the “City defendants”). Appellants’ claims relate to two earlier lawsuits

involving the City’s past discrimination against African-Americans and Hispanics (collectively,

“minorities”) in hiring and promoting by the CPD. The district court approved in those cases a

consent decree that implemented a race-based plan to advance the hiring and promotion of minority

police patrol officers.

        In the proceedings below, the City defendants were granted summary judgment. Appellants

sought review, asking this court to reverse. As explained below, we affirm the district court’s

judgment. On Appellants’ reverse discrimination claim, the City’s history of racial discrimination

against minorities – as evidenced by its own admission of discrimination, judicial findings, and
statistical disparities – provided the City defendants with a compelling interest in implementing the

CPD’s temporary race-based hiring plan. Given several important features of the plan, especially

its sunset provision and flexibility, we also find that the plan was narrowly tailored, and thus

survives the strict judicial scrutiny required by the Supreme Court’s equal protection jurisprudence.

On Appellants’ disparate treatment claim, we affirm the district court’s finding that the claims are

without sufficient support in the record.



                                        I. BACKGROUND

A.      The Shield Club Lawsuits and Resulting Consent Decree

        Although the Appellants do not, strictly speaking, contest the constitutionality of the consent

decree except as it applied to them in 1993-1994, their claims necessarily take aim at several of its

foundations. Thus, we must review the historical record supporting the consent degree. Appellants

take issue only with the hiring practices of the CPD, and not its promoting practices; accordingly,

only the evidence involving the CPD’s hiring history will be reviewed, unless otherwise noted.

        In 1966, minorities constituted 6.2% of the CPD police force. In 1972, that figure had risen

to 8.1%. In comparison, minorities made up approximately 40% of the City’s population according

to 1970 census figures. The Shield Club, an organization comprised of minority police officers who

represented the interests of minority officers and applicants, filed a lawsuit against the City, the

CPD, and other defendants alleging racial discrimination in the hiring and promoting of minority

police patrol officers.1

        As one of their claims, the Shield Club-plaintiffs maintained that the CPD’s entrance


The Shield Club-plaintiffs filed a subsequent lawsuit against City defendants. The two cases –
1

Nos. C72-1088 and C77-346 – were consolidated.

                                                  2
examination discriminated against minorities. The district court agreed, finding that the high failure

rate of minorities compared to the much lower failure rate of non-minorities supported plaintiffs’

prima facie case of discrimination. Shield Club v. City of Cleveland, 

, 253 (N.D.

Ohio 1972). The district court also found that the examinations were not validated for job

performance. 

this evidence and no effective rebuttal by the defendants, the district

court concluded that the examinations had a racially discriminatory impact. 

The district

court’s conclusion was buttressed by the historical workforce disparities between minorities and

non-minorities. See 

        As a result, the district court directed the CPD to appoint minorities at a rate no less than

18% of its incoming police patrol officers. 

rate was equal to the percentage of

examination passers who were minorities. 

to the deficiencies in the examination,

the district court directed the defendants to create an examination that was job-related. 

        Two years later, the district court found that the CPD’s screening procedures used to

determine which applicants, of those who passed the entrance examination, could be appointed as

officers had a racially disparate impact on minorities. Shield Club v. City of Cleveland, No. C 72-

1088, 

, at *2 (N.D. Ohio July 6, 1974). The court ordered defendants to develop and

implement racially-neutral screening procedures. 

Faced, however, with a serious shortfall

of officers and without racially-neutral selection procedures available, the district court also directed

the City to hire a specific percentage of minorities as officers. In determining the target, the district

court noted that of the 400 highest ranking applicants based on the 1974 examination, approximately

39.5% were minorities. The court therefore instructed the City that if a significant departure in

officer hiring (more than 2% to 3% below 39%) developed, it would take another look at whether

the City was engaging in racial discrimination. 

                                                   3
       In 1976, the district court determined that the Chief of Police had a racially discriminatory

purpose in perpetuating under-representation of minorities in certain minority-dominated districts

of the City. Shield Club v. City of Cleveland, No. C72-1088, 

, at *12 (N.D. Ohio Oct.

20, 1976). The court specifically found a “continuing marked under-representation” of minorities

in these districts, without any explanation or justification by the defendants. 

1977, the parties submitted a consent decree to the district court. The parties

entered into the decree to “effectuate an effective prospective remedy designed to eliminate all

vestiges of race discrimination within the City of Cleveland Police Department.” Consent Decree

¶ 1, JA at 586. As to minority officer hiring, the parties stipulated that the CPD would utilize “only

such selection criteria . . . as are non-discriminatory and demonstrably job-related.” 

JA at

588. They established a minority-hiring target of not less than 35.8%, a figure “based upon external

labor market figures contained in the 1970 census.” 

JA at 589. In order to accomplish

this target, the CPD agreed temporarily to hire no less than three minorities for every four non-

minority hires (the “3:4 hiring ratio”). 

JA at 589.

       According to its charter, the City did not employ a pure “examination rank order” system

for hiring officers. Rather, it used what is called the “one-in-three” rule (the “1:3 rule”), which gave

the CPD discretion in selecting one candidate from a group of three qualified candidates. Those

candidates not appointed on the first review remained on the list in their rank order and were

considered again, along with the next highest-ranking candidate, for the next available opening. If

a candidate had been certified and not chosen on four separate occasions, that person was removed

from the eligibility list. The consent decree permitted the CPD to continue to use the 1:3 rule, 

JA at 588; however, to ensure race played a minimal role in hiring, but also that the 3:4 hiring

ratio was met, the CPD could maintain separate lists of qualified minority and non-minority

                                                   4
candidates, 

JA at 590. Qualified minorities were compared against each other,

qualified non-minorities were compared against each other, but minorities were not compared

against non-minorities.

        The district court held a hearing to determine whether to approve the consent decree. The

court specifically addressed whether the Shield Club-plaintiffs had proven their claim of a historic

pattern of racial discrimination. While the court had earlier determined that the record had not

convinced it that the Shield Club-plaintiffs had carried their burden, it noted during the approval

hearing that this finding was “not engraved in stone.” JA at 959. The district court specifically

focused on Paragraph 3 of the consent decree. In that paragraph, the parties agreed that the “entire

record” reflected a “history of race discrimination in the hiring practices and in the promotion

practices” of the CPD. 

of this admission by the defendants, the district court found it

unnecessary to hold additional proceedings “to go all through what evidence was spread on this

record for five years.” JA at 960. The district court concluded that the record was sufficient to reach

“an intelligent and objective opinion of the probabilities of ultimate success should the claim be

litigated,” accepting as fact the defendants’ admission of racial discrimination. 

(quoting

Flinn v. FMC Corp., 

, 1173 (4th Cir. 1975)). The district court placed its imprimatur

on the consent decree in November 1977. At the time, minorities made up 9.2% of the City’s police

officers.2
2
 Subsequent to the entry of the consent decree in the Shield Club cases, the City admitted to a pattern
of racial discrimination in its fire department. The Vanguards of Cleveland, an organization of
African-American and Hispanic firefighters, brought similar claims against the City as those
presented in the Shield Club cases. The City likewise entered into a consent decree in that case,
which was contested and upheld. See Local No. 93, Int’l Ass’n of Firefighters v. City of Cleveland,

(1986); Vanguards of Cleveland v. City of Cleveland, 

(6th Cir. 1994).
As the Supreme Court characterized the City’s position, the City decided to negotiate with the
Vanguards rather than engage in “another round of futile litigation.” Int’l Ass’n of 

.

