Up to the early 1900s, the U.S. did not have an income tax, and tariffs were one of the primary means of raising money for the U.S. Treasury. And because this is not something that was true in the lifetimes of most Americans, we have gotten used to not thinking about it, or thinking about it as an illegitimate use for tariffs. And that the legitimate uses for tariffs are ... related to national security or to protection of specific industries and jobs. But in fact, our predecessors very much saw tariffs as a perfectly legitimate revenue-raising strategy. ...

So my contention is that in the current moment, because of the changes to our tax structure and the skyrocketing levels of our deficit, it's going to be very hard for any future president from any political party to forego revenue. So we shouldn't expect a complete and instant rollback because that's a political leader giving up revenue, which they're just not going to do. So that's point 1.

Point 2 is ... that we've seen from the Trump tariffs that there are some industries and sectors that perceive that they have benefited, and it's very hard, and again politically challenging, to pull that back.

I see her points here and there is some logic to them. However, on the issue of giving up revenue in the midst of high budget deficits, I think it's worth noting that the Trump tariffs have coincided with a period of well above average budget deficits. As the first Section 232/301 tariffs were being imposed under Trump, the budget deficit started increasing, and the trend of both higher tariffs and higher budget deficits has continued since then.

In addition, Heather takes us back to the early 1900s and before, when tariff revenue played a bigger role. But given the size of the federal government today, there's no level at which tariffs can be set that will make more than a modest dent in the deficit/debt.

All of this suggests that high tariffs are not an effective way of addressing budget deficits or the debt, and what we need with U.S. fiscal policy at the moment is a broader rethink of tax revenues and spending. Tariff revenues may not be totally irrelevant here, but they are a somewhat marginal component of the discussion.

Even if I'm right about all this in an objective sense, though, how will things play out in political debates? What will future legislators and executive branch officials think? And what will the voters want here (to the extent we can figure that out)?

My points above should be fairly easy to communicate with simple charts showing the tariff increases along with the rising budget deficit and debt, so I'm hopeful this can work politically for a future president who may be interested in pursuing the removal of these tariffs. As noted, tariff revenue is simply not the main factor in dealing with budget deficits/debt, and the focus of the tariff debate should be on the broader impact of the tariffs on the economy and on relations with other countries (which, in my view, hasn't been great so far).

With regard to Heather's point about whether any changes can be "instant," it is true that tariff removal on the proverbial "day 1" of the presidency might be difficult. There are processes that need to be followed under tariff provisions such as Section 301, and if a future administration were making an effort to follow the rules when removing tariffs, it might want to be sure everything was done properly. Nevertheless, if the administration were inclined to go in that direction, it could probably do things pretty quickly.

As to "complete" removal, this could be tough because of the China element. Politically speaking, removal of China-specific tariffs will be a challenge. Someone will have to come up with a coherent alternative policy in relation to China in order to get those tariffs removed (this can be done, but it won't be easy). However, I would think most of the non-China Trump tariffs could be rolled back over time. Yes, as Heather notes, there are special interests who have benefitted and will want to keep them in place. But if you'll permit me a bit of wishful thinking, we may be at a moment where people are fed up with the government catering to special interests.

Of course, subsidies aren't the only factor suppressing Chinese consumption, according to this view. Michael Pettis, with whom I sometimes spar over these issues, recently listed the following domestic policy areas in which China could make policy changes in order to achieve greater domestic consumption (subsidies are at the end of the list):

· Faster wage growth transfers income from employers to workers.

· Larger social transfers shift income from the state to retirees and the poor.

· Lower income taxes or consumption taxes increase the GDP share of household disposable income.

· Pro-union legislation, or the elimination of residency restrictions, transfers income from employers to workers.

· Stricter environmental protection transfers income from industrial polluters to local households.

· Reducing subsidies to manufacturing, infrastructure, or favored borrowers shifts resources away from producers and toward households.

After presenting this list, he stated: "Although these policies appear very different, they all accomplish a similar macroeconomic objective: raising the household share of GDP while reducing the producer share, thereby reducing the excess saving that ultimately manifests itself as current account surpluses."

Just to state clearly what I think will be obvious to trade policy experts, there is an important difference between industrial subsidies and the other items on the list: For subsidies, when China joined the WTO it agreed to abide by the obligations of the SCM Agreement and its Accession Protocol, whereas China did not sign on to an international agreement for the other items. As a result, there isn't much scope for governments to press China on the other issues. Obviously, any government can raise whatever issue it wants with any other government. But I'm not sure the Chinese government is going to pay much attention to what foreign governments have to say on these other issues. (I doubt the U.S. government would take action if it were pressed by other governments on all this). I suppose the idea is to impose some sort of sanction in order to pressure China, but I'm skeptical that China would respond positively to that.

By contrast, because China did sign on to an international agreement on subsidies, there is a clear avenue for others to exert pressure in a constructive way. China's record on compliance with adverse WTO dispute settlement decisions is decent, so why not try to enforce the rules on subsidies? The arguments I've heard against bringing WTO complaints here are not very persuasive. Yes, these complaints can take a while, but (1) if the complaints had been brought back in 2018 when the conversation around these issues picked up, we could have made a great deal of progress already and (2) if we don't start the complaints now, we could still be talking about all the same things in 2034. And the alternative approaches that have been used in the meantime haven't gotten us anywhere.

Turning to "the need to strengthen household consumption" (as Pettis puts it), while I think that Chinese government subsidies and other non-market practices are a real problem, the broader logic tying producer subsidies to household consumption is not as convincing. It's certainly possible that consumption would rise if Chinese government subsidies to producers were reined in, but the actual consumption outcomes depend on what exactly is done with the money previously given to producers. Would consumption go up a bit if the Chinese government transferred more money to retirees and the poor? Probably. But it's worth emphasizing that the debate over transfers to retirees and the poor happens everywhere, and is a very sensitive one in domestic politics. It's difficult to influence how other governments approach this, and my sense is that the views of the Chinese government about these policies are firm and not easily changed.

But that doesn't mean people should ignore the subsidies. While reducing subsidies to industry may or may not translate into much greater consumer spending, these subsidies do have a real impact on trade flows. As part of the effort to create a level playing field, it makes sense to enforce the rules on subsidies that every government agreed to. Thus, other governments should press China on complying with WTO rules on subsidies (and the behavior of state-owned enterprises as well). This shouldn't be particularly controversial, as it just means enforcing the international rules in this area. While there's no guarantee of a big payoff in terms of increased Chinese consumption as a result, it would help make everyone feel like the trading system is operating under terms that are fair.

The latest idea in this space is @BernieSanders' proposal to deposit 50% of the shares of leading AI companies into a new sovereign wealth fund.

While it could mean $$, the bigger reason to consider this is greater democratic control over our era's most disruptive technology.

I wasn't looking to engage on the merits of public equity in general (I'm against it!), but the international aspects of the AI part of his argument intrigued me, so I asked him whether his argument for public equity here applied only to the U.S. government or to other governments as well:

With regard to AI companies in particular, is your argument that the US government should take equity stakes in domestic AI companies, but no other government should do this? Or do you think other governments should also adopt this policy and take equity stakes in domestic AI companies?

His replied as follows:

Any country could certainly demand this of any AI company as a condition for operating domestically.

But a sustainable policy regime internationally (which we do not even attempt to get into) would probably involve some type of exclusivity or "apportionment" where home countries get more of a claim.

I followed up with a question about whether his suggestion might lead to some economic nationalism for people choosing between AI companies:

Do you think such a policy regime is likely to lead to some degree of national boundaries for AI, under which citizens of a country use the services of national AI companies because they (1) share in the profit from these companies & (2) have more trust in the way these companies handle their data?

And he answered:

Possibly. But I think we're already seeing some degree of that desire, even without the policy.

Todd and I disagree on a lot in terms of what economic policy should look like, but sometimes we agree on what things do look like, and here I think he's right about what we are seeing at the moment. There is a growing awareness of the ties between the leading tech companies and the governments of the countries where they are based, including how governments can access consumer data and how they can exert control over who uses the products/services. It seems to me that an absence of trust in these governments is likely to intensify the push for homegrown alternatives. (In the U.S., the main concern is China. In other parts of the world, there is growing concern about the U.S. as well.)

Tech sovereignty vs. concerns with specific governments

This is playing out in the policy debate as "tech sovereignty," with some companies pitching their services as having a national orientation, which they hope will make them more trustworthy for the citizens of the nation(s) where they are based. I've seen this in the social media world with both Canadian and EU companies taking this approach.

But the situation goes beyond pure nationalism. In addition to a call to use domestic providers of services, there can be a broader pitch to consumers that is more directly focused on privacy and other guarantees, as well as an assurance that use won't be cut off by the government where the company is located. Many consumers will prefer to use products/services for which they can trust the government that is setting the rules. Having data and operations hosted in particular countries can therefore be a selling point to consumers around the world.

While I think Todd is right that all of this is happening already, I would say that a shift to public ownership could further stimulate this trend. For example, Europeans who are currently worried about using OpenAI's ChatGPT – e.g., for privacy reasons or a fear of becoming dependent on ChatGPT and the U.S. government then cutting them off – would likely be even more concerned if the U.S. government owned shares in OpenAI.

These issues will have an impact on the international competition among AI companies, a key theme of which today is the U.S.-China competition to "win" the AI race. While it will be interesting to see which country's industry will have the most advanced AI, I think that is only part of the race that's happening here. Currently, U.S. AI companies have the most capable models, but it seems likely that eventually we will get to a point where the basic models can do everything most people want or need, and the advantage held by the leading models will be less noticeable and valuable. Anthropic's Claude may be the best, but for the average person second or third best will be more than enough. At that point, the choice of model for many people is likely to be based on factors such as trust in the company (and cost, of course), with trust linked to the country in which the company is located and the policies/regulations that are in place there. And public ownership of the company would likely play a role in the level of trust as well.

Public ownership vs. regulation of the AI sector

As to Todd's main point, I'm not sure public ownership necessarily means "greater democratic control." Certainly it means strong bureaucratic oversight, but I don't think that's the same thing. A small number of experts working in government who make decisions on complicated technical issues without much transparency is not necessarily the democratic ideal. A key question here is, greater democratic control as compared to what? Would public ownership be better than detailed and comprehensive legislation and regulation of private companies?

To me, what is needed here is a sensible version of the latter. But the companies don't want that, and I'm not sure legislators and regulators are able to provide it at the moment. At some point they will figure it out, but it may take a crisis or two to get us there. In the meantime, there is plenty of room for experimentation among governments around the world, and hopefully some of them can move things forward a bit.

As is probably obvious to this blog's readers from 20 years of me posting, in many areas I'm a skeptic on regulation. But here the potential harms are wide ranging, and we are likely to need something pretty extensive. AI regulation will have to address a much broader set of concerns than, say, automobile regulations do (of course, we now have to deal with AI in automobiles!). From what I can tell, the industry doesn't seem like it has a handle on this, so it's going to be up to legislators and regulators. Perhaps the courts can help too (a suggestion I recently came across for applying strict liability here was interesting), and there is plenty of litigation going on, but it would be nice to have a coherent policy rather than a hodgepodge arising from court rulings.

My own experience with AI is still pretty limited, so what do I know?!

As for me personally, since I last posted about this, I've actually used AI tools for a couple tasks: Trying to figure out some problems with our website code and coming up with a market value for a house. It seemed moderately useful for those tasks, although with the website problems it took a developer looking deep inside the code to get things working again. (I should note, though, that at one point the developer was using AI to help diagnose the problem, so I shouldn't totally dismiss AI in that case).

With regard to law/policy thinking and writing, I've heard people say they use AI for research or brainstorming or drafting. I'm never going to use it for any of those tasks because they are things I enjoy doing the way I do them now. Perhaps AI will someday be able to eat pepperoni pizza for me, but that's an experience I want to keep enjoying and I'm never going to let AI eat my pizza! I feel the same way about my current process for thinking and writing.

But I can see how someone might not always want to do one or more aspects of these tasks, and how outsourcing that to AI might make sense. For that reason, I'm not an AI-denialist. I see the value, although I think it is perhaps overstated. In addition, I suspect that when people have to pay the actual costs of AI tools instead of getting them at the current subsidized price being offered by the AI companies, things might shift a bit. And there are big copyright issues still to be dealt with here, which could change what AI companies can offer. For these and other reasons, I'm happy to stay mostly on the sidelines for now and let the companies and governments work out what AI will turn out to be.

“I had to store my friend’s dead body in the freezer.” A fisher shared this with me and my colleagues at the Office of the U.S. Trade Representative (USTR) through an interpreter. “Every time I opened it to store the catch, I saw his body until we finally reached shore.” His friend had worked extreme hours for months under abusive conditions, without a single port of call. When he fell ill, he did not receive the medical treatment he needed. The vessel never diverted for his care, or to bring his body home. It kept to its schedule. 

What the fisher described was forced labor as defined by the International Labor Organization (ILO) Convention 29 on Forced Labor, 1930: 

“all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily.”

The fisher was part of a delegation of forced labor survivors from distant-water fishing vessels that a labor rights organization accompanied to Washington, DC, in 2025. They were brought to the agency to meet with us, tell their stories, and offer solutions for preventing forced labor. The fishers understood that trade enforcement done right could protect workers like them.  

Trade policy can be a force for good in workers’ lives. I know because I spent 25 years as a senior government official advancing labor rights at the intersection of trade and labor, including combating child labor and forced labor. As deputy assistant U.S. trade representative for labor affairs, I led the global forced labor portfolio before my retirement from federal service.

I carried that portfolio under Ambassador Katherine Tai’s worker-centered trade policy. It was a policy built on years of advocacy by the labor and anti-trafficking community to link trade and labor rights. Under her leadership, we secured the first G7 Trade Ministers’ Statement on Forced Labor and the only U.S. government trade strategy on forced labor. Under Ambassador Jamieson Greer, I helped negotiate forced labor protections into Agreements on Reciprocal Trade, which trading partners signed to secure relief from tariffs imposed under the International Emergency Economic Powers Act. I watched him hold the line on labor rights provisions. I was still at USTR when the pivot to Section 301 of the Trade Act of 1974 took place earlier this year, but left before that work went further. 

Pretext or Principle?

You did not need to be on the inside to know forced labor was not the real problem driving the 301 tariffs, levied on 60 economies, made effective July 24. After the Supreme Court struck down the “Liberation Day” tariffs in February, temporary tariffs under Section 122 served as a stopgap measure lasting 150 days. Those Section 122 duties were set to expire on the same day that Section 301 emerged as the replacement tool for the same set of tariffs. Unfair trade practices related to “failure to impose and effectively enforce” a forced labor import ban (“import ban”) became the new legal justification. 

Worker organizations have long called for Section 301 to address labor rights as an unfair trade practice. Congress made denial of worker rights an actionable unfair trade practice in 1988. The AFL-CIO filed Section 301 petitions against China in 2004 and 2006. USTR declined to accept either petition. It is not surprising, then, that many beyond labor advocates see the use of Section 301 as pretext. They watched the sequence of tariff justifications unfold in real time. 

The USTR fact sheet claims President Trump is “tackling modern-day slavery at its source.” In one sense, that is defensible. USTR is requiring trading partners to block goods made with forced labor. In another sense, it is disingenuous. USTR’s tariffs are designed to incentivize a government to “impose and effectively enforce” a forced labor import ban. But according to its notice of action, USTR carved out a set of exemptions. Among those exemptions, USTR will forgo tariff leverage if the product cannot be “grown or produced in sufficient quantities or at reasonable prices in the United States or obtained from other sources,” regardless of whether tariffing it would pressure a government to act. While not the same, the Section 301 tariff consumptive demand exemption echoes the Section 307 consumptive demand clause, which gutted the Tariff Act of 1930’s forced labor import ban for 85 years until advocates won its repeal in 2015.

