The Yale Law Journal


Arbitration Asymmetries in Class Actions

15 Sep 2021
January 23, 2022

abstract. Courts frequently deny class certification when confronted with “arbitration asymmetries”: cases where the class representative is not bound to arbitrate claims, but class members may be. The result? Courts enforce illegal or nonexistent arbitration agreements. To avoid such patent injustice, this Essay advances an alternate approach to arbitration asymmetries.


As arbitration agreements have become increasingly commonplace,1 a question has arisen for class-action litigation: when a defendant asserts that putative class members are subject to arbitration agreements, can a putative class representative who is not bound to arbitrate her claims still certify the class? I describe cases that present this question as cases with “arbitration asymmetries.” In these cases, the putative class representative’s good luck often becomes a class-action misfortune. Where a class representative has escaped an arbitration agreement for any number of reasons (such as a litigation waiver or unique factual circumstances), courts tend to conclude that the representative is atypical or inadequate to represent the class. In short, courts decline to certify class actions with arbitration asymmetries.

Consider the example of Tan v. Grubhub, Inc.2 There, a driver for Grubhub—an app that facilitates food delivery—alleged that Grubhub had misclassified him as an independent contractor and thus denied him the benefit of certain wage-and-hour protections that only apply to formal employees. Because individual recovery in wage-and-hour cases tends to be minimal, the driver sought to represent a class of other Grubhub delivery drivers so that legal expenses could be spread broadly. Further, if the driver prevailed in such a class action, the penalty to Grubhub would be multiplied, ensuring that the company would be meaningfully held to account for its misconduct. Yet Grubhub’s employment contract contained an arbitration provision. And while this posed no obstacle for the driver because he had opted out of this provision, the class he sought to represent had not. On the basis of this arbitration asymmetry alone, the court declined to certify the class.3

Importantly, it is possible that none of the putative class members were actually bound to arbitrate their claims. But the court never determined whether the putative class members in the Tan litigation were actually subject to arbitration agreements with Grubhub and were therefore precluded from joining the class. The mere fact that the putative class members were potentially bound, whereas the putative class representative was not, in and of itself defeated class certification. In fact, the driver at the helm of the Tan litigation had sought to demonstrate that Grubhub’s arbitration agreements were unenforceable on various grounds common to the class. The court even acknowledged that Grubhub’s delivery drivers may have lacked adequate notice of the arbitration provision’s terms.4 But the court ultimately refused to entertain these arguments, stating that the driver could not challenge the enforceability of an arbitration agreement that did not apply to him.5

As this example shows, courts deny class certification in arbitration asymmetry cases regardless of the strength of the defendant’s asserted arbitration rights vis-à-vis putative class members. Courts reason that the putative class representative lacks standing to litigate the arbitration issue, even if she raises colorable arguments that no valid arbitration agreement exists. They also hesitate to pass judgment on putative class members’ rights prior to class certification, when the putative class members are not yet formal parties to the case. In other words, questions of justiciability lurk behind courts’ refusal to certify class actions with arbitration asymmetries.

Outcomes like this are a boon for defendants. A defendant can shield itself from class proceedings merely by presenting evidence of an arbitration agreement with unnamed class members, even if the agreement is illegal, invalid, or otherwise unenforceable. Equally troublingly, a defendant can manufacture an arbitration asymmetry to avoid class litigation by strategically waiving its arbitration rights as to the putative class representative and thereby setting her apart from the putative class. In effect, arbitration asymmetry jurisprudence offers employers a means to opt-out of class litigation.

This also provides defendants with the benefits of mandatory arbitration without requiring them to actually demonstrate the existence of enforceable arbitration agreements. After all, the primary purpose of mandatory arbitration agreements for most defendants is the avoidance of class actions because class-actions are a powerful mechanism for holding monied interests accountable. Without access to the class-action device, workers, consumers, and other small-claims litigants have no incentive to bring suit in either a judicial or arbitral forum. Those individuals are unable to vindicate their legal rights.

But arbitration asymmetries need not hobble class actions. This Essay argues that courts should give class representatives an opportunity to contest the validity of arbitration rights asserted against putative class members before declining to certify a class.6 Courts may adjudicate the arbitration issue at either the class certification stage or after a class has been certified without running afoul of either Rule 23 or the case-or-controversy requirement of Article III. Only if absent class members have indeed waived their rights to a judicial forum by agreeing to arbitrate their claims should a court determine that an action may not proceed as a class.

