Congress charged the Equal Employment Opportunity Commission (EEOC) with administering and enforcing T itle VII of the Civil Rights Act of 1964, 42 U.S.C. §§ 2000e et seq. Plaintiff Thomas Liu alleges, on behalf of himself and a putative class of affected non-white Uber Technologies (Uber) drivers, that Uber’s use of a star-rating system to terminate drivers violates Title VII because it disparately impacts non-white drivers. The district court dismissed Liu’s Third Amended Complaint (TAC) on the grounds that, even at this early stage, Liu lacked sufficient statistical or other evidentiary support to plausibly plead his claim. In the EEOC’s view, the court applied an unnecessarily high pleading standard to Liu’s complaints and thereby erred.

The EEOC has a substantial interest in the proper interpretation of Title VII, including the threshold question of what suffices, at the pleading stage, to constitute a plausible violation of the statute. Accordingly, the EEOC files this brief pursuant to Federal Rule of Appellate Procedure 29(a).

Whether the district court erred when it held that Liu’s TAC did not give rise to a plausible inference that non-white drivers are disparately impacted by Uber’s star-rating system.

Statement of the Case

A. Statement of the Facts

Uber is a transportation service that “engages drivers across the country to transport riders.” ER-40 ¶ 9. Uber asks riders to rate their drivers on a scale of one to five stars after a ride is completed. Id. ¶ 11. Uber terminates, or deactivates, drivers when their aggregate passenger rating falls below a certain threshold. ER-40-41 ¶¶ 12-14. Plaintiff and putative class representative Thomas Liu is an Asian-American man from Hawaii who worked as an Uber driver in San Diego until Uber deactivated him “because his average star rating fell below 4.6.” Id. ¶¶ 14-15.

Liu filed a timely charge of race discrimination with the EEOC. ER-55 ¶ 45. After receiving a dismissal and notice of right to sue in August 2020, Liu filed a class-action complaint, and three amended complaints, purporting to represent a class of “all non-white Uber drivers across the country who have been terminated (or put at risk of termination) based upon Uber’s star rating system.” ER-52 ¶ 41.

B. District Court’s Decisions

The district court dismissed all four of Liu’s complaints for failure to state a valid claim under Federal Rule of Civil Procedure 12(b)(6).

Liu’s initial complaint alleged generally that Uber relies on a racially discriminatory passenger star-rating system to terminate drivers. According to the complaint, social-science research recognizes that “reliance on customer evaluation systems may lead to discriminatory impact on racial minorities.” ER-146 ¶ 13. Liu further alleged that Uber “has long known that relying on a system that depends on passenger evaluations is discriminatory,” given the fact that Uber initially refused to allow passengers to tip drivers through its app, citing concerns that rider bias would affect drivers’ tips. ER-146 ¶¶ 12. The court dismissed the complaint, noting that “one could readily imagine a plaintiff successfully stating a disparate impact claim for a policy of terminating drivers solely on the basis of customer ratings without accounting for likely racial bias,” but Liu’s complaint was “too sparse and poorly drafted.” ER-142. The court pointedly faulted Liu for alluding to “social science research” without citing or describing specific research. Id.

Liu filed an amended complaint in which he elaborated upon the allegations in his original complaint by citing social-science research that specifically used Uber as a “case study to explore how bias may creep into evaluations of drivers through consumer-sourced rating systems.” ER-129-131 ¶ 20. According to that research, such systems are “highly likely to be influenced by bias on the basis of factors like race or ethnicity.” Id. Liu also cited an article discussing driver concerns about bias in the rating system. ER-128-29 ¶ 18.

The court dismissed the amended complaint, again noting that “Liu’s overall theory is certainly plausible,” but faulting him for “not adequately alleg[ing] facts to support each element of a disparate impact claim.” ER-99. Liu did “more than enough to support an inference that, if there were indeed a racial disparity in driver terminations, it would likely have been caused by Uber’s star rating system.” ER-101. But, the court said, “the complaint essentially skips over the first element—plausibly alleging that a disparity actually exists, in the sense that minority Uber drivers are disproportionately terminated for low ratings compared to white drivers.” ER-100.

Liu then filed a second amended complaint in which he included the results of a survey his attorney conducted of 4,093 Uber drivers asking (1) whether they had been terminated by Uber because of their star rating, and (2) their race. ER-86-87 ¶¶ 24-25. Liu alleged that his attorney consulted with an expert in economics regarding the results of the survey and the expert found a 1 in 10,000 chance that the survey would produce the results it did “had race played no role in drivers being deactivated based on their star ratings.” ER-88 ¶¶ 28-29. The court dismissed the second amended complaint, holding that the new pleading did not “cure the defects that doomed his complaint the last time around.” ER-58. Focusing on the methodology of Liu’s survey, the court found the survey “essentially meaningless” because it “uses the wrong denominator” by surveying only “drivers who identify as being terminated” as opposed to a sample of the overall Uber driver population. ER-59.

Liu then filed his third and final amended complaint (the TAC), in which he clarified that the survey his attorney conducted included both drivers who had been terminated and drivers who had not. ER-45 ¶ 26 n.4. It also reported the results of follow-up questions his attorney sent to the respondents. Id.

The court dismissed the TAC as well. The court began its analysis by explaining that “[a] plaintiff in a disparate impact case must allege three elements: (1) a significant disparity with respect to employment for the protected group, (2) the existence of a specific employment practice or set of practices, and (3) a causal relationship between the identified practice and disparity.” ER-3 (citing Freyd v. Univ. of Or., 990 F.3d 1211, 1224 (9th Cir. 2021)). Applying this standard, the court concluded that Liu had “plausibly alleged that racial discrimination could affect customer ratings, including in the rideshare industry,” but that he failed to “adequately allege that this legitimate concern about racial discrimination actually manifested itself in driver terminations at Uber.” Id. Therefore, the court held, Liu did not “satisfy the first element” of a disparate-impact prima facie case. Id. The court acknowledged that “it is hardly fanciful to suspect that Uber’s practice of terminating drivers based on customer ratings negatively affects minority drivers,” but that “Liu must include non-conclusory allegations about the impact of the challenged practice at the actual company he is suing.” ER-5.

Liu timely appealed the district court’s order dismissing the TAC.

Argument

Title VII was enacted to “achieve equality of employment opportunities,” Griggs v. Duke Power Co., 401 U.S. 424, 429 (1971), by “root[ing] out discrimination in employment,” EEOC v. Shell Oil Co., 466 U.S. 54, 77 (1984). Title VII’s prohibition of employment practices that cause an unjustified disparate impact is an essential component of that effort. See Griggs, 401 U.S. at 429-30; see also Civil Rights Act of 1991, Pub L. No. 102–166, § 3, 105 Stat 1071 (purpose of the 1991 Act includes “to confirm statutory authority and provide statutory guidelines for the adjudication of disparate impact suits under title VII”); 42 U.S.C. § 2000e-2(k) (codifying burden of proof in disparate-impact cases).

A successful disparate-impact claim often relies on sophisticated forms of proof from the plaintiff, including statistical analysis of information that is only available to the plaintiff through discovery. E.g., Freyd, 990 F.3d at 1224-26 (discussing plaintiff’s statistical showing at summary judgment). At the pleading stage, however, the plaintiff’s burden to plausibly allege disparate-impact liability is substantially lower than the evidentiary burden the plaintiff will face at summary judgment or trial. A successful disparate-impact complaint does not require such evidence, but rather asks the plaintiff to allege facts from which the court can plausibly infer that the employer’s challenged practice caused a disparate impact.

In this case, Liu’s complaint needed to allege facts giving rise to a plausible inference that Uber’s use of the star-rating system caused a disparity in non-white drivers terminated for low star ratings. Because Liu satisfied this standard, including by citing social-science research dealing specifically with bias in the Uber-passenger context, the court erred by requiring more of him at the Rule 12(b)(6) stage.

I. A disparate-impact complaint need only allege facts giving rise to a plausible inference that the employer’s challenged policy caused a disparate impact.

A. A disparate-impact plaintiff does not have to plead a prima facie case to plausibly allege a claim.

Title VII prohibits employment practices that “cause[] a disparate impact on the basis of race, color, religion, sex, or national origin” that are not “job related . . . and consistent with business necessity.” Lewis v. City of Chicago, 560 U.S. 205, 212 (2010) (quoting 42 U.S.C. § 2000e-2(k)(1)(A)(i)). At summary judgment or trial, a plaintiff may establish a prima facie disparate-impact claim under Title VII “by showing that the employer ‘uses a particular employment practice that causes a disparate impact’ on one of the prohibited bases.” Id. (emphasis omitted). The employer may then rebut the prima facie case by showing that the challenged practice is necessary to its business. Ricci v. DeStefano, 557 U.S. 557, 578 (2009). The plaintiff may still prevail if he can show that an alternative practice is available to satisfy the business necessity. Id.

At the pleading stage, however, the plaintiff need not plead a prima facie case. See Austin v. Univ. of Or., 925 F.3d 1133, 1136-37 (9th Cir. 2019). The Supreme Court held as much in the specific context of employment-discrimination claims in Swierkiewicz v. Sorema N.A., 534 U.S. 506, 508 (2002). That is because the prima facie case is an evidentiary standard, not a pleading standard. Id. at 510. Moreover, “the precise requirements of a prima facie case can vary depending on the context,” and “[b]efore discovery has unearthed relevant facts and evidence, it may be difficult to define the precise formulation of the required prima facie case in a particular case.” Id. at 512.

Under Federal Rule of Civil Procedure 8(a)(2), a complaint must contain “a short and plain statement of the claim showing that the pleader is entitled to relief.” The Supreme Court elaborated upon this liberal pleading standard in Bell Atlantic Corp. v. Twombly, 550 U.S. 544 (2007), and Ashcroft v. Iqbal, 556 U.S. 662 (2009), explaining that, “[t]o survive a motion to dismiss, a complaint must contain sufficient factual matter, accepted as true, to ‘state a claim to relief that is plausible on its face.’” Iqbal, 556 U.S. at 678 (quoting Twombly, 550 U.S. at 570). A claim is plausible when the facts alleged in the complaint allow for “the reasonable inference that the defendant is liable for the misconduct alleged.” Id. (citing Twombly, 550 U.S. at 556). The complaint’s factual allegations must establish “more than a sheer possibility that a defendant has acted unlawfully.” Id. But they need not establish a “probability” of liability. Twombly, 550 U.S. at 556. Importantly, “a well-pleaded complaint may proceed even if it strikes a savvy judge that actual proof of those facts is improbable, and that a recovery is very remote and unlikely.” Id. at 556 (quotation marks omitted).

The Court’s decisions in Iqbal and Twombly did not alter the holding in Swierkiewicz that discrimination plaintiffs are not required to allege a prima facie case to state a valid claim. See Twombly, 550 U.S. at 569 n.14; see also Austin, 925 F.3d at 1135, 1137 (applying Swierkiewicz to Title IX claims and rejecting the argument that Iqbal abrogated Swierkiewicz); Sheppard v. David Evans & Assoc., 694 F.3d 1045, 1050 n.2 (9th Cir. 2012) (ADEA plaintiff is “not required to plead a prima facie case of discrimination” (emphasis in original) (citing Swierkiewicz, 534 U.S. at 508-11)).