                                                  5
        Near the end of 1984, the Shield Club-plaintiffs sought to modify and extend the consent

decree. By this time, minorities made up 20.8% of the CPD police force. The defendants initially

opposed such modification and extension. The district court determined that the defendants had met

their obligations as to promotions, but that there remained a question of fact on whether the decree

should be modified and extended as to officer hiring. The parties then submitted an Amended

Consent Decree (the “ACD”) dealing solely with CPD hiring practices. Upon review of the decree,

the district court approved it.

        The ACD expressed the parties’ intent to “finally and fully resolve these actions so as to

preclude any further requests for extension and/or modification.” ACD Preamble, JA at 856. The

decree again provided that the defendants would use “only such selection criteria . . . as are non-

discriminatory and demonstrably job related” and that the CPD would continue to use the 1:3 rule.

JA at 857.

        Specifically as to minority hiring, the ACD required the following:

        Temporarily and until such time as 33% of the police officers employed by the City
        are minorities or until December 31, 1992 (eight years), whichever time or event
        occurs first, defendants shall hire no less than three minority police patrol officers
        for every four non-minority police patrol officers, . . . .

JA at 858. In the event the CPD failed to hire a minimum of 70 patrol officers in any year

during the eight-year period, the 3:4 hiring ratio would continue “in full force and effect for one

additional year for each year in which the City shall fail to hire the minimum number of police

patrol officers, unless the City has otherwise achieved the 33% level.” 

JA at 859. The

decree permitted the CPD to use separate eligibility lists for non-minority and minority candidates

in conjunction with the 1:3 rule. 

JA at 859.

        Throughout the eight-year period, the CPD never reached the target of a 33% minority police

                                                  6
force. Moreover, the CPD failed to hire at least 70 police patrol officers in 1986 and 1991.

Accordingly, the CPD was bound under the ACD to continue using the 3:4 hiring ratio until the

earlier of December 31, 1994, or the date when minorities made up 33% of the police force. The

CPD indeed met the 33% target in mid-1994, so the district court concluded that the defendants had

fully complied with the requirements of the decree on that date and that the decree expired under

its own terms.



B.      Appellants’ Present Lawsuit

        Appellants filed their lawsuit in May 1994, shortly before the ACD expired. Appellants are

non-minority applicants for the position of police patrol officer with the CPD. A portion of them

comprise a sub-class of applicants who allege that as a result of the ACD’s extension into 1993 and

part of 1994, they were never considered for the position even though minority candidates who

ranked lower than they did on the eligibility list were considered. The remaining Appellants are

individuals who were considered for the position in 1993 and 1994, but were rejected by the CPD

based on qualifications that were allegedly not applied to minority applicants who were lower on

the eligibility list, but still hired.

        Appellants argue that the application of the City’s race-based hiring plan, as it was applied

in 1993 and 1994, was unconstitutional. Specifically, Appellants claim that the ACD should not

have been applied beyond 1992, and that such application resulted in less qualified minority

applicants being hired as police patrol officers. They argue that had the parties and district court in

the Shield Club cases attempted to develop the type of findings required under the constitution, the

district court would have found that application of the plan in 1993-1994 was not justified by a

compelling governmental interest, nor was it narrowly tailored to achieve such an interest.

                                                  7
Furthermore, Appellants contend that they received less-favorable treatment than minorities during

the candidate review and selection process.

       In their amended complaint, Appellants set forth five separate causes of action: (1) racial

discrimination in the CPD’s selection of police patrol officers in violation of the Civil Rights Act

of 1871, 42 U.S.C. §§ 1981 and 1983, and the Equal Protection Clause of the Fourteenth

Amendment; (2) loss of property interest in consideration for hire based on examination rank in

violation of the Civil Rights Act of 1871, 42 U.S.C. § 1983, and the Due Process Clause of the

Fourteenth Amendment; (3) civil conspiracy in violation of the Civil Rights Act of 1871, 42 U.S.C.

§ 1985(3), and the Equal Protection and Due Process Clauses of the Fourteenth Amendment; (4)

violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e; and (5) violation of Ohio’s

common law of fraud. On the City defendants’ motion, the district court granted them summary

judgment, finding that Appellants did not have standing to contest the consent decree or its

application. We reversed and remanded to the district court for further proceedings. Rutherford v.

City of Cleveland, 

, 911 (6th Cir. 1998).

       After engaging in discovery, the parties filed motions for summary judgment. The district

court granted summary judgment to the City defendants in two separate decisions. Appellants now

seek appellate review on their equal protection and Title VII claims, as well as their disparate

treatment claims.3

                                          II. ANALYSIS

A.     Summary Judgment Under Fed. R. Civ. P. 56(c)


3
 Appellants conceded below that summary judgment was appropriate on their common-law fraud
and civil conspiracy claims. Appellants have not pursued on appeal their loss of property claims
related to the CPD police patrol officer positions they were not offered. Accordingly, these
claims are not before us on appeal.

                                                  8
       We review the factual findings of the district court for clear error, see Brunet v. City of

Columbus, 

, 410 (6th Cir. 1993), and its legal conclusions, including whether summary

judgment is appropriate, de novo, see Zambetti v. Cuyahoga Cmty. College, 

, 255 (6th

Cir. 2002) (citation omitted). Summary judgment is appropriate “if the pleadings, depositions,

answers to interrogatories, and admissions on file, together with the affidavits, if any, show 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). Under Rule 56(c), a fact is material only if its resolution

will affect the outcome of the lawsuit. Anderson v. Liberty Lobby, Inc., 

, 248 (1986).

Whether a factual issue is genuine requires consideration of the applicable evidentiary standards.

Hedrick v. W. Reserve Care Sys., 

, 451 (6th Cir. 2004).

       As the moving parties, the City defendants bear the initial burden of showing that there exists

no genuine issue of material fact to litigate. Celotex Corp. v. Catrett, 

, 323 (1986). The

court reviews the evidence and draws all reasonable inferences in favor of Appellants as the non-

moving parties. 

(citation omitted). Appellants must show, however, more

than a “mere . . . scintilla of evidence” to defeat summary judgment; “there must be evidence on

which the jury could reasonably find” in their favor. Copeland v. Machulis, 

, 479 (6th

Cir. 1995) (citation omitted). Moreover, if the City defendants carry their burden of proof, the

burden then shifts to Appellants to “produce evidence that results in a conflict of material fact” to

be resolved by the fact-finder. Cox v. Kentucky Dep’t of Transp., 

, 150 (6th Cir. 1995).



