NextFin News - The US goods deficit with China shrank to $15.03 billion in September, down roughly $2 billion from the prior month, and China has slipped to fourth place among America's largest goods-deficit partners. Yet China's trade surplus hit a record of nearly $1.2 trillion in 2025, and its share of global exports kept climbing even as US tariffs on Chinese goods averaged 47.5 percent. The contradiction is the story: tariffs are winning the bilateral battle while losing the structural war. The problem is not a price gap that a tariff can close; it is an economic model that turns domestic savings into surplus production and ships the difference abroad.
The Trump-Xi summit at the White House this week produced a truce extension to January 10, not a settlement. Treasury Secretary Scott Bessent and China's Vice Premier He Lifeng met in New York ahead of the leaders' meeting, and Trade Representative Jamieson Greer said outcome details would follow. But the deeper issue survives any temporary ceasefire. As long as China saves roughly 43 percent of GDP, consumes a historically low share of national income, and directs credit into factories that the home market cannot absorb, the surplus will find a way out — through Vietnam, through Mexico, through Southeast Asia — with or without tariff relief.
The Numbers Say Tariffs Are Working. The Deficit Says Otherwise.
On the surface, the tariff campaign has delivered exactly what its architects promised. The US goods deficit with China fell to $15.03 billion in September, and China now ranks behind Ireland, Mexico, and Vietnam among the countries with which the US runs the largest goods gaps. The share of Chinese-origin products in total US imports has collapsed from about 21 percent in 2018 to roughly 8 percent in the second quarter of 2025. Bilateral trade between the two economies totaled an estimated $494.6 billion in 2025, down 25.1 percent from 2024. Chinese exports to the US fell 10.4 percent year on year in April alone, according to Chinese customs data.
But the overall US trade deficit tells a different story. The US goods-and-services deficit widened to $88.6 billion in July 2026, up from a revised $71.2 billion in June. Mexico now runs a goods deficit with the US of $17.38 billion, Vietnam $16.6 billion, and Ireland $18.22 billion — each larger than the gap with China. Vietnam's exports to the US surged 53 percent year on year to more than $20 billion in January 2026 alone, even as US imports from China fell 46 percent. Mexico's imports from Vietnam more than doubled in the first half of 2026, a pattern that has raised transshipment suspicions among Mexican officials.
This is not leakage. It is geometry. A tariff is a wall built around one door; the surplus is water, and it flows to the lowest point. Capital Economics estimates that only about one-eighth of the decline in China's direct exports to the US has been offset by rerouting through third countries so far — but that offset was a third or more during the first trade war, and supply chains adjust with a lag. Southeast Asia's share of US inbound container bookings has already climbed from 20 percent to 29 percent over two years. The four main Southeast Asian lanes — Vietnam, Malaysia, Thailand, Indonesia — were up 28.3 percent year to date in 2026, with the most recent nine-week pace running 34 percent above the comparable 2025 period.
The arithmetic is unforgiving: shrinking the deficit with China while the overall deficit grows means the deficit has not disappeared. It has changed its passport.
Why the Surplus Is a Feature of the Model, Not a Bug
The second question is why China keeps exporting more even when its biggest customer raises the price of admission. The answer sits in the structure of the Chinese economy, not in the tariff schedule.
China saves about 43.24 percent of GDP — one of the highest rates among major economies. Households save because the social safety net is thin: healthcare, education, and elder care costs are largely private, and the property market, which once served as the household balance sheet, remains in contraction. With domestic consumption restrained, factories built with state-directed credit produce more than Chinese households can buy. The excess has to go somewhere. It goes abroad, at prices that undercut competitors.
The price evidence is stark. Chinese goods export prices have fallen by roughly 25 percent since 2022, while export prices elsewhere have been broadly flat. Chinese firms are selling more for less — a strategy that has kept China's share of global export volumes rising from about 13 percent in 2018 to roughly 18 percent at the start of this year, holding steady even through the tariff escalation of the second Trump term. The cost of that strategy is visible at home: almost one in three Chinese manufacturers is now loss-making, and industrial profit margins have fallen from 7.8 percent in 2017 to 4.3 percent in 2024, recovering only slightly to 4.4 percent in 2025.
"As the second largest economy in the world, China is simply too big to generate much growth from exports," said Kristalina Georgieva, Managing Director of the International Monetary Fund, speaking in Beijing. The IMF has warned that China's manufacturing dominance risks exacerbating global trade tensions and has urged Beijing to shift toward domestic consumption.
Beijing knows this. The Politburo's "anti-involution" campaign, launched to curb the vicious price competition that has eroded profits across sectors from electric vehicles to solar panels, is an admission that the old model is eating itself. Export tax rebates on photovoltaic products are being phased out, and a licensing system with quality standards now restricts certain steel exports. But the response is calibrated to avoid a jobs shock, not to rewire the model. T. Rowe Price does not expect broad-based capacity cuts on the scale of the 2015 supply-side reforms; the approach is gradual and market-oriented. Roughly 50 unprofitable mainland Chinese EV makers are expected to scale back operations in 2026 — a consolidation, not a transformation.
The International Monetary Fund estimates that social-spending reforms and an overhaul of the hukou household-registration system could lift consumption by up to 3 percent of GDP, while resolving the property crisis could cost around 5 percent of GDP. Reliance on state support and industrial policy is already curbing productivity growth by as much as 1.2 percent. These are not small adjustments. They require transferring income away from the state and corporate sector toward households — a redistribution of power that the current political economy resists.
