Harper v. Mayor and City Council of Baltimore, 359 F. Supp. 1187 (D. Md. 1973)
May 2, 1973
MAYOR AND CITY COUNCIL OF BALTIMORE, a municipal corporation, et al., Defendants,
Patrick A. McCarthy et al., Defendants and Intervening Defendants,
William Gregory Sheffield et al., Intervening Defendants.
United States District Court, D. Maryland.
*1188 *1189 *1190 *1191 Kenneth L. Johnson, Baltimore, Md., Jeffry A. Mintz, New York City, for plaintiffs.
Samuel M. Campanaro, H. Thomas Howell, Richard T. Sampson, Baltimore, Md., for intervenors.
George L. Russell, Jr., A. Douglas Owens, Gerald S. Klein, Baltimore, Md., for defendants.
*1192 MEMORANDUM OPINION AND ORDER
JOSEPH H. YOUNG, District Judge.
Four black employees of the Baltimore City Fire Department have brought suit individually and as representatives of all similarly situated and affected persons, seeking injunctive and declaratory relief from patterns and practices of racial discrimination allegedly employed by the defendants, Mayor and City Council of Baltimore, members of the Board of Fire Commissioners and Civil Service Commission. Twenty-three white firemen have intervened as defendants.
By order of October 20, 1972, the Court certified this suit as a class action on behalf of all black employees of the Baltimore City Fire Department, all black former employees and all potential black employees and applicants. The actions of defendants on which the complaint is based, if proven, would make injunctive or declaratory relief appropriate to all such individuals. Rule 23(b) (2), Federal Rules of Civil Procedure.
Jurisdiction is founded upon 28 U.S.C. § 1343(3), which confers on this Court, without regard to the amount in controversy, jurisdiction of any action to redress the deprivation under color of any state law or regulation, of any right, privilege or immunity secured by the Constitution, or by any federal civil rights statute. Plaintiffs' complaint charging racially discriminatory employment practices properly invokes such jurisdiction by stating a claim for relief under 42 U.S.C. §§ 1981, 1983 and 1988, 28 U.S.C. § 2201 and the Thirteenth and Fourteenth Amendments. Whitner v. Davis, 410 F.2d 24 (9th Cir. 1969); Carter v. Gallagher, 452 F.2d 315 (8th Cir. 1971); Bennett v. Gravelle, 323 F. Supp. 203 (D.Md.1971). Venue is proper in this District. 28 U. S.C. § 1391.
Plaintiffs' amended complaint sets out numerous specific grievances. They concern overt discrimination against the plaintiff class through a long-standing refusal to hire any blacks to work in the Fire Department and through segregation and harassment after they were hired. Additionally they allege discrimination against plaintiffs in more covert ways, including maintenance of hiring procedures designed to minimize entry of blacks into the Department and promotion procedures designed to hinder the progress of those blacks who do enter the Department. The complaint places more than the intentions of defendants in issue. Plaintiffs are entitled to treatment by this government employer free from discrimination of any sort. The Court must evaluate the position of blacks in the Department and determine whether responsibility for any disadvantaged position can be laid to defendants. Carter v. Gallagher, supra; Chance v. Board of Examiners, 458 F.2d 1167 (2d Cir. *1193 1972); Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971).
Black employees in the Department do not comprise he same percentage of the Department as they do of the City. A survey by the Fire Department shortly after suit was filed disclosed that only 13.7% of the Department employees were black, and a survey taken just before trial commenced disclosed that the percentage rose imperceptibly in the year that this action was pending. The percentage of blacks is even lower in the officer ranks. By contrast, the census figures indicate that in 1970 47% of the residents of the City of Baltimore were black. Intervenors contend that, since the Fire Department is not limited in its hiring to City residents, more appropriate figures for comparison are the racial percentages of the Baltimore Standard Metropolitan Statistical Area. Blacks comprised slightly less than 24% of that population in 1970.
If there were significance to the community-department comparison alone, it could not be doubted that plaintiffs have made a dramatic showing. For though new firefighters can reside anywhere within 30 miles of the City, the record shows that over two-thirds of the applicants come from the City itself. Therefore, to accurately reflect the racial composition of the actual hiring area, the black percentage of the Department would have to be closer to 47% than 24%. It is 14%. But dramatic and frustrating though it may be, this discrepancy does not prove a violation of the Civil Rights Acts, as plaintiffs contend. It does not evidence that the defendants maintained practices or procedures which caused the statistical discrepancy. And most importantly, it gives no clue as to where the problem of disparate treatment might lie.
Black entrance into the Baltimore City Fire Department is a recent story. Blacks were for years totally excluded from Fire Department employment, and the years after their initial entry were marked by segregation, ostracism and harassment.
When blacks were first admitted to the Department in 1953, it was only at the behest of a local civil rights organization. Until 1953, the Board of Fire Commissioners (fire board) had simply refused to appoint black applicants (whose names were conveniently designated in red ink by the Civil Service Commission on the list of eligible applicants.)
As can be imagined, those people who countenanced and perhaps encouraged the exclusion of blacks from the Department were not willing to accept black firemen as equals once they entered. The record in this case is replete with testimony of individual firemen to widespread segregation of firehouse facilities and to an atmosphere of harassment and ostracism. One witness called by defendants, a Battalion Chief in 1953 and thus in a position to observe conditions in many of the firehouses, testified that, though segregation did not occur in all of the firehouses, it did occur in most. The fire board caused the removal of "Reserved" signs in 1956, but the segregation which the signs had signaled continued thereafter, though perhaps not on the same scale. In some houses, the segregation that persisted was tacit, and individual blacks who defied the house policy and used "white" facilities sometimes escaped sanctions. In other houses, segregation was practiced openly, as when blacks reporting for temporary duty at a segregated house were shuffled off to another because the "black bed" was occupied. This situation persisted in more and more isolated instances until the late 1960's. During the period of the worst racial segregation, blacks were made the butt of house pranks to a disproportionate degree and were frequently ostracized from quasi-official house clubs.
In the situation just described, it is not hard to imagine that it was difficult for black firemen to carry on in the Department or work their way to higher positions. Indeed, some of the harassment was directly connected with ability to attain promotions on an equal basis, causing blacks to leave the Department in far greater proportions than whites during the 1950's.
*1195 Defendants and intervenors make much of the fact that no causal relation between segregation, harassment, ostracism and heightened attrition rates has been shown. But a psychiatrist called by defendants conceded that one natural reaction to such treatment would be to leave the Department. And the Court does not need statistics or psychiatric testimony to draw a conclusion that is consistent with everything we know of human nature blacks were treated unfairly and many of them left the Department for that reason. One witness, an ex-employee and a prominent member of the community, testified to having left the Department for just that reason. His description of the frustrations and cruelties of racial discrimination was both moving and convincing. And he left in 1964.
Responsibility for each of these instances of discrimination rests squarely with the defendants. It was the City of Baltimore which permitted the collusion of the Board of Fire Commissioners and the Civil Service Commission which resulted in black exclusion from the Fire Department prior to 1953. It was the fire board, appointed by the Mayor of Baltimore and responsible under the City charter for Fire Department affairs which condoned segregation in the use of Fire Department facilities and victimization of blacks in many forms in fire house affairs. It was the City and the Board which failed to take any effective corrective action throughout the 1950's when the link between discriminatory treatment and high attrition of black firemen should have signaled the urgent need for such action. The defendants, by what they did and, charged with the responsibility to act, by what they did not do, deprived the plaintiffs of the equal protection of the laws guaranteed by the Fourteenth Amendment and violated 42 U.S.C. § 1981 and § 1983. Carter v. Gallagher, supra; Allen v. City of Mobile, 331 F. Supp. 1134 (S.D.Ala.1971) aff'd 466 F.2d 122 (5th Cir. 1972).
Those defendants should not be heard to claim laches, or limitations, since the discriminatory treatment, of which the pre-1953 exclusion was only one part, continued well into the period covered by the statute. While it is *1196 sufficient for relief that the effects of past practices have continued within the statutory period, it should be noted that the practices here complained of themselves are not so dated as to be without the reach of Section 1983.
As representatives of all potential black employees and applicants for employment, one of the principal grievances of plaintiffs is "the utilization of tests which discriminate against blacks as a condition precedent to their employment." Amended complaint II, 9(d). The procedure which determines entrance to the Department is the common merit-system method. The Civil Service Commission conducts entrance tests for the Fire Department and certifies to the Department a list of all applicants who have passed those entrance procedures in rank order. The Board of Fire Commissioners then selects new firemen from this certified list of eligibles. In making this selection, the fire board may depart from strict numerical selection within the confines of the "rule of five". The Civil Service Commission *1197 publicizes the tests and puts all prospective applicants on notice that they must meet certain Commission rules for Fire Department entrance, such as height limitations and a high school diploma. The Commission does not advertise the limitation on residence which the Department places upon its employees, apparently because that limitation is within the province of the Board rather than the Commission.
The selection process includes a multiple choice written test ("firefighter" test), a physical examination and a background check for police record and character. The process included until recently a "penmanship" examination, designed to test for writing and spelling ability. This exam was abandoned in 1971, the same year that an "agility" component was added to the testing procedure. No complaint is made that either the physical or the background check discriminate against the plaintiff class, nor has complaint been lodged against the high school diploma prerequisite to entrance. Excluding credits for veterans, the multiple choice written test was, until 1971, the only factor in the entrance procedures which could effect rank on the eligibility list. (In 1971, the "agility" test score was factored into the total score.) Other facets of the entrance procedures have played a screening function. If one did not pass the "penmanship" exam, for example, one was not eligible to take the written exam. For this reason it would be misleading to concentrate solely on the effect of the written test. But since the number of applicants who are eligible to take the entrance exam is invariably much greater than the number of openings in the Department, the impact of the written test in sorting among those applicants is obvious. And the Civil Service Commission does not seem to have treated the written entrance examination as the important selection device it is.
During the last decade, the Commission has drawn on two sources for its entrance exam. One of these has been the Public Personnel Association (PPA), a national commercial organization from which the Commission purchased several examinations. The other source has been the solitary work of a commission personnel technician. The employee upon whom this responsibility fell lacked professional training in test construction. And he was, as a Commission employee, unfamiliar with Fire Department work. Working principally on his own, he constructed a test that was used as an alternate to the PPA exam from 1962 to 1969 when the Department purchased additional entrance exams from the Public Personnel Association. In 1971, the Commission employee was again given the job of constructing a Fire Department entry level test. This time he had the advice of a study panel which had been appointed by the fire board in an effort to lessen the racial impact of the entrance tests. Though he took their advice on many points it was not expert advice, and the actual choice of questions remained largely his own. In making this selection, he relied heavily *1198 on the types of questions he had observed in the PPA examinations.
