TABLE OF CONTENTS

The thriving mobile-based ride-sharing and food-delivery business in the United States has proven to be fertile grounds for litigation. Lawsuits, at times in the form of class actions, have been launched by drivers claiming to have been misclassified as independent contractors rather than employees by companies such as Uber, Lyft, and DoorDash. Since the drivers’ agreements with these companies include an arbitration clause as well as a class proceedings waiver, the courts have thus far referred these actions to individual arbitration. This outcome has been commonly perceived as detrimental to the drivers, who lose both their day in court and their ability to sue as a class.

This Essay sets out to counter this perception and the broader assertion that arbitration inherently favors the interests of corporate employers over individual employees. Its starting point is a recent decision by a federal district court holding DoorDash to its contractual obligation to arbitrate the claims of over 5,000 individual drivers. The essay argues that this decision reflects the neutrality of arbitration as a dispute resolution mechanism in the employment context and generally. Moreover, the decision suggests that companies such as DoorDash stand to benefit from using class arbitration, despite businesses’ traditional disdain for class proceedings. Class arbitration has been used in the United States since the 1980s, although some have forecasted its demise following several discouraging Supreme Court decisions. The essay argues that, at least in the context of disputes between drivers and companies such as DoorDash, class arbitration in fact presents an attractive option for both parties and that we may soon see these disputes ushering in a new era of class arbitrations.

* * *

Introduction

“Chutzpa is that quality enshrined in a man who, having killed his mother and father, throws himself on the mercy of the court because he is an orphan.”1

Transportation network companies (“TNCs”) operating in the American mobile-based ride-sharing and food-delivery market2  have increasingly been the object of lawsuits brought by their drivers that allege employment violations and other claims.3  These lawsuits have been filed in courts across the United States, often in the form of class actions, despite the fact that the plaintiff drivers have entered into agreements with TNCs that mandate resolving future disputes through arbitration and prohibit class proceedings. Unsurprisingly, TNCs have responded by filing motions to stay the court proceedings and to refer the drivers’ claims to arbitration. Since these stays would leave each driver to pursue arbitration individually against the relevant TNC, drivers have routinely objected to this outcome. They have argued that enforcing these arbitration clauses would place an unfair financial burden on drivers, would unduly delegate “gateway issues” (such as the enforceability of these clauses) to the arbitrator,4  and would be unconscionable because the clauses are contained in contracts of adhesion, which drivers had no opportunity to negotiate. But consistent with the Federal Arbitration Act’s (FAA) requirement that courts enforce valid arbitration agreements, these arguments from drivers have largely failed.

The FAA, enacted in 1925, governs the enforcement of arbitration agreements in both federal and state courts. It establishes “a liberal federal policy favoring arbitration agreements.” Congress adopted the FAA to counteract courts’ then-prevailing “disapproval” and “hostility” toward arbitration. The FAA thus aimed to “place an arbitration agreement upon the same footing as other contracts . . . and to overrule the judiciary’s longstanding refusal to enforce agreements to arbitrate.” As Congress hoped, this judicial hostility toward arbitration has indeed subsided, with courts increasingly recognizing the advantages of arbitration––its informality, speed, and cost effectiveness––including in the employment context. Hostility has gradually intensified, however, in the court of public opinion. Anti-arbitration arguments have increasingly gained traction, particularly with respect to arbitration agreements between employers and employees.

A detailed discussion of such arguments is beyond the scope of the present Essay, and I do not mean to suggest that these arguments are necessarily without merit. Rather, this Essay sets out to provide a modest counterpoint to the assertion that arbitration is a dispute resolution mechanism that inherently favors the interests of corporate employers at the expense of individual employees. It focuses on the particular context of TNC-driver arbitration because this thriving business has proven to be fertile ground for novel legal issues.5  The Essay takes as its starting point the recent federal district court decision in Abernathy v. DoorDash, Inc., in which the court compelled DoorDash to arbitrate with more than 5,000 individual drivers.

