Shares of AT&T, Verizon and T-Mobile fell in pre-market trading Tuesday after a Reuters report detailed how much wireless spectrum SpaceX has quietly assembled for a ground network to complement its Starlink satellite service.
SpaceX, according to the report, spent $19.6 billion last year on two transactions that brought it 65 megahertz of wireless spectrum across the AWS-3, AWS-4 and H-Block bands, purchased from EchoStar. Combined with its global mobile satellite services licenses, the company has put together the pieces it says it needs to build a full terrestrial mobile network, with cell sites on the ground and satellite backhaul in the sky.
The disclosure rattled the three largest U.S. carriers, whose shares slipped before the open on fears that a company with SpaceX’s capital, engineering culture and fresh public-market funding could replicate the disruption it brought to launch services. The phrase doing the rounds on trading desks was “satellite phone disruption,” the idea that Starlink’s satellite-direct-to-phone capability, joined with a real ground network, could undercut the carriers on coverage and price.
The deals themselves are a study in how the wireless market’s pieces have moved around. EchoStar, the parent of Dish Network, spent years accumulating spectrum in auctions, then struggled to build a network of its own; selling to SpaceX handed the satellite company a ready-made portfolio at a time when the FCC’s rules required the spectrum to be put to use. SpaceX’s willingness to pay $19.6 billion, more than most analysts had expected the assets to fetch, showed how much a dedicated spectrum position is worth to a company entering the market from space.
What SpaceX would do with the spectrum is the question. The company has said little publicly about its wireless plans beyond its satellite business, but the pieces point to a network that combines Starlink’s global coverage with terrestrial cells in cities and suburbs, where satellite signals are weakest. The technology, called direct-to-device when satellites talk to phones, has been advancing quickly, and SpaceX has already tested the capability with partners. Adding its own ground network would let it offer a complete service without depending on the carriers it would be competing against.
The carriers have pushed back, arguing that spectrum purchased for satellite use carries conditions, and that repurposing it for a terrestrial network would require regulatory approval. The FCC, which has been friendly to SpaceX in past fights, would have the final say, and any decision would face years of legal challenge from the incumbents. Analysts said the regulatory path is the biggest unknown in the company’s plan, and that the carriers’ lawyers will treat every step as a battle.
The broader question is whether the wireless market, long divided among three dominant carriers, can absorb a fourth competitor with a different cost structure. SpaceX’s advantage is that its satellites already cover the country, and its ground network would be an addition rather than a foundation. The incumbents’ advantage is their existing customers, their tower positions and their decades of regulatory experience. Both sides know the fight is coming; Tuesday’s pre-market dip was the market’s first estimate of what it will cost.
For a company that listed publicly only this summer, the spectrum strategy is a statement of ambition. SpaceX has moved from launching rockets to operating the world’s largest satellite network to, if the pieces fall into place, competing for the most valuable wireless market on earth. The $19.6 billion it paid for spectrum is a wager that the era of the satellite-plus-ground network has arrived, and that the carriers’ hold on American wireless is not as secure as it looks.
The spectrum purchases also explain a puzzle that has followed SpaceX since its listing: why a satellite company would sit on cash while rivals built ground infrastructure. The answer, the Reuters report makes clear, is that SpaceX has been assembling the pieces of a full network, and the spectrum was the scarce one. Wireless spectrum is a finite national resource, allocated by auction and regulated by the FCC, and the AWS-3, AWS-4 and H-Block bands SpaceX bought are among the most valuable in the country, suited for the high-capacity mobile services that the carriers themselves use.
What happens next depends on two things: the FCC’s willingness to let SpaceX use spectrum originally earmarked for satellite services in a terrestrial network, and the carriers’ response in court and in the market. The incumbents have deep experience in both arenas, and they have made clear they will treat any attempt to repurpose the spectrum as a threat to their businesses. But they also face a competitor with different economics: SpaceX’s satellites are already in orbit, its capital base is deep, and its tolerance for long regulatory fights has been proven in a decade of battles over Starlink.
