HILJ Online Symposium: Greg Shill Responds to Christopher Whytock

HILJ Online Symposium: Greg Shill Responds to Christopher Whytock

[Greg Shill is a Visiting Assistant Professor at the University of Denver Sturm College of Law.]

This post is part of the HILJ Online Symposium: Volumes 54(2) & 55(1). Other posts in this series can be found in the related posts below.

I thank Professor Christopher Whytock for engaging with the ideas in my article, Ending Judgment Arbitrage: Jurisdictional Competition and the Enforcement of Foreign Money Judgments in the United States, 54 Harv. Int’l L.J. 459 (2013), and the Harvard International Law Journal and Opinio Juris for hosting this symposium. Whytock has published widely on transnational litigation and judgment enforcement. Ultimately, I think his response misreads or overstates the article’s claims in some places (and in others we may simply have a difference of opinion), but the sister-state dimension of transnational judgment enforcement has thus far not attracted much scholarly attention and I am delighted to see his thoughtful and serious commentary in this forum.

I. Judgment Arbitrage & Whytock’s Criticisms

Briefly, the article focuses on the enforcement of foreign-country judgments in the United States. By its nature, this process creates the potential for clashes between domestic and foreign legal systems. In a typical case, a local court, often in the U.S., is asked to order a local defendant to satisfy a judgment rendered by a foreign court, under foreign law. Thus, unsurprisingly, scholars to date have tended to focus on the conflict between foreign sources of law and systems of justice on the one hand and their American counterparts on the other—the international-level conflict. One mission of the article is to explore domestic—i.e., sister-state—conflicts that result from the judgment-enforcement process.

To collect on a foreign judgment in the U.S., a plaintiff must first domesticate it. This entails a two-stage process: the judgment must first be recognized and then enforced. Federalism and the Erie doctrine are key to this process: (1) recognition is governed by state law, specifically forum law, (2) recognition standards differ widely from state to state, and (3) states have an obligation to enforce one another’s judgments. Thus, I argue, plaintiffs can exploit sister state differences in recognition law by first obtaining recognition in a state that is receptive to foreign judgments and then enforcing in a state that might not have recognized the foreign judgment in the first place. My article gives this phenomenon the name “judgment arbitrage,” and closes by proposing a federal statute to address it. The upshot of the statute is to allow states to resist judgment arbitrage by declining to enforce judgments they would not have recognized in the first place.

As the title of his response suggests, Whytock is skeptical that judgment arbitrage exists. While my article is the first study of the strategy, it acknowledges the work of other commentators who have alluded to it (e.g., p. 478 n.83.), a practice guide that endorses its use (p. 478), and a Texas decision, Reading & Bates Construction v. Baker Energy, 976 S.W.2d 702 (Tex.App.-Hous. 1st Dist. 1998)), that recognized and rejected a bald-faced attempt to employ it (p. 490 n.157). So I’m comfortable claiming that it is available as a matter of law. Further, given its advantages, I would be surprised if it were not used in practice to some extent, though it’s probably sufficiently expensive and complex to be rare (on which more shortly).

The three forums relevant to transnational judgment litigation are:

F1: the forum that renders a judgment on the merits (e.g., Pakistan),
F2: the forum called upon to recognize the judgment rendered by F1 (e.g., New York), and
F3: the forum called upon to enforce the recognition judgment rendered by F2. For example, New York or Massachusetts. See pp. 476-79.

F2 and F3 are usually the same forum, e.g., New York. In fact, so much so that in most cases the cases are collapsed into a single proceeding. I call the phenomenon “classic cross-border litigation” (p. 477). There is no potential for sister-state conflicts: in this example, New York law controls both recognition and enforcement.

Formally, however, recognition and enforcement are two distinct proceedings, and nothing obligates plaintiffs to bring both suits in a single forum. In addition, states enforce one another’s judgments almost automatically. This creates the possibility of judgment arbitrage: if the plaintiff gets a recognition judgment from a friendly F2 state and selects a more protective F3 state for enforcement, he may effectively exploit the difference in recognition law between sister states. For example, New York law is friendly to foreign judgment-creditors whereas Massachusetts has one of the most restrictive enforcement regimes in the country (see pp. 492-98). The arbitrage opportunity arises by virtue of the two-step recognition-and-enforcement system.

