IN RE: UBER TECHNOLOGIES (2024)

SUPPLEMENTAL ORDER

Before the Panel: The Ninth Circuit has invited the Judicial Panel on Multidistrict Litigation (the Panel) to respond to Uber's (Uber Technologies, Inc.; Rasier, LLC; and Rasier-CA, LLC) petition seeking a writ of mandamus from our order centralizing this litigation in the Northern District of California. See Order, Uber Techs., Inc. v. United States Judicial Panel on Multidistrict Litig., No. 23-3445 (9th Cir. Dec. 14, 2023), ECF No. 10. Our initial transfer order fully sets forth the basis for our decision to centralize the actions in this docket. We offer the following supplemental observations to aid the Ninth Circuit in its review of this matter.

I.

We first provide a brief description of the Panel's statutory mandate and operation, as this background may be helpful when considering the present litigation and defendants’ arguments against transfer. The Panel was created by the passage of 28 U.S.C. § 1407 in 1968. Section 1407 gives the Panel a broad charge: it “may” transfer civil actions pending in two or more districts to a single district for coordinated or consolidated pretrial proceedings ¹ if the actions involve “one or more common questions of fact.” 28 U.S.C. § 1407(a). Such transfer can only be made if the Panel determines that it would serve “the convenience of parties and witnesses” and that it will “promote the just and efficient conduct of such actions.” Id.

No party is entitled to centralization as a matter of right. We determine whether centralization is appropriate on a case-by-case basis considering the statutory criteria. See, e.g., In re Bear Creek Techs., Inc., (’722) Patent Litig., 858 F. Supp. 2d 1375, 1379 (J.P.M.L. 2012) (“Centralization of any litigation ․ is not automatic, and will necessarily depend on the facts, parties, procedural history and other circumstances in a given litigation.”) (citing In re CVS Caremark Corp. Wage & Hour Emp't Practices Litig., 684 F. Supp. 2d 1377, 1379 (J.P.M.L. 2010) (“[W]e do not ‘rubber stamp’ in any docket ․”)). See also In re Equinox Fitness Wage & Hour Emp't Practices Litig., 764 F. Supp. 2d 1347, 1348 (J.P.M.L. 2011) (denying unopposed motion for centralization because “the Panel has an institutional responsibility that goes beyond simply accommodating the particular wishes of the parties”). Indeed, in exercising the wide discretion granted to it by Section 1407,² the Panel denies a large portion of the motions for centralization it considers. For instance, in 2022, we denied approximately one-third of motions for centralization that we considered. In 2019, that number was almost half. See U.S. Jud. Panel on Multidistrict Litig., 2022 Calendar Year Statistics at Slide 8 (Motions for Centralization Granted and Denied), available at https://www.jpml.uscourts.gov/sites/jpml/files/JPML_Calendar_Year_Statistics-2022.pdf.

Granting a motion for centralization is not an indication of our views about the underlying merits of the claims at issue in transferred cases. See, e.g., In re Fluoroquinolone Prods. Liab. Litig., 122 F. Supp. 3d 1378, 1380 (J.P.M.L. 2015) (“The Panel is not authorized to engage in an assessment of the merits of the actions.”); In re Kauffman Mut. Fund Actions, 337 F. Supp. 1337, 1339–40 (J.P.M.L. 1972) (“The framers of Section 1407 did not contemplate that the Panel would decide the merits of the actions before it and neither the statute nor the implementing Rules of the Panel are drafted to allow for such determinations.”). We place the ultimate progress of pretrial proceedings—including overseeing discovery, deciding jurisdictional motions (e.g., motions to remand to state court), and determining the validity of claims—in the hands of the transferee judge. Because MDLs often involve some case-specific discovery and pretrial proceedings arising from the unique circumstances and claims in individual constituent cases, transferee judges frequently are called upon to design “a pretrial program that will ensure that the needs of each party for any unique discovery or individual judicial attention can be accommodated concurrently with the conduct of common pretrial matters.” In re Transocean Tender Offer Sec. Litig., 415 F. Supp. 382, 384 (J.P.M.L. 1976). See also In re FedEx Ground Package Sys., Inc., Emp't Practices Litig. (No. II), MDL No. 1700, 2011 U.S. Dist. LEXIS 107272, at *3 (J.P.M.L. Sept. 22, 2011) (“[I]n most instances the transferee judge has an acute sense about the procedural steps necessary to advance the litigation in the fairest and most efficient way.”); Manual for Complex Litigation (Fourth) § 22.8 (2004) (discussing various methods of organizing discovery in mass tort cases, including MDLs).

