Full opinion text
MEMORANDUM AND ORDER APPROVING CONSENT DECREE PATEL, District Judge. This employment discrimination action was originally brought by the United States against the City and County of San Francisco (“the City”) citing the illegal practices of the San Francisco Fire Department (“SFFD”). Various individuals and organizations have since intervened as plaintiffs and San Francisco Firefighters Local 798 (“the Union”) has intervened as a defendant. After over four years of litigation, following a change in the political administration of the City which necessitated the postponement of trial, plaintiff-inter-venors and the City reached a settlement of both the individual and the class-wide claims of discrimination against women and minorities and memorialized it in a proposed consent decree. The United States declined to participate in the negotiations leading to settlement. The Union, although it took part in settlement negotiations, objects to the terms of the decree and did not sign the agreement. The case is now before the court for approval of the consent decree. Approval of the decree is opposed by the United States and by the Union. The court has also considered the objections filed by members of the public in opposition to the provisions of the decree. In addition, a fairness hearing was held on December 22, 1987 at which counsel for all the parties argued the merits of the decree and the court entertained the statements of various firefighters both for and. against the decree. Having considered the submissions of the parties and the objections filed with the court and raised at hearing, for the following reasons, the court approves the consent decree in settlement of all class-wide and individual claims for relief raised both in this action and in the concurrent state action, City and County of San Francisco v. Fair Employment and Housing Commission [FEHC], No. A024145. The Stipulation of Undisputed Facts [hereinafter “Stipulation”] filed August 15,1986 and the Statement of Undisputed Facts [hereinafter “Facts”] filed November 12, 1987 are deemed facts found by the court and incorporated herein except as otherwise noted. BACKGROUND The SFFD hired no Black firefighters before 1955, allowed no women to apply before 1976 and hired no women until August 1987. Facts paras. 5, 2 and 1. It is unsurprising then that this action is not the first challenge faced by the City to the employment practices of the SFFD. This litigation must be understood in conjunction with an earlier federal suit, Western Addition Community Organization [WACO] v. Alioto, C 70-1335 WTS and a concurrent state action, City and County of San Francisco v. Fair Employment and Housing Commission [FEHC] No. A024145. The WACO suit challenged only the SFFD’s entry-level hiring practices and only on grounds of racial discrimination. The FEHC suit challenged only promotional practices. The current litigation addresses both entry-level and promotional practices as well as problems of both gender and race discrimination in employment. I. Previous Federal Litigation: The Entry-Level Test Although Blacks represented 14% of the City’s population in 1970, of the 1800 uniformed firefighters employed by the SFFD only four were Black. Western Addition Community Org. [WACO] v. Alioto, 330 F.Supp. 536, 538 (N.D.Cal.1971) [hereinafter “WACO /”]. That year the NAACP in concert with several community groups filed a class action suit in this court against the City pursuant to the Civil Rights Act, 42 U.S.C. §§ 1981 and 1983, to challenge the validity of the format used for the 1968 H2 entry-level firefighter test. WACO I, 330 F.Supp. at 537. The 1968 H2 test determined eligibility for entry-level positions in the SFFD. The test comprised six components: physical measurement; athletic ability; medical examination; qualifications appraisal; employment, character and background check; and a written examination. Success on the first five components determined that an applicant was qualified. Only the written examination score determined an applicant’s rank on the Civil Service list of eligi-bles. Id. at 538. According to Civil Service rule, hiring was by rank order from the eligibility list. The WACO plaintiffs challenged only the written component of the 1968 H2 test. The WACO plaintiffs charged that the 1968 H2 written exam had an adverse impact on minority applicants without having any relationship to the requirements of the job. See Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). The results of the written portion of the 1968 H2 test were as follows: Tested Passed Pass Rate Total 1883 662 35% Black 101 12 12% Hispanic 69 24 35% Other 1713 626 37% See WACO I, 330 F.Supp. at 538; Facts para. 13. Thus, the pass rate for Black applicants on the 1968 H2 written exam was about one-third that of White applicants. Facts para. 13. In 1971 Judge Sweigert found that the plaintiffs had demonstrated that the 1968 H2 test had an adverse impact on minorities and that the City had failed to carry its burden to show any connection between the qualities tested on the written exam and the job requirements of an entry-level firefighter. WACO I, 330 F.Supp. at 539-40. In defense of the exam, the City responded only that the SFFD had made efforts to recruit minority applicants for the 1968 H2 test. The court ruled that “whatever may have been the good intentions of defendants, there is a prima facie case for predicating employment discrimination.” Id. at 539. A revised H2 entry-level test was administered in September 1971. The 1971 H2 test comprised three parts given in two stages. The first stage required the applicant to pass a written examination by answering 90 out of 130 questions correctly. If the applicant passed the written exam, he then had to pass an athletic ability exam and an oral exam. The results of the second stage exams determined an applicant’s rank on the eligibility list. Western Addition Community Org. [WACO] v. Alioto, 340 F.Supp. 1351, 1352 (N.D.Cal.1972) [hereinafter “WACO II”]. The WACO plaintiffs again came before the court in 1972 seeking to enjoin, on the grounds of adverse impact, the use of the resulting eligibility list. The results of the written portion of the 1971 H2 test were as follows: Tested Passed Pass Rate Total 1741 831 48% White 1187 676 57% Black 333 67 20% Hispanic/Asian 221 88 40% WACO II, 340 F.Supp. at 1353. Thus, instead of eliminating adverse impact, the 1971 H2 test only perpetuated the 1968 H2 test’s approximately three-to-one passing advantage of White over Black applicants. The court held that the 1971 H2 entry-level test had an adverse impact on minority applicants and was not shown to be job-related. Specifically, the court found that the City was unable to validate the 1971 H2 test because it had not, in the first instance, conducted a job analysis. Because there was no way of knowing what the job required, there was no way to determine whether the test was reasonably related to job requirements. See WACO II, 340 F.Supp. at 1356. After noting that the City’s own experts had “virtually confessed” that validation was impossible and affirming the City’s good intentions while bemoaning its failure “to comply with the fairly well-established requirements of the law in this field,” the court enjoined the use of the 1971 H2 entry-level firefighters exam and the resulting eligibility list. WACO II, 340 F.Supp. at 1356; Facts para. 14. In 1973 the City