Creating a High-Standard Remedy

These tariffs face lawsuits. Whether or not one agrees with these tariffs, the transformation of the forced labor import ban landscape has been swift. Governments are taking action to adopt and enforce forced labor import bans. As of August 5, 2026, 25 governments and the European Union have import bans on the books or on the way, many moving quickly to avoid tariffs. But speed should not supersede a systemic approach to designing these bans. 

Ending forced labor is shared work, and the labor community’s energy and expertise are essential to turn commitments on paper into enforcement in practice. Advocates remember what happens when a well-intentioned trade measure harms the people it was meant to protect. The U.S. Child Labor Deterrence Act of 1992, otherwise known as the “Harkin Bill,” proposed a ban on imports made with child labor. In the fall of 1993, expecting it to pass, the Bangladeshi garment industry dismissed tens of thousands of working children, with documented cases of children ending up in more exploitative conditions. That bill never became law. If these tariffs are to stay in place, they must work for workers. 

If the Section 301 tariffs are to be a credible deterrent, they must incorporate the lessons of the labor community. USTR should insist that governments build those lessons into their own approach. As Martina Vandenberg, President of the Human Trafficking Legal Center, and Ayla Francis Foster, Director for Policy & Government Relations at Humanity United Action, noted in their MS NOW opinion piece, “When forced labor import bans are real, they make it possible to envision a world without any safe harbor for goods tainted with forced labor.” 

USTR should set a high standard for the Section 301 remedy floor with a transparent offramp to zero tariffs and benchmarks that governments can meet. Drawing on U.S. Customs and Border Protection’s Forced Labor Enforcement Operational Guidance for Importers, Laura Murphy’s An International Blueprint for Forced Labor Import Bans, and the ILO Protocol of 2014 to the Forced Labor Convention, 1930, the remedy should, at a minimum, require evidence-based priority targeting of goods, clear and specific evidentiary thresholds, and a public list of implicated entities. It should also fund enforcement and require remediation that corrects labor violations, including back pay, regardless of migration status. Lower rates should go only to economies that recognize U.S. determinations of high-risk goods, act on them, and demonstrate that forced labor goods have been stopped at the border. 

Enforce Existing Trade Authorities

Existing trade authorities to combat forced labor also need consistent and timely enforcement. The Uyghur Forced Labor Prevention Act entity list had remained unchanged since January 2025 until the U.S. Department of Homeland Security added 43 companies on July 31, 2026. The politicization that led to lifting the Section 307 Withhold Release Order blocking Central Romana sugar, as New York Times reporting showed, should never have occurred. Slashed staff and funding for labor rights and anti-trafficking programs that level the playing field for both American workers and those abroad must be restored at the border and at the “source” itself: the recruitment corridors and workplaces where exploitation begins. Reckless disregard for forced labor throughout global supply chains requires consequences at home and abroad. 

This Moment Is for the Labor and Anti-Trafficking Movement

The recent commemoration of World Day Against Trafficking in Persons is a reminder to the labor and anti-trafficking community: claim this moment and make it a win for workers. Many feel sidelined as governments rush out import bans to avoid tariffs, without consulting workers or the advocates who spent years learning what makes them work. That frustration is fair. That said, the cause belongs to the movement that has pushed for decades to protect workers from and through trade. Governments and businesses must be held to account — to protect, to respect, and to remedy — as laid out in the United Nation Guiding Principles on Business and Human Rights. 

For more than a century dating back to the Keating-Owen Child Labor Act of 1916, survivors, workers, organizers, and unions have fought to keep goods made with abusive labor out of American commerce. They marched in the Global March Against Child Labor for the ratification of ILO Convention 182 on the Worst Forms of Child Labor. They filled the streets at the Battle of Seattle in 1999 to demand that trade not ignore working people. They wrote enforceable labor provisions into free trade agreements, including the hard-won Rapid Response Mechanism in the United States-Mexico-Canada trade agreement. They won the passage of the Uyghur Forced Labor Prevention Act. They then took it global, winning their fight to press the European Union to shut its market to forced labor goods. The labor and anti-trafficking movement built the foundation these tariffs rest on. That history is why they should not yield now. 

The stakes are high. Whether on sea or land, workers and children find themselves working involuntarily under coercive conditions. The ILO, Walk Free, and the International Organization for Migration estimate 28 million people are in forced labor on any given day. The worker who died at sea was one of them. The fisher honored his friend’s memory by sharing his testimony with us. We can honor their bravery by demanding these tariffs be used legitimately, and that a remedy be designed for the workers it pledged to protect.

Charita Libao Castro, PhD, MSW, is a former deputy assistant U.S. trade representative for labor affairs, where she served as the global forced labor lead

I had been wondering if this process might provide some new information on the Trump administration's various trade deals, and I've been watching for documents related to the trade policy reviews of governments with whom these deals have been negotiated. Japan's latest review took place at the end of May, and, as I hoped, there were some questions on these issues. Below are the ones I came across that I thought might be of interest to people in relation to the Trump trade deals.

First up were some questions on the details of the U.S.-Japan framework trade deal and its consistency with WTO obligations. On this issue, China asked the following:

During the period under review, Japan reached a trade agreement framework with the United States. When will Japan reduce tariffs for the United States, and will such reductions be implemented on a Most-Favoured-Nation (MFN) basis? When will this framework be notified to the WTO, and will it conflict with Japan's obligations under the WTO?

Japan's reply was:

Under the Agreement between Japan and the United States on July 22, 2025, Japan made no tariff concessions. The Agreement is not legally binding. It is not a regional trade agreement under the WTO Agreement and is not subject to notification.

China later followed up with a question about the nature of the deal as "legally binding" or not:

In its response to the first batch of questions, Japan stated that the framework agreement is not legally binding. Could Japan further clarify the reasons why the agreement is not legally binding? If it carries no legal binding effect, may either party decline to perform its stipulated obligations?

Japan elaborated on the point as follows:

The Agreement between Japan and the United States on July 22, 2025 was not intended to establish legal obligations for either country. Japan intends to steadily implement the Agreement, and will continue to call on the United States to do the same.

Taiwan asked a similar question on the point about consistency with WTO obligations:

Noting the bilateral tariff arrangements between Japan and the United States in 2025, including the application of a baseline 15% reciprocal tariff and preferential treatment under Section 232 measures for certain sectors, could Japan elaborate on how such arrangements are consistent with WTO principles, particularly the most-favoured-nation (MFN) obligation, and how Japan assesses their implications for non-discrimination and predictability in the multilateral trading system?

Japan gave an answer similar to what it said above:

Under the Agreement between Japan and the US on July 22, 2025, Japan made no tariff concessions, and did not commit to implement measures inconsistent with the rules on the market access under the WTO Agreement.

Kazakhstan asked a variation of this question, with some additional aspects thrown in:

Japan's report describes the bilateral tariff agreement concluded with the United States on 22 July 2025, under which a baseline reciprocal tariff rate of 15% applies to Japanese goods. Kazakhstan requests Japan to clarify how this bilateral arrangement is consistent with the Most-Favoured-Nation (MFN) obligation under the WTO Agreement, and what measures Japan is taking to ensure that the agreement does not result in trade diversion to the detriment of other WTO Members, including Central Asian countries exporting comparable goods to the United States market.

Japan's reply focused only on the first part of the question:

Under the Agreement between Japan and the US on July 22, 2025, Japan made no tariff concessions, and did not commit to implement measures inconsistent with the rules on the market access under the WTO Agreement.

Next there were some questions about Japan's market access promises related to automobile regulations and subsidies. For reference here, on the regulations, a White House fact sheet on the U.S.–Japan Strategic Trade and Investment Agreement says: "Longstanding restrictions on U.S. cars and trucks will be lifted, granting U.S. automakers access to the Japanese consumer market; U.S. Automotive standards will be approved in Japan for the first time ever." China asked the following on this issue:

According to the framework trade agreement with the United States, Japan will accept US manufactured passenger vehicles certified under US safety standards without requiring additional domestic testing. Please explain Japan's considerations and grounds for waiving domestic testing requirements on US-manufactured passenger vehicles. Will Japan grant the same exemption treatment to passenger vehicles from other members that meet equivalent safety standards?

Japan's reply was:

The road vehicle safety regulations of Japan apply equally to vehicles imported from all countries without distinction.

The measure has been implemented following an evaluation of the FMVSS and the means by which compliance with such standards is ensured.

And on subsidies, a Joint Statement on the Framework Agreement between the United States and Japan on July 22, 2025, released on September 4, 2025, states that Japan committed to the following: "Providing Clean Energy Vehicle Introduction Promotion Subsidies for American cars." On this issue, China asked:

According to the framework trade agreement with the United States, Japan specifically states that Japan will provide Clean Energy Vehicle Introduction Promotion Subsidies for US manufactured passenger vehicles. Who is eligible for this subsidy, and what are the criteria for receiving it?

Japan replied:

Clean Energy Vehicle Introduction Promotion Subsidies provides financial support to local governments, other organizations (excluding independent administrative agencies), and individuals to cover part of the costs required for introducing clean energy vehicles. As application requirements, the program requires that the vehicle be new and private use, among other conditions.

China followed up by asking:

According to the framework trade agreement with the United States, Japan specifically states that Japan will provide Clean Energy Vehicle Introduction Promotion Subsidies for US manufactured passenger vehicles. Can similar products from other WTO members also receive this subsidy?

Japan then replied:

Only vehicles that are mass-produced and whose specifications or model types have been preapproved by the subsidy implementing organization, based on applications submitted by the manufacturer or equivalent entity, are eligible for Clean Energy Vehicle Introduction Promotion Subsidies Vehicles manufactured in countries other than the United States are also eligible to receive this subsidy.

Next up, there was a historical look at the trade deal signed between Japan and the U.S. during the first Trump administration. Here, China asked:

Could Japan please explain the reasons why the 2020 trade agreement on goods with the United States, together with their related rules of origin, were not notified to the WTO?

Japan replied:

We look forward to notification and discussion, taking into account that the Agreement includes the reference to additional negotiations between the Parties.

Russa also asked about this:

Please clarify whether Japan has already notified the Japan-United States Trade Agreement of December 2019 to the WTO? If this is not the case, please clarify when the missing notification will be provided.

Japan's reply was:

In light of Russia's aggression against Ukraine, Japan does not think it is appropriate to engage with Russia in a business-as-usual manner in the WTO. Therefore, we cannot respond to this question.

And there were questions about the foreign investment commitments Japan recently made to the U.S. in a Memorandum of Understanding. China first asked the following (the question seems to have gotten a little bit garbled in the WTO document, and I'm just quoting it as is):

As the Report by the Secretariat mentioned that "a pledge by Japan to invest up to USD 550 billion into the United States under a dedicated Memorandum of Understanding""The Memorandum is an "administrative understanding" between the two parties and does not create legally binding obligations under domestic or international law. It may be modified or terminated by either party with written notice". Could Japan please explain how Japan consider the implementation of this Memorandum in a new circumstance following the United States Supreme Court ruling in late February 2026?

Japan's reply was:

The Strategic Investment Initiative set forth by the Governments of Japan and the United States contributes to promoting mutual benefits, ensuring economic security, and promoting economic growth between Japan and the United States. Japan intends to steadily implement the Agreement between Japan and the United States, and at the same time, will continue to call on the United States to do the same.

China then followed up with:

As the Report by the Secretariat mentioned that "Investment proposals are selected by the US President (based on recommendations from an Investment Committee chaired by the Secretary of Commerce, with input from a bilateral Consultation Committee.)""Cash flows generated by projects are distributed in two stages. Initially, returns are shared equally, 50% to each party, net of US taxes, until the "Deemed Allocation Amount" is satisfied. Once this threshold is met, distributions shift to a preferential allocation of 90% to the United States and 10% to Japan for all subsequent cash flows.""Japan retains discretion to decline funding, but such decisions trigger consequences: forfeiture of entitlement to distributions under the original allocation formula and application of a revised formula until the United States recovers the shortfall ("Catch-up Amount")". Could Japan please explain how Japan assess the impact of the distribution mechanism on long-term fiscal returns and commercial interests of Japanese investors?

Japan's reply was:

First of all, this description in the Report by the Secretariat is written in the memorandum of understanding (MOU) between the government of Japan and the Government of the United States of America with respect to strategic investments. Regarding funding for each project under the Strategic investment initiative, JBIC (Japan Bank for International Cooperation) will invest and provide loans, and commercial banks will also provide loans guaranteed by NEXI (Nippon Export and Investment Insurance). The MOU also outlines the modality for distributing the available cash flow. Until the aggregate cash flow amounts equal to the principal and the interest funds provided by Japan, 50% to the US and 50% to Japan. This part reflects the fact that loans constitute a significant portion of Japan's funding. For the remaining amount, 90% to the US and 10% to Japan, taking into account various contributions from the US side such as arranging leases for the US federal land, water, and power/energy, as well as facilitating off-take arrangements and expediting regulatory processes. What "long-term fiscal returns" and "commercial interest of Japanese investors" mean seems unclear to us but following the MOU's modality, the principal and the interest funds will be provided appropriately from the available cash flows.

Taiwan also asked about these investment pledges:

With reference to Chapter 5, Section 5.2.2, paragraphs 5.6 and 5.7 of the Japanese report, which state that Japan has committed to promoting USD 550 billion in investment in the United States, with a focus on industries related to economic security, please explain the implications of such outward investment policies for the global investment landscape and the reconfiguration of international supply chains.

Japan replied:

As the paragraph 5.7 of the Japanese report mentions, the purpose of this strategic investment initiative is to build resilient supply chains that benefit both Japan and the US through Japanese investment in the US. Although we are not sure about what 'the reconfiguration of international supply chain' exactly means, we expect these projects under this initiative will strengthen critical supply chains in Japan and the US.

Finally, there were some questions about Japan's promise to buy more U.S. rice. The Joint Statement on the Framework Agreement between the United States and Japan on July 22, 2025 sets out the following commitment: "Expedited implementation of a 75% increase of U.S. rice procurements within the Minimum Access rice scheme." China asked about this as follows:

Whether Japan's unilateral commitment to increase imports of rice from the United States by 75% within the Minimum Access quota violates the GATT principles of Most-Favored-Nation treatment and non-discriminatory treatment for state trading enterprises, thereby reducing export opportunities for rice from other WTO Members to Japan?

Japan replied:

The import of rice under the Minimum Access commitment is conducted through tenders based on the WTO rules, taking into account the needs of domestic consumers, the production and export capacity of exporting countries, and domestic and international supply and demand trends. Import volumes by country are the results of such tenders, and no preferential treatment is given to any specific country.

China also asked:

During the review period, Japan reached a framework trade agreement with the United States, committing to expediting implementation of a 75% increase of US rice procurements within Minimum Access TQs. Could Japan please clarify what is the current quantity of Japan's MA quota allocated to the United States?

Japan replied:

The import of rice under the Minimum Access commitment is conducted through tenders based on the WTO rules, taking into account the needs of domestic consumers, the production and export capacity of exporting countries, and domestic and international supply and demand trends. Import volumes by country are the results of such tenders, and no preferential treatment is given to any specific country.

China also asked:

Japan has committed to increasing rice imports from the United States by 75%. Does this additional volume come from the existing Minimum Access (MA) quota, or has Japan allocated an extra MA quota volume specifically for the United States?

Japan replied:

The import of rice under the Minimum Access commitment is conducted through tenders based on the WTO rules, taking into account the needs of domestic consumers, the production and export capacity of exporting countries, and domestic and international supply and demand trends. Import volumes by country are the results of such tenders; no preferential treatment is given to any specific country, and no extra MA quota volume, has been allocated for the United States.

