Case Information
*1 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 1 of 69 Page ID
#:24903 UNITED STATES DISTRICT COURT CENTRAL DISTRICT OF CALIFORNIA ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) Case No. CV 17-6848-DMG (PLAx) CHRISTY PALMER, VARTAN PIROUMIAN, BRIAN COX, and JEAN- ORDER RE MOTION FOR CLASS CERTIFICATION [256], MOTION TO CLAUDE FRANCHITTI, EXCLUDE TESTIMONY OF DR. Plaintiffs, PHILLIP JOHNSON [281], MOTION v. TO EXCLUDE TESTIMONY OF DR. RONIL HIRA [285], MOTION FOR
COGNIZANT TECHNOLOGY PARTIAL SUMMARY JUDGMENT SOLUTIONS CORPORATION and [266], MOTION FOR SANCTIONS COGNIZANT TECHNOLOGY [298]
SOLUTIONS U.S. CORPORATION,
Defendants. ) ) ) ) )
Before the Court are (1) Plaintiffs’ Motion to Certify a Class [Doc. # 256]; (2) Defendant Cognizant’s Motion to Exclude the Expert Testimony of Dr. Philip Johnson [Doc. # 281]; (3) Cognizant’s Motion to Exclude the Expert Testimony of Dr. Ronil Hira [Doc. # 285]; (4) Cognizant’s Motion for Partial Summary Judgment [Doc. # 266]; and *2 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 2 of 69 Page ID
#:24904 (5) Plaintiffs’ Motion for Sanctions [Doc. # 298]. All five motions are fully briefed. The Court held a hearing on the motions on September 9, 2022. Having carefully considered the parties’ arguments, the Court issues the following omnibus ruling.
I.
RELEVANT PROCEDURAL BACKGROUND
This action was filed in this Court on September 18, 2017. Compl. [Doc. # 1]. After the Court resolved two motions to dismiss, see First MTD [Doc. # 42], Second MTD [Doc. # 58], Plaintiffs Christy Palmer, Vartan Piroumanian, Edward Cox, and Jean-Claude Franchitti filed the operative Third Amended Complaint (“TAC”) on January 14, 2021. [Doc. # 145.] In their TAC, Plaintiffs allege that Defendants Cognizant Technology Solutions Corporation and Cognizant Technology Solutions U.S. Corporation (together, “Cognizant”) discriminate against non-South Asians and non-Indians in hiring, promotion, and terminations. Id . at ¶ 1. Plaintiffs assert disparate treatment claims under 42 U.S.C. section 1981 and Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e, et seq. , and a disparate impact claim under Title VII. . at ¶¶ 2, 84, 89, 94. Cognizant filed its operative Answer on January 29, 2021. [Doc. # 147.]
Plaintiffs now seek to certify the following Classes and Subclass: Termination Class
Between September 18, 2013 to the date of class certification, all individuals who are not of South Asian race or Indian national origin and who were terminated while employed within a Cognizant Class Band in the U.S., excluding any individuals bound by an agreement to arbitrate termination claims with Cognizant.
Bench Termination Sub-Class
Between September 18, 2013 to the date of class certification, all individuals who are not of South Asian race or Indian national origin and who were terminated from the bench while employed within a Cognizant Class Band in *3 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 3 of 69 Page ID
#:24905 the U.S., excluding any individuals bound by an agreement to arbitrate termination claims with Cognizant.
Hiring Class
Between September 18, 2013 to the date of class certification, all individuals who are not of South Asian race or Indian national origin and who applied to a Cognizant Class Band position in the U.S. and were not offered a position.
See Mot. for Class Certification at 3 [Doc. # 256]. Plaintiffs also seek an order (a) appointing Brian Cox as the Class Representative of the Hiring Class, (b) appointing Brian Cox and Vartan Piroumanian as Class Representatives of the Termination Class and Bench Terminations Subclass, and (c) appointing Kotchen & Low LLP and Yadegar, Minoofar & Soleymani LLP as Class Counsel.
II.
ADMISSIBILITY ISSUE
Cognizant interposes evidentiary objections to much of Plaintiffs’ evidence in support of class certification, on the ground that the evidence is inadmissible. The Ninth Circuit has previously squarely held that a district court abuses its discretion “by declining to consider [evidence] solely on the basis of inadmissibility.” Sali v. Corona Reg’l Med. Ctr. , 909 F.3d 996, 1006 (9th Cir. 2018). Cognizant acknowledges this holding, but contends that the Ninth Circuit abrogated it in Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC , 31 F.4th 651 (9th Cir. 2022). See Cognizant’s Evidentiary Objections [Doc. # 270].
In its
en banc
decision in
Olean
, the Ninth Circuit resolved the outstanding question
of what evidentiary standard a plaintiff seeking class certification must satisfy, concluding
that a plaintiff must establish the elements of Rule 23 are present by a preponderance of
the evidence.
See Olean
,
*4 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 4 of 69 Page ID
#:24906 [i]n carrying the burden of proving facts necessary for certifying a class under Rule 23(b)(3), plaintiffs may use any admissible evidence. See Tyson Foods, [ Inc. v. Bouaphakeo , 577 U.S. 442, 454-55 (2016)] (explaining that admissibility of evidence at certification must meet all the usual requirements of admissibility and citing to Rules 401, 403, and 702 of the Federal Rules of Evidence).
Id . at 665. It is upon this sentence that Cognizant relies to argue that all of Plaintiffs’ evidence in support of certification must be admissible. Cognizant interposes a range of evidentiary objections to Plaintiffs’ evidence, including, for example, objections under Federal Rule of Evidence 403. But other than this sentence, there is no indication in the Olean decision that the Ninth Circuit intended to abrogate its prior determination that evidence presented in support of class certification need not be admissible in its current form. Instead, the Court in Tyson Foods reasoned that the “permissibility” of using any type of evidence in a class action depends “on the degree to which the evidence is reliable in proving or disproving the elements of the relevant cause of action.” Tyson Foods , 577 U.S. at 455. The Court cited to Federal Rules of Evidence 401 (regarding relevance), 403 (regarding the relationship between probative value and potential prejudice), and 702 (regarding expert witnesses).
Taken in context, the statement in
Olean
regarding the use of admissible evidence
can be harmonized with the Ninth Circuit’s decision in
Sali
. In
Sali
, the court stated that a
district court “evaluating challenged expert testimony in support of class certification [. . .]
should evaluate admissibility under the standard set forth in
Daubert
.”
#:24907 reason, the Court does not understand the en banc panel in Olean to have overruled Sali . This Court thus reads the Ninth Circuit’s statements regarding admissibility in Olean in light of those earlier decisions, and emphasizes the reliability and relevance of evidence rather than the form of the evidence. Cognizant’s evidentiary objections directed strictly at admissibility are thus OVERRULED .
III.
FACTUAL BACKGROUND
A. Cognizant Services and Staffing Model
Cognizant is an international corporation, headquartered in the United States, that provides different kinds of information technology and consulting services. Westphal Decl. ¶ 3 [Doc. # 272-11]. It serves many different industries, including banking and financial services, healthcare, life sciences, and communications. Id . Cognizant is organized into multiple overlapping business units, including “vertical” units of industry- specific practice groups and “horizontal” units that provide a particular type of service across multiple industries. Id .
Cognizant has more than 40,000 employees in the U.S. Westphal Decl. ¶ 3. When Cognizant contracts with a client, it staffs individuals to serve the client. Answer at ¶ 16. These individuals are drawn from either external applicants or current Cognizant employees. Id . These current employees may be working for another client project, or they may be drawn from the “bench,” pools of employees waiting to be deployed. .
Some roles at Cognizant are “billable,” in that their time can be billed to clients. Employees in these roles typically move from client project to client project. See Westphal *6 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 6 of 69 Page ID
#:24908 Decl . at ¶ 7. Other roles at Cognizant are “nonbillable.” These include employees who work in corporate departments such as Human Resources and Legal Services. Some nonbillable employees also provide nonbillable functions to particular business units, such as sales. Id.
Cognizant organizes its U.S. workforce into 19 “job bands,” or levels of seniority. Johnson Report ¶ 18, Fig. 4 [Doc. # 256-5]; see also Westphal Decl. ¶ 6 (17 “commonly used job bands”). The higher job bands are more junior (Band 95 is “Interns”); the lowest job band is Band 10, for Cognizant’s Chief Executive Officer. Id . The “Class Bands” make up the following job descriptions: Associate, Senior Associate, Manager, Senior Manager, Associate Director, Director, and Senior Director. Johnson Report at 12, Fig. 4.
About 88% of employment during the Class Period was in jobs in the Class Bands. [2] Johnson Report ¶ 19. More than 98% of South Asian employees and 99% of employees who required a visa to work in the U.S. were employed in the Class Bands. Id. More senior roles, such as Vice President and Chief Executive Officer, are excluded from the Class Bands, as are bands with the descriptions “Unleveled” and “Unknown,” Interns, Administrative Executives, and various Programmers. . [3]
B. Hiring
Cognizant has processes for hiring externally and for “hiring” from the bench or between projects within Cognizant. [4] Multiple departments within Cognizant are responsible for hiring employees and placing them on client projects, as described herein. *7 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 7 of 69 Page ID
#:24909
1. The Bench
When a billable employee’s work on a particular project ends, the employee moves either directly to a new project or to the bench. Padmanabhan Decl. ¶ 5 [Doc. # 272-9]. Employees who are on the bench will try to find a placement on a new project, sometimes with the assistance of other employees from the Talent Supply Chain, a department within Cognizant that helps allocate U.S. talent to billable positions on client projects. Id. at ¶¶ 5, 3. The Talent Supply Chain was previously called Global Workforce Management. Id. Employees on the bench continue to receive their full salaries from Cognizant. Westphal Decl. ¶ 3.
2. Hiring From the Bench Talent Supply Chain does not “hire” employees from the bench; instead, it “facilitates” hiring. Padmanabhan Decl. at ¶ 9. Since 2018, it has done this in part through an internal system called Talent Marketplace, through which benched employees can view and apply for open positions within the company. Id . at ¶ 6. Hiring managers can also use Talent Marketplace to find employees who are a match for open roles. Id . Talent Supply Chain uses Talent Marketplace to attempt to match open positions and available employees. Id . Talent Marketplace does not allow hiring managers and Talent Supply Chain employees to view the work authorization status of an employee. Id . Before Talent Marketplace was introduced, Cognizant used two other systems to attempt to place employees from the bench. From 2015 to 2018, Cognizant used a platform called C- Deploy. Id . at ¶ 8. From 2013 to 2014, Talent Supply Chain employees sent weekly emails with the names and information of benched employees to hiring managers, who could then choose to set up interviews with the employee. .
Employees can indicate through Talent Marketplace whether or not they are open to relocation. Padmanabhan Decl . at ¶ 6. But Talent Supply Chain and hiring managers *8 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 8 of 69 Page ID
#:24910 sometimes recommend employees for roles in different locations, even if the employee has indicated a location preference in Talent Marketplace. Id . at ¶ 7. The Talent Supply Chain and hiring managers tell employees that being willing to relocate increases the roles for which they may be considered. Id. Plaintiffs also assert that many positions are posted to Cognizant’s internal electronic staffing system (“ESA”), these positions are filled before they are posted to Talent Marketplace, and many benched employees therefore do not see the bulk of positions that become open within the company. [ See Doc. # 323-22.]
3. External Hiring Cognizant also, of course, hires external candidates. Cognizant’s Assistant Vice President of Workforce Strategy, Eric Westphal, attests that Cognizant “strongly prefers to staff new client projects with existing U.S.-based employees, either those just rolling off another project or those on the U.S. bench.” Westphal Decl. ¶ 8; accord Padmanabhan Decl. ¶ 14 (Cognizant prefers to staff existing U.S. employees). This is because “staffing a new role with an existing employee is the fastest way to fulfill client demand, which is particularly important when a project is time-sensitive for a client.” Westphal Decl. ¶ 8. Westphal also reasons that prioritizing the placement of U.S. employees who are on the bench is “financially prudent,” given that these employees continue to receive their salaries while they are on the bench. Id .
