Further Thoughts on Necessity as a Reason to Defer Sovereign Debt Ob...

Mitu and I posted some preliminary thoughts about the defense of necessity, which might be raised as a basis for allowing sovereign borrowers to defer debt service during the crisis. I wanted to follow up on some of the open issues. A few are technical, addressing some potential objections to the defense. I’ll deal with these first and close with a more fundamental question: What good does this potential defense really do for a sovereign? In thinking through that question, my premise is that many sovereigns will need a temporary standstill on debt service during the crisis. For proposals to this effect, see here, here, and here. (Others will eventually need a debt restructuring, but that’s a topic for another day.) But of course private creditors must agree to a standstill on payments. Those who don’t might sue or file arbitration claims, which will potentially put the sovereign's assets at risk and will certainly consume time and resources to defend. [Last sentence edited for clarity.]

Some background

Necessity is a rule of customary international law. As expressed in Article 25 of the International Law Commission’s draft Articles on Responsibility of States for Internationally Wrongful Acts, a state can invoke necessity to excuse its non-performance of an “international obligation” if non-performance is the only way to address “a grave and imminent peril,” as long as non-performance does not seriously impair an essential interest of the “State or States towards which the obligation exists.” Even if these conditions are satisfied, the state cannot invoke necessity to excuse the violation of an international obligation that “excludes the possibility of invoking necessity.” (Put differently, the doctrine purports to treat necessity as a default rule.) Nor can a state invoke the defense if it has contributed to the state of necessity. Finally, even if the defense is available, non-performance is excused only while the threat persists. The state must resume performance when the crisis ends, and it may have to pay compensation for any loss caused by its non-compliance.

It may not be obvious, but this is a remarkably crabbed conception of “necessity.”

For example, under U.S. law I am allowed to enter your property without your consent if this is necessary to save my life or avoid some other serious harm, although I will likely have to compensate you for any harm I do in the process. This is so even if the danger I was trying to avoid was “the result of [my] own tortious conduct or contributory negligence.” (Quoting here from the comments to section 197 of the Restatement (Second) of Torts, and ignoring some nuance and exceptions.) This makes sense, right? The whole point is that, in the circumstances covered by necessity, to respect the other party’s rights would be the greater of two evils. The doctrine’s fundamental insight is, like, that’s the wrong thing to do.* Pretty simple and intuitive. Because the rights-holder is entitled to compensation for any loss, it isn’t in a good position to argue that the violator should just suck it up and endure whatever grievous harm is threatened.

Measured against this conception, the CIL understanding of necessity is exceptionally narrow. The sovereign can’t have contributed to its conduct. Non-compliance must be the only way to prevent the threatened harm. And the harm itself must be existential, or nearly so. All of these limits protect (arguably, over-protect) against the obvious moral hazard involved in a rule that allows a sovereign to escape its obligations.

Another limitation comes from the implication that the necessity defense is a default rule, unavailable if the obligation being violated “excludes the possibility of invoking necessity.” This is not an obvious candidate for a default rule, if only because it is implausible to think that, in a crisis, a state will honor the commitment to prioritize less important obligations. Why should a state be able to sell its right to prioritize the welfare of its residents over other, concededly less important obligations? But the puzzle recedes somewhat when we consider that the doctrine’s early roots often involved state-to-state disputes (i.e., “international obligations”) and armed conflict. Here, for instance, is the example given in the commentary to Article 25: “[C]ertain humanitarian conventions applicable to armed conflict expressly exclude reliance on military necessity.” Okay, that’s a context in which it would make sense to treat necessity as a default rule. But presumably we should be cautious about treating necessity as a default in all contexts. For example, if the defense is available in contract disputes involving private creditors (see below), it’s less obvious that it should be viewed as a default rule that the contract can obviate.

Some technical questions

With this background, we can think through some of the technical questions that may come up if a foreign sovereign invokes necessity to defend against a lawsuit filed by a private creditor in a national court. To my knowledge, this scenario is very uncommon. I don’t know of any directly relevant law in the U.S. Venezuela has raised necessity as a basis for seeking a stay in a lawsuit filed by Casa Express in the Southern District of New York. No ruling on that as yet, and the judge can decide the case without ruling on necessity specifically. Anyway, I’ll focus on U.S. law, since that’s what I know best. (Cue jokes about how little that is…)

My bond is governed by New York law. Why are you babbling about customary international law?

Casa Express argues in opposing Venezuela’s stay motion that customary international law is irrelevant when the contract is governed by New York law. I suspect many private creditors share this belief. It’s probably wrong.

