Parties to commercial disputes frequently encounter and litigate the threshold issue whether their dispute is within the scope of an arbitration clause and must therefore be arbitrated rather than litigated in court. Even before this threshold issue may be decided, however, there is a more fundamental question: Who gets to decide whether a dispute is subject to arbitration—a court, or the arbitration panel itself? American courts have struggled with the question for decades—and the struggle continues. Other jurisdictions have a much more straightforward approach.
American Approach
Enactment of the FAA and Presumption of Arbitrability
In the United States, the Federal Arbitration Act requires federal district courts to stay judicial proceedings or compel arbitration as long as the parties have a valid agreement to arbitrate that encompasses their dispute. 9 U.S.C. §§ 3, 4. However, the Act does not address whether, or when, the question of arbitrability should be decided by the arbitration panel rather than the court.
Passed by Congress in 1924, the Act reversed more than a century of American common law reflecting traditional English hostility to arbitration agreements, which were considered unenforceable attempts to divest the courts of their jurisdiction. See Scherk v. Alberto-Culver Co., 417 U.S. 506, 510-11 & n. 4 (1974). The Congressional report accompanying the Act explained that English courts, jealous of their jurisdiction, “refused to enforce specific agreements to arbitrate,” and that this “jealously survived for so long a period that the principle became firmly embedded in the English common law and was adopted with it by the American courts.” H.R. Rep. No. 96, 68th Cong., 1st Sess., 1-2 (1924). By reversing this historical hostility to arbitration agreements, Congress hoped to address “the costliness and delays of litigation.” Id. at 2. This was not, however, the main purpose of the Act; instead, “passage of the Act was motivated, first and foremost, by a congressional desire to enforce agreements into which the parties had entered.” Dean Witter Reynolds, Inc. v. Byrd, 470 U.S. 213, 220 (1985).
In giving effect to the Act, the United States Courts of Appeals gradually adopted a “federal policy that, when construing arbitration agreements, every doubt is to be resolved in favor of arbitration.” Dickinson v. Heinold Securities, Inc., 661 F.2d 638, 643 (7th Cir. 1981). Indeed, so widespread was the adoption of this “policy” that by 1981 the Seventh Circuit felt comfortable declaring it “axiomatic.” Id. On the basis of this lower court consensus, the Supreme Court itself ruled in 1983 that “as a matter of federal law, any doubts concerning the scope of arbitrable issues should be resolved in favor of arbitration.” Moses H. Cone Mem. Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 24-25 (1983)
First Options and the “Reverse Presumption”
The “presumption of arbitrability” seemed to suggest that the arbitration panel, rather than the court, would be empowered to determine the arbitrability of disputes unless the parties expressly agreed that arbitrability would be decided by a court. After all, if “doubts concerning the scope of arbitrable issues” were to be resolved in favor of arbitration, then any doubt regarding who was to decide arbitrability should likewise be resolved in favor of arbitration, at least absent a clear agreement to the contrary.
The Supreme Court rejected this conclusion in the 1995 case, First Options of Chicago, Inc. v. Kaplan, 514 U.S. 938 (1995). First Options involved a dispute between a stock-clearing firm and a couple who, through their wholly owned investment company, had become indebted to the stock-clearing firm in the 1987 stock market crash. Id. at 940. The stock-clearing firm initiated an arbitration against the couple on the basis of an agreement signed only by their investment company, which the stock-clearing firm had taken control of and subsequently liquidated. Id. Although the couple objected to the arbitration, the arbitration panel determined that it had the power to adjudicate the parties’ dispute, which it decided in the stock-clearing firm’s favor. Id. at 941. The federal district court confirmed the award, but the court of appeals reversed it, ruling that the courts were obligated to independently determine whether the dispute was arbitrable, and concluding that the parties’ dispute was not. Id.
The Supreme Court affirmed. It began by noting that the question of who decides arbitrability, like any other question of arbitrability, was ultimately a question of the parties’ intent: “Just as the arbitrability of the merits of a dispute depends upon whether the parties agreed to arbitrate that dispute, so the question ‘who has the primary power to decide arbitrability’ turns upon what the parties agreed about that matter.” Id. at 943. If the parties have agreed to submit the question of arbitrability to the arbitrators, then the arbitrators may decide that question, and their determination will be set aside only under the narrow parameters set forth in F.A.A. Section 10. Id. If the parties have not agreed to submit the question of arbitrability to the arbitrators, however, “then the court should decide that question just as it would decide any other question that the parties did not submit to arbitration, namely, independently.” Id. The Court explained that these two results “flow inexorably from the fact that arbitration is simply a matter of contract between the parties; it is a way to resolve those disputes—but only those disputes—that the parties have agreed to submit to arbitration.” Id.
