Devas and Subject Matter Jurisdiction
In an earlier post on this platform, I discuss what a personal jurisdiction requirement for suits confirming a foreign arbitral award would look like. CC/Devas v. Antrix Corp. Ltd. surfaced but did not resolve the issue. Here I want to compare an alternative approach to confirmation suits, a possible rule of subject matter jurisdiction.
In the Supreme Court, Antrix’s counsel, the acclaimed Supreme Court litigator Carter Phillips (whom I count as a friend, not that such relationships matter in arguing about the law, and who first drew my attention to the case), focused on the specific language of Section 1605(a)(6), the arbitration exception to the Foreign Sovereign Immunities Act (FSIA). It provides that a claim, to come within the exception, must concern “a subject matter capable of settlement by arbitration under the laws of the United States.” This raises the question of what limits U.S. law might impose on arbitration agreements consistently with its international commitments.
The relevant law of the United States, Phillips argued, starts with the Federal Arbitration Act, otherwise known as Chapter One of Title 9 of the U.S. Code. Section 1 of the Act defines “commerce” for purposes of that title as limited to interstate commerce within the United States or foreign commerce with the United States. Section 2 provides for the validity under the Act of, among other things, “a contract evidencing a transaction involving commerce.” Section 202, found in Chapter Two of Title 9 (which implements the New York Convention as U.S. law), provides that an “arbitration agreement or arbitral award arising out of a legal relationship . . . which is considered as commercial, including a transaction, contract, or agreement described in section 2 of this title, falls under the Convention.”
Stringing these provisions together, an arbitration agreement with no ties to either interstate commerce or the foreign commerce of the United States does not fit within Section 1, thus lies outside Section 2, and consequently does not, as a matter of U.S. law, fall under the Convention. Put simply, an award arising out of a purely domestic transaction of a foreign state, lacking any nexus to the United States, is not capable of settlement under current U.S. law.
The argument goes beyond the definition of FSIA’s arbitration exception. It would bar confirmation any New York Convention award, whether involving a foreign state or not, if the dispute subject to arbitration lacks a nexus to U.S. interstate commerce or foreign commerce involving U.S. persons or territory. Indeed, because the New York Convention in most cases provides the means for confirming investment treaty awards, this subject-jurisdiction rule would also bar suits to confirm investment-treaty awards absent a nexus between U.S. interstate or foreign commerce or a specific statutory provision outside Title 9.
The argument raises several interesting issues. Section 1’s limited definition of commerce, when married to Section 2’s determination of what counts as a valid arbitration agreement, makes perfect sense in the Federal Arbitration Act, which deals with domestic U.S. arbitration. But Chapter Two, including Section 202, implements the New York Convention, which addresses international arbitration. Chapter Two doesn’t offer a separate definition of commerce or otherwise indicate that it dispenses with Chapter One’s U.S. nexus, and Section 202’s specific reference to Section 2 points in the opposite direction. Yet the Convention itself is not limited to commercial disputes with ties to the United States. How might a court reconcile the two Chapters for purposes of determining what the “laws of the United States” regard as arbitrable?
A possible pro-confirmation argument is to assert that no matter what Chapter Two provides, the New York Convention counts as a law of the United States and manifestly dispenses with a U.S. nexus. But the Court in Medellín v. Texas, a case having nothing to do with arbitration, used the Convention as an example as a non-self-executing treaty, meaning a treaty that has no legal effect in the United States other than through the statute that implements its. Gary Born, a leading international arbitration practitioner (and yes, also a friend with whom I’ve practiced law and shared teaching duties) argues that the New York Convention is self-executing. He dismisses the Medellín observation as a throw-away remark, not a precedent.
Were Born correct on the issue of self-execution, the next question would be whether the treaty, to the extent it regulates matters that fall outside the scope of domestic or foreign commerce within the meaning of the Constitution’s Article I, unconstitutionally extends the powers of the federal government. Justice Holmes, writing the majority opinion in Missouri v. Holland, upheld a treaty that exceeded the enumerated Article I powers of Congress on the ground that the treaty power, grounded in Article II, permits more. But that decision is more than a century old and, typical of Holmes’s opinions, does not really explain itself. Most people who have considered the question believe that the Constitution must implicitly limit the treaty power in some way, but no clear consensus exists over what interests these limits must protect. It remains an open question whether an Article II treaty must only conform to the protections that the Constitution extends to individual liberties, or also account for the Constitution’s federalism principles.
A related question is, whatever the constitutionality of the New York Convention under the treaty power, whether the courts should interpret the legislation implementing it, here Chapter Two of Title 9, as not intruding into traditional areas of State competence absent a clear indication from Congress to do so. Bond v. United States, a far more recent decision of the Court, embraces this interpretive presumption. Here the recognition and enforcement of foreign legal decisions, whether judicial or arbitral, can be seen as a traditional State function, even if the Constitution’s Full Faith and Credit Clause and the Federal Arbitration Act also contribute to the applicable law.