B.     Race-Based Classifications Must Survive Strict Judicial Scrutiny

       Appellants state that they do not question the constitutionality of the City’s race-based hiring

plan – embodied in the original consent decree and the ACD – as it was applied prior to 1993. They

                                                  9
insist instead that their claims arise from application of the ACD in 1993-1994. Yet, if the City’s

affirmative action plan was unconstitutional as originally implemented and amended, it is hard to

fathom how it could be constitutional in 1993-1994. Conversely, if the plan was constitutional in

1977 and 1984, this would be some support for finding it constitutional as it applied in 1993-1994

(given its automatic sunset provision). Moreover, Appellants repeatedly attack the reasoning and

factual record which underpin the original consent decree and the ACD. Thus, while focusing on

the ACD as it applied in 1993-1994, we also address issues touching squarely on the basic

constitutionality of the original consent decree and ACD.

        We review the CPD’s hiring decisions, based in part on race, under the Equal Protection

Clause of the Fourteenth Amendment.4 Wygant v. Jackson Bd. of Educ., 

, 273 (1986).

“The Equal Protection Clause prohibits discrimination by government which either burdens a

fundamental right, targets a suspect class, or intentionally treats one differently than others similarly

situated without any rational basis for the difference.” Trihealth, Inc. v. Bd. of Comm’rs, 

, 788 (6th Cir. 2005). Here, the fundamental right of Appellants “to be treated with equal dignity

and respect” by a public employer without reference to their race was implicated by the City’s

temporary race-based hiring plan. City of Richmond v. J.A. Croson Co., 

, 493 (1989).

        Given this fundamental right of equal protection, it is now beyond question that all racial

preferences instituted by a government actor – even those actions sanctioned, like here, under a

consent decree – must pass strict judicial scrutiny. Grutter v. Bollinger, 

, 326 (2003);

Adarand Constructors, Inc. v. Peña, 

, 227 (1995); 

; United

Black Firefighters Ass’n v. City of Akron, 

, 1008 (6th Cir. 1992). The race of the


4
 The Equal Protection Clause guarantees that no State shall “deny to any person within its
jurisdiction the equal protection of the laws.” U.S. Const. amend. XIV, § 1.

                                                   10
challengers to a government’s preference system has no bearing on our analysis: “any person, of

whatever race, has the right to demand that any governmental actor subject to the Constitution

justify any racial classification subjecting that person to unequal treatment under the strictest judicial

scrutiny.” 

; see also 

(“[T]he level of scrutiny does

not change merely because the challenged classification operates against a group that historically

has not been subject to governmental discrimination.”) (citations omitted).

        To satisfy strict scrutiny, a race-based remedial plan must be narrowly tailored to accomplish

a compelling governmental purpose or interest. United States v. Paradise, 

, 167 (1987).

Although a high standard to meet, “[s]trict scrutiny is not ‘strict in theory, but fatal in fact.’” 

(quoting 

). “When race-based action is necessary to

further a compelling governmental interest, such action does not violate the constitutional guarantee

of equal protection so long as the narrow-tailoring requirement is also satisfied.” 

C.      Burdens of Proof

        Where a race-based remedy is subject to strict judicial scrutiny, as here, the party defending

the remedy bears the initial burden of demonstrating “a strong basis in evidence” that a compelling

governmental interest exists which justifies the remedy. Associated Gen. Contractors of Ohio v.

Drabik, 

, 735 (6th Cir. 2000) (citing 

, 500). This burden

has also been referred to as the “convincing evidence” standard. 

; United

Black 

. If the defending party meets its initial burden, then the

burden shifts to the party challenging the remedy to prove its unconstitutionality. Aiken v. City of

Memphis, 

, 1162 (6th Cir. 1994); 

; see also Wygant, 

at 277-78.5



D.     Compelling Governmental Interest

       As we have previously observed, “[t]here is no question that remedying the effects of past

discrimination constitutes a compelling governmental interest.” Associated Gen. 

(citing 

; United Black 

).

Societal discrimination alone is insufficient, though – the City defendants must make “some

showing of prior discrimination” on their part to justify the “limited use of racial classifications in

order to remedy such discrimination.” 

; see also 

(recognizing a compelling interest in a government’s effort to remedy past discrimination for which

it was responsible). Thus, if the City defendants come forward with a strong basis in evidence of

past discrimination in the hiring policies or practices of the CPD, they will have met their initial

burden.

5
 There is some dispute on whether the government should instead bear the ultimate burden to
prove the constitutionality of its affirmative action plan in all cases. See, e.g., Concrete Workers
of Colo., Inc. v. City & County of Denver, Colo., 

(2003) (Scalia, J., dissenting
from denial of writ of certiorari) (stating that government should bear ultimate burden of
proving constitutionality of race-based affirmative action plan); Bass v. Bd. of County Comm’rs,

, 1116 (11th Cir. 2001) (stating that “Supreme Court precedent governing equal
protection claims places the burden on a defendant to prove that an affirmative action plan
satisfies strict scrutiny”); Middleton v. City of Flint, 

, 404 (6th Cir. 1996) (stating
that city had to “prove” it had a compelling interest and that the remedy was narrowly tailored).
The Supreme Court’s recent decision in Gratz v. Bollinger appears to lend support to this
position, at least in the education context. 

, 270 (2003) (“To withstand our strict
scrutiny analysis, respondents must demonstrate that the University’s use of race in its current
admissions program employs ‘narrowly tailored measures that further compelling governmental
interests.’”) (quoting 

). Even if the ultimate burden of proof were to be
placed on the City defendants, we find that they have provided sufficient evidence to meet this
burden. Thus, we leave to another day the question of whether parties challenging a remedial
race-based plan bear the ultimate burden of proof (as do most parties challenging government
action) or rather that burden should be placed on the government-defendants given the important
equal protection issues involved.

                                                  12
        Establishing a strong basis in evidence is not an easy burden to meet. “The only cases found

to present the necessary ‘compelling interest’ sufficient to ‘justif[y] a narrowly tailored race-based

remedy’ are those that expose . . . ‘pervasive, systematic, and obstinate discriminatory conduct.’”

Associated Gen. 

(quoting 

). “[A]n

amorphous claim that there has been past discrimination in a particular industry” is not enough.

Vogel v. City of Cincinnati, 

, 599 (6th Cir. 1992) (quoting 

).

        To begin, the City’s own admission of its history of discriminating against minorities in CPD

hiring – a position it fought against in litigation for a number of years – is persuasive evidence of

“pervasive, systematic, and obstinate discriminatory conduct.” This is not a case in which a

government body makes a finding that societal discrimination affects a particular industry, such as

in Croson and Adarand. Nor is this a case in which a government body expressly maintains that it

did not engage in unlawful racial discrimination. See, e.g., 

. This is a case,

rather, in which a governmental defendant admits its own culpability – against its own interests at

the time – in past racial discrimination. These are critical distinctions. See 

(recognizing distinction between affirmative action plans based on generalized discrimination in an

industry versus discrimination by a city in hiring its own police force).

        The City’s admission was not simply made in a newspaper article, press release, or in some

other voluntary, public forum, but in a consent decree approved by and entered as a judgment of the

district court. A race-based hiring plan embodied in a consent decree “stands somewhere in between

a voluntary affirmative action program (as in Wygant) and a remedial plan that a court has imposed

after making a formal finding of intentional discrimination (as in United States v. Paradise . . . ), for

a consent decree is a hybrid that has ‘attributes both of contracts and of judicial decrees.’” Donaghy

v. City of Omaha, 

, 1459 (8th Cir. 1991) (quoting Int’l Ass’n of Firefighters, 

at 519). On the one hand, as a form of “contract founded on the agreement of the parties, . . . [i]t

should be construed to preserve the position for which the parties bargained.” 

