James v. Uber Technologies Inc.

• Court: District Court, N.D. California
• Decided: January 26, 2021
• Docket: 3:19-cv-06462
• Status: Unknown

Opinion

1 2 3 4 UNITED STATES DISTRICT COURT 5 NORTHERN DISTRICT OF CALIFORNIA 6 7 CHRISTOPHER JAMES, et al., Case No. 19-cv-06462-EMC

8 Plaintiffs, ORDER GRANTING IN PART AND 9 v. DENYING IN PART PLAINTIFFS’ MOTION FOR CLASS 10 UBER TECHNOLOGIES INC., CERTIFICATION 11 Defendant. Docket No. 56

12 13 14 Plaintiffs Christopher James and Spencer Verhines are current or former Uber drivers who 15 contend that they and a putative class of approximately 4,828 other Uber drivers are Uber’s 16 employees and therefore eligible for various protections under the California Labor Code. See 17 Docket No. 81 (Amended Consolidated Class Action Complaint (“Am. Compl.”)) ¶ 23. Plaintiffs 18 raise various wage-and-hour claims under California law and seek various forms of relief, 19 including under California’s Unfair Competition Law (UCL) and the federal Declaratory 20 Judgment Act (DJA). Id. 21 Pending before the Court is Plaintiffs’ motion for class certification pursuant to Federal 22 Rule of Civil Procedure 23. See Docket No. 56 (“Mot.”). For the following reasons, Plaintiffs’ 23 motion is GRANTED in part and DENIED in part. 24 I. BACKGROUND 25 The Court and the parties are well acquainted with the background of this case, so it is not 26 set forth in detail here. On May 19, 2020, Plaintiffs filed the pending motion to certify Class, see 27 Mot, and two days later Uber filed a motion to dismiss Plaintiffs’ consolidated class action 1 On June 30, 2020, the Court dismissed, with leave to amend, the consolidated amended 2 complaint’s claims that Uber failed to provide paid sick leave as required by section 246 of the 3 California Labor Code, and any UCL claims premised on violations of section 246. See Docket 4 No. 74 (“Order on MTD”). The Court also dismissed Thomas Colopy as a named Plaintiff in this 5 case without prejudice to his claims. See id. 6 On July 14, 2020, Messrs. James and Verhines (hereinafter, “Plaintiffs”) filed the operative 7 amended consolidated class action complaint (hereinafter, “Amended Complaint”) alleging as 8 follows. Plaintiffs are residents of California who drive for Uber. Am. Compl. ¶¶ 8–9, 17–18. They 9 bring this case as a putative class action on “behalf of . . . all other individuals who have worked as 10 Uber drivers in California who have not released all of their claims against Uber.” Id. ¶ 10. They 11 assert claims related to their alleged misclassification as independent contractors, including (1) failure 12 to reimburse business expenses, (2) failure to pay minimum wage, (3) failure to pay overtime, (4) 13 failure to provide properly itemized pay statements, (5) failure to provide paid sick leave, and (6) 14 unlawful business practices. See id. Plaintiffs seek damages dating back to February 28, 2019, as 15 well as declaratory and injunctive relief, which would require Uber to reclassify its drivers as 16 employees. Id. ¶ 7. 17 II. LEGAL STANDARD 18 Although expressly authorized by Rule 23, the “class action is ‘an exception to the usual 19 rule that litigation is conducted by and on behalf of the individual named parties only.’” Wal– 20 Mart Stores, Inc. v. Dukes, 564 U.S. 338, 348 (2011) (quoting Califano v. Yamasaki, 442 U.S. 21 682, 700–01 (1979)). “In order to justify departure from that rule, ‘a class representative must be 22 part of the class and possess the same interest and suffer the same injury as [her fellow] class 23 members.’” Id. (quoting E. Tex. Motor Freight Sys., Inc. v. Rodriguez, 431 U.S. 395, 403 (1977)). 24 Accordingly, before certifying a class, the Court “must conduct a ‘rigorous analysis’ to 25 determine whether the party seeking certification has met the prerequisites of Rule 23.” Mazza v. 26 Am. Honda Motor Co., Inc., 666 F.3d 581, 588 (9th Cir. 2012) (quoting Zinser v. Accufix Rsch. 27 Inst., Inc., 253 F.3d 1180, 1186, amended 273 F.3d 1255 (9th Cir. 2001)). The Supreme Court 1 Behrend, 569 U.S. 27, 33 (2013) (quoting Wal-Mart, 564 U.S. at 349). Rather, the party seeking 2 certification must “affirmatively demonstrate” her compliance with the requirements of both Rules 3 23(a) and 23(b). See Wal-Mart, 564 U.S. at 349. 4 Rule 23(a) permits plaintiffs to sue as representatives of a class only if (1) “the class is so 5 numerous that joinder of all members is impracticable” (“numerosity” requirement); (2) “there are 6 questions of law or fact common to the class” (“commonality” requirement); (3) “the claims or 7 defenses of the representative parties are typical of the claims or defenses of the class” 8 (“typicality” requirement); and (4) “the representative parties will fairly and adequately protect the 9 interests of the class” (“adequacy” requirement). Fed. R. Civ. P. 23(a)(1)-(4). The purpose of 10 Rule 23(a)’s requirements is largely to “ensure[ ] that the named plaintiffs are appropriate 11 representatives of the class whose claims they wish to litigate,” and to “effectively limit the class 12 claims to those fairly encompassed by the named plaintiff’s claims.” Wal-Mart, 564 U.S. at 349 13 (quoting Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 156 (1982)). 14 If each of the Rule 23(a) requirements are satisfied, the purported class must also satisfy 15 one of the three prongs of Rule 23(b). Here Plaintiffs seek certification under Rule 23(b)(3), 16 which requires the Court to find that “questions of law or fact common to class members 17 predominate over any questions affecting only individual members” (“predominance” 18 requirement), and “that a class action is superior to other available methods for fairly and 19 efficiently adjudicating the controversy” (“superiority” requirement). Fed. R. Civ. P. 23(b). 20 The underlying merits of the case, while admittedly relevant at the class certification stage, 21 should not overly cloud the Court’s certification analysis—the only question presently before the 22 Court is whether the requirements of Rule 23 are met. See Comcast, 569 U.S. at 33–34. The fact 23 that certain elements of proof may favor the defendant on the merits does not negate class 24 certification; the issue is whether the proof is amenable to class treatment. Indeed, once a class is 25 certified, the party prevailing on the merits can benefit from certification, be it Plaintiffs or 26 Defendant. 27 Moreover, “[n]either the possibility that a plaintiff will be unable to prove [her] 1 original decision to certify the class wrong, is a basis for declining to certify a class which 2 apparently satisfies the Rule.” Blackie v. Barrack, 524 F.2d 891, 901 (9th Cir. 1975). Indeed, 3 even “after a certification order is entered, the judge remains free to modify it in the light of 4 subsequent developments in the litigation.” Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 160 5 (1982). Ultimately, whether or not to certify a class is within the discretion of the Court. See 6 Levya v. Medline Indus. Inc., 716 F.3d 510, 513 (9th Cir. 2013); United Steel, Paper & Forestry, 7 Rubber, Mfg. Energy, Allied Indus. & Serv. Workers Int’l Union, AFL–CIO CLC v. ConocoPhilips 8 Co., 593 F.3d 802, 810 (9th Cir. 2010). 9 III. DISCUSSION 10 This order proceeds as follows. First, the Court will apply the Rule 23(a) criteria 11 (numerosity, commonality, typicality, and adequacy) to Plaintiffs’ claim that they are/were Uber’s 12 employees, rather than independent contractors, and for each of their five substantive law claims: 13 failure to (1) reimburse business expenses, (2) pay minimum wage, (3) pay overtime, (4) provide 14 properly itemized pay statements, and (5) provide paid sick leave. 15 Second, the Court will consider whether Plaintiffs have met their burden under Rule 16 23(b)(3), which requires them to establish that the employment classification question, and all of 17 their substantive claims, can be resolved with reference to predominately common proof 18 (predominance) and that prosecuting their claims in a class action is superior to other available 19 methods (superiority). 20 A. Rule 23(a) Requirements 21 1. Ascertainability and Numerosity 22 Before analyzing numerosity under Rule 23(a)(1), courts typically require a showing that 23 the class to be certified is ascertainable. See Daniel F. v. Blue Shield of Cal., 305 F.R.D. 115, 24 121–22 (N.D. Cal. 2014); 7A Charles Alan Wright et al., Federal Practice and Procedure § 1760 at 25 142–47 (3d ed. 2005). To be ascertainable, the definition of the class must be “definite enough so 26 that it is administratively feasible for the court to ascertain whether an individual is a member” 27 before trial, and by reference to “objective criteria.” Daniel F., 305 F.R.D. at 122; see also 1 June 9, 2015) (discussing ascertainability requirement). Put differently, the Court must identify 2 “the persons (1) entitled to relief, (2) bound by a final judgment, and (3) entitled under Rule 3 23(c)(2) to the ‘best notice practicable’ in a Rule 23(b)(3) action.” Daniel F., 305 F.R.D. at 121 4 (quoting Manual for Complex Litigation, Fourth § 21.222 (2004)). 5 Plaintiffs seek to certify a class of drivers who have driven for Uber in the state of 6 California since February 28, 2019 and who opted out of Uber’s arbitration agreement. 7 Membership in this class is objectively ascertainable from Uber’s business records. See O’Connor 8 v. Uber (O’Connor II), No. C-13-3826 EMC, 2015 WL 5138097, at *8 (N.D. Cal. Sept. 1, 2015). 9 Uber does not dispute that it maintains business records with respect to each of its drivers, nor is 10 there any dispute that those records will reveal whether each putative class member drove for Uber 11 in the state of California since February 28, 2019. In fact, Uber has already identified 4,828 12 putative class members,1 which represent “all individuals who completed at least one ride on the 13 Uber app in California between February 28, 2019 and August 31, 2020, attempted to opt out of 14 arbitration, and have a California driver’s license.” Docket No. 94 (Decl. of Justin McCrary in 15 Support of Opp’n to Mot. (“McCrary Decl.”)) ¶ 11, n. 4. The ascertainability requirement is 16 therefore satisfied here. 17 A plaintiff satisfies the numerosity requirement if “the class is so large that joinder of all 18 members is impracticable.” Hanlon v. Chrysler Corp., 150 F.3d 1011, 1019 (9th Cir. 1998) 19 (quoting Fed. R. Civ. P. 23(a)(1)). While no court has set the precise number of class members 20 that are needed to satisfy the numerosity requirement, there is general recognition that Rule 21 23(a)(1) is satisfied when the proposed class contains one hundred or more members. See, e.g., 22 Wang v. Chinese Daily News, 231 F.R.D. 602, 607 (C.D. Cal. 2005) (recognizing there is a 23 presumption of numerosity where the proposed class contains one hundred or more members), 24 reversed on other grounds by 737 F.3d 538 (9th Cir. 2013); Ikonen v. Hartz Mountain Corp., 122 25 F.R.D. 258, 262 (S.D. Cal. 1998) (finding a proposed class of forty members sufficient to satisfy 26

