Christy Palmer, Vartan Piroumian, Edward Cox, and Jean-Claude Franchitti v. Cognizant Technology Solutions Corporation and Cognizant Technology Solutions U.S. Corporation

Opinion

CHRISTY PALMER, VARTAN ) Case No. CV 17-6848-DMG (Ex) ) PIROUMIAN, EDWARD COX, and ) JEAN-CLAUDE FRANCHITTI, ) FINDINGS OF FACT AND Plaintiffs, ) ) CONCLUSIONS OF LAW RE v. ) PLAINTIFFS’ DISPARATE IMPACT ) SOLUTIONS CORPORATION and ) COGNIZANT TECHNOLOGY ) ) SOLUTIONS U.S. CORPORATION, ) Defendants. This matter is before the Court following an October 4, 2024 jury verdict in favor of Plaintiffs Vartan Piroumian and Brian Cox,1 and against Defendants Cognizant Technology Solutions Corporation and Cognizant Technology Solutions U.S. Corporation (collectively, “Cognizant”) on Plaintiffs’ claim that Cognizant engaged in a pattern or practice of intentional discrimination against non-South Asian and non-Indian employees. [Doc. # 666 (“Verdict”).] Plaintiffs brought both disparate treatment and disparate impact claims. The legal claim for disparate treatment was tried to the jury from September 24, 2024 to October 4, 2024, and the equitable claim for disparate impact is the subject of this Order. I. Plaintiffs Brian Cox and Vartan Piroumian,2 representing themselves and over 2,000 former employees of Cognizant, initiated this class action lawsuit in 2017, alleging that Cognizant engaged in a pattern or practice of discriminatory employment practices that favored employees of South Asian race and Indian national origin, in violation of 42 U.S.C. section 1981 (“Section 1981”) and Title VII of the Civil Rights Act of 1964 (“Title VII”). The Section 1981 Class in this action includes “[a]ll individuals who are not of South Asian race or Indian national origin and who, between September 18, 2013 to the date of class certification [October 27, 2022], were terminated from the bench while employed within a Cognizant Class Band in the U.S., excluding any individuals bound by an agreement to arbitrate termination claims with Cognizant.” Class Cert. & MSJ Ord. at 69 [Doc. # 384].3 There is also a Title VII Subclass, which includes “[a]ll

1 Brian Cox was not originally a party to this action, but was substituted as a Plaintiff following the deaths of his father, Edward Cox, in March 2021, and his stepmother, Ann Cox, in May 2022. [Doc. ## 229, 265.] 2 Christy Palmer and Jean-Claude Franchitti also are named Plaintiffs in this action, but they do not represent the Class.

3 Page citations herein refer to the page numbers inserted by the CM/ECF system. individuals who are not of South Asian race or Indian national origin and who, between December 15, 2016 to the date of class certification [October 27, 2022], were terminated from the bench while employed within a Cognizant Class Band in the U.S., excluding any individuals bound by an agreement to arbitrate termination claims with Cognizant.” Id. The Court bifurcated the trial in this action into two phases—Phase I and Phase II—in accordance with the framework set forth in Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 339 (1977). Final Pretrial Conference Order (“FPTCO”) at 2 [Doc. # 638]. Following the nine-day Phase I trial on Plaintiffs’ disparate treatment theory, the jury returned a unanimous verdict in favor of Plaintiffs, finding that: (1) Cognizant had engaged in a pattern or practice of intentional discrimination against non-South Asian employees (on the basis of race) and non-Indian employees (on the basis of national origin) who were terminated from “the bench” and (2) Cognizant’s conduct met the standard for punitive damages. See Verdict. Plaintiffs’ disparate impact claim, which is based on the same underlying facts as their disparate treatment claim, is now before the Court. The parties agreed that the class disparate impact claim could be decided solely on the basis of the existing trial record and supplemental briefing from the parties. [Doc. ## 680 (“JSR”), 684 (“DI Brief”), 691 (“DI Opp.”), 695 (“DI Reply”).] Having carefully reviewed the evidence and the arguments of counsel, as presented at trial on the disparate treatment claim and in the parties’ subsequent written submissions, the Court issues the following findings of fact and conclusions of law pursuant to Rule 52 of the Federal Rules of Civil Procedure. // // // // // II. FINDINGS OF FACT4 A. The Parties 1. Defendants are Cognizant Technology Solutions Corporation and Cognizant Technology Solutions U.S. Corporation (collectively, “Cognizant”). FPTCO at 3. 