A Chemical Nobody Had Heard Of
On 3 October, five days before EU trade commissioner Maroš Šefčovič landed in Beijing, China’s commerce ministry opened an anti-dumping investigation into p-nitrotoluene. It is a European chemical intermediate used in dyes, pharmaceuticals and pesticides. Few people in Brussels could name the firms that make it, and that was the point. The probe is small, legally routine and precisely timed. It reminds the EU, on the eve of the second EU–China Trade and Investment Council on 8–9 October, that Beijing can open a case against any European sector whenever it likes.
In June, Brussels gave China until October to deliver “tangible results.” Beijing answered before the talks even began: it expects Europe to back down. On the numbers, it is misreading the balance of costs. EU leaders, who review the outcome in Brussels on 15–16 October, should act on that.
A €1bn-a-Day Problem With a Deadline
The arithmetic driving Brussels is stark. In 2025 the EU exported €199.6bn of goods to China and imported €559.4bn, a deficit of €359.8bn, according to Eurostat data. Gunnar Wiegand, formerly the EU diplomatic service’s top Asia official, expects close to €400bn this year. The gap is still widening: Chinese exports to the EU rose 15.3% in the first eight months of 2026, while EU sales to China grew 6.2%.
Stay ahead of the geopolitical week.
MD Briefing delivers expert analysis across five global fronts — the Indo-Pacific, energy, geoeconomics, European security, and the Middle East — every Monday morning. Free.
The talks launched on 29 June carry three EU demands: curbs on subsidised overcapacity in cars, steel and batteries; fair access for European firms, including to public contracts; and reliable rare-earth supplies. The test case is cars. Chinese plug-in hybrids escaped the EU’s 2024 duties on electric vehicles, and their exports to Europe rose from about 3,800 a month in October 2024 to 50,000 in July 2026. Brussels wants limits. Beijing has refused voluntary curbs and opposes quotas.
The rare-earth clock hangs over everything. China’s sweeping October 2025 export controls are suspended only until 10 November 2026. Washington’s truce now runs to 10 January 2027, a question MD examined in August. Europe, which gets about 98% of its rare-earth imports from China, has had no such extension.
Beijing Is Betting on a Blink. The Numbers Disagree.
Read China’s moves over the past six months as one sequence and the message is coherent. In April and July it banned dual-use exports to 21 European entities in Germany, Czechia, Poland, France and the Netherlands, the second batch in explicit retaliation for EU sanctions on Russia. It has publicly rejected export restraint. It extended rare-earth relief for Washington but not for Brussels. Then it opened the p-nitrotoluene case on the eve of the talks, mirroring the EU’s own September probe into Chinese PVC. Each step is too small to justify escalation and too pointed to miss. Together they rest on one assumption: that a divided and exposed Europe will settle for process. That means a new working group, faster rare-earth licences, perhaps a minimum-price deal on a few car models.
The assumption is grounded in experience. When the EU imposed its electric-vehicle duties in 2024, China hit European brandy and pork, and Brussels drifted toward price deals rather than hard caps. Even now, Spain and Slovakia remain the main holdouts against tougher action, and Spain is heading into an election with its pork exporters’ memories intact.
The conventional view is that a trade war is “lose-lose” and that Europe, the more exposed party, loses more. That view misreads where the costs fall.
Start with China. Its economy now runs on exports. In the first eight months of 2026, property investment fell 19.9% and infrastructure investment 7.2%, while industrial output rose 5.2%, according to GMF’s analysis. That gap between what China makes and what it buys at home has to go somewhere. With US tariffs walling off much of the American market, Europe is where it lands, which is why Chinese sales to the EU grow at 15% a year. A quota on hybrids, safeguards on chemicals and machinery (temporary import limits the EU can impose within months) and a fast-acting retaliation tool would hit exactly the growth Beijing relies on. No other market of Europe’s size can absorb €560bn of goods. Pushing them into emerging economies invites the same defences there: G20 finance leaders, minus China, agreed in September to act against “non-market” distortions.
Now Europe. Its exposure is real but concentrated. Of €199.6bn in exports to China, machinery accounts for €45.3bn, electrical equipment €29bn, vehicles €16.4bn and pharmaceuticals €13.6bn. Retaliation would hurt German machine-builders and carmakers, and French and Spanish farmers. But those sales are already shrinking: EU exports to China fell 6.5% last year. The cost of inaction is larger and spreads further. Wiegand warns of deindustrialisation and job losses, with Germany and the Czech Republic most exposed. The Commission has already opened 30 trade-defence investigations in a year, nearly triple the usual rate. Brussels has stopped pretending the problem will fix itself.
The strongest objection is rare earths. With 98% of its imports coming from China, a full cutoff would reach European car and defence assembly lines within weeks. China’s April 2025 controls showed how fast magnet shortages travel through a supply chain. This is Beijing’s real lever, and it knows it.
But backing down does not disarm that lever. The suspension expires on 10 November whatever the EU decides this month. China extended relief to Washington because Washington held tariffs it could trade, not because Washington was conciliatory. A Europe that reaches November with nothing on the table has nothing to exchange. Rare earths are also a one-shot weapon. A cutoff would confirm every European fear and turbocharge the diversification Brussels is already building, including a proposed rule requiring firms in critical sectors to keep at least three suppliers. MD has tracked how far that effort has to go, but a cutoff would end the debate about whether to fund it. Leverage like this works best unused.
So the real choice facing EU leaders is not between a deal and a trade war. It is between a credible tool now, while the rare-earth clock is running, and a weaker hand in November.
Three Ways the Next Five Weeks Go
Base case: managed escalation (roughly 55%). The Beijing talks end on Friday with process, not limits: faster rare-earth licensing for EU firms and continued dialogue on cars, but no export cap. On 15–16 October, leaders endorse the Franco-German rapid-response instrument in principle and back safeguards on hybrids and chemicals. The Commission’s first measures land by the end of October, with a broader toolkit in December. China answers with more calibrated probes, most likely against European farm goods, and extends the rare-earth suspension for Europe on a short leash. The key assumption is that both sides prefer pain they can control.
Downside: the November cliff (roughly 25%). The EU imposes a hybrid quota, and Beijing lets the 10 November suspension lapse for Europe alone while Washington’s truce runs to January. Licences slow, and by December magnet inventories at carmakers and defence suppliers, some already under China’s dual-use bans, run thin. Pressure from German industry and Spain splits the Council, and the EU retreats into minimum-price deals. That is exactly the outcome Beijing is betting on. Firms with magnet-heavy supply chains should stress-test their first quarter of 2027 now, not after the summit.
Upside: a proof of concept (roughly 20%). Beijing worries that EU quotas could become a template for other markets, and it wants Europe quiet while its US truce runs. It accepts a car-only arrangement: volume or minimum-price commitments on plug-in hybrids, modelled on the voluntary export restraints Japan accepted from Europe in the 1980s. The deal would be small, but it would be the first time China has accepted managed trade with Europe. That would change the terms of every negotiation that follows.
The Price of Blinking
China’s timing tells us what it expects: that Europe will treat a chemical probe and a hanging rare-earth deadline as reasons to settle for dialogue. The numbers point the other way. Beijing needs the European market more than ever because it has fewer places to put its surplus. Europe’s biggest vulnerability, rare earths, is not removed by retreat; it is only postponed to 10 November.
Watch the European Council conclusions on 15–16 October. If leaders give the Commission a rapid-response instrument with a date attached, Beijing’s bet has failed. If the text speaks only of “continued engagement,” China will have learned that a probe into a chemical few Europeans had heard of is enough to move a bloc of 450 million people. It will not need to learn that lesson twice.