Part I sets the stage by assessing the prevailing approach to the arbitration asymmetry problem. It examines the flaws in courts’ existing reasoning and attempts to lay out their underlying concerns. Part II demonstrates that the current treatment of arbitration asymmetries is untenable because it unfairly disadvantages class plaintiffs and threatens to immunize defendants from legal liability—underlining the importance of this Essay’s contribution. Part III makes the case for why a class representative not bound by an arbitration agreement nevertheless has standing to argue that the putative class members are not bound to arbitrate their claims. Part IV contends that a court may adjudicate a defendant’s arbitration rights as to absent class members either at the class certification stage or after a class has been certified.

I. the prevailing approach to arbitration asymmetries and its underlying jurisdictional puzzles

A. The Prevailing Approach

By and large, courts have determined that a putative class representative who is not bound to arbitrate her claims may not certify a class where a defendant has asserted that putative class members are bound by arbitration agreements.7 Relying on either or both of the typicality or adequacy requirements of Rule 23,8 courts often conclude that the putative class representative is “in a unique position” vis-à-vis the class and thus atypical,9 or “not an adequate representative . . . like other potential class members might be” because she is unable to adjudicate the arbitration issue.10

Arbitration asymmetries arise in a variety of circumstances. Unlike members of the proposed class, a putative class representative may have opted out of an arbitration agreement.11 Alternatively, particular factual circumstances may give the class representative a unique defense to the defendant’s assertion of arbitration rights—for example, she may never have come into contact with the arbitration agreement at issue.12 But a proposed class may equally run into this problem if the defendant corporation simply waived its arbitration rights as to the putative class representative.13 In these cases, the defendant has essentially placed the putative class representative in a “class of one” because the defendant either: (1) consented to the judicial forum with respect to the putative class representative’s individual claims or (2) litigated in a fashion inconsistent with an intention to exercise arbitration rights14 and is now barred from asserting their arbitration rights against the named plaintiff.15

Courts decide that classes may not be certified under these circumstances while candidly admitting that they do not and cannot know whether putative class members have in fact entered valid and enforceable arbitration agreements.16 As the court in Jensen v. Cablevisions Systems Corp. explained, where a defendant asserts arbitration rights against putative class members, the “mere potential that the relevant arbitration provision is valid is sufficient to preclude a named plaintiff who [is not bound by] the provision from representing a class largely made up of individuals that may be subject to the agreement.”17 Speculative language runs throughout the arbitration asymmetry decisions, with courts determining that putative class members’ claims are “likely barred,”18 “potentially subject,”19 or “may be bound by”20 arbitration agreements.

Sometimes the hypothetical language is dropped altogether and the validity of the arbitration agreement is simply assumed. For example, in Panzer v. Verde Energy USA, Inc., a putative class representative alleged that he had not received an arbitration agreement, whereas members of the proposed class had.21 The court questioned whether the named plaintiff was adequate as a class representative because “the class he seeks to represent did agree to arbitrate,” and concluded that “[the class representative’s] interests [did] not align with those whom he [sought] to represent.”22 The court made these statements despite the fact that, at the time, the defendant had not met its burden to show that putative class members were actually bound by arbitration agreements.

In fact, many courts take this tack even where colorable arguments are presented that the purported arbitration agreements are unenforceable. In Panzer, for example, the plaintiff had challenged the validity of the arbitration agreement in question on grounds of unconscionability and lack of consideration—arguments the court simply declined to consider.23 Similarly, in Jensen, the plaintiff asserted that the corporate defendant had failed to incorporate a reference to any arbitration provision in its terms of service.24 The court refused to entertain this argument, stating that it was not obligated to determine “the validity of the [arbitration] agreement,” but only “whether the presence of class members that [were] potentially subject to the provision satisfie[d] the requirements of Rule 23.”25 A determination as to the validity of the arbitration agreement, the court held, would be “procedurally improper and analogous to an advisory opinion.”26

B. The Misfit of Rule 23’s Typicality and Adequacy Rubrics

Although courts often cite Rule 23’s typicality and adequacy requirements in their reasoning, those requirements make for strange policemen here.27 The usual concerns underlying the typicality and adequacy requirements do not apply to cases involving arbitration asymmetries.

To start, typicality is a “permissive standard[],”28 meant to ensure that “the interests of the class representative align with those of the class, so that by prosecuting his own case he simultaneously advances the interests of the absent class members.”29 The key inquiry is whether the class representative’s claims “arise from a similar course of conduct and share the same legal theory” as those of the putative class members.30 Where an arbitration asymmetry arises, it is seldom contested that the named plaintiff’s claims meet this requirement.

Although defenses are relevant to the typicality inquiry, the traditional formulation focuses on whether the “proposed representative may face significant unique or atypical defenses to her claims.”31 That is because “there is a danger that absent class members will suffer if their representative is preoccupied with defenses [that are] unique to [her].”32 But of course, a class representative who seeks to demonstrate that putative class members are not bound by arbitration agreements would hardly derail or distract from the interests of the class action. Rather, answering that question would be central to the class action’s success. In sum, the typicality requirement should not bar class actions with arbitration asymmetries.

