This Commentary begins, innocently, with a report of a new decision in the US Ninth Circuit Court of Appeals, about choice of law in the arbitrability realm. I realize you are watching lot of daytime TV these days. The Edge of Night, etc. So I will try to make it interesting. Settle in, remove your mask, stay awhile.
This story actually begins with the Outokumpu case. You remember Outokumpu, from last June. It is famous for a few reasons. One is that Justice Thomas wrote the opinion for a unanimous US Supreme Court. How often has that happened in 30 years? It’s also famous, in our world, because the Court held that the New York Convention doesn’t prevent a Court from compelling arbitration for or against a non-signatory based on equitable estoppel. What’s sauce for the FAA Chapter 1 domestic goose, held the Court, is also sauce for the FAA Chapter 2 international gander. Birds of a feather. No conflict, no problem.
Well, maybe a technical problem. Because the 11th Circuit US Court of Appeals, from whence Outokumpu emerged, was not the only such US Court – the 9th Circuit was another — to have held that the New York Convention not only required the courts of Contracting States to enforce arbitration agreements in writing within the coverage of the Convention upon the application of one of the parties, but also that the Convention by implication prohibited such courts from enforcing such agreements upon the application of a non-party who invokes estoppel principles in support of enforcement. So, in the 9th Circuit case that is the text for today’s sermon, the Supreme Court granted certiorari, and sent the case back to the 9th Circuit to figure out what happens next in light of Outokumpu. The 9th Circuit heard new arguments, and then held that, in a federal district court, federal common law, not state law, controls the question of whether non-signatories can enforce arbitration agreements based on equitable estoppel when federal statutory claims are asserted in the district court in the underlying dispute. And so the 9th Circuit held that the District Court did not err, in the first go-round, in denying the motion to compel arbitration. One judge dissented, elaborately, on the question of what law governs the equitable estoppel/arbitrability issue. The position taken was that the choice of law for the non-signatory arbitrability issue ought probably to be the law of India, and that a remand to the District Court to re-decide the motion to compel arbitration after considering anew the choice-of-law issue should have been the outcome. So here we are, with material for a Commentary. Setty v. Shrinivas Sugandhalaya LLP, 2020 WL 192820 (9th Cir. Jan. 20, 2021).
Where I would like to take you in this Commentary eventually, dear Readers, is to the question of what a US-seated arbitrator should do if confronted with a jurisdiction objection raising the equitable estoppel issue, in an international case between parties of non-US nationalities. That is a particular challenge because in the USA, maybe more than in other beehives of international arbitration activity, because the US jurisprudence on arbitrability mostly concerns how judges should act or refrain from acting. Translating that into guidance for arbitrators, for whom US arbitration law is lex arbitri, can be a major challenge.
So let’s first sort out the different views of the Panel majority and the dissent in the Setty case. That should qualify you to sit on the District Court in some reliably sunny place, say Phoenix or LA or San Diego. Even Seattle, if you like coffee and rain. But since you are, or aspire eventually to be, an international arbitrator sitting in such places from time to time, let’s eventually try to sort out what a case like Setty means for you.
Setty is about a family business in India selling incense. Due to a family feud, one branch of the family ended up with its incense business in Mumbai, the other in Bangalore. The only question for the 9th Circuit, after the post-Outokumpu remand from the Supreme Court, was what law applies to the question of using equitable estoppel as a basis for a non-signatory to enforce an arbitration clause against a signatory, to require the signatory to arbitrate not litigate its claims against the non-signatory. Is it the law of India, or federal common law? The arbitration clause is found in a “partnership deed”, and the family members who run the Mumbai operation are not parties to the deed. They are parties to a federal district court lawsuit brought against them by the Bangalore business, claiming trademark infringement under US federal trademark laws. Even though the partnership deed evidently was made in India between partners who were of Indian nationality and provided for arbitration governed by the law of India, the Ninth Circuit swiftly dismissed the partnership deed as irrelevant to the choice of law issue, reasoning that, under 9th Circuit precedent, what it called a “threshold issue” of arbitrability arising under FAA Chapter 2 and the New York Convention is governed by federal common law. The Panel cited a 2016 9th Circuit case holding that when it is for a court to decide whether a binding agreement to arbitrate exists in an FAA Chapter Two case, i.e. whether an agreement to arbitrate falling under the New York Convention was formed, that issue is determined based on federal common law. So the first problem with the Setty Panel majority’s reasoning is that it relies on a precedent that involved whether any arbitration agreement ever was made, not whether a non-party could enforce such an agreement that indisputably was made.