(citations omitted). On the other hand, as a form of judgment, “[w]hile the entry of an affirmative

action consent decree does not guarantee that the decree serves a remedial purpose . . . , the

heightened judicial oversight inherent in a properly entered decree helps attain that end.” 

. Here, both aspects – contractual and judicial – weigh in favor of giving real effect

to the City’s admission of past discrimination.

       A party can also demonstrate a strong basis in evidence by showing that a court made a

finding of past discrimination. 

; see also Boston Police Superior Officers

Fed’n v. City of Boston, 

, 20 (1st Cir. 1998) (explaining that a “strong basis” in evidence

can be based on a “contemporaneous or antecedent finding of past discrimination by a court or other

competent body”) (citation omitted). As recounted above, the district court made several findings

of discrimination prior to 1977 against the City. During its review of the consent decree in 1977,

the district court concluded that the record supported the City’s admission of a history of race

discrimination within the CPD. It subsequently reaffirmed this position by entering the ACD as a

judgment of the court in 1984.

       A review of the statistical evidence also supports the City’s admission and the district court’s

findings of discrimination. While evidence of “mere statistical disparities” is not enough by itself

to show a compelling interest, Associated Gen. 

(citing 

), a prima facie case of discrimination can be made where “a gross disparity exists

between the expected percentage of minorities selected [for hire] and the actual percentage of

minorities selected [for hire],” United Black 

(emphasis added); see

also 

(plurality op., O’Connor, J.) (explaining that a prima facie case of

                                                  14
discrimination can arise “[w]here there is a significant statistical disparity between the number of

qualified minority contractors willing and able to perform a particular service and the number of

such contractors actually engaged by the locality or the locality’s prime contractors”); 

(“It is settled that appropriate statistical evidence setting forth a prima facie case of

discrimination is sufficient to provide a strong basis in evidence to support a public employer[’s]

affirmative action plan.”) (internal quotations omitted; emphasis in original); 

(“Evidence of wide statistical disparities . . . may justify an affirmative action policy adopted by a

public employer.”) (citation omitted).

       The record is replete with gross statistical disparities between the treatment of minorities and

non-minorities by the CPD. For example, the district court found that in the 1970s, minorities

constituted 23% of those taking the entrance examination, but represented 64% of those who failed.

Minorities had a failure rate of 26.3% compared to a rate of 4.5% for non-minorities. Shield 

. Throughout the 1970s, 1980s, and early 1990s, the percentage of minorities

in the CPD officer force tracked below – often quite significantly – the level of minorities in the

general population, the percentage of examination takers who were minorities, and the percentage

of examination passers who were minorities. This court has found that similar disparities supported

a finding of racial discrimination. See, e.g., 

.

       Appellants cite to this court’s decision in Middleton for support of their position that the

statistical disparities do not warrant a finding of compelling interest. That case is distinguishable

on a number of important points. First, as the court noted at the very beginning of its analysis, the

City of Flint’s plan was not presented as part of a motion for a judicially approved consent decree,

but rather was voluntarily conceived as part of the normal political process. 

. Thus, there was no heightened judicial review of the plan before its implementation nor any

                                                  15
prior judicial findings of discrimination. The city relied upon general population statistics rather

than also considering the relevant labor pool. 

The court was further concerned about

several serious deficiencies in the city’s expert statistical analysis. 

these shortfalls exist

here.6

         Appellants also argue that even if there was a compelling interest justifying the race-based

remedy in 1977 and 1984, surely there was no such interest in 1993 and 1994. As evidence, they

point to the convergence of the CPD’s minority police force percentage to the 33% target. Yet, in

determining whether the governmental body had a compelling interest, a reviewing court should

focus on the evidence of discrimination existing at the time the body enacted the race-based remedy.

See Shaw v. Hunt, 

, 910 (1996) (holding that “the institution that makes the racial

distinction must have had a strong basis in evidence to conclude that remedial action was necessary,

before it embarks on an affirmative-action program.”) (internal quotation omitted, emphasis in

original); In re City of Memphis, 

, 350-51 (6th Cir. 2002) (noting same). This and other

courts have “considered evidence of prior discrimination occurring years before” enactment of a

race-based remedy, sometimes finding the evidence convincing, see, e.g., 

(considering evidence of discrimination nine years before affirmative action plan instituted; upheld

plan), Stuart v. Roache, 

, 453 (1st Cir. 1991) (considering evidence of discrimination

ten years before affirmative action plan instituted; upheld plan), and other times finding it
6
 This court’s decision in Long v. City of Saginaw is likewise distinguishable. In that case, we
reversed a district court’s grant of summary judgment to city officials over an amendment to the
city’s race-based hiring plan for its police department. 

(6th Cir. 1990). We found that
there was no compelling interest for the plan because, in part, there had never been a complaint of
discrimination lodged against the city, nor had there been any adjudication or formal findings of
discrimination against the city in its hiring of officers. 

Without some record of
discrimination, we concluded that the city’s admission against interest was insufficient by itself to
justify the remedial measures. 

The city’s statistical evidence also suffered from
numerous defects which are not present here. 

                                                   16
unconvincing, see, e.g., 

(finding evidence of discrimination 14 years before

enactment of affirmative action plan “too remote”). Here, the historical and contemporaneous

evidence of discrimination in the 1970s and 1980s provided the Shield Club parties and the district

court with a strong basis in evidence for the necessity of the original consent decree in 1977 and its

amendment in 1984.

        Finally, a valid race-based remedy is not limited to alleviating the current practice of racial

discrimination, but can also address the “lingering effects” of such discrimination. See 

; see also Dean v. City of Shreveport, 

, 456-57 (5th Cir. 2006) (explaining

that courts should consider whether “lingering effects of past discrimination still necessitate a race-

conscious remedy”) (citing 

; Police Ass’n of New Orleans v.

Cannatella, 

, 1168-69 (5th Cir. 1996)). This is necessarily the case in which the race-

based remedy extends beyond the immediate moment into the future, when presumably the practice

of discrimination has halted, but its effects are still felt. As discussed in more detail infra Sections

II.E.2 and 4, the vestiges of racial discrimination had not been sufficiently relieved by 1992 and

therefore the City defendants remained bound by the terms of the ACD.

        Thus, with the admission by the City of past racial discrimination, supported by the findings

of the district court and a review of the statistical evidence, the City defendants have shown

convincing evidence of a compelling interest in increasing the number of qualified minorities among

the CPD’s ranks. We now turn to determine whether the City defendants crafted a measured, narrow

remedy designed to attain that end.



E.      Narrowly Tailored Remedy

        As explained by the Supreme Court, “[t]he purpose of the narrow tailoring requirement is

                                                  17
to ensure that ‘the means chosen “fit” th[e] compelling goal so closely that there is little or no

possibility that the motive for the classification was illegitimate racial prejudice or stereotype.’”

