The Precursor Problem: Europe's Pharmaceutical Supply Chain Security Gap

Why Europe Makes the Pill but Not the Molecule — and how the 2 April Section 232 proclamation, the 12 May Critical Medicines Act deal and China's untriggered export-control option are repricing it.

Start with one distinction, because it reorders the entire risk map. Europe can formulate, test, package, and distribute medicines at scale. What it has largely lost the capacity to do is make the molecule those medicines are built from. A 2021 European Commission study found that 80% of the active pharmaceutical ingredients imported into the Union come from just five countries — China, the United States, the United Kingdom, Indonesia and India — with China alone accounting for 45% of the total. For generic medicines, the older Falsified Medicines Directive assessment put roughly 90% of APIs as sourced from China and India combined. China supplies an estimated 80–90% of the world's antibiotic API.

That is the headline dependency. The binding one sits a layer deeper, and it is the layer most boards never map. Even when Europe diversifies its sourcing to India — the "pharmacy of the world" — it does not escape China; it merely adds a step. This is the dependency behind the dependency, and it is the same structural signature we documented in The Critical Minerals Trilemma (5 May): the leverage lives at the processing tier, not the finished product. In 2026, three architectures are repricing that exposure simultaneously — and, as with every dependency we have mapped this year, they do not align.

Europe can formulate, package and distribute medicines at scale. What it has largely lost is the capacity to make the molecule — the active ingredients and precursors that increasingly trace back to Asian reactors.

The Dependency Behind the Dependency

The finished-medicine view flatters Europe. Step upstream and it inverts. The clearest illustration comes from the EU Institute for Security Studies, which in April 2026 traced three everyday molecules through the full value chain. At the API stage, the antidiabetic metformin shows a 33% Chinese production share, the antibiotic amoxicillin 20%, and cefpodoxime 13% — medium exposure, manageable on paper. Take the precursor chemicals each one is built from into account, and the Chinese capacity share rises to 83% for metformin, 71% for amoxicillin, and 94% for cefpodoxime. A supply chain that looks Indian, or even European, at the pharmacy counter resolves back to the same handful of Chinese reactors two chemical tiers up. The precursor is where the leverage lives — and where almost no board is looking.

Paracetamol makes the point physically. The continent's most-consumed analgesic had no API production left in Europe at all until France's Seqens, backed by France 2030, moved to restart domestic synthesis. Large-volume antibiotics depend on fermentation capacity that consolidated in China after European plants closed on cost and environmental grounds through the 2000s and 2010s. The political system has registered the danger: in spring 2025, eleven EU health ministers warned in an open letter that Europe's dependence on Chinese medicines had become the weak point in the continent's own defence.

The dependency behind the dependency: for three everyday medicines, Chinese capacity share leaps from the active-ingredient stage to the precursor stage — up to 94%. Source: EUISS, April 2026.

The operational lesson is the one boards most often miss: you cannot stockpile your way out of a continuous-manufacturing dependency. A strategic reserve of finished doses buys weeks. It does not replace a reactor. Replacing a Russian-gas dependency with a Chinese-mineral one is portfolio diversification, not autonomy, as we argued in The European Sovereignty Arithmetic (10 May). The pharmaceutical version is identical: a finished-goods reserve held on top of an imported-precursor base is resilience theatre, not security.

The Export-Control Option Beijing Has Not Triggered

Here is the fact European health ministries are underpricing. Beijing has, over the past three years, built and demonstrated a complete export-control apparatus — and has so far chosen not to point it at medicine.

The chronology is familiar from the minerals file: gallium and germanium licensing in 2023, and in April 2025, in direct response to the Liberation Day tariffs, a licensing regime on seven medium and heavy rare-earth elements. The machinery — licensing by molecule, customs-code targeting, reversible by simple announcement — is now mature and proven to work as strategic leverage. Pharmaceutical APIs and their chemical precursors sit outside that regime today. That absence is being read by too many treasuries as a structural fact. It is not. It is an unexercised option. The same mechanism that placed dysprosium under licensing could place an antibiotic precursor or a fermentation intermediate under the identical instrument with the same legal footing — and the precursor concentration figures above describe precisely where such a move would bite hardest. Boards treating the current freedom of pharmaceutical inputs as permanent are mispricing a lever Beijing holds and has shown it will use.

