48 Fair empl.prac.cas. 282, 40 Empl. Prac. Dec. P 36,107 Melba J. Taylor v. Hudson Pulp and Paper Corporation

CLARK, Circuit Judge,

dissenting:

The record in this case clearly reveals that appellants met their prima facie burden of demonstrating sex discrimination in Hudson’s promotion practices. This claim should be remanded to the district court for presentation of Hudson’s case. With respect to appellants’ discrimination in placement (transfer) claim, the district court failed to consider (or reveal how it considered) significant, relevant evidence. It also made critical errors of law. Rather than conducting our own fact-finding to determine whether the district court’s judgment should be affirmed, we should remand this claim to the district court for proper consideration of appellants’ futility argument.

I. Promotions

To understand appellants’ promotion claim, it is necessary to know something about the way promotions at Hudson were to work under the relevant collective bargaining agreement. When a vacancy occurred in a permanent position, the person with the most temporary seniority in the *1463 vacant position was to be offered the promotion. If no one held a temporary slot in the position or if the temporary employees declined promotion, the person with the most job seniority in the position previous in the line of progression to the vacant position would be promoted into the available job unless that person signed a waiver form declining the promotion. There was no need to apply for promotion or indicate any interest in it. The person next in line (in terms of job seniority) was automatically to be asked.

The bag catcher position, historically assigned to women, and the chute loader position, historically assigned to men, both fed into the baler position as of the merger of the lines of progression in the Grocery Bag Department in 1974. When a baler vacancy occurred, the most senior bag catcher or chute loader should have been promoted to baler, unless someone was already working as a temporary baler and was therefore entitled to preference.

In September, 1976, seventeen men were promoted to permanent baler positions. Fifteen of these were temporary balers and so properly given priority in promotion. 1 However, two of the promotions were given to men whose job seniority entitled them to no preference over appellants. One had temporary seniority as a “broke handler” dating back to January 30, 1976. He had moved into broke handling from chute loading and had mill seniority only as of July, 1975. The other had temporary seniority as a “bypass rich and joyce for broke” and had mill seniority as of July, 1975. He had also been a permanent chute loader when placed in his temporary position. Appellants all had job seniority as bag catchers dating back to September, 1970. None was offered a baler position at the relevant time. As these men had neither temporary seniority as baler entitling them to preference in promotion nor sufficient job seniority as chute loaders to entitle them to promotion over appellants, appellants have clearly made out a prima facie case of discrimination in promotion falling within the limitations period.

The majority acknowledges the evidence presented by appellants to demonstrate that the two baler promotions were dis-criminatorily made, but fails to accord it any significance, saying only: “Taking the evidence regarding this issue on the whole, the district court’s finding on this issue was not clearly erroneous.” Majority Op. at 1461. I cannot understand the majority’s approach to this issue. The district court did not discuss appellants’ promotion claim, let alone make any findings with respect thereto. How, then, can the majority determine that the district court’s finding on this issue is not clearly erroneous?

The district court discussed only the transfer claim. In that context, the district court found that no plaintiff had requested an intradepartmental transfer and that no vacancy had occurred during the limitations period. Record Excerpts at 63-64. To the extent that these may be considered findings bearing on the promotion claim, the majority’s discussion reveals them to be clearly erroneous. Majority Op. at 1461. There is no dispute that there was no such thing as an “intradepartmental transfer” under the 1974-1977 collective bargaining agreement and that an employee was not required to request a promotion. Furthermore, there were at least seventeen vacancies in the permanent baler position during the limitations period. There is simply no basis for affirming the district court’s dismissal of the promotion claim. “Taking the evidence regarding this issue on the whole,” fifteen nondiscriminatory promotions do not cure or in any way legitimate two discriminatory promotions.