If Cognizant must hire externally, a “requisition request” is sent to Cognizant’s Talent Acquisition Group. Westphal Decl. ¶ 9. The Talent Acquisition Group is the unit in charge of recruiting and screening external candidates. Id. The Talent Acquisition Group recruits candidates through multiple channels, including employee referrals, campus recruitment and job fairs, apprenticeships and training programs, and online job platforms like LinkedIn and Indeed. Id . at ¶ 10. External applicants can apply for jobs directly through Cognizant’s online Career Site. . Talent Acquisition Group can also search Taleo, Cognizant’s external applicant data system, for candidates with the necessary skills to fill particular roles. Id. at ¶ 11. New positions for which Cognizant seeks an external candidate go into Taleo, and information for all applicants to external positions also goes *9 Case 2:17-cv-06848-DMG-GJS Document 384 Filed 10/27/22 Page 9 of 69 Page ID
#:24911 into Taleo. Id . Internal applicants—that is, Cognizant employees who are on the bench— are not intended to apply for new roles through Taleo. Id. at ¶ 12. Instead, they apply “on their own” or with assistance from Talent Supply Chain. Id . [6]
Through Taleo, Cognizant keeps track of positions to which particular external candidates apply. Westphal Decl. ¶ 11. Taleo has a field labeled “InternalCandidate,” which can have a “true” or a “false” value. Westphal Decl. ¶ 13. Westphal testifies that this field does not, however, reflect whether a candidate was an internal candidate at the time they applied. Id . Instead, Westphal says this field is updated only after an external candidate is hired, and the “true” value populates across all positions to which that candidate applied. Id . Cognizant’s employee database, PeopleSoft, communicates that information to Taleo to “close the loop” on these candidates. Id . Westphal testifies the purpose of this is to inform hiring managers that the formerly external candidate has now been hired. Id . [7]
When an external candidate is applying for an open job at Cognizant, she may interview with multiple Cognizant employees, and for project-based positions, she often also interviews with the client. Westphal Decl. ¶ 14; see also Padmanabhan Decl. ¶ 4. The Talent Acquisition Group coordinates interviews with hiring managers. Westphal Decl. ¶ 9. Across all job openings, both internal and external, about 40% explicitly indicate that a client interview is required, but Westphal attests that in his experience, the actual percentage is much higher. Id . at ¶ 15. After all interviews are completed, the business team decides who to hire, often with input from the client, then informs the recruiter. . at ¶ 14. There is no single gatekeeper who screens all hires, nor a single decisionmaker who reviews or approves all external hires. Id .
4. Hiring To the Bench External applicants are sometimes hired directly to the bench in anticipation of a gap between client demand and Cognizant’s existing supply of potential workers. Padmanabhan Decl. ¶ 17.
C. Terminations
Cognizant employees may be terminated for many different reasons. Many of these reasons are self-explanatory, including terminations for poor performance and terminations with separation agreements. Westphal Decl. ¶¶ 21-23. Cognizant employees may also be terminated, however, for failure to find a new project after being placed on the bench. The details of this process require further explanation.
1. Terminations From the Bench In general, when employees have been on the bench for more than five weeks, they are terminated. Padmanabhan Decl. ¶¶ 11-12. There are exceptions to this rule—for example, new hires are exempt from this “bench timeline,” as are other employees on a case-by-case basis. Id . at ¶ 11. Employees who are hired directly to the bench are not subject to a bench timeline for six months after they are hired. Id . at ¶ 17. The “bench timeline” may also be extended if, for example, an employee obtains an interview shortly before their bench timeline is set to expire. Id . at ¶ 12. Cognizant’s legal counsel review all bench terminations. . at ¶ 13.
Before January 1, 2014, terminations from the bench were coded as “CDP” in internal databases. Since that date, there have been six CDP codes for bench terminations, each describing a different reason for the termination. Westphal Decl. ¶ 23. The code that is most at issue here is “CDP – Location Preference.”
D. Visa-Holding Employees
From 2013 to 2019, United States Citizenship and Immigration Services (“USCIS”) reported that Cognizant received the highest number of H-1B visas of any employer in the country. See Kotchen Decl. ISO Class Cert. ¶ 5, Ex. 6 [Doc. ## 256-1, 256-9].
Cognizant acknowledges that, in order to hire enough staff to fill information technology and STEM roles, it “supplements” its U.S. workforce with nonimmigrant visa holders. Westphal Decl. ¶ 16. Cognizant primarily hires holders of H-1B and L visas. Id . For H-1B visa sponsorship, Cognizant “almost always” requires that the sponsored individual be a current employee of a Cognizant affiliate overseas. Id . at ¶ 17. L1 visas are permitted only for intra-company transfers who have worked for Cognizant for between two and four years. Id . Certain jobs at Cognizant are not open to visa-holding employees, including in project-based jobs, for which the client determines whether visa-holding employees are eligible. Id . at ¶ 18. It is Cognizant’s practices pertaining to visa-holding employees, and the impacts of those practices on members of the proposed classes, that are primarily at issue here.
1. Visa-Readiness Policy Plaintiff Franchitti is a former Assistant Vice President at Cognizant. Franchitti Decl. ISO Class Cert. ¶ 2 [Doc. # 256-3]. One of his job duties was to sign invitation letters in support of visa applications. Id . at ¶ 3. Beginning in about May 2014, he became concerned about “false representations” in those letters. Id . He asserts that it is his understanding that the purpose of Cognizant’s “fraudulent” visa practices was to ensure Cognizant received the maximum number of visas possible, in order to create a supply of “travel-ready” individuals in India available to be rotated onto projects in the U.S. . Westphal, Cognizant’s Rule 30(b)(6) witness, testified that “travel-ready resources are resources in India with viable visas,” Kotchen Decl. ISO Class Cert., Ex. 21 (Westphal Depo.) at 3:25-4:1 [Doc. # 256-24], although it appears that this may more accurately refer to individuals who are overseas, not only those in India.
Plaintiffs have presented evidence that Cognizant sought visas for individuals with particular skills, even where there was not yet a project to which those individuals would be assigned, so that those individuals could then be quickly staffed on a project when one arose. In a May 2013 email, another Cognizant employee apparently asks Franchitti “to address the team to explain the strategy regarding visa-readiness for potential U.S. opportunities.” [Doc. # 256-28.] The employee went on:
The message to the team has been that the idea is to get associates visa-ready, so if a suitable opportunity arises in the US we can move quickly. In theory this could be a good opportunity for some UK associates, help utilization, and potentially ‘everybody wins’. UK HR, who have a vested interest in any strategy impacting UK&I associates, have requested a meeting beforehand so they fully understand the implications and can consider how they can be of assistance.
Id. In a 2015 email, a Cognizant employee appears to assert that he or she “will look into” a plan to “increase the pool of visa ready associates – run it as a separate project thereby reducing subcon intake for standard skills.” [Doc. # 256-29.] Various documents produced by Cognizant illustrate this policy in practice. [ See, e.g. , Doc. # 256-30 (August 2017 email acknowledging that the location at which a visa-holding employee will be working sometimes changes after the “H Visa” application is filed); Doc. # 256-31 (November 2019 email proposing plan for TSC to project future need for visas as a solution to having “insufficient visa ready resources”); Doc. # 256-33 (October 2011 email describing practice of filing a visa petition without necessarily knowing where the employee will be placed, and describing understanding that “it is a normal practice” for a U.S.-based manager “from the vertical/horizontal” to provide contact information on such a petition); Doc. # 256-33 (April 2013 email describing process of filing visa petitions even without project to which to assign the associate, and instructing another employee that “per JC [Franchitti],” the employee could “ignore” an instruction not to process petitions without an assignment pending); Doc. # 256-35 (May 2016 email instructing an associate to “start working on getting yourself travel ready to come to the U.S.” even though there was no specific opening at that time).] A 2017 “H-1B Visa Planning” presentation states:
We plan to file 12,000 H-1B petitions . . . Assuming a 30% selection in the lottery, this will create a supply of 3000 new visas .
[Doc. # 256-32.] [8] Plaintiffs have also presented evidence that Cognizant sought out job candidates who already held visas, or already had approved visa petitions sponsored by Cognizant’s competitors. [ See, e.g. , Doc. # 255-40 (2017 Visa Planning presentation recommending hiring H-1B visa holders “from the market in India and transfer[ring] their petition to Cognizant”); Doc. # 256-45 (2014 email describing similar plan); Doc. # 256- 37 (same).] It is clear that individuals at Cognizant kept track of which associates were travel-ready. [ See Doc. # 256-50 (noting current number of travel-ready associates and estimating how many will be needed to fulfill demand); Doc. # 256-51 (identifying current number of travel-ready associates); Doc. # 256-52 (same).]
2. Visa Utilization Policy Plaintiffs have also produced substantial evidence that Cognizant has or had a policy to induce associates with an approved visa petition sponsored by Cognizant to use that petition. [ See, e.g. , Doc. # 256-54 (2014 presentation regarding need for visa-ready associates to travel); Doc. # 256-55 (2014 email recapping “Town Hall” at which supervisors were instructed to “deploy visa ready associates within 180 days of approval”).] [9] Individuals with approved petitions not “deployed” by their supervisors within 180 days are either moved to the bench for deployment by a centralized hiring manager or the business unit that failed to timely deploy the individual bears the cost of the petition. [Doc. # 256-55.] The centralized hiring manager (previously Global Workforce Management, now Talent Supply Chain) is responsible for mapping benched travel-ready individuals overseas to U.S. positions. [ See, e.g. , Doc. # 256-58 (2019 email with list of travel-ready associates in India, listing the associate’s skill family and the date of their visa expiration); Doc. # 256-59 (2018 email including same information); Doc. # 256-60 (2020 email regarding associates with U.S. or Canadian visas who have not traveled, by vertical business unit).]
Some of the evidence adduced by Plaintiffs also indicates an explicit preference for visa holders, although some evidence of this preference is also susceptible to the reading that it is intended to exclude employees without authorization to work in the U.S. [ See, e.g. , Doc. # 256-66 (position open to “visa ready” associates only); Doc. # 256-68 (request for employee to staff client, specifying “in-house resources only” and “visa holder preferable”); Doc. # 256-69 (looking for individuals who are visa ready; internal and external profiles to be considered); Doc. # 256-70 (“interested associates that are US travel ready, or have a valid work permit here, please reach out to me immediately”); but see Doc. # 256-67 (complaint from former Cognizant employee Shane Gundlach, who asserts he was instructed that he could hire only from the Indian visa ready pool).] Cognizant sometimes provided additional training in order to match travel-ready associates to open positions. [ See, e.g. , Doc. # 256-71 at 5 (advising “cross-skill[ing]” to match associates to positions); Doc. # 256-72 (recommending training travel ready associates with relevant background knowledge and “bring[ing] them onsite”); Doc. # 256-77 (describing plan to “upskill[] offshore H resources” for certain positions).]
Plaintiffs also assert that Cognizant implemented a so-called “3X3 BU” policy, by which senior resources were rotated off projects and replaced with visa ready associates. Plaintiffs’ evidence of this policy is a presentation stating that a “3-3-3 Policy” provides for (1) rotation of associates who remain unbilled for 3 months to a new project, (2) assignment of visa ready associates who have not traveled within 3 months, and (3) rotation of associates who have been on the same project for 3 years to a new project, as well as references to replacement of more senior associates (i.e., those who have been on a project for 18-36 months) with newer ones. [ See Doc. ## 256-75, 256-76 at 3, 256-77 at 15.] Plaintiffs’ evidence does support their contention that Cognizant rotates more senior employees off accounts to be replaced with newer employees. [ See Doc. ## 256-78 at 12, 256-79 at 2.]
Westphal attests there is no policy instructing recruiters or interviewers to prefer or prioritize visa holders or South Asian candidates. Westphal Decl. ¶ 14. He also states that Cognizant does not have a “visa readiness” or “visa utilization” policy. Id. at ¶ 19. Instead, he says Cognizant tracks the visas it is allotted by the government so those visas can be used to address talent gaps in Cognizant’s existing U.S. workforce or through external U.S. hires. Id . According to Westphal, “it is always Cognizant’s priority to fully staff its existing U.S. workforce before turning to other available options.” Id . Cognizant disclaims any “rotation policy” or “3x3 BU policy.” Padmanabhan Decl. ¶ 18.
3. Statistical Disparities Analysis by Plaintiffs’ expert, Dr. Phillip Johnson, shows that in 2013, nearly 80% of Cognizant employees in the United States were visa-holders. Johnson Report ¶ 16. The percentage of Cognizant employees in the United States who held visas declined precipitously over time. Id . at 10, Fig. 2 (showing more than 70% visa holders between 2014 and 2016, more than 60% visa holders in 2017 and 2018, and more than 20% visa holders in 2019 and 2020). Approximately 99% of those visa-holders were South Asian. Id . at ¶ 16. Likewise, approximately 99% of those visa-holders were either born in India or held an Indian passport. .; see also Westphal Decl. ¶ 17 (“Because most of Cognizant’s non-U.S. workforce is located in India, the majority of Cognizant’s visa- holding employees are from India.”).
Dr. Johnson conducted an analysis of involuntary terminations from within the Class Bands using internal “Termination Reason Codes” obtained from Cognizant. Approximately 42% of these involuntary terminations are coded “CDP-Location Preference.” See Johnson Report at 14, Fig. 5. This is by far the largest share of the involuntary terminations analyzed by Dr. Johnson. The next most-common reason is “Separation Agreement,” which makes up approximately 28% of involuntary terminations, then “CDP-Skill Mismatch,” which make up approximately 7% of involuntary terminations. Location Preference terminations make up an even larger share of terminations specifically from the bench: approximately 83%. See id . at 15, Fig. 6. The next most-common reason for bench terminations is “CDP-Skill Mismatch,” which makes up approximately 13% of bench terminations, then “CDP-Lack of Assignment,” which makes up approximately 2%. Id.
From 2013 to 2016, South Asian employees made up about 90% of Cognizant’s workforce in the Class Bands. The percentage declined from 2016 to 2020, but remained above 70% of Class Band employees in 2020. See Johnson Report at 21, Fig. 8.
