RAZAK v. UBER TECHNOLOGIES, INC.

• Court: District Court, E.D. Pennsylvania
• Decided: June 4, 2024
• Docket: 2:16-cv-00573
• Status: Unknown

Opinion

IN THE UNITED STATES DISTRICT COURT FOR THE EASTERN DISTRICT OF PENNSYLVANIA

ALI RAZAK, et al. CIVIL ACTION

v. NO. 16-573

UBER TECHNOLOGIES, INC., et al.

MEMORANDUM RE: MOTIONS FOR JUDGMENT AS A MATTER OF LAW

Baylson, J. June 4, 2024 Following a hung jury, both Plaintiffs (ECF 280) and Defendants (ECF 282) in this case filed post-trial motions for judgment as a matter of law. For the reasons set forth below, both motions are DENIED. The Court will convene a second jury trial as of June 10, 2024. I. INTRODUCTION AND PROCEDURAL HISTORY

Plaintiffs are three UberBLACK drivers that operated in the Philadelphia area between 2013 and 2018. Plaintiffs allege that Uber “misclassified” Plaintiffs and other similarly situated drivers as independent contractors, rather than employees, thus precluding Plaintiffs from certain benefits and compensation under the Fair Labor Standards Act (“FLSA”), 29 U.S.C. §§ 201-219, the Pennsylvania Minimum Wage Act (“PMWA”), 43 P.S. §§ 333.101-333.115, and the Pennsylvania Wage Payment and Collection Law (“WPCL”), 43 P.S. §§ 260.1-260.45. ECF 299 at ¶ 1. This case has a lengthy procedural history, and the following is a short summary. After the case was filed in Court of Common Pleas of Philadelphia County and removed to this Court, ECF 1, extensive discovery took place. Following this Court’s denial of a number of pretrial motions, see, e.g., ECF 94, Defendants filed a motion for summary judgment as of January 26, 2018, ECF 114. In a thirty-eight-page memorandum dated April 11, 2018, this Court granted Defendants’ motion. ECF 124. The Third Circuit reversed, vacated, and remanded the case, finding that there were a number of material factual disputes that prevented summary judgment. Razak v. Uber Techs., Inc., 951 F.3d 137, 145 (3d Cir.), amended, 979 F.3d 192 (3d Cir. 2020) (“[W]here there are genuine questions of material fact that need resolution, these questions must go to a fact-finder.

This case presents such genuine disputes of material facts”). Following what the Parties reported were extensive but unsuccessful settlement discussions, a trial took place in this Court beginning on March 4, 2024. ECF 243. As stipulated by the Parties, that trial was limited to “the threshold liability question of whether the three individual Plaintiffs were Defendants’ employees under the FLSA, PMWA, and/or WPCL.” ECF 146. Likewise, the Parties limited the relevant time period to “only events occurring . . . prior to January 11, 2018.” Id. After a five-day trial, the jury could not unanimously agree on whether Plaintiffs had proved—by a preponderance of the evidence—that Plaintiffs were employees of Defendants. ECF 269. Following the jury’s report of a deadlock on this ultimate issue, the Court decided to submit

specific questions to the jury on (1) the six “economic reality” factors that guide the “misclassification" determination under the FLSA and PMWA, as set forth by the Third Circuit in Donovan v. DialAmerica Mktg., Inc., 757 F.2d 1376 (3d Cir. 1985), and (2) the ten factors that guide a similar holistic analysis under Pennsylvania’s WPCL, as endorsed by the Third Circuit in Williams v. Jani-King of Philadelphia Inc., 837 F.3d 314 (3d Cir. 2016). ECF 269. When the Court proposed use of this supplemental form, both parties initially objected. Defendants maintained that objection throughout trial, whereas Plaintiffs ultimately informed the Court that Plaintiffs did not “have an objection to this general process.” Trial Tr. at 5, Mar. 11, 2024, ECF 277. The Court went ahead with the poll, which showed that a majority of the jury, on the majority of the questions, favored Uber’s position that Plaintiffs had failed to prove that they were “employees.” ECF 269. Ultimately though, because the Court had not given clear advance notice to counsel that it was going to use such a poll, the Court determined that—even if it had been permissible to use an

advisory jury for a damages claim under the FLSA—the poll could not have had any legal significance at that time and place in this case. Trial Tr. at 10, Mar. 11, 2024, ECF 277; see also Bereda v. Pickering Creek Indus. Park, Inc., 865 F.2d 49, 52 (3d Cir. 1989).1 The Parties—both of whom had previously filed Rule 50(a) motions—each moved for Rule 50(b) post-trial relief in due course.2 ECFs 280, 282. While the Court has previously indicated to the Parties that the Court intended to deny those motions, or potentially hold the motions until after a second trial, the Court has not expressly ruled on them. Thus, those motions are currently before the Court. II. LEGAL STANDARD

Federal Rule of Civil Procedure 50(b) allows a district court to enter judgment as a matter of law, upon renewed motion after a hung jury, “only if, as a matter of law, the record is critically

1 That said, the potential insights gleaned from this more granular polling led the Court to require each party to submit jury interrogatories in advance of the second trial. See ECFs 301, 304. To the extent the jury provides unanimous answers to individual factual questions, this may provide the basis for a final judgment on the merits, even if the jury is unable to reach a verdict on the ultimate conclusion as to whether Plaintiffs have proved they were employees under the FLSA, PMWA, and WPCL. 2 Notably, in between summary judgment and trial, there were some important changes in the undisputed facts. Most relevant here, Defendants advised the Court that Defendants had ceased providing UberBLACK services in Philadelphia as of 2022 and had no intention of resuming UberBLACK services. Plaintiffs nonetheless pursue this case because if they are successful in demonstrating they are employees, Plaintiffs allege they will be entitled to substantial damages for the period between 2013 and 2018, the relevant time period for this case. See ECF 148. deficient of that minimum quantity of evidence from which a jury might reasonably afford relief.” Trabal v. Wells Fargo Armored Serv. Corp., 269 F.3d 243, 249 (3d Cir. 2001); see also Stewart v. Walbridge, Aldinger Co., 882 F. Supp. 1441, 1443 (D. Del. 1995); Greco v. Nat’l R.R. Passenger Corp., 2005 WL 3591196, at *4 (E.D. Pa. Dec. 30, 2005).

This remedy is to be invoked “sparingly,” Marra v. Phila. Hous. Auth., 497 F.3d 286, 300 (3d Cir. 2007), as a district court may grant judgment as a matter of law “only if, viewing the evidence in the light most favorable to the nonmovant and giving it the advantage of every fair and reasonable inference, there is insufficient evidence from which a jury reasonably could find liability,” Le Page’s, Inc. v. 3M, 324 F.3d 141, 145-46 (3d Cir. 2003) (citation and internal quotation marks omitted). In so doing, the court “may not weigh evidence, determine the credibility of witnesses or substitute its version of the facts for that of the jury.” Parkway Garage, Inc. v. City of Philadelphia, 5 F.3d 685, 691-92 (3d Cir. 1993), abrogated on other grounds by United Artists Theatre Cir., Inc. v. Twp. Of Warrington, 316 F.3d 392 (3d Cir. 2003); see also LePage’s, 324 F.3d at 146 (“[R]eview of a jury’s verdict is limited to determining whether some

evidence in the record supports the jury's verdict.”). Likewise, “conflicting evidence which could reasonably lead to inconsistent conclusions will not justify a judgment notwithstanding the verdict . . . . It is the function of the trier of fact alone . . . to evaluate contradictory evidence and to draw inferences therefrom.” Fireman's Fund Ins. Co. v. Videfreeze Corp., 540 F.2d 1171, 1178 (3d Cir. 1976), cert. denied, 429 U.S. 1053 (1977) (citations omitted). Thus, “[n]ormally, when the evidence is contradictory, a JNOV is inappropriate.” Bonjorno v. Kaiser Aluminum & Chemical Corp., 752 F.2d 802, 811 (3d Cir. 1984). III. PARTY CONTENTIONS

