For the better part of a decade, the European defence debate ran on a single complaint: Europe would not spend. That complaint is now obsolete. Germany’s defence budget has climbed to €108.2 billion for 2026, the EU has allocated €150 billion in low-interest SAFE loans for defence procurement, and European states have become the world’s largest arms-importing region, absorbing a third of global arms imports in 2021–25. The question that matters has changed. It is no longer whether Europe will pay for its own defence, but what kind of defence-industrial order the money is building.
The answer emerging from procurement decisions in 2026 is uncomfortable for both camps in the familiar argument. The “buy European” camp can point to real progress in moving spending away from American suppliers. The Atlanticist camp can point to the systems Europe still cannot build. Neither is describing the more consequential trend. European rearmament is being Europeanised in label and nationalised in substance. The money is flowing less to a common European defence industrial base than to a collection of national ones, each protected by its own capital, while the continent’s most strategically significant capabilities remain American.
A three-tier dependency
It helps to stop treating “European dependence on the United States” as a single variable. The procurement picture is better understood in three layers.
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.
At the top sit the strategic enablers: fifth-generation aircraft, long-range air and missile defence, maritime patrol and deep-strike missiles. Here dependence on Washington is barely shrinking. SIPRI’s latest data show that the United States supplied 58 per cent of the arms imported by European NATO members in 2021–25, and its researchers note that European buyers kept turning to American combat aircraft and long-range air defence even as EU-backed intra-European orders grew. Germany’s own procurement planning, widely read as a turn away from American suppliers, still includes around 15 additional F-35s and four P-8A Poseidon maritime patrol aircraft.
In the middle sits the volume layer: armoured vehicles, artillery, ammunition, short- and medium-range air defence, and drones. This is where the “Europeanisation” story is real. Politico’s reporting on Berlin’s near-term procurement plan found that only around 8 per cent of roughly €83 billion in contracts was earmarked for US firms. But the same planning documents show where much of the rest goes: roughly half of the orders on Berlin’s wider wish list are slated for German companies, with Rheinmetall and Diehl at the centre. Europe is buying European. More precisely, Europe’s largest spender is buying German.
At the bottom sits the component layer, which rarely makes headlines. Here dependence runs not only toward the United States but toward China. The European Parliament has estimated that up to 60 per cent of drone components assembled in Europe come from third countries, and a single American-origin chip can bring an entire system under US export controls. A “European” drone is often a European assembly of non-European parts.
The headline statistic on American market share mostly measures the top layer. The more important story of 2026 is unfolding in the middle one.
How SAFE became a national instrument
SAFE was designed to push that middle layer toward integration. Its loans were meant to finance common procurement. It requires that at least 65 per cent of component value originate in the EU, EEA/EFTA states or Ukraine, and for critical products it demands that the prime contractor hold full design autonomy, free of third-country restrictions. On paper, it is the most serious attempt yet to use the EU’s borrowing power to consolidate a fragmented market.
In practice, a transitional clause did much of the work. The regulation allowed procurements by a single member state to qualify if contracts were signed by 30 May 2026, with only a commitment to extend the benefits to others later. The effect was predictable: capitals moved fast to lock in national contracts before the window closed. Poland, the largest beneficiary at €43.7 billion, signed its loan agreement in May and said 89 per cent of the contracts would go to Polish companies. In the final days before the deadline, Warsaw aimed to sign dozens of contracts worth around 100 billion zloty, with the government presenting the programme as a boost for thousands of domestic firms.
None of this breaks SAFE’s rules, and it would be unfair to treat it as bad faith. Poland faces the most acute threat environment in the EU, its government had to work around a presidential veto at home to use the money at all, and domestic production capacity is a legitimate security asset. But it does mean that the EU’s flagship integration instrument has, in its first and largest tranche, functioned primarily as cheap financing for national industrial policy. Common procurement is now the rule. Whether it becomes the practice will depend on how much is left to buy jointly once the largest national programmes are already under contract.
FCAS and the limits of co-development
If SAFE shows how joint money gets nationalised, the collapse of the Future Combat Air System shows what happens when co-development meets national industrial interest head-on. In June, Berlin and Paris abandoned the plan to build a joint next-generation fighter after mediation between Dassault and Airbus failed. Only the combat cloud, the networking layer meant to link aircraft, drones and sensors, is to continue as a shared effort.
The failure is usually narrated as a corporate feud over workshare and intellectual property. That is accurate but incomplete. The two governments wanted different aircraft. France needs a fighter capable of carrying nuclear weapons and operating from a carrier, while Chancellor Merz said Germany needed neither. When requirements diverge at the root, industrial disputes are the symptom rather than the cause. The lesson for European defence integration is sobering. The most ambitious collaborative programme of the past decade did not die for lack of money. It died during the most generous period of European defence spending in a generation.
There is, however, a more constructive reading. If what survives is the combat cloud, Europe may be stumbling toward a more realistic model: separate national platforms tied together by shared standards and a common digital architecture. Integration at the level of connectivity rather than ownership is less ambitious, but it may be the only kind European politics can currently sustain.
The case for fragmentation, and its limits
It would be too easy to dismiss national preference as parochialism. There are serious arguments for it.
The first is speed. Joint programmes are notoriously slow, and governments that see a Russian threat on a three-to-five-year horizon have reason to buy from suppliers they can direct and audit themselves. The second is resilience. The war in Ukraine has shown the value of distributed production that is harder to disrupt than a few large, concentrated plants. A Europe with ammunition and drone producers spread across many states is, in one sense, more survivable than one built around a single champion. The third is politics. Rearmament requires sustained public consent for higher spending, and voters are more likely to back defence budgets that produce local jobs and local factories.
The limits are just as real. Duplicated production lines mean higher unit costs and shorter runs. Multiple national variants of similar systems complicate logistics, training and spare parts in any coalition war. And national champions, once built, become constituencies that resist later consolidation. Every factory opened with SAFE money in 2026 is a future political argument against a cross-border merger.
The British case shows the same tension from the outside. London’s attempt to join SAFE broke down over the size of the entry fee Brussels demanded, leaving British firms with only limited third-country access. Whatever the merits of each side’s position, the episode showed the EU defining “European” defence partly in terms of club membership and financial contribution rather than military capability alone. That is an understandable instinct for an instrument backed by the EU budget, but a costly one for a continent whose largest military power outside the Union sits on the wrong side of the line.
What to watch
The test of whether Europe’s rearmament is building a common industrial base will not come in 2026, when contracts are signed and announcements are made. It will come later, in three places.
The first is the post-deadline SAFE pipeline: whether the remaining procurement produces genuine multi-country orders, or national purchases with a nominal partner attached. The second is the component layer: whether European funding begins to build supply in chips, sensors, engines and propellants, the inputs that currently tie “European” systems to third-country export controls. The third is the top tier: whether Europe starts developing credible alternatives in long-range air defence and deep strike, or quietly accepts that its most strategically decisive systems will remain American for another generation.
Europe has answered the question it spent a decade asking. It is now paying for its own defence. The question it has not yet faced is harder: whether it is building one defence-industrial base or many, and whether, in a crisis, that difference will matter more than the money.