The Essay argues, first, that this decision illustrates how arbitration operates as a neutral and nonpartisan dispute resolution mechanism, even for employment disputes. Second, the Essay argues that the Abernathy decision has the potential to revive a form of arbitration that has effectively been declared “dead”: class arbitration. Class actions generally permit one or more parties to assert claims on behalf of similarly situated individuals. Whether by way of litigation or arbitration, such actions serve to vindicate individual rights in situations where “the claims would be too small to bring economically if each person were required to assert his claim.” For this reason, businesses traditionally have disfavored class actions. Finally, the Essay suggests that the Abernathy decision may usher in a new era of class arbitrations by demonstrating that businesses, or at least TNCs, stand to benefit from such class-wide resolution just as much as their drivers.

Abernathy v. DoorDash, Inc.

The recent federal district court decision in Abernathy v. DoorDash, Inc. illustrates––contrary to popular opinion––that arbitration clauses in TNC-driver agreements can in fact be more advantageous for drivers than for TNCs. The 5,010 plaintiff drivers6  in this case had each entered into a contract with DoorDash that contained a “Mutual Arbitration Provision” applying “to any and all disputes arising out of or relating to this Agreement” and requiring the arbitrations to be administered by the American Arbitration Association (AAA).7  The provision also stipulated that the parties “both waive their right to have any dispute or claim brought, heard or arbitrated as, or to participate in, a class action, collective action and/or representative action.” The AAA’s Employment Arbitration Rules in turn require each individual to pay a filing fee of $300 and the responding company to pay a filing fee of $1,900.

The drivers had filed individual demands for arbitration with the AAA, claiming they had been improperly classified as independent contractors rather than employees in violation of statutes such as the Fair Labor Standards Act and the California Labor Code. They had collectively paid over $1.2 million in filing fees. DoorDash was then called upon to pay its share of the fees in accordance with the AAA Rules. But Doordash refused to do so. It claimed that “there are significant deficiencies with the claimants’ filings,” and that “Doordash is under no obligation to, and will not at this time, tender to AAA the nearly $12 million in administrative fees.”

After Doordash refused to honor its contractual obligation to arbitrate, the drivers sued in the federal district court to compel DoorDash to do so. DoorDash sought to stay the suit until final approval of a related class action involving some of the same drivers, in which a class settlement was pending approval.8  Originally, however, DoorDash had sought to dismiss that parallel action on the ground that the drivers were obligated to arbitrate their claims. The irony was not lost on the district court, which proceeded to deny DoorDash’s motion to stay.

Finding that the electronic arbitration agreements entered into by the 5,010 drivers were “valid, cover the claims in suit, and require arbitration before the AAA,” the court granted the plaintiffs’ motion to compel and ordered DoorDash to immediately commence AAA arbitration. DoorDash’s “hypocrisy” was aptly captured by the court:

For decades, the employer-side bar and their employer clients have forced arbitration clauses upon workers, thus taking away their right to go to court, and forced class-action waivers upon them too, thus taking away their ability to join collectively to vindicate common rights. The employer-side bar has succeeded in the United States Supreme Court to sustain such provisions. The irony, in this case, is that the workers wish to enforce the very provisions forced on them by seeking, even if by the thousands, individual arbitrations, the remnant of procedural rights left to them. The employer here, DoorDash, faced with having to actually honor its side of the bargain, now blanches at the cost of the filing fees it agreed to pay in the arbitration clause. No doubt, DoorDash never expected that so many would actually seek arbitration. Instead, in irony upon irony, DoorDash now wishes to resort to a class-wide lawsuit, the very device it denied to the workers, to avoid its duty to arbitrate. This hypocrisy will not be blessed, at least by this order.

The district court’s decision in Abernathy serves as a warning that courts will enforce valid arbitration agreements even when a party, be it individual or corporate, has buyer’s remorse. This suggests that arbitration may be a neutral and nonaligned dispute resolution mechanism, rather than a mechanism inherently skewed in favor of corporate interests. Moreover, even though TNC-driver arbitration agreements currently prohibit class proceedings, this does not mean that drivers’ individual arbitrations must proceed as isolated silos. Drivers may be able to file joint claims or consolidate their claims, depending on the particularities of the arbitration agreement and the applicable institutional arbitration rules. It may also be increasingly easier for drivers to file multiple individual arbitration claims through services that facilitate mass filings. Finally, drivers may be able to “discuss their claims with one another, pool their resources to hire a lawyer, seek advice and litigation support from a union, solicit support from other employees, and file similar or coordinated individual claims.” Rather than serving the unilateral interests of TNCs, arbitration can, as the Abernathy decision highlights, operate to level the playing field between TNCs and drivers. Arbitration stands to do so even more effectively in the form of class proceedings, an alternative addressed in the next section.