27 1 Uber explains that this number might be an overestimation because it includes folks who 1 numerosity). 2 Uber is not disputing that the numerosity requirement is satisfied here, nor could it, given 3 that it has already identified almost five thousand putative class members. McCrary Decl. ¶ 11. 4 Even if that number is an overestimation, there is little doubt that at least one hundred of the 5 roughly five-thousand individuals who drove for Uber in California since February 2019 will meet 6 the class definition. The numerosity requirement is therefore also satisfied here. 7 2. Commonality 8 In order to satisfy Rule 23(a)(2)’s commonality requirement, a plaintiff must “affirmatively 9 demonstrate” that their claims depend upon at least one common contention the truth or falsity of 10 which “will resolve an issue that is central to the validity” of each one of the class members’ 11 “claims in one stroke.” Wal-Mart, 564 U.S. at 350. That is, the lawsuit must call upon the court 12 or jury to decide at least one factual or legal question that will generate a common answer “apt to 13 drive the resolution of the litigation.” Id.; see also id. at 359 (holding that “[e]ven a single 14 [common] question” will suffice to satisfy Rule 23(a) (quoting Nagareda, The Preexistence 15 Principle and the Structure of the Class Action, 103 Colum. L. Rev. 149, 176, n. 110 (2003)). 16 In this case, whether Uber misclassified its drivers as independent contractors is a common 17 question that satisfies the commonality requirement. This Court has previously found that “the 18 common legal issue of whether all class members should be classified as employees or 19 independent contractors is one whose answer would not only be ‘apt to drive the resolution of the 20 litigation,’ but could in fact be outcome determinative.” O’Connor II, 2015 WL 5138097, at *8 21 (quoting Guifu Li v. A Perfect Franchise, Inc., No. 5:10-CV-01189-LHK, 2011 WL 4635198, at 22 *7 (N.D. Cal. Oct. 5, 2011)). In O’Connor II, the central question—which this Court determined 23 was capable of class-wide determination—was whether employees were misclassified as 24 independent contractors under the common-law multi-factor test laid out in the California 25 Supreme Court’s decision in S.G. Borello & Sons, Inc. v. Department of Industrial Relations, 769 26 P.2d 399 (Cal. 1989). See id. at *5–*6. 27 After this Court’s O’Connor II decision, however, the California Supreme Court adopted a 1 Operations W. v. Superior Court, 416 P.3d 1 (Cal. 2018), reh’g denied (June 20, 2018). Under 2 the so-called “ABC” test, workers are presumptively considered to be employees unless the 3 “hiring entity” establishes that the worker in question satisfies three conditions:

4 (a) that the worker is free from the control and direction of the hirer in connection with the performance of the work, both under the 5 contract for the performance of the work and in fact; and

