SpaceX wants to go from satellite internet company to phone company. On Thursday, it took its biggest step yet.
The company agreed to buy Grain Management’s nationwide 800 MHz spectrum portfolio. That’s up to 14 megahertz of paired low-band spectrum, the kind of airwaves that carry through walls. The deal needs FCC approval. Neither side disclosed a price.
Grain’s own press release called it a “definitive agreement.” SpaceX posted about it on X at 3:19 p.m. CT Thursday. Its message was blunt: the portfolio “will pave the way for @Starlink to become a major mobile carrier in the US.”
What SpaceX Is Actually Buying
Spectrum is the invisible real estate every wireless signal travels on. Low-band spectrum is the beachfront property. Its long waves travel far and push through buildings, trees and hills.
SpaceX says that’s the layer its satellite spectrum is missing. In a statement, the company explained how the pieces fit together. “While Starlink Mobile’s global 2 GHz mid-band spectrum will provide high-bandwidth capacity in the United States, this new low-band spectrum will provide a coverage layer that ensures Starlink Mobile’s signal penetrates through obstacles, such as walls, and can provide service to customers’ devices even when they are in buildings,” SpaceX said, as quoted by Via Satellite.
SpaceX also says most existing phones already support the band. That matters. Nobody has to buy a special handset.
A Spectrum Block With a Strange History
This slice of 800 MHz has had an odd journey. The Justice Department made T-Mobile sell it as a condition of the Sprint merger. The FCC has called the licenses “underutilized in recent years.” Grain bought the licenses from T-Mobile in a deal announced in early 2025 and completed in August. Grain paid $2.9 billion in cash plus some of its own 600 MHz spectrum, Via Satellite reported.
In July, the FCC approved Grain’s purchase with strings attached. The agency wanted the airwaves used, fast. It allowed the buyer to meet its deadlines with terrestrial towers, satellites or a mix.
SpaceX‘s filing leans on that flexibility. Under the baseline rules, service must reach at least 33% of the population, or 25% of the geography, by July 1, 2029. The bar rises to 66% of the population, or 50% of the geography, by July 1, 2034. SpaceX describes a “flexible, hybrid architecture” and says it plans radios on towers, rooftops and other structures. It also asked for expedited approval.
Why the Timing Matters
This didn’t come out of nowhere. Days earlier, the FCC approved SpaceX’s plan for a 15,000-satellite direct-to-device constellation. Via Satellite says the satellites will fly in very low orbits, between 326 and 335 kilometers. SpaceX currently serves direct-to-phone customers with roughly 650 satellites.
SpaceX President Gwynne Shotwell told investors the next generation will be a leap. “You could look at the Starlink Mobile leveraging this next-generation satellite and the EchoStar spectrum as being 100 times better than what we’ve got right now,” she said, per Via Satellite. She said second-generation satellites should start flying next year, with service by the end of next year.
Put the pieces together and the strategy is clear. EchoStar’s mid-band spectrum feeds the satellites. Grain’s low-band spectrum fills in the walls and the dead zones. It’s a two-layer network.
The three big carriers saw it coming. On Oct. 1, AT&T, T-Mobile and Verizon formally launched their own joint venture to line up satellite coverage from multiple providers, with Paul Roth as interim CEO. Their goal, AT&T CEO John Stankey said in May, is to “keep pricing in check, so there isn’t a bottleneck of any particular single provider.”
Wall Street Flinched
Investors reacted fast. Analyst Vish Nandlall noted that Verizon, AT&T and T-Mobile each lost roughly 7% in after-hours trading. Tower companies moved the other way. The Wall Street Journal valued the deal at around $8 billion, RCR Wireless reported.
Elon Musk, per the same outlet, said the deal would enable “complete phone coverage in America.” He also called it an “earthquake” for “those who understand the spectrum wars.”
The Reality Check
Not everyone buys the earthquake talk. Nandlall called the selloff “a crazy market reaction to a relatively small amount of radio spectrum.” His core point: “the spectrum provides coverage rather than much capacity.”
Coverage means a signal reaches you. Capacity means thousands of people in a city can stream video at once. Low-band spectrum is great at the first and limited at the second. Nandlall cited Bernstein Research, which estimates a competitive U.S. network could cost $50 billion to $130 billion to build.
JPMorgan reached a similar view. It said the deal makes Starlink Mobile’s long-term opportunity “more credible.” But the firm sees limited near-term risk to the carriers, given the time, infrastructure and capital needed to build a competitive terrestrial network.
Our read: both camps are right. SpaceX doesn’t need to build a nationwide tower network overnight to hurt the big three. It can start with rural areas, vehicles and Starlink broadband bundles. It could lease towers instead of building them. Even a small price cut across a customer base of millions rewrites carrier math. But a true fourth national carrier is a long road, and nobody has yet shown how SpaceX pays for it.
The Bigger Picture
Remember where this started. Starlink began as a broadband play. Now SpaceX says its satellite network serves more than 12 million customers in more than 170 countries, according to its own FCC filing. It launches its own satellites on its own rockets. Few phone companies can say that. Starlink (Mobile) is on its way to an industry interrupting uprising.
That vertical stack is the real advantage. Every Falcon 9 that lifts a Starlink batch adds capacity at SpaceX’s cost, not a vendor’s price. Each launch also feeds the story investors are betting on.
