Washington holds back the Section 301 excess capacity report that would recommend a 7.5 percent duty on Chinese goods, preserving leverage days before a Trump and Xi summit and leaving importers to budget for a tariff that exists only on paper
WASHINGTON, SEPT. 19, 2026
The Trump administration is expected to withhold the release of a long anticipated trade report on foreign manufacturing overcapacity, and the tariffs that would flow from it, until after next week’s summit between President Donald Trump and Chinese President Xi Jinping, according to reporting by Bloomberg News published on September 17 and by the trade publication Inside U.S. Trade, which first disclosed the delay.
The decision, if it holds, postpones what would be one of the most consequential tariff actions of the year. Bloomberg reported that the administration had intended to publish the excess capacity report before the leaders met and that the document would recommend an additional 7.5 percent duty on Chinese goods. That increment would lift the second term tariff burden the administration has layered onto Chinese imports to roughly 20 percent, a level Beijing has previously described as consistent with the trade truce the two governments reached earlier in the current cycle of negotiations.
The Office of the U.S. Trade Representative and the White House did not respond to requests for comment from Bloomberg, and the reason for the delay has not been stated publicly. For American importers, exporters, customs brokers and corporate treasurers, the practical effect is familiar and unwelcome: a tariff that is neither imposed nor withdrawn, hanging over fourth quarter purchase orders and 2027 sourcing plans with no effective date attached.
What the excess capacity investigation is
The pending action grows out of a set of investigations the administration opened in March 2026 under Section 301 of the Trade Act of 1974. Section 301 is the statute that gives the U.S. Trade Representative authority to investigate and respond to foreign acts, policies and practices that are unreasonable or discriminatory and that burden or restrict U.S. commerce. It is the same authority that produced the original List 1 through List 4 tariffs on Chinese goods beginning in 2018, and the same authority the administration used in July of this year to impose duties tied to forced labor enforcement failures.
The excess capacity inquiries covered more than a dozen major trading partners rather than China alone. Reporting on the March initiation indicated that USTR opened proceedings touching China, the European Union, South Korea, Vietnam, Taiwan and Japan among others, with the agency identifying different sectors of concern for each economy. Semiconductors, steel and other capital intensive industries featured prominently in the sectoral flags.
The underlying theory is one that has circulated in Washington trade policy circles for years and that has bipartisan adherents. The argument holds that state directed investment, subsidized credit, preferential land and energy pricing, and tolerance for loss making producers create manufacturing capacity far in excess of what domestic demand can absorb. That surplus output is then exported at prices that domestic producers in market economies cannot match, distorting global prices and hollowing out industrial bases abroad. Advocates of the approach say conventional antidumping and countervailing duty cases, which are filed product by product and country by country, are too slow and too narrow to address a problem that manifests across dozens of sectors at once.
Critics counter that Section 301 was designed to address specific foreign practices harming specific U.S. commercial interests, not to serve as a general purpose instrument for reshaping global industrial policy, and that a broad excess capacity action invites retaliation and litigation without a clear endpoint.
Why the timing matters so much
The delay is best understood against the legal upheaval that has reshaped U.S. tariff policy this year.
On February 20, 2026, the Supreme Court ruled 6 to 3 that the International Emergency Economic Powers Act does not authorize the president to impose tariffs, invalidating the country specific reciprocal duties that had been the centerpiece of the administration’s trade program. The decision removed, at a stroke, the legal foundation for a substantial share of the tariff wall built over the preceding year.
What followed was a rapid pivot to statutes with firmer footing. Section 232 of the Trade Expansion Act of 1962, which authorizes import restrictions on national security grounds, has been used aggressively across metals, pharmaceuticals, drones and other sectors. Section 301, with its investigation and determination procedures, has been used to build a parallel structure of country specific duties that can survive judicial review in a way the emergency powers tariffs did not.
Transport Topics, republishing the Bloomberg report, noted that the expected excess capacity levies, taken together with the forced labor duties already in force, are anticipated to bring effective tariff rates back toward the levels of the country specific duties the Supreme Court struck down. In other words, the excess capacity report is not a marginal addition. It is a load bearing element in the administration’s effort to rebuild, on sturdier statutory ground, what the Court took away.
That makes the decision to hold it back before a leaders’ summit legible. A tariff not yet announced is a threat that can be traded. A tariff already imposed is a fact that must be unwound, at political cost, if it is to be used as a concession. Trump has repeatedly used pending tariff decisions as negotiating instruments, and the pattern is consistent here.
U.S. and Chinese officials are expected to take up a broad agenda at the summit, including the war in Iran, artificial intelligence and trade, according to the Bloomberg account. The breadth of that agenda cuts both ways for tariff watchers. It raises the possibility that the 7.5 percent figure is traded away or reduced in exchange for movement on nontrade files, and it equally raises the possibility that a breakdown on any one file hardens the tariff outcome.