The crux of Whytock’s critique is a pair of related claims, one empirical and the other theoretical. First, he claims that I argue “non-U.S. plaintiffs ‘routinely’ practice a three-step strategy called ‘judgment arbitrage’” while “clearly fail[ing] to demonstrate” that it is in fact “‘routine’” or to identify any examples of “real-world” judgment arbitrage. Second, he questions the availability and relevance of judgment arbitrage by arguing that its three steps are “theoretically unlikely to occur with much frequency.” I will respond to these claims in turn.

II. Empirical Challenges

Whytock is mistaken when he says the article argues that judgment arbitrage is routine. In fact, the article contrasts judgment arbitrage, which is nonstandard, with the usual model of transnational judgment litigation, which I call “classic cross-border litigation” (p. 477). In the “classic” procedure, plaintiffs litigate the merits of their claims in one country (F1) and seek recognition and enforcement in another (F2 and F3—which are usually the same forum). It is this model of judgment litigation that I claim plaintiffs “routinely” practice (p. 470).

Although I do not claim judgment arbitrage is the dominant model and acknowledge its use is hard to detect, I believe is available and is likely practiced to some degree. In Reading & Bates, a prominent example of failed judgment arbitrage I discuss in the article (p. 490 & n.157), the plaintiffs tried to enforce their Canadian judgment via judgment arbitrage, but were none too subtle about it. Apparently, the plaintiffs feared their Canadian judgment would not have been recognized under Texas recognition law but believed it would be under Louisiana recognition law, so they went to Louisiana court first and secured recognition of the Canadian judgment. They then filed a pair of actions in Texas court: suits to recognize the Canadian judgment (as a prelude to enforcement) and to enforce the Louisiana recognition judgment. The Texas court consolidated the actions and in somewhat colorful language denied recognition to the Louisiana judgment on the grounds that the plaintiffs were attempting a naked end-run around Texas recognition law (“we refuse to allow [the judgment creditor] to enforce its Canadian judgment ‘through the back door’” in Texas, Reading & Bates, 976 S.W.2d at 715).

Whytock sees Reading & Bates as convincing evidence that attempts at judgment arbitrage will fail. I don’t think the opinion can bear that much weight. It is a single example of a particularly obvious effort at judgment arbitrage being rebuffed by a Texas state court. In the future, sophisticated plaintiffs would be wise not to come forward with foreign and sister-state judgments at the same time lest they risk alerting the court to their strategy. But even so, for reasons I detail in Part III of the article (e.g., at pp. 484-85), other courts may well decide this result is incorrect as a matter of law and may not follow it.

Nevertheless, the facts of Reading & Bates confirm that some in the plaintiffs bar are aware of differences in sister-state recognition law and seek to exploit those differences to their advantage. It’s not a foregone conclusion that other state courts would interpret their prerogatives as aggressively as the Texas state court. It’s equally possible that another court would interpret various sources of law, like the Uniform Enforcement of Foreign Judgments Act (adopted by 47 states), to require the enforcement of a sister-state recognition judgment subject only to the limited defenses specified in the statute, which do not include judgment arbitrage.

Whytock extends his observation about Reading & Bates to a few prominent failed attempts by plaintiffs to manipulate transnational litigation—the Chevron/Ecuador, Dole/Nicaragua, and Shell/Nicaragua cases—and from this concludes that our legal system stands ready to repel attempts at judgment arbitrage. I am not so sure. As Whytock notes, U.S. courts in these cases were “presented with evidence of corruption or fraud”—often overwhelming evidence, I would add. (As I note in the article, I represented Chevron in U.S. proceedings on the Ecuadorian judgment in 2011.) That evidence was not always easy to find. In the Chevron case, the effort to unearth evidence of scientific and litigation fraud took years and cost an absolute fortune. And in that case, the professional misconduct was pervasive and egregious—the plaintiffs made a documentary about their litigation strategy and filmed instances of their own misconduct! These revelations, which only came to light after a global discovery battle, led an S.D.N.Y. judge to issue a historic 497-page civil RICO judgment last month against the lead American attorney for the plaintiffs and his team.