6 (b) that the worker performs work that is outside the usual course of the hiring entity’s business; and 7 (c) that the worker is customarily engaged in an independently 8 established trade, occupation, or business of the same nature as that involved in the work performed. 9 10 Id. at 34 (emphasis added). The legislature has since codified the ABC test with the passage of 11 Assembly Bill No. 5 (AB 5), and the test is now embodied in section 2775 of the California Labor 12 Code. If the hiring entity fails to prove any one of these conditions, the plaintiffs will prevail on 13 the merits of the question of their status as employees. 14 As a preliminary matter, Uber vigorously argues that the ABC test does not apply to this 15 case because Uber is not a “hiring entity” under section 2775(b)(1) of the California Labor Code. 16 See Docket No. 92 (“Opp’n”) at 5–6; Cal. Lab. Code § 2775(b)(1) (“[P]erson providing labor or 17 services for remuneration shall be considered an employee rather than an independent contractor 18 unless the hiring entity demonstrates that all of the following conditions are satisfied.” (emphasis 19 added)). Uber’s reasoning is that “the drivers do not render services to [Uber]; rather, drivers are 20 [Uber’s] customers, who render services to [Uber’s] other customers, the riders, using the two- 21 sided platforms [Uber] developed.” People v. Uber Techs., 270 Cal. Rptr. 3d 290, 307 (Ct. App. 22 2020), as modified on denial of reh’g (Nov. 20, 2020). The California Court of Appeal recently 23 rejected this argument, holding that “[r]eading the term ‘hiring entity’ in context, we think the 24 phrase is used in Dynamex and in section 2775 for its neutrality, so that it covers both employment 25 status and independent contractor status, and thus does not presuppose an answer one way or 26 another.” Id. at 308 (emphasis added). Indeed, the People v. Uber Court correctly pointed out that 27 Uber’s “hiring entity” argument poses a “false dichotomy” because “drivers’ services performed 1 third party, benefitting each one.” Id. For example, the delivery services that drivers performed in 2 Dynamex could have been characterized as benefiting both Dynamex, the corporate dispatcher, 3 and the entities sending and receiving packages. Id. There is little doubt that even if customers 4 could be deemed a “hiring entity,” Uber is a “hiring entity” as well. Therefore, the Court 5 concludes that the ABC test applies to this case because Uber is a “hiring entity” under section 6 2775.2 Moreover, Uber’s argument that it is not a “hiring entity” under Dynamex itself presents a 7 legal question common to all members of the class. 8 Having determined that the ABC test applies, the question that will generate a common 9 answer “apt to drive the resolution of the litigation” is whether the class member drivers were 10 misclassified as independent contractors because Uber failed to satisfy one or more prongs of the 11 ABC test. Wal-Mart, 564 U.S. at 350. Indeed, if the court or the jury determines that Uber 12 satisfied all three prongs of the ABC test, this class action will have reached its end. Guifu Li, 13 2011 WL 4635198, at *7 (“If . . . Plaintiffs have been properly classified as independent 14 contractors, the Court need not consider the additional claims that Plaintiffs have raised.”). 15 Inversely, if the court or the jury determine that Uber did not satisfy any one of the prongs, then 16 Plaintiffs are to be—as a class—classified as employees. See Costello v. BeavEx, Inc., 810 F.3d 17 1045, 1059 (7th Cir. 2016) (“Because the [ABC] test is conjunctive, if [the hiring entity] cannot 18 satisfy just one prong of the test, its couriers must be treated as employees.”). Either way, the 19 outcome of the ABC test is “apt to drive the resolution of the litigation.” Guifu Li, 2011 WL 20 4635198, at *7. Therefore, the commonality requirement is satisfied here. 21 3. Typicality and Adequacy 22 Rule 23(a)(3) requires that “the [legal] claims or defenses of the representative parties [be] 23 typical of the claims or defenses of the class.” Fed. R. Civ. P. 23(a)(3). Representative claims are 24 2 Uber also argues that at least some of Plaintiffs’ claims—for example, those based on expense 25 reimbursements in 2019—are governed by S.G. Borello because Dynamex only applies to wage order claims and AB 5 did not take effect until January 1, 2020. Opp’n at 7. But whether 26 Dynamex applies to this claim—a legal issue in dispute itself—presents a common question. In any event, even if Borello applied, commonality is satisfied as this Court already determined in 27 O’Connor II that “the worker classification claim presents a common issue capable of class-wide 1 “typical” if they are “reasonably co-extensive with those of absent class members; they need not 2 be substantially identical.” Castillo v. Bank of Am., NA, 980 F.3d 723, 730 (9th Cir. 2020) 3 (quoting Hanlon, 150 F.3d at 1020). Thus, the “test of typicality is whether other members have 4 the same or similar injury, whether the action is based on conduct which is not unique to the 5 named plaintiffs, and whether other class members have been injured by the same course of 6 conduct.” Hanon v. Dataproducts Corp., 976 F.2d 497, 508 (9th Cir. 1992). Moreover, courts 7 may evaluate whether a named plaintiff is typical by determining whether she is “subject to unique 8 defenses which threaten to become the focus of the litigation.” Id. “Class certification should not 9 be granted if there is a danger that absent class members will suffer if their representative is 10 preoccupied with defenses unique to [her].” Id. 11 Rule 23(a)(4) requires that the putative class representative must “fairly and adequately 12 protect the interests of the class.” Fed. R. Civ. P. 23(a)(4). A named plaintiff satisfies the 13 adequacy test if the individual has no conflicts of interest with other class members and if the 14 named plaintiff will prosecute the action vigorously on behalf of the class. See Ellis v. Costco 15 Wholesale Corp., 657 F.3d 970, 985 (9th Cir. 2011). 16 As other courts and commentators have noted, the typicality and adequacy inquiries tend to 17 significantly overlap. See, e.g., Newberg on Class Actions § 3:32 (5th ed. 2015) (“Due to the 18 related nature of the two requirements and the frequency with which they are challenged on the 19 same grounds, many courts address the typicality and adequacy requirements in a single 20 inquiry.”). For instance, a named plaintiff who is subject to unique defenses (i.e., may not satisfy 21 typicality) may also have a conflict of interest with her fellow class members (i.e., be an 22 inadequate class representative). In light of this overlap, and because Uber and the parties largely 23 treat the two issues together in their briefs, the Court analyzes the representative Plaintiffs’ ability 24 to represent their fellow class members with respect to all of their claims. 25 Uber principally argues that Plaintiffs are not adequate because “independent studies” 26 show that the vast majority of drivers oppose reclassification and would be worse off if they are 27 1 reclassified as employees.3 Opp’n at 20. This Court already rejected this argument in O’Connor 2 II, concluding that “even if Uber had demonstrated some real tension between the goals of the 3 class representatives and some statistically significant percentage of the class members, courts 4 have refused to find inadequacy on these grounds.” 2015 WL 5138097, at *13. Indeed, district 5 courts in the Ninth Circuit routinely refuse to deny class certification where some—even most—of 6 the putative class members oppose it. See e.g., Guifu Li, 2011 WL 4635198, at *9 (“The fact that 7 all proposed class members may not like each other, or even that some potential class members 8 may prefer their current employment situation, is not sufficient to defeat adequacy.”); Norris– 9 Wilson v. Delta T. Grp., Inc., 270 F.R.D. 596, 606 (S.D. Cal. 2010) (“Just because potential class 10 members disagree with the spirit of an action doesn’t mean it shouldn’t be certified. It will almost 11 always be the case that some putative class members are happy with things as they are.”). Indeed, 12 as Judge Conti correctly explained in Smith v. Cardinal Logistics Management Corp., where 13 putative employees seek to invoke the protections afforded under California labor laws, the Court 14 “must be mindful” of the fact that “the protections conferred by [these laws] have a public purpose 15 beyond the private interests of the workers themselves.” No. 07-2104 SC, 2008 WL 4156364, at 16 *7 (N.D. Cal. Sept. 5, 2008) (quoting S.G. Borello, 769 P.2d at 399); see also Department of 17 Labor Administrator’s Interpretation No. 2015–1, 2015 WL 4449086, at *1 (July 15, 2015) 18 (noting as a public policy matter that “[m]isclassification also results in lower tax revenues for 19

