Washington and Beijing move to unwind duties on roughly $60 billion of two-way trade, but the carve-outs tell importers more than the headline number does
WASHINGTON, Oct. 2, 2026. The United States and China have agreed to roll back tariffs on approximately $60 billion of bilateral trade under a reciprocal framework that trade officials on both sides are describing as a confidence-building step rather than a settlement, leaving the architecture of the four-year tariff war almost entirely intact.
Under the arrangement announced following a presidential summit in Washington between President Donald Trump and President Xi Jinping, each government identified roughly $30 billion of what both sides characterized as non-sensitive goods for more favorable tariff treatment, measured against 2024 trade values. The Office of the United States Trade Representative released the American list late on Sept. 28, and China’s Ministry of Commerce published its corresponding schedule the same weekend. Implementation on both sides remains pending the completion of domestic legal procedures, and neither government has published effective dates or the precise rate reductions that will apply line by line.
For importers and exporters trying to plan fourth-quarter and first-quarter sourcing, that gap between announcement and implementation is the central operational fact. The political signal is clear. The entry-level guidance is not.
What the two lists actually cover
The asymmetry between the two schedules is striking and it says a good deal about the structure of the relationship.
The American schedule covers 77 tariff categories of Chinese-origin goods. The list is dominated by consumer household merchandise: bed linens and blankets, tableware, small kitchen appliances including coffee makers and toasters, garden umbrellas, artificial flowers, fireworks, holiday decorations, billiard balls and children’s car seats. Toys are the single largest category by value, accounting for roughly $14.4 billion of 2024 imports on their own, according to trade data cited in reporting on the agreement.
The Chinese schedule, by contrast, runs to 1,619 tariff categories. It is weighted toward American agriculture and primary products: corn, wheat, sorghum, meat, dairy, vegetable oils, fish and seafood, logs and wood products, together with cosmetics, medical devices, scientific instruments and a long tail of specialty items that includes live horses, donkeys, whales and dolphins. China’s Commerce Ministry said that for roughly 90 percent of the listed products, duties would revert to most-favored-nation baseline rates.
The arithmetic of those two lists is the story. The American concessions touch approximately 7 percent of Chinese exports to the United States. The Chinese concessions touch approximately 21 percent of American exports to China. USTR Jamieson Greer framed the latter figure generously, saying the deal was “unlocking improved market access” for close to 30 percent of American exports destined for the Chinese market, and that the administration would continue to pursue “fair, balanced, and reciprocal trade with China” while holding Beijing to its commitments.
Both statements can be true at once, and both reflect the same underlying imbalance: the United States buys far more from China than it sells, so a dollar-for-dollar swap necessarily covers a much larger share of the smaller flow.
The soybean problem
The most consequential item on the Chinese list is the one that is not on it.
Soybeans, historically the largest single American agricultural export to China, were excluded from the tariff relief. Beijing’s 10 percent retaliatory duty on American soybeans remains in force. That omission is not a drafting oversight. Chinese crushers entered the autumn with soybean inventories at roughly fifteen-year highs and have spent two seasons rebuilding supply relationships with Brazilian and Argentine exporters, who now hold structural advantages in both price and harvest timing that American growers cannot easily claw back.
What the agreement offers American agriculture instead is a volume commitment. The American Soybean Association said China has committed to purchasing 25 million metric tons annually, a figure association president Scott Metzger said “provides critical stability” for growers planning acreage decisions. The broader package is also understood to support an agricultural purchase pledge in the region of $17 billion.
The distinction between tariff relief and purchase commitments matters enormously for exporters. A tariff reduction is a durable change in the landed cost of a product and it benefits every seller. A purchase commitment is a political undertaking that can be met through state trading enterprises, that may concentrate benefits among a handful of large shippers, and that historically has been met partially or not at all. American farm groups have lived through both versions and read the difference carefully.
Energy illustrates the same pattern. China agreed to import 10 million metric tons of American coal annually across 2027 and 2028, a commitment of roughly 20 million tons in total. Liquefied natural gas and crude oil, which carry far larger dollar values and far more political weight, were left out entirely.
What was left untouched
The agreement is bounded in ways that should discipline expectations.
Semiconductors are not covered. Electric vehicles are not covered. Neither are the export control regimes on advanced computing hardware, the outbound investment screening rules, the entity listings, the Section 301 architecture that underpins much of the duty structure on Chinese industrial goods, or the critical minerals and rare earth licensing disputes that have shadowed the relationship through 2026. One analyst characterized the package as “less the demolition of a tariff wall than the opening of a safety valve,” a description that matches the evidence.
Alongside the tariff schedules, the two governments agreed to establish communication channels for artificial intelligence incidents and to open approval pathways for American financial services firms operating in China. Those are meaningful in their own right, particularly for the financial sector, but they do not change the cost basis of a single container.