2. Cognizant is an American multinational corporation that provides information technology, consulting, and staffing services to customers worldwide. Id. 3. Cognizant employs over 40,000 employees in the United States. Id. at 4. 4. Plaintiffs are Brian Cox, Vartan Piroumian, Jean-Claude Franchitti, and Christy Palmer. Id. at 2. Cox and Piroumian represent the Class in this action. Trial Tr. (9/24/24) 1675:18-21 [Doc. # 670]. 5. Brian Cox’s late father, Edward Cox (hereinafter “Mr. Cox”), began working for Cognizant in January 2014 as an infrastructure engagement manager/program manager supporting infrastructure services, staffing, and delivery for North American clients. FPTCO at 4. 6. Mr. Cox was of Caucasian race and Hispanic national origin. Trial Tr. (10/10/24) 3110:25; see also Class Cert. & MSJ Ord. at 19. 7. Mr. Cox was sent to the corporate deployment pool, also known as the “bench,” in January 2017. FPTCO at 4. 8. Mr. Cox interviewed for multiple open roles, but despite having already performed a similar role, he was told that he was not qualified for the positions. Trial Tr. (9/30/24) 2592:2-7, 2606:6-8 [Doc. # 674]; see also Class Cert. & MSJ Ord. at 20. 9. Mr. Cox was terminated from Cognizant, without severance, on April 3, 2017. FPTCO at 5. 10. Plaintiff Vartan Piroumian began working for Cognizant on April 1, 2012 as an enterprise architect. Id. 4 To the extent any of the Court’s findings of fact may be considered conclusions of law or vice versa, they are so deemed. 11. Piroumian is of Caucasian race and American national origin. Id. 12. Piroumian was terminated from Cognizant on August 2, 2017. Id. 13. Both Mr. Cox and Piroumian were terminated from the bench. Id. B. Cognizant’s Staffing Model 14. When Cognizant contracts with a client, it staffs individuals to serve the client. These individuals may be external applicants or current Cognizant employees working on another client project or waiting to be deployed from the bench. FPTC0 at 4. 15. When billable employees complete their work on a particular project, they either move directly to a new project or go to the bench. Trial Tr. (9/30/24) 2556:1-10. 16. In general, when an employee has been on the bench for more than five weeks, they are terminated. Id. 17. While on the bench, employees may utilize an internal system known as the Talent Marketplace to view and apply for open positions. Trial Tr. (10/2/24) 3062:21- 3063:12 [Doc. # 676]. 18. The “Talent Supply Chain” then uses Talent Marketplace to match available employees with open positions. Id. at 3037:6-8. 19. Cognizant’s Talent Supply Chain, Global Mobility, and Business Operations teams were all involved in carrying out the staffing policies relevant to this action. See Trial Exhibit (“TX”) 219 at 5. C. Cognizant Implemented “Visa Readiness” and “Visa Utilization” Policies During the Class Period. 20. Cognizant’s “Visa Readiness” policy refers to Cognizant’s policy of applying for a significant number of H1-B visas based on projected growth and future needs to create a pool of “travel ready” employees to fill U.S. roles in the future. Cognizant’s implementation of this policy is evidenced by various internal communications and presentations. See, e.g., TX-1 (email explaining that “the idea is to get associates visa-ready, so if a suitable opportunity arises in the U.S. we can move quickly”); TX-58 at 4 (presentation on “H-1B Visa Planning”); TX-232 (email discussing need for “more aggressive placement of travel ready resources”); TX-295 at 2 (email suggesting that “mobile talent deficit” could be resolved by “hir[ing] candidates with H1B visa in India working for other employers”); TX-322 (email outlining process to add associates from India to the U.S. workforce since Cognizant “didn’t get as many visas [as] it wanted”). 21. Cognizant’s “Visa Utilization” policy encompasses multiple other policies at Cognizant, including the rotation policy, “operational excellence” initiatives, and visa chargeback policy. Through these policies, Cognizant’s Global Mobility, Talent Supply Chain (“TSC”), and Business Opportunities (“BU Ops”) teams work to maximize utilization of Cognizant’s pool of “travel ready” visa workers. Cognizant’s implementation of this policy is evidenced by various internal communications and presentations. See, e.g., TX-219 at 5 (presentation explaining “joint effort between Global Mobility, TSC and the BU Ops teams to drive visa utilization”); TX-324 (“Visa Utilization” presentation). D. The Statistics 22. Plaintiffs hired an expert, Dr. Phillip Johnson, to conduct a statistical analysis of hiring and termination disparities at Cognizant. Trial Tr. (9/25/24) 2032:8-10. 