20 3 In its opposition, Uber lists harms that will befall the putative class members if they are reclassified as employees, including that it would (1) “prevent hundreds of thousands of drivers 21 from earning any income through the app;” (2) cause drivers to make less money; (3) cause drivers to lose the flexibility to work whenever they want; (4) cause drivers to lose their ability to use 22 other ride share applications; (5) prevent drivers who operate transportation business from relying on employees and subcontractors; and (6) “jeopardize the emergency federal benefits available to 23 drivers as self-employed workers during an unprecedented fall in demand for rides due to the pandemic.” Opp’n at. 21–23. But the only basis that Uber submits for these claims are 24 declarations from its employees and several current drivers. These declarations are insufficient, because as this Court noted in O’Connor II, “it [is doubtful] that most Uber drivers or declarants 25 correctly understand the pertinent legal difference between being an employee and an independent contractor, or the potential consequences of this [litigation].” 2015 WL 5138097, at *13. 26 Moreover, these declarations are insufficient to definitely establish “that a victory for Plaintiffs in this lawsuit would require Uber to use ‘less flexible’ work schedules going forward.” Id. In any 27 event, as discussed above, the fact that some class members may oppose the suit and the relief 1 government and an uneven playing field for employers who properly classify their workers”). “It 2 would be antithetical” to the public interest embodied in California’s Labor Code to permit a 3 portion of Uber’s workforce “to frustrate the attempt by others to assert rights under California 4 labor law solely because [they] are satisfied with their current jobs.” Smith, 2008 WL 4156364, at 5 *7; Cf. Tony & Susan Alamo Found. v. Sec’y of Lab., 471 U.S. 290, 302 (1985) (“If an exception 6 to the [Fair Labor Standards Act] were carved out for employees willing to testify that they 7 performed work ‘voluntarily,’ employers might be able to use superior bargaining power to coerce 8 employees to make such assertions, or to waive their protections under the Act.”). Moreover, if 9 there are class members who truly object to the goals of this lawsuit, they are free to opt-out of the 10 putative class after it is certified. See Dalton v. Lee Publ’ns, Inc., 270 F.R.D. 555, 560–61. (S.D. 11 Cal. 2010). 12 Uber also argues that Messrs. James and Verhines are not typical Uber drivers for several 13 reasons. First, Uber contends that Messrs. James and Verhines cannot represent the interests of 14 most Uber drivers because they use the application full time. The Court has already noted in 15 O’Connor II that whether “some Uber drivers drive only part-time while others drive full-time . . . 16 is irrelevant for the class-certification analysis under Borello.” 2015 WL 5138097, at *17 n. 15. 17 Similarly here, the amount of time that drivers spend driving is irrelevant under the ABC test 18 because, under prong A, “the relevant question is Uber’s right to control its drivers’ schedules. 19 Because it uniformly has no such control, it is not surprising that there are significant differences 20 between class members with respect to the actual number of hours they spend driving for Uber.” 21 Id. 22 Second, Uber argues that Messrs. James and Verhines are not typical Uber drivers because 23 they do not use competitors’ applications and are not independent business owners who employ or 24 subcontract other drivers. These distinctions are potentially legally significant under prong C of 25 the ABC test, which requires the court or the jury to determine if the putative employee is 26 customarily engaged in an independently established occupation, trade, or business. Dynamex, 27 416 P.3d at 34. As an initial matter, as this Court recognized in O’Connor II, “to the extent that 1 drivers . . . the argument is really a commonality or predominance argument masquerading as a 2 typicality argument.” 2015 WL 5138097, at *10. That is because, “[i]f legally material 3 differences between class members are so substantial that the predominance or commonality tests 4 cannot be satisfied, then the typicality test likely cannot be satisfied either.” Id. Here, as will be 5 discussed more fully below, the fact that some Uber drivers use competing applications and work 6 for third-party transportation companies defeats predominance under prong C of the ABC test to 7 the extent plaintiffs seek to certify the entire class with respect thereto. But that does not 8 necessarily preclude certification of a class under prongs A and B of the ABC test.4 9 Third, and finally, Uber argues that Mr. Verhines is not typical because he is seeking 10 injunctive relief compelling drivers to be reclassified as employees even though he testified that he 11 no longer wants to drive for Uber. Reply at 25 (citing Am. Compl. ¶¶ 7, 78–79). Uber 12 mischaracterizes Verhines’s testimony; he simply expressed an aspiration to quit driving for Uber: 13 “I prefer not to have to work with them at all anymore, but I may have to. I don’t know.” Docket 14 No. 120 (Decl. of Shannon Liss-Riordan in Support of Reply (“Liss-Riordan Decl.”)), Ex. 9 at 15 131:8, 232:22-23. This is insufficient to find that Mr. Verhines is atypical. 16 Accordingly, the Court concludes that Messrs. James’s and Verhines’s claims are typical 17 of their fellow class members’ claims, and that both are adequate class representatives. 18 B. Rule 23(b)(3) Requirements 19 Having satisfied the Rule 23(a) criteria for all of their claims, Plaintiffs must next 20 demonstrate “the superiority of maintaining a class action and show that the questions of law or 21 fact common to class members predominate over any questions affecting only individual 22 members.” Castillo, 980 F.3d at 730 (quoting Mazza, 666 F.3d at 596 (quoting FRCP 23(b)(3))). 23 1. Predominance 24 “The predominance test of Rule 23(b)(3) is ‘far more demanding’ than the commonality 25 test under Rule 23(a)(2).” Villalpando v. Exel Direct Inc, 303 F.R.D. 588, 607 (N.D. Cal. 2014) 26 (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 624 (1997)). Predominance is satisfied 27 1 only “when common issues ‘represent a significant aspect of the case and they can be resolved for 2 all members of the class in a single adjudication.’” Edwards v. First Am. Corp., 798 F.3d 1172, 3 1182 (9th Cir. 2015) (quoting 7A Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, 4 Federal Practice & Procedure § 1777 (2d ed.1986)). “[T]he focus of the predominance inquiry” is 5 whether “a proposed class is ‘sufficiently cohesive to warrant adjudication by representation.’” 6 Amgen Inc. v. Conn. Ret. Plans & Tr. Funds, 568 U.S. 455, 469 (2013) (quoting Amchem, 521 7 U.S. at 623). But the rule “does not require a plaintiff seeking class certification to prove that each 8 element of their claim is susceptible to class-wide proof,” so long as one or more common 9 questions predominate. Castillo, 980 F.3d at 730 (quoting Amgen, 568 U.S. at 469); see also 10 Tyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1045 (2016) (“When ‘one or more of the 11 central issues in the action are common to the class and can be said to predominate, the action may 12 be considered proper under Rule 23(b)(3) even though other important matters will have to be 13 tried separately.’” (quoting 7AA C. Wright, A. Miller, & M. Kane, Federal Practice and 14 Procedure § 1778, pp. 123–124 (3d ed. 2005)). 15 Here, the Court must perform the predominance analysis twice. First, the Court will 16 consider whether common questions predominate with respect to the threshold employment 17 misclassification claim. “If they do not, then the inquiry ends there and class certification should 18 be denied.” Guifu Li, 2011 WL 4635198, at *12. Second, if “common questions predominate the 19 classification inquiry,” the Court should then consider Plaintiffs’ individual substantive claims “to 20 determine whether they also pass the predominance test.” Id.; see also Norris–Wilson, 270 F.R.D. 21 at 606 (following same analytical framework); Villalpando, 303 F.R.D. at 608 (same); Soto v. 22 Diakon Logistics (Del.), Inc., No. 08–cv–33–L(WMC), 2013 WL 4500693, at *8 (S.D. Cal. Aug. 23 21, 2013) (same). 24 a. Whether Class Members Are Employees or Independent Contractors 25 Uber argues that individual issues predominate as to each of the prongs of the ABC test. 26 These arguments fall flat as to prongs A and B. However, as explained more fully below, 27 individual issues pertaining to whether drivers engaged in an independently established business 1 adjudicating prongs A and B, which means that prong C will have to be adjudicated for each 2 plaintiff on an individual basis. 3 First, Uber argues that individual issues predominate as to prong A—whether the drivers 4 are “free from the control and direction of [Uber] . . . both under the contract . . . and in fact,” 5 Dynamex, 416 P.3d at 34; Cal. Lab. Code § 2775(b)(1)(A)—because since January 2020 Uber 6 changed its contract to give drivers much more flexibility and independence, creating several 7 “variations in drivers experience with the app,” Opp’n at 9–10. For example, Uber explains that 8 under the current contract “drivers can now view the estimated fare, destination, and other key 9 information before deciding to accept or decline a dispatch,” which has resulted in “wide variation 10 in drivers’ accepting the suggested price or departing upwards or downwards.” Id. at 9. Similarly, 11 Uber explains that the current contract allows Uber drivers (1) to develop relationships with 12 particular riders who can specifically request them, (2) use the Uber application infrequently 13 without penalty, (3) simultaneously use competing applications like Lyft, (4) contract directly with 14 passengers outside of Uber, (5) hire employees to drive on their behalf, (6) and decline ride 15 requests without facing any penalties. Id. at 10. Therefore, Uber argues that a class cannot be 16 certified because some drivers avail themselves of these flexible options more than others, leading 17 to varying experiences while using the application. 18 The problem with this argument is that all the drivers are subject to the same uniform 19 terms of Uber’s standardized “Platform Access Agreement,” even if only some of them take 20 advantage of the new flexible terms of that agreement. See Docket No. 93 (Decl. of Brad 21 Rosenthal in Support of Opp’n to Mot. (“Rosenthal Decl.”)) ¶ 50. Uber is not arguing, for 22 example, that only some drivers are offered the option to use the application infrequently, or that 23 only some drivers are allowed to simultaneously use the Lyft app. In other words, a jury could 24 very well review the terms of Uber’s current standard agreement and conclude “in one stroke” that 25 Uber satisfies prong A of the ABC test because it does not exercise sufficient control or direction 26 over the drivers. Wal-Mart, 564 U.S. at 350. As this Court noted in O’Connor II, “this factor will 27 likely weigh in [Uber’s] favor on the merits, [but] the fact that Uber admits that it exercises a 1 class certification stage because it proves that this factor can be adjudicated on a class-wide basis.” 2 2015 WL 5138097, at *17. Therefore, the varying experiences of drivers on the app matters not 3 for class certification purposes because whether Uber drivers are given sufficient flexibility such 4 that they are not under Uber’s control for purposes of prong A is susceptible to common and class- 5 wide proof, namely, the terms of Uber’s current agreement. See Id. at *19 (“To the extent that 6 Uber has uniformly retained a right to discharge its drivers at will in its standardized form 7 contracts, this factor weighs heavily in favor of class certification.”); Ayala v. Antelope Valley 8 Newspapers, 59 Cal. 4th 522, 534 (2014) (explaining evidentiary importance of form employment 9 contracts). The issue under prong A is Uber’s right to control whether drivers use the application 10 or not (and conversely, drivers’ right to choose among options made available by Uber); the same 11 questions apply to all drivers and are thus commonly shared. See Villalpando, 303 F.R.D. at 608 12 (recognizing that “uniform contracts are a significant focus of the ‘right to control’ inquiry”) 13 (citing Alexander v. FedEx Ground Package Sys., Inc., 765 F.3d 981, 989–94 (9th Cir. 2014) and 14 Ruiz v. Affinity Logistics Corp., 754 F.3d 1093, 1102 (9th Cir. 2014)). 15 Uber further argues that class certification is inappropriate under prong A because Uber 16 has used three different agreements over the relevant class period, each of which exercised varying 17 control over Uber’s drivers. See Opp’n at 8, n. 3; Rosenthal Decl. ¶ 50 (“During the relevant class 18 period, there were two other basic platform use agreements that were used: the November 25, 19 2019 Technology Services Agreement and the December 11, 2015 Technology Services 20 Agreement.”). For example, although the current agreement allows drivers full discretion as to 21 when to drive, previous agreements required drivers to fulfill at least one ride per month or risk 22 deactivation. Compare Rosenthal Decl., Ex. 1 §1.2, with id., Ex. 2 § 2.1, and id., Ex. 3 § 2.1. 23 Similarly, prior agreements allowed Uber to deactivate drivers if they failed to maintain a 24 minimum average star rating, which is no longer the case under the current agreement. Compare 25 id., Ex. 1, with id., Ex. 2 § 2.52, and id. Ex. 3 § 2.52. 26 This argument is unpersuasive for at least three reasons. One, whether drivers are required 27 to fulfill at least one ride, or maintain a minimum rating, does not determine the outcome of the 1 control—over its drivers under all of the agreements, despite their differences. Two, Plaintiffs 2 contend that the current agreement incorporates by reference Uber’s Community Guidelines, 3 which state that “[i]f your rating is lower than the minimum average rating in your city, we will let 4 you know. And drivers . . . that don’t meet the minimum average rating for their city may lose 5 access to the Uber apps.” Liss-Riordan Decl., Ex. 4 (Uber Community Guidelines); id., Ex. 11 6 (“Rosenthal Dep.”) at 179:2-180:3 (confirming that drivers are automatically deactivated if their 7 ratings fall below a threshold determined by Uber). If that is true, then Uber’s characterization 8 that the agreements are materially different is incorrect. Three, and most importantly, the 9 existence of multiple agreements is insufficient to establish that individual issues predominate 10 under prong A because the Court could easily certify subclasses consisting of drivers who 11 provided services under each of the three operative agreements throughout the class period. See 12 Fed. R. Civ. P. 23(c)(5) (“When appropriate, a class may be divided into subclasses that are each 13 treated as a class under this rule.”); Cf. Rodriguez v. Hayes, 591 F.3d 1105, 1123 (9th Cir. 2010) 14 (“To the extent there may be any concern that the differing statutes authorizing detention of the 15 various class members will render class adjudication of class members’ claims impractical or 16 undermine effective representation of the class, it may counsel the formation of subclasses.”); 17 Second, Uber argues that individual issues predominate as to prong B—whether the drivers 18 perform “work that is outside the usual course of [Uber’s] business” Dynamex, 416 P.3d at 34; 19 Cal. Lab. Code § 2775(b)(1)(B)—because some drivers purchased a flat subscription for a certain 20 number of leads within a set time period and pay no additional per-ride fee to Uber. Opp’n at 11. 21 Uber therefore contends that, at least for drivers who purchased a subscription, “Uber’s revenue is 22 not dependent on whether drivers complete trips, which renders their work outside the usual 23 course of Uber’s business.” Id. But, as the California Court of Appeal reasoned in People v. 24 Uber, subscriptions do not alter the basic fact that providing transportation is Uber’s core 25 business:

