I’ve been in this industry long enough to know the most interesting question is rarely what happened. It’s why rational actors made a specific move at a specific moment.
Last week gave us three important signals that, taken together, say a lot about where mobile is heading:
- The AT&T/T-Mobile/Verizon joint venture announcement
- The FCC’s approval of the $40 billion EchoStar spectrum sale to SpaceX and AT&T
- And SpaceX’s S-1 filing, submitted to the SEC on May 20, timing that was very likely intentional
Individually, each story matters. Together, they tell a much bigger one. At its core, this is a story about satellite mobile strategy, spectrum control, and the future of direct-to-device connectivity. Let’s unpack it.
Understanding the Spectrum Battle Behind Satellite Mobile Connectivity
Before getting into motivations, it’s worth taking a quick detour through spectrum because the spectrum layer underneath all of this is what makes the broader story make sense.
When a U.S. wireless operator deploys mobile service, it does so using spectrum exclusively licensed by the FCC. That exclusivity is the foundation of the U.S. wireless investment model. It’s what gave carriers confidence to spend hundreds of billions building networks: they operate in a protected lane, and nobody else can use it without permission.
Satellite works differently, and this distinction matters.
If a satellite operator wants to connect directly to a standard mobile phone from orbit — what the industry broadly calls direct-to-device (D2D) — there are two ways to access spectrum.
The first is called Supplemental Coverage from Space (SCS). Under this model, the satellite operator uses a carrier’s terrestrial spectrum license. That means the satellite company has to negotiate access with whichever carrier owns that spectrum. The carrier controls the terms, the access, and ultimately the leverage.
That’s why SpaceX’s “Starlink Direct to Cell” partnership with T-Mobile — and AT&T’s arrangement with AST SpaceMobile — are structured the way they are. The carriers own the spectrum, so they hold the negotiating power.
(A quick terminology note, because I know some readers will catch this: SpaceX brands its service as “Direct to Cell” or D2C. D2D is the broader umbrella term covering all satellite-to-handset services, including D2C, AST SpaceMobile, Skylo, and others. Both terms are relevant here.)
The second path is fundamentally different. The ITU established dedicated satellite spectrum allocations years ago under what’s known as Mobile Satellite Service (MSS). These include bands such as L-Band and S-Band, including AWS-4 spectrum.
Operators that control MSS spectrum don’t need permission from terrestrial carriers to reach mobile users. That independence is becoming one of the most important competitive advantages in the emerging satellite mobile connectivity market.
Until last year, EchoStar held a significant MSS spectrum portfolio — specifically AWS-4 licenses — which it operated terrestrially under the DISH brand. Not particularly successfully, but it still controlled the licenses.
When SpaceX agreed to acquire that spectrum portfolio for $19.6 billion in equity and cash consideration, it wasn’t simply buying bandwidth. It was buying the ability to reach mobile users without needing to negotiate with AT&T, T-Mobile, or Verizon.
The FCC approved the transaction in May 2026, although the full license transfer won’t officially close until November 2027. In practical terms, though, the key regulatory hurdle has already been cleared.
Amazon’s acquisition of Globalstar spectrum assets follows the same logic. This is not a coincidence. It’s a pattern.
Why AT&T, Verizon, and T-Mobile Are Aligning on Satellite Mobile Strategy
Once you look at the spectrum dynamics, the carrier joint venture becomes far less surprising, and far more interesting.
Publicly, the announcement was framed around improving rural coverage and addressing dead zones. That’s real enough.
But the more honest interpretation is this: three major carriers watched SpaceX successfully push a nearly $20 billion MSS spectrum acquisition through the FCC, then file an IPO registration statement eight days later, and concluded that the moment required a coordinated response. The fact they made this announcement before having a definitive agreement is telling.
The wireless industry has seen this kind of disruption before. When Apple launched iOS and Google followed with Android, carriers lost control of the software layer, and with it, a meaningful part of the customer relationship. For the past fifteen years, carriers have largely remained the infrastructure and billing providers while gradually becoming disintermediated from the experience layer.
MSS spectrum creates the structural possibility for a satellite operator to establish a direct customer relationship that bypasses terrestrial carriers entirely.
Not everywhere. Not immediately. There are still major technical and regulatory hurdles: handset support, spectrum coordination, coverage density, and device ecosystem alignment all matter.
But the threat is structural, not hypothetical. As satellite mobile services become more capable, carriers face the risk of losing control over customer relationships in much the same way they did during the smartphone era.
I’ll put it plainly: the mere fact that AT&T, T-Mobile, and Verizon made a joint public announcement in the same week is significant.
Anyone who has worked in telecom long enough knows how difficult it is to get those companies aligned on anything. When they move together, it usually means the shared threat has become more urgent than the competitive positioning.
Now, whether this JV actually results in meaningful infrastructure is still an open question. The regulatory and antitrust path will not be simple.
But the announcement itself matters because it establishes a negotiating position and begins shaping industry norms around D2D access before SpaceX’s V2 Mobile constellation materially changes the competitive landscape.
What SpaceX’s S-1 Reveals About Its Satellite Mobile Strategy
SpaceX filed its registration statement on May 20, just eight days after the FCC approved the EchoStar spectrum acquisition. Companies do not file IPO registration statements while major regulatory uncertainty is unresolved.
That timing alone tells you something, and several other details in the filing are especially important. Starlink already has partnerships with more than 30 mobile network operators globally and approximately 7.4 million active mobile users across roughly 30 countries. T-Mobile has openly acknowledged that current Direct to Cell usage with Starlink remains negligible — essentially a corner-case traffic scenario today. That’s accurate.