The forced labor precedent shows how fast this can move
If importers want a template for how an excess capacity action would be executed, the forced labor proceedings of this year provide one, and the timeline is instructive.
USTR initiated sixty investigations on March 12, 2026, under Section 302(b)(1) of the Trade Act, examining whether various economies had failed to impose and effectively enforce a prohibition on the importation of goods produced with forced labor. On June 2, 2026, the agency determined that actionable conduct existed in every one of those investigations, publishing a comprehensive report alongside a Federal Register notice proposing remedial tariffs and inviting comment. The resulting duties took effect at 12:01 a.m. Eastern time on July 24, 2026.
That is roughly four and a half months from initiation to collection across sixty economies simultaneously. It is an aggressive pace by historical Section 301 standards, and it was achieved by running the determinations in parallel and by using a single consolidated notice rather than sixty separate proceedings.
The excess capacity investigations were initiated one day earlier, on March 11, 2026, covering a smaller but more economically significant group of partners. On the forced labor timetable, determinations and a proposed action would have been expected around the start of June, with duties in force by late July. The fact that the report has not issued more than six months after initiation suggests either substantive difficulty or deliberate sequencing, and this week’s reporting points toward the latter.
For importers, the lesson is that the interval between publication of a determination and the imposition of duties can be as short as six to eight weeks. That is not enough time to redesign a supply chain. It is barely enough time to accelerate in transit shipments and reprice a catalogue. Preparation has to precede the notice.
The numbers behind the exposure
The macroeconomic backdrop helps size what is at stake.
The Penn Wharton Budget Model, in an update published September 9, 2026, put the average effective U.S. tariff rate at 6.7 percent as of July 2026, with China facing the highest effective rate among major trading partners at 22.8 percent, a figure the model described as a marked decline from earlier months. The decline reflects both the Supreme Court’s invalidation of the reciprocal duties and the trade truce arrangements that followed.
The Tax Foundation, in its own tracker, estimates the effective tariff rate reflecting revenues actually collected at 7.2 percent for calendar year 2026, and the applied tariff rate at 11.8 percent, against 1.5 percent in 2022. The organization estimates that tariffs currently imposed and scheduled will raise approximately 1.4 trillion dollars for the federal government over the 2026 to 2035 window on a conventional basis.
Against those baselines, an additional 7.5 percentage points on Chinese goods is material but not transformative at the aggregate level. Its bite is concentrated. For importers whose bill of materials is heavily Chinese and whose products fall outside the sectoral Section 232 regimes, 7.5 points is the difference between a manageable margin compression and a repricing exercise. For firms already absorbing Section 232 metals or component duties on top of existing Section 301 lists, the stacking effect is what matters, and stacking has been the defining compliance problem of the year.
Reaction: relief, frustration and suspicion
Reaction to the reported delay has split along predictable lines, though with less heat than earlier episodes in the tariff cycle.
Importer groups and retail associations have generally welcomed any postponement, on the straightforward grounds that a later effective date is a cheaper effective date and that the fourth quarter is the worst possible moment to reprice goods already on the water. Customs brokers and trade compliance practitioners have been more equivocal. Several have noted publicly over the past year that the operational cost of repeated announcements, suspensions, partial revocations and refund programs now rivals the cost of the duties themselves.
That frustration is not abstract. Following the Supreme Court decision, U.S. Customs and Border Protection stood up a Consolidated Administration and Processing of Entries system to handle refunds of invalidated IEEPA duties. Court filings and practitioner alerts through the middle of the year indicated that more than 95 billion dollars had been queued for refund through that system, with more than 40 billion dollars expected to have been disbursed by the end of June 2026. The government has separately appealed aspects of the Court of International Trade’s refund order to the Federal Circuit, contesting the trade court’s authority to compel refunds on entries that are both liquidated and past the window in which CBP can reprocess them.
Domestic producer groups in steel, solar, batteries and machine tools have taken the opposite view of the delay, arguing that every month without an excess capacity remedy is another month of import penetration that will be difficult to reverse. Their argument is that capacity, once installed abroad and idled domestically, does not come back on a policy timetable.
There is also a strand of commentary, visible across trade law firm alerts this week, that treats the delay as evidence of internal disagreement about the merits rather than pure summit choreography. The excess capacity theory requires USTR to make findings about the industrial policies of allies as well as adversaries. Extending the same remedy logic to the European Union, Japan, South Korea and Taiwan, all of which are simultaneously being courted for semiconductor and critical mineral investment commitments, is politically awkward in a way that a China only action would not be.
