More on international arbitration in the wake of Devas
The United States in the International Arbitration System
Residing as I do for the time being in Shenzhen, located twelve time zones away from the East Coast and burdened by limited internet access, engaging with the latest Supreme Court decisions presents challenges. Yet when the Court hands down a case in which I took part, if only as an uninvited and perhaps unwanted amicus curiae, I feel compelled to have my say. The following represents a deep(ish) dive into the law of international arbitration framed by a recent decision that contributes next to nothing to this law but opens the door to some interesting and possibly important issues. I develop the argument for applying the U.S. rules for personal jurisdiction to all confirmation claims under the international arbitration system. In a following post I will discuss another issue raised but not resolved in the Court, whether U.S. law requires confirmation claims to have some nexus to either interstate commerce or the foreign commerce of the United States.
Recently the Supreme Court decided CC/Devas v. Antrix Corp. Ltd., an international arbitration case to which I had contributed an amicus brief. My reaction to the decision appears on Transnational Litigation Blog. I say there that Justice Alito’s opinion for a unanimous court is a damp squib, an emission that only resolves an issue on which the parties fully agreed (that the lower court had relied on an indefensible interpretation of the personal jurisdiction provision of the Foreign Sovereign Immunities Act (FSIA)). The opinion leaves all the remaining, mostly more important, questions for resolution (or not) on remand. In that post, I ponder what the Ninth Circuit might do now that the case has returned to it.
Here I want to go big picture. I speculate about the case’s implications for the general state of the international arbitration system and the role of the United States in sustaining it. In particular, I generalize beyond those instances where sovereign immunity, and the particular U.S. statute governing that status, is involved. Should a person seeking recognition of an international arbitration award within the terms of U.S. law (Chapter Two of the Federal Arbitration Act plus the New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards) have to establish personal jurisdiction over the person subject to the award to get confirmation of the award? If no, then interpreting FSIA differently seems problematic. If yes, then giving foreign state-owned companies the same protection as foreign private companies makes sense.
The core problem is whether an arbitration agreement should function something like a universal injunction. The latter is a procedural device through which a person aggrieved by something the United States has done (or failed to do) can sue anywhere in the United States to override the offending action. As many suits can be filed as there exist plaintiffs with standing to oppose the government, which means opponents can bring suits around the country. Once one suit succeeds, the government is completely bound (pending appeal and ultimate resolution by the Supreme Court), no matter how many others fail. Some members of the Court, perhaps a majority, find this procedure troubling and are looking for ways to rein it in. In the legal academy, Sam Bray has done much of the heavy lifting in exposing the problems with universal injunctions.
A reading of the New York Convention, the main treaty governing international commercial arbitration, produces something very similar, except with money judgments rather than equitable relief. The treaty, which virtually every country in the world has joined, obliges a state party to confirm an award arising from an arbitration almost anywhere, as long the dispute is commercial in nature and the confirming state does not regard the award as domestic, that is made under its own legal system. A state may (but not must) withhold confirmation for several reasons, including if the award conflicts with the agreement’s terms of reference, if the home state of the arbitral award sets it aside, if confirmation would conflict for its procedural rules, or if the award offends a public policy of the confirming state. Once confirmed, the award becomes a domestic judgment, a significant step in turning the rights of the person who procured the award (the award creditor) into money.
Importantly, a decision denying confirmation in one state has no legal effect anywhere else. The award creditor can convert a single confirmation decision into full satisfaction of the award, even if courts elsewhere have deemed the award invalid. Even a conclusive judicial decision to set aside the award under the laws governing the arbitration does not cancel the power of another judicial system to confirm. Confirmation provides a basis for collection against any legal entitlement (a tort or property claim, not just a property interest) of the award debtor that the award creditor can track down in the confirming state. In Devas, the award creditor wants to use confirmation in Washington as a basis for subrogating to Antrix’s contract claim in a Virginia bankruptcy case, even though Antrix has had no contacts with either Washington or Virginia. Moreover, it chases the money even though the Indian courts have ruled that the award, a product of Indian law, is invalid as a matter of Indian law.
A confirmed award can lie embedded in that country’s legal system for years (subject to the local statute of limitations for court judgments), awaiting any future event that brings an entitlement with its reach. Defenders of this approach say it gives the award creditor the level of enforcement that the system requires to function effectively. I suggest, to the contrary, that it provides the award creditor too many bites at the apple, given the fallibility of arbitral tribunals and the shortcomings of confirming courts.