Similarly, even if a common set of facts among a group of cases might yield different legal outcomes under the law of different states, the facts relevant to the varying legal questions posed often overlap substantially across many jurisdictions. Consequently, “it is ‘within the very nature of coordinated or consolidated pretrial proceedings in multidistrict litigation for the transferee judge to be called upon to apply the law of more than one state.’ ” In re CVS Caremark, 684 F. Supp. 2d at 1378 (quoting In re Air Crash Disaster at John F. Kennedy Int'l Airport on Jun. 24, 1975, 407 F. Supp. 244 (J.P.M.L. 1976)). See also MANUAL FOR COMPLEX LITIGATION (FOURTH) § 22.634 (“Multiple tort cases frequently involve claims and defenses asserted under various federal and state laws.”). Moreover, if the transferee judge determines that coordinated or consolidated pretrial litigation of any action or group of actions no longer is beneficial, he or she may suggest to the Panel that we remand those actions to their transferor courts. See Panel Rules 10.1–10.2.

II.

Turning to the present litigation, the Panel weighed the parties’ extensive arguments, made in briefing and at oral argument, regarding whether centralization of the actions against Uber was appropriate. We explicitly found that Section 1407 was satisfied because (1) the cases “involve[d] common questions of fact,” and (2) transfer would “serve the convenience of the parties and witnesses and promote the just and efficient conduct of” the litigation. In re Uber Techs., Inc., Passenger Sexual Assault Litig., ––– F. Supp. 3d ––––, 2023 WL 6456588, at *1 (J.P.M.L. Oct. 4, 2023).³ We concluded that, on balance, centralization of these cases for common pretrial proceedings was warranted because it would provide important efficiencies for the parties, witnesses, and the judiciary. See id.

As with any collection of cases brought by different plaintiffs, we recognized that these actions involve individualized factual issues arising out of the unique circumstances of each plaintiff's allegations and claims. See id. Even so, almost 80 actions were pending when we reached our decision and, given the state court's forum non conveniens decision, the litigation seemed likely to grow.⁴ Thus, we viewed centralizing the cases before a single judge for pretrial proceedings as the most efficient route for the litigation. Given the potential complexity of this litigation, we assigned this MDL to the Honorable Charles R. Breyer, who we emphasized “has unparalleled experience as a transferee judge.” In re Uber, 2023 WL 6456588, at *2. We were, and remain, confident that Judge Breyer will be able to draw on his wealth of experience to determine how coordinated proceedings regarding common issues can best be utilized.

The only arguments we did not consider at the time of centralization are those that Uber now appears to advance for the first time on appeal. First, Uber argues that centralization would create “a unique situation whereby multidistrict litigation may continue to persist indefinitely because the only common fact—alleged sexual assault by drivers using the Uber App—is not something Uber can ever foreclose entirely.” Pet. for Mandamus at 22. As an initial matter, we disagree that alleged sexual assault by drivers using the Uber App is the only common fact among these actions. See infra note 5. Furthermore, this argument would not have persuaded us to reach a different conclusion. We regularly order centralization in similar situations—for example, in product liability litigation involving a medical device or consumer or pharmaceutical product that is still on the market. Moreover, plaintiffs in these actions do not claim that Uber can prevent all sexual assaults by its drivers, but rather allege that defendants failed to put into place procedures and safeguards that would reduce the number of such assaults.

Second, Uber argues that “the non-California drivers are likely to argue that they cannot be joined or impleaded into an action that is part of a multidistrict litigation in a court where they are not subject to personal jurisdiction of the forum in which the case was originally filed.” Pet. for Mandamus at 26. As Uber concedes, the Panel is not encumbered by considerations of personal jurisdiction in rendering a transfer decision. See id.; In re Delta Dental Antitrust Litig., 509 F. Supp. 3d 1377, 1379–81 (J.P.M.L. 2020). Moreover, to the extent Uber asserts that any of these cases—the vast majority of which have been filed in the Northern District of California—have been filed in an improper venue, centralization provides an opportunity to brief the issue efficiently before a single judge.

III.

Uber's petition states that “[m]ultidistrict litigation is reserved for civil actions in which common questions of fact will predominate over individual factual issues.” Pet. for Mandamus at 19 (quotations omitted). This is a misstatement of the legal standard for centralization under Section 1407. The statute does not contain a predominance requirement. See 28 U.S.C. § 1407(a) (“When civil actions involving one or more common questions of fact are pending in different districts, such actions may be transferred to any district for coordinated or consolidated pretrial proceedings.”) (emphasis added). See also In re Katz Interactive Call Processing Patent Litig., 481 F. Supp. 2d 1353, 1355 (J.P.M.L. 2007) (“Transfer under Section 1407 does not require a complete identity or even a majority of common factual or legal issues as a prerequisite to transfer.”). The Panel has found in certain instances that, where numerous individualized issues overwhelmed any common factual issues, efficiency would not be served by centralization. See, e.g., In re Electrolux Dryer Prods. Liab. Litig., 978 F. Supp. 2d 1376, 1377 (J.P.M.L. 2013) (finding that, upon an established record of procedurally advanced cases, including “numerous trials,” individualized factual issues would overwhelm common ones and, therefore, the Panel was “not persuaded that Section 1407 centralization is necessary either to assure the convenience of the parties and witnesses or for the just and efficient conduct of this litigation”); In re Belviq (Lorcaserin HCl) Prods. Liab. Litig., 555 F. Supp. 3d 1369, 1370 (J.P.M.L. 2021) (“[I]ndividualized factual issues concerning causation will predominate and diminish the potential to achieve significant efficiencies in an MDL.”). To the extent predominance is a factor, it is only one factor the Panel considers when it determines whether centralization of a litigation will result in significant efficiencies for the parties, witnesses, and the courts.

Here, Uber characterized this litigation as involving claims that hinge upon the individual conduct of third-party drivers and the unique circumstances of each plaintiff and each alleged assault. Plaintiffs, in contrast, claimed that at the core of this litigation are their common allegations that Uber's corporate culture, policies, and practices enabled sexual predators to become Uber drivers, and that Uber misrepresented to customers that they would be safe with Uber. We were persuaded that the common factual issues relating to these allegations were sufficient to warrant centralization, and that the efficiencies created by the coordinated treatment of common discovery and pretrial proceedings were likely to be substantial despite any individualized factual issues presented by the cases. See In re Uber, 2023 WL 6456588, at *1. This assessment was made by balancing the statutory factors of convenience, efficiency, and common factual questions—not by mechanically weighing the number of common facts ⁵ against individual ones. Indeed, it is possible that an MDL involving a single common issue could result in significant efficiencies. Engrafting a predominance requirement into the statute that is found nowhere in the text of Section 1407 would substantially change the discretion purposefully afforded the Panel and radically alter the approach taken by the Panel since its inception.

IV.

Uber also asserts that its Terms of Use forbid centralization because they foreclose plaintiffs from seeking centralization altogether. Private contracts among federal court litigants do not circumscribe the Panel's authority to centralize civil actions pending in federal court. Uber likens the relevant clause in its Terms of Use to a forum selection clause, but contractual forum selection clauses “ ‘do not limit the Panel's authority with respect to the selection of [a] transferee district,’ or, by the same token, our authority to transfer tag-along actions to an existing MDL.” In re Park W. Galleries, Inc., Mktg. & Sales Practices Litig., 655 F. Supp. 2d 1378, 1379 (J.P.M.L. 2009) (quoting In re Med. Res. Sec. Litig., MDL No. 1247, 1998 U.S. Dist. LEXIS 15832, at *3 (J.P.M.L. Oct. 7, 1998)). “When civil actions satisfy the criteria set forth in 28 U.S.C. § 1407(a), the statute authorizes the Panel to centralize those actions (as well as any subsequently identified tag-along actions) in ‘any district.’ ” In re Park W., 655 F. Supp. 2d at 1379.

Furthermore, Uber's Terms of Use do not limit our authority to centralize these actions because we are authorized to consider centralization of litigation on our own accord. See 28 U.S.C. § 1407(c) (“Proceedings for the transfer of an action under this section may be initiated by (i) the judicial panel on multidistrict litigation upon its own initiative ․”). Indeed, the Panel notified the parties of this authority during briefing of the motion for centralization. See Hearing Session Order, MDL No. 3084 (J.P.M.L. Aug. 11, 2023), ECF No. 25 (“IT IS FURTHER ORDERED that the Panel may, on its own initiative, consider transfer of any or all of the actions in those matters to any district or districts.”). Thus, Section 1407 authorizes the Panel to identify and centralize for pretrial proceedings multiple actions that would otherwise cause congestion in the federal courts, notwithstanding the parties’ private agreements.