Another from China:

If Japan's commitment to increase rice imports from the United States by 75% is to be fulfilled within the existing Minimum Access (MA) quota, does that mean the export opportunities for rice from other countries to Japan have been crowded out?

And Japan's reply:

The import of rice under the Minimum Access commitment is conducted through tenders based on the WTO rules, taking into account the needs of domestic consumers, the production and export capacity of exporting countries, and domestic and international supply and demand trends. Import volumes by country are the results of such tenders, and no preferential treatment is given to any specific country. Therefore, export opportunities for rice from any country to Japan have not been crowded out.

One more from China:

The Secretariat report indicates that the import of rice under the Minimum Access commitment is conducted through tenders. Could Japan explain how it ensures that increasing rice imports from the United States by 75% through such procedures will not prejudice the interests of other WTO Members?

Japan's reply:

The import of rice under the Minimum Access commitment is conducted through tenders based on the WTO rules, taking into account the needs of domestic consumers, the production and export capacity of exporting countries, and domestic and international supply and demand trends. Import volumes by country are the results of such tenders, and no preferential treatment is given to any specific country. Therefore, by ensuring export opportunities for any WTO Members, we are not prejudice to their interests.

They start by noting that "the Chinese yuan is no doubt undervalued," but then say "the emphasis on the exchange rate as the lever of adjustment [for imbalances] is misplaced." As set out in a G7 expert report and an IMF policy paper on global imbalances, the yuan-dollar exchange rate is not a culprit but rather an outcome, "an undesirable but predictable consequence of a particular configuration of domestic policies." They explain further that:

A country that suppresses household consumption while simultaneously facing a collapse of property investment will run persistent current-account surpluses and will, other things equal, have a weak currency. A country with insufficient private savings and unsustainably large fiscal deficits will run persistent current-account deficits and will, other things equal, have a strong real exchange rate against other other currencies. The currency is misaligned because the underlying policy mix produces too much or too little saving. The exchange rate is a symptom, not the disease.

Then on the China side, they say:

... [China's] surpluses are real, large and a legitimate concern for the rest of the world, including for a European economy that cannot serve as the absorber of last resort. What China needs to do—and what is in its own long-term interest, given its ageing population and an investment model overly reliant on the tradable sector—is to raise the share of household income in GDP, expand social insurance so that families feel able to spend, and stop financing tradable-sector expansion at the expense of consumption. Do those things, and a real appreciation of the yuan will follow.

And on the U.S. side, they say:

... America will also need to tackle its unsustainable fiscal policy, as well as its persistently low private savings. That, too, cannot be addressed by exchange-rate gimmicks.

They conclude with this:

The G7 and the IMF locate the problem correctly: it lies in the constellation of domestic macroeconomic choices on both sides of the imbalance. A policy package, whereby China pivots to consumption and services-led growth, and America reins in fiscal deficits, is less spectacular than a grand currency bargain. But it has the advantage of being both effective and achievable.

When it comes to getting China to move on issues such as expanded social insurance and higher consumer spending, I have no sense of how to do this. What might the Chinese leadership be willing to do? How can they be nudged in a particular direction? Are Chinese citizens willing to spend more? I'd be interested in what China experts have to say on all this.

On the U.S. side, I feel like I understand the issues better, but it's still not clear how to get anyone to budge. We now have about 5 and a half years of Trump as president, and it's pretty clear that fiscal responsibility is not on his agenda. The question is, what would it take for some future administration to embrace this? It's strange to have to ask this, because from what I can tell fiscal responsibility is somewhat popular. If a presidential candidate were to say "I'm going to cut spending and raise taxes because we need to bring down the budget deficit and reduce the debt," I think that general pronouncement could get decent support. You can't satisfy everyone, of course, and the specific spending and tax adjustments would be a challenge to work out, but I can imagine that at least a small majority would recognize this policy as sensible and support it. Instead, though, we have been getting pro-cyclical, deficit spending stimulus and presidents who are not very popular. There may be a lesson in there for anyone who is willing to learn it.

Getting back to exchange rates, if I understand the authors' main point correctly, they are saying that it doesn't make sense to push for a currency adjustment without also undertaking some changes to domestic policies/consumer behavior, and I think that makes sense. I do wonder, though, what happens if, in practice, you can only get partial changes on the domestic side. I raise this because I think a shift in savings patterns may be hard to achieve, as this could be more cultural than policy-driven. In that case, the exchange rate may need some adjustment, and it would be nice if both countries would be amenable to allowing that to happen.

Today, Ambassador Jamieson Greer is taking final action, at President Trump’s direction, under Section 301 of the Trade Act of 1974 by imposing a 25% tariff on certain goods of Brazil.  This follows a yearlong investigation by USTR that determined that certain Brazilian measures related to digital trade and electronic payment services; unfair, preferential tariffs; anti-corruption interference; intellectual property protection; ethanol market access; and illegal deforestation are unreasonable and burden or restrict the commerce of American farmers, workers, innovators, and exporters. ...

Not surprisingly, Brazil was unhappy about this. One aspect of the Brazilian response was the filing of a WTO consultations request (which also includes a challenge to the Section 301 forced labor tariffs). In its request, Brazil alleges violations of GATT Article I:1 and Article II:1(a) and (b), as well as DSU Article 23.1 and Article 23.2(a).

If Brazil follows through with a panel request and the complaint is litigated, it will be interesting to see what arguments the U.S. presents as a defense in connection with the various substantive determinations USTR made in the two Section 301 investigations. Regardless of the U.S. legal response, though, based on what we know of the facts here, I think the likely outcome is that the panel will find violations of GATT Articles I and II and reject the U.S. defenses (I'm skeptical that Brazil will win the DSU claims though).

But then what happens? With the Appellate Body still not functioning, and the U.S. not a party to the MPIA, the U.S. will be able to appeal the panel report "into the void." This move would prevent Brazil from following the formal process under the DSU to get authorization for retaliation.

However, Brazil has a new law in place that could undermine the effectiveness of such a move. As Geraldo Vidigal and Melina Coelho explained in a piece last year:

During the Appellate Body era, Brazil’s regime for trade retaliation was intrinsically connected to multilateral dispute settlement. As a matter of domestic law, express trade retaliation could only be adopted following an authorisation by the WTO DSB. Once the Appellate Body’s inoperation seemed permanent, Brazil enacted a new retaliation regime — a ‘Revised WTO Retaliation Law’93 — under which domestic authorities could apply economic retaliatory measures even when a final decision had not been issued by the DSB in a dispute involving Brazil.

Under the Revised WTO Retaliation Law, the Brazilian Chamber of Foreign Trade is entitled to decide on the suspension of concessions or other obligations adopted by Brazil. This decision may derive from the authorisation by the WTO DSB, as per the previous law. But, under the Revised Law, the committee may adopt these measures without WTO authorisation, provided that four conditions are fulfilled: (i) a WTO panel report confirms, in whole or in part, the allegations presented by Brazil as the complaining party; (ii) there is an appeal by the WTO member, as the respondent; (iii) the appeal cannot be heard by the Appellate Body or the report cannot be approved by the WTO DSB; and (iv) sixty days elapse after Brazil notifies the respondent WTO member of its intention to suspend concessions or other obligations. The Revised WTO Retaliation law also incorporates the proportionality requirement, stating that “the suspension of concessions or other obligations shall not be greater than the nullification or the harm caused to the country’s trade benefits by the said WTO member.”94

Under Brazil’s Revised WTO Retaliation Law, therefore, the Executive Committee does not need to wait for a final authorisation by the DSB, considering its possible paralysis owing to an appeal ‘into the void’. The Executive Committee is authorised to proportionally retaliate against the other disputing party, including by suspending IPRs. The instrument further provides that, in the event that the Appellate Body reestablishes its activities and modifies the recommendations issued by the Panel (or the DSB revokes its authorisation), the retaliatory measures shall be withdrawn.95

In effect, this means Brazil has explicitly given itself the ability under its domestic law to overcome an "appeal into the void" and move forward with retaliation, even without authorization under the DSU. Of course, governments could always do something like this, but establishing a formal legal mechanism makes this easier to accomplish when they consider that the action is needed.

If it does happen, I can imagine the Trump administration will complain about it, arguing that it shows other governments are not as concerned with international law as they sometimes claim. And I know there is a debate about whether what Brazil would be doing here is permitted under international law. But putting aside the legalities, if Brazil does follow through on this, perhaps it shows that "the void" only exist if you allow it to, and it can be closed (filled in?) with appropriate actions.

What would be the practical impact if Brazil follows this approach? It may just be symbolic, as the litigation will take a while and retaliation might be implemented by Brazil even before a panel ruling. But symbolism can matter, and a ruling that these Section 301 tariffs violate WTO rules could have some relevance for the broader political arguments going on in trade, if not now than perhaps in the future.

The legislation responds to this growing global pattern uncovered by the House Judiciary Committee: foreign governments using coercive state powers to target successful American-owned companies while protecting favored domestic competitors. In Europe, regulators have now fined Google more than $1 billion under the Digital Markets Act (DMA) for how it presents its own services in search results and operates its Play Store. This is part of a regulatory regime backed by fines of up to 10 percent of worldwide annual revenue and up to 20 percent for repeat violations. In South Korea, authorities subjected American-owned Coupang to dozens of investigations, thousands of document demands, and a record-setting fine. In Brazil, officials have advanced DMA-style legislation that would impose special restrictions on overwhelmingly large U.S.-headquartered technology platforms. 

What does Baumgartner have in mind by "economic discrimination"? The legislation defines it as follows:

ECONOMIC DISCRIMINATION.—Any alien who, while serving as a government official of any foreign government, initiates, directs, conducts, engages in, or issues one or more investigations, enforcement actions, licensing determinations, fines, fees, tax assessments, or other legal, regulatory, or administrative burdens against a United States person that are, individually or taken together, more severe, more frequent, or less procedurally favorable than those initiated, directed, conducted, engaged in, or issued against a similarly situated party that is not a United States person, is inadmissible.

That strikes me as a very broad scope. It seems like it is focused on disparate impact, and I'm not sure what room there would be to make an intent-based argument in defense.

How would this standard be applied? A fact sheet elaborates on the real world examples cited in the press release above:

A Growing Global Pattern

● European Union. The European Union’s Digital Markets Act imposes special obligations on designated “gatekeeper” companies, most of which are American. Violations can result in fines of up to 10 percent of worldwide annual revenue, rising to 20 percent for repeat violations, as well as possible structural remedies. House Judiciary Committee oversight has raised concerns that the law’s thresholds and obligations disproportionately burden American technology companies while insulating European competitors.

● South Korea. House Judiciary Committee investigators found that South Korean authorities subjected American-owned Coupang to a broad government campaign involving: more than ten government agencies; dozens of unrelated investigations involving more than 4,000 document requests and at least 652 employee interviews; and a fine exceeding $410 million, the largest imposed on a single company in South Korea. The Committee report concluded that Coupang received disproportionately hostile regulatory treatment not faced by similarly situated Korean competitors.

● Brazil. Brazil has proposed Digital Markets Act-style legislation that would allow regulators to designate large digital companies for special obligations lasting as long as ten years. The proposal could require interoperability, data sharing, heightened merger scrutiny, and restrictions on common business practices, with fines reaching 20 percent of gross revenue in the affected line of business. House Judiciary Committee oversight has warned that the proposal appears designed primarily to reach large U.S.-headquartered platforms.

I know these examples are commonly cited by groups in the U.S., but it is important to note that the foreign governments at issue here all claim there are legitimate policy reasons for their actions. While there may be a disparate impact on American companies, they argue, that is not enough to find the actions to constitute "discrimination," as the criteria under the legislation/regulation at issue are objective and the intent is not to protect domestic producers from foreign competition. In the view of the foreign government officials, these actions are all legitimate and evenhanded.

I would also note that while the Baumgartner legislation seems focused on tech regulators, it is written in a way that would give it a much broader reach. For example, anyone responsible for any sort of tariff, including trade remedies, could be covered.

I have no idea whether this legislation will become law. And if it did, I have no idea how it would be applied. Would U.S. immigration courts suddenly have to act like trade dispute panels and make determinations about whether discrimination exists as a result of governments' enforcement of their regulations? That doesn't seem like a great role for them to be playing.

Executive Summary

This paper offers general observations on the key mechanisms and WTO consistency of the Proposal for the EU Cybersecurity Act 2 (‘CSA2.0’). The proposed CSA2.0 lays down the ICT supply chain security mechanism and the European cybersecurity certification mechanism, under which the European Commission may designate certain third countries as posing cybersecurity concerns, by taking account of the cyber-related legal systems and practices of such countries. Providers of ICT products and services that are established in, or have certain affiliation with, the designated third countries constitute ‘high-risk suppliers’. Most notably, the proposed CSA2.0 prohibits ‘high-risk suppliers’ from providing ICT products and services for ‘key ICT assets’ used by EU entities in 18 critical sectors and denies their access to cybersecurity certification and other market opportunities in the EU. If adopted, the proposed CSA2.0 will have significant economic impact on sectors where foreign exporters play a major role on the markets, including those related to telecommunications, solar inverters, battery energy storage, and connected and automated vehicles. 

While the potential impact on trade is profound, it is surprising and disappointing that the EU seems to have largely ignored the WTO consistency issues inherent in the proposed CSA2.0, as they have never properly and sufficiently addressed these issues in the proposal and other accompanying documents during the legislative process. China’s Ministry of Commerce, however, has questioned the WTO consistency of this legislative proposal in its comments submitted to the European Commission, pointing to violations of various obligations under the WTO agreements. If adopted, the CSA2.0 may trigger a major WTO dispute due to the systemic concerns shared by many affected countries. Against this background, this paper provides preliminary thoughts on the WTO consistency of the CSA2.0, with a particular focus on the non-discrimination obligations under the GATT 1994, the GATS, and the TBT Agreement.

The non-discrimination obligations under the GATT 1994 and the GATS are clearly applicable to the measures targeting products and services of the ‘high-risk suppliers’, as defined by the proposed CSA2.0, while the applicability of the TBT Agreement is less straightforward and requires further demonstration by the potential complainant. One of the key issues concerning those measures’ consistency with the said obligations is the ‘likeness’ test under the WTO agreements, which can be shown through the comparison of relevant factors or presumed if the measure is origin-based. Either way, a complainant is likely to be able to show that the products are ‘like’.

On the basis that the affected products are ‘like’, this paper further observes that under Article I of the GATT 1994, the blanket prohibitions established by the CSA2.0 fail to accord to ICT products of high-risk suppliers from a Member, immediately and unconditionally, the advantages accorded to those of other suppliers from other countries. Under Article III:4 of the GATT 1994, it is also not difficult for a complaint to show that the imported ICT products of high-risk suppliers are treated less favourable than the ‘like’ domestic ICT products of other suppliers, as the former’s products are outright excluded from the EU market. The same is true for claims under Articles II and XVII of the GATS. On the other hand, however, to establish violations of the non-discrimination obligations under Articles 2.1 and 5.2 of the TBT Agreement may be trickier, given the doctrine of legitimate regulatory distinction underlying these clauses.

When a complainant successfully establishes violations of the non-discrimination obligations under the GATT 1994 and the GATS, the EU will likely argue that the CSA2.0 is justified under the applicable exceptions of these WTO agreements. For the general exceptions, even assuming that any of the grounds for justification are relevant, which does not appear to be the case, it would be challenging for the EU to establish that such exclusionary measures introduced by the CSA2.0 are ‘necessary’ for protecting the stated objective of cybersecurity and/or the application of these measures does not constitute arbitrary discrimination or unjustifiable discrimination. For the security exceptions, it is clear from the treaty text as consistently interpreted in prior WTO disputes that political or economic differences between Members themselves are not sufficient to constitute an ‘emergency in international relations’. 