So Whytock may be right that outright fraud is likely be rebuffed where good evidence of it can be uncovered, as in Chevron. And perhaps Reading & Bates-style gamesmanship will be rejected too (though there has been no comparable decision in the intervening 16 years). But these predictions overlook both the enormous discovery efforts required and the probability that more sophisticated plaintiffs’ lawyers will be more careful in the future.

Whytock’s complaint that the reported cases do not reveal many clear-cut examples of judgment arbitrage echoes a frustration I voice in the article. Far from claiming that the practice is “routine,” I detail this research challenge by noting that “the nature of [judgment arbitrage] makes reliable statistics on its use difficult to obtain,” “[d]ebtors ordered to pay enormous damages awards by foreign courts often settle out of court rather than take their chances at resisting domestication,” and that outside high-profile cases “the existence of [such] a settlement can often be kept confidential or at least quiet” by corporations keen not to encourage further lawsuits (pp. 478-79). These factors would complicate any inquiry that relied on Westlaw searches as the basis for counting instances of judgment arbitrage.

Nevertheless, I don’t mean to minimize the research challenge posed by the paucity of concrete evidence. If the article were a policy brief, or an empirical study, I think that would be damaging. Instead, however, the article’s focus is primarily theoretical and doctrinal. It is the first to introduce the concept of judgment arbitrage and the first to extend the theory of regulatory competition to judgment recognition. It is also the first to develop different models of “receptive” and “protective” recognition regimes. Whytock appears to take this research challenge as fatal to the model and a sign that judgment arbitrage can safely be dismissed as a threat.

We may have a methodological disagreement here. For the reasons I mention above, I do not believe the absence of reported decisions is as revealing as Whytock assumes, but regardless, the article’s contributions (if any!) do not depend on judgment arbitrage being proven frequent. Similarly, the article’s statutory proposal is mainly a tool to advance the literature; it’s unsurprising that as a report to Congress Whytock finds that discussion wanting. Perhaps in future projects scholars can use interviews of general counsel and the plaintiffs’ bar and other methods beyond traditional legal research to supplement the claims of Ending Judgment Arbitrage with empirical data, which could then help inform policy.

III. Theoretical Challenges

Whytock’s second criticism is that judgment arbitrage is “theoretically unlikely to occur with much frequency” because the “existing legal barriers [are] quite formidable.” I disagree.

In addition to a foreign forum that renders a judgment on the merits (F1), the judgment arbitrage model requires:

F2: a forum to recognize the judgment rendered by F1 (e.g., New York), and
F3: a different forum to enforce the recognition judgment rendered by F2 (e.g., Massachusetts).

Whytock contends that plaintiffs pursuing an arbitrage strategy would be thwarted at the recognition stage (F2) because defendants could get the court to bounce the recognition action via forum non conveniens (FNC). He then contends that even if a plaintiff did procure a recognition judgment in a receptive state and then tried to enforce in a more restrictive state, the enforcement court would not be obligated to enforce the recognition judgment. The law here is not terribly well developed, but I think the available evidence weighs against both these arguments, especially the second.

A. Forum Non Conveniens

Whytock argues that defendants can bounce recognition actions from receptive to protective U.S. jurisdictions via FNC. He writes: “when recognition is sought neither to establish a judgment’s preclusive effect in pending in-state litigation nor as a prelude to enforcement of the judgment against in-state assets of the defendant,” defendants can pursue dismissal on FNC grounds. I agree that without any assets or other contacts in the state, it would be hard to justify a recognition action. But a major judgment debtor will have assets and contacts in many states, even if it’s a foreign corporation (e.g., Honda). Whytock also assumes that FNC is available to resist the recognition of a foreign-country judgment, which is not beyond doubt. Even assuming FNC were available, however, I don’t think it would undermine judgment arbitrage against multinationals and other likely defendants.