Defendants move for Rule 50(b) relief on each of Plaintiffs’ claims. ECF 282 at 1. Plaintiffs, by contrast, limit their cross-motion to their state law claims under the PMWA and WPCL.3 ECF 280. Notably, Plaintiffs’ motion hinges on the novel legal argument that the Pennsylvania Supreme Court—if given the opportunity—would “likely” adopt a new, worker- friendly classification test for these two statutes. ECF 280 at 2. This Court already rejected that argument at trial, Trial Tr. at 168, Mar. 7, 2024, ECF 275, but it takes this opportunity to expand on that ruling below. A. Defendants’ Contentions

Defendants contend that “[n]o reasonable jury could find that Plaintiffs were ‘employees’ under the FLSA, PMWA, and WPCL, which was the sole issue presented at trial.” ECF 282-1 at 3. In so doing, Defendants note that “while the Third Circuit held that disputed issues of fact precluded summary judgment, the evidence presented at trial—which has superseded the summary judgment record[]—is even more one-sided in favor of independent contractor status” for each of the six “economic reality” factors. Id. (citing to Dupree v. Younger, 598 U.S. 729, 734 (2023)). On the “right to control,” Defendants assert that (1) “[t]he evidence at trial revealed that the parties’ contracts explicitly disavow Uber’s right to control Plaintiffs in the performance of their services providing transportation using the Uber App,” and (2) “actual practice was consistent with the contracts,” with significant trial testimony indicating that Plaintiffs had the freedom to

3 Plaintiffs do not move for judgment as a matter of law on their FLSA claims, instead requesting the Court declare a mistrial due to lack of juror unanimity. Id. at 1. choose their own hours, select their own routes, and to work for Uber’s competitors. ECF 282-1 at 4-5 (record citations omitted). Defendants further argue that much of Plaintiffs’ counterevidence regarding Uber’s right to control amounted to inadmissible hearsay, as Plaintiffs repeatedly testified that “unidentified”

Uber personnel had imposed various rules, restrictions, or quality assurance measures. Id. at 6. Likewise, Defendants contend that “[m]any of the so-called ‘rules’ of which Plaintiffs complain were required by law, and thus could not constitute ‘control’ under the economic realities test.” Id. (citing to 29 C.F.R. § 795.110(b)(4)). On “opportunity for profit or loss,” Defendants assert the evidence at trial demonstrated that each of the three Plaintiffs (1) “owned and operated their own businesses and engaged in activities (like advertising) to drum up additional business outside of the relationship with Uber,” and (2) “could also be strategic when using the Uber App in order to maximize profit.” Id. at 7. On “investment in equipment or materials,” Defendants again direct the Court to the fact that Plaintiffs “paid for and owned their vehicles” and “made other substantial investments,

including in advertising.” Id. at 7. Defendants also assert that Plaintiffs’ attempts to highlight the “relative” investment by Uber in its platform misconstrue the scope permissible considerations for this factor. Id. at 7-8. On “special skills,” Defendants argue the record reflects that Plaintiffs are “professionally licensed driver[s]” who provided a “higher end” service,” and that “Plaintiffs demonstrated specialized skills related to their compliance with the commercial licensing requirements and generating revenue.” Id. at 8-9. On “degree of permanence,” Defendants reiterate Plaintiffs can and did provide transportation services outside of Uber, along with the fact that “Plaintiffs had complete control of their schedules.” Id. at 9-10. Finally, on “integrality,” Defendants argue that “Uber is a technology company,” rather

than “a transportation company like Plaintiffs’ operations,” and thus that “Plaintiffs are only ‘integral’ to Uber in the same way riders are; both are customers of Uber who utilize Uber’s technology services for a fee. But customers, obviously, are not employees.” Id. at 10-11. B. Plaintiffs’ Contentions

Plaintiffs counter that the trial record here not only precludes this Court from granting Defendants’ motion, but also necessitates Rule 50(b) relief for Plaintiffs on their PMWA and WPCL claims. ECF 280 at 1-2. As noted, a key tenet of Plaintiffs’ argument is that Plaintiffs believe there “is a real chance that the Pennsylvania Supreme Court will hold that the ABC test applies to these state law claims” and that “Uber has clearly failed to satisfy Prong B of the ABC test.” Id. at 9. That test, Plaintiffs explain, “places the burden on the alleged employer to prove that the worker is an independent contractor and presumes an employment relationship unless the employer can demonstrate each of the following three factors: (A) that the worker is free from the control and direction of the hiring entity in connection with the performance of the work, both under the contract for the performance of the work and in fact; and

(B) that the worker performs work that is outside the usual course of the hiring entity’s business; and

(C) that the worker is customarily engaged in an independently established trade, occupation, or business of the same nature as the work performed.” Id. at 10 (citing to Dynamex Operations West, Inc. v. Superior Court, 4 Cal. 5th 903, 957 (Cal. 2018)). And here, Plaintiffs contend, the trial record unequivocally reflects that (1) Uber was a transportation company, rather than merely a technology company, and (2) “without the drivers,

there would be no [transportation] business.” Id. at 13. In so arguing, Plaintiffs ask this Court to credit the fact that “[n]one of the jurors voted that this consideration supported independent contractor status.”4 Id. at 12. Likewise, Plaintiffs note that “other courts have reached the same conclusion that Uber is in the business of providing transportation” and “not merely a technology company.” Id. at 14 (citing to People v. Uber Technologies, Inc., 56 Cal. App. 5th 266, 292 (2020); James v. Uber Technologies, Inc., 2021 WL 254303, at *10 (N.D. Cal. Jan. 26, 2021); Namisnak v. Uber Technologies, Inc., 444 F. Supp. 3d 1136, 1143 (N.D. Cal. 2020; O’Connor v. Uber Technologies, Inc. 82 F. Supp. 3d 1133, 1141 (N.D. Cal. Mar. 11, 2015); see also Rogers v. Lyft, Inc., 452 F. Supp. 3d 904, 911 (N.D. Cal. 2020); Cotter v. Lyft, Inc., 60 F. Supp. 3d 1067, 1078- 79 (N.D. Cal. 2015)).5

As for the six Donovan factors and Defendants’ cross-motion, Plaintiffs similarly assert that sufficient evidence at trial would support a reasonable jury finding in Plaintiffs’ favor, thus precluding Rule 50 relief for Defendants.

4 More precisely, because the Court did not instruct the jury on Plaintiffs’ desired ABC test, Plaintiffs direct this Court to one of the ten WPCL factors, which asked the jury to consider “whether the work is part of the regular business of the employer,” Jani-King, 837 F.3d at 320, and which Plaintiffs deem analogous to prong B of the ABC test. 5 Plaintiffs further assert that Defendants have also independently failed to satisfy both prongs A and C of the ABC test, relying on the various forms of Uber’s right to control that Plaintiffs summarize in their Rule 50(a) motion. Id. at 15 (citing to ECF 258). That evidence substantially overlaps with the evidence Plaintiffs highlight in response to Defendants’ Rule 50 motion, which the Court summarizes immediately below. On Uber’s “right to control,” Plaintiffs contends the trial record here demonstrates that Uber: (1) maintained the right to terminate Plaintiffs for substandard service; (2) implemented a rating system that demanded near perfection from drivers; (3) warned and deactivated drivers with low ratings each week; (4) terminated drivers for 71 different potential “infractions”; (5) logged

infractions and kept notes in driver personnel files; (6) constantly surveilled Plaintiffs; (7) required drivers to accept a minimum number of trips; (8) required drivers to use luxury vehicles that were no more than five years old; (9) subjected drivers to its own background checks; and (10) imposed a stricter dress code for black car drivers. ECF 288 at 7-15. On “opportunity for profit or loss,” Plaintiffs contends that driver profits turned only on “hustle,” rather than actual entrepreneurial or strategic skill. Id. at 15. Specifically, Plaintiffs direct this Court to the testimony of former Uber customer representative liaison Tara Murray, who stated that “it’s all about, you know, hustle and knowing what you -- you know, knowing your job well.” Id. at 15 (record citations omitted). Plaintiffs further highlight trial evidence indicating that (1) Uber unilaterally set fares and rates; (2) controlled trip assignments and precluded drivers

