Full opinion text
MEMORANDUM WARRINER, District Judge. This is a class action for declaratory, injunctive and other relief brought against the City of Richmond, Virginia and several of its employees. The complaint alleged that the Richmond Fire Bureau engages in acts and practices that discriminate against blacks in their employment because of their race. Jurisdiction is conferred upon this Court by 28 U.S.C. § 1343. The plaintiff class was certified as all Negroes who have, within the applicable period of the statute of limitations, been discriminated against with respect to employment practices by defendants and who (a) are presently employed in the fire department of the City of Richmond or (b) formerly have been employed in the fire department of the City of Richmond. Prior to trial, plaintiffs dropped their claims under the post-Civil War Civil Rights Acts and their constitutional claims. The case was tried only under the Civil Rights Act of 1964 on the question of liability by the Court. Title VII of the Civil Rights Act of 1964 was made applicable to local government employment by passage of the Equal Employment Opportunity Act of 1972 on 24 March 1972. The two named plaintiffs in this action, Roscoe Friend and Theodore Fuller, filed a charge with the Equal Employment Opportunity Commission on 11 November 1974, pursuant to 42 U.S.C. § 2000e-5. The statute requires that a charge be filed with the Commission within one hundred and eighty days after the alleged unlawful employment practice occurred. 42 U.S.C. § 2000e-5(e).' Therefore, only those acts and practices which occurred on or after 11 May 1974 may be a basis for a finding of liability in this case. See United Air Lines, Inc. v. Evans, 431 U.S. 553, 97 S.Ct. 1885, 52 L.Ed.2d 571 (1977). It is abundantly proven and not denied that defendants or their predecessors in office practiced overt and officially-sanctioned racial discrimination up to the year 1963. Such racial discrimination in earlier years was not considered to be in violation of the law of the land; in fact, it was generally deemed to be in obedience to State law and not prohibited by federal law. This nation’s evolving concept of decency, morality and constitutional law underwent revolutionary changes during the decades following Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954). The fire department of the City of Richmond, though clearly not leading the way, followed in the train of that revolution. In 1950, the first Negroes were hired as firemen in the City of Richmond. They were organized into a separate fire company, Company No. 9, located in a Negro neighborhood. White officers were assigned as their superiors. Change followed as Negro officers were trained and assigned to the segregated fire company. Then, in 1963, the black firemen were dispersed among other units in all areas of the city. As time went on black firemen became assimilated into the fabric of the fire department. The battle had been won but mopping up operations continued. The present case is one such operation. I Plaintiffs have compiled extensive statistics on the Fire Bureau’s disciplinary policies relating to white and black firemen. Particularly, plaintiffs contend that the Fire Bureau discriminates on the basis of race by using garnishments as the basis of disciplinary action and by taking more severe action against blacks than against whites who are involved in traffic accidents. The two arguments do not follow the same path. Plaintiffs contend that the mere use of garnishments as a basis of disciplinary action is per se a violation of Title VII, since blacks have their wages garnished more often than whites. Plaintiffs contend that the greater frequency of “chargeable” incidents among black firemen who are involved in traffic accidents is evidence of discrimination by those Fire Bureau personnel who determine whether an incident is chargeable or non-chargeable. A. Garnishments Plaintiffs’ contend that defendants practice racial discrimination by using the fact that a fireman’s wages have been garnished as a basis for employment decisions. The evidence purported to show that while only 2% of the white firemen had their wages garnished in the period from 1972 to 1976, 40% of the black firemen had their wages garnished in the same period. Plaintiffs’ Exhibit No. 104B. Dr. Gittleson, the plaintiffs’ statistics expert, testified at trial that this difference was statistically significant. Record, page 213. Plaintiffs’ evidence on the number of garnishments for black and white firemen is contained in Plaintiffs’ Exhibits Nos. 103 and 104. Plaintiffs’ Exhibit No. 103 contains notations made on the basis of observations from Fire Bureau personnel files. Plaintiffs’ Exhibit No. 104 is a summary of the data found in 103. Dr. Gittleson testified that he had no knowledge of the meaning or accuracy of the data he was given, which was taken directly from Plaintiffs’ Exhibit No. 104. To the extent there are any inaccuracies in these exhibits, Dr. Gittleson’s testimony must be discounted. Plaintiffs’ Exhibit No. 104B shows a total of 41 garnishments of black firemen from 1 July 1972 to March 1977. However, analysis of the documents on which this compilation is based shows that 15 of these garnishments involved the same fireman, one A. Irving. Plaintiffs’ Exhibit No. 103, page 71. Five of the 41 involve one Irving S. Friend. Plaintiffs’ Exhibit No. 103, page 34. Thus, almost half of all garnishments of black firemen are attributable to two individual firemen. The Court cannot say what Dr. Gittleson’s testimony would be on the statistical significance of the so-called “garnishment rate” between black and white firemen if the garnishments attributable to these atypical individuals were removed from the sample. Another serious error appears from analysis of Plaintiffs’ Exhibit No. 103. On page 34 of that Exhibit there appears an entry for one Irving, Andrew, who was hired on 3 July 1965. Among other things, the entry records garnishments for Mr. Irving dated 27 October 1975 and 20 August 1975. On page 71 of that Exhibit there appears an entry for one A. Irving, hired 3 July 1965, with garnishments on 27 October 1975 and 8 August 1975. It is apparent that these two entries concern the same individual, and that at least two garnishments have been counted twice. These errors cast a pall over the plaintiffs’ whole case on garnishments and disciplinary actions. Dr. Gittleson’s testimony on the statistical significance of data compiled from Exhibits 103 and 104 can only be significant if the data are correct. In addition to the errors above noted, defendants presented the testimony of Chief Julian L. Madison, who said he could not find in Plaintiffs’ Exhibit No. 103, upon which Dr. Gittleson depended, any instance where a fireman was suspended for 60 days for a garnishment or any instance where one lost 3 vacation days for a garnishment as claimed by plaintiffs. The first exception refers to an entry for Robert Myers on page 8 of Exhibit 103. The entry “Lost 60 days of work” appears immediately below the entry “Counselled 6/73 for Garnishment.” According to Chief Madison, counselling is not a disciplinary action in the Fire Bureau, but only an admonition intended to prevent the necessity for future disciplinary action. Record, page 82. It is inconsistent that Myers would be suspended for 60 days for a garnishment and at the same time be noted as “counselled” for the same offense. The likely explanation is that the two entries are separate, and plaintiffs erred in confusing them in Exhibit 104B. The entry indicating a loss of 3 vacation days for a garnishment appears on page 53 of Exhibit 103 under the name John Raspberry, Jr. The entry “3 vac days 12/2, 3, 4, 1975” appears immediately over the entry “garnish. 