What happens when it is unclear whether the parties have agreed to submit the question of arbitrability to the arbitrators? The presumption of arbitrability would seem to counsel that any doubt should be resolved in favor of finding an intent to arbitrate. Not so, said the Supreme Court: where there is silence or uncertainty regarding who should decide arbitrability, “the law reverses the presumption” and requires “clear and unmistakable evidence” that the parties agreed to have the arbitrators decide. Id. at 944-45 (internal quotation marks, alterations, and citations omitted). Recalling the primary purpose of the Federal Arbitration Act—enforcement of private agreements—the Court explained that reversing the presumption of arbitrability was necessary to ensure that parties were not forced to arbitrate matters they had not agreed to arbitrate. Id. at 945.
Howsam and “Questions of Arbitrability”
Although First Options purported to answer the “fairly simple” question of who decides questions of arbitrability, id. at 943, it raised another question that has proven equally problematic: what constitutes a “question of arbitrability”? The Supreme Court took up that question in Howsam v. Dean Witter Reynolds, Inc., 537 U.S. 79 (2002). The plaintiff in Howsam initiated arbitration before National Association of Securities Dealers (NASD). Id. at 81-82. The NASD submission agreement, which plaintiff executed, contained a blanket six-year limitations period on all arbitration claims. Id. The defendant filed suit in federal court seeking to enjoin the arbitration on the ground that the dispute was more than six years old and was therefore ineligible for arbitration under the NASD rules. Id. The district court dismissed the complaint, ruling that the applicability of the six-year limitations period was a matter for the arbitration panel rather than the court. Id. The Court of Appeals reversed, ruling that the applicability of the limitations period raised a question regarding the arbitrability of the dispute and noting—citing First Options—that questions of arbitrability are presumptively for the court. Id.
The Supreme Court reversed. Recognizing the holding of First Options, the Court noted that “one might call any potentially dispositive gateway question a ‘question of arbitrability,’” since its resolution would determine “whether the underlying controversy will proceed to arbitration.” Id. at 83. The Court explained that in the context of determining who decides arbitrability, “the phrase ‘questions of arbitrability’ has a far more limited scope.” Id. Questions of arbitrability are raised—and must be resolved by the court, rather than the arbitrator—“in the kind of narrow circumstances where contracting parties would likely have expected a court to have decided the gateway matter, where they are not likely to have thought that they had agreed that an arbitrator would do so, and consequently, where reference of the gateway dispute to the court avoids the risk of forcing parties to arbitrate a matter that they may well not have agreed to arbitrate.” Id. at 83-84. As examples of such questions of arbitrability, the Court identified “dispute[s] about whether the parties are bound by a given arbitration clause” and “disagreement[s] about whether an arbitration clause in a concededly binding contract applies to a particular type of controversy.” Id. at 84.
By contrast, the Supreme Court held that questions of arbitrability are not raised in “circumstances where parties would likely expect that an arbitrator would decide the gateway matter.” Id. As examples of such “procedural” questions, the Court identified disputes regarding whether prerequisites or conditions precedent to arbitration had been satisfied, as well as “allegations of waiver, delay, or a like defense to arbitrability.” Id. at 84-85 (internal quotation marks and citations omitted). Because the NASD six-year limitation at issue in Howsam was effectively a defense to arbitrability, it constituted a “procedural” issue that was “presumptively for the arbitrator, not for the judge.” Id. at 85.
Howsam’s framework has proven difficult to apply in practice. In Rent-A-Center West, Inc. v. Jackson, for example, the plaintiff sued his former employer for employment discrimination, and the defendant responded by seeking to compel arbitration pursuant to a written arbitration agreement that expressly delegated to the arbitrator the “exclusive authority to resolve any dispute relating to the interpretation, applicability, enforceability or formation of” the arbitration agreement itself. 561 U.S. 63, 65-66 (2010). The plaintiff opposed arbitration on the ground that the arbitration agreement was unconscionable under controlling state law. Id. at 66. A split court ruled that the plaintiff’s unconscionability argument was for the arbitrator, not the courts, since it was directed at the arbitration agreement as a whole and not at the specific provision—which the court characterized as an antecedent, severable agreement—vesting the arbitrator with authority to adjudicate challenges to enforceability of the agreement as a whole. Id. at 70-73. Characterizing the entire arbitration agreement as a single facet of a broader employment agreement, the dissent argued that under First Options the plaintiff’s unconscionability argument was a question of arbitrability requiring judicial resolution. Id. at 80-81 (Stevens, J., dissenting).
Rent-A-Center demonstrates continued uncertainty regarding how courts are to identify questions of arbitrability. The same uncertainty was raised in the Second Circuit’s 2014 decision in NASDAQ Group, Inc. v. UBS Securities, LLC, 770 F.3d 1010 (2d Cir. 2014). There, a broker-dealer sought to arbitrate claims against the NASDAQ exchange related to the exchange’s alleged mis-handling of the Facebook IPO. Id. at 1016-17. Although the parties had an agreement containing a broad arbitration clause that would otherwise have constituted “clear and unmistakable” evidence of an intent to vest the arbitration panel with authority to determine arbitrability, the clause was expressly made subject to a set of NASDAQ rules that arguably immunized the exchange from liability for the sorts of claims asserted by the broker-dealer. Id. at 1031-32. Rather than construing the NASDAQ rules as raising the sort of defense to liability that Howsam suggested would be subject to resolution by the arbitrator, the Second Circuit ruled that reference to the rules in the arbitration clause itself raised an ambiguity regarding whether the parties had in fact intended to vest the arbitrator with authority to determine the arbitrability of the specific claims asserted. Id.
In short, while First Options purported to answer the question of who decides arbitrability, it appears that the focus has simply shifted from the question “who decides” to the question “what is being decided.” Despite the Supreme Court’s efforts to resolve that issue in Howsam, it is clear that confusion and uncertainty remains.
Comparative Perspectives
The struggle to determine who decides questions of arbitrability is largely an American phenomenon, the result of the Federal Arbitration Act’s silence on the issue and the American courts’ subsequent need to develop rules of decision in light of perceived Congressional intent and fundamental principles of contract law. Other jurisdictions have managed to avoid the lasting uncertainty catalogued above by codifying the rules of decision.
In France, the Code of Civil Procedure expressly vests arbitration tribunals with “exclusive jurisdiction” to determine their jurisdiction. Code of Civil Procedure, Art. 1465. The Code also provides that the existence of an arbitration agreement divests the courts of jurisdiction entirely, except where the arbitration panel “has not yet been seized of the dispute” and the arbitration agreement is “manifestly void or manifestly not applicable”—an exception that is strictly interpreted. Id. Art. 1448, para. 1; see also, e.g., Fouchard, “La coopération du président du tribunal de grande instance à l’arbitrage,” Rev. Arb. 1985, at 27. Application of this principle of “compétence-compétence”—jurisdiction to determine jurisdiction—means that even where the arbitration tribunal’s jurisdiction is in question, the arbitration tribunal itself enjoys “chronological priority” to decide the issue and the courts remain divested of jurisdiction unless the parties mutually consent to judicial intervention. See Code of Civil Procedure, Art. 1448, para. 2 (court “may not decline jurisdiction on its own motion,” giving parties the ability to jointly consent to judicial intervention); see generally E. Gaillard, “L’effet négatif de la compétence-compétence,” in Mélanges J.-F. Poudret, 1999, at 387.
Whereas the American approach to the question of who determines arbitrability is directed primarily at vindicating the parties’ contract, the French approach places a greater premium on preventing dilatory tactics and encouraging the centralized and efficient resolution of all disputes surrounding the subject of the arbitration. See C. Seraglini, J. Ortscheidt, Droit de l’arbitrage interne et international, 2013, Lextenso eds. at 679. This does not mean, however, that the courts have no role—it simply means that whereas American courts exercise chronological precedence, acting as a gate-keeper to arbitration, the French courts act as a back-stop, exercising a limited review of the arbitration panel’s decision—including its decision regarding arbitrability—once that decision has been delivered. See Code of Civil Procedure, Art. 1492, para. 1 (allowing a court to “set aside” domestic award where the “arbitral tribunal wrongly upheld or declined jurisdiction”); id. Art. 1520, para. 1 (same with respect to international awards).
English law strikes something of a balance between the American and French approaches. Like the French Code, the English Arbitration Act expressly empowers an arbitration panel to “rule on its own substantive jurisdiction,” including “what matters have been submitted to arbitration in accordance with the arbitration agreement.” Arbitration Act (1996), § 30(1). Unlike the French approach, however, the English courts are not divested of jurisdiction, id. § 9, and may be called on to make a determination as to the jurisdiction of the arbitral tribunal (and hence the arbitrability of the parties’ dispute)—but only when all parties agree to seek such a ruling, or when the tribunal itself allows a party to do so, id. § 32. Otherwise, a party must await issuance of an award before challenging the arbitration panel’s arbitrability determination. See id. § 67(1).
Conclusion
Given the continued uncertainty regarding how to determine who should decide “gateway” questions of arbitrability in the United States, parties wishing to ensure resolution of such questions by a specific decision-maker—whether the court or the arbitrator—should spell out their preference as clearly as possible. Conversely, parties seeking to challenge a decision-maker’s authority to decide arbitrability should be attentive to circumstances that might implicate unresolved aspect of this vexing issue. Parties who seek to avoid litigation over this threshold question—and who are comfortable with having an arbitration panel determine its own jurisdiction in the first instance—may wish to consider contracting for arbitration in France or the United Kingdom, where judicial involvement is generally delayed until completion of the arbitration itself.