International Economic Law and Policy Blog

Changing China's Trade Practices: Hope vs. Unilateral Tariffs vs. WTO Complaints

Former Biden administration official Peter Harrell recently argued that with regard to reform of China's economic practices (and political system), the original plan was to hope that economic ties would be enough, but when that didn't work the Trump administration imposed tariffs for this purpose: 

For many years, the strategic paradigm of U.S. trade policy toward China was defined by the hope that economic ties would persuade China to continue on a course of gradual economic and political liberalization. By the mid-2010s, it was clear that this paradigm had failed, prompting Trump to impose sweeping tariffs in a bid to generate negotiating leverage to compel China to change its economic model. China responded to the tariffs, initially with retaliatory tariffs on U.S. exports like agricultural products, and ultimately Trump offered tens of billions of dollars in assistance to U.S. farmers to offset the impacts of Chinese tariffs. Later, China offered a handful of concessions as part of a “Phase 1” trade deal, largely to avoid a threatened future tariff escalation. However, China steadfastly refused to make more fundamental changes to its economy and market, and there is no evidence that China’s willingness to reform has increased in the years since.

People hoping for China to evolve on its own in terms of economic and political liberalization may have been looking at South Korea and Taiwan as models. That wasn't totally unreasonable, but the parallels between South Korea/Taiwan and China are limited, so a similar path for China may not have been likely.

With regard to economic liberalization specifically, rather than wait for China to change on its own, a more active strategy of pushing for economic reform was probably necessary in the years after China joined the WTO. That would have been a good foreign policy priority for the U.S. at that time.

Actually, it would still be a good priority for the U.S., as I'm not sure Chinese economic liberalization is a U.S. priority even now. As I recently wrote in a short piece for the Baker Institute, it seems clear that the Section 301 tariffs have not accomplished much on these issues. Perhaps they were intended for this purpose, but unilateralism of this sort has not had much success in the past, and the evidence is accumulating that it still does not work very well. Section 301 tariffs might be a way to protect U.S. producers from Chinese competition, but they aren't doing much to induce policy change in China.

As an alternative, I think WTO complaints would be more effective, although of course it does not seem likely that USTR will bring such complaints any time soon. Australia, the EU, and Japan have had a few recent complaints against China, but the big systemic issues related to non-market practices have not not addressed. It's hard to figure out what everyone's strategy is here, but as I see it, China made some broad commitments when it joined the WTO, and it seems like it could be useful to try to hold China to them. Here's how I concluded the Baker piece:

Sometimes “toughness” is judged by attitude, but results are what matter most. It is increasingly clear that U.S. unilateralism is coming up short on effectiveness. China committed to obligations tougher than any unilateral measures, and yet successive U.S. administrations have underutilized WTO enforcement — with the Biden administration totally abandoning it. 

This is a mistake. If we are serious about pushing China to play an appropriate role in the world trading system — one commensurate with its status as an “upper-middle-income country” — we should make this a priority and hold China to the obligations it agreed to.

SIEL Call for Papers: Navigating New Horizons: International Economic Law in a Changing World

This is from SIEL:

The Ninth Biennial Global Conference of the Society of International Economic Law (SIEL) will take place on 9-11 July 2025 in Taipei (Taiwan), in collaboration with National Taiwan University (NTU). SIEL’s Ninth Biennial Global Conference will be in person and address the pressures of escalating geopolitical tensions, environmental challenges, unbridled economic nationalism, profound digital transformations, and growing inequalities are reshaping not just economies, but societies globally.

More information, including how to submit a paper, panel or poster is available at https://sielnet.org/siel-2025-announcement-of-date-and-venue/. 

Still Puzzling over Privacy Exceptions in Digital Trade Chapters/Agreements

There seems to be a lot of experimentation going on these days with exceptions for the protection of personal data/privacy in digital trade agreements/chapters. I've talked about the some of these exceptions in prior posts, but to be honest I think I missed some nuances as I was looking at the bigger picture of data flows obligations and exceptions. In this post, I'm going to focus on several examples of exception provisions that refer to concerns around privacy, and then one additional example of exceptions that could cover privacy without mentioning it explicitly.

First up is the EU-Japan Economic Partnership Agreement. A recently added Annex to that agreement establishes obligations related to data flows in paragraphs 1 and 2, and then right after these obligations it sets out the following public policy exceptions:

...

3. Nothing in this Article shall prevent a Party from adopting or maintaining measures inconsistent with paragraphs 1 and 2 to achieve a legitimate public policy objective², provided that the measure:

(a) is not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination between countries where like conditions prevail, or a disguised restriction on trade; and
(b) does not impose restrictions on transfers of information that are greater than necessary to achieve the objective.³

4. Nothing in this Article shall prevent a Party from adopting or maintaining measures on the protection of personal data and privacy, including with respect to cross-border transfers of information, provided that the law of the Party provides for instruments enabling transfers under conditions of general application⁴ for the protection of the information transferred.

...

-----------------

^{2 For the purpose of this Article, "legitimate public policy objective" shall be interpreted in an objective manner and shall enable the pursuit of objectives such as the protection of public security, public morals, or human, animal or plant life or health, or the maintenance of public order or other similar objectives of public interest, taking into account the evolving nature of digital technologies.
3 For greater certainty, this provision does not affect the interpretation of other exceptions in this Agreement and their application to this Article and the right of a Party to invoke any of them.
4 For greater certainty, in line with the horizontal nature of the protection of personal data and privacy, "conditions of general application" refer to conditions formulated in objective terms that apply horizontally to an unidentified number of economic operators and thus cover a range of situations and cases.}

What's confusing here is that paragraph 3 looks like a fairly typical public policy exception, but then right after it is paragraph 4, which covers one specific public policy (personal data and privacy) and works differently than paragraph 3.

One key difference is that paragraph 3 makes explicit reference to inconsistencies with paragraphs 1 and 2, whereas paragraph 4 does not. That omission has to mean something, but what exactly? Does it limit the application of paragraph 4 as an exception to the data flows obligations of paragraphs 1 and 2 in some way?

Another difference is that paragraph 4 doesn't have typical qualifiers such as a necessity test or a requirement that it not be administered in a manner that constitutes arbitrary or unjustifiable discrimination. It does, however, have one that says: "provided that the law of the Party provides for instruments enabling transfers under conditions of general application for the protection of the information transferred." This seems narrower than a general non-discrimination obligation. Does this suggest that a government has more discretion under paragraph 4 than under paragraph 3? If so, how much discretion? Is it a simple means-ends test that considers objectively whether the measures at issue are "on the protection of personal data and privacy," and then turns to the proviso for "enabling transfers under conditions of general application"?

Next up is the EU-NZ FTA digital chapter, which I discussed here (and which is what got me thinking about privacy exceptions).  The data flows obligations are in Article 12.4, and then Article 12.5 on protection of personal data and privacy states:

Protection of personal data and privacy

1. Each Party recognises that the protection of personal data and privacy is a fundamental right and that high standards in this regard contribute to enhancing consumer confidence and trust in digital trade.