23. First, Dr. Johnson compared the rate at which Cognizant’s non-South Asian employees in the United States were involuntarily terminated during the Class Period against the rate at which Cognizant’s South Asian employees in the United States were involuntarily terminated. TX-407 at 28. 24. Dr. Johnson’s analysis found that non-South Asians were approximately seven times more likely to be subject to an involuntary termination, with a disparity of 96 standard deviations. 5 Id. at 6. 25. According to Dr. Johnson, a disparity of 96 standard deviations represents an “extremely low probability of a[n] occurrence by chance, less than one in a billion.”6 Id. at 5; Trial Tr. (9/25/24) 2044:7-9. 26. Second, Dr. Johnson compared the rate at which non-South Asian employees in the United States were terminated from the bench against the rate at which South Asian employees in the United States were terminated from the bench. TX-407 at 29. 27. Dr. Johnson’s analysis found that non-South Asian and non-Indian employees were 8.4 times more likely to be terminated from the bench than South Asian and Indian employees, with a disparity of 75.12 standard deviations. Id. at 6; Trial Tr. (9/25/24) 2046:5-22. 28. Given that the jury found in favor of Plaintiffs on all issues at the disparate treatment trial, the jury implicitly credited Dr. Johnson’s findings and testimony, despite Cognizant’s challenges thereto during cross-examination. See Jury Instr. No. 16 (“Expert Opinion”). E. Plaintiffs’ Experiences 29. The personal experiences of Mr. Cox, Piroumian, Franchitti, and Palmer illustrate the impact shown in Dr. Johnson’s statistics.

5 “A standard deviation is a measure of the likelihood that an outcome is the result of chance as opposed to some systematic difference and so the [ ] higher standard of deviation means it being a more unlikely occurrence to have occurred just because of chance.” Trial Tr. (9/25/24) 2043:2-6. 6 “Standard deviations of greater than 1.96 correspond to less than 5% probability (or one in twenty) of the disparity being observed by chance, standard deviations of greater than 3.29 correspond to less than 0.1% probability (or one in one thousand), standard deviations of greater than 6.12 correspond to less than 0.0000001% probability (or one in a billion).” TX 407 at 20 n.47. 30. For example, it is undisputed that Mr. Cox was removed from his role as part of Cognizant’s “operational excellence” and “margin optimization” programs. See TX-700 at 30; Trial Tr. (9/30/24) 2606:6-8. 31. Similarly, in accordance with Cognizant’s rotation policy, Palmer was rotated off the UHC project and replaced by a visa-holding, South Asian employee, despite their difference in qualifications. See Trial Tr. (9/30/24) 2645:20-2646:15; TX- 30 at 1-2 (rotation policy presentation explaining that associates higher on the “value chain” should be rotated off projects and replaced by visa-holding employees “with a lower grade and experience”). 32. Piroumian’s and Franchitti’s respective testimony at trial were analogous, as they both described the impact of the visa policies on staffing Cognizant’s GTO group for elite architects. See Trial Tr. (9/27/24) 2488:6-2490:2 (Piroumian testified that he was removed from projects and replaced by Indian visa-holding employees), 2402:10-2406:11 (Franchitti testified about the impact of the visa readiness policy on his ability to staff members of the GTO group [like Piroumian] to projects). F. The Jury’s Findings 33. Given that the verdict forms did not require the jury to make any express findings as to why it found for Plaintiffs on the disparate treatment claim, the Court looks to the jury instructions and the jury’s verdict to discern the jury’s implicit determinations. 34. The Court instructed the jury that for Plaintiffs to prevail on the disparate treatment claim, they must prove the following by a preponderance of the evidence: a. that race- or national origin-based discrimination was Cognizant's regular practice, rather than something unusual; and b. that Cognizant acted with discriminatory intent. Jury Instr. No. 23. 