The manner of preparation of the penmanship and agility tests need concern us less. The former was utilized to ensure that all new firemen were literate enough to accurately record incoming communications. Statements of the type one might expect to hear over the Department communications network were dictated and the examinees were required to set them out in writing without exceeding a certain number of spelling errors. The penmanship test was abandoned before the 1971 entrance examination at the instigation of the aforementioned study panel. The same panel also caused inauguration of the agility test. The panel, whose function it was to make the entrance process yield more blacks, felt that it was important to add to the written test a "non-intellective component." The agility test consisted of situps, pushups, pullups and a jog. It was valued at 40% of the total score, the written test at 60%. The scoring was designed in such a way that a good performance on the agility portion could salvage a below-passing score on the written test.
The Court has had presented to it complete racial statistics on only the 1971 examination. The primary reason for this is that the race of each examinee in 1971 was recorded and was available both to plaintiffs and defendants for development of their cases. For that 1971 examination, examinees were not screened by penmanship exams, and the eligibility ranking was a combination of the written test score and the new agility test score. But the statistical data that is available on the results of the firefighter examinations in previous years indicates that the 1971 test, despite its variations in procedure, is a representative example of the racial impact of the firefighter examinations over the years.
In 1971, of 538 applicants, 172 were black. 426 names were certified to the Fire Department as eligible for appointment. Of these 426, 112 were black. The figures indicate that while approximately 32% of the applicants were black, only 26% of those certified as eligible for appointment were black. The part the written test played in this impact is demonstrated by the fact that, while only 15 of 366 whites who took the written test scored below 60% (for a failure percentage of 4―%), 35 of the 172 black examinees did so (failure percentage of 20%). Because of their ranking on the eligibility list a product of both written and agility scores blacks garnered only 15 of 92 appointments, 16%. The most significant comparison in this picture is that, while 32% of those who sought appointment were black, Civil Service Commission entrance procedures yielded an appointment group with only half that black percentage.
The Court's ability to make a similar comparison for other years is hampered chiefly by the fact that racial statistics regarding applicants or employees were *1199 not kept by the Department for many years, at least not on an official basis. Racial identification of all relevant information for a few other years has been made. For every year in which the black percentage of those applying to the Department is known, the percentage of those appointed who are black has been significantly smaller. For the June, 1956 exam, 40% of the applicants were black and only 24% of those appointed were black. In January, 1957, 44% of the applicants were black and only 28% of the appointees. In 1960, 40% of the applicants were black and only 24% of the appointees. And in 1963, 47% of the applicants were black and only 24% of the appointees. The only conclusion possible is that some aspect of the entry procedure has had a substantially adverse impact on blacks. As in 1971, there is evidence for earlier years that the written test played a significant part in this impact. And there is further evidence that when the penmanship exam was used as a screening device, it too had a significant impact on black success. For each of the years in which adverse impact of the entry procedures on black applicants was documented, black failure rates on both the penmanship portion and the written portion were higher than those for whites. On the written portion, unlike the penmanship portion, the actual score is as important as whether one fits above or below the passing line. But the failure rate is at least one indicia of what a comparison of actual scores might show. An inference that exam failure rate by race is indicative of exam scores is borne out by the 1971 experience both high black failure rate and low ranking of black test passers. To reiterate, every year for which we have information regarding the percentage of applicants to the Department who were black, the percentage of black appointees was much lower. And both the written and penmanship tests contributed to this result. The fact that the adverse impact demonstrated by that discrepancy survived the elimination of the penmanship exam in 1971, corroborates what the statistics for the earlier years indicated that the written exam played an independent and significant role in diminishing black appointments.
Statistics for appointments to the Department for every year since 1953 indicate that the years discussed above for which application figures were also *1200 available are clearly representative of the general experience. But the conclusion regarding defects in the entry procedures of the Fire Department was made independently of any assumptions regarding a desirable percentage of black entrants. The "problem" that has been defined, quite simply, is the fact that blacks get hired in a lower percentage than they apply.
Employment tests which are shown to eliminate a disproportionate percentage of one racial group must be demonstrably accurate measures of job performance. Griggs v. Duke Power Co., supra. Whether intentional or not, the use made of the results of these entry tests has hampered the employment opportunities of black people. That use is a violation of 42 U.S.C. § 1981 and § 1983, unless it can be shown that the tests sorted out the best applicants for appointment. Chance v. Board of Examiners, supra; Carter v. Gallagher, supra; Castro v. Beecher, 334 F. Supp. 930 (D.Mass.1971), modified 459 F.2d 725 (1st Cir. 1972). This burden of validation rests with the defendants.
Before addressing the question of whether the defendants have met that burden, one further point about the admission of blacks to the Fire Department should be mentioned. The Fire Department has not always employed residents who live as far from the City as 30 miles. Until 1955, in fact, the Department hired only City residents. The Department changed its restriction to residency within three miles of the City that year, then to five miles in 1958, ten air miles in 1963, and finally, in 1966, the Department relaxed the residency restriction to 30 miles. There has been some testimony that the liberalization of the limitation was occasioned by desire for a greater number of applicants. But more persuasive are the records of the fire board, which indicate that expansion was due, quite simply, to the Department's desire to abide the wishes of its members, some of whom wanted to reside at ever greater distances from the City. Though restriction on residence for initial employment could have been retained without burdening those already employed who wished to move, the Board chose not to make the distinction between employees and applicants. And non-city residents took advantage of the relaxed residency restriction by applying for employment in the City Fire Department in great numbers. The year after the change from City residency to residency within three miles of the City, 5% of those on the eligible list were non-city people. And the evidence indicates a steady trend toward increased non-city presence on the eligibility list as the residency requirement was further relaxed. Since 1966 when the restriction was eased to 30 miles from the City, the non-city presence on the eligibility list has been greater every year than for any year that preceded 1966. The non-city examinees as a group score slightly higher on the written exams than the City people as a group, indicating an even greater *1201 dominance of appointment than the eligibility percentages reveal. And, as should be evident from the significant difference between the black percentage of the City and SMSA populations, the entry of more and more non-city people into the Department brought with it a decreasing percentage of black entrants. Non-city applicants to the Department, unlike their City counterparts, have been overwhelmingly white. Were it not for the adverse impact of the entry tests on blacks, of course, the percentage of black entrants would still have matched the percentage of black applicants. Restricting initial employment to City residents, which defendants seem to suggest, while it would result in more black firefighters, would not rid the procedures of their illegality. Only a showing of job relation can accomplish that.
Testing instruments which adversely affect one racial group must be shown to be valid instruments. Simply defined, test validity is a showing that there is a greater probability that high scorers will perform well on the job than will low scorers. The degree of probability is the degree of test validity.[33A] Cooper and Sobol, Seniority and Testing under Fair Employment Laws, 82 Harv. L.Rev. 1598 (1969). The Court's duty to ensure that tests are "valid" ushers in considerations of psychology and psychometrics, about which the Court has been ably informed by all sides to this dispute.
The origin of the various entry level tests used by the Fire Department has already been briefly described. Their mode of development leaves much to be desired. The tests prepared by the Civil Service Commission were prepared without benefit of a sound job description, a fact which seriously hampers, if it does not destroy, the tests' ability to "measure the person for the job and not the person in abstract." Griggs v. Duke Power Co., supra, 401 U.S. p. 436, 81 S. Ct. p. 856. And the personnel technician who composed the tests for the Commission did not draw on the expertise of the Fire Department training school staff, or of any other person knowledgeable about the qualifications *1202 for a successful firefighter, as he did for promotion exams. There is little need to detail further the transgressions of sound test construction practice which resulted in the locally prepared Fire Department exams. The Civil Service Commission staff head aptly summarized the status of these tests when he testified that he could not say whether or not the tests had rational validity.
Somewhat the same situation prevails with respect to the PPA tests. The Civil Service Commission made specific inquiry to the PPA to determine whether a validation study had ever been made of their firefighter test and received a negative reply. Defendants introduced the report of an industrial psychologist, based on his study of the content of the 1971 firefighter exam, to the effect that the exam meets the requirements of the EEOC. But the EEOC guidelines he made reference to relate to the manner in which a test is to be developed rather than to evaluation of the content of a specific test. In fact, the guidelines approve content validation studies only when empirical studies are not technically "feasible," 29 C.F.R. 1607.5(a), a term which is defined as insufficient minority group members to run an empirical study, or lack of an unbiased measure of job performance. There is no indication that an empirical study was, by that definition, infeasible.
The defendants' more serious efforts to demonstrate the validity of their firefighter tests involved both content analysis by a firefighting expert and an empirical validation study by an industrial psychologist. The former analysis was made only of the 1971 exam. The expert described the test as of the general mental ability and mechanical aptitude type and concluded that the test was definitely job related. His comments amount to articulation of an assumption that the kind of aptitude which written general aptitude tests measure is a valid indicator of success as a firefighter. He cannot be faulted for that assumption. It is common enough. But the law does not afford public employers the luxury of reliance on an untested assumption when the tests which proceed from that assumption adversely affect one racial group. The crucial issue is whether the same type of psychological processes are involved in answering general aptitude questions of the type found in the test and in performing the duties of a firefighter. The firefighting expert agreed that he was not equipped to answer that question.
The empirical validation study consisted of statistical analysis of correlation between firefighter test scores and scores on a test administered at the end of the six month fire school, which is the first assignment of all entering firemen. The fire school test scores were chosen as a criterion because they are objective, and because it was felt that, since fire school is the first Fire Department assignment, success in fire school is, by definition, success on the job. It is clear that the exams administered at the end of fire school are based strictly on material that is presented during fire school. (It is just as clear that not all of the knowledge or skills one is expected to attain at fire school can be tested by a written instrument.) Defendants' study indicates a statistically significant correlation between score on the written entrance exams and score on the fire school exam. Because the firefighter and fire school exams are of such different types, this validation study does not, at least, have the vice of testing the validity of performance on one exam through correlation with performance *1203 on an identical exam. However, there is much to be said for the proposition that success on an examination cannot be used as a criterion of validity for another examination, without distorting the correlation with test-taking ability variables. Though the testimony offered by plaintiffs to the effect that blacks as a class have poorer test-taking abilities than whites as a class is extremely weak, a survey of the literature regarding test validation suggests that using one test as a criterion for validity of another test is not a rigorous approach to validation. Given the haphazard origin of the firefighter tests used by defendants, and the Court's own perusal of the quality of those exams, it can only be concluded that if the tests were valid, it would only be by sheerest chance. That being so, the Court cannot accept a questionable procedure for empirical validation as sufficient to meet defendants' burden. The careless manner in which the tests were compiled or acquired makes effective validation all the more necessary.
All new firemen enter the Department as firefighters. There are five officer ranks above that of firefighter, and several specializations that a firefighter may be promoted to but which do not serve as prerequisites to officer promotion. The two dominant specializations are pump operator and emergency vehicle driver. There are time in grade prerequisites to eligibility for promotion to those specialities and each of the officer positions. All promotions are by competitive exam. Score on the competitive exam accounts for 60% of the final promotion score, supervisor's efficiency ratings account for 30% and seniority in grade accounts for 10%. Plaintiffs object to the use of what they consider discriminatory promotion tests and efficiency ratings, alleging that the entire promotion system has the purpose and effect of perpetuating the effects of past discrimination.