*1464 It is true that appellants have demonstrated only two instances of discrimination within the limitations period while appellants number six. Appellants have alleged that the failure to promote women in the Grocery Bag Department is continuing discrimination. They need only demonstrate one instance of present discrimination within the period to be entitled to recover for other instances of the continuing discrimination that occurred before the onset of the limitations period. In Havens Realty Corp. v. Coleman, 455 U.S. 363 , 102 S.Ct. 1114 , 71 L.Ed.2d 214 (1982), the Supreme Court held, in the Fair Housing Act context, that “[w]here a plaintiff ... challenges not just one incident of conduct vio-lative of the Act, but an unlawful practice that continues into the limitations period, the complaint is timely when it is filed within 180 days of the last asserted occurrence of that practice.” Id. at 382-83 , 102 S.Ct. at 1125 . In Havens Realty, plaintiffs had alleged five separate instances of housing discrimination, only one of which fell within the 180-day limitations period. The other discriminatory acts had been suffered by a different plaintiff prior to the 180-day period, yet the Supreme Court held that all the claims were timely because they were all instances of one continuing violation.

The Court pointed out that staleness concerns disappear when a discriminatory practice is shown to have continued into the limitations period. It further reasoned that “wooden application” of the statute of limitations “ignores the continuing nature of the alleged violation” and “undermines the broad remedial intent of Congress.” Id. Although the Court was not interpreting Title VII’s statute of limitations, the reasoning of the Court applies equally in the Title VII context. Indeed, the Supreme Court cited a Title VII case in explaining that application of a statute of limitations should reflect the broad remedial purpose of the law. Zipes v. Trans World Airlines, Inc., 455 U.S. 385 , 102 S.Ct. 1127 , 71 L.Ed.2d 234 (1982). Moreover, when plaintiffs show that discrimination continued into the limitations period, the court is assured that they challenge present discrimination rather than the present effect of past discrimination. See United Air Lines, Inc. v. Evans, 431 U.S. 553 , 97 S.Ct. 1885 , 52 L.Ed.2d 571 (1977).

In this circuit, a plaintiff who has never filed an administrative charge may proceed in an action with a plaintiff who has properly filed charges if the issues they raise are similar. See, e.g., Griffin v. Carlin, 755 F.2d 1516, 1532 (11th Cir.1985) (class action); Crawford v. United States Steel Corp., 660 F.2d 663, 665-66 (5th Cir. Unit B 1981) (multiple plaintiff non-class action). It is not clear from the facts or discussions in these cases whether nonfiling plaintiffs are limited to litigating claims that arose during the limitations period relevant to the charges that were filed. However, the courts reasoned that the purpose of requiring that charges be filed with the EEOC— to give the EEOC notice of and an opportunity to investigate discriminatory practices — is served when one charge is filed. Where, as here, the discriminatory actions that occurred prior to the limitations period were but instances of a practice shown to continue into the period, the same reasoning supports allowing all the plaintiffs to proceed on the charges that were timely filed. The EEOC was notified of and given a chance to investigate the discriminatory practice — failure to promote. “There is no question but that, if the employer has a formal rule or policy which discriminates in ... allocation of jobs ..., and the system is maintained into the period, the system or practice can be attacked despite the fact that the plaintiff was not denied a particular job ... within said charge-filing period.” B. Schlei & P. Grossman, Employment Discrimination Law 1050 (2d ed. 1983) (citing cases); see also C. Sullivan, M. Zimmer & R. Richards, Federal Statutory Law of Employment Discrimination 279-80 (1980).

There can be no dispute that the claims of appellants arose from the same discriminatory practice. All worked in the same department at the same position. All were denied the opportunity to be promoted to baler because, despite the formal merger of *1465 the bag catcher position into the baler line of progression in 1974, bag catchers (all women) were never offered the promotions due them under the collective bargaining agreement. Because two of the EEOC charges were timely filed, the district court had jurisdiction over the promotion claims of all six appellants. As pointed out earlier, appellants have demonstrated that all fifteen of the men promoted from temporary to permanent baler in September of 1976 were promoted to temporary baler in 1975 and early 1976 over appellants, who had more job seniority than any of the fifteen men receiving the transfers. There can be no doubt that each of these women has made out a prima facie case of sex discrimination in promotions, and each is entitled to return to the district court for the completion of the trial.

II. Transfers

The district court found that appellants failed to establish a prima facie case because they did not prove that they had requested transfers or that vacancies had occurred during the limitations period. Record Excerpts at 64. The finding regarding vacancies (not discussed in the majority opinion) is clearly erroneous in light of evidence that thirty-six entry-level openings outside the Grocery Bag Department were filled by men during the limitations period. See Appellants’ Brief at 18a-23a (Table B); Plaintiffs’ Exhibit 8. The majority affirms the district court's finding with respect to the failure to apply and rejects appellants’ claim that they did not apply for transfers because of a justifiable belief that such application would have been futile. In reaching its decision, the majority conducts substantial fact-finding that it should not, as an appellate court, conduct. I cannot agree with its decision to do so or with its factual conclusions.

Appellants’ futility argument is based on International Brotherhood of Teamsters v. United States, 431 U.S. 324 , 97 S.Ct. 1843 , 52 L.Ed.2d 396 (1977). In that case, the Supreme Court recognized that:

[t]he effects of and injuries suffered from discriminatory employment practices are not always confined to those who were expressly denied a requested employment opportunity. A consistently enforced discriminatory policy can surely deter job applications from those who are aware of it and are unwilling to subject themselves to the humiliation of explicit and certain rejection.

Id. at 365 , 97 S.Ct. at 1870. The Court held that nonapplicants could make out a prima facie case of race discrimination under Title VII. In applying Teamsters, a court looks to the existence of a belief in futility (the state of mind of the plaintiff) and the reasonableness of the belief (an objective inquiry). See Garrett v. Okaloosa County, 734 F.2d 621, 625 (11th Cir.1984); Reed v. Lockheed Aircraft Corp., 613 F.2d 757, 761-62 (9th Cir.1980).

The district court made no finding as to the state of mind of appellants. 2 The district court instead framed the inquiry in objective terms: would it have been futile to apply? Record Excerpts at 64. The appropriate question is: Did appellants justifiably believe that application for transfer would have been futile because of Hudson’s discriminatory practices? Framing the question correctly is crucial to identifying the evidence relevant to the fact-finding determination. Focusing on justifiable belief broadens the inquiry to include evidence of the plaintiff’s career-long experiences with the employer’s discrimination in-all areas, as well as of the employer’s actual practice during the limitations period, because such experiences would affect a female employee’s expectations with regard to her transfer opportunities.

Any finding implicit in the district court’s discussion that a belief in futility was not justified (or, for that matter, that application would not have been futile) is clearly erroneous. The court first found that only the testimony of Howell Gordon supported *1466 appellants’ futility argument. Gordon, a supervisor in the Grocery Bag Department from 1962 until January, 1977, testified that even after the formal merger of the men’s and women’s lines of progression in 1974, he continued to assign women only to historically women’s jobs (bag catcher or inspector) and men only to historically men’s jobs (baler or machine operator). This practice continued through the limitations period.

This testimony, alone sufficient to demonstrate that appellants’ belief in futility was justified and that the district court was clearly erroneous in finding that it would not have been futile to apply for men’s jobs, was corroborated and supplemented with further evidence that Hudson continued to discriminate up to and through the limitations period. Male and female employees testified that they were unaware that the segregated lines of progression had been merged in 1974. As I have already mentioned, documentary evidence corroborates Gordon’s testimony, and the employees’ perception, by demonstrating that men were promoted into the baler position ahead of women with far more years job seniority. Appellants introduced Weekly Work Schedules denominating certain jobs as “men’s” and others as “women’s.” These separate listings of men’s and women’s jobs continued through the relevant period and into 1980. Hudson also maintained separate “Extra Boards” — on which laid-off employees were to indicate an interest in temporary jobs in departments other than those in which they permanently worked — for men and women.

The 1974-1977 collective bargaining agreement (as well as all such agreements through 1983) provided that female employees would be entitled to two break periods per shift not given to men, with the following exception:

Female employees who, on or after December 1, 1965, move on to jobs not previously available to females, will not have available to them the two fifteen (15) minute rest periods referred to above.

Plaintiff’s Exhibit 7-G at 60. Only men were allowed to work overtime, at overtime rates, to clean up in the Grocery Bag Department. Appellants’ “snapshot” statistics demonstrate that fifteen years after enactment of Title VII, and four years after the actionable period in this case, Hudson’s work force remained almost completely segregated by sex. 3

All this evidence relates to discriminatory practices contemporaneous to the actionable period that would influence a female employee’s perception of her mobility within the Hudson plant. Because the appellants’ perceptions were shaped by previous *1467 experiences with their employer, 4 the aforementioned evidence must be viewed against the backdrop of Hudson’s long history of sex segregation and discrimination.