2. Each Party may adopt or maintain measures it deems appropriate to ensure the protection of personal data and privacy, including through the adoption and application of rules for the cross-border transfer of personal data. Nothing in this Agreement shall affect the protection of personal data and privacy afforded by the Parties' respective measures.

3. Each Party shall inform the other Party about any measures referred to in paragraph 2 that it adopts or maintains.

4. Each Party shall publish information on the protection of personal data and privacy that it provides to users of digital trade, including:

(a) how individuals can pursue a remedy for a breach of protection of personal data or privacy arising from digital trade; and

(b) guidance and other information regarding compliance of businesses with applicable legal requirements protecting personal data and privacy.

As I said as an addition to the earlier post on the EU-NZ FTA, while an exception doesn't have to be labeled as an exception, if it is not labeled as such, then it would be useful to tie it to the obligation for which it provides an exception in a clear way. For example, the exception could be in the same article as the obligation, and then be tied textually with wording such as "nothing in this article shall prevent a party from taking a measure to ... ". While this is not strictly necessary, it would still be helpful to have a tie between the supposed exception and the obligation. Here, paragraph 2 says "[e]ach Party may adopt or maintain measures it deems appropriate to ensure the protection of personal data and privacy, including through the adoption and application of rules for the cross-border transfer of personal data." That language does have a tie to the cross-border data flows obligations of Article 12.4, so that is an element in favor of Article 12.5 being treated as an exception.

On the other hand, instead of using "nothing in this agreement shall prevent" language, Article 12.5, paragraph 2, sentence 1 talks about what a party "may" do. That seems a bit softer to me, and more like a "right to regulate" provision, which tend not to have much impact. In addition, sentence 2 reads more like a traditional exception, albeit with some different wording: "Nothing in this Agreement shall affect the protection of personal data and privacy afforded by the Parties' respective measures." That's a variation on the traditional "nothing in this agreement shall prevent"-style language, and also seems a bit softer to me. What does it mean to say the agreement shall not "affect" these things? Does this involve some kind of "means-ends" test as well? Would the adjudicator have to determine whether the measures protect personal data and privacy, and then ensure that the agreement does not "affect" (undermine?) this protection?

Third, we have the March 2024 Draft Chair's Text of the WTO E-Commerce JSI, which uses a version of the EU-NZ text but relies on the more traditional "nothing shall prevent" language and labels the provision as an "exception":

Article 7: Personal Data Protection Exception

Nothing in this Agreement shall prevent a Party from adopting or maintaining measures on the protection of personal data and privacy, including with respect to cross-border data transfers, provided that the law of the Party provides for instruments enabling transfers under conditions of general application¹ for the protection of the data transferred.

1 For greater certainty, "conditions of general application" refer to conditions formulated in objective terms that apply horizontally to an unidentified number of economic operators and thus cover a range of situations and cases.

Of course, there are no data flows obligation in the JSI text at the moment, so that changes the role and impact of this exception.

Finally, let's look at the RCEP. First, in the e-commerce chapter, the RCEP offers the following instruction to adopt personal information/privacy protections:

Article 12.8: Online Personal Information Protection

1. Each Party shall adopt or maintain a legal framework which ensures the protection of personal information of the users of electronic commerce.^{7, 8}

8 For greater certainty, a Party may comply with the obligation under this paragraph by adopting or maintaining measures such as comprehensive privacy or personal information protection laws and regulations, sector-specific laws and regulations covering the protection of personal information, or laws and regulations that provide for the enforcement of contractual obligations assumed by juridical persons relating to the protection of personal information.

But this kind of provision is different from ones that contain data flows obligations and personal data/privacy exceptions.

In addition, there are a number of exception provisions of relevance here, although they don't mention privacy explicitly, and I'm including these provisions in order to illustrate that privacy could be covered by broader public policy/security exceptions provisions. In the e-commerce chapter, we have the following obligations and exceptions:

Article 12.14: Location of Computing Facilities

...

2. No Party shall require a covered person to use or locate computing facilities in that Party’s territory as a condition for conducting business in that Party’s territory.¹¹

3. Nothing in this Article shall prevent a Party from adopting or maintaining:

(a) any measure inconsistent with paragraph 2 that it considers necessary to achieve a legitimate public policy objective,¹² provided that the measure is not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination or a disguised restriction on trade; or

(b) any measure that it considers necessary for the protection of its essential security interests. Such measures shall not be disputed by other Parties.

-----

^{12 For the purposes of this subparagraph, the Parties affirm that the necessity behind the implementation of such legitimate public policy shall be decided by the implementing Party}

...

Article 12.15: Cross-border Transfer of Information by Electronic Means

... 

2. A Party shall not prevent cross-border transfer of information by electronic means where such activity is for the conduct of the business of a covered person.¹³

3. Nothing in this Article shall prevent a Party from adopting or maintaining:

(a) any measure inconsistent with paragraph 2 that it considers necessary to achieve a legitimate public policy objective,¹⁴ provided that the measure is not applied in a manner which would constitute a means of arbitrary or unjustifiable discrimination or a disguised restriction on trade; or

(b) any measure that it considers necessary for the protection of its essential security interests. Such measures shall not be disputed by other Parties.

---

^{14 For the purposes of this subparagraph, the Parties affirm that the necessity behind the implementation of such legitimate public policy shall be decided by the implementing Party.}

And then there is this in the exceptions chapter:

Article 17.13: Security Exceptions

Nothing in this Agreement shall be construed:

...

(b) to prevent any Party from taking any action which it considers necessary for the protection of its essential security interests:

...

(iii) taken so as to protect critical public infrastructures⁷ including communications, power, and water infrastructures;

------

7 For greater certainty, this includes critical public infrastructures whether publicly or privately owned.

The RCEP public policy and security exceptions could be used to cover privacy issues, and provide discretion along the same lines as what the GATT exceptions do. But perhaps of more importance here is that dispute settlement does not apply to the RCEP e-commerce chapter: "No Party shall have recourse to dispute settlement under Chapter 19 (Dispute Settlement) for any matter arising under this Chapter."

So what's the big takeaway from all this? I guess one thing is that if litigation on these issues ever comes, adjudicators are going to have some challenges in defining the scope of these privacy exceptions. The wording differs from traditional trade agreement exceptions in a number of ways, and adjudicators will have to decide what those differences mean. I would imagine they will end up at a place with a good deal of deference to domestic regulators, while stopping short of saying the exceptions are totally self-judging. But where specifically they end up and how they get there will be interesting.

Guest Post: IMO Carbon Pricing Mechanism, the Equitable Transition of International Shipping, and WTO Law

This is a guest post from Goran Dominioni, Assistant Professor, School of Law and Government, Dublin City University 

International maritime transport is the lifeblood of the global economy, accounting for about 80 percent of international trade (by volume). In July 2023, the International Maritime Organization (IMO)—the UN agency that regulates international maritime transport—reached an agreement to reduce greenhouse gases (GHGs) from the sector. The resulting 2023 IMO GHG Strategy envisions making the sector climate-neutral by or around 2050 and includes relatively stringent GHG interim checkpoints for 2030 and 2040.  

