On 17 July 2026, at Augustusburg near Cologne, Chancellor Merz and President Macron raised the alarm over the widening EU–China trade gap and tasked their economy, finance and foreign ministers with a joint roadmap by September. Merz stressed dialogue over overcapacity, explicitly not a trade war; Macron pressed for faster, emergency-grade Commission action. The desk's read is that this crosses a threshold. What began as a single electric-vehicle anti-subsidy case has broadened, on the Commission's own findings, into a systemic problem — Chinese state support and overcapacity across EVs, solar, wind-turbine components, advanced and mature-node semiconductors, steel and batteries. The EU's 2025 goods deficit with China was about €360 billion. Brussels is assembling a defensive architecture in response: tighter steel safeguards, a possible new "overcapacity" instrument, expanded carbon-border tariffs, fresh state-aid authorisations, local-content procurement rules and closer trade-defence coordination with Washington. Most of this is proposed or partly enacted, not fully in force, and this piece labels each accordingly. The structural question the map poses is not whether Beijing has the capacity — it plainly does — but whether twenty-seven governments with divergent China exposures can hold a common line long enough for any of it to bite.
Start with the balance sheet, not the summit photo. In 2025 the European Union ran a goods trade deficit with China of roughly €360 billion — a figure Paris and Berlin now render, for effect, as about €1 billion a day. That gap is not a rounding error in a healthy relationship; it is the arithmetic of a manufacturing base being outproduced. And the pattern behind it is no longer confined to one product. China accounts for close to 30 percent of global manufacturing output while consuming perhaps 13 percent of it — the definition of an economy whose surplus must be exported or absorbed by someone else. Europe is discovering it is the someone else.
1. What the 17 July meeting actually signalled
The facts of the meeting are firm. On 17 July 2026, Merz and Macron held a joint Franco-German cabinet meeting at Augustusburg castle near Cologne and used it to raise the alarm over the trade imbalance with China, agreeing to draw up a common "roadmap" by September and tasking their economy, finance and foreign ministers to write it. Merz stressed dialogue over industrial overcapacity and said explicitly that he does not seek a trade war; Macron took the sharper tone, urging the European Commission to respond faster — with emergency measures rather than years-long investigations — when Chinese overcapacity threatens European industry. Established
Read that split tone carefully, because it is the whole story in miniature. The two governments closest to the problem, and most aligned in naming it, still stood a visible distance apart on what to do — Berlin protective of an export economy that sells cars and machine tools into China and fears retaliation, Paris readier to reach for hard defensive tools. If the Franco-German motor cannot fully align, the twenty-five other members — from a Hungary courting Chinese battery plants to a Spain repairing relations with Beijing — will align less. The alarm is genuine and shared. The remedy is neither. Assessed
2. The threshold: from a case to a category
For two years the European response to Chinese industrial policy had a shape: the discrete anti-subsidy investigation. The electric-vehicle probe was its emblem — a formal injury inquiry that ended, in late 2024, with countervailing duties layered on top of the standard 10 percent car tariff, reaching into the mid-thirties of a percent for the least cooperative producers. That is the case-by-case model: prove the subsidy, prove the injury, set the duty, move on. Established
What changed by mid-2026 is that the Commission's own findings stopped describing a set of cases and started describing a condition. Brussels has concluded that Chinese state subsidies and overcapacity create unfair-competition conditions threatening European industrial capacity not only in EVs but across solar panels, wind-turbine components, and advanced and mature-node semiconductors — with steel and batteries drawn into the same frame. Established That the accumulation of these findings amounts to a structural threshold — a shift from policing individual products to confronting a rival industrial model — is the desk's framing, and we label it as analysis, not as a Commission declaration. Assessed
The reason the distinction matters is procedural, and it is where Macron's impatience bites. The case-by-case tool is slow: an injury investigation runs a year or more, by which time a wave of subsidised imports has already reset the market. A structural problem outruns a structural remedy built for one product at a time. That mismatch — fast capacity against slow process — is exactly what the emerging European toolkit is trying to close, and exactly what makes it hard.
3. The defensive architecture — and what is actually built
Here the labelling has to be strict, because the gap between what is announced, what is proposed, and what is in force is where analysis most often launders intention into fact. The measures under discussion divide into three tiers.
Enacted, and already biting. The EV countervailing duties are live. On clean-energy hardware, the EU has moved to bar public subsidy from projects using Chinese battery-storage power-conversion systems from mid-2026, and has signalled the end of subsidy support for energy projects relying on Chinese inverters — a market where roughly three-fifths of European imports are Chinese — on both industrial and cyber-security grounds. Established These are real, and they are narrow — hardware carve-outs and product duties, not a system.
