624 F.2d 379
25 Fair Empl.Prac.Cas. 1077,
23 Empl. Prac. Dec. P 31,068
Porter HILTON et al., Plaintiffs, Appellants,
WYMAN-GORDON COMPANY et al., Defendants, Appellees.
United States Court of Appeals,
Argued Feb. 1, 1980.
Decided June 27, 1980.
Richard K. Latimer, Boston, Mass., with whom Henry F. Owens, III, Boston, Mass., and Joyce W. Poulin, Cambridge, Mass., were on brief, for plaintiffs, appellants.
Charles Donelan, Worcester, Mass., with whom James E. Wallace, Jr., George B. Sanders, Jr. and Bowditch & Dewey, Worcester, Mass., were on brief, for defendants, appellees.
Before COFFIN, Chief Judge, CAMPBELL and BOWNES, Circuit Judges.
COFFIN, Chief Judge.
Appellant Porter Hilton brought this action on behalf of himself and a class "composed of Negro persons who are employed, or might be employed, by Wyman-Gordon Company", a Worcester, Massachusetts manufacturer of heavy metal forgings. Appellants claimed that Wyman-Gordon maintained employment policies and practices which discriminated against members of the class on account of their race in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., and the Civil Rights Act of 1866, 42 U.S.C. § 1981. The gravamen of appellants' complaint is that they were hired as janitors or porters in the defendant's Housekeeping Department, while white workers were hired for better paying jobs in other departments with greater opportunity for advancement. After a nine day trial on issues of liability before the court sitting without a jury, judgment was entered for the defendant and this appeal followed. We affirm.
The evidence produced at trial consisted of three days of statistical proof and expert testimony for both sides and six days of nonstatistical proof including the testimony of individual plaintiffs, representatives of the defendant, and nonparty employees concerning the hiring and promotion practices within the company. The statistical evidence indicated that the percentage of minorities working in the company as a whole and within most departments was greater than the percentage of minority workers in the relevant Worcester area work force. It also indicated that the relative "overrepresentation" of minorities in the Housekeeping Department was greater than that of the other departments. The nonstatistical evidence, including findings of credibility made by the district court, showed that an appreciable number of blacks preferred housekeeping jobs to jobs in some other departments. The evidence also showed that the distribution of workers among the several departments was largely the result of numerous factors, including employee preferences, job qualifications, and economic conditions affecting job availability.
Appellants contend that the district court erred in issuing an "anticipatory ruling" that questions of liability would be determined as of February 28, 1969, at the earliest. Appellants argue that this is a "continuing violation" case involving class claims of ongoing discrimination and that the limitations period under Title VII remains open back to the Act's effective date, July 1, 1965. As for the § 1981 claim, appellants argue that the appropriate limitations period adopted by the district court should have been the Massachusetts six year period for contracts actions, M.G.L. c. 260 § 2, rather than the two year tort statute, M.G.L. c. 260 § 2A.
Whether the district court's "anticipatory ruling" was in fact error is, we think, irrelevant to our decision. Appellants do not point to a single instance in which the ruling affected the presentation of their claim. The district court exhaustively considered all of appellants' evidence, extending back even beyond 1965, and it found that there was no discrimination post-1964. The court therefore never found it necessary to determine finally the applicable limitations period.
Pursuant to International Brotherhood of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977), appellants attempted to establish a prima facie case of discrimination largely on the basis of statistical evidence. In the course of this attempt, a question arose concerning the level of statistical probability that appellants should be required to demonstrate. The district court concisely posed the question as follows:
"If one were viewing a purely chance event such as a coin tossing, theoretically, in the long run, heads and tails would come down an equal number of times. In practice, experience shows that this is never exact. Hence the question is, assuming there are enough occurrences to constitute a fair sample, how large a departure from the mathematically expected norm may be attributed simply to chance, and when there is to be thought a significant variance, sufficient to suggest that some factor other than chance was operating."
In answering this question, the district court decided to adopt the .05 level of significance, i. e., a level of precision that allows a variance from norms to be chalked up to randomness unless the probability of such a variance occurring is one out of twenty or less. This is a figure that was proposed by experts for both parties, that is commonly accepted in nonlegal contexts, and that is used by the Equal Employment Opportunity Commission in examining the validity of purportedly discriminatory employment tests, see Hallock, The Numbers Game The Use and Misuse of Statistics in Civil Rights Litigation, 23 Villanova Law Review 5, 11-13 (1977).