7/2/75.” It is difficult to believe that the Fire Bureau would allow 5 months to pass before taking disciplinary action. In fact, Chief Madison testified that Raspberry lost 3 vacation days for failing to mark off duty properly and produce a doctor’s certificate saying he was sick. It appears that in this case also, Exhibit 104B is in error. Plaintiffs failed to explain any of these errors. In their post-trial brief, plaintiffs say, “In an attempt to point out inaccuracies, defendants called Acting Chief Madison as a witness. The witness failed to point out any inaccuracies.” Plaintiffs’ Post-Trial Brief, page 17. As above noted, this claim is patently incorrect. Had the errors not been discovered plaintiffs’ exhibit would have led the Court to believe that the punishment given blacks for garnishments was far more severe than the punishment given whites. These errors are serious and their existence makes the Court uneasy in placing firm reliance on the statistical evidence. According to Chief Finnegan, after a fireman receives one garnishment he is counselled on his financial affairs. Subsequent garnishments result in written reprimands which become a part of a fireman’s records. No disciplinary action beyond a reprimand is used. Record, pages 432-38. Chief Finnegan also testified that he does not consider one or two garnishments in determining who should be promoted. Record, page 483. Except as mentioned at page 444 of the Record, noted infra, Chief Finnegan expressed the view that the Fire Bureau should take no disciplinary action with respect to garnishments. Record, page 442. Finnegan testified that when a fireman’s wages are garnished it places an administrative burden on the Fire Bureau. He also testified that he would prefer a rule under which three garnishments within one year would result in automatic termination for a fireman. Chief Finnegan felt this would protect the image and morale of the Fire Bureau. Record, page 444. It appears from the testimony that no black has been penalized in pay or promotion because of a garnishment. No evidence of such a penalty being imposed has been educed by plaintiffs. Nor is there evidence that the Fire Bureau uses its disciplinary policy to discriminate against blacks when it issues a reprimand to a fireman after his second garnishment. The Fire Bureau has no control over the private financial affairs of its employees, nor any control over the financial affairs of its employees’ creditors. The inconvenience and bad community relations created by firemen who fail to pay their debts and have their wages garnished is ample justification for the mild disciplinary measures taken by the Fire Bureau. Further, failure to pay one’s debts is classic evidence of irresponsibility, and the Court is of the opinion that Chief Finnegan is entitled to consider it and give it whatever weight he deems appropriate in determining who should be promoted; provided, however, that his doing so is not a pretext for racial discrimination. In the absence of evidence that Chief Finnegan’s consideration of excessive garnishments (Record, page 484) in the promotion procedure is a mere pretext for racial discrimination, the Court cannot find any violation of Title VII in the Fire Bureau’s manner of handling employee garnishments. B. Traffic Accidents When a fireman is involved in a traffic accident while on duty and operating City equipment, the Fire Bureau’s Accident Review Committee investigates the accident and determines whether the fireman involved should be held responsible. Plaintiffs contend that black firemen are found chargeable for traffic accidents at a higher rate than white firemen, and that this is proof of racial discrimination. The statistics which plaintiffs believe establish racial discrimination are found in Plaintiffs’ Exhibit No. 104A. This exhibit shows that of 22 vehicular accidents involving black firefighters since 1972, only 5, or 24% were found to be non-chargeable. Of 76 vehicular accidents involving white firemen 48, or 63% were found to be nonchargeable. The rate of favorable action for blacks was only 38% the rate for whites. Dr. Gittleson testified at trial that this difference was highly significant in statistical terms. Record, page 222. Though the percentage difference is substantial the actual numbers involved are very small. A shift in only nine cases involving blacks over a five year period from chargeable to non-chargeable would have resulted in identical percentages. The Court cannot base a finding of racial discrimination on such sparse data. Furthermore, analysis of Plaintiffs’ Exhibit No. 103 reveals that only 16 of the accidents involving black firemen occurred within the period covered by this lawsuit. It may be expected that where there are a large number of traffic accidents compiled over a long period of time, that black and white firemen will be expected to have a similar number of accidents for which they can be held responsible. Under those circumstances, statistics that show disparate treatment for blacks would be proof of discrimination. Here, however, the sample size is simply too small. Cf. Mayor v. Educational Equality League, 415 U.S. 605, 94 S.Ct. 1323, 39 L.Ed.2d 630 (1974). Even if these statistics supported a finding of adverse impact, defendants have met the burden such a finding would have imposed upon them to show a compelling business necessity that requires that this procedure be followed. Robinson v. Lorillard Corp., 444 F.2d 791 (4th Cir. 1971), cert. dis., 404 U.S. 1006, 92 S.Ct. 573, 30 L.Ed.2d 655 (1971). It is clearly essential to the operation of a fire department that firemen whose negligence or incompetence cause accidents with fire equipment be singled out and disciplined. The Accident Review Committee system defendants have established is a reasonable means to this end. Plaintiffs have not even attempted to show that any other system would serve the same purpose but eliminate any possible adverse impact. See, McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). Therefore, the Court finds no violation of Title VII in the defendants’ system of disciplining employees involved in traffic accidents. II Plaintiffs contend that the testing procedures used by defendants to determine who shall be hired and promoted are in violation of Title VII. Under Title VII no matter how well intentioned defendants might be, if the result of their acts or omissions in hiring, firing, promotion and the like has an adverse impact upon a protected class, then the burden is placed upon defendants, by law, to show that the adverse impact was a necessary but unintended result of job-related requirements. Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971). Thus, under Title VII we are not limited to the area of “fault liability.” Instead, liability may be found for wholly unintended and undesired results of racially neutral acts or omissions. Congress directed the Act to the consequences of racial discrimination not merely the intent to discriminate on the basis of race. Griggs, supra. However, the burden on the defendant of showing that a particular test or requirement is job related arises only after the plaintiff has made out a prima facie case of discrimination. A. Entry-Level Testing Plaintiffs allege that the entry-level test used by the Richmond Fire Bureau discriminates against black applicants. The most telling evidence of this discrimination would be a showing that the percentage of blacks employed by the Fire Bureau was significantly smaller than the percentage of blacks in the population from which applicants are drawn. In Harper v. Mayor and City Council of Baltimore, 359 F.Supp. 1187 (D.Md.1973), aff’d, 486 F.2d 1134 (4th Cir. 1973) for example, it was shown that while 47% of the residents of the City of Balti-more were black, and 24% of those residing within 30 miles of the City of Baltimore were black, only 14% of the firemen employed by the City of Baltimore were black. A similarly dramatic discrepancy in the percentage of black firemen in Richmond compared to the percentage of blacks in the population of the Richmond Standard Metropolitan Statistical Area would be a strong indication that the recruiting and testing procedures used by the City of Richmond tended to exclude blacks. There is no such showing. On the contrary, the statistics show that the proportion of blacks in the Richmond Fire Bureau is very close to the proportion of blacks in the Richmond Standard Metropolitan Statistical Area. Chief Finnegan testified at trial that the Richmond Fire Bureau has 549 employees, of whom 136 are black. Record, page 336. Thus, approximately 24.8% of the employees of the Fire Bureau are black. Chief Finnegan further testified that the employees of the Richmond Fire Bureau reside throughout the Richmond metropolitan area, with 24.75% living in the City of Richmond, 3.5% living in Goochland County, 34.28% living in Henrico County, 17.4% living in Hanover County, and 15.45% living in Chesterfield County. Record, pages 334-335. The City of Richmond, Virginia Standard Metropolitan Statistical Area includes all the above counties plus Charles City County and Powhatan County. Only a very small percentage of the employees of the Richmond Fire Bureau are from these latter two counties. According to the 1970 census, 26.1% of the population of the Richmond Standard Metropolitan Statistical Area is black. Thus, the proportion of blacks in the Richmond Fire Bureau is only 1.3% less than the proportion of blacks in the population from which applicants for positions in the Fire Bureau are drawn. However, plaintiffs base their argument not on a statistical discrepancy between the proportion of blacks in the population and the proportion of blacks in the general population, but upon the results of the administration of the Fire Bureau’s entry-level test on 1 November 1975. 145 blacks and 53 whites took the Fire Bureau’s entry-level test on 1 November 1975. Of these, 69 blacks and 38 whites passed and were certified as eligible for hiring. Of those eligible to be hired, 27 blacks and 17 whites were hired. Thus 18.6% of the black applicants and 32.1% of the white applicants were hired. The Federal Executive Agency Guidelines on Employee Selection Procedures, as adopted by the Department of Justice, indicate that a selection rate for any group that is less than 80% of the rate for the group with the highest rate will generally be regarded as evidence of adverse impact. 41 Fed.Reg. 51734 (1976) (to be codified in 28 C.F.R. § 50.14). The selection rate 'for black applicants in the most recent selection procedure was only 57.9% of the rate for white applicants. Even if'the 80% guideline is not taken as a hard and fast rule, but only as a rule of thumb, this is clearly an adverse impact upon the black applicants for employment. Furthermore, it is apparent that this adverse impact can be attributed to the entry-level examination itself. Of those blacks who passed the test, 39.1% were hired, compared with 44.7% of the whites. The selection rate for blacks was thus 87.5% of the selection rate for whites, substantially above the rule of thumb for determining adverse impact. The test used by the City of Richmond was the Firefighters B-l(m) test, developed by the International Personnel Management Association. The City added to Bl(m) a physical condition test. Record, page 104. The latter test tended to increase the percentage of blacks qualifying. Nelson L. Sutton, III, senior personnel technician of the City of Richmond testified that the B-l(m) test had been validated, or shown to be a job-related test, in 55 jurisdictions in California. Record, pages 107-110. Sutton further testified that the validation study in California met the E.E.O.C. guidelines for such studies. Record, page 107. In addition, Sutton testified that a job-analysis study indicated .that the job of a firefighter in the jurisdictions in which this validation study had been conducted was the same as the job of a firefighter in Richmond. Record, page 109. This testimony was not contradicted by plaintiffs, and the Court accepts it as true. Plaintiffs, however, vigorously object to the use of a validation study conducted elsewhere to validate a test used in Richmond. Plaintiffs contend that the E.E.O.C. Guidelines on Employee Selection Procedures, 29 C.F.R. Part 1607, dó not permit an employer to rely upon vicarious validation where a validity study is feasible. 29 C.F.R. § 1607.7. Defendants admit this contention but argue that the Federal Executive Agency Guidelines on Employee Selection Procedures, as adopted by the Department of Justice, should apply to this case. The Department of Justice guidelines are the product of the deliberations of the Equal Employment Opportunity Coordinating Council, established by Congress in 1972 and given a mandate to eliminate inconsistency among the agencies of the Federal Government responsible for enforcing equal employment opportunity 42 U.S.C. § 2000e-14. While a majority of the Council endorsed the new guidelines, the E.E.O.C. has chosen to stand firm with its old guidelines. Defendants argue that because the Department of Justice would have the responsibility pursuant to 42 U.S.C. § 2000e-5(f)(1) to bring a civil action against the City of Richmond for violations of Title VII, and because the Department of Justice has accepted the new guidelines, that the Department of Justice guidelines, not the E.E.O.C. guidelines, are the appropriate standard in this case. The Department of Justice guidelines permit the use of validation studies conducted in other jurisdictions. The Court is of the opinion that whatever effect these various guidelines have upon the agencies promulgating them, they are not legally binding upon either the parties to this suit or the Court. The Court can and will use both sets of guidelines as an aid to determining if either party has carried its burden of proof under Title VII, but although these guidelines are of great value as the interpretation of the law by government agencies charged with the duty of enforcing the law, the Court will not be bound by one set of guidelines over another, but only by what the Court deems as reasonable in this ease. There is evidence from the defendants that the Department of Justice standards are more modern and reflect the most recent advances in psychometrics and the latest thinking of the experts in the field of employment testing, particularly the American Psychological Association. Record, page 129. This evidence is not disputed, and