Dr. Johnson compared Cognizant’s demographics to those reflected in American Community Survey (“ACS”) data, specifically to the data relating to the Computer Systems Design and Related Services Industry in the North American Industry Classification System (“NAICS”). Johnson Report ¶ 34. Dr. Johnson concluded that the share of South Asian employees hired by Cognizant was “far above” that in the relevant benchmark data: 85% of employees in the benchmark companies were non-South Asian, compared to 10- 25% of employees within the Class Bands at Cognizant. . at ¶ 36. Dr. Johnson concluded that the likelihood that this statistical disparity was due to chance was “less than one in a billion.” Id . at ¶ 37. Dr. Johnson concluded the same was true as to new hires. See id . at ¶ 39.
Dr. Johnson also compared terminations of South Asian employees with non-South Asian employees. Dr. Johnson’s analysis showed that non-South Asian employees within the Class Bands were about seven times more likely to be involuntarily terminated than South Asian employees. Johnson Report ¶ 42. He determined the likelihood that this disparity was due to chance was one in a billion. Id . Non-South Asian employees were eight times more likely to be terminated from the bench than South Asian employees. Id . at ¶ 43. Analysis by Cognizant’s expert, Dr. Justin McCrary, shows, however, that South Asian employees are more likely to be actually placed on the bench. See McCrary Report ¶ 149, Ex. 23 [Doc. # 272-2]. Cognizant’s expert also shows that the percentage of South Asian employees working at Cognizant in the Class Bands has declined during the Class Period, from about 90% of employees in the Class Bands to just over 70%. Id . at Ex. 25. E. Allegations of Discrimination
A number of former employees have filed charges of discrimination against Cognizant. [ See Doc. # 256-87.] On February 5, 2020, EEOC Commissioner Chai Feldblum issued a letter of determination in which he found “reasonable cause to believe that since December 4, 2014, [Cognizant] discriminated against a nation-wide class of non- Indian applicants by refusing to hire them on the basis of their race and national origin and against a nation-wide class of non-Indian employees on the basis of race and national origin by removing them from assignments that they were qualified for, refusing to assign them to new projects, and terminating them when they remained unassigned.” Kotchen Decl. ISO Class Cert. ¶ 5, Ex. 1 at 2 (EEOC Letter of Determination) [Doc. # 256-4]. Commissioner Feldblum thus determined that Cognizant had violated Title VII. . The most specific allegations of intentional discrimination were allegations that employees were excluded from meetings because of their race. [ See, e.g. , Doc. # 256-95 (Christy Palmer told the EEOC she was excluded from meetings and social events because she was the only non-South Asian employee at her work location).]
Plaintiffs’ retained expert, Dr. Ronil Hira, asserts that Cognizant and its competitors (which Dr. Hira calls “India-centric IT consulting firms”) rely on workers with H-1B visas in order to maximize their profits and out of “deep-seated cultural preferences.” See Hira Report ¶¶ 12-18 [Doc. # 256-104]. Dr. Hira states that so-called India-centric IT consulting firms sponsor visas proactively in order to avoid delays in filling positions and then, having sponsored visas for their employees, are incentivized to utilize those workers. Id . at ¶¶ 10-11. Dr. Hira also asserts that these firms, which he says include Cognizant, start out with large South Asian workforces, and “generally prefer” to hire employees in the U.S. “who are culturally similar to its [sic] offshore employees and to management.” Id . at ¶ 13. He also explains that “very different cultural norms surrounding the employer- employee relationship in India,” as opposed to the United States, operate to encourage hiring South Asian employees. Id . at ¶ 14. Specifically, he identifies “significantly reduced worker expectations in terms of working conditions, pay, resource support, and even safety.” Id . He describes the H-1B visa program as “indenturing.” Id .
Some portion of the dynamic identified by Dr. Hira is supported by Cognizant’s internal documents. In particular, a March 2020 chat between two Cognizant employees produced in discovery shows the employees describing a differential between the salaries of Indian employees hired in the U.S. and the salaries of “whites etc.” [ See Doc. # 256- 98.] One of the employees wrote: “An open secret: we hired H1s for the longest because they were cheaper labor that would relocate anywhere and everywhere and as often as needed.” Id . The other employee replied: “Yeah see . . . that’s race discrimination. Cognizant is going to pay dearly one day.” . The first employee again wrote: “OPEN SECRET. Am I going to get fired for opening up? This has been our core biz model.” Id .
Westphal testified in his personal capacity that he understands that visa-holders might have less job mobility than non-visa-holders because of the need for an employer to sponsor their visas. [Doc. # 256-24 at 5.] [ See also Doc. # 256-103 (internal email stating that the recruiting team does not prefer visa holders, but that some managers “will gravitate towards visa workers (mostly because they believe they can push them to work extra hours),” and that Cognizant’s insistence on willingness to relocate as well as Cognizant’s pay rates both pose obstacles to recruiting U.S. workers).]
F. Edward Cox’s Experience
Edward Cox was a U.S. citizen, and described himself as “of American national origin and ancestry.” He also described himself as Hispanic. See Edward Cox Statement in Support of EEOC Charge (“Edward Cox Statement”) ¶ 1 [Doc. # 256-92]. He received two years of schooling in Computer Information Systems from the University of Texas at San Antonio, and two years in Oil and Natural Gas Engineering at Texas A&M University. Id . He began to work in the information technology field in 1984, and started at Cognizant in 2014 in a Director level role. Id . at ¶¶ 7-8. During his first year at Cognizant, he reported to Anil Bhandari, who is South Asian, and Cox asserted that during client-facing meetings, Bhandari would talk over Cox and refuse to allow him to speak to the client. Id . at ¶ 8. Cox also asserted that Bhandari often spoke in Hindi while discussing client-related work and sales opportunities. Id . Cox later reported to several other individuals, all of whom were South Asian. Cox asserted these individuals took little interest in Cox or his work. Id . at ¶ 9. He asserted he was excluded from a meeting by one supervisor, and that his managers and team members often spoke Hindi socially and during meetings, preventing Cox from participating fully. . at ¶ 11.
In January 2017, Cox was placed on the bench. He interviewed for multiple open roles at Cognizant, but despite having already performed a similar role, was told he was not “vetted” or qualified for the positions. Edward Cox Statement at ¶ 12. The roles for which he interviewed were filled with less qualified South Asian employees. Id . Cognizant terminated Cox from the bench on April 3, 2017, with no severance. Id .
In an additional EEOC Charge, Cox asserted that in November 2017 he was contacted by Dave Stark, a Cognizant employee, regarding a senior-level role on Mr. Stark’s team. Second Attachment to Cox EEOC Charge ¶ 9 [Doc. # 256-93]. Cox asserted he told Mr. Stark he was interested, and interviewed with two members of Mr. Start’s team, who told him they believed he was qualified and that they looked forward to Cox joining the team. Id . A Senior Manager at Cognizant informed Cox that Cognizant intended to make Cox an offer, and provided him with information about his compensation package. Id . at ¶ 10. In January 2018, Cox was again informed that Cognizant was working through the offer approval process. Id . Cox never received an offer of employment from Cognizant. . at ¶ 11.
On March 28, 2021, Edward Cox passed away. [Doc. # 158.] On December 10, 2021, the Court granted Plaintiffs’ motion to substitute his wife, Ann Cox, as a plaintiff in this action. [Doc. # 229.] On May 22, 2022, Ann Cox passed away. [Doc. # 265.] On June 16, 2022, while Plaintiffs’ Motion for Class Certification was pending, the Court granted the parties’ joint request to substitute Brian Cox, Edward Cox’s son and the executor of Ann Cox’s estate, as the plaintiff in this action. [Doc. # 282.]
In a declaration filed in support of Plaintiffs’ Reply ISO Class Cert., Brian Cox asserts that he spoke at length with his stepmother about Edward Cox’s claim before her death. Brian Cox Decl. ¶ 2 [Doc. # 323-30]. He asserts he also spoke with his father about the discrimination Edward believed he experienced at Cognizant, including the distress Edward felt when he lost his job at Cognizant in 2017. Id . at ¶ 3. Brian attests that he is invested in participating in the litigation in a meaningful way. Id . at ¶ 5. He also asserts that he is familiar with the claims at issue in this case, has read some of the pleadings, and communicates often with Plaintiffs’ lead counsel. Id . at ¶ 6.
G. Vartan Piroumanian’s Experience
Vartan Piroumanian began working at Cognizant as an Enterprise Architect on April 1, 2012, and was terminated from the bench on August 2, 2017. Piroumanian Decl. ISO Class Cert. ¶¶ 2, 4 [Doc. # 256-2]. He is aware that he was replaced on three of his projects by individuals who traveled from India. Id . at ¶ 3. He was told by others that he was replaced by individuals who traveled from India on 10 of his other projects. .
On June 20, 2017, Piroumanian sent an email in which he told the recipient “I am at Cognizant for another four weeks as I'm on CDP ... big surprise, eh?! I will be participating in the law suit that Daniel Kitchen [sic] is preparing.” Smith Decl. ISO Class Cert. Opp., Ex. 27 [Doc. # 273-9].
IV.
DAUBERT MOTION TO EXCLUDE REPORT OF DR. PHILLIP JOHNSON
A. Legal Standard
Under Daubert v. Merrell Dow Pharmaceuticals , 509 U.S. 579 (1993), and its progeny, a court must
assess [an expert’s] reasoning or methodology, using as appropriate such criteria as testability, publication in peer reviewed literature, and general acceptance, but the inquiry is a flexible one. Shaky but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof . . . Expert opinion testimony is relevant if the knowledge underlying it has a valid connection to the pertinent inquiry. And it is reliable if the knowledge underlying it has a reliable basis in the knowledge and experience of the relevant discipline.
Alaska Rent-A-Car, Inc. v. Avis Budget Grp., Inc.
,
B. Discussion
Under Federal Rule of Evidence 702, a qualified expert witness may offer opinion testimony if “(a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts of the case.” Dr. Johnson was retained by Plaintiffs to offer his opinion of whether Cognizant’s employment data shows that
Cognizant’s employment practices resulted in disparities disfavoring non- South Asian and non-Indian individuals and, in particular, whether the data indicates that Cognizant’s employment practices resulted in disparities disfavoring individuals who were not of South Asian race or Indian national origin, including in hiring, termination/retention, and promotion decisions.
Johnson Report ¶ 7.
Dr. Johnson first performed an analysis to identify which of Cognizant’s employees to categorize as “South Asian.” He used birth country and passport to identify individuals of Indian, Pakistani, Nepalese, Bangladeshi, Afghan, Sri Lankan, or Bhutanese nationality, then analyzed the last names of Cognizant employees who did not fit into this category based on their names to assess which of these employees might be South Asian. Id. at ¶¶ 28-30. Dr. Johnson compared the percentage of new hires at Cognizant who were not South Asian during the Class Period with the percentage of new hires in an industry benchmark groups of other companies who were not South Asian. Id . at ¶¶ 32-41. He also compared the rate at which non-South Asian employees in the U.S. were involuntarily terminated during the Class Period against the rate at which South Asian employees in the U.S. were involuntarily terminated. He finally compared the rate at which non-South Asian employees in the U.S. were terminated from the bench against the rate at which South Asian employees in the U.S. were terminated from the bench. Id . at ¶¶ 42-48.
Importantly, Dr. Johnson did not testify as to what caused the disparities he identified in his report. Indeed, he testified that he viewed the question of what caused the disparities he identified as outside the scope of his report. See Johnson Depo. at 22:6-11 [Doc. # 281-2]. But when asked, he testified that he “think[s] it’s fair to say” and “believe[s]” that there is a causal relationship between Cognizant’s policies and practices and the disparities he identified in his report. . at 16:15-18.
Cognizant does not argue that Dr. Johnson is not qualified. Rather, Cognizant argues that his testimony is inadmissible because it is unhelpful and unreliable. Cognizant contends that Dr. Johnson’s report is unreliable because (1) he compared apples to oranges in conducting his hiring analysis, by (a) neglecting to assess whether the companies in his benchmark industry to which he compared Cognizant, or the jobs performed at these companies, were truly comparable, and (b) failing to use actual applicant data in evaluating potential hiring disparities); (2) he did not control for any other variables that could explain the observed disparities in hiring and termination; and (3) he performed his analysis in an arbitrary way, by (a) categorizing certain termination codes as “involuntary” despite evidence to the contrary, and (b) treating Cognizant workers from India who transfer to the United States as “new hires” for purposes of measuring hiring disparities, but not treating these workers as “terminations” when they subsequently leave U.S. positions. Cognizant finally contends the report is not probative because the “broad statistical averages” which Dr. Johnson produces “say nothing about whether individual class members shared a common experience.” The Court addresses each contention in turn.
