Over the last decade, the Supreme Court’s jurisprudence on the Federal Arbitration Act (“FAA”) has created an impression, at least in the popular press, that the Supreme Court is “pro-arbitration.” As The New York Times wrote in 2013, “[i]n the eight years since Chief Justice Roberts joined the Court, it has…made arbitration the favored way to resolve many disputes.” Adam Liptak, Corporations Find a Friend in the Supreme Court, N.Y. Times, May 4, 2013, at BU1. Based on this perception, many believe that the Court is more likely than not to enforce arbitration clauses and uphold arbitration awards on appeal. Although this view is not completely inaccurate, it obscures some important nuances of the Supreme Court’s jurisprudence over the past 10 years. Rather than operating on the basis of a single “pro-arbitration principle,” the Supreme Court’s FAA jurisprudence actually reveals two distinct principles at work. The first is what might be called the “Consent Principle,” which states that “[a]rbitration is strictly a matter of consent, and thus is a way to resolve those disputes —but only those disputes—that the parties have agreed to submit to arbitration.” Granite Rock Co. v. Int’l Broth. of Teamsters, 561 U.S. 287, 299 (2010) (citations and quotation marks omitted). This principle arises because of the fundamental fact that “an arbitrator derives his or her powers from the parties’ agreement to forgo the legal process and submit their disputes to private dispute resolution.” Stolt-Nielsen S.A. v. AnimalFeeds Int’l Corp., 559 U.S. 662, 682 (2010). Without consent, there can be no arbitration. The second principle is what the Supreme Court calls the “national policy favoring arbitration,” as embodied in the FAA itself. Buckeye Check Cashing, Inc. v. Cardegna, 546 U.S. 440, 443 (2006). This principle arises because “the FAA was designed to promote arbitration,” with the goal of “encourage[ing] efficient and speedy dispute resolution.” AT&T Mobility LLC v. Concepcion, 131 S. Ct. 1740, 1749 (2011) (citations omitted). So long as these two principles are both pointing in the same direction and thus operating in tandem — such as when the Supreme Court holds that state law prohibitions on agreements to arbitrate are preempted, see, e.g., Marmet Health Care Center, Inc. v. Brown, 132 S. Ct. 1201 (2012) (per curiam) —the Supreme Court’s decisions might justly be called “pro-arbitration.” Just as often, however, these two principles are opposed —that is, vindicating the agreement of the parties would limit the use of arbitration to resolve disputes or require overturning an arbitrator’s award. These latter cases cannot unambiguously be called “pro-arbitration.” Moreover, it is these latter cases that provide the most insight into the Supreme Court’s FAA jurisprudence, and teach the majority of the lessons that are of interest to those in the world of reinsurance arbitration. Questions of Arbitrability: Together the Consent Principle and the Policy in Favor of Arbitration Triumph Over State Law The category of cases in which the two principles most often operate in tandem are those that decide whether a particular dispute is arbitrable. This category of cases includes those in which the question is whether the arbitrability of the dispute is itself arbitrable. That is, when a litigant challenges the enforceability of an arbitration clause in a contract, who decides if the arbitration clause is enforceable —the court, or an arbitration panel? The Supreme Court’s jurisprudence has established some basic principles for resolving this question, based on § 4 of the FAA, which provides that a court must order arbitration after it is satisfied that “the making of the agreement to arbitrate” is not at issue. 9 U.S.C. § 4. Under that language, the Court had long ago held that courts must decide any challenges to the validity of the arbitration clause itself, whereas challenges Sidley Insurance & Reinsurance Law Report | 2016 11 to the validity of the contract as a whole must be arbitrated. Prima Paint Corp. v. Flood & Conklin Mfg. Co., 388 U.S. 395 (1967). In the last decade, several cases tested whether this principle applied also to state courts, particularly in the face of state laws seeking to assign the decision away from arbitrators. In the first, Buckeye Check Cashing, Inc. v. Cardegna, the Court reaffirmed that “regardless of whether the challenge is brought in federal or state court, a challenge to the validity of the contract as a whole and specifically to the arbitration clause, must go to the arbitrator.” 546 U.S. 440, 449 (2006). The second, Preston v. Ferrer, went yet further and held that this principle applies even in the face of a state law assigning exclusive original jurisdiction to a state administrative agency. 