The IMO is now working on adopting a basket of measures to reduce GHGs in alignment with the new target. The basket of measures is scheduled for adoption in Spring 2025 and is expected to include a GHG pricing mechanism (e.g., a GHG levy or a feebate scheme). If adopted, this could be the first global GHG pricing mechanism that is not simply an offsetting scheme.  

In a previous round of negotiations, India and Saudi Arabia made a submission to the IMO outlining possible incompatibilities of an IMO GHG pricing mechanism with WTO law. Responding to a request by the IMO Secretariat, the WTO Secretariat then identified WTO law provisions that could be relevant in implementing an IMO GHG pricing instrument. To date, however, the WTO Secretariat has not provided a conclusive answer on the matter (perhaps not surprisingly). 

In a recent article, I analyzed the risk of WTO member states challenging national or EU measures to implement or enforce an IMO GHG pricing mechanism on the grounds that they violate GATT and GATS provisions.  

Currently, several proposals are tabled for discussion at the IMO. Although the precise character of a IMO GHG pricing mechanism is yet to be determined, it is possible to make a preliminary assessment on the basis of these proposals. Below, I focus on three key questions:  

• Who could challenge national or EU measures to implement or enforce an IMO GHG pricing mechanism? 
• What GATT and GATS provisions could be violated by national or EU measures adopted to implement or enforce an IMO GHG pricing mechanism?  
• If exemptions are included in the design of the GHG pricing mechanism, would national or EU measures adopted to implement or enforce it become easier to challenge?  

Overall, the analysis indicates that the risks of a successful challenge are low as only a few countries would be in a position to bring a challenge, and the grounds for challenging the EU/national measures would be very limited. However, these degree of risk will depend on how the instrument is designed and on the procedure used to adopt the IMO GHG pricing mechanism. I discuss some of the details below.  

Who could challenge national or EU measures to implement or enforce an IMO GHG pricing mechanism? 

At the last round of IMO negotiations at the Marine Environmental Protection Committee (MEPC), IMO member states indicated a preference for adopting the IMO GHG pricing mechanism via an amendment of Annex VI of the International Convention for the Prevention of Pollution from Ships (MARPOL Annex VI). If this path is followed, only countries that have not ratified MARPOL Annex VI would be in a good position to bring a challenge, as normally countries can not challenge measures adopted to implement or enforce a convention they have ratified. As of January 2024, 71 countries have not ratified MARPOL Annex VI. Of these, very few have a history of bringing disputes to the WTO.  

Note that vessels and goods of countries from non-MARPOL Annex VI countries would (most likely) still be subject to the IMO GHG price because the IMO tends to operate based on the ‘no more favourable treatment’ (IMO NMFT) clause, according to which port states under an IMO convention will apply the convention also to vessels registered in countries not part of that convention.  

Usually, the IMO tries to operate on a consensus basis. However, when consensus can not be reached, IMO member states can decide through a voting procedure. In previous negotiations, GHG measures were adopted despite China, Chile, Brazil, Kuwait, and Saudi Arabia voting against them. If the GHG pricing mechanism is adopted through a voting procedure, the countries that oppose its adoption may try to bring a WTO challenge. However, their position would be weak, as they have accepted to be bound by the procedural rules for amending MARPOL Annex VI.  

If the IMO GHG pricing mechanism is adopted through a new convention (which looks unlikely at this point), the number of countries that could bring a challenge is difficult to predict, as it depends on the number of countries that ratify the convention.  

What GATT and GATS provisions could be violated by national or EU measures adopted to implement or enforce an IMO GHG pricing mechanism?  

Based on the current proposals tabled for discussion at the IMO, there are a few potential ways in which EU/national measures could violate GATS or GATT provisions. A key reason for this is that all these proposals envision that the GHG price will be paid by vessels directly into an international fund without being levied by the port or flag state. Because of this feature, many of the provisions identified by the WTO Secretariat as potentially relevant for developing an IMO GHG pricing mechanism are unlikely to apply. These include GATT Art I:1, GATT Art III:2, GATT Art VIII.  

Research indicates that a GHG pricing mechanism could reduce the profitability of shipping companies. One of the critical points of contention in IMO negotiations on a GHG pricing mechanism is whether the price should apply universally (i.e., to all emissions from vessels of a specific size or larger) or only to ships that do not meet a preestablished GHG fuel intensity target. The GHG fuel intensity of the fleet of different countries may vary systematically, depending on the type of vessels and routes covered (on some routes, zero and near-zero GHG bunker fuels may be less available than on others). Potential negative impacts of the GHG pricing mechanism on the fleet of different countries may vary depending on how the GHG fuel intensity target is set. Countries could question how such a target was chosen and bring a challenge for violation of MFN under the GATS. Measures that implement or enforce a universal GHG pricing mechanism are less contestable as they apply the same price per tonne of GHG.  

In IMO, the GHG pricing mechanism will also increase the price of goods traded by sea. While potentially low on average, impacts will likely be uneven, with high-weight, low-value goods and countries less well-connected with their export markets likely to be negatively impacted more. Work is ongoing at the IMO to quantify such impacts. To the extent that some countries are impacted more than others, they may argue that measures that implement or enforce the GHG pricing mechanism violate Article XI:1 GATT (quantitative restrictions on imports/exports).  

Any potential violation of MFN under the GATS and Article XI:1 GATT could be justified under Articles XIV GATS and Article XX GATT. In this respect, the GHG pricing mechanism should be, inter alia, designed to be climate-effective and have the lowest possible trade restrictiveness. To this end, a share of the significant revenues that could be raised through the GHG pricing mechanism (up to 40-60 billion US dollars a year until 2050) could be used to reduce disproportionately negative impacts on states.  

If exemptions are included in the design of the GHG pricing mechanism, would national or EU measures adopted to implement or enforce it become easier to challenge?  

Some IMO member states—including China, Brazil, and Argentina—proposed a GHG pricing instrument that is less stringent for vessels that call at ports of certain developing countries. This adjustment would result in lower or no payment made by vessels that serve the selected ports. This feature of the proposal is meant to help address equity-related concerns in the energy transition of shipping, such as avoiding disproportionately negative impacts on developing countries.  

Including a partial exemption of this type in the GHG pricing instrument will likely make EU/national measures that implement or enforce it more susceptible to a successful challenge under the GATT and the GATS for two reasons. On the one hand, such exemptions could be infringing the MFN principle under the GATS. On the other hand, exemptions (even partial ones) risk undermining the climate effectiveness of the GHG pricing instrument. Besides lowering the climate stringency of the instrument overall, they also risk enabling avoidance opportunities for vessels. The GHG effectiveness of the instrument is one of the criteria for the applicability of the justifications under Article XIV GATS and Article XX GATT. Furthermore, despite calls to change this approach, WTO jurisprudence tends to equate differentiation with discrimination under Article XX GATT.  

In the IMO context, the LDC waiver is unlikely to be helpful in implementing exemptions for LDCs. There are various reasons for this, including the fact that the waiver would likely apply only to a share of the fleet registered in LDCs. In addition, such a waiver would contravene the IMO NMFT clause. Similarly, the enabling clause is unlikely to help implement exemptions for LDCs. If vessels pay the GHG price directly into a fund periodically, Article I GATT is unlikely to apply.  