1. Proper Basis for Comparison In order to determine whether a larger share of Cognizant employees in the Class Bands are South Asian relative to other similar companies, Dr. Johnson compared the demographics of employees in the Class Bands at Cognizant with employment data from other companies in related industries. Using data from the American Community Survey (“ACS”), an annual survey conducted by the United States Census Bureau, he compared Cognizant’s demographics with those of other companies in the Computer Systems Design and Related Services Industry as designated by the North American Industry Classification System (“NAICS”). See Johnson Report at ¶¶ 34-25. Cognizant objects to Dr. Johnson’s use of the Computer Systems Design and Related Services Industry as a benchmark, without performing additional or more granular analysis of any similarities between Cognizant and the other companies included in that industry group, or similarities between the jobs included in the Class Bands and the jobs performed by employees included in the survey. Indeed, in his deposition, Dr. Johnson acknowledged that the ACS data, and the particular subset of that data on which he relied, “might not be an ideal comparator,” although he also stated that he “do[es]n’t know that there is an ideal comparator.” Johnson Depo. at 37:4-6. Cognizant contends that Dr. Johnson should have used Cognizant’s actual applicant pool as the basis for his analysis of hiring disparities.
At the outset, “there is no requirement that a statistical showing of disproportionate
impact must always be based on analysis of the characteristics of actual applicants.”
Bodnar v. Motorola, Inc.
,
Another district court in this circuit has already considered exactly this question in a case against one of Cognizant’s competitors, and declined to exclude the plaintiffs’ evidence. See Buchanan v. Tata Consulting Services, Ltd , No. 15-CV-01696-YGR, 2017 WL 6611653, at *11 (N.D. Cal. Dec. 27, 2017). In Buchanan , the court reasoned that because applicants to Tata Consulting Services (“TCS”), Cognizant’s competitor, might be aware of TCS’s allegedly discriminatory hiring practices, the applicant pool might not accurately reflect the potential pool of applicants. The court therefore rejected TCS’s argument that the only proper basis for comparison was the actual applicant pool. Plaintiffs’ expert in that case likewise relied on ACS data from the Computer Systems Design and Related Services Industry as his basis for comparison. The Buchanan court concluded that the issues raised by TCS went to weight, not admissibility. Id .
In this case, there is evidence that the actual applicant pool may have been tainted by Cognizant’s alleged practice of preferring visa holders in hiring. [ See, e.g. , Doc. # 256- 103 (internal email acknowledging Cognizant’s struggle to attract applicants who are green card holders or U.S. citizens).] In addition, the parties appear to agree that the data contained in Cognizant’s Taleo database does not capture all applicants, because “new hires” who are in fact overseas transfers from other Cognizant affiliates are not included. Thus, the Court rejects Cognizant’s objection to Dr. Johnson’s use of an industry proxy rather than Cognizant’s actual applicant pool.
The question remains, however, what the appropriate benchmark is. Cognizant
attacks Dr. Johnson’s choice of the ACS data regarding other companies in the Computer
Design and Related Services Industry, on the basis that this broad category contains too
many workers with jobs that are too different from those performed by Cognizant
employees in the Class Bands. Cognizant points to
Loeffel Steel Prod., Inc. v. Delta
Brands, Inc.
,
2. Failure to Control for Relevant Variables
Cognizant contends that Dr. Johnson’s failure to control for other possible variables
that might have been the reason for an employee’s termination renders his analysis
inadmissible. In particular, Cognizant contends that, had Dr. Johnson controlled for the
location preferences of applicants, his conclusions would have been quite different.
[21]
But
in
Obrey v. Johnson
, the Ninth Circuit rejected the contention that a study “based entirely
on statistical disparities” should be
excluded
in a Title VII “pattern or practice” case,
merely on the basis that it failed to account for other relevant factors that might have
influenced hiring.
3. Interpretation of Cognizant Data Cognizant also objects to Dr. Johnson’s treatment of certain terminations as “voluntary” or “involuntary.” Dr. Johnson coded “CDP – Location Preference” and “Job Abandonment” as involuntary terminations, and “Negative Reference/Background Check” as voluntary. The “CDP – Location Preference” code refers to when a benched employee is terminated because he or she has refused to relocate for a new position. The “Job willingness to relocate is a critical factor in bench terminations, for which Dr. Johnson should have controlled in his analysis. . at ¶ 135. Plaintiffs, and Dr. Johnson, specifically deny that Dr. Johnson provided an opinion as to
causation. To the extent Cognizant seeks to distinguish Obrey on the basis that the statistical analysis in that case did not opine on causation, the same is true here.
Abandonment” code is used when an employee stops coming to work and responding to outreach. The “Negative Reference/Background Check” code is used when a new hire is subsequently terminated because of problems with his or her background check or references. Cognizant contends that the first two reasons for termination are not involuntary, but are the result of choices the employee made. Cognizant argues the final reason should be treated as involuntary when it is not the result of a choice the employee made. Cognizant contends that Dr. Johnson’s decisions regarding how to treat these termination codes were arbitrary and illogical, and thus that Dr. Johnson’s opinion should be excluded.
The question of whether certain of these codes—in particular, “CDP – Location Preference,” which makes up the lion’s share of the terminations at issue here—should be coded as voluntary or involuntary is the subject of Plaintiffs’ motion for sanctions, discussed infra . But it is evident from the documents produced in this action that Cognizant’s own internal treatment of these codes is inconsistent. [ Compare Doc. # 307- 13 (chart created by Cognizant for this litigation indicating that CDP – Location Preference is a voluntary termination) with Doc. # 307-9 (internal summary of bench exits treating CDP – Location Preference as an involuntary termination) and Doc. # 307-10 (same).] Cognizant’s own inconsistent treatment of these codes at the very least undermines its contention that Dr. Johnson’s decision to treat these codes as involuntary was arbitrary or illogical. Dr. Johnson’s reasonable resolution of this question—which is evidently difficult for even Cognizant to resolve—does not render his opinion “unreliable nonsense,” and the Court declines to exclude Dr. Johnson’s report on this basis.
Cognizant also points out that Dr. Johnson’s report treats international Cognizant employees who transfer into U.S. roles as new hires for purposes of his analysis, but declines to include transferred workers who complete U.S. assignments in his termination analysis. Cognizant contends that this is an example of Dr. Johnson attempting to maximize South Asian representation in hiring, but minimize South Asian representation in terminations. Dr. McCrary performed the same analysis Dr. Johnson did, but coded “Assignment Completion” (the code used when a transfer employee completes an assignment in the U.S.) as an involuntary termination. Using Dr. McCrary’s analysis, in half of the Class Period years, non-South Asian employees were terminated at lower rates than South Asian employees, and in six of the eight Class Period years, non-South Asian employees were terminated from the bench at lower rates than South Asian employees. McCrary Report ¶¶ 140-41. The “Assignment Completion” code used in the McCrary Report does not appear in Dr. Johnson’s report, and Cognizant admits it does not appear on any of its lists of termination codes. In its Reply, Cognizant concedes that “Assignment Completion” is in fact “not, strictly speaking, a ‘termination’ code in the data.” Reply ISO Mot. to Exclude Dr. Johnson at 24 n.2 [Doc. # 330]. There is nothing arbitrary or illogical about treating a termination differently from the end of an assignment if the latter is not a termination. Cognizant’s failure to identify “Assignment Completion” as a termination code renders reasonable Dr. Johnson’s failure to treat it as such.
4. Other Bases for Exclusion Finally, Cognizant contends that Dr. Johnson’s report does not demonstrate commonality under the Rule 23(a) analysis because he performed an aggregate analysis, rather than disaggregating the different business units or job types within Cognizant. But Plaintiffs contend that Cognizant’s “visa readiness” and “visa utilization” policies existed across business units and job types throughout the Class Bands. In light of the evidence that the policies identified by Plaintiffs were company-wide, it was reasonable for Dr. Johnson to use aggregate, company-wide data.
The Court also disagrees with Cognizant’s characterization of Dr. Johnson’s report as “simple math.” Although Dr. Johnson’s analysis is not exceptionally complex, his report required the selection of relevant benchmarks for comparison and the aggregation and analysis of data in addition to the performance of statistical analysis to compute the standard deviation of the data analyzed from what is otherwise predicted by the benchmark data.
For these reasons, the Court DENIES Cognizant’s motion to exclude Dr. Johnson’s testimony.
V.
DAUBERT MOTION TO EXCLUDE TESTIMONY OF DR. RONIL HIRA
A. Discussion
Cognizant seeks to exclude Dr. Hira’s testimony on the grounds that it is (a) unhelpful to the trier of fact, since Plaintiffs cite it only for the conclusion that Cognizant has engaged in wage theft as to its Indian employees (a conclusion as to which, Cognizant contends, Dr. Hira is unqualified to offer an opinion), (b) methodologically unsound, and (c) is not Cognizant-specific. The Court discusses each objection in turn.
1. Helpfulness Cognizant contends that Dr. Hira’s conclusion, cited in Plaintiffs’ Motion for Class Certification, that Cognizant engages in “wage theft” is irrelevant, and that Dr. Hira’s report is unhelpful to the issue of class certification. While the Court agrees that Cognizant’s treatment of its H-1B-visa-holding employees is not the subject of Plaintiffs’ motion or lawsuit, many of Dr. Hira’s conclusions do offer additional support for Plaintiffs’ contention that Cognizant systematically prefers visa-holders. Dr. Hira’s conclusions about how a business model reliant on H-1B visa holders would help maximize profits lend support to Plaintiffs’ case for commonality because they help to explain why a company might pursue a “visa readiness” or “visa utilization” policy, and thus substantiate Plaintiffs’ contention that Cognizant had such a policy.
In addition, Dr. Hira’s opinion that cultural preferences—not only unconscious bias, but also supervisors’ preferences relating to the relationship between supervisors and supervisees—might lead managers to develop a preference for Indian workers even in local hiring is obviously relevant to Plaintiffs’ claims. Both of these opinions are relevant to class certification, and would assist a trier of fact in determining liability on a class-wide basis. The Court therefore concludes that Dr. Hira’s testimony is relevant, at least as to these issues. By contrast, many of Dr. Hira’s more Cognizant-specific conclusions are couched in language that reflects his limited study of Cognizant itself. See, e.g. , Hira Report ¶ 13 n.20 (noting that non-public documents “reflect” Cognizant’s disproportionate hiring of Indian employees, even in local hiring, and citing examples). These opinions are circumscribed in their scope, and in their persuasiveness. The Court does not exclude them on this basis, but these limitations may limit their weight.
2. Methodology “A trial court . . . has broad latitude in determining whether an expert's testimony is reliable [. . . and] in deciding how to determine the testimony's reliability.” Hangarter v. Provident Life & Acc. Ins. Co. , 373 F.3d 998, 1017 (9th Cir. 2004) (citations omitted) (affirming decision to admit expert testimony regarding claims adjustment standards based on the expert’s knowledge and experience) (emphasis in original).
Dr. Hira is a tenured professor of political science at Howard University, where he specializes in “outsourcing” and “offshoring.” See Hira Report, App. B ( curriculum vitae ). He has published extensively about these topics, over the course of several decades, in both academic and popular publications. See id. He holds a Ph.D. in public policy from George Mason University. Id . He has also testified before Congress a number of times and been solicited as an expert on the topics of outsourcing and offshoring by a range of entities. . His expert opinion does not cite extensively to academic literature or empirical data about outsourcing or offshoring. Instead, he appears to rely primarily on his own expertise and understanding in opining on the industry generally, and on a selection of public and non- public documents created by Cognizant for his opinions on Cognizant specifically.
Cognizant objects that Dr. Hira’s expert report is based only on a small selection,
handpicked by Plaintiffs’ counsel, of the documents produced in this case. But that is
untrue: Dr. Hira’s expert report is based on decades of research about the industry and the
visa regime under which Cognizant operates,
in addition to
the non-public, Cognizant-
specific documents produced in this action. Dr. Hira also relied on public information
about Cognizant, including Cognizant’s own 10-K and USCIS data about H-1B visa
approvals.
See, e.g.
, Hira Report ¶¶ 4-7. To the extent that the internal Cognizant
documents actually relied on by Dr. Hira are contradicted by other documents in the record,
that is a proper subject for cross-examination, not a basis for exclusion.
See Hangarter
,
Cognizant also contends that Dr. Hira’s testimony is distinguishable from that
offered in
Hangarter
and similar cases cited by Plaintiffs because the claims adjuster in
Hangarter
testified based on his own experience. But it is obviously not the case that an
expert must testify based either on studies produced following the scientific method or on
his own experience.
See, e.g.
,
Jinro Am. Inc. v. Secure Invs., Inc.
,
VI.
PLAINTIFFS’ MOTION FOR SANCTIONS
Plaintiffs move for sanctions on the basis that Cognizant manufactured evidence that the terminations code “CDP – Location Preference” is treated as a “voluntary” rather than “involuntary” termination internally. [Doc. # 298].
A. Background
A number of internal documents Cognizant produced in discovery treat “CDP – Location Preference” as an involuntary termination code. But a spreadsheet created by Cognizant as part of this litigation treats this code as a “voluntary” termination. Plaintiffs contend that Cognizant “fabricated” this evidence in order to transform these terminations from firings into resignations. This is significant for Plaintiffs because 82% of bench terminations at issue in this case—2,926 terminations—use this code as the reason for the employee’s termination. Plaintiffs request that the Court (a) enter default judgment as to liability on behalf of the proposed classes against Cognizant, (b) strike Cognizant’s briefs that rely on the allegedly falsified evidence, or (c) strike the arguments that rely on the allegedly falsified evidence.
