This Title VII case arose out of Safeway, Inc.'s closure of its "pre-pakt" plant in Denver, Colorado. Plaintiffs/cross-appellants, long-term female employees at the pre-pakt plant, alleged that they were discriminated against on the basis of sex when they were laid off and not rehired when the plant closed in June, 1984. The case was bifurcated and the issue of liability was tried to the United States District Court for the District of Colorado. The court found that Safeway had not engaged in disparate treatment of plaintiffs but did conclude that Safeway had a policy which had a disparate impact based on sex. Pursuant to a subsequent court order, Safeway reinstated plaintiffs commencing August 7, 1988. After awarding attorneys' fees and costs against Safeway, the court entered judgment for plaintiffs in the total amount of $612,569.10, plus post-judgment interest. Additionally, the court ordered Safeway to pay $17,633.25 to six of the plaintiffs' pension plans. Safeway appeals the finding of disparate impact and plaintiffs cross-appeal the finding of no disparate treatment. We affirm the finding of no disparate treatment and reverse the finding of disparate impact.

Plaintiffs Dolores Rojo, Martha Ortega, Josphine Duran, Ray Ann Lobato, Bertha Medina, Mary Martinez, Cora Charboneau and Yvonne Gutierrez worked as packers in Safeway's Pre-Pakt Plant until its closure on June 22, 1984. As packers, they sorted and packaged vegetables along an assembly line for sale in Safeway's retail stores. Their years of service to Safeway ranged from 12 to 20 years.

Plaintiffs were eight of fifteen female packers employed at the plant. The eighteen males employed at the plant all worked as warehousepersons, keeping the assembly lines where plaintiffs worked supplied with goods. All thirty-three pre-pakt plant employees were union members and were covered by a collective bargaining agreement. While historically, the jobs of packer and warehouseperson were sex-segregated, in the 1970's Safeway encouraged males to work as packers and females to work as warehousepersons. Few in fact did, although one of the plaintiffs, Cora Charboneau, had been a warehouseperson for three years until an injury forced her to resume packing. Generally speaking, the job of warehouseperson required more strenuous physical work than the job of packer.

Safeway decided in 1984 to close the pre-pakt plant because outside packing houses could provide the same service at a lower cost. The closure of the plant resulted in the abolition of all but three positions. Those positions were in the banana-ripening operation. Pursuant to the relevant union contract, the banana-ripening jobs went to the three most senior of the qualified warehousepersons in the pre-pakt plant.1

This case involves Safeway's policies and practices for placing employees who have been laid off. Safeway has a policy that laid-off employees retain their company-wide seniority only if they are rehired within 30 days of their termination. Accordingly, employees who were placed in new positions within 30 days of the pre-pakt plant closure retained their company-wide seniority, whereas those who were not lost that seniority. Vacation time and sick leave are computed on the basis of company-wide seniority. It was therefore important to those employees wishing to be rehired that they be rehired within the 30 day period.

Safeway had no established written policy for rehiring employees who lost their jobs when a plant closed. There was evidence, however, that Safeway's "formal" verbal policy was to assist laid-off employees through the Central Employment Office, placing them in order of seniority and granting to them priority over new hires for jobs for which they were qualified. Such employees were to be placed in whatever jobs were available in any of Safeway's three divisions (the distribution center, the supply plants and the retail stores).2 There was also evidence that, in this case, there was an informal "word-of-mouth" policy pursuant to which Norm Scherzer, the distribution center manager, helped find new jobs for pre-pakt employees who approached him. We consider the evidence relating to each policy in turn.

With respect to the formal policy involving the Central Employment Office, job application forms were distributed to plant employees prior to its closure. All of the plaintiffs filled out job applications and were interviewed by Ray Simmons, a representative from the Central Employment Office. Plaintiff Ortega's application indicated she could not work nights or weekends, that she sought full-time work and that she could not lift and stack more than 30 pounds.3 Plaintiff Rojo's application indicated she sought full time work and could not do heavy lifting.4 Plaintiff Charboneau indicated on her application that she sought any full-time position. Plaintiff Martinez stated in her application that she wanted a full-time job as a meat wrapper or in one of the retail stores, that she preferred the day shift and that she could not do heavy lifting.5 Plaintiff Lobato indicated she sought a full-time job as a meat wrapper or in the deli or bakery department of a retail store. Plaintiff Duran said she wanted a full-time position as a meat wrapper, bakery clerk, deli clerk or warehouse sanitation person.6 Plaintiff Medina expressed an interest in any full-time position.7 Plaintiff Gutierrez indicated she would prefer a meat wrapper position in a retail store, but that she would take any full-time or part-time position.8

Based upon the application form filled out by each plaintiff, as well as the interview, Simmons was supposed to place plaintiffs in currently available positions or positions as they became available.9 Under normal Safeway procedures, each application was to be kept active until the applicant was placed. There was testimony that, as part of these procedures, the Central Employment Office interviewer (Simmons) was to mark the interview applications with the phrase "hold for rehire." This designation indicated that the application should be sent to the employment relations manager so the particular individual could be rehired. If the employment relations manager cleared the applicant for rehire, he would complete a new form called a "rehire approval form."

There was some testimony that failure to follow these procedures meant that the application would be placed in an "inactive" file. R. Vol. III at 415-16. However, Simmons also testified that the rehire approval "was part of the process" but was not essential to placement in a job. R. Vol. IV at 488. Further, Simmons testified that the pre-pakt employee applications "were placed into an active file, and they were given consideration as ... openings became available...." Id. at 481.

Plaintiffs allege that Safeway followed all of these procedures for male pre-pakt employees, but failed to follow them for female pre-pakt employees. The applications of several of the male pre-pakt employees were indeed marked "hold for rehire."10 None of the plaintiffs' applications were so marked.

Plaintiffs also claim that Simmons failed to explain to them what retail jobs were available, despite their expressed interest in such jobs, and the fact that at least some of the part-time retail jobs would eventually become full-time with higher pay.11 Three of the four male pre-pakt employees who sought new jobs through the Central Employment Office were placed in such jobs. None of the female pre-pakt employees except Charboneau were so placed.

With respect to the informal word-of-mouth policy for placing pre-pakt employees, there was testimony that rumors of the closure of the pre-pakt plant circulated for some time prior to its actual closure. Both before and after the closure, employees approached Scherzer concerning job opportunities in the distribution center. No one was specifically told to see Scherzer. Those who did so sought out Scherzer for placement assistance on their own initiative.12 2] At least five of the male employees approached Scherzer about new jobs; plaintiff Charboneau was the only female who approached Scherzer. Scherzer helped the five male employees find new jobs in the distribution center, and he helped one female, Alberta Silva (the only person whom Scherzer approached directly), find a janitorial job in the center.13 He helped place plaintiff Charboneau in an order selector position in the distribution center's variety warehouse, which she held for only a short time because she was unable to meet the production demands of the job. She then held two successive temporary janitorial positions, followed by a three month lay-off. She was then placed in a permanent janitorial position, which she held as of the time of trial.

As it turned out, the only available entry level jobs14 in the distribution center (order selector positions, janitorial positions, parts clerk positions, and pallet repair jobs) required repeated lifting of more than 40 pounds, and the only available entry level jobs in Safeway's retail stores (i.e. bakery clerk, deli clerk, meat wrapper, net and cleaver, and courtesy clerk) were almost exclusively part-time positions, some of which would, however, become full-time jobs with higher pay within three to five years. Thus, although plaintiffs hotly dispute this, there was evidence that at the time the pre-pakt plant closed, there were no jobs available for which plaintiffs were eligible and which suited plaintiffs' (except for Charboneau and Gutierrez) expressed preference for full-time work which did not require heavy lifting.

Of eight male pre-pakt employees who sought continued employment with Safeway, seven were placed in new jobs either through Scherzer or the Central Employment Office within the 30 day period following the pre-pakt plant's closure. Additionally, four other male employees with rights under the existing union contract were placed in new jobs. Of the thirteen female pre-pakt employees who sought continued employment, two were placed in new jobs within the thirty day period, and one of the two had rights under the existing union contract.

The district court made the following findings regarding Safeway's overall hiring in the Denver area in 1984:

During 1984, Safeway hired 1395 new employees of whom 53 were hired by the warehouse and 1309 were hired by the retail stores. In the retail stores during that period, Safeway hired 54 butcher block or net & cleaver employees, 77 deli clerks, 30 meat wrappers and 33 bakery workers. In addition, Safeway in 1984 hired 894 courtesy clerks and 110 all purpose food clerks.

Nevertheless, during the period June 22, 1984, through December 31, 1984, Safeway filled no full-time bakery clerk, deli clerk or meat wrapper positions in its Metro Denver retail stores. One full-time net & cleaver clerk was hired during this period, a female named Danielle Ware. Sixty-two percent of all employees hired by Safeway during that time frame were women. In 1984, 22 of 33 (67%) of the retail bakery employees were women; 75 of 77 (97%) of the deli clerks were women; 21 of 30 (70%) of the meat wrappers hired were women; and 45 of 54 (83%) of the net & cleaver employees hired were women. All of these positions except one net & cleaver job were part-time.

Memorandum Findings of Fact, Conclusions of Law and Order at 6-7. These findings are not disputed.

In August, 1984, several of the plaintiffs filed discrimination charges with the Colorado Civil Rights Commission. The Commission subsequently found probable cause to believe that the charge of discrimination was true with respect to seven of the eight plaintiffs. There was no finding one way or the other with respect to Gutierrez. Later that month, two of Safeway's affirmative action representatives met with plaintiffs Medina, Lobato, Duran and Gutierrez to discuss the charges and to inquire of plaintiffs what jobs they wanted. The same plaintiffs met again with two Safeway representatives about one week later.

In October, 1984, some of the plaintiffs met with Safeway affirmative action representatives and Norm Scherzer. Plaintiffs expressed an interest in janitorial positions at the distribution center.15 Plaintiffs Medina, Lobato and Duran testified at trial that they were told that they would be considered for such janitorial positions only if they agreed to sign written releases of their discrimination claims. Scherzer and one of the affirmative action representatives denied that any such condition was imposed, and Safeway continues to deny that the job offers were in any way conditioned on dropping plaintiffs' discrimination charges.

On October 29, 1984, one of the affirmative action representatives sent a letter to plaintiff Gutierrez asking her to call Scherzer if she was interested in an order selector position which had become available. She apparently never responded. She testified that she did not respond because she had the impression she would have to drop her discrimination case. R. Vol. II at 106-07.16

Safeway finally offered full-time janitorial jobs to plaintiffs Medina, Duran, Rojo, Ortega and Gutierrez in July and August, 1985. Apparently, plaintiffs Martinez and Lobato were offered full-time jobs in April, 1986. While plaintiffs insist the offers were made in order to induce settlement of the discrimination charges, the district court specifically found "there was no credible or persuasive evidence that the offers of full-time janitorial positions made to several of the plaintiffs in July 1985, were conditioned on those plaintiffs dropping their discrimination charges." Memorandum Findings of Fact, Conclusions of Law and Order at 8. Plaintiffs declined the job offers.17

The district court found no disparate treatment of plaintiffs. After concluding that plaintiffs had established a prima facie case of discrimination, the court held that Safeway had articulated legitimate, nondiscriminatory reasons for plaintiffs' treatment, and that plaintiffs failed to show that those reasons were a pretext. In sum, "[w]hile Safeway certainly could have exerted greater efforts to help the plaintiffs find new jobs, I am unable to conclude that its failure to do so constituted intentional discrimination." Id. at 28.

With respect to plaintiffs' disparate impact claim, the court found that plaintiffs had met their burden of demonstrating a significant disparate impact on a protected class. It then held that:

Safeway had a practice of failing adequately to counsel its laid-off employees concerning placement in suitable jobs. Unlike their female counterparts, the male Pre-Pakt employees were not disadvantaged by this practice because they either: (1) did not limit themselves to unavailable jobs as most of the plaintiffs did; and/or (2) profited from an informal network of male friends and contacts that enabled them to learn from Norm Scherzer about available jobs before the announcement of the plant's closing.

Id. at 33. It further found "that a causal relationship existed between the practice of inadequate counseling and the discriminatory result." Id. Thus, plaintiffs met their burden of establishing a prima facie case of disparate impact discrimination. Having concluded that Safeway failed to articulate any "business necessity to justify its practice of inadequately counseling the laid-off Pre-Pakt employees," id. at 34, the district court found for plaintiffs on their disparate impact claim.

Plaintiffs were ordered reinstated as of August 7, 1988 in part-time meat wrapper, net and cleaver, and deli clerk positions. The court also assigned seniority dates as if plaintiffs had been rehired in July, 1984 (within 30 days of the plant closure).

Title VII prohibits discrimination by employers on the basis of race, color, religion, sex, or national origin. 42 U.S.C. §§ 2000e-1 to 2000e-17. This case involves allegations both of disparate treatment and disparate impact on the basis of sex. Disparate treatment occurs where "the employer simply treats some people less favorably than others" because of their sex or other protected status. Drake v. City of Fort Collins, 927 F.2d 1156, 1159 (10th Cir.1991) (quoting Coe v. Yellow Freight Sys., Inc., 646 F.2d 444, 448 (10th Cir.1981) (quoting International Bhd. of Teamsters v. United States, 431 U.S. 324, 335, 97 S. Ct. 1843, 1854-55, 52 L. Ed. 2d 396 (1977))). "A claim of disparate impact exists when 'employment practices that are basically neutral in their treatment of different groups in fact fall more harshly on one group than another....' " Id.; see also Williams v. Colorado Springs, Colo. School Dist. No. 11, 641 F.2d 835, 839 (10th Cir.1981). We address each in turn.

When alleging disparate treatment on the basis of sex, the plaintiff must prove by a preponderance of the evidence that the defendant had a discriminatory motive or intent. Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 986, 108 S. Ct. 2777, 2784, 101 L. Ed. 2d 827 (1988). This may be done either by direct proof of discriminatory intent, or, more commonly, by the "series of shifting evidentiary burdens that are 'intended progressively to sharpen the inquiry into the elusive factual question of intentional discrimination.' " Id. (quoting Texas Dep't of Community Affairs v. Burdine, 450 U.S. 248, 255 n. 8, 101 S. Ct. 1089, 1094 n. 8, 67 L. Ed. 2d 207 (1981)); see also McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S. Ct. 1817, 36 L. Ed. 2d 668 (1973); Drake v. City of Fort Collins, 927 F.2d at 1159-61; Cunico v. Pueblo School Dist. No. 60, 917 F.2d 431, 436 (10th Cir.1990); Ramsey v. City and County of Denver, 907 F.2d 1004, 1007-08 (10th Cir.1990).

Thus, pursuant to the shifting burden of proof scheme of McDonnell Douglas and subsequent cases, plaintiffs must first establish a prima facie case of discrimination. Drake v. City of Fort Collins, 927 F.2d at 1159; see also Watson v. Fort Worth Bank & Trust, 487 U.S. at 986, 108 S.Ct. at 2784; Texas Dep't of Community Affairs v. Burdine, 450 U.S. at 248, 252-53 n. 6, 101 S. Ct. at 1089, 1093-94 n. 6 (1981); McDonnell Douglas Corp. v. Green, 411 U.S. at 802, 93 S.Ct. at 1824; McAlester v. United Air Lines, Inc., 851 F.2d 1249, 1260 (10th Cir.1988).

Once plaintiffs establish a prima facie case of discrimination, "the burden of production shifts to defendants to rebut the presumption of discrimination." Drake v. City of Fort Collins, 927 F.2d at 1160 (citing Carey v. United States Postal Serv., 812 F.2d 621, 624 (10th Cir.1987)). Safeway can rebut that presumption by producing "some evidence that it had legitimate, nondiscriminatory reasons for the decision." Watson v. Fort Worth Bank & Trust, 487 U.S. at 986, 108 S.Ct. at 2784; see also Drake v. City of Fort Collins, 927 F.2d at 1160; McAlester v. United Air Lines, Inc., 851 F.2d at 1260. Its articulation of those reasons must be "clear and reasonably specific." Drake v. City of Fort Collins, 927 F.2d at 1160; see also Texas Dep't of Community Affairs v. Burdine, 450 U.S. at 258, 101 S.Ct. at 1096.

The burden of proof shifts once again if the defendant succeeds in rebutting the presumption of discrimination raised by the plaintiff's prima facie case. Now, "the plaintiff must prove by a preponderance of all the evidence in the case that the legitimate reasons offered by the defendant were a pretext for discrimination." Watson v. Fort Worth Bank & Trust, 487 U.S. at 986, 108 S.Ct. at 2784. Plaintiffs in this case may do this either directly by showing that Safeway acted with actual discriminatory motives or indirectly by showing that its proffered reasons for its treatment of plaintiffs are unworthy of belief. Drake v. City of Fort Collins, 927 F.2d at 1160.

The Supreme Court has cautioned, however, that this shifting burden of proof scheme is only intended to assist in marshalling and presenting relevant evidence. "The ultimate burden of persuading the trier of fact that the defendant intentionally discriminated against the plaintiff remains at all times with the plaintiff." Watson v. Fort Worth Bank & Trust, 487 U.S. at 986, 108 S.Ct. at 2784 (quoting Texas Dep't of Community Affairs v. Burdine, 450 U.S. at 253, 101 S.Ct. at 1093-94). Further, the ultimate question in a Title VII disparate treatment case is whether the defendant intentionally discriminated against the plaintiff. United States Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 714-15, 103 S. Ct. 1478, 1481-82, 75 L. Ed. 2d 403 (1983). Thus, when such a case is fully tried, as here, we need only consider that ultimate question--whether plaintiffs proved that the defendant intentionally discriminated against them. The subsidiary steps in the McDonnell Douglas proof scheme become irrelevant. Id. at 715, 103 S.Ct. at 1481-82 ("Where the defendant has done everything that would be required of him if the plaintiff had properly made out a prima facie case, whether the plaintiff really did so is no longer relevant."); McAlester v. United Air Lines, Inc., 851 F.2d at 1260 ("[T]his court need not review whether McAlester made a prima facie case where the district court did not dismiss McAlester's claim on United's motion at the conclusion of McAlester's case."); Pitre v. Western Elec. Co., 843 F.2d 1262, 1266 (10th Cir.1988) ("Once the district court makes the ultimate determination as to whether the employer violated [Title VII], its legal conclusions in regard to the first two stages of evaluating the evidence become irrelevant on appeal."). In addressing that ultimate question, evidence of the disparate impact of an employer's practices may be relevant. Scales v. J.C. Bradford and Co., 925 F.2d 901, 906 (6th Cir.1991); Green v. USX Corp., 896 F.2d 801, 807 (3d Cir.), cert. denied, --- U.S. ----, 111 S. Ct. 53, 112 L. Ed. 2d 29 (1990).

Finally, in reviewing the district court's conclusion that plaintiffs failed to show disparate treatment, we bear in mind that a finding of intentional discrimination, or a finding of no intentional discrimination, is subject to the clearly erroneous standard of review. Anderson v. City of Bessemer City, 470 U.S. 564, 573, 105 S. Ct. 1504, 1511, 84 L. Ed. 2d 518 (1985); Cunico v. Pueblo School Dist. No. 60, 917 F.2d at 436; Ramsey v. City and County of Denver, 907 F.2d at 1010; Pitre v. Western Elec. Co., 843 F.2d at 1266. Thus, we will affirm the district court's factual findings on this matter "unless our review of the entire evidence leaves us 'with the definite and firm conviction that a mistake has been committed.' " Cunico v. Pueblo School Dist. No. 60, 917 F.2d at 436 (quoting United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S. Ct. 525, 541-42, 92 L. Ed. 746 (1948)). "Where there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous." Anderson v. City of Bessemer City, 470 U.S. at 574, 105 S.Ct. at 1511.

Plaintiffs argue the district court erred in several major respects in concluding that plaintiffs failed to show intentional discrimination. First, they claim the court ignored crucial evidence of intentional discrimination in the practices of the Central Employment Office. Second, they assert the court erred in concluding that plaintiff Charboneau was not the victim of intentional discrimination in that male pre-pakt employees with less seniority than she were placed in jobs ahead of her. Finally, they argue that other findings of the court in support of its conclusion that Safeway had not intentionally discriminated against plaintiffs were clearly erroneous. We affirm.

A. Central Employment Office Procedures.

Plaintiffs assert the district court erred: (1) in failing to infer intentional discrimination from the fact that plaintiffs' employment applications were treated differently from the male pre-pakt employees' applications in that the males' applications were marked "hold for rehire" and the plaintiffs' were not; (2) in failing to infer intentional discrimination from the statistical disparity between the number of male pre-pakt employees placed by the Central Employment Office and the number of female employees so placed; and, (3) in failing to address Safeway's alleged failure to explain its different treatment of plaintiffs' applications and in failing to find that Safeway's stated reasons for not rehiring plaintiffs were false.

We cannot conclude that the district court's apparent refusal to find intentional discrimination in the fact that plaintiffs' applications were not coded "hold for rehire" is clearly erroneous. As we have indicated, the evidence was conflicting as to whether this difference in treatment had any effect on plaintiffs' applications.18 "Where there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous ... even when the district court's findings do not rest on credibility determinations, but are based instead on physical or documentary evidence or inferences from other facts." Anderson v. City of Bessemer City, 470 U.S. at 574, 105 S.Ct. at 1511-12.

Similarly, we see no clear error in the district court's failure to infer intentional discrimination from the statistical disparity between the number of female pre-pakt employees placed in new jobs by the Central Employment Office as contrasted with the number of males so placed. As Safeway points out, while the district court did not specifically address the numbers of male and female pre-pakt employees placed through the Central Employment Office, it did specifically note the total placement numbers. Indeed, in holding that plaintiffs had established a prima facie case of discrimination, the court stated that

seven out of the eight male employees who did not have seniority job rights under existing union contracts, and who sought new jobs with Safeway, received jobs within the 30-day period following the Pre-Pakt closure.... Four additional male employees were placed under existing union contracts. Yet only two out of the thirteen female employees who sought new jobs with Safeway received full-time job offers in 1984.

Memorandum Findings of Fact, Conclusions of Law and Order at 21. While it is true that "discriminatory intent or motive ... can be inferred from the mere fact of differences in treatment," McAlester v. United Air Lines, Inc., 851 F.2d at 1260, in the circumstances of this case, the district court's refusal to draw that inference is not clearly erroneous.

We also find no error in the district court's refusal to find that Safeway's articulated reasons for its treatment of plaintiffs were false or pretextual. As the district court noted, Safeway's stated reasons for not placing plaintiffs were that plaintiffs were offered jobs, but rejected them, and that plaintiffs' self-imposed restrictions on the types of jobs they would take removed them from consideration for the available jobs.

Plaintiffs do not dispute that they were only eligible for certain entry level jobs--order selector, janitor, pallet repair, and part-time retail clerk positions. The order selector, janitor, and pallet repair jobs all required heavy lifting. Thus, plaintiffs Ortega, Rojo, Martinez, Duran, and Medina removed themselves from eligibility for those jobs because of their expressed inability or unwillingness to do heavy lifting. Plaintiff Lobato removed herself from eligibility for those jobs because of her stated interest in a retail store job. Those were also the jobs in which Scherzer placed male employees who approached him. Thus, those plaintiffs also had removed themselves from eligibility for the kind of assistance Scherzer was providing.

All of these plaintiffs also removed themselves from consideration for the part-time retail jobs because of their expressed interest in full-time work.19 Plaintiffs testified that they would have taken part-time retail positions, had they been told that those positions could become full-time within a few years. Nonetheless, plaintiffs presented no evidence that Safeway's decision to take them at their word in their preference for full-time work evidenced intentional discrimination.20

The situation is slightly different with respect to plaintiff Gutierrez. Gutierrez stated that she would prefer a meat wrapper position in a retail store, but that she would take any full-time or part-time position. She did, however, call Simmons in early July and tell him she wanted to be considered for a meat wrapper position, which led Simmons to believe that was the only position she wanted. As the district court found, Safeway filled no full-time meat wrapper position between June 22 and December 31, 1984. There was evidence that a part-time meat wrapper position was filled by a female outside hire with no previous grocery store experience on July 7, 1984--i.e., 3 days after Gutierrez had called to express an interest in a meat wrapper position. Safeway has not explained why Gutierrez was not offered that meat wrapper position, although the interview evaluations of Gutierrez and the woman who received the position are slightly different.21 The evidence also established, as the district court noted, that 70% of the meat wrappers hired in 1984 were women.

Additionally, in concluding that Gutierrez was not the victim of intentional discrimination, the district court placed some reliance on the fact that she was offered two positions--an order selector position in October, 1984 and a janitorial position in July, 1985--both of which she rejected.22

We agree with the district court that Gutierrez presents the strongest prima facie case of discrimination. It is unclear why she was not offered the part-time meat wrapper position filled on July 7. Nonetheless, the vast majority of meat wrappers hired during the relevant time period were women. "Proof that [a] work force was racially balanced or that it contained a disproportionately high percentage of minority employees is not wholly irrelevant on the issue of intent when that issue is yet to be decided." Furnco Constr. Corp. v. Waters, 438 U.S. 567, 580, 98 S. Ct. 2943, 2951, 57 L. Ed. 2d 957 (1978). While we recognize that Safeway's placement of another female in the particular meat wrapper position, combined with the fact that 70% of the meat wrappers hired were women, does not by itself exclude a finding of intentional discrimination on the basis of sex against Gutierrez, we do not believe that the circumstances of this case render the district court's finding of no intentional discrimination against Gutierrez clearly erroneous.

In sum, we affirm the district court's conclusion that "[w]hile Safeway certainly could have exerted greater efforts to help the plaintiffs find new jobs, I am unable to conclude that its failure to do so constituted intentional discrimination." Memorandum Findings of Fact, Conclusions of Law and Order at 28.

B. Treatment of Charboneau.

Charboneau was placed by Norm Scherzer in a full-time order selector position on July 19, 1984. She argues she was the victim of intentional discrimination in placement in her job, in particular job assignments, and in "bumping" rights.23 We affirm the district court's conclusion that she was not.

At the time the pre-pakt plant was closed, Charboneau had twenty years of seniority. She was the only female pre-pakt employee to approach Scherzer about a new job. She claims that she was purposefully placed at the "heavy end" of one of the most physically demanding order selector positions (order selector in the variety warehouse), and was therefore predictably unable to meet the production demands of the job.24 Scherzer testified that at the time Charboneau was placed in the variety warehouse, assignments to different sections of the warehouse were made by bid based on seniority. As a beginning employee in that warehouse, she was assigned to the section where the heaviest items (and therefore the heaviest lifting) were located. R. Vol. V at 649.

Simmons testified that the variety warehouse was, generally speaking, one of the less physically demanding positions in the distribution center. R. Vol. IV at 462. Scherzer's testimony was somewhat confusing on that point. Initially, he agreed that "at least for a new order selector, the variety warehouse would have been one of the more difficult order selector assignments." R. Vol. V at 684. Yet he then testified:

A. Variety warehouse is no harder than those other warehouses. As a matter of fact, it's lighter than the produce warehouse, it's more pleasant to work in that the frozen food warehouse, and it is no more difficult than the grocery warehouse.

. . . . .

A. What I'm trying to say is, variety warehouse selector, in the beginning section, lifts no more weight in an eight-hour day, if as much, as a grocery warehouseman or a produce warehouse order selector, that's what--the point I'm trying to make. The end section that Ms. Charboneau is in was the heaviest section in the variety warehouse but that section being no heavier than the produce warehouse order selector, the grocery warehouse order selector, that's the point I'm trying to make.

Id. at 685-86. From all the evidence presented, we cannot say that the district court clearly erred in concluding that Charboneau was not the victim of intentional discrimination in her initial placement as an order selector in the variety warehouse or in the particular job assignments she received within that warehouse.

Charboneau also claims she was intentionally discriminated against in that Scherzer declined to bump less senior male employees from janitorial jobs in favor of Charboneau, although he had bumped a less senior male in favor of a more senior male employee. In particular, she charges Scherzer could have bumped two males, Ray Billups and Joe Martinez, out of janitorial jobs they were placed in on June 24 and August 6, respectively, and replaced them with the more senior Charboneau. Instead, when Charboneau was terminated from the order selector position on August 16, she was placed in two successive temporary janitorial positions. She was then laid off from October 22 until January, at which time she received the permanent janitorial position she held at trial.

Once again, considering all of the evidence, we cannot conclude that the district court's finding that Charboneau was not the victim of intentional discrimination in bumping rights was clearly erroneous. Plaintiffs argue that "Safeway's records plainly revealed that Scherzer utilized a bumping procedure to move Don Svaldi into the position of another employee, Bob Lucero, on the basis of seniority." Opening Brief of Plaintiffs-Cross Appellants at 28. In fact, while an employee form for Lucero contains the notation "was bumped by senior employee," Exh. 38 at 2, Scherzer testified that he did not bump Lucero simply because of seniority. R. Vol. III at 359. Further, during the 30 day period in which she could have bumped Billups, Charboneau was still in the order selector position in which Scherzer thought she was interested and in which he thought she would succeed. Viewing the entire record, it is simply not clear that Safeway intentionally gave males the benefit of the bumping procedure, but did not do so for females.25 There was therefore no clear error in the district court's finding.

C. Other Findings.

We reject plaintiffs' argument that other findings of the district court are clearly erroneous and require reversal of its judgment on disparate treatment. Plaintiffs challenge the court's finding that Safeway no longer segregated male and female pre-pakt employees such that females were packers and males were warehousepersons. While it is true that at the time the plant closed in 1984 packers were exclusively female and warehousepersons were exclusively male, there was also testimony that Safeway did not exclude either sex from performing either job and had in fact encouraged females to become warehousepersons. There was also testimony that the sex segregated lunch rooms resulted from the fact that the lunch rooms also functioned as locker rooms. We cannot say that the district court's conclusion that Safeway no longer engaged in sex segregation in its pre-pakt plant was clearly erroneous.

Plaintiffs next challenge the district court's finding that minorities were represented in both management and non-management positions at Safeway. They argue this finding was erroneous in that it should have been based on the promotion practices at the pre-pakt plant only, not Safeway overall, and it ignored the fact that only one female had been promoted to a supervisory position in the pre-pakt plant. That finding was not critical to the district court's conclusion and would not affect our holding in this case. We therefore need not consider it.

Finally, they argue the court erred in basing its finding of no intentional discrimination in part on plaintiffs' rejection of unconditional job offers and in part on the fact that the majority of employees hired into the retail clerk positions sought by plaintiffs were female. We have already noted that the percentage of females holding the jobs plaintiffs claim they should have gotten may be relevant to the existence of discriminatory intent. The fact that plaintiffs rejected later job offers is not, in our view, critical to the district court's finding of no intentional discrimination. Therefore, any possible error in its consideration of the offers is immaterial.

A claim of disparate impact, unlike a claim of disparate treatment, does not require a finding of intentional discrimination. Indeed, "the necessary premise of the disparate impact approach is that some employment practices, adopted without a deliberately discriminatory motive, may in operation be functionally equivalent to intentional discrimination." Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 987, 108 S. Ct. 2777, 2785, 101 L. Ed. 2d 827 (1988); see also Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 109 S. Ct. 2115, 104 L. Ed. 2d 733 (1989); Drake v. City of Fort Collins, 927 F.2d 1156, 1161 (10th Cir.1991).

To establish a prima facie case of disparate impact discrimination, plaintiffs must show that a specific identifiable employment practice or policy caused a significant disparate impact on a protected group. Wards Cove Packing Co. v. Atonio, 490 U.S. at 656, 109 S.Ct. at 2124; Drake v. City of Fort Collins, 927 F.2d at 1161.26 Plaintiffs may rely on statistics to show that the challenged practice or policy has the requisite disparate impact. Wards Cove Packing Co. v. Atonio, 490 U.S. at 650, 109 S.Ct. at 2121; Hazelwood School Dist. v. United States, 433 U.S. 299, 307-08, 97 S. Ct. 2736, 2741-42, 53 L. Ed. 2d 768 (1977); Drake v. City of Fort Collins, 927 F.2d at 1161; see Green v. USX Corp., 896 F.2d 801, 804-05 (3d Cir.), cert. denied, --- U.S. ----, 111 S. Ct. 53, 112 L. Ed. 2d 29 (1990); Allen v. Seidman, 881 F.2d 375 (7th Cir.1989); Hawkins v. Bounds, 752 F.2d 500, 503 (10th Cir.1985). However, as we will explore more fully below, any statistical analysis must involve the appropriate comparables, see Wards Cove Packing Co. v. Atonio, 490 U.S. at 650-55, 109 S.Ct. at 2121-24, and must "cross a threshold of reliability before it can establish even a prima facie case of disparate impact." Allen v. Seidman, 881 F.2d at 378; see also Shutt v. Sandoz Crop Protection Corp., 934 F.2d 186 (9th Cir.1991).27 Additionally, an employer's use of subjective criteria in making employment decisions is susceptible to challenge under disparate impact principles. Wards Cove Packing Co. v. Atonio, 490 U.S. at 648, 109 S.Ct. at 2120; Watson v. Fort Worth Bank & Trust, 487 U.S. at 989-91, 108 S.Ct. at 2786-87.

Once plaintiffs establish a prima facie case of disparate impact based on sex, the burden shifts to the employer, Safeway, to produce evidence of the "business justification" for the challenged practice. Wards Cove Packing Co. v. Atonio, 490 U.S. at 658, 109 S.Ct. at 2125. As the Supreme Court made clear in Wards Cove, the burden of persuasion remains with the plaintiff. The employer's burden at this stage is to "produc[e] evidence of a business justification for his employment practice.... 'The ultimate burden of proving that discrimination against a protected group has been caused by a specific employment practice remains with the plaintiff at all times.' " Wards Cove Packing Co. v. Atonio, 490 U.S. at 659, 109 S.Ct. at 2126 (quoting Watson v. Fort Worth Bank & Trust, 487 U.S. at 997, 108 S.Ct. at 2790 (O'Connor, J.)) (emphasis in original).28

Furthermore, in producing evidence of business justification, the "dispositive issue is whether a challenged practice serves, in a significant way, the legitimate employment goals of the employer." Wards Cove Packing Co. v. Atonio, 490 U.S. at 659, 109 S.Ct. at 2125-26. The challenged practice or policy need not be " 'essential' or 'indispensable' to the employer's business for it to pass muster...." Id. Rather, the "touchstone of this inquiry is a reasoned review of the employer's justification for his use of the challenged practice." Id.

If Safeway meets its burden of producing evidence of the business necessity or business justification for the challenged practice, plaintiffs may still prevail if they can "persuade the factfinder that 'other tests or selection devices, without a similarly undesirable [discriminatory] effect, would also serve the employer's legitimate [hiring] interest[s].' " Id. at 660, 109 S.Ct. at 2126 (quoting Albemarle Paper Co. v. Moody, 422 U.S. 405, 425 (1975)). Further, those alternative practices "must be equally effective ... in achieving [the employer's] legitimate employment goals." Id. 490 U.S. at 661, 109 S. Ct. at 2127. This may be a difficult burden for plaintiffs to meet, however: "the judiciary should proceed with care before mandating that an employer must adopt a plaintiff's alternative selection or hiring practice in response to a Title VII suit." Id.

Applying those principles to this case, we reverse the district court's conclusion that plaintiffs proved disparate impact based on sex. The district court held that plaintiffs had "met their burden of demonstrating a significant disparate impact on a protected class." Memorandum Findings of Fact, Conclusions of Law and Order at 29. The court based this finding on the following facts:

88% of the male Pre-Pakt plant employees who did not have job seniority rights under existing union contracts, and who sought new jobs with Safeway, received jobs within the 30-day period following the Pre-Pakt plant's closure. In contrast, only 15% of the female Pre-Pakt plant employees who did not have job seniority rights under the existing union contracts, and who sought new jobs with Safeway, received full-time job offers in 1984.

Id. We hold that, under Wards Cove, that was not the proper comparison.29

In Wards Cove, "the Supreme Court emphasized the importance of identifying the proper groups for comparison in a disparate impact analysis." Abbott v. Federal Forge, Inc., 912 F.2d 867, 873 (6th Cir.1990). In Wards Cove itself, "the proper comparison [was] between the racial composition of [the at-issue jobs] and the racial composition of the qualified ... population in the relevant labor market." 490 U.S. at 650, 109 S. Ct. at 2121 (quoting Hazelwood School Dist. v. United States, 433 U.S. 299, 308, 97 S. Ct. 2736, 2742, 53 L. Ed. 2d 768 (1977)). Thus, a simple comparison of the percentage of non-white workers in skilled jobs and the percentage of non-white workers in unskilled jobs was insufficient to establish a disparate impact on non-whites caused by certain employment practices.

The district court in this case did not have the benefit of the Supreme Court's decision in Wards Cove when it issued its opinion. It therefore did not, in our view, make the proper statistical analysis in determining whether plaintiffs showed a disparate impact based on sex.

We start by recognizing that it is difficult to determine exactly how Wards Cove applies to different factual patterns. In particular, here we must determine the relevant comparison groups in a rehire situation, and we must identify the "at-issue" jobs. Safeway argues that the "at-issue" jobs are the retail store jobs plaintiffs now say they would have taken had they been properly counseled about the prospects of those jobs becoming full-time. The undisputed evidence shows that the vast majority of people hired into those jobs in 1984 were women.

In our view, those jobs could be the at-issue jobs for disparate impact analysis if it is clear that rehires seeking to be placed in those jobs are treated no differently from new hires seeking those jobs. Thus, if rehires seeking a retail store job go through the identical application process as new hires seeking such jobs, the relevant comparison would be between the percentage of females in those jobs and the percentage of qualified female applicants, including female rehires, for those jobs. The evidence in this case shows that rehires and new hires seeking retail store jobs did in fact go through the identical process--all filled out applications and were interviewed by a Central Employment Office representative.30 If that is so, it would be difficult to find disparate impact where the vast majority of those hired into those jobs are women.

On the other hand, plaintiffs urge us to narrow our consideration to the laid-off pre-pakt employees only. They allege that the essence of this case is that laid-off men were treated more favorably in placement in jobs for which they were qualified than were the laid-off women. In this analysis, the at-issue jobs would be defined as jobs which the laid off employees sought and for which they were qualified.

Even under this analysis, however, it is error to simply compare the raw statistics showing that a higher percentage of men were rehired than women. Although, as plaintiffs continue to urge, both the male and female pre-pakt employees were unskilled workers, they were qualified for different jobs. The men were qualified for and sought out jobs requiring heavy lifting in the distribution center, both through the informal word-of-mouth procedure involving Norm Scherzer and through the more formal Central Employment Office procedures. The women, by contrast, removed themselves from consideration for those jobs because of their expressed preference for work that did not require heavy lifting.31 Thus, even if we look simply at the pre-pakt men and women who were rehired, plaintiffs have failed to prove a disparate impact based on sex because the men who were rehired were qualified for jobs for which the women were not qualified, and self-imposed limitations on the type of work they would do removed the women from consideration for the only jobs--retail store jobs--for which they were qualified.

Even were we to assume a disparate impact on women, we would reverse the district court's conclusion that a specific identifiable employment practice caused the disparate impact. The district court found, and plaintiffs assert, that the employment practice which caused the disparate impact was Safeway's poor counseling services. Safeway's counseling practices were not, however, the cause of plaintiffs' failure to be placed in the part-time retail jobs they claim they wanted. It was plaintiffs' stated preference for full-time work that caused them to not be placed in retail jobs. Further, it was their stated preference for full-time work that did not require heavy lifting that caused them to not be placed in the types of jobs in which the male pre-pakt employees were placed.

We therefore reverse the district court's conclusion that plaintiffs proved a disparate impact based upon sex.

For the reasons stated above, we affirm the order of the district court with respect to disparate treatment and reverse the order with respect to disparate impact. We therefore reverse the judgment in favor of plaintiffs.