presented the court with a new written test, which it claimed to have properly validated. The court found that although the City had conducted a job analysis substantially in accordance with federal guidelines, it had failed entirely to validate the written test empirically (as required by the guidelines if feasible) and further that it had not shown the adequacy of the attempted content validation. Western Addition Community Org. [WACO] v. Alioto, 360 F.Supp. 733, 739 (N.D.Cal.1973) [hereinafter “WACO III”]. Judge Swei-gert refused to approve the written exam and issued the following admonishment to the City: The time has come — especially after the City has twice failed to satisfactorily demonstrate the validity of its Fireman H-2 written examination tests — for the City to seriously, rather than superficially and speciously, face up to the problem presented to it by new federal law by further efforts to rebut or correct the already legally established presumptive discriminatory effect of its selection procedures. WACO III, 360 F.Supp. at 739 (emphasis in original). The court declined at that time to order affirmative race-conscious relief. It did order, however, that the written portion of the proposed test not be used for cut-off or ranking purposes and also ordered the City to proceed apace to validate empirically either the test under consideration or a new test. WACO III, 360 F.Supp. at 740-41. In June 1973, the WACO plaintiffs moved for and were granted affirmative relief. Western Addition Community Org. [WACO] v. Alioto, 369 F.Supp. 77 (N.D.Cal.1973) [hereinafter “WACO IV”], appeal dismissed as moot, 514 F.2d 542 (9th Cir.) (per curiam), cert. denied, 423 U.S. 1014, 96 S.Ct. 446, 46 L.Ed.2d 385 (1975). The court found that the City had failed three times to demonstrate that the challenged H2 tests were job-related. The appropriate remedy for such repeated failures to comply with the law and with court orders was affirmative relief “even though it cannot be established that the discriminatory practices were intentional or invidious.” WACO IV, 369 F.Supp. at 79. The court directed its exasperation at the City’s Civil Service Commission (“the Commission”), pointing out that the Commission’s failure to comply with the order in WACO III had resulted in the SFFD being over eleven percent understaffed. The Commission’s dilatoriness in these matters and apparent stubborn insistence upon arguments and alternatives which this court has repeatedly found unacceptable, have created an intolerable situation; the adequacy of Fire Department manpower for the safety of the City is coming into question; all Fireman H-2 applicants, both minority and non-minority, have been kept in a state of uncertainty for several years and the rights of minority applicants to a more prompt correction of the City’s still unvalidated selection procedures are involved. WACO IV, 369 F.Supp. at 80. Judge Swei-gert ordered one-for-one White to minority hiring from the 1973 H2 list (identified as list E-25), regardless of rank, until the list of qualified minority applicants (consisting of those who had answered at least 50 of the 100 questions on the written exam correctly and passed the oral and athletic tests) was exhausted. Id. at 80-81; Facts para. 27. In early 1974, the SFFD substantially modified the ongoing evaluation of participants at the Fire College. The Fire College, established in 1957, provides several weeks of required training and practical instruction to newly appointed H2 firefighters. Between 1957 and 1972, only two firefighters were ever dismissed based on inadequate performance at the Fire College. In 1974, however, the Fire College program was revised to include weekly testing of ability to manipulate fire equipment and a procedure was instituted for dismissal of those who failed the program. Stipulation paras. 56-61. The WACO litigation was terminated by consent decree in May 1977. Facts para. 31. The decree was to be in effect for five years. The decree reiterated the court’s earlier finding that the plaintiffs had made out a prima facie case of race and national origin discrimination. In addition, the decree noted the court’s finding that the City had failed to carry its burden to explain the disparate impact of its selection procedures upon minority applicants. See Western Addition Community Org. [WACO] v. Alioto, No. C-70-1335 WTS, slip op. at 2 (N.D.Cal. May 18, 1977) [hereinafter “WACO V”]. The decree established guidelines for the use of exam components from the newly announced (1976) H2 test. It included no provision for affirmative relief on the grounds that “[i]n view of the agreement of the parties and the defendants’ efforts of improving the selection process for entry-level firefighter, there is no need for nor would it be appropriate to impose a further ratio or quota hiring order upon defendants.” WACO V, slip op. at 9. The decree did state as a goal of the parties, however, that the H2 eligible list should reflect 40% minority representation. Id. at 9-10. The 1976 H2 test used the 1973 written exam with a cut-off score of 70% and a weight of 40% in ranking for the eligibility list. In addition, applicants were required to pass a physical agility test (“PAT”) that was weighted 60% in ranking. According to Civil Service rule, appointments from the resulting list of eligibles (list E-26) were to be made in rank order. Facts para. 30. The pass rates for the written portion of the 1976 H2 test were as follows: Tested Passed Pass Rate Total 3386 2642 78% White 1995 1761 88% Black 562 306 54% Hispanic 462 319 69% Asian 289 202 70% Am. Ind./Other 78 54 69% See Facts para. 32. The rate of Black to White applicants passing the 1976 H2 test was 61%, the rate of Hispanic to White applicants passing the test was 78% and the rate of Asian to White applicants passing the test was 80%. In 1978, the WACO plaintiffs raised an objection to the use of the 1976 test. Judge Sweigert rejected their objection because the City had comported with the express terms of the consent decree. See Western Addition Community Org. [WACO] v. Alioto, No. C-70-1335 WTS, slip op. at 2 (N.D.Cal. June 9, 1978) [hereinafter “WACO VI”]. As of November 1987, fourteen years after Judge Sweigert’s 1973 order granting affirmative relief and ten years after the entry of the WACO consent decree, over 55% of the minority firefighters working in the SFFD had been hired pursuant to Judge Sweigert’s 1973 order. Facts para. 28. Thus, while the WACO consent decree ameliorated some of the disparate impact on minorities of the SFFD’s entry-level hiring practices, it was by no means a complete resolution. It is therefore no surprise that the City has since had to defend the hiring and promotional practices of the SFFD before both state and federal courts. II. Prior State Court Litigation: The Promotional Test All positions above the rank of H2 firefighter (except that of Chief) are filled by promotion from within the SFFD. Of the 352 permanent supervisory positions in the SFFD, none are held by women, none are held by Asians, four (1%) are held by Blacks and 16 (5%) are held by Hispanics. Only one member of a minority group holds a rank above captain. See Facts paras. 