Courts’ use of “adequacy” also fails to make sense as a device for disqualifying class representatives in the context of arbitration asymmetries. As the Supreme Court explained in Amchem Products, Inc. v. Windsor, “the adequacy inquiry under Rule 23(a)(4) serves to uncover conflicts of interest between named parties and the class they seek to represent.”33 The interests of absent class members and the putative class representative must therefore be “aligned” for the adequacy requirement to be met.34 Additionally, both the class representative and class counsel must “understand that they are acting in a representative capacity and will prosecute the action throughout its duration fairly, vigorously, and competently on behalf of the class.”35

Again, the existence of arbitration agreements that may bind putative class members does not create a “conflict[] of interest” with the named class representative.36 Rather, both putative class members and the class representative share an interest in ensuring that the defendant does not prevail in enforcing illegal or nonexistent arbitration agreements that would threaten the class action. And just because a putative class representative is not subject to an arbitration agreement does not indicate that she cannot “fairly, vigorously, and competently” litigate a class action’s central claims.37 An arbitration asymmetry should thus not create adequacy concerns.

C. The Underlying Article III Puzzles

The misfit of Rule 23’s typicality and adequacy requirements suggests that something else motivates the prevailing approach to arbitration asymmetries. And where courts include more in-depth analysis of the arbitration asymmetry issue, it becomes apparent that Article III jurisdictional puzzles are at play. Courts are reticent to probe the validity of arbitration agreements as to putative class members because they are unsure of two things: (1) whether a class representative has standing to litigate the arbitration rights of absent class members; and (2) whether a court may exercise jurisdiction over the rights of putative class members prior to certification, given that putative class members are not yet considered “parties” before the court. Courts may decline certification on typicality or adequacy grounds, but it is Article III—not Rule 23—that poses the true roadblock to class actions with arbitration asymmetries.

With respect to standing, many courts reason that since the putative class representative has somehow escaped the arbitration agreement in question, she lacks standing to contest its enforceability as to absent class members.38 For example, in Conde v. Open Door Marketing, LLC, a district court held that the named plaintiffs were not “typical” of the class because, unlike putative class members, they had not signed an arbitration agreement with the defendants.39 The court stated that the named plaintiffs were not “personally affected by the arbitration agreements,” had “no interest in the enforceability of the arbitration agreement itself,” and thus “lack[ed] the ability to challenge the agreements on behalf of individuals who did.”40 As illustrated by Conde, what renders a putative class representative atypical or inadequate is not necessarily a difference as to arbitration rights (which many courts concede may not exist if the purported arbitration agreement is unenforceable or invalid41). Rather, the animating concern is that the class representative lacks the personal stake necessary to litigate the arbitration rights of absent class members. At bottom, it is the Article III standing requirement—not Rule 23—that gives these courts pause.

Courts also hesitate to adjudicate the arbitration rights of putative class members prior to the certification of a class. Before class certification, putative class members are not yet formal “parties” to the litigation.42 As a result, courts hold that they lack jurisdiction over the claims of putative class members altogether.43 Consider, for instance, the Eleventh Circuit’s holding in In re Checking Account Overdraft Litigation, an action brought against Wells Fargo and Wachovia Bank alleging that they unlawfully charged the plaintiffs overdraft fees.44 The banks had waived their arbitration rights as to the named plaintiffs, but had moved to compel arbitration of putative class members’ claims prior to class certification. The district court denied the motion to compel arbitration, and the banks appealed. The Eleventh Circuit vacated the district court’s order, finding that “the District Court lacked jurisdiction to resolve th[e] question.”45 “Absent class certification,” the Eleventh Circuit reasoned, “there is no justiciable controversy between [the defendants] and the unnamed putative class members.”46 As such, the district court in that case could not “purport[]” to rule on the “hypothetical claims that might be raised in the future by hypothetical plaintiffs” without stepping outside the bounds of Article III’s case-or-controversy

Adjudicating the arbitration rights of putative class members thus presents a “conundrum.”48 “The validity [or] enforceability of [an alleged arbitration agreement] cannot be adjudicated until after class certification proceedings” because before that point, a court technically lacks jurisdiction over the claims of absent class members.49 But a certified class may not include “individuals who are bound by arbitration agreements” because they have waived their right to a judicial forum.50 Wary of “issu[ing] a ruling regarding the enforceability” of an arbitration agreement as to “class members who are not before th[e] Court,”51 courts simply assume the validity of the arbitration agreements and decline to certify proposed classes that present the arbitration-asymmetry problem.

