CITY CTY. OF SAN FRANCISCO v. FAIR EMP. HOUS

OPINION

HANING, J.

This action arises from an accusation filed December 29, 1980, by the Director of the California Department of Fair Employment and Housing (Department) accusing the City and County of San Francisco (City) of racial discrimination against Black firefighters in the administration and implementation of the 1978 promotional examination for the position of H-20 lieutenant. After a lengthy administrative proceeding the Fair Employment and Housing Commission (FEHC) determined that the City discriminated against Black firefighters in violation of California's Fair Employment and Housing Act (Gov. Code, § 12900 et seq.) by using a promotional examination with an adverse impact on Blacks which was not proven to be job related. The City sought a writ of administrative mandamus to relieve it from FEHC's decision and remedial order. (Code Civ. Proc., § 1094.5; Gov. Code, § 11523.) The superior court agreed the examination had an adverse impact on Black candidates for promotion, but disagreed with FEHC's determination that the City failed to prove the examination's validity. This appeal is taken from the superior court's grant of administrative mandamus vacating FEHC's decision.

The City was joined in its request for administrative mandamus by San Francisco Fire Fighters, Local 798, and certain individual firefighters who passed the H-20 promotional examination and were eligible for promotion to lieutenant. These parties have filed briefs in support of the City's position on appeal, and all were parties to the administrative proceedings. Future reference to the City's arguments include those of Local 798 and the named firefighters. Local 798 and the named firefighters have also appealed from the superior court's denial of their motions for attorney fees.

The FEHC and the Department are joined in this appeal by amici curiae San Francisco Chapter of the International Association of Black Professional Firefighters, the NAACP, Mexican-American Legal Defense and Educational Fund and the Asian Law Caucus. Future reference to FEHC's arguments include those of the Department and amici.

In 1978 the City's civil service commission prepared a new written examination for the position of H-20 lieutenant. The position of H-20 lieutenant is the first promotive rank above entry-level H-2 firefighter, and any H-2 firefighter in the San Francisco Fire Department who had been in that rank for one year was eligible to take the H-20 lieutenant's promotional examination.

On October 14, 1978, 577 firefighters took the examination. Of these 577 examinees, 54 (or 9.35 percent) were Hispanic, 33 (or 5.7 percent) were Black, 14 (or 2.42 percent) were other minorities, and 476 (or 82.5 percent) were White. The examination consisted of 149 multiple choice questions, 34 fill-in questions, 2 case problems and 1 written communications problem. It was essentially a job knowledge test covering, among other topics, firefighting strategy and technique, arson detection, equipment maintenance, rescue and emergency medical procedures, and report writing. The test score was the exclusive means of determining which applicants would be promoted to lieutenant.

A ranked eligibility list was prepared based on the results of the examination, listing the 250 highest scoring firefighters in descending order of their test scores. Of the 250 persons on the eligibility list, 226 were White, 17 were Hispanic, 6 were Black, and 1 was a member of another minority group. As these figures reflect, the pass rate for White firefighters (47.8 percent) was nearly two and one-half times that for Black firefighters (18.18 percent).

The critical factor in being promoted, however, was not to be placed on the list, but to be placed at a high standing on the list. The parties stipulated that only the top 160 firefighters had a realistic chance of promotion during the 4-year life of the eligibility list. Of the top 160 positions on the eligibility list, 146 were White, 11 were Hispanic, and 3 were Black. Promotion is subject to a number of variables: attrition within the existing lieutenant ranks, attrition within the ranks of the candidates, departmental growth, funding, etc. However, it is certain that eligibility for promotion is based on successful passage of the examination. Hence, our focus is on eligibility for promotion, i.e., the 250 highest scoring firefighters.

An accusation was filed by the Director of the Department on December 29, 1980, accusing the City and various City officials of discriminating against Black firefighters in promotional opportunities in violation of California's Fair Employment and Housing Act. (See Gov. Code, §§ 12921, 12940.) Beginning June 29, 1981, FEHC commenced hearings before an administrative law judge to hear the Department's complaint. After 11 days of testimony the administrative law judge filed his "Proposed Decision" with FEHC in which he recommended dismissal of the Department's accusation. The administrative law judge concluded "the [e]xamination was job related and content valid to a degree far in excess of that established as adequate under the case law governing the subject." The FEHC decided not to adopt the proposed decision of the administrative law judge and elected to decide the case itself on the administrative record. (Gov. Code, § 11517.)

An accusation is the administrative equivalent of a civil complaint. It states in ordinary and concise language the acts or omissions charged and the statutes alleged to have been violated. (Gov. Code, § 12965, subd. (a).) The Department and FEHC are statutorily empowered to enforce and administer the Fair Employment and Housing Act. (Gov. Code, §§ 12930, 12935.) The Department acts as prosecutor on the accusation before FEHC; FEHC acts as the adjudicatory agency. (See State Personnel Bd. v. Fair Employment Housing Com. (1985) 39 Cal.3d 422 428-429 [ 217 Cal.Rptr. 16, 703 P.2d 354].)

