Taiwan’s export map has shifted sharply. On a Chinese-mainland-only basis, excluding Hong Kong, exports to the U.S. overtook exports to the mainland in 2024. The shift appears mainly driven by high-end electronics, chips, AI servers and related supply-chain demand. Exports to the Chinese mainland still grew by a healthy 16%, so this is more a story of a U.S. surge than of collapsing mainland demand.
The 2026 figure should be read only as a projection based on recent quarterly trends. It points to a much wider gap, but the situation remains highly fluid. Technology-export rules, U.S.–China negotiations, the possible Trump-Xi meeting, the current tariff truce and wider political noise could all affect the trajectory before year-end.
Germany gets most of the attention, but Bratislava deserves a look too. Slovakia is one of Europe’s deepest car hubs, and China had become one of its biggest export markets. More recently, Bratislava posted the weakest employment outlook in Slovakia for Q1 2026, which is hardly surprising given the scale of the fall in car exports to China. It is equally telling that Slovakia’s investment agency travelled to China in March to pitch the country as a gateway for Chinese automotive investment into Europe.
China’s share of Slovakia’s car exports fell to 4.5% in 2025, from 8.4% in 2022. The brighter news is that Slovak car exports to the UK are thriving, up 86% since Brexit.
Before Trump’s second term, EU passenger cars entering the U.S. faced a tariff of 2.5%. Trump then added a 25% auto tariff, bringing the effective rate to 27.5%. The 2025 EU-U.S. framework reduced that burden to 15%. Trump now says he wants to raise the tariff on EU cars to 25%.
For Europe’s carmakers, this would be a second front. They are already losing ground in China. A renewed U.S. tariff shock would put pressure on the other major pillar of their export business.
In 2025, the EU still exported about €31bn worth of cars to the U.S., split between roughly €20bn in combustion vehicles and €11bn in electric and hybrid vehicles. That is below the 2023 peak, but still a very large export flow. Trump’s new 25% tariff threat therefore targets a sizeable EU business. His stated aim is also to push more EU carmakers to produce in the U.S., although several already do.
The drop in EU auto exports to the U.S. is broad-based. One detail stands out: the category that has fallen least, and has therefore resisted Trump’s tariffs better, is petrol cars with large engines, above 3,000 cc. In 2025, the EU still exported more than €5bn worth of these vehicles to the U.S.
The big jump in 2023 and 2024 does not mean controls disappeared. It more likely reflects a rush by Chinese chipmakers to buy what was still allowed: especially lithography machines, the ASML equipment that uses ultraviolet light to print tiny circuit patterns onto silicon wafers. The most advanced machines were already restricted, but China could still buy some less advanced, though still highly sophisticated, tools before the rules tightened further.
The U.S. imposed major semiconductor-manufacturing controls in October 2022, tightened them again in October 2023, and added further restrictions in December 2024. The Netherlands followed with its own licensing requirement from September 2023, broadened it in September 2024, and tightened it again from April 2025.
Our reading is that 2026 will bring fresh pressure on lithography machinery, together with wider efforts to close servicing loopholes for Chinese chipmakers.
Despite the talk of China+1 and China+2, EU smartphone imports remain heavily China-centred. India and Vietnam matter, and India’s 2024 jump is visible, but China still supplied about two-thirds of EU shipments in 2025. The data show some diversification at the edge, not a decisive shift away from China. In 2025, EU smartphone imports were worth nearly €40bn.
China defines rare earths as state-owned resources and restricts mining, smelting and separation to state-approved companies. It is now doing something more bureaucratic, but still worth noting: turning regulations into enforceable penalty practice.
The new draft does not change the strategic direction. It makes the system easier to enforce. That matters because China’s rare-earth policy is not only about export licences at the border.
Control starts upstream: who may mine, who may process, and how violations are punished. Often, that upstream control matters as much as, or even more than, export controls themselves.
EU exports of agri-food and beverages to China rose sharply after 2018 and peaked in 2020, just as China’s African swine fever (ASF) crisis created exceptional demand for imported pork. As China rebuilt its pig herd and domestic supply recovered, that temporary export windfall faded. By 2025, EU agri-food and beverages exports to China were about one-third below their peak five years earlier. Our estimate is that, in 2026, they will be back around their pre-ASF level.
The first chart shows China’s Intellectual Property (IP) receipts as a share of U.S. receipts. On that measure, China has clearly become more visible. In 2015 and 2016, China’s receipts were only about 1% of the U.S. level. By 2025, they had risen to around 7%, after briefly reaching about 9% in 2022. That is not convergence, but it is not trivial either. China has moved from near invisibility to a small but measurable position.
But ratios can flatter the story, so the second chart is the necessary reality check. In absolute terms, the gap remains enormous. China’s rise over the decade is real and important, but U.S. IP receipts rose from about US$111bn in 2015 to US$189bn in 2025, with particularly strong gains again in 2024 and 2025.
While China has improved its relative position, the U.S. lead in absolute dollars remains overwhelming.
In Q1, China reduced its services trade deficit by 17% compared with the same period last year. Most, if not all, of the improvement came from travel services: foreign visitors spent more in China, while Chinese travellers spent less abroad.
China’s outbound travel spending has largely recovered from the pandemic shock, but Q1 2026 does not show fresh momentum. Spending fell from about US$70bn in Q1 2025 to US$63bn in Q1 2026, despite longer holidays. That still leaves outbound spending close to pre-pandemic levels, but the direction is notable: more holiday time did not translate into higher overseas travel spending.
This contrasts sharply with inbound travel to China. Spending by foreign visitors in China increased by 36% in Q1 2026 compared with the same period in 2025. As a result, China reduced its travel-services deficit by about US$11bn in Q1 compared with Q1 2025.
The PMI employment subindex is not a job-count measure. It is a monthly signal from firms on whether they are hiring, holding steady, or cutting staff. A reading below 50 means employment conditions are weakening. The subindex improved slightly in April but remained in contraction territory.
We have been looking at several indicators to assess the impact of the current Strait of Hormuz situation as of April 27. But the data we have used can only offer hints. It is backward-looking and present-diagnostic, not predictive.
Bunker fuel prices are a good example. The chart helps us read the recent shock, not predict the next move. The 7-day average shows that bunker fuel prices have come down from their March peak. The 21-day average shows that the shock was large enough to keep average prices elevated into April.
But the chart itself does not tell us what comes next. It tells us where the pressure has been, and how much of it is still visible in recent prices.
Three years ago, we moved our weekly SOAPBOX newsletter to Substack. Since then, we have found a broad and engaged community of readers, with far less noise than seems to be the standard on many other social tools.
We want to thank everyone reading, sharing, commenting, correcting and encouraging.