V.

The purpose of Section 1407 is not only to consider the convenience and desires of litigants, but also to promote the just and efficient conduct of the litigation. See 28 U.S.C. § 1407(a); In re Equinox Fitness, 764 F. Supp. 2d at 1348 (“[T]he Panel has an institutional responsibility that goes beyond simply accommodating the particular wishes of the parties.”). In determining that centralization is appropriate for cases involving allegations that Uber failed to implement appropriate safety precautions to protect passengers from sexual assaults, the Panel considered efficiency from both the standpoint of the parties and that of the judiciary itself. After identifying several common factual questions, we concluded that, in light of the number of involved parties, counsel, and courts, centralization is appropriate to ensure that pretrial proceedings are conducted efficiently in these cases.

FOOTNOTES

1.   The Panel refers to such transfers as “centralization” to distinguish this procedure from other forms of coordination or consolidation, such as consolidation under Federal Rule of Civil Procedure 42(a). In practice, centralization is the first step in a process of coordination and consolidation that continues with the transferee judge. After centralization, the transferee judge determines how best to conduct pretrial proceedings given the unique features of the cases before him or her. In proceedings involving mass torts, for example, it may be necessary to identify issues that are common to certain subsets of cases and decide those issues on a consolidated basis, while also ensuring that the MDL as a whole proceeds according to a unified, coordinated case management plan.

2.   The broad grant of discretion afforded the Panel in Section 1407(a), combined with the limitation of appellate review of Panel decisions to applications for an extraordinary writ in Section 1407(e), has resulted in a relative dearth of appellate decisions on the Panel's transfer orders. It appears that no writ overturning centralization or inclusion of a tag-along action in an MDL has ever issued. See, e.g., In re Regents of the Univ. of Cal., 964 F.2d 1128 (Fed. Cir. 1992); In re State of New Mexico, No. 21-1121 (4th Cir. Jun. 16, 2021), ECF No. 29. But cf. In re Food Lion, Inc., Fair Labor Standards Act “Effective Scheduling” Litig., 73 F.3d 528, 532–33 (4th Cir. 1996) (directing the Panel to retransfer cases it had previously remanded to their transferor courts “to provide for review by one appellate court”).

3.   When the issue of centralization is before the Panel, the underlying district courts often slow or stay the cases while centralization is being considered. For this reason, among others, the Panel endeavors to provide decisions on motions to centralize as soon as possible following its bi-monthly hearing sessions to ensure that the litigation progresses. While our order in the present litigation identified several common factual issues among the actions, it was not an exhaustive list.

4.   Cf. Pet. for Mandamus at 8 (“[S]hould the [state court's forum non conveniens] order be affirmed, it is fair to expect that many of those plaintiffs also will file federal actions against Uber”). See also Mem. in Support of Pls.’ Mot. for Transfer at 1–2, MDL No. 3084 (J.P.M.L. Jul. 14, 2023), ECF No. 1-1 (“Given the nationwide scope of Defendants’ transportation services (and the recent Forum Non Conveniens Order in the California Judicial Council Coordination Proceeding No. 5188 (“Uber JCCP”)), it is likely that hundreds or thousands of additional actions will soon be filed ․”).

5.   Based on the briefing of the parties, oral argument by counsel, and our review of the constituent cases, common factual issues in these cases included but were not limited to: (1) Uber's policies for vetting, training, and monitoring of its drivers; (2) Uber's representations about its safety and hiring policies; (3) any knowledge Uber may have had concerning the pervasiveness of sexual assault by its drivers; (4) Uber's practices for gathering information about sexual assault and sexual harassment on its platform; (5) Uber's practices for responding to and investigating sexual assault and harassment complaints; (6) Uber's policies on cooperating with law enforcement in connection with sexual assault and harassment complaints; (7) Uber's policies regarding disciplining drivers about whom it received complaints; and (8) safety measures that were, or could have been but were not, implemented.

KAREN K. CALDWELL, Chair