An interesting wrinkle, however, is that awards based on investment arbitration also rely on Chapter Two of Title Nine for recognition suits. These awards rest on investment treaties, mostly bilateral, that bind states directly to provide compensation. A long line of Supreme Court opinions, most recently Dames & Moore v. Regan, depict the power to protect U.S. nationals from foreign states through resolution of their investment claims as an element of the inherent national sovereignty. As Curtis Bradley (OK, a former colleague and ongoing collaborator in many projects, a dear friend) has argued forcefully, and I think brilliantly, the concept of inherent national sovereignty complements the text of the Constitution and has its greatest force in addressing federalism, rather than separation-of-powers issues.
It might follow that the Bond presumption should apply to run of the mill commercial arbitration cases such as Devas, but not to claims against sovereigns (as distinct from their agencies and instrumentalities) based on investor rights. Development of the international law safeguarding foreign investors matters greatly to the national government, whether the disputes arising from the treaties involve U.S. persons or others. Only the national government, and not the States, have the competence to act on behalf of the United States in the making of this body of law. Confirmation of the awards that vindicate the rights of foreign investors thus do not entrench on a traditional area of State authority, and a statute addressing these awards would not need a restrictive gloss motivated by deference to State interests.
What this argument does is give a specific gloss to the words “a subject matter capable of settlement by arbitration under the laws of the United States” that depends on the nature of dispute. Regular commercial arbitration comes within the competence of U.S. law in its present form only if the dispute has some connection to U.S. commerce, either interstate or foreign. Disputes based on the protection of foreign investors from states, by contrast, are arbitrable under the laws of the United States because of the nation’s stake in the development of this body of law, independent of the nationality of disputants. This is a national interest derived from a conception of sovereignty that is constitutional, but not derived directly from the constitutional text.
Even with this exception for investment treaties, supporters of the international arbitration system would argue that reading U.S. law as requiring some nexus to the domestic or foreign commerce of the United States in order for a foreign arbitral award to qualify for confirmation would breach the New York Convention and thus undermine a valuable international project. Yet the Convention’s Article V(2)(a) allows a state party to withhold confirmation whenever the subject of an arbitration dispute “is not capable of settlement by arbitration under the law of that country.” The nexus requirement conforms to a literal reading of this proviso. For most jobbing lawyers, that should be answer enough. Still, given the institutional importance of the international system that rests on the New York Convention, it is fair to ask whether a nexus requirement conforms to the treaty’s overall spirit.
There are at least two lenses through which to discern the purposes of the Convention. Given its focus on the connection between arbitration and commerce, the Convention might be seen as primarily about facilitating business transactions, largely by reducing the legal risk associated with disputes connected to business relationships. Alternatively, the Convention, like many other private international law treaties that promote the unification of legal rules across borders, might be about uniform law as such, with the impact on business transactions only a secondary concern.
If the Convention is mostly about international business, that is transactions that connect in some way to the movement of services, goods, people or capital across national borders in pursuit of gain, then excluding purely domestic transactions would not impair its objects and purpose. The Devas transaction, for example, possesses an international effect only with a lot of imaginative reasoning. An Indian service-provider (Antrix) contracted to provide an Indian customer (Devas) access to a satellite network that the Indian state planned to locate in a region of outer space controlled, under international law, by India. The only connection of the transaction to international commerce is that, if had succeeded, Devas could have marketed its telecommunication services to non-Indian customers. Unless all commerce is fundamentally international, perhaps by analogy to the old Wickard v. Filburn principle defining interstate commerce for purposes of the U.S. Constitution, the contract had no substantial connection to international commerce.
When constitutional arguments arise, I try to duck and run. My own view is in the post-modern knowledge economy based on connectivity and transparency, the view that everything is interstate, and indeed international, makes sense. Hence I resist the logic in National Federation of Independent Business v. Sebelius, the Obamacare decision that significantly cut back on Wickard. Still, life is too short to worry about constitutional law, so I will accept for argument’s sake that the limits found in Section 1 of the Federal Arbitration Act are real and meaningful and extend to an assessment of the purposes of the New York Convention.
Consider instead an understanding of the Convention as promoting uniformity in the worldwide rules for arbitral award confirmation, no matter what the subject of the arbitrated dispute. Perhaps, but why? Uniformity unlinked to particular substantive objects has few benefits and may entail significant costs.
My views may be suspect, as some decades ago I published a notorious and controversial critique of the uniformity project in the context of international commerce. In a nutshell, most commercial transactors, including those engaged in international commerce, prefer legal predictability, understood as clarity and stability, over formal uniformity of rules. Given the diversity of preferences among states, an internationally negotiated attempt at uniformity can succeed only with great generality and abstraction that conceals that diversity. As a result, the international search for uniformity produces elevated legal risk and ambiguity, exactly what transactors do not want. Unifications projects benefit the law reformers who run them and grab the glory, but at the expense of international commerce.
