SpaceX Spent $17B on Spectrum the FCC Won't Let It Use From the Orbit It Needs.

The Spectrum Deal That Rewrites Starlink's Business Case

SpaceX closed a $17 billion deal on September 8, 2025, to buy EchoStar's H-block and AWS-4 spectrum licenses in the United States. EchoStar's own investor relations statement confirmed the sale covers its "full portfolio of AWS-4 and H-block spectrum licenses" and that SpaceX intends to use those licenses to "develop and deploy a next-generation Starlink Direct-To-Cell constellation."

That single transaction reframes what Starlink is. Until now, the service has been a broadband play, with ground terminals beaming internet to rural homes and enterprise sites. The EchoStar spectrum gives SpaceX the licensed frequencies to talk directly to unmodified smartphones, which is a different product, a different regulatory posture, and a different revenue model.

The price tag itself signals how seriously SpaceX is treating the shift. Octus reported the transaction at roughly $19 billion when including related international authorizations and the AT&T spectrum component, while LinkedIn and EchoStar's own announcement put the U.S. license sale at $17 billion. Either figure represents one of the largest private spectrum acquisitions in the sector's history, and it puts SpaceX in control of the MSS (mobile satellite service), AWS-4, and H-block bands that domestic carriers have long eyed for terrestrial 5G.

The regulatory angle matters as much as the dollars. EchoStar stated publicly that the sale, combined with the AT&T deal, is expected to "resolve the Federal Communications Commission" inquiries into its spectrum holdings. For SpaceX, that means acquiring licenses that already survived FCC scrutiny, a faster path than applying for new allocations and one that hands the company a position AST SpaceMobile and other would-be D2D players have to negotiate around rather than replicate.

What the FCC's 15,000-Satellite Order Actually Authorizes

The FCC's December 2024 order brought SpaceX's total authorized Gen2 count to 15,000 satellites. The commission granted an additional 7,500 spacecraft on top of the 7,500 already approved for the Gen2 system. But the orbital mechanics locked inside that authorization matter more than the raw satellite count.

SpaceX asked to fly these satellites at very low Earth orbit altitudes, specifically 340, 345, 350, and 360 kilometers. Those are VLEO altitudes, roughly 100 km lower than where the operational Starlink fleet orbits today. Flying lower cuts signal round-trip latency and increases link budget margin for direct-to-device connections with unmodified smartphones. It also drags satellites through thicker atmospheric drag, requiring more propulsion fuel and shortening orbital lifetimes.

The FCC said no to the lower orbits. The commission rejected SpaceX's bid to deploy thousands of satellites in that 340–360 km VLEO range, deferring that portion of the application. SpaceX walked away with authorization for 7,500 additional Gen2 satellites, but not at the altitudes it wanted for maximum phone connectivity performance. The order grants in part and defers in part, with conditions attached.

Despite the altitude restriction, the FCC framed the Gen2 expansion as a direct enabler of mobile service. The commission's order states the expansion will let SpaceX deliver high-speed, low-latency internet globally, including "enhanced mobile and supplemental coverage from space." That regulatory language matters: it embeds direct-to-device capability into the official purpose of the Gen2 constellation, not just fixed broadband terminals.

On its own site, SpaceX emphasizes the throughput gains per satellite over first-generation systems, promising more bandwidth and increased reliability for end users. The hardware improvement per spacecraft is real. Gen2 satellites carry significantly more capacity than the v1.5 fleet. But the VLEO altitