Summary
holding in a Title VII disparate impact claim that courts should be wary of "incomplete or inapplicable analyses, simplistic percentage comparisons, and small sample sizes"
Summary of this case from Doe v. Bluecross Blueshield of Tenn., Inc.Opinion
No. 06-4454.
October 20, 2008.
On Appeal from the Southern District of Ohio.
BEFORE: GUY, BATCHELDER, and McKEAGUE, Circuit Judges.
The plaintiffs appeal the district court's directed verdict for the defendant on their disparate-impact gender-discrimination claims and its denial of their subsequent motion for a new trial. For the reasons that follow, we AFFIRM.
I.
Planes Moving Storage provides transportation and relocation services for residential, corporate, and commercial movers. In September 2001, Planes determined that a reduction in force ("RIF") would be necessary at its Cincinnati facility. Prior to the RIF, Planes's non-management workforce comprised 120 women and 86 men, split into several departments. Management comprised 18 women and 35 men. So, there were 259 total employees (53% female).
Based on Planes's assessment of economic conditions and business prospects" the same assessment that led it to implement the RIF in the first place — Planes confined the RIF to certain departments: Customer Service; Credit Collections; Operations; and Billing Rating. Of the 101 total employees in these departments, 90 were women and 11 were men, meaning that the departments were 89% female. Meanwhile, Planes excluded from the RIF its other departments: Warehouse; Movers Packers; and Drivers. These departments consisted of 30 women and 75 men (105 total), meaning that they were only 29% female. Planes delegated to the individual department manager's the decision of which employee(s) from their departments to lay off, using criteria of conduct, performance, reliability, and seniority. Ultimately, Planes laid off 12 women and one man.
Four of the laid-off employees — Linda Shollenbarger, Helena Davidson, Melanie Anderson, and Lisa Zecher — sued Planes in federal court, alleging gender discrimination on theories of disparate treatment and disparate impact. At the close of the evidence, Planes moved the court for directed verdict on the disparate impact theory. The plaintiffs had claimed that Planes's selection of only certain (predominantly female) departments for the RIF and delegation of complete discretion to the department managers to select the employees to be laid off disparately impacted female employees. The plaintiffs argued that statistics demonstrated their prima facie case of disparate impact, but the court disagreed and granted directed verdict for Planes. Meanwhile, the disparate treatment claim went to the jury, which returned a verdict for Planes. The plaintiffs appeal only the directed verdict on the disparate impact claim; they do not appeal the jury verdict.
Two of the plaintiffs had also alleged age discrimination, but the court dismissed those claims on summary judgment prior to trial. The plaintiffs have not appealed the summary judgment, and age discrimination is not at issue.
II.
"We review the grant of a directed verdict de novo, using the same test as the district court." Jones v. Fed. Fin. Reserve Corp., 144 F.3d 961, 967 (6th Cir. 1998). "[W]e must determine whether sufficient evidence was presented to raise a material issue of fact for the jury[,][and][a] directed verdict is proper only when no reasonable juror could find for the nonmoving party." Id.
To establish a prima facie case of disparate impact, a plaintiff must: (1) identify the "particular employment practice"; (2) show a disparate impact on a protected group; and (3) prove that the employment practice caused the disparity. 42 U.S.C. § 2000e-2(k)(1)(A)(i); see Meacham v. Knolls Atomic Power Lab., 554 U.S. ___, 128 S.Ct. 2395, 2405, 171 L.Ed.2d 283 (2008) ("The plaintiff is obliged to do more" than "merely alleg[e] a disparate impact, or point to a generalized policy"; the plaintiff must "isolate and identify the specific employment practices that are allegedly responsible for any observed statistical disparities." (quotation and editorial marks omitted)); Wards Cove Packing Co., Inc. v. Atonio, 490 U.S. 642, 656-57, 109 S.Ct. 2115, 104 L.Ed.2d 733 (1989) (superseded by statute on other grounds); Connecticut v. Teal, 457 U.S. 440, 446, 102 S.Ct. 2525, 73 L.Ed.2d 130 (1982); Griggs v. Duke Power Co., 401 U.S. 424, 432, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971).
Most frequently, plaintiffs show a disparate impact by the use of statistics. See Int'l Bhd, of Teamsters v. United, States, 431 U.S. 324, 339 n. 20, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977). But, courts must be careful to evaluate the proffered statistical analyses in light of the total circumstances present in a given case. Id. at 339-40, 97 S.Ct. 1843. Incomplete or inapplicable analyses, simplistic percentage comparisons, and small sample sizes produce statistical analyses with little probative value. See, e.g., New York City Transit Auth. v. Beazer, 440 U.S. 568, 582-87, 99 S.Ct. 1355, 59 L.Ed.2d 587 (1979), Mayor of Phila, v. Educ. Equal. League, 415 U.S. 605, 620-21, 94 S.Ct. 1323, 39 L.Ed.2d 630 (1974). Finally, the Supreme Court has flatly rejected "bottomline" comparisons:
[A] nondiscriminatory `bottom line' and an employer's good-faith efforts to achieve a nondiscriminatory work force, might in some cases assist an employer in rebutting the inference that particular action had been intentionally discriminatory: Proof that a work force was [] 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. But resolution of the factual question of intent is not what is at issue in this case. Rather, petitioners seek simply to justify discrimination against respondents on the basis of their favorable treatment of other members of respondents' [protected] group. Under Title VII, a[] balanced work force cannot immunize an employer from liability for specific acts of discrimination.
Teal, 457 U.S. at 454, 102 S.Ct. 2525 (citations, quotation marks, and editorial marks omitted).
