If I'm counting the 2026 announcements correctly, there have been 21 safeguard investigations initiated by WTO Members so far this year. This puts us on pace to see the most investigations initiated by WTO Members in a year, surpassing the 34 in 2002. The figures for 2022, 2023, 2024, and 2025 were 4, 12, 16, and 14, respectively. (The WTO spreadsheet doesn't have data on the GATT-era).

What to make of this development? While I'm not a fan of protectionism in general, safeguards are probably my favorite kind because they constitute – as people say, although I can't find a quote – "honest protectionism." There are no suspect methodologies or accusations of "unfair trade." Rather, safeguard measures are simply an acknowledgement that the domestic industry is having trouble competing with imports and needs some temporary protection to help it out.

It will be interesting to see if any of the safeguard measures that result from these investigations are challenged in WTO dispute settlement. Some of the past WTO disputes on safeguard measures were a source of controversy due to interpretations by the Appellate Body. With dispute settlement now having been disrupted by the blockage of Appellate Body appointments and the emergence of the MPIA, there may be an opportunity to revisit the jurisprudence. We have already seen that the MPIA was willing to revisit the AD Agreement standard of review. It's possible that a fresh look at various elements of the GATT Article XIX/Safeguards Agreement analysis could lead to new thinking as well.

Of the WTO Members notifying safeguard investigations so far this year, the following are parties to the MPIA: Australia (fabricated structural steel), Canada (certain wood goods / certain vegetable goods), the EU (grain-oriented electrical steel), New Zealand (certain aluminium extrusions), and the Philippines (ceramic tiles / rice). Depending on the outcomes of the investigations by these WTO Members, other MPIA-party WTO Members faced with the resulting safeguard measures might bring formal WTO complaints, which could give the MPIA arbitrators a chance to weigh in here.

Products covered by the other notified investigations are:

• office paper (South Africa)
• motor car pneumatic tyres (Russia)
• soda ash (India)
• certain juices, nectars and non-alcoholic beverages (Madagascar)
• certain tableware, kitchenware, and household and packaging articles (Madagascar)
• certain types of rice (Morocco)
• fibreboards (Zimbabwe )
• doors (Zimbabwe)
• tinplate (Russia)
• dry pastries and breakfast cereals (Madagascar)
• plastic pipes and tubes and accessories (Madagascar)
• polyethylene terephthalate resin (Türkiye)
• terephthalic acid (Türkiye)
• other paper and paperboard (Türkiye)

If there's a pattern here in the types of products being targeted, I can't make it out, and I'm not sure what explains this surge in safeguard investigations. Exports by China are surely a part of it, but there could be other factors as well.

📢 Call for Papers! The Block Center for the Study of International Business Law at Brooklyn Law School invites scholars to submit proposals for the 2026 International Business Law Scholars Roundtable, taking place on Friday, October 9, 2026.

Researchers working in international business, economic, and financial law are encouraged to apply. Selected scholars from outside the New York City area may receive up to $500 in travel reimbursement.

Read more details and submit your 500-word proposal by July 15th, 2026 at this link.

So how are these export opportunities working out so far? The earlier post focused on exports of specific categories of products, and in this post I'm going to follow up by looking at some Bureau of Economic Analysis (BEA) data on exports broken down by country. In particular, I'll look at overall U.S. goods exports to the top 10 country destinations (which constitutes around 60% of U.S. goods exports).

If you just want to go straight to the export figures, there's a table at the bottom of the post that compares data on January through April goods exports over the period from 2024 - 2026, for each of the top 10 export destinations (note that the data is for value of exports rather than volume, and is in nominal terms, so you would expect the figures to rise a bit during the period to reflect inflation). Some notable points in the data are as follows:

• Exports to Canada are slightly down, which probably in part reflects Canadian consumer attitudes in response to the rhetoric of President Trump and other U.S. government officials
• Exports to China are significantly down, which is likely due in part to Chinese retaliatory tariffs
• Exports to Switzerland are significantly up, which may reflect the non-monetary gold exports that I mentioned in the earlier post
• Exports to Hong Kong are up by about the same amount as exports to China are down, which may reflect transshipment of goods that are ultimately destined for China but which are avoiding the Chinese tariffs (although it could be some non-monetary gold too).
• Exports to the UK are up, and this could be non-monetary gold as well.

To really get into the details of these export patterns, we need to combine the product data with the country data. I'll try to come back to that soon.

Here's the table of export figures by country:

Source: Data derived from Bureau of Economic Analysis trade data

Recent IMF research found that persistent trade imbalances harm deficit economies and benefit surplus ones by reallocating productivity gains. ...

The IMF acknowledged recently that imbalances are “concentrated and persistent” and driven at least in part by surplus country policies. In its most recent Article IV report, the IMF raised the alarm on the US current account deficit (driven primarily by the trade deficit), noting that the resulting negative net international investment position “raises the risk of an eventual disorderly external rebalancing.”

But, to address this problem, the IMF recommends untenable and outrageous solutions: large-scale tax increases (including a 10 percent federal sales tax) and austerity measures (including deep cuts to popular entitlement programs). ...

Jamieson says the IMF "recommends untenable and outrageous solutions" on taxes and spending. Let's take a look at what they said exactly. On p. 28 of the Article IV report, "Box 5. Potential Tax and Spending Policies to Lower the Federal Debt" provides the following:

A combination of options will be needed on both the revenue and expenditure side of the budget to bring the debt-GDP ratio down over the medium term. These include:

• Scaling back poorly targeted tax expenditures. These include not taxing the value of employer-provided health care, capital gains exemptions for individuals selling their principal residence, and income tax deductibility for mortgage interest and state and local taxes. Removing these tax expenditures would increase progressivity and generate around 1.4 percent of GDP per year in revenues.

• Closing the “carried interest” provision whereby income earned from partners in an investment fund can be treated as capital gains and taxed at a 23.8 percent rate (rather than at the 37 percent top marginal rate).

• Eliminating “step up basis” for capital gains which allows the value of inherited assets to be reset at the date of death so that any capital gains that has accrued during the life of the original owner are effectively never subject to capital gains tax.

