By R. David Gallo, Law Clerk at the U.S. Court of International Trade
On January 18, 2017, the Second Circuit issued an opinion in the case of CBF Indústria de Gusa S/A et al. v. AMCI Holdings, Inc., 846 F.3d 35 (2d Cir. 2017).1 In Gusa, a party sought to enforce a foreign arbitral award against an entity that was not a signatory to the underlying arbitration agreement. Gusa is the latest in a line of Second Circuit cases to tackle this difficult issue. Gusa is also the latest case to invoke New York Convention (“Convention”) Article V(2)(a) and apply U.S. law – the law of the enforcing jurisdiction – to the question of whether a foreign arbitral award can be enforced against a non-signatory. The below discussion explores some of the reasons why this approach is untenable.
Sarhank and the Article V(2)(a) Defense to Enforcement
In Gusa, the Second Circuit cited its prior decision in Sarhank Group v. Oracle Corp. for the proposition that “general principles of [U.S.] contract law” govern whether a foreign arbitral award can be enforced against a non-signatory in a U.S. court. Sarhank concerned an Egyptian arbitral award, rendered against Oracle, despite the fact that a subsidiary of Oracle, and not Oracle itself, had been the signatory to the underlying arbitration agreement. In reversing the district court’s decision to enforce the award, the Second Circuit relied upon Article V(2)(a) of the Convention, which provides that recognition and enforcement of an arbitration award may be refused if “[t]he subject matter of the difference is not capable of settlement by arbitration under the law of [the enforcing] country.” The court reached Article V(2)(a) by framing the non-signatory inquiry as one about “arbitrability.” In other words, if I didn’t agree to arbitrate with you, then any dispute you and I have is not “capable of settlement by arbitration” or not “arbitrable.” But Article V(2)(a) and the concept of “arbitrability” are not generally understood in this manner by the international arbitration community. As Gary Born explains:
classic examples of non-arbitrable subjects include certain disputes concerning consumer claims; criminal offenses . . . The non-arbitrability doctrine rests on the notion that some matters so pervasively involve public rights, or interests of third parties, which are the subjects of uniquely governmental authority, that agreements to resolve such disputes by ‘private’ arbitration should not be given effect.2
In line with this understanding, commentators Garfinkle and Herlihy stated that, unlike other legal systems, “in U.S. case law an ‘arbitral’ matter often connotes a matter falling within the scope of the parties’ arbitration agreement in any given case. . . . In [Sarhank], the Second Circuit incorrectly grafts this [ ] concept of ‘arbitrability’ into Article V(2)(a) of the Convention . . .”3 Nevertheless, the Second Circuit has continued to invoke Article V(2)(a) to address the non-signatory issue, most recently in Gusa. See, e.g., VRG Linhas Aereas S.A. v. MatlinPatterson Global Opportunities Partners II L.P., 717 F.3d 322 (2d Cir. 2013).
An examination of the logic and structure of the Convention further reveals why applying Article V(2)(a) in non-signatory enforcement cases is questionable. The Convention defenses can be said to fall into two basic categories: those that address a defect in the arbitration or award and those that protect fundamental interests of the enforcing jurisdiction. The defenses in Article V(1) essentially contemplate circumstances that were problematic at the time and should have been addressed or avoided before reaching enforcement. Appropriately, the Article V(1) defenses generally require the enforcing jurisdiction to apply the same law or rules that were applicable in the first instance. To illustrate, if the arbitration agreement was not valid under the law to which it was subject (Art. V(1)(a)), the arbitration should not have proceeded. The enforcing jurisdiction can act as a last line of defense to right this wrong. Put differently, if the parties and arbitrators do everything “right”, then the Article V(1) defenses will likely not apply.
By contrast, the Article V(2) defenses contemplate awards which may have been flawlessly rendered, but which nevertheless conflict with fundamental interests of the enforcing jurisdiction. The drafters of the Convention provided a public policy exception, Article V(2)(b), to protect the mandatory principles of the enforcing country. Because the public policy exception looks to the law of the enforcing jurisdiction, there may be cases where the parties and the arbitral tribunal do everything right, but the award nevertheless violates the public policy of a jurisdiction where enforcement is later sought. Such cases are appropriately rare.