Thus, to survive a motion to dismiss, a disparate-impact plaintiff need only allege facts supporting a reasonable inference that the practice he challenges caused a disparate impact. See Frappied v. Affinity Gaming Black Hawk, LLC, 966 F.3d 1038, 1054 (10th Cir. 2020) (acknowledging “plaintiffs do not need to establish a prima facie case [of disparate impact discrimination] in the complaint”); Mandala v. NTT Data, Inc., 975 F.3d 202, 208-09 (2d Cir. 2020) (same);Wu v. Special Couns., No. 14-7159, 2015 WL 10761295, at *1 (D.C. Cir. Dec. 22, 2015) (“[T]o sustain a disparate impact claim . . . neither prima facie proof nor detailed factual allegations are necessary to withstand a Rule 12(b)(6) motion[.]”).

B. A disparate-impact plaintiff can plead a plausible claim in at least two different ways.

Beyond the baseline requirements of Rule 8(a)(2), there is no precise formula that a disparate-impact plaintiff must replicate to plead a plausible claim. Indeed, the courts have paved at least two distinct paths to a successful disparate-impact complaint.

First, a plaintiff could identify an employment practice and plead an actual disparity in the employer’s workforce, giving rise to a plausible inference that the employer’s policy caused the disparity. In Chaidez v. Ford Motor Co., 937 F.3d 998 (7th Cir. 2019), for instance, the plaintiff’s complaint alleged that the employer’s pre-employment testing process resulted in a disparate impact on Hispanic and Latino applicants. The Seventh Circuit held that the complaint stated a sufficient disparate-impact claim because it both identified the testing process and alleged that the employer’s workforce did not reflect the racial demographics of the surrounding area. Id. at 1007; see also Frappied, 966 F.3d at 1054-56 (allegations regarding the number of employees laid off over the age of forty, compared to the number of individuals laid off under the age of forty, “ma[d]e plausible” plaintiff’s allegation that termination policies had a disparate impact based on age).

This path will be available to those plaintiffs with pre-discovery access to relevant information about the employer’s workforce. The Chaidez plaintiffs, for example, attached to their complaint photographs of the employer’s recent classes of new hires to support their allegation regarding the racial makeup of the employer’s workforce. Chaidez, 937 F.3d at 1007.

Second, a plaintiff who lacks pre-discovery access to information about the employer’s workforce could identify the challenged practice and point to real-world conditions suggesting that the practice will result in a disparity. There are limited appellate opinions addressing motions to dismiss disparate-impact claims post-Iqbal/Twombly, but a few examples from the employment context and from the related Fair Housing Act disparate-impact context help to illustrate what constitutes a plausible claim under the second formulation.

In Reyes v. Waples Mobile Home Park Limited Partnership, 903 F.3d 415 (4th Cir. 2018), a Fair Housing Act case, the plaintiffs alleged that a Fairfax, Virginia mobile-home park’s requirement that all occupants provide documentation of legal status disproportionately ousted Latino families from their rental homes. To support an inference that the policy disparately impacted Latinos, the plaintiffs alleged that Latinos constitute 64.6% of the undocumented population in Virginia, and that undocumented immigrants constitute 36.4% of the Latino population, compared to just 3.6% of the non-Latino population. Id. at 428. Reversing the district court’s dismissal of the complaint, the Fourth Circuit held that these allegations plausibly alleged that the challenged policy disproportionately affected Latinos. Id.

Notably, the Reyes court did not require the plaintiffs to plead any information about the number of Latino families who were actually affected by the implementation of the policy, even though the suit was filed approximately one year after the policy was implemented. Id. at 420-21, 429 (plaintiffs adequately alleged policy “was likely to cause Latino tenants . . . to be disproportionately subjected to eviction”). Rather, the real-world condition that, state-wide, undocumented individuals were disproportionately likely to be Latino gave rise to a plausible inference that the policy disproportionately affected Latinos in Fairfax County.

Similarly, in Carson v. Lacy, 856 F. App’x 53 (8th Cir. 2021), the Eighth Circuit held that the district court erred when it required a Title VII disparate-impact plaintiff to “allege [the defendant employer] had disproportionately fewer black custodians as a result of using felony background checks.” Id. at 54. According to the court of appeals, the plaintiff’s allegation that Black people are incarcerated at a higher rate than White people in Arkansas was sufficient to plausibly allege that the defendant’s policy of disfavoring applicants with felony convictions excluded Black applicants “from the opportunity of employment at a higher rate than white applicants.” Id.