Once the plaintiff has established a prima facie case, the burden shifts to the employer to articulate and demonstrate a legitimate business justification for the challenged practice. 42 U.S.C. § 2000e-2(k)(1)(A)(i). And, reciprocally, if the defendant establishes such a justification, the burden shifts back to the plaintiff to show that "other tests or selection devices, without a similarly undesirable . . . effect, would also serve the employer's legitimate [business] interest[s]." Watson v. Fort Worth Bank Trust, 487 U.S. 977, 998, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988) (quoting Albemarle Paper Co. v. Moody, 422 U.S. 405, 425, 95 S.Ct. 2362, 45 L.Ed.2d 280 (1975)); see 42 U.S.C. § 2000e-2(k)(1)(A)(ii) 2000e-2(k)(1)(C).
The plaintiffs first challenged Planes's "particular employment practice" of selecting only certain (predominantly female) departments for the RIF. In directing verdict for Planes, the district court determined that the plaintiffs' statistical analyses were insufficient to demonstrate either a disparate impact on female employees or a causal connection with the challenged employment practice. Therefore, the court held that the plaintiffs had failed to establish a prima facie case.
The plaintiffs contend that the statistics do show a disparity and we agree. At this step in the analysis — the prima facie step — Planes's reasons for selecting certain departments is immaterial; the only questions at this point are whether there was an identifiable disparity and, if so, whether the challenged employment practice (i.e., the selection of certain departments) could have caused the disparity. Based on a rudimentary statistical analysis, we answer both in the affirmative.
If Planes had randomly selected one employee for layoff from its entire non-management labor pool (i.e., all departments), it would have had a 58% chance (120/206) of selecting a woman. By targeting only certain departments, the likelihood of selecting a woman increased to 89% (90/101). More telling is that the odds of selecting 12 women from the affected departments is 23%, whereas the odds of selecting 12 women from the entire non-management labor pool is 0.1%. We find this to be a sufficient disparity to demonstrate a disparate impact from the decision.
0.23 = 90/101 x 89/100 x 88/99 x 87/98 x 86/97 x 85/96 x 84/95 x 83/94 x 82/93 x 81/92 x 80/91 x 79/90
0.1 = 120/206 x 119/205 x 118/204 x 117/203 x 116/202 x 115/201 x 114/200 x 113/199 x 112/198 x 111/197 x 110/196 x 109/195
Thus, the burden shifts and we must consider whether Planes set forth a legitimate business justification. Planes explained that its declining business necessitated the RIF and that some departments were affected more that others; specifically, those employees who dealt most directly with customers were the most affected. In addition, the predominantly male, unaffected departments were staffed largely with seasonal workers (typically high school and college students) who had already left at the end of the peak summer season. And, there was no decline in the business being done by the warehouse. We conclude that the challenged employment practice of subjecting only certain departments to the RIF had a legitimate business justification.
Planes explained that some predominantly female departments were not included in the RIF (i.e., Accounting, Administration, Marketing, and Claims).
Because Planes clearly met its burden of showing a legitimate business justification, the burden shifts back to the plaintiffs to show that "other tests or selection devices, without a similarly undesirable . . . effect, would also serve the employer's legitimate [business] interest." Watson, 487 U.S. at 998, 108 S.Ct. 2777; 42 U.S.C. § 2000e-2(k)(1)(A)(ii) 2000e-2(k)(1)(C). The plaintiffs attempt to rely on the magistrate judge's statement that "there wasn't any exploration of alternatives to these layoffs at all." But, this is a misunderstanding of the standard and, hence, completely irrelevant. The plaintiffs were obligated to prove equally effective alternatives and" although they offer alternatives to a RIF in general — they offer no alternative to subjecting only the particular, selected departments to the RIF. The purpose of this step is not to second guess the employer's business decisions, it is to show — by pointing to obviously ignored alternatives — that the "particular employment practice" was actually pretext for discrimination. See Albemarle, 422 U.S. at 425, 95 S.Ct. 2362 ("Such a showing would be evidence that the employer was using its tests merely as a `pretext' for discrimination."). These plaintiffs have not proven such pretext.
The plaintiffs also challenged Planes's employment practice of "permitting the department managers to determine which employees would be selected for the RIF . . . without any formal review." The district court determined that the plaintiffs' statistical analyses were insufficient to demonstrate a disparate impact on female employees or a causal connection with the employment practice. Therefore, the court held that the plaintiffs had failed to establish a prima facie case.
The same statistics that establish the prima facie case in the aforementioned claim disprove it in this claim. That is, accepting that Planes's targeting of certain departments was legitimate, Planes had an 89% random chance (90/101) of selecting a woman for the layoff (and a 23% chance of selecting 12 women) because the selected departments were 89% female. The RIF was 92% female (12/13), which is perfectly consistent with a random selection from an 89% pool. Statistically, 12 is the most likely number of women from this pool, as anything less would diverge from the basic statistical probability. Therefore, this statistical result does not demonstrate disparity, much less a significant disparity that can be connected causally to the challenged employment action.
Nevertheless, even if the plaintiffs could prove their prima facie case, Planes offered legitimate business justifications for selecting each of the individual plaintiffs. Ms. Anderson had proven herself a difficult employee, resistant to change. Ms. Shollenbarger had low seniority in her department, was struggling to learn the job, and was a below-average performer. Ms. Zecher had attendance problems and had previously been suspended for a drug incident. And Ms. Davidson was rude and had received low quality scores in customer ratings. The plaintiffs did not rebut these assessments and did not offer anything to demonstrate that this was mere pretext.
III.
For the foregoing reasons, we AFFIRM the district court's judgment.