the Court finds it very significant. In any case, the Court finds it credible, as Mr. Sutton testified, that the job of a fireman in Richmond is very similar to the job of a fireman in certain California cities. Given this, the Court concludes that the City of Richmond carried its burden of showing the Firefighters B-l(m) test to be job-related by proof that the test had been validated as job-related according to E.E. O.C. guidelines in a validation study in California, and that proof that the job for which the test was validated was substantially the same as the job for which the test was being used in Richmond. But the City did not rely exclusively upon the validation study conducted in California to prove that the Firefighters B-l(m) test was job-related. Mr. Sutton testified at trial that he conducted a validation study in Richmond, using both black and white Richmond firemen. Record, pages 112-122. Sutton further testified that the validation study he conducted showed the Firefighters B-l(m) test to be valid and job-related under the E.E.O.C. guidelines. Record, page 180. Sutton testified that the study he conducted was a concurrent criterion-related validation study. A sample of 38 white and 36 black firefighters was given the Firefighters B-l(m) test. At the same time, each firefighter was rated by his immediate supervisor on twelve characteristics found to be important to a firefighter’s tasks. Plaintiff’s Exhibit No. 48. Sutton found that the relationship between test performance and rating performance was sufficient to show that the test was job-related. Record, page 114. Plaintiffs submitted the deposition testimony of Dr. Richard S. Barrett, a recognized authority on employment testing, in support of their contention that the City’s validation study did not meet the professional standards for such studies. Plaintiff’s Exhibit No. 116. Dr. Barrett testified that the Richmond study was inadequate in that it used only one supervisory rating for each man, that no specific training was held for the raters, and that the statistics developed by the study did not permit a determination that the test was free of unfair adverse impact. Plaintiff’s Exhibit No. 116, pages 18-22. Even if Dr. Barrett’s criticisms are true, both the Department of Justice and the E.E.O.C. guidelines provide for the use of a test which is not at the moment fully supported by the required evidence of validity, if there is substantial evidence of validity and if there is in progress a study which is designed to produce the additional data needed to determine validity within a reasonable time. 141 Fed.Reg. 51737 (1976) (to be codified in 28 C.F.R. § 50.14), 29 C.F.R. § 1607.9. Mr. Sutton testified that a predictive validity study would be done as soon as a sufficient sample of firemen who had taken the test and had been hired had been gathered. He estimated that this would occur within a year of the trial date, 18 April 1977, or at most two years. Record, page 122. Under these circumstances, use of the Firefighters B-l(m) test is justified under the E.E.O.C. guidelines, and so falls within the Congressional grant of immunity for any act done “in good faith, in conformity with, and reliance on any written interpretation or opinion of the Commission . . . ” 42 U.S.C. § 2000e-12(b). As noted, the City of Richmond changed its entry level test battery to include a test of physical agility. This test battery was administered in December of 1976 and January of 1977, according to Sutton. Record, page 124. Sutton indicated that 81 blacks and 55 whites took this new test. No statistics on how many of these applicants were finally hired were presented to the Court. Sutton, however, indicated that the black applicants averaged 53.56 on the test, while the white applicants averaged 57.96. Record, page 128. Although these averages are reasonably close to one another, it is impossible for the Court to tell if there was any adverse impact on black applicants from the most recent administration of the Fire Bureau’s new entry-level test battery. Finally, the plaintiffs have failed to show how they, or any member of the plaintiff class, has been aggrieved by the Fire Bureau’s entry-level selection procedures. All members of the plaintiff class either are presently employed by the Fire Bureau or formerly have been employed by the Fire Bureau. Thus, all members of the plaintiff class have successfully passed over the hurdle of the Fire Bureau’s entry-level selection procedures. If anyone has been discriminated against, (and the Court finds that none have), it has not been the members of the plaintiff class. Therefore, no liability to plaintiffs can be found on the basis of the entry-level test. B. Promotional Testing Plaintiffs contend that the promotional tests used by the Richmond Fire Bureau violate Title VII in that they discriminate against black employees because of their race. Plaintiffs point to the fact that, while 24.8% of the employees of the Richmond Fire Bureau are black, only 8.1% of the fire lieutenants and 2.9% of the fire captains are black, and no black employee of the Richmond Fire Bureau has ever risen above the rank of fire captain. The only explanation of this disparity offered by defendants was that of Chief Finnegan who testified at trial that blacks have never scored high enough on the tests to be certified as eligible for promotion in sufficient numbers. Record, page 340. There can be little doubt that this is attributable to the Fire Bureau’s past history of racial discrimination. To the extent that such discrimination occurred before 11 May 1974, however, it cannot be the basis of liability in this case. The Supreme Court recently said: Racial discrimination by public employers was not made illegal under Title VII until March 24, 1972. A public employer who from that date forward made all its employment decisions in a wholly nondiscriminatory way would not violate Title VII even if it had formerly maintained an all-white work force by purposefully excluding Negroes. [Hazelwood School District v. United States, 433 U.S. 299, 308, 97 S.Ct. 2736, 2742, 53 L.Ed.2d 768 (1977).] The requirement that a charge be filed with the E.E.O.C. within 180 days of the alleged discriminatory act also serves to limit an employer’s liability: Respondent is correct in pointing out that the seniority system gives present effect to a past act of discrimination. But United was entitled to treat that past act as lawful after respondent failed to file a charge of discrimination within the 90 days then allowed by § 706(d). A discriminatory act which is not made the basis for a timely charge is the legal equivalent of a discriminatory act which occurred before the statute was passed. [United Air Lines, Inc. v. Evans, 431 U.S. 553, 558, 97 S.Ct. 1885, 1889, 52 L.Ed.2d 571, 578 (1977).] In any event, the evidence presented by plaintiffs is insufficient to prove that the promotional test given prior to the effective date of the Act discriminated unfairly against blacks. No more than 12 blacks took any promotional test given by the Fire Bureau prior to the effective date of the Act. Plaintiffs’ Exhibits Nos. 15-29. The total number of blacks taking any promotional test in the Fire Bureau during the ten-year period preceding 1 November 1975 is at most 24. Record, page 210. These numbers are simply too small to be the basis of any finding of statistical adverse impact. There was much dispute