The slice of radio spectrum that knocked roughly 6% off AT&T, Verizon and T-Mobile in extended trading Thursday has had a long and slightly embarrassing life. It is 817 to 824 MHz paired with 862 to 869 MHz, the old Nextel band, and it passed to Sprint and then to T-Mobile in the 2020 merger, where it was promised to DISH as part of the price of approval. DISH never paid. In 2023 its parent company conceded it could not fund the agreed $3.59 billion, and the option lapsed. T-Mobile then sold the band to Grain Management, a private-equity firm, for $2.9 billion in cash plus Grain’s own 600 MHz licenses, a swap the FCC approved in July and the companies closed in August.
On Thursday, Grain agreed to sell 100% of that 800 MHz portfolio to SpaceX. Neither side disclosed a price. The Wall Street Journal reported, and Reuters relayed, that SpaceX will pay about $8 billion in cash, citing people familiar with the matter. That figure is still single-sourced, so treat it as the best number available rather than a confirmed one.
Most of the coverage has framed this as Elon Musk declaring war on the phone companies. The more useful frame is the receipt. Sixty-odd days ago T-Mobile owned these exact airwaves. Today they are the indoor-coverage piece Starlink Mobile was missing, and if the reported price holds, a private-equity firm turned $2.9 billion of cash into roughly $8 billion without building anything on the band at all.
It is worth sitting with how everyone read the T-Mobile swap at the time, because the read was reasonable. When the FCC signed off in July, Recon Analytics analyst Roger Entner told Light Reading that “the biggest winner is clearly T-Mobile,” because the deal moved “spectrum that is de facto unusable for them” and handed back 600 MHz the carrier could actually put to work. That was true for T-Mobile’s network. It was a statement about what the band was worth to a terrestrial carrier that already had plenty of low-band. It said nothing about what the band was worth to a satellite company trying to reach phones inside buildings.
The same FCC order is what changed the answer. The commission granted Grain waivers to use the band for supplemental coverage from space, rejected Grain’s proposed six- and twelve-year buildout plan in favour of three- and eight-year deadlines, and, as Bloomberg Law reported in August, required Grain to seek bids to develop spectrum the order itself described as underutilized. In plain terms: Washington made the band useful to satellites, then told its new owner to sell it or build on it fast. Bloomberg put Grain’s asking price at about $6 billion, with SpaceX and AST SpaceMobile both circling. If the Journal’s number is right, SpaceX paid about a third over that ask, and AST did not get the band.
What SpaceX actually bought is narrower than the share prices suggest. The company called it “a license portfolio of up to 14 megahertz of paired spectrum in the 800 MHz band.” Low-band signals travel through walls and tree cover, and most phones already in American pockets support 800 MHz, so Starlink Mobile customers would not need special hardware. SpaceX already holds mid-band 2 GHz spectrum from its 2025 EchoStar purchase, and this week the FCC approved 15,000 direct-to-device satellites for the Starlink Mobile constellation. Low-band plus mid-band plus a constellation is the outline of a carrier.
The carriers know it. Last week T-Mobile, AT&T and Verizon formed a joint venture for satellite and direct-to-device coverage in underserved areas, CNBC noted, and Starlink was conspicuously not invited. SpaceX’s answer, seven days later, was to buy the one band T-Mobile used to own that makes a satellite signal work indoors. You could not script a pettier sequel.
Is 6% the right price for the threat? Our read: right direction, early timing.
The bear case on SpaceX here is real, and the most credible version comes from TMF Associates’ Tim Farrar, who told CNBC that it is “still a very limited amount of spectrum and to get reliable building penetration in urban areas SpaceX would have to deploy towers on the ground.” Fourteen megahertz is a sliver next to what each of the big three holds. Satellite capacity per square mile is thin in dense cities. And SpaceX’s own stock did not treat Thursday as a coronation: it fell 4% in regular trading before recovering about 2.5% after hours.
That is exactly why the tower owners did not sell off with their biggest tenants. A Starlink that wants indoor coverage in cities needs ground sites, and American Tower, Crown Castle and SBA Communications rent ground sites. The bull case circulating among tower analysts Thursday night was simple: a SpaceX ground network would be a fourth paying customer on the same steel. Buying spectrum is not a build commitment, so that trade rests on optionality. But it is the right instinct about where the money flows if Musk means what SpaceX said, that this spectrum will “pave the way for Starlink Mobile to become a major mobile carrier in the US.”
Where the carriers should actually worry is not Manhattan. It is the places where their coverage maps are thinnest and their marketing is loudest. Starlink had 12 million subscribers worldwide at the end of June, and Deutsche Bank estimates about 4.1 million of them are in the U.S. Those are households that already pay SpaceX every month. Bundling a phone line into that bill in rural America does not need towers on every corner. It needs exactly the kind of low-band signal SpaceX just bought.
Now the part nobody seems bothered by.
If the reported price is accurate, Grain made about $5 billion on top of its cash outlay in roughly two months. Grain also gave up its 600 MHz licenses in the swap, so the true margin is smaller than the headline gap. It is still a spectacular return for holding a public resource through one regulatory decision.
That value did not come from Grain. It came from the FCC deciding this band could serve satellites. The public’s airwaves created the premium, and a middleman collected it.
We think the FCC got the competition policy right and the economics wrong. Opening low-band to satellite carriers is good for rural customers and good for pricing. Letting the windfall land with whoever happened to hold the licenses that week is not.
There is still a lever. The SpaceX purchase needs FCC approval.
The commission should hold SpaceX to the same three-year interim buildout deadline it imposed on Grain, in writing, in the approval order.
Otherwise this band spends another decade as a trading chip.
And T-Mobile, which got the 600 MHz it wanted, will spend that decade competing against the spectrum it sold.