7-9. The first rank in the SFFD above H2 entry-level firefighter is H20 fire lieutenant. Before 1982, the H20 test for promotion to the rank of lieutenant consisted solely of a written job knowledge exam. Facts para. 33. The 1978 lieutenant’s exam tested for knowledge of firefighting strategy and technique, arson detection, equipment maintenance, rescue and emergency medical procedures, and report writing. See City and County of San Francisco v. Fair Employment and Housing Comm’n [FEHC], 191 Cal.App.3d 976, 980-81, 236 Cal.Rptr. 716 (1987) [hereinafter “FEHC”]. Rank on the eligibility list (which listed only the top 250 applicants) was determined by a combination of an applicant’s score on the written exam, his seniority, his record of meritorious service and his “clean record” points. Facts para. 33. For the 1978 H20 test, an applicant’s score on the written exam counted 88% and the remaining factors counted 12% in determining rank on the eligibility list. Facts para. 34. The results of the 1978 H20 test for fire lieutenant were as follows: Pass Apmt Tested Passed Rate Aptd Rate Total 577 250 43% 153 27% White 473 226 48% 140 30% Black 34 6 18% 3 9% Hispanic 54 17 31% 10 19% Asian/Other 16 1 6% 0 See Facts para. 35. Thus, the ratio of the pass rate of the minority subgroup to the pass rate of White firefighters on the 1978 H20 test was 38% for Black firefighters, for Hispanic firefighters it was 65% and for Asian/Other firefighters it was 13%. Between June and September of 1980, Black firefighters filed ten complaints with the California Department of Fair Employment and Housing (“the Department”) alleging that the SFFD’s 1978 H20 test discriminated on the basis of race. Facts para. 36. The Department issued a formal complaint and an administrative hearing ensued. FEHC, 191 Cal.App.3d at 981, 236 Cal.Rptr. 716. The Fair Employment and Housing Commission (“the FEHC”) determined on the basis of the hearing record that the 1978 H20 test had an adverse impact on minority applicants and that the City had failed to show that the test was sufficiently job-related to be valid. Id. at 982, 236 Cal.Rptr. 716. Although the City obtained a writ of administrative mandamus from a California superior court vacating the decision of the FEHC, the Court of Appeal subsequently concluded that the FEHC’s findings were supported by substantial evidence and reversed the superior court’s judgment. Id. at 982, 990, 994, 236 Cal.Rptr. 716. The decision of the Court of Appeal became final and this court has held that the decision has preclusive effect in this litigation. Order of May 25, 1988 (confirming bench order of October 21, 1987). The Court of Appeal held that the Department had established a prima facie case of discrimination by demonstrating a significant disparity between the passing rates of White and Black applicants. The significance of the disparity in passing rates was determined by reference to the Uniform Guidelines on Employee Selection Procedures (“the Uniform Guidelines”), 29 C.F.R. § 1607.1 et seq. FEHC, 191 Cal.App.3d at 986-87, 236 Cal.Rptr. 716. The rule of thumb under the Uniform Guidelines provides that a selection rate (among minority applicants) of less than 80% of the rate of the group with the highest selection rate (usually White applicants) is considered to be strong evidence of adverse impact. 29 C.F.R. § 1607.4(D). The 38% passage rate of Black as compared with White firefighters on the 1978 H20 test was obviously far below the 80% rate suggested by the Uniform Guidelines. The Court of Appeals found the likelihood that the adverse impact was happenstance to be slight. FEHC, 191 Cal.App.3d at 987, 236 Cal.Rptr. 716. The court rejected the arguments raised by the City to account for the disparity in passing rates. First the City argued that job experience rather than race accounted for most of the disparity in passage rates of Black and White applicants. The court found that the Department’s statisticians had accounted for differences in seniority, service as a temporary lieutenant and prior simulator training experience only to find that these factors could not explain the disparity in passing rates. FEHC, 191 Cal.App.3d at 988-89, 236 Cal.Rptr. 716. The City also tried to argue that the Black applicants were less capable than their White counterparts because 73% of them had been hired pursuant to the WACO IV hiring order. See 369 F.Supp. at 81. The court found that this argument was not only without factual support, it was in fact contradicted by evidence that all the applicants were highly experienced and had demonstrated success by surviving a stringent training program. FEHC, 191 Cal.App.3d at 989, 236 Cal.Rptr. 716. Finally, the court dismissed the City’s argument that the Black candidates had failed to prepare adequately for the test. Id. Once the court determined that there was substantial evidence to support the FEHC’s finding of adverse impact, it turned to the question of whether the City had demonstrated that the test was job-related. It concluded that there was substantial evidence to support the FEHC’s findings that the 1978 H20 test was unrelated to the job and wholly unvalidated. Id. at 990, 236 Cal.Rptr. 716. The FEHC had found that although supervision was the primary component of the fire lieutenant’s job, ability to supervise was nowhere listed in the job description, nor was it tested for in the 1978 H20 test. The City had presented no evidence demonstrating a correlation between higher test scores and job performance. Id. The Court of Appeal pointed out that such empirical validation could have been obtained easily because many of the questions on the 1978 H20 test had been taken from earlier H20 tests. Id. at 990-91, 236 Cal.Rptr. 716. The FEHC concluded that written tests are an inappropriate measure of supervisory skills and the Court of Appeal agreed. Id. at 990, 236 Cal.Rptr. 716. Recognizing the overlap between the state case and this litigation, the Court of Appeal sought to avoid the possibility of inconsistent awards and so vacated the relief ordered by the FEHC and deferred relief pending the outcome of this litigation. Id. at 944, 236 Cal.Rptr. 716. III. Current Litigation: The H2, H20 and H4 Tests All regular entry-level hiring in the SFFD between the time of the termination of the WACO litigation and the filing of this action was carried out pursuant to lists E-25 and E-26, under the supervision of the WACO court. Facts paras. 18, 27, 31; WACO V, slip op. at 10. The SFFD administered a new H2 entry-level test in 1982-83, and a new H20 fire lieutenant promotional test in 1984. Also in 1984, the SFFD administered an H4 fire inspector promotional test. After the results of the 1982-83 H2 test were published, the United States and plaintiff-inter-venors brought separate suits in this court under Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., and the State and Local Fiscal Assistance Act of 1972, 31 U.S.C. § 6701 et seq., for the City’s failure to correct the effects of past discrimination and for its continued use of invalid hiring procedures that had an adverse impact on women and minorities. The United States sought relief for Black, Asian and Hispanic applicants. The plaintiff-intervenors sued on behalf of women and Blacks and added claims for racial harassment. Subsequently, Asian and Hispanic intervenors joined the action. In addition, there were individual claims for relief. The cases were consolidated in 1986. The consolidated complaints focused on current hiring practices, specifically on the adverse impact of the 1982-83 H2 test for entry-level firefighter, the 1984 H20 test for promotion to fire lieutenant and the 1984 H4 test