Image: 123rf

SpaceX’s proposed acquisition of Grain Management’s 800 MHz spectrum portfolio strengthens Starlink Mobile’s push into terrestrial wireless. But coverage is not capacity, and building a competitive nationwide network could cost tens of billions of dollars.

In sum – what to know:

Low-band boost – Proposed acquisition gives SpaceX access to up to 14 MHz of paired 800 MHz spectrum, complementing its satellite plans and strengthening its ability to deliver solo terrestrial coverage.

The capacity catch – Analyst Vish Nandlall warns that spectrum alone cannot support busy urban networks. Bernstein estimates a competitive nationwide mobile network could cost up to $130bn to build.

New incumbent threat – But SpaceX could pressure AT&T, Verizon, and T-Mobile through pricing and rural, while leasing towers or buying wholesale capacity rather than building a nationwide network from scratch.

More news from Starlink. SpaceX has agreed to acquire a nationwide portfolio of 800 MHz spectrum from investment firm Grain Management, strengthening Starlink Mobile’s push into terrestrial wireless. SpaceX chief Elon Musk claimed the deal will enable “complete phone coverage in America”. He also said, in a social-media back-and-forth, that, for “those who understand the spectrum wars”, it represents an “earthquake” deal, effectively to take on the status quo among the traditional telco set. 

The agreement follows fresh FCC approvals for SpaceX to deploy 15,000 direct-to-device (D2D) satellites, as covered here yesterday. But as also discussed, questions still remain over capacity limits, infrastructure costs, and the economics of competing with established mobile operators.

Grain said on October 8 that SpaceX would acquire 100 percent of its nationwide 800 MHz spectrum portfolio, subject to FCC approval and other closing conditions. The companies did not disclose financial terms. The Wall Street Journal has valued the transaction at around $8 billion. Grain acquired the portfolio from T-Mobile in August in exchange for cash and its own 600 MHz spectrum holdings. The subsequent sale to SpaceX gives the satellite operator access to up to 14 MHz of paired spectrum in the 800 MHz band.

There is some irony, of course. The lower the frequency, the further the signal goes. T-Mobile swapped 800 MHz for 600 MHz, effectively, and will be happier with it for strategic reasons; but its old spectrum has gone directly to a well-heeled potential new rival, with grand ambitions – and lots of in the way of market eyeballs and media airtime. SpaceX says the licences will provide a coverage layer complementing its existing spectrum holdings, helping Starlink Mobile to connect devices indoors and across areas where terrestrial cellular coverage is limited. 

On October 7, the FCC approved a request by SpaceX to deploy a constellation of 15,000 satellites for D2D services, supporting mobile satellite services and ‘supplemental coverage from space’ (SCS). The commission also waived standard leasing requirements for certain spectrum SpaceX acquired from EchoStar, allowing the company to pursue satellite coverage without being required to rely on a separate terrestrial operator for those arrangements. The approvals strengthen SpaceX’s ability to combine satellite and terrestrial spectrum under its own control. 

The Grain transaction would extend that strategy with additional low-band spectrum. The deal has rattled the US mobile sector. AT&T, Verizon and T-Mobile shares fell sharply in after-hours trading on Thursday, while European telecoms stocks also came under pressure on Friday. Tower companies – including American Tower, Crown Castle, and SBA Communications – moved in the opposite direction, as noted by analyst Vish Nandlall, as investors weighed the prospect of greater competition against potential demand for terrestrial infrastructure.

But coverage is not the same as capacity, as Nandlall, one of the sharper commentators on Starlink Mobile, mapped on the back of an envelope in these pages this week. In a social media post, Nandlal again cautioned against equating spectrum ownership with a viable national mobile network. The acquisition gives SpaceX a useful low-band coverage layer, he argued, but it does not solve the capacity challenge in densely populated areas. 

A satellite signal that reaches a rural home or penetrates a building is not necessarily capable of delivering the throughput and consistency of a busy terrestrial network. Nandlall said: “Verizon, AT&T and T-Mobile each lost roughly seven percent in after-hours trading… That’s a crazy market reaction to a relatively small amount of radio spectrum… Combined with its next-gen satellites, SpaceX can begin assembling a hybrid mobile network. But the spectrum provides coverage rather than much capacity.”

It won’t support busy urban mobile data traffic, he said – “without terrestrial infrastructure”, however SpaceX builds or buys such a thing. But building a terrestrial network looks like a tall order. Nandlall cited figures from financial research firm Bernstein Research that it will cost anywhere between $50 billion and $130 billion to build a competitive network. BNP Paribas has previously put the cost of a US build at $100 billion, based on 50,000-70,000 sites – rising to 120,000 with the C-band holdings, should SpaceX collect from the upper C-band auction next April. 

“At 20 million subscribers, recovering $50 billion over 10 years requires $21 per subscriber per month, before operating costs, financing, or any return,” he reasoned. “That’s a difficult starting point for a low-price competitor.” But he also noted that SpaceX might do plenty of damage to the likes of AT&T, T-Mobile, and Verizon in the meantime. “SpaceX [also] doesn’t have to take many customers to damage the incumbents’ economics… For every 100 million subscribers, a $1 reduction in the monthly bill removes $1.2 billion in annual revenue [from the market].”

Noting the market’s gamble on tower companies following the deal’s announcement, he added: “SpaceX also has choices that a conventional fourth mobile operator wouldn’t have. It can start with rural customers, connected vehicles and Starlink broadband bundles. It can lease existing towers or purchase wholesale cellular capacity instead of immediately building a duplicate national network. Investors appear to be pricing in both the competitive threat and the infrastructure SpaceX will still need. I wouldn’t dismiss SpaceX’s ambitions, but I wouldn’t confuse acquiring spectrum with having a competitive mobile network either.” 

Very clearly, SpaceX is moving beyond its established role as a satellite connectivity provider and towards a more integrated mobile proposition. The FCC approvals and the proposed 800 MHz acquisition both reduce its reliance on traditional mobile operators, giving it greater control over spectrum and the network architecture. The latest deal does not establish that Starlink Mobile can replicate the capacity of terrestrial networks, or that it can do so at a price that makes a standalone service commercially attractive.

Tower agreements, wholesale arrangements, retail pricing, and service performance will determine that. Again, chapeau Vish.