It has been noted that "from the moment of the agency's rejection [of the administrative law judge's proposed decision], it serves no identifiable function in the administrative adjudication process or, for that matter, in connection with the judicial review thereof." ( Compton v. Board of Trustees (1975) 49 Cal.App.3d 150, 158 [ 122 Cal.Rptr. 493].)

The FEHC concluded the City had engaged in discrimination against Black firefighters on the basis of their race. This conclusion was based on FEHC's finding that the 1978 promotional examination for lieutenant had an adverse impact on Black firefighters and that the examination was not proven to be job-related. Specifically, FEHC determined that ability to supervise was the primary component of the job and was not tested, that higher scores were not shown to be correlated with job performance, and that there was no evidence the City had evaluated alternative procedures with less discriminatory impact. FEHC enjoined the City from making appointments from the H-20 eligibility list and ordered the immediate promotion with backpay of four Black firefighters who made the eligibility list but had not yet been promoted. FEHC further required the promotion of six Black firefighters to the position of lieutenant within one year, and the appointment of one Black for every five non-Blacks "until such time as a job-related selection procedure had been administered by [the City] for the position of H-20 lieutenant."

The City sought a writ of administrative mandamus (Code Civ. Proc., § 1094.5) claiming, inter alia, the examination was sufficiently related to a lieutenant's duties to justify any disproportionate impact it might have upon Black applicants. The superior court agreed, finding an "absence of credible evidence to support [FEHC's] finding that the examination was not job related. . . ." The writ issued on July 5, 1983, vacating FEHC's decision and permitting permanent appointments to the rank of H-20 lieutenant from the eligibility list devised from the 1978 promotional examination. This appeal followed.

FEHC and the Department have not sought to enjoin permanent promotions to lieutenant pending this appeal and apparently do not challenge these promotions now.

(1) Preliminarily, we must address Local 798's threshold argument that FEHC lacked jurisdiction to take any action on the Department's administrative complaint. This argument is based on Government Code section 12960, requiring a verified complaint to be filed with the Department within "one year from the date upon which the alleged unlawful practice" occurred. In this case, the challenged examination was administered October 14, 1978, and the resulting list of persons eligible for promotion to lieutenant was adopted by the San Francisco Civil Service Commission on March 9, 1979. However, the Director of the Department did not file her administrative complaint until June 19, 1980. Local 798 contends the "unlawful practice" complained of was the adoption of the eligibility list, therefore, FEHC was barred from proceeding on the complaint based on the one-year limitation of Government Code section 12960.

FEHC rebuts this argument claiming the "unlawful practice" is of a continuing nature, beginning with the administration of the examination and continuing through the life of the eligibility list, which expired by its own terms on March 8, 1983. Since it is undisputed the fire department made promotions from the eligibility list during the year preceding the filing of the complaint, FEHC claims the complaint was timely filed.

In support of this argument, FEHC relies on the doctrine of "continuing violations" developed in cases examining the timeliness of complaints filed with the Equal Employment Opportunity Commission to establish a violation of title VII of the Federal Civil Rights Act of 1964 (see 42 U.S.C.A. § 2000e et seq.) As most recently explained in Williams v. Owens-Illinois, Inc. (9th Cir. 1982) 665 F.2d 918, cert. den., 459 U.S. 971 [74 L.Ed.2d 283, 103 S.Ct. 302], "the relevant strain of continuing violation doctrine is that a systematic policy of discrimination is actionable even if some or all of the events evidencing its inception occurred prior to the limitations period." ( Id., at p. 924.) Williams goes on to explain "[s]uch continuing violations are most likely to occur in the matter of placements or promotions. A minority employee who is not promoted in 1973, for example, and is subject to a continuing policy against promotion of minorities, may then file a timely charge in 1976, because the policy against promoting him or her continued to violate the employee's rights up to the time the charge was filed." ( Ibid.; see also, Guardians Ass'n of New York City v. Civil Serv. (2d Cir. 1980) 633 F.2d 232, 247-251, cert. den., 463 U.S. 1228 [77 L.Ed.2d 1410, 103 S.Ct. 3568] [police department's reliance on eligibility list reflecting performance on discriminatory examination constituted continuing violation].) We find these federal decisions reflect the same interests implicated in the instant controversy and agree the doctrine of continuing violation is relevant to the resolution of this matter. Because promotions continued to be made from the challenged eligibility list during the year preceding the filing of the complaint, we conclude it was timely filed.