Appellants objected to the district court's decision and they contend that the .05 figure, while perhaps appropriate in other contexts, is in the context of this case arbitrary and unduly restrictive. They point out that, for example, a .10 figure would indicate that there was only a one out of ten chance that randomness was the explanation for a given disparity in the racial make-up of various departments of the company. Such an indication, appellants argue, should be adequate to establish a prima facie case.
In so arguing, appellants fail to suggest how the district court's adoption of the .05 figure harmed them, even assuming it was error. If the district court had completely adopted appellant's statistical theory, this would only have demonstrated that chance was not the cause of the disparity. To follow appellants to their goal, the establishment of a prima facie case, we would have to assume further that such a statistical demonstration would itself, in the absence of any corroborative nonstatistical proof, be sufficient to establish a prima facie case of race-based discrimination in the context of the suit. While such an assumption is by no means dictated by analogous Supreme Court precedent, see International Brotherhood of Teamsters v. United States, supra, 431 U.S. at 339, 97 S.Ct. at 1856 ("But, as even our brief summary of the evidence shows, this was not a case in which the Government relied on 'statistics alone'."),1 we are willing to entertain it in this instance for the purpose of demonstrating the flaw in appellants' argument.2
This flaw lies in the fact that fully adopting appellants' reasoning only brings us to the point of finding that they made out a prima facie case. The district court, however, specifically found that even assuming that appellants' statistics established a prima facie case, it "was so clear that defendant did not discriminate, that I would make an ultimate finding in its favor, even if for some reason, it had the burden."3 The district court supported this statement with additional factual findings, based on the testimony of witnesses for both parties, that employee preferences and job qualifications were major causes of the disparity. Viewed in this context, appellants' argument concerning the use of statistics is merely academic.
The United States Department of the Census has prepared certain statistical projections, the Standard Metropolitan Statistical Area projections (SMSA projections), which show the percentage racial composition of the Worcester area work force in each job category. The district court noted that on the basis of these projections, blacks were proportionately "over-represented in virtually all job categories and departments at Wyman-Gordon. Appellants assert that the district court erroneously relied upon this fact as evidence establishing that appellee did not discriminate. Appellants point out that they are not challenging defendants' aggregate hiring of blacks, but rather claiming that given the total number of blacks hired, a disproportionate number of them are assigned to the Housekeeping Department, a "dead-end job".
Appellants misconstrue the impact of the district court's reliance4 on the SMSA projections. The district court merely used the projections to demonstrate the narrow focus of plaintiffs' claim. Based on their own statistical evidence (which was their chief evidence aside from the testimony of individuals whom the district court found lacking in credibility), the court concluded that "plaintiffs are reduced to a single, allegedly disparate fact: that there were, from time to time, substantial (but steadily declining) overages in the hiring and complement of the housekeeping department." In so reasoning, the district court was simply restating appellants' own theory of the case. The fact that this restatement highlighted the weakness of the evidentiary support for that theory is by no means a reason to find error.
EEOC Investigatory Findings
Appellants introduced as evidence a report of the EEOC critical of appellee. The district court examined the report, but found "that whatever presumptive weight attached to this report was dissipated after lengthy discovery and a nine-day trial at which both sides have had, and have exercised, a full opportunity to present all aspects of the case." Specifically, the district court criticized the report as follows:"In addition, I find the report superficial in its reasoning. Defendant is criticized for having no blacks in housekeeping in (the Grafton plant) (in point of fact there were more blacks in Grafton than had applied to go there), without any indication that blacks were purposely excluded from Grafton or why they were prejudiced thereby, and without apparent consideration of the fact that, as was pointed out to the EEOC, and as plaintiffs' own witnesses brought out at trial, blacks chose not to go there, although offered the opportunity, because of transportation difficulties. In relating statistics, no mention was made of Worcester SMSA figures. Particularly important, defendant was condemned for the alleged effects of a departmental seniority system without even a suggestion that that system was purposely discriminatory . . . . (T) here are enough untruths, or only partial truths, so that I choose not to accept as probative any specific findings that were not independently supported."
