Washington holds back the Section 301 structural excess capacity determination and its recommended 7.5 percent duty on Chinese goods until after the Sept. 24 Trump and Xi meeting, leaving sixteen economies and thousands of importers waiting on a decision that was supposed to land this week.
WASHINGTON, Sept. 18, 2026
The Trump administration is expected to postpone the announcement of new tariffs targeting trading partners’ alleged structural excess manufacturing capacity until after next week’s summit between President Donald Trump and Chinese President Xi Jinping, according to a Bloomberg report published Thursday that cited people familiar with the matter.
The delay is significant for reasons that go well beyond the calendar. The administration had intended to release a trade report on excess capacity ahead of the leaders’ meeting, and that report was expected to recommend a 7.5 percent tariff on Chinese products. Bloomberg reported that such a duty would restore this administration’s second term tariffs on China to roughly 20 percent.
Trump is expected to host Xi in Washington on Sept. 24, more than four months after the two leaders last met in Beijing. Bloomberg reported that United States and Chinese officials are expected to discuss Iran, trade, and artificial intelligence, and that the final tariff rate remains unsettled.
The reason for the delay was not disclosed. The most widely offered reading among trade practitioners is the simplest one. An unannounced tariff is leverage. An announced tariff is a fact that the other side must respond to, and responding to it publicly is harder than negotiating around the threat of it privately.
Sixteen economies, not one
The excess capacity proceeding is frequently described in shorthand as a China action. It is not.
On March 11, 2026, the Office of the United States Trade Representative initiated investigations under Section 301 of the Trade Act of 1974 into structural excess capacity or production in certain manufacturing sectors across sixteen economies. According to USTR’s initiation notice and subsequent law firm analyses from White and Case, Davis Wright Tremaine, Brownstein, and Mayer Brown, the covered economies are China, the European Union, Singapore, Switzerland, Norway, Indonesia, Malaysia, Cambodia, Thailand, Korea, Vietnam, Taiwan, Bangladesh, Mexico, Japan, and India.
The sectoral scope is extraordinarily broad. The notice identified aluminum, automobiles, batteries, cement, chemicals, electronics, energy goods, glass, machine tools, machinery, non ferrous metals, paper, plastics, processed food and beverages, robotics, satellites, semiconductors, ships, solar modules, steel, and transportation equipment.
Read together, the country list and the sector list cover a very large fraction of what the United States imports. Three of the four largest sources of American imports are on the list. So are most of the economies that received investment under the China plus one relocation strategy of the past decade.
The procedural record is complete. A public docket opened on March 17. Written comments and hearing requests were due April 15. Public hearings ran in Washington from May 5 to May 8. Under Section 301, USTR must determine whether the identified acts, policies, and practices are unreasonable or discriminatory and whether they burden or restrict United States commerce. An affirmative determination permits the President to impose tariffs and other restrictions or to enter negotiations with the government concerned.
The determination is what has now slipped.
The architecture this proceeding is meant to hold up
To understand why a 7.5 percent recommendation matters more than the number suggests, it helps to trace how the administration arrived here.
The Supreme Court ended the first architecture in February. In a six to three decision issued Feb. 20, 2026, the Court held that the International Emergency Economic Powers Act does not authorize the President to impose tariffs of indefinite scope, affirming the Federal Circuit’s August 2025 ruling. That decision eliminated the country specific reciprocal tariffs and the trafficking and immigration tariffs at a stroke, and triggered a refund process that the Court of International Trade has valued at roughly 165 billion dollars.
The administration responded with a bridge and a replacement.
The bridge was Section 122 of the Trade Act of 1974, the balance of payments authority, which permits a surcharge of up to 15 percent for a maximum of 150 days without congressional action. A 10 percent baseline went into effect on Feb. 24 and expired on July 24 when the statutory clock ran out.
The replacement is Section 301. Unlike IEEPA, Section 301 is a purpose built trade statute with a developed administrative record requirement, a hearing process, and decades of judicial gloss. It is slower and more procedurally demanding, but it is far harder to challenge. Korea’s Yonhap news agency, in reporting carried by The Korea Times, noted explicitly that the administration carried out the excess capacity and forced labor investigations under Section 301 as it pushed to roll out new duties to replace the country specific emergency tariffs that the Supreme Court struck down in February.
The first piece of that replacement is already in force. In parallel with the excess capacity probe, USTR opened a second Section 301 investigation in March into roughly 60 economies over whether their governments had taken sufficient steps to prohibit the importation of goods produced with forced labor. USTR announced its final determination in June and imposed the resulting duties effective July 24, 2026, at 10 percent for most goods from 15 trading partners and 12.5 percent for most goods from 45 others. Korea and Japan are among the 60 partners covered, at up to 12.5 percent.
The excess capacity action is the second piece. Until it lands, the post IEEPA tariff structure is incomplete.