(quoting 

). There is no single “best” remedy –

courts and other government bodies enjoy some discretion in remedying racial discrimination.

(“While a remedy must be narrowly tailored, that requirement does not

operate to remove all discretion from the District Court in its construction of a remedial decree.”);

(noting that “there is no universal answer to the problem of remedying

racial discrimination” and that trial courts have sound discretion in fashioning remedy). Yet, as the

district court in the Shield Club case succinctly noted, “[o]nly tough choices, each freighted with

some weakness and the risk of individual or group inequities” face the body charged with remedying

past discrimination. Shield 

.

        In determining whether the City’s race-based remedial measure was narrowly tailored, the

court must look to the following factors: “the necessity for the relief and the efficacy of alternative

remedies; the flexibility and duration of the relief, including the availability of waiver provisions;

the relationship of the numerical goals to the relevant labor market; and the impact of the relief on

the rights of third parties.” 

(citation omitted). These are not elements –

the ACD need not satisfy each and every item to survive strict scrutiny. See Ashton v. City of

Memphis, 

, 1056 (W.D. Tenn. 1999). Rather, after reviewing the ACD in light

of all the factors, the court must be satisfied that the remedy is narrowly tailored to meet its remedial

purpose. See 

fact that the ACD was reviewed and approved by the district court

weighs in favor of finding that it was narrowly tailored. See 

(finding

relevant the “heightened judicial oversight” of the affirmative action plan in determining whether

it was narrowly tailored).

                                                   18
1.     Necessity for the Race-Based Remedy and Efficacy of Alternative Remedies

       To determine whether the CPD’s race-based hiring plan was necessary, it is important to

consider first the goal of the plan. The parties in the Shield Club lawsuits (by agreeing to the

consent decree) and the district court (by entering the decree as a judgment) intended to ameliorate

the effects of the City’s past racial discrimination against African-Americans and Hispanics by

temporarily increasing the number of these minorities hired as police patrol officers. Accordingly,

the plan was appropriately focused on only the group of minorities proven to have been subjected

to discriminatory practices. 

(explaining that remedy must be “linked to

identified discrimination” and not suffer from “gross overinclusiveness”).

       Appellants argue, however, that application of the remedial plan in 1993-1994 was not

necessary for two reasons. First, Appellants argue that its application was not necessary in light of

the City’s use of job-validated examinations and screening procedures for a number of years prior

to 1993-1994. They also assert that the CPD hired “substantially more than the number of

minorities between 1985 and 1992 than required by the terms” of the ACD. Given this, Appellants

maintain, the plan was unnecessary in 1993-1994 to eradicate any remaining vestiges of

discrimination. Both arguments fail on close examination.

       A job-validated examination “identifies those skills important to a particular job and tests

those skills.” 

. The use of validated, non-discriminatory hiring procedures can,

under certain circumstances, be an acceptable alternative to race-based relief. See 

. For example, when a governmental body makes decisions based solely on the rank-order

results from a job-validated examination, the race of the examination taker will not factor into the

hiring decision. Over time, the effects of past discrimination should eventually dissolve away as the

effects of the meritocracy begin to dominate.

                                                 19
       In this case, however, the job-validated examinations and screening procedures were not

effectively remedying the effects of past discrimination at the time the City entered the ACD.

Although the patrol officer examination was validated in 1974, by 1984 the City’s proportion of

minority officers had only reached 20.8%. The 3:4 hiring ratio simply sped up the City’s progress

towards achieving its goal of a police workforce made up of 33% minority officers. The Supreme

Court has held that hiring ratios are permissible as a means to regulate the speed of progress towards

fulfilling a race-based plan’s goal. 

.7

       In support of its second argument, Appellants point out that the City hired an average of 84.6

police patrol officers per year during the 1985-1992 period, exceeding the ACD’s annual minimum

of 70 officers. By application of the 3:4 hiring ratio, this meant that the CPD hired more minorities

(on average) than it would have if it had hired only 70 officers per year during that period. Thus,

according to Appellants, there was no need for the remedial measures in 1993-1994.

       As recognized by the district court, however, there is no basis in the ACD for this

calculation. Appellants contest here the interpretation of the ACD, rather than its constitutionality.

Although they have standing to challenge the ACD’s constitutionality, 

,

as non-parties to the decree, they do not have standing to enforce their understanding of its terms,

; 

. The City defendants were bound to follow the

terms of the ACD, not some hypothetical plan now suggested by the Appellants. See W.R. Grace

& Co. v. Local Union 759, Int’l Union of The United Rubber, Cork, Linoleum & Plastic Workers,

, 766 (1983) (“It is beyond question that obedience to judicial orders is an important


7
 We also point out that the that the CPD did not hire police patrol officers based purely on their
examination score rankings – it had a measure of discretion in choosing candidates by operation of
the 1:3 rule. Thus, there is no guarantee that use of a validated examination alone would have
ameliorated the past effects of racial discrimination in the workforce.

                                                 20
public policy. An injunction issued by a court acting within its jurisdiction must be obeyed until the

injunction is vacated or withdrawn.”) (citation omitted)).

       Moreover, a brief review of Appellants’ argument shows that it is without merit. The CPD

continued to operate under the ACD in 1993-1994 because it failed to hire 70 officers in two

different years during the initial 8-year period. Appellants are correct that, even with this two-year

shortfall, the average number of officers hired per year exceeded 70. Thus, under their argument,

the fact that the consent decree extended the 3:4 hiring ratio for two years even though the average

hires exceeded 70 per year shows that the decree was not narrowly tailored.

       Other than pointing to the average annual-hire figure, Appellants do not show how the 70-

officer minimum rule somehow violates the constitution. In fact, had the ACD included a rule

consistent with Appellant’s position – i.e., that the average number of hires per year should control

– such a rule could have plausibly extended the race-based hiring plan well beyond the original

expiration date, thus making the ACD more, not less, susceptible to attack on “narrowly tailored”

grounds.8,9

8
 To illustrate, assume 75 police officers were hired each year from 1985 to 1991; however, in
1992, assume the City placed a freeze on all officer hiring for one year. Also assume that from
1993 forward, the CPD again hired 75 officers per year. Under the ACD, the 3:4 hiring ratio
would have been extended for one year. Under a rule in which the average must be 70 in the
final year for the decree to expire, however, it would take 15 years, beginning in 1985, for the
average to reach 70.
9
  In his concurring opinion, Judge Rogers states that the ACD was “open-ended depending on the
actions of the City,” and concludes that “[a] consent decree that permits affirmative action as an
ongoing option, rather than as a limited requirement, should not so easily be maintained by relying
on the discriminatory situation faced at the time of the original decree.” In this particular case,
however, we find that the minimum-70-officer rule made sense for several reasons, including: (a)
the possibility of financial hardships causing the City to fall short in remedying past discrimination
(which was one of the reasons for amending the original consent decree); and (b) the possibility
(given the City’s prior history of discrimination) that the City might try to thwart the goals of the
consent decree by simply freezing hiring or hiring only a relatively few number of officers during
the eight-year period of the consent decree. See, e.g., McNamara v. City of Chicago, 

,
1223 (7th Cir. 1998) (noting hiring freeze by the city immediately after it agreed to hire a higher

                                                 21
2.      Duration of Race-Based Remedy

        Appellants next argue that the remedial measures extended too far in time in comparison

with the discrimination the measures were meant to remedy. Courts view with disfavor those

affirmative action plans that are not temporary and do not terminate when the identified racial

imbalances have been eliminated. See, e.g., 

; Detroit Police Officers Ass’n

v. Young, 

, 228 (6th Cir. 1993); Paradise v. Prescott, 

, 1531 (11th Cir.