Proposed, and moving. On steel, the Commission has proposed a sharp tightening of its safeguard regime — cutting tariff-free import quotas by close to half and roughly doubling out-of-quota duties, with "melt-and-pour" origin rules to stop rerouting through third countries — against a backdrop in which global steel overcapacity could approach 721 million tonnes by 2027, several times total EU consumption. This is a legislative proposal that must clear the Parliament and member states; it is not yet law. Assessed Alongside it sits the idea of a broader "overcapacity instrument" that would let Brussels act against distortion without proving product-by-product injury each time — powerful in principle, unbuilt in practice. Assessed
Proposed, and further off. The remaining pillars named around the Franco-German push — extending the Carbon Border Adjustment Mechanism deeper into downstream goods so that carbon-priced tariffs reach more Chinese exports; new state-aid authorisations to let governments back domestic manufacturers; local-content procurement rules of the "made in Europe" type requiring European assembly and content thresholds for subsidised purchases; and accelerated coordination with Washington on trade defence — are intentions and draft frameworks, not enacted policy. Assessed CBAM's core is operational; its expansion to cover the products that matter most here is prospective. The procurement rules, where floated, phase in toward the end of the decade. The US coordination is aspiration in a transatlantic relationship that has been anything but predictable on trade.
Stack the three tiers and the shape is clear: the enacted layer is real but narrow, and the ambitious layer — the part that would actually match a structural problem — is the part not yet built.
4. The map's real question: capacity versus cohesion
The Cartographer's discipline is to begin from the map and the balance sheet, not the communiqué, and both point to the same fault line. Beijing's advantage is cohesion of purpose: one industrial strategy, one financing system, one direction of travel, capable of flooding a sector faster than any injury investigation can adjudicate it. Europe's disadvantage is the mirror image: twenty-seven fiscal authorities, divergent exposures, and a trade-defence machinery designed for deliberation, not speed. Assessed
The exposures do not point the same way. Germany's carmakers and machine-tool firms sell into China and dread retaliation; France's industrial base is more defensive and readier for tariffs; central European economies host Chinese battery and EV investment they will not readily antagonise; southern members have courted Chinese capital and rapprochement. A trade war needs a pain threshold every member can bear, and China's retaliatory options — rare earths, agricultural purchases, market access for European premium goods — are calibrated precisely to find the seams between them. Assessed
That is why the honest read is not "Europe is launching a trade war." It is that Europe has recognised a structural imbalance and is reaching for structural tools it has not finished building, with a coalition it has not finished assembling, against an adversary organised to exploit exactly that lag. The September roadmap is the near-term tell. If it produces concrete, enforceable Commission action rather than another statement of shared concern, the threshold will have been crossed in deed as well as in framing. If it produces a communiqué, Beijing's capacity will have overrun a divided Brussels once more — quietly, and on schedule.
Europe's China problem has changed category. It is no longer a sequence of anti-subsidy cases but a structural confrontation with Chinese industrial overcapacity across EVs, solar, wind components, mature- and advanced-node chips, steel and batteries — a shift the Merz–Macron alarm of 17 July, and the Commission's own findings, both register. That much is established. What is not yet built is the defensive architecture to meet it: the enacted measures (EV duties, clean-tech subsidy carve-outs) are narrow, and the ambitious ones (an overcapacity instrument, expanded CBAM, state aid, procurement rules, US coordination) are mostly proposed, not law. The decisive variable is not Beijing's capacity, which is not in doubt, but Europe's cohesion, which is. Judge this fight by whether a fragmented Union can convert alarm into enforceable common action faster than Chinese exports reset its markets.
By 30 June 2027, the European Union will have formally adopted or entered into force at least one new structural China-facing trade-defence measure that goes beyond the single-product anti-subsidy model — specifically one or more of: (a) the tightened steel safeguard regime enacted into law; (b) a new "overcapacity" or anti-distortion instrument allowing action without product-by-product injury proof; or (c) a formal CBAM expansion extending carbon-border charges to downstream goods materially exposing Chinese exports. At least one of the three is adopted or in force within the window.
Confidence: Medium (Assessed). The basis is the cadence already visible — the July steel proposal, the September Franco-German roadmap, and explicit Commission and member-state intent — set against the EU's structural slowness and internal divergence, which is the main way this resolves wrong. It also resolves wrong if a negotiated de-escalation with Beijing (a currency/market-access understanding of the kind Merz floated) defers hard measures, or if the roadmap yields only declarations. Resolution date: 30 June 2027, scored against adopted EU legal acts and official Commission notices, not press briefings or draft proposals.