The American track no longer threatens; it has landed. The 2 April Section 232 proclamation taxes imported pharmaceuticals and APIs at up to 100% — fusing tariff, pricing and reshoring policy in one instrument.

Washington's Proclamation: Industrial, Pricing and Geopolitical Leverage in One Instrument

The American track no longer threatens; it has landed. On 2 April 2026 — one year to the day after the Liberation Day tariffs — a Section 232 proclamation imposed a default 100% ad valorem duty on imported patented pharmaceuticals and associated APIs, the culmination of an investigation opened in April 2025. The rate is tiered, and the tiers reveal the intent. Companies with a Commerce-approved onshoring plan drop to 20%; companies that also sign a Most-Favored-Nation pricing agreement with HHS drop to 0%. The European Union, Japan, South Korea and Switzerland face a 15% rate; the United Kingdom 10%. The duties bite in two waves — 31 July 2026 for firms without onshoring commitments, 29 September 2026 for the rest.

What matters for European planners is the fusion. This single instrument runs tariff policy, drug-pricing policy and reshoring policy through one valve — the dual-architecture pattern we keep returning to, now embodied in one proclamation. Two second-order effects flow directly to Europe. First, the structure pulls diversification capital toward the United States, not the EU: the way to reach 0% is to onshore in America, which competes for exactly the manufacturing investment Europe needs for its own buildout. Second, MFN pricing compresses margins on the molecules it touches — and a margin structure that makes a low-revenue generic uneconomic does not stay contained in the US market; it thins the global manufacturing base for that molecule, and Europe, a price-disciplined buyer, inherits the resulting shortage. Reshoring incentives and price compression are not opposing forces here. Together they redraw who makes what — and Europe is, once again, not the jurisdiction setting the terms.

Europe excels at the visible end of the chain — formulation, fill-finish, finished doses. The Critical Medicines Act funds that tier; the molecule beneath it stays largely Chinese.

Brussels' Answer: The Critical Medicines Act, and Its Limits

Brussels has moved, and faster than the previous decade would have predicted. The first Union list of critical medicines appeared in December 2023; the Critical Medicines Alliance launched in April 2024; the Commission proposed the Critical Medicines Act in March 2025. The Council fixed its position on 2 December 2025, Parliament on 20 January 2026, and on 12 May 2026 the Cyprus presidency reached a provisional trilogue agreement — now awaiting formal endorsement by both institutions. The toolkit is recognisable: supply-chain diversification, incentives and strategic projects for EU manufacturing, joint procurement, stockpiling, and a decisive shift in public tenders away from lowest-price toward resilience and most-economically-advantageous-tender criteria.

The architecture is real. The limits rhyme with those the European Court of Auditors exposed across the minerals file in its September 2025 special report. The procurement reform and strategic-projects funding address the visible end of the chain — formulation, fill-finish, availability of finished doses. The molecule and precursor tiers, where the actual dependency lives, are the least funded of all. Reshoring fill-finish is comparatively cheap and politically legible. Reshoring multi-step synthesis and fermentation is capital-intensive, permitted on the same five-to-ten-year clocks that plague every European industrial buildout, and economically unviable without the price floors and offtake guarantees the Act gestures at but does not fund at the scale the gap demands. Seqens' paracetamol restart proves it can be done — and is a reminder of how rare the example still is. Whether the Act closes the molecule gap by 2030, or merely re-onshores the end of a chain that stays Chinese at its base, is the open question the trilogue text does not settle.

Three Scenarios for European Pharmaceutical Supply Chain Security, 2026–2030

Scenario A — Managed Diversification (base case, ~45–50%). The Critical Medicines Act, once adopted, reshores a narrow critical list; a "China + India + selective EU" structure emerges; formulation and fill-finish capacity returns while the API and precursor tiers stay largely Asian. Europe achieves genuine resilience on a short, defined list of essential molecules — and accepts continued dependence on the long tail. Real progress, delivered at insufficient scale. The familiar signature.