A Revival of Class Arbitration?

The district court’s decision in Abernathy may also encourage TNCs such as DoorDash to allow for classwide claims in their arbitration agreements. While class proceeding waivers such as those currently contained in DoorDash’s driver agreement have been upheld as valid, that does not mean that class arbitration is without value in appropriate circumstances. Contracts of adhesion, in particular, provide a setting in which class arbitration “may offer a better, more efficient, and fairer solution.” Indeed, had DoorDash allowed for such a procedure in its arbitration agreement, it might not have found itself required to pay millions in fees to arbitrate thousands of individual claims.

Abernathy serves as a warning that courts will enforce valid arbitration agreements even when a party, be it individual or corporate, has buyer’s remorse.

Class arbitration is not a new concept. It has been employed in the United States since at least the 1980s. After several inconsistent judgments, the U.S. Supreme Court has clarified that class arbitration may not be compelled where the arbitration agreement is silent or ambiguous on the availability of this procedure. Moreover, the Supreme Court has described the differences between individual arbitration and class arbitration as “crucial” and “fundamental,” suggesting that arbitration loses its advantages over litigation in a class format that “sacrifices the principal advantage of arbitration—its informality—and makes the process slower, more costly, and more likely to generate procedural morass than final judgment.” The Court has also pointed to “serious due process concerns by adjudicating the rights of absent members of the plaintiff class.” Some commentators view these decisions as signaling the “demise of class arbitration” in the United States.

Notwithstanding the discouraging tone of these Supreme Court rulings, there is little doubt that class arbitration is permitted in principle under the FAA where “there is a contractual basis for concluding that the part[ies] agreed to do so” and that, in such circumstances, it is to be enforced by the courts. Moreover, while class arbitrations may indeed differ in some respects from individual arbitrations, these differences are not necessarily disadvantageous. Class arbitrations may be longer and more costly––an unavoidable function of their complexity. But at the same time, unlike individual arbitration, class arbitration may be public rather than confidential.9

 This is an important advantage of class arbitrations, which “are invested with a public interest and often affect a large number of potential class members beyond the named class representative who is purporting to represent their interests.” Beyond these differences, the fundamental distinctive features of arbitration––for instance, parties’ ability to choose their arbitrators, governing laws, and the preferred finality of the proceedings––are entirely transferable from individual to class arbitration. Therefore, it is questionable whether class arbitration is indeed entirely devoid of the advantages of individual arbitration. As noted by one commentator, “it has not yet been proven that parties to large-scale disputes are in fact better off in court than in arbitration.” In any event, class arbitration continues to take place, including in the employment context, notwithstanding predictions of its demise. The AAA’s class action arbitration docket, for instance, shows 582 class actions filed between 2003 and 2020—330 of which are employment disputes.

The AAA’s Supplementary Rules for Class Arbitrations (“Supplementary Rules”) also directly address many of the due process concerns associated with the use of class arbitration. First, to ensure the proficiency and appropriate experience of the arbitrators, the Supplementary Rules provide that at least one of the arbitrators in a class arbitration shall be appointed from a national roster of class arbitration arbitrators maintained by the AAA. To protect the due process rights of absent members of the plaintiff class, the Supplementary Rules require the arbitral tribunal to render, as a threshold matter, an award on the construction of the arbitration clause, in which it is to determine whether the applicable arbitration clause permits the arbitration to proceed on behalf of a class. Significantly, following the issuance of the award, the tribunal shall stay all proceedings for at least thirty days to allow any party to seek a court’s judgment confirming or vacating the award.