35. The Court also instructed the jury that: “A plaintiff is not required to produce direct evidence of intentional discrimination. Intentional discrimination may be inferred from the existence of other facts. In some situations, discriminatory intent can be inferred from evidence of a pattern or practice. Such situations occur only where there is a stark or clear pattern, unexplainable on grounds other than race or national origin.” Id. 36. The jury also found that punitive damages were available to Plaintiffs. See Verdict. 37. The Court instructed the jury that for punitive damages to be available to Plaintiffs, they must prove by a preponderance of the evidence “that the defendant's conduct that harmed the plaintiff was malicious, oppressive, or in reckless disregard of the plaintiff’s rights.” a. Conduct is malicious if it is accompanied by ill will, or spite, or if it is for the purpose of injuring the plaintiff. b. Conduct is in reckless disregard of the plaintiff's rights if, under the circumstances, it reflects complete indifference to the plaintiff's safety or rights, or if the defendant acts in the face of a perceived risk that its actions will violate the plaintiff's rights under federal law. c. An act or omission is oppressive if the defendant injures or damages or otherwise violates the rights of the plaintiff with unnecessary harshness or severity, such as by misusing or abusing authority or power or by taking advantage of some weakness or disability or misfortune of the plaintiff. Jury Instr. No. 27. 38. Having found for Plaintiffs on both the disparate treatment claim and availability of punitive damages, the jury must have implicitly found that Plaintiffs proved by a preponderance of the evidence that race- and/or national origin-based discrimination was Cognizant’s regular practice, that Cognizant acted with discriminatory intent in carrying out this practice, this practice was malicious, oppressive or in reckless disregard of Plaintiffs’ rights, and it caused harm to at least some Plaintiffs. // // Disparate impact claims challenge “employment practices that are facially neutral in their treatment of different groups but that in fact fall more harshly on one group than another and cannot be justified by business necessity.” Stout v. Potter, 276 F.3d 1118, 1121 (9th Cir. 2002) (quoting Int’l Brotherhood of Teamsters v. United States, 431 U.S. 324, 335 n.15 (1977)). A plaintiff seeking to advance a disparate impact claim must demonstrate: (1) a specific employment practice that (2) causes a significant discriminatory impact. The plaintiff must also establish that the challenged practice is either (a) not job related or (b) inconsistent with business necessity. Even if the practice is job related and consistent with business necessity, though, a plaintiff may still prevail by showing that the employer refuses to adopt an available alternative practice that has less disparate impact and serves the employer’s legitimate needs. Freyd v. Univ. of Oregon, 990 F.3d 1211, 1224 (9th Cir. 2021) (internal citations omitted). A. The Court is bound by the jury’s implicit findings. 1. “[W]here legal claims tried by the jury and equitable claims tried by the court are ‘based on the same set of facts, the Seventh Amendment requires the trial judge to follow the jury’s implicit or explicit factual determinations.’” Sanders v. City of Newport, 657 F.3d 772, 783 (9th Cir. 2011) (quoting Miller v. Fairchild Indus., Inc., 885 F.2d 498, 507 (9th Cir. 1989)); Los Angeles Police Protective League v. Gates, 995 F.2d 1469, 1473 (9th Cir. 1993) (reversing district court’s denial of equitable relief because it “engag[ed] in factfinding contrary to the implicit findings of the jury verdict”). // // B. Plaintiffs may pursue both disparate treatment and disparate impact claims. 2. Disparate treatment and disparate impact claims are distinct claims, and plaintiffs’ success on one claim does not guarantee their success on the other. See Raytheon Co. v. Hernandez, 540 U.S. 44, 52 (2003). But these two claims are not mutually exclusive, and they may be brought together in the same action. See Fragante v. City and County of Honolulu, 888 F.2d 591, 594 (9th Cir. 1989) (“A plaintiff may bring an action against an employer under a disparate treatment and/or disparate impact theory”). 