Cases in which courts have dismissed disparate-impact complaints are also instructive. In Mandala v. NTT Data, Inc., 975 F.3d 202 (2d Cir. 2020), the plaintiffs, who had applied to work at NTT but were rejected when the company learned of their prior convictions, brought a disparate-impact claim against NTT. They alleged that NTT’s blanket policy of not hiring individuals with a felony conviction disproportionately harmed Black applicants because, nationwide, Black people are more likely to have a felony conviction than White people. The Second Circuit held that the complaint failed to plausibly allege a disparate-impact claim under Title VII because the plaintiffs did not adequately allege that the nationwide felony-conviction trend also arose within the pool of individuals qualified for skilled jobs at NTT. Id. at 212.[2] In order to cure their pleading, the court said, the plaintiffs needed either to rely on population statistics that were more reflective of NTT’s applicant pool or explain why national statistics regarding felony convictions are likely to be representative of NTT’s applicant pool. Id. at 211-12; see also Mandala v. NTT Data, Inc., 988 F.3d 664, 668 (2d Cir. 2021) (Sullivan, J., and Nardini, J., concurring in denial of rehearing en banc) (suggesting that plaintiffs could cure their pleading by alleging that racial disparities in criminal convictions persist in population with “some college education”).

Notably, the court did not require the plaintiffs to include allegations regarding the actual impact of NTT’s policy on the racial makeup of its workforce. It did not require, for instance, that plaintiffs describe at the pleading stage a racial imbalance in NTT’s workforce. Instead, the court required the plaintiffs to allege more precise real-world conditions that would then “plausibly suggest that the challenged practice actually has a disparate impact.” 975 F.3d at 210 (emphasis in original).

Finally, the D.C. Circuit’s treatment of two separate disparate-impact claims presented in Boykin v. Fenty, 650 F. App’x 42 (D.C. Cir. 2016), one of which was dismissed under Rule 12(b)(6) and the other under Rule 56, illustrates the distinct requirements at each stage of disparate-impact litigation. Specifically, Boykin shows that a complaint must allege facts suggesting a disparate impact, whereas at summary judgment, the plaintiff must proffer evidence demonstrating that impact.

The Boykin plaintiffs, approximately forty-two predominantly Black and Hispanic former residents of a D.C. homeless shelter, challenged the closure of that shelter, alleging that the closure had a disparate impact based on disability under the Americans with Disabilities Act and based on race under the Fair Housing Act. The Americans with Disabilities Act claim was dismissed on the pleadings; the Fair Housing Act claim was dismissed at summary judgment. The court affirmed the dismissal of the plaintiffs’ claim that closing low-barrier housing shelters had a disparate impact based on disability because “[t]he complaint failed to allege facts suggesting that the closure affected a greater proportion of disabled individuals . . . as it did not, for instance, include an allegation that disabled homeless individuals are more likely to rely on low-barrier shelters than non-disabled homeless individuals.” Id. at 44 (emphasis added). The court affirmed the grant of summary judgment on the plaintiffs’ race-based claim because the plaintiffs “did not provide sufficient evidence to support a finding that the challenged policy actually disproportionately affected a protected class.” Id. at 45 (emphasis in original).

In each of these cases, the courts have not required the disparate-impact plaintiffs to plead an evidentiary basis sufficient to demonstrate a disparate impact resulting from the challenged policy. Instead, they have required the plaintiff to allege facts making it plausible that a disparate impact arises from the challenged policy. In other words, they do not require plaintiffs to allege what is “actually happening,” but rather what is plausibly happening as a result of the confluence of real-world conditions and the challenged policy. See also Meyer v. Bear Rd. Assocs., 124 F. App’x 686, 688 (2d Cir. 2005) (holding plaintiffs plausibly alleged that challenged policy “actually or predictably leads to” disparate impact); Lee v. Hertz Corp., 330 F.R.D. 557, 561 (N.D. Cal. 2019) (complaint adequately alleged that employer policy would have a disparate impact on Latinos where it did not allege any facts regarding actual Latino applicants denied jobs, but instead alleged that Latinos were convicted of crimes more frequently than White people and that employer disqualified applicants with criminal histories); Nat’l Fair Hous. All. v. Travelers Indem. Co., 261 F. Supp. 3d 20, 34 (D.D.C. 2017) (disparate impact claim sufficiently pleaded where complaint plausibly alleged members of protected class were “more likely to be harmed” by challenged policy).