The only support Whytock cites for his suggestion that judgment debtors can and should move to dismiss recognition actions on FNC grounds is In re Arbitration between Monegasque De Reassurances S.A.M. v. Nak Naftogaz of Ukraine, 311 F.3d 488 (2d Cir. 2002), a case borrowed from the area of arbitral award enforcement. But for FNC purposes Monegasque De Reassurances is unlikely to be relevant to efforts at judgment arbitrage.

In Monegasque De Reassurances, a Monaco plaintiff brought suit in an American court to enforce a Moscow arbitration award against Ukranian defendants (including the Ukrainian government). The court emphasized that (1) “the jurisdiction provided by the [New York] Convention is the only link between the parties and the United States” and (2) “[m]oreover, there would be great inconvenience in litigating in the United States the complex issues involved in this case.” Id. at 499 (emphasis added). Neither situation would apply in most cases of judgment arbitrage.

For example, say a class of Bangladesh plaintiffs won a Bangladeshi judgment against Wal-Mart for a garment factory fire and sought to enforce that judgment in the U.S. To the Monegasque De Reassurances court’s first point, it would not be difficult to show a link to whatever U.S. forum the plaintiffs chose. Wal-Mart has assets and contacts all over the U.S., and it would be hard for any given U.S. state to bounce a recognition action to another state on FNC grounds. Second, in most actions to recognize a foreign judgment it is unlikely to be much easier to litigate any “complex issues” in any one U.S. state than in any other. Say Wal-Mart alleges that the Bangladeshi court system is corrupt or that trial witnesses were intimidated. It would be hard for Wal-Mart to show that courts in any one U.S. state would be better suited to oversee the resolution of these questions than those in any other.

Thus, even if the case can be extended from arbitral awards to judgments, FNC may not be available where the defendant is American or has significant U.S. ties. Regardless, the case does not demonstrate that FNC would undercut judgment arbitrage, because it would still be hard to argue that one U.S. forum is better than another for the purpose of recognizing a foreign-country judgment.

B. Full Faith and Credit

Finally, Whytock argues that the judgment arbitrage model fails at the enforcement stage because states are not constitutionally obligated to recognize one another’s judgments. Recall that judgment arbitrage is only available (if at all) because the F3 state must enforce the F2 state’s recognition judgment. If states could simply decline enforcement of one another’s judgments, the possibility of judgment arbitrage would be more remote.

The trouble is that they usually cannot—but not, I argue, because of any constitutional requirement. The Full Faith and Credit Clause is “‘not an inexorable and unqualified command’” (p. 489 (quoting Pink v. A.A.A. Hwy. Exp., 314 U.S. 201, 210 (1941)). For this reason the article’s proposed solution is not a constitutional amendment, but a federal statute. This statute would be plainly unconstitutional if states were constitutionally obligated to enforce one another’s judgments. Part III.B.1 of the article thus briefly lays out the argument that states are under no constitutional obligation to recognize one another’s judgments. Instead, the problem—as I argue there and in Part III.A—is a federal statute and uniform act (28 U.S.C. § 1963 and the Uniform Enforcement of Foreign Judgments Act, which is law in 47 states) that effectively dictate the same result. Both laws establish a registration procedure that streamlines the enforcement of judgments across district and state lines: the plaintiff records the judgment with the court clerk and then proceeds to enforce, subject only to sharply limited defenses (e.g., defective process).

Given that Part III.B.1 argues that states are not constitutionally obligated to enforce one another’s judgments, Whytock’s claim that the article fails to “critically examine[]” the “assumption” that U.S. states are required by full faith and credit to enforce one another’s recognition judgments is a little puzzling. The article’s position is clear—they are not constitutionally required to—but as I discuss at pp. 488-91 the precise quantum of obligation that one state owes to another’s judgments and other acts is contested, perhaps more so than Whytock’s argument allows. Regardless of the constitutional position, state and federal courts remain obligated under registration statutes and Whytock provides no reason why they would create an exception where the underlying judgment originates overseas. In fact, the doctrine rejects such a possibility. Under the law of every state, once a foreign-country judgment has been recognized by the local forum, it becomes a judgment of the local forum. The foreign-ness of the judgment disappears.