• Phasing in a federal consumption tax. Such an increase in indirect taxes would generally be regressive and so would need to be combined with a well-designed social assistance program to cushion the impact on poor households. As an example, a 10 percent, broad-based VAT would yield around 2 percent of GDP in revenues per year.

• Raising the federal excise tax on gasoline and diesel. The federal tax on gasoline and diesel is not subject to indexation and has remained unchanged in nominal terms since 1993 (at 18.4 cents for gasoline and 24.4 cents for diesel). Doubling the tax on both gasoline and diesel would yield around 0.15 percent of GDP per year. A carbon tax could also be considered.

• Raising the corporate tax rate and fully moving to a cashflow tax. If combined, such a change in the corporate tax system could both raise revenue and reduce the marginal disincentive to invest. Each 5 percentage point increase in the corporate income tax rate would yield around 0.3 percent of GDP per year.

• Reducing imbalances in the social security system. Indexing social security benefits to chained CPI would save around 0.1 percent of GDP per year. Subjecting earnings greater than US$250,000 to social security payroll taxes would increase progressivity and yield around 0.4 percent of GDP per year.

• Lowering public healthcare outlays. A range of policies to mitigate the impact of population aging on the federal healthcare deficit have the potential to significantly reduce Medicare outlays and increase the program’s efficiency (see Box 6).

Reasonable people can disagree on these specific suggestions. For instance, I would say that removing the tax exclusion for employer-provided health care is a great idea, and a change that is crucial in order for the U.S. health care system to become something other than the horrible mess it is now; on the other hand, I'm more skeptical of a "10 percent, broad-based VAT." Others might have a different view on each of these.

Regardless of what you might think of each item, though, it's good that the IMF is putting specific proposals out there. It's clear that a change in direction is necessary for U.S. fiscal policy, and it's only going to happen if someone finds a way to get the issue into the political conversation. In domestic politics right now, these problems are mostly being ignored. For almost a decade now, U.S. fiscal policy has involved very high deficit spending, acting as a stimulus to keep the economy growing. You can do that for a while, but at some point you have to adopt a more fiscally responsible set of tax and spending policies.

In the meantime, a side effect of the fiscal irresponsibility is going to be trade deficits that are higher than they otherwise would be. If balanced trade is your goal, then U.S. fiscal policies need to be put in the mix of solutions. Whether the IMF's specific recommendations are appropriate ones will, of course, be the subject of disagreement among people who have always disagreed about these things. Regardless, it's clear that the U.S. needs to reduce its budget deficit, and that one of the effects of doing so will be to lower the trade deficit (although there are many other factors in play here, so it's hard to say with any precision how much of an impact this would have).

Senior U.S. officials have held preliminary discussions with major artificial intelligence companies about the potential for the federal government to acquire some shares in their firms, according to three people familiar with the matter.

Sam Altman, the CEO of OpenAI, has discussed the idea with senior Trump administration officials periodically since the president began his second term, said two of the sources, all of whom spoke on the condition of anonymity to reflect private deliberations. Altman first pitched the concept directly to President Donald Trump in a conversation in early 2025, and has discussed it again with senior administration officials in recent weeks as a way to more broadly distribute the economic benefits of AI to the public, they said.

While planning is ongoing and details are in flux, discussions have centered on having the firms voluntarily cede the shares to the government, the people said. The returns on the investment could then be directed to public purposes, one of the people said, such as distributing a dividend payment to all American households.

In comments he made on Friday, Trump seemed open to this idea.

If this actually happens, it will be interesting to see what the final arrangement looks like, but regardless I think this policy is likely to generate the following international response: Foreign governments could decide to pour resources into developing their own domestic AI sector. There are a couple reasons for this.

First, if a person's use of services from OpenAI and other American AI companies means that money gets sent to Americans but not to non-Americans, non-Americans are going to have a disincentive to use the service. Such a financial outcome would feel totally unfair to non-American users, and they are likely to seek out alternative services.

Second, there will be added privacy concerns with the U.S. government being involved in these companies. There is already a good deal of distrust about the U.S. government getting access to data held by U.S. tech companies. U.S. government ownership of shares in these companies makes this an even bigger concern.

It seems to me, then, that in response to demands from their citizens (and nascent industries), many foreign governments are almost certain to move in the direction of developing their own AI industries to compete with American government-owned AI companies, kicking off an industrial policy competition in this sector.

How difficult will it be for a foreign government to do this successfully? That's hard to say. Building an industry that is as good as the American one could be tough. The American industry has both a head start and greater resources. On the other hand, it's much easier to follow than to lead, so I think a fair number of other countries could develop a sector that offers a pretty good service, even if not quite at the American level at first.

And if they follow the U.S. model that is being suggested, it will be easy to market the product domestically by telling people there is money in it for them if they use the domestic service, and also that by using the domestic service they can help it improve and catch up to the Americans.

At this point, let me just add a general comment that will not come as a surprise to this blog's readers: In my view, state capitalism of this sort would be a terrible policy for the U.S. government (and every other government) to adopt, and everyone should turn away from this approach immediately. It will lead to worse services, worse business practices, and worse government policies in this sector, and should be avoided. Whatever your views on AI in general (I'm more on the skeptic side, as I think of AI as probably useful for some things but a bit overhyped*), I don't think government ownership of this industry is the right road to go down. What we want instead is a competitive industry that gives companies an incentive to behave better, and is subject to regulation that will prevent the range of abuses and harms that have been widely reported and should be taken more seriously.

Finally, let me note that if the real reason for U.S. government ownership here is that the AI sector has no path to profitability and the government is effectively being asked to "socialize the losses," this whole thing makes even less sense.

* _{The caveat to this is that I don't use AI at all (except for when a Google search forces it on me), so it's possible I don't know what I'm talking about! But I've read enough about how it works, and also I ran this by some people who are actual users/experts, so I feel confident enough to post this. Feel free to tell me in the comments how I'm getting it all wrong though!}

First, the switch from the old platform last September led to the loss of threading in comments that had been posted up to that point, which made it difficult to follow those conversations. I was recently able to salvage the original version of a good portion of those older comments and restore the threading, so things should look better now (although I think I've reached the limits of what can be accomplished and some comments from the old platform will remain jumbled).