Before proceeding to the merits of this appeal, we must also determine whether the "substantial evidence" test or "independent judgment" test guides our review of the administrative mandamus judgment herein. (2a) Under the substantial evidence test the appellate court and the trial court perform the identical function of reviewing the administrative record to determine whether FEHC's decision was supported by substantial evidence. In performing this function, all reasonable doubts must be resolved in favor of FEHC's findings and decision. ( Topanga Assn. for a Scenic Community v. County of Los Angeles (1974) 11 Cal.3d 506, 514 [ 113 Cal.Rptr. 836, 522 P.2d 12]; Pescosolido v. Smith (1983) 142 Cal.App.3d 964, 970 [ 191 Cal.Rptr. 415].) (3) Under the independent judgment test the appellate court reviews the administrative record to determine whether the trial court's findings are supported by substantial evidence. (See Patterson v. Central Coast Regional Com. (1976) 58 Cal.App.3d 833, 842 [ 130 Cal.Rptr. 169].)

(4) Which test applies hinges on a single factor: If the administrative decision substantially affects a fundamental vested right, the court must exercise its independent judgment. If, on the other hand, the decision does not substantially affect a fundamental vested right, the substantial evidence test applies. ( Berlinghieri v. Department of Motor Vehicles (1983) 33 Cal.3d 392, 395-396 [ 188 Cal.Rptr. 891, 657 P.2d 383]; Strumsky v. San Diego County Employees Retirement Assn. (1974) 11 Cal.3d 28, 44-45 [ 112 Cal.Rptr. 805, 520 P.2d 29].) (5) An employer's right to establish its employment practices and procedures and to impose conditions of employment has not been considered a fundamental vested right. ( American National Ins. Co. v. Fair Employment Housing Com. (1982) 32 Cal.3d 603, 607 [ 186 Cal.Rptr. 345, 651 P.2d 1151]; County of Alameda v. Fair Employment Housing Com. (1984) 153 Cal.App.3d 499, 503 [ 200 Cal.Rptr. 381].) Consequently, FEHC's findings on questions of fact will be sustained if supported by substantial evidence on the record considered as a whole. ( LeVesque v. Workmen's Comp. App. Bd. (1970) 1 Cal.3d 627, 637 [ 83 Cal.Rptr. 208, 463 P.2d 432].) (2b) However, in determining whether FEHC committed any errors of law, the trial and appellate courts perform "essentially the same function" and are not bound by FEHC's legal conclusions. ( Lewin v. St. Joseph Hospital (1978) 82 Cal.App.3d 368, 387 [ 146 Cal.Rptr. 892] ; see also Lowe v. Civil Service Com. (1985) 164 Cal.App.3d 667, 676 [ 210 Cal.Rptr. 673].)

The named firefighters who achieved positions on the eligibility list sufficiently high to be promoted to lieutenant assert a fundamental vested right "not [to be] discriminated against" in promotional opportunities. (See Kerrigan v. Fair Employment Practice Com. (1979) 91 Cal.App.3d 43, 51-52 [ 154 Cal.Rptr. 29], cert. den., 444 U.S. 930 [62 L.Ed.2d 187, 100 S.Ct. 273].) They request that the independent judgment standard of review be applied to their interest in this matter. Courts must decide on a case-by-case basis whether an administrative decision affects a fundamental vested right thereby requiring independent judgment review. ( Bixby v. Pierno (1971) 4 Cal.3d 130, 144 [ 93 Cal.Rptr. 234, 481 P.2d 242].) As has already been pointed out, the promotions to lieutenant have been allowed to proceed in the rank order of the eligibility list, and FEHC does not seek demotions. (Cf. Dawn v. State Personnel Bd. (1979) 91 Cal.App.3d 588, 589-590 [ 154 Cal.Rptr. 186].) Therefore, FEHC's decision has not substantially interfered with the named firefighters promotional opportunities, and a full and independent review of the administrative record on their behalf is not required.

After reviewing the administrative record and making factual findings in light thereof, FEHC determined that the written examination for the position of H-20 lieutenant, though neutral on its face, in fact operated to exclude Blacks, a group protected against employment discrimination by the Fair Employment and Housing Act. FEHC also determined that the written examination was not sufficiently job-related to justify its discriminatory impact. The proper analytical framework for reviewing FEHC's determination was prescribed by Griggs v. Duke Power Co. (1971) 401 U.S. 424 [28 L.Ed.2d 158, 91 S.Ct. 849]. In that case the United States Supreme Court held that an employer was barred under title VII of the Federal Civil Rights Act of 1964 (42 U.S.C.A. § 2000e et seq.) from requiring a high school diploma or achievement of satisfactory scores on two standardized intelligence tests as a condition of employment in a skilled laborer job. The court concluded that even though there was no evidence the requirements had been instituted with a racially discriminatory intent, their use was barred under title VII. "If an employment practice which operates to exclude [Blacks] cannot be shown to be related to job performance, the practice is prohibited." ( Id., at p. 431 [28 L.Ed.2d at p. 164].)