for promotion to fire inspector. The plaintiff-intervenors also focused on charges of racial harassment. In February 1986 this court preliminarily enjoined the City from hiring from the E-27 list generated by the 1982-83 H2 test. See Order of February 12, 1986 at 15. The 1982-83 H2 test consisted of a written exam and a physical agility test (“PAT”). The passing score on the written test was 60%. The written portion was to count 30% and the PAT was to count 70% toward ranking. After this suit was filed, the City re-weighted the test, considering the written exam only as qualifying test without ranking weight and giving the PAT 100% weight as a ranking device. Facts para. 45. The E-27 list consisted of 190 applicants in rank order. Facts para. 44. The results of the 1982-83 H2 written exam were as follows: Tested Passed Pass Rate Total White Black Hispanic Asian 05 05 cn *£> -3 ^ cn co oo h- to to CO CO 05 05 CJT CO -3 05 OO — JOOM to CO H 05 CO C- O OO OO 05 CO OO CP CO See Facts para. 47. The passing rate on the 1982-83 H2 written exam for Black applicants was 59%, for Hispanic applicants was 74% and for Asian applicants was 69% of the passing rate for White applicants. Only those who passed the written test were permitted to take the PAT test. The results of the 1982-83 H2 PAT test were as follows: Tested Passed Pass Rate Total 2061 1286 62% White 1332 886 67% Black 282 173 61% Hispanic 245 138 56% Asian Women 79 18 23% Men 1982 1268 64% See Facts para. 48. Compared with White applicants, the passing rate on the 1982-83 H2 PAT for Black applicants was 91%, for Hispanic applicants was 84% and for Asian applicants was 73%. The passing rate for women was 36% of that for men. Of the top 190 scorers, however, which were the applicants certified to the eligibility list, 66% (125) were White, 19% (36) were Black, 5% (9) were Hispanic, 8% (15) were Asian and none were women. Facts para. 49. In June 1986 this court granted partial summary judgment on the claims that the 1982-83 H2 written exam adversely impacted Blacks and that the 1982-83 H2 PAT adversely impacted women. See Order of June 23, 1986 at 8-9. On the eve of trial in October 1986, the City declared that it would not defend the validity of the H2, H20 and H4 tests. Later that month the Commission cancelled the H2 and H20 tests along with all eligibility lists. In early November the Commission did the same with respect to the H4 test. In response to the City’s withdrawal of the tests and its decision to give up defending their validity, the United States and plaintiff-intervenors moved for summary judgment. In February 1987 the court granted plaintiffs’ motion for summary judgment, issued a permanent injunction and provided for interim hiring. See United States v. City and County of San Francisco, 656 F.Supp. 276 (N.D.Cal.1987) [hereinafter “Davis /”]. The court found, in light of the City’s admissions, that “the H2, H4, and the H20 selection and promotion procedures fall within the definition of employment practices proscribed by Title VII. These facts clearly establish a prima facie violation of law_” Davis I, 656 F.Supp. at 282. Further, the court determined that the City’s withdrawal of the H2, H20 and H4 tests did not bar the award of injunctive relief, “particularly ... in light of the City’s plans to immediately develop a new series of hiring and promotional examinations. Given the City’s historical inability to develop valid employment procedures, the possibility of future violations must be precluded by assuring that the City’s new examinations finally satisfy the requirements of Title VII.” Id. at 287 n. 14. The court rejected the argument that by withdrawing the challenged tests, the City had “immunized itself from the future use of discriminatory hiring practices.” Id. at 287. [T]he City has only deepened cause for concern by conceding that it has failed once again to develop valid employment procedures for hiring and promotions within the Fire Department, thus adding yet another chapter to an unenviable record of behavior reaching back nearly two decades. For the court to ignore the SFFD’s consistent and ongoing inability to utilize nondiscriminatory employment procedures, and to decline to enter judgment and award injunctive relief to assure that this sorry history will finally come to an end, would be inexcusable. Id. The court then issued a permanent injunction, large portions of which have been incorporated in the consent decree. Compare Davis I. 656 F.Supp. at 289-292 with Consent Decree Ex. A. The injunction, which remains in effect for five years, prohibits violations of Title VII by the SFFD or its toleration of such violations by its employees; requires the institution of recruitment and training programs; requires new test development according to the Uniform Guidelines; sets out a schedule for test development deadlines; requires semiannual progress reports to the court; and requires record keeping. Davis I, 656 F.Supp. at 289-92. Finally, pending the development of a new H2 entry-level firefighter test, the court provided for provisional interim hiring to avoid a staffing emergency that might endanger the public welfare. Id. at 292-93. With one exception, this court never received evidence or made any findings with respect to plaintiff-intervenors’ charges of racial harassment. The City and plaintiff-intervenors, however, have stipulated that the SFFD’s procedure for handling charges of racial harassment is different from that used for other rules violations. Facts para. 54. In addition, the parties stipulated that prior to October 1986 only two White firefighters had ever been suspended as a result of a complaint of racial harassment filed by a Black firefighter. Facts para. 55. Although no evidence was ever received at a full trial regarding plaintiff-inter-venors’ charges of racial harassment, the court did on one occasion receive sworn testimony regarding incidents of racial harassment. In January 1988 plaintiff-in-tervenors moved the court to find the City in contempt of court for failing to comply with the term of the permanent injunction forbidding racial harassment. See Davis I, 656 F.Supp. at 289-90 (permanent injunc-tive relief para. 1). Specifically, the City was charged with failing to investigate or punish members of the SFFD for incidents of racial harassment and encouraging harassment of minorities by maintaining an attitude of tolerance toward such incidents. The catalyst for the motion was widely reported incident in which two minority firefighters (one Asian and one Black) returned to their firehouse desks to find a swastika hanging from a nearby wall. At the court’s request, the court-appointed monitor filed under seal a report on the swastika incident as well as other recently reported incidents of racial harassment. Before the hearing on the contempt motion, the report was made available only to the court, counsel for the City and plaintiff-in-tervenors, and the Chief. The SFFD’s Chief Edward Phipps testified at a hearing on the motion on January 14, 1988. In response to questioning by attorneys for plaintiff-intervenors, the City and the Union, as well as by the court, Chief Phipps explained the procedure for dealing with charges of harassment and testified that it was his duty to investigate charges of discrimination and racial harassment. The Chief admitted, however, that he was unfamiliar with both the SFFD’s internal reports and the report of