Let me also note with respect to the comments that if there are multiple comments on a post, the current platform defaults to sorting them by "best." You can adjust this to "newest" or "oldest," but it starts out with "best." If I understand it correctly, "best" is judged by likes and dislikes on the comments. I don't find any of this very useful, but at least as of now there doesn't seem to be a way to disable it.

Second, just a reminder that if you are signed up for email notifications of new posts, they sometimes end up in the spam folder. So if you happen to be reading this post and are wondering why you aren't getting the email notifications you signed up for, please check your spam folder and tag the IELP blog emails as "not spam" or "not junk" (or however your email system describes these things).

 At the March 2026 Ministerial Conference of the World Trade Organization (WTO), India blocked the adoption of the Investment Facilitation for Development Agreement (IFDA), a plurilateral initiative backed by 131 members, including 94 developing countries. India stood alone after South Africa and Türkiye withdrew their objections.

The IFDA is, by design, modest in scope. It does not address market access (including for firms from specific countries), investment protection, or investor–state dispute settlement; nor does it restrict the ability of governments to formulate their own investment policies. Instead, it focuses on improving transparency, streamlining administrative procedures, and fostering cooperation between governments and investors.

These are not abstract ideals; they are practical reforms that many countries—including India—have already pursued domestically. India’s own policy trajectory mirrors the Agreement’s core objectives.

If India has already validated these reforms at home, why resist their codification at the global level?

The answer lies less in substance and more in principle. India’s opposition reflects a deeper concern about the rise of plurilateral agreements within the WTO framework.

The WTO has been anchored in consensus-based decision-making. Plurilateral agreements, negotiated among subsets of members, risk diluting this principle by creating parallel rule-making tracks that exclude dissenting countries. From a systemic standpoint, plurilateralism could fragment the WTO.

Yet there is a difference between resisting a trend and isolating oneself from it.

India’s solitary opposition to the IFDA carries reputational and strategic costs. When a major economy stands apart from a proposal supported by a broad coalition of developing and developed economies, it can appear hesitant about transparency or broader rule-making efforts.  

This perception sits uneasily with India’s parallel ambition to position itself as a leading destination for global investment. Over the past decade, India has invested significant political and economic capital in improving its business and investment climate. But global investors assess not only domestic policies; they also read signals from international engagement. A veto at the WTO risks undermining the credibility of reforms undertaken at home.

More importantly, when a critical mass of WTO members moves ahead with plurilateral agreements, the resulting standards often become de facto global benchmarks. In this case, the 131 participants in the IFDA accounted for over 75% of world imports (2025) and over 70% of world inward FDI flows (2024).

This dynamic is already visible in areas such as digital trade, where subsets of countries have advanced negotiations despite the absence of full consensus. Over time, these rules shape global value chains, regulatory expectations, and investment flows. Countries that remain outside such frameworks do not escape their influence.

How, then, to move forward?

A first option is for India to lift its veto on the IFDA’s integration into the WTO rulebook as a plurilateral agreement while safeguarding its systemic concerns.  

During the Ministerial, Minister Piyush Goyal indicated how this could be done, namely by establishing clear guardrails and legal safeguards. These could be enshrined in a Declaration accompanying the adoption of the IFDA, approved by consensus among all WTO members.

Such a Declaration could:

·      Reaffirm the core safeguard embedded in the WTO’s founding Marrakesh Agreement: that decisions must be taken by consensus.

·      Clarify that the IFDA is a standalone agreement, and its adoption does not establish a precedent for other plurilateral initiatives.

·      Confirm that any investment-related issue going beyond investment facilitation as contained in the IFDA—such as market access, investment protection, investor-state dispute—is separate and distinct from the IFDA.

·      Underline that nothing in the IFDA affects the rights and obligations of non-participants.

·      Reiterate that the IFDA is open to all WTO members to join in the future.

·      Require that the implementation of IFDA measures needs to be done in a non-discriminatory manner regarding foreign investors from all WTO members.

The second option is straightforward: India joins the IFDA, thus allowing the Agreement to be adopted by consensus. This would align India’s international stance with its domestic reform agenda and send a clear signal of commitment to transparency and predictability. Given that the Agreement does not constrain policy space on sensitive issues such as market access or investment protection, the costs of participation are limited, while the reputational gains could be significant. A Declaration as outlined above could facilitate this option, although joining the Agreement would in any event give India a direct role in its further development.

Whatever option is chosen, India can uphold multilateralism without resorting to obstruction, by engaging constructively in shaping emerging trade rules.

With the WTO at a turning point, India faces a choice not between multilateralism and plurilateralism, but between influencing the future of global trade or being shaped by decisions made without it.

· Karl P Sauvant is Senior Fellow, CCSI, Columbia University, and former Director of UNCTAD’s Investment Division; Rajesh Aggarwal is Visiting Professor, ICRIER, New Delhi.

Briefly, the report finds that all 60 economies under investigation, including the European Union, Canada, and Mexico, either failed to effectively enforce a forced labor prohibition or failed to impose any legal prohibition on the importation of goods produced wholly or in part with forced labor. Citing Section 301(b) of the Trade Act, the report notes that the Trade Representative must now “take all appropriate and feasible action …to obtain the elimination of that act, policy, or practice.” USTR proposes 10% as the rate of additional duties for one group of countries, and 12.5% as the rate for another group. To make this argument, USTR’s report simultaneously needs Customs and Border Protection (CBP) to be effective (to sustain the commerce-burden claim) and ineffective (to explain why forced labor still pervades the U.S. supply chain).

Researchers and practitioners have already devoted considerable time trying to understand the report’s implications for trade and labor governance, including why an initiative as critical to market integration as a forced labor import ban is being weaponized to justify yet another round of tariffs. I look forward to hearing from others, perhaps during the written comments period, which closes on July 6, 2026, or hearings on July 7, 2026. In this post, I offer a critique of some of the core elements of USTR’s reasoning related to forced labor import bans and their effects on destination and source economies.