552 U.S. 346 (2008). Thus, the validity of a contract between an actor and his talent agent must be decided by the arbitrator, even though the California Talent Agencies Act assigned the decision to the state Labor Commissioner. A third case, Rent-a-Center v. Jackson, took another pro-arbitration step beyond Buckeye, holding that a challenge to the validity of an arbitration agreement itself may still go to an arbitrator if the parties’ contract specifies that this “gateway issue” is itself arbitrable (at least so long as a party does not challenge the validity of the clause stating that the gateway issue is arbitrable). 561 U.S. 63, 69-70 (2010). In all three cases the Court’s holding vindicated both the parties’ contracts — which provided for arbitration to resolve their disputes — and the federal policy in favor of arbitration over conflicting state laws. More recently, the Supreme Court has repeatedly asserted these principles against state courts refusing to recognize that the FAA preempts state law. In Marmet Health Care Center, Inc. v. Brown, 132 S. Ct. 1201 (2012), the West Virginia Supreme Court had held that an arbitration clause in a nursing home admittance agreement that was applicable to personal injury or wrongful death claims was unenforceable under state public policy. In a 9-0, per curiam opinion, the Supreme Court reversed, holding that the FAA preempted the state’s rule, again citing both the need to enforce the parties’ agreement and the “federal policy in favor of arbitral dispute resolution.” Id. at 1203 (citation omitted). Two other cases with 9-0, per curiam decisions similarly reversed state supreme court decisions refusing to enforce arbitration clauses on state law grounds. See Nitro-Lift Techs., L.L.C. v. Howard, 133 S. Ct. 500 (2012); KPMG LLP v. Cocchi, 132 S. Ct. 23 (2011). To be sure, questions of arbitrability do not always present situations in which the Consent Principle and the policy in favor of arbitration point in the same direction. For example, the Supreme Court’s most recent decision in DIRECTV v. Imbrugia, No. 14-462 (Dec. 14, 2015), represents yet another example of the Court overturning a state court’s invalidation of a binding arbitration provision in a consumer contract, this time including a class action waiver. Once again, the Court invoked the “federal policy favoring arbitration” to hold that the class arbitration waiver was effective despite contrary California law. Id. slip op. at 10. The Court acknowledged, however, that a contractual term making the arbitration clause ineffective where it was invalidated by “the law of your state” probably reflected the parties’ understanding that class arbitration waivers would be ineffective in California. Id. at 1-2. The policy in favor of arbitration nonetheless trumped that understanding and the Consent Principle. In another example, Granite Rock v. International Brotherhood of Teamsters, the underlying question was whether a collective-bargaining agreement (“CBA”) containing a no-strike provision was in force at the time of a strike by the union that was a party to the agreement. Because the CBA contained an arbitration provision, under § 4 of the FAA all questions arising under the contract would typically go to an arbitrator after a court had satisfied itself that a valid contract had been formed. See 9 U.S.C. § 4. In Granite Rock, however, the dispute was when a contract took effect, not precisely whether a contract had ever been formed, and the arbitrability of that question was unclear. 12 Is the Supreme Court “Pro-Arbitration”? In most questions of arbitrability presented to the Supreme Court, the two principles point in the same direction, requiring the Court to vindicate both the parties’ agreement to arbitrate and the FAA’s endorsement of arbitration over conflicting state laws. The labor union had argued, and the court of appeals had held, that the federal policy in favor of arbitration made the result clear. The court of appeals noted that the FAA’s “permissive policies in respect to arbitration counsel that any doubts concerning the scope of arbitral issues should be resolved in favor of arbitration.” Granite Rock, 561 U.S. at 298 (quotation marks and citation omitted). Thus, because the question of when a contract was formed was distinct from the question of whether it was formed —the only question the FAA reserved for judicial determination —the court of appeals sent the dispute to arbitration. The Supreme Court reversed, however, noting that the federal policy in favor of arbitration “cannot be divorced from the first principle that underscores all of our arbitration decisions: Arbitration is strictly a matter of consent.” Id. at 299. Moreover, if the contract was not in force on the date of the strike, the parties could not