at trial as to whether it would be meaningful to combine these samples for an aggregate analysis or whether the number 30 was a “magic number” for sample size in statistical analysis. Dr. Barrett, the plaintiffs’ expert, testified that a sample of 30 persons was a “rule of thumb” in statistical testing. Plaintiffs’ Exhibit No. 116, page 57. Plaintiffs’ other expert, Dr. Gittleson, testified that while some- scientists perforce must work with small samples, that the larger the sample the greater the reliability of the findings. Record, pages 230, 261. Based on this testimony, the Court finds that the sample size of blacks taking promotional examinations in the Richmond Fire Bureau prior to 1975 is too small for a statistical analysis sufficiently reliable to prove adverse impact. Since 11 May 1974, the relevant date to this action, the Fire Bureau has administered only one set of promotional tests. These tests were administered on 1 November 1975. The results of these tests are found in Plaintiff’s Exhibit No. 58. No blacks took the test for Battalion Chief. Only one black took each of the tests for Deputy Battalion Chief and Fire Captain. Neither of these two individuals was certified eligible for promotion. However, the number of black employees taking these tests is too small to establish that there was an adverse impact upon blacks taking the test. In the absence of any proof of bad faith on the part of the Fire Bureau in administering or grading the tests, there is no sufficient evidence to shift the burden of proof to defendants to show that these tests for upper level positions were job-related. See Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977), fn. 20. The promotional test for fire lieutenant was given to 67 black firemen and 209 white firemen. Of these, 3 blacks and 11 whites were promoted. Thus, 4.5% of the blacks and 5.3% of the whites were promoted. The selection rate for blacks was 84.9% of the selection rate for whites, well over the 80% rule of thumb for adverse impact suggested by the Department of Justice guidelines. Plaintiffs contend, however, that the fire lieutenant test discriminated against blacks in that of the 67 black firemen tested, only 14, or 20.9% were found eligible to go through the Assessment Center procedure following a written examination, while 75, or 35.9% of the whites passed the written examination. Thus, on the written portion of the fire lieutenant test, the pass rate for blacks was only 58.2% of the pass rate for whites, substantially less than the 80% rule of thumb for adverse impact. The Court is of the opinion that it is the entire selection procedure, not any given segment of it, that must be examined for adverse impact under Title VII. The fact that any stage in a selection procedure has an apparent adverse impact upon blacks could be nullified by a corrective procedure which would have an apparent adverse impact upon whites, so that the final result would show no racial bias. The fire lieutenant promotion procedure followed by the Fire Bureau in 1975 follows this pattern. Plaintiffs’ Exhibit No. 114 shows that in 1975 there were 347 white and 134 black firefighters in the Richmond Fire Bureau. Of these, 67, or 50% of the blacks and 209, or 60.2% of the whites took the fire lieutenant test. The rate for the blacks was 83.1% of the rate for the whites, a figure above the 80% rule of thumb for adverse impact. Apparently, whatever standards existed, if any, as to who could take the test did not bar blacks disproportionately from taking the test. The test consisted of a preliminary written test and, for those who successfully completed the written portion, an oral Assessment Center procedure. The assessors certified 19 firefighters to the- Fire Chief as eligible for promotion, and from that list the Chief chose 14 to be fire lieutenants. Of the 14 blacks and 75 whites who passed the written examination, 3 blacks or 21.4% and 16 whites, or 21.3% were certified as eligible for promotion after passing through the Assessment Center procedure. This is substantially the same rate for each race, and the Assessment Center procedure showed no adverse impact upon blacks whatsoever. Record, page 293. Of the 3 blacks and 16 whites certified as eligible for promotion by the assessors, Chief Finnegan promoted all 3 or 100% of the blacks, and 11, or 68.8% of the whites, to the rank of fire lieutenant. This is substantially less than the 80% rule of thumb for adverse impact, but is irrelevant to this case inasmuch as there are no whites in the plaintiff class. It is clear, therefore, that only one stage in the promotion procedure for fire lieutenant, the written test, showed an ádverse impact on black candidates. On similar facts, the Sixth Circuit held that the plaintiff in a Title VII case failed to demonstrate a prima facie case, despite clear evidence that the defendants’ written test had a disproportionate adverse impact on blacks where the total hiring process of the defendant showed no adverse impact. Smith v. Troyan, 520 F.2d 492 (6th Cir. 1975), cert. den., 426 U.S. 934, 96 S.Ct. 2646, 49 L.Ed.2d 385 (1976). In Smith, the City of East Cleveland, Ohio, used the Army General Classification Test as part of its hiring process for police officers. The statistics as set out in the district court opinion (Smith v. City of East Cleveland, 363 F.Supp. 1131, 1145 (N.D. Ohio, 1973), showed that in 1973 the average score for whites on the test was 106.4, while black applicants scored an average of 83.2. The same year, 22% of the blacks who took the test received a score of over 100, while 71% of the whites scored over 100. Nevertheless, the City hired blacks in approximately the same proportions as it had black applicants. Blacks comprised 33% of all applicants and 29% of all persons hired. The Court of Appeals held that, under the circumstances, the burden of showing the test to be job-related did not shift to the defendants: That blacks fared less well than whites on the AGCT, a “subtest” in the process of hiring East Cleveland police officers, is insufficient in itself to require defendants to justify the AGCT as being job-related. Carried to its logical extreme, such a criterion would require the elimination of individual questions marked by poorer performance by a* racial group, on the ground that such a question was a “sub-test” of the “subtest.” [520 F.2d at 498.] The Court finds that the fire lieutenant promotional procedure taken as a whole does not show an adverse impact on blacks, and that the preliminary written test given 1 November 1975 is merely a “subtest” in the promotion scheme. As such, the Court concludes that a showing that the written subtest had an adverse impact on black candidates is not sufficient to shift the burden of proof to defendants to show the test to be job-related. Even if a showing of adverse impact for the written subtest were sufficient to shift the burden to defendants to show the test to be job-related, the Court is of the opinion that defendants carried the burden. The defendants presented the testimony of Dr. Robert J. Filer, president of Psychometric Consultants, Inc. Dr. Filer’s firm was employed by the City of Richmond to develop valid and job-related tests and procedures for selecting personnel for promotion in the Fire Bureau. Dr. Filer testified at trial that the promotional tests and procedures he developed for the City of Richmond, including the written subtest for