26 While these details relating to how drivers are compensated might to a limited extent bear on whether the drivers are free from Uber’s 27 direction and control or whether the drivers are engaged in an outside the usual course of its business under prong B. Quite to the 1 contrary, according to the People, “Drive Pass . . . financially incentivizes the Driver to accept every dispatched ride [and thus] . . . 2 is further evidence of why Drivers are within Uber’s usual course of business—to provide rides.” 3 4 270 Cal. Rptr. 3d at 316 (emphasis added). In other words, drivers complete trips for Uber’s 5 customers, regardless of how Uber charges its fee to drivers for using the Uber application (per- 6 trip or in bulk). 7 Uber also argues that “other central pieces of the Prong B” analysis require individualized 8 proof. For example, Uber points out that many drivers display its competitors’ logos or advertise 9 their own transportation companies on their cards and business cards, and that many of them 10 “engage in gig work only in their off-time from their other independent primary professions.” 11 Opp’n at 11. But the appropriate question under prong B is whether the drivers’ work of driving 12 customers from one place to another is central to Uber’s business of providing customers rides 13 from one place to another—not whether drivers “continuously performed” that work. Opp’n at 12. 14 This Court has repeatedly held that Uber and Uber’s drivers are both in the business of 15 transportation. See O’Connor v. Uber Techs., Inc., (O’Connor I), 82 F. Supp. 3d 1133, 1141 16 (N.D. Cal. 2015) (“[I]t is clear that Uber is most certainly a transportation company.”); O’Connor 17 II, 2015 WL 5138097, at *27 (“Uber’s business is the business of transportation. . . . It is equally 18 clear that Uber drivers are also in the business of transportation.”). What drivers do when they are 19 not driving for Uber—much like what a part-time employee does on their own free time—does not 20 change the fact that when they are doing work for Uber they are driving Uber’s customers around, 21 which is “within [Uber’s] usual course of business.” Dynamex, 416 P.3d at 34. 22 Third, and finally, Uber correctly argues that individualized evidence is required to 23 determine, under prong C, whether “the worker is customarily engaged in an independently 24 established trade, occupation, or business of the same nature as that involved in the work 25 performed” for Uber. Dynamex, 416 P.3d at 36; Cal. Lab. Code § 2775(b)(1)(C). In O’Connor II, 26 this Court declined to certify a “mega-class of all Uber drivers who have ever driven for Uber 27 since 2009” because there was “tremendous (and likely material) variance between those class 1 indirectly through a distinct third-party business that had itself contracted with Uber to provide 2 driving services—and those who did not.” 2015 WL 5138097, at *22 (emphasis added). The 3 variance was caused by the fact that drivers who owned or worked for third-party transportation 4 companies “did not drive exclusively Uber,” which meant that it was “at least plausible that a jury 5 could find that” they were “engaged in a distinct occupation or business” such that they would be 6 classified as independent contractors under the S.G. Borello test.5 Id. at *23. This Court 7 nonetheless certified a class that excluded drivers who owned or worked for third-party 8 transportation companies, leaving open the possibility that, on a later date, “Plaintiffs could 9 demonstrate that some such drivers could participate in a class action via an appropriately defined 10 subclass or subclasses where there are no material variations within such subclass.” Id. (emphasis 11 added). 12 A couple of months later, the O’Connor Plaintiffs filed a motion precisely to certify a 13 subclass of Uber drivers laboring for third-party transportation companies. See O’Connor v. Uber 14 Technologies (O’Connor III), 311 F.R.D. 547, 552–53 (N.D. Cal. 2015). This Court denied their 15 motion because there were “potential significant variations” within that subclass, “such as the 16 number of hours that a driver drove for Uber clients versus other clients—that could cause a jury 17 to reach different results” among those drivers. Id. at 553. Importantly, the Court refused to 18 certify this subclass because the O’Connor plaintiffs had “not offered a concrete proposal” for 19 ascertaining which drivers were members of this subclass (or classes):