from making in-formed business decisions; (3) decided what fees to charge customers; and (4) controlled business expenses through automatic deductions. Id. at 16-18. On “investment in equipment or materials,” after noting that certain courts have credited the “relative” investment as between a plaintiff and an alleged employer, Plaintiffs argue the trial record here reveals that “Uber has spent billions of dollars building and operating its platform,” which “towers over what Plaintiffs spent on their vehicles and other business expenses.” Id. at 18. On “special skills,” Plaintiffs assert the Third Circuit in this case already held that driving a limousine does not count as a special skill. Id. at 18-19 (citing to Razak, 951 F.3d at 147 (“Although there may be a distinction between ‘driving’ and ‘replicat[ing] the limousine experience,’ . . . this is not enough to overcome the presumption that driving is not a special skill.”)). On “degree of permanency,” Plaintiffs contend that each of the three Plaintiffs testified they “worked long hours over extended periods,” which demonstrates more than a mere

“transitory” relationship with Uber. Id. at 19. Finally, on “integrality,” Plaintiffs note that—at summary judgment—this Court already held that “Uber drivers are an essential part of Uber’s business as a transportation company,” id. at 20 (citing to Razak v. Uber Techs., Inc., 2018 WL 1744467, at *19 (E.D. Pa. Apr. 11, 2018) (Baylson, J.)), and that other courts have reached the same conclusion, id. (citing to O’Connor v. Uber Techs., Inc., 82 F. Supp. 3d 1133, 1142 (N.D. Cal. 2015); People v. Uber Technologies, Inc., 56 Cal. App. 5th 266, 292 (2020); Rogers v. Lyft, Inc., 452 F. Supp. 3d 904, 911 (N.D. Cal. 2020)). IV. DISCUSSION

At the outset of its analysis, the Court highlights that numerous policymakers, academics, and international tribunals have already devoted significant attention to the confounding legal question of worker status in the “gig economy.” Indeed, commentators worldwide have argued that existing legal frameworks are ill- equipped to properly classify rideshare workers, with many advocating for a “shift in labor law.” Alexander Kondo, Abraham Singer, Labor Without Employment: Toward A New Legal Framework for the Gig Economy, 34 ABA J. Lab. & Emp. L. 331, 336 (2020). Certain proposals have pushed for new, distinct worker designations, such as a “labor without employment” classification. Id. Others have championed a “default rule” within existing frameworks; one which would afford the employee designation to most gig workers. Miriam A. Cherry, Antonio Aloisi, “Dependent Contractors” in the Gig Economy: A Comparative Approach, 66 Am. U. L. Rev. 635, 640 (2017). The remainder have called for a middle ground approach, such as a gig economy-specific misclassification test. See, e.g., Alex Kirven, Whose Gig Is It Anyway? Technological Change, Workplace Control and Supervision, and Workers' Rights in the Gig Economy, 89 U. Colo. L. Rev. 249, 254 (2018) (comment).

Internationally, a diverse patchwork of regulatory and judicial solutions further highlight the complexity of this issue. In the United Kingdom and Switzerland, for example, the countries’ highest courts have ruled that Uber drivers are employees. See Uber BV v. Aslam [2021] IRLR 407; Uber loses appeal as top Swiss court rules company is an employer, Reuters (June 3, 2022, 9:52 AM), https://www.reuters.com/business/autos-transportation/uber-loses-appeal-top-swiss- court-rules-company-is-an-employer-2022-06-03/. By contrast, Australian courts have consistently found Uber drivers to be independent contractors, while emphasizing the case-by-case nature of the analysis. See, e.g., Zhuge v Uber Austl. Pty. Ltd. [2024] FWC 911 (Austl.); Kaseris v Rasier Pacific V.O.F [2017] FWC 6610; Pallage v Rasier Pacific Pty Ltd [2018] FWC 2579; see also Lucy Trevelyan, The Gig Economy: A New Global Battleground, 72 Int. Bar Assos. Glob.

Insight 45 (2018) (summarizing various countries’ legal responses to the rise of gig economy) The economics literature similarly stresses the pivotal role that worker classification issues play in shaping market forces. See, e.g., Lawrence Mishel, Uber and the Labor Market: Uber Drivers’ Compensation, Wages, and the Scale of Uber and the Gig Economy, Econ. Pol’y Institute 1-28 (2018); Seth D. Harris & Alan B. Krueger, Is Your Uber Driver an Employee or Independent Contractor?, 20 Persps. on Work 30-33, 80 (2016). While many commentators have argued that a shift in worker classification could be fatal to the gig economy business model, others have posited that reclassification may incentivize companies like Uber and Lyft “to raise the level of their hiring, compensation and benefit practices to support a more appropriate competitive strategy,” ultimately encouraging further competition and innovation by creating a more reliable product that increases both demand and the quantity of available drivers. Stan Malos, Gretchen Vogelgesang Lester and Meghna Virick, Uber Drivers and Employment Status in the Gig Economy: Should Corporate Social Responsibility Tip the Scales?, 30 Emp. Resps. and Rts. J. 239-251 (2018).

Despite these various proposals and empirical insights, however, the actual legal landscape—at least in the Eastern District of Pennsylvania—has not changed since the Third Circuit’s opinion in this case. This case still requires an “either/or” conclusion—either Plaintiffs have proven by a preponderance that they are employees, or they have failed to do so, rendering them independent contractors. This Court is not free to invent a third worker category, or to adopt the types of “hybrid” legal frameworks advocated for above. However enticing those solutions might be, embracing them would ignore many years of precedent and justified reliance by companies providing Uber- like services. Indeed, many individuals voluntarily and happily embark on the position of being a driver, recognizing some of the control that will govern the economic reality of the relationship.

This is a topic for a legislature to tackle, rather than a judge sitting on a dispute such as this case. A. Applicable Law Thus, this Court must address the Parties’ requests for Rule 50 relief as the law currently stands. Yet, as noted above, Plaintiffs predicate their motion on the argument that Pennsylvania law is in flux, asserting that “there is a real chance that the Pennsylvania Supreme Court will hold that the ABC test” applies to Plaintiffs’ PMWA and WPCL claims. ECF 280 at 9. This Court rejected that argument at trial, explaining briefly that (1) Dep’t of Labor & Indus. v. Stuber, 822 A.2d 870 (Pa. Commw. Ct. 2003), aff’d, 580 Pa. 66 (2004) controls claims under the PMWA, and (2) the Third Circuit’s decision in Williams v. Jani-King of Philadelphia, Inc. “predict[ed] that the Pennsylvania Supreme Court would employ [the ten-factor test charged to the jury here] in the context of the WPCL,” 837 F.3d 314, 321 (3d Cir. 2016); a prediction the Pennsylvania Supreme Court has yet to debunk. But because Plaintiffs continue to contend this Court’s conclusions and corresponding jury instructions were incorrect, the Court begins its Rule

50 analysis by expanding on these earlier rulings. To frame the inquiry, recall that Plaintiffs have brought claims under three similar but distinct statutes: (1) the federal FLSA, (2) the PMWA, and (3) the WPCL. ECF 299 at ¶ 1. Thus, for each statute, this Court must answer two questions. First, the Court must discern each statute’s substantive test for distinguishing employees from independent contractors. Second, the Court must ask whether each statute places the initial burden on Plaintiffs to prove employee status, or alternatively on Defendants to disprove it. In other words, the Court essentially confronts a three- by-two matrix of inquiries—i.e., for each of the three statutes, the Court must discern the relevant substantive test and the existence (or lack thereof) of a plaintiff-friendly presumption. i. FLSA

a. Substantive Test

The proper substantive test under the FLSA is not in dispute here, so the Court addresses it only briefly. Indeed, the Court need look no further than the Third Circuit’s decision in this very case. On appeal here, the Third Circuit explained the FLSA defines “employer” to include “any person acting directly or indirectly in the interest of an employer in relation to an employee,” and “employee” to include “any individual employed by an employer.” Razak, 951 F.3d at 142 (citing to 29 U.S.C. §§ 203(d), (e)(1)). In view of the “circularity of the[se] definitions,” however, the Third Circuit has long employed the “economic realty” test to discern employees from independent contractors. Id. at 143. This case is no different. For an FLSA claim, the holistic determination turns on the following six factors: 1) the degree of the alleged employer’s right to control the manner in which the work is to be performed;

2) the alleged employee’s opportunity for profit or loss depending upon his managerial skill;

3) the alleged employee’s investment in equipment or materials required for his task, or his employment of helpers;

4) whether the service rendered required a special skill;

5) the degree of permanence of the working relationship; [and]

6) whether the service rendered is an integral part of the alleged employer’s business.