Several years after Griggs, the high court outlined the respective burdens of proof in title VII disparate impact cases. (6a) "To establish a prima facie case of discrimination, a plaintiff must show that the facially neutral employment practice had a significantly discriminatory impact. If that showing is made, the employer must then demonstrate that `any given requirement [has] a manifest relationship to the employment in question,' in order to avoid a finding of discrimination. [Citation.] Even in such a case, however, the plaintiff may prevail, if he shows that the employer was using the practice as a mere pretext for discrimination." ( Connecticut v. Teal (1982) 457 U.S. 440, 446-447 [73 L.Ed.2d 130, 137, 102 S.Ct. 2525]; Dothard v. Rawlinson (1977) 433 U.S. 321, 329 [53 L.Ed.2d 786, 797, 97 S.Ct. 2720]; see also Albemarle Paper Co. v. Moody (1975) 422 U.S. 405, 425 [45 L.Ed.2d 280, 300-301, 95 S.Ct. 2362].)

(7) It has been recognized that the "antidiscriminatory objectives" and "overriding public policy purposes" of title VII and California's Fair Employment and Housing Act are identical. ( County of Alameda v. Fair Employment Housing Com., supra, 153 Cal.App.3d at p. 504; see also Kerrigan v. Fair Employment Practice Com., supra, 91 Cal.App.3d at p. 48, fn. 1.) In fact, California courts have often turned to federal title VII authority to interpret analogous provisions of the Fair Employment and Housing Act. (See, e.g., Price v. Civil Service Com. (1980) 26 Cal.3d 257, 270-277 [ 161 Cal.Rptr. 475, 604 P.2d 1365]; County of Alameda v. Fair Employment Housing Com., supra, 153 Cal.App.3d at p. 504.) In this case, the standards established by FEHC for evaluating a facially neutral selection criterion which has a discriminatory impact on a protected group are identical to federal standards under title VII. None of the parties dispute the applicability of title VII analysis for resolving the issues presented in this case, and we rely on those principles.

FEHC is authorized to interpret the Fair Employment and Housing Act by adopting "suitable rules, regulations, and standards." (Gov. Code, § 12935, subd. (a)(1).) Here FEHC promulgated title 2, California Administrative Code, section 7287.4, subdivision (a) adopting the federal Uniform Guidelines on Employee Selection Procedures. (29 C.F.R. § 1607 et seq.) In 1978, the Equal Employment Opportunity Commission, the Civil Service Commission, the Department of Labor and the Justice Department jointly adopted the Uniform Guidelines to provide "a framework for determining the proper use of tests and other selection procedures." (29 C.F.R. § 1607.1(B).)

(6b) To establish a prima facie case of discrimination in a case alleging disparate impact, the Department was required to show that the H-20 lieutenant's examination in question selected candidates in a "significantly discriminatory pattern." ( Dothard v. Rawlinson, supra, 433 U.S. at p. 329 [53 L.Ed.2d at p. 797].) If the Department establishes this discriminatory pattern the burden shifts to the City to show the "impact is only an incidental product of criteria that genuinely predict[s] or significantly correlate[s] with successful job performance. . . ." ( Contreras v. City of Los Angeles (9th Cir. 1981) 656 F.2d 1267, 1277, cert. den., 455 U.S. 1021 [72 L.Ed.2d 140, 102 S.Ct. 1719].) There was no claim in the instant case of intentional discrimination against Black candidates, nor allegations of purposeful discriminatory acts in the way the test was administered. The Department's claim was simply that the written test had a highly discriminatory impact in screening out Black candidates.

(8a) Both FEHC and the trial court found that the Department had shown sufficient disparity between the passage rates of White and Black candidates to establish a prima facie case of discrimination. The City disagrees. For determining adverse impact the Uniform Guidelines, which have been adopted in cases brought under California's Fair Employment and Housing Act (see fn. 8, ante), set 80 percent as a "rule of thumb." Thus, a written examination may be considered to have adverse impact if it has a "selection rate for any race, sex, or ethnic group which is less than four-fifths (4/5) (or eighty percent) of the rate of the group with the highest rate. . . ." The Uniform Guidelines go on to provide that a selection rate that "is less than [80 percent] of the rate for the group with the highest rate will generally be regarded . . . as evidence of adverse impact." (29 C.F.R. § 1607.4(D).)

An analysis of the examination results reveals the passage rate for Black firefighters was only 38.8 percent of that for White firefighters — substantially below the 80 percent rate established in the Uniform Guidelines. (9) It has been held that statistical disparities alone may constitute prima facie proof of discrimination. ( New York Transit Authority v. Beazer (1979) 440 U.S. 568, 584 [59 L.Ed.2d 587, 601-602, 99 S.Ct. 1355]; Hazelwood School District v. United States (1977) 433 U.S. 299, 307-308 [53 L.Ed.2d 768, 777, 97 S.Ct. 2736].) (8b) The easiest statistic to understand is that 47.8 percent of the White firefighters passed the H-20 lieutenant examination as compared to 18.18 percent of the Black firefighters. In other words, there was a variance in the passage rate of two and one-half times. Similar passage rates have been found statistically significant in establishing a prima facie case of discrimination. (See Clady v. County of Los Angeles (9th Cir. 1985) 770 F.2d 1421, 1429, cert. den., 475 U.S. 1109 [89 L.Ed.2d 915, 106 S.Ct. 1516] [minority passage rate of 55.6 percent to majority's 75.7 percent constituted adverse impact]; Contreras v. City of Los Angeles, supra, 656 F.2d at p. 1274 [minority passage rate of 26.1 percent to majority's 63.9 percent constituted adverse impact]; Craig v. County of Los Anqeles (9th Cir. 1980) 626 F.2d 659, 661-662, cert. den., 450 U.S. 919 [67 L.Ed.2d 345, 101 S.Ct. 1364] [minority failure rate of 33 percent to majority's 13 percent alone held to establish adverse impact].)