3. Indeed, although evidence of discriminatory intent is not required for disparate impact claims as it is for disparate treatment claims, it can bolster a plaintiff’s disparate impact case. See Ave. 6E Invs., LLC v. City of Yuma, Arizona, 217 F. Supp. 3d 1040, 1055 (D. Ariz. 2017) (citing Casa Marie, Inc. v. Superior Court, 988 F.2d 252, 270 n.20 (1st Cir. 1993)). 4. Cognizant’s argument that the jury’s verdict on disparate treatment prevents Plaintiffs from prevailing on their disparate impact claim therefore is incorrect. See Antonio v. Wards Cove Packing Co., Inc., 810 F.2d 1477, 1480 (9th Cir. 1987) (en banc), cert. denied, 485 U.S. 989 (1988) (“In a class action suit, commonly known as a ‘pattern or practice’ case, plaintiffs typically assert claims both of disparate treatment occasioned by an employer’s practices and of disparate impact produced by those practices”). C. Cognizant’s Visa Readiness and Visa Utilization policies are facially neutral, specific employment practices. 5. “Plaintiffs generally cannot attack an overall decision-making process in the disparate-impact context, but must instead identify the particular element or practice within the process that causes an adverse impact.” Stout v. Potter, 276 F.3d 1118, 1124 (9th Cir. 2002). Typically, this will be accomplished by providing “statistical evidence showing that an employment practice selects members of a protected class in a proportion smaller than their percentage in the pool of actual applicants.” Id. at 1122. 6. Here, Plaintiffs challenge Cognizant’s Visa Readiness and Visa Utilization policies, as specifically identified in Cognizant’s own presentations and communications. See, e.g., TX-30 (indicating that holding a visa is a requirement for entering the “rotation list” as part of Visa Utilization policy); TX-41 at 12, 14 (describing need to “further improve visa utilization in India”); TX-219 at 5 (highlighting “joint effort” to “drive visa utilization”); TX-261 at 8 (explaining Cognizant’s plan to “create an additional pool of visa ready associates” by, in part, identifying additional potential associates in India); TX-316 at 2 (email expressing “need to leverage offshore travel ready resources more”); TX-325 at 6 (presentation listing “ensure robust rotation of visa ready associates” as an action item). 7. Cognizant’s Visa Readiness and Visa Utilization policies are facially neutral because visa status is not a protected trait. See Class Cert. & MSJ Ord. at 58. D. Cognizant’s policies caused a disparate impact on non-South Asian and non- Indian employees. 8. Non-visa holding employees were both sent to the bench and terminated from the bench at significantly higher rates than their visa holding coworkers. See TX-38 at 4 (30% termination rate from the bench for non-visa employees vs. 3% for visa employees); TX-43 (showing that in 2015, for example, 69% of U.S. citizens on the bench were redeployed and 23% were terminated, while 88% of visa holders on the bench were redeployed and 1% were terminated). Further, approximately 99% of the visa-holding employees at Cognizant were of Indian national origin, and approximately 88% of Cognizant’s workforce was South Asian at most times during the class period. See Trial Tr. (9/26/24) 2242:22-25; TX-279 at 7; see also Trial Tr. (9/25/24) 1900:4-7 (“at least over 80%” of Cognizant’s Asian employees were specifically Indian and/or South Asian). This is precisely the type of overrepresentation that the disparate impact analysis was intended to address. See Sw. Fair. Hous. Council, Inc. v. Maricopa Domestic Water Improvement Dist., 17 F.4th 950, 963 (9th Cir. 2021) (explaining that comparing the demographics of the populations that were affected and unaffected by a suspect policy is the appropriate method for assessing disparate impact), 964 (finding disparate impact where 38.9% of African Americans were affected by policy even though they comprised only 2.9% of the district’s customer base). 