the monitor regarding the incidents in question. Further, the Chief said that although he had the power to suspend immediately, he had never done so, preferring to counsel firefighters regarding their behavior and wait for an official report. The Chief admitted that the SFFD had not committed sufficient resources to enable it to investigate quickly and punish charges of racial harassment. The Chief could not recall whether he had ever imposed discipline in response to a racial grievance. The Chief admitted he had heard reports that members of the SFFD had taken up collections and compensated firefighters suspended for disciplinary infractions. He could not say whether storing as opposed to displaying a swastika in a firehouse would warrant disciplinary action. Finally, the Chief expressed less concern about the posting of a racially inflammatory letter at one of the firehouses than about whether it was written on-duty or off-duty. The court found that with respect to the treatment of minority members, the SFFD was “out of control.” Further, the court found that the SFFD was clearly lacking either the willingness or the ability to carry out the court’s injunction and the SFFD’s internal order prohibiting racial harassment. The court found that the delays in investigating charges of harassment suggested an atmosphere of toleration for such incidents. The court issued an injunction which holds all members of the SFFD above the rank of lieutenant personally responsible for implementation of the SFFD’s anti-harassment policy as embodied in its General Order of October 1987 and as incorporated in the court’s order. See Order filed February 5, 1988 [hereinafter “Davis //”], notice of appeal filed March 7, 1988. The court ordered these persons to ensure, within ten days, that each firehouse under their supervision was inspected and that all racially or sexually inflammatory materials were removed and destroyed. Davis II, slip op. at 4. Failure to comply with provisions of the order will subject these supervisors to a finding of contempt and the imposition of a fine or imprisonment or both. Id. The order also provides specific time limits for processing and acting on complaints of harassment. Id. at 4-5. Finally, the order provides that no person disciplined by the SFFD shall be reimbursed for any resulting loss of pay. Id. at 5. The injunction is to remain in effect until this court orders otherwise upon determining that the SFFD has shown that it is taking responsibility for compliance with the civil rights laws as seriously as its firefighting responsibilities. THE CONSENT DECREE The consent decree, filed with the court on November 12, 1987, is attached as Appendix A to this order and will not be detailed here. For purposes of discussion, however, it will be helpful to outline briefly its broad provisions. The stated purpose of the consent decree is “to eradicate the present effects of past employment practices with respect to minorities and women.” Consent Decree (“C.D.”) at 8. The consent decree satisfies all claims of private plaintiffs and plaintiff-intervenors brought in this action. Id. at 5. The decree incorporates, as modified by the parties, relevant portions of the injunc-tive relief awarded by this court in Davis I. C.D. at 6 and Ex. A. The decree provides that the City shall continue to require that all applicants for the position of H2 entry-level firefighter be bona fide residents of the City and County of San Francisco at the time of and for a period of four months prior to the time of their application and through the date of their appointment. The parties specify that the City and County of San Francisco is the relevant geographic labor market. Id. at 6. The decree provides for the constitution of and specifies the powers and duties of a test development committee. Id. at 7. The decree sets out long term hiring goals of 40% minority and 10% female representation. Such goals are made subject to the availability of qualified applicants. To help assure a pool more accurately reflective of female and minority representation in the community served, the consent decree provides for specific recruitment efforts. Id. at 8-9. Failure to meet a goal cannot be justified on the basis of an approved test unless the test comports with the Uniform Guidelines on Employee Selection Procedures and is shown to eliminate or minimize adverse impact consistent with those guidelines. Id. at 8. The provisional firefighters hired pursuant to Davis I are to be given a special test no later than October 1, 1988. Those who pass will receive permanent appointments retroactive to their hiring date. C.D. at 9. All hiring goals are targets. Failure to meet a goal must be justified to the court. The goals presuppose hiring of 500 new firefighters over the life of the consent decree. Of the 10% goal for women in entry-level hiring, at least 50% should be minority women, and a higher goal may be imposed should there be sufficient qualified candidates. The decree imposes a 55% goal for minorities in entry-level hiring. The minimum goal for each minority group protected by the decree will be the percentage reflected in the 1980 and 1990 census. Currently these percentages are: 19% Asian, 10% Black and 11% Hispanic. If the City decides to test less than all those who apply, the testing pool must reflect the applicant pool and be selected at random. Id. at 10-11. To remedy adverse impact in promotions the decree makes several provisions. First, six named Black provisional lieutenants will be promoted to permanent lieutenant retroactive to March 9,1979 with back pay. A special H30 test for promotion to captain will be given to these six by April 1, 1988. The results of this exam are to be integrated with the existing captain’s list. Those who fall within the range of those already promoted will be immediately promoted. The others will remain on the list until in the regular course they are promoted or the list expires. Id. at 11-12. Second, the goal for promotions is that they come to reflect minority representation in the applicant pool. The applicant pool for promotions is defined as those who both are eligible and apply. Id. at 12. Third, the consent decree sets out a fairly complicated schedule for the administration of promotional tests designed to insure that there is an increasing number of minority and female applicants in the promotional pool. All eligibility lists are to expire after two years. Id. at 13-15. The consent decree sets up a special procedure for grieving tests and selection procedures. Id. at 15-16. Finally, the consent decree provides that within 60 days of execution by all parties, the SFFD shall promote (in addition to the six provisional lieutenants earlier named) 11 Black firefighters, 8 Hispanic firefighters and 8 Asian or Filipino firefighters from the 1984 H20 list to the position of fire lieutenant. The SFFD may also promote an additional 48 lieutenants from the 1984 H20 list. Of those promoted, 25% shall be minorities. If the SFFD promotes all 48 allowed, 5 shall be Black, 5 shall be Hispanic and 2 shall be Asian or Filipino. No one found by the court to have violated federal civil rights laws shall be promoted pursuant to these provisions of the consent decree. Id. at 16-17. The consent decree provides for human relations training. Id. at 17. It also provides a specific process for grieving human relations charges. Id. at 18-20. The consent