Argument preview: Class arbitration agreements and the sounds of silence
on Mar 22, 2013 at 4:42 pm
In its 2010 decision in Stolt-Nielsen S.A. v. AnimalFeeds Int’l Corp., the Court held that arbitrators exceed their authority under the Federal Arbitration Act (FAA) when they allow class arbitration absent a basis in the arbitration agreement for concluding that the parties affirmatively agreed to so provide. Thus, even in a case in which a class action would have been available in a judicial proceeding, and in which the arbitration agreement does not include an express waiver of class arbitration [the substantive validity of which the Court is currently considering in the Italian Colors case], the mere fact that the parties agreed to submit a particular dispute to arbitration was not sufficient proof, standing alone, of their intent to allow class-wide arbitration. The question that Justice Alito reserved in footnote 10 of Stolt-Nielsen, and which has divided the lower courts, is what would suffice to demonstrate the parties’ intent to allow class arbitration: Must the agreement to arbitrate expressly recognize the possibility of class-wide adjudication, or could such intent be inferred from the text of the underlying contract and/or parol evidence?
That is the question on which the Justices granted certiorari in Oxford Health Plans LLC v. Sutter – in which they are scheduled to hear oral argument next Monday. But lest it seem as if Sutter merely raises a minor question of Stolt-Nielsen’s scope, the reality is that very few arbitration agreements include express provisions authorizing class-wide adjudication. Thus, what is ultimately at stake in Sutter is the future of class arbitration – and whether it will ever be permissible under the FAA without express contractual language so providing.
I. Background
The respondent in this case, Dr. John Ivan Sutter, is a primary care physician who, through a 1998 contract, agreed to provide services to members of petitioner Oxford Health Plan LLC’s managed care network in exchange for compensation at predetermined reimbursement rates. The 1998 contract also included an arbitration clause, which provided that:
No civil action concerning any dispute arising under this Agreement shall be instituted before any court, and all such disputes shall be submitted to final and binding arbitration in New Jersey, pursuant to the Rules of the American Arbitration Association with one arbitrator.
In 2002, Sutter accused Oxford of a series of improper denials, underpayments, and delays in reimbursing his and other physicians’ claims, and filed a class action against Oxford and other health insurers in New Jersey state court. Oxford moved to compel arbitration, and the state court agreed. Before the arbitrator, the parties raised the question of whether the arbitration clause authorized class-wide arbitration. The arbitrator ruled that it does, characterizing the clause as “much broader even than the usual broad arbitration clause;” it was “unique in [his] experience and seem[ed] to be drafted to be as broad as can be.” As the Third Circuit subsequently explained, “The arbitrator thus determined that the clause’s first phrase . . . embraces all conceivable court actions, including class actions. Because the clause’s second phrase sends ‘all such disputes’ to arbitration, he reasoned that class disputes must also be arbitrated.” The arbitrator subsequently certified a class, which Oxford unsuccessfully challenged in an action brought under section 10 of the FAA in district court.
After the Supreme Court’s decision in Stolt-Nielsen, Oxford renewed its challenge to the arbitrator’s approval of class-wide arbitration. Notwithstanding Stolt-Nielsen, though, the arbitrator reaffirmed his original reasoning, explaining that his decision was not simply based upon the absence of a class waiver (which Stolt-Nielsen forbids), but on the (in his view) unique text of the arbitration clause, which expressly authorized arbitration of “[a]ll such disputes.” Because class actions are clearly within the ambit of “such disputes,” the arbitrator concluded, there was enough evidence of the parties’ agreement to allow class arbitration under Stolt-Nielsen.
Oxford again brought suit to challenge the arbitrator’s decision under section 10 of the FAA, and again lost in the lower courts. In affirming the district court’s denial of Oxford’s motion (and its confirmation of the arbitrator’s decision), the Third Circuit stressed the deference due to an arbitrator’s factual and legal analysis under the FAA. After all, as Judge Fuentes explained, the relevant provision of the FAA only authorizes vacatur of arbitral awards “where the arbitrators exceeded their powers.” To be sure, under Stolt-Nielsen, “an arbitrator lacks the power to order class arbitration unless there is a contractual basis for concluding that the parties agreed to that procedure.” Nevertheless, the court concluded, even under Stolt-Nielsen, an arbitrator’s “decision to order class arbitration is within his authority so long as it stands on a contractual basis.” In other words, because the arbitral award could only be set aside if the arbitrator exceeded his authority, the fact that the arbitrator grounded his class arbitration ruling in the text of the contract, and not just in inferences from its existence, disposed of Oxford’s challenge under the FAA – regardless of whether the arbitrator’s interpretation was correct.