II. policy arguments against the prevailing approach

A new approach to arbitration asymmetries is not only legally defensible, but urgent as a policy matter. Courts’ current approach to arbitration asymmetries gives effect to arbitration agreements regardless of their enforceability—meaning that, in practice, defendants are able to enforce illegal arbitration provisions and keep viable claims out of court. Perhaps just as importantly, the current approach enables defendants to shield themselves from class proceedings. Defendants need only suggest that putative class members are bound by arbitration agreements, knowing that courts will not subject those agreements to scrutiny. The result is that defendants can avoid legal liability altogether.52

Courts already enforce arbitration agreements that prevent litigants from aggregating claims, even where doing so deprives litigants of their substantive rights. In AT&T Mobility LLC v. Concepcion, the Supreme Court held that arbitration agreements may be enforced where a take-it-or-leave-it consumer contract includes an arbitration agreement with a class action ban.53 Concepcion involved claims worth $30.22.54 In such circumstances, as one circuit judge acknowledged in another case, the “realistic alternative to a class action is not [a multitude of] individual suits, but zero individual suits, as only a lunatic or a fanatic sues for $30.”55 Similarly, in American Express Co. v. Italian Colors Restaurant, the Court held that an arbitration agreement may be enforced even where the litigants could not “effectively vindicate [their] rights in the arbitral forum.”56

Even if litigants could prevail in arbitration, the reality is that arbitration hardly ever takes place. Though the Supreme Court imagines arbitration as an alternative forum for dispute resolution, studies have demonstrated that as a practical matter, “almost no consumers or employees ‘do’ arbitration at all.”57

At least in other contexts, defendants must first meet their burden of proving that an arbitration agreement is valid and enforceable before keeping claims out of court.58 Prior to granting a motion to compel arbitration, a court must decide whether “(1) a valid agreement to arbitrate exists, and (2) the particular dispute falls within the scope of that agreement.”59 And in adjudicating motions to compel arbitration, courts must provide the “party opposing arbitration . . . ’the benefit of all reasonable doubts and inferences that may arise.’”60 But in the arbitration-asymmetry context, courts simply presuppose the enforceability of the arbitration agreement in question. This might be thought of in terms of a burden-of-error allocation: by declining to certify a class in these circumstances, courts place the burden of error on plaintiffs. After all, if the arbitration agreements were not enforceable, class treatment would be appropriate. One could imagine a different allocation, whereby courts certify classes despite a possible arbitration asymmetry and defer adjudicating the agreement’s enforceability until afterward.61

Perhaps most disconcertingly, the current approach to arbitration asymmetries equips defendants with the ability to block class actions through voluntary action within the lawsuit. Arguments about a putative class representative’s standing to litigate an arbitration agreement or the court’s jurisdiction over putative class members’ claims apply even where the defendant manufactures the arbitration asymmetry. The approach thus has the “disastrous effect of enabling defendants to essentially opt-out of Rule 23.”62

For example, a putative class representative might have, at one point, been subject to an arbitration agreement, just like putative class members. But, as discussed above, a defendant may waive its arbitration rights as to the putative class representative and then claim that because she is no longer “personally affected by the arbitration agreement[],” she “lack[s] the ability to challenge the agreements on behalf of individuals who [are].”63 Similarly, a defendant could bring flimsy arguments to compel arbitration of a putative class representative’s claims and then, if the motion is unsuccessful, use the named plaintiff’s success as a sword against class action treatment. This would create a win-win situation for defendants: either the putative class representative’s claims or the putative class members’ claims will be shut out of court. In both cases, the defendant becomes immunized from class liability in a judicial forum. Moreover, these processes of “picking off” putative class representatives could occur successively. Each time a named plaintiff stepped into the role of putative class representative, the defendant could waive its arbitration rights as to that individual and undermine the viability of the class. Such gamesmanship should not be permitted.

By allowing purported arbitration agreements to hamper class actions, courts vindicate the purposes underlying the trend toward mandatory arbitration without requiring defendants to actually secure enforceable arbitration agreements with putative class members. While companies seldom compel individuals to arbitrate their claims, “it is quite common for arbitration clauses to be invoked to block class actions.”64 Indeed, the Consumer Financial Protection Bureau determined that arbitration clauses are already “design[ed] . . . to block class actions in court.”65 Most arbitration agreements today now include class action waivers.66 But it appears that defendants need not labor over fine print: the mere possibility that putative class members signed arbitration agreements is now enough to prevent them from aggregating their claims and vindicating their rights in a judicial forum.