Id. at 142-43 (citing to Donovan, 757 F.2d at 1382). As such, this Court has no trouble concluding that it properly instructed the jury on these factors. b. Burden

Likewise, the Razak Court expressly answered the burden inquiry for FLSA claims. In relevant part, the Court explicitly noted that “[t]he burden lies with Plaintiffs to prove that they are employees.” Id. at 143 (citing to Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, 686–87 (1946)). Therefore, this Court is convinced it also correctly instructed the jury here that Plaintiffs bore the burden of proving employee status under federal law. ii. PMWA

The proper substantive test under the PMWA is also relatively clear cut. Again, the Third Circuit’s decision here is a useful starting point, as the Court of Appeals held that “[a]lthough Plaintiffs’ case includes claims under the PMWA, Pennsylvania state courts have looked to federal law regarding the FLSA for guidance in applying the PMWA.” Id. at 142. In support of that contention, the Razak Court relied on Dep’t of Labor & Indus. v. Stuber, where the Pennsylvania Commonwealth Court expressly concluded that “because [the PMWA and

FLSA] have identity of purpose . . . federal case law, and the ‘economic reality’ test employed by the federal courts, is the appropriate standard to use.” 822 A.2d 870, 873 (Pa. Commw. Ct. 2003). Crucially, the Pennsylvania Supreme Court then affirmed “on the basis of the Commonwealth Court opinion.” Com. v. Stuber, 580 Pa. 66 (2004). Thus, in this Court’s view, the Pennsylvania Supreme Court has adopted the economic reality test for PMWA claims, and that decision has never been overturned or amended. Nonetheless, both pre- and post-trial, Plaintiffs have asserted that the Pennsylvania Supreme Court’s recent decision in In re Amazon.com, Inc., 255 A.3d 191 (Pa. 2021) dictates a different result. This contention is misguided. At issue in Amazon was whether the PMWA entitled Amazon employees to compensation for time spent on post-shift antitheft security

screenings. Id. at 193-94. Notably, the Pennsylvania Supreme Court considered this question following the Supreme Court’s 2014 decision in Integrity Staffing Solutions v. Busk, where the high Court concluded that—under the FLSA—this time was not compensable. 574 U.S. 27, 29- 31 (2014). Yet, the Amazon Court rejected the applicability of Busk. After noting that “the FLSA, by its own terms, specifically permits states ‘to enact more beneficial wage and hour laws’ than provided by the FLSA,” the Amazon Court explained that Busk was not even “of persuasive value in interpreting the PMWA in this instance, given the stark contrast in the governing provisions of the two laws.” 255 A.3d at 201 (citations omitted) (emphasis added). The Court clarified that while “the high Court’s decision in Busk rested on its interpretation of the federal PTPA, which specifically classifies ‘activities which are preliminary to or postliminary to’ a worker’s principal activity or activities as non-compensable under the federal FLSA,” Pennsylvania has “never statutorily adopted the federal PTPA’s specific classification of certain employee activities as

being exempt from compensation, even though the PMWA has been amended six times since its initial passage in 1968.” Id. at 201-202 (emphasis added). In so ruling, the Court actively rejected the applicability of Stuber, “disavowing” Amazon’s attempt to construe Stuber as holding “that the PMWA and federal FLSA mirror each other in all respects and must always be interpreted similarly.” Id. at 202 n.11. Plaintiffs advocate for the same result here, latching onto the Amazon’s Court Stuber proviso. That attempt is unavailing. Even if the Amazon Court’s admonishment of Stuber supports Plaintiffs’ position here in some generalized sense, Plaintiffs ultimately read it too broadly. Indeed, immediately preceding this commentary, the Amazon Court clarified that although Stuber “does not stand for the proposition that [courts] must construe the PMWA and the

federal FLSA in a parallel fashion,” where “the statutory provisions of the enactments parallel each other, federal decisions construing the federal FLSA may be considered in interpreting the PMWA.” Id. (emphasis added). Accordingly, nothing about the Amazon Court’s discussion of Stuber indicates that Stuber itself was wrongly decided. To the contrary, the Amazon Court actively distinguished Stuber on the grounds that the PMWA and FLSA “define the terms ‘employ,’ ‘employer’ and ‘employee’ in a manner which is ‘virtually identical for purposes of the case sub judice,’” which made it “appropriate [for the Stuber Court] to adopt the ‘economic reality test’ used by federal courts under the FLSA to determine whether a worker should be classified as an employee or an independent contractor under the PMWA.” Id. at 199. Thus, Amazon more narrowly suggests that the respective FLSA and PMWA “hours worked” statutory provisions at issue in that case did not have the same “identity of purpose” as

the “employee” provisions at issue in Stuber. Id. And because Plaintiffs cannot direct this Court to any additional case law that might further cast doubt on Stuber, this Court cannot find for Plaintiffs simply because they “expect that, when the question is presented,” the Pennsylvania Supreme Court will now “adopt a more lenient test for establishing independent contractor misclassification than federal law.” ECF 151 at 2.6 b. Burden

The PMWA’s burden inquiry requires a slightly more nuanced analysis. To begin with, as Defendants rightly note, the Third Circuit’s decision here did not indicate distinct federal and state law burdens.7 The Court of Appeals held only that (1) “[t]he minimum wage and overtime wage

6 To the extent Plaintiffs invoke Lowman v. Unemployment Comp. Bd. of Rev., 661 Pa. 29 (2020) in reference to the PMWA, Lowman was (1) decided before the Third Circuit’s amended decision in Razak and (2) pertains to an entirely differently statute, failing to even mention Stuber. Indeed, Lowman involved a claim under Pennsylvania’s distinct Unemployment Compensation Law (“UCL”) and was decided on statutory-text grounds alone. It is therefore inapposite to the PWMA. Somewhat incredibly, the Lowman Court itself even noted that “Razak is not helpful in our analysis.” Id. at 55 n.17. The inverse remains true here. This Court discusses Lowman in more detail below, in relation to the WPCL. 7 Likewise, Defendants direct this Court to numerous instances where Pennsylvania courts have stated that PMWA “Plaintiffs must prove . . . the plaintiff was an employee.” Beauregard v. Broadway Elec. Serv. Corp., 2022 WL 2293969, at *5 (W.D. Pa. June 24, 2022) (emphasis added); see also, Gitzen v. S&S, Inc., 2021 WL 6286382, at *3 (W.D. Pa. Nov. 23, 2021) (noting that “a plaintiff must show that . . . the plaintiff was an employee as defined by the act”) (emphasis added); Williams v. Sweet Home Healthcare, LLC, 325 F.R.D. 113, 116 (E.D. Pa. 2018) (“Plaintiffs and all class members must show that they were employees rather than independent contractors”) (emphasis added). provisions at issue all require that Plaintiffs prove that they are ‘employees,’” and (2) “Pennsylvania state courts have looked to federal law regarding the FLSA for guidance in applying the PMWA.” Razak, 951 F.3d at 142 (emphasis added). Yet, as Plaintiffs astutely counter, in Stuber, the Commonwealth Court noted that “both