In addition to this preliminary interpretation of the arbitration clause, the arbitral tribunal must also determine whether the arbitration should proceed as a class arbitration. In particular, the tribunal should consider whether “the questions of law or fact common to the members of the class predominate over any questions affecting only individual members” and whether “a class arbitration is superior to other available methods for the fair and efficient adjudication of the controversy.”10  As part of making this determination, the tribunal must also decide, in accordance with enumerated criteria, whether one or more members of a class may act in the arbitration as representative parties on behalf of all members of the class.11

Where the arbitral tribunal determines that a class arbitration is appropriate, it must issue an award certifying a class arbitration. This award defines the class, identifies the class representative(s) and counsel, and sets forth the class claims, issues, or defenses, as well as when and how members of the class may be excluded from the class arbitration. As with the preliminary award interpreting the arbitration clause, following the issuance of this second award, the tribunal is to stay the proceedings for thirty days to permit any party to seek a court’s judgment confirming or vacating the award. The tribunal is then to provide all members who can be identified through reasonable effort with the “best notice practicable” of the details of the class arbitration. Finally, the Supplementary Rules also preserve judicial support and supervision of class arbitrations. For instance, they allow a party to initiate court proceedings relating to a class arbitration without being deemed to have waived its right to arbitrate.

Arbitration caters to the vested interest of no one. It is a creature of contract, not of politics.

These provisions of the Supplementary Rules satisfy the due process criteria set out by the Supreme Court to protect non-named class members in actions for monetary damages, described by one commentator as including “that absent class members be afforded notice of the suit, an opportunity to be heard and participate in the litigation, and a chance to opt out, and . . . that the named plaintiff adequately represent at all times the interests of the absent class members.” Moreover, there seems to be no reason why arbitrators, and particularly those operating pursuant to institutional arbitration rules tailored specifically for class arbitrations, would be unable to protect due process interests and deal with potential abuses in the class process as successfully as they do in individual arbitrations. Finally, class arbitration also safeguards parties’ due process rights by preserving class members’ right to opt out for any reason at all.

Conclusion

The district court in Abernathy appropriately vindicated the drivers’ right to invoke the arbitration clause contained in their agreements with DoorDash. Hopefully, this decision will rectify what seems to be an increasingly popular, yet one-sided, perception of arbitration as a dispute resolution mechanism that inherently favors employers’ interests over those of employees. In fact, arbitration caters to the vested interest of no one. It is a creature of contract, not of politics. As such, arbitration agreements are to be enforced according to their terms as agreed upon by the parties, no more and no less. And as Abernathy makes clear, a party, be it corporate or individual, cannot unilaterally opt out of these terms in the event that they become inconvenient.

If DoorDash and other TNCs conclude that individual arbitrations are unworkable, both they and their drivers stand to benefit from pursuing class arbitration. This dispute resolution avenue makes good sense insofar as drivers are parties to substantially the same agreement with the same TNC and bring forward substantially the same claims against it. Class arbitration between TNCs and drivers would no doubt be complex, as would class litigation. This does not mean, however, that class arbitration is not possible or would not be effective. The AAA Supplementary Rules, specifically designed for class arbitrations, illustrate how such arbitrations can be administered efficiently and with appropriate due process safeguards. Ultimately, the desirability of class arbitration, “as Sir Winston Churchill said of democracy, must be evaluated, not in relation to some ideal but in relation to its alternatives. If the alternative in a case of this sort is to force hundreds of individual [drivers] each to litigate its cause with [a TNC] in a separate arbitral forum, then the prospect of classwide arbitration, for all its difficulties, may offer a better, more efficient, and fairer solution” to both sides.    

Regardless of whether class or individual arbitration is used, it is not a tool for evading the consequences of a breach of contract or statute. Quite the contrary, arbitration aims to ensure that any such breach is promptly addressed and compensated for in an effective and efficient forum. In order to preserve the legitimacy of arbitration in the resolution of employment disputes in particular, and commercial disputes in general, counsel, arbitral institutions, and the courts should resist any attempt by a party to circumvent its contractual obligation to arbitrate.

The takeaway for DoorDash and other TNCs from the district court’s decision in Abernathy is therefore twofold: First, in order to reap the benefits of arbitration in a situation of multiple similar claims brought by drivers, TNCs should seriously consider employing class arbitration. Second, whatever form of arbitration TNCs agree to, they should assume that they will be held to it. In other words, “[d]on’t cross the river if you can’t swim the tide.”

