On 7 September 2026, Bloomberg reported that Orange, Deutsche Telekom, Vodafone and Telefónica are in early, preliminary talks on a joint bid for the EU's 2 GHz satellite-to-phone spectrum. Industry coverage of the report says no final decision has been made on forming the consortium or submitting a bid. That caveat matters. This is a negotiating signal, not a deal, and the rules it responds to are still only a proposal.
What the Commission has actually proposed
On 27 May 2026 the European Commission proposed a Regulation setting up an EU-level selection procedure for mobile satellite services (MSS) spectrum in the 2 GHz band. It would replace the 2008 Decision, which expires in May 2027. The incumbent licences are held by Viasat and EchoStar.
According to Greenberg Traurig's summary, the band would be split in three. One third would go to an EU operator for government, security and defence use, integrated with the IRIS² constellation. The commercial two thirds would be divided between a share reserved for new EU entrants and a share open to EU and non-EU bidders, with per-operator caps on non-EU players. Press accounts describe the split differently in places, so the final text deserves a careful read. Rights would not carry over on a change of holder, and holders could request a two-year extension of expiring authorisations. The same summary reports that the Commission's budget planning points to adoption in late 2027 or 2028, selection in 2028 to 2029, and new rights in 2029.
The strongest case for the Commission's approach
The argument for reservation is serious. Satellite-to-phone service is becoming a layer of critical communications, and a government that depends on a non-EU provider for emergency or defence connectivity has handed that provider leverage. Spectrum is also scarce and, once assigned, hard to reallocate. A band reserved for European operators and a sovereign constellation is a reasonable hedge against a market in which one American company already dominates low-Earth-orbit launch and capacity. The caps on any single non-EU bidder, and the rule that licences cannot be quietly sold on, are sensible anti-hoarding tools.
Where proportionality should push back
The weakness is the gap between the sovereignty goal and the consumer outcome. Reserved slices help only if someone actually builds and sells a service on them. According to the reporting on the talks, SpaceX lacks the spectrum rights to provide direct-to-device services independently in the EU and works through partners, including Deutsche Telekom. In France, no operator has been named as a Starlink Mobile partner, and no French launch date has been given. French users therefore have no firm date for satellite coverage of any kind on their phones from this band, from Starlink or from anyone else.
A four-carrier consortium raises its own competition question. The same companies that would jointly bid are the incumbents on the ground in much of Europe. Pooling them could lower the cost of an expensive satellite programme and give Europe a credible alternative. It could also mean that the competing bidders in an auction are replaced by one bloc. Regulators will want to know whether the consortium offers wholesale access to smaller carriers and MVNOs, or whether satellite coverage becomes a differentiator held by four firms. The report also notes that Vodafone's existing 50/50 joint venture with AST SpaceMobile would need to adjust its ownership structure to bid, which shows how tangled the commercial alliances already are.
There is also a timing mismatch. The licences expire in May 2027, but new rights are not expected until 2029. The two-year extension is the pragmatic bridge, and it should be granted without conditions that interrupt services already in use. A cliff-edge would hurt the incumbents' customers, not the policy debate.
What France's regulator is already doing
France is not waiting for Brussels on every front. In July 2026, ARCEP, the French regulator, opened a public consultation on reassigning frequencies in the L band for mobile satellite services, including direct-to-device uses, with responses due by 30 September 2026. That is a separate band from the 2 GHz spectrum at issue here, but it shows that national regulators are building their own access routes in parallel. The practical lesson is that satellite-to-phone policy will be set at two levels at once, and France will want both to point the same way.
What a proportionate design would look like
- Keep at least one genuinely contestable block. If the open commercial share is shrunk by caps and eligibility tests until only European champions can win it, the auction becomes a negotiation among a small group.
- Attach build-out and wholesale obligations to the reserved slices. Reserving spectrum without service milestones protects a position, not a user.
- Test the consortium on competition grounds, not suspicion. A joint bid is legitimate. The right questions are about access terms and whether members also remain in separate deals with other providers.
- Fix the transition early. The extension should be predictable so that current service is not disrupted before new licensees can launch.
The open internet argument is simple. Phones that connect to the sky are most useful where terrestrial networks do not reach, such as rural France, mountains and coastlines, and the people there care about coverage and price, not the flag on the satellite. Europe is right to want capability it controls. It should not mistake closing the band for building the service.
Bottom line
The consortium story is preliminary, and the Commission's text is still a proposal that Parliament and Council can change. Judge both by one test: whether they produce working, competitively priced coverage for users, and whether the final rules leave room for a rival to win spectrum on merit.