Over the course of my (very long) professional life, international arbitration has grown enormously. It has become the preferred method of resolving business disputes throughout the global economy, excepting only financial transactions. The system works remarkably well, and the New York Convention has been a worthy as well as successful project. When I talk about the fallibility of awards and national confirmation proceedings, I do not mean to suggest that either are inherently suspect. Rather, I make only the minimalist claim that any legal proceeding may go awry. If this is a concern, then there is a possible case for imposing safeguards. It depends, of course, on whether the risk from fallible arbitrations and consequent confirmation proceedings is greater than the cost of the contemplated safeguard.
In the United States, the requirements of personal jurisdiction provide a check on profligate civil suits. The minimum contacts standard in civil litigation, first announced as a requirement of constitutional due process in International Shoe Co. v. Washington, requires a plaintiff to demonstrate that the defendant has attributes that make a suit in the selected forum appropriate, “whether arising from the respondent’s residence, his conduct, his consent, the location of his property or otherwise.” As the reference to property indicates, an exception exists where a defendant owns property in that jurisdiction. Shaffer v. Heitner indicates that a person with rights under a conclusive court judgment can assert what’s called quasi-in-rem jurisdiction over a defendant’s property wherever that property can be found, even if the defendant has no other ties to the forum. The lower courts for the most part extend this rule to award creditors. Otherwise, though, personal jurisdiction requires either consent or some connection between the defendant and the forum, typically called minimum contacts. The recent history of the Court have been less than helpful in clarifying how this works in practice, but that sad fact is orthogonal to my argument here.
Some of the amici briefs in Devas, authored by prominent figures in the international arbitration community, argue that, as to international arbitral awards, the personal-jurisdiction barrier to confirmation goes too far. They observe that consent by itself suffices to establish personal jurisdiction and argue that entering into an arbitration agreement under the New York Convention constitutes constructive consent to all conceivable means for enforcing the resulting award. This consent, however much a legal fiction, is enough to dispense with other standards for personal jurisdiction. If an award rests on an otherwise valid arbitration agreement, then, confirmation does not require any relationship between the award debtor and the forum entertaining the claim.
In my view, that argument proves too much. It is accepted, at least in the United States, that consent to arbitration in any of the several States implies consent to confirmation of the resulting award in that State. It does not follow, however, that consent to arbitration in a jurisdiction, whether a U.S. State or a foreign country, implies consent to confirmation in any state that is a party to the New York Convention, even though that Convention may be seen as part of the law governing the arbitration. As the D.C. Circuit ruled in Creighton Ltd. v. Gov’t of the State of Qatar, “an implicit waiver of personal jurisdiction in a defendant's agreement to litigate or to arbitrate in a particular jurisdiction is applicable only within that jurisdiction.”
Moreover, arbitration agreements can and often do provide for express consent to subsequent enforcement proceedings. It is not much short of malpractice to fail to includes such a provision in an agreement to arbitrate. As Linda Silberman has observed, “Parties can always include a clause affirmatively consenting to future enforcement of any award in the arbitral agreement, . . . if the parties choose not to include such a clause, then perhaps the courts should not impose one . . .” In the absence of such a clause, inferring consent is a motivated, policy-driven choice, not an inevitable step.
The overlap between personal-jurisdiction and sovereign-immunity law bolsters this argument. Both personal jurisdiction and sovereign immunity, which bar a court from hearing otherwise valid claims, can be waived. Yet both the U.S. courts of appeal and the British High Court have rejected the contention that consent to an arbitration agreement, through either a contract or a treaty, qualifies as a waiver of immunity. National laws that provide for (and limit) the confirmation of an arbitration award against a state or state-owned entity, such as FSIA and the United Kingdom’s State Immunity Act 1978, treat consent and entering into an arbitration agreement as legally distinct categories for purposes of immunity. These enactments, although applicable only to states and state-owned entities, suggest a broader point: an agreement to arbitrate a dispute does not necessarily entail willingness to submit to a subsequent confirmation suit anywhere in the world. There is no reason to embrace such a broad conception of constructive consent unless to increase the impact of arbitral awards in the face of fallibility risk.
If an agreement to arbitrate, without more, does not count as constructive consent to submit to confirmation proceedings anywhere in the world, then a possible limiting principle emerges. It might provide that an award creditor may seek confirmation wherever it can find property belonging to the award debtor or where the debtor otherwise has sufficient contacts with the forum to satisfy the local personal jurisdiction rule. Shaffer seems to impose a property-suffices-for-jurisdiction rule, and the exigencies of international arbitration require this. Otherwise, award debtors could stash their assets in a place where they do not otherwise have contacts and wait for the statute of limitations on the award (normally shorter than that for local judicial judgments) to run.
But, following the language of Shaffer, a judgment of confirmation could not be used as a basis for attachment of or subrogation to any non-property rights of the award debtor, such as tort, contract or tax refund claims. To pursue these non-property entitlements, a personal-jurisdiction limiting principle would provide that the award creditor may seek confirmation only in places where, under the forum’s rules, personal jurisdiction exists, and only to the extent of that personal jurisdiction. If jurisdiction rests only on the presence of property, then the confirmation award can do nothing more than address the award creditor’s rights in that property.