The time in grade requirements serve as obvious hurdles for those who were not given the opportunity until late in the Department's history to become firemen. A new fireman cannot be tested for emergency vehicle driver during his first year or pump operator until he has been a firefighter for three years, and cannot be tested for Lieutenant, the first officer step, before serving five years in the Department. Each successive officer step requires two years in grade in the next lower officer position, until one attains the rank of Deputy Chief from which promotion to Chief of the Department requires only six months in grade. These time in grade requirements bar testing for the higher position and not just promotion to it. If a test is held *1204 only once a year, as many are, the time in grade requirement could actually amount to six years for Lieutenant or three years for other officer positions, depending on date of entry and date of testing.
The time in grade requirements provide no competitive advantage to any fireman, in the sense of advantage in a head to head contest. They do remove certain people from the contest. Because blacks could not enter the Department until 1953, and because of heightened attrition rates for blacks in the 1950's, there are fewer blacks able to compete for higher level positions. Time in grade requirements place a premium on experience and blacks are not in a position to have as much experience as whites because of the discriminatory practices of defendants. By retarding the progress of all firemen, the requirements reinforce patterns of racial disparity from rank to rank that originated in illegality. This injury is not individualized in the traditional sense. Although plaintiffs have not pointed out a black fireman who applied to the Department before 1953, and who now suffers because time in grade requirements hold him back, it is obvious that each black fireman has been injured by discrimination against other blacks who may have been his cohorts or superiors. The practices of the Department in the past have been, and many of the problems and misunderstandings in it today are, racial.
There is no rule of law that requires the Department to apportion its officer positions proportionally between the races. But the law does require that no barrier to such proportion be erected. The time in grade requirements draw out the effects of the old discrimination. In so doing they harm every black in the Department and must, therefore, be shown to be called for by business necessity. Robinson v. Lorillard, 444 F.2d 791 (4th Cir.), cert. denied, 404 U.S. 1006, 92 S. Ct. 573, 30 L. Ed. 2d 655 (1971).
The requirement that discriminatory employment practices be justified with a showing of business necessity reflects the law's general approach to employment equality, of which the previously discussed test validation requirement is but a part. An employment *1205 practice, whether a test or not, which adversely affects minority groups must be eliminated if it is not necessary to the conduct of the business. Griggs v. Duke Power Co., supra; Robinson v. Lorillard, supra; United States v. St. Louis-San Francisco Ry. Co., 464 F.2d 301 (8th Cir. 1972). But while courts have come to depend on science in determining test validity, no such assistance is available in passing on the necessity of other employment practices. When the "business" is protecting the lives and property of a dependent citizenry, a Court should be especially leery of reconstructing employment practices which are ostensibly designed to increase the professional competence of the Department. Vulcan Society v. Civil Service Commission, 5 FEP cases 699 (S.D.N.Y.1973). The employer's resistance to change in the employment practices is much more appealing than if it were presented in the guise of lost profits or increased price of the product. Robinson v. Lorillard, 444 F.2d, at 799; Johnson v. Pike Corp., 332 F. Supp. 490 (C.D.Cal.1971).
The record provides little guidance on the matter of justification for the time in grade requirements. Testimony from two previous members of the Board of Fire Commissioners indicated their support for reducing the requirements. But their desire to lower the period resulted more from a belief that it would help blacks than from studies or personal knowledge indicating that it would not decrease Department efficiency. The current Chief of the Department and the Executive Secretary of the fire board personally believe that a decrease to three years in the time in grade requirement for Lieutenant would be desirable, and the Department currently has such a proposal under consideration. No inference can be drawn from the Department's inertia on the other time in grade requirements, since the unions' interest in this matter leaves the Department something less than a free agent.
A former senior chief in the Fire Department testified that the time in grade requirements were necessary and even indicated that a somewhat longer period was called for between Captain and Battalion Chief. Defendant's firefighting expert stated generally that a time in grade requirement was necessary to gain important technical knowledge and develop personal qualifications. The testimony by individual plaintiffs that has concerned the value of holding acting positions, supports the time in grade requirements. The testimony indicates there are skills involved in the higher positions that a junior employee is not likely to possess.
Most of the evidence supporting reduction in time in grade requirements points toward reduction of the five year requirement only, yet it is the shorter requirement for the higher ranks that more directly impinges on black efforts to gain immediate representation at higher levels. Moreover, each of the time in grade requirements is but a formality *1206 to the average fireman for whom promotion normally takes much longer.
The same figures which document this irrelevance of time in grade requirements to most firemen also demonstrate that reducing the time in grade requirements would not lower the floodgates to large numbers of inexperienced officers. It is true that one inexperienced officer at the fireground is one too many. But those individuals who are able to compensate for the decreased seniority credits and lower efficiency ratings that are concomitant with short time in grade, cannot be said to be of anything but high caliber if we are to have any faith in the promotion system. Because the time in grade requirements reinforce patterns for which discrimination is responsible and because no serious effects from their careful modification are likely to occur, the Court concludes that some such modification must be made.
A potentially more serious impediment to large numbers of blacks is the written promotion exam. Plaintiffs rely on statistical data from the City's Community Relations Commission and expert testimony for support of the proposition that blacks have fared poorly in promotion, and that poor performance of blacks is inevitable on written tests in general, or these tests in particular. The Community Relations Commission work proved unusually uninformative and the expert testimony was flatly contradicted by the facts.
Because of the controversy generated during trial regarding evidence, including expert testimony, introduced by plaintiffs which was said to defile blacks, it would be well to note specifically those portions of the evidence which demonstrate clearly that blacks have done at least as well as whites on the Fire Department promotion exams. Due to the comparatively recent entry of blacks into the Department and the time in grade requirements, the only promotion exams which blacks have taken in significant numbers are the pump operator, Lieutenant and Captain exams. In examination for each of these categories, black and white performance has been substantially equal. For the most recent six years, blacks have passed the pump operator written exam at a 24.7% rate, whites at a 24.9% rate. For six selected years, blacks have passed the Lieutenant exam at a 17.2% rate, whites at a 16.2% rate. And for six selected years, blacks have passed the Captain exam at 77.3% rate and whites at a 49.9% rate. Because typically there are more passers than there are open positions, a comparison of scores on the exams is probably more informative than the comparison of pass rates. But the pass rates suggest scores on the promotion exams are also comparable. Blacks do just as well as whites whether *1207 analysis is made of pass rates or averages of the scores.
The accuracy and import of the promotion test statistics which defendants have presented is not seriously disputed by plaintiffs. They indicate in their brief that "there is little indication that one group performs better or worse than the other." But they contend that for two reasons the performance of blacks would be better than that of whites if the tests were fair.
Their principal argument is that the adverse impact of the entrance examination has masked the effect of the promotion exams. Plaintiffs could not contend that only the more qualified blacks survived the entrance exam without admitting the validity of that examination. A contention, on the other hand, that the entrance examination screened out all but the best black test takers fails for a different reason. The equal performance of both races on the promotion exams might only indicate that those promotion exams effectively neutralized test-taking skills. One would have to postulate the invalidity of the promotion examinations in order to conclude that equal pass rates implied adverse impact. The Court will not shift the burden to defendants to prove the validity of the promotion exams on the basis of such a postulate.
Chance v. Board of Examiners, 330 F. Supp. 203 (S.D.N.Y.1971), 458 F.2d 1167 (2d Cir.), on which the plaintiffs must almost exclusively rely for this proposition, is distinguishable from the instant situation in two very important respects. First, in Chance all of the tests involved were of the same basic type. In the instant suit, the entrance test is meant to discern aptitude whereas the promotion tests are directed in large part toward a demonstration of acquired knowledge. That difference makes it impossible to assume that the defects which were found in the entrance exam also inhere in the promotion examinations. Secondly, the court in Chance dealt with a pyramiding of tests, each having an adverse impact. It was for this reason that the court described a "magnification" of statistical discrepancies. Plaintiffs here are not asking the Court through reference to the test that served a screening function to make a finding of adverse impact from moderate statistical discrepancy. Rather, plaintiffs ask the Court to hypothesize an adverse impact on the basis of statistics which show a clearly equal performance by both races. As noted above, the Court does not feel the law compels or justifies such a decision.
*1208 The other reason plaintiffs suggest that the Court should infer adverse impact from equal performance on the tests is that black Fire Department officers, as a group, have more formal education than their white counterparts and that common sense suggests they should do better on the promotion exams for that reason. But while common sense may suggest that additional formal education may be an ingredient in success as a fire officer, it is certainly not the only ingredient. And if there is one thing which the testimony of psychometricians in this case has made clear, it is that each element of success in the position must be weighed and compared with all others in determining what role it should play in the testing process. To single out one factor and suggest that it should be reflected in high pass rates is to suggest the construction of an invalid test. If the plaintiffs' argument is that additional formal education necessarily has, though it should not, a marked effect on test performance, then this contention also is untenable unless the Court assumes the invalidity of the promotion exams. For the reasons stated above, the Court will not do so.
There is one other matter pertaining to performance on the promotion tests which should be mentioned. The discrimination against blacks that occurred in the Department on a wide scale in the early years took the form, in some cases, of deprivation of study material and unequal assignment of acting positions. Since promotion exams test acquired knowledge, access to study materials could materially affect performance on the exams. And there is convincing testimony that holding an acting position is excellent preparation for the exam for that higher position. The denial of access to study materials seems to have been a part of the general ostracism of blacks in some firehouses. Therefore, though the testimony does not reflect the fact, we may assume that the practice may once have been widespread. But the Court is unable to determine the impact on promotions from this problem, which happily does not seem to have existed in more recent years. The Department has taken steps to make study material available to all members of the Department on an equal basis. That being the case, and especially in light of the consistent results of Fire Department promotion exams over a span of years, the Court cannot conclude that blacks would have done better on the promotion exams but for this discrimination.
Denial of acting positions is an allegation that has been sparsely documented. Acting positions are generally given to the senior man. Some of the specific charges of denial have been shown to be baseless, and there is testimony that arbitrary parcelling out of acting positions has been at the expense of some whites as well as some blacks. The Court finds that there was no denial of equal rights to acting positions except in isolated instances, and that victims of such out-of-order appointments were sometimes whites.
Plaintiffs attack the system by which supervisor's efficiency or performance ratings are assigned to each fireman. The efficiency rating procedure has been suspended for administrative reasons while a new system is enacted. The end product of the old system has been the subject of extensive proof.
*1209 Plaintiffs allege that an analysis of the percentages of black and white candidates for promotion who have maximum efficiency ratings indicates a substantial discrepancy in favor of whites. Defendants have introduced data indicating that there has been no statistically significant difference between the averages of efficiency ratings of such candidates. They argue that maximum efficiency ratings have no special significance and point out that, since efficiency ratings generally reflect time in grade, the relatively junior blacks could be expected to experience efficiency ratings below the maximum more frequently than whites.