The prevailing approach to arbitration asymmetries has political and economic contours. Because so few individuals avail themselves of arbitration, the enforcement of arbitration provisions has a “claim-suppressing” effect.67 By disappearing valid monetary claims, arbitration facilitates an upward transfer of wealth to increasingly concentrated corporate power.68 Meanwhile, the class action is a mechanism for “counter[ing] concentrated power.”69 By aggregating small claims, class actions vindicate the rights of individuals who have no incentive to bring suits individually.70 Without Rule 23’s cost-sharing mechanism, those claims are unlikely to ever see the light of day.71 By enforcing arbitration provisions and making the class device inaccessible, courts’ current approach to arbitration asymmetries thus has a doubly claim-suppressing effect. Fewer and fewer litigants will have the opportunity to vindicate their rights and obtain just outcomes.

Workers and consumers are the categories of individuals at once most vulnerable to arbitration agreements and most dependent on the class-action device. A Grubhub delivery driver, for example, has limited ability to negotiate her terms of employment. Likewise, a consumer contracting for internet services or a cellphone possesses little to no bargaining power. Further, when workers and consumers are wronged, they typically end up with “negative-value” claims: the cost of litigation exceeds any expected monetary reward.72 And low-income workers and consumers are particularly burdened because they are mostly likely to be the victims of abusive practices—whether consumer scams or illegal workplace exploitation.73 Research shows that “employers engage in misconduct against low-wage and unskilled workers because they have less bargaining power and are less likely to sue.”74 Similarly, predatory business practices “specifically target[]” low-income groups as consumers.75 Where class-action procedures are foreclosed, it is the most marginalized members of society that suffer.

Courts’ prevailing approach to arbitration asymmetries equips defendants with the ability to kneecap class actions and adds potency to arbitration agreements—even those that would not stand up to scrutiny in court. This exacerbates inequality, tilting the scales in favor of defendants that already benefit from consolidated economic power. As a result, it is critically important, as a both normative matter and a matter of policy, that courts adopt a new approach to arbitration asymmetries—one that does not unfairly advantage defendants.

III. the standing solution: class representatives’ “private attorney general” interest in litigating putative class members’ arbitration rights

Courts’ underlying concerns about arbitration asymmetries are unfounded. First, a putative class representative has standing to advance a claim that absent class members are not bound by arbitration agreements.76 To see why, it is useful to consider a different fact pattern that also implicates a putative class representative’s ability to argue on behalf of absent class members: cases in which a putative representative’s individual claims have become moot, but the putative representative continues to pursue the claims of putative class members anyway.

United States Parole Commission v. Geraghty is an example of one such case.77 There, the Supreme Court asked “whether a trial court’s denial of a motion for certification of a class may be reviewed on appeal after the named plaintiff’s personal claim has become ‘moot.’”78 A prisoner challenged certain parole guidelines that prevented his release on statutory and constitutional grounds, filing a putative class action on behalf of similarly situated prisoners.79 Based on the merits of Geraghty’s individual claims, the district court first denied his motion for class certification and subsequently ruled that the parole guidelines could remain intact.80 Geraghty appealed both decisions, but was mandatorily released from prison in the meantime.81 The U.S. Parole Commission argued that because Geraghty’s individual claims were now moot, the appellate court lacked Article III jurisdiction to entertain his appeal of the denial of his motion for class certification.82

The Supreme Court held that the denial of his motion for class certification was not moot. Geraghty still had standing to litigate that appeal. “A plaintiff who brings a class action presents two separate issues for judicial resolution,” stated Justice Blackmun, writing for the majority.83 “One is the claim on the merits; the other is the claim that he is entitled to represent a class.”84 True, Geraghty’s claim on the merits had been resolved and he no longer possessed the “personal stake” necessary to assure the Court of jurisdiction over his individual claims.85 But Justice Blackmun saw fit to examine Geraghty’s “‘personal stake’ in the class certification claim” separately.86

That inquiry led the Supreme Court in Geraghty to the purposes behind both the class-action device and the Article III case-or-controversy requirement. Class actions, Justice Blackmun noted, were developed in part to “provi[de] . . . a convenient and economical means for disposing of similar lawsuits[] and [to] facilitat[e] . . . the spreading of litigation costs among numerous litigants with similar claims.”87 And Rule 23 also gives putative class representatives “the right to have a class certified if the requirements of the Rules are met”88—an interest, Justice Blackmun explained, that was more akin to that of a “private attorney general”89 than a “personal stake.”90 This “private attorney general” interest could satisfy the purposes behind Article III’s “‘personal stake’ requirement.”91 After all, “the purpose of the ‘personal stake’ requirement is to assure that the case is in a form capable of judicial resolution,” which entails “sharply presented issues in a concrete factual setting and self-interested parties vigorously advocating opposing positions.”92 Despite the fact that Geraghty’s merits claim had expired, “[t]he question whether class certification is appropriate remain[ed] as a concrete, sharply presented issue.”93 Therefore, “vigorous advocacy [could] be assured through means other than the traditional requirement of a ‘personal stake in the outcome’”94 and could be satisfied by Geraghty’s “private attorney general”95 interest in certifying a class.