sides agree[d] that . . . there is a presumption that the individual is an employee, (a presumption the employer must rebut).” Stuber, 822 A.2d at 871. The Stuber Court also mentioned, in a footnote, that “under both the Act and our state unemployment law,” there is a “statutory presumption of an employment relationship that the employer must overcome.” Id. at 873 n.5. But these passing references do not ultimately afford Plaintiffs their desired presumption. Indeed, the counterfactual context of this footnote is critical. In full, this footnote reads: Even applying other law, the outcome would likely not change. For example, under Pennsylvania unemployment law, the inquiries are whether the claimant was free of the employer’s control, i.e., does the employer direct the job and manner of performance, and whether the claimant was engaged in an independently established trade, i.e., did the claimant have a proprietary interest in the business or was he free from control by the employer. Sharp Equipment Co. v. Unemployment Compensation Board of Review, 808 A.2d 1019 (Pa.Cmwlth.2002). The inquiry, additionally, involves determining whether the claimant could perform the work for anyone, or whether the nature of the services is such that they could only be performed for the employer. Moreover, under both the Act and our state unemployment law, there is a statutory presumption of an employment relationship that the employer must overcome. Id. When these principles are applied to the facts in the case sub judice, a different outcome from the one we reach here is far from assured.

Id. (emphasis and bold added). It therefore amounts to nothing more than dicta regarding a hypothetical scenario about “applying other law;” namely the dissimilar and inapplicable Unemployment Compensation Law (“UCL”), 43 P.S. § 753. Indeed, Sharp Equipment Co., the lone case to which the Superior Court cites here, involves only a claim under the UCL. And unlike the PMWA, the text of the UCL expressly includes such a presumption. id.; see also 43 P.S. § 753 (“Services performed by an individual for wages shall be deemed to be employment subject to this act, unless and until it is shown to . . .”)

(emphasis added). Thus, even if the Stuber Court had been suggesting “the Act” to which the Court referred was the PMWA, rather than the UCL, and that Act somehow created a “statutory presumption,” id. at 873 n.5 (emphasis added), this position would simply be incorrect. The Stuber Court cites no relevant authority for this proposition. Nor do Plaintiffs here, as their voluminous pre- and post-trial briefing similarly direct this Court to other Pennsylvania statutes—namely the UCL—where the statutes’ texts expressly presume employment status. See, e.g., ECF 151 at 4 (citing to Lowman v. Unemployment Comp. Bd. of Rev., 661 Pa. 29, 65 (Pa. 2020); 43 P.S. §§ 753(1)(2)(B)). Simply put, had the Pennsylvania “legislature wanted to include” a presumption within the PMWA, “it certainly knew how to do so.” Clearwater Constr., Inc. v. Northampton Cnty. Gen.

Purpose Auth., 166 A.3d 513, 521 (Pa. Commw. Ct. 2017); see also, e.g., Kirtz v. Trans Union LLC, 46 F.4th 159, 165 (3d Cir. 2022), aff’d sub nom., 601 U.S. 42 (2024) (“where Congress wanted to use a different or narrower definition [within a statute], it knew how to do so”). Indeed, the UCL itself is excellent evidence of the legislature’s capacity to craft such a presumption. Yet, the PMWA does not contain one. So, this Court will not create one now simply because “Plaintiffs expect that, when the question is presented, the Pennsylvania Supreme Court will adopt a more lenient test for establishing independent contractor misclassification than federal law.” ECF 151 at 2. Reinforcing this conclusion, this Court notes that the Stuber Court expressly warned against overreliance on the UCL when interpreting the PMWA. In relevant part, Stuber explained that “[w]hile such [Pennsylvania unemployment and tax] laws should not be entirely discounted, we must remain cognizant that they were not enacted for precisely the same purpose as the

Minimum Wage Act.” Stuber, 822 A.2d at 872. It would therefore be strange indeed for the Stuber Court to have explicitly discounted the UCL’s effect, and then simultaneously and sub silencio imported the UCL’s textual presumption into the PMWA. Likewise, this Court notes that—in the twenty years since Stuber was decided—no subsequent court has even mentioned the existence of a presumption, let alone definitely established one. Had Stuber so clearly created the presumption that Plaintiffs now contend it does, one would think this issue would have manifested in the interim decades. Such silence further reaffirms the Court’s decision at trial to reject Plaintiffs’ requested jury instruction.8

8 Moreover, even if this Court were to find that some type of presumption existed here, Plaintiffs misconstrue the implications of any such presumption. While Plaintiffs initially requested a jury instruction that “place[d] the burden on Defendants to establish that the Plaintiffs were independent contractors,” ECF 151 at 3 (emphasis added), Plaintiffs then more narrowly asserted that the PMWA raised a “presumption that an individual is an employee,” id. at 11 (emphasis added); see also ECF 212 at 1 (“the burden is on the employer to rebut the presumption that workers are employees”). Yet, the distinction between burdens and presumptions is important. As the Federal Rules of Evidence make clear, although “a presumption imposes on the party against whom it is directed the burden of going forward with evidence to rebut or meet the presumption,” it simply “does not shift to such party the burden of proof in the sense of the risk of nonpersuasion, which remains throughout the trial upon the party on whom it was originally cast.” McCann v. Newman Irrevocable Tr., 458 F.3d 281, 287 n.4 (3d Cir. 2006) (citing to Fed R. Evid. 301) (emphasis added). Indeed, “once the burden of production to rebut a presumed fact is met, the presumption evaporates from the case.” Ricco v. Goston, 2020 WL 3642863, at *3 (E.D. Pa. July 6, 2020). Otherwise stated, “at no time does the burden of persuasion shift.” Id. As such, to the extent such a presumption existed here, Defendants needed only to “rebut [this] presumed fact.” Ricco, 2020 WL 3642863, at *3. And, as our deadlocked jury demonstrates, the record is replete with evidence that rebutted any employer presumption here. Thus, even if this Court did err here, any such error would have been harmless because “there is a high probability that the iii. WPCL

Finally, this Court turns to the WPCL. Here, the simplest answer is that the Third Circuit’s decision in Williams v. Jani-King of Phila., Inc. expressly “predict[ed] that the Pennsylvania Supreme Court would employ [the ten-factor test this Court charged the jury on here] in the context of the WPCL.” 837 F.3d 314, 321 (3d Cir. 2016). Nonetheless, Plaintiffs assert that (1) even if this Court relies on Jani-King’s multi-factor test, the Third Circuit left open a question regarding relative burdens or presumptions, and (2) intervening Pennsylvania case law has abrogated Jani-King. For the following reasons, however, both contentions are unavailing. a. Burden

As with the PMWA, Plaintiffs argue that this Court must import the UCL’s textual presumption into the WPCL. Again, the Court cannot agree. To be sure, “[i]n interpreting the meaning of employee under the WPCL, Pennsylvania courts have looked to similar statutes such as the Pennsylvania Unemployment Compensation Act and the Pennsylvania Workers’ Compensation Act.” Jani-King, 837 F.3d at 320. But Plaintiffs’ suggestion that this Court is compelled to import the entirety of the UCL into the WPCL is belied by both precedent and logic. Indeed, Plaintiffs simply ignore that we may look to both the UCL and the WCA. And the WCA, unlike the UCL, lacks any sort of express textual presumption. To the contrary, the Pennsylvania Supreme Court has held that, under the WCA, “it is a claimant’s burden to establish an employer/employee relationship in order to receive benefits.” Universal Am-Can,

[erroneous instruction] did not affect the outcome of the case.” O’Brien v. Middle E. F., 57 F.4th 110, 121-22 (3d Cir. 2023) (citations omitted). Ltd. v. W.C.A.B. (Minteer), 563 Pa. 480, 485 (2000) (emphasis added). The Supreme Court has also actively distinguished the WCA from the UCL on the very grounds that the latter presumes employment status. In re Perrone, 587 Pa. 388, 406 n.2 (2006) (Eakin, J.) (dissenting on other grounds).