It is true that there is a statistically significant correlation between efficiency ratings and seniority in grade. Statistical evidence to this effect was personalized by the efficiency history of several witnesses which showed a progression as the individuals gained seniority in grade. Fire Department procedures encourage this correlation by requiring efficiency ratings to be given at a minimum level the first time a newly promoted person is graded. But the record indicates that black candidates for promotion have more seniority in grade on the average, not less, than the whites with whom they compete for promotions. Because of the correlation between efficiency ratings and seniority we would expect blacks to experience maximum efficiency ratings more frequently than whites rather than less frequently. Moreover, if the discrepancy in maximum ratings were explainable by these short periods in grade, there is no reason not to expect a difference in average ratings as well. And there is no such difference. The conclusion must be that while blacks have not been graded lower on the average than whites, they have been graded differently. Apparently, blacks have experienced neither high nor low scores with the frequency of whites. Thus, the efficiency ratings themselves do not answer the question of whether black promotion opportunity has suffered because of the efficiency rating system.
The seniority situation is somewhat akin to efficiency in that figures regarding the potential impact of this factor are unsatisfying. The Court is forced to conjecture about whether this potential is sufficient to have had an adverse impact on black promotion. One-tenth of the total promotion grade depends on seniority in grade. Seniority credits are computed on a scale of 100, with 70 being a minimum. The 30 other seniority points are gained at the rate of two per year for the first five years, one per year for the next five, and one-half per year thereafter. Ten percent of the total seniority points are added into the total promotion score. Because of the inverse graduation of the seniority credits, it is difficult to generalize about their effect. Blacks who entered the Department five years after whites would be penalized in competition for promotion, but that penalization would depend on when the individuals were *1210 competing. For example, a white who entered the Department in 1944 would have had seniority credit of 8.75 on the 1959 Lieutenants' exam. A black who entered the Department in 1954 would have had seniority credit of 8 on that same exam. There would be less of a discrepancy in competition for the same promotion five years later. Thus, the fact that there could be an effect is apparent, but the dimension of that potential impact is not at all clear.
There are large numbers of whites in the Department who entered before the first blacks were admitted. The fact that one entered the Department prior to 1953 does not imply that one is benefitting in promotion competition from the earlier entry. Seniority for promotions is job seniority. But there is evidence that large numbers of pre-1953 entries are holding the same position they held 20 years ago, and thus are enjoying the fruits of entry into the Department at a time when it excluded blacks. And an officer need not have attained his present position prior to 1953 to be benefitting from seniority that blacks were not able to acquire. If a Lieutenant attained his present position in 1958, for example, he attained it before any black could, due to the combined effect of the pre-1953 exclusion and the time in grade requirement. All white firemen in these categories are benefitting unjustly from their seniority in competition with blacks for promotion.
Defendants have effectively demonstrated that this potential for substantial adverse impact from either seniority or efficiency has not been realized, as of yet. If promotions had been based strictly on written test score, there would have been, over a seven-year period, three additional black pump operators, and, over an eight-year period, one additional black Captain and no additional black Lieutenants. While adding one more black Captain to the current complement of five would be a significant percentage increase, it cannot be said that in a Department with well over 100 Captains, an increase of one black Captain substantially alters the racial make-up of that rank. And the same is true for the three additional black pump operators. But the potential for impact on the racial composition of the Department inheres in the seniority and efficiency figures discussed above. With seniority it can be said that blacks are in a position to be victimized through competition with whites who gained their job seniority at a time when blacks were excluded from the Department or were ineligible for the job because of the prior exclusion. With efficiency, it can be said only that blacks are graded differently. The potential for substantial adverse impact may never be realized, but the possibility that it will may diminish the expectations of some black firemen, and should lead this Court to guarantee that it will not. Moreover, though these promotion factors *1211 have played little part in the present racial make-up of the Department, it is clear that several black firemen have lost promotions due to the seniority situation and differing efficiency ratings. For these individuals, the fact that the practices do not bear heavily on their class as a whole is no consolation. Defendants have not proven that either the efficiency rating system or the use of seniority credits toward promotion is necessary to the conduct of the business of the Fire Department, a burden which they bear.
Promotion in the Baltimore City Fire Department is a composite of success on written examinations, ratings by one's supervisor, and longevity in grade. The application of all these factors has resulted, 20 years after the first black fireman entered the Department, in racial compositions within the Department that vary widely among the ranks. Excepting the new emergency vehicle driver position, the black percentage of each position decreases as one progresses up the line from firefighter to Chief.
As a method of evaluating the racial impact of the overall promotion system, defendants' statistician has computed the expected racial composition of the Department if promotions had been administered fairly since 1953, and compared that table of expectations with the present composition of the Department. If the emergency vehicle driver position is excluded (as it should be since there has been only one test for this new position and most holders of the title were grandfathered in without entering into competition for the position), the resulting discrepancies between the races in the various positions are not statistically significant.
But these calculations, based on comparison of only those men who entered since 1953, are faulty. They presuppose that no blacks can be expected to have progressed to the point of those whites who entered before that date, and thereby posit departmental longevity as the most important variable in determining the rank one holds within the Department. That principle itself is a serious indictment of a Department which prevented blacks from having more than 20 years longevity. Thus, the statistically insignificant differences in Department racial composition, if one controls for entry date, really only describe phenomena already noted: that performance on the written promotion exams has been about the same and that the efficiency and seniority components of promotion have not appreciably altered this essential equality. The defendants' calculations describe a system that has been operating objectively. But they disregard the inescapable fact that white control of the upper echelons of the Department is disproportionate to white presence in the Department and that this is due to the total exclusion of blacks until 1953 and to a system so structured that the effects of the exclusion *1212 are reflected in the composition of the Department 20 years later. The time in grade requirements exert a drag on black ascendance in the Department. The efficiency system has been applied differently to blacks than to whites. The seniority system has penalized some blacks and has potential for doing so to many more, perhaps even to the extent of distorting the objective operation of the promotion system. This system does not amount to the equal protection of the laws mandated by 42 U.S.C. § 1981 and § 1983. Allen v. City of Mobile, 331 F. Supp. 1134 (S.D.Ala.1971), 466 F.2d 122 (5th Cir. 1972); Carter v. Gallagher, supra; Griggs v. Duke Power Co., supra.
Defendants have continued to contest the propriety of the Court's decision to cause this action to be maintained as a class action on behalf of the class described at the outset of this opinion. The gist of their continuing objection is a supposed failure of the four officer litigants to effectively represent all members of the class. Had the Court not been assured by the conduct of this litigation that its previous determination to have it maintained as a class suit was proper, that determination would have been modified or amended. Rule 23(c) (2), Fed.R.Civ.P. This observation is not made in needless refutation of defendants' contention, nor in an attempt to alter later determinations of the res judicata effect of this decision. See Cherner v. Transitron Electronic Corp., 221 F. Supp. 48 (D.Mass.1963). Rather, the Court wishes to make clear that its attempt to tailor relief to those members of the class who have borne the brunt of a particular inequity, should not be interpreted as a belief that the litigants were not a coherent class or the named plaintiffs did not provide effective representation. The fact that an action is grounded in wrongs common to the class members does not mean that certain of the class members may not have been harmed in ways others were not. For the Court to wield its equitable powers without regard to where the harm hit would be improper. In employment situations, courts have often been called upon to tailor their injunctive relief narrowly, for example to make the precise adjustment to seniority systems called for by the evidence. United States v. Chesapeake and Ohio Ry. Co., 471 F.2d 582 (4th Cir. 1972); United States v. St. Louis-San Francisco Ry. Co., 464 F.2d 301 (8th Cir. 1972); Hicks v. Crown Zellerbach Corp., 319 F. Supp. 314, modified 321 F. Supp. 1241 (E.D.La.1970). Such precision in formulating a proper injunction does not indicate that the class was improperly formed. "[T]he Court under F.R.Civ.P. 23 has the duty, and ample powers, both in the conduct of the trial and relief granted to treat common things in common and to distinguish the distinguishable." Jenkins v. United Gas Corp., 400 F.2d 28, 35 (5th Cir. 1968).
Employment discrimination suits are frequently pursued on behalf of classes composed of both employees and non-employees. See e. g., Commonwealth v. O'Neill, 348 F. Supp. 1084 (E.D.Pa. 1972). In such cases, the relief that is afforded at the entry level necessarily benefits the prospective employees more directly than those already hired. The Court must make its remedy comport with the specific harm that has been shown. Intervenors argue that the Court should not compensate blacks in the Fire Department for harm that may have befallen other black persons. Plaintiffs counter that racial discrimination is inherently class discrimination and thus relief is proper if it goes to any member of the class. Both positions are too facile. Class actions afford no excuse for indiscriminate relief. And yet, as previously noted in discussion of the time in grade requirements, the nature of racial discrimination and the conditions in the Fire Department may combine to render injuries directly suffered by some blacks indirectly harmful to others. The Court has fashioned relief *1213 with both of these principles in mind.
This Court has neither the intention nor the authority to supervise the operation of the Fire Department. When requested, it does have the responsibility to see that the procedures for hiring and promoting firemen are within the bounds proscribed by applicable constitutional and statutory provisions.
In carrying out that responsibility, the Court would be remiss in not commenting on several aspects of this litigation. Above the din of battle, one clarion call comes through the service and expertise provided the residents and business community of Baltimore City by the Fire Department. With one voice all witnesses members and nonmembers of the Department have recognized the high quality of fire protection available to the City and the fact that racial problems which have beset the Department's personnel matters have in no way affected the performance of the Department. The relief provided herein does not imply a lowering of those high standards, but rather an improvement of those standards to make certain they determine, without discrimination, who is and who is not qualified, hopefully resulting in bolstering the morale of all firemen by removing impediments to unity in the Department.
High standards of performance were also evident in the work of all counsel in presenting the issues and evidence necessary to enable the Court to consider the merits of the case. The Court is especially indebted to the City Solicitor, who, with unusual perseverance, expertise and good humor, sought out the facts that the plaintiffs could not, or would not, produce, enabling the Court to arrive at an informed conclusion.
The entrance procedures must be purged of discrimination through validation of the firefighter test. The new testing procedures must center on a validated examination, excluding the penmanship examination and the agility component in the final score. The penmanship examination has been proven to be superfluous in the last two years, and the agility test stems from old and misguided efforts to increase the number of blacks in the Department, efforts which have no place in the non-discriminatory procedures the Department must inaugurate.
Until the Fire Department validates the firefighter tests or acquires tests which are valid, hiring cannot be based entirely on the results of such tests. But though using the test to make fine distinctions between men ranking closely on the list would be inappropriate, the test is a logical and adequate method of screening the applicants to the Department. The six month fire school that immediately follows entrance into the Department will ensure that the screening process will produce qualified firefighters.