In this context, equity-related considerations might be better addressed via the use of revenues from the GHG pricing mechanism, as argued in IMO negotiations by various countries—including Belize and many Pacific SIDS—and the World Bank. As mentioned above, these revenues can be substantial and could provide a new and additional source of climate finance. In a recent World Bank report, my co-authors and I have proposed a framework to use revenues to enable an effective and equitable transition.  

Alan Sykes on Making WTO Appeals Discretionary

Stanford law prof Alan Sykes has posted a new paper entitled "The Utility of Appellate Review at the WTO and Its Optimal Structure." One part of the paper that I found particularly interesting was his call for any appeals body to have discretion over what appeals to hear:

C. Establishing a Discretionary (Certiorari-Like) Process

... But what about a possible alternative involving discretionary review – an appellate entity with the power to accept or reject appeals based on their perceived merit and importance? Such a system would mirror the jurisdiction of the U.S. Supreme Court, which selectively accepts appeals viewed as especially important through writs of certiorari, often reflecting significant legal conflicts in the decisions of lower courts. The great virtue of such a system, as I will elaborate below, is that it allows appellate review to be deferred until such time as issues are fully “ripe” for resolution, both jurisprudentially and politically.

...

2. The Advantages of Discretionary Appellate Jurisdiction

In contrast to the guaranteed right of immediate appeal in the existing DSU, the analysis to this point hints at some clear advantages to a system of discretionary appeals. First, it reduces the case load of the appellate tribunal. The tribunal can devote more attention to the hard cases, and ideally may allow it to hear cases en banc (like the U.S. Supreme Court), bringing to bear the broader perspective of a more diverse group of judges and the benefits of a larger deliberative body.

Second, the appellate tribunal can wait until issues have been confronted by multiple panels. Again with reference to the U.S. Supreme Court, certiorari is often denied until several lower courts have addressed a systemically important issue. And if all of them resolve the issue the same way, appellate review may ultimately prove unnecessary. Where conflicting decisions arise, the divergent opinions of adjudicators in the lower tribunals often aid in assessing the strength of the parties’ positions as a legal matter. Removing the time pressure from the appellate process and waiting for a broader collection of “inputs” from lower tribunals can reduce the risk of an appellate decision that clashes with the best interpretation of the treaty framework.

Finally, and perhaps most importantly, a discretionary system of appeals allows the Appellate Body time to gauge carefully the reaction of the membership to dispute panel decisions on difficult issues. Decisions are discussed at DSB meetings, and those debates can highlight issues that are politically destabilizing and may require artful compromise. Likewise, the flow of information from national capitals can sensitize the appellate tribunal to conflicts that may elicit systemically dangerous responses. Hard issues can be left unsettled until opinions among members coalesce on the best approach.

In short, a discretionary system of appeals enables the appellate tribunal to become more aware of, and sensitive to, institutional political pressures. Its interpretations of treaty text can then become savvier politically, and it can better avoid decisions that key players cannot accept, subject to the constraint that decisions must still be textually plausible. Where the membership seems hopelessly at odds, the appellate tribunal can simply duck the issues and allow disparate panel decisions to stand.

This approach, of course, is much like the approach to hard constitutional and statutory issues in the U.S. Supreme Court. The relevant legal texts are often hopelessly vague or incomplete (“due process,” “unreasonable search and seizure,” “equal protection,” “restraint of trade”…). Like the Appellate Body, the Supreme Court often pretends that correct answers can be found in text or prior precedent, but in reality, decisions take account of evolving politics on challenging issues.

Of course, under a discretionary appellate process, some challenging issues will remain unsettled for a long time. Panel decisions that are not accepted for appeal will become binding between the disputants, but the same issue may be relitigated several times before different panels with different results. This prospect is a virtue and not a vice in my view, however, as it enables disputants who believe deeply in their positions additional chances to vindicate them and makes it possible for rulings with systemically problematic political consequences to be revisited.

As far as I know, the issue of discretion on whether to hear appeals has not come up in the WTO dispute settlement reform discussions, but maybe it should, and maybe this call by Alan Sykes will get the negotiators thinking about it.

I was thinking along these lines myself when I suggested a while back that the Appellate Body does not have to weigh in on every appealed issue, but rather can just uphold panel rulings without further comment. Arguably, they could do that under existing rules, but if people aren't convinced the rules allow for that, the DSU could be amended to make it clear that the Appellate Body has this power.

In addition, and I'll have more to say about this at some point soon, having looked into the archives a bit, it's worth noting that discretion over which appeals would be heard was something that the Uruguay Round DSU negotiators talked about in some detail. Thus, having the current dispute settlement reform discussions delve into this would not be unprecedented.

If the dispute settlement reform negotiators do end up thinking about this, here are a couple issues they might want to consider:

• Who exactly gets to decide if the appeal will be heard? If it's a standing body, maybe it would be a certain number of the judges, just like under U.S. Supreme Court practice where 4 out of 9 is sufficient to grant certiorari. In the WTO context, maybe it could be 3 of the 7 Appellate Body Members or 4 of the 10 MPIA arbitrators. If it's an ad hoc body things are more complicated, but maybe if 3 judges were appointed to hear an appeal, at least 1 of the 3 would have to decide to grant the appeal.
• Is the discretion over the appeal as a whole, with the entire appeal accepted or rejected? Or is it over each individual point of appeal, so the judges can pick and choose which issues to hear? 

Guest Post: Time to regulate cross-border subsidies? Who decides?

This is a guest post from Luca Rubini, Senior Research Fellow in International Law, University of Turin

This brief post has been triggered by yesterday’s delivery of the Opinion of Advocate General Capeta in the Fiberglass case currently pending before the European Court of Justice.

The key legal issue is whether EU anti-subsidy law (and WTO subsidy law) allow countervailing so-called cross-border subsidies, i.e. subsidies granted by Country A to companies in Country B and exporting their goods to Country C.

The EU Commission has started doing so recently, opining that the notion of subsidy in EU CVD laws (and WTO subsidy laws) can be construed to include such support measures. The Commission’s practice has been provisionally green-lighted by the General Court of the European Union last year (Case T-480/20). The issue is currently subject to dispute settlement in the WTO (DS 616) where Indonesia challenged the EU.

Last summer I wrote a little think piece where I was criticising the decision of the General Court.

The possibility to countervail cross-border subsidies is not a new claim. Seasoned US lawyers will remember that similar attempts had been made in the early 1980s and rejected by the US DOC (see the very illuminating account of Gary Horlick on this). Anecdotal evidence suggests that the issue was not touched in the Uruguay Round negotiations that led to the current WTO subsidy disciplines (I recently interviewed a key negotiator who, to my query about whether they did consider to include cross-border subsidies in the notion of subsidy, replied: "Who would have even thought about suggesting this? They would have said: this guy is crazy! Subsidizing a company out of your jurisdiction!").

I guess the main key take-away from trade law is that subsidy laws are one of the most hard-fought and controversial areas. And right at the centre of all these controversies you have always had the definition of subsidy. The definition that we have in the WTO ASCM is the best that could have been agreed. It has its nuances and deficiencies. Perhaps a better definition could be negotiated and drafted today, in the light of the current challenges including cross-border subsidies. But it is difficult to reconstruct WTO rules now – and the Appellate Body knows it!