This approach makes sense, given that plaintiffs often will not have access to the data necessary to plead the actual disparate impact of the challenged policy before discovery. See Schmitt v. Kaiser Found. Health Plan of Wash., 965 F.3d 945, 959 n.8 (9th Cir. 2020) (recognizing, in Affordable Care Act case, that plaintiffs may be unable to plead numeric impact of policy pre-discovery and could rely instead on showing that policy is “likely to predominately affect disabled persons”); Gilligan v. Jamco Dev. Corp., 108 F.3d 246, 250 (9th Cir. 1997) (rejecting requirement that plaintiff plead a prima facie case in FHA context because it “would require plaintiffs to plead facts they have no way of knowing”); Mandala, 975 F.3d at 210 (acknowledging that racial composition of workforce in certain jobs is “not always available, particularly before discovery”); Id. at 217 (Chin, J., dissenting) (same); cf. Chaidez, 937 F.3d at 1007 (plaintiffs will “need to utilize the discovery process to support their allegations with statistical and comparative evidence” and defendant can “present contrary evidence at the summary judgment stage or at trial”).

Requiring plaintiffs to plead information they have no way of knowing would render a number of disparate-impact claims impossible to plead, undermining Title VII’s antidiscrimination mandate. See supra 7-8. The EEOC therefore respectfully urges this Court not to impose such a requirement.

II. Liu adequately alleged a disparate impact arising from Uber’s use of passenger ratings to terminate drivers.

In his amended complaint and each complaint thereafter, Liu undisputedly identified a specific, allegedly discriminatory practice: Uber’s policy of terminating drivers when their passenger ratings fall below a certain threshold. He then alleged that driver ratings are likely to be infected by racial bias, based on two sources of real-world information. First, Liu cited and quoted from social-science research that analyzed Uber’s consumer-sourced rating system—the very system he challenges—and concluded that “[c]onsumer-sourced ratings like those used by Uber are highly likely to be influenced by bias on the basis of factors like race or ethnicity.” ER-42-44 ¶ 21. Second, Liu cited Uber’s own judgment that “passengers discriminate against racial minorities” and that “allowing [app-based] tipping would therefore discriminate against minority drivers in the wages they would receive.” ER-41 ¶ 18. The documented risk of Uber riders’ bias infecting their interactions with drivers, including driver ratings, combined with Uber’s challenged policy, gives rise to a plausible inference that minority drivers are disproportionately terminated for low passenger ratings.

As the cases discussed above show, the district court required too much of Liu when it asked him to show that a disparate impact “actually manifested itself,” ER-3. rather than to allege facts from which a disparate impact could plausibly be inferred. See supra section I.B; see also Jenkins v. N.Y.C. Transit Auth., 646 F. Supp. 2d 464, 469 (S.D.N.Y. 2009) (rejecting argument that complaint had to allege “that an employer’s policy has had a statistically significant, disproportionate, and negative effect on one protected group as compared to others”). It is not until summary judgment that a plaintiff is required to adduce evidence from which a jury could find the alleged impact “actually manifested itself.”

The district court’s reliance on and multiple citations to Freyd, 990 F.3d at 1224, a summary-judgment decision that said nothing about pleading standards, further suggest that it incorrectly viewed Liu’s claim through the lens of a prima facie case. See ER-3, 100. Indeed, the court held that Liu satisfied “the second and third elements” of a prima facie case, but faulted him for “not satisfy[ing] the first element” by not pleading a “significant disparity with respect to employment for the protected group.” ER-100-101; see also ER-101 (Liu cannot “plow ahead to discovery before successfully pleading the first element of a disparate impact claim.”).