As the next and essentially final step in its analysis, the Panel majority in Setty relied on a 1986 9th Circuit decision in which it was held – in a domestic arbitration context – that when a party to an arbitration agreement seeks to arbitrate US federal statutory claims (there, under the federal securities laws’ anti-fraud provisions) against non-signatories, federal common law rather than state law governs the arbitrability issue. Appellant in Setty had contended this conclusion was incorrect, invoking a US Supreme Court decision in 2009, Arthur Andersen LLP v. Carlisle, 556 U.S. 624, where the Court held in a domestic FAA case that the relevant state contract law determines whether a non-signatory of the arbitration agreement may require its enforcement. But per the 9th Circuit majority in Setty, prior 9th Circuit decisions had only applied Arthur Andersen as the choice of law rule in cases involving state substantive law claims that came before the federal district court based on diversity of citizenship jurisdiction, and (by implication) not in cases raising federal statutory substantive law questions and no in cases involving arbitration itself as a federal question under FAA Chapter Two. The affirmative limitation of Arthur Andersen to a non-federal claims and domestic arbitration agreements is, for the 9th Circuit, evidently a new development in Setty. And whereas the federal common law version of equitable estoppel for purposes of arbitrability requires that the claims be “intertwined” with the contract containing the arbitration agreement, and the District Court had already found that there was no such intertwining, the Panel re-affirmed its earlier judgment that sustained the District Court’s denial of the motion to compel arbitration.
So, you may ask, tell us more about Arthur Andersen. OK. Simplified, the Supreme Court in Arthur Andersen held that when a party to a federal lawsuit seeks a stay of the action under FAA Section 3, claiming the dispute is arbitrable, and the arbitrability issue is a non-signatories issue (i.e. who is bound by a valid arbitration agreement) that issue is determined according to the ordinary, generally applicable principles of state contract law that apply to contracts generally. The litigation in an Ohio federal court underlying Arthur Andersen did not involve federal statutory causes of action, and this was a domestic FAA case where diversity of citizenship provided the needed independent basis for federal subject-matter jurisdiction. But nothing in the Court’s opinion states expressly that a choice of federal arbitrability law would have been required had there been federal statutory claims. And nothing said expressly in Arthur Andersen limited its holding to FAA Section 3 motions (to stay litigation pending arbitration) affecting purely domestic arbitration agreements.
Not only your Commentator, but also the dissenting 9th Circuit panelist in Setty, thinks the Setty panel majority’s attempt to pigeon-hole Arthur Andersen to domestic arbitration of non-federal substantive claims was not justified. And not only your Commentator, but also the dissenting 9th Circuit panelist in Setty, believes the Supreme Court would transpose the notion of “state law” in Arthur Andersen in the context of a dispute between non-US parties to mean the law governing the enforceability of the arbitration agreement by or against a non-party as determined by the applicable choice-of-law principles in the forum where enforcement is sought. So, to conclude the first section of this Commentary, I would say to stay very tuned for a petition for writ of certiorari to the US Supreme Court in the Setty case. An elaboration of how the choice-of-law principle of Arthur Andersen applies in the setting of international arbitration and/or in the setting of underlying US federal statutory substantive law claims may attract the interest of the Court.
