In March, Planet Labs put a delay on imagery of Iran. Ninety-six hours, then fourteen days, then on 4 April an indefinite withhold at the request of the US government, with distribution handled case by case. Vantor tightened access over parts of the same region.

Nothing about this was irregular. Shutter control has been a condition of US commercial remote sensing licenses since 1992, and both operators accepted it when they took their licenses. Every satellite stayed in orbit and kept working. What changed was who could use them.

This is what sovereignty arguments usually miss. The debate is conducted in terms of ownership (whose satellites, whose constellation, whose flag on the fairing), while the aspect that determines availability in a crisis is something else entirely. A capability can be domestically owned and still unavailable. It can be foreign-owned and reliably available. The difference is decided at four separate layers, and each one has its own answer.

The Materials Layer

Spacecraft depend on a short list of materials for which there is no substitute at all: specific metals for power generation, magnets, optics and structures. This is the least discussed layer because it is the furthest from the mission, and it is the layer we have limited control over.

As example, satellites fly multi-junction solar cells grown on germanium substrates, which is what makes high efficiency possible. One spacecraft can consume thousands of germanium wafers, a large platform ten of thousands, and global output runs around 130 tonnes a year, only a fraction of which reaches space-grade purity. There is no design workaround available on a program timescale; thus, a licensing decision several steps upstream impacts directly the schedule.

The exposure is shared, and so are the responses. There are limited Western suppliers of space-grade multi-junction cells, and production-share data suggest all of them rely to varying degrees on Chinese-origin germanium upstream. China has licensed germanium and gallium exports since 2023 and has widened the regime steadily since, most recently reaching processing technology itself. Industry is answering across jurisdictions with Rocket Lab pushing silicon as a scalable alternative for high-volume constellations, Azur Space building capacity in the United States, and Europe’s aerospace carbon fibre coming from a Japanese company’s plants in south-west France.

The Components Layer

Radiation-hardened processors, RF front ends, sensors, actuators: the qualified pieces that make a spacecraft flight-worthy. Unlike raw materials, these are governed less by geology than by law, specifically by export control and by the fact that many of them have only a handful of qualified suppliers.

Single-source qualification turns a commercial relationship into a political dependency. If a part exists in one country under an export license, that country’s foreign policy becomes part of the program risk, whatever the contract says. Replacing it is not a procurement decision but a multi-year engineering effort with flight heritage implications.

The European Commission’s non-dependence programme has published what is missing (radiation-hard FPGAs at advanced nodes, GaN devices for mm-wave, substrate capability and irradiation test facilities inside Europe). It requires applicants to demonstrate that their supply chains are free of ITAR or equivalent restrictions and to declare exposure to US export rules explicitly. An example shows the timescale involved in becoming sovereign: rad-hard high-accuracy accelerometers were a single-source item from one American supplier until ESA funded an Irish firm to build a European equivalent. US programs carry their own concentrated qualified sources in rad-hard microelectronics and specialty optics, and the same requalification cost when a supplier changes hands.

The Manufacturing Layer

Factories, integration halls, test facilities and launch. What used to be a question of whether anyone would order enough to justify a production line has become a question of whether the lines can be built fast enough to absorb the orders now arriving.

Europe has committed serious money here, and the delivery dates are close together: tens of billions at the EU and national level, a record ESA subscription and a consolidation of the prime base through the ongoing Airbus-Leonardo-Thales merger. The Bundeswehr’s planned constellations suggest something in the order of a hundred satellites a year. Germany’s next-generation military satcom constellation targets initial capability in 2029. IRIS² (expanded on 7 August to 348 satellites) targets first launches in the same year. When several programs need the same specialist supplier in the same quarter, capacity becomes a timeline risk.

As the primes consolidate with full order books, the pressure shifts one level down, to the Tier-2 and Tier-3 firms. They generally have the engineering knowledge; what they lack is the capital for the single machine that limits throughput and a procurement signal early enough to justify buying it. Scale is not a complete answer here either. The US industrial base is deeper in workforce, tooling and supplier depth, and it is still absorbing the strain of proliferated architectures, with launch manifests saturated years out.

The Control Layer

Finally, the set of licenses, contracts and legal powers that determine who may task an asset, who may be told to stop and who is served first when demand exceeds supply, tops the layers.

This is the layer that decides what happens in the week that matters most. It is also where the commercial-versus-sovereign framing breaks down, because states routinely control assets they do not own; what differs is the instrument. The United States works through licensing authority, as in the March and April examples. Provider discretion is a second channel: Starlink service in Ukraine has been available for communications, while offensive applications have required authorization, and Kyiv was still routing requests for broader permission through Washington as of July. The EU is building market and regulatory instruments. Russian doctrine, for its part, treats commercial satellites supporting military operations as legitimate targets by function.

For anyone specifying a space service, that reduces to four practical questions. Which state can lawfully compel this provider, and how? In a contested surge, whose tasking is served first, and is that contractual or discretionary? Who can degrade, delay or withhold, and with what notice? And if the provider is compelled tomorrow, what is the fallback, and has anyone exercised it?

A satellite owned in one jurisdiction and built from parts licensed in another fails the first question at the supply-chain level. A foreign constellation tasked through a domestically controlled ground segment can pass the second and third while still failing the first. The answers are per layer, and programs are more useful to their users when they say which layers they actually control.

What the Market is Converging On

If sovereignty is layered, then partnerships are being designed to buy specific layers rather than whole systems, and four models are visible in signed contracts. A national operator can be built around foreign technology, as Germany did in contracting Rheinmetall’s joint venture with Finland’s ICEYE for radar reconnaissance data. Processing can be brought home while the sensor stays abroad, as in the Rheinmetall-Vantor venture placing a US spatial intelligence platform inside German command-and-control. Capacity can be pre-committed on someone else’s asset, as France did with an eight-year call-off for OneWeb capacity, including funded work to harden the service, as cover until IRIS² is running. And access can be held in reserve, as the US does through its Commercial Augmentation Space Reserve, which pre-negotiates peacetime contracts with pre-priced surge and reaches full operational capability this autumn.

In essence, sovereignty has never been a binary question. Dependencies across layers are shaping how the ecosystem is built to meet future needs, and each nation is currently navigating the spectrum of risks and controls to determine its own strategy, which in turn shapes the industry.