The EU's trading partners—most notably China—have warned that if the CSA2.0 is adopted in its current form, they will enact corresponding countermeasures against EU businesses. Moreover, if adopted, it is likely that China will initiate WTO dispute settlement proceedings against the EU and its Member States, which implement the CSA2.0. China could bring a WTO complaint against, in particular, Germany and France, which both have substantial trade volumes with China and played a decisive role in the CSA 2.0 proposal's adoption. Note that China and the EU are parties to the Multi-Party Interim Appeal Arbitration Arrangement (MPIA), and that, therefore, any WTO dispute between them will be brought to a legally binding resolution.

Introduction

To maintain momentum in the ongoing WTO Reform negotiations, the European Union circulated on 13 July 2026 three new discussion papers on subsidies and industrial policy (WT/GC/REFORM/W/5), foundational issues (WT/GC/REFORM/W/6), and the governance and decision-making (WT/GC/REFORM/W/7). Each of these papers makes a valuable and inspired contribution to the discussion on how to adapt the multilateral trading system and its principal institution, the WTO, to the geopolitical and geoeconomic realities of the 2020s. They reflect the European Union’s longstanding view that international trade must be rules-based. Recently, however, the European Union has adopted, or is considering, several trade-related measures of which the consistency with WTO law is, at the very least, debatable. Be that as it may, in the internal policy discussions on these measures (adopted or under consideration), the question of their WTO consistency is remarkably muted or virtually absent. This is odd for a WTO Member which professes to be a strong supporter of rules-based international trade. The current rules may be in urgent need of change, but they are still the ones that apply and should be abided by until new rules are in place. A telling example of this absence of consideration of WTO consistency in the internal policy discussions is the European Commission’s Proposal for the EU Cybersecurity Act 2 (‘CSA2.0’)[2]. As the European Commission points out in the Explanatory Memorandum of its Proposal, in recent years ‘cyberattacks have surged and became more sophisticated, targeting critical infrastructure, businesses, and the general public’.[3] It is clear that the threat to cybersecurity has significantly worsened, and that the current EU cybersecurity legislation and policy instruments, primarily including the EU Cybersecurity Act of 2019 (‘CSA1.0’)[4], the Directive (EU) 2022/2555 of 14 December 2022 (‘NIS 2 Directive’)^[5] , and the 5G Cybersecurity Toolbox, fall short of addressing this heightened threat. The transposition of the NIS 2 Directive into the domestic law of EU Member States was slow and is uneven among Member States. On 8 July 2026, the European Commission announced its decision to refer Ireland, Spain, France and the Netherlands to the European Court of Justice for failing to notify measures transposing the NIS 2 Directive into national law.[6] The 5G Cybersecurity Toolbox is a non-binding policy document, merely requesting Member States to assess the national risk profiles of their 5G network suppliers on inter alia the likelihood of interference by non-EU governments, and to restrict or exclude ‘high-risk suppliers’ from participating in critical and sensitive assets. Member States have taken divergent attitudes and approaches to implementing the non-binding 5G Cybersecurity Toolbox. To achieve the EU’s stated objective of effectively protecting Europe from cyber threats, mandatory cybersecurity instruments are considered necessary. This is where the proposed CSA2.0 comes in.

Key Features of the Proposed CSA2.0

The CSA2.0, as proposed, establishes two principal mechanisms that are of particular importance to international trade in ICT products and services, namely the ICT supply chain security mechanism, and the European cybersecurity certification mechanism.

As to the first of these mechanisms, the proposed CSA2.0 establishes a ‘trusted ICT supply chain framework’ designed to address non-technical risks, i.e., risks not related to hard- or software vulnerabilities of ICT, products or service, but to geopolitical, legal, and systemic dependencies in ICT supply chains on third countries. Central to this ‘trusted ICT supply chain framework’ are five related concepts, namely ‘critical sectors’, ‘essential and important entities’, ‘key ICT assets’, ‘countries posing cybersecurity concerns’, and ‘high-risk suppliers’. Of these concepts, the first three reveal the very wide scope of application of the proposed CSA2.0. Eighteen economic sectors are defined as ‘critical sectors‘, including inter alia energy, transport, health, finance, water, digital infrastructure & ICT services, public administration and manufacturing. ‘Essential and important entities’ are all large and medium-sized companies operating in any of the 18 ‘critical sectors’. ‘Key ICT assets’ are ICT products manufactured or services provided by ‘essential and important entities’. The fourth concept, i.e., ‘countries posing cybersecurity concerns’, refers to third countries that, according to the European Commission, pose a non-technical risk to ICT supply chains, as indicated by security risk assessments or other public sources. In identifying ‘countries posing cybersecurity concerns’, the Commission shall consider four elements: (1) whether any laws or practices from that third country require entities under its jurisdiction to report software or hardware vulnerabilities to authorities before such vulnerabilities are known to have been exploited; (2) the absence of effective judicial remedies and independent democratic oversight mechanisms capable of correcting such security concerns; (3) substantiated information about malicious cyber activities or campaigns carried out by threat actors operating from the territory of that country; and (4) the lack of ability or willingness of the third country to cooperate with the European Commission or Member States to address such risks. Finally, the fifth concept, i.e., ‘high-risk suppliers’, is defined in the proposed CSA2.0 as referring to: (1) an entity established in a third country designated as posing cybersecurity concerns; (2) an entity controlled by such a third country; (3) an entity controlled by an entity established in such a third country; and (4) an entity controlled by a national of such a third country.[7] In short, the European Commission designates manufacturers of ICT products or providers of  ICT services as ‘high-risk suppliers’ based on their place of establishment, ownership and control structure. In addition, albeit under exceptional circumstances only, the European Commission can directly designate an entity as a ‘high-risk supplier’ without that entity having any affiliation with a country designated as posing cybersecurity concerns.[8]

Under the CSA2.0, as proposed, the European Commission may adopt three kinds of exclusionary measures adversely affecting ICT products and services of ‘high-risk suppliers’. First, the Commission may prohibit European providers of electronic communication networks from using, installing, or integrating in ICT components from ‘high-risk suppliers’ in operation of ‘key ICT assets’, and require that components of ‘high-risk suppliers’ already installed shall be phased out within no more than 36 months.[9] Second, the Commission may prohibit ‘essential and important entities’ operating in the 18 ‘critical sectors’ from using, installing, or integrating ICT components from ‘high-risk suppliers’ in operation of ‘key ICT assets’.[10]  Third, the Commission may exclude ICT components supplied by a specific entity from being used, installed or integrated by ‘essential and important entities’, without engaging in separate identification of ‘key ICT assets’ or designation of ‘countries posing cybersecurity concerns’.[11] In short, the CSA2.0, as proposed, empowers the Commission to prohibit European ‘essential and important entities’ from using, installing, or integrating ICT components from ‘high-risk suppliers’ in operation of ‘key ICT assets’. Furthermore, ‘high-risk suppliers’ are also precluded from participating in: (1) European standardisation activities in the area of cybersecurity; (2) public procurement procedures in relation to provision of ICT components used in ‘key ICT assets’; and (3) the EU funding programs related to provision of ICT components used in ‘key ICT assets’.^[12]  Overall, the CSA2.0, as proposed is designed to preclude from the EU market ‘high-risk suppliers’ and their ICT products and services which the EU considers to exhibit ‘non-technical risks’ associated with particular third countries.

Apart from the ‘ICT supply chain security mechanism’, the proposed CSA2.0 also establishes a second mechanism, which is of particular importance to international trade in ICT products and services, namely the ‘European cybersecurity certification mechanism’. This mechanism aims to create a unified certification regime across the EU and strengthens the European Cybersecurity Certification Framework (‘ECCF’), which was established under the CSA1.0. The ECCF provides for a mechanism to evaluate and attest that ICT products, services and processes meet the applicable cybersecurity requirements. Certification under the ECCF is generally voluntary, unless otherwise specified in EU or national law.^[13]  EU Members States and the European Commission may make certification mandatory for certain types of ‘essential and important entities’.^[14]  Also, under the EU Cyber Resilience Act of 2024, EU cybersecurity certification is mandatory certification for critical products with digital elements, listed in Annex IV of this Act.^[15] A manufacturer of ICT products or provider of ICT services may either apply for and obtain a conformity certificate or issue a statement of conformity, both of which have the effect of presumed conformity with the relevant technical requirements, and should be recognised across all EU Member States. Besides general rules fleshing out the European cybersecurity certification regime, the CSA2.0 introduces three targeted restrictions on ‘high-risk suppliers’ regarding cybersecurity certification. First, ‘high‑risk suppliers’ are not entitled to apply for or be a holder of any European cybersecurity certificates.[16] Second, ‘high-risk suppliers’ are not entitled to become accredited conformity assessment bodies or authorised attestation providers.[17] Third, holders of a European cybersecurity certificate are prohibited from using, installing, or otherwise integrating ICT components from ‘high‑risk suppliers’ in their certified ICT products and services that are identified as ‘key ICT assets’.[18] In short, the CSA2.0, as proposed, excludes ICT products and services of ‘high-risk suppliers’ from the EU market by denying them certification or a statement of compliance.

Markets and Market Players Potentially Affected by the Proposed CSA2.0

As noted above, the scope, of the proposed CSA2.0 is very broad as it affects the 18 critical sectors identified by the NIS 2 Directive. Its potential impact on non-EU manufacturers of ICT products and suppliers of ICT services is therefore significant. First and foremost, the CSA2.0, as proposed, affects the digital communication and telecommunication market which was valued at EUR 1 trillion in 2023, and represents around 4.7% of the European GDP.^[19]  Key players in the telecommunication market include Nokia from Finland, Ericsson from Sweden, Huawei and ZTE from China, as well as Cisco from the US. Note in this regard, that eight European countries source more than 50% of their 5G RAN equipment from Chinese suppliers. In particular, in Germany, 59% of 5G RAN equipment in 2022 was supplied by Chinese suppliers.^[20]

The CSA2.0 would also affect the solar photovoltaic and battery energy storage market as its broad definition of ICT products also includes solar inverters, which are critical components of a photovoltaic system. In 2023, 70% of solar inverters installed in Europe were supplied by Chinese suppliers.[21] Between 2015 and 2023, Europe imported a total of 350 GW of solar inverters, of which 64% (225 GW) were supplied by Chinese companies, and 32.5% (114 GW) were supplied by Huawei.[22] Chinese enterprises have also expanded rapidly in the EU battery energy storage market, with companies such as Sungrow, CATL, BYD, and Huawei continuously securing major orders in Europe or planning to establish local manufacturing facilities. It is reported that Chinese manufacturers account for more than 80% of residential battery storage installations in Europe.[23]  In addition, China remains the main source of batteries import to the EU in 2025, and strengthened its position covering 85.5% of the EU external import needs in 2022-2024 (up from 81.9% in 2021-23).[24]

Also, the connected and automated vehicle (CAV) market would be affected by the CSA2.0 as the automobile industry, which accounts for 8% of European manufacturing value added and represents 6.1% of total EU employment^[25], falls under one of the ‘critical sectors’, namely manufacturing. In recent years, China, Turkey and the UK are the biggest vehicle exporters to the EU. Especially, China’s exports to the EU have shown a remarkable growth, rising from the 8th largest exporter of vehicles in 2020, to the highest volume exporter to the EU in 2024, with EUR 781 billion worth of motor cars and vehicles, and EUR 784 billion worth of motor vehicle parts exported.^[26]

In short, the potential adverse impact of the proposed CSA2.0 is significant and this is especially so for Chinese manufacturers of ICT products and providers of ICT services, as they are important players on the affected markets.

Questions Regarding the WTO Consistency of the Proposed CSA2.0

The proposed CSA2.0 and particularly its principal mechanisms discussed above raise important questions regarding its WTO consistency. These questions concern the consistency with the non-discrimination and market access obligations of the GATT 1994, the GATS and the TBT Agreement. Considering that the European Union professes to be a champion of rules-based trade, it is surprising and frankly disappointing, but perhaps a sign of the times, that the European Commission in its CSA2.0 proposal and accompanying documents pays scant, if any, attention to the WTO consistency of the CSA2.0. There is not even a reference to the WTO or WTO law in the text of the draft CSA2.0 or the CSA2.0 Explanatory Memorandum.[27] The CSA2.0 Impact Assessment refers twice to the EU’s obligations under WTO law. On p. 93, the European Commission states:

‘As for the impact on international competitiveness and trade, the considered options are compatible with the EU’s legal commitments under the World Trade Organisation (WTO) Agreements as well as bilateral, multilateral and plurilateral agreements. The options D.1–D.3 provide for a necessary and proportionate intervention to ensure cybersecurity of critical ICT supply chains. Whilst measures taken under these options could have a trade restrictive effect for certain goods and services, these measures would be justifiable in view of the overall legitimate objective to ensure the security of critical ICT supply chains in the EU.’ [28]

This is but a bold assertion of WTO consistency, without any analysis or supporting reasoning.  The second reference to WTO obligations is to Article 2.9 of the TBT Agreement. The European Commission notes on p. 343 of the CSA2.0 Impact Assessment that:

‘Under the TBT Agreement, WTO Members have the obligation to notify, through the WTO Secretariat, draft measures that may have a significant effect on trade of other Members and are not based on relevant international standards. In such a case, the Commission shall also submit a WTO TBT notification.’ [29]

Note that on 20 July 2026, six months after the European Commission published its CSA2.0 proposal, the EU had still not notified the WTO Secretariat of this draft measure.

Contrary to the Commission’s bold assertion of WTO consistency, China’s Ministry of Commerce (MOFCOM) has, unsurprising, cast serious doubt on the WTO consistency of the CSA2.0. In its comments on the CSA2.0, submitted to the European Commission on 17 April 2025, MOFCOM noted:

A victim of protectionism and unilateral bullying itself, the EU has been calling for upholding the rules-based multilateral trading system and opposing unilateral bullying and coercion. Nonetheless, the Proposal goes the opposite way.[30] 

MOFCOM argues that the CSA2.0 violates: the MFN and national treatment obligations of Articles I and III of the GATT 1994 and Articles II and XVII of the GATS; the prohibition of quantitative restrictions of Article XI of the GATT; the obligation of Article VI of the GATS that all measures affecting trade in services must be administered in a reasonable, objective and impartial manner; Article 3 of the SCM Agreement, which prohibits subsidies contingent upon the use of domestic over imported products; Articles 2.1 and 5.1 of the TBT Agreement, which impose MFN and national treatment obligations  on technical regulations and conformity assessment procedures; Articles 2.2 and 5.2 of the TBT Agreement, which require that technical regulations and conformity assessment procedures are not more trade-restrictive than necessary to fulfil a legitimate objective; Article 2.9 of the TBT Agreement on the notification of draft technical regulations; and finally, Article 39 of the TRIPS Agreement relating to the protection of undisclosed information.[31] MOFCOM further argues that none of the above alleged inconsistencies can be justified under general or national security exceptions of Article XX or XXI of the GATT 1994 or Articles XIV and XIV bis of the GATS. MOFCOM seems to throw everything but the kitchen sink at the proposed CSA2.0. Its list of alleged WTO violations is long but could be even longer. One might inter alia also argue that the proposed CSA2.0 is inconsistent with Article X:3(a) of the GATT 1994, which requires measures affecting trade in goods to be administered in a reasonable, objective and impartial manner; Article XVI of the GATS , which prohibits market access barriers in sectors for which the European Union has made market access commitments; Article 2.4 of the TBT Agreement, which requires that technical regulations are based on international standards; and Article 2.5 of the TBT Agreement, which requires that national certification bodies use relevant guides or recommendations regarding conformity assessment procedures issued by international standardizing bodies. MOFCOM may have good reasons for not including the ‘missing’ claims of inconsistency, but these reasons are not clear. If CSA2.2 is adopted as it is currently proposed, China is likely to challenge its WTO consistency and bring a formal complaint against the European Union or against the European Union and its Member States. As far as the EU Member States are concerned, the complaint would focus on the measures implementing the CSA2.0. China may, however, not be the only WTO Member bringing a formal complaint. This is because of the way in which the CSA2.0 would limit trade and the justification for such limitation invoked by the European Union. The WTO consistency of the CSA2.0 raises important systemic issues. Any future EU – CSA2.0 dispute would certainly have a record number of third parties. Also note that since both the European Union and China are both MPIA parties, this dispute could not end up in legal limbo due to an appeal to the paralysed Appellate Body, but will result in a legally binding resolution of the dispute.

