Großwald Curated No. 44 — Paid on delivery

The market stopped paying for the promise

Rheinmetall fell as much as eighteen per cent on Wednesday, on the news that Berlin had cancelled the F126 frigate the firm had bid to lead. What makes it a re-rating rather than a bad day is the arithmetic: the company shed more than ten billion euros of market value over a contract worth at most 12.8 billion — and one it had not yet won. The selloff carried into Hensoldt, Thales, Saab, Leonardo and BAE; it is the same reversal this column has tracked since the rearmament trade turned off its war-boom highs in the spring, with institutions unwinding the overweights they built when the sector doubled. The analysts' verdict was about credibility, not capacity: a prime that had been valued on the expectation of capturing a disproportionate share of new programmes was marked down the moment one such promise failed to convert.

Into exactly that cold tape, KNDS published its intention to float — a Frankfurt–Paris listing valued at twelve to fifteen billion euros against a record order book of 33.1 billion, first trading expected in July. The valuation is the same discipline at work: the figure once mooted for KNDS ran as high as twenty-five billion, and the listing arrives at little more than half of it, offered to institutions only, retail kept out — a minority slice of a Franco-German fortress sold at a discount. The deeper tell is in the mechanics: no new stock is issued, so the company raises nothing. Only existing shares change hands, and the proceeds go to the sellers — the French state and the founding Wegmann families. At the top of a rearmament boom, Europe's largest land champion is going public not to fund a single new production line but to recycle its ownership and pass control to two governments. That is how a utility lists, not a growth company. The market is no longer paying primes a growth multiple for the size of the backlog; it wants the booking, and it is prepared to treat the incumbent as a controlled, dividend-bearing, slow-moving thing while it waits.

The customer was already buying that way

The F126 cancellation that broke Rheinmetall's week was, from the state's side, the proof premium paid in steel. Berlin walked away from six developmental 10,000-tonne frigates — the largest German warships since the war, their bill grown past eighteen billion euros with no hull in the water before 2032 — wrote off some two billion already spent, and will buy instead up to eight TKMS MEKO A-200 ships for about 11.6 billion, first delivery 2029. The MEKO is half the displacement and none of the ambition; what it is, that the F126 was not, is a design already at sea with navies abroad. A hull that exists beat a hull that was promised. The same instinct ran at the bottom of the stack, where France fast-tracked an order for five thousand reconnaissance drones from a two-year-old supplier whose product the army had already flown on exercise — proven first, procured fast.

This is not new for the customer; it is the logic No. 43 traced to the combat-proven outsider, now turned by the German state against its own flagship and its own champion. What it kills is a specific thing: the sovereign developmental mega-programme, the decade-out promise that was the national champion's stock-in-trade. The casualty list of the past month reads as one sentence — the Franco-German FCAS fighter dead, the joint MGCS tank stalled nine years in, the F126 cancelled, the SAMP/T still not credited with a ballistic-missile kill in combat over Ukraine. For a generation Europe paid a premium for these programmes and tolerated their slippage as the price of autonomy. This week it did the opposite twice in a day: cancelled the flagship and denied the lead, because slippage is now disqualifying and the in-service alternative is on the table.

What is proven is held by the people Europe does not control

Follow the proof premium to where it is actually paid and the sovereignty problem appears. Europe is not short of combat-proven kit as such: its Caesar howitzers have been validated up and down the Ukrainian front; the Leopard tank carries the same pedigree. The proof it lacks sits in the categories the war invented — the cheap interceptor, the attritable drone, the ground robot, the one-way deep-strike round, mass made disposable — and there the combat record belongs to the people Europe does not control: to Korea (the interceptors Rheinmetall is localising), to Turkey (the drones), to Ukraine (almost everything the front has tested) — and, in the naval case, to the export incumbent whose proof is its order book abroad, the MEKO already sailing for South Africa, Egypt and Algeria. The week's marquee venture round is the same shape: Stark, a Berlin loitering-munition maker founded in 2024, raised five hundred million euros led by Sequoia and Founders Fund at a valuation past 3.5 billion, and a day later Germany's ARX and Ukraine's Roboneers signed a joint venture to mass-produce a Ukrainian ground robot — capital from American and European funds, the proof from the Ukrainian front. The funds now say the quiet part aloud: battlefield validation in Ukraine is the due diligence. European defence-tech took a record 8.7 billion dollars in 2025, and Helsing is raising toward an eighteen-billion valuation on the same logic. ARX is the detail that gives the game away: its own first systems, built to a European specification, failed in combat until Ukrainian input helped redesign them. The proof was never the European engineering; it was the war. And where it is not the war, it is foreign capital outright.

This locates the KNDS exception precisely. Europe paid up for sovereign control there — golden shares, board vetoes, a ten-year ownership floor — because it could: the Leopard, the Leclerc and the Caesar are mature designs Europe already owns and has long since proven, so proven and European do not pull apart. They pull apart the moment the capability is new and the threat is immediate — the cheap interceptor, the attritable drone, the deep-strike round — where the only proof on offer is someone else's. The trade-off, stated flat: Europe can pay for proof, or it can pay for sovereign control of the design, and this week, wherever the need was urgent, it chose proof. The honest caveat is that even the proof premium smuggles promise, and sometimes the proof itself is thin. Stark and ARX are valued on combat records of present kit but priced on volume not yet built — tens of thousands a year against hundreds fielded — and Stark's Virtus missed all four targets in allied live-fire trials last November, one drone veering off and burning out in woodland, before Berlin booked it in February and the funds tripled its value in June, its hit rates having since recovered. Fire Point's claim of a sub-million-euro ballistic interceptor by year-end, on a German radar and European seekers, is likewise a claim and not yet an intercept. The regime has not abolished the promise; it has narrowed it to promises made by those who can point to proof of something adjacent.

Europe has answered the budget question, and walked into a harder one

One regime change runs under all three sections. This week European defence stopped paying for promises and began paying on delivery — the same discipline in the share price, the procurement decision and the term sheet. What Curated No. 33 called combat-provenness — a perishable, mispriced property — is what this week's buyers began to pay a premium for; whether they have also priced the perishability, the fact that a 2026 proof answers a 2026 threat and no later one, is the question the regime leaves open. That settles, almost in passing, a different question — the one that has dominated since 2022. The argument about how much Europe will spend — two per cent, three and a half, five — is effectively over; the spending is coming. The live question is what the money will buy, and the answer, this week, is proof: the delivered, the in-service, the combat-tested, and a sharp discount on everything still drawn on paper.

The trouble is what that discipline exposes. Europe has spent three years treating speed and sovereignty as one project — rearm fast and rearm sovereign, money the solvent for both. The proof premium prises them apart, because in the new categories that matter most now — cheap mass, made to be expended — money buys proof and the proof is mostly foreign. Every clean buying decision this week chose fast-and-proven over slow-and-sovereign: the export hull over the national frigate, the Korean interceptor over the unbuilt European one, the Ukrainian robot over the home design that failed under fire. Sovereign control survived only where the product was old enough that nothing had to be given up to get it.

So the wager under the week is narrow and concrete: whether Europe can collapse the trade-off by making its own proof — building combat-tested kit at home fast enough that proven and European stop being rival demands. If it can, the proof regime rebuilds sovereignty on a firmer footing than the promise regime ever did. If it cannot, the most disciplined defence market Europe has run in a generation will have spent its way into a better-delivered dependence. The reading is not in any communiqué; it is in the KNDS order book a year out, the units shipped against this week's venture pledges, and the first European interceptor that connects.