When the top 160 candidates who had a realistic chance at promotion are isolated, the figures become even more disparate. Black candidates will be promoted at only 29.6 percent of the promotion rate for White candidates. This rate is also substantially below the 80 percent established in the Uniform Guidelines. (See Connecticut v. Teal, supra, 457 U.S. at p. 443, fn. 4 [73 L.Ed.2d at p. 135] [disparate impact found where minorities passed written examination at approximately 68 percent of passage rate for White]; but see fn. 3, ante.)

The probability that this disproportionate passage rate is a result of chance is slight. The United States Supreme Court in Castaneda v. Partida (1977) 430 U.S. 482, 496-497, footnote 17 [51 L.Ed.2d 498, 512, 97 S.Ct. 1272], adopted a statistical analysis to determine whether the difference between the expected and observed values is statistically significant. As a general rule, if the difference between expected value (if a race-neutral selection device was used) and the actual value is greater than two or three standard deviations, the deviation can be regarded as caused by some factor other than chance. The expert for the Department, a Ph.D. in statistics, ran a two-sample binomial test to compare the passage rates of the top two hundred fifty candidates. The results revealed that the passage rate of Black firefighters was 3.37 standard deviations below the expected rate. The probability of observing a disparity this large by chance was one in one thousand. Even when the passage rate for the top 160 candidates was considered, the passage rate was still 2.64 standard deviations below the expected rate, a figure which the Department's expert characterized as "statistically significant." (See Equal Employment Opportunity Com'n v. Am. Nat. Bank (4th Cir. 1981) 652 F.2d 1176, 1192-1193, cert. den., 459 U.S. 923 [74 L.Ed.2d 186, 103 S.Ct. 235].)

This standard has been adopted by the Uniform Guidelines, 29 Code of Federal Regulations, section 1607.14, subdivision (B)(5).

A two-sample binomial test is a statistical procedure which compares the actual difference in observed passage rates to that which would be expected in a random selection. (See Hameed v. Intern. Ass'n of Bridge, etc. (8th Cir. 1980) 637 F.2d 506, 513, fn. 8.)

Local 798 questions FEHC's reliance on the top 250 candidates and insists the focus should be on the top 160 candidates who would actually be promoted. In the next breath, it claims that when the top 160 candidates are isolated, the number of Blacks in this group (3) is too small to provide meaningful statistics. There are two responses to this argument. First, as pointed out in Connecticut v. Teal, supra, the United States Supreme Court has "consistently focused on employment and promotion requirements that create a discriminatory bar to opportunities. [It] has never read [Title VII] as requiring the focus to be placed instead on the overall number of minority or female applicants actually hired or promoted." ( Id., 457 U.S. at p. 450 [73 L.Ed.2d at p. 139].) In any event, allowing an employer to avoid judicial scrutiny because the number of minorities actually promoted was too small would allow it to benefit from its own discriminatory practices.

The City attacks the Department's statistical proof by pointing to the City's expert's conclusion that job experience, rather than race, accounted for most of the disparity in performance between Black and White candidates. They also point to the atypical nature of the pool of Black applicants; and testimony that Blacks failed to adequately prepare for the examination. It is clear from its findings that FEHC relied on evidence presented by the Department to refute the significance of each of these factors. For example, FEHC found persuasive the results of a two-sample T-test prepared by the Department's expert comparing the examination scores of White and Black firefighters with comparable seniority. The results of this test indicated the average score for Blacks was still 3.52 standard deviations below the expected average. Using regression analysis, the Department's expert tried to determine whether other factors in addition to seniority, such as service as a temporary lieutenant, or prior simulator training experience, might account for the observed differences in examination results. Test scores for Blacks, after removing these factors, were still 79.1 points below the expected. The probability of a disparity this large occurring at random is less than one chance in one thousand. FEHC concluded that even after isolating the effect of job experience, a statistically significant disparity remained between the performance of Black and White candidates; and that finding is supported by the record.

Of the 33 Blacks who took the H-20 examination, 24 (72.7 percent) were hired by the fire department pursuant to the one minority, one White hiring ratio effectuated during the Western Addition Community Organization v. Alioto (N.D.Cal. 1973) 360 F. Supp. 733, litigation.

A two-sample T-test is a statistical procedure which is used to compare the average of two groups to determine whether the disparity is statistically significant.