Critical here, the WPCL—like the PMWA and the WCA—lacks any sort of express statutory presumption.9 Yet, Plaintiffs now ask this Court to conjure one out of thin air. That would, again, require this Court to conclude that the Pennsylvania Supreme Court has sub silencio endorsed the proposition that the WPCL somehow engrafts the UCL’s text onto itself, despite (1) the Supreme Court reaching precisely the opposite conclusion for the WCA, and (2) having expressly disavowing a connection between Razak and the UCL in the “self-employment” context, as discussed in more below. So here too, Plaintiffs fail to appreciate “the legislature . . . knew how to” create a presumption had it so desired, Clearwater Constr., Inc., 166 A.3d at 521; Kirtz, 46 F.4th at 165, but chose not to do so. b. Substantiative Test

That is not to say, however, we cannot harmonize Jani-King’s directive to consider Pennsylvania courts’ interpretations of both the UCL and the WCA when analyzing the WPCL. Indeed, once we cast aside Plaintiffs’ convoluted argument regarding presumptions, it becomes a simple exercise. First, as noted, Jani-King—following numerous Commonwealth Court opinions— expressly directs this Court to adopt the Universal Am-Can multi-factor test for the WPCL. Jani- King, 837 F.3d 320-21. That is a straightforward, one-to-one comparison. Courts may look to the

9 Indeed, the WPCL does not define “employee” at all. Carpenter v. Pepperidge Farm, Inc., 2024 WL 2103257, at *1 (3d Cir. May 10, 2024) (non-precedential). WCA when interpreting the WPCL, and the Pennsylvania Supreme Court has long held the WCA uses the ten-factor test on which this Court charged the jury here. Universal Am-Can, Ltd., 563 Pa. at 489-90. Second, as Plaintiffs seemingly fail to appreciate, Pennsylvania courts have also routinely

folded these factors into the text-based test for overcoming the UCL’s employee presumption. Specifically, numerous Pennsylvania courts have explicitly linked the Jani- King/Universal Am-Can factors to section 753’s call to consider whether a worker “has been and will continue to be free from control or direction over the performance of such services both under his contract of service and in fact.” See, e.g., May I Help You, Inc. v. Unemployment Comp. Bd. of Rev., 2010 WL 9513597, at *3-4 (Pa. Commw. Ct. Mar. 16, 2010); Thompson v. Unemployment Comp. Bd. of Rev., 2010 WL 9514438, at *4 n.5 (Pa. Commw. Ct. Apr. 13, 2010); Kurbatov v. Dep’t of Lab. & Indus., Off. of Unemployment Comp., Tax Servs., 29 A.3d 66, 70 (Pa. Commw. Ct. 2011); Dull v. Dep’t of Lab. & Indus., 2010 WL 9512524, at *3 (Pa. Commw. Ct. Mar. 5, 2010); York Newspaper Company v. Unemployment Compensation Board of Review,

160 Pa.Cmwlth. 475, 480-81 (Pa. Commw. Ct. Dec. 14, 1993); see also Stauffer v. Unemployment Comp. Bd. of Rev., 74 A.3d 398, 404-05 (Pa. Commw. Ct. 2013) (applying a similar totality test); Quality Care Options v. Unemployment Comp. Bd. of Review, 57 A.3d 655, 659-60 (Pa. Commw. 2012) (same); CE Credits OnLine v. Unemployment Comp. Bd. of Review, 946 A.2d 1162, 1168- 69 (Pa. Commw. 2008) (same); Beacon Flag Car Co. v. UCBR, 910 A.2d 103, 108 (Pa. Commw. 2006) (similar); Lowman v. Unemployment Comp. Bd. of Rev., 661 Pa. 29, 66-67 (2020) (“In assessing the control factor, the Commonwealth Court has identified additional indicia of control”); id. at 69-70 (“all relevant factors presented in a given case should be considered when determining whether a business is independently established.”) As such, a far more reasonable reading of those Pennsylvania cases directing courts to interpret the WPCL by way of the UCL and WCA is that a court may draw on the relevant factors common to both statutes. That avoids a Hobson’s choice regarding presumptions and standards, given that the two relevant statutes are at odds.10

c. Lowman

Plaintiffs’ entire argument hinges on the proposition that the Pennsylvania Supreme Court’s recent decision in Lowman v. Unemployment Comp. Bd. of Rev., 661 Pa. 29, 35 (2020) applies directly to the WPCL, thus abrogating the above analysis. In Lowman, however, the Pennsylvania Supreme Court addressed a relatively narrow question: “the appropriate test to determine whether a claimant who is otherwise entitled to receive unemployment compensation benefits due to a separation from employment becomes ineligible for those benefits as a result of being self-employed pursuant to Section 402(h) of the Unemployment Compensation Law (the “Act”),43 P.S. §§ 751–919.10, the self-employment exclusion.” 661 Pa. 29, 35 (2020) (emphasis added). As noted above, the Lowman Court concluded that 43 P.S. § 753(l)(2)(B)—i.e., the text of the UCL itself—“contains the appropriate test for determining whether or not an individual is in self-employment.” Id. That text, again, expressly includes a presumption, and sets forth a form of the “a-c” test for which Plaintiffs have advocated:

10 Moreover, to the extent this Court were required to make such a choice, the WCA is the natural option. See, e.g., Stuber, 822 A.2d at 872–73 (“[w]hile such other [unemployment compensation] laws should not be entirely discounted, we must remain cognizant that they were not enacted for precisely the same purpose as the Minimum Wage Act.”); Lowman, 661 Pa at 55 n.17 (noting that, unlike Lowman, the current case involves “consider[ing] independent contractors for purposes of assessing federal and state wages”) (emphasis added). Services performed by an individual for wages shall be deemed to be employment subject to this act, unless and until it is shown to the satisfaction of the department that (a) such individual has been and will continue to be free from control or direction over the performance of such services both under his contract of service and in fact; and

(b) as to such services such individual is customarily engaged in an independently established trade, occupation, profession or business.

Id. at 38 (citing to 43 P.S. § 753(l)(2)(B)) (emphasis added).

Apart from the obvious distinction that the WPCL does not include this textual test, a few other points bears mentioning. First, and somewhat incredibly, the Lowman Court expressly disclaimed any connection between its decision and this Court’s prior decision in this very case: Uber also compares Lowman’s entrepreneurialism to drivers in Razak v. Uber Techs., Inc., 2018 WL 1744467 (E.D. Pa. 2018), who were free from control and “formed their own companies, [ ] purchased numerous vehicles, hired helpers, advertised, and provided transportation services for competitors and the public at large,” and were considered independent contractors for purposes of assessing federal and state wages. We note that the Internal Revenue Service uses the “common law” test to determine if an employer is required to withhold income, Social Security, and Medicare taxes on behalf of a worker. That test includes twenty factors, grouped into three categories related to the degree of control exercised by the employer: behavioral control, financial control, and the relationship of the parties. Catherine Tucciarello, The Square Peg Between Two Round Holes: Why California's Traditional Right to Control Test Is Not Relevant for On-Demand Workers, 13 Seton Hall Circuit Rev. 351, 355 (2017). Given the substantial dissimilarities between the common law test and the two factor analysis set forth in § 753(l)(2)(B), Razak is not helpful in our analysis. Id. at 55 n.17 (emphasis added). On that basis alone, it is entirely unpersuasive for Plaintiffs to now suggest Lowman maps onto the WPCL (or the PMWA), or that the Pennsylvania Supreme Court would so find in the future. Second, the Lowman Court repeatedly made clear that its decision was limited to the term

“self-employment,” “as that term is used in Section 802(h).” Id. at 60, 62, 67 n.27, 79 (emphasis and bold added). Indeed, the Lowman Court was even careful enough to “express no opinion” on the further question whether “the use of a ‘positive steps’ analysis [was] part of the test for self- employment embodied in Section 753(l)(2)(B) where the personal services are performed by an individual in a stand-alone context,” Id. at 62 n.24; a much finer distinction than what this Court confronts here. In other words, the Lowman Court addressed a very specific scenario for this singular statutory provision. Given that narrowness, it again strains credulity that this decision— about a particular provision of a distinct statute—now governs the WPCL, especially without any explicit or implicit suggestion from the Pennsylvania Supreme Court to that effect.11 Third, and critical here, the Lowman Court repeatedly cited positively to Universal Am-

Can, the very case that sets forth the relevant ten factors that guide a WCA analysis. Id. at 48, 78 n.32. This directly contradicts the notion that the Lowman Court somehow rendered Universal Am-Can no longer good law as applied to the WCA (and thus the WPCL). Had the Lowman Court concluded as much, presumably it would have made this clear.