have consented to arbitrate a dispute arising from that strike. Accordingly, the date of contract formation was — like the question of whether a contract existed —a threshold question that required judicial determination. Id. In other words, the Consent Principle triumphed over the federal policy in favor of arbitration. Still, cases like Granite Rock are few and far between. In most questions of arbitrability presented to the Supreme Court, the two principles point in the same direction, requiring the Court to vindicate both the parties’ agreement to arbitrate and the FAA’s endorsement of arbitration over conflicting state laws. The Supreme Court’s arbitrability decisions have therefore gone a long way to creating the popular impression that the Supreme Court is “pro-arbitration.” Conflicts in Federal Law: Together the Consent Principle and the Policy in Favor of Arbitration Triumph Over Substantive Federal Rights Another class of cases in which the Consent Principle and the federal policy in favor of arbitration point the same way are those in which the FAA is pitted against another federal statute. These cases generally present the question of whether an agreement to arbitrate is trumped by the need to ensure the vindication of a substantive right granted by another federal statute. For example, in 14 Penn Plaza LLC v. Pyett, 556 U.S. 247 (2009), the Court faced the question of whether individual union members could consent to arbitrate their claims under the Age Discrimination in Employment Act (“ADEA”) through a collective bargaining agreement. To resolve the question, the Court first invoked the National Labor Relations Act’s provision that a union can bind its members through collective bargaining. Id. at 255. Because of that provision, the individual union members had effectively consented to arbitration, and “having made the bargain to arbitrate, the party should be held to it” absent congressional intent to preclude a waiver of the right to litigate in court. Id. Moreover, Congress’s decision to grant a substantive right, such as in the right not to be fired because of age in the ADEA, was insufficient to foreclose the operation of an arbitration clause, given the FAA’s endorsement of arbitration as an adequate forum for the vindication of such rights. Id. at 265-67. The Court thus wove together both the Consent Principle and the federal policy in favor of arbitration to justify its enforcement of the arbitration agreement. In CompuCredit v. Greenwood, 132 S. Ct. 665 (2012), the Court took these principles one step further to establish a kind of “clear statement rule” for Congress to override the FAA’s directive to enforce the right to arbitrate. See id. at 672 (“Had Congress meant to prohibit these very common provisions in the [Credit Repair Organizations Act], it would have done so in a manner less obtuse than what respondents suggest.”). In the absence of such clear congressional intent, the “liberal federal policy favoring arbitration agreements” and the need “to enforce agreements to arbitrate according to their terms” would ensure that arbitration agreements would be enforced. Id. at 669. Finally, in American Express Co. v. Italian Colors Restaurant, 133 S. Ct. 2304 (2013), the Court endorsed the same principles even when the practical effect of an arbitration Sidley Insurance & Reinsurance Law Report | 2016 13 …when combined, the Consent Principle and the federal policy in favor of arbitration are a powerful combination, resulting in another class of “pro-arbitration decisions.” agreement was to preclude the enforcement of federal antitrust law. Thus, even though Italian Colors Restaurant had demonstrated that the cost of individual arbitration was far more than any one individual litigant could reasonably expect to recover, and that no litigant would thus bring an individual arbitration proceeding, the Court enforced a class arbitration waiver. Id. Once again, the power of the parties’ consent to that provision, combined with the FAA’s strong policy in favor of enforcing such agreements, carried the day, even though it meant that the right of an antitrust victim to sue was completely vitiated. Accordingly, when combined, the Consent Principle and the federal policy in favor of arbitration are a powerful combination, resulting in another class of “pro-arbitration decisions.” Judicial Review of Arbitration Awards: The Two Principles Collide In a third class of cases, the Consent Principle and the policy in favor of arbitration conflict, and the outcome of the conflict has been mixed. The first case in this category involved an arbitration agreement in which the parties agreed on a broad standard of review, under which the Court could overturn an award “(i) where the arbitrator’s findings of facts are not supported by substantial evidence, or (ii) where the arbitrator’s conclusions of law are erroneous.” Hall Street Assoc., L.L.C. v. Mattel, Inc., 552 U.S. 576, 580 (2008). The Consent Principle obviously pointed in the direction of upholding this agreement. The Court, however, held that the FAA provided the exclusive grounds for overturning an arbitration decision through the FAA’s expedited confirmation procedures. 