Veteran U.S. trade lawyers Warren Maruyama and Alan Wolff have a new paper out on "Saving the WTO from the national security exception". In their conclusion, they set out three solutions to the "impasse over justiciability of national security claims": (1) allowing Members to "opt out" of dispute settlement for national security cases; (2) relying on a revised non-violation nullification or impairment (NVNI) remedy; and (3) automatic rebalancing.
First up is the opt out:
A sweeping way to avoid the problem is to allow WTO members to opt out of dispute settlement for this class of cases. Were a member to opt out for national security cases, no WTO member could question the invocation by the opting-out member of the exception. To provide balance and symmetry, the member opting out would not be allowed to bring a dispute settlement challenge against other members’ invocations of the exception. While a logically possible outcome, this solution would run contrary to the overall objective of restoring binding dispute settlement for all matters, of having a rules-based international trading system, and of maintaining a balance of concessions, since it would be taken as invitation for the aggrieved member to respond to any extent that they saw fit—in effect, to over-retaliate.
You might think that this is the U.S. position, but this thought is muddled by the U.S. view that NVNI claims, which I am about to turn to, should be the remedy for measures based on national security.
Second, Maruyama and Wolff say that "[a] more promising solution lies in insulating national security claims from panel review while not shielding members from the consequences of invoking the exception." In this regard, they note that "[t]he United States suggested something along these lines, but since the WTO’s nonviolation doctrine has fallen into the trade law’s dustbin and the idea was put forward as part of a US defense of President Trump’s Section 232 tariffs on steel and aluminum, it does not appear to have gotten much attention." They further explain this idea as follows:
In stating that it would not accept adjudication on the issue of national security, the United States suggested that adversely affected countries bring a nonviolation case against it; not to review the legitimacy of the claim, but to restore the balance of concessions:
From the beginning of the trading system, it has been the U.S. view that the appropriate remedy where a Member is impacted by another Member’s essential security measures is to seek a non-violation / nullification or impairment claim. The reality is we designed a WTO where rebalancing could take place without interfering with a Member’s sovereign responsibilities in the area of national security.
Invocation of a nonviolation approach would revive a WTO doctrine that has languished in recent years. The nonviolation nullification and impairment (NVNI) doctrine offers an alternative, and a potentially more workable alternative. It also has major advantages in dealing with politically charged disputes that realistically cannot be resolved through legal proceedings and panel or AB findings.
The judgment of a country invoking the national security exception under Article XXI would be nonjusticiable, just as the drafters of Article XXI apparently intended. However, members adversely affected could be authorized, after litigation or as part of a rebalancing procedure, to suspend substantially equivalent concessions on the grounds that such actions nullify or impair the benefits of trade concessions. The handling of nonviolation claims in these circumstances could be accelerated through a shortened and simplified dispute settlement procedure. A member would be free to act in its essential security interests and the validity of such measures would be nonjusticiable. If one or more members alleged that their trade interests were harmed by the measures, the Director-General would immediately appoint a panel to consider the matter with a 60-day fuse to reach a preliminary determination, with relief retroactive to the date of that determination (if upheld on appeal, and if there is once again an appellate stage to dispute settlement). Since the underlying rationale for the national security claim would be nonjusticiable, the only issue for the panel would be whether the action impairs a tariff concession or otherwise has a negative impact on imports or exports of others, which will almost certainly be the case, and the panel would undertake an assessment of the amount of resulting trade damage. Once that is determined, the parties would seek a negotiated agreement on compensation. This could take place within a shortened time frame, after which, absent agreement, the injured party would be free to suspend equal concessions in order to restore a reciprocal level of benefits.