It is not my ambition with this paper to dive into a detailed analysis of all of possible claims of WTO inconsistency of the proposed CSA2.0. I will limit myself to a few preliminary observations on the possible inconsistency with the non-discrimination obligations under the GATT 1994, the GATS and the TBT Agreement and on whether such violations could be justified. If the CSA2.0 were found inconsistent with the non-discrimination obligations and such inconsistency were be justifiable, the European Union would have to redesign the core elements of this measure and any finding of inconsistency with other WTO provisions of secondary importance.

For claims of inconsistency with the non-discrimination obligations to be successful, a complainant would first have to show the applicability of the relevant obligations. The applicability of Articles I or III of the GATT 1994 is clear. As it currently stands, the CSA2.0 is a ‘law … affecting the sale … of products’ to which Article III:4 of the GATT 1994 applies, and Article I of the GATT 1994 applies to ‘all matters referred to in paragraphs 2 and 4 of Article III’. Also Articles II and XVII of the GATS would apply to the CSA2.0 as the latter is a ‘measure by a Member affecting trade in services’ within the meaning of Art. I of the GATS; none of the EU’s exemptions from the MFN treatment under Article II:2 is relevant; and, as reflected in its Services Schedule, the European Union has made national treatment commitments in the services sectors and regarding the modes of supply covered by the CSA2.0. The applicability of the non-discrimination obligations under Articles 2.1 and 5.2 of the TBT Agreement may be less obvious. For example, for Article 2.1 of the TBT Agreement to apply, the measure at issue must be a technical regulation. Pursuant to Annex 1.1 of the TBT Agreement, a technical regulation is ‘a document which lays down product characteristics or their related processes and production methods.’ Does the CSA2.0 lay down the product characteristics or their related PPMs of the ICT products affected by it? To date, no WTO adjudicator has addressed the question whether a measure of this kind constitute a technical regulation within the meaning of Annex 1.1. This question is still to be decided.

Having shown that some or all the relevant non-discrimination provisions apply, the complainant would subsequently have to show that the products or services at issue, i.e., the ICT products or services of ‘high-risk suppliers’ and ICT products or services of other suppliers, are ‘like’ products or services. According to well-established case law, products and services may be presumed to be ‘like’ when the measure at issue distinguishes between products or services solely on the basis of their origin. The CSA2.0, as proposed, distinguishes between the ICT products and services on the basis of the origin of the product or service, namely on the basis of whether the product is manufactured or the service provided by a supplier with a link to a country posing cybersecurity concerns. A complainant can thus argue that the ‘likeness’ of the ICT products and services may be presumed. Alternatively, a complainant may, of course, address the ‘likeness’ issue in full. The determination of the ‘likeness’ of products or services is, according to firmly established case law, a determination of the nature and extent of the competitive relationship between the products or services at issue. This determination is made based on factors such as – when products are concerned – physical characteristics, end use, consumer tastes and preferences and customs classification, and – when services are concerned – the characteristics of the services and service suppliers and consumer preferences regarding these services and service suppliers. None of these factors is considered determinative in and of themselves and other factors may also be of relevance. I will limit my comments to the ‘likeness’ of the products at issue, but I note that similar considerations may be relevant for the ‘likeness’ of the services at issue. It is clear is that both the end use and the customs classification of ICT products supplied by ‘high-risk suppliers’ and those supplied by other suppliers are the same. The physical characteristics would also be the same, as ‘physical’ should be understood as referring to properties pertaining to the product itself. As to consumer tastes and preferences, note that the consumer of the ICT products at issue would, generally speaking, not be ordinary citizens but entities with advanced technical knowledge of the products purchased. The tastes and preferences of these entities are likely to be determined on objectively established technical differences between the ICT products at issue, rather than non-technical differences. Taking all these factors into consideration, a complainant is likely to be able to show that the competitive relation between the ICT products at issue is sufficiently strong for these products to be considered ‘like’ under Article I and III:4 of the GATT 1994 and Article 2.1 of the TBT Agreement.

Having established the likeness of the products and services at issue, a complainant must show, under Article I of the GATT 1994, that the ICT products of ‘high-risk suppliers’ from a Member are not accorded, immediately and unconditionally, the advantages accorded to ‘like’ ICT products of other suppliers from other countries. This is the case. Under Article II of the GATS, a complainant must show that services of ‘high-risk suppliers’ from a Member are accorded less favourable treatment than is accorded to ‘like’ ICT services of other suppliers from other countries. Under Articles III:4 of the GATT 1994 and XVII of the GATS, the complainant must show that the imported ICT products or services of ‘high-risk suppliers’ are treated less favourably than the ‘like’ domestic ICT products or services of other suppliers. Showing treatment less favourable within the meaning of Article III:4 of the GATT 1994 or Articles II and XVII of the GATS requires the complainant to show the CSA2.0 distorts the conditions of competition between the products and services at issue to the detriment of the imported ‘like’ ICT products or services of ‘high-risk suppliers’. Also, this is clearly the case as the latter ICT products or services are outright excluded from the EU market. Establishing whether ICT products of ‘high-risk suppliers’ are accorded less favourable treatment within the meaning of Article 2.1 of the TBT Agreement is trickier, because, in the case of de facto discrimination, even if the CSA2.0 would have a detrimental impact on the conditions of competition, it would not be considered to accord treatment less favourable if that detrimental impact stems exclusively from a legitimate regulatory distinction. The question a complainant would have to address is whether the distinction made in casu, namely whether ICT products are supplied by ‘high-risk suppliers’ or not, is a legitimate regulatory distinction. The comments made below regarding the general exceptions under Article XX of the GATT 1994 are of relevance to this question.

Assuming a complainant is successful in establishing that the CSA2.0, as proposed, is inconsistent with Articles I and/or III:4 of the GATT 1994 and/or Articles II and/or XVII of the GATS, it would be for the European Union to show that these inconsistencies can be justified under Articles XX or XXI of the GATT 1994 or Articles XIV or XIV bis of the GATS. The grounds of justification of otherwise GATT- or GATS-inconsistent measures are exhaustively set out in Articles XX of the GATT 1994 and Article XIV of the GATS respectively. Few, if any, of these grounds appear to be relevant. Moreover, for certain of these grounds, the European Union would be required to show that the CSA2.0 is ‘necessary’ to protect itself from cybersecurity risks. The CSA2.0 will only be considered ‘necessary’ if none of the possible alternative measures, identified by the complainant, is a less trade restrictive measure that is reasonably available and that would achieve the level of protection the European Union wants to achieve. Assuming the European Union can show that the CSA2.0 is ‘necessary’, the latter would be provisionally justified under Articles XX of the GATT 1994 or XIV of the GATS. However, to be justified, the application of the CSA2.0 would need to meet the requirements of the chapeau of both provisions, i.e., the application of the CSA2.0 may not constitute arbitrary discrimination or unjustifiable discrimination. In this regard, the following questions may need to be addressed: (1) whether the European Union applies the CSA2.0 in an overly rigid and inflexible manner without due regard to the actual conditions in the country of origin of the ICT product or service; (2) whether the European Union tried to address its concerns regarding cybersecurity through multilateral negotiations before having recourse to unilateral measures, such as the CSA2.0; and (3) whether in its application, the CSA2.0 reveals discrimination that cannot be reconciled with, or is not rationally related to, the policy objective, i.e. the protection of cybersecurity, pursued by the European Union. Showing that the CSA2.0 is applied in a manner that does not constitute arbitrary or unjustifiable discrimination may be tricky for the European Union.

The European Union may also try to justify the otherwise GATT or GATS inconsistent CSA2.0 by invoking the national security exceptions of Article XXI of the GATT 1994 and Article XIV bis of the GATS. Note in this regard that in recent years the European Union strongly asserted that the national security exceptions should not be overstretched or abused. Arguably of some relevance in this case is paragraph b of either of these provisions, and, in particular, subparagraph (iii) thereof. Subparagraph (iii) concerns measures taken in times of war or other emergency in international relations. The first of these circumstances, ‘war’, is thankfully not present, but can it be argued that cyber insecurity, threats and attacks constitute an emergency in international relations? The Panel in Russia – Traffic in Transit (2019) defined an emergency in international relations as ‘a situation of armed conflict, or of latent armed conflict, or of heightened tension or crisis, or of general instability engulfing or surrounding a state’. The Panel in that case further clarified that political or economic differences between Members are not sufficient, in and of themselves, to constitute an ‘emergency in international relations’, unless they affect defence and military interests, or the maintenance of law and public order interests.

The above considerations on the consistency of the CSA2.0, if adopted as currently proposed, with the non-discrimination obligations of the GATT 1994, the GATS and the TBT Agreement are just a broad and incomplete overview of the challenges that either a complainant or the European Union must overcome to show that the CSA2.0 is WTO-inconsistent or -consistent respectively. The European Commission’s proposal is currently moving through the long EU legislative process and is currently before the Council of Ministers. A progress report by the Presidency of the Council, discussed on 7 June 2026, listed the issues raised by Member States regarding the proposed CSA2.0. The WTO consistency of the CSA2.0 is not explicitly mentioned but may well have been discussed. The CSA2.0 is also being discussed in the European Parliament (‘EP’), where the proposal has been assigned to the Industry, Research and Energy Committee (and not the International Trade Committee). Marketa Gregorova from Czechia and member of the Greens/EFA Group in the EP (and also a member of the EP International Trade Committee), has been appointed as rapporteur. It will be interesting to see whether Ms. Gregorova will address the WTO consistency in her report and/or whether this issue will be raised in the discussions in committee or in plenum. In a European Parliament’s Briefing on the CSA2.0, dated 16 June 2026, it is noted: 

Non-EU trade representatives and vendors have raised concerns about the legal underpinnings of the non-technical risk framework and its application. China's Minist[ry] of Commerce stressed how the proposal introduces highly subjective and arbitrary 'non-technical risk' in the name of cybersecurity and supply chain security. It notes how such exclusion policies could disrupt market operations and conflict with WTO rules and multilateral trade norms. It suggests removing provisions related to 'third countries posing cybersecurity concerns' and 'non-technical risks', and delete or substantially revise the criteria for identifying [‘high risk suppliers’], so that supply chain cybersecurity measures remain technology-neutral, evidence-based and proportionate in scope.^[32] 

As MOFCOM stated in its comments on the CSA2.0 submitted to the European Commission on 17 April 2025, China ‘stands ready to conduct in-depth and frank exchanges’ on cybersecurity, but it did also warn the European Union that if the CSA2.0 is adopted as proposed, ‘China will have to respond in kind by carrying out corresponding countermeasures against the EU and EU businesses with reference to the EU’s logic and measures. [33] The European Union would act with wisdom if they were to re-consider the proposed CSA2.0 to ensure its WTO consistency and thus uphold the rules-based multilateral trading system of which it professes to be the champion. 

[1] Chair Professor of International Economic Law, Xi’an Jiaotong University School of Law; former Member and Chair of the WTO Appellate Body; Professor Emeritus of International Economic Law, World Trade Institute, University of Bern; and external legal advisor, King & Wood.

[2] Proposal for a Regulation of the European Parliament and of the Council on the European Union Agency for Cybersecurity (ENISA), the European cybersecurity certification framework, and ICT supply chain security and repealing Regulation (EU) 2019/881 (The Cybersecurity Act 2), COM(2026) 11 final, dated 20 January 2026, https://digital-strategy.ec.europa.eu/en/library/proposal-regulation-eu-cybersecurity-act.

[3] Ibid., p. 1.

[4] Regulation (EU) 2019/881 of the European Parliament and of the Council, Official Journal of the European Union of 7 June 2019, L 151/15, https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32019R0881.

[5] Directive (EU) 2022/2555 of the European Parliament and of the Council of 14 December 2022  on measures for a high common level of cybersecurity across the Union, amending Regulation (EU) No 910/2014 and Directive (EU) 2018/1972, and repealing Directive (EU) 2016/1148 (NIS 2 Directive), https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32022L2555. This Directive replaced  

[6] European Commission, Press Release, 8 July 2026, https://ec.europa.eu/commission/presscorner/detail/en/ip_26_1499.

[7] Article 2(39) of CSA2.0.

[8] Article 103(6)(7) of CSA2.0.

[9] Articles 110 and 111 of CSA2.0.

^[10] Article 103.1 of CSA2.0.

[11] Article 103.7 of CSA2.0

[12] Article 100.4 of CSA2.0.

[13] Article 71.3 of CSA2.0.

[14] Article 24 of the NIS2 Directive.

[15] Article 32.4 of Regulation (EU) 2024/2847 of the European Parliament and of the Council of 23 October 2024 on horizontal cybersecurity requirements for products with digital elements and amending Regulations (EU) No 168/2013 and (EU) No 2019/1020 and Directive (EU) 2020/1828 (Cyber Resilience Act).

[16] Article 100.4(b) of CSA2.0.

[17] Article 100.4(c) and (d) of CSA2.0.

[18] Article 103.1 of CSA2.0.

[19] State of Digital Communications 2025, https://connecteurope.org/insights/reports/state-digital-communications-2025

[20] The Market for 5G RAN in Europe: Share of Chinese and Non-Chinese Vendors in 31 European Countries, https://strandconsult.dk/the-market-for-5g-ran-in-europe-share-of-chinese-and-non-chinese-vendors-in-31-european-countries/

[21] Commission Staff Working Document Impact Assessment Report, SWD(2026) 11 final, https://digital-strategy.ec.europa.eu/en/library/proposal-regulation-eu-cybersecurity-act

[22] Ibid..

[23] Anna Darmani Tous Isabel Nieto, ‘Chinese Suppliers Solidify Control of Europe’s Home Battery Market’ (pv magazine Global, 29 October 2025), https://www.pv-magazine.com/2025/10/29/chinese-suppliers-solidify-control-of-europes-home-battery-market.

[24] Battery Technology in the European Union - 2025 Status Report on Technology Development, Trends, Value Chains and Markets, 60, https://setis.ec.europa.eu/battery-technology-european-union-2025-status-report-technology-development-trends-value-chains-and_en.

[25] EU Coordinated Risk Assessment-Connected and Automated Vehicles, 30 January 2026, https://digital-strategy.ec.europa.eu/en/policies/nis-cooperation-group

[26] CSA2.0 Proposal (n 2), https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026PC0011.

[27] https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026PC0011

[28] Commission Staff Working Document Impact Assessment Report, SWD(2026) 11 final, dated 20 January 2026, at https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52026SC0011&qid=1782442565955

[29] Ibid.

[30] Feedback from the Ministry of Commerce of China on the EU’s Proposal for a revised Cybersecurity Act, dated 17 April 2025, at https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/14578-The-EU-Cybersecurity-Act/F33393498_en.

[31] Ibid.

[32] European Parliament Briefing on EU Legislation in Progress, Cybersecurity Act Revision (CSA2), dated 16 June 2026, p. 10, https://www.europarl.europa.eu/RegData/etudes/BRIE/2026/789345/EPRS_BRI(2026)789345_EN.pdf.

[33] MOFCOM (n 30).