  • 1Embury v. King, 361 F.3d 562, n. 22 (9th Cir. 2004) (quoting Leo Rosten, The Joys of Yiddish 94 (1971).
  • 2The three main companies are Uber Technologies Inc, Lyft, Inc, and DoorDash Inc. Uber’s share of the U.S. ride-hailing market in 2018 was estimated at 69 percent. Lyft’s share of the market in 2018 was estimated at 29 percent. DoorDash’s share of the food delivery market in the U.S. in 2018 was estimated at 12 percent.
  • 3In addition to the cases discussed later in the piece, see, for example, Johnston v. Uber Technologies, Inc., 2019 WL 4417682 (N.D. Cal. 2019) (concerning Uber’s termination of its services in Austin, Texas); Peterson v. Lyft, Inc., 2018 WL 6047085 (N.D. Cal 2018) (concerning background checks). Lawsuits have also been filed against TNCs by passengers, see, for example, National Federation of the Blind of California v. Uber Technologies, Inc., 103 F.Supp.3d 1073 (N.D. Cal. 2015) (concerning accessibility) (the case settled, see National Federation of the Blind of California v. Uber Technologies, Inc., 2016 WL 10920461 (N.D. Cal. 2016)); Doe v. Uber Technologies, Inc., 184 F.Supp.3d 774 (N.D. Cal. 2016) (concerning sexual assault).
  • 4See, for example, Mohamed v. Uber Technologies, Inc., 848 F.3d 1201 (9th Cir. 2016); Edwards v. Doordash, Inc., 888 F.3d 738, 746 (5th Cir. 2018).
  • 5As one court has noted, “app based ride-sharing is a disruptive business model in search of a legal theory. The courts that have dealt with litigation arising out of ride-sharing technology have struggled to find an appropriate legal doctrine to fit these novel commercial relationships.” Razak v. Uber Technologies, Inc., 2017 WL 4052417, *1 (E.D. Pa. 2017). See also generally Jeremy Kidd, Who’s Afraid of Uber?,20 Nevada L.J. 581 (2020).
  • 6There were 5,879 plaintiff drivers in the action altogether, but the court found that 869 of them did not have a valid agreement with DoorDash since they had failed to submit declarations setting forth “the identifying information he or she used to register with DoorDash” and “at least referencing in an ascertainable way the specific arbitration he or she clicked through.” Abernathy v. DoorDash, Inc., 2020 WL 619785, *4 (N.D. Cal. 2020).
  • 7The not-for-profit American Arbitration Association (AAA)-International Centre for Dispute Resolution (ICDR) is “the largest private global provider of alternative dispute resolution (ADR) services in the world.” American Arbitration Association, online: https://www.adr.org/mission.
  • 8Abernathy, 2020 WL 619785, at *2 (referring to Marciano v. DoorDash, Inc., No. CGC-18-567869 (Cal. Super. Ct. 2018)). As noted by the court, the hearing on the motion for preliminary approval of the settlement was recently vacated and the action was designated as a complex case. Id. DoorDash has already settled at least one class action lawsuit in the U.S. alleging that its drivers were misclassified as independent contactors for $5 million, Gibbs Law Group, “DoorDash Class Action Lawsuit (2020)”, online: https://www.classlawgroup.com/employment/doordash-driver-class-action-lawsuit/.
  • 9For instance, the AAA’s Supplementary Rules for Class Arbitrations provide that: “The presumption of privacy and confidentiality in arbitration proceedings shall not apply in class arbitrations. All class arbitration hearings and filings may be made public, subject to the authority of the arbitrator to provide otherwise in special circumstances.” American Arbitration Association, Supplementary Rules for Class Arbitrations, at art. 9(a), online: https://adr.org/sites/default/files/document_repository/Supplementary%20Rules%20for%20Class%20Arbitrations.pdf.
  • 10Id. at art. 4(b). In making these findings, the arbitral tribunal should consider: “(1) the interest of members of the class in individually controlling the prosecution or defense of separate arbitrations; (2) the extent and nature of any other proceedings concerning the controversy already commenced by or against members of the class; (3) the desirability or undesirability of concentrating the determination of the claims in a single arbitral forum; and(4) the difficulties likely to be encountered in the management of a class arbitration.”
  • 11Id. at art. 4(a): “The arbitrator shall permit a representative to do so only if each of the following conditions is met: (1) the class is so numerous that joinder of separate arbitrations on behalf of all members is impracticable; (2) there are questions of law or fact common to the class; (3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; (4) the representative parties will fairly and adequately protect the interests of the class; (5) counsel selected to represent the class will fairly and adequately protect the interests of the class; and (6) each class member has entered into an agreement containing an arbitration clause which is substantially similar to that signed by the class representative(s) and each of the other class members.”