fire lieutenant, were content valid according to the standards of the American Psychological Association and the E.E.O.C. Record, pages 279-289. Defendants have also introduced into evidence Dr. Filer’s extensive job analysis reports and recommendations. Defendants’ Exhibit No. 12. Dr. Filer’s expert testimony that the written subtest for fire lieutenant was valid and job-related according to E.E.O.C. guidelines for content validity stands unchallenged. Plaintiffs’ post-trial brief contains the astounding statement: “There was absolutely no effort made by the defendants to demonstrate that any promotional test was job-related.” Plaintiffs’ post-trial brief, page 20. Plaintiffs attacked Dr. Filer’s Assessment Center procedure with great vigor, but that procedure did not result in any adverse impact at all. Plaintiffs did not advance any alternative methods of selection which would serve the City as well and yet eliminate any adverse impact. On the basis of Dr. Filer’s testimony and the exhibits introduced by defendants, the Court finds that the City of Richmond complied with the E.E.O.C. guidelines in testing the validity and job-relatedness of the written subtest for fire lieutenant. The, Court, therefore, concludes that the use of that test did not violate Title VII. Ill Plaintiffs contend that the Fire. Bureau discriminates against blacks in awarding fitness ratings. These ratings are given by each fireman’s immediate supervisor, and are supposed to represent the supervisor’s estimate of the fireman’s performance over the period rated. The ratings available to the supervisor are “unacceptable,” “acceptable,” “more than acceptable,” and “outstanding.” Plaintiffs’ Exhibit No. 105D shows the following distribution of fitness ratings, drawn from Fire Bureau files since 1972: More Than Unacceptable Acceptable Acceptable Outstanding Blacks Less than 1% 56% 43% Less than 1% Whites 0% 25% 72% 3% If receiving a “more than acceptable” or “outstanding” rating is considered a favorable employment decision, the proportion of blacks receiving a favorable decision is only 58.7% of the proportion of whites receiving such a decision. This is substantially less than the 80% rule of thumb for adverse impact. Fitness ratings, however, are not the same as an employment or promotional test. To compare the ratings of a fireman who has seen ten or twenty years’ experience with the ratings of a rookie fireman is like comparing neophyte lawyers to seasoned veterans of the law. This is especially true in the light of General Order # 80, propounded by Chief Finnegan on 22 August 1973. In General Order # 80, Chief Finnegan says, “All employees are to be compared to the standards set for the position. People should not be compared to people, but, rather people to the job standards.” A rookie fireman may be “outstanding” compared to other rookies, but merely “acceptable” when measured against the standards established for the position. Similarly, the veteran may be average among firemen of his experience but “outstanding” measured against his job-standard. Plaintiffs’ Exhibit No. 112A shows the fitness ratings for white and black firemen broken down by seniority in five-year brackets. Before the period 1961-65 there were insufficient blacks hired by the Fire Bureau to constitute a fair sample. Therefore, these years have been eliminated from the following table: It is clear from these tables that seniority plays an important role in the distribution of fitness ratings. An analysis of these tables shows that blacks received ratings of “more than acceptable” or higher at the following proportions to the comparable white ratings: However, these tables reflect fitness ratings going back to 1972. Plaintiffs’ Exhibit No. 112. Only ratings given after 11 May 1974 may be used to find liability in this case. Accordingly, the Court has abstracted from Plaintiffs’ Exhibit No. 112 all those ratings given in 1975 and after. The following tables are the result of that analysis: White Firefighters and Officers Year Hired 61-65 66-70 71-76 Acceptable 15.2% 16.1% 59.6% More Than Acceptable 77.3% 81.1% 39.9% Outstanding 7.6% 2.7% 0.5% Black Firefighters and Officers Year Hired 61-65 66-70 Acceptable 40.0% 20.7% More Than Acceptable 60.0% 78.3% Outstanding 0.0% Once again comparing the black to the white rates of receiving “more than acceptable” or higher ratings, the Court arrived at the following proportions: Year Hired Rate 61-65 70.7% 66-70 94.5% 71-77 50.2% The analysis of these statistics requires that certain facts be kept in mind. First, the statistics for ratings of blacks hired between 1961 and 1965 are based on only 30 ratings during the period 1975 to 1977. Second, by far the largest number of ratings of blacks was in the category of firemen hired between 1971 and 1977. The overwhelming majority of these men were newly-hired employees when they were rated in 1975, 1976 and 1977. Analysis of Plaintiffs’ Exhibit No. 64, a computer listing of Fire Bureau personnel for the years 1971 to 1976, shows that of 72 white firefighters hired between 1971 and 1976, 43, or 59.7%, were hired in 1974,1975 or 1976. By contrast, of 66 black firemen hired between 1971 and 1976, 58, or 87.9%, were hired in 1974,1975 or 1976. Thus, almost 90% of the black firemen hired after 1970 were rated for their first time during the period covered by this lawsuit, while only 60% of the white firemen were rated for their first time during the same period. This would easily explain the disparity between white and black ratings in the group hired after 1971, since, as has been seen, experience is an important factor in determining a fireman’s rating. This leaves only the ratings received by firemen hired between 1965 and 1970 from which the Court may get a picture of how the Fire Bureau rates experienced blacks against whites of comparable experience. In the years 1975, 1976 and 1977, blacks hired by the Fire Bureau between 1965 and 1970 received “more than acceptable” or better ratings 79.2% of the time. In the same years, whites hired by the Fire Bureau between 1965 and 1970 received “more than acceptable” or better ratings 83.8% of the time. The rate for blacks receiving “more than acceptable” or better ratings was 94.5% of the rate for whites receiving the same rating. Thus, there was no adverse impact shown on blacks with respect to their fitness rating for the only group for which we have reliable comparable statistics. If the ratings for firemen hired between 1961 and 1965 are combined with those of the firemen hired between 1966 and 1970, the comparable ratio for the larger sample is 89.1%, slightly smaller than 94.5% but still well above the 80% rule of thumb for showing adverse impact. Even if the statistical evidence had conclusively shown an adverse impact on black firemen from the fitness ratings, the evidence shows that these ratings are considered in the promotion process only after a fireman has passed the written examination and the assessment center procedure and been certified as eligible for promotion. At that point, Fire Bureau Chief Finnegan reviews the records, including the ratings, of the fireman and determines who shall be promoted. There is no evidence of adverse impact on black firemen at that stage in the promotion procedure. Indeed, the statistics show adverse impact on white candidates at that stage. Even if there were adverse impact, it is difficult to imagine a more job related evaluation