If Article V(2)(a) is understood to mean that courts do not have to enforce awards that concern “subjects of uniquely governmental authority”, then this defense takes on the characteristics of the Article V(2)(b) public policy exception. An award may be rendered after the parties and arbitrators did everything right, but enforcement may be refused in order to protect fundamental interests of the enforcing jurisdiction. This makes sense, given the nature of what is at stake for the enforcing jurisdiction.
Under the Sarhank approach, the Article V(2)(a) defense does not appear to concern any fundamental interests of the enforcing jurisdiction. Rather it concerns the scope of the arbitration agreement, which is to be resolved by reference to “general principles of [U.S.] contract law”, despite the fact that the arbitral tribunal would almost certainly have had no occasion to apply U.S. contract law to the question. If the tribunal pulled a non-signatory into an arbitration by correctly applying the law with the closest connection to the contract, or the place of incorporation, or the allegedly fraudulent behavior, or the arbitration, then why would the Convention provide for non-enforcement just because the non-signatory has assets in the U.S. and U.S. contract law leads to a different outcome on the non-signatory question? If U.S. courts believe that the drafters did in fact provide for non-enforcement under these circumstances, they should at least explain how this conclusion fits within the Convention as a whole.
In sum, under the Second Circuit’s interpretation, Article V(2)(a) does not resemble the Article V(1) defenses because it does not require that the parties or tribunal did anything “wrong” according the rules or law they were bound to follow. Nor does it resemble the Article V(2)(b) public policy exception, as it does not appear to protect any fundamental principles of the enforcement jurisdiction. One conclusion to be drawn is that the non-signatory question may be one of the many issues that are not to be litigated or re-litigated in enforcement proceedings. By framing the non-signatory question as one of Article V(2)(a) “arbitrability,” courts apply U.S. law to the inquiry and address the issue with fresh eyes, without regard to whether U.S. law has a logical connection to the question or whether the drafters of the Convention intended for enforcing jurisdictions to second-guess arbitral tribunals on this issue.
Perhaps a strong argument can be made that the enforcing jurisdiction does in fact have the strongest connection to the question of whether a non-signatory with assets within its borders can be held liable for a foreign arbitral award. Perhaps it is the best view that non-signatories must be protected from unjust enforcement of foreign arbitral awards. But if U.S. courts want to advance these views, they should do so while meeting the terms of the Convention head on. If the Second Circuit wants to protect non-signatories, there may be more defensible alternatives available. For example, in a case such as Gusa, where the non-signatory was neither a party to the arbitration nor named in the award, the court could likely bypass the Convention and determine whether it can enforce the award by reference to “legal principles concerning enforcement of awards or judgments under applicable state law.”4
As it stands, current jurisprudence rests largely on assuming that Article V(2)(a) is applicable to the question of “who,” or “what,” is reached by a particular agreement to arbitrate. This issue can and should be revisited by the Second Circuit – an important jurisdiction for the enforcement of foreign arbitral awards. In doing so, the Second Circuit can further its critical, stated objective of reducing “parochialism and uncertainty in[ ] international arbitration.”5
The opinions expressed herein are the author’s own and do not reflect the views of the U.S. Court of International Trade.
1 On March 2, 2017, the Second Circuit reversed and superseded its January 18 opinion on grounds not relevant to this discussion. CBF Indústria de Gusa S/A et al. v. AMCI Holdings, Inc., 850 F.3d 58, 75–76 (2d Cir. 2017).
2 Gary Born, International Commercial Arbitration at 768 (2009).
3 Barry H. Garfinkle & David Herlihy, Looking for Law in All the Wrong Places: The Second Circuit’s Decision in Sarhank Group v. Oracle Corporation, Mealey’s International Arbitration Report, June 2005.
4 Brief of the Association of the Bar of the City of New York as Amicus Curiae in Gusa, Case No. 15-cv-1133, Doc. No. 166 at 3.
5 Smith/Enron Cogeneration Ltd. v. Smith Cogeneration Int’l, Inc., 198 F.3d 88, 96 (2d Cir. 1999).