decree provides for the appointment and sets out the powers and responsibilities of a monitor who will insure compliance with the decree. Id. at 20-21. The decree does not dispose of claims for attorneys’ fees. Id. The consent decree shall be in effect for seven years. Id. at 22. LEGAL STANDARD Voluntary compliance is the preferred means of achieving the objectives of equal employment opportunity embodied in Title VII. Officers for Justice v. Civil Service Comm’n, 688 F.2d 615, 625 (9th Cir.1982), cert. denied, 459 U.S. 1217, 103 S.Ct. 1219, 75 L.Ed.2d 456 (1983). See Johnson v. Transportation Agency, 480 U.S. 616, -, 107 S.Ct. 1442, 1457, 94 L.Ed.2d 615 (1987); Kirkland v. New York State Dep’t of Correctional Servs., 711 F.2d 1117, 1128 (2d Cir.1983) (“voluntary compromises of Title VII actions enjoy a presumption of validity”), cert. denied, 465 U.S. 1005, 104 S.Ct. 997, 79 L.Ed.2d 230 (1984). There are several reasons for the preferred status accorded amicable settlements of litigation and plans voluntarily undertaken by employers in the absence of litigation. In addition to reasons of judicial economy and the savings of the costs of litigation, voluntary action may produce more favorable results because it is less likely than a court order to engender opposition and resentment. Kirkland, 711 F.2d at 1128 n. 14. Further, “[t]he value of voluntary compliance is doubly important when it is a public employer that acts, both because of the example its voluntary assumption of responsibility sets and because the remediation of governmental discrimination is of unique importance.” Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 290, 106 S.Ct. 1842, 1855, 90 L.Ed.2d 260 (1986) (O’Con-nor, J., concurring). With the hope that this settlement will end the need for the court’s intervention and allow the SFFD to set about healing the wounds inflicted over long years of discrimination and during the course of this litigation, the court will review the terms of the consent decree not only for the fairness of its terms, but also to insure its statutory and constitutional validity. The standards for evaluating fairness and validity are discussed below. I. The Standard of Review for Fairness A class action cannot be compromised or settled without court approval. Fed.R.Civ.P. 23(e); Officers for Justice, 688 F.2d at 623. Thus, the first standard by which the court must evaluate the settlement reached by the parties in this case is whether the terms of the consent decree are fundamentally fair, adequate and reasonable. See Officers for Justice, 688 F.2d at 625. This determination should be reached by focusing primarily on the terms of the decree. See Moore v. City of San Jose, 615 F.2d 1265, 1271 (9th Cir.1980). In determining whether the settlement is fair, however, the court should also consider an amalgam of factors including but not limited to: the strength of the plaintiffs’ case; the risk, expense, complexity and likely duration of further litigation; the extent of discovery and stage of the proceedings; the experience and views of counsel; the presence of a governmental party; and the reaction of interested parties to the settlement. See Officers for Justice, 688 F.2d at 625. II. The Standards for Constitutional and Statutory Evaluation In addition to evaluating the basic fairness and adequacy of the settlement, because the City is a party to the consent decree and because the decree provides for affirmative race-conscious relief, the court must find that the settlement satisfies two legal standards. There is no longer any question that “government bodies constitutionally may adopt racial classifications as a remedy for past discrimination.” 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). However, an affirmative action plan entered into by a municipality implicates state action and so must comport with the equal protection clause of the fourteenth amendment in addition to Title VII. Local Number 93, Int’l Ass’n of Firefighters v. City of Cleveland, 478 U.S. 501, 517 n. 8, 106 S.Ct. 3063, 3073 n. 8, 92 L.Ed.2d 405 (1986). See Johnson, 107 S.Ct. at 1446 n. 2; Wygant, 476 U.S. at 272 & n. 3, 106 S.Ct. at 1846 & n. 3; Higgins v. City of Vallejo, 823 F.2d 351, 352 (9th Cir.1987). For reasons that remain somewhat elusive, the standards applicable to analysis under Title VII and the equal protection clause are not coterminous. Johnson, 107 S.Ct. at 1449 n. 6. More often than not, the substantive requirements of equal protection are more stringent than those of Title VII and therefore the case to be made for a settlement incorporating race-conscious relief is more difficult for a public than for a private employer. Although the substantive requirements of the law may differ, the framework of analysis is the same for both equal protection and Title VII analysis. An affirmative action plan must be evaluated according to a two part test under either Title VII or equal protection. First, there must be adequate justification for the use qf affirmative action. See Johnson, 107 S.Ct. at 1452; United States v. Paradise, 480 U.S. 149, 172, 107 S.Ct. 1053, 1067, 94 L.Ed.2d 203 (1987); Wygant, 476 U.S. at 274, 106 S.Ct. at 1847; Sheet Metal Workers, 106 S.Ct. at 3050; United Steelworkers v. Weber, 443 U.S. 193, 208, 99 S.Ct. 2721, 2729, 61 L.Ed.2d 480 (1979). Second, if the plan is justified, the court must then determine that it does not unnecessarily burden the rights of non-minority employees. Johnson, 107 S.Ct. at 1452; Paradise, 107 S.Ct. at 1067; Wygant, 476 U.S. at 274, 106 S.Ct. at 1847; Sheet Metal Workers, 106 S.Ct. at 3050; Weber, 443 U.S. at 208, 99 S.Ct. at 2729. See also Higgins, 823 F.2d at 356-57, 358. A. Justification Analysis The Supreme Court has not agreed on the standard to be applied to affirmative action plans under the equal protection clause. Paradise, 107 S.Ct. at 1064 & n. 17. Courts have therefore exercised caution by applying strict scrutiny, that is, by requiring the plan to be “narrowly tailored” to serve a “compelling governmental purpose.” Id. at 1064; see Higgins, 823 F.2d at 358. The compelling governmental purpose language is not useful in the context of evaluating an affirmative action plan because that purpose is always the same. The purpose of the plan is always to remedy past or present discrimination. For a governmental entity this is always a compelling purpose. Paradise, 107 S.Ct. at 1065. The real issue to be resolved is the adequacy of the evidence of discrimination used to justify the plan. Equal protection requires evidence of past discrimination by the particular employer at issue in order to justify affirmative race-conscious relief. See Sheet Metal Workers, 106 S.Ct. at 3050. A plurality of the Court in Wygant stated that the Court has required “some showing of prior discrimination by the governmental unit involved before allowing limited use of racial classifications in order to remedy such discrimination.” 