Although the Supreme Court’s decision-making on mootness and class actions has been checkered96 and the “private attorney general”97 concept of the class action is arguably outdated, Geraghty nevertheless remains good law today.98 For example, the Second Circuit relied on Geraghty last year in Jin v. Shanghai Original, Inc., holding that where a class representative prevails on the merits of his individual claims following pretrial decertification of his class, he retains standing to appeal the decertification order.99 In reaching its decision, the Second Circuit acknowledged that interests beyond those of the named plaintiff must be “considered when questions touching on justiciability are presented in the class-action context”—including “the responsibility of named plaintiffs to represent the collective interests of the putative class.”100

Unlike Geraghty, whose merits claim simply expired,101 Jin obtained complete relief for his individual claim through a successful judgment on the merits.102 Still, the Second Circuit determined that he retained a “private attorney general” interest in litigating the motion for class certification.103 The Second Circuit explained that “[t]he private attorney general concept relates to the objectives of the class action device, which include deterring misconduct through private enforcement of vital public policies.”104 The facts underlying Jin offered a prime example: Jininvolved small claims arising from violations of New York State Labor Law. Independent of class certification, individual class members “lack[ed] incentive” to litigate their claims and “would not [have] obtain[ed] relief.”105 And because the class was largely comprised of immigrants, fear of retaliation served as an additional “impediment for class members,” further justifying the need for representative litigation.106

Geraghty and the notion of a “private attorney general interest” for putative class representatives provide insight into how courts can resolve the problem of standing in the context of arbitration asymmetries. This doctrine directs courts to take a functional approach107 to “questions touching on justiciability . . . in the class-action context,”108 considering both the purposes behind the class-action device and the justiciability requirements of Article III in addressing arbitration asymmetries. And courts must recognize a key principle from Geraghty: putative class representatives have a freestanding interest in certifying a class sufficient to satisfy the Article III case-or-controversy requirement, regardless of their underlying “personal stake” in the claims at issue and even where no class has been formally certified.109

With these background principles in mind, it becomes clear that a putative class representative does not need to have a “personal stake” in the validity of an arbitration agreement in order to argue that it may not be enforced against putative class members. Where an arbitration asymmetry is at issue, a putative class representative maintains the “concrete adverseness”110to the defendant that is required for standing. That adversity is preserved both by the putative class representative’s “personal stake” in her underlying claim against the defendant and by her “private attorney general interest” in litigating the class action. That is to say, the argument for justiciability here is even stronger than in Geraghty or Jin, because here the putative class representative’s “personal stake” in her individual claims against the defendant remains live. Surely in these circumstances, a putative class representative arguing that putative class members are not bound by arbitration agreements meets the “imperatives of a dispute capable of judicial resolution”— “sharply presented issues in a concrete factual setting and self-interested parties vigorously advocating opposing positions.”111

The approach outlined here also better comports with the purposes underlying the class-action device. As Geraghty and Jinboth recognized, class actions are meant to facilitate cost sharing among small-claims litigants in order to deter misconduct that no one would otherwise have an incentive to challenge.112 Much as in the strategic mooting context—where flexible understandings of justiciability prevent gamesmanship on the part of defendants attempting to avoid legal liability113—enabling putative class representatives to litigate the enforceability of arbitration agreements on behalf of absent class members ensures that the decks are not stacked against a plaintiff class.

The subsequent history of the Eleventh Circuit’s In re Checking Account decision is an informative case study on how this approach may play out in practice. In that case, named plaintiffs were permitted to argue that arbitration agreements purportedly binding putative class members were not valid.114

Ironically, In re Checking Account is often cited for the proposition that arbitration asymmetries create an irresolvable standing problem because the opinion contains the following oft-quoted line: “[T]he named plaintiffs lack standing to assert any rights the unnamed putative class members might have to preclude [the defendant] from moving to compel arbitration because the named plaintiffs have no cognizable stake in the outcome of that question.”115 Yet, as explained above, the opinion was written in the context of an appeal from the district court’s denial of the defendant’s motion to compel arbitration against putative class members before class certification proceedings had even commenced. The Eleventh Circuit made clear that the issue had just been litigated prematurely, not that it could not be litigated at all:

[A]s a practical matter, whether [the defendant] can compel the unnamed putative class members to arbitrate their claims may be highly relevant to the named plaintiffs, given that the answer to that question may effectively decide the viability of their class action as such. However that issue is properly litigated via a motion to certify a class, not in defense of a decision the District Court had no jurisdiction to make.116

The opinion thus recognized that a putative class representative does have standing to litigate the enforceability of arbitration agreements as to putative class members because of their stake in certifying a class action.

