A planned 7.5 percent overcapacity tariff would push US duties on Chinese goods to the very ceiling of the trade truce, just as new export data shows Beijing’s shipping machine running hotter than ever ahead of the September 24 Trump-Xi summit
WASHINGTON, September 9, 2026
Two weeks before President Donald Trump is scheduled to welcome Chinese President Xi Jinping to Washington, the administration is finalizing a new 7.5 percent tariff on Chinese imports that would test the outer limit of the fragile trade truce between the world’s two largest economies. The measure, which would penalize Beijing for what Washington calls structural industrial overcapacity, arrives against a backdrop of fresh Chinese customs data, released Monday, showing exports growing 25 percent year over year in August, the fastest pace of the year and a statistic almost tailor made to reinforce the administration’s argument that China is flooding world markets with underpriced goods.
The overcapacity tariff, first reported by Bloomberg News in late August and confirmed by people familiar with the deliberations who spoke to the Associated Press, would be imposed under Section 301 of the Trade Act of 1974, the trade statute that has become the load bearing wall of US tariff policy since the Supreme Court demolished the administration’s original framework in February. It would come on top of tariffs of 10 to 12.5 percent announced in July against 60 economies accused of failing to enforce bans on goods made with forced labor, bringing the administration’s second term duties on Chinese goods to roughly 20 percent.
That number is not arbitrary. Twenty percent is precisely the ceiling that China’s Commerce Ministry says Washington committed to respecting when the two sides extended their trade truce. On July 27, the ministry confirmed that the United States had agreed to cap its replacement tariffs on Chinese goods at 20 percent, with the rate then standing at 12.5 percent. The planned 7.5 percent overcapacity levy would fill the remaining headroom exactly, taking the administration to the maximum the truce allows without technically breaching it.
A tariff engineered to the millimeter
The precision is the point. According to the people familiar with the deliberations cited in the Associated Press reporting, administration officials selected the 7.5 percent figure specifically because they believe it would not endanger the one year truce or the planned White House summit, now expected on September 24. Trump could still change his mind, those people stressed, and the White House, the Office of the US Trade Representative and the Chinese embassy in Washington all declined to comment on the deliberations.
The choreography reflects a delicate balancing act. Trump wants to arrive at the summit having demonstrated maximum pressure on Beijing, and the overcapacity investigation gives him a legally defensible instrument for doing so. At the same time, neither capital wants a rupture. The truce, first reached in late 2025 and extended for one year through November 10, 2026, restored a measure of predictability to a relationship that had earlier featured tariff rates well above 100 percent, and both governments have been carefully assembling deliverables for the leaders’ meeting. Reuters reported on September 4 that the two sides are preparing a mid September dialogue on artificial intelligence safety, a signal that the summit agenda is meant to stretch well beyond tariffs.
Beijing, for its part, has been preparing its rebuttal for months. China’s Commerce Ministry published a lengthy report, titled China’s Position on the So called Excess Capacity Issue, rejecting the premise of the American investigation and arguing that China has never sought a large trade surplus. Chinese officials warned through August that new tariffs would draw a response, though analysts note that Beijing has so far calibrated its own countermeasures to stay within the truce framework as well.
From Supreme Court defeat to Section 301 rebuild
The overcapacity tariff is best understood as the latest chapter in the administration’s rebuilding project after its most significant legal defeat. On February 20, 2026, the Supreme Court ruled 6 to 3 in Learning Resources v. Trump that the president could not use the International Emergency Economic Powers Act to impose sweeping tariffs on imports from nearly every trading partner, striking down the centerpiece of the administration’s first year trade agenda. Within hours, the White House pivoted to Section 122 of the Trade Act of 1974, a balance of payments provision, to impose a temporary tariff on all countries, and set about constructing a more durable legal architecture.
The core of that architecture is Section 301, which allows the US Trade Representative to impose duties in response to foreign practices found to be unreasonable or discriminatory and to burden US commerce. Unlike the emergency powers statute, Section 301 has survived decades of litigation and was the basis for the original China tariffs of 2018 and 2019, which courts upheld.
In March, USTR launched twin Section 301 investigations of unprecedented scope. One targeted forced labor enforcement across 60 economies. The other, initiated March 12 and formally noticed in the Federal Register on March 17, examined structural excess capacity and production in manufacturing sectors across China and 15 other major trading partners, a list that included the European Union, Japan, South Korea, India, Vietnam, Taiwan, Mexico and others. The investigation found, in USTR’s words, evidence that China’s goods trade surplus is driven by increasing excess manufacturing capacity, and that in some sectors Chinese excess capacity has driven global overcapacity.