But these partnerships are not really about current traffic volumes. They are laying the foundation for a global satellite-to-cellular ecosystem that could reshape how mobile coverage is delivered.
The S-1 also clarifies how Starlink’s mobile constellation is structured. Today, Starlink operates roughly 650 dedicated V1 Mobile satellites separate from its broadband constellation. Those satellites currently support light data, SMS, and OTT voice services. The next-generation V2 Mobile satellites are where things become more consequential. Those satellites are designed for broadband data and IoT connectivity directly to handsets.
Deployment is expected to begin via Starship in 2027, and full commercial Gen2 service in the U.S. depends on two things happening together:
- V2 Mobile satellites successfully reaching orbit
- Completion of the EchoStar license transfer in November 2027
That makes the real competitive inflection point more likely in the 2028–2029 timeframe. That timing matters. This is not a crisis arriving next quarter. But it is close enough that carriers are not wrong to start positioning now.
The Financial Logic Behind SpaceX’s Satellite Mobile Push
This is where the S-1 becomes especially revealing.
Starlink’s Connectivity segment is currently the only profitable part of SpaceX. In 2025, it generated $4.4 billion in operating income on $11.4 billion in revenue, representing nearly 50% year-over-year growth. Those are impressive numbers, until you look at the broader company structure sitting above it. The Space segment — including rockets and Starship R&D — posted a $657 million operating loss in 2025.
Meanwhile, the AI segment, which now includes xAI and X, burned through $6.4 billion in operating losses during the same year. Overall, SpaceX reported a consolidated net loss of $4.9 billion for 2025, with an accumulated deficit of $41.3 billion as of March 31, 2026.
In other words, Connectivity is carrying the company. That context changes how you interpret everything else in the filing.
Now add another important pressure point: ARPU. Starlink subscriber ARPU declined from $91 per month in 2024 to $81 in 2025, then down again to $66 in Q1 2026. The S-1 explicitly says management expects that decline to continue as international expansion accelerates and lower-priced plans become a larger share of the subscriber mix.
That means more subscribers, but lower revenue per user. This is exactly why mobile matters so much to the SpaceX business model. Revenue per megabyte in mobile remains materially higher than in fixed broadband.
The EchoStar acquisition gives SpaceX an independent path into mobile. And when you consider what the Connectivity business is now supporting — including a Starship program consuming roughly $3 billion annually in R&D, alongside an AI business still in heavy investment mode — simply serving as a supplemental rural dead-zone layer for T-Mobile is not enough.
To justify the economics, SpaceX needs scale. A lot of it. There’s no formal subscriber target disclosed in the S-1. But the constellation economics tell their own story.
The long-term architecture reportedly calls for roughly 9,600 satellites with useful lifespans of only three to five years, creating continuous replacement demand. At the same time, Starship economics improve as launch cadence scales.
Put those together, and the math increasingly requires meaningful mobile revenue growth to support the capital structure. The broadband growth trajectory is already substantial — 10.3 million subscribers growing 105% year-over-year. But mobile is where the long-term unit economics likely need to go.
What This Means for Operators, Investors, and Broadband Providers
For Mobile Operators
The JV is the right instinct. The real question is execution. Coordination agreements that lead to actual shared specifications, deployment plans, and investment commitments meaningfully change the leverage dynamic. Announcements without operational follow-through do not. The important thing to watch over the next twelve months is what actually gets filed, funded, and built. That’s the real signal.
For Tower Companies and Infrastructure Investors
On balance, this is probably net positive. When carriers face competitive pressure from satellite, they typically don’t slow terrestrial investment. If anything, they accelerate it in order to make coverage quality strong enough that satellite substitution becomes irrelevant for most everyday use cases.
And there’s another practical reality here: Direct to Cell still struggles indoors. Reliable indoor performance will continue requiring terrestrial densification, WiFi offload agreements, and ground infrastructure partnerships. Satellite alone is not a complete mobile architecture.
For Broadband Operators
The longer-term implication is the emergence of a converged competitor offering both fixed broadband and mobile connectivity from a single provider. And that platform may improve faster than terrestrial networks can realistically be built. The timeline is not immediate. But the direction is increasingly clear, and “wait and see” is becoming a strategy with a narrowing window.
Why Spectrum Control Is Driving the Future of Satellite Mobile Connectivity
The principle connecting all of this is actually pretty simple, even if the mechanics are complicated: Spectrum is control. Not in an abstract sense, but in a very contractual, regulatory, and strategic one.
Once you understand who controls which spectrum layers — and under what terms — the motivations behind these announcements become much easier to read. The companies that control spectrum will have a significant advantage in shaping the future of satellite mobile connectivity and direct-to-device services.
Game theory only works if you understand what each player is optimizing for.
SpaceX is optimizing for the free cash flow trajectory needed to support a company carrying a $41 billion accumulated deficit while simultaneously funding Starship, scaling xAI, and operating X.
The carriers are optimizing to avoid repeating the customer relationship outcome they experienced during the smartphone platform transition.
And the FCC is trying to manage a spectrum transition involving a company unlike any the telecom industry has dealt with before.
That’s the story. The rest is press release.
I work with operators, investors, and infrastructure companies navigating exactly this kind of strategic complexity. If you’re evaluating where your business fits as satellite becomes a more meaningful layer in the connectivity stack, I’d welcome the conversation.