11 Moreover, on the facts, the Lowman Court made clear that its decision was an incredibly narrow one. The Majority expressly noted that “the Dissent's contention that results may vary depending upon the ‘non-Uber activities’ of different drivers does not constitute a criticism of the analysis set forth herein, but rather merely a recognition that individual decisions must be made in specific cases based upon the unique facts presented in each circumstance.” Id. at 78 n.32 (emphasis added). On that carveout alone, it is actively misleading for Plaintiffs to suggest, as they have pre- , during-, and post-trial—that Lowman fully decides this case. So even if Lowman’s test applied, which it does not, judgment as a matter of law would be inappropriate. Thus, while Lowman is the most recent decision from the Pennsylvania Supreme Court, in this Court’s view, it simply does not control the burdens or relevant factors that guide a misclassification analysis under the WPCL. Indeed, just last month, the Third Circuit reaffirmed the use of the Jani-King test for WPCL claims in Carpenter v. Pepperidge Farm, Inc. 2024 WL

2103257, at *1-2 (3d Cir. May 10, 2024). This latest endorsement of Jani-King, while non- precedential, further bolsters this Court’s confidence in its decision to instruct the jury on those factors here. B. Neither Party is Entitled to Judgment as a Matter of Law Here And once this Court sets aside Plaintiffs’ arguments regarding alternative state-law misclassification tests, it becomes clear this Court may not grant either side’s Rule 50 motion. In this Court’s humble view, the trial record here is replete with a definite “hybrid” of facts that cut in both directions for most, if not all, of the economic reality and WPCL factors. This fissure, in turn, precludes a finding that either Party’s case at trial was so “critically deficient of that minimum quantity of evidence from which a jury might reasonably afford relief.” Trabal, 269

F.3d at 249. As explained above, this deadlock flows largely from the intractable nature of the gig- economy itself. Where, as here, factfinders must balance (1) the worker flexibility inherent to app- based ridesharing platforms, and (2) those platform’s attempts at quality control and standardization, a diverse set of reasonable inferences and conclusions abound. i. Right to Control

On Uber’s “right to control,” for instance, Defendants skillfully argued at trial that Plaintiffs had the “freedom to decide whether, when, and how to use” the Uber app. Trial Tr. at 65, Mar. 4, 2024, ECF 272. Indeed, Defendants elicited significant testimony and documentary evidence to that effect. See ECF 282-1 at 4-5 (summarizing evidence)); but see Trial Tr. at 227, Mar. 6, 2024 (“if you’re on the online more than twelve hours, it will go offline”). Yet, even if the Court accepts Defendants’ framing of the record here as undisputed and uncontradicted—which the Court does not—Plaintiffs compellingly counter that ample evidence

at trial indicated that, despite Plaintiffs’ perceived scheduling and operational freedoms, Uber had the right to exercise significant control over Plaintiffs once they actually decided to use the Uber app. In other words, although Plaintiffs may have had significant freedom to choose when and where to work, Uber had the right to control numerous aspects of the working environment. For example, the record here reflects that Uber not only implemented a rating system for drivers, but also heavily monitored those ratings and consistently warned drivers with low ratings about the potential for deactivation. See., e.g., PX-101 (internal “Uber Philly Quality Playbook” encouraging Uber personnel to “warn/deactivate bad drivers often (daily, at-least weekly)”). Indeed, the following trial testimony from Plaintiff Sabani is instructive: Q. And so, Mr. Sabani -- Kenan, do you remember getting this email where Uber told you that they were increasing the Uber rating for you from 4.6 to 4.7? A. Yes. Q. And so, here it says what this means for you that if you fall -- and just paraphrasing -- if you fall below 4.7, what might happen? A. Deactivation. Q. Right. And was that a big deal for you? A. It was. Q. And so, when you receive an email telling you that you've got to at least maintain a 4.7, how did that impact the way that you drove? A. I was extra careful.

Trial. Tr. at 9-10, Mar. 6, 2024, ECF 274; see also PX-145 (January 16, 2016 message from partnersphilly@uber.com to Kenan Sabani stating that “[p]artners below 4.7 will be warned about their ratings, and give the chance to improve over their next 50 trips,” but that “[p]artners who do not sustain a 4.7 over those 50 trips will no longer be able to offer UberBLACK rides on the Uber platform.”); (internal presentation listing potential “infractions”). Likewise, at trial, Defendants repeatedly emphasized the fact Uber expressly disclaimed— via contract—the right to control Plaintiffs. JX-11 at § 2.4 (noting that “Uber does not, and shall

not be deemed to, direct or control ... Drivers ...”). While that may be true, other portions of the Parties’ various agreements could lead a reasonable factfinder to different conclusions. For instance, in a June 2014 “Driver Addendum Related to Uber Services,” Uber “reserve[d] the right to immediately deactivate Subcontractor’s access to the Software and Service” for various reasons, including, but not limited to, (1) a “subcontractor’s refusal to fully complete a trip after acceptance of a trip request . . . without waiver by the User or Uber,” or (2) any “intentional misrepresentation by [subcontractor] to a User or Uber, including intentionally taking an indirect route to User’s specified destination.” JX-005; see also JX-004. Such provisions belie the notion that Plaintiffs had complete autonomy—in contract or in practice—when it came to route selection or trip cancellations. To be sure, these restrictions do not immediately establish an employer-employee

relationship. Nor do they automatically tip the “right to control” factor in Plaintiffs’ favor. Nonetheless, they are undoubtedly relevant considerations from which a factfinder might draw a range of inferences that are favorable to Plaintiffs. The same sort of balancing act applies to the broader “Technology Services Agreement.” Defendants again highlight that—in signing this contract—Plaintiffs expressly acknowledged they were “independent contracting parties.” JX-011 at ¶¶ 2.4, 13.1. Again, Defendants are correct this contractual reality is a permissible consideration, as it potentially reflects the Parties’ own understanding of the nature of their working relationship. Jani-King, 837 F.3d at 323 (“The provisions of an agreement may be evidence of what the actual practice or working relationship is.”); Holder v. Bingnear, 1991 WL 7381, at *4 (E.D. Pa. Jan. 18, 1991) (“While the language of the agreement between the parties is not totally decisive on the issue . . . it is, however, a factor to be considered”) (citations omitted). Yet, numerous courts, along with the Department of Labor, have discounted the significance of such conclusory contractual labels. Jani-King, 837 F.3d at 323