It would be very dangerous for any WTO Member to start reconstructing this definition in fundamentally new ways, certainly not contemplated by the negotiators and drafters of the rules – and, most importantly, not necessarily agreed by other WTO Members.

This would be an additional example of dangerous unilateralism which cannot bade well in the current difficult times. Hypothetically, it would be even better to introduce a completely new tool to tackle cross-border subsidies (see the EU Foreign Subsidy Regulation) rather than pretending that the subsidy rules that have been agreed multilaterally can be unilaterally re-interpreted and stretched to include completely new forms of subsidies. Such reinterpretation would be a much worse injury to the rule of law than introducing normative innovations.

The Opinion of the Advocate General is an interesting read (see in particular paras. 27-64). Let’s immediately go to the two key questions.

First, does EU law cover cross-border subsidies? The answer, according to the AG, is yes!

"46.  In conclusion, the Basic Anti-Subsidy Regulation is not limited to subsidies granted by the government of a State in its ‘own’ territory.”

Secondly, and perhaps more importantly, is this interpretation in line with WTO law?

"62.   Consequently, I consider that Article 1.1(1)(a) of the SCM Agreement does not preclude treating as a subsidy an arrangement whereby one WTO member provides a financial contribution to a product manufactured in the territory of another WTO member."

Yes, you read well. The WTO subsidy definition does NOT preclude to include cross-border subsidies.

It is now likely that the ECJ will follow the AG and firmly conclude that the EU Commission can countervail cross-border subsidies. My understanding is that a similar approach has been aired across the Atlantic in the US (but, perhaps crucially, by suggesting a change in the law rather than reconstructing it judicially).

Realistically, the ball is in the WTO dispute settlement court now. Once again, lacking the Members’ initiative, the pressure is put on a WTO Panel. My guess on what the Panel will decide? Considering the debacle of the Appellate Body, it may do well in adopting a self-restrained approach – which may incidentally pave the way to a clash between ECJ and WTO dispute settlement (and it is clearly the former that has to give way to the latter).

My two cents (but usually I get it wrong!)

U.S. Subsidies vs. Chinese Subsidies / Good vs. Bad Subsidies

At a press briefing with Ambassador Tai on the announcement that the Biden administration would maintain the Section 301 tariffs on China and add tariffs on a few additional products, a reporter asked the following question regarding U.S. vs. Chinese electric vehicle subsidies:

The U.S. government is subsidizing the EV industry here pretty significantly.  How is that different from what the Chinese government is doing?

Tai responded as follows:

So, if you take a look at the Chinese economic model, what you see is a system of state support that is built to dominate and take over entire industries.  So, this is what I mean by, “We’ve seen the pattern” — steel and aluminum, solar panels, batteries, EVs. 

Today, China’s production capacity in steel is 55 percent of world capacity — that’s in one country; aluminum, 60 percent; EVs, 60 percent; solar, 80 percent; and in certain critical minerals, 85 to 95 percent.  So, those are subsidies with an aim to cornering the world market and achieving dominance and creating dependency. 

The types of support that we’re talking about here are defensive in nature.  They’re about creating the space to compete, the space to thrive, the space to survive the kind of onslaught that we are seeing across the board.

A lot of people have been trying to figure out how to distinguish good subsidies from bad subsidies. Is there something in what Tai said that can help provide the basis for doing so? Here are a couple possibilities:

• Maybe it's the intent that matters: Subsidies are a problem if they are "built to dominate and take over entire industries" or if they are provided "with an aim to cornering the world market and achieving dominance and creating dependency."
• Maybe it's the level of global capacity that is the problem (e.g. "steel is 55 percent of world capacity"): If it gets too high, that's a concern. On this point, here's something Jennifer Hillman and Chad Bown wrote a while back:

Because prohibited subsidies have both a clearer and faster remedy than merely actionable subsidies, expanding their list could add teeth to the ASCM. Currently, ASCM Article 3 limits prohibited subsidies to export subsidies or subsidies contingent on the use of domestic products over imports. If certain subsidies that are considered more trade distortive, such as those leading to substantial global overcapacity, could be defined and added to Article 3, it would strengthen the ASCM.

•  Maybe if they are "defensive in nature" it's OK, i.e., if they are provided in response to other governments' subsidies.

Could these factors be used in evaluating whether subsidies are good or bad? I've also heard people suggest that the real problem is "pervasive" subsides, so maybe the amount of subsidies should be considered as well. It would be interesting to see WTO Members discuss all this. 

At the same time, I would note that the existing SCM Agreement "adverse effects" rules -- along with the narrower prohibited subsidies rules of course -- might also be a good way to distinguish between good and bad subsidies. These rules provide a remedy based on the effect that subsidies have on foreign competitors. In a sense, the rules give Members a way to play defense when they are harmed by another government's subsidies.

If the goal here is to discipline subsidies, maybe it's worth using the WTO's adverse effects rules more. We can then see if there are still gaps that need to be filled in with additional rules.

It may take a while, but ultimately I think we will need to settle on some international rules in this area that everyone can live with. Otherwise the negative economic impact of unilateral trade retaliation and subsidy races is going to keep growing and become a serious problem.

Director of WTO Rules Division: Apply Soon!

If you are a trade remedies expert contemplating a career change, here's a great opportunity:

The Secretariat of the WTO is seeking to fill a position of Director to lead the Rules Division.

The Rules Division is responsible for all aspects of WTO work, legal and technical, in the areas of trade remedies (anti-dumping, countervailing, and safeguard measures), subsidies, trade-related investment measures (TRIMS), civil aircraft, and state trading enterprises. The Rules Division facilitates and actively assists in negotiations, consultations, and monitoring of implementation; provides legal and other assistance to dispute settlement panels considering disputes; and provides implementation assistance, counselling, and expert advice to WTO Members, in each of these areas.

General Functions

1.         Leading the Division, including defining the strategic objectives of the Division; setting the objectives for, and motivating, a multidisciplinary staff of lawyers, economists, trade specialists, and support staff to meet those objectives; providing guidance to staff in carrying out their duties; reviewing the output and performance of the staff; supporting the development of the knowledge and skills of staff; and ensuring appropriate staff training plans are in place and executed;

...

More at the link. But the deadline is May 12, so hurry if you are interested!

Anticipating the First USMCA RRM Panel Determination: A Look Back

By now, many will have heard that the first panel constituted under the USMCA Rapid Response Labor Mechanism (RRM) recently completed its determination as to whether there has been a Denial of Rights at the San Martín mine.  At the moment, we are waiting for the public release of the panel’s findings. 

While we wait, this post, and at least one more that will follow it, will flag some of the issues that arose in the proceedings – both on procedure and substance for those who may not be readily tracking. 