 The Findings

Section 301(b) of the Trade Act authorizes the Trade Representative to take action to address acts, policies, and practices of a foreign country that are “unreasonable or discriminatory and burden[] or restrict[] [U.S.] commerce….”

Before applying that standard, the report divides the 60 economies into two distinct buckets. It finds that those in the first bucket (Canada, Ecuador, the entire European Union, Indonesia, Mexico and Pakistan) have “failed to effectively enforce a forced labor import prohibition” while those in the second bucket (54 other economies, including Australia, Singapore, and Switzerland) have “failed to impose a legal prohibition on the importation of goods produced wholly or in part with forced labor and to effectively enforce such a prohibition.”

The report then finds that the failure of all 60 economies to impose and effectively enforce a forced labor import prohibition is actionable. Referring back to the statute, the report argues that such a failure is “unreasonable” because it:

 (1) undermines the universal aim of eliminating forced labor; (2) permits firms that avail themselves of forced labor to produce goods at a lower cost and thereby distort market conditions for firms that do not use forced labor; (3) undermines the profitability of firms that do not use forced labor; and (4) contributes to the circumvention of existing forced labor import prohibitions.

That failure also:

[B]urdens or restricts U.S. commerce by subjecting U.S. producers to unfair competition from forced labor goods in both export markets and the U.S. market, and by displacing foreign goods produced without forced labor or forced labor inputs from their domestic market to the United States and other markets.

The report’s thesis, as I understand it, is as follows. U.S. forced labor import bans require U.S. producers to exercise, under relevant statutes, “reasonable care” in importing their goods into the United States, including through costly due diligence. Importers in the 60 named economies do not have to exercise reasonable care, or pay associated costs. When responsible U.S. producers export their goods, they have to compete with foreign producers in other countries, despite having higher production costs, giving foreign competitors an unfair competitive advantage in export markets. Moreover, when foreign producers import goods produced in whole or part by forced labor, they sell goods at artificially depressed prices in their domestic market, which in turn can pressure producers (presumably, the good actors) in those countries to look for markets elsewhere, “thereby perpetuating the trade in forced labor goods.” The U.S. economy, with its compliant production practices, is presumably one such external market.

 The Critique

USTR’s findings and reasoning require a lot from us without offering us much in return. They require us to agree that: (1) U.S. enforcement authorities, namely CBP, effectively enforce forced labor import bans in the United States; (2) CBP’s effective enforcement compels compliance with forced labor import prohibitions in the United States; (3) U.S. forced labor import bans make it more costly for U.S. companies to produce goods with supply chain inputs than foreign competitors; and (4) U.S. exporters do not similarly export inputs produced in whole or part by forced labor. I take each argument in turn.

 1. The United States Does Not Effectively Enforce Its Forced Labor Import Bans

USTR’s Section 301 findings rise or fall on its argument that CBP effectively enforces forced labor import bans under Section 307 of the Trade Act and the Uyghur Forced Labor Prevention Act (UFLPA). The report argues: “Economies like the United States that effectively enforce laws that prohibit importation of forced labor goods are a natural destination for these producers, as they would be less subject to price distortions caused by forced labor goods.” If CBP’s enforcement is deficient, however, none of the interconnected consequences described above come to fruition. USTR argues in support of its thesis that CBP has “at present 55 WROs and eight Findings in place with respect to various goods whose entry into the United States is prohibited under the U.S. forced labor prohibition.” There are (at least) two problems with this argument.

First, the premise of the report is that forced labor is prevalent in the supply chain, requiring an all-hands approach to banning imports produced with it. That is correct, and the strongest portions of the report draw on empirical data showing how forced labor continues to permeate production in trade and distort markets. In particular, the report aptly relays polysilicon and cotton case studies to show how forced labor inputs transit through third economies to reach U.S. markets despite existing prohibitions.

In fact, USTR makes such a strong case for the prevalence of forced labor that it sets up CBP’s 55 WROs and eight Findings to appear rather inadequate. If the U.S. import market is truly facing a barrage of inputs and goods produced by forced labor (as a consequence, presumably, of weak import bans elsewhere), why has CBP only issued 55 WROs?

While USTR’s report paints these empirics in the best light possible, a quick review of congressional testimony sheds better light on CBP’s enforcement scorecard. It is safe to write that Congress, the entity that designed and enacted the forced labor import bans, hasn’t given the agency high marks. During a congressional hearing in 2024, for instance, Representative Dan Bishop observed that “CBP’s detention rate is just a sliver of the billions of dollars of textile products the U.S. imports annually, emphasizing the continuing challenge in effectively enforcing the law.” That inadequate detention rate, he elsewhere observed, compounds the “lack of visibility into CBP’s decisions [and thus] represents ‘the epitome of failure of government.’”

Representative Bishop is not alone. In a public letter to CBP in 2023,  seven U.S. Senators, including Senators Bill Cassidy and Ron Wyden, wrote: “Recent reports of textile and apparel mill closures in the United States raise serious concerns as the lack of effective customs enforcement has been cited repeatedly as a key factor contributing to declining demand,” and argued: “Insufficient enforcement can create a pathway for banned Xinjiang cotton to infiltrate regional supply chains and undermine efforts to enforce the Uyghur Forced Labor Prevention Act.” CBP’s track record, according to Congress, fails to align with USTR’s characterization of effective enforcement.

Second, USTR’s figures are questionable for two reasons: (1) the link offered in the footnote, like many in the report, is wrong and (2) even a correct link reveals that at least four of the active WROs have been superseded by the UFLPA’s rebuttable presumption and thus no longer function as independent enforcement tools.

USTR has dug itself into an analytical hole here. It needs to convince us that the lack of effective forced labor import bans elsewhere has led to pervasive forced labor in the supply chain, including the U.S. supply chain, harming U.S. producers. But it also needs us to agree that CBP is effectively enforcing the ban to keep those imports out of the United States. Instead of addressing this tension head on, the report under-analyzes CBP’s track record and over-credits CBP’s enforcement statistics.