476 U.S. at 274, 106 S.Ct. at 1847. In Paradise and Sheet Metal Workers, the Court seemed to require more, holding that a finding of egregious, persistent and systematic past discrimination, specifically persistent and systematic discrimination was sufficient to justify court-ordered affirmative action. Paradise, 107 S.Ct. at 1065; Sheet Metal Workers, 106 S.Ct. at 3050 (race-conscious relief also may be justified “to dissipate the lingering effects of pervasive discrimination”). In both cases, without specifying what showing was required as a minimum, the Court held that a finding amounting to intentional past discrimination justified the district court’s award of an affirmative race-conscious remedy. There is a distinction drawn, however, between the evidence of past discrimination needed to justify court-ordered relief and that required for voluntary action by a public employer. In Wygant, a plurality of the Court would have required less than evidence of intentional discrimination to justify a public employer’s voluntarily adopted affirmative action plan. Specifically, Justice Powell joined by three other Justices would have required only a “strong basis in evidence,” for a finding of past discrimination sufficient to justify a public employer’s voluntarily undertaken plan. Wygant, 476 U.S. at 277, 106 S.Ct. at 1848. In her concurrence, Justice O’Connor expanded on the “strong basis in evidence” or, as she calls it, the “firm basis” standard. See id. at 286, 106 S.Ct. at 1853 (O’Connor, J., concurring). She wrote that “a contemporaneous or antecedent finding of past discrimination by a court or other competent body is not a constitutional prerequisite to a public employer’s voluntary agreement to an affirmative action plan.” Id. at 289, 106 S.Ct. at 1854. Her policy concern was that if public employers were compelled to demonstrate their own history of illegal discrimination in order to justify the adoption of voluntary affirmative action programs, they would necessarily be discouraged from voluntarily complying with their civil rights obligations. See id. at 290,106 S.Ct. at 1855. Justice O’Connor suggested that statistical evidence of disparity sufficient to support a prima facie case under Title YII would justify (that is, constitute a “firm basis” for believing remedial action was required) voluntary affirmative action consistent with the Constitution. See id. at 292, 106 S.Ct. at 1856. See also Johnson, 107 S.Ct. at 1461 (O’Con-nor, J., concurring). A majority of the Court is agreed that a gross statistical disparity alone may satisfy the prima facie standard. See Hazelwood School District v. United States, 433 U.S. 299, 307-08, 97 S.Ct. 2736, 2741-42, 53 L.Ed.2d 768 (1977). See also Johnson, 107 S.Ct. at 1453 n. 11 (opinion of Brennan joined by Marshall, Blackmun, Powell and Stevens, JJ.) and at 1461 (O’Connor, J., concurring). Statistical disparities, however, provide a reliable indicator of past discrimination by a particular employer only in cases where past societal discrimination is not a factor that needs to be screened out. The only cases where societal or industry-wide discrimination need not be screened out are those in which the remedy sought to be justified affects entry-level or non-skilled hiring, or where the remedy relates to promotions but the applicant pool is wholly internal. For entry-level or nonskilled positions, the probative statistical comparison is between the percentage of minorities in the employer’s work force and the percentage of minorities in the relevant labor market. See Hazelwood, 433 U.S. at 308 n. 13, 97 S.Ct. at 2742 n. 13. Where, as in the SFFD, promotions are made entirely from the lower ranks and there is therefore no outside labor pool, the probative statistical comparison is between the percentage of minorities in each of the upper ranks and the percentage of minorities in the qualifying rank below. Because all hiring is entry-level and non-skilled there is no need to screen for the effects of societal discrimination (in access to training and skills) that may not be used to justify the adoption by or imposition on a public employer of affirmative race-conscious relief. When societal or industrywide discrimination has been a factor in constraining the employment advancement of minorities, it will be impossible to establish a prima facie case by statistical comparison alone. This is because in order to screen out the effects of general discrimination (as opposed to discrimination for which the employer is directly responsible), the probative statistical comparison is between the percentage of minorities in the employer’s skilled work force and the percentage of minorities in the relevant labor force who possess the required skills. See Hazelwood, 433 U.S. at. 308 & n. 13, 97 S.Ct. at 2742 & n. 13. Where minorities have historically been denied access to training, see, e.g., Weber, 443 U.S. at 198 n. 1, 99 S.Ct. at 2725 n. 1, this statistical comparison generally will not reflect a disparity. See Johnson, 107 S.Ct. at 1452 n. 10. Where generalized societal discrimination has prevented minorities from acquiring the skills necessary to become part of the qualified labor force, the statistical comparison is too crude to establish a prima facie case. Ironically, the lack of imbalance between the percentage of minorities in the employer’s work force and the percentage in the qualified labor force prevents public employers “in precisely those [areas] in which discrimination has been most effective” from voluntarily adopting affirmative race-conscious remedies for employment discrimination. Id. In order to justify such action consistent with the Constitution, a public employer would need to come up with non-statistical evidence of past discrimination, see id. at 1453 n. 11, and that could only be evidence of intentional discrimination. See International Bhd. of Teamsters v. United States, 431 U.S. 324, 339 & n. 20, 97 S.Ct. 1843, 1857 & n. 20, 52 L.Ed.2d 396 (1977). It is reasonable to predict that in these circumstances, public employers will be significantly deterred from voluntarily complying with their obligations under the civil rights laws, because to do so would expose them to liability for past discrimination. By contrast, the statutory standard under Title VII does not require evidence of past discrimination by a specific employer to justify that employer’s voluntary adoption of an affirmative action plan. See Johnson, 107 S.Ct. at 1452-53. Last term in Johnson, the Court took its earlier ruling in Weber respecting private employers and extended it to public employers. The Court held that voluntarily adopted race-conscious remedies for past discrimination could be justified by public agencies under Title VII by showing only the existence of a “manifest imbalance” reflecting under-representation of minorities or women in “traditionally segregated job categories.” Johnson, 107 S.Ct. at 1452 (quoting Weber, 443 U.S. at 197, 99 S.Ct. at 2724). With respect to training programs or to jobs that demand no special skills, a manifest imbalance can be established by comparing “the percentage of minorities or women in the employer’s workforce with the percentage in the area labor market or general population.” Id. With respect to jobs demanding special skills or training, however, “the comparison [of the percentage of minorities or women in the employer’s skilled work force] should be with those in the labor force who possess the relevant qualifications.” Id. Clearly, if there is sufficient evidence to justify the consent decree in this action under the constitutional standard, it will also pass muster under Title VII’s statutory standard. Accordingly, the court will apply only the constitutional “firm basis” standard to determine whether the terms of the decree are justified. B. Burdens Analysis Several factors are relevant in determining whether the justifiable remedy is nonetheless too burdensome to innocent