1 The Court rejected Hall Street’s argument based on the Consent Principle, and held instead that the FAA’s provisions on overturning an arbitration decision should be read “as substantiating a national policy favoring arbitration with just the limited review needed to maintain arbitration’s essential virtue of resolving disputes straightaway.” Id. at 588. Moreover, “[a]ny other reading opens the door to the full-bore legal and evidentiary appeals that can render informal arbitration merely a prelude to a more cumbersome and time-consuming judicial review process, and bring arbitration theory to grief in post arbitration process.” Id. (citation, alteration, and quotation marks omitted). In other words, the FAA prevents contracting parties from agreeing to what the Court considered to be an inferior arbitration process. The policy in favor of arbitration thus trumped the parties’ agreement. The opposite principle prevailed in Stolt-Nielsen S.A. v. Animalfeeds Int’l Corp., 559 U.S. 662 (2010), which reviewed an arbitrator’s decision that an arbitration agreement allowed class arbitration. Two aspects of the agreement were relevant: first, the parties had agreed that the question of class arbitration should itself be submitted to an arbitration panel. Id. at 668. Second, they stipulated that their settlement agreement was silent on the question of class arbitration, meaning that “there’s been no agreement that has been reached on that issue.” Id. (citation omitted). The arbitration panel attempted to fill this vacuum by citing “a consensus among arbitrators that class arbitration is beneficial in a wide variety of settings.” Id. at 673. Based on this conclusion about what the Court called “public policy,” the panel allowed class arbitration. In this case, however, the Court held that a policy favoring arbitration could not trump the Consent Principle, and that “[t]he panel’s conclusion is fundamentally at war with the foundational FAA principle that arbitration is a matter of consent.” Id. at 684. Because the parties had stipulated that their agreement was silent on the question of class arbitration, they could not have agreed to class arbitration, and no policy in favor of arbitration could overcome that lack of consent. Moreover, the arbitrator could not 1 The FAA specifies that an award may be vacated through expedited judicial review only where the award was procured by “corruption, fraud, or undue means,” where there was “evident partiality or corruption” in the arbitrators, where the arbitrators are guilty of misconduct in refusing to hear evidence, or where the arbitrators exceeded their powers or failed to issue an award on the subject matter submitted to them. See 9 U.S.C. § 10(a). The FAA also lists a limited number of circumstances in which a court can modify an arbitration award. See 9 U.S.C. § 11. 14 Is the Supreme Court “Pro-Arbitration”? The Court’s opinion therefore clarified the ruling in Stolt-Nielsen: the problem there was not that the arbitrator’s decision lacked a sufficient basis in the contract, but instead that it lacked “any contractual basis.” independently reach a conclusion based on “public policy” in the absence of some foundation in the parties’ contractual agreement. Ironically, this conclusion led the Court to ignore the other aspect of the parties’ agreement, which had been to let the panel decide class arbitrability. The Court’s decision usurped this power by holding that the only reasonable decision the panel could make was that there was no class arbitration. Id. at 677. Nonetheless, on the whole Stolt-Nielsen represents a vindication of the Consent Principle over the policy in favor of arbitration. Moreover, that principle led the Court in this case to a decision one might fairly describe as “anti-arbitration.” The Court’s decisions in Hall Street Associates and Stolt-Nielsen are hard to square. In the first case, the Court held that the FAA limits the parties’ ability to consent to particular arbitration procedures. In the second, the Court held that the FAA makes consent the “foundational principle” of arbitration. The third case in this category represents a sort of compromise between the two positions. In Oxford Health Plans LLC v. Sutter, 133 S. Ct. 2064 (2013), the Court again faced an arbitrator’s decision that an arbitration agreement allowed class arbitration. This time, however, the