1985). As this court has noted, “[n]arrow tailoring . . . implies some sensitivity to the possibility that

a [race-based] program might someday have satisfied its purposes.” Associated Gen. 

. It is also important to understand, however, that “[r]emedial action takes time . .

. and discrimination may linger for many years in an organization.” Boston 

(internal quotations omitted).

        To determine whether the length of any race-based hiring plan is properly attuned with the

workforce shortfall caused by discrimination, the reviewing court should consider the extent of the

shortfall and the reasonable likelihood that it will be ameliorated promptly. Unlike some other plans

struck down or questioned by this court,10 the ACD was not intended to be a long-term or permanent

remedy, as it contained a flexible, self-executing sunset provision. This provision ensured that the

decree would last no longer than necessary to remedy (or approach remedying) the workforce

minority shortfall caused by past racial discrimination. The decree further provided for an

percentage of minorities for fire department; when freeze ended, percentage of minorities remained
unchanged). While the reasonableness of the rule does not, by itself, make the decree narrowly
tailored, it is certainly a factor in its favor.
10
   See, e.g., Assoc. Gen. 

-38 (noting with disapproval the lack of a
sunset provision in legislation providing racial preferences in state contracts); 

(finding that the city’s failure to make any effort to limit the duration of remedy “cut[]
against” finding it was narrowly tailored).

                                                   22
automatic, temporary extension in the event that: (a) the CPD did not hire at least 70 officers in any

particular year; and (b) the 33% target had not yet been met. In this case, the plan continued for

slightly less than two years as a result of this provision. Thus, this is not a plan which was “timeless

in [its] ability to affect the future.” 

.

       Appellants point to the fact that in 1992, the percentage of minorities in the CPD (29.6%)

was closely approaching the 33% target. They assert that the CPD had substantially met the target,

and therefore the ACD’s continuation into 1993-1994 was unjustified. This argument would be

stronger if the workforce convergence had occurred in the beginning or middle of the plan’s

intended duration, rather than near the end. Of course, any responsible race-based hiring plan should

have as its ultimate goal the attainment of the target at the natural end of the life of the plan. The

fact that the percentage of minorities on the CPD’s police force was approaching the 33% target near

the end of the plan is a testament to the plan’s efficacy, not its unconstitutionality. Furthermore, the

fact remains that although the ACD was successful in increasing the percentage of minorities in the

CPD, the 33% target was not met until mid-1994. Once the target was met, the plan terminated of

its own accord and did not reach beyond the length of time necessary to alleviate the remaining

vestiges of identified discrimination.



3.      Flexibility and Impact on Third Parties

       Appellants next claim that the City’s affirmative action plan did not contain sufficient

flexibility. On the contrary, the ACD included several provisions giving the CPD flexibility in

hiring officers. First, the decree precluded the City from hiring unqualified minority applicants.

This is an important feature of any well-designed race-based hiring plan. 

(noting with approval that the affirmative action plan required “the hiring only of qualified blacks

                                                  23
and women; it [did] not require the selection of unqualified blacks or women over qualified white

males” (emphasis in original)). The City graded the examinations blindly, without reference to the

race of applicants. If qualified minority candidates were not available from a given list, the decree

provided that another entrance examination be offered and lists of qualified candidates be drawn

from that separate examination. Appellants have provided no evidence that the CPD hired

unqualified candidates in contravention of the decree.

       Nor did the decree mandate that the CPD had to hire a certain number of minorities each and

every year. The 70-officer minimum rule gave the CPD the option of hiring fewer than 70 officers

in a given year if, for example, there was an insufficient number of qualified candidates that year.

The CPD could then make up this shortfall in a later year, if necessary. This feature provided the

CPD with flexibility in meeting its hiring needs with only qualified candidates. See 

(noting with approval a similar measure).

       Furthermore, the decree did not require that incumbent non-minorities be terminated to make

room for minority applicants. Of course, “initial employment opportunities coupled with hiring

goals may burden some innocent individuals.” 

(quoting 

). This type of remedy, however, does “not impose the same type of intrusive injuries that

layoffs, which result in loss of job expectancy, security, and seniority, involve.” Id.; see also

(noting with approval that 1:1 hiring requirement did not require layoff

and discharge of non-minority employees); 

(finding that the denial of future

employment is not as intrusive as the loss of an existing job). Finally, non-minorities were not

barred from hire – the CPD maintained separate lists of qualified minority and non-minority

candidates and hired candidates from both lists. 

(noting with approval that

affirmative action plan did not “present a complete bar to non-minorities”). With these provisions,

                                                 24
the ACD contained sufficient flexibility with minimal impact on third parties to meet equal

protection concerns.



4.     Relationship of Numerical Target with Relevant Labor Market

       The court must also consider the relationship between the ACD’s numerical target and the

relevant labor market. The “relevant labor market” is defined as the number of minority applicants

who were actually qualified for the job – the “qualified labor pool.” For jobs requiring no special

qualifications, the relevant labor pool is often simply the pool of minorities within the local general

population. See 

. The CPD requires special qualifications for its police

patrol officers, however, such as passing entrance and physical examinations, psychological and

medical evaluations, and a background investigation. Accordingly, the court must look to the pool

of minorities actually qualified to be police patrol officers when comparing this pool to the 33%

target. See 

(noting that “where special qualifications are necessary, the relevant

statistical pool for purposes of demonstrating discriminatory exclusion must be the number of

minorities qualified to undertake the particular task”).

       The parties offer two different definitions of the qualified labor pool: (a) of those people who

qualified to take and actually took the examination, the percentage who were minorities (the City

defendants’ qualified labor pool); and (b) of those people who passed the examination, the

percentage who were minorities (Appellants’ qualified labor pool). While both proposed definitions

find support in the record, the distinction between the two proffered definitions is, in this case, one

without a difference. Even under Appellants’ stricter definition, there is a sufficiently close

relationship between the qualified labor pool and the ACD’s numerical target.



                                                  25
       There exists no set formula for determining whether the 33% target fits “close enough” to

the qualified labor pool. The question, rather, is whether the target represents “a plausible lower-

bound estimate of a shortfall in minority representation” among the CPD resulting from racial

discrimination. 

(citing, among other cases, this circuit’s decision in

). The parties provide minority examination flow data for four separate years:

1982, 1989, 1992, and 1994. The percentage of examination passers who were minorities range

from a low of 30.6% to a high of 47.3% (the percentage range of minority examination takers equals

41.0% to 52.3%). Specifically, the percentage of passers in 1982 who were minorities was 47.3%;

in 1989, 45.6%; in 1992, 30.6%; and 1994, 33.7%.