 2. CBP’s Enforcement Does Not Compel Compliance

Part of USTR’s Section 301 investigation centered on whether the 60 economies effectively enforce their forced labor import bans, raising an obvious question about what “effectively enforce” means. According to the report: “USTR considers that an economy fails to effectively enforce a prohibition if the economy is deficient in compelling observance of its forced labor import prohibition, if any.” To show how CBP compels observance, the report merely recites the number of WROs, Findings, and examinations the agency has conducted since 2016, indicating that CBP modified 16 WROs and Findings after importers “demonstrated they had remediated the indicators of forced labor” that led to the enforcement action.

Again, USTR’s statistics are not doing the empirical work it needs here. According to the Global Slavery Index, a total of $196.6 billion worth of imports at risk of being produced by forced labor make it into the United States. Neither foreign exporters nor domestic importers have been compelled into observance. Against that backdrop, the USTR’s paltry 16 actions stand starkly.

Furthermore, USTR offers no methodology for how it counts remediation actions. It does not break this figure down by sector or importer, nor does it provide a basis for treating 16 modifications across a decade as evidence of systemic compliance.

Perhaps acknowledging that arguments about compulsion require empirics on compliance behavior that exceed USTR’s expertise, the report helpfully points the reader to eight elements that can “also assist in examining whether an economy has the tools necessary to effectively enforce its forced labor import prohibition.” Space does not permit me to do a full assessment of CBP’s procedures, so I will stick to the first element: “A statutory definition of forced labor grounded in international law.”

I want to review this element because of its importance to broader trade and labor governance efforts. USTR’s invocation of international law to benchmark effective forced labor ban legislation is highly problematic in two respects.

First, while the U.S. forced labor import ban legislation reproduces the International Labor Organization’s definition of forced labor, it departs from that definition in the details. The ILO’s definition is codified under ILO Convention No. 29. The Convention states:

For the purposes of this Convention the term forced or compulsory labour shall mean 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.

Meanwhile, Section 307 similarly states:

'Forced labor', as herein used, shall mean all work or service which is exacted from any person under the menace of any penalty for its nonperformance and for which the worker does not offer himself voluntarily.

However, Convention No. 29 expressly includes “the imposition of forced labor for the benefit of private individuals, companies or associations.” U.S. legislation fails to honor that definition. When the question of ratifying Convention No. 29 was before the U.S. Tripartite Advisory Panel on International Labor Standards (TAPILS), the panel concluded:

Convention 29 cannot be ratified without amending U.S. law and practice…[TAPILS] concluded that the trend of states to subcontract the operation of prison facilities to the private sector in the United States conflicted with the requirements of Convention 29 relating to circumstances under which the private sector may profit from prison labor.

Consequently, the U.S. definition of “forced labor” is not grounded in international law.

Second, the deviance in the definition of forced labor has salient effects on trade. Private prisons in the United States made billions of dollars in profits last year owing to ramped up immigration detention. Prisoners who “volunteer” to work make $1 a day. The goods they produce, including food items, make it into the stream of commerce.

 3. U.S. forced labor import bans do not make it demonstrably more costly for U.S. companies to produce goods with supply chain inputs than foreign competitors

USTR’s report claims that due diligence costs in the United States are equivalent to a 2.5 percent tariff and attributes it to a RAND study commissioned by the Department of Homeland Security. It uses this statistic as evidence that U.S. producers are at a competitive disadvantage because producers in the offending economies “do not have to undertake due diligence to ensure that their goods are produced free of forced labor.”

USTR mischaracterizes what the RAND Study measures. The study’s actual calculation is entirely different. It calculates the tariff equivalent of the UFLPA’s observed trade effect, that is, the total reduction in import volumes the UFLPA caused.

USTR misattributed a macroeconomic policy effect to a microeconomic firm-level cost category. The UFLPA’s total trade-reducing effect has nothing to do with the specific cost burden on complying businesses. The UFLPA is a statute, not an enforcement action. Its trade-reducing effect reflects U.S. policy choices. However, Section 301(b) requires a finding that foreign conduct burdens U.S. commerce; USTR has instead substituted evidence of a U.S. statute’s trade effects to satisfy that requirement.

I am not suggesting that these data do not exist, merely that USTR has failed to offer them. An empirical survey of U.S. importers testing due diligence costs before and after 2016 would be probative, if USTR has time between now and the rather predictable litigation that will surely arise as this action unfolds.

 4. USTR’s Argument Falsely Assumes U.S. Exporters Are Compliant

The report’s comparative disadvantage argument assumes that U.S. companies do not engage in forced labor practices. This is a separate argument from above. If companies in the United States use forced labor to produce goods they export, then those companies are placing destination country companies at a similar competitive disadvantage.

Tellingly, USTR made no effort to show that U.S. exports comply with forced labor legislation. It merely contrasts the United States with another economy that “permits or does not police the use of forced labor to produce or provide services.” If the United States is guilty of the same, the comparative disadvantage argument either fails or is significantly weakened. Sectors of concern, the report notes, include agriculture, construction, manufacturing, and mining and quarrying. The report argues that “When these goods enter into an economy that prohibits the domestic use of forced labor, they directly undercut domestic producers and importers of legitimate goods.”

USTR is partly right. Its conclusion that forced labor practices undermine compliant companies is well documented. But its failure to engage with the U.S. forced labor practices in those same sectors raises significant red flags. Reports documenting forced labor in the United States are abundant. They suggest that between 50,000 and 1.1 million people are working in forced labor in the United States right now. They work in agriculture, where immigrant workers and young children are exposed to hazardous work without pay, including in export sectors. They work in manufacturing, where unaccompanied migrant children are forced into debt bondage. Private prison labor produces inputs in the global supply chain.

If U.S. exporters, like exporters elsewhere, defy forced labor prohibitions to produce tradable goods, then USTR’s comparative disadvantage argument requires rethinking. Before it can claim unreasonable interference with fair conditions of trade, USTR must think deeply about how U.S. corporate practices play into the so-called “race to the bottom.”