The trade truce itself was extended by two months, to Jan. 10, 2027. That extension is arguably the most important near-term provision in the entire package, because it postpones the snapback of a much larger tariff structure that would otherwise have landed in the middle of the first-quarter buying season.
Where the tariff burden actually stands
The scale of what remains can be measured. Modelled estimates from Global Trade Alert’s United States Tariff Estimates series, which computes applied rates at the eight-digit tariff line level, put the trade-weighted average American tariff on Chinese-origin goods at 27.7 percent as of Oct. 2, across roughly $434 billion of covered trade flows. The simple average across tariff lines is considerably higher at 36.5 percent, and the maximum applied rate on a single line reaches 151.8 percent.
The distribution is more revealing than the average. Of the 8,848 flows covered in that estimate, 56.3 percent sit in the 30 to 50 percent band and a further 11.5 percent sit above 50 percent. Only 1.2 percent of flows carry a zero rate, and those account for 12.6 percent of trade value, reflecting a small number of high-value exempted categories. Roughly a quarter of trade value faces duties between 30 and 50 percent, and another 14.5 percent faces duties above 50 percent.
Against that backdrop, a $30 billion reduction in covered American imports, concentrated in consumer housewares and seasonal goods, moves the aggregate very little. It will, however, move specific lines a great deal, and for the firms sitting on those lines the effect is not marginal at all. An importer of holiday decorations or bed linens whose landed cost has been carrying an additional thirty or forty points of duty is looking at a materially different margin structure for the 2027 season, assuming the relief lands in time to affect orders that are being placed now.
Stakeholder reaction
Reaction has split along predictable lines, with the interesting variation coming within sectors rather than between them.
Agricultural groups welcomed the Chinese schedule while flagging the soybean exclusion as a serious gap. The volume commitment on soybeans gives growers something to plan against, but it leaves the tariff differential against Brazilian and Argentine supply in place, which is the variable that determines whether Chinese crushers buy American beans when they are not under political instruction to do so.
Consumer goods importers and retail trade associations have been the most visibly relieved. The American list reads almost as though it were drafted from a retail buying calendar, and for firms that have spent three years re-engineering sourcing out of China at considerable cost, it raises an uncomfortable strategic question: whether the relief is durable enough to justify slowing or reversing diversification programs that are only partially complete. The honest answer, with a truce that expires on Jan. 10, 2027, is almost certainly no.
Manufacturers in sectors left off both lists have been quiet, which is itself informative. Steel, aluminum, copper, semiconductors, autos, industrial machinery and pharmaceuticals remain governed by Section 232 and Section 301 regimes that this agreement does not touch.
Customs brokers and trade compliance professionals, meanwhile, are focused on a narrower and more immediate issue: the absence of implementation guidance. Until a Federal Register notice or a CBP CSMS message specifies effective dates and the tariff subheadings affected, no entry can be filed on the new basis, and no importer can responsibly tell a buyer what a first-quarter landed cost will be.
Economic impact
The macroeconomic effect of the package, taken alone, is modest. Sixty billion dollars of covered trade against a bilateral goods relationship that has run well north of $500 billion in recent years, with relief concentrated in low-margin consumer categories, does not shift aggregate price levels or trade balances in any way that will be visible in quarterly data.
The sectoral effects are a different matter. In categories where duties have been carrying thirty to fifty points, a reversion toward most-favored-nation treatment can swing gross margin by several hundred basis points. In toys specifically, where the import value is large and the retail price points are sensitive, the pass-through to shelf prices could be noticeable by the second half of 2027 if the relief holds and if it is implemented in time to affect the production and shipping cycle for the next holiday season. That second condition is doing a lot of work. Orders for holiday 2027 are placed in the first and second quarters of 2027, which means the implementation timetable, still unpublished, determines whether the tariff cut reaches consumers next year or the year after.
For American exporters, the agricultural and wood products lines are the most immediately actionable. A reversion to most-favored-nation rates on corn, wheat, sorghum, meat and dairy restores price competitiveness against South American and Australasian suppliers in a market where American share has eroded since 2022. Whether that translates into volume depends on Chinese demand conditions and on purchasing decisions that remain substantially administered rather than commercial.
The more significant economic variable is the truce extension itself. Businesses on both sides have been operating under the possibility of a tariff snapback that would have raised rates well above current levels. Removing that risk from the fourth-quarter calendar has a real option value, even though it never appears as a line item.
What importers and exporters should do now
For American importers of Chinese-origin goods, the immediate task is classification work, not celebration. The relief is defined at the tariff category level, and whether a given stock keeping unit falls inside or outside a covered category will turn on classification decisions that many importers have not revisited since the duties were imposed. Firms should map their Chinese-origin bill of materials against the published USTR list now, flag the lines that appear covered, and confirm the eight-digit or ten-digit classification against the actual product rather than against a legacy entry summary.