Following the Eleventh Circuit’s opinion, the district court certified the putative class on remand.117 And after certification, the defendants once again moved to compel arbitration against absent class members.118 Despite the fact that the named plaintiffs had secured a waiver to the arbitration agreement and were not subject to its terms, the district court permitted them to argue against the motion—by claiming, for example, that the purported arbitration agreements were illusory and unconscionable—without questioning whether the named plaintiffs had a sufficient “personal stake” to litigate those claims.119 On appeal, the Eleventh Circuit affirmed the district court’s
decision.120 There again, the fact that named plaintiffs were litigating the arbitration rights of absent class members did not cause concern. No one raised any question of jurisdictional impropriety. Once the class was certified, the standing concern slipped away.

Another informative example is Milbourne v. JRK Residential America, LLC.121 There, a district court addressed the issue of a class representative’s standing to litigate the arbitration rights of absent class members head on. After the class had been certified, the defendant moved to compel arbitration of absent class members. The defendant argued that the class representatives lacked standing to contest the motion because they, unlike the putative class members, had not signed an arbitration agreement.122 The court held that the class representatives could contest the motion without running afoul of Article III: “[T]he issue of whether some class members may be subject to the defense of mandatory arbitration does not alter the fact that there is an actual controversy between the defendant and the class members—it merely determines the forum in which that controversy will be decided.”123 Moreover, standing could hardly pose a barrier where the defendant “cannot[] dispute that each class member, including the [n]amed [p]laintiffs, has an actual, justiciable claim against [the defendant].”124

The ease with which the standing question falls away once a class has been certified might have something to do with the unique nature of class-action litigation. “Once certified, the class as a whole is the litigating entity,”125 and the class representative takes on a “private attorney general”126 function with not only a capacity, but a “responsibility . . . to represent the collective interests of the putative class.”127 That is, the personal and collective interests of the class representative and absent class members merge.

Interestingly, the court in Milbourne agreed with other federal courts that “the issue of whether arbitration may be available as to some class members . . . [is] relevant to the class certification analysis under [Rule] 23, rather than standing.”128 The court also suggested that an arbitration asymmetry “admittedly affects the typicality of the [n]amed [p]laintiffs’ claims and their ability to adequately and fairly represent the class as a whole.”129

But it is not at all clear why that should be the case. As discussed above, the typicality requirement assures a reasonable degree of class cohesiveness with respect to the underlying claims, while the adequacy requirement weeds out cases where a class representative’s interests are antagonistic to the class. The reason courts deny class certification in the face of arbitration asymmetries is because they fear that putative class representatives “lack the ability to challenge the [arbitration] agreements on behalf of [absent class members].”130 But if once a class is certified, a class representative may properly litigate the enforceability of those arbitration agreements, then (working backward) there should be no obstacle to certifying a class in the first place.

In summary, courts should not decline to certify classes with arbitration asymmetries because they fear that the class representative lacks standing to litigate the arbitration issue as to absent class members. Class representatives have a justiciable interest in ensuring that a defendant not enforce an invalid or illegal arbitration agreement against absent class members.

IV. the jurisdictional solution: courts’ power to ascertain whether putative class members are bound by arbitration agreements

The differing treatment of arbitration asymmetries in the pre- and postcertification contexts draws attention to a second jurisdictional question: when may a court adjudicate whether unnamed class members are bound by arbitration agreements?

The concern underlying the question of “when” to address arbitration asymmetries is that before a class is certified, putative class members are not yet considered “parties” to the litigation.131 Rule 23’s class certification requirements are designed to “protect absent class members,”132 such that only after the requirements are met may members of the proposed class be “legally b[ound]” by the outcome.133 Prior to class certification, a court lacks jurisdiction over the claims of putative class members altogether.134 Even if a putative class representative had standing to argue that an arbitration agreement was unenforceable as to putative class members, courts have expressed a reluctance to rule on these agreements precertification because it would amount to a “procedurally improper . . . advisory opinion.”135 This follows from the notion that a court may not rule on the “hypothetical claims that might be raised in the future by hypothetical plaintiffs” without stepping outside the bounds of Article III’s case-or-controversy requirement.136

The notion that courts cannot exercise jurisdiction over putative class members’ claims prior to certification can cut in either direction. A court, as in Jensen, may use it as a justification for declining to certify a class, despite being unsure whether the arbitration agreement is valid and would indeed preclude putative class members from litigating claims in a judicial forum.137 Conversely, a court, as in In re Checking Account, may rely on its precertification lack of jurisdiction over putative class members’ claims to defer ruling on the effect of an arbitration agreement until after a class has been certified.138

From a normative perspective, the latter—deferring ruling on the arbitration question—is clearly the more compelling approach. Whereas the court in Jensen disclaims issuing an “advisory opinion,” its decision not to certify a class rests on pure speculation. In order to find that a class representative is atypical or inadequate due to an arbitration asymmetry, a court must presuppose the enforceability of the arbitration agreement in question. Moreover, as described at greater length in Part II above, it equips defendants with the ability to block class actions and avoid meritorious legal disputes. The In re Checking Account approach, by contrast, ensures that putative class members are not excluded from the class unless they are actually bound by valid arbitration agreements.