 Conclusion

Forced labor, whether in the United States or abroad, is abhorrent. It distorts the supply chain through perverse incentives to maximize profits at the expense of fundamental human rights. Curiously, USTR’s report characterized the prohibition against forced labor as “universal,” noting “the consensus within the international community that forced labor goods should not exist.” Its arguments seem to advocate for a rule of customary international law that could entice all governments, as a matter of erga omnes obligations, to enforce the prohibition under a shared legal interest. Rather than take that approach, USTR has tied itself (and brought us along for the ride) in circles, requiring it to argue, on the one hand, that forced labor continues to permeate the U.S. supply chain and, on the other, that CBP has effectively enforced U.S. forced labor import bans so as to compel compliance among importers. USTR’s poor reasoning and misleading use of data raise more questions than the report answers.

In this post, I'll go through some of the key issues I'm thinking about in the context of this review:

• Will the USMCA be extended for another 16 year term?
• What's the timing of the review process?
• What are the main substantive issues being discussed?
• Are separate bilateral deals really a possibility?
• Where is Congress in all this?
• What non-trade issues are kind of, sort of a part of these talks?
• Will Canada and the U.S. find a way to be friends again?

To extend or not to extend

I don't want to make people wait, so let's start with the big question on everyone's mind. As part of the Article 34.7 "review and term extension" process, the USMCA "shall terminate 16 years after the date of its entry into force, unless each Party confirms it wishes to continue this Agreement for a new 16-year term ... ." So will the agreement be extended for another 16-year term as part of the upcoming review? I've been skeptical for a while and I'm still skeptical now, for the following reasons.

I've never been a trade negotiator, but from what I understand from those with direct experience, deadlines can be a useful way to force governments to come to agreement. Often this means setting somewhat arbitrary deadlines, just to get people moving, although in some cases you have an election in a particular country that you have to work around, so the deadline is more real.

In the USMCA review and extension context, we do have a deadline of sorts. The problem is, it's somewhat of a soft deadline, in the sense that not much happens if the parties don't extend the USMCA as part of the upcoming review. If they don't extend, the agreement stays in place for 10 more years, and they can just go on negotiating during that time, either though the annual reviews that follow or outside of them. As a result, the incentives for forcing the governments to make a decision to extend may not be there at the moment. Yes, the failure to agree on an extension will create some uncertainty, but 10 years is a long time and I'm not sure how much people will be panicking just yet.

To state the obvious here, the views within the U.S. – and in particular the Trump administration – are the most important on this point (Canada and Mexico would probably be happy to sign on to an extension that kept the agreement as is – Canada said as much here). But the Trump administration is looking to extract various additional concessions, and does not seem particularly eager to extend without getting something more. My instinct here is that the Trump administration might be willing to extend if they got everything they asked for, but they are asking for so much that I'm not sure Canada and Mexico can give it in terms of their own domestic politics.

Also, while there are U.S. companies with an interest in keeping the USMCA in place, and might put pressure on the Trump administration to do so, given the 10 years left on the current term these companies won't be sufficiently panicked yet. Instead, some of them may be happy to wait for a change in who is in charge in U.S. politics, thinking that this process might go better for them in those circumstances.

So my prediction is no extension in July or soon thereafter, but I make it with the same degree of certainty about anything in the policy world at the moment, which is to say very little!

(I need to add a parenthetical here about the possibility of a U.S. withdrawal. I can imagine this will be threatened from time to time, but I'd be very surprised if it actually happened. The chances are so low that I'm not even giving withdrawal its own section in this post!)

The timing of the review

As mentioned in the intro, July 1 is the six-year anniversary of the USMCA entering into force, and Article 34.7.2 tells us that on "[o]n the sixth anniversary of the entry into force of this Agreement, the Commission shall meet to conduct a 'joint review' of the operation of this Agreement, review any recommendations for action submitted by a Party, and decide on any appropriate actions." But what exactly does the timing look like here? Deputy USTR Jeffrey Goettman recently said "we'll make a decision on July 1." But presumably even if they make that decision on July 1 (and I can imagine it might see a bit of a delay), they are unlikely to wrap up everything with the USMCA review on that day, and we should give them some time to meet and discuss things further. The U.S. and Mexico have already scheduled a negotiating round "[d]uring the week of July 20," so it doesn't seem like a quick outcome is being anticipated.

If on July 1 (or shortly thereafter) they are not ready to agree to extend, and this is stated publicly, where do we stand then in terms of the likely timing going forward? Note that Article 34.7.4 seems to provide for an open-ended amount of talking even if there is no decision to extend:

... If one or more Parties did not confirm their desire to extend this Agreement for another 16-year term at the conclusion of a given joint review, at any time between the conclusion of that review and expiry of this Agreement, the Parties may automatically extend the term of this Agreement for another 16 years by confirming in writing, through their respective head of government, their wish to extend this Agreement for another 16-year period.

The "at any time" language provides a lot of leeway here. Let's say the parties get to July 28 or so and recognize that they won't reach an agreement to extend this time around. Well, presumably that would conclude the review, but the "at any time" language makes it sounds like the parties can still keep talking and at some later point – say, just before the midterm elections if someone wanted a political victory to announce – could decide to extend. So, if the parties end up going this route, we could see a continuous negotiation that lingers on and which could lead to a term extension decision at any time. Or things could just linger on without an extension or continuous negotiations, and we could end up with the annual joint reviews that Article 34.7 also provides for.

Summing up here, I feel like there's some uncertainty on the timing of key aspects of the review, and it's not clear to me what each of the three governments has in mind, whether they are all on the same page, and how things will proceed.

Main topics under discussion

Last December, U.S. Trade Rep. Jamieson Greer submitted to Congress a list of what the administration saw as the key issues in the USMCA review. Are these the issues the administration cares most about? Or are these issues just what the administration thinks Congress wanted to hear that they care most about? Hard to say! But certainly it's a good starting point for thinking about what the main topics of discussion will be.

At the outset, though, it's worth noting a broad objective in the USMCA review that reflects the administration's overall trade policy: A concern about trade deficits. As Greer put it at a CFR event in late May in response to the question "[w]hat is it that you want to see changed vis-a-vis Mexico?":

Well, President Trump is concerned about our deficit with Mexico. It has grown over the past. It's one of the few where their deficit has grown with us. So while our overall trade in goods deficit, ... Mexico has been one of the big winners of American diversification from China and Asia, right? They've been one of the big winners. So while our trade deficit in goods has gone down over the past 12 months, the share of imports from Mexico has gone up. They've taken some of that. All that being said, we would like to see a broader and more balanced distribution of that production, including in the United States. So what do we want to see? We want to see that deficit go down. I think that over the course of these negotiations, we are going to be talking about rules of origin in a way that enhances U.S. content in these goods.