The Court cannot honor plaintiffs' request for hiring quotas. When racial classifications are imposed by the government, other than as remedies to court suits, they are subject to the "most rigid scrutiny." Korematsu v. United States, 323 U.S. 214, 216, 65 S. Ct. 193, 89 L. Ed. 194 (1944). There have been suggestions that such classifications are per se illegal. Loving v. Virginia, 388 U.S. 1, 13, 87 S. Ct. 1817, 18 L. Ed. 2d 1010 (1967), concurring opinion; McLaughlin v. Florida, 379 U.S. 184, 198, 85 S. Ct. 283, 13 L. Ed. 2d 222 (1964), concurring opinion. And although the full Court has not accepted such a categorical prohibition, the standards which the Supreme Court has applied to this suspect classification seem as rigorous in practice as the flat ban would be.
Despite its concern with racial classifications, the Supreme Court has not dealt with the granting of favors or the creation of rights on a racial basis *1214 as part of a court's remedial powers. In Swann v. Charlotte-Mecklenburg, 402 U.S. 1, 91 S. Ct. 1267, 28 L. Ed. 554, (1971), the Court stressed the traditional breadth of a trial court's remedial powers in fashioning a cure for the illegality of segregation. But the duty of a court to root out the vestiges of school segregation is not precedent for compelling public employers to hire on a racial basis, even as a remedy for past discrimination. Carter v. Gallagher, supra, 452 F.2d at 325. Racial quotas in education impose no burden on anyone, since no one has a right to attend a segregated school. On the contrary, all prospective employees have the right to consideration for public employment without regard to race. As a result of the suspicion with which the law views racial classifications, and the fact that the Supreme Court has yet to address the matter, racial employment quotas may not be valid ingredients in relief. "The constitutional status of such quotas is by no means a clear or settled question." Developments in the Law-Employment Discrimination and Title VII of the Civil Rights Act of 1964, 84 Harv.L.Rev. 1109, 1115 n. 19 (1971). A conclusion of unconstitutionality is not necessary to cause the Court to reject a remedy so at odds with the spirit of the Civil Rights Acts. Classifications such as these "must be viewed in light of the historical fact that the central purpose of the Fourteenth Amendment was to eliminate racial discrimination emanating from official sources." McLaughlin v. Florida, supra, 379 U.S. at 192, 85 S. Ct. at 288.
When the government acts through legislation, the Supreme Court requires a showing of compelling state interest to justify such distinctions based on race. In another context, the Fourth Circuit has interpreted requirements that a discriminatory action be shown to be "necessary" as including a demonstration that no acceptable alternative to the action exists. Robinson v. Lorillard, supra; Lea v. Cone Mills, 438 F.2d 86 (4th Cir. 1971). Technical description of an appropriate standard is not necessary here. The Court simply concludes that the law's rigid scrutiny of racial classifications must be an element of the Court's exercise of its remedial powers. And in this case, no sufficiently compelling need exists for the imposition of quotas. Such quotas would reinforce the view that it is the race of the applicants that is important, rather than their qualifications. That was the operative view in the Baltimore City Fire Department less than a generation ago. Because effective relief can be granted without resort to this device, the Court will avoid it.
*1215 The rejection of racial hiring quotas does not mean that the Court's ability to ameliorate the effects of years of discriminatory hiring is exhausted. In contrast to hiring quotas, alteration of the permissible hiring area is a remedy that is both in keeping with the civil service merit hiring system and suggested by the facts of this case. The Court has already indicated the clear impact which opening the Department to non-city applicants has had on black entrance into the Department. A remedy which corrects this cause of imbalance logically follows and has the additional advantage at the entrance level of reinforcing the importance of personal qualifications, rather than race.
The fire board once restricted its hiring to city residents. This manner of exercising its appointment discretion seems to have been well within its power to further the public welfare through appropriate appointing requirements. There is now a constitutional dimension to the matter, however. Shapiro v. Thompson, 394 U.S. 618, 89 S. Ct. 1322, 22 L. Ed. 2d 600 (1969). And though a trial court's remedial powers are sometimes thought free of constitutional restraint, as when courts mandate hiring on the basis of race, this Court does not wish to offend fundamental rights with its decree. But residency requirements for public employment, drawn for a specific and important purpose and restricted only to the extent called for by the purpose, have no constitutional infirmity. Krzewinski v. Kugler, 338 F. Supp. 492 (D.N.J.1972). This is especially true in light of the more sensitive nature of hiring based on race, the logical alternative to a residency restriction.
In promotion, as in hiring, the decree attempts to correct the impact of discrimination on the racial composition of the Department by enabling blacks to *1216 capitalize on the competitive promotion system. What was said about hiring quotas is at least as applicable to promotion quotas in this case. Black employees have done as well in promotion competition as have whites. The tests, the biggest part of promotion, are fair. Court ordered promotions would be a windfall to employees who cannot complain that they personally should be higher in the Department. That relief would also be unfair to those who otherwise could compete equally for the positions. It is more proper to ensure that the potential for impact on promotions of the seniority system is removed and to charge the fire board with ensuring that the new efficiency system is applied equally to blacks and whites. And most importantly, it is proper to reduce the drag on quick ascendancy in the Department that is caused by inordinate time in grade requirements. These changes are best effected immediately, before the present system is reinforced through another round of promotions from the present eligibility lists. The sixteen officer positions which were left unfilled by court order, will expand the number of promotions for which the enlarged group of eligible firemen can compete equally.
As recompense for the discrimination which has resulted in their holding lower positions, the plaintiffs also seek back pay. Hiring discrimination and a promotion structure which stresses longevity have been identified as the reasons those class plaintiffs presently employed are clustered in the lower echelons of the Department. Back pay does not seem to comport with discrimination of that type. But the normal discretion of a trial court on this issue has been recently circumscribed. In Moody v. Albermarle, 474 F.2d 134 (4th Cir. 1973), the Fourth Circuit held that a district court does not have discretion, absent special circumstances, to deny an award of back pay if discrimination is shown. The decision in Moody concerned a Title VII suit. The Fourth Circuit's decision, however, was based on the broad aims of the statute, and it cannot be said that Title VII is directed at different ends than are § 1981 and § 1983.
Unlike the situation with attorneys fees, discussed below, the decision of the Fourth Circuit as to back pay is not buttressed by decisions in other courts in non-Title VII cases. In fact, back pay in § 1983 cases seems to be the exception rather than the rule, and the Court has seen no opinion by an appellate court which suggests that the decision is anything but discretionary with the trial court. Though back pay is equitable relief, and not to be considered damages, Robinson v. Lorillard, supra; Moody v. Albermarle, supra; Johnson v. Georgia Highway Express, Inc., 417 F.2d 1122 (5th Cir. 1969), the fact remains that the bite of this remedy may be deeper than would be a purely prospective injunction. Given the nature of this remedy, the fact that the broad aims of Title VII are fleshed out in a detailed statutory scheme may be reason to distinguish actions brought under the cryptic sections applicable here.
A related consideration is the fact that we are not dealing here with the classic case of racial discrimination dealt with in Moody. In that case it appeared that the effects of prior segregation were maintained by department seniority and other procedures which inhibited department transfer. This was the situation *1217 in Robinson v. Lorillard, supra, upon which the court in Moody relied. A third Fourth Circuit case which suggests that back pay awards are mandatory is Brown v. Gaston Dyeing, 457 F.2d 1377 (4th Cir. 1972). The deprivation was even more clear cut in Brown the individual had applied for a job and had been told he was ineligible due to his race. By contrast, much of the discrimination in the instant case has had an inestimable effect. This Court cannot say that a given black applicant would have passed a valid entrance exam. The Fourth Circuit evidenced an awareness of this problem of ill-defined effects of discrimination in Robinson v. Lorillard, supra. In response to defendants' argument that the back pay award would be too nebulous, the court noted that "[p]resumably the awards will be based on fixed wage rates and ascertainable periods of work." Robinson v. Lorillard, supra, 444 F.2d at 804. Limiting a trial court's discretion in awarding back pay may result in inequity or chaos unless the discrimination which occasions the back pay has produced a discrete harm to ascertainable individuals. None of the forms of discrimination the Court has found present here fall into that category.
A further problem in the computation of back pay is limitations, discussed above. Though a court can enjoin practices which, though not discriminatory in the abstract, maintain past discrimination, the general rule that limitations begin to run when the injury occurs still stands. The three-year Maryland statute of limitations is the most nearly analogous to § 1983 actions, McIver v. Russell, supra; Hall v. Asher, supra. It may be that § 1981 claims are also most nearly analogous to those based upon a violation of Article 23 of the Maryland Declaration of Rights and thus a three-year limitation period applies to § 1981. 5B Md.Ann. Code art. 57, § 1 (1972 replacement volume). If any back pay were to be awarded, it would be in order only for more recent instances of discrimination. But back pay requests are less appealing in § 1983 suits than in Title VII litigation. And the specific types of discrimination that have been found in this case amount to the special circumstances that the court in Moody had in mind. The Court concludes that no back pay can be granted.
Plaintiffs' request for punitive damages is simply not well taken. Those members of the plaintiff class who suffered under segregation are best compensated by a court order that will ensure that all employees are treated equally.
Plaintiffs also request payment of their litigation expenses. Were this a Title VII case, the availability of attorneys fees could be disposed of by citation of the clear Fourth Circuit rule that attorneys fees must be awarded. Lea v. Cone Mills Corp., supra; Robinson v. Lorillard, supra. The Fourth Circuit drew on Newman v. Piggie Park Enterprises, 390 U.S. 400, 88 S. Ct. 964, 19 L. Ed. 2d 1263 (1968) for that rule. The Piggie Park suit was under the public accommodations title of the 1964 Civil Rights Act. The rationale of that decision, that plaintiffs were vindicating a public policy that Congress considered of the highest priority, was deemed by the appellate court to be equally applicable to Title VII litigation. And though one facet of this "private attorneys general" approach is that the litigants cannot personally recover damages and thus may be deterred from vindicating the public interest by the expense of suit, the availability of damages has not altered the Fourth Circuit rule. See Brown v. Gaston Dyeing Machine Co., supra, a suit based on both Title VII and § 1981.
Federal courts have the discretion as a part of their equitable powers to grant attorneys fees, even absent statutory authorization. Sprague v. Ticonic National Bank, 307 U.S. 161, 59 S. Ct. 777, 83 L. Ed. 1184 (1939). The lack of statutory authorization is particularly unpersuasive where, as in §§ 1981 *1218 and 1983, the statutes do not include detailed remedies. Fleischmann Distilling Corp. v. Maier Brewing Co., 386 U.S. 714, 87 S. Ct. 1404, 18 L. Ed. 2d 475 (1967). And where the suit is one to rectify racial discrimination, the better view is that the court's exercise of discretion to grant fees need not depend on a showing of bad faith on the part of defendants. Lee v. Southern Home Sites Corp., 444 F.2d 143 (5th Cir. 1971); Knight v. Auciello, 453 F.2d 852 (1st Cir. 1972); Sims v. Amos, 340 F. Supp. 691 (M.D.Ala.1972); Wyatt v. Stickney, 344 F.2d 387 (M.D.Ala.1972); NAACP v. Allen, 340 F. Supp. 703 (M. D.Ala.1972). Plaintiffs have effectuated a strong congressional policy by maintaining this suit, and they are entitled to attorneys fees regardless of defendants' good or bad faith.