(further noting that “labels” in an independent contractor agreement “are not determinative”); Safarian v. American DG Energy Inc., 622 F. App’x 149, 152 (3d Cir. 2015) (non-precedential) (citations omitted) (“This [FLSA] inquiry is not governed by the ‘label’ put on the relationship by the parties or the contract controlling that relationship, but rather focuses on whether ‘the work done, in its essence, follows the usual path of an employee.’”); 29 C.F.R. § 795.105 (“Labeling employees as ‘independent contractors’ does not make these protections inapplicable.”). Similarly, although Defendants may be correct that a company’s compliance with applicable laws and regulations does not reflect “control” for FLSA purposes, Plaintiffs direct this Court to a number of additional rules and restrictions imposed by Uber that went beyond what the

law required. For instance, certain evidence indicates that (1) Uber’s car specifications were more restrictive than those of local regulators, Trial Tr. at 142, Mar. 5, 2024, ECF 273; (2) Uber required Plaintiffs to pass two background checks—one for the local regulators and one for Uber, JX-111 (explaining that Uber’s “background check is more thorough, and is a requirement to remain an active driver with Uber.”); and (3) Uber warned drivers of long lists of committable “infractions” that were not necessarily violations of applicable law or regulations, PX-224; PX-248; PX-227 (deactivating a driver for, among things, “bad attitude”); JX-253-009 (presentation containing infractions); PX-304 (noting that “[e]ach city has a maximum cancellation rate, based on the average cancellation rate of drivers in that area.”). In short, the record here—as highlighted be this non-exhaustive set of examples—contains a veritable jumble of contested, disputed facts, the amalgamation of which permit numerous reasonable inferences regarding the extent of Uber’s right to control Plaintiffs. ii. Remaining Factors

And while the “highly relevant” nature of the “right to control” factor may, by itself, preclude Rule 50 relief for either party here, Razak, 951 F.3d at 145, the remainder of the record reinforces this pattern of contrasting evidence for each economic reality and WPCL factor. On “opportunity for profit or loss,” for instance, the jury was forced to balance evidence that Plaintiffs had at least some ability—and in fact did—work for Uber’s competitors, D-426, D- 442, against evidence indicating that Uber occasionally attempted to frustrate those efforts. See, e.g., PX-302 (prohibiting “activities conducted outside of the monitored system of the Uber app— like anonymous pickups,”). Perhaps more importantly, that balancing act also extended to intra- app strategic opportunities. While Defendants’ stressed at trial that Plaintiffs could avoid certain pickup areas as desired (such as sporting events), Trial Tr. at 73, Mar. 4, 2024, ECF 272, Plaintiffs

countered that Uber intentionally deprived drivers of information before trip acceptance (such as rider destination), Trial Tr. at 268, Mar. 5, 2024, ECF 273, potentially blunting Plaintiffs’ strategic maneuvering capabilities. Similarly, on “degree of permanence,” the trial record—much like record at summary judgment—reflected that “[o]n one hand, Uber [could] take drivers offline, and on the other hand, Plaintiffs [could] drive whenever they choose to turn on the Driver App, with no minimum amount of driving time required.” Razak, 951 F.3d at 147. Each strand of evidence created a series of reasonable and conflicting inferences for the jury to consider. Likewise, even for those factors this Court and the Third Circuit may have suggested tip in a specific direction at summary judgment—such as the lack of a “special skill” or the “integrality” factors—the voluminous record here is sufficiently mixed that this Court is deeply skeptical that it may “substitut[e] its version of the facts for that of the jury.” Parkway Garage, Inc., 5 F.3d at

691-92. Particularly as to “integrality,” the Third Circuit here expressly concluded this factor would further turn on whether Plaintiffs could prove that Uber was “transportation company that employs drivers,” rather than was “a technology company that supports drivers’ transportation businesses.” Razak, 951 at 147 n.12. What’s more, the Court of Appeals explicitly deemed this issue “a disputed material fact that should be resolved by a fact-finder.” Id. Yet, after vigorous arguments and a mountain of evidence from each Party,12 the jury seemingly could not resolve this dispute. ECF 269. Accordingly, this Court is persuaded that the convoluted and disputed set of “historical facts” here, Brock v. Mr. W Fireworks, Inc., 814 F.2d 1042, 1044 (5th Cir. 1987), along with the extensive universe of possible “reasonable inferences,” Martin v. Selker Bros., 949 F.2d 1286,

1293 (3d Cir. 1991), could support a reasonable jury finding for either Party here, thus precluding Rule 50 relief for both. C. Invited Error Doctrine Finally, and somewhat tangentially, the Court notes that Defendants have separately argued this Court may rely on the doctrine of “invited error” to enter judgment based on the jury’s non- unanimous polling sheet, the answers to which tended to favor Defendants. ECF 279 at 12. This

12 Specifically, Plaintiffs highlighted the numerous communications from Uber to its drivers that stressed the drivers’ importance to the company, PX-216, whereas Defendants directed the jury to various testimony, marketing materials, contracts, and internal documents that characterize Uber as a mere technology platform for connecting riders and drivers, Trial Tr. at 16, Mar. 7, 2024, ECF 275. doctrine dictates that “when a litigant takes an unequivocal position at trial, he cannot on appeal assume a contrary position simply because the decision in retrospect was a tactical mistake, or perhaps a candid but regretted concession encouraged,” Lesende v. Borrero, 752 F.3d 324, 337 (3d Cir. 2014) (citations omitted).

For two reasons, that argument is unavailing. First, nothing about Plaintiffs’ eventual consent to this Court’s use of the poll was unequivocal. True, at various points, Plaintiffs endorsed the possibility of an advisory jury, including on the final day of trial. Trial. Tr. at 5, Mar. 11, 2024, ECF 277. Nonetheless, the trial record is replete with ongoing and fluid discussions regarding the jury verdict ratio that each Party would be willing to accept. See e.g., id. at 5, 22. Plaintiffs repeatedly stated they would accept either a 5-3 or 6-2 verdict, id. at 5, whereas Defendants remained steadfast at 7-1, id. at 17. Those discussions, at a minimum, created sufficient ambiguity as to Plaintiffs’ continued willingness to rely on an advisory jury, thus precluding the invited error doctrine. Second, even if this Court were to ignore that ambiguity, Defendants have repeatedly

asserted they are entitled to a jury trial here under the Seventh Amendment. ECF 189. The Court agrees with that position, and it has never officially held otherwise. Indeed, in administering the poll at trial, the Court made sure to cabin its legal significance, inviting the parties to brief the issue post-trial. Trial Tr. at 9, Mar. 11, 2024, ECF 277 (“This is not a verdict. This is a questionnaire under Rule 49. Now, what I would do with this would be subject to briefing and argument.”). Thus, the Court is convinced it has not yet made the error allegedly invited by Plaintiffs. The Court will not do so now because—as Defendant have long argued—the Court would be walking directly into a Seventh Amendment violation. V. CONCLUSION AND ORDER

For the foregoing reasons, both Parties’ motions for post-trial relief are DENIED. As a result of the research in the foregoing pages of this memorandum, it is clear to this judge that the “gig economy” is really a “hybrid.” There are aspects of both employer/employee relationships and also independent contractor relationships, using traditional common law standards. However, the FLSA, and the appellate precedents, require an “either/or” decision, and district judges do not have power to ignore the statutory language and appellate precedents. Thus, Congress and/or state legislatures should consider an alternative jurisprudence allowing an adjudication in hybrid cases where a fact finder can allow for equitable damages and/or an allocation of damages that are not available under the FLSA and related state laws at this time. The Court will convene a second jury trial as of June 10, 2024. At trial, the Court will instruct the jury in accordance with this memorandum. For Plaintiffs’ FLSA and PMWA claims, the Court will charge the jury on the six economic reality factors, as set forth by the Third Circuit in Donovan. For Plaintiffs’ WPCL claims, the Court will charge the jury on the ten factors

highlighted by the Third Circuit in Jani-King. For each of these claims, the burden is on Plaintiffs to prove, by a preponderance of the evidence, that they were employees, rather than independent contractors. While the Court is weary of this “either-or” resolution in view of the practical realities of the gig economy, it is what the law requires. An appropriate order follows. BY THE COURT:

s/ Michael M. Baylson _______________________________ MICHAEL M. BAYLSON United States District Court Judge

O:\CIVIL 16\16-573 Razak v Uber Technologies\16cv573 memorandum 6.4.24.docx