The forced labor track moved first. USTR determined on June 2 that the identified practices were actionable, received more than 1,600 public comments, heard testimony from more than 100 witnesses in July hearings, and imposed tariffs of 10 percent on most goods of 15 trading partners and 12.5 percent on most goods of 45 others, effective July 24. The overcapacity track is now reaching its own decision point, and administration officials have been racing to publish the investigation’s findings before the September 24 summit so that the new tariff can be announced around the leaders’ meeting.
Beijing’s export machine complicates the optics
Monday’s Chinese trade data will make it harder for Beijing to argue the overcapacity concern is manufactured. According to figures released September 8 by China’s General Administration of Customs and analyzed by the research firm Trivium China, Chinese exports grew 25.0 percent year over year in August, accelerating from 23.9 percent in July. Analysts noted that growth continues to be driven by price rather than volume, with elevated memory chip prices playing an outsized role, but the headline number lands at an awkward moment for Chinese diplomats arguing that their country’s industrial policy poses no threat to global markets.
The longer arc is more striking still. China’s trade surplus reached a record of nearly 1.2 trillion dollars in 2025, a figure without precedent in modern economic history. Massive capacity across autos, solar panels, batteries, cement and steel has drawn complaints not only from Washington but from Brussels, New Delhi and emerging market capitals that fear their own industrial bases will be hollowed out by Chinese exports redirected from the American market. Although China’s leadership has publicly prioritized rebalancing the economy toward domestic consumption, weak internal demand keeps pushing Chinese firms outward.
That global dimension is why the administration’s overcapacity investigation named 16 economies rather than one. It is not yet clear whether determinations against the other 15 targets are imminent, and the Associated Press reporting suggests the China decision is moving on its own accelerated track, timed to the summit. Trading partners from Seoul to Brussels are watching closely, aware that the same legal machinery pointed at Beijing today can be rotated toward them tomorrow.
A statute with history
Section 301 is not a novel weapon; it is the oldest one in the modern arsenal. The statute authorized the original China tariffs of 2018 and 2019, imposed after a USTR investigation into forced technology transfer and intellectual property practices, and those duties survived every legal challenge mounted against them, including a Court of International Trade case brought by thousands of importers. The Biden administration kept the tariffs and, in 2024, raised them sharply on strategic sectors, taking electric vehicles to 100 percent, semiconductors and solar cells to 50 percent, and lithium ion batteries to 25 percent. When the Supreme Court struck down the emergency powers tariffs this February, Section 301 was the obvious foundation on which to rebuild, precisely because it had been stress tested.
What is new is the scale and speed of its use. The traditional Section 301 process, a single country investigation running a year or more, has been compressed and multiplied. The forced labor investigation covered 60 economies and moved from initiation to tariffs in roughly four months. The overcapacity investigation covers 16 economies and appears poised to produce its first determination against China in under seven months, a pace trade lawyers describe as unprecedented. Critics, including some who supported the underlying policy goals, argue that compressed timelines shortchange the notice and comment process and invite legal challenge; a July analysis by the Congressional Research Service flagged open questions about whether forced labor and overcapacity findings fit comfortably within the statute’s definition of actionable practices. New lawsuits are already accumulating at the Court of International Trade, as SCOTUSblog noted in a July survey of post ruling litigation, though courts have historically given USTR wide latitude under the statute.
The administration is betting that even if individual actions are trimmed at the margins, the architecture holds. Unlike the emergency powers tariffs, which the Supreme Court found exceeded the president’s delegated authority, Section 301 duties rest on findings, hearings and a documented record. That procedural scaffolding is slower to build, which is why the tariff landscape of 2026 has been assembled in layers rather than announced in a single Rose Garden event. But it is also far harder to knock down.
The view from Beijing
Chinese officials have spent the summer preparing domestic and international audiences for this moment. The Commerce Ministry’s white paper on the so called excess capacity issue argues that China’s cost advantages reflect scale, infrastructure and engineering talent rather than subsidies, and that the real source of global imbalances is underinvestment elsewhere. State media commentary has framed the coming tariff as evidence that Washington negotiates in bad faith, pocketing truce concessions while inventing new pretexts for pressure.
Yet Beijing’s actions have been more measured than its rhetoric. China has continued purchasing American agricultural goods under the truce framework, has kept its own retaliatory tariffs within agreed bounds, and has confirmed rather than denied the 20 percent ceiling understanding, a signal that it prefers to hold Washington to the cap rather than tear up the agreement. Chinese negotiators are also managing their own economic headwinds: a property sector still deleveraging, deflationary pressure at the producer level, and youth unemployment that makes export manufacturing jobs politically precious. A stable, capped tariff relationship with the United States, even at 20 percent, is worth more to Beijing right now than a satisfying round of escalation.