Regression analysis is used to measure the importance of a set of factors (independent variables) as determinants of a single dependent variable.

Furthermore, the suggestion that Black candidates hired pursuant to court mandate in Western Addition Community Organization v. Alioto, supra, 360 F. Supp. 733, were somehow less qualified than their White counterparts has no support in the record. (Compare Paxton v. Union Nat. Bank (E.D.Ark. 1981) 519 F. Supp. 136, 159.) To the contrary, evidence showed that all candidates who sat for the H-20 examination had worked as an H-2 firefighter for at least one year and had survived a rigorous training program including weekly testing on the content of firefighting manuals. There is nothing in the record to support the inference that Black candidates were less capable than White candidates. In a similar vein, the City charges Black candidates failed to adequately prepare for the examination. There was some evidence this occurred; however, the FEHC found this evidence unconvincing and we cannot say its finding was unsupported.

Beginning in the spring of 1977, a series of classes to prepare candidates for the October 1978, H-20 lieutenant's examination were conducted by the San Francisco Fire Department. Of the 33 Blacks who took the examination, 6 took the above classes. Furthermore, special preparation classes for minority firefighters, taught by a Black H-20 lieutenant, were cancelled due to lack of participation. However, based on the testimony of several Black candidates about their personal study regimen, FEHC found that Black firefighters "adequately prepared for the examination according to their own particular study habits."

As the foregoing discussion makes obvious, FEHC's conclusion that the H-20 lieutenant's examination had a disparate racial impact is fully supported by the record. The critical issue in this case, therefore, is whether the City proved the examination was job-related.

(10) Once the racially adverse impact of the examination was established, the burden of proof shifted to the City to prove the examination job-related. ( Dothard v. Rawlinson, supra, 433 U.S. at p. 329 [53 L.Ed.2d at p. 797]; Albemarle Paper Co. v. Moody, supra, 422 U.S. at p. 425 [45 L.Ed.2d at pp. 300-301].) In order to meet its burden the City must demonstrate a business necessity for use of the examination. "`The touchstone is business necessity,' [citation]; a discriminatory employment practice must be shown to be necessary to safe and efficient job performance to survive a Title VII challenge." ( Dothard v. Rawlinson, supra, 433 U.S. at pp. 331-332, fn. 14 [53 L.Ed.2d at p. 799], citing Griggs v. Duke Power Co., supra, 401 U.S. at p. 431 [28 L.Ed.2d at p. 164].) "The test is whether there exists an overriding legitimate business purpose such that the practice is necessary to the safe and efficient operation of the business. Thus, the business purpose must be sufficiently compelling to override any racial impact; the challenged practice must effectively carry out the business purpose it is alleged to serve; and there must be available no acceptable alternative policies or practices which would better accomplish the business purpose advanced, or accomplish it equally well with a lesser differential racial impact." ( Robinson v. Lorillard Corporation (4th Cir. 1971) 444 F.2d 791, 798, fns. omitted; Blake v. City of Los Angeles (9th Cir. 1979) 595 F.2d 1367, cert. den., 446 U.S. 928 [64 L.Ed.2d 281, 100 S.Ct. 1865].) Mere "[a]dministrative convenience is not a sufficient justification for the employer's [discriminatory] practices." ( Blake v. City of Los Angeles, supra, at p. 1376.)

(11a) FEHC made several crucial findings in the area of job relatedness. It found: (1) "[S]upervision is the primary function for the position of H-20 fire lieutenant. . . ."; (2) ". . . ability to supervise is neither listed as a knowledge, skill, or ability in the job description. . . ."; (3) "[ability to supervise] was [not] tested for on the 1978 H-20 examination. . . ."; and (4) "[the City] presented no evidence showing a correlation between higher test scores and the performance of job incumbents. The content validation studies offered as exhibits by the [City] do not in any way document such [a] relationship." FEHC also determined that written tests are not appropriate measures of supervisory skills.

(2c) FEHC's findings, if supported by substantial evidence are, as previously discussed, binding on us. The function of a reviewing court in a case of this type is to affirm or set aside a decision made by an administrative body, such as FEHC, which has been given the authority to conduct hearings, take evidence, make findings, and ultimately make decisions in its field of expertise. (See Environmental Defense Fund, Inc. v. Coastside County Water Dist. (1972) 27 Cal.App.3d 695, 703 [ 104 Cal.Rptr. 197] .) FEHC's findings will not be disturbed on appeal if they are supported by inferences which are fairly drawn from the evidence. ( Cansdale v. Board of Administration (1976) 59 Cal.App.3d 656, 664 [ 130 Cal.Rptr. 880].) In summary, FEHC's findings come before this court with a strong presumption as to their correctness. ( Smith v. Regents of University of California (1976) 58 Cal.App.3d 397, 405 [ 130 Cal.Rptr. 118].) (11b) We conclude the record contains substantial evidence in support of the relevant findings. The testimony of Mr. Nulph, who prepared the job analysis, and that of the firefighters who challenged the examination, establish the importance of the supervisory component of the job. The job description does not, in fact, list ability to supervise as a necessary component; rather, it lists "knowledge of supervisory principles." A cursory reading of the examination reveals that ability to supervise is not tested, and we doubt that any paper and pencil test can measure this ability. We find no evidence that the City made any attempt to correlate examination scores with satisfactory job performance. Since many of the questions on the examination in dispute were taken from past examinations, such a correlation could have been obtained, if it existed.