nonminorities. These factors are the same under the “narrowly tailored” inquiry of equal protection analysis as they are under Title VII because the inquiries are conceptually identical. The factors to be considered in evaluating the permissible burden of justified affirmative relief include: whether the plan unnecessarily trammels the rights of non-minorities; whether the plan acts as an absolute bar to non-minorities; whether the plan requires “blind hiring” of minority applicants without regard to qualifications; whether the plan uproots legitimate expectations of non-minorities; whether the plan requires rigid quotas or flexible goals; and whether the plan is temporary or designed to maintain a balanced work force. See, e.g., Johnson, 107 S.Ct. at 1454-56; Paradise, 107 S.Ct. at 1067; Higgins, 823 F.2d at 357. The court will consider these factors in evaluating the terms of the consent decree. DISCUSSION The settlement of this case was hard fought and hard won. San Francisco has long prided itself on its reputation for celebrating rather than discriminating on the basis of human diversity. No party more than the City wants the SFFD to end this lengthy, divisive litigation and get on with the business of serving the San Francisco community. I. Review for Fairness The first question to be addressed by the court is whether the settlement reached is fair, adequate and reasonable. See Officers for Justice, 688 F.2d at 625. For the City, the risks of continuing with this litigation were many, and plaintiff-intervenors’ case was strong. The City repeatedly has been admonished by courts and repeatedly has failed to produce entry-level and promotional employee selection procedures that are valid under Title VII. As a result, the SFFD has required court supervision in order to hire and promote uniformed firefighters for close to the last twenty years. This court has already entered partial summary judgment against the City on the liability issues. See Davis I, 656 F.Supp. at 288. Further, the court recently found that the SFFD is either unwilling or unable to enforce the provisions of the civil rights laws. See supra, p. 1298. In addition, the City had substantial indication from the reports of the monitor that there was a good case to be made against the SFFD for racial harassment. Indeed, at the fairness hearing, the City revealed that its primary motive in negotiating and agreeing to the consent decree was to avoid the risk that the court would make a finding of intentional discrimination after trial and award relief more onerous than that imposed by the terms of the decree. Finally, although discovery was virtually complete, not only would a trial on the harassment issues have been long, complex and costly, but whatever the result, general publication of the evidence at trial, which this court previewed at various pretrial proceedings, would almost certainly have been extremely embarrassing and demoralizing to the SFFD. The objections of the United States and the Union are addressed specifically to the standards mandated under Title VII and the Constitution and will be discussed below. The objections filed by individual firefighters and members of the public, however, do go to the general issue of the fairness of the consent decree and so it is appropriate to consider them at this juncture. One hundred thirteen persons took the time and effort to file written objections with the court. The great majority filed on forms provided by the Union. Many of those who filed objections had given the competing interests considered thought and some offered alternative proposals. Some, however, do not understand that the fundamental premise of compromise is that no party gets everything it wanted; each has to give something up. Others fail to understand that it is both fair and reasonable to require those who have been the passive beneficiaries of past discrimination to bear some of the burden in remedying the harm caused to others. The court cannot here answer each objection. However, the objections fall into certain broad categories that will be addressed. Almost all the objections concerned the changes in promotional practices, as opposed to hiring practices, made in the decree. Many objected that those minorities promoted to H20 fire lieutenant pursuant to the decree did not qualify on the merits by passing a test. This argument, of course, fails to account for the fact that the City has conceded that the H20 tests previously used have not been shown to be job-related. Thus, those who passed the test are not demonstrably more qualified to perform the job of fire lieutenant than those who failed. A related objection is that departmental morale will be adversely affected by the promotion of unqualified personnel. The court hastens to assure and to remind the members of the SFFD' that the decree specifically provides that no one who is not qualified shall be hired or promoted according to its terms. Others objected to the loss of educational credits which will no longer be taken into account in promotions. These credits, however, have not been shown to be job-related. Similarly, some firefighters objected to the failure of the affirmative remedies to take seniority into account; two men objected to consideration of intra-minority seniority. However, to take seniority into account, in view of the SFFD’s persistent discrimination in hiring, would only serve to confirm rather than to remedy past discrimination. Some claimed the SFFD has already done its best; others objected that the City had caved in and should have done more to defend the tests. These assertions are belied by irrefutable evidence of a series of exams having an adverse impact on women and minorities. Still others complained that the testing schedule was unrealistic. This may already have proven too true. However, it is not grounds for invalidating the decree. The terms of the consent decree are fair. The firefighters hired pursuant to the provisional hiring order in Davis I, 656 F.Supp. at 292-93, will be given permanent status retroactive to their hiring dates if they pass a special test to be given no later than October 1 of this year. This test will be additional insurance of their qualifications for the job, although their tenure alone strongly suggests they are qualified. The promotional remedies are likewise fair. To remedy the adverse impact of the 1978 H20 fire lieutenant’s exam, the next six Black firefighters from the 1978 list will be promoted with back pay retroactive to March 9, 1979. They will also be afforded the opportunity to take a special H30 test for the rank of captain in an effort to put them in the same position for promotion they would have been in had they been appointed lieutenants in 1979. In addition to these six men, the decree calls for the promotion of 27 minority firefighters from the invalid 1984 H20 fire lieutenant list and of up to 48 more lieutenants of which 25% shall be minority firefighters. This will act to stem the shortage of lieutenants and to get qualified minority firefighters quickly into the upper ranks after twenty years of virtual exclusion. It will also be an opportunity for promotion for a number of firefighters, many of whom will be White, who have long waited for promotion. As the City pointed out at the fairness hearing, these promotions will not be blind. Race will be only one consideration among others, including each candidate’s work history and p