than the performance rating of an employee by his immediate supervisor on the job. Plaintiffs have failed to point out any alternative rating system which would serve the same purpose and eliminate the alleged, though non-existent, adverse impact. In view of these findings, and the lack of credible evidence of bad faith on the part of defendants, the Court finds that the supervisory ratings used by the fire department do not constitute a violation of Title VII. IV Plaintiffs seek to prove their allegations of racial discrimination by means of affidavits filed by leave of Court. The Court has reviewed the affidavits with care and has divided them into two categories. The first category consists of those affidavits which relate acts of racial discrimination that took place before 11 May 1974. The facts set out in these affidavits are not actionable, but they are read, insofar as they are otherwise admissible, to shed light on the intent of defendants in later similar or related circumstances. Their weight and value is commensurate with the weight and value given to a past history of an official policy of racial discrimination. This past history must always be read in the light of changing social and legal values. The Court has no writ to punish for past sins; its authority permits only remedies for violations of law within the applicable period of limitations. The Court finds on the basis of the first category of affidavits that incidents evidencing racial bigotry and discrimination among firemen and officers occurred prior to 11 May 1974. Such discrimination was not always officially sanctioned, but among the more egregious incidents was the refusal of the former Fire Chief to permit a thoroughly qualified officer, Captain Arthur L. Page, to serve as an acting battalion chief and “drive the red car.” Captain Page is a Negro and the humiliating refusal to permit him to serve as an acting battalion chief was based solely on his race. After the present Fire Chief, defendant Finnegan, assumed control of the Fire Bureau in July 1972, he discontinued this gross abuse. The second category of affidavits consists of those which relate events that occurred after 11 May 1974. Much of what is contained in these affidavits is of little value in determining facts. It is conclusory in nature and based on hearsay. By agreement of counsel all affiants were present in Court arid could have been called by either party. Plaintiffs rested on the affidavits and, with two exceptions, defendants chose not to cross examine the affiants. Defendants objected to the admissibility of these affidavits on the ground of hearsay and their objection was well founded as to large portions of many of these affidavits. Order on Pre-Trial Conference, page 16. Instead of cross examining plaintiffs’ witnesses, defendants chose to testify themselves in refutation of the matters raised in the affidavits. Defendants Finnegan and Leidinger testified as to the racial policies and practices of the Fire Bureau and of the City of Richmond. Nineteen affidavits filed by plaintiffs described events occurring within the actionable period. The Court has reviewed each one carefully and analyzed the allegations of each one individually. Affiant Chuck L. Davis alleges that he received very poor efficiency ratings while whites that did not perform as well as he did in training received much higher ratings. Davis does not say how he knows the whites got better ratings. Nor does he give the Court any facts to support his assertion that those who received better ratings had performed no better than he had. Davis further states that he was rated by a lieutenant who had had no supervisory experience over him and so had no basis for a rating. He does not indicate that in this he was discriminated against on the basis of race, or that white firemen were always rated by lieutenants who had had supervisory experience over them. If Davis were rated by someone who knew nothing of his performance, he would surely be aggrieved, especially if he received a lower rating than was just under the circumstances. But that is not a violation of Title VII. Finally, Davis alleges that a white fireman altered an official record to transfer Davis and other blacks to temporary duty while white firemen received “credit” for the transfers. Davis further alleges that when he complained of this practice, he was transferred to another firehouse and the whites responsible “were neither reprimanded or transferred.” Affidavit of Chuck L. Davis, page 2. Defendant Finnegan testified that the white firemen responsible for this incident were punished by a reduction in pay, one for a year and two others for 30 days. Record, pages 354-356. Davis was unable to explain why this information was omitted from his affidavit when testifying at trial. Record, page 616. This omission of an important fact casts a pall on Davis’ affidavit, despite his denial that he was intentionally trying to mislead, the Court. The fact that Chief Finnegan had not disciplined these three white firemen for their conduct would have been strong evidence of his acquiescence in discriminatory practices by his subordinates. However, the evidence shows that Finnegan disciplined these men severely. The one-year reduction in pay was, by Finnegan’s estimate, the equivalent of a $1,000 fine. Record, page 356. The other two white firemen involved, who knew of the incident and failed to report it were both reduced in pay for 30 days. Thus, the incident described by Davis, instead of tending to show discrimination on the part of the fire bureau, tends to show that the fire chief was sensitive to discrimination by his subordinates and was willing to act effectively when discrimination was shown to exist. Affiant Theodore R. Fuller alleges that after he was assigned to the fire prevention division, he was harassed by whites who put grease on his automobile door and on the garage door. Although Fuller does not say so, the Court infers that the unidentified whites who harassed Fuller are employees of the Fire Bureau. Fuller states that his complaint to “the Captain” was ignored. Fuller does not say whether or not he pursued his complaint up the chain of command to defendant Finnegan. Nor does he say that similar complaints from white firemen were not ignored. Fuller further states that he placed nineteenth on the written portion of the fire lieutenant test, then dropped to forty-ninth on the list after going through the Assessment Center. Although he clearly feels that this was because of his race, this is a mere conclusion. Fuller does not indicate whether or not any white firemen suffered similar drops in standing. Finally, Fuller alleges that after his transfer to the fire prevention division, he received all “acceptable” ratings while whites transferred at or near the same time received mostly “more than acceptable” ratings. Fuller does not allege that the whites who received the better ratings performed no better than he performed. Nor does he state how he discovered the confidential ratings of the other firemen. Nor does he allege that he deserved a higher rating than the one he received. Affiant Steve I. Foreman alleges that black fire inspectors are assigned to black areas, and that the Fire Bureau makes assignments to beds and lockers on the basis of race. He recites no specific instances of any of these abuses. These conclusions on the part of the affiant are of no use to the