Why treat property rights differently from other kinds of legal entitlements? The distinction might smack of legal formalism. It’s not clear that courts always do this. Some have labeled any legal entitlement, such as an inchoate contract claim, as “property” for purpose of asserting quasi-in-rem jurisdiction. It’s not absolutely clear that Shaffer, addressing the point in a footnote in a case where the issue was not present, ruled out this extension. Yet the distinction between property and other legal entitlements makes sense here, when considering how the rules for confirming international arbitral awards work.
A property right constitutes a legal claim against the whole world, indicating that its holder must stand guard against challenges originating anywhere. Requiring people to keep track of and defend their property rights seems reasonable as a general matter. The rules of adverse possession rest on that premise. Property rights thus include any legal entitlement that its holder can assert against anyone and thus anyone might infringe.
Other legal interests, such as inchoate claims not yet reduced to a conclusive judgment, are specific to particular persons (the claimant and either the tortfeasor or contractual counterparty). Claims by third parties to these entitlements are more likely to come as a surprise to their beneficiaries. Accordingly, the legal system reasonably might require less of the beneficiary when it comes to defending non-property entitlements from claims by third persons (those who are not themselves party to the claim).
If so, the question of when an award debtor may be compelled to defend its legal interests by opposing a confirmation suit turns on the kind of interests that the confirmation proceeding puts at risk. It presumes that the law should adjust the burden of vigilance assigned to the beneficiary of an entitlement depending on the beneficiary’s reasonable expectations about the entitlement. Some entitlements are sufficiently definite to support a greater duty of care to protect one’s interest. Interests typically classified as property, as well as final and conclusive domestic judicial judgments that have much of the character of property, should fall into this category. Less definite legal interests, such as a power to claim compensation for a supposed tort, an alleged breach of contract or a possible improper tax assessment, seem sufficiently uncertain to justify lower levels of precaution, at least where those interests dwell in places to which the potential beneficiary has no ties.
This argument does not pin down several important concepts. It recognizes a category of “property,” inspired by Shaffer, but does not say in marginal cases what interest should or should not count as property, other than to suggest it has something to do with rights against the whole world and reasonable expectations about vigilance. Neither does it specify what should count as a “tie” to a place, much less defend the current state of the Court’s personal-jurisdiction jurisprudence. The point, rather, is that imposing some limits on an award creditor’s right to seek confirmation in a particular jurisdiction makes sense, and that those limits should have something to do with whether the award debtor should reasonably be expected to protect all its interests in that jurisdiction or only some. It maintains that the presence of property, however defined, or ties, whatever that means, has a lot to do with those reasonable expectations.
Why talk of vigilance at all? A premise of the argument, as in the universal jurisdiction controversy, is that an initial arbitration may be fallible and that a court considering confirmation must have some means to detect its faults. The arbiters may address an issue that they lack the authority (legal competence) to resolve. The respondent might not get fair notice of the proceeding. A competent court of the domestic legal system on which the arbitration rests may find the award invalid as a matter of local law. An international arbitration regime that does not provide means of addressing these defects would be an abomination. These are the rules that U.S. courts generally apply when considering the confirmation of a foreign (non-U.S.) judicial judgment.
Confirming courts cannot be counted on to uncover the award’s defects without the help of the award debtor, yet providing that help is expensive. The award debtor must first learn about the confirmation proceeding, hire local counsel to launch its defense, and assemble the evidence to expose the award’s flaws. It must decide, in particular, whether the costs of attacking the award is worth it, if victory will protect from seizure only whatever entitlements it has in that particular forum and will have no legal effect in any other place. A reasonable limiting principle would let the debtor, without legal risk, pass up on defending its interests in places where confirmation would threaten the kinds of entitlements for which it normally would not be expected to exercise utmost vigilance in protecting.
To come back down to earth from this theoretical digression, does the actual international arbitration system with which we live, the one based largely on the New York Convention, tolerate limits tied to personal jurisdiction? Some of the amicus briefs in Devas contend that imposing a personal jurisdiction test, understood to encompass the presence of judgment debtor’s property, on confirmation suits would violate U.S. international obligations and undermine the New York Convention system. Having agreed to engage that system, the argument goes, all parties to an arbitration agreement made under its auspices must face the risk of confirmation proceedings in any state that belongs to the system (again, virtually the whole world). This, they say, is what the treaty requires.