1. Historic Treatment of Code Documents produced by Cognizant show that Cognizant has historically treated “CDP – Location Preference” as an involuntary termination code. For example, in a series of 2018 emails regarding the firm’s attempt to address a particular attrition metric, Cognizant employees describe “CDP – Location Preference” as one of the involuntary termination codes in “HCM,” and thus excludes employees terminated for that reason from the potential list of departing employees whose departures should be included for purposes of the metric. See Kotchen Decl. ISO Sanctions ¶ 5, Ex. 2 [Doc. # 298-3]. This treatment of the “Location Preference” code is reproduced across a range of evidence produced by Cognizant. See Kotchen Decl. ISO Sanctions, Ex. 3 (slide deck dated 2017) [Doc. # 298- 4], Ex. 4 (slide deck dated 2016) [Doc. # 298-5], Ex. 9 (slide deck dated 2019) [Doc. # 298-10], Ex. 11 (2015 emails describing 2014 terminations) [Doc. # 298-12], Ex. 13 (2017 email regarding attrition reporting) [Doc. # 298-14]. The HCM database does not include, however, a field that reflects whether a particular termination code is “voluntary” or “involuntary.”
2. Preparation of “Exhibit 25” The document at issue in Plaintiffs’ sanctions motion, which the parties refer to as Exhibit 25, was an exhibit discussed at the deposition of Eric Westphal. Plaintiffs explain that Cognizant created the document as part of a compromise to limit the burden on Cognizant in preparing Westphal for Cognizant’s Rule 30(b)(6) deposition. On July 17, 2021, Plaintiffs noticed a deposition of Cognizant’s person most knowledgeable concerning, among other issues, “the terminations data maintained by Cognizant . . . including but not limited to, the meaning of each termination reason/code (e.g., CDP-Lack of assignment, CDP-Other Performance Related, Conversion to CWR, Work Authorization Expired, Voluntary Separation Program, etc.), the process for entering the termination reasons into the system, the individual(s) who determines the reason for termination.” See Kotchen Decl. ISO Sanctions, Ex. 24 [Doc. # 298-25]. In advance of this deposition, Plaintiffs requested, and Cognizant agreed to prepare, a chart providing “an explanation of each code (including which code(s) indicate terminations from the bench), and whether each termination reason is voluntary or involuntary.” Id ., Ex. 26 [Doc. # 298-27]. A Cognizant Human Resources employee named Kerstin Aukerman created the chart, based in part on prior documents. Aukerman Decl. ¶¶ 4, 6 [Doc. # 328-1]. The specific columns included in the chart were included at the request of counsel. Id . at ¶ 3. She relied on her own understanding of the codes, and where she was unsure about how to describe a code, she confirmed with Sharmilee Santhanaraman, the Director of Human Resources. Id . at ¶ 7.
The chart that resulted is Exhibit 25, which provides a description of Cognizant’s termination codes and identifies them as “voluntary” or “involuntary.” See Kotchen Decl. ISO Sanctions, Ex. 25 [Doc. # 298-26]. Cognizant describes “CDP – Location Preference” as the code used when “the offered Cognizant job required relocation which the employee chose not to accept,” and categorized it as “voluntary.” .
Eric Westphal was deposed as Cognizant’s person most knowledgeable pursuant to Federal Rule of Civil Procedure 30(b)(6). In his deposition, he relied on Exhibit 25 for his explanation of Cognizant’s “Location Preference” code. See Kotchen Decl. ISO Sanctions, Ex. 39 (excerpts from Westphal Depo.) at 7:12-18 [Doc. # 298-40]. Just before, he had testified that “a voluntary separation is someone, you know, an associate resigning of their own volition,” and that an involuntary separation is “being separated for performance or . . . going through the bench process and not finding another project, you may be released.” . at 4:24-5:9. Plaintiffs did not ask Westphal about Exhibit 25’s treatment of the “Location Preference” code as a voluntary termination code.
B. Legal Standard
District courts have “inherent power” to impose sanctions in order to manage their
cases, ensure the orderly administration of justice, and enforce compliance with
orders.
ChromaDex, Inc. v. Elysium Health, Inc.
, 535 F. Supp. 3d 906, 911 (C.D. Cal.
2021) (citing
Int'l Union, United Mine Workers of Am. v. Bagwell
, 512 U.S. 821, 831
(1994). “That authority includes the ability to fashion an appropriate sanction for conduct
which abuses the judicial process.”
Goodyear Tire & Rubber Co. v. Haeger
, 137 S. Ct.
1178, 1186 (2017). But a court imposing sanctions pursuant to its inherent power may
impose sanctions based only on a finding of either “a willful violation of a court order [or
. . .] bad faith.”
Am. Unites for Kids v. Rousseau
,
C. Discussion
Plaintiffs’ sanctions motion fails on the simple ground that there is no evidence of bad faith in Cognizant’s creation of Exhibit 25. Based on the evidence presented in support of Plaintiffs’ motion—and Cognizant’s failure to present contradictory evidence—it appears that Cognizant uniformly treats terminations for “Location Preference” as involuntary internally. Cognizant’s characterization in Exhibit 25 of these terminations as voluntary flies in the face of the weight of the evidence, but that does not render it falsified . To the extent Cognizant relies on Exhibit 25 in its papers to raise doubts about Dr. Johnson’s expert opinion, the Court has already addressed that issue. See Section IV.B.4, supra (discussing impact of Cognizant chart on admissibility of Dr. Johnson’s analysis).
In its Opposition to Plaintiffs’ Motion for Class Certification, Cognizant relies on Exhibit 25 for the idea that “Location Preference” is “in fact” a voluntary termination. See Class Cert. Opp. at 34 (“’location preference’ . . . is, in fact, a voluntary termination based on the individual’s decision to reject relocation”) (citing to Exhibit 25); see also id. at 36 n.7 (citing to Exhibit 25 for the proposition that “Location Preference” and other codes are voluntary termination codes), 24 (citing to Exhibit 25 for the idea that “there is extensive variation in why an employee may terminate while on the bench”). In light of the circumstances of the document’s creation for this litigation, the Court does not consider Exhibit 25 to be particularly persuasive evidence of the variety of circumstances under which a “Location Preference” termination could occur. The Court does not find evidence, however, of bad faith in Cognizant’s creation and use of this document. The Court thus DENIES Plaintiffs’ motion for sanctions. The Court will neither strike Cognizant’s briefs nor its arguments, but will take Cognizant’s evidence for what it is worth.
VII.
MOTION FOR PARTIAL SUMMARY JUDGMENT
Cognizant moves for partial summary judgment on Edward Cox’s disparate treatment and disparate impact claims. MSJ [Doc. # 266].
A. Undisputed Facts
Edward Cox was neither of South Asian race nor of Indian national origin. See SUF 1-2. He was a Director at Cognizant from January 2014 to April 3, 2017. SUF 4. He was terminated from the bench. SUF 42. In November 2017, David Stark, another Director at Cognizant, contacted Edward about a potential non-billable role at Cognizant as a Regional Development Center ECCS Leader in Dallas. SUF 5-6. At that time, Cognizant was establishing a large employment center in Dallas. SUF 44-45. In December 2017, Edward interviewed with several members of Stark’s team, and was told Cognizant intended to make him an offer. SUF 51-52. Edward never received a formal offer letter. SUF 11. On January 10, 2018, a Cognizant employee emailed Stark concerning Edward’s potential role, seeking confirmation of various information, but later that day stated he “could not approve” the offer. SUF 55. Also in early 2018, Stark’s supervisors decided to look at internal candidates before external candidates were considered. SUF 14; see also Smith Decl. ISO MSJ Reply ¶ 3 (asserting that Cognizant responded to an interrogatory by stating that leadership subsequently “informed Mr. Stark that, in any event, he would need to consider the bench and other internal resources for any such role before seeking an external hire”) [Doc. # 312-4]. Stark was instructed to “disengage” from Edward. SUF 58. But Cognizant continued to look for a candidate to fill the role. SUF 71. A Cognizant supervisor identified an internal candidate who might be a potential fit for the role, but that person never formally applied and was not hired. Swallows Decl. ¶ 9 [Doc. # 266-1]. Cognizant ultimately did not create the job for which Cox was being considered. SUF 19. Cognizant did continue to recruit for other positions at the Director and Assistant Director level at Cognizant in the Dallas area. See SUF 61. Edward Cox died on March 28, 2021. SUF 3.
B. Legal Standard
Summary judgment should be granted “if the movant shows that there is no genuine
dispute as to any material fact and the movant is entitled to judgment as a matter of law.”
Fed. R. Civ. P. 56(a);
accord Wash. Mut. Inc. v. United States
,
The moving party bears the initial burden of establishing the absence of a genuine
issue of material fact.
Celotex Corp. v. Catrett
, 477 U.S. 317, 323 (1986). Once the
moving party has met its initial burden, Rule 56(c) requires the nonmoving party to “go
beyond the pleadings and by her own affidavits, or by the ‘depositions, answers to
interrogatories, and admissions on file,’ designate ‘specific facts showing that there is a
genuine issue for trial.’”
Id.
at 324 (quoting Fed. R. Civ. P. 56(c), (e));
see also Norse v.
City of Santa Cruz
,
C. Discussion
Plaintiff Brian Cox, on behalf of his father, asserts claims for disparate treatment under both Section 1981 and Title VII and for disparate impact under Title VII. Cognizant moves for summary judgment on all three claims.
1. Proper Framework
As an initial matter, the parties disagree about whether Cox’s disparate treatment
claim should be analyzed under the
McDonnell Douglas v. Green
,
Both Teamsters and McDonnell Douglas provide a blueprint for a burden-shifting framework to adjudicate claims of employment discrimination under Title VII. Under the Teamsters framework, plaintiffs asserting a pattern-or-practice claim must first establish that a pattern or practice of discrimination pervaded at a company. As the Supreme Court confirmed in Wal-Mart Stores, Inc. v. Dukes ,
In a pattern-or-practice case, the plaintiff tries to establish by a preponderance of the evidence that discrimination was the company's standard operating procedure, the regular rather than the unusual practice. If he succeeds, that showing will support a rebuttable inference that all class members were victims of the discriminatory practice, and will justify an award of prospective relief, such as an injunctive order against continuation of the discriminatory practice.
Dukes
,
The McDonnell Douglas framework provides a burden-shifting framework that applies in individual cases. Under McDonnell Douglas , a plaintiff establishes his prima facie case by showing that:
(1) the plaintiff belongs to a protected class, (2) he was performing according to his employer's legitimate expectations, (3) he suffered an adverse employment action, and (4) similarly situated employees were treated more favorably, or other circumstances surrounding the adverse employment action give rise to an inference of discrimination.
Reynaga v. Roseburg Forest Prod
.,
In all the cases cited by either party, courts have rejected the application of the
McDonnell Douglas
framework to a motion for summary judgment against proposed class
representatives asserting pattern-or-practice claims of discrimination. In
Colindres v.
Quietflex Mfg.
, No. CIV. A. H-01-4319,
the existence of a valid individual claim does not necessarily warrant the conclusion that the individual plaintiff may successfully maintain a class action. It is equally clear that a class plaintiff's attempt to prove the existence of a companywide policy, or even a consistent practice within a given department, may fail even though discrimination against one or two individuals has been proved.
In
Buchanan
, an employment discrimination class action against Cognizant’s
competitor TCS (asserting similar claims based on similar factual allegations), the district
judge likewise denied as premature a motion for partial summary judgment on the named
plaintiff’s individual claim, shortly after certifying the named plaintiff as a class
representative.
See Buchanan
, 2017 WL 6611653, at *23 (citing
Colindres
, 2004 WL
3690215, at *6).
Accord Thiessen v. Gen. Elec. Cap. Corp.
,
Another district court has concluded, however, that it is proper to consider a motion
for summary judgment as to a named plaintiff’s claim in a proposed
Teamsters
class action
where defendants moved for summary judgment before a class had been certified. In
Brewer v. Holder
, 20 F. Supp. 3d 4 (D.D.C. 2013), the district court adjudicated the
defendant’s motion for summary judgment as to the named plaintiffs’ pattern-or-practice
claims before ruling on class certification. The
Brewer
court reasoned that “[i]f a specific
plaintiff has not suffered the injury as a result of the alleged discriminatory practice, then
he or she is ‘simply not eligible to represent a class of persons who did allegedly suffer
injury.’”
In light of the pending motion for class certification currently before the Court, and the Court’s duty to carefully scrutinize the adequacy of the proposed class representative (especially under the unusual circumstances presented by this case), the Court concludes that it is proper to consider Cognizant’s motion for partial summary judgment at this stage. Like the court in Brewer , however, the Court will proceed as if Plaintiffs had succeeded in establishing liability under Phase One of the Teamsters framework, and thus established a presumption that Cognizant’s hiring decisions were discriminatory. Yet, as noted herein, the Court would have reached the same conclusions proceeding under the McDonnell Douglas framework.