The quote comes from Wallace McClure, a legal scholar who worked at the U.S. State Department for many years and published a book in 1924 called "A New American Commercial Policy, As Evidenced by Section 317 of the Tariff Act of 1922." You can read and download the whole book here. Below I have excerpted a long discussion of what "discrimination" means and what some examples are (footnotes omitted, and hopefully copying and pasting from an OCR'd book did not leave any errors). As I said in my earlier post, I have doubts as to whether these tariffs will be imposed, but if it does happen at some point, and they end up being litigated in court, the parties will have a lot of material to work with in interpreting and applying the term "discrimination" in this context.

One short passage from the McClure book that might be of particular relevance is the following:

Somewhat similar is the purpose of stipulating that a discriminatory practice is not to be actionable unless it is "unreasonable". There are certain practices which, though clearly discriminatory, are commonly accepted as justifiable because of the existence of peculiar or unusual circumstances.

In the case of U.S. Section 338 tariffs imposed on Canada, a key point in contention in any litigation in U.S. courts could be whether the circumstances here involve "practices which, though clearly discriminatory, are commonly accepted as justifiable." With regard to whether the Canadian actions constitute discrimination, the following justifications could be offered: (1) the Canadian actions on motor vehicles and alcoholic beverages are in retaliation for earlier U.S. tariffs; and (2) Canadian discrimination in favor of EU products through its dairy trade regime is permitted under a bilateral trade agreement that is consistent with WTO rules. The question is, what would a U.S. court think of these justifications?

Here's the rest of the discrimination sections from McClure:

3. WHAT IS A "DISCRIMINATION"?

In a carefully prepared address upon the administrative features of the pending tariff bill, Senator Smoot explained at length the proposed Flexible Tariff Policy, reaching the conclusion that "the elastic tariff provisions provided effective protection against discriminations for American overseas commerce."

The Tariff Commission, in formulating its policy and making its recommendations under Section 317, may be presumed to have laid down full and clear definitions of the expressions, contained in the law, which limit the application of the authorized defensive duties. Perhaps the most important problem arising in this connection, and certainly the immediately obvious one, is concerned with the meaning of the phrases "unreasonable charge, exaction, regulation, or limitation which is not equally enforced upon the like articles of every foreign country" and "discriminates in fact against the commerce of the United States such manner as to place the commerce of the United States at a disadvantage compared with the commerce of any foreign country." In other words, the definition of "discrimination", within the meaning of Section 317, is an essential question – and one not without difficulties. Experience has shown that an extensive margin exists between practices that are clearly consistent with strict equality of treatment and practices that are clearly discriminatory.

In the statement made by the House Managers to accompany the report of the Conference Committee on the tariff bills of the House and the Senate, occurs the following:

"... The Senate amendment inserts a new section giving the President discretionary powers to impose additional duties or prohibition upon imports from any country discriminating against the overseas commerce of the United States.

This section follows the precedent established by a maximum and minimum provision of the Payne-Aldrich Act, which had for its purpose the obtaining of equality of treatment for American overseas commerce. The Senate amendment, however, is more flexible than the provision of the Payne-Aldrich Act and is designed to reach every form of discrimination, direct or indirect, whereby American commerce is placed at a disadvantage as compared with the commerce of any foreign country."

Attention is called particularly to the statement that the language employed "is designed to reach every form of discrimination, direct or indirect." It was evidently expected that a very broad and inclusive definition of "discrimination" would be used by the administrative officers of the Government in interpreting and enforcing the law.

This impression is confirmed by an examination of the text of the Act. In subdivision (a) the additional duties are made applicable when another country imposes any unreasonable charge on a product of this country which is not equally enforced upon the like articles of every foreign country; and also when another country discriminates in fact against the commerce of the United States.

Such language seems to indicate that legalistic distinctions of all kinds are to be ignored and that the defensive duties are to be applicable whenever a different and larger burden is placed on American as compared with any other external commerce. "In fact" doubtless possesses, however, its familiar connotation of "material", "real" or "actual", as distinguished from "theoretical" or "as a matter of law". The use of the words suggests the intention of the framers of the Section to confine its application to the service of practical utility, that is, to obtain for American exporters substantial equality of treatment, omitting action in regard to practices that, though discriminatory on paper, do not interfere with the practical flow of trade. Somewhat similar is the purpose of stipulating that a discriminatory practice is not to be actionable unless it is "unreasonable". There are certain practices which, though clearly discriminatory, are commonly accepted as justifiable because of the existence of peculiar or unusual circumstances. A good example in point is the freedom of trade until recently permitted between the corner of Switzerland in which Geneva is located and the adjoining districts of France. The arrangement for this overleaping of political frontiers originated in the time of Napoleon and, in eliminating the arbitrary political boundaries that divide a single economic area, appears to have proved very acceptable to both French and Swiss inhabitants. No one would contend that, because of it, Switzerland and France have unreasonably discriminated against the United States in favor of each other.

Inequalities that would otherwise be unreasonable may be considered permissible for reasons such as the requirements of sanitation or public safety.

The expression "unequal imposition or discrimination" occurs several times in Section 317, however, obviously with intent ordinarily to use the terms synonymously and to make any "unequal imposition" a "discrimination". That an exact interpretation of "discrimination" is the intention of the Section is suggested, moreover, by the abandonment in its favor of provisions contained in the original House Bill authorizing practices which would themselves have been discriminatory in operation. The House Managers, continuing the passage quoted above, said:

"Sections 301 and 303 of the House bill provide for special negotiations whereby exclusive concessions may be given in the American tariff in return for special concessions from foreign countries. Section 302 of the House bill places in the hands of the President power to penalize the commerce of any foreign country which imposes on its imports, including those coming from the United States, duties which he deems to be "higher and reciprocally unequal and unreasonable." Under the Senate amendment, however, the United States offers, under its tariff, equality of treatment to all nations, and at the same time insists that foreign nations grant to our external commerce equality of treatment; and the House recedes with an amendment rewriting subdivisions (e) and (f) and making further clerical changes."

Finally the breadth of meaning of discrimination is established by the definition of "foreign country" as "any empire, country, dominion, colony, or protectorate, or any subdivision thereof within which separate tariff rates or separate regulations of commerce are enforced." The language of this definition seems to leave no doubt that an unequal imposition upon the commerce of the United States when compared with impositions upon the commerce of a country's own colonies, if such colonies possess separate tariff laws or regulations of commerce, is a discrimination within the terms of Section 317.

4. EXAMPLES OF DISCRIMINATORY PRACTICES AND POLICIES

An examination of the present world commercial situation reveals numerous and varied discriminations, many of which seriously interfere with the commerce of the United States.

First, there is the existence, just referred to, of preferences between parts of an Empire. For instance, Canada accords to Great Britain and Great Britain accords to Canada treatment in respect to import duties that is more favorable than the corresponding treatment which either gives to the United States. Each of these two countries has its wholly separate customs laws and, although they are the best customers of the United States and the United States is one of the best customers of each of them, each imposes upon imports from the United States duties that are unequal to and heavier than the duties which it places upon similar goods imported from the other. As already indicated, such treatment constitutes a discrimination against the United States. The same is true in the case of the preferential export duties which certain colonies grant to the mother country.

Second, there are countries, notably France, which have enacted double-column schedules of import duties. The lower schedule commonly represents the rates which the tariff policy and budgetary needs of the country really require. The higher schedule is commonly expected to be used for bargaining purposes and as a defense or threat against countries which do not grant their lowest rates. Such is the case with France, which, notwithstanding the fact that its products receive equality of treatment in the American market, imposes maximum rates upon numerous products from this country. There is no question as to the liability of France to the imposition of additional duties under Section 317. Countries having this so-called maximum-minimum tariff system commonly bind themselves by treaty to accord all or portions of their minimum rates.

Third, there are countries, notably Switzerland and pre-war Germany, which have developed double-schedule tariff systems upon a plan substantially different from that just described. Their tariff laws as enacted by their legislatures contain one uniform schedule of rates; but, by entering into treaties providing on the one hand for lower rates and on the other hand for most-favored-nation treatment, they have gradually constructed what is in effect a system not dissimilar, with reference to the present discussion, to the maximum-minimum arrangement. The "general-conventional" system may be just as discriminatory under the terms of Section 317 as is the French practice. In either case the lowest duties could be granted to another country either freely or in return for reciprocal favors or concessions. There appears to be nothing in the language of Section 317 which forms the basis of any distinction as to its application between free favors and favors accorded for a consideration. Presumably, therefore, its defensive duties are applicable to discriminations against the United States resulting from either sort of preference to third countries.

Fourth, there are countries which have entered into one or more reciprocity treaties with other countries for the exchange of favors or concessions which are not generalized, that is, are not accorded to countries other than the two between which the particular reciprocity treaty is in force. An interesting example is Haiti, which has a single reciprocity treaty - with France - under the terms of which, in return for French minimum duties for certain of its products, it accords substantial special reductions to numerous goods of French origin and heavy reductions to French liquors. American goods which compete with those of France in the market of Haiti are undoubtedly placed at a disadvantage by the existence of this arrangement, which must be assumed to constitute a discrimination against the United States and so to render Haiti liable to the enforcement of the defensive duty provision of Section 317.

Finally, it is necessary to consider a large number of minor provisions of law and incidental practices that result in unequal burdens to American commerce. From time to time many little ways of favoritism crop out, often not even intimated by the text of the law. A thing so intangible as an unofficial act of a customs officer which would result in the delay of goods from a particular country, while hastening the passage through the customs house of the goods of a rival country, might conceivably be of genuine importance where competition is close or the market insufficient for two cargoes. Under such circumstances every little advantage counts in making sales. Some instances of concealed and other miscellaneous inequalities may appropriately be examined:

(a) Referring to the discriminations against American commerce which Section 317 was designed to combat, Mr. Smoot, in his above-quoted address to the Senate, mentioned the practice of certain countries,

"giving a separate classification to and levying a higher rate upon cottonseed oil than upon olive, palm, or other competing oils, or so adjusting their automobile duties that those types of cars which we export are subject to the highest rates."

This practice is one of long standing. It was the cause of much diplomatic correspondence when, following the enactment of the maximum-minimum provision of the Tariff Act of 1909, serious effort was made to obtain for cottonseed oil, a distinctively American product, a parity of treatment with edible oils ordinarily originating in other countries. Italy at the present time, notwithstanding its treaty assurance of most-favored-nation treatment to American products, imposes a higher duty upon cottonseed oil than upon other oils which are used for identical purposes, but which are not important exports of the United States. It is probable that the resulting unequal burden upon the American product is a discrimination within the meaning of Section 317. In other words, the competitive use to which a product is put, rather than its name and extractive source, would seem, within narrow limits and in perfectly clear cases, to be the decisive factor in respect to the operation of the defensive duties. Otherwise, ingenious refinements in tariff nomenclature might entirely defeat the purpose of the Section.

(b) A more debatable case is found in the practice of levying especially high duties upon products that are more important exports from the United States than from other countries but which do not compete with other products of a similar variety commonly originating elsewhere than in the United States. A case in point was furnished when Italy, in 1921, increased its import duty upon typewriters from seventy-five to four hundred lire gold per quintal. Reverting to the preceding paragraph, an intermediate example would result if, instead of differentiating between cottonseed and similar oils, all such oils were highly taxed and oil-bearing seeds and nuts were admitted freely or at low rates of duty. The invocation of Section 317 with respect to these practices seems improbable, though the one is and the other would be, in a sense, discriminatory against the commerce of the United States.

(c) Still another variation of what seems to be the common principle of the preceding examples is found where a product of the United States is dutiable at higher rates than the same product of another country differing slightly in process of manufacture or in constituent elements. The butter content of condensed milk manufactured in the United States and sold abroad is normally seven and eight-tenths per centum. The British Guiana Customs Duties Ordinance, 1922, inaugurated a drastic increase in the duty on condensed milk containing less than ten per centum of butter fat. Italy levies a tax of sixteen lire gold per quintal upon natural vaseline and thirty lire gold upon vaseline containing paraffine-a distinction which appears to affect adversely the importation of vaseline from the United States. Recent reports from another country have given account of a change in classification, as a result of which a certain brand of American-made hats appears to be subject to higher duties than similar hats of different makes, from whatever place originating. Whether or not these unequal burdens are discriminations within the meaning of Section 317 probably depends on whether they are bona fide regulations for the promotion of health, the avoidance of adulteration and the correction of erroneous classifications on the one hand or, on the other hand, instances of disguised favoritism. The determination of this question will obviously be difficult in many cases.

(d) It is a common practice among commercial nations to include, for the purpose of making assessments for ad valorem duties, the cost of transportation. This practice results in unequal impositions upon the commerce of more remote as compared with nearby countries. Similarly, the packing in which goods are contained is commonly dutiable and long-distance shipments, which require heavier packing, are consequently subjected to added burdens. The comparatively isolated location of the United States renders its commerce peculiarly vulnerable to these inequalities. This country itself levies duties upon packing, however, and the inclusion in valuation of such items as freight is a long-accepted practice which is recognized in at least one international convention - the final act of the Conference of Berlin as amended at Brussels in 1890. Moreover, the official valuations for the Chinese tariff have been fixed by the international commission, upon which the United States was represented, on the basis of the values of goods delivered at Shanghai. These two examples of unequal burdens could hardly be considered unreasonable discriminations by the American Government. It is interesting to note, however, that Switzerland, which is able to import automobiles brought in by their own power from France and Italy, was recently induced to lower its duties upon the heavy crating necessary for automobiles shipped overseas from the United States.

(e) Specific duties are collected by La Luz and Las Palmas, Canary Islands, on all merchandise entering their ports, at rates varying according to zones of origin. The first zone consists of Spain and its possessions; the second of the remaining ports of Europe, the Mediterranean Sea and part of the Atlantic coast of Africa; the third of the remainder of the world. The products of the United States thus pay the highest rates: whether they are discriminated against within the meaning of Section 317 forms a nice question with the chances of decision apparently favoring the negative.

(f) The parcel-post regulations of Venezuela require the collection of a fee of five cents for handling each package brought in from foreign countries except the United States, with respect to which the corresponding duty is thirty cents. This appears to be a palpable discrimination against the latter country.

(g) In certain countries – particularly in Central Europe — the emergencies of the war and reconstruction periods have developed regulations forbidding the importation or exportation, except under license, of the commodities named in extensive lists. Agreements to license the exchange of fixed quotas or contingents" of specified commodities have been concluded between countries. Where goods cannot cross the frontier except under license, the grant of which may be left to the discretion of administrative officials, opportunities for favoritism are obviously many. Other such opportunities may arise in the determination of limited quotas of goods for competing states. There is evidence that American commerce has suffered some detriment in this way, but such cases are difficult to prove. They seem undoubtedly discriminatory.

(h) The Comisión Exportadora de Yucatán is an official branch of the socialist government of that state, which has a complete monopoly of the production and marketing of Mexican sisal (henequen). A recent report to the Comisión, presented by its attorney, contained this statement:

"Since it is our aim to develop the consumption of our sisal hemp, our institution has undertaken to facilitate all the operations carried through in Europe, one of the many facilities afforded being that of assigning a lower price than the one quoted in the United States, taking into consideration the greater cost of transportation of the merchandise, there being no doubt that if, to the price obtainable in the United States, we should add the excess rate of freight for transportation to Europe, we would be placed in an unfavorable position in competing with the other fibers sent there."

In view of the recent immense progress of state socialism in certain important countries, the question whether a practice of the sort described is a discrimination, within the meaning of Section 317, seems of great potential moment.