I have doubts about using the existing NVNI remedy against national security measures because these cases are hard to win. As set out by the Film panel, the general NVNI standard is as follows:
10.41 ... The text of Article XXIII:1(b) establishes three elements that a complaining party must demonstrate in order to make out a cognizable claim under Article XXIII:1(b): (1) application of a measure by a WTO Member; (2) a benefit accruing under the relevant agreement; and (3) nullification or impairment of the benefit as the result of the application of the measure
The main difficulties with NVNI claims, as I see it, are the following elements (as the Film panel explained them):
10.30 In a case of non-violation nullification or impairment pursuant to Article XXIII:1(b), Article 26.1(a) of the DSU and GATT jurisprudence confirm that this is an exceptional remedy for which the complaining party bears the burden of providing a detailed justification to back up its allegations.
...
10.76 ... in order for expectations of a benefit to be legitimate, the challenged measures must not have been reasonably anticipated at the time the tariff concession was negotiated.
Maruyama and Wolff seem to want a revision to the NVNI standard for the purposes of challenges to national security measures. If they can get the new faster process they are looking for, and in doing so remove the higher burden of proof and the "reasonably anticipated" requirement (and perhaps also clarify the "benefit" issue), it might work. My concern here is that once you open this standard up for discussion, you will hear a lot of different views and it will be hard to reach consensus. Nevertheless, if someone can convince the Biden administration to accept this approach, I think it would be a big improvement over where we are now.
I'm skeptical that the administration would agree to this, though. If I understand the U.S. view correctly, and I'm not totally sure I do, they want to applying existing NVNI standards, under which, as noted, it is hard to succeed on a claim.
To be clear, I fully support an effort to rework NVNI to make it helpful here, a project that would probably include a deep dive into the origins, purpose, and nuances of the NVNI remedy. Based on what I've seen, however, I worry that there are not going to be clear answers and everyone will get bogged down.
Finally, a third approach suggested by Maruyama and Wolff is automatic rebalancing along the lines of what I've suggested:
Alternatively, if a WTO reform agenda actually gets underway, the WTO could adopt a renegotiation procedure along the lines of GATT Article XXVIII, in which national security actions could be notified to the WTO and subject to accelerated procedures for compensation negotiations. If necessary, where trade compensation could not be agreed, the member whose trade was adversely affected could rebalance the level of trade concessions itself by imposing retaliatory trade measures. There would be the potential for panel review if the rebalancing is deemed excessive. This would largely eliminate the need for a panel process. This would be somewhat analogous to GATT Article XXVIII, which provides for withdrawal, renegotiation, and rebalancing of previously agreed tariff concessions. If a member determines that a previously agreed tariff concession is no longer workable, GATT Article XXVIII provides a procedure for it to notify other members of its intent to withdraw the concession. This is followed by a fast-track trade negotiation to determine if the parties can agree on appropriate compensatory reductions in other tariffs to restore the balance of concessions. If the parties cannot agree, the member can still withdraw the concession, but the injured parties have a right to rebalance by raising their tariffs on an equivalent amount of trade. Under this scenario, if a WTO member invokes the Article XXI national security exception to impose import restrictions, countries whose trade is adversely affected would have the right to immediately withdraw substantially equivalent concessions (that is, limit the imports from the invoking country in like amount) without any panel proceeding or other authorization from the WTO if an agreement on compensation could not be reached within a relatively brief period, e.g., 60 days.
If the invoking country felt the response was excessive, it could bring a case to the WTO for adjudication on the issue of what was needed to rebalance the preexisting level of concessions. In our view, the right of an equivalent response should be automatic, without requiring additional litigation, other than perhaps on whether the level of response was calibrated correctly. Of course, the member invoking Article XXI could offer trade compensation and avoid others acting to limit its trade. That would be a matter for negotiation.
This approach would offer a cleaner, simpler, and quicker mechanism for resolving national security disputes without the baggage of having panels issue findings on whether such measures are justified by war, an international emergency, or traffic in arms, while more effectively discouraging spurious invocations of Article XXI by imposing a clear, timely price for members who invoke it.
Having proposed something along these lines myself, it won't surprise anyone to hear that I like this approach! Developing a process for it is sure to be complicated, but I think it could work. Inu and I tried to flesh it out a bit here.