The collective bargaining agreements through 1966 set forth separate starting rates and base rates for men and women, with women’s rates lower than men’s. Women’s jobs within each department were designated as such in the collective bargaining agreement. This blatant sex segregation continued until January, 1966, six months after the effective date of Title VII. 5 Thereafter, most sex-specific job titles were removed from the contract, but some remained: “Matron,” “Broke Man,” “Chute Man,” “Pasteman.” There is no evidence that Hudson made any further effort to eliminate sex segregation before 1974.

Practices in the Multiwall Department, where appellants worked until 1970, reflected the sex segregation formally established in the collective bargaining agreements. Women worked at certain jobs that, until January, 1966, had sex-specific titles. Men worked at other jobs. When the Multiwall Department was relocated to a distant city and appellants were forced to transfer to new jobs, they were told that they could transfer only to the Grocery Bag Department or the Tissue Converting Department, the two departments that had women-only lines of progression. The male Multiwall employees transferred to jobs in many other departments.

After their transfers, Hudson continued to communicate to appellants that women were allowed to fill only certain positions. As Howell Gordon testified, job assignments in the Grocery Bag Department continued to be made according to sex through January, 1977. Indeed, the district court found that “sex was involved to some extent in where individuals were placed, from the period of 1970 to at least 1974.” Record Excerpts at 62.

Ignoring most of the evidence that Hudson continued to discriminate through the limitations period, the district court relied on three pieces of evidence to find that it would not have been futile to apply to transfer to a male department or job: (1) the merger of the bag catcher (female) and machine operator (male) lines of progression in the 1974 collective bargaining agreement; (2) the successful transfer of a few women into historically men’s jobs (three had transferred prior to the commencement of the limitations period; another two transferred during the limitations period); (3) the 1974 publication of an article in the Hudson newsletter indicating that women could transfer into historically men’s jobs. Rejecting Mr. Gordon’s testimony — direct evidence that sex discrimination continued through the limitations period 6 —the district court found that application would not have been futile and that appellants had failed to establish a prima facie case of sex discrimination. It did not find that appellants did not believe application would have been futile. Nor did it find that any such belief would not have been justified.

The majority, also ignoring most of the evidence that Hudson continued to discriminate, 7 finds that the evidence upon which *1468 the district court relied supports the conclusion that “[c]onsiderable evidence exists ... that by the onset of the actionable period Hudson had abandoned any alleged discriminatory practices, was undertaking to remedy the effects of past practices, and that appellants were aware that Hudson was in fact actively encouraging women to apply for positions which had previously been occupied exclusively by men.” Majority Op. at 1459 (emphasis added). 8 To do so, it must find that appellants were aware of the steps taken by Hudson, a leap the district court did not make, and infer that appellants could not justifiably have retained a belief in futility in the face of this awareness.

Its effort is unconvincing. None of the appellants had seen the “Hey Lady, Wanna Be a Welder” article. Furthermore, although each appellant testified that she “knew” one or the other of the women who had successfully transferred into “male” positions, each also testified that she did not know about the transfers. It appears from the testimony that appellants’ acquaintance with the transferees was minimal and is more accurately reflected in stating that appellants knew who the transferees were than in stating that appellants knew them. Even if it can reasonably be inferred that appellants were aware that a very few women had successfully transferred, Teamsters demonstrates that evidence of visible tokenism should not defeat a futility claim. In that case, the Supreme Court in no way indicated that the fact that five Spanish-surnamed individuals had been hired as line drivers prior to filing of the law suit would preclude all Spanish-sur-named nonapplicant class members from participating in the superseniority remedy. See Teamsters, 431 U.S. at 337, 362-71 , 97 S.Ct. at 1855, 1868-73 ; see also Bing v. Roadway Express, Inc., 485 F.2d 441, 451 (5th Cir.1973) (“Certainly a few ... have the courage to fight ‘the system,’ but it is equally certain that others must have been intimidated and discouraged’ by [the employer’s] practices.”). Finally, paper merger of historically sex-segregated lines of progression does not render unjustified a belief in the futility of application for transfer when testimony by the person responsible for putting paper into practice reveals that the merger was not effectuated during the relevant period. One does not have to credit the testimony of male and female (plaintiff and non-plaintiff) witnesses that they were unaware of the formal merger to find that the paper merger in no way supports rejection of the futility argument.