There are also narrower possible challenges, such as an argument that focuses on the actions taken by affected governments to address the specific unfair trade practices at issue here. If a government were to improve its laws/enforcement related to imports made with forced labor, it could argue that its tariff should be lowered. In this context, note that as USTR moved from proposed to final tariff rates in the investigation, a handful of countries got a slightly lower tariff, falling from 12.5% to 10%, on the basis of actions taken along these lines. As the Presidential memorandum released by the White House states:

... the Trade Representative has informed me that following consultation with certain economies in these investigations and publication of the Notice of Determinations, additional economies have imposed forced labor import prohibitions (Cambodia, Guatemala, Honduras, India, Sri Lanka, and Trinidad and Tobago) or undertaken commitments regarding forced labor import prohibitions in an Agreement on Reciprocal Trade (Jordan).  As a result of these actions, the Trade Representative has advised me that the goods of these economies should be tariffed at the 10 percent rate to further encourage these economies to effectively enforce such prohibitions, and, in the case of Jordan, to enact and effectively enforce its commitments regarding forced labor import prohibitions.

Similarly, USTR's Federal Register notice explains:

Following government-to-government consultations pursuant to Section 303 of the Trade Act and the publication of proposed actions in the June 5, 2026 FRN, additional economies have imposed forced labor import prohibitions—Cambodia, Guatemala, Honduras, India, Sri Lanka, and Trinidad and Tobago; or undertaken commitments regarding forced labor import prohibitions in an ART—Jordan.

So the precedent exists: By addressing the unfair trade practices at issue, a government can get a lower tariff rate.

But will that precedent continue to be followed now that the final rates are set? My sense is that some governments will keep going with their efforts to improve their laws dealing with imports made with forced labor, either because they think it's a good idea or to try to satisfy U.S. demands. After they do so, presumably they will go to USTR (or President Trump directly) and ask for tariff relief. Will they get it?

The argument against an adjustment happening is that Section 301 does not provide any process for adjustments, so USTR doesn't have to do anything. The statute gives USTR lots of discretion, and the courts are likely to recognize this and stay out of USTR's business. As a result, the Trump administration – whose actual goal here is to impose a broad set of tariffs on worldwide imports – won't take any action to lower the tariffs.

The argument for an adjustment happening is that it's just plain common sense to make this adjustment, and maintaining the tariffs after the unfair trade practice has been eliminated would be arbitrary and capricious (to put it in Administrative Procedure Act terms). You can't impose a Section 301 remedy on unfair trade practices that no longer exist. And as the Presidential memorandum states, removing the unfair trade practices is the only purpose here: "Each tariff action directed in this memorandum is only for the purpose of obtaining the elimination of the specific economy’s act, policy, or practice found actionable under section 301 and not for any other purpose." If that purpose has been satisfied through elimination of the unfair trade practice, the tariff action no longer serves a purpose. Furthermore, the absence of an adjustment process spelled out in Section 301 shouldn't prevent USTR from making adjustments when it is reasonable to do so, and the courts won't let USTR get away with avoiding adjustments if that is the Trump administration's plan.

So how is this all going to play out? To find out, we'll need to have some government move ahead with its reforms, and then look for ways to make the case to the Trump administration that it has made improvements to its forced labor import laws or the enforcement thereof. And then, well, anyone's guess is probably as good as mine as to what happens next. I am more bullish on the argument for the courts stepping in to force USTR to change the rates than some people I've talked to are, but I don't have great confidence in any particular outcome here. If someone would ask Trump administration officials what their plan is in these circumstances, that could help shed a bit of light on the situation.

On July 21, 2026, the U.S and Jordan concluded and signed a new Agreement on Reciprocal Trade. Jordan has now officiated preferential access to the US market. The U.S has pursued these legally binding deals with a host of other countries with the purpose of re-shaping trade terms. The Agreements on Reciprocal Trade are not conventional free trade agreements. They are short, asymmetric instruments that use tariff access, and the threat of renewed tariff pressure, to secure commitments on customs enforcement, technical standards, digital trade, critical minerals, export controls, sanctions cooperation, forced labor, state-owned enterprises and “third country” trade practices.

The Agreement on Reciprocal Trade modified and advances the US-Jordan Free Trade Agreement (FTA) signed in 2001. The 2001 FTA successfully eliminated all standard bilateral duties by 2010. However, the Agreement on Reciprocal Trade was necessary to navigate recent across-the-board U.S. tariffs and modern economic realities not captured 25 years ago, such as digital data flows. The Agreement on Reciprocal Trade deal follows a robust trading period, with total bilateral trade hitting $5.34 billion in 2025. It coincides with heavy commercial ties, including Royal Jordanian Airlines' recent $1.4 billion purchase of Boeing Dreamliners and Jordanian pharmaceutical firm Hikma’s $1 billion U.S. investment plan.

Under the new Agreement on Reciprocal Trade, the U.S. commits to provide the preferential rate of duty in effect under the U.S. – Jordan FTA for goods that qualify as originating under the rules of origin under the U.S. – Jordan FTA, or the most-favored-nation rate of duty in effect, as applicable.[1] Key provisions of the Agreement on Reciprocal Trade include a commitment by the U.S. that it will reduce and cap reciprocal customs duties on Jordanian imports at 10 percent. This provides relief from the stiffer 20 percent reciprocal tariffs imposed by the U.S. in April 2025 reduced to 10% (for all countries) under Section 122 on February 24 and set at 12.5% under the section 301 action on forced labor as of July 24. In return, Jordan will maintain duty-free market access for nearly all U.S. exports, including agricultural products and vehicles. The Agreement on Reciprocal Trade binds Jordan to stricter enforcement of environmental protections, enhanced labor rights, and a prohibition on importing goods produced by forced labor within five years.[2] It should be noted that the tariff levels imposed on Jordan since April 2025 are a flagrant violation of the US-Jordan FTA, which provided that Jordanian exports to the U.S would enter duty-free when they qualified as originating under the FTA rules of origin.

More controversial provisions include article 4.1(2) (third-country enforcement clause) and whether it violates GATT or GATS. The purpose of this article is to eliminate transshipment and duty evasion by third-country entities trying to bypass standard U.S. tariffs. China, which currently commands a 19% share of Jordan’s domestic import market, frequently routes raw materials or semi-finished items through friendly trade partners to exploit lower tariff rates. For critical manufacturing sectors—most notably Jordan’s multi-million-dollar garment and textile industry—Article 4.1(2) requires comprehensive tracking of the production pipeline. The enforcement of Article 4.1(2) forces a shift in corporate compliance for companies operating out of industrial zones like Irbid or Zarqa. These companies must provide the complete upstream supply chain trail or face immediate retroactive tariff hikes up to the previous 20% default rate.

The Agreement on Reciprocal Trade prevents Jordan from implementing taxes on digital services being provided within its borders or customs duties on electronic transmissions, thereby denying Jordan what would be a source of revenue.[3] The Agreement on Reciprocal Trade requires Jordan to ensure unimpeded data flows to the U.S., including by recognizing the “adequacy” of U.S. data protection frameworks in its domestic privacy law.[4]

Like any other trade agreement, there is a trade-off and concessions made to secure access to the U.S. market. To secure this 10% tariff cap, Jordan conceded significant elements of its regulatory autonomy, particularly within the digital economy, technology governance, and domestic supply chain oversight. Through joint, unannounced customs inspections and mandatory data sharing, the U.S. has integrated its economic security priorities directly into Jordan's border enforcement. The Agreement on Reciprocal Trade between the U.S. and Jordan functions heavily as an anti-circumvention tool. The Agreement would make Jordan actively police its own industrial zones to ensure that Chinese and other non-market raw materials are not simply re-routed to exploit the U.S. market. In sum, the Agreement on Reciprocal Trade between the U.S. and Jordan is not "reciprocal" but totally one sided like the many other ARTs forced on other US trading partners. For other countries like Canada and Mexico, the new Agreement on Reciprocal Trade between the U.S. and Jordan would be the template for CUSMA review and what countries would expect from the U.S. entering trade negotiations.

[1] Annex I, art. 3, Agreement between the U.S and Jordan on Reciprocal Trade, available at < https://www.whitehouse.gov/wp-content/uploads/2026/07/Jordan-Annex-I.pdf >.

[2] Arts. 2.9 and 2.10, Agreement between the U.S and Jordan on Reciprocal Trade.

[3] Art. 3.3, Agreement between the U.S and Jordan on Reciprocal Trade.

[4] Art. 3.2, Agreement between the U.S and Jordan on Reciprocal Trade.

Section 338 provides executive tariff powers to enforce a policy of equal treatment for the commerce of the United States in foreign markets. The President is authorised to employ duties when he or she determines, as a matter of fact, that another country has discriminated against the United States’ commerce compared with the commerce of any other country. The law empowers the President to reset unequal competitive conditions. What constitutes an unreasonable limitation, or has the effect of discrimination, is left largely to the President.

Building on Simon’s last post, I want to explain how the design of the President’s discretionary power within this law marked a critical turn in U.S. trade policy – a turn towards unconditional MFN, underscoring reciprocity after the First World War. However, the rationale for and demand for a law like Section 338, which was formulated before a multilateral trading system was in place, does not fit comfortably in a world of multilateral MFN governed by the GATT/WTO. In the oddest turn of events, the United States is using Section 338 to demand unconditional MFN from the world while simultaneously urging WTO Members to reform the WTO to allow for selective discrimination or conditional MFN.

Section 338: Presidential Powers and Determinations of Discrimination

Section 338 is quite lengthy, but the relevant part concerning additional duties is the following text at paragraph (a):

The President when he finds that the public interest will be served shall by proclamation specify and declare new or additional duties as hereinafter provided upon articles wholly or in part the growth or product of, or imported in a vessel of, any foreign country whenever he shall find as a fact that such country—

(1) Imposes, directly or indirectly, upon the disposition in or transportation in transit through or reexportation from such country of any article wholly or in part the growth or product of the United States any unreasonable charge, exaction, regulation, or limitation which is not equally enforced upon the like articles of every foreign country; or

(2) Discriminates in fact against the commerce of the United States, directly or indirectly, by law or administrative regulation or practice, by or in respect to any customs, tonnage, or port duty, fee, charge, exaction, classification, regulation, condition, restriction, or prohibition, in such manner as to place the commerce of the United States at a disadvantage compared with the commerce of any foreign country.

The President must use section 338 power clearly and precisely; at a minimum, this should require clarity in the object of, and scrutiny for, comparison. The President must make a fact-based determination that there have been unreasonable limitations or discriminations placed upon US commerce that disadvantage it as compared to the commerce of another country. The language of section 338 specifically directs the President to make a determination; it is when ‘he shall find as a fact’ that a trading partner imposes ‘any unreasonable charge, exaction, regulation, or limitation which is not equally enforced upon the like articles of every foreign country’ or ‘discriminates in fact against the commerce of the United States’ such that US commerce is placed ‘at a disadvantage compared with the commerce of any foreign country.’ As such, it is wholly insufficient for the President to bypass this fact-based inquiry and simply call out ‘other countries’ without engaging with fact-finding.

At the same time, the law does not impose evidentiary standards upon the President. Instead, the Tariff Commission, a technical, bipartisan, expert group (established in 1916), would supply Congress and the President with investigations and reports concerning trade relations. Sections 332 to 338 lay out this process concerning today’s International Trade Commission (since 1975).

It’s worth a separate post to consider whether the Trump administration can cite specific cases of discrimination in three sectors and then impose more sweeping duties to pressure Canadian concessions. As I will explain below, Section 338 was seen as an MFN power intended to equalise conditions. If so, there remains a broader, more complex question about whether the duties, per subparagraph (d) of Section 338, should be limited to strictly ‘offset commercial disadvantages’ based on evidence of discrimination. A brief analogy here might be the GATT nullification or impairment language.

The 1922 Precursor to the 1930 Section 338 Powers

Section 338 has its origins in Section 317 of the Tariff Act of 1922.[1] The powers granted in Section 338, as initially conceived, were subject to fiery Congressional debates. I have been working on a book project that includes an examination of the United States’ position on MFN in 1922, as the switch to unconditional MFN occurred in tandem with these discretionary Presidential powers. 

From 1921 to 1922, Congress spent months discussing the constitutionality, scope, and execution of a new tariff bill. In drafting the tariff law, the Senate and House debated various aspects including: delegating legislative power to the executive branch, how the President would make determinations, what constituted discrimination, what constituted a foreign country, the scope of defensive powers, and how it conflicted with reciprocal treaties.

In one discussion on possible guardrails to untethered power, the following quote from a senator encapsulates the tension:

If the time has arrived when we cannot trust the President of the United States to use his very best judgment in a matter of such extreme importance [as tariff rate setting] to the happiness and contentment of every class of citizens of the country, then God help the future of the Republic.[2]

Congress attempted to impose some limits to the President’s powers, as intense criticisms mounted. One senator cautioned that the 1922 bill would establish a power capable of abuse, one which the framers of the Republic intended to deny.[3] In the Baltimore Sun, an editorial dismissed the ‘indefinite propositions’ that would allow presidential power ‘capable of making or breaking particular industries.’[4] And that the bill would set up ‘a political machine under which no business affected, directly or indirectly, by tariff rates will fare to be out of the favor of the administration.’[5] The Baltimore Sun editorial made clear the proposal was ‘capable of grave abuse’ and was ‘bureaucacy run mad.’[6] Perhaps one of the most damning responses from a senator was the following:

You are putting in the hands of one man the greatest power for political corruption and political advantage that has ever been conferred since the beginning of the Government.[7]

How would the President be a fact finder?

In 1922, senators believed that the President’s powers acknowledged ‘proper restrictions’ and discussed a ‘rule laid down which will stand the test of the courts to change the rates within reasonable limitations so that they may be adjusted to changed conditions.’[8] That said, one senator explained that the President should only exercise such power based on investigation and recommendations from an ‘expert authority.’[9] This meant that Congress should do more than debate an intention to seek agency support, but set a legal requirement to do so. Even when Senator Smoot (of the subsequent Smoot-Hawley tariff) debated the role of the Tariff Commission as part of the President’s powers in 1922, he confirmed, ‘the President of course would go immediately to the Tariff Commission,’ as well as other relevant departments. To which his fellow senator replied, ‘Then there ought not to be any objection to providing that there should be this investigation by this body of experts and a public report.’[10]

Having enumerated the powers of the Tariff Commission (as it was at the time) in the law, and recognising the vital resource for investigations into disadvantages in competition, President Harding signed an Executive Order on October 7, 1922 (#3746) which confirmed:

It is ordered, that all requests, applications, or petitions for action or relief under the provisions of Section 315, 316, and 317 of Title III of the Tariff Act approved September 21 1922 [the predecessor to the Tariff Act of 1930], shall be filed with or referred to the United States Tariff Commission for consideration and for such investigation as shall be in accordance with law and the public interest, under rules and regulations to be prescribed by such Commission.’

The takeaway is that while the President retained flexibility, there was a plan for support from the Tariff Commission. Though the original intention was to take the politics out of tariffs in exceptional circumstances, the Commission would be at least some part of the picture. The Sixth Annual Report of the Tariff Commission further set out the rules of procedure for the Commission regarding section 317 of the 1922 Tariff Act, concerning protection of foreign trade against discrimination. The Commission reiterated the organising principle of equality of treatment, and confirmed that Congress had rejected a concessional method of tariff bargaining. Instead, section 317 afforded flexibility to the President to combat discrimination based on findings of fact as to effects upon U.S. commerce.

What was the relationship between Section 317 and the MFN clause?