9. Therefore, non-visa holding, non-South Asian or Indian employees were terminated from the bench at a rate that was significantly disproportionate to their representation in Cognizant’s employee population. Compare TX-279 at 18 (“Audit Readiness” presentation noting that “[t]he significant population of Asians in our U.S. workforce will most likely raise questions concerning our employment practices”), 22 (demographic survey results showing that over 85% of U.S. workforce was Asian) with Trial Tr. (9/25/24) 1925:24-1927:6 (Abigail Israel discussing TX-245 and comparing involuntary attrition rates with employee demographics, noting that, for example, 72% of Black employees were involuntarily terminated, as compared to less than 1% of Asian employees). 10. These disproportionate rates of termination, along with statistically significant standard deviations, see supra II(D), are “sufficiently substantial” to raise “an inference of causation.” See Paige v. California, 291 F.3d 1141, 1145 (9th Cir. 2002) (quoting Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 995 (1988)). 11. When compared with other cases in which courts have found disparate impact based on statistical disparities, it is evident that Cognizant’s facially neutral visa policies had a disparate impact on non-South Asian and non-Indian Cognizant employees. See, e.g., Guerrero v. California Dep’t of Corr. & Rehab., 119 F. Supp. 3d 1065, 1077 (N.D. Cal. 2015), aff’d in part and rev’d in part other grounds, 701 F. App’x. 613 (9th Cir. 2017) (affirming district court’s finding of disparate impact where challenged policy resulted in Latinos being 2.4 times more likely to suffer an adverse impact than non-Latino counterparts); Reynolds v. Sheet Metal Workers Loc. 102, 498 F. Supp. 952, 967 (D.D.C. 1980), aff’d 702 F.2d 221 (D.C. Cir. 1981) (finding disparate impact where there was a 12.5% disparity, with approximately 2 to 5 standard deviations). E. Cognizant’s discriminatory policies were not justified by any business necessity. 12. After plaintiffs meet their burden of proving a prima facie case of discriminatory impact, the employer then bears the burden of showing that the challenged practice is job-related and consistent with business necessity. See Contreras v. City of Los Angeles, 656 F.2d 1267, 1275 (9th Cir. 1981). 13. Here, however, the jury already implicitly rejected Cognizant’s business necessity defense. Not only is the Court bound by the jury’s implicit findings, the evidence at trial overwhelmingly supports that conclusion. As the Court explained in its Order denying Plaintiffs’ request for a ruling on their disparate impact claim following the June 2023 Phase I mistrial, “at least one factual issue—whether Cognizant had a valid reason why it engaged in practices that resulted in disproportionate hiring of Indian and South Asian employees in the U.S.—is common to both [the disparate treatment and disparate impact] claims.” [Doc. # 615 at 1.] Thus, by finding that Cognizant’s disparate treatment was “unexplainable on grounds other than race or national origin,” the jury also implicitly found that Cognizant did not have a legitimate business necessity for their discriminatory visa policies. See Allison v. Citgo Petroleum Corp., 151 F.3d 402, 424 (5th Cir. 1998) (“[T]he business necessity defense to disparate impact claims and the legitimate nondiscriminatory reason defense to disparate treatment claims are not ‘so distinct and separable’ from one another that they may be considered separately by multiple factfinders without violating the Seventh Amendment” (quoting Gasoline Prod. Co. v. Champlin Refining Co., 283 U.S. 494, 500 (1931))).7

7 Because Cognizant’s visa policies were not justified by any legitimate business necessity, the Court need not address whether alternative policies were available. See Griggs v. Duke Power Co., 401 U.S. 424, 432 (1971). 1 IV. In light of the foregoing, the Court concludes that Cognizant’s “Visa Readiness,” Visa Utilization,” and related policies and practices had a disparate impact on non-South Asian and/or non-Indian employees, resulting in the disparate termination of non-South Asian and/or non-Indian employees from the bench during the Class Period of December 2016 to October 27, 2022. The Court hereby ORDERS as follows: 1. by December 19, 2025, the parties shall meet and confer regarding the Hogistics of Phase Two of this case; and 2. by January 9, 2026, the parties shall file a stipulation and proposed order regarding the conduct of Phase Two proceedings or, if there is no agreement, a proposed motion briefing schedule for the Court to resolve the manner in which Phase Two will ! proceed. 1T IS SO ORDERED. DATED: December 5, 2025 Fister M. GEE CHIEF UNITED STATES DISTRICT JUDGE

-15-