Twenty percent, and what sits beneath it
Bloomberg’s reporting that a 7.5 percent excess capacity duty would take China to roughly 20 percent is worth unpacking, because the figure is a sum of layers rather than a single rate.
Chinese goods entering the United States today face, in varying combinations depending on the tariff line, the column one most favored nation rate, the legacy Section 301 China tariffs from the 2018 and 2019 actions and the 2024 four year review, the new Section 301 forced labor tariff, and where applicable Section 232 duties on steel, aluminum, copper, automobiles, pharmaceuticals, and unmanned aircraft systems. Section 232 duties stack independently of the Section 301 layers.
Adding 7.5 percent on top of that is not a modest adjustment for importers whose products sit in multiple categories. A steel intensive machinery component from China could plausibly carry a Section 232 metals duty, a legacy Section 301 List 3 duty, a forced labor Section 301 duty, and the new excess capacity duty, all on the same entry line.
That stacking problem is the reason the trade bar has been pressing USTR for a consolidated rate schedule rather than a sequence of standalone actions. No such consolidation has been announced.
Reactions and the view from allied capitals
The muted official response to the delay tells its own story.
Seoul’s interest is direct. Korea is named in both March investigations, sits in the forced labor tier at up to 12.5 percent, and separately concluded a bilateral tariff arrangement with Washington that exchanged a 350 billion dollar investment commitment for a 15 percent rate on Korean autos and auto parts rather than the 25 percent Section 232 rate. That package comprises 150 billion dollars in shipbuilding cooperation and 200 billion dollars across semiconductors, pharmaceuticals, critical minerals, energy, artificial intelligence, and quantum computing, with an annual investment cap of 20 billion dollars.
The Korea Economic Institute of America and Korean outlets including Seoul Economic Daily have reported friction over project scope, with the combined scale of projects sought by the United States side, including a gas fired combined cycle plant in Encinal, Texas, eight large nuclear reactors, and liquefied natural gas development in Alaska, exceeding both the strategic investment ceiling and the total package agreed last year. A new excess capacity tariff on Korean shipbuilding, steel, batteries, or semiconductors would sit awkwardly alongside an agreement premised on Korean capital building exactly those industries inside the United States.
Tokyo faces a similar structure. Brussels, Mexico City, Taipei, and New Delhi are each exposed through different sectors.
Beijing has not commented specifically on the delay. China’s Foreign Ministry has maintained its general position, articulated by spokesman Guo Jiakun in a separate context this week, that it opposes extraterritorial measures lacking a basis in international law and United Nations Security Council authorization.
The economics of overcapacity
The substantive case the administration is building rests on a genuine and widely acknowledged phenomenon.
Global manufacturing capacity in several of the listed sectors materially exceeds global demand. Steel is the canonical example, with world capacity persistently above consumption for more than a decade. Solar module capacity expanded far faster than installation rates. Battery cell capacity was built against demand forecasts that have since been revised down. Shipbuilding capacity is concentrated in a handful of yards operating with state support.
Where the policy argument becomes contested is in the attribution. The administration’s theory is that these gluts are the product of deliberate government policy, including directed credit, preferential land and energy pricing, tax preferences, and tolerance of loss making expansion, and that the resulting exports burden United States commerce by depressing prices below the level at which American producers can sustain investment.
Critics of the approach make two arguments. The first is that overcapacity is a normal feature of capital intensive industries with long build cycles and is not by itself evidence of unfair practice. The second is that a tariff on the symptom does not address the cause, and that taxing imports from sixteen economies simultaneously simply raises input costs for American manufacturers without reducing global capacity by a ton.
Both arguments will feature in the litigation that any affirmative determination will attract.
What the delay means for importers
The immediate practical consequence of a delay is that nothing changes at the border. No new duty applies. Entries continue to be filed and liquidated under the existing rate structure. Importers who have been accruing reserves against a possible excess capacity duty can leave those reserves where they are.
The medium term consequences are less comfortable.
Section 301 actions are generally prospective. They apply to goods entered, or withdrawn from warehouse for consumption, on or after the effective date stated in the implementing Federal Register notice. That means a delay is genuinely a reprieve, not a deferral of accrued liability. Goods landed between now and the eventual effective date will not be reached.
That creates an obvious incentive to pull forward shipments. It is an incentive importers should treat carefully. Front loading inventory carries working capital cost, warehousing cost, obsolescence risk, and in some categories the risk that the pulled forward goods arrive into a demand environment softened by the same tariff that prompted the pull forward. The 2018 and 2019 rounds produced a well documented inventory whipsaw in several consumer categories, and the lesson from that period is that only importers with genuine visibility into the effective date benefited.
Second, the absence of a determination is not the absence of a proceeding. The record closed in May. The analytical work is done. USTR can move from a completed record to a published determination and an effective date on a short timeline. Importers should assume that once the summit concludes, the window between announcement and effectiveness could be measured in weeks rather than months.