       Appellants point to the fact that in 1992, the percentage of passers who were minorities

(30.6%) was almost identical to the percentage of minorities on the police force (29.6%). The year-

to-year fluctuations in the qualified labor pool, however, are of little to no relevance to whether the

plan was narrowly tailored. While there may be long-run demographic trends which drive the

average of the percentage of examination passers who are minorities close to 33% (or, more likely,

a higher percentage equal to the percentage of minorities residing in the City), there is nothing to

suggest that the examination pass rate in 1992 had any direct or causal relationship to the minority

police force figure of that same year. The annual pass rates undoubtedly exhibited a certain element

of randomness from year-to-year, while the minority police force percentage was being driven to

33% by operation of the 3:4 hiring ratio.

       The better comparison in this case is between the ACD’s target of 33% and the range of

minority examination passers over the years. Given that the annual pool of qualified candidates

from which the CPD had to choose ranged between 30.6% to 47.3% minority candidates (using the



                                                  26
Appellants’ more conservative definition) over the relevant years, a workforce target of 33% was

sufficiently related to the qualified labor pool, if not a bit conservative.

        The fact that the 33% target was at the lower bound of the conservative qualified labor pool

further illustrates that the 29.6% minority workforce in 1992 was not “close enough” to the target

for the CPD to have ignored the terms of the ACD and stopped using the 3:4 hiring ratio in 1993-

1994. The standard is not whether the CPD had come close to satisfying the terms of the ACD, but

rather “whether the vestiges of past discrimination had been eliminated to the extent practicable.”

(quoting Bd. of Educ. of Oklahoma City Pub. Sch. v. Dowell, 

,

250 (1991)) (emphasis added). Certainly, the percentage of minorities in the CPD workforce was

increasing throughout the 1980s and early 1990s. Even so, the CPD had not yet reached the 33%

target, the level at which the parties agreed, the district court approved, and we confirm, was a

reasonable lower-bound measure of the relevant qualified labor pool. Within the context of the

ACD as a whole – a temporary plan, the City’s admitted history of discrimination, and a modest

target – we find that the CPD did not err in following the terms of the ACD in 1993-1994 even

though it was approaching the target in 1992.

        Finally, Appellants take issue with the 3:4 hiring ratio, arguing that the rate of hire (42.9%)

far exceeded the 33% target. As a matter of simply comparing the percentages, they are correct.

Yet, Appellants ask us to compare apples to oranges. The 33% figure represents the CPD police

force minority target; the 3:4 hiring ratio represents the speed to which the target was to be met. As

Section II.E.1, there is nothing unconstitutional about the 3:4 hiring ratio as a

reasonable rate of hire to meet the 33% minority workplace target. 

(approving 1:1, or 50%, hiring rate to meet 25% minority workplace target).



                                                  27
F.     The ACD Satisfies Equal Protection Concerns

       Given the modest workforce target, the sunset provision, and the various flexibility

provisions, we find that the ACD was narrowly tailored to remedy the City’s past discrimination

against minorities. The City defendants did not violate the Appellants’ rights under the Equal

Protection Clause or Title VII of the Civil Rights Act by following the ACD in 1993-1994.11

Accordingly, the district court properly granted judgment to the City defendants on these claims.



G.     Appellants’ Disparate Treatment Claims

       In addition to their claim that the ACD, as a whole, violated their constitutional and statutory

rights, several of the individual Appellants also argue that they faced disparate treatment by the City

defendants. Specifically, they argue that the CPD’s use of the 1:3 rule shielded less-qualified

minority applicants from competition against them. They also allege they were victims of reverse

discrimination because the City defendants used background checks and selection procedures

against them in a disparate manner.

       Appellants do not suggest here that any of the minority candidates hired in 1993 or 1994

were not qualified. They insist, rather, that for various reasons they were more qualified than some

of the minority candidates who were ultimately hired. As explained by the district court, however,

“[d]egrees of qualification” are not the proper focus here.

       As already discussed, rather than relying on a pure rank-order rule, the City hired officers

using the 1:3 rule in conjunction with separate minority and non-minority candidate lists. Thus,

within the framework of the 1:3 rule, minorities were compared against other minorities, and non-
11
  Although a public employer’s obligations under the Equal Protection Clause and Title VII are not
identical, 

, Appellants have not argued that their Title VII claims relating to
the ACD’s application in 1993-1994 differ in any way from their equal protection claims.

                                                  28
minorities were compared against other non-minorities.12          Any leniency afforded minority

candidates on the minority eligibility list would affect only other minority candidates, and not the

individual Appellants. It is only when one steps outside the 1:3 framework and views the ACD as

a whole that it can be said that Appellants were treated differently than minority candidates. Yet,

as explained above, the different treatment of minority and non-minority candidates under the

remedial hiring plan met constitutional muster.

       On Appellants’ claim of reverse discrimination based on the City’s background checks and

screening procedures, the district court correctly found that Appellants failed to show that the CPD

was “that unusual employer who discriminates against the majority.” Murray v. Thistledown Racing

Club, Inc., 

, 67 (6th Cir. 1985) (citation omitted). Appellants were not similarly situated

to minorities who were allegedly treated more favorably. Accordingly, the district court properly

granted summary judgment to the City defendants on Appellants’ disparate treatment claims.



                                       III. CONCLUSION

       For the reasons provided above, we AFFIRM the district court’s grant of summary judgment

in favor of the City defendants.




 This feature distinguishes the present case from Voels v. New York, 

12


(S.D.N.Y. 2002), relied upon by Appellants in their reply brief. The government actor in that
case also used a 1:3 hiring rule, but with the critical difference being that separate lists of
minorities and non-minorities were not maintained. Thus, in any group of three, minorities and
non-minorities were likely competing directly against each other.

                                                  29
       KAREN NELSON MOORE, Circuit Judge, concurring. I join in the majority’s opinion

but write separately to respond to the concurring opinion’s suggested narrowing of the category of

individuals entitled to relief due to past discrimination. The concurring opinion could be read to

state that only a plan benefitting those individuals who suffered discrimination themselves or who

were “around at the time the discrimination took place” will survive strict scrutiny analysis; it

further states that “a limited remedy of, for example, up to 10 years” would be permissible for the

latter group. Concurring Op. at 1-2. This part of the concurrence is based upon the following

statement from City of Richmond v. J. A. Croson Co., 

, 508 (1989): “But the interest

in avoiding the bureaucratic effort necessary to tailor remedial relief to those who truly have suffered

the effects of prior discrimination cannot justify a rigid line drawn on the basis of a suspect

classification.” (emphasis added). However, Croson does not imply that those who suffer the

effects of prior discrimination must have been “around at the time the discrimination took place.”

Such a conclusion would fail to acknowledge the key distinction between eliminating facially

discriminatory policies and the continuing effects of historically institutionalized racial

discrimination; it would also fail to acknowledge that ten years will often be an unrealistic amount

of time in which to eradicate the effect of years — often many decades — of discrimination. “As

long as significant specified effects linger, affirmative action may be justified despite the

implementation of valid selection procedures. Public employees cannot escape their constitutional

responsibilities merely by adopting facially-neutral policies that institutionalize the effects of prior

discrimination and thus perpetuate de facto discrimination.” Ensley Branch, NAACP v. Seibels, 

, 1575 (11th Cir. 1994) (citing United States v. Fordice, 

, 729 (1992)).