Thus, we are presented with an examination process for promotion to lieutenant which fails to measure the primary function of that job. The instant case is nearly identical to Firefighters Institute, etc. v. City of St. Louis (8th Cir. 1977) 549 F.2d 506, cert. den., 434 U.S. 819 [54 L.Ed.2d 76, 98 S.Ct. 60], wherein a promotional examination for captain was challenged. The position of captain in the St. Louis Fire Department at that time was virtually identical to that of H-20 lieutenant in San Francisco; i.e., supervisory ability was the "one major job attribute that separates a firefighter from a fire captain. . . ." ( Id., at p. 511.) Citing the Uniform Guidelines and other cases which dealt with the issue, the Eighth Circuit concluded that the failure to test for supervisory ability prevented the examination from achieving content validity. The decision also notes that the experts in that case agreed that "there is no good pen and paper test for evaluating supervisory skills." ( Id., at p. 512.) The FEHC in the instant case arrived at the same conclusion, and their conclusion is supported by the record.

In light of FEHC's findings, the statutory and regulatory guidelines and the cases cited herein, we conclude the City failed to carry its burden of justifying the use of the 1978 H-20 lieutenant's examination for selection of lieutenants. (12) We also conclude that the extreme passage of time and intervening events have altered the situation significantly so that FEHC's original remedial order is no longer appropriate.

We do not invalidate the examination in question as a measurement of firefighting knowledge (except for questions 148 through 157, inclusive, which were not designed to measure knowledge of firefighting skills). It is undisputed that the 1978 H-20 examination validly measures knowledge of firefighting skills and techniques. We mention this because the City contends that lieutenants must possess a thorough knowledge of firefighting, and as a screening device, or merely one component of the selection process, this examination performs that function. (See, e.g., Firefighters Institute, etc. v. City of St. Louis, supra, 549 F.2d 506; Guardians Ass'n of New York City v. Civil Serv. (2d Cir. 1980) 630 F.2d 79, cert. den., 452 U.S. 940 [69 L.Ed.2d 954, 101 S.Ct. 3083].) Indeed, in a service so vitally concerned with the safety of life and property, in which critical decisions must frequently be made quickly, it is reasonable to require supervisors to possess a high degree of knowledge of the firefighting skills and techniques employed by the persons they are asked to supervise. Neither we nor FEHC need, nor should we, decide how the City should measure the supervisorial component of the job. However, the assessment center approach suggested in the City of St. Louis case is apparently available in San Francisco as well. Oral interviews by members of outside fire departments might also be utilized, as well as on-the-job demonstrations. (See, e.g., Baker v. City of Detroit (E.D. Mich. 1979) 483 F. Supp. 930, 968-969.) However, as we discuss infra, the selection process is now under the supervision of the federal district court in related litigation.

First, we are confronted with ongoing proceedings in federal district court brought in 1984 challenging the fire department's hiring and promotional practices. ( United States v. City and County of San Francisco, et al., N.D. Cal., No. C-84-7089-MHP; Davis, et al. v. City and County of San Francisco, et al., N.D. Cal., No. C-84-1100-MHP.) These actions, which were brought under title VII of the Federal Civil Rights Act (42 U.S.C.A § 2000e et seq.), encompass a number of examinations given by the City for both entry-level and promotional positions, including the 1978 H-20 lieutenant's examination challenged herein. This litigation also challenged a new selection procedure for H-20 lieutenant which was developed by the City and administered in 1984. Because of the instant proceeding, the federal court stayed the claims with respect to the 1978 H-20 examination. Nevertheless, the federal plaintiffs' claims with respect to the 1984 H-20 examination and the fire department's promotional policies in general are an integral part of the federal action and were vigorously contested. However, after lengthy settlement negotiations and on the eve of trial, the City indicated to the federal court in October 1986 that it no longer intended to defend the challenged examinations, including the 1984 H-20 examination. Consequently, in late 1986 the San Francisco Civil Service Commission cancelled all procedures with regard to the 1984 H-20 examination. The selection process for lieutenants is now proceeding pursuant to guidelines and a schedule established by the federal court.