The district court’s error is also evident in its acknowledgement that Liu did identify “a business practice and show[ed] that it could plausibly cause a disparate impact.” ER-100. The court should have stopped there, as that is all that Rule 8(a)(2) and Supreme Court precedent require of Liu.

At summary judgment, Liu will have to proffer evidence from which a reasonable jury could find a disparate impact, and Uber will have an opportunity to test the soundness of Liu’s evidentiary showing, including the methodology of any statistical evidence that Liu puts forward. Uber will also have a chance to show that the challenged practice is “job related . . . and consistent with business necessity.” 42 U.S.C. § 2000e-2(k)(1)(A)(i). But at the pleading stage, such evidence is not required. See Watson v. Fort Worth Bank & Tr., 487 U.S. 977, 994-97 (1988) (discussing plaintiff’s evidentiary burden and defendant’s opportunity to test the reliability of that evidence); see also Wu, 2015 WL 10761295, at *2 (“Although Wu need not present statistical evidence in his complaint, he must be able to present such evidence at some point.”); and see Chaidez, 937 F.3d at 1007 (plaintiffs “will need to utilize the discovery process to support their allegations with statistical and comparative evidence”).

Conclusion

For the foregoing reasons, the judgment of the district court should be vacated and the case remanded for further proceedings.

Respectfully submitted,

GWENDOLYN YOUNG REAMS

Acting General Counsel

JENNIFER S. GOLDSTEIN

Associate General Counsel

ELIZABETH E. THERAN

Assistant General Counsel

s/Georgina C. Yeomans

GEORGINA C. YEOMANS

Attorney

Equal Employment

Opportunity Commission

Office of General Counsel

131 M St. N.E., 5th Floor

Washington, D.C. 20507

202-921-2748

georgina.yeomans@eeoc.gov

Certificate of Compliance

[Insert Form 8]

Certificate of SERVICE

I certify that on April 3, 2023, I electronically filed the foregoing motion in PDF format with the Clerk of Court via the appellate CM/ECF system. I certify that all counsel of record are registered CM/ECF users, and service will be accomplished via the appellate CM/ECF system.

s/Georgina C. Yeomans

GEORGINA C. YEOMANS

Attorney

Equal Employment

Opportunity Commission

Office of General Counsel

131 M St. N.E., 5th Floor

Washington, D.C. 20507

202-921-2748

georgina.yeomans@eeoc.gov

[1] The EEOC takes no position on any other issue in the case, including whether Liu, or Uber drivers generally, are employees or independent contractors. Although Uber raised this issue in its motion to dismiss Liu’s original complaint, R.16 at 6-8, the district court did not rule on it, observing only that “[i]f Liu chooses to file an amended complaint, he may also wish to do a better job of supporting his contention that Uber drivers should be classified as employees under federal law.” ER-142-43. The issue did not come up again during the litigation, and plaintiff states in his opening brief before this Court that it is not at issue on appeal. See Plaintiff-Appellant’s Opening Br. at 2 n.1. If the issue should later be raised on appeal, the EEOC takes no position on it.

[2] Judge Pooler and four other Second Circuit judges dissented from the denial of rehearing en banc in Mandala. 988 F.3d 664 (2d Cir. 2021) (Pooler, J., dissenting from denial of rehearing en banc). Judge Pooler argued that, in assuming that racial disparities in conviction rates that are apparent nationwide might not hold true for NTT’s qualified applicant pool, the Second Circuit drew an inference against the plaintiffs, which it is not permitted to do on a motion to dismiss. Id. at 670-71; see also id. at 682 (Chin, J., dissenting from denial of rehearing en banc); 975 F.3d at 215 (Chin, J., dissenting) (making this point in dissent from the panel opinion).