The Court will order relief from defendants only. Though the individual white firemen who intervened in this suit were beneficiaries of the discriminatory policies of defendants, they did not have the authority in and of themselves to alter the aspects of the entrance and promotion procedures which are the basis for relief.
The foregoing shall constitute the Court's findings of fact and conclusions of law in accordance with the provisions of Rule 52, Federal Rules of Civil Procedure.
Accordingly, it is this 2nd day of May, 1973, by the United States District Court for the District of Maryland,
1. The Fire Department shall hire no additional firefighters on the basis of results of written examinations except as herein provided. There shall be no formal spelling and penmanship screening test, and standing on the firefighter eligibility list shall be determined by written score and veterans points only. All written examinations for firefighter shall be validated before use in accordance with generally accepted professional standards for determining test validity. Applicants for employment who are city residents shall be given preference in hiring over non-city residents as long as there exists a sufficient number of qualified city applicants. There is no requirement that the hiring preference be extended to require continued residence in the city. Prior to the first administration of a validated firefighter examination, the Department shall personally notify all previously unsuccessful black applicants at their last known address that a non-discriminatory examination is to be given.
2. Pending validation of a firefighter test, the Department may hire from a list of those persons who took the firefighter examination administered on March 3, 1973, and passed the examination with a score of 85 or higher. Selection from this list of qualified applicants shall be made at random, excluding non-city residents unless the pool of qualified city residents is exhausted.
3. No further promotions shall be made from any promotion eligibility list for the rank of Battalion Chief and below, including all specializations. The 16 positions left vacant as a result of court order will be filled through normal procedures as a result of new examinations.
4. In all cases in which black employees participated in promotion competition, no credit toward promotion shall be given for seniority gained prior to 1953 for promotion to Lieutenant and below, nor for seniority gained prior to 1958 for promotion to Captain, nor for seniority gained prior to 1960 for promotion to Battalion Chief, nor for seniority gained prior to 1962 for promotion to Deputy Chief, nor for seniority gained prior to 1963 for promotion to Chief of the Department.
5. The Board of Fire Commissioners shall ensure that blacks and whites are graded equally by promulgation of regulations governing supervisor's efficiency ratings.
6. Time in grade requirements for promotion to Lieutenant shall be no longer than three years. The time in *1219 grade prerequisite for promotion to other officer positions shall be no longer than one year.
7. Attorneys fees and costs shall be awarded to plaintiffs in an amount to be determined by the Court upon application of the interested parties.
8. The preliminary injunction entered October 18, 1972, as amended March 12, 1973, and the corresponding eligibility list extensions, are vacated.NOTES
 The order provided that the action would be maintained under Rule 23(b) (2), Fed. R.Civ.P., rather than under Rule 23(b) (3). Class action suits under subdivision (b) (2) may append claims for monetary relief to requests for injunctive and declaratory relief. Robinson v. Lorillard, 444 F.2d 791 (4th Cir. 1971). That subdivision was added in 1966 in part to make it clear that civil rights suits for injunctive or declaratory relief can be brought as class actions. When an action meets the criteria of both subsections 23(b) (2) and 23(b) (3), Rule 23(b) (2) is to be preferred so that the res judicata effect of the action may be preserved. 3B Moore, Fed. Practice 23.72 ; Wright and Miller, 7A Federal Practice and Procedure § 1775. Though not required by the provisions of Rule 23(c) (2), the Court determined that notice of the pendency of the action was in order both to advise the class members of their right to intervene and to ensure that the Court's views as to the adequacy of plaintiffs' representation were shared by the class members. The notice included individual notification to present and former black employees. No members of the class opted to intervene and none dispute the effectiveness of representation by the named plaintiffs.
 DIX-1. Refinements in the census figures for sex and age alter only slightly the percentage of these labor pools which blacks compose.
 DX-1 indicates that since the residency limit was expanded to 30 miles, approximately one-third of those certified as eligible for entrance to the Department have consistently been non-city residents. Because that same exhibit also indicates the scores of non-city test passers tend to be higher than average, it is reasonable to conclude that their relative advantage in performance on the test extends to greater passing rates. And thus, while non-city residents represent one-third of test passers, they no doubt represent a smaller percentage of applicants.
 Plaintiffs argue both that the population discrepancy proves discrimination as a matter of law, and that the discrepancy establishes a prima facie case of discrimination that shifts the burden of proof to the defendants. Courts which have given great weight to such a statistical discrepancy have not gone further than according the statistics the status of a prima facie case. See Carter v. Gallagher, 452 F.2d 315 (8th Cir. 1971); Western Addition Community Organization v. Alioto, 330 F. Supp. 536 (N.D.Cal.1971); Bridgeport Guardians v. Civil Service Commission, 354 F. Supp. 778 (D.Conn. 1973). A population comparison ignores the fact that all segments of the population may not be equally qualified for the positions in question, Chance v. Board of Examiners, 330 F. Supp. 203 (S.D.N.Y. 1971), and different groups within the population may have different levels of desire for the particular job, Castro v. Beecher, 334 F. Supp. 930 (D.Mass.1971). And most importantly, acceptance of the idea that discrepancies between racial composition of the community and the plant or department alone make out a prima facie case of discrimination leads inevitably toward a narrowing of the Court's options in fashioning a remedy. If the problem is to be demonstrated by the mere fact of a discrepancy, then the solution logically must amount to an order to bring the employment statistics into line with the population statistics, lest the Court mandate a continuing prima facie violation. Hiring in that manner in the first instance is not required by law. No citizen has a constitutional right to have public employees perfectly reflect the racial composition of the hiring unit. Castro v. Beecher, 334 F. Supp. 930 (D.Mass.1971), modified 459 F.2d 725 (1st Cir. 1972); Stebbins v. State Farm Mutual Insurance Co., 5 FEP cases 142 (4th Cir. 1972). And courts must be careful not to give credence to such misconceptions through their evidentary requirements. To hold that a discrepancy between the employment population and the community population is prima facie a violation of law would not be wise, and it is not required. Logan v. General Fireproofing Co., 3 FEP cases 854 (4th Cir. 1971).
 The Department was organized in 1858. PX-7.
 One plaintiff testified that he did not even become aware of promotion exam study material which was kept in his fire house until after he assumed a position of authority in the house. Yet that material was apparently freely available to whites in the house. The material in question was forwarded to the house by the fire school for the express purpose of facilitating in-house preparation for promotion exams, and it was clearly marked to indicate that it was not to be the personal property of anyone.
 The drop out rate figures which have been presented describe the attrition problem by entry group rather than by year of departure. We cannot as confidently say that undue black attrition was a problem of the 1950's as we can say that it was a problem with those black firemen who entered during the 1950's. There are two sources of data on attrition rates. PX-55 lists attrition rates by entering year groups. The overall difference in attrition rates for all the years presented, 1953 to 1970, has been determined to be statistically significant. 31% of entering blacks left the Department during that period, and only 17% of entering whites. DX-34H and 34I analyze attrition rates for various shorter periods. For the two periods covering the 1950's, black attrition rates were greater than white at a statistically significant level. The slightly greater white attrition rate for the 1960's year groups is not statistically significant. Correction of one error in the defendants' calculations (failure to include one 1957 entry list from the raw data contained in DX-36A) would alter the percentages but not the percentage differences. A year by year analysis of the plaintiffs' figures indicates corroboration of the fact that the disproportionate black attrition rate was a problem of the 1950's year groups.
 As already noted, the attrition data compares cohorts, firemen who entered the Department together. Thus the data controls for variations in attrition rates among different ranks in the Department.
 Charter of Baltimore City, Art. VIII, § 45 (1964 rev.). "The Board [of Fire Commissioners] shall have control, regulation and supervision of the Department, the personnel and properties thereof and all matters relating to the same, and shall perform such other duties as may be required by ordinance not inconsistent with the Charter."
 It has been argued that the segregation practiced in the Department was of a kind with that practiced throughout the City. But some distinction should be made between the clear duty of a public entity and the practices of more private organizations. The repudiation of Plessy v. Ferguson, 163 U.S. 537, 16 S. Ct. 1138, 41 L. Ed. 256, in 1954 by Brown v. Board of Education, 347 U.S. 483, 74 S. Ct. 686, 98 L. Ed. 873, was well understood to have implications beyond public schools. Those implications were forcefully brought home to the City of Baltimore one year later by Dawson v. Mayor and City Council of Baltimore, 220 F.2d 386, (4th Cir. 1955) aff'd per curiam, 350 U.S. 877, 76 S. Ct. 133, 100 L. Ed. 774. Segregation persisted in the Baltimore Fire Department for more than a decade after these decisions established it to be a violation of the Fourteenth Amendment.
 Lacking a statute of limitations of their own, the Civil Rights Acts are governed by the most nearly analogous state statute of limitations. Young v. I. T. & T., 438 F.2d 757 (3rd Cir. 1971); Crosswhite v. Brown, 424 F.2d 495 (10th Cir. 1970); McIver v. Russell, 264 F. Supp. 22 (D. Md.1967). Because this § 1983 suit is not one "for which the sole remedy is in equity," Holmberg v. Armbrecht, 327 U.S. 392, 395, 66 S. Ct. 582, 584, 90 L. Ed. 743 (1946), it is appropriate to apply limitations rather than the equitable doctrine of laches even though plaintiffs seek only equitable relief. Swan v. Board of Higher Education, 319 F.2d 56 (2d Cir. 1963).
Maryland law establishes a three year limitation for actions based on Article 23 of the Maryland Declaration of Rights and actions based on simple contract, and most other common law actions, 5B Md. Ann.Code, art. 57 § 1, and a twelve year limit for actions based on "specialities," art. 57 § 3. Though a suit based upon a statute is a suit upon a speciality, Roland Electrical v. Black, 163 F.2d 417 (4th Cir. 1947), suits which owe their existence to a statute are not necessarily regarded as "based on" that statute. The distinction is not purely semantic. There is strong precedent for the conclusion that actions based on Article 23 of the Maryland Declaration of Rights, which concerns the deprivation of life, liberty or property except in accordance with the law of the land, are more nearly analogous to § 1983 than are specialities. Hall v. Asher, 355 F. Supp. 808 (D.Md., 1973); McIver v. Russell, supra. But whether the twelve year or three year limit is chosen, the segregation that existed in the Department would still be properly a part of this suit. Action by the fireboard was necessitated by the vestiges of segregation as late as 1970. PX-20. Suit was filed in 1971.