Scenario B — Forced Substitution (~30%). A shock forces the issue on a compressed timeline: a Chinese decision to extend licensing to a pharmaceutical precursor, a disruptive interaction between the 31 July / 29 September Section 232 waves and existing shortages, or a chokepoint event of the kind we analysed in Why It Matters (18 April). Emergency stockpiling and procurement scrambles follow; generic prices spike; Europe captures narrow reshoring wins at high cost and under duress. The probability has risen as external shocks become the operating environment rather than the exception.

Scenario C — Bifurcated Pharma Bloc (~20%). The tail. US reshoring and MFN pricing reorder global manufacturing and R&D economics; Beijing exercises the export-control option on a pharmaceutical input the way it did on rare earths; Europe is caught between two regimes it does not set, and structural generic shortages become a recurring feature rather than an episodic one. Mispriced by markets — not because it is likely, but because its probability is no longer trivial and its consequences are systemic for public-health budgets.

Weak Signals Worth Tracking

A short watchlist, each capable of shifting the probabilities: any addition of an API, intermediate or precursor to a Chinese export-control announcement — the rare-earth playbook applied to medicine, and the single clearest signal of Scenario C; the company-by-company outcomes of Section 232, in particular which firms reach MFN-and-onshoring deals for the 0% rate and where they build; the formal adoption text of the Critical Medicines Act and whether binding manufacturing provisions survive intact; spot shortages in antibiotics, contrast agents and oncology generics as leading indicators; and the second-order effect of MFN pricing on which low-revenue molecules retain a viable global manufacturing base at all.

What This Means, Concretely, For Boards

For institutions across pharmaceuticals, distribution, insurance and the public-procurement chain, four disciplines now apply.

Discipline 1 — Map exposure at the precursor level, not the supplier level. A European or Indian supplier is not diversification if its API or chemical precursor resolves to a single Chinese tier. The metformin and cefpodoxime examples are the template: bill-of-materials exposure must be traced two chemical steps upstream, or it is not mapped at all.

Discipline 2 — Distinguish stockpiling from supply security. Strategic reserves of finished doses are a timing buffer, not a supply solution. The board metric that matters is upstream manufacturing optionality, not weeks-of-cover.

Discipline 3 — Price the 2 April proclamation into cost of goods now. The 15% EU rate, the 31 July and 29 September effective dates, and the MFN-pricing spillover into generic availability are inputs to landed cost and forward supply — not abstract trade policy. Model the company-level tier each key supplier will land in.

Discipline 4 — Treat the Critical Medicines Act as a strategic timeline, not a compliance event. The shift away from lowest-price procurement, the joint-procurement mechanism and the strategic-projects regime will reshape financing costs, offtake structures and competitive position over three to five years. Institutions that map the sequence now will position ahead of the consensus; those waiting for the Official Journal will react to it.

The era in which pharmaceutical supply chain security could be treated as a procurement function is over. It is now a jurisdictional exposure with consequences for public-health continuity, cost of capital and strategic autonomy — and it is the exposure Europe has been slowest to cost. Three architectures are visible. None of them, on its own, closes the molecule gap before 2030. The question for any board, treasury or health ministry is which combination best fits its specific exposure — and whether its planning has even framed the question.

---

This briefing is not investment advice, financial advice or legal advice. It draws on the European Commission's 2021 API dependency study and Critical Medicines Act file, the Council of the EU's 12 May 2026 provisional-agreement announcement, the European Court of Auditors' September 2025 special report, EUISS analysis of April 2026, the US White House Section 232 pharmaceutical proclamation of 2 April 2026, Chinese export-control measures of 2023–2025, and the published CES Intelligence analyses on the Critical Minerals Trilemma, the European Sovereignty Arithmetic, the Beijing Stabilisation, and geopolitical trade risk.

From critical-input dependency to dual-track trade controls, we provide the independent intelligence required to navigate 2026. Request a secure consultation.