2. Disparate Treatment Claim
“To establish disparate treatment under Title VII, a plaintiff must offer evidence
that gives rise to an inference of unlawful discrimination, either through the framework set
forth in [
McDonnell Douglas
] or with direct or circumstantial evidence of discriminatory
intent.”
Freyd v. University of Oregon
,
In general, showing that there was in fact a vacancy is necessary to establish an
employment discrimination claim. In
Chavez v. Tempe Union High Sch. Dist. No. 213
,
565 F.2d 1087 (9th Cir. 1977), and
Valent v. Summers
, 208 F.3d 216 (6th Cir. 2000)
(unpublished), the district court concluded (and the court of appeal affirmed) that, where
the job plaintiff sought had already been filled at the time the plaintiff applied (albeit
without the plaintiff’s knowledge), the plaintiff’s claim for hiring discrimination could not
go forward.
See Chavez
,
Still, under the particular circumstances of this case, the Court concludes that the
fact that the position was never actually filled is not dispositive. This is because, after
rejecting Edward Cox, Cognizant
did
continue to consider another applicant.
See
Swallows
Decl. ¶ 9. Although the existence of a job vacancy is a critical element of an employment
discrimination claim, this is because the lack of a vacancy is one of the “most common
legitimate reasons on which an employer might rely to
reject
a job applicant.”
See
Teamsters
,
The question remains, however, whether Cognizant continued to consider other individuals with Edward Cox’s qualifications . It is uncontroverted that the other individual Cognizant considered was an internal employee, which Cognizant contends renders that individual better-qualified than Edward Cox.
Cognizant has provided a plausible explanation for its preference for internal
candidates and, in general, a preference for internal candidates may constitute a legitimate
and non-discriminatory reason for declining to hire a particular candidate.
See, e.g.
,
Robinson v. PPG Indus., Inc.
, Case No. CV 19-4033-ODW (RAOx),
3. Disparate Impact Claim
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
,
that a facially neutral employment practice has a significantly discriminatory impact upon a group protected by Title VII. This showing consists of two parts: the plaintiff 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
,
Cognizant contends that Brian Cox lacks standing to pursue his disparate impact claim because he cannot show that his father suffered an injury-in-fact as a result of Cognizant’s failure to rehire him—that is, he cannot show causation. Cognizant’s basis for this argument is, again, that (a) Cognizant decided to hire an internal candidate, and thus Edward Cox’s race or national origin must not have been the cause for Cognizant’s failure to hire him, and (b) Cognizant never in fact hired anyone for the position. Cognizant also contends, for the same reasons, that Brian Cox cannot make out a prima facie disparate impact claim because he cannot demonstrate that Cognizant’s allegedly discriminatory practice caused Edward Cox’s injury. For the reasons discussed supra with respect to the disparate treatment claim, the Court concludes that a triable issue of fact remains as to whether Cognizant’s rejection of Cox was the result of discrimination. Accordingly, the Court DENIES Cognizant’s MSJ as to disparate impact.
VIII.
MOTION FOR CLASS CERTIFICATION
A. Legal Standard
Federal Rule of Civil Procedure 23 provides the standard for certification of a class
action. Plaintiffs must meet all of the requirements under Rule 23(a) and must also satisfy
at least one of Rule 23(b)’s requirements.
Valentino v. Carter-Wallace, Inc.
,
Courts refer to the Rule 23(a) requirements by the following shorthand:
“numerosity, commonality, typicality and adequacy of representation.”
See Mazza v.
American Honda Motor Co., Inc.
, 666 F.3d 581, 588 (9th Cir. 2012). If the four
prerequisites of Rule 23(a) are satisfied, a court must also find that the plaintiffs can “satisfy
through evidentiary proof” at least one of the three subsections of Rule 23(b).
Comcast
Corp. v. Behrend
,
Rule 23 is more than a pleading standard, and it requires the party seeking class
certification to “affirmatively demonstrate his compliance with the Rule.”
Dukes
, 564 U.S.
at 350 (internal quotation omitted). Thus, a court must conduct a “rigorous” class
certification analysis.
Id.
at 350-51 (quoting
Gen. Tel. Co. of Sw. v. Falcon
,
Here, the Court will address numerosity, then commonality and predominance together, and then will turn to adequacy, typicality, and superiority.
B. Numerosity
Plaintiffs satisfy the numerosity requirement, because Cognizant does not contest that there are sufficiently numerous proposed class members that joinder of all members would be impracticable. See Class Cert. Opp. [Doc. # 272-1].
1. Hiring Class The parties agree that the proposed Hiring Class contains approximately 169,000 applicants. See Class Cert. Opp. at 15. This decidedly satisfies numerosity as to the proposed Hiring Class.
2. Terminations Class Plaintiffs’ evidence shows that 4,000-4,200 non-South Asian employees were involuntarily terminated from the Class Bands during the Class Period. Johnson Report ¶ 48. This meets the numerosity requirement as to the proposed Terminations Class.
3. Bench Terminations Subclass Plaintiffs’ evidence shows that 2,200-2,300 non-South Asian employees were involuntarily terminated from the bench from the Class Bands during the Class Period. Johnson Report ¶ 48. This too satisfies numerosity as to the proposed Bench Terminations Subclass.
C. Commonality and Predominance
Because the commonality and predominance factors go hand in hand, the Court will
consider them together. The commonality requirement is satisfied if “there are questions
of law or fact common to the class.” Fed. R. Civ. P. 23(a)(2). “Commonality requires the
plaintiff to demonstrate that the class members ‘have suffered the same injury.’”
Dukes
,
In an employment discrimination case, plaintiffs seeking certification of a class
should identify a
specific
employment practice that is challenged.
See Dukes
, 564 U.S. at
357 (emphasis added). The employment practice must “touch and concern all members of
the class.”
Id
. at 359 n.10;
see also Ellis
,
Once plaintiffs have established that common questions exist, they must show that those common questions will predominate over individualized issues. The predominance inquiry “tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.” Tyson Foods, Inc. v. Bouaphakeo , 136 S. Ct. 1036, 1045 (2016) (internal citations and quotations omitted). This requires “careful scrutiny” of “whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues.” Id . (internal citations and quotations omitted). If “one or more of the central issues in the action are common to the class and can be said to predominate, the action may be considered proper under Rule 23(b)(3) even though other important matters will have to be tried separately, such as damages or some affirmative defenses peculiar to some individual class members.” .
“Considering whether questions of law or fact common to class members
predominate begins, of course, with the elements of the underlying cause of action.”
See
Olean
,
1. With respect to each class/subclass, did Cognizant engage in a pattern or practice of discriminating on the basis of race and/or national origin? If so, does Cognizant’s conduct meet the standard for an award of punitive damages?
2. With respect to each class/subclass, did Cognizant’s Visa Readiness Policy and/or Visa Utilization Policy (and related practices) disparately impact class members? If so, was each such policy justified by a business necessity that could not have been accomplished by a less discriminatory alternative? Plaintiffs contend that the first question should be resolved by a jury and the second question by the Court.
1. Hiring Class Two other district courts in this circuit have considered class certification motions in similar cases, and both have declined to certify a hiring class on the basis that common questions did not predominate over individualized issues.
In
Handloser v. HCL Technologies, Ltd.
, the plaintiffs asserted claims against a
Cognizant competitor, HCL Technologies, based on similar allegations: the plaintiffs
asserted that HCL’s use of an alleged “visa prioritization” policy, combined with an alleged
policy of screening out non-South Asian candidates through “culture fit” interviews,
resulted in “gross statistical disparities” that demonstrated discrimination.
Handloser v.
HCL Techs. Ltd.
, No. 19-CV-01242-LHK, 2021 WL 879802, at *6 (N.D. Cal. Mar. 9,
2021). The court concluded that the plaintiffs had not established commonality “for at
least seven reasons.” . For a number of positions, the court concluded that the
candidate’s race or national origin could not have been the uniform reason for their
rejection: (1) over 1,000 job requests during the proposed class period specifically
excluded visa holders from consideration, (2) for roughly 50% of job requests during the
proposed class period, HCL did not fill an open position with any candidate because the
client withdrew the request or the client filled the request with a non-HCL candidate, (3)
for as many as two thirds of relevant job requests, HCL’s client “[would] have reached
their own, independent decision regarding an applicant.”
Id
. at *7. The court also found
that, because the hiring process for different jobs varied so widely, there was insufficient
“glue” holding the class together.
Id
. a *7-8 (citing
Dukes
,
At most, Plaintiffs point to the fact that Defendants utilize a shared database for internal job requests . . . and a standard process by which internal job requests are processed and filled. . . . However, the fact that Defendants utilize a shared structure for considering job candidates and a shared database for logging such requests does not establish that Defendants use a centralized or standardized decision-making process with respect to employment decisions.
Id . (citations omitted).
In
Buchanan
, the district court certified a class of individuals who were terminated
from the bench at TCS, a Cognizant competitor, but declined to certify a hiring class.
See
Plaintiffs here point to policies similar to those described by the plaintiffs in
Handloser
and
Buchanan
. In both those cases, the court concluded that similar visa
readiness and visa utilization policies were insufficient to tie together a hiring class. In this
case, Plaintiffs have one additional piece of important evidence: the EEOC has concluded
that Cognizant discriminated against a class of non-South Asian individuals who were not
hired by Cognizant. “An EEOC determination, prepared by professional investigators on
behalf of an impartial agency, has been held to be a highly probative evaluation of an
individual's discrimination complaint.”
Plummer v. W. Int'l Hotels Co.
,
There are an enormous number of hiring decisions at issue in this case. Even if every hiring manager was bound by the “visa utilization” and “visa readiness” policies, it is clear that additional factors would have gone into a large number of the hiring decisions at issue in this case. These additional factors include basic questions, such as whether an individual met the minimal qualifications for the position in question, and more complex questions, such as the quality of the applicant’s interview and how much influence the client had over an applicant’s rejection. Like in Handloser , a number of positions are not open to visa- holding employees, and for project-based jobs, the client—not Cognizant—determines whether a position may be taken by a visa-holding employee. See Westphal Decl. ISO Class Cert. Opp. ¶ 18. At least 40% of positions require a client interview. See id . at ¶ 14. In other words, even if a factfinder concluded that Cognizant did have the “visa utilization” and “visa readiness” policies that Plaintiffs allege, that determination would not resolve an element of either Plaintiffs’ disparate treatment or disparate impact claims, because individualized inquiries would remain as to whether one of these other issues was the intervening cause of Cognizant’s failure to hire any given class member. And like in Handloser (and other nationwide employment discrimination class actions, including Dukes ), the sheer size of the class here makes it more difficult to say that the same common policy could have impacted each hiring decision.
Plaintiffs contend that, unlike in Handloser , Cognizant’s visa readiness and visa utilization policies serve a gatekeeping function, and exclude certain candidates even from consideration. For this reason, Plaintiffs argue, questions about an applicant’s performance in an interview or the role a client played in a hiring decision would not cut off the causal chain between Cognizant’s allegedly discriminatory behavior and the harm to potential class members.
Although this argument appears compelling at first blush, it is undermined by the facts of this case. Cognizant’s expert calculated that 59% of the proposed hiring class members only applied for a position for which a non-South Asian applicant was hired. See McCrary Report ¶ 119. This means that these proposed class members applied only for positions for which either no South Asian applicant was even considered, or for which non- South Asian applicants competed successfully against South Asian applicants. This statistic helps to rebut Plaintiffs’ contention that Cognizant’s allegedly discriminatory practices screened out non-South Asian applicants. It also makes clear that the question of how Cognizant’s alleged visa readiness and visa utilization policies might have influenced proposed class members’ hiring is quite individualized.
The Court therefore concludes that individualized issues will predominate over common questions as to the proposed hiring class.
2. Terminations Class The proposed terminations class is significantly smaller than the proposed hiring class: only 4,000-4,200 individuals. The risk of individualized issues predominating over common questions is also lessened by the limitation of the proposed class to involuntary terminations. Still, the Court again concludes that individualized issues are likely to predominate over common questions. In particular, Plaintiffs have adduced no evidence that the alleged visa policies infected Cognizant’s terminations of non-South Asian individuals for poor performance, or that Cognizant disproportionately entered into separation agreements with non-South Asian employees. The Johnson Report shows that, during the Class Period, 42% of involuntary terminations were for “CDP – Location Preference.” The next most common reason was “Separation Agreement” (28.3%), “CDP – Skill Mismatch” (6.6%), then “Misconduct” (3.9%), “Job Abandonment” (3.5%), and “Unsatisfactory Performance” (3.4%). Johnson Report, Fig. 5. It is clear from this data that bench terminations made up the bulk of involuntary terminations, but that reasons less plausibly tied to Cognizant’s visa policies drove many of the remaining terminations. The Court thus concludes that individualized issues will predominate over common questions as to the proposed terminations class.
