In Schein v. Archer and White, 139 S.Ct. 524 (2019), the U.S. Supreme Court addressed a narrow aspect of a perennial question that arises in arbitration: Who, as between courts and arbitrators, should resolve objections to arbitrability made at the outset of an arbitration proceeding? Before discussing Schein, it is worth explaining the nature of that question.

Questions of Arbitrability

Sometimes parties disagree about whether a particular dispute properly belongs in arbitration. This disagreement has been characterized by U.S. courts as one about “arbitrability.” A party might assert that a dispute is not arbitrable on any number of grounds: the arbitration clause does not cover the dispute; a condition precedent to arbitration (e.g., mediation) was not met; the contract is invalid on grounds of illegality. The question arises as to who, as between courts and arbitrators, should resolve such objections at the front-end of the process. Two important considerations underlie the “who decides” question: the legitimacy of the arbitration process and its viability.

The legitimacy of the arbitration process rests on the consent of the parties to an arbitration agreement, since it is from that agreement that arbitrators derive their authority to resolve the merits of a dispute. An objection that a dispute is not arbitrable is necessarily a challenge to arbitral authority and, therefore, to the legitimacy of an arbitral tribunal resolving that dispute. Since an arbitrability objection places at issue the legitimacy of an arbitral tribunal to resolve the merits of a dispute, it is reasonable to inquire as to the legitimate basis for an arbitral tribunal, rather than a court, to resolve that objection.

The viability of the arbitration process rests in part on its ability to function without undue court involvement. There is an inverse relationship between the degree of court involvement in the arbitration process and the attractiveness of that process as a method of dispute resolution; the greater the court involvement in resolving arbitrability objections, the less appealing the arbitration process.

The bearing of these two considerations—legitimacy and viability—on the “who decides” question can be illustrated by examining the two ends of the spectrum of potential solutions to that question.

At one extreme, if courts had exclusive authority to resolve in advance all arbitrability objections, cases would inevitably get swept into the courts that properly belong in arbitration. While such court involvement might bestow on arbitration legitimacy—any and all arbitrability objections would have been scrupulously addressed by the courts before an arbitration proceeding even gets going—the effect would be to undermine the viability of the arbitral process. After all, who would want to arbitrate if an arbitrability objection triggered costly and time-consuming court proceedings to determine whether a case gets into arbitration in the first place? Moreover, if all arbitrability objections had to be resolved in advance by courts, an inevitable result would be tactical challenges designed to delay or derail the arbitration process.

At the other extreme, if it fell to arbitrators to all resolve arbitrability questions, court involvement in the arbitration process would diminish, but cases would inevitably be swept into arbitration that do not legitimately belong there. Take, for example, a case where A forged B's signature on an arbitration agreement. Imagine that case was allowed to skate into arbitration based simply on A's bare assertion that it had agreed to arbitrate with B, on the basis that the arbitral tribunal alone was authorized to determine whether that agreement was forged. Since a tribunal's authority is derived from the consent of the parties, in such a case, the tribunal would lack the authority to make the precise determination it is asked to make (is the arbitration agreement forged?) because a forged agreement could not supply that authority.

Much of the law on the “who decides” question can be viewed through the lens of how it balances the competing considerations of legitimacy and viability.

In many countries, the allocation of authority between courts and arbitrators has been addressed through the adoption by statute of the doctrine of “competence-competence,” which provides that an arbitrator has the jurisdiction to decide on her jurisdiction. For example, Article 16 of the UNCITRAL Model Law, which has been adopted many countries, provides that “[t]he arbitral tribunal may rule on its own jurisdiction, including any objections with respect to the existence or validity of the arbitration agreement.”

In the United States, the allocation of the authority between courts and arbitrators is not dealt with by statute; the “who decides” question is not explicitly addressed in the Federal Arbitration Act (FAA). Rather, it is the subject of a series of decisions by the Supreme Court that has given rise to two doctrines allocating authority between arbitrators and courts: the separability doctrine and the delegation doctrine.

The Separability Doctrine

The separability doctrine is a legal fiction that provides that when parties enter into a contract containing an arbitration clause, they enter into two separate agreements: the underlying contract (without the arbitration clause) and the arbitration clause itself. Without the separability doctrine, if arbitrators decide that a contract is invalid, it necessarily follows that any arbitration clause contained within it is likewise invalid, which would, in effect, vitiate the arbitrators' authority to have made that precise invalidity determination in the first place. The separability doctrine avoids this outcome by holding that an arbitration clause survives any invalidity finding as to the underlying contract.

In Prima Paint v. Flood & Conklin Mfg., 388 U.S. 395 (1967), the Supreme Court held that the separability doctrine also has implications for the “who decides” question. The court held that challenges to the validity of an underlying contract should be resolved by arbitrators on the ground that, even if such challenges were successful, an arbitrator's authority under the arbitration clause would be unaffected. By contrast, challenges to the validity of the arbitration clause itself should be resolved by courts, since any such challenge places in issue an arbitrator's authority to resolve the merits of any dispute on the basis of such clause. The court justified this approach, in part, by relying upon the “congressional purpose that the arbitration procedure, when selected by the parties, be speedy and not subject to delay and obstruction in the courts”—something central to the viability of the arbitration process.