Judgment arbitrage is probably feasible in enough cases, where enough money is at risk, to make an impact on transnational litigation. I’m very grateful to Whytock for his thoughtful critique, and I hope he and others will continue writing about judgment arbitrage and competing models of recognition law.

I also think the model is useful as a vehicle for analyzing sister-state conflicts of law more broadly. Beyond transnational litigation, the availability of judgment arbitrage illuminates some unexamined aspects of the federal system that merit further attention.

As the article notes, many important social issues like “abortion, gun control, immigration, and the rights of same-sex couples . . . involve horizontal conflicts of state law that are mediated to varying degrees by the federal government” (p. 464-65). Compared to these issues, judgment litigation is not controversial. That no fundamental rights are involved—plus the fact that judgment recognition law is already a mess—may make the area suitable for a federal effort to encourage states to experiment, to create “better law” (however they define it). With a little luck, the results could yield insights into how to manage more serious conflicts among the laws and values of sister states in the future.

Print Friendly, PDF & Email
Notify of
Kevin M. Clermont
Kevin M. Clermont

I think that the theoretical risk of judgment laundering might appear greater if registration statutes are more fully considered. Greg refers to the effect of registration in F−3 as increasing the risk. But registration in F−2 really augments the risk, as it makes F−2 into the ultimately receptive forum:
A crafty plaintiff might register the judgment of the foreign country (F−1) in a Florida state court (F−2), which is one of the places that allow registration of foreign-country judgments.  Registration is easier than suing, not requiring jurisdiction and being unlikely to generate opposition. The plaintiff could then chase assets by registering the Florida registration in any other state (F−3). Because registration is under the radar, enforcement in F−3 might proceed without ever having to face the difficult open questions: whether full faith and credit has an exception for laundered judgments—and whether to accept the more dubious proposition that an F−2 registered judgment is a “judgment” for full faith and credit purposes entitled to be registered in F−3, even if we assume that F−2 had personal jurisdiction and gave adequate notice.
Practicing lawyers in NYC have revealed that they have tried to use this Florida route and plan to step up its usage.


[…] Professor Christopher Whytock, an expert on international litigation at UC Irvine, wrote a very thoughtful response to the article. I replied. […]

Ted Folkman

This is a very interesting paper! I do, though, question the premise. Suppose the defendant is a major US corporation. If such a corporation becomes subject to a judgment of a court in any US state, it is difficult to believe that the corporation would shift assets from one US state to another in order to avoid enforcement or would refuse to pay if its assets were located in another state. Indeed, in light of procedures like the procedures available under FRCP 69 and analogous rules in many state courts, it’s not really clear that a responsible US corporation could shield its assets in that way. But if that’s right, then I wonder whether the premise of your paper is practically meaningful. It seems to me the real issue is, as Chris suggests, whether doctrines like forum non conveniens (or, I would add, personal jurisdiction) provide adequate assurance that the US company is only being sued in an appropriate US jurisdiction.

Greg Shill

Thank you, Kevin and Ted, for your excellent comments. The Florida registration procedure does offer plaintiffs an attractive mechanism for the reasons Kevin notes. That process is still subject to the defenses of the 1962 Act (see Fla. Stat. 55.604 & 55.605), assuming the debtor shows up, but if he does not or loses, then the creditor will have won a US judgment that is likely to be enforced nearly automatically by 47 US states, via the Uniform Enforcement of Foreign Judgments Act. I share Kevin’s skepticism that a Florida-style registration strategy could withstand closer scrutiny – either in F-2 or in any F-3 state presented with the F-2 registered judgment, where F-2 lacked jurisdiction – so if I were a plaintiff that might temper my enthusiasm a little. But for now it appears to be a practical option in some cases. I know Kevin will be following that issue and I will too. If I understand Ted’s point correctly, the real issue is that there are some states that have receptive recognition regimes, i.e., if the plaintiff got one receptive state to order recognition and enforcement of the judgment, then regardless of where its assets were located wouldn’t the company… Read more »