Courts generally understand CIL to be part of federal law, which preempts contrary state law. That isn’t the only way to understand it, as Curt Bradley and Jack Goldsmith famously argued. One might instead think that CIL is relevant, in the absence of a statute or treaty, only when incorporated into state law. But either way, a generic governing law clause of the sort found in most contracts—“any dispute arising under or in connection with this contract will be governed by the law of New York”—leaves room for the parties to invoke relevant CIL. New York law has long been understood to incorporate customary international law when relevant (e.g., in the Bergman case, an important early case on diplomatic immunity). So even under the narrower view of CIL, a governing law clause designating New York law includes CIL.

Again, however, courts usually understand CIL to be part of federal law. Does a generic governing law clause implicitly exclude this aspect of federal law? Almost certainly not. The closest analog here is to international contracts for the sale of goods, which potentially could be governed either by a state’s version of U.C.C. Art. 2 or by the Convention on Contracts for the International Sale of Goods (CISG), a federal treaty. With very few exceptions, courts interpret generic governing law clauses to incorporate the CISG, on the theory that the law of the designated state includes any supervening federal law. My colleague John Coyle has written insightfully about the interpretation of choice of law clauses here.

What part of “international” don’t you understand?

Recall that the classic statement of the necessity defense in ILC Article 25 refers to non-performance of an “international obligation”—i.e., an obligation owed by one state to another. Indeed, some of the elements of the defense become hard to apply or irrelevant when a private creditor asserts a contract claim against a state. For example, in such a dispute, what does it mean to ask whether the state’s non-compliance with the contract “seriously impair[s] an essential interest of the State or States towards which the obligation exists, or of the international community as a whole”? There is no obligation to another state, so the first part of the question is non-sensical. Oddities like this don’t automatically mean the defense is unavailable; the doctrine is expansive enough to accommodate contract disputes with private creditors. But they highlight the doctrine’s history as a tool for resolving inter-state disputes.

Is the necessity defense available in a contract dispute with a private creditor? The German Constitutional Court said that it was not (English translation by Google). To reach this outcome, the court had to figure out how to distinguish all of the ICSID arbitrations in which Argentina was allowed to assert the defense in response to claims by private creditors. In those disputes, however, the creditor’s home country had entered a bilateral investment treaty with Argentina. Technically, the creditor asserted a claim under the treaty, not a contract claim, and treaty compliance is an international obligation. This distinction was enough for the majority of the Constitutional Court (over a vehement dissent).

I find this result quite odd. I have to concede, of course, that Article 25 has inter-state disputes in mind (hence, “international obligations”). Likewise, the history of the necessity defense is one of inter-state disputes. To be sure, many of these disputes involved claims by private creditors, such as this early-20th century arbitration involving Venezuela’s unpaid obligations to a French company. But in every case (at least, every case I’m aware of) the private creditor was able to bring its claim only because its home state had negotiated an inter-state agreement allowing it to do so. This history arguably lends support to the German court’s decision.

Still, there is a strong argument for extending the necessity defense to this context. I don’t assign too much significance to the ILC’s use of “international obligations” in Article 25. Legal rules are expressed in language that reflects their historical context; this does not necessarily imply that the rule intends to exclude its application in new contexts. I also think we should be careful not to read too much into the doctrine’s history of being applied against private creditors only in disputes that qualified as inter-state. Throughout most of this history, the doctrine of absolute immunity barred private suits against sovereigns. It is not as if tribunals routinely refused to apply the doctrine in private contract disputes. Those disputes almost never occurred. Moreover, although technically in an inter-state context, tribunals routinely considered the defense—mostly rejecting it, but at least considering it—in disputes involving private creditors aggrieved by the state’s non-performance of a contract. Now that private creditors can sue foreign sovereigns in national courts, little in its history suggests that the necessity defense should be unavailable in this functionally-identical category of disputes.

Indeed, denying states the necessity defense in contract disputes with private creditors results in strange priority distinctions among creditors. It effectively subordinates official creditors to most private creditors, because only the former risk having their claims deferred. It also subordinates one group of private creditors to another, functionally-identical group. This is because private claimants who pursue arbitration under a BIT can see their claims deferred; creditors who file lawsuits need not worry. These consequences would be worth tolerating if they made sense, or if the law of necessity clearly required them. But they don’t, and it doesn’t.