Underlying the trade deficit issue, of course, is the Trump administration's goal of bringing manufacturing back to the U.S., in autos and other products, and this is a fundamental U.S. objective in the USMCA review. (A proposal to require 50% U.S. content in USMCA-qualifying autos reflects this idea).

As for more specific topics to be discussed in the review, in his remarks at the CFR event, Greer mentioned Mexico's "external trade policies," noting that Mexico has already "raised tariffs on non-FTA partners, particularly in Asia, because it's in Mexico's interest. Mexico competes with a lot of these Asian economies for the US market, and so that's an area I think of common agreement ... ." Obviously China is the main target here.

China also lies behind various other "economic security" issues: The Trump administration would like to see Canada and Mexico coordinate more with the U.S. on export controls and investment screening (the Biden administration wanted this as well).

And China plays a role in U.S. demands to tighten up rules of origin to limit non-USMCA content in imports that benefit from zero tariffs on USMCA-qualifying goods, including autos and other products.

And finally on China, there will be talks about cooperation on critical minerals.

Summing up those last few paragraphs, China will be a big deal in the USMCA review!

Beyond China, there are traditional issues related to U.S. market access for agriculture, such as dairy in Canada and corn in Mexico, both of which were litigated under the USMCA's state-state dispute settlement mechanism.

And then there are a variety of other laws and regulations that U.S. interest groups have raised concerns about and the U.S. government has already been pushing:

- Mexican energy regulation
- Mexican labor laws
- Mexican treatment of geographical indications
- Canada's online streaming law
- Canadian provincial bans on U.S. alcohol

Clearly, there are a lot of issues on which the Trump administration will be pressing Canada and Mexico. But which of these are most important to the Trump administration? How do they rank all of these? What tradeoffs will they be willing to make? That's very hard to say looking in from the outside. (And they may not know yet either!)

Taking a quick look in the other direction, what do Canada and Mexico want out of all this? One big issue is how the various Section 301 and Section 232 tariffs are applied to them. The forced labor Section 301 investigation targets both Canada and Mexico; the excess capacity Section 301 investigation includes Mexico but not Canada. And the Section 232 tariffs have an impact on a wide range of Canadian and Mexican products.

Canada and Mexico would like as big an exclusion from these tariffs as they can get, on steel, aluminum, autos, lumber and other products. As the Canadian government put it in relation to Section 232, "discussions with the United States on addressing sectoral tariffs will be essential." (The proposed Section 301 tariffs related to forced labor concerns do not cover "USMCA-compliant goods of Canada or Mexico.")

Finally, for Canada, anti-dumping/countervailing duties on lumber are always a concern, and for Mexico these kinds of duties on seasonal produce are an issue, so that may come up as well.

Are separate bilateral deals possible?

As there was with the original NAFTA renegotiation, there has been talk of separating the USMCA into separate bilateral deals between (1) the U.S. and Canada and (2) the U.S. and Mexico (Canada and Mexico have the CPTPP which applies between them). I think it's unlikely things will go this route, but if you wanted to make the case that it could, you might point to the following.

First, Mexico and the U.S. started their USMCA review discussions well before Canada and the U.S. did. Here are some USTR press releases on the Mexico-U.S. developments from March 5, March 18, April 20, May 27, and May 29. The May 27 link sets out the following negotiating timeline:

On May 28-29, Deputy United States Trade Representative Ambassador Jeff Goettman will lead a U.S. delegation to Mexico City for the first bilateral negotiating round with Mexico, which will feature negotiations on economic security and rules of origin for key industrial goods. On June 16-17, the two countries will hold a second negotiating round in Washington, D.C., which will also include discussions on agriculture and a level playing field. During the week of July 20, the United States and Mexico will hold a third negotiating round in Mexico City.

By contrast, formal Canada-U.S. talks have yet to take place, and clearly Mexico and the U.S. started talking earlier and are further along.

Second, as is well-known, and which I'll come back to later in the post, the Canada-U.S. relationship is struggling to deal with some heated rhetoric at the highest political levels. I won't go through all the details, but everyone knows the main points here.

Could all of this lead to separate bilateral deals? At a CFR event in late May, Greer was asked the following by former U.S. Trade Rep. Mike Froman:

... how serious are the fissures with Canada, the rupture with Canada? And can you envisage USMCA being transformed into separate agreement with Mexico, separate agreement with Canada, or no agreement with Canada?

Greer responded:

Well, I would say that, the team right now is in Mexico, my team, and they're negotiating with Mexico on a bilateral basis. I speak with some regularity to my Canadian counterparts. Our sense is that we have with Canada some trade challenges, which ... some people might think, "Oh, those are just irritants." To us, they're significant, and the reality is we've spent the past year and a half going to countries, telling them we have to have some level of tariff on the globe to deal with this giant deficit that we're dealing with, to try to reshore, etc. And most countries have, grudgingly, but they've said, "We understand your policy. We understand, so we're going to negotiate with you. We're going to remove some of these tariffs and non-tariff barriers, etc." Canada's approach has been different. They and China retaliated against the United States. Two countries in the world ... China and Canada. So they're just in a different spot, and it's hard to see necessarily where that ends. I will say, we have trade in energy and minerals and other things, fertilizers between the U.S. and Canada, that really has not been affected. It's been untouched, that's gone without any trouble. I think those are areas of common economic benefit, I think. When it comes to some of these manufactured goods, we have a different view.

Will any of this matter for the USMCA review? As discussed further below, the personal animosity between some people in the U.S. and Canada at the moment is beyond anything I've ever seen in that relationship. In the NAFTA renegotiation, things got a bit touchy between the Canadian and U.S. trade officials, but this time around the touchiness has filtered down to the people. A significant number of Canadian citizens are very angry, and that has had a real impact on how government officials approach things.