In addition to the evidence recited by the district court, the majority leans heavily on its own inferences drawn from the fact that a superseniority agreement for the benefit of women such as appellants was negotiated and signed by Hudson and the unions during the limitations period. The fact that a superseniority plan was in the works during the limitations period may be relevant to the futility assessment in two ways. First, it may be considered to have demonstrated to appellants that Hudson was willing to allow them to break into formerly all-male positions. Second, it may be offered to support the contention that appellants refrained from applying to take advantage of the upcoming transfer benefits rather than because of Hudson’s entrenched discrimination. The second point *1469 goes to whether any justifiable belief actually caused the failure to apply, which will be discussed below.

The first point reflects on whether the appellants’ asserted belief that they could not transfer into male-only jobs was justified. If appellants knew about the negotiations, perhaps they should have realized that Hudson was willing to allow previously forbidden transfers. The majority suggests that appellants were aware that the designate plan was being discussed during the limitations period. However, appellants testified that they were unaware that the designate seniority plan had been adopted until Hudson held meetings in the fall of 1976 to inform the employees of the terms of the agreement. It is true that an affidavit given by Melba Knowles in 1982, which states that there was widespread hostility toward the designate seniority plan during the time the plan was being discussed and negotiated, conflicts with her testimony. The affidavit is the only evidence that any of the appellants knew or should have known of the plan before September, 1976. It does not prove that Knowles was aware of the plan throughout the limitations period, the conclusion that must be reached to defeat her futility argument. Nor does it reveal the level of detail of any knowledge that may be inferred to have existed. There is no evidence that appellants “attended numerous union meetings during the period the negotiations took place and at which these negotiations were discussed.” Majority Op. at 1459-1460. Appellants’ testimony disclosed only that some of them were active in the union at some unspecified point in their twenty-plus year careers with Hudson and that some of them attended an unspecified number of meetings at unspecified times during 1975 and 1976. There was no evidence presented that the designate plan was discussed at any meeting, let alone at any meeting attended by an appellant. Thus, the determination whether appellants knew anything about the designate plan that should have communicated to them Hudson’s willingness to allow women into men’s jobs is purely a matter of credibility assessment. The district court made no finding one way or the other, and the majority is in no position to resolve a credibility dispute.

Assuming, however, that appellants were fully aware that Hudson and the unions were negotiating a plan that would allow women and blacks to transfer into previously segregated jobs, the question remains whether this knowledge would render incredible or unjustified any belief in the futility of applying for a transfer prior to the effective date of the plan. On the facts of this case, given appellants long experience with Hudson’s discriminatory practices, I can only conclude that the credibility of appellants’ belief that they could not transfer is not weakened by such knowledge. Rather than understanding Hudson’s participation in negotiations to mean that they were free to transfer into men’s jobs, women such as appellants may have understood negotiations to represent a battle over the means by which they would be admitted to forbidden territory; implementation of the plan, and not Hudson’s willingness to negotiate (if any can be presumed), would signal the end of discrimination. Thus, it would be reasonable for appellants to have believed that they would be allowed to transfer to “men’s” jobs only through, and because of, the designate program. Their knowledge that the plan might take effect ought not be equated with belief they could transfer prior to the effective date.

Hudson’s history of sex discrimination and appellants’ extensive and intimate familiarity with it, as well as the evidence of discrimination occurring within the limitations period, establishes that appellants’ belief that they would not be allowed to transfer out of women-only jobs was credible and justified. However, I realize that to reach this conclusion I have, had to indulge in the same kind of fact-finding for which I have taken the majority to task. As the district court did not make findings one way or the other on the issue of the existence and justifiability of appellants’ belief in futility, and erred in finding appli *1470 cation would not be futile, I would remand to the district court to make findings consistent with the correct legal standard. I cannot join in the majority’s factual determinations 9 or the legal conclusions resulting therefrom.