Court in finding facts. In his testimony at trial, Chief Finnegan flatly denied assigning fire inspectors to any area of the city on the basis of race. Finnegan testified that Inspector Foreman, a black inspector, was the “roving inspector,” who covered the whole city. Finnegan also testified that the Fire Bureau had black inspectors assigned to white neighborhoods, and white inspectors assigned to black neighborhoods. Record, pages 361-63. The Court finds on the basis of this testimony that black fire inspectors are not assigned to black areas of the city on the basis of race. Foreman’s statement with regard to bed assignments in the Fire Bureau are considered with those of affiant Wilder. Affiant William R. Talley alleges that he placed seventh on the written portion of the fire lieutenant test given in 1975, but dropped to fifty-fifth on the eligibility list after the Assessment Center. He clearly feels that this was because of his race, but this is only a conclusion on his part. Talley states that the assessors in the Assessment Center were not qualified to be assessors because they knew nothing of the technical, managerial or supervisory aspects of firefighting. Dr. Filer, the defendants’ expert, testified at trial that the assessors were trained members of the staff of the City Personnel Department. Record, page 308. This conflict bears upon the validity of the Assessment Center procedure and does not involve any acts of racial discrimination by the fire bureau. If the assessors were poorly selected their disability was not shown to be selectively inflicted on any one race. Talley further states that the material given at the Assessment Center was incomplete, and that others had a complete test and he did not. He does not state whether or not he complained of this irregularity at any time. It is possible that Talley is referring to the fact that nine questions on the written portion of the fire lieutenant’s exam were not graded, due to a breach in the security of the test. See Plaintiff’s Exhibit No. 54. This was an unfortunate incident, but it is not evidence of discrimination. Affiant Leroy Henderson alleges that whites who were less qualified than he and who placed lower on the certification list were hired before he was hired. Although Henderson clearly realizes that the fact that he had been placed on probation for assault played some part in the delay, he feels the true reason for the Fire Bureau’s delay in hiring him was because of his race. He does not say whether or not less qualified blacks were also hired between March, 1975 and May, 1975 when Henderson was hired. The Court finds that even if Henderson’s testimony were not conclusory and based on hearsay, a two month delay in hiring was not unwarranted where the applicant had a criminal record. Affiant Cornelius Hood, Jr. alleges that Acting Lieutenant Wheller put a two-year fireman in charge of the company over Hood with 11 years experience. Although Hood does not say so, it may be inferred that both Wheller and the two-year fireman are white. Affiant Andrew L. Irving alleged that Lieutenant Seiz, a white male, allowed Irving to act as lieutenant only three times in two years. There is no evidence to show how often the average fireman qualified to act as lieutenant is given that opportunity, although Irving characterized his opportunities as coming “very seldom.” Irving clearly feels that his lack of opportunity to act as lieutenant was because of his race. In response to Hood’s and Irving’s allegations Chief Finnegan testified that each company has an approved list of firefighters that may act as lieutenant. The list is drawn up by the lieutenant and the captain of the company. Once a firefighter is on the approved list, according to Chief Finnegan, he will rotate on the list, dropping to the bottom after he has served as acting lieutenant. Record, page 407. Finnegan testified that he had investigated the lists at the fire companies, and found a non-disproportionate distribution of black and white firefighters on the lists. Finnegan also testified that it is his policy that every man will act as soon as he possibly can, and that he makes a practice of looking at the duty rosters to see that his policy is being carried out. Finnegan also testified that no one has ever complained to him about being left out of the acting lieutenant rotations because of race. Record, page 409. Irving also alleges that the credit union engages in racially discriminatory practices against black firemen in its lending policy. The credit union is not a party to this action, but acquiescence in such practices by defendants would be evidence of discrimination. Chief Finnegan testified at trial that he had no reason to believe that the credit union engaged in discriminatory policies and practices in making loans. Record, page 473. Affiant Everett Leon Jasper alleges that the Fire Bureau allows junior whites to act as lieutenants in preference to senior blacks. Jasper says that when he questioned this policy, he was allowed to act. Jasper does not say whether or not this incident occurred during the time covered by this lawsuit. In any case, it appears from Jasper’s affidavit that the Fire Bureau is sensitive to complaints from black firemen about possible discrimination. Jasper also says that Captain Pappas, a white man, is prejudiced against blacks. Jasper does not say whether he has complained about Captain Pappas to higher officers in the Fire Bureau. Jasper further says that since 1972 the white officers have harassed black firemen about their hair styles and mustaches. The form that this harassment took is not set out. Chief Finnegan testified at trial that regulations for hair length were the same for all firemen, and that Afro haircuts were tested to insure that they did not interfere with gas mask tightness and fit. Record, pages 465-467. Jasper further states that he was rated “outstanding” for his 1975 performance, but that a new lieutenant in 1976 gave him a low rating even though his performance had remained the same. Jasper does not say whether the new lieutenant’s higher standards applied to all firemen or only to black firemen or only to Jasper himself. Jasper’s affidavit also contains the following statement: Chief G. E. Gray was talking to him [the affiant] one day and volunteered the statement “I carry a cross too.” He assumed that that was the Chief’s way of telling him that he [the Chief] was a member of the Klu Klux Klan. It is possible that Jasper’s assumption, given the context of the conversation, which the Court cannot know, is a perfectly reasonable one. But as it appears in Jasper’s affidavit, Chief Gray’s remark could just as easily refer to the burdens which Gray carries as a Christian, obedient to Jesus’ command, “If any man will come after me, let him deny himself and take up his cross, and follow me.” Mt. 16:24. Jasper further says that he and other blacks are not trained properly, while white firefighters are trained properly. The fact that Jasper received an “outstanding” rating for 1975 indicates that either this training is not necessary or Jasper was able to get the needed training in spite of his white superiors or that his complaint is groundless. No facts are adduced to indicate that training was restricted on a racial basis. Jasper also says that Chief Richard Palmer has a general reputation as a racist, and often makes unprovoked racist comments to black firemen. Jasp