As readers will recall, this panel was constituted last August, following an exchange of views between the United States and Mexico concerning the situation at the mine in Zacatecas.  The United States, having received a petition in May 2023 from the AFL-CIO, the United Steelworkers, and a Mexican union, concluded that there was a Denial of Rights occurring at the mine.  Mexico, however, found no Denial of Rights because, in Mexico’s view: (1) there had been a denial but it occurred prior to entry into force of the USMCA so fell outside the scope of the RRM; (2) the denial alleged by the United States did not implicate legislation that complies with Annex 23-A of the USMCA as required by the Agreement; and, (3) the San Martín mine is not a “Covered Facility” within the meaning of the Agreement.

 Procedural Innovations and Difficulties

Let’s start with a procedural overview.  In my last post about the panel, I noted some challenging questions that arose – given that this was the first RRM panel – in the early application of the rules and in the panel’s “confirmation” of the petition before it.  After that early stage, throughout last fall, the two governments filed their written submissions.

The panel also sought “written briefs” from “non-governmental entities that are directly involved in the dispute”.  It further stated that the Secretariat could “inform selected entities” of this opportunity.  It appears the panel accepted submissions from the company owning the mine (IMMSA), from the AFL-CIO, from Los Mineros (the aggrieved union), and from the other group of workers in question (Los Trabajadores Coaligados).  These participants line up unsurprisingly where you would expect them: IMMSA and Los Trabajadores Coaligados supporting Mexico’s position and Los Mineros and AFL-CIO supporting the United States.

(I will note here that one of the biggest challenges in following the panel’s work is the lack of available documents.  For what is likely a variety of reasons, the documents on the Secretariat website for this dispute are limited to a partial selection.  I will be sure to correct this post if we learn more.) 

We know that the panel carried out a “verification” involving the mine in late February 2024 and held a hearing, which was livestreamed, in Mexico City on February 28-29. 

Some details about the verification came out during the hearing. The panel chair stated at the hearing that the parties had told the panel that the panel had broad investigative powers for purposes of the verification.  The panel members understood that they could “go out and seek information to verify what was really happening”.  Comments at the hearing suggested that the panel at least interviewed several persons associated with the mine.  Perhaps we will learn more about the details of the verification in the panel’s determination.

Throughout these steps, some entities have voiced concerns with the panel process.  For instance, members of the legal team for IMMSA protested that they were not permitted to be present when the panel carried out its verification and spoke with their client. 

Finally, the panel did not quite complete its work within the very short timeline set out in the Agreement (30 days to determination after the verification), but the panel appears to have given the parties its decision within a few weeks of that date – an extension to which the parties likely agreed. 

A Multi-Layered Dispute

There are a few points that are worth making up front for clarity.  This is a state-to-state dispute settlement mechanism concerning activities at a particular worksite.  The panel was not asked to evaluate acts by the Mexican state or whether Mexico breached the Agreement.  Rather, the panel is asked to “determine whether there has been a Denial of Rights”.  A “Denial of Rights” occurs when “workers at a Covered Facility are being denied the right of free association and collective bargaining” under certain Mexican laws (leave which laws to one side for a moment).  This distinction seemed to get a bit muddled in the pleadings.  And this distinction has an impact on the panel’s work, especially with respect to what principles of international law may or may not apply. 

I think the easiest way to understand the U.S. argument is to think of it as having two principal claims: (1) that the mine is operating during a strike and such operation denies workers the right of free association and collective bargaining under Mexican law; and (2) that the mine is bargaining with a different group of workers from the legitimately elected union and that denies workers the right of free association and collective bargaining under Mexican law.

Again, the panel’s task under the Agreement is to determine whether there has been a Denial of Rights at the mine.  It could find that the United States has substantiated the facts of claim 1 and that that is a Denial under the Agreement, and the same of claim 2.  The panel does not need to determine who is responsible for any Denial (unless that’s part of the legal test under Mexican law).  Under the USMCA, either there has been or there has not been a Denial of Rights, or the panel cannot reach that question because of some preliminary flaw in its constitution or jurisdiction. 

According to the short statement offered by Mexico, the panel found that the conduct in question fell outside the scope of its review as prescribed by the USMCA because the conduct must be after the entry into force of the Agreement and the conduct must be subject to the laws that were part of the 2019 Mexican labor reform. 

This brings me to highlight now three issues that led to some interesting arguments at the February hearing.  Of course, these also came up in the submissions, including in the IMMSA submission.  I’ll refer to those on occasion but concentrate on what was said at the hearing. 

The first interesting issue has to do with disaggregating questions of timing from other matters in the U.S. claims.  We do not yet know, from Mexico’s short statement, on precisely what grounds the panel reached its decision.  Was it what the parties characterized as the panel’s jurisdiction ratione temporis?  That is, did the acts by the mine that the United States claimed to constitute a Denial of Rights fall within the jurisdiction of the panel given when they occurred?  Or was it a question of the relevant Mexican law in place in 2023, and if so, how should one characterize such a question as a matter of international law under the Agreement?  These questions in turn can be broken down into various sub-questions, but I’m reluctant to get into all their iterations if we are just days away from seeing how the panel reached its outcome.  Of course, it could have been both reasons or neither.

The second interesting issue that arose was about whether the mine is a Covered Facility under the Agreement.  As a reminder, the Agreement provides that Covered Facility is one that is “in a sector that produces manufactured goods, supplies services, or involves mining, and that (a) produces a good or supplies a service traded between the United States and Mexico or (b) produces a good or supplies a service that competes in the territory of a Party with a good or service of the other Party.”  Much more to be said on this in a future post.  It seems like maybe the panel did not reach the question based on what little we know about the determination, but as Jesse Kreier already flagged on X at the time of the hearing, the discussion among many participants in the submissions and in the hearing went down a road very similar to that traversed in the Guatemala—Labor dispute involving the “manner affecting trade” language in the CAFTA-DR.  

The third interesting composite issue that got considerable attention at the hearing was about (a) the purpose of the panel, (b) the purpose of the RRM (and labor issues in a trade treaty), and (c) the relationship between the USMCA and the Mexican court system.  At times, each of these three points arose separately, and at other moments, they were intertwined.  A future post will elaborate on the importance of these points to the broader literature on trade agreements and labor provisions.

Questions for Future Panels?

Even before we have the panel’s decision in hand, we can anticipate some questions that its decision may precipitate.  For one, given that the panel determined that there has not been a Denial of Rights, there will be no test of the remedies provision of the Agreement. Perhaps the outcome of the second RRM panel's review will create that test.  As the former head of USTR’s Labor Office already noted, that question is a big one given the nature of the facility under review: a call center.  What penalties would the United States seek to impose there and with what implications?  And what remedy if the call center provides its services only outside the United States?  We may not reach answers to those matters there either if the United States and Mexico reach a settlement in that dispute; some media have suggested they might be seeking to settle.  (No information has been made available yet by the Secretariat as to the establishment of a panel [EDIT: In the hours between the completion of this post and its upload to the site, the Secretariat has now posted the names of the members of the panel] or its confirmation of the petition following the United States’ request that one be established on April 16).  Note also that that second situation has a slightly different posture: Mexico found a Denial of Rights had occurred at the Atento Servicios facility in Hidalgo but it also found the facility had taken the necessary measures to remediate the Denial. 