The Delegation Doctrine

In Howsam v. Dean Witter Reynolds, 123 S. Ct. 588 (2002), the court, invoking the metaphor of a “gateway” to arbitration, distinguished “gateway questions” to be resolved by the courts from those to be resolved arbitrators. It stated that gateway matters that raise “'question[s] of arbitrability' are for a court to decide.” These involve those “narrow circumstance[s] where contracting parties would likely have expected a court to have decided the gateway matter …,” which involve such issues as the validity (but not the formation) of the arbitration agreement or its scope. By contrast, gateway matters that go to arbitrators are those “where parties would likely expect that an arbitrator would decide the gateway matter.” These include “'procedural' questions which grow out of the dispute and bear on its final disposition” and defenses of “waiver, delay, or a like defense to arbitrability.”

The delegation doctrine is best understood as an exception to the general rule articulated by the court that gateway questions raising issues of arbitrability are presumptively for courts to decide: there is an exception to this general rule when the parties clearly and unmistakably agree to delegate arbitrability questions to the arbitrators. First Options v. Kaplan, 514 U.S. 938 (1995). Some courts have held that parties can express such clear and unmistakable intent by adopting arbitration rules that delegate the authority to resolve such questions to arbitrators. See, e.g., Contec v. Remote Solution, 398 F.3d 205 (2d Cir. 2005) (AAA Rules); Shaw Group v. Triplefine, Int'l, 322 F.3d 115, 121 (2d Cir. 2003) (ICC Rules).

The 'Wholly Groundless' Exception

The Supreme Court's recent decision in Schein resolved a circuit split on whether—in cases where the delegation doctrine applied (i.e., there was clear and unmistakable evidence that the parties had delegated arbitrability questions to the arbitrators)—courts could nonetheless resolve such questions where the argument in favor of the arbitrability of the dispute was “wholly groundless.” In Schein, Justice Kavanaugh, writing for the court, rejected the “wholly groundless” exception, stating “[w]hen the parties' contract delegates the arbitrability question to an arbitrator, the courts must respect the parties' decision as embodied in the contract.”

Schein concerned an agreement between Archer and White (Archer), on the one hand, and Pelton and Crane, on the other, to arbitrate their disputes “except for actions seeking injunctive relief and disputes related to trademarks, trade secrets or other intellectual property of [Schein].” The parties agreed to apply “the arbitration rules of the American Arbitration Association.”

Archer sued Schein, the successor to Pelton and Crane, in the District Court for the Eastern District of Texas seeking, among other things, injunctive relief. Schein moved to compel arbitration. Archer argued that the case, in part, was not arbitrable based on the arbitration clause's exclusion of actions seeking injunctive relief. Schein responded that Archer's objection should be decided by the arbitrator not the court on the ground that the AAA Rules contained a delegation provision. The Fifth Circuit affirmed the District Court's denial of Schein's motion to compel on the ground that Schein's argument in favor of arbitration was “wholly groundless.”

The Supreme Court vacated the Fifth Circuit decision and remanded the case. Central to Justice Kavanaugh's reasoning was a textual argument: the FAA “does not contain a 'wholly groundless' exception, and we are not at liberty to rewrite the statute … .” This reasoning is curious. As I noted above, the “wholly groundless” doctrine operates as an exception to the delegation doctrine, which itself is an exception to a general rule that gateway arbitrability questions are presumptively for courts to decide. But both that general rule and the delegation exception are the product of judicial decisions, not the FAA. In these circumstances, it is hardly a surprise that the FAA makes no mention of a “wholly groundless” exception; the general rule and the delegation exception (to which the “wholly groundless” doctrine would itself be an exception) are not mentioned in the FAA either.

Justice Kavanaugh also justified his decision by relying on considerations that implicate the viability of the arbitration process, i.e., avoiding undue court involvement. Archer had argued that it was a waste of time and money for courts to send a “wholly groundless” arbitrability question to an arbitrator, since the arbitrator would inevitably find the dispute not to be arbitrable. Justice Kavanaugh rejected this argument, in part, on the ground that having a wholly groundless exception “would inevitably spark collateral litigation (with briefing, argument, and opinion writing) over whether a seemingly unmeritorious exception to arbitration is wholly groundless, as opposed to groundless. We see no reason to create such a time-consuming sideshow” (emphasis in original).

It is important to stress that, in rejecting the “wholly groundless” exception, Justice Kavanaugh was careful to note that the fact that arbitrability objections could be resolved by arbitrators at the “front-end” of the arbitration process, did not negate the authority of the courts under the FAA to review the awards of arbitrators at the “back-end” to determine whether the arbitrators had exceeded their powers.

In the opinion of this author, the court reached the correct decision in Schein. In Schein, there was no dispute that the parties had entered into a valid arbitration agreement. Thus, the requisite consent that is essential to arbitral legitimacy was present. To be sure, the parties disagreed about whether their arbitration agreement covered their particular dispute. But if Schein were correct that the parties, by agreeing to the AAA Rules, had delegated arbitrability questions to the arbitrators, it follows the arbitrators, not the courts, must decide whether the arbitration clause applies to the particular dispute between the parties. This minimizes court involvement and thus promotes the viability of the arbitration process.

It bears noting that Justice Kavanaugh made it clear that the court expressed no view on whether an arbitration clause selecting the AAA Rules had the effect of delegating arbitrability questions to arbitrators. Justice Kavanaugh suggested that that is the first question that may be addressed by the Fifth Circuit upon remand.

John Fellas is a partner at Hughes Hubbard & Reed in New York.