Necessity isn’t CIL in *this context*

In its dispute with Venezuela, Casa Express argues that, sure, necessity is a general rule of CIL or whatever, but it’s not a rule of CIL in this context. Or something like this. The argument is confusing (pp. 17-18 here). A rule of CIL exists when it represents the general and consistent practice of states, and when states observe the practice because they think it is obligatory. Everyone agrees that the doctrine of necessity satisfies these requirements. But, Casa Express argues, it has not been generally accepted that necessity can be used as a defense to a contract claim by a private creditor.**

This is clever, but more than a bit cynical. The argument invites the court to mistakenly construe uncertainty about the scope of a rule as evidence that the rule itself is not widely accepted. But no one believes that a rule achieves the status of CIL only when it has been generally and consistently observed in the particular (and ultimately unique) context implicated by each dispute. To be sure, some purported CIL rules might be framed in such abstract terms that a skeptic might wonder whether the rule actually describes any consistently-observed practice. In that case, one might question whether the rule merits the status of CIL. But this isn’t the case with the doctrine of necessity. Yes, there is a dispute about the scope of the rule, and in particular about whether the defense is available in contract disputes brought by private creditors. That uncertainty will eventually be resolved, likely through adjudication. But not if Casa Express has its way.

A closing comment on the importance of necessity

There’s plenty more to talk about regarding the merits of the necessity defense, but I’ll leave it there. Let me close, though, with a more fundamental question, which is whether the necessity defense can play a meaningful role in creating an effective debt standstill. Even if the defense can be raised, it’s hardly a slam dunk. So won’t lots of private creditors file lawsuits or arbitration claims rather than agree to a standstill?

Maybe? Probably? I dunno. But if the necessity defense can be raised, there is reason to think the volume of litigation will be materially lower.

I’m agnostic about how much litigation we can expect to see if a country declares a standstill. Let’s assume the standstill is implemented as proposed here (also linked above) by Patrick Bolton, Lee Buchheit , Pierre-Olivier Gourinchas, Mitu Gulati, Chang-Tai Hsieh, Ugo Panizza, and Beatrice Weder di Mauro. This means that official sector actors will publicly endorse the need for a standstill and declare a state of necessity to exist. The delay and uncertainty inherent in suing a sovereign will be even greater here, and not just because COVID-19 has closed courts and lengthened the litigation queue. It’s not unreasonable for a state to seek a temporary standstill on payments from its creditors. As a matter of litigation optics, creditors who refuse and instead file lawsuits will be starting behind the 8-ball. To put the point a bit less cynically, I suspect that the large majority of commercial and financial creditors are willing to play a role in ending the crisis; they will mostly expect fair treatment and clear guidance on the plan for emerging from the crisis. For those who choose instead to file suit, judges have a great deal of soft power, which they can use to control the litigation timeline and to encourage compromise. Because of this, I may be more optimistic than many that the volume of litigation will be manageable, whether or not states can assert (much less win on) the defense of necessity.

Still, if the necessity defense is taken seriously—as I think it should be—it can significantly reduce expected litigation payouts. At least in the sovereign debt context, creditors are usually assured of winning on the merits. The primary uncertainties involve the time and expense of enforcing a money judgment. But with necessity, the merits are suddenly in play. I don’t like to handicap these things, and the necessity defense is hardly a slam-dunk, but in the remarkable circumstances of this global pandemic, it’s no long-shot either. And recall: the state won’t be asking for debt forgiveness. The defense simply entitles it to a delay. The risk that a court will accept the defense will provide at least some additional disincentive to sue.

Some might see the limited relief afforded by the doctrine of necessity—delay, not forgiveness—as a weakness. But it can just as easily be seen as a virtue. If a court believes that the present circumstances are truly unique—such that accepting the defense will not seriously undermine the enforceability of debt contracts generally—then the defense arguably strikes an appropriate balance between the state’s need to protect its residents and the concerns of creditors.

*We can put this in efficiency terms. The “access plus compensation for loss” result roughly approximates the deal you and I would strike if we had time to bargain over my access to your property (although our hypothetical deal would have given you a bit of profit). Of course, you would have been in position to drive a pretty hard bargain, but the law doesn’t generally allow a person to exploit another’s exigency in that way.  

** It also argues that 28 U.S.C. 1606 bars consideration of necessity by providing that, when a foreign state is not immune, it is liable “in the same manner and to the same extent as a private individual.” According to Casa Express, this means that, unless the defense of necessity is available to private defendants, it can’t be available to a foreign state. As a textual matter, I suppose one can read the statute this way. But why? The reading isn’t required by the text or legislative history (which reflects a central concern with remedies) and is bizarre. The FSIA does not assimilate standards of conduct applicable to foreign states to the standards applicable to private parties. If the underlying substantive law makes a defense available to states but not private actors, the statute has no problem with that. Why would it?