Third, exclusion processes matter enormously and are not guaranteed. The 2018 and 2019 China actions eventually produced a formal exclusion mechanism, and USTR has extended certain of those exclusions into November 2026. The forced labor action in July did not come with a comparable general exclusion process. Whether the excess capacity action includes one is unknown, and it is a question worth raising with counsel and with trade associations now, while the determination is still unpublished.
Fourth, classification discipline pays. With a sectoral action of this breadth, the annex will define coverage by Harmonized Tariff Schedule subheading. Products that sit near a classification boundary may fall inside or outside the action depending on a determination that was made years ago and never revisited. A binding ruling request, filed before an action is announced, is far more credible than one filed the week after.
Exporters and the reciprocal question
American exporters have their own reason to watch the outcome.
Sixteen economies is a large enough group that coordinated response becomes plausible in a way it is not when a single partner is targeted. The European Union has an enforcement regulation and an anti coercion instrument available. China has demonstrated willingness to use export licensing on critical minerals and rare earth processing as a counterweight. India has previously imposed retaliatory duties in response to the Section 232 metals actions.
United States agricultural exporters, aircraft manufacturers, medical device firms, and spirits producers have all been on retaliation lists in previous rounds and should expect to be on them again.
What to watch
The Sept. 24 meeting is the fulcrum. Bloomberg’s reporting indicates the agenda covers Iran, trade, and artificial intelligence, a combination that suggests the tariff question is one variable in a broader negotiation rather than the sole subject.
Three outcomes are plausible. The determination could issue shortly after the summit at or near the reported 7.5 percent, applied broadly across the sixteen economies. It could issue at a lower rate or with a narrower sectoral annex as part of an understanding reached in Washington. Or it could be held indefinitely, functioning as standing leverage in the manner the administration has used other pending actions this year.
The one outcome that appears unlikely is withdrawal. The excess capacity proceeding is a structural component of the post IEEPA tariff architecture, not an opportunistic threat. Having built the record, the administration has little reason to discard it.
For importers, the planning assumption should be that a duty is coming, that its timing is political rather than procedural, and that the interval between announcement and effectiveness will be short.
The sectors most likely to be named first
If the administration narrows the annex rather than applying a flat rate across all twenty one listed sectors, the sequencing is reasonably predictable from the record.
Steel and aluminum would almost certainly appear, though the practical effect is muted because Section 232 duties of 50 percent already apply to a wide range of steel and aluminum articles and derivatives. The Section 232 program was also overhauled earlier this year so that duties apply against the full customs value rather than only the metal content, which raised effective rates on derivative products substantially. Layering a Section 301 excess capacity duty on the same goods produces a rate that invites litigation over double counting.
Solar modules and cells are a strong candidate. The sector has the clearest global capacity overhang, the most documented state support, and an existing American trade remedy history running through antidumping and countervailing duty orders and the earlier safeguard action. The complication is that the same tariffs raise the installed cost of domestic generation capacity at a moment when electricity demand from data centers is growing rapidly.
Batteries and battery materials present the same tension in sharper form. The administration has spent two years encouraging domestic cell manufacturing, and those plants depend on imported cathode and anode material, separators, and processing equipment. A broad duty raises the cost of the very investment the policy is meant to induce.
Shipbuilding is politically attractive and commercially marginal. The United States imports very few commercial vessels directly, so a tariff would function largely as a signal. The more consequential shipbuilding measures have run through separate port fee and procurement channels.
Semiconductors are the most difficult. Coverage would have to define whether the duty attaches to bare die, packaged devices, or downstream products containing them, and each choice produces a different set of winners and losers among American firms. The administration has handled semiconductor trade policy through Section 232 and export controls rather than Section 301 to date, and extending the excess capacity theory to chips would be a notable departure.
Machinery, machine tools, and robotics would be the quiet surprise. Coverage there reaches capital equipment purchased by American manufacturers, which raises the cost of domestic capacity expansion in exactly the sectors the policy claims to defend.
Compliance housekeeping that pays for itself
Whatever the annex ultimately says, several steps are worth completing during the waiting period because they have value regardless of outcome.
Reconcile your Harmonized Tariff Schedule classifications against actual product specifications rather than against historical practice. Classification drift is common in companies that have grown by acquisition or changed suppliers, and a sectoral action defined by subheading turns a stale classification into a direct financial exposure.
Confirm country of origin for every stock keeping unit, and document the analysis rather than relying on supplier attestation. Where processing occurs across multiple jurisdictions, the substantial transformation analysis should be written down with supporting bill of materials and value added data.
Audit your customs valuation methodology. Verify that assists, royalties, proceeds of subsequent resale, and packing are being treated correctly, and assess whether a first sale for export structure is available and defensible.
Quantify the exposure. Build a model that applies candidate rates to last twelve months of entry data by tariff line and origin. Boards approve contingency plans far more readily when the number is specific.
Engage through trade associations while the determination is unpublished. Once an action is announced, the only remaining avenues are an exclusion process that may not exist and litigation that will take years.