3. Bench Terminations Subclass Plaintiffs have shown, however, that common questions will predominate over individualized issues as to the proposed Bench Terminations Subclass.
This proposed class is a more manageable size—fewer than 3,000 individuals—and
has unique features that make a determination of liability well-suited to classwide
adjudication. The proposed class is limited to individuals who were involuntarily
terminated from the bench, which avoids the potential need for individualized inquiries
about whether an individual was terminated for poor performance. These individuals’
failure to find a new placement is more plausibly related to Cognizant’s visa policies, while
avoiding potential questions about their basic qualifications, since these class members had
already been hired by and worked for Cognizant by the time they were placed on the bench.
Plaintiffs have adduced evidence that Cognizant worked to avoid the placement of visa-
holding individuals on the bench at all, and also created systems to move those employees
off the bench quickly.
[36]
The EEOC’s determination that Cognizant discriminated against
non-South Asian individuals in making bench terminations also weighs in support of
certification (although, for the reasons stated above regarding the proposed hiring class,
the EEOC determination is not dispositive).
[37]
Accord Buchanan
,
Plaintiffs’ alleged visa readiness and visa utilization policies are the type of facially
neutral policies that may support a disparate impact claim under Title VII.
See, e.g.
,
Ricci
v. DeStefano
, 557 U.S. 557, 576-77 (2009). Plaintiffs have also adduced substantial
evidence of discriminatory intent. The facts that, at least during one part of the proposed
class period, most management positions were held by South Asians, employees who did
not speak the same language as or share the culture of the majority of employees were
excluded from participation in meetings and social events, and the evidence of an internal
understanding that the reason the visa program was part of Cognizant’s business model
was because H-1B employees—99% of whom were South Asian—“were cheaper labor
that would relocate anywhere and everywhere and as often as needed,” all support an
inference of intentional discrimination. In addition, the simple fact that 99% of employees
with visas in the U.S. were South Asian supports an inference of, but does not alone prove,
intentional discrimination.
See, e.g.
,
Ave. 6E Invs., LLC v. City of Yuma, Ariz.
, 818 F.3d
493, 507–08 (9th Cir. 2016) (on motion to dismiss, “statistics on the disparate impact
caused by [a city’s] decision,” in conjunction with other factors, “tend to make the
disparate-treatment claims plausible.”);
Schechner v. KPIX-TV
,
that treats this as a “voluntary” reason for termination was created for this litigation, a fact that significantly reduces its probative value.
Cognizant contends that individualized inquiries will predominate because
Cognizant will have the right to present evidence that any policies or practices which
Plaintiffs are able to establish were justified by a business necessity. This argument does
not cut, however, against finding that common questions will predominate over
individualized issues: this is an element of all the class members’ claims that can be
resolved with common evidence, including a common defense such as business necessity.
“When one or more of the central issues in the action are common to the class and can be
said to predominate, the action may be considered proper under Rule 23(b)(3) even though
other important matters will have to be tried separately, such as damages or some
affirmative defenses peculiar to some individual class members.”
Olean
,
Cognizant also contends that individualized issues will predominate because
Cognizant will still be entitled to show that each class member was denied an employment
opportunity (or, here, terminated) for lawful reasons. Under
Teamsters
, these issues are
deferred to Phase Two, after the common question of whether Cognizant proceeded under
a pattern or practice of discrimination has been answered.
See Teamsters
,
Cognizant argues, however, that a recent Ninth Circuit decision,
Bowerman v. Field
Asset Servs., Inc.
,
On appeal, the plaintiffs’ counsel did not contest the defendant’s assertion that the class could not prove through common evidence that class members had actually worked overtime hours or had reimbursable business expenses, and the Ninth Circuit noted that at the bellwether trial, the class had relied primarily on the testimony of the class members and not on contemporaneous documentary evidence of overtime hours or business expenses. 39 F.4th at 661. The Ninth Circuit reversed the class certification decision because even if plaintiffs could prove through common evidence that the class members had been misclassified, they could not prove through common evidence that the defendant had a common policy of requiring overtime work or requiring class members to incur unreimbursed necessary business expenses. Id . at 662. Moreover, even if plaintiffs had shown they were capable of establishing liability on a classwide basis in Bowerman (which the Ninth Circuit held they were not), the facts of the bellwether trial also showed that the mere calculation of damages would have required “excessive difficulty,” because proving damages by individual class members’ testimony years after the alleged overtime work occurred proved unwieldy. . at 663.
Bowerman is simply inapposite here, because the Teamsters framework does not divide employment discrimination cases into a liability phase and a damages phase: it divides such cases into two liability phases, with the second phase expressly intended to resolve individualized defenses to liability. There is no dispute that in this case, and in all cases proceeding under the Teamsters framework, Phase Two will involve liability determinations, and Teamsters expressly authorizes such a division. The result of Bowerman does not counsel against finding predominance and commonality here. Cognizant’s real contention is that consideration of these individualized issues at Phase Two will be unmanageable. The Court will address this issue in its discussion of superiority, infra .
a. Statute of Limitations Cognizant finally contends that Plaintiffs’ proposed class is limited by the statute of limitations. Title VII requires that discrimination claims be filed with the EEOC within 180 days of the alleged violation, or 300 days if the charging party files his or her charge with an appropriate state agency, or when the EEOC refers the charge to a relevant state agency. See 42 U.S.C. § 2000e-5(e); see also Douglas v. California Dep't of Youth Auth. , 271 F.3d 812, 823 n.12 (9th Cir.), amended , 271 F.3d 910 (9th Cir. 2001) (describing application of the same statutory provision to ADA claims). See 42 U.S.C. § 12117(a). Cognizant seeks to limit Plaintiffs’ proposed class to violations that occurred within the 300-day period preceding Edward Cox’s filing of his EEOC charge. See Opp. at 29 n.4.
Plaintiffs do not respond to this contention in their Reply, which constitutes a
concession of that argument.
See Ramirez v. Ghilotti Bros. Inc.
,
4. Punitive Damages
Plaintiffs also seek certification as to the availability of punitive damages. To
recover punitive damages for a Title VII claim, “an employer must at least discriminate in
the face of a perceived risk that its actions will violate federal law.”
Kolstad v. Am. Dental
Ass'n
, 527 U.S. 526, 536 (1999). Plaintiffs’ evidence that Cognizant proceeded in its
course of conduct despite warnings from employees responsible for affirmative action
demonstrates the existence of common questions as to this issue as well. Moreover,
because the question of the availability of punitive damages focuses on the defendant’s
conduct, individualized issues will not predominate as to liability.
Accord Ellis v. Costco
Wholesale Corp.
,
At the hearing on this matter, Cognizant argued that certification of the availability
of punitive damages would violate Cognizant’s due process rights under
State Farm Mut.
Auto. Ins. Co. v. Campbell
,
D. Typicality
Typicality requires a showing that “the claims or defenses of the representative
parties are typical of the claims or defenses of the class.” Fed. R. Civ. P. 23(a)(3). The
purpose of this requirement “is to assure that the interest of the named representative aligns
with the interests of the class.”
Wolin v. Jaguar Land Rover N. Am., LLC
,
1. Brian Cox Edward Cox suffered a similar injury to that of other members of the Bench Terminations Subclass in that he was terminated from the bench after failing to secure a new position, allegedly as a result of Cognizant’s visa policies. Brian Cox has been properly substituted as a Plaintiff in this action, and stands in his father’s shoes. The Court has found no case law suggesting that Brian Cox is anything but an appropriate class representative.
2. Vartan Piroumanian Piroumanian also suffered a similar injury to that of other members of the Bench Terminations Subclass in that he was terminated from the bench after failing to secure a new position. Cognizant contends, however, that Piroumanian cannot represent the class for two reasons.
First, Cognizant notes that the Court previously dismissed for lack of exhaustion Piroumanian’s disparate impact claims arising out of Cognizant’s practice of applying for visas. See First MTD at 7 [Doc. # 42]; Second MTD at 3 [Doc. # 58]. But the Court concluded that Piroumanian had properly exhausted his claims relating to termination from the bench. See First MTD at 7-8. Because the Court certifies only a bench termination class, and because Piroumanian’s claims regarding termination from the bench did survive Cognizant’s motions to dismiss, the Court does not find that Piroumanian is inadequate on this basis.
Second, Cognizant argues that Piroumanian is an atypical class representative because of comments he made at the time of and after his termination that call into question his motives in pursuing his claim. Cognizant highlights (a) Piroumanian’s statement shortly before his termination from the bench that he would be at Cognizant for “another four weeks,” and that he intended to participate in this lawsuit, and (b) derogatory statements about Indians Piroumanian has made.
Cognizant argues that Piroumanian’s email statement while he was on the bench that he would be at Cognizant for another four weeks indicates that he was not trying to find a new position at Cognizant while he was on the bench and that, instead, he allowed himself to be terminated from the bench in order to manufacture standing to bring this lawsuit. The Court does not find this reasoning persuasive. Even if Piroumanian did make a conscious choice not to actively apply for new positions within Cognizant while he was on the bench—and this email is scant evidence that he did—there is no indication that he turned down a position that he was offered. Indeed, Plaintiffs contend that South Asian employees are less likely to be terminated from the bench even if they merely sit on their hands than are non-South Asian employees, because Cognizant works harder to place its visa-holding South Asian employees from the bench. Piroumanian’s termination from the bench does not make him atypical for purposes of Rule 23, even if it was for lack of trying to get off the bench. Cf. Nghiem v. Dick's Sporting Goods, Inc. , 318 F.R.D. 375, 382 (C.D. Cal. 2016) (finding atypical the claim of a TCPA lawyer who opted into text alerts in order to manufacture a claim under the TCPA because there was too much danger the plaintiff would be “preoccupied” with defenses unique to him, such as whether he opted in in good faith and whether he suffered an actionable invasion of privacy).
Cognizant also adduces evidence that Piroumanian made several derogatory comments about Indian individuals and companies “run by Indians” in 2017 and 2019. In a 2017 email sent from Piroumanian’s Cognizant email to Franchitti, Piroumanian called a South Asian colleague who was asking for receipts to support a request for reimbursement (and who had apparently previously given Piroumanian a poor review) a “mother fucker” and “a loser.” Smith Decl. ISO Class Cert. Opp., Ex. 21 [Doc. # 273-6]. In a 2019 email, apparently unhappy with a response from a recruiter seeking to recruit him for a position, he wrote “I asked you a simple question. You should answer it. I’m not one of your Indian visa workers.” See id. , Ex. 22 [Doc. # 273-7]. In another 2019 email, likewise unhappy with the hourly rate offered by a recruiter, he responded “Ha ha ha ha . . . !!! Insulting . . . Must be run by Indians . . . Hey dude, Namaste . . .” ., Ex. 23 [Doc. # 273-8]. Cognizant contends that these statements call into question Piroumanian’s motivation in bringing his claim here, and could derail the class’ case by injecting the issue of Piroumanian’s motivation into this action. Piroumanian’s comments reflect poorly upon him, but they do not appear to create the possibility of a conflict within the class. The Court therefore concludes that Piroumanian’s claims are typical of the class claims.
E. Adequacy
The Rule 23(a)(4) adequacy determination turns on the answers to two questions:
“(1) do the named plaintiffs and their counsel have any conflicts of interest with other class
members and (2) will the named plaintiffs and their counsel prosecute the action vigorously
on behalf of the class?”
Hanlon v. Chrysler Corp.
,
Cognizant contends that Brian Cox is not an adequate representative for either the Hiring Class or the Terminations Class because (a) he lacks the requisite knowledge to adequately represent the classes, and (b) he (and his stepmother and father before him) abdicated all control of this case to counsel. The Court considers only Cox’s adequacy as a representative of the Bench Terminations Subclass.
There does not appear to be any conflict between Brian Cox and the proposed Bench Terminations Subclass, and it is clear that Plaintiffs and proposed Class Counsel have and are likely to continue to vigorously litigate this action. In a May 17, 2022 filing, Brian Cox avers that he spoke at length with his stepmother about his father’s claim before she passed away. Brian Cox Decl. ISO Notice of Impending Death ¶ 2 [Doc. # 257-1]. He attests that he intends to pursue his father and stepmother’s wishes to pursue this action to its conclusion. . at ¶¶ 2-3. In a supplemental declaration filed in support of Plaintiffs’ Reply in Support of Class Certification, Brian Cox supplies additional facts regarding his knowledge of this action. [ See Doc. # 323-30.] Moreover, most of the basic facts pertinent to Edward Cox’s claim are undisputed. Those facts that are disputed pertain to the class as a whole. In light of these circumstances, the Court concludes that Brian Cox is an adequate class representative.
Cognizant also contends that proposed Class Counsel is inadequate because of various deficiencies in their representation of Plaintiffs thus far. The Court is satisfied that proposed Class Counsel are adequate. Kotchen & Low and Yadegar, Minoofar & Soleymani LLP have vigorously litigated this action for years, are deeply familiar with both the law and the facts of this case, and have devoted substantial resources to this action. There is no indication that proposed Class Counsel will not continue to assiduously pursue this action on behalf of the Class.