Second, importers should resist the temptation to reverse diversification programs. The truce expires on Jan. 10, 2027. The relief covers categories that both governments explicitly designated as non-sensitive, which is another way of saying they were selected precisely because they carry little leverage. Nothing in the structure of the agreement suggests that the sensitive categories, where the real duty burden sits, are moving.
Third, importers with entries filed under rates that may be revised should preserve their protest and post-summary correction options. Until implementation guidance specifies whether any relief applies retroactively, and the default assumption should be that it does not, entries continue to be liquidated at current rates.
For American exporters to China, the priority is confirming that Chinese customs treatment has actually changed before pricing against it. Chinese implementation also awaits domestic legal procedures, and the Commerce Ministry schedule is a statement of intent rather than an operative tariff instrument. Exporters should ask their Chinese consignees to confirm the applied rate on a test entry before committing to forward pricing.
Fourth, for agricultural exporters specifically, the distinction between the tariff schedule and the volume commitments should drive planning. The tariff changes are usable by anyone. The volume commitments are likely to be executed through channels that concentrate benefits, and growers should not assume that a national purchase figure translates proportionally into their own order book.
The calendar ahead
Two dates dominate. The first is whenever implementation guidance appears, which both governments have left open and which determines when any of this becomes operational. The second is Jan. 10, 2027, when the extended truce expires and the full negotiation reopens.
Between now and then, the structural disputes remain exactly where they were: technology controls, industrial subsidies, semiconductor access, critical minerals licensing and the Section 301 framework. The $60 billion package addresses none of them. It was not designed to.
What it does accomplish is to demonstrate that the two governments can execute a reciprocal, documented, line-by-line de-escalation on a defined scope. For a relationship that has spent four years ratcheting in one direction, that is not nothing. It is also, as the analyst quoted above put it, a safety valve rather than a demolition, and importers who treat it as the latter will be exposed when the January deadline arrives.
Background: how the wall was built
The structure the two governments are now chipping at was assembled in layers, and understanding those layers explains why a $60 billion package leaves the aggregate burden close to where it was.
The oldest layer is the Section 301 action on Chinese technology transfer and intellectual property practices, which dates to 2018 and still governs duties on a wide span of Chinese industrial and consumer goods. Successive reviews have adjusted coverage and rates, and an exclusion process has opened and closed repeatedly, but the underlying action has never been withdrawn.
Layered on top of that are the Section 232 national security actions on steel, aluminum, copper and derivative articles, which apply by product rather than by country and therefore capture Chinese-origin goods alongside everyone else’s. Those regimes were restructured again in April 2026, and the derivative product lists have expanded steadily, pulling in finished goods whose metal content is a small fraction of their value.
Above both sits the emergency tariff architecture that defined 2025 and much of 2026, which the Supreme Court substantially reshaped in its ruling on the International Emergency Economic Powers Act earlier this year. That decision forced a reconstruction of the legal basis for a large share of the duty structure and triggered a refund process that importers and Customs are still working through. The practical result was not a lower tariff wall but a differently founded one, rebuilt on authorities that are harder to challenge.
Finally there is the trade remedy layer, the antidumping and countervailing duty orders that accumulate case by case and that operate independently of any presidential action or bilateral negotiation. Those orders do not appear in summit communiques and they are not traded away in packages like this one, but they land on specific importers with great force.
The $60 billion package touches the first layer lightly, in categories both sides agreed were expendable. It does not touch the others at all. Any importer reading the announcement as the beginning of a general unwinding should map their own exposure across all four layers before concluding that relief is coming.
Scenarios through January
Three paths are worth planning against.
In the first, implementation guidance appears within weeks, the covered lines revert to most-favored-nation treatment in time to affect first-quarter ordering, and the January truce deadline is extended again as part of a broader negotiation. This is the path both governments appear to be steering toward, and it would make the package genuinely useful to the consumer goods sector.
In the second, implementation slips. Legal procedures on both sides take longer than expected, guidance lands in December or January, and the relief arrives too late to influence the 2027 holiday production cycle. The tariff reduction still shows up in landed costs eventually, but a full season later than the announcement implies. Importers who priced forward on the assumption of early relief would be exposed.
In the third, the January deadline passes without an extension. The truce lapses, the broader tariff structure snaps back, and the $60 billion package becomes a historical footnote to a renewed escalation. Nothing in the current posture of either government suggests this is the likely outcome, but it is the scenario that determines how much diversification spending should be paused on the strength of this announcement, and the answer is very little.
Firms with meaningful Chinese-origin exposure should be building their fourth-quarter plans against the second scenario and their capital allocation decisions against the third.