Except that it doesn’t. The New York Convention does let state parties dispense with personal jurisdiction rules in a confirmation proceeding if they choose. Thus a country such as France, which generally (excepting litigation within the EU) embraces a concept of extravagant jurisdiction that allows its nationals to assert legal claims against anyone in the world, may (and for all I know, does) allow any and all confirmation suits in cases where the award creditor is French and the award debtor is not an EU national. The Convention, however, does not compel this approach.
First, Article III of the Convention requires a party to confirm the awards to which it applies “in accordance with the rules of procedure” of the confirming state. Requirements for personal jurisdiction are procedural, not substantive, and apply to all civil proceedings. States may take different approaches to personal jurisdiction, and neither the New York Convention nor general international law obliges states to adopt the French conception of extravagant jurisdiction when confirming arbitral awards.
Second, Article V(2) allows a state to refuse confirmation either because the dispute is, under its domestic law, not arbitrable or because the award violates its public policy. Limits on personal jurisdiction can express a state’s public policy, especially if they apply to all confirmation proceedings, not just those involving arbitral wards. More broadly, both the arbitrability and public-policy exclusions allows a state to withhold confirmation because of peculiar judgments expressed in its legal system, rather than hewing to a universal and uniform standard of good arbitration practice.
The embrace of some limits on personal jurisdiction, applied to all proceedings to confirm an arbitral award, advances the goal of thwarting abusive claims, not hostility to arbitral awards as such. In the context of U.S. law, it rests on a view that any legal claim, to attain the status of a legal entitlement protected under the Full Faith and Credit Clause, must be fixed through a proceeding that complies with general rules of personal jurisdiction, as well as the other aspects of due process. It thus is wrong to argue, as some of the amici briefs in Devas do, that a U.S. court should treat an arbitral award as equivalent to a U.S. judicial judgment. That puts the cart before the horse: the award becomes a judgment only after a confirmation proceeding, which itself should comply with normal procedural standards that govern civil suits. Indeed, the New York Convention allows a refusal to confirm an arbitral award on grounds, such as local public policy, that do not apply to confirmation of State judgments under the Full Faith and Credit Clause.
The remaining objections involve efficiency and efficacy. Forcing judgment creditors to wait for fixed entitlements to appear before springing into action slows down recovery and diminishes the value of awards. An award may expire due to the applicable statute of limitations (three years in the United States) before any property shows up. In David-Goliath contexts, the rich and powerful award debtor can grind down the plucky award creditor with procedural objections and endless litigation. Why add yet another procedural barrier to justice for the already overburdened Davids of the international arbitration world?
A simple answer is that, in that world, separating the Davids from the Goliaths is no easy thing. Because arbitration has become the dominant form of dispute resolution for the lion’s share of cross-border commerce, excepting only financial transactions, it is impossible to detect any systematic imbalance in the composition of the award debtor and creditor classes. If anything, it seems reasonable to expect that richer and more powerful actors would win more often, leading to an overweighting of Goliath award creditors relative to debtors.
A last, perhaps peripheral issue entails a collateral benefit associated with U.S. confirmation suits, relative those in the rest of the world. U.S. discovery rules apply in all federal civil suits, including those for confirmation of an arbitral award. To my knowledge, no other legal system has such generous discovery for litigants as the United States provides. If nothing else, allowing an award creditor to bring a U.S. confirmation suit means arming it with a unique tool to uncover where else in the world the debtor might be vulnerable to the award’s enforcement.1
It cannot be the case, however, that the international arbitration system depends for its flourishing on a unique and, from a comparative perspective, idiosyncratic litigation tool found in only one (admittedly important) country. If generous discovery were so important, surely we would see it in other countries, if only for arbitral confirmation proceedings. Moreover, the trend in the United States is in the opposite direction. ZF Automotive US, Inc. v. Luxshare, Ltd., a 2022 Supreme Court decision, denied parties in international commercial arbitration disputes access to U.S. discovery with respect to the arbitral proceedings themselves. Allowing an award creditor access to a U.S. court only for the purpose of getting discovery runs against the spirit, if not the letter, of that decision.
To conclude, there are good arguments for applying the normal rules of personal jurisdiction to any suit for the confirmation of an international arbitration award. Whether or not Congress has the constitutional authority to change the rules when a foreign state-owned company is the award debtor, present law most likely does not do this. This outcome is consistent with the international arbitration system, and its embrace would bolster, not undermine, the settlement of commercial disputes through the global economy.
This is enough for today, but I want to consider one more issue on this Substack. In the Supreme Court, counsel for Antrix proposed a different limitation on FSIA’s arbitration exception, one focused on the present of a nexus between the dispute subject to arbitration and the foreign commerce of the United States. I will explore some of the implications of that possibility in a later post.
1. I am indebted to Carter Phillips, counsel for respondent in Devas, for reminding me of this point.