20 “For example, Plaintiffs have not submitted any proof that they could objectively identify all drivers who drove for Uber more than 21 30 hours per week. Alternatively, if a subclass were to be defined by the percentage of rides given to Uber customers versus customers 22 obtained from other sources, Plaintiffs have not shown that they could objectively determine whether a driver was more like Ezzikhe 23 or Gebretensia (i.e., drive solely for Uber) or more like Enriquez or Alshara (i.e., drive for Uber about 30% of the time).” O’Connor II, 24 2015 WL 5138097, at *24. Despite being given the opportunity, Plaintiffs again fail to meet this burden. 25 26 5 The first S.G. Borello factor—“whether the one performing services is engaged in a distinct 27 occupation or business”—mirrors prong C of the ABC test—“whether the worker is customarily 1 Id. at 552. Therefore, the Court concluded that the SG Borello test could not be applied on a class- 2 wide basis to Uber drivers who also worked for third-party transportation companies because it 3 would be impossible to determine whether they engaged in a distinct occupation or business. Id.; 4 see also Bowerman v. Field Asset Servs., Inc. (Bowerman I ), No. 13–cv–0057–WHO, 2014 WL 5 4676611, at *10 (N.D. Cal. Sept. 17, 2014) (initially denying class certification for want of 6 predominance, and remarking that “[i]t is critical to me” that some proposed class members “have 7 independent businesses and do not work full time for [Defendant]”); Sotelo v. Media News Grp., 8 Inc., 143 Cal. Rptr. 3d 293, 308 (Cal. App. 2012) (affirming denial of class certification where the 9 trial court had found a materially significant difference between proposed class members with 10 respect to the “distinct business” Borello factor). 11 Here, the variations that made it impossible to certify a class that included Uber drivers 12 who owned or worked for third-party transportation companies in O’Connor II and III also pertain 13 to prong C. Indeed, as Uber underscores in its opposition, some of the putative class members in 14 the instant case also “use Uber’s competitors’ apps at the same time that they use Uber and build 15 their own clientele base.” Opp’n at 12. In addition, some drivers incorporate their own 16 businesses, obtain commercial driver’s licenses, advertise under their own name, target their 17 driving periods during surge pricing on the Uber application, and identify as self-employed on 18 their tax returns. Id. A finder of fact could plausibly conclude that some of these putative class 19 members are engaged in an independently established business under prong C of the ABC test, 20 while others are not. This requires an intensive fact-based inquiry on a case by case basis. There 21 is no way to resolve this question for all of these drivers, one way or the other, “in one stroke.” 22 Wal-Mart, 564 U.S. at 350. 23 Unlike the SG Borello test, however, the ABC test is conjunctive, which means that drivers 24 could be categorized as Uber’s employees even if some of them are engaged in an independently 25 established trade under prong C, as long as prongs A and B are not satisfied. Therefore the Court 26 will certify the proposed class so that the trier of fact can adjudicate prongs A and B on a class- 27 wide basis, leaving prong C to be decided on an individual basis. If the trier of fact finds that 1 employees irrespective of prong C. Conversely, if the trier of fact decides that Uber satisfies 2 prongs A and B, the jury would have to determine on an individual basis whether each individual 3 driver also satisfies prong C. Either way, class certification is appropriate, and would 4 considerably advance the litigation, because adjudicating two out of the three prongs of the ABC 5 test using class-wide proof is much more efficient than adjudicating all prongs on an individual 6 basis. 7 Accordingly, the predominance requirement is satisfied for the misclassification question 8 with respect to the adjudication of prongs A and B of the ABC test. 9 b. Plaintiffs’ Substantive Claims 10 Even if “common questions predominate the threshold question of employee 11 classification,” the Court must still consider whether “Plaintiffs’ individual claims . . . also pass 12 the predominance test.” Guifu Li, 2011 WL 4635198, at *12; Soto, 2013 WL 4500693, at *8. The 13 Court concludes that only plaintiffs expense reimbursement and itemized pay statements claims 14 satisfy the predominance requirement and are thus suitable for class-wide determination. 15 Accordingly, Plaintiffs will have to adjudicate their minimum wage, overtime, and paid sick leave 16 claims on an individual basis. 17 i. Expense Reimbursement 18 Plaintiffs are only seeking reimbursement of vehicle and phone-related expenses, see 19 Reply at 9, which this Court has already found to be appropriate for class certification in 20 O’Connor III. See 311 F.R.D. at 567 (“The Court finds that certifying vehicle-related and phone 21 expenses will not cause individualized issues to predominate.”). Uber does not raise any new 22 arguments that would disturb that conclusion. 23 Accordingly, the predominance requirement is satisfied for Plaintiffs’ expense 24 reimbursement claim because a court or a jury could determine, on a class-wide basis, that Uber’s 25 failure to reimburse drivers for phone-and-vehicle-related expenses violates California’s expense 26 reimbursement laws. See Castillo, 980 F.3d at 730 (“Individual differences in calculating the 27 amount of damages will not defeat class certification where common issues otherwise 1 (“Our precedent is well settled on this point. In Yokoyama, we held that ‘damage calculations 2 alone cannot defeat certification.’ That is, the ‘amount of damages is invariably an individual 3 question and does not defeat class action treatment.’” (quoting Yokoyama v. Midland Nat. Life Ins. 4 Co., 594 F.3d 1087, 1094 (9th Cir. 2010)). 5 ii. Itemized Pay Statements 6 Uber’s only argument against certification of Plaintiffs’ itemized pay statement claims is 7 that they are “derivative of an employee’s right to wages themselves,” such that they rise or fall 8 with Plaintiffs’ other wage and hour claims. Naranjo v. Spectrum Sec. Servs. Inc., 253 Cal. Rptr. 9 3d 248, 265–66 (Ct. App. 2019); see also Valadez v. CSX Intermodal Terminals, Inc., 298 F. 10 Supp. 3d 1254, 1269 (N.D. Cal. 2018), reconsideration denied, No. 15-CV-05433-EDL, 2019 WL 11 1975460 (N.D. Cal. Mar. 15, 2019) (“Defendant acknowledges that the wage statement violation 12 claim ‘succeeds or fails’ based on employment status.”). 13 Here, the predominance requirement is satisfied for Plaintiffs’ itemized wage statements 14 claims because it is also satisfied for the misclassification claim. See Johnson v. Serenity Transp., 15 Inc., No. 15-CV-02004-JSC, 2018 WL 3646540, at *15 (N.D. Cal. Aug. 1, 2018), reconsideration 16 denied, No. 15-CV-02004-JSC, 2018 WL 9782170 (N.D. Cal. Oct. 12, 2018), and aff’d, 802 F. 17 App’x 250 (9th Cir. 2020) (“If the drivers were employees and not independent contractors—a 18 question the Court has already decided is subject to common proof—the wage statement and 19 waiting time claims are also subject to common proof. Accordingly, Plaintiffs’ wage statement 20 and waiting time claims are suitable for common resolution.”). In other words, if the trier of fact 21 decides that Uber’s drivers are employees, then Uber was required to provide them itemized pay 22 statements. Determining whether itemized pay statements were provided consistent with 23 California Labor Code § 226(a) and Wage Order 9-2001, can be determined on a class-wide basis 24 since Uber’s practice in this regard is uniform. 25 Accordingly, the predominance requirement is satisfied for Plaintiffs’ itemized pay 26 statements claims. 27 iii. Minimum Wage and Overtime 1 suggested fares based on the time and distance traveled and offers surge pricing to account for 2 real-time fluctuations in supply and demand.” Opp’n at 13. Similarly, some drivers only work 3 during peak times, receive significant tips from riders, and most importantly, as of July 2020, set 4 their own fares. See Rosenthal Decl. ¶ 37. Other drivers purchase a set number of leads for a flat 5 fee (a subscription), which allows them to retain the entirety of the fare passengers pay thereafter. 6 Rosenthal Decl. ¶¶ 42–43. Given the varying modalities that Uber uses to pay its drivers, Uber 7 contends that Plaintiffs’ minimum wage and overtime claims are not suitable for class-wide 8 determination because “the Court would have to conduct exhaustive individualized analyses just 9 to determine each driver’s hourly pay.” Opp’n at 14. 10 Under Ninth Circuit law, to show predominance “plaintiffs must be able to show that their 11 damages stemmed from the defendant’s actions that created legal liability.” Vaquero, 824 F.3d at 12 1154 (emphasis added) (quoting Pulaski v. Middleman, LLC v. Google, Inc., 802 F.3d 979 987–88 13 (9th Cir. 2015)); see also Castillo, 980 F.3d at 730 (“[I]f the plaintiffs cannot prove that damages 14 resulted from the defendant’s conduct, then the plaintiffs cannot establish predominance.” 15 (quoting Vaquero, 802 F.3d at 1154)). 16 Uber’s argument that “‘a case-by-case analysis would be required to determine’ Uber’s 17 liability as to each individual driver, precluding class-wide adjudication,” Opp’n at 15, is 18 unconvincing because its deliberate choice to not have a pay policy that guarantees minimum 19 wage and overtime pay applies to all drivers, regardless of how they use the application or how 20 much money they make driving for Uber. In other words, the jury could determine on a class- 21 wide basis that Uber violated California’s minimum wage and overtime laws precisely by not 22 having a pay policy or pay formula that guarantees drivers are paid minimum wage for all hours 23 driven and overtime for every hour above forty per week spent driving. Mot. at 15, n. 15. 24 The Ninth Circuit’s decision in Vaquero is instructive on this point. See 824 F.3d at 1154– 25 55. There, the plaintiff was a sales associate claiming that the defendant paid its sales associates 26 only a commission, even though it required them to complete many tasks that were unrelated to 27 sales, which violated California’s law that “proscribes compensation through commission for 1 Co., 978 P.2d 2, 10 (Cal. 1999)). The Vaquero Court concluded that the plaintiffs had satisfied 2 predominance because