What it means for importers and exporters
For U.S. companies, the immediate task is scenario planning rather than reaction, because there is as yet nothing to react to.
Classification and origin discipline comes first. An excess capacity action under Section 301 would almost certainly be implemented as a Chapter 99 subheading applied to specified Harmonized Tariff Schedule lines, in the pattern of every other 301 action. That means exposure is determined by tariff classification and by country of origin under the substantial transformation test, not by where a supplier’s invoice is issued or where final assembly happens to occur. Companies that have not recently audited their HTS classifications against the lines flagged in the March initiation notices are carrying unmeasured risk.
Stacking is the real number. Any new 301 duty would sit on top of existing 301 lists, most favored nation duties, and any applicable Section 232 sectoral tariffs. The administration’s 2026 adjustments to the metals regimes, which moved several programs to apply duties on full customs value rather than on metal content alone, have already made stacking calculations materially harsher for derivative products. A 7.5 percent headline understates the landed cost effect for goods that also carry a metals derivative duty.
First sale and valuation planning becomes more attractive at the margin. As ad valorem rates climb, the return on legitimate customs valuation planning rises with them. First sale for export, where a multi tier transaction structure supports it, and careful treatment of assists, royalties and post importation adjustments are the conventional levers. So are foreign trade zones and duty drawback for goods that are re exported, though 301 duties have historically been drawback eligible in ways that 232 duties are not, a distinction worth confirming against current CBP guidance rather than assuming.
Contract language matters more than it did. Purchase agreements signed before the Supreme Court ruling frequently allocated tariff risk by reference to duties in effect on the date of order. Agreements signed since have tended to include explicit change in law provisions with price adjustment mechanics. Companies negotiating 2027 supply contracts now should assume further statutory substitution rather than tariff stability.
Exporters face the mirror image. Any broad U.S. action on excess capacity invites retaliation, and retaliation in recent cycles has been aimed with precision at politically exposed U.S. export sectors, particularly agriculture. U.S. soybean, sorghum, pork and dairy exporters have been through this pattern before and have institutional memory of how quickly order books can empty. Aerospace, semiconductors and medical devices are the other perennial targets.
Sector by sector, the exposure is uneven
Aggregate effective rates conceal the distribution that actually determines corporate outcomes, and the excess capacity framing makes the distribution unusually predictable.
Because the theory targets subsidized overcapacity, the sectors most likely to appear in any product annex are the capital intensive ones where state directed investment is easiest to document. Steel and steel derivatives, aluminum and its downstream products, solar cells and modules, lithium ion cells and battery components, electric vehicles and their subassemblies, legacy node semiconductors, shipbuilding inputs, machine tools, industrial robotics and large scale chemical intermediates are the recurring names in the policy literature and in the March initiation notices.
Several of those categories are already subject to Section 232 metals duties on a full customs value basis following this year’s adjustments, and several are already on existing Section 301 lists. That is why stacking, rather than any single rate, is the operative concept. An importer of a steel intensive machinery component from China could plausibly face a most favored nation rate, an existing Section 301 list rate, a Section 232 derivative duty assessed on full customs value, and a new excess capacity duty, all on the same entry line.
Consumer facing sectors are comparatively less exposed under an excess capacity theory than they were under the reciprocal tariff regime the Supreme Court struck down, because apparel, footwear, toys and household goods are not typically framed as products of subsidized industrial overcapacity. That is cold comfort for those sectors, which remain exposed to other instruments, but it does mean that a company’s excess capacity risk correlates with the industrial character of its import book rather than with its total Chinese sourcing share.
The waiting game
The most probable near term sequence, on the current reporting, runs as follows. The summit takes place next week. The excess capacity report is released at some point after it, with a recommended rate that may or may not be the 7.5 percent figure Bloomberg described. A Federal Register notice follows with a comment period, an effective date, and a product annex. Practitioners then have somewhere between two weeks and two months to reclassify, reprice and, where possible, accelerate shipments.
That sequence is not guaranteed. The administration could decline to publish the report at all if the summit produces a framework it values more than the tariff. It could publish a narrowed version covering fewer economies. It could publish the full version and immediately suspend application to negotiating partners, a technique it has used repeatedly.
What is reasonably clear is that the direction of travel for U.S. tariff policy has not changed. The Supreme Court removed one legal instrument. The administration has responded by building the same wall out of different statutory bricks, using Section 232 for sectors it deems security relevant and Section 301 for conduct it deems unfair. The excess capacity report is the next brick. Its delay is a scheduling decision, not a reversal.
For importers, the counsel that follows is unglamorous. Know your classifications. Model the stacking. Read the annexes when they come, not the headlines. And price 2027 on the assumption that the tariff environment will be at least as demanding as today’s, because nothing in this week’s reporting suggests otherwise.