The wild card is the interaction with export controls. Beijing’s dominance in rare earths, gallium, germanium and battery chemicals gives it leverage that tariffs cannot touch, and its new restrictions on chipmaking materials, the ones Japan protested on September 8, demonstrate a willingness to use chokepoints selectively. If the overcapacity tariff lands badly in Beijing, the response is more likely to arrive through a licensing delay on critical minerals than through a counter tariff.
Stakeholders brace for the stack
For American importers, the practical concern is arithmetic. The 7.5 percent overcapacity duty would not exist in isolation. It would stack on top of the 12.5 percent forced labor tariff, the residual Section 301 duties dating to 2018 and 2019 that run as high as 100 percent on electric vehicles and 50 percent on semiconductors and solar cells, Section 232 tariffs on steel, aluminum, copper and their derivative products, and most favored nation base rates. Industry analysts calculate that cumulative duties on some Chinese semiconductor products would approach 70 percent once the new layer is added.
Retailers and consumer goods importers, who spent 2025 whipsawed by rates that spiked above 145 percent before the truce brought them down, describe the current environment as painful but plannable. A 20 percent ceiling, they note, is at least a number that can be built into sourcing models. The greater fear among trade groups is a collapse of the truce itself, which would revive the threat of triple digit rates and another round of emergency supply chain scrambles.
US manufacturers competing with Chinese imports, by contrast, have urged the administration to use every point of headroom the truce allows. Steel, solar and machinery producers argue that Chinese overcapacity suppresses global prices to levels at which no market economy producer can earn a return, and that tariffs are the only tool that changes Beijing’s calculus. Organized labor has echoed that view, particularly in autos and shipbuilding.
Farm groups sit in a third camp, anxious that any escalation invites Chinese retaliation against agricultural exports, the classic pressure point Beijing reaches for first. Soybean and pork exporters remember the 2018 and 2019 cycle vividly and have lobbied the administration to keep agriculture off the escalation ladder heading into the summit.
The economics of a 20 percent wall
Economists are divided over what the full 20 percent stack will mean. The tariffs of the past eighteen months have pushed measurable but so far contained increases into US consumer prices, with the February Section 122 levy and the July forced labor duties feeding through to goods prices over the summer. Yale Budget Lab analyses of the post Supreme Court tariff landscape have estimated that the replacement architecture, while narrower than the struck down reciprocal scheme, still represents the highest average US tariff level in generations.
The overcapacity duty adds a modest increment to that baseline, which is precisely how the administration wants it read in Beijing. But the cumulative effect on bilateral trade is no longer modest. Chinese shipments to the United States have fallen as a share of total Chinese exports every quarter since early 2025, as Beijing redirects goods toward Southeast Asia, Latin America, Africa and Europe. That redirection, in turn, feeds the overcapacity complaints of third countries and gives Washington diplomatic running room to argue its case is global rather than bilateral.
For China, the calculus involves more than tariffs. The two governments are negotiating across a broad front that includes export controls on rare earths and chipmaking materials, technology licensing, agricultural purchases, and now artificial intelligence governance. On September 8, the same day as the export data, the Associated Press reported that Japan was protesting new Chinese restrictions on exports of a key chipmaking material, a reminder that Beijing has escalation instruments of its own and is willing to use them against American allies even while courting a summit.
What happens on September 24
The Washington summit will be the first face to face meeting between the two leaders since the truce was extended, and both sides have incentives to declare success. A plausible landing zone, trade analysts suggest, involves the United States formally announcing the overcapacity tariff at or just before the summit, China responding with restraint and a reaffirmation of the truce, and both leaders unveiling cooperative deliverables on artificial intelligence safety, fentanyl precursors and possibly aircraft purchases.
The risk case is that precision engineering fails. The 20 percent cap exists in Chinese statements about American commitments, not in any published bilateral text, and the two governments have disagreed before about what was actually agreed. If Beijing treats the overcapacity tariff as a breach rather than a maximization, the truce that expires November 10 may not survive to its own deadline, let alone be renewed.
For US importers and exporters, the operating assumption should be that the 7.5 percent duty arrives this month and that the 20 percent stack becomes the durable baseline for China sourcing through at least 2027. Companies should re run landed cost models, review tariff engineering and origin planning options, and watch the November 10 truce expiration as the next structural risk date on the calendar. The summit may produce warm words and signed memoranda. The tariff wall, at its new full height, will remain.