I am also troubled by the majority’s treatment of the question whether appellants have demonstrated that they did not apply because of their belief in the futility of application. The majority states that appellants did not identify jobs they actually wanted or in which they expressed an interest during the limitations period but relied on their designate program bids to prove their interest. Majority Op. at 1460. However, appellants do not claim that their willingness to transfer under the designate program proves that they would have transferred prior to implementation of the program if they had not believed that application would be futile. Rather, each testified that she would have transferred if she had realized she would be allowed to do so. This evidence is sufficient to show that appellants’ failure to apply was caused by Hudson’s discrimination. Appellants need not demonstrate that they actually expressed desire for particular jobs, for the plaintiff’s task is to prove that “he would have applied for the job had it not been for those practices.” Teamsters, 431 U.S. at 368 , 97 S.Ct. at 1871 ; see also Cox, supra, at 1560. Although the Court in Teamsters indicated in a footnote that such proof might take the form of evidence of an informal inquiry, expression of interest or unexpressed desire, id. 431 U.S. at 371 n. 58, 97 S.Ct. at 1873 n. 58, its discussion makes clear that these are all just means of identifying the “choice an employee would have made had he previously been given the opportunity freely to choose____” Id. at 371, 97 S.Ct. at 1873. An employee who can show that she would have transferred into a “man’s” job if not for discriminatory practices that led her to believe such a step would be out of the question has demonstrated that she would have chosen to transfer. 10

Appellants had all worked for Hudson for over twenty years as of the beginning of the limitations period. For over twenty years, Hudson had communicated to appellants that certain jobs were for women, and women only, while others — the better-paying jobs — were for men and men only. Because these women had worked only for Hudson for all or the great majority of their working years, their belief that they could not work in men’s jobs may be attributed to their experience with Hudson’s discriminatory practices. Their testimony that they would have transferred into “men’s” jobs if they had known they could should be credited as proof that Hudson's discrimination caused their failure to apply for transfers.

*1471 Even under the majority’s interpretation of appellants’ burden, several of the appellants should be found to have succeeded in carrying it. Vivian Earls applied and was rejected for a printer position sometime in 1975 or 1976 (before commencement of the limitations period). 11 Carrie Branham testified that she had been interested in the jobs of baler and machine operator. Jessie Mathews testified that she would have moved into the tissue pulp tester job if given the opportunity. Melba (Taylor) Knowles asked the labor relations manager, Bob Garrett, if she could transfer to the lab when Multiwall was phased out in 1970. She also asked Carl Kight, a union official, about becoming a baler in early 1976.

The majority fails to acknowledge this evidence and rejects appellants’ testimony that they would have applied to transfer to the better-paying “men’s” jobs if they had realized women were allowed to do so, finding instead that “evidence ... exists that appellants delayed any contemplated requests for transfers not because of the prospect of discriminatory rejections, but rather because of the unattractive policy Of loss of departmental seniority upon transfers to another line.” Majority Op. at 1461.

The evidence cited by the majority, appellants’ EEOC charges, does not, however, contain assertions that “they could not transfer to other jobs because such transfers would entail a loss of departmental seniority and the risk of unacceptable layoffs.” Id. The charges state that they were not permitted to transfer because they were women. They also state that if they were allowed to transfer they would be subject to layoff despite their twenty-plus years of mill seniority. They did not say or imply that they would not transfer because of the loss of seniority. Obviously, that they felt it was unfair and discriminatory that they should always remain at a disadvantage to more junior male employees does not tell us whether they would have been willing to risk departmental seniority to transfer to better-paying jobs. There is no other evidence that they were deterred by neutral disincentives in the collective bargaining agreement. If we are going to infer anything from the bare fact that the collective bargaining agreement contained disincentives to transfer, we may as well infer that appellants were deterred by the loss of two fifteen-minute breaks per shift, a gender-specific disincentive. See supra, p. 1461. We would not be required to reject appellants’ futility argument were we to do so. I see no reason to choose one inference over the other.