In 1922, when Congress debated the scope of Section 317, senators debated the intersection of reciprocity treaties with MFN clauses and the new executive powers. Two issues to flag. First, if other countries entered into preferential arrangements, could the President find this to be discriminatory against U.S. commerce, even if the United States had its own preferential arrangements? Second, if the United States had signed MFN-based treaties, then what was the reason for the additional powers to target discrimination? Inevitably, both issues were blurred in debates, raising intense questions about the appropriate U.S. commercial policy in the postwar world. One senator quipped, ‘I had supposed, Mr President, that after the horrible war through which we have passed we were going to try to cultivate amicable relations with our neighbors and friends and not go around carrying a chip on our shoulders constantly.’[11]

That these issues were identified and yet appear unresolved in debates suggests some short-sightedness on the behalf of the architects. Even those senators largely in support of tariff law reform took issue with the potential of imposing penalty duties on other countries for entering into reciprocal arrangements that expanded trade as between themselves. There was extensive debate that while such conditioned reciprocal preferences may fall outside MFN clauses, they may still be subject to offset duties under section 317.[12] Put another way, while preferences may not constitute discrimination in law, it might, under section 317, constitute discrimination in fact.[13] Wallace McClure's assessment was that the insertion of the words 'in fact' in the law may 'have been intended to deprive the word "discriminates" of any legal connotation that might cling to it' (84). Moreover, the term ‘foreign country’ included ‘any empire, country, dominion, colony, or protectorate.’ In 1922, the Commission took note that the ‘colonies of the different powers, both as export markets, and as sources of raw materials’ had grown in importance to the United States, and that many discriminatory export duties were ‘found almost exclusively in colonies’ (6).

Some senators highlighted the double standards, considering the United States had its own preferences, such as with Cuba.[14] The lead architects of the law attempted to assure the critics that if there was no discrimination, it would not fall under the rule.[15] Yet, for those keen to maintain sweeping powers, the scope of the term ‘discrimination’ did not seem to cause worry. For them, the provision sent a ‘notice to the world that America intends to be treated equally with all other countries.’[16] The term ‘discrimination’ as constituting a disadvantage to U.S. commerce remained sweeping.

Unilateral, unconditional MFN against Multilateral MFN

Whether in 1922 or 1930, Congress enacted the executive tariff powers before the development of the multilateral trading system. As such, in a world without a multilateralising MFN, the United States sought to enforce the principle of equality of treatment as between its commerce and that of all other countries. The only way to do so was to acknowledge some outer limits, concerning potential exclusive privileges and preferential arrangements, such as the case of Cuba. The choice to press for breadth of scope over precision reflected the postwar uncertainty and inevitable protectionist interests of Congress at the time. If there were hypocrisies and potential issues in the law, this seemed a small price to pay. It did not seem to matter if the discrimination was justifiable or not, what mattered was the effects – and here flexible powers were entrusted with the President. ‘Taking the tariffs out of politics,’ it was believed, would contribute to the United States preserving its negotiated agreements.[17] If this sounds familiar, it is because this remains the great value of MFN at the WTO.

Moving forward in time, the relationship between MFN and reciprocity altered with the completion of the GATT. Indeed, the Trump administration’s use of Section 338 appears to stand in contravention of Article XXIV of GATT 1994, and all Members’ commitments to negotiating free-trade areas. Canada may therefore negotiate closer trade relations with the United States and the European Union separately while committing to the conditions maintained by Article XXIV. Yet, under section 338, the United States has asserted that the President possesses a power that pierces through the Canada-European agreement in an effort to equalise conditions on cheeses. In a way, the United States’ use of section 338 mirrors foreign investors’ efforts to use MFN clauses to import greater substantive protections from one investment treaty to another. The United States appears to be using its domestic law to reset preferences to its advantage.

Yet one of the most striking parts of the 1920s congressional debate was that the architects seemed to reject the possibility that the section 317 powers (the section 338 equivalent) would conflict with the MFN clauses in reciprocal treaties. Using section 338 now would. While the United States has already questioned the utility of unconditional MFN for the, at least immediate, future, it likewise exposes itself to another challenge – policing a world of preferences. But partners should beware that, if left unchallenged, the use of section 338 could require all governments to continuously pay for reciprocity. Another immediate consequence would be that all non-US agreements, whether formal or not, could become evidence of discrimination. The most ironic end to this story is that if the United States wants equal advantages, then the solution would be more MFN, not less.

[1] The sections are similar. One important modification is that section 338 applies to articles imported in vessels of foreign countries as discrimination against the commerce of the US. See Tariff Commission Annual Report no. 14 (1930), p. 2.

[2] August 10, 1922, C.R. 62-11-11186.

[3] August 11, 1922, C.R. 62-11-11223.

[4] C.R. 62-11-11197. Citing the Baltimore Sun of July 9 1922, Taxation by Executive Fiat.

[5] C.R. 62-11-11197. Citing the Baltimore Sun of July 9 1922, Taxation by Executive Fiat.

[6] C.R. 62-11-11207.

[7] C.R. 62-11-11207.

[8] C.R. 62-11-11193.

[9] C.R. 62-11-11193.

[10] C.R. 62-11-11193.

[11]C.R. 62-11- 11244.

[12] C.R. 62-11-11246.

[13] C.R. 62-11-11246.

[14] C.R. 62-11-11245.

[15] C.R. 62-11-11245.

[16] C.R. 62-11-11246.

[17] C.R. 62-11-11251.

First, the White House tells us that "[t]he tariffs will take effect 30 days after signing" the proclamations, but I'm skeptical that will actually happen. My best guess is that the administration is using them more as negotiating leverage in the talks underway right now as part of the 6-year review of the USMCA. In my view, what we may see is a series of 30-day (or so) extensions that keep the Section 338 tariffs hanging over the U.S.-Canada trade relationship for a while. (Will that approach have an impact on Canadian negotiators? I have doubts, but nonetheless it seems consistent with how the Trump administration approaches trade negotiations.)

Second, if the tariffs were to be imposed, I think it would be difficult to bring a successful challenge in U.S. court. As trade lawyers John Veroneau and Catherine Gibson explained a while back:

Section 338 permits the president to impose “new or additional duties” on countries that have discriminated against commerce of the United States. Section 338 authority is triggered when the president finds that a foreign country has either (1) imposed an “unreasonable charge, exaction, regulation, or limitation” on U.S. products which is “not equally enforced upon the like articles of every foreign country”; or (2) “[d]iscriminate[d] in fact” against U.S. commerce “in respect to customs, tonnage, or port duty, fee, charge, exaction, classification, regulation, condition, restriction or prohibition” so as to “disadvantage” U.S. commerce as compared to the commerce of any foreign country.

The three Presidential tariff proclamations (motor vehicles, alcoholic beverages, dairy) lay out the case for how exactly Canada has been engaging in "discrimination or an unreasonable and unequal imposition" on U.S. commerce, and the Trump administration is right: Canada has been doing so. Of course, Canada has defenses to this discrimination. For motor vehicles and alcoholic beverages, the defense is that it is retaliating against U.S. tariffs; and on dairy, the defense is that the discrimination in favor of EU producers, which have better terms under CETA than U.S. producers have under the USMCA, is allowed under GATT Article XXIV. But looking at the Section 338 statute, I don't see that a justification for discrimination is permitted. If discrimination exists, the President can impose tariffs under Section 338. (Section 338 is part of the Tariff Act of 1930, and I suppose it's possible there may be some drafting history out there on what exactly people had in mind by "discrimination." I'll leave that for the legal historians to illuminate.)

One other aspect of a possible lawsuit that people may point to is the requirement of an ITC investigation. On this issue, a CRS report notes that:

... Section 338 charges the ITC with “ascertain[ing]” and informing the President of relevant instances of discrimination:

"It shall be the duty of the [ITC] to ascertain and at all times to be informed whether any of the discriminations against the commerce of the United States enumerated in . . . this section are practiced by any country; and if and when such discriminatory acts are disclosed, it shall be the duty of the commission to bring the matter to the attention of the President, together with recommendations.271"

This provision, together with Section 338’s placement in Part II of the Tariff Act of 1930 (concerning the ITC), may raise a question as to whether the ITC must find that discrimination has occurred before the President may impose tariffs.272 By authorizing the President to impose tariffs “whenever he shall find as a fact” that discrimination has occurred,273 however, Section 388 [sic] does not appear to condition the President’s authority on such a finding by the ITC.

I agree with what CRS seems to have in mind: The opening language of the statute ("The President when he finds that the public interest will be served shall by proclamation specify and declare new or additional duties as hereinafter provided upon articles wholly or in part the growth or product of, or imported in a vessel of, any foreign country whenever he shall find as a fact that such country ... ") arguably makes Presidential fact-finding sufficient here.

As a final point, Veroneau and Gibson state: "We cannot explain why Section 338 disappeared from public view after 1949 even though it has remained in the statute books." One answer may be that, practically speaking, Section 338 was superseded by the inclusion of the non-discrimination principle in trade agreements, starting with the GATT in 1948. While there may have been gaps in trade agreement coverage for other issues, non-discrimination was well covered, so there was no need to use Section 338 for this. And then when people had concerns about the effectiveness of the GATT, they came up with Section 301 of the Trade Act of 1974 to deal with these issues. So, maybe there were just better known options, and people focused on what they knew.

The paper begins by identifying the "current challenges" with the "commercial activities of State enterprises":

2.3. When not properly regulated, State-owned and State-invested enterprises active in commercial markets may distort competition. The weight of State enterprises in the global economy is increasing: in 2023, among the 500 largest firms in the world, over 200 were State-controlled, up from 50 in the year 2000.

2.4. They can unbalance the playing field thanks to State interventions such as regulatory advantages or access to privileged information. Their dominant market positions, often reinforced by legal monopolies or historic incumbency, can also stifle innovation and deter private investment, while cross-subsidisation between commercial and non-commercial activities skews further competition.

2.5. Such distortions not only disadvantage other (privately-managed) competitors but can also lead to inefficiencies, reduced consumer choice, and misallocation of resources in the broader economy. The same is true for private entities entrusted or directed by the State, especially those acting as providers of preferential funding or of raw materials and inputs used in the production process. Raw materials and other inputs provided on non-market terms cause market distortions in the final product, making it cheaper than it would be absent these distortions. They also cause distortions in the markets of the raw materials, whose access to foreign competitors becomes constrained or whose international prices increase as a result.

It then considers the "gaps and shortcomings in WTO rules" on these issues:

2.6. This concern is only partially addressed by the WTO. State Trading Enterprises¹ are governed by GATT Article XVII, which subjects this specific set of State enterprises to disciplines on commercial considerations, non-discrimination and transparency. There is no specific WTO agreement addressing the potentially distortive conduct of other State enterprises in the marketplace, beyond the rules on provision of subsidies in the Agreement on Subsidies and Countervailing Measures (ASCM) which apply to public bodies and to private bodies entrusted or directed by the State.

2.7. Several WTO Members have taken specific commitments on State intervention and State enterprises in their WTO Accession Protocols.² These commitments vary in substance. Some include commitments to privatise certain State enterprises, or to require State enterprises to make purchases and sales solely in line with commercial considerations and refrain from discriminating, for instance, in access to infrastructure, and prohibit non-commercial mandates such as employment guarantees. Others are limited to general commitments to limiting State enterprises' privileges and promoting private sector competition.

2.8. However, these commitments are currently not actively monitored or enforced. This should be corrected.

2.9. Next to WTO accession protocols, many more WTO Members, including the EU, are bound by commitments on State enterprises in around 100 Free Trade Agreements, with a variety of definitions and disciplines. The EU's FTAs typically include obligations to commercial considerations and non-discrimination, establish core principles of competitive neutrality as concerns ownership, and are complemented by transparency obligations. Other regional trade agreements such as CPTTP and USMCA also include disciplines to limit the trade distortive impacts of State enterprises.

2.10. Taking together the WTO accession commitments and the Free Trade Agreements, it can be seen that a large number of WTO Members already share a set of international disciplines regulating the behaviour of State enterprises. While the scope of these disciplines varies, they share the goal of preventing State enterprises from distorting competition through unfair State support, non-commercial behaviour, discrimination, regulatory advantages and lack of transparency.

2.11. In other words, for many WTO Members commercial considerations and non-discrimination are already a widely shared international norm for the behaviour of State enterprises in the marketplace.
------------------------------------------------------------
_{1 Working definition agreed by Members: "Governmental and non-governmental enterprises, including marketing boards, which have been granted exclusive or special rights or privileges, including statutory or constitutional powers, in the exercise of which they influence through their purchases or sales the level or direction of imports or exports."
2 Afghanistan, Cambodia, Cape Verde, China, Comoros, Kazakhstan, Laos, Liberia, Montenegro, Mongolia, Nepal, Russia, Tajikistan, Ukraine, Vietnam.}

Thus, according to the EU, it is a problem that there are no general WTO obligations in this area, with the only current rules being those in a few Members' accession commitments and in FTAs.

With regard to a "possible way forward," the EU offers the following suggestions:

2.12. This set of existing commitments should be reviewed and, as appropriate, consolidated at the WTO, so as to ensure a more systematic monitoring (drawing inspiration from the monitoring of STEs carried out by the Working Party on State Trading Enterprises), reporting, and more uniform application. This would be separate from the ASCM, as the issue at stake does not limit itself to subsidies.

2.13. A first step towards consolidation of these rules would be a mapping exercise that could be coordinated by the facilitator for level playing field issues. The resulting overview could then serve as a basis for developing a model schedule of commitments for acceding Members, as well as further targeted and consolidated rules for existing Members. The new set of WTO rules would address the identified gaps as regards the conduct of State enterprises in the marketplace by establishing obligations on commercial considerations, non-discrimination and transparency, in line with the shared international norms already found in Free Trade Agreements and WTO accession protocols.

In my view, the EU proposal here is a good one. I've always thought it was odd that there are general GATT obligations that apply to state-trading enterprises, but nothing similar for state-owned or state-invested enterprises. This gap in the rules should be closed.

As the EU suggests, it would be useful in this context to look at what some governments have already agreed to on state-owned and state-invested enterprises as part of their accession commitments. In footnote 2 above, the EU points to the following governments that have made these commitments: Afghanistan, Cambodia, Cape Verde, China, Comoros, Kazakhstan, Laos, Liberia, Montenegro, Mongolia, Nepal, Russia, Tajikistan, Ukraine, Vietnam. Let's look at China for an example of the commitments.

Paras. 43 through 49 of China's Accession Working Party Report cover "State-Owned and State-Invested Enterprises." Para. 46 says the following:

46. The representative of China further confirmed that China would ensure that all state-owned and state-invested enterprises would make purchases and sales based solely on commercial considerations, e.g., price, quality, marketability and availability, and that the enterprises of other WTO Members would have an adequate opportunity to compete for sales to and purchases from these enterprises on non-discriminatory terms and conditions. In addition, the Government of China would not influence, directly or indirectly, commercial decisions on the part of state-owned or state-invested enterprises, including on the quantity, value or country of origin of any goods purchased or sold, except in a manner consistent with the WTO Agreement. The Working Party took note of these commitments.

The para. 46 commitments (along with those in paras. 47 and 49) are made enforceable through para. 342 of the Working Party Report and para. 1.2 of the Accession Protocol.

On their face, these Chinese commitments seem like a sensible approach to the issue. No doubt the wording could be updated and tweaked a bit, but it's a good start.

However, as the EU paper notes, the accession commitments in this area "are currently not actively monitored or enforced." I'm not sure why the monitoring and enforcement hasn't been done to date, but I agree with the EU that "[t]his should be corrected." It would be useful to have a better sense of how these commitments work in practice, and attempts at enforcement would help with that.

With regard to extending these commitments to other governments by creating a general WTO obligation along these lines, an obvious question is: What are the chances of success for negotiating such an extension? When market-distorting practices are discussed in the WTO context, there is often a concern that China won't agree to additional disciplines on its behavior, which will be an impediment to completing any negotiations. But here, China is already subject to the discipline of its own commitments, so Chinese objections shouldn't be an issue. In fact, China should want to have these "WTO-Plus" obligations extended to others so as to level the playing field. That doesn't mean reaching agreement will be easy, of course, as getting agreement from all the other Members that use some version of state enterprises will be a challenge. But it does remove one big stumbling block.