Ok so now let’s consider how all of this arcane US law on arbitrability affects you as an arbitrator if you must decide arbitrability. Suppose the facts of the Setty case (Mumbaians v. Bangaloreans, incensed), and that the partnership deed identifies no seat of arbitration but does provide for ICC arbitration. The parties make submissions to the ICC Court concerning the place of arbitration. The Claimant prefers Mumbai and the Respondents prefer Seattle (where the federal district court case leading to Setty was heard, after a transfer from Alabama). The ICC Court picks Seattle, and the Parties eventually, by agreement, pick YOU!
When you must rule on an objection to the jurisdiction of the Tribunal, with the objecting parties urging that the non-signatories cannot rely on equitable estopped principles to enforce the arbitration agreement, what choice of law rules shall you apply? The answer in an ICC arbitration may be less than clear. Article 21 of the ICC Rules states that with respect to the law applicable to the merits of the dispute, in the absence of agreement of the parties the Tribunal shall apply the laws or rules of law that it considers to be appropriate. But is the choice of law rule at issue applicable to the merits? It seems that the issue of jurisdiction is preliminary to the merits, and the ICC Rules contain no express reference to law applicable to the issue of arbitral jurisdiction. If your case were under the ICDR Rules, on the other hand, Article 31.1 provides, without limitation only to “the merits,” that the Tribunal in the absence of party agreement “shall apply such law(s) or rules of law as it determines to be appropriate.” Under the literal text of ICDR Art. 31.1, if you reject the Panel majority approach in Setty and instead apply a traditional contacts-and-connections based choice of law rule, your choice of law rule is protected by the agreement of the parties to arbitrate the jurisdiction issue under the ICDR Rules, and the Federal Arbitration Act protects the agreement, so the courts ought not second guess you.
With only slightly less confidence, I believe the same outcome is achieved under the ICC Rules. Article 42 of the ICC Rules provides that “[i]n all matters not expressly provided for in the Rules, the Court and the arbitral tribunal shall act in the spirit of the Rules….” And of course the empowerment of the arbitrator to decide upon her own jurisdiction is embedded in those Rules. Your ability to select a choice of law rule, in relation to the extension of the arbitration agreement to non-signatories, seems to be protected by the agreement to arbitrate jurisdiction under the ICC Rules, and in turn by the FAA whose core principle is to enforce the agreement. Add to this the observation – which I make from recollection and not from a systematic review of Supreme Court decisions got this Commentary, but I believe it to be correct – that the Supreme Court’s interpretations of the requirements of the FAA pertaining to arbitrability – apart from allocation of power between courts and arbitrators — have concerned how courts shall address arbitrability when it is required to be decided by a court. That is to say, the US jurisprudence on arbitrability is not mainly concerned, perhaps concerned almost not at all, with what rules of law the arbitrator decides to apply to arbitrability issues, unless the arbitration agreement imposes rules she must apply and it is claimed that she failed to apply them.
If this characterization of US arbitration law is accurate, then unlike many leading arbitral jurisdictions the United States really does not have an American arbitration law approach to the law applicable to issues of arbitrability that binds arbitrators. When you instinctively apply New York law because the contract provides for a New York seat and New York substantive law, be aware that you are not enforcing an implied choice of law on arbitrability. Instead you are adopting one of several perfectly supportable approaches to the choice of law issue. And so it is a useful bank of knowledge for the US-seated international arbitrator, or the arbitrator applying a contract that generally selects New York or other US state law, to have in mind generally how other major arbitral jurisdictions treat the question. Here are some basics.
1.The Law of England: The UK Supreme Court chimed in on October 9, 2020 with a decision called Enka v Chubb ([2020] UKSC 38). Per this decision, a general choice of law clause in the contract will also govern the arbitration agreement if the parties have not chosen other law applicable specifically to the arbitration agreement. If no choice of law is made, there will be a presumption that the law of the seat of arbitration governs the arbitration agreement, a presumption derived from the UK conflict rule that points to the law with which the arbitration agreement is most closely connected. Caveat: this was a case concerning the scope of arbitrable disputes and not about who other than a signatory may enforce or be bound by the clause.