 Quarles v. Philip Morris, Inc., 279 F. Supp. 505 (E.D.Va.1968); Griggs v. Duke Power, 401 U.S. 424, 430, 91 S. Ct. 849, 853, 28 L.Ed.2d 158: "Under the Act, practices, procedures, or tests neutral on their face, and even neutral in terms of intent, cannot be maintained if they operate to `freeze' the status quo of prior discriminatory employment practices." The Supreme Court was referring to Title VII, but the Fourteenth Amendment prohibition against discrimination by public employers imposes a standard at least as stringent. Baker v. Columbus Municipal Separate School District, 329 F. Supp. 706 (N.D.Miss.1971). It would be anomalous to apply the Griggs requirement only to private employers. "The anomaly would only be emphasized by the recent passage of the Equal Employment Opportunity Act of 1972, which broadened Title VII to include state and city public employers." Chance v. Board of Examiners, 458 F.2d 1167, 1176-1177 (2d Cir. 1972). The Court considers broad principles of equal employment opportunity judicially developed under Title VII as generally applicable to this § 1983 suit.
 Charter of Baltimore City, Article VII § 115(e) (1964 rev.)
 Id.: "Certification by the commission shall include the five persons standing highest on the appropriate list." Utilization of the "rule of five" may have facilitated the exclusion of blacks from the department which persisted until 1953, especially since the names of blacks were entered on the list in red ink. The "rule of five" applies seriatum down the list so that it has the effect, where a long list of names is certified, of permitting the appointing officer to choose any name from that list. There was testimony that the fire board in 1947 ordered all appointments to the Department to be made in strict numerical order. The method by which the appointing authority in the Department maintained black exclusion for six years after the Department's abrogation of the "rule of five" is not explained by the record, though there is no doubt that the exclusion continued. The fire board still requires its selecting panel to explain deviation from strict numerical selection.
 See DX-65, the announcement for the 1973 examination. The Charter gives the Civil Service Commission the power to prescribe "reasonable requirements" for applicants for the Fire Department and other City jobs. Physical requirements are specifically included, education requirements are not. Charter of Baltimore City, Article VII § 115(c) (1964 rev.).
 See DX-62. The residency requirement varied over the course of a decade from City residency to residency within 30 miles of the City, which is the present requirement.
 A veterans credit is specifically provided by the Charter, Article VII, § 115 (f). Analysis of the 1971 firefighter examination revealed that almost the same percentage of black and white applicants to the Department benefited from the veterans credits.
 PX-71B1b and 71B1c.
 The resulting examination is found at PX-71B3.
 The Transcript of testimony of Dr. Barbara Lyles, February 1, 1973, p. 46. A black psychiatrist called by defendants took issue with this recommendation of the panel which he felt pandered to racial myths. He cited it as one example of reinforcement of such myths by people, blacks as well as whites, who have advocated the cause of the black firemen. Though the psychiatrist went far beyond psychiatry in much of his testimony, it is clear that racial and ethnic stereotypes were never far below the surface of this case, and may have been a part of the problem in the Department and the shortcoming of efforts at reform. Some of these myths, such as poor black performance on written exams and control of the Fire Department by one ethnic group, have been effectively debunked by the evidence in this case. Others will no doubt continue to plague the Department, as they do the society.
 Press release for the 1971 firefighter exam, attached to the 1970 list of applicants to the Department in DX-35A.
 No exhibit sets out this comparison of application rate and appointment rate in toto. All figures for appointments come from DX-34A. The black application rates come from PX-58A. They were derived from analysis of Civil Service Commission exam histories which did not include racial identification. That identification was made by a team of black firemen under the supervision of a member of the Baltimore Community Relations Commission based in large part on street addresses. Because no census tracts were used, the identification was error prone. But the supervising member of the Community Relations Commission testified that where there were doubts as to racial identification, the individuals were indicated to be white. On the 1971 exam, where Civil Service Commission figures are available as a check, it is apparent that the group of black firemen did underestimate the number of black applicants. Compare PX-26 with PX-24A. Because the black percentage of applicants, but not of appointments, were underestimated, the resulting discrepancies between application rates and appointment rates are probably larger than indicated.
After trial of this suit was completed, plaintiffs move to enjoin appointments from an eligibility list derived from a new firefighter examination administered during the course of trial. At the hearing on this motion, the statistical results of the new examination were presented. They corroborate prior experience. 45% of the applicants were black. If appointments had been made as expected from the top 161 names on the list, based on written scores alone, only 12% of those offered appointment would have been black.
 PX-58. Statistics in this exhibit were drawn from information developed by the racial identification procedure discussed in footnote 24 above.
 The agility component of the 1971 exam obviously did not help blacks appreciably. Both parties are apparently of the view that the non intellective component of the 1971 test helped neither race.
 The possibility that discrimination within the Department and by the entrance tests has resulted in depression of the rate at which blacks apply to the Department for employment is belied by the continuing high percentage of black applicants. It has been noted that, for the years 1956, 1957, 1960 and 1963, black application rates exceeded 40% each year and were in excess of even the City black population. In 1971 and 1973, black application rates were greatly in excess of the black percentage of the potential hiring area, which by then had been extended to 30 miles from the City. See PX-63 and DX-12. In fact, the 1973 black application rate compares favorably with the black population of the City. In the face of these statistics and the efforts of the Department to recruit more black applicants, the Court will not speculate that the relatively high application rates are less than they might have been.
 DX-34D. The experience in 1973 corroborates the earlier data. 35% of the top scorers on the 1973 firefighter exam reside outside the City. Plaintiffs' Motion Exhibit 2.
 The trend toward an increasing percentage of non-city residents on the eligibility lists is almost the inverse of the trend toward a decreasing percentage of black appointees. DX-34A and 34D. Had City residency remained a requirement, we might have expected black application rates to rise with the City's black population and black appointment rates to rise accordingly.
 Defendants' post trial brief, p. 47: "[I]f the number of blacks selected for firefighter is to increase in the future, some consideration as to place of residency should be given."
[33A] Courts have treated validity more as a question of status than of degree. In Griggs, the Supreme Court held an employer to a showing that its tests had a "manifest relationship to the employment in question," 401 U.S., at 433, 91 S. Ct. at 854.
 Both the EEOC "Guidelines on Employee Selection Procedures," 29 C.F.R. Part 1607, and the United States Civil Service Commission testing standards, 37 Fed.Reg. 198 (Oct. 12, 1972) require such a careful job description. Therefore, for this matter it is not necessary that the Court choose between the two. Intervenors argue that the Commission standards are the more germane to this suit, since we are dealing with the peculiarities of a merit employment system. Plaintiffs argue that this Court should accept the EEOC guidelines, as so many other Courts have done. The Supreme Court has noted that the EEOC guidelines are entitled to great deference. Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971). But the Court stopped short of formally adopting them for the use of federal courts, even in Title VII cases. Its endorsement in Griggs followed from the conclusion that the guidelines, on the point there considered, were consistent with Congressional intent. Thus this Court's intention to minimize conflicts between § 1983 suits and Title VII suits does not mean that it need accept uncritically the EEOC guidelines any more than it need accept the United States Civil Service Commission guidelines. The argument that the Commission standards are more germane has no appeal except on issues where the nature of merit hiring creates a special situation.
 Other courts have found PPA tests to have an adverse impact on blacks, Bridgeport Guardians v. Civil Service Commission, 354 F. Supp. 778 (D.Conn.1973); Allen v. City of Mobile, 331 F. Supp. 1134 (S.D.Ala.1971). In the former case the court also found the tests invalid. In the later case they were determined to to be valid. In neither suit were the tests in question firemen's tests.
 If the discrepancy in test-taking ability were a fact, then it would follow that, generally speaking, a lower score for a black would predict the same level of job performance as a higher score by a white. And this idea of differential validity was not shown to have any merit.
 The EEOC guidelines suggest that "training time" is a permissible criterion. 29 C.F.R. 1607.5(b) (3). The United States Civil Service Commission standards more directly state that tests may serve as valid criteria for other tests. But neither of these sources suggest that this criterion has worth unless there has been reliance on a careful job analysis in test formulation. A more reliable validation method could include such tests as a criterion, perferably not the only one, for a test developed in a sound fashion. This was the approach of the Educational Testing Service in a lengthy study of test differential validity. In a situation factually similar to this one, a validation study utilizing training school test performance as a criterion was rejected as inadequate. Commonwealth v. O'Neill, 348 F. Supp. 1084 (E.D.Pa.1972), modified 5 FEP cases 713 (3rd Cir. 1973).
 The emergency vehicle driver position was established in 1972. Firefighters serving in that capacity at the time of formal creation of the new position were excused from the competitive examination. DIX-6.
 The efficiency ratings were weighted at 40%, with a corresponding decrease in the importance of the mark on the written exam, until 1966.
 In the Fourth Circuit, "it go[es] without saying that a practice is hardly `necessary' if an alternative practice better effectuates the intended purpose or is equally effective, but less discriminatory." Robinson v. Lorillard, 444 F.2d 791, 798 (4th Cir. 1971). Other circuits, including the Eighth, United States v. St. Louis-San Francisco Ry. Co., 464 F.2d 301 (1972), the Fifth, United States v. Jacksonville Terminal Co., 451 F.2d 418 (1971), and the Second, United States v. Bethlehem Steel, 446 F.2d 652 (1971), have also amplified the business necessity rule with the requirement that the employer demonstrate that the objectionable practice is the only way to accomplish his legitimate objective. The less drastic alternative requirement has not met with wholesale acceptance in § 1983 cases. In Castro v. Beecher, 334 F. Supp. 930 (D.Mass. 1971), the court specifically rejected that approach, and the Court of Appeals approved. 459 F.2d, at 733. In both Commonwealth v. O'Neill, 348 F. Supp. 1084 (E.D.Pa.1971) and Chance v. Board of Examiners, 330 F. Supp. 203 (S.D.N.Y. 1971), 458 F.2d 1167 (2d Cir. 1971), the court never got to the issue of the requirement of showing no less drastic alternative because the test was not even shown to be valid, the first step. The distinction between the Title VII cases and the § 1983 cases seems to to be that the latter cases have raised essentially dissimilar issues, issues for which the less drastic alternative test seems too harsh. In fact, though the EEOC guidelines require such a showing as part of validation, 29 C.R.F. 1607 the Court is aware of no court decision in which the validity of a test has been weighed with that in mind.
The issue would lend itself less to haphazard resolution in § 1983 cases if it were considered along with the general standard for review of government actions which put a burden on race. See, generally, Bridgeport Guardians v. Civil Service Commission, 354 F. Supp. 778 (D.Conn. 1973). The question is whether courts are to apply the compelling state interest test or the rational relationship test, or something in between, to actions by the government, such as employment testing, which discriminate but not intentionally. If the compelling interest test is not called for, then it would be difficult to justify a requirement that the government employer examine each other avenue to discover whether there may be one which is as effective, yet less burdensome to racial minorities.
 The interest of any union in a matter affecting seniority rights should not need elaboration. When the issues first arose in the Fire Department, the representative of the fire officers union made it clear that his organization would take a position on any reduction in the time in grade requirements. PX-6.
