IRIS²'s €15.6 Billion Bet: Europe Buys a Constellation It Can't Yet Launch

In three weeks, the European Union turned its flagship satellite programme from a plan into a purchase order. On 1 September, Germany's OHB signed a contract worth close to €1 billion to build eighteen satellites in medium Earth orbit. Nine days later, Belgium's Aerospacelab won €2.4 billion to supply 264 spacecraft platforms — the largest single contract in the history of European space. Spain's Hispasat took the ground segment at more than €1.6 billion.

Three awards. Five billion euros. Zero satellites launched.

The programme behind them is IRIS², short for Infrastructure for Resilience, Interconnectivity and Security by Satellite. On 7 August the European Commission signed an implementation agreement with the SpaceRISE consortium that lifted the constellation to 348 satellites and the estimated bill to €15.6 billion. When the concept was presented in 2022, it carried a €6 billion price tag and a service date of 2027. Both numbers have since moved in the wrong direction. The useful question is whether a network that arrives in 2029 can still win the users it was designed to serve.

Sovereignty Is the Product

The programme is not a consumer broadband business. Its customers are governments, defence ministries, emergency services and the operators of critical infrastructure — users who need connectivity they control rather than connectivity they rent.

Europe does not have that today. Its secure government traffic rides largely on systems it does not own, most of them American. For a bloc that has spent a decade talking about strategic autonomy, that is the gap it exists to close. It is the EU's third flagship constellation, after Galileo for navigation and Copernicus for Earth observation.

The architecture is deliberately mixed. The baseline runs 330 satellites in higher low Earth orbit (LEO) at roughly 1,200 kilometres, plus 18 in medium Earth orbit (MEO). Sixty-six of the higher-LEO spacecraft were added in the August agreement specifically for defence, security and emergency services, with hardened designs.

The money follows the same logic. The funding structure is a public-private partnership: the EU commits about €6 billion across its multiannual budget, the European Space Agency adds €550 million, and the SpaceRISE consortium — Eutelsat, Hispasat and SES — contributes €4.1 billion. Those three operators then run the system under a twelve-year concession that stretches to 2036.

That structure is the point. Unlike Galileo, the programme is mostly privately financed and privately operated, with the state as anchor customer rather than sole owner. For an investor, an anchor tenant with a treaty behind it is close to the best demand signal a capital-intensive network can carry.

The politics reinforced it. At the International Space Summit in September, the EU's defence and space commissioner, Andrius Kubilius, pressed member states not to fragment a planned "European Space Shield" into separate national projects. The message was that sovereign capacity will be bought once, at European scale.

The Bill, the Delay and the Launch Gap

Then there is the arithmetic.

IRIS² was sold at €6 billion in 2022. By the time the concession was signed in December 2024, the figure had become €10.6 billion. In August 2026 it reached €15.6 billion. Four years, roughly two and a half times the original estimate, and first launches now scheduled for 2029, with government services from about 2030.

The delay is not cosmetic. The constellation enters a market where the incumbents are already in orbit and already selling capacity. A sovereign system that switches on in 2030 has to justify why its users waited, when those same users can buy commercial capacity today.

Europe also has a launch problem. The constellation needs satellites delivered on a schedule, and Europe's own launch capacity is only now recovering. Isar Aerospace reached orbit in September — a real milestone, and the first commercial European rocket to do it — but one small launcher is not a deployment machine. Every month of delay is a month of competitive drift.

There is a cautionary tale closer to home. Earlier in September we covered SpaceSail's $1 billion raise for a zero-revenue constellation, and the unhappy economics of building orbital infrastructure before the customers arrive. The programme has better customers and looser cost discipline. Both facts matter.

The procurement critique is louder in Brussels than outside it. The Commission's revised terms drew published criticism over high costs and the removal of an earlier quota system that had spread work more evenly across member states — the kind of change that concentrates value in fewer primes and fewer countries.

None of this makes it a bad project. It makes it an expensive, late project with strategic customers and uncertain returns — a different investment proposition from the one its price tag implies.

How the Three Awards Stack Up

The September contracts divide cleanly, and each one tells a different part of the story. OHB's award is the sovereign core: the small MEO shell that carries government and defence traffic. Aerospacelab's is the industrial bet: 264 platforms at volume, the largest purchase Europe has ever made from a single satellite builder. Hispasat's is the invisible half of every constellation — the antennas, control systems and ground links that decide whether a satellite in orbit is actually connected to anything.

Set the three values side by side and the shape of the programme becomes visible. Most of the money sits in the LEO layer, which is also where commercial competition is fiercest. The sovereign layer — the 18 MEO satellites that hold government traffic — is the smallest line item and the most defensible one.

Contract values announced August–September 2026 · European Commission, SES, Aerospacelab

That distribution explains the bull and bear cases at once. The bear case says Europe is spending sovereign money on a commercial layer it will never win. The bull case says the contracts are already awarded, the anchor tenant is a treaty organisation, and the alternative — renting secure capacity from a foreign operator — is politically closed.

As we wrote in September, satellite manufacturing is racing to catch constellation demand. It is now the clearest test of whether that race can be won on European soil, and the first real answer arrives when the launch contracts are signed.

Sources