It is our belief that parallel proceedings in both federal and state forums scrutinizing the fire department's promotional practices and potentially resulting in comprehensive relief orders, raise the real possibility of each forum ordering conflicting relief. Such would only add further complication to this complex situation, not to mention additional burdens on the public fisc. We should avoid conflicting promotional policies and/or examinations which, left unchecked, could lead to the creation of "state" and "federal" lieutenants within the City's fire department. While the framework of title VII clearly envisions that an initial resort to a state administrative agency in an employment discrimination matter does not deprive an individual of the right to pursue a federal title VII claim, this does not mean that the individual is entitled to duplicative relief in both forums, or that the employer should be subjected in inconsistent remedial orders. (See Bowe v. Colgate-Palmolive Company (7th Cir. 1969) 416 F.2d 711, 715; Voutsis v. Union Carbide Corp. (2d Cir. 1971) 452 F.2d 889, 893, cert. den., 406 U.S. 918 [32 L.Ed.2d 117, 92 S.Ct. 1768].) Traditional concepts of comity and judicial restraint must guide the imposition of any remedy in the instant action. This is particularly appropriate here, since the United States is a party only in the federal actions.

At the heart of FEHC's original relief order was the requirement that the City develop and administer a job-related selection procedure for the position of H-20 lieutenant. The federal district court has taken affirmative steps to see that such an examination is developed by December 31, 1987, with administration of the examination no later than March 31, 1988. The federal court will oversee test development, test validation, test administration and all other procedures to bring the City's selection procedure for H-20 lieutenant into compliance with the law.

We feel this aspect of relief should be deferred pending the outcome of the federal proceedings in order to avoid conflicting relief. The federal court is in the process of determining whether the City's administration and cancellation of the examinations violated title VII and to determine whether the City's record of discriminatory practices is sufficiently egregious and persistent to warrant affirmative race- and/or sex-conscious relief. Title VII has vested broad equitable power in the federal courts to fashion "the most complete relief possible." ( Albemarle Paper Co. v. Moody, supra, 422 U.S. at p. 421 [45 L.Ed.2d at p. 298].) The federal courts have "`not merely the power but [also] the duty to render a decree which will so far as possible eliminate the discriminatory effects of the past as well as bar like discrimination in the future.'" ( United States v. Paradise (1987) 480 U.S. 149, ___ [94 L.Ed.2d 203, 231, 107 S.Ct. 1053, 1073], quoting Louisiana v. United States (1965) 380 U.S. 145, 154 [13 L.Ed.2d 709, 715, 85 S.Ct. 817].) The federal court has signalled its intention to assure the expeditious development of valid entry-level and promotional examinations, to preclude the reinstitution of previous discriminatory employment practices, and to ensure that lawful employment procedures are instituted in the future. We have confidence that the federal court is in the best position to order broad prospective relief. Therefore, the imposition of accelerated promotional goals or other action to eradicate the effects of past discrimination should be left to the federal forum.

Since the parties are not challenging the promotions which have already been made from the 1978 H-20 eligibility list, no remedial order should be made which nullifies these promotions. Neither should the City be required to promote persons in the absence of legitimate vacancies. In the event the City finds itself in need of additional lieutenants, FEHC can devise a method for interim, temporary selection in the event the parties are unable to agree thereon and the federal court has not undertaken to issue alternative procedures.

While we recognize that our directions upon remand may temporarily impinge upon FEHC's broad statutory authority to fashion remedies, we are convinced that the ongoing proceedings in federal court should be given deference. We are acutely aware that Black and White firefighters qualified for promotion continue to wait for the City to develop a valid selection procedure for the position of H-20 lieutenant. However, the City is currently under a duty to comply expeditiously with the federal court's mandate to develop a selection procedure for H-20 lieutenant which is job related and which will allow future promotions to be made on a nondiscriminatory basis. If the federal court should, for any reason, decline or fail to provide remedial relief, FEHC can then proceed to do so.

In light of representations made by FEHC in an earlier phase of this appeal, we conclude that it agrees that relief should be deferred pending the result in the federal actions. In supplemental briefing, FEHC stated: "The Commission agrees that conflicting relief must be avoided and that `the federal court is in the best position to evaluate the present employment practices of the fire department and order comprehensive relief if necessary.' [Citation.] A remand to the Commission will permit the federal court to order appropriate relief, while still preserving the Commission's statutory authority." FEHC has assured this court that "[t]he Commission has already decided that upon remand, it will defer to the pending federal court action since the federal court is fully prepared to evaluate the validity of a subsequent examination, and its effect, if any, on the appropriate remedy for the violation at issue herein. Commission review of this later examination is thus unnecessary in light of the pending federal court action, and the federal court is of course, fully able to order appropriate relief. [Citing authorities.]"

That part of the judgment denying Local 798 and the named firefighters attorney fees is affirmed. In all other respects, the judgment is reversed and remanded with instructions to recall the writ of mandate, and to enter a new order denying the application for same. The superior court shall remand the matter to FEHC with instructions to hold additional hearings and issue a new order consistent with this decision.

Low, P.J., and King, J., concurred.

Petitions for a rehearing were denied May 26, 1987, and the petitions of all parties for review by the Supreme Court were dismissed as moot on August 13, 1987.