More important is what this San Martín panel’s determination means for RRM-watchers.  Of course, the RRM system is not a precedential system.  Still, many observers – including U.S. trading partners contemplating their own RRMs – are anxious to consider the panel’s reasoning.

USTR's View of Non-Discrimination

Here is an excerpt from the U.S. statement at the last DSB meeting, in which it discusses the EU - Palm Oil panel's interpretation of non-discrimination in TBT Agreement Article 2.1:

- The United States is disappointed that the Panel in this dispute at times simply adopted the Appellate Body’s previous erroneous interpretations of the TBT Agreement.

- For instance, with respect to Article 2.1 of the TBT Agreement, the Panel adopted the interpretation of recent Appellate Body reports that any detrimental impact on like imports will constitute a breach unless the detrimental impact “stems exclusively from a legitimate regulatory distinction.” Under this erroneous approach, any detrimental impact could constitute a breach – not because the impact is related to the foreign origin of the product, but because the measure was not constructed so as to eliminate any detrimental impact not exclusively related to the regulatory distinction.

- This test – first introduced by the Appellate Body and followed here by the Panel – lacks any basis in the text of Article 2.1 of the TBT Agreement. Article 2.1 is directed at controlling origin-based discrimination. Under a proper understanding of the provision based on its text, in its context, the focus is on whether the alleged different and detrimental treatment is explained by factors unrelated to the foreign origin of the product, such as whether the measure and distinction at issue bears a rational relationship to the regulatory purpose invoked.

- Yet, as a consequence of the Appellate Body’s test, adjudicators are placed in the inappropriate position of balancing the detrimental impact of a measure against its contribution to the objective at issue – a responsibility that should rest properly with policymakers.

(footnote omitted)

I am very interested in USTR's position on non-discrimination, and if I understand them correctly, I am sympathetic to their general concerns here. But I have some thoughts and questions.

To me, it's important to do both an impact/effect analysis and an intent/purpose/aims analysis under TBT Agreement Article 2.1, and it sounds like USTR agrees with that. (Let's not get into how GATT non-discrimination obligations and exceptions should divide up these two sets of analysis, because opening that can of worms will distract us.)

The issue then becomes how to carry out each part of the analysis.

On the impact part, what I would look at here is the degree of detrimental impact (or disparate impact or detrimental effect if you prefer) on imported products that results from a measure which contains a regulatory distinction that treats particular categories of similar but not identical products differently (e.g., liquor products with different levels of alcohol content). Let's say you have a measure which has that kind of impact on imported products on a de facto basis (i.e., worse treatment of imported products is not set out explicitly in the law, but the result of the measure when it is applied is that more imported products than domestic products fall into the product category that gets the worse regulatory treatment). In these circumstances, I would calculate the percentage of imported products that receive the worse treatment and the percentage of like domestic products that receive the worse treatment. If the difference in impact on imported and domestic products is appreciable (say, 60% of imported products get the worse treatment but only 40% of domestic products do), that would lend support to a finding of discrimination. And the higher the percentage difference, the more weight you would give to this factor in the overall consideration.

Turning to intent, in my view, the best way to characterize the intent inquiry is something along the lines of a design, structure etc. analysis, where you dig in to the wording of the measure and assess how it will operate in the market in question based on that wording (the GATT Article III:2, second sentence "so as to afford protection" analysis in the Appellate Body report in Chile - Alcohol offers a good example of what I have in mind). In my view, this gives the adjudicator the right amount of flexibility in examining, on an objective basis, the intent of the measure. (I don't mind looking at subjective intent too, but again, that's a big can of worms, so forget I said that!)

After determining the degree of detrimental impact and the degree of discriminatory intent, you would then come to an overall conclusion on whether the measure violates Article 2.1.

Now let's get to the U.S. statement. At the DSB meeting, USTR was critical of the "stems exclusively from a legitimate regulatory distinction" language used by the Appellate Body in formulating the TBT Agreement Article 2.1 standard. I was always a bit uncomfortable with that language myself, partly because it seemed like it might be applied very strictly and therefore it would be hard for Members to defend their non-protectionist measures. I feel like USTR was expressing similar concerns.

As an alternative, USTR puts forward a "rational relationship to the regulatory purpose invoked" standard. Here I would want to know more about how such a standard would be applied. How would the existence of a "rational relationship" be assessed? Is it the design, structure etc. test? Something else? USTR says "the focus [should be] on whether the alleged different and detrimental treatment is explained by factors unrelated to the foreign origin of the product." This is similar to the language used by the Appellate Body in Dominican Republic - Cigarettes when interpreting GATT Article III:4 ("the existence of a detrimental effect on a given imported product resulting from a measure does not necessarily imply that this measure accords less favourable treatment to imports if the detrimental effect is explained by factors or circumstances unrelated to the foreign origin of the product"), but people found that standard confusing and the Appellate Body abandoned it. I'd like to hear more about what USTR thinks it means and how it would be applied. USTR suggests that the "rational relationship" standard is an example ("such as") of the "unrelated to the foreign origin of the product" approach. Are there other examples?

Also, USTR says the following about the "stems exclusively" standard: "adjudicators are placed in the inappropriate position of balancing the detrimental impact of a measure against its contribution to the objective at issue – a responsibility that should rest properly with policymakers." I'm not sure what they mean here. In particular, I'm not sure how you can evaluate the impact and intent factors without the measure's contribution to its stated objective coming up at some point. 

Overall, I think these are good discussions to have because this is a core principle of the trading system and we really need to get it right. But I'd want to hear more about how exactly USTR would like to see its standard applied (maybe by having them take some real world examples and walking everyone through it). And I'm curious to see how other governments would react to the U.S. view.

The U.S. Non-Violation Nullification or Impairment Argument in the USMCA GE Corn Dispute

In its rebuttal submission in the Mexico - GE Corn USMCA dispute, the U.S. has set out a non-violation nullification or impairment claim as an alternative to its violation claims (the U.S. did not elucidate the non-violation claim in its initial written submission, but it did indicate in fn. 136 of that submission that the claim was within the terms of reference, and it mentioned the claim in footnotes in the panel request).

As background, the USMCA provides for the following types of claims:

Article 31.2: Scope

Unless otherwise provided for in this Agreement, the dispute settlement provisions of this Chapter apply:

(a) with respect to the avoidance or settlement of disputes between the Parties regarding the interpretation or application of this Agreement;

(b) when a Party considers that an actual or proposed measure of another Party is or would be inconsistent with an obligation of this Agreement or that another Party has otherwise failed to carry out an obligation of this Agreement; or

(c) when a Party considers that a benefit it could reasonably have expected to accrue to it under Chapter 2 (National Treatment and Market Access for Goods), Chapter 3 (Agriculture), Chapter 4 (Rules of Origin), Chapter 5 (Origin Procedures), Chapter 6 (Textile and Apparel Goods), Chapter 7 (Customs Administration and Trade Facilitation), Chapter 9 (Sanitary and Phytosanitary Measures), Chapter 11 (Technical Barriers to Trade), Chapter 13 (Government Procurement), Chapter 15 (Cross-Border Trade in Services), or Chapter 20 (Intellectual Property Rights), is being nullified or impaired as a result of the application of a measure of another Party that is not inconsistent with this Agreement.