Still, one might protest, a court should not be required to turn a blind eye to the possibility that an arbitration agreement binds putative class members and precludes them from participation in the class. Even granting that an arbitration asymmetry creates no typicality or adequacy concerns, it may implicate numerosity—the requirement that the members of a class be sufficiently numerous as to make ordinary litigation impractical. If the arbitration agreement were enforceable against putative class members, it could leave the class representative in a “class of one.”139

Indeed, the In re Checking Account litigation itself reveals the inefficiencies of waiting to adjudicate the arbitration question until after class certification. After the district court certified a class, the defendant subsequently moved to compel arbitration of absent class members’ claims, and the court determined that the arbitration agreements were
valid.140 The claims of “all members of the certified classes other than the named [p]laintiffs” were dismissed—effectively dismantling the class action.141 Of course, no one was excluded from the class without the court first determining that they were, in fact, bound by valid arbitration agreements. And the class representatives were given an opportunity to contest the validity of the purported arbitration agreements. Still, the In re Checking Account litigation suggests that courts may reasonably want to adjudicate the validity of arbitration agreements earlier, before a class has been formally certified.

The nature of Rule 23 class certification proceedings should make such an inquiry appropriate without creating jurisdictional difficulties. Judges routinely “make whatever factual and legal inquiries are necessary under Rule 23”142 before certifying class actions, a process that sometimes involves evaluating the contours of the claims of putative class members. Consider, for example, a case where a defendant asserts that a class fails to meet the numerosity requirement of Rule 23 because the majority of persons within the class definition would be barred by a statute of limitations. A court in that circumstance might apply law to facts to determine the impact of this possible defense on the proposed class. The court would be unlikely to decline to certify the class on the grounds that it could not wade into the statute-of-limitations question at all. “When defendants opposing class certification raise a legal defense that may defeat [class certification], the district court cannot assume its validity but should make a threshold determination on the legal merits.”143

In other words, putative class members in the midst of class certification proceedings are only before the court in a liminal sense. Although a court cannot bind these parties prior to class certification, the existence of and proper forum for their claims is relevant to whether class treatment is appropriate. Here, the understanding of the putative class representative’s “private attorney general” interest in certifying a class may again be helpful. A putative class representative’s personal stake in the underlying litigation and “private attorney general” stake in certifying a class ensures that the enforceability of arbitration agreements as to putative class members may be presented in a justiciable fashion at the certification stage. The resulting legal determination is not an “advisory opinion,” because it concretely impacts whether the putative class representative indeed meets Rule 23 requirements and is entitled to certify a class. Class actions, including motions for class certification, “are prosecuted through representatives.”144 And that means that the rights of absent parties not directly before the court come under judicial scrutiny during the class certification process.

In any event, Jensen’s fatalistic approach to arbitration asymmetries is in no way required by Article III. A court may defer adjudicating a defendant’s arbitration rights as to absent class members until after a class has been certified. Or it may evaluate the enforceability of those rights in the context of a motion for class certification, insofar as the issue may be concretely presented by a putative class representative because it impacts her ability to certify a class. What is clear is that a court need not throw up its hands in the face of an arbitration asymmetry and, on that basis alone, decline to certify a class.


Although courts faced with arbitration asymmetries make reference to Rule 23 requirements, their rulings are motivated by underlying justiciability concerns. This Essay argues that class representatives have standing to argue that absent class members are not bound by arbitration agreements and that courts are not powerless to adjudicate such disputes. With those jurisdictional anxieties sorted, Rule 23 concerns about typicality and adequacy melt away and courts may freely certify classes when faced with an arbitration asymmetry. Indeed, as a policy matter, it is imperative that they do so. Otherwise, courts will continue to make just outcomes inaccessible to low-income litigants and allow corporate misconduct to go unchecked.

Thanks to the Forum editors for their wonderful edits and thoughtful suggestions. And thanks to the Yale Law Journal, generally, for supporting my fellowship year at Public Justice, without which this essay would not have been possible. I am deeply grateful to Karla Gilbride and Leah Nicholls for their helpful conversations and care in guiding me through this area of the law, as well as the entire Access to Justice team at Public Justice. Many thanks!