Appellants mount a feeble attack on these findings of the district court primarily by challenging the credibility of one of defendant's witnesses who criticized the report. We would have thought it needless to point out that determinations of credibility are for the trial court. In any event, as the above excerpt from the court's opinion demonstrates, its findings rested on much more than the critical testimony of a defense witness. Appellants also attempt to defend the EEOC's failure to consider defendant's intent by pointing out that the relevance of intent in employment discrimination cases was not made clear until the Supreme Court's decision in Teamsters, decided after the EEOC report was produced. While this chronology may explain the EEOC's error, it does not mean that the district court should adopt that error as its own.5
Appellants also cite to our dictum in Blizard v. Fielding, 572 F.2d 13, 16 (1st Cir. 1978) that an EEOC investigatory finding is entitled to "great deference" from the trial court. As the context of that language in Blizard makes clear however, EEOC findings "will not bar a trial de novo " in which "both parties are entitled to an inquiry by a court into the charges." Moreover, as this case makes abundantly clear, while EEOC findings in general may be significant evidence, their probative force in individual cases varies considerably and is left to the determination of the trial court.
Appellants devote the last seven pages of their brief to the claim that the district court erred in failing to find that appellee's seniority system violated Title VII. Appellants contend that the issue here is not subject to the clearly erroneous standard. Appellants, however, concede that even if the seniority system, which they agree is neutral on its face, "locked-in" blacks to less desirable jobs,6 it would not constitute a violation of Title VII unless its adoption was the result of an intention to discriminate. International Brotherhood of Teamsters v. United States, 431 U.S. 324, 355-56, 97 S.Ct. 1843, 1864-65, 52 L.Ed.2d 396 (1977). Thus, contrary to appellants' argument, the issue whether appellee intended to discriminate is most certainly governed by the clearly erroneous standard. See Sweeney v. Bd. of Trustees of Keene State College, 604 F.2d 106 (1st Cir. 1979). Here, there was more than ample evidence including the testimony of appellee's officers and employees, to support the district court's finding that there was no intent to discriminate.7 The fact that appellants can point to other evidence giving rise to contrary inferences (some of which are questionable) is irrelevant.
Of the First Circuit, sitting by designation
The court in Teamsters did refer to "the use of statistical proof . . . to establish a prima facie case of racial discrimination in jury selection cases, see, e. g., Turner v. Fouche, 396 U.S. 346 (, 90 S.Ct. 532, 24 L.Ed.2d 567); Hernandez v. Texas, 347 U.S. 475 (, 74 S.Ct. 667, 98 L.Ed. 866); Norris v. Alabama, 294 U.S. 587 (, 55 S.Ct. 579, 79 L.Ed. 1074)", and stated that "(s)tatistics are equally competent in proving employment discrimination." 431 U.S. at 339, 97 S.Ct. at 1856. However, even reading this language as stating that the use which may be made of statistics in the two contexts employment discrimination and jury selection cases is identical (a reading which is somewhat undercut by the court's subsequent statement that statistics' "usefulness depends on all the surrounding facts and circumstances"), in the jury selection cases as well the findings of prima facie discrimination rested on more than statistics alone, see, e. g., Turner v. Fouche, 396 U.S. 346, 360, 90 S.Ct. 532, 540, 24 L.Ed.2d 567 (1970)
We do note, however, that in this case such an assumption is particularly suspect since the statistics themselves give rise to conflicting inferences. Thus, for example, one could infer from the statistics that the defendant never discriminates against blacks, but that he does discriminate against whites in the hiring of workers for the Housekeeping Department. Obviously, this inference is not the one which appellants draw. The point is, however, that at best the statistics are ambiguous
Of course the burden of proof always remained with appellants, Sweeney v. Bd. of Trustees, Keene State College, 439 U.S. 24, 99 S.Ct. 295, 58 L.Ed.2d 216 (1979)
The word "reliance" is itself a bit of a mischaracterization of the court's use of the projections. In determining that appellee did not discriminate, the district court relied primarily upon the extensive testimony of numerous witnesses for both sides. It used the SMSA projections merely to show how weak were the statistical inferences upon which appellants relied
Appellants do not even attempt to defend the report's reasoning pertaining to Grafton
The district court found that it did not
In this regard, it is helpful to quote a portion of the district court's reasoning:
"There are many departments, housekeeping being only one. Whatever plan defendant adopted would affect all departments and all employees. The alleged improper target, housekeeping blacks, constituted well under 2% of the total payroll. In the absence of evidence of extraordinary prejudice against them, it would be entirely unreasonable to assume that an employer would adopt a company-wide plan, not for business reasons, but in order to restrict the mobility of a small number of blacks in housekeeping. It would be particularly unreasonable to believe this in the present case in light of the fact that defendant was voluntarily there was no union at the time instituting changes whose effect was markedly to increase the mobility of housekeeping blacks, as was demonstrated by the fact that in the ensuing months almost 30% of the housekeeping blacks transferred out, almost none having done so before. I find defendant's actions were bona fide.