Shockingly, not everyone in the international law reform community embraces my argument. Still, I stick by it today, and reference another one advanced by the distinguished practitioner and sometime law teacher Delphine Nougayrède (yes, another friend with whom I have both practiced law reform in Russia and taught in the United States). In the course of her extended transactional practice in Russia after the demise of the Soviet Union, Nougayrède observed how Russian law enabled significant business actors to outsource their essentially domestic business disputes to outside actors, principally the British courts. This practice deprived Russia’s domestic courts of the opportunity to build legitimacy by adjudicating these conflicts themselves. Abandoned as a legal backwater, the courts failed to meet the needs of an emerging domestic private sector.
The Russian instance involves foreign courts rather than arbitration, but the point remains the same. Channeling domestic business disputes to an international dispute resolution system, here the New York Convention regime, deprives domestic courts of the chance to take up a valuable function in society, as well as removing an incentive for the government to invest in the domestic rule of law. Disputes that matter, namely those between the rich and powerful, go abroad, leaving the courts as places where the rich and powerful have their way with regular people.
The point is not that outsourcing domestic business disputes is always bad. In some instances it may be the lesser evil. My argument, rather, is that uniformity in legal process as an abstract goal, not tied to substantive ends such as nurturing international commerce, can come at a cost. This point does not compel, but it may explain, why the adoption of a foreign commerce nexus as an element of arbitrability under U.S. law might advance, rather than impede, the overall goals of the New York Convention regime, properly understood.
What I have argued so far justifies the application of a nexus requirement to confirmation suits that excludes purely domestic arbitral awards from other countries. It does not, however, address cases where an arbitrated dispute arises out of an international transaction, just not one with a U.S. nexus. As noted, the Convention seems to allow the United States to regard such disputes as “not capable of settlement by arbitration” under its laws. But this understanding of U.S. law might undermine the system of international arbitration in one important respect.
If a party to international agreements can, in anticipation of losing a dispute, move its property to a country where no duty to confirm an award exists, it could sit out the statute of limitations for seeking confirmation (three years in the United States) and then get away scot-free. Frustrating such moves is an important systemic goal. Yet the nexus rule implied by U.S. law seems to give award debtors a free pass if the dispute does not implicate any transaction touching on U.S. territory or involving U.S. persons.
Last year I published on Transnational Litigation Blog Comity and Kleptocrats, a critique of a New York court’s decision to bar the victims of a foreign fraud from making a claim against the U.S. real estate in which the fraudster had invested the fruits of his misconduct. (The decision was for a failure to state a claim, so the fraud has to be assumed even though not proved.) Additional thoughts appeared on this platform. In essence, I argued that the United States should not enter into the international competition for the provision of money laundering services. This means that, even in the absence of a criminal proceedings, U.S. courts should not set up arbitrary barriers to victims of financial crimes who can prove the wrongful origins of property now in the United States. We take for granted that if someone steals property (say looting antiquities) and brings their ill-gotten gains to the United States, the victim should be entitled to establish its status as rightful owner and retrieve what’s theirs. The same principle should apply if the proceeds of misconduct are invested in U.S. property.
A debtor who evades a lawful legal judgment, here an arbitral award, by moving its property to a place of legal impunity acts wrongfully, even if not necessarily criminally. In some contexts, such a move could constitute criminal fraud or money laundering. Criminal or not, it is not a practice that the United States should encourage.
Squaring aversion to fraudulent conversion with the imposition of a nexus limitation on confirmation of arbitral awards is difficult. It is reasonable for the United States to decline to apply its domestic law to the arbitration of disputes that have no connection to its subjects or territory. But once property enters the United States, our country acquires an interest in upholding the rule of law with respect to that property. What is at least a plausible reading of U.S. law, and perhaps the best, neglects that interest.
The tidiest way out of the conundrum would be a quick statutory fix. Congress could amend Title 9 to authorize confirmation proceedings based on quasi in rem jurisdiction with respect to any property imported into the country after the commencement of an arbitral proceeding covered by the New York Convention involving its owner. Rules like those defining fraudulent transfers in bankruptcy or tax could apply.
A less tidy solution would be to restrict the meaning of “not capable of settlement by arbitration” by abandoning the nexus test found in Section 1 of the Federal Arbitration Act. Perhaps Congress should have addressed the issue when it adopted Section 202 and not limited federal jurisdiction over arbitral awards by direct reference to the prior statute. Perhaps judicial reworking of Chapter Two, indifferent to the actual choice of legislative language, might repair the damage. Still, over the last forty years and more the Supreme Court has repeatedly called for legislative action, and not judicial creativity, as the preferred method for fixing gaps in our statutes.
As a teacher and not a judge, I’m paid to raise problems more than solve them. The subject-matter-jurisdiction issue for confirmation of international arbitral awards will not go away, and either Congress or the courts will have to address it. I hope this post will help them, but I’m not holding my breath.