Here we are talking about a claim under Article 31.2(c) that "a benefit [the U.S] could reasonably have expected to accrue to it under Chapter 2 (National Treatment and Market Access for Goods), ... [or] Chapter 9 (Sanitary and Phytosanitary Measures) ... is being nullified or impaired as a result of the application of a measure of another Party that is not inconsistent with this Agreement."

Here's an excerpt from the U.S. submission with the "reasonably could have expected" element of the claim, to give people a flavor of the overall argument:

A. The United States Reasonably Could Have Expected to Accrue Benefits Under Chapter 2 and Chapter 9.

254. The ordinary meaning of the terms in the phrase “a benefit that [a Party] could reasonably have expected to accrue” is broad. Dictionary definitions of “benefit” include “[a]dvantage, profit, good”, as well as “[a] natural advantage or ‘gift,’” while “reasonably” may be defined as “[a]ccording to reason; with good reason, legitimately; justly, properly, fairly.” Thus, a “benefit” the United States “could reasonably have expected to accrue” could be understood as an advantage (including a natural advantage) or good that the United States legitimately or fairly could have expected.

255. U.S. exports of corn have long moved freely to Mexico, and—until recently—Mexico facilitated this access, including for GE corn. While the North American Free Trade Agreement (“NAFTA”) permitted Mexico to regulate U.S. access to its corn market via a tariff-rate quota until 2007, Mexico opted to open its markets to U.S. corn more than the NAFTA required, particularly during the latter years of this transition. Beginning in 2008, the NAFTA lifted all formal restrictions, allowing U.S. corn to enter Mexico free of all tariffs and quotas. Thereafter, U.S. exports to Mexico of corn and corn-based products blossomed, and according to USDA, U.S. exports of white corn and yellow corn rose from less than 190 million dollars in 1994 to more than $3 billion dollars in 2018. The USMCA continued this tariff-free and quota-free trade in corn, and in 2022—when GE products accounted for over 93 percent of corn planted in the United States—the United States exported more than $4 billion in corn to Mexico.

256. As this prior trade suggests, Mexico permitted the importation and sale of GE corn for decades, and has been one of the countries with the most authorizations for importing and selling GE crops for use in human food and animal feed. As discussed in Section II.D of the U.S. Initial Submission, Mexico’s principal legal instruments governing the importation and sale of agricultural biotechnology products in Mexico date from 2005, and in the years following the promulgation of these instruments Mexico regularly reviewed and approved authorization applications for GE events for food and feed use in Mexico. During this time Mexico issued over 200 event authorizations across 11 different GE crops—and the number of authorizations for corn alone nearly equals the number of authorizations for the other ten GE crops combined. In fact, prior to its sudden, recent reversal of policy, Mexico repeatedly touted the benefits of biotechnology, for example, noting that biotechnology can solve agricultural problems quickly and with minimal risk, meet critical food and healthcare needs, and contribute to sustainable development.

257. In negotiating what became the USMCA, the United States, Mexico, and Canada sought an ambitious outcome that would update the NAFTA and generate important economic opportunities for all three countries. After the USMCA entered into force, Mexico called the USMCA one of the most ambitious instruments it had negotiated on SPS matters, and noted that the USMCA SPS provisions improved disciplines provided for in the WTO SPS Agreement. Notably, Mexico also described the USMCA SPS Chapter as ensuring the protection of life and health and with the aim to advance decision-making with a scientific basis and to avoid unnecessary obstacles to trade. With respect to the National Treatment and Market Access Chapter, Mexico noted that its purpose was to provide greater certainty and transparency to the commercial exchange of goods between the USMCA parties, in order to facilitate and promote trade in North America.

258. The history of Mexico’s actions in the period leading to the successful renegotiation of the NAFTA supports the reasonableness of the U.S. expectation. Mexico permitted the importation and sale of corn, including GE corn, for decades before the USMCA entered into force. Mexico itself touted the benefits of GE corn as well as its safety, authorizing numerous GE events. Mexico also encouraged the tariff-free and quota-free entry of that GE corn. Mexico further lauded the ambition of the USMCA outcomes on SPS and market access. Accordingly, it was reasonable for the United States to expect that Mexico would not completely reverse itself by banning the use of GE corn in dough and tortillas and instructing Mexican government agencies to gradually substitute the use of GE corn in all products for human consumption and for animal feed. Instead, the United States could—and did—reasonably expect that the volume and value of U.S. exports to Mexico of corn, including GE corn, would continue under Chapter 2 and Chapter 9 after USMCA entered into force.

(footnotes omitted)

Let me start with a few thoughts on non-violation claims in general, and then I'll get to the claim in this case.

Back when the GATT -- with its broad principles but gaps when it came to specific subjects (such as subsidies) -- was the main source of international trade governance, the non-violation nullification or impairment remedy had a clearer role. If a particular subject wasn't covered by the rules, and thus there was no clear way to address the trade impact of a measure in that subject area which was unexpectedly put in place, a non-violation complaint could provide a remedy.

But as the GATT (and then the WTO) expanded in scope, there were fewer things not covered, and therefore fewer measures that could be characterized as unexpected. If a particular subject, such as SPS measures, had a set of detailed rules, then it could be argued that any measure that is consistent with the rules should have been expected. The logic here is that if there is a governing regime in place, and if a government was complying with that regime when it adopted the measure, shouldn't the measure be expected? For this reason, I've been skeptical of the continuing utility of the non-violation remedy in the GATT/WTO after the establishment of the WTO and its expansive set of rules.

On the other hand, it could be argued that the non-violation remedy was preserved in the text, so it must have some purpose, and the expanded substantive coverage should not totally eliminate the possibility of using it. Nevertheless, I still have questions about the scope of the non-violation remedy in the WTO context. And the limited number of claims over the years, and the absence of any recent successful claims, suggests others have these questions too.

That brings us to the USMCA. In this context, as set out above, the non-violation nullification or impairment provision explicitly applies to the subject matter chapters, including the SPS chapter at issue here, so the things I said about the GATT non-violation remedy and whether specific subjects are covered don't seem to apply. But that leaves me with some questions: What did the USMCA drafters -- borrowing a bit from the CUSFTA and NAFTA drafters -- have in mind when they explicitly created a non-violation remedy that applies to the SPS chapter (among other chapters)? In what circumstances did they think the non-violation remedy should be used, either in SPS cases or in general? I had a whole theory about how this remedy emerged during the League of Nations talks, but that doesn't apply very well here because of the broad coverage of the USMCA. 

With all that in mind, how should the non-violation remedy be applied in a situation like the one here, in which a domestic regulation is the measure that is affecting trade flows? Should the U.S. have anticipated that Mexico might change its mind at some point on approving genetically engineered food products and adopt measures -- which are USMCA-compliant or else we wouldn't even get to the non-violation claim -- such as the ones at issue here? Does the well-known and long-standing controversy over these products affect the analysis of what is reasonable to expect?

I wish the GE Corn panel luck in sorting this all out if they decide there is no violation and thus have to deal with the non-violation claim.