 Plaintiffs have tried to draw a distinction between the value of acting positions as preparation for the exam for the higher position and the value of practice in the higher position as preparation for the position itself. There is circularity in that distinction. If holding the position on a temporary basis actually prepares for the exam for the position, then the exam can only be a true reflection of the demands of the job.
 DX-52. On the average, firefighters who receive promotions require approximately ten years to reach Lieutenant and officers five years to make the next higher grade.
 Amended Complaint, II, 9(e).
 DX-40(c). A performance test is also a component of the pump operator promotion process. There has been testimony that blacks have been prejudiced by their inability to prepare equally for this phase of the promotion process. And there is evidence that white pass rates on the performance component were somewhat higher in the earlier years. But pass rates on the performance tests have been 100% for both races for six years. And there has been no explanation of why, if the normal patterns of discrimination caused lower black pass rates at one time, that adverse impact so abruptly stopped. The Court cannot conclude from these figures and scant testimony that discrimination on the performance test, strictly historical at most, ever really existed.
 DX-40(e) and (g). Only 22 blacks took the test for Captain during the years studied, whereas 363 whites took such examinations during the same period. However, statistical tests which take sample size into account indicate that the superior black success rate on Captain exams is statistically sound.
 DX-41(a), (b) and (c).
 Plaintiffs' Post Trial Brief, p. 32.
 Expert testimony presented by plaintiff to the effect that the manner in which the promotion tests were prepared ensured their invalidity was more than counterbalanced by defendants' firefighting expert's study of the content validity of the promotion exams, DX-29, even if that study had not been buttressed by an empirical validation study. DX-41(e), (f) and (g). Standing alone the empirical validation study would not have sufficed for validation of an exam having adverse impact. Moody v. Albermarle, 474 F.2d 134 (4th Cir 1973). As discussed below, there is evidence that blacks have been assigned efficiency ratings differently than have whites. In light of this discrepancy, it does not seem wise to make efficiency ratings the criterion for a validation study. But there was no demonstration of adverse impact. And the evidence that has been presented regarding validation simply makes an inference of invalidity from equal performance even less permissible than it would otherwise have been.
 The examinations were for supervisory level positions in the New York City school system.
 Chance v. Board of Examiners, 330 F. Supp., at 210.
 The Court is aware of no case which has used the "multiplier effect" analysis to find adverse impact from a test on which performance was equal, but for which an invalid test served a screening function. In Bridgeport Guardians v. Civil Service Commission, supra, the Court rejected the argument that the promotion tests were invalid under Chance, but the rejection was due to the tentative nature of the small sample-size statistics, rather than to a conviction that equal performance on the higher exam was sufficient. In Commonwealth v. O'Neill, 348 F. Supp. 1084 (E.D.Pa.), 4 FEP cases 1286 (3rd Cir. 1972), rehearing en banc 5 FEP cases 713 (1973), though a substantial disparity in test passing rates at the entrance level was clear, confusion over the significance of promotion level statistics led to reversal of the preliminary injunction as to promotion. If the "multiplier effect" analysis were thought applicable to that situation, it would seem that the Court of Appeals would have upheld the promotion level injunction despite the confusion as to the degree of impact of the promotion exams.
 PX-52A and 54B.
 DX-41D. Since the efficiency ratings are attacked as a discriminatory ingredient of the promotion process, it is appropriate that the evidence has been limited to a comparison of efficiency ratings of candidates for promotion who have passed the written test.
 PX-56A and DX-52. Plaintiffs' calculations for time between promotions from Lieutenant to Captain were made at a time when there was only one black Captain. Similar calculations by defendants still are based on only five black Captains. Disregarding that time period, it appears that, on the average, blacks take 1.2 years longer to become pump operator, and 1.5 or 2.5 years longer, depending on which statistics are accepted, to attain the rank of Lieutenant. The difference in promotion time before Lieutenant was shown to be statistically significant.
 All parties agree that the restricted range of efficiency ratings necessarily diminishes any impact they may have. Plaintiffs' Post Trial Brief, p. 11.
 PX-55A. PX-1 reveals total Department employment of 2236, of whom about 2180 were uniformed firemen, white and black. Over 600 men entered before 1953. Thus close to one-third of all white firemen entered the Department before the first black.
 Comparison of a roster of employees and an enumeration of post-1953 entries indicates that less than half of the 600 employees who entered since 1953 have been promoted to the officer level. This means that over 300 whites currently enjoy an advantage over blacks because of their pre-1953 entry date in competition for Lieutenant alone. PX-1, DX53B. It is just as likely that large numbers of white Lieutenants gained their present rank before 1958, large numbers of Captains before 1960 and so on.
 The fact that the effect of seniority and efficiency were not determined separately does not diminish the worth of this data. The effect of one will not mask the effort of the other, since there is a high correlation between the two.
 Defendants contend that black firemen do not apply for promotion as frequently as white firemen. DX-55. Were that so, then diminished expectations of black firemen due to use of efficiency and seniority in promotion competition might be more than conjecture. But the evidence which shows a lower percentage of blacks than white applying for promotion may only indicate that proportionately more blacks than whites have not been in the Department long enough to qualify for promotion.
 The Court cannot identify those blacks who have been denied promotion due to use of the pre-1953 seniority, or even accurately estimate their number. Evidence previously discussed indicates that very few blacks would have been promoted if only test score, rather than all three factors, had been relied on, and it has been noted that it is probably accurate to simultaneously consider the effects of seniority and efficiency. However, the calculations were made disregarding all seniority rather than only pre-1953 seniority. The more proper inquiry for these purposes would have been to disregard only pre-1953 seniority. Since time between promotions for blacks is longer than for whites, the calculations may have concealed the true impact of pre-1953 seniority by disregarding legitimate advantage from post-1953 seniority.
 DX-53A. These rank-to-rank discrepancies are not themselves illegal, but do require explanation. United States v. Chesapeake and Ohio R. Co., 471 F.2d 582 (4th Cir. 1972).
 DX-53 B and C.
 All relief concerns the uniformed employees of the Department. Plaintiffs' allegations regarding the non-uniformed staff were not borne out by the evidence.
 Griggs v. Duke Power Co., 401 U.S. 424, 91 S. Ct. 849, 28 L. Ed. 2d 158 (1971), did not concern the constitutional dimension to this problem of discrimination through hiring quotas, as some courts have suggested. The Court in Griggs dealt exclusively with the anti-preference provision of Title VII, 42 U.S.C. § 2000e-2 (j). Moreover, that decision involved substantive illegality, not limitations on remedial powers.
 A basic difficulty in the field of equal protection is that government inaction may sometimes insure the inequality. Norwalk CORE v. Norwalk Redevelopment Agency, 395 F.2d 920 (2d Cir. 1968), is a fair example of this complexity. Because of the inequities of the housing market, the court there concluded that provision by the public housing authority for displaced blacks on a par with that for whites would have fostered inequality rather than curtailed it. In that housing situation, the benefits followed almost directly from the opportunity for them. And so the court could not deal with unequal opportunity without also apportioning the end product on the basis of race. But a racial employment quota does not amount to an attempt to make members of both races equal at the starting point of the search for jobs. Rather, since the actual employment is the end result of the "race", an employment quota is a statement that the only effective way to even out the competition is to eliminate it.
 Courts are in substantial disagreement as to the propriety, or legality, of racial employment quotas as remedies for discrimination in public employment. In the leading case on the subject, Carter v. Gallagher, 452 F.2d 315 (8th Cir. 1971), the full bench modified the earlier disavowal of quotas, except as to "absolute preference" quotas. In Castro v. Beecher, 334 F. Supp. 930 (D.Mass.1971), 459 F.2d 725 (1st Cir. 1972), the Court of Appeals reversed the district court's refusal to institute a hiring quota, but the quota instituted by the appellate court extended only to those plaintiffs who had taken and failed an invalid test. In Chance v. Board of Examiners, 330 F. Supp. 203 (S.D.N.Y.), 458 F.2d 1167 (2d Cir. 1971), the court of appeals characterized the district court's position as a "specific rejection" of employment quotas. In Commonwealth v. O'Neill, 4 FEP cases 1286 (3rd Cir. 1972), the Court of Appeals for the Third Circuit vacated the temporary hiring ratio that the district court had imposed on the ground that it mandated hiring unqualified individuals. The court stated that it was "unnecessary at this time to determine the constitutionality of an order imposing a mathematically-fixed quota system upon the hiring and promotion procedures of a department of municipal government." On rehearing by the full panel, the hiring order of the district court was upheld without discussion by an equally divided court. 5 FEP cases 713 (3rd Cir. 1973). In Bridgeport Guardians v. Civil Service Commission, 354 F. Supp. 778 (D.Conn. 1973), the court imposed a hiring ratio at both the entry and promotion levels. The court saw hiring and promotion "goals" as the alternative to quotas and felt that they amount in the end to the same thing. In NAACP v. Allen, 340 F. Supp. 703 (M.D.Ala.1972), the court imposed a 1:1 hiring ratio until 20% of the state police force was black. In Shield Club v. City of Cleveland (N.D. Ohio, Jan. 16, 1973) the quota was considerably more narrow. Pending validation of the entrance test, the court required those hired to include a minority proportion equivalent to the proportion of the minority persons who had passed the test. In Arrington v. Massachusetts Bay, 306 F. Supp. 1355 (D.Mass.1969), the court refused to order preference hiring of minorities, reasoning that giving preference to them would disadvantage the others on the list.
 Raney v. County Commissioners, 170 Md. 183, 183 A. 548 (1936); McQuillen, The Law of Municipal Corporations, § 12.80 (3d ed. 1949). Ordinances to the same effect have been routinely upheld, and the standard for fire board discretion without benefit of ordinance seems no more stringent. Berg v. City of Minneapolis, 274 Minn. 277, 143 N.W.2d 200 (1966); Kennedy v. City of Newark, 29 N.J. 178, 148 A.2d 473 (1959); Salt Lake City Firefighters v. Salt Lake City, 22 Utah 2d 115, 449 P.2d 239 (1969).
 Memorandum opinion and order of October 18, 1972, as amended March 12, 1973.
 Cases finally disposed of in which back pay was not awarded, though it would have been at least as appropriate as in the instant case, include Carter v. Gallagher, 452 F.2d 315 (8th Cir. 1971); Castro v. Beecher, 334 F. Supp. 930 (D.Mass.1971), 459 F.2d 725 (1st Cir. 1972); Allen v. City of Mobile, 331 F. Supp. 1134 (S.D. Ala.1971), 466 F.2d 122 (5th Cir. 1972); Chance v. Board of Examiners, 330 F. Supp. 203 (S.D.N.Y.) 458 F.2d 1167 (2d Cir. 1971); NAACP v. Allen, 340 F. Supp. 703 (M.D.Ala.1972), and Bridgeport Guardians v. Civil Service Commission, 354 F. Supp. 778 (D.Conn.1973).