arbitrator had explicitly referenced the text of the parties’ agreement as the basis of his conclusion. The Court found that decision to be poorly reasoned —a concurring opinion noted that “[w]ere we reviewing the arbitrator’s interpretation of the contract de novo, we would have little trouble concluding that” it was incorrect, id. at 2071 (Alito, J., concurring) — but nonetheless upheld the arbitrator’s decision. The Court’s opinion therefore clarified the ruling in Stolt-Nielsen: the problem there was not that the arbitrator’s decision lacked a sufficient basis in the contract, but instead that it lacked “any contractual basis.” Id. at 2069 (majority opinion). In other words, so long as an arbitrator claims to be basing her decision on the parties’ agreement — even if that claim is debatable —the Court will uphold the decision. Id. at 2070 (“So long as the arbitrator was ‘arguably construing’ the contract…a court may not correct his mistakes…”). One can read this outcome as an attempt to harmonize the Consent Principle and the policy in favor of arbitration. That is, an arbitrator must at least pay lip service to the parties’ agreement by advancing a rationale that is at least plausibly interpreting the parties’ agreement. Because of the policy favoring arbitration, however, a court will not look behind the veil to pick apart this rationale. Oxford Health Plans thus attempts to mediate the conflict between the two principles by establishing a framework for the review of arbitration decisions that accounts for both. Lessons for Reinsurance Arbitrations What lessons can those involved with reinsurance arbitration learn from this description of the Supreme Court’s cases? The first lesson is that it is too simplistic to say that the Court is “pro-arbitration.” When the question is whether a particular dispute is arbitrable, it is true that the Court will generally uphold an arbitration agreement against conflicting state law. Moreover, it is now well established that the need to provide a forum for the effective vindication of a substantive federal right will only rarely justify a refusal to enforce an arbitration agreement. The basis for such decisions is not, however, a simple “pro-arbitration” orientation. Instead, such decisions are based on a dual adherence to both the federal policy favoring arbitration, as embodied in the FAA, and the basic principle that arbitration is a matter of consent. Where these two principles point in the same direction, parties can be confident that the law will uphold their arbitration agreements. When the two principles conflict, as they often do when the question is whether a court should uphold an arbitration decision, however, the outcome will not be so clear. In these cases, the policy in favor of arbitration counsels upholding an arbitrator’s decision. When doing so does violence to some aspect of the parties’ agreement, however, a Sidley Insurance & Reinsurance Law Report | 2016 15 Is the Supreme Court pro-arbitration? Yes, it is, but it is also proenforcement of the parties’ contractual agreements, and the two principles do not always point in the same direction. conflict will ensue. Parties should be aware that when the two principles are thus opposed, and in the absence of on point precedent, the outcome of the case will be more uncertain. The Court’s jurisprudence in this area also provides an obvious lesson for arbitrators, who should ensure that their decisions are grounded as much as possible in the contractual agreement of the parties. If an arbitrator fails to at least arguably interpret the contract, a general policy in favor of arbitration will not be sufficient to ensure that the decision holds up on appeal. Such decisions will be vulnerable to attack in court. This lesson is particularly important for reinsurance arbitrators, who often base decisions on custom and practice in the insurance industry. To be sure, such decisions are usually not as untethered as the “public policy” concerns of the arbitrator in Stolt-Nielsen. For one thing, reinsurance contracts themselves generally point to consideration of custom and practice when they include provisions relieving the arbitrators from following the strict rule of law and allowing them to consider “equitable” principles. Nonetheless, arbitrators would still be wise to attempt to tether their decisions to a particular contractual provision whenever possible. Is the Supreme Court pro-arbitration? Yes, it is, but it is also pro-enforcement of the parties’ contractual agreements, and the two principles do not always point in the same direction. A full understanding of the Supreme Court’s jurisprudence over the past decades requires an appreciation of both principles and the situations in which one or the other triumphs.