The Second-Order Effect: Tariffs Tax the Wrong Problem
Here is the second-order consequence that the tariff debate misses. A tariff on Chinese goods is designed to change the price signal. But the Chinese export machine is not responding to price signals — it is responding to a structural need to externalize overcapacity. When the price of selling to the US rises, the rational response for a firm that must keep producing is not to cut output; it is to cut the route. That is why transshipment enforcement has become the newest front: the US has imposed punitive tariffs of 40 percent on goods deemed transshipped through Vietnam, with similar measures aimed at Cambodia at 49 percent and Thailand at 36 percent. US Customs is deploying AI-powered origin-verification systems. Treasury officials describe a truce extended to January 10 while negotiations continue.
But enforcement is a game of whack-a-mole against a supply chain that has spent two decades optimizing for exactly this kind of arbitrage. Every origin check creates an incentive to add one more intermediate stop. Every bilateral deal with a third country pushes the assembly step one jurisdiction further away. The USMCA agreement is up for renegotiation, and Mexico — caught between placating Washington and preserving its own export platform — has begun imposing its own tariffs on Chinese goods. The result is not less Chinese content in US-bound goods; it is more Chinese content with more paperwork.
The inflationary logic of tariffs also runs into the deflationary logic of Chinese overcapacity. Tariffs are meant to raise import prices and pull production home. But Chinese exporters, competing against each other in a saturated domestic market, have been cutting prices by 25 percent. They can absorb part of the tariff and still undercut. The party that pays is not only the American consumer; it is also the Chinese manufacturer already operating on thin or negative margins. The question is which side runs out of cash first — and with one in three Chinese manufacturers loss-making, the strain is visible.
Yet this pressure does not automatically produce rebalancing. A loss-making firm in China does not exit the way it would in a market economy. Local governments protect employment. Banks roll over loans. The central government tolerates low profitability to preserve social stability. That is why the "anti-involution" campaign stops short of the creative destruction that would actually clear capacity. The surplus persists because the mechanism that would eliminate it — firm exit — is politically constrained.
The Strongest Case for Tariffs — and Why It Falls Short
The most serious defense of the tariff strategy is not that it will eliminate the trade deficit. It is that it achieves something narrower and, to its proponents, more important: decoupling. The numbers support this read. China's share of US imports has fallen from 21 percent to 8 percent. Bilateral trade is down a quarter. Critical supply chains have moved. National-security hawks argue that reducing dependence on Chinese manufacturing in strategic sectors is worth the economic cost, and that transshipment enforcement will steadily close the loopholes as origin-verification technology improves.
This argument is coherent, and it is the one that policymakers are actually making. But it concedes the central point: the trade problem is being "solved" by shrinking the relationship, not by fixing the imbalance. The global surplus remains; it is simply intermediated through more countries. And the decoupling argument does not answer the inflation-and-supply-chain question that matters to markets: if Chinese overcapacity is not absorbed by Chinese consumers, and the US and Europe keep raising walls, where does the output go? The answer so far has been emerging markets — Africa, Southeast Asia, Latin America — at falling prices. That is a growth model that exports deflation as much as goods, and it is not stable indefinitely.
There is also the question of whether Beijing is already doing enough. The anti-involution campaign, the export-rebate cuts, the two-year output-growth targets for ten key industries that are lower than 2024's — these are real policy moves, and supporters of the tariff approach point to them as evidence that pressure works. The counter is that these measures manage symptoms. They do not touch the savings rate, the hukou system, or the flow of credit to state-favored industries. Until they do, capacity will rebuild whenever prices recover.
What to Watch: The Signals That Would Change the Call
The base case is that the surplus persists above $1 trillion annually through 2026, even as the bilateral deficit with the US stays small. Natixis forecasts Chinese export growth of about 3 percent in 2026, down from 5.5 percent in 2025, with the surplus remaining above $1 trillion. BNP Paribas expects exports to remain "a big growth driver" in 2026. The IMF projects China's GDP growth slowing to around 4.5 percent in 2026 from roughly 5 percent in 2025.
Three signals would falsify this structural-surplus thesis. First, if China's household consumption share of GDP rises by three or more percentage points within two years — the IMF's own estimate of what hukou and social-spending reform could deliver — the model would be genuinely rebalancing, and the surplus would shrink at the source. Second, if the annual trade surplus falls below $600 billion while Chinese export prices stabilize or recover, that would show overcapacity is clearing rather than migrating. Third, if the US monthly goods deficit with China stays below $20 billion for four consecutive months while the deficits with Vietnam, Mexico, and ASEAN stop rising, the rerouting thesis would be weaker than the decoupling thesis.
None of these is likely in the near term. The truce extension to January 10 kicks the hard decisions into an election year. The anti-involution campaign is gradual by design. And the US deficit is a macroeconomic identity — the gap between national saving and national investment — that no tariff schedule can rewrite.
The short-term read is that markets should expect continued volatility around trade headlines, with any escalation hitting risk assets and any truce extension providing relief. The medium-term read is that Chinese exporters will keep absorbing tariff costs through lower margins, pressuring profitability but not volumes. The long-term read is that the only durable fix is a Chinese consumption boom — and that requires a political choice Beijing has not made.
Tariffs can shrink a bilateral deficit. They cannot shrink a savings rate. Until China spends more of what it earns, the world will keep buying what China makes — just through more middlemen, at lower prices, and with a trade war that never quite ends.
Explore more exclusive insights at nextfin.ai.