Thus, the Third Circuit joined the First and Second Circuits, which had also upheld class arbitration determinations by arbitrators after Stolt-Nielsen despite the absence of express contractual language to that effect. In contrast, cases in both the Fifth and Eleventh Circuits had held, more narrowly, that the contract must evince far more specific and compelling indicia of the parties’ attempt to agree to class arbitration. Certiorari was therefore impelled by the need to settle the issue Stolt-Nielsen had reserved.
II. The briefing
Not surprisingly, Oxford’s brief paints the Third Circuit’s efforts to distinguish Stolt-Nielsen as an exercise in splitting hairs. The purpose of Stolt-Nielsen, it argues, is to preclude class arbitration absent at least some manifestation of the parties’ intent to authorize it. And Stolt-Nielsen itself rejected the argument that such intent could be “inferred” from the mere agreement to submit disputes to arbitration. Thus, Oxford reasons, the arbitrator erred in suggesting that the specific language of the arbitration clause here at issue was somehow more explicit with regard to class arbitration than the arbitration clause at issue in Stolt-Nielsen. As in Stolt-Nielsen, nothing beyond the agreement to arbitrate suggested any intent to allow class arbitration, as such. Although “Stolt-Nielsen arguably leaves open the possibility that affirmative agreement to class arbitration might sometimes be shown by evidence beyond the text of an arbitration clause,” that is not in fact what the arbitrator found here. Instead, the arbitrator relied on the broad language of the arbitration clause – language that, Oxford suggests, is “wholly unremarkable,” and does little more than cement the parties’ agreement to arbitrate, i.e., exactly what Stolt-Nielsen held to be insufficient indicia of the parties’ intent.
In response, Sutter focuses on the deference due to arbitrators under the FAA. Even if the arbitrator’s reading of the contractual language isn’t the most obvious one, he argues, there is no doubt that, unlike in Stolt-Nielsen, the arbitrator based his decision on the text of the arbitration agreement, and not on its mere existence. Thus, nearly three-quarters of Sutter’s merits brief is devoted to arguing that the arbitrator acted within the scope of his authority, and that his decision therefore cannot be set aside under section 10 of the FAA. To that end, Sutter relies upon the “Enterprise Wheel standard” for when arbitral opinions are based upon the terms of the underlying contract, i.e., whether “it draws its essence . . . from the agreement.” And although reasonable minds may differ as to the arbitrator’s motives in this case, under Enterprise Wheel, Sutter concludes, “a court can find that an arbitrator exceeded his authority in construing a contract only if the arbitrator’s decision is unambiguously based solely on considerations other than the agreement itself.” Whereas such a finding was at the heart of Stolt-Nielsen, here, Sutter suggests, there is clear indication that the arbitrator’s decision was based at least in part on the text of the agreement—and so the arbitrator is entitled to deference.
In reply, Oxford concedes that, in Stolt-Nielsen, unlike here the parties had stipulated to the lack of actual agreement to permit class arbitration. Nevertheless, Oxford concludes, “an arbitral award imposing class arbitration must be vacated as ultra vires unless it rests on a proper contractual basis. . . . And an arbitrator may not find such an agreement—as the arbitrator in this case did—based on nothing more than the agreement to arbitrate itself.”
III. Taking stock
On a blank slate, it would be easy to see the logic of Sutter’s argument for distinguishing Stolt-Nielsen. Under the plain terms of the relevant FAA provision, an arbitral award cannot be overturned so long as the arbitrator acts within his authority. And so long as the arbitrator’s determination that the parties agreed to allow class arbitration is reasonably based upon the text of the contract (as opposed to its mere existence, as in Stolt-Nielsen), there can be little question that he is, in fact, acting within his authority – even if his interpretation of the contract fails to persuade on the merits.
Ultimately, however, the central problem for Sutter is that the slate is far from clean. It can hardly be gainsaid that the Supreme Court in recent years has routinely evinced hostility to class arbitration – in Stolt-Nielsen itself, in AT&T Mobility LLC v. Concepcion, and, if the oral argument is any indication, in this Term’s Italian Colors case. And although the decision below (and Sutter’s arguments in support thereof) offer entirely plausible grounds for distinguishing Stolt-Nielsen, what this case may ultimately come down to is whether the Justices truly believe that class arbitration should never be permissible absent the most express evidence of the parties’ intent to allow it, or whether Stolt-Nielsen was instead a far narrower decision than most have thought, simply requiring that class arbitration findings actually be based upon the language – and not just the existence – of the parties’ agreement to arbitrate.