Nevertheless, I tend to agree with Prime Minister Carney's assessment of the USMCA negotiations: "These things have their own rhythm, and they also have what's happening above the surface and what's happening below the surface." It may be that some statements by government officials for public consumption are more of a negotiating tactic, and things aren't quite as bad as they sound.

But you never know, and the occasional talk about turning the USMCA into bilateral agreements has me wondering (and worried!) about the following scenario.

I had been assuming that under USMCA Article 34.7, the decision on whether to extend the agreement would be purely a trilateral one. Either the three parties agree to extend, or they don't. But given the recent rhetoric, I'm thinking the Trump administration could actually try to inject a bit of bilateralism into the renewal decision, either as a threat to Canada or just as a preferred outcome.

Let's say the U.S. and Mexico reach agreement on the various issues they have been talking about. Could these two countries then sign a protocol that says something like "in the event that the USMCA is not extended, and terminates in 2036, the USMCA will continue as a bilateral agreement between the U.S. and Mexico"?

I can imagine that Trump administration officials are looking into possibilities like this. Would Mexico go along with it though? They have been talking to the Canadians about all this as well, and it would certainly not be an ideal outcome from their perspective. At the same time, they would still have a trade deal with Canada through the CPTPP.

Anyway, I don't think this is likely, but if we are thinking about all possible outcomes here, we may have to put this on the table.

Where is Congress in all this?

This question about Congress applies to many things in trade policy over the past year and a half or so. Will the USMCA review be the event that wakes Congress up? Are they going to hold any hearings? Will they just put out press releases and letters to the administration, with some generalities about what they want to see here? (So far I've come across letters from a group of Dem Senators, a group of Dem House Ways and Means members, a bipartisan group of Senators on agricultural market access, and a group of House GOP members of Congress on enforcement of the Mexican VAT.)

Will members of Congress mostly just send these letters and lobby quietly behind the scenes? Whatever their plan is, they better start speaking up soon if they want to have an impact on the decision on whether to extend. But maybe they don't see a non-extension as a big deal at this moment either, given the 10-year soft deadline noted above.

Non-trade issues

As much as many of us in the trade policy world might like to, we can't separate out non-trade issues from trade issues. For Canada, some of the key non-trade issues being contested right now relate to defense.

One of these is Canada's potential participation in the Trump administration's Golden Dome project. Back in 2025, Trump said: "I told Canada, which very much wants to be part of our fabulous Golden Dome System, that it will cost $61 Billion Dollars if they remain a separate, but unequal, Nation, but will cost ZERO DOLLARS if they become our cherished 51st State ... They are considering the offer!" I'm not sure if that reflects actual costs, but regardless, it would cost Canada something. Is paying some amount here worth it for Canada to get better USMCA review terms?

Along the same lines, there are also some purchases of planes at issue right now. Canada is trying to decide between buying U.S. F-35s and Swedish Gripens; and there is a radar plane purchase up for grabs too, with Canada apparently going with Saab over U.S. suppliers.

I don't know exactly how the Canadian government is thinking about all this, but I can imagine they are using these military spending decisions as leverage to try to get a more favorable trade deal. (There was also a U.S. pause with the U.S.–Canada Permanent Joint Board on Defence (PJBD), which may fit in here somehow, I think.)

And then on the Mexican side, you have the perennial "cracking down on cartels" issue, as well as the more recent "oil sales to Cuba" issue.

These issues may not come up directly between the USMCA trade negotiators, but they will be there in the background and could influence the USMCA review outcomes.

Tough rhetoric and bad feelings in Canada-U.S. relations

Trump and his administration must be aware of the effect the 51st state talk has had in Canada. Whatever their level of awareness, though, they have not toned things down very much, with the U.S. ambassador to Canada being quite aggressive with his rhetoric.

Canadians have reacted to all this with moves such as the Ontario ad touting Reagan's tariff policies or Carney's speech in Davos. In response, the Trump administration and its surrogates have been very critical of the Canadian reaction. Here's Deputy USTR Rick Switzer speaking at an event in April:

What I'll say is that the countries that are led by serious leaders, and I'll say President Sheinbaum is a serious leader in Mexico, when we came in and we were talking, Ambassador Greer was just there, we're talking to Mexico about this, and the President has spoken directly about this, they have been very serious. Look, we know the United States is our most important economic partner, that it is the policy of President Sheinbaum, that the United States and Mexico will have a positive economic relationship, and that we know that we'll have some friction, ... but we're going to figure it out, right? This is the bottom line.

There are other economies who decided to make it personal. And you know, I think Carney has made it personal. I think it's political malpractice for the prime minister of ... Canada to pit politically himself against any president. I don't care what the president is, who the president is, what party they represent, it's political malpractice. Canada is dependent upon the US economy. That's just a fact, right? That's not hubris. That's not something that Canada needs to be concerned about. It's not something Canada can change, right? The fundamental fact is geography wins out. Canada is located where it's located. They can't move shop. They can have a weak economy that is underperforming and not doing well, and Carney can feel superior. Or they can have an economy that participates as a partner to the US economy. And Carney can do what a grown-up should do, which is figure it out, and come like President Sheinbaum and decide that the United States and Canada will have a positive economic relationship. It's my job as a person who's [supposed] to protect Canadian jobs, Canadian citizens, and the Canadian economy, to not let my ego and my feelings dictate what's best for my own economy.

See also Greer's remarks quoted above.

I'm going to offer three points here:

• All of this is bad, and it will take some time to repair the damage to the broader Canada-U.S. relationship.
• If the administration's view is that the Canadians should not respond to Trump's statements, I would say this is unrealistic given Canadian domestic politics. Even if taking the high road were the right approach, I don't think Canada's politics would allow it. (American politics certainly does not allow for this sort of thing!)
• The impact of all this tension means that trade is affected in a way that goes beyond government-enacted trade barriers that can be dealt with through trade agreements. When you have Canadians choosing not to buy U.S. goods or travel to the U.S., tariffs and trade barriers become less important. It's not about market access anymore, it's about whether a market for those goods and services exists.