2. The Law of Singapore: At least pending any new decision, where the influence of Enka v Chubb might be felt, Singapore law declines to treat the contract’s general choice of law clause as a choice made for the arbitration agreement where no specific choice for the latter has been expressed. It was so held by the Singapore Court of Appeal in BNA v. BNB et ano. ([2019] SGCA 84). The “proper law” of the underlying contract (what we US types call the substantive law applicable) is however presumptively the proper law of the arbitration agreement in the absence of an express choice for the latter. As to what circumstances might bring about rebuttal of this presumptive choice, first on this list is where the proper law of the contract would deny enforcement of the arbitration agreement despite other clear manifestations of the parties’ intent to arbitrate. Stay alert for new developments in Singapore: the presumption sustained in BNA was drawn from UK case law prior to Enka v Chubb, notably in an English Court of Appeal case called (for short) Sulamerica. Does the “Sulamerica Presumption” survive Enka? It’s Topic #1 at the outdoor bar at the Raffles Hotel. Listen closely to the chatter! And caveat reader once more: this is not a case about enforcement by or against non-signatories!
3. The Law of Switzerland: Treading carefully now in the potential quicksand of civil law tradition, a federal arbitration statute without a decisive single construction on the question by the nation’s highest court, and a vast and influential academic and professional literature, an educated guess at the Swiss law position is as follows: One key provision of the Swiss federal arbitration statute (referring to relevant portions of the Swiss Private International Law Act (PILA)) treats an arbitration agreement as valid if it is valid under one of three sources of law: “the law chosen by the parties, or the law governing the subject-matter of the dispute, in particular the main contract, or [under] Swiss law.” (PILA Art. 178(2)). In Europe they call this “in favorem validitatis”; which at your kitchen table means (roughly) “whatever works.” But here’s the thing. Clause “validity” is obviously just a small piece of larger arbitrability puzzle, so it is robustly debatable whether the Swiss position on clause validity applies to non-signatory issues (what in European lexicon pertains to the “extension of” the arbitration clause). But wait, the plot thickens : If you go down a ways in the PILA you get to Art. 187 (1) which is a “special conflict of laws provision just for international arbitration” (said an actual reliable source, Pierre Karrer, in 2014). It says “The Arbitral Tribunal shall decide the case according to the rules of law chosen by the parties or, in the absence thereof, according to the rules of law with which the case has the closest connection.” For some scholars (e.g. Prof. Karrer) this test inevitably leads to applying the law of the seat in the absence of a different express choice. But what if the seat in Switzerland is not chosen by agreement, but by the ICC Court, as a compromise choice wanted by neither party? Happily there is clarity in the Swiss law on extension of the arbitration clause to non-signatories if the conclusion on choice of law is that Swiss law applies: a concept effectively equivalent to common law equitable estoppel prevails – “behavior with respect to the conclusion or the performance of the principal agreement that either clearly demonstrates its (implied) intent to be bound by the arbitration clause, or that under the general principle of good faith may and must have been understood by the other party in such a way that the non-signatory intended to join the principal agreement, including the arbitration clause being part thereof.” (I quote Georg von Segesser paraphrasing the Swiss Supreme Court in a 2009 Kluwer blog post).
4. The Law of France: Choice of law is easy in France because no national law stands to be chosen. The arbitration clause is considered to be governed by general principles focused on the common intent of the parties that are not associated with any national legal system. Voila, another good reason to go to Paris, if ever again we can.
It is a distinctive aspect of American arbitration law there really is not an “American law approach” to be added to this list. Effectively, American arbitration law prescribes to some extent how judges decide issues of arbitrability, but does not prescribe rules about how arbitrators decide issues of arbitrability so long as they respect the agreement of the parties. So it is open to arbitrators sitting in the US to consider all of foregoing approaches to choice of law applicable to the arbitration agreement. And it is clearly open to arbitrators sitting in the US to treat US case law like the Setty case as a unique law-of-the-forum applicable to judicial proceedings and not even particularly instructive about how arbitrators sitting in the US should approach their arbitrability task.