                                                  30
       ROGERS, Circuit Judge, concurring. I concur entirely in the majority opinion. I write

separately to acknowledge two troubling aspects of our decision, and to identify an additional

consideration supporting our judgment.

       Between 1972 and 1977, the City defended a lawsuit in which the Shield Club argued that

the City had discriminated in its hiring, transfers, and promotions of minority police officers. The

City maintained throughout that it had never discriminated against minorities. Nevertheless, after

years of litigation and findings of intentional discrimination by the district judge, the City entered

into a consent decree governing hiring and promotions within its police department.

       In 1984, the Shield Club moved to modify and extend the consent decree. The City opposed

the motion, but it ultimately agreed. A decade later, despite faithfully following the terms of the

ACD, the City once again was forced to defend against allegations of racial discrimination when the

appellants brought the present lawsuit in May 1994, just one month before the ACD reached its goal

in June 1994. This time, however, the City was targeted not for discriminating against minorities,

but for discriminating against nonminorities by following the terms of the ACD. The plaintiffs

argued that the ACD should not have been extended beyond 1992, and thus that the City had been

acting illegally for the subsequent year and a half in complying with the ACD.

       Had this case instead been brought in 1992 to modify or terminate the consent decree, the

reasoning of the majority opinion might well have supported continuation of the decree. Two

factors, however, give me pause in reaching that conclusion.

       First, the ACD arguably had been in place too long to be narrowly tailored. In an affirmative

action plan seeking to remedy past discrimination, there are essentially three possible groups that

stand to benefit: (1) those actually discriminated against, (2) those not actually discriminated against

but who are members of the same group as those discriminated against and who were around at the

                                                  31
time the discrimination took place (justifying a limited remedy of, for example, up to 10 years), and

(3) other members of that group at any time (as an extreme example, 75 years after the

discrimination). In this case, the minority recruits are mostly members of the third group because

few of them would have been old enough to be hired as a police officer when the discrimination took

place in the 1970s.

       In City of Richmond v. J.A. Croson Co., the Supreme Court’s analysis brings into question

whether a plan benefitting only group three survives strict scrutiny:

       [T]he Richmond Plan’s waiver system focuses solely on the availability of MBE’s;
       there is no inquiry into whether or not the particular MBE seeking a racial preference
       has suffered from the effects of past discrimination by the city or prime contractors.


       Given the existence of an individualized procedure, the city’s only interest in
       maintaining a quota system rather than investigating the need for remedial action in
       particular cases would seem to be simple administrative convenience. But the
       interest in avoiding the bureaucratic effort necessary to tailor remedial relief to those
       who truly have suffered the effects of prior discrimination cannot justify a rigid line
       drawn on the basis of a suspect classification.

, 508 (1989) (emphasis added). In addition, Justice Stevens noted:

       The class of persons benefitted by the ordinance is not, however, limited to victims
       of such discrimination—it encompasses persons who have never been in business in
       Richmond as well as minority contractors who may have been guilty of
       discriminating against members of other minority groups. Indeed, for all the record
       shows, all of the minority-business enterprises that have benefitted from the
       ordinance may be firms that have prospered notwithstanding the discriminatory
       conduct that may have harmed other minority firms years ago.

(Stevens, J., concurring). Similarly, the minority recruits that benefitted from the ACD

extension were hired even though they may have never suffered from the effects of racial

discrimination. Of course, on the other hand, it is impossible to know who would have been hired

under a nondiscriminatory system when, during the pendency of the consent decree, a separate

                                                  32
remedial quota system was in place. Perhaps some of the young minority beneficiaries of the ACD

in the early 1990s who would not have been hired without the ACD would have been hired had there

never been any discrimination.

        Second, the consent decree was open-ended depending on the actions of the City. By

continually hiring fewer than seventy police officers per year, the City could have extended the ACD

indefinitely. A consent decree that permits affirmative action as an ongoing option, rather than as

a limited requirement, should not so easily be maintained by relying on the discriminatory situation

faced at the time of the original decree. The reasonableness of the 70-per-year rule in terms of

protecting minorities from city evasion of consent decree requirements does not necessarily imply

that the rule is narrowly tailored for purposes of protecting the interests of nonminority applicants

adversely affected by the decree.

        While these considerations may in the end not have been sufficient to warrant modification

of the ACD two years before it reached its goal, they at least raise serious concerns as to whether

the decree continued to be narrowly tailored at that point. We do not need to resolve the issue

because, in my view, a somewhat more deferential approach is required when, instead of moving

to modify the decree, the officers sued after the fact for retroactive relief based on the failure of the

City to violate the decree.

        To award damages against the City because it failed to violate a consent decree would be

troubling for a number of reasons. First, it would be inequitable to force the City to pay damages

for faithfully implementing the affirmative action plan that it had challenged for years in court. The

case would be different if the City entered into a consent decree that was clearly illegal at the time,

or if the City subsequently ignored a new decision or statute that made it clear that its conduct was

illegal. But when the City entered into the ACD, the Supreme Court had yet to announce in Croson

                                                   33
that strict scrutiny should be used in evaluating state and local affirmative action programs. 

. There is little doubt that, in 1984, the affirmative action program was legal under

existing caselaw. Even after United States v. Paradise, 

(1987), Croson, and

subsequent cases were decided, it was not immediately clear that the extension of the ACD was

unconstitutional.

       Second, permitting retrospective relief would place parties like the City in a problematic

legal position. A party is bound by a court order even if the party strongly disagrees with the order.

If we award damages for failure to violate a court order, we are in some sense telling the party that

it should have violated the order. Perhaps when compliance with a court order clearly violates the

constitutional rights of third parties, a party should violate the order and use its legal position to

defend the enforcement of contempt sanctions against it. But where the illegality of the order is not

so clear, certainly a party ought to comply. This shows the respect for the system that is necessary

on the part of responsible participants in our ordered republic.

       Instead of violating a court order perceived to be illegal, a party should generally litigate to

modify the order. One might certainly argue that the City should have sought a modification of the

ACD as soon as the argument became tenable that the decree was not narrowly tailored. But such

an obligation to seek to modify cannot require a litigant (here the City) to challenge a decree

continuously during its pendency. Such an obligation also arguably disrespects the system,

especially where the judicial order resulted from vigorous litigation, as here, and where the change

in factual and legal circumstances was not clearly marked. Thus, for us to reverse the district court

in this case and impose retrospective relief for failure to violate the decree, we should determine not

only that the decree would properly have been modified had modification been timely sought, but

also that the unconstitutional nature of compliance with the decree was sufficiently clear to the City

                                                  34
as to require the City to have returned to court in 1992 to lift the decree. In my view, for the reasons

given in the majority opinion, we cannot say that the ACD was so clearly unconstitutional as to

require the City sua sponte to have sought modification in 1992, and we need not decide the closer

question of whether such a modification would have been required had it been sought by the officers

at the time.




                                                  35