3 Defendants’ consciously chosen compensation policy deprived the class members of earnings in violation of California’s minimum 4 wage laws. In a wage and hour case, unlike in an antitrust class action, the employer-defendant’s actions necessarily caused the 5 class members’ injury. Defendants either paid or did not pay their sales associates for work performed. No other factor could have 6 contributed to the alleged injury. 7 Id. at 1154–55. Same here, Uber’s consciously chosen policy of not guaranteeing minimum wage 8 or overtime deprived its drivers of earnings in assured compliance with California’s minimum 9 wage and overtime laws. This straight-forward question of liability is suitable for class-wide 10 determination. 11 Importantly, the Vaquero court also explained that predominance was satisfied “even if the 12 measure of damages proposed [] is imperfect,” because “it cannot be disputed that the damages (if 13 any are proved) stemmed from Defendants’ actions.” Id. at 1155; see also Pulaski, 802 F.3d at 14 987 (“We explained that Comcast stood for the proposition that ‘plaintiffs must be able to show 15 that their damages stemmed from the defendants’ actions that created the legal liability.’” (quoting 16 Leyva v. Medline Indus. Inc., 716 F.3d 510, 514 (9th Cir. 2013))). In doing so, the Vaquero court 17 reiterated the longstanding principle that “damage calculations alone cannot defeat class 18 certification.” Id. (quoting Yokoyama, 594 F.3d at 1094); see also Leyva, 716 F.3d at 514 (“[T]he 19 presence of individualized damages cannot, by itself, defeat class certification under Rule 20 23(b)(3).”); Blackie, 524 F.2d at 905 (“The amount of damages is invariably an individual 21 question and does not defeat class action treatment.”). 22 Even if Plaintiffs’ damages stem from Uber’s pay policy, however, class-wide adjudication 23 of Plaintiffs’ minimum wage and overtime claims is made problematic by Plaintiffs’ insistence 24 that “all time that drivers are on the [Uber] app and thus ready and willing to accept ride requests 25 is compensable time.” Mot. at 15, n. 15. This Court already recognized in Yucesoy v. Uber 26 Technologies, Inc., that determining whether an employee’s waiting or “on-call” time is 27 compensable for purposes of overtime and minimum wage violations “is a fact-specific inquiry 1 call. No. 15-CV-00262-EMC, 2016 WL 493189, at *5 (N.D. Cal. Feb. 9, 2016). In other words, 2 “[t]he classification of ‘on-call’ time is a highly fact-driven, context-dependent inquiry, which is 3 ill-suited to class treatment.” Ladore v. Ecolab, Inc., No. CV1109386GAFFMOX, 2012 WL 4 12861141, at *12 (C.D. Cal. Apr. 11, 2012), amended, No. CV 11-9386 FMO (JCX), 2013 WL 5 12246503 (C.D. Cal. Aug. 2, 2013). 6 Under Ninth Circuit and California law, courts weigh several factors to determine if an 7 employee should be compensated for on-call time:

8 (1) whether there was an on-premises living requirement; (2) whether there were excessive geographical restrictions on 9 employee’s movements; (3) whether the frequency of calls was unduly restrictive; (4) whether a fixed time limit for response was 10 unduly restrictive; (5) whether the on-call employee could easily trade on-call responsibilities; (6) whether use of a pager could ease 11 restrictions; and (7) whether the employee had actually engaged in personal activities during call-in time. 12 13 Berry v. County of Sonoma, 30 F.3d 1174, 1183 (9th Cir. 1994) (quoting Owens v. Loc. No. 169, 14 Ass’n of W. Pulp and Paper Workers, 971 F.2d 347, 350 (9th Cir. 1992)); Mendiola v. CPS Secs. 15 Sols. Inc., 340 P.3d 355, 360 (Cal. 2015) (same). 16 Here, as in Yucesoy, the trier of fact’s determination of compensability of waiting time is 17 complicated by the need to balance the Owens factors to determine if the putative class members’ 18 time is compensable. This individualized determination turns on “how often these [ride] requests 19 [come] in, how many of the requests they must accept, and the magnitude of the risk of 20 deactivation if requests are not accepted. Without such information, it is unclear what ability 21 drivers have to conduct personal business while logged onto the app.” 2016 WL 493189, at *6. 22 See also Ladore, 2012 WL 12861141, at *12 (denying class certification because Ladore did “not 23 offer[] any evidence to suggest that all putative class members’ Weekend Duty was all alike such 24 that it would all be or all would not be compensable overtime under such a fact-driven analysis.”). 25 Presumably, some of the drivers in Plaintiffs’ putative class “engaged in personal activities”— 26 including paid work for third parties—while waiting to receive ride requests on the Uber app; 27 others may not or may do so only occasionally. Determining Uber’s liability for wages cannot be 1 manifold. 2 Accordingly, the Court cannot conclude that the predominance requirement is satisfied for 3 Plaintiffs’ minimum wage and overtime pay because Plaintiffs refuse to limit those claims to the 4 time that they spent actually driving for Uber. 5 iv. Paid Sick Leave 6 Uber first challenges the certification of Plaintiffs’ paid sick leave claims by arguing that 7 there is no class-wide evidence to establish that drivers gave “reasonable advance notification” of 8 a need for sick leave. Opp’n at 16–17 (citing Cal. Lab. Code § 246(m)). This argument has no 9 merit because “Uber does not dispute that, at the time this case was filed, it did not offer sick pay, 10 Plaintiffs need not plead that they requested it in order to state their claim.” Order on MTD at 6 11 (emphasis added). As Plaintiffs point out, it would be nonsensical to require a worker to request 12 permission to take paid sick leave from an employer that does not offer it. Reply at 13. 13 Uber also argues that Plaintiffs’ sick leave is not appropriate for class-wide determination 14 because “the amount of paid sick leave to which any driver may be entitled is directly tied to the 15 number of hours each driver works . . . which . . . is not subject to common proof.” Opp’n at 17. 16 Uber again conflates liability with damages. As with Plaintiffs’ minimum wage and overtime 17 claims, there is no dispute that Uber does not have a paid-sick-leave policy, such that any driver’s 18 inability to obtain sick leave necessarily “stemmed from” Uber’s choice not to offer it. Pulaski, 19 802 F.3d at 987. The amount of sick leave that each driver was entitled to is an “individualized 20 damages” calculation that “cannot, by itself, defeat class certification under Rule 23(b)(3).” 21 Leyva, 716 F.3d at 514. 22 As with Plaintiffs’ minimum wage and overtime claims, however, the Court cannot 23 determine if Uber denied Plaintiffs’ paid sick leave on a class-wide basis because Plaintiffs insist 24 that all of the time they spent on the application (including waiting time) should count towards 25 determining their sick leave. That determination is extremely difficult to make as to each driver 26 for reasons stated above. Accordingly, given Plaintiffs’ insistence, the predominance requirement 27 is not satisfied for Plaintiffs’ paid sick leave claims either. 1 c. Arbitration Opt-Out 2 Finally, Uber argues that it would require fact-intensive individualized inquiries to 3 determine which drivers opted out of its arbitration agreement. Opp’n at 17–18. First, as 4 Plaintiffs’ point out, this argument is particularly disingenuous given that the Court just certified a 5 settlement class of such individuals, and Uber conceded that the group was readily ascertainable 6 for settlement purposes. See O'Connor v. Uber Techs., Inc. (O’Connor IV), No. 13-CV-03826- 7 EMC, 2019 WL 4394401, *1 (N.D. Cal. Sept. 13, 2019). 8 Second, Uber does not identify what factual inquiries are necessary to determine if a driver 9 opted out of Uber’s arbitration agreement, other than checking to see if the person signed the opt- 10 out provision or not. In fact, Uber has already identified 4,828 putative class members, which 11 represent “all individuals who completed at least one ride on the Uber app in California between 12 February 28, 2019 and August 31, 2020, attempted to opt out of arbitration, and have a California 13 driver’s license.” McCrary Decl. ¶ 11, n. 4. To be sure, Uber explains that this might be an 14 overestimation because it includes folks who attempted, but where not eligible, to opt out. Id. But 15 Uber does not explain what it will take to weed those folks out and why it cannot easily be done 16 on a class-wide basis. 17 Accordingly, this Court will certify a class of Uber drivers who opted out of Uber’s 18 arbitration agreement between February 28, 2019 and December 16, 2020 19 2. Superiority 20 Finally, in addition to satisfying all four requirements of Rule 23(a) and the predominance 21 requirement of Rule 23(b)(3), Plaintiffs must also show that “a class action is superior to other 22 available methods for fairly and efficiently adjudicating the controversy.” Fed. R. Civ. P. 23 23(b)(3). With respect to the Court's “superiority” analysis, Rule 23(b)(3) suggests the Court 24 consider:

25 (A) the class members’ interests in individually controlling the prosecution or defense of separate actions; 26 (B) the extent and nature of any litigation concerning the 27 controversy already begun by or against class members; the claims in the particular forum; and 1 (D) the likely difficulties in managing a class action. 2 3 Fed R. Civ. P. 23(b)(3)(A)-(D). 4 Uber’s superiority challenge is focused entirely on the last Rule 23(b)(3) factor: the 5 difficulty of managing Plaintiffs’ class action. The Ninth Circuit has held that “a class action filed 6 under [Rule] 23(b)(3) must be ‘superior to other available methods’ of adjudication in light of any 7 ‘difficulties likely to be encountered in the management of a class action.’” Six (6) Mexican 8 Workers v. Ariz. Citrus Growers, 904 F.2d 1301, 1304 (9th Cir. 1990) (quoting Fed. R. Civ. P. 9 23(b)(3)). Manageability “encompasses the whole range of practical problems that may render the 10 class action format inappropriate for a particular suit,” Eisen v. Carlisle & Jacquelin, 417 U.S. 11 156, 164 (1974), which typically includes “potential difficulties in notifying class members of the 12 suit, calculation of individual damages, and distribution of damages,” Six (6) Mexican Workers, 13 904 F.2d at 1304. 14 Uber does not raise any real manageability concerns; it simply recycles its predominance 15 arguments, claiming that the case is unmanageable because “there are a host of individual issues 16 that would require mini-trials to resolve.” Opp’n at 18. As discussed above, the only issues that 17 will require individualized analysis are prong C of the ABC test and the individual minimum 18 wage, overtime, and paid sick leave claims. The Court can resolve everything else on a class-wide 19 basis. In fact, as discussed above, because common issues predominate with respect to prongs A 20 and B of the ABC test, it will “be far more efficient to resolve the question of employment status 21 on a class-wide, rather than individual, basis.” Dalton, 270 F.R.D. at 565; see also Estrada v. 22 FedEx Ground Package Sys., Inc., 64 Cal. Rptr. 3d 327, 338 (Ct. App. 2007). Similarly, 23 efficiency counsels in favor of litigating the merits of class members’ expense reimbursement and 24 itemized pay statement claims on a class-wide basis. 25 Accordingly, there are no foreseeable manageability issues because the Court can address 26 numerous common questions with common answers in a single trial. If Plaintiffs prevail at that 27 trial, then each individual Plaintiff can thereafter pursue their minimum wage, overtime, and paid 1 Fund v. Las Vegas Sands, Inc., 244 F.3d 1152, 1163 (9th Cir. 2001) (certifying class for purposes 2 of addressing common questions because “individualized issues are few, and most of them are 3 likely to be relatively easy” to adjudicate on an individual basis). This approach is clearly the 4 superior method of resolving these cases for the Plaintiffs, Uber, and the Court. 5 C. Proposition 22 6 1. Background 7 Prop 22 passed on November 3, 2020, repealing AB 5 with respect to app-based drivers 8 and declaring these drivers to be independent contractors, as long as the network company (in this 9 case, Uber) provides those drivers with specific wage and hour protections.6 The text of Prop. 22 10 was codified in Division 3, Chapter 10.5, of the California Business and Professions Code. 11 Specifically, under section 7451 of the California Business and Professions Code, “an app- 12 based driver is an independent contractor and not an employee or agent with respect to the app- 13 based driver’s relationship with [Uber] if the following conditions are met:

14 (a) The network company does not unilaterally prescribe specific dates, times of day, or a minimum number of hours during 15 which the app-based driver must be logged into the network company’s online enabled application or platform. 16 (b) The network company does not require the app-based driver to 17 accept any specific rideshare service or delivery service request as a condition of maintaining access to the network company’s 18 online enabled application or platform.

19 (c) The network company does not restrict the app-based driver from performing rideshare services or delivery services through 20 other network companies except during engaged time.

21 (d) The network company does not restrict the app-based driver from working in any other lawful occupation or business. 22 23 Cal. Bus. & Prof. Code. § 7451. 24 Under section 7453 of the same code, Uber is required to guarantee a “net earnings floor,” 25 comprised of 120% the minimum wage for “engaged hours” plus a $.30 per-mile compensation 26

27 6 Prop 22, Exempts App-Based Transportation and Delivery Companies from Providing Employee 1 for vehicle expenses (based on number of “engaged” miles). See id. § 7453. Prop 22 took effect 2 on December 16, 2020. 3 2. Effect On This Litigation 4 As a preliminary matter, the Court concludes that Prop 22 does not apply retroactively. 5 The Supreme Court and the California Supreme Court have long held that statutes are 6 presumptively prospective absent a clear intent that the statute be applied retroactively. See 7 Landgraf v. USI Film Prods., 511 U.S. 244, 265 (1994) (“The presumption against retroactive 8 legislation is deeply rooted in our jurisprudence, and embodies a legal doctrine centuries older 9 than our Republic.”); United States v. Sec. Indus. Bank 459 U.S. 70, 79–80 (1982) (“The principle 10 that statutes operate only prospectively, while judicial decisions operate retrospectively, is familiar 11 to every law student . . . retrospective operation will not be given to a statute which interferes with 12 antecedent rights . . . unless such be the unequivocal and inflexible import of the terms, and the 13 manifest intention of the legislature.”); Evangelatos v. Sup. Ct., 753 P.2d 585, 597 (Cal. 1988) 14 (“[There is a] common understanding that legislative provisions are presumed to operate 15 prospectively, and that they should be so interpreted ‘unless express language or clear and 16 unavoidable implication negatives the presumption.’” (quoting Glavinich v. Commonwealth Land 17 Title Ins. Co., 209 Cal. Rptr. 266, 272 (Cal. App. 1984)). This same principle applies to ballot 18 initiatives. See, e.g., Robert L. v. Sup. Ct., 69 P.3d 951, 900-01 (Cal. 2003) (“In interpreting a 19 voter initiative . . . , we apply the same principles that govern statutory construction.”) (quoting 20 Horwich v. Sup. Ct., 980 P.2d 927 (Cal. 1999)); Evangelatos, 753 P.2d at 598 (same); Strauss v. 21 Horton, 207 P.3d 48, 120-21 (Cal. 2009) (same). Unlike A.B. 5, Prop 22 does not contain any 22 statement that it is a declaration of current existing law, or any express retroactivity provision.7 23 Therefore, Proposition 22 does not apply retroactively. 24 Because Prop 22 does not apply retroactively, at most it forecloses damages in this case for 25 7 The text of Proposition 22 may be found at https://vig.cdn.sos.ca.gov/2020/general/pdf/topl- 26 prop22.pdf. The official title and summary of Proposition 22, “Changes Employment Classification Rules for App-Based Transportation and Delivery Drivers”, issued by Attorney 27 General of California Xavier Becerra, on January 2, 2020, may be found at 1 conduct that occurred after December 16, 2020. That is to say, Prop 22 limits the class period. 2 Plaintiffs nonetheless argue that the passage of Prop 22 does not necessarily limit the class 3 period in this case because “it remains to be seen whether Proposition 22 will apply to Uber.” See 4 Docket No. 136 (“Plaintiffs’ Supp. Br.”) at 4. Indeed, the statute is written so that Uber has to 5 first show that it complies with section 7451’s requirements before its drivers can be classified as 6 independent contractors. This showing, according to Plaintiffs, “will be a fact-based question 7 common to the class.” Id. at 4. Accordingly, Plaintiffs ask the Court to add whether Uber 8 satisfies section 7451’s requirements to the list of questions that can be decided in this case on a 9 class-wide basis. But the amended complaint, as currently written, contains no allegations that 10 Uber does not comply with section 7451’s requirements, which is what Plaintiffs must allege (and 11 eventually prove) in order to reclassify Uber’s drivers as employees and be able to bring wage and 12 hour claims against Uber after December 16, 2020. Therefore, it would be premature for the 13 Court to certify that question for determination on a class-wide basis. 14 Accordingly, the Court will end the class period on December 16, 2020, as Plaintiffs have 15 conceded that, assuming Uber complies with section 7451’s requirements, they will not be able to 16 show that Uber drivers are employees after that date. In deciding whether Uber misclassified 17 drivers as independent contractors between February 28, 2019 and December 16, 2020, the court 18 or the jury can apply the ABC test because Prop 22 was not yet the law of the land. 19 IV. CONCLUSION 20 For the foregoing reasons, the Court GRANTS in part Plaintiffs’ Rule 23(b)(3) motion to 21 certify a class of Uber drivers who drove for Uber in the State of California between February 28, 22 2019 and December 16, 2020, and who opted out of Uber’s arbitration agreement. The trier of 23 fact will first consider on a class-wide basis whether this class of drivers satisfies prongs A and B 24 of the ABC test. If these drivers do not satisfy either prong, then the trier of fact will proceed to 25 consider, also on a class-wide basis, this class’s expense reimbursement and itemized wage 26 statement claims. 27 The Court also DENIES in part Plaintiff’s Rule 23(b)(3) motion as it pertains to prong C 1 The parties are ordered to meet and confer regarding the contents and logistics of class 2 notice and other relevant procedural details in advance of the next case management conference, 3 to be held at 10:30 a.m., February 18, 2021. At that conference, the parties shall be prepared to 4 discuss any proposal regarding further class certification. A joint case management statement is 5 due by February 11, 2021. 6 This order disposes of Docket No. 56. 7 8 IT IS SO ORDERED. 9 10 Dated: January 26, 2021 11 12 ______________________________________ EDWARD M. CHEN 13 United States District Judge 14 15 16 17 18 19 20 21 22 23 24 25 26 27