F. Superiority
The factors the Court considers in determining superiority of the class action device are:
(A) the class members’ interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action.
Fed. R. Civ. P. 23(b)(3);
see Zinser v. Accufix Research Inst., Inc.
,
Cognizant raises two primary contentions regarding superiority. First, Cognizant
contends that, because individual recovery under Title VII may be substantial and because
attorneys’ fees may be available, the class action device is not superior here. But Plaintiffs
assert a pattern-or-practice claim, which—as described
supra
—requires proof of a broad
practice. The class action device is a superior method of proving claims such as those
presented in the Bench Terminations Subclass.
See, e.g.
,
Simpson v. Dart
,
Second, Cognizant contends that litigation of Phase Two of Plaintiffs’ claims would
be unwieldy, and that Plaintiffs have presented no workable plan for resolving Cognizant’s
individual defenses as to all class members. But there is no requirement under Rule 23 that
a district court articulate how a case would be tried.
See Chamberlan v. Ford Motor Co.
,
IX.
CONCLUSION
In light of the foregoing, the Court:
1. DENIES Cognizant’s motion to exclude the testimony of Dr. Phillip Johnson; 2. DENIES Cognizant’s motion to exclude the testimony of Dr. Ronil Hira; 3. DENIES Plaintiffs’ motion for sanctions;
4. DENIES Cognizant’s MSJ as to Brian Cox’s individual claim; and 4. GRANTS in part and DENIES in part Plaintiffs’ motion for class
certification.
The Court certifies the following Class as to Plaintiffs’ Section 1981 claim: Between September 18, 2013 to the date of class certification, all individuals who are not of South Asian race or Indian national origin and who 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.
The Court certifies the following Subclass as to Plaintiffs’ Title VII claim:
Between December 15, 2016 to the date of class certification, all individuals who are not of South Asian race or Indian national origin and who 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.
The Court appoints Brian Cox and Vartan Piroumanian as Class Representatives, and appoints Kotchen & Low LLP and Yadegar, Minoofar & Soleymani LLP as Class Counsel.
IT IS SO ORDERED.
DATED: October 27, 2022
DOLLY M. GEE UNITED STATES DISTRICT JUDGE
Notes
[1] The Court’s effort to harmonize Ninth Circuit precedents regarding the use of admissible
26
evidence is consistent with the Ninth Circuit’s rule in the context of evaluating other types of substantive
motions.
See, e.g.
,
Fraser v. Goodale
,
[2] Dr. Johnson measures this in terms of “employee-years,” i.e. , the sum of years each employee spent in the company over a period of time. See Johnson Report ¶ 18 n.17.
[3] Positions below Associate level are not open to visa-holding employees. Westphal Decl. ¶ 18. 25
[4] As noted herein, employees who are on the bench are Cognizant employees who continue to 26 receive their full salaries. “Placement” may, therefore, be a more accurate description than “hiring” when an employee moves from the bench to a project. Still, the process of such placements has many of the 27 hallmarks of a hiring process, including applications and interviews, as discussed below. The Court thus 28 describes these placements as “hiring” for purposes of this Order.
[5] Non-billable employees generally do not go to the bench. Kotchen Decl. ISO Class Cert., Ex. 9 28 (Westphal Depo.) at 14:6-10 [Doc. # 256-12].
[6] Plaintiffs contend that because internal applicants can only apply for 10 jobs at a time through Talent Marketplace, they sometimes applied through Taleo. [ See Doc. # 256-80 at 6 (showing that applicants can apply only for 10 jobs through Talent Marketplace).] But Plaintiffs do not offer evidence to support this contention.
[7] This testimony does not appear irreconcilable with Plaintiffs’ contention that benched applicants sometimes apply for jobs through Taleo: the description provided by Mr. Westphal suggests that an internal candidate would still show up as such in Taleo.
[8] The emphases are in the original document. It is unclear who “we” is.
[9] Plaintiffs submitted an undated document entitled “Visa Utilization Policy v2.2” with their motion. It is not clear when this document was created, or by whom, or whether it was a draft or a final 27 product. The Court thus does not rely on this document, but it was produced by Cognizant, and it appears 28 to outline a policy similar to that described elsewhere in the record.
[10] This complaint is hearsay, but it is plausible that Gundlach’s testimony could be introduced at
26
trial. Indeed, Gundlach was apparently deposed in this matter.
See
Kotchen Decl. ISO Class Cert., Ex.
102 [Doc. # 256-105]. The Court thus declines to exclude this evidence.
See JL Beverage Co., LLC v.
27
Jim Beam Brands Co.
,
[11] Whether or not “Location Preference” is properly coded as “voluntary” or “involuntary” termination is the subject of Plaintiffs’ motion for sanctions, as well as a part of Cognizant’s Daubert motion.
[12] Dr. Hira’s report is the subject of a Daubert motion, addressed infra .
[13] This conversation appears to have taken place in a Cognizant chat function, and punctuation has 28 been added for clarity.
[14] Because the Court discusses Edward Cox at length with respect to Cognizant’s Motion for Summary Judgment, but Brian Cox at length with respect to the Motion for Class Certification, the Court 27 will sometimes refer to Edward Cox by his full name or as “Edward,” and to Brian Cox by his full name 28 or as “Brian.”
[15] Plaintiffs do not seek to certify a promotions class. See Mot. for Class Certification at 3. Thus, although Dr. Johnson was retained to provide an opinion regarding disparities in promotions, and discusses his promotions analysis in his report, the Court does not address that analysis herein.
[16] Dr. Johnson appears qualified to testify on the issues for which he offers his opinion. Dr. 26 Johnson holds a PhD in economics from the University of California, Los Angeles, and before joining the consulting firm for which he currently works, he was a professor at the Instituto Tecnológico Autónomo 27 de México. He continues to publish regularly, and has served as an expert witness in a number of actions 28 alleging employment discrimination. See Johnson Report, Ex. 1 ( curriculum vitae ).
[17] Cognizant also contends, in a footnote, that Johnson’s methods for identifying which employees 23 are South Asian—which rely in part on determining whether the employee’s name indicates South Asian ethnicity—is “fairly arbitrary and likely to lead to misidentifications.” Johnson Mot. at 12 n.1. For 24 example, Cognizant suggests, Johnson’s approach might incorrectly identify a non-South Asian individual 25 who marries and adopts the last name of a South Asian spouse as South Asian. While Johnson’s method may be somewhat imprecise, there is no indication that it is likely to result in statistically significant 26 discrepancies or that it is more likely to be over-inclusive than under-inclusive (a South Asian individual may just as easily adopt a non-South Asian spouse’s surname and thus be missed by Johnson’s analysis). 27 Cognizant does not ask the Court to exclude Johnson’s report on this basis, and the Court declines to do 28 so.
[18] This information was available through Cognizant’s Taleo system.
[19] The other case relied on by Cognizant,
State Farm Fire & Cas. Co. v. Electrolux Home Prod.,
20
Inc.
,
[20] The fact that the plaintiffs’ expert in Buchanan used the same data as the benchmark suggests that use of this data has some general acceptance within the economic community, which also supports its 26 admissibility. 27
[21] Dr. McCrary explains that “benched South Asian employees and visa-holding employees are 28 more than twice as likely to be willing to relocate,” and that employees who were willing to relocate were five times less likely to be terminated from the bench. McCrary Report ¶ 134. Dr. McCrary contends that
[23] Judge Rogers persuasively addressed similar criticisms of the plaintiffs’ expert report in
26
Buchanan
, and concluded that it was reasonable for the plaintiffs’ expert to treat a transfer employee’s
first project in the United States as the transfer’s date of hire.
See
[24] The fact that a standard deviation “can be calculated using online tools” does not place its 28 calculation outside the scope of expert testimony.
[25] Dr. Hira’s specific conclusion that Cognizant engages in “wage theft” does appear to be beyond the scope of his expertise in business and immigration practices. The Court does not rely on this 27 conclusion, and Plaintiffs should not attempt to elicit testimony regarding Dr. Hira’s opinion of 28 Cognizant’s wage and hour practices at trial.
[26] Cognizant also contends that Dr. Hira’s report is “riddled with errors.” None of the errors Cognizant cites are major errors, and none are critical to his broad conclusions. The Court thus declines to exclude Dr. Hira’s report on this basis.
[27] Cognizant analogizes the testimony offered by Dr. Hira regarding the business practices of what
Dr. Hira calls “India-centric IT consulting firms” to the stereotyping testimony offered by an expert in
Jinro
. In
Jinro
, the “expert” offered testimony at trial about the purported corruptness of Korean
businesses, “inviting the jury to assume the Korean litigant fits the stereotype.”
[28] Plaintiffs contest this description, pointing to evidence that certain Cognizant employees whose terminations were coded for “CDP – Location Preference” did not, in fact, turn down offered positions because of their location preferences. See, e.g. , Equal Employment and Opportunity Commission (“EEOC”) Letter of Determination re: Richard Stewart [Doc. # 256-4 at 13] (concluding that “credible evidence support[ed Stewart’s] contention that he was not considered for positions for which he was eligible and willing to take. . . . At no time was the prospect of relocation made to [Stewart]”). Plaintiffs’ sanctions motion does not depend, however, on the question of whether employees actually “chose” to turn down offered positions.
[29] To the extent Cognizant seeks to persuade, its characterization of “Location Preference” terminations as “voluntary” appears aimed at raising individualized issues that arguably would preclude certification.
[30] This summary of the facts is undisputed for purposes of this motion, unless otherwise stated. Facts are drawn from Cognizant’s Statement of Undisputed Facts (“SUF”), as set forth with its Reply 27 [Doc. # 312-1]. To the extent the parties’ purportedly disputed facts are not in fact controverted by the 28 evidence, the Court therefore cites to them as uncontroverted facts.
[31] The parties dispute whether a formal requisition for the role was ever created and whether Stark had authority to hire for the position, because this dispute is not material to this Court’s disposition of the motion, the Court does not address that dispute herein.
[32] Historically, Title VII disparate treatment claims and Section 1981 claims have been treated 23 alike. See, e.g. , Surrell v. California Water Serv. Co. , 518 F.3d 1097, 1103 (9th Cir. 2008) (“When analyzing § 1981 claims, we apply the same legal principles as those applicable in a Title VII disparate 24 treatment case.”). A recent Supreme Court decision held that Section 1981 requires a higher showing 25 from plaintiffs: Section 1981 must prove “that, but for race, [they] would not have suffered the loss of a legally protected right.” Comcast Corp. v. Nat'l Ass'n of Afr. Am.-Owned Media ,140 S. Ct. 1009 , 1019 26 (2020); s ee also id. at 1017 (explaining that Title VII’s “motivating factor” causation standard does not apply in Title VII cases). Cognizant does not argue, however, that the Title VII disparate treatment claim 27 and Section 1981 claim should be treated separately here, or how Comcast might change the outcome of 28 this motion. The Court thus addresses them together.
[33] Plaintiffs contend that demonstrating an actual vacancy is not the plaintiff’s burden in a
25
Teamsters
case.
See, e.g.
,
Teamsters
,
[34] Even under McDonnell Douglas , the Court would conclude that Cognizant failed to meet its 26 burden to establish the absence of a triable issue of fact. Specifically, the fact that Cognizant employees actively recruited Edward Cox for the position, and that his application progressed far beyond mere 27 submission before being rejected by different Cognizant employees, creates a triable issue of fact as to 28 whether Cognizant’s preference for an internal candidate was pretextual.
[35] The court had previously declined to exclude the expert’s analysis under Daubert . See id . at 28 *10 (declining to exclude expert opinion but noting that inclusion of “expats” went to weight of opinion).
[36] The Court rejects Cognizant’s suggestion that the Court discount Dr. Johnson’s opinion because 25 of his failure to reach a conclusion about whether Cognizant’s visa policies caused the statistical disparities he identifies in his expert report. Although Plaintiffs may not rely only on statistical evidence 26 to make their case, they have not attempted to do so here. 27
[37] As noted supra with respect to Plaintiffs’ sanctions motion, the Court does not find persuasive 28 Cognizant’s argument that “location preference” terminations from the bench are actually voluntary. It is apparent that all internal documents treated this as an involuntary termination. The only piece of evidence
[38] Plaintiffs also sought an order appointing Cox as the class representative for the proposed hiring class, but since the Court has concluded that common questions will not predominate over individualized 27 issues as to the proposed hiring class, Plaintiffs’ request for appointment of Cox as representative of that 28 class is moot.
[39] Cognizant does not contend that Brian Cox is an inadequate class representative because his
father, who was the subject of the alleged discrimination in question, has passed away. Multiple courts
have permitted substitution of a class representative’s executor or heirs after settlement.
See, e.g.
,
In re
Pharm. Indus. Average Wholesale Price Litig.
, 277 F.R.D. 52, 60 (D. Mass. 2011) (finding executor
adequate where class was certified before settlement). In
Shamberg v. Ahlstrom
,
[40] Most of these facts can also be proven through Cognizant’s internal documents, which would constitute admissions. For this reason, the fact that Edward Cox is not available to provide testimony also does not counsel against a finding of adequacy.