Moreover, there is no evidence that appellants delayed their applications to take advantage of the designate program or had any reason to do so. See Majority Op. at 1460. Nothing in the record or, more particularly, in the Special Seniority Procedures indicates that a woman who transferred or applied to transfer prior to the effective date of the designate program would be barred from participating, or in any way disadvantaged if allowed to participate, in the designate program. There is no basis for the majority’s inference that anticipation of the designate program, rather than appellants’ belief in futility, deterred them from applying for transfer into “men’s” jobs during the limitations period. There being no evidence that appellants were deterred from applying by other gender-neutral considerations, I would accept their testimony that they would have applied for “men’s” jobs had they not been led to believe they would not be allowed to take such jobs.

III. Conclusion

Ultimately, the question to be answered in this case is whether appellants were victims of intentionally discriminatory practices of which the EEOC was timely apprised. My review of the record leaves me with the definite and firm impression that the district court erred in failing to find that appellants’ evidence raised a rebut- *1472 table presumption that they had indeed suffered such discrimination. That Hudson took steps to end some of its discriminatory practices does not lessen or eliminate its liability for the discrimination experienced by appellants. An employer may not “substitute” present efforts to remedy discrimination for liability for past discrimination. Majority Op. at 1462. Neither does the law support the district court’s finding that “any discriminatory practices prior to 1974 was [sic] cured by the [1974-1977] collective bargaining agreement, which on its face is facially neutral.” Record Excerpts at 63. We may not dismiss a sex discrimination case against an employer because we fear it will “chill bona fide and constructive measures to abide [sic] the national policy [embodied in Title VII].” Majority Op. at 1462. As the Supreme Court recognized in Teamsters:

The company’s later changes in its hiring and promotion policies could be of little comfort to the victims of the earlier post-Act discrimination, and could not erase its previously illegal conduct or its obligation to afford relief to those who suffered because of it.

341 U.S. at 341-42, 97 S.Ct. at 1858. Regard for an employer’s attempt to change its ways should not influence our assessment of its liability for actions that are not as worthy of regard.

Appellants have demonstrated that there occurred during the limitations period vacancies for which they were qualified and for which they would have applied if not for their justified belief, based on Hudson’s discriminatory practices, that to do so would be futile. I conclude that the district court committed legal and factual errors in analyzing appellants’ transfer claim and in holding that plaintiffs had not established a prima facie case. Those errors should require us to remand this case for fact-finding on the issues whether Hudson’s past and present discriminatory practices led appellants to believe they would not be allowed to transfer and, if so, whether this belief in the futility of application actually explains appellants’ failure to apply. Furthermore, as the record clearly shows that appellants established a prima facie case of continuing sex discrimination in promotions, I would reverse the district court’s dismissal of that claim and remand for completion of trial. The majority determines otherwise and so I respectfully dissent.

. However, the original placement of the fifteen men in temporary baler positions was clearly discriminatory. These men, who had not only job but mill seniority that dated back only to 1975 or 1976, were made temporary balers be fore women who had job seniority dating back as far as the 1950’s. No woman was offered a promotion. Because the original discrimination occurred before the limitations period, it cannot form the basis of appellants’ prima facie case unless other instances of discrimination in promotion occurred within the period. See in-jra at 1464-65.

. In fact, all testified that they did not think that they were allowed to work anywhere in the plant but in the jobs historically assigned to women.

. Although appellants do not rely on their statistical evidence to establish prima facie proof of discrimination, the majority’s possibly misleading discussion of the proper use of ‘‘snapshot’’ statistics in proving discrimination should be noted. The Supreme Court in Teamsters, supra, held that such statistics may be sufficient to establish a prima facie case of discrimination. 431 U.S. at 339, 97 S.Ct. at 1856; see also Griffin v. Carlin, 755 F.2d 1516, 1525 (11th Cir.1985). Thus, it is the defendant’s burden to refute “snapshot" statistics by offering more refined and accurate statistics such as “flow” statistics showing employment patterns as they occurred over time. See Movement for Opportunity and Equality v. General Motors Corp., 622 F.2d 1235, 1245 (7th Cir.1980). The defendant may not successfully rebut stark statistical evidence of segregation by pointing to a declining work force, as the majority contends, for it is the employee turnover rate, not the overall number of employees, that determines the number of vacancies. Teamsters, 431 U.S. at 341 , 97 S.Ct. at 1857 . So long as the number of new hires were less than the number of departures, the employer could hire in significant numbers while experiencing an overall decline in work force.

If snapshot statistics may form the basis of a prima facie case of discrimination, there appears to be no reason to consider such statistical evidence irrelevant to the justifiability of a belief in futility due to the employer's discrimination. If a court may infer that a rigidly segregated work force reflects discriminatory practices, it must grant that similar inferences drawn by employees are justified. As the statistics presented in this case are as stark as those before the Court in Teamsters, I would give them considerable weight in assessing whether appellants’ belief — that applying for transfer into historically male jobs would be futile in light of the employer’s discriminatory practices — was justified.

. Melba Taylor (now Knowles), Jessie Mathews, Vivian Earls, Gertrude Burroughs, Delphine Carter and Carrie Branham were hired in 1953, 1953, 1949, 1950, 1954 and 1953, respectively.

. Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e was enacted on July 2, 1964 and went into effect on July 2, 1965.

. Because plaintiffs presented direct evidence of discrimination, it would have been Hudson’s burden to prove that it would have made the same decisions even in the absence of the discriminatory motivation, Cox v. American Cast Iron Pipe Co., 784 F.2d 1546, 1559 (11th Cir. 1986); Lee v. Russell County Board of Education, 684 F.2d 769, 774 (11th Cir.1982).

. Although the majority acknowledges evidence that tends to justify belief in futility and, more generally, to bolster appellants’ prima facie case, it concludes that the district court’s (unspecified) findings were not clearly erroneous. Majority Op. at 1461. If the findings to which the majority refers are the findings with respect to justified belief in futility, its discussion of the evidence misses the point. It does not matter that discriminatory practices took place before *1468 the actionable period so long as the nonappli-cant’s belief in futility arising from those practices justifiably continued into the limitations period. As many of the instances of discrimination not falling within the limitations period occurred immediately prior to the relevant period (not to mention the practices that continued throughout the period), these instances, at least, should be weighed in support of appellants’ position.

. Considering the evidence recounted above that Hudson continued to discriminate past the actionable period, the majority’s statement is bald hyperbole. However laudable Hudson’s initial efforts to end its discriminatory practices and desegregate its work force, the fact remains that its attitude toward discrimination, as revealed in its practices, remained equivocal at best. Far from "[cjonsiderable evidence" that Hudson "abandoned” discrimination, there is considerable evidence that Hudson wanted to have its discrimination and avoid lawsuits, too. Any interpretation of the record that sees active encouragement of desegregation is likewise wishful thinking.

. Once again I must point out that the majority’s sweeping conclusions with respect to factual issues are unsupported by the record. In particular, I refer to the reiteration that “considerable evidence has been presented ... that Hudson had abandoned any alleged discriminatory practices long before the onset of the actionable period ...,” Majority Op. at 1461, as well as the statement that "the evidence is overwhelming that an application for transfer would not have been futile; such applications for transfer were actively solicited by the appellees and, when received, were granted." Majority Op. at 1462. The factual questions are far more difficult than the majority would have them. Furthermore, there is no evidence that all applications for transfer were granted.

. The majority’s emphasis on proof of interest or desire places beyond the scope of Title VII the most entrenched forms of sex discrimination. Sex discrimination has been marked by its segregation of women into work roles different (not merely separate) from those performed by men. The woman who has been most thoroughly victimized by sex discrimination will believe she is limited to those jobs traditionally assigned to women. She will not realize she wants a "man's" job. Yet, when a woman’s desire for a job is not translated into a formal application because the sex discrimination with which she lives has become internalized to the point she believes herself to have no choices, she is as much a victim of discrimination as is she who can envision herself working at a “man’s” job and turns that vision into desire. To hold that the first woman is not protected by Title VII "putts] beyond the reach of equity the most invidious effects of employment discrimination — those that extend to the very hope of self-realization.” Teamsters, 431 U.S. at 367 , 97 S.Ct. at 1871 .

. Evidence of interest expressed prior to the actionable period is relevant. Teamsters, 431 U.S. at 371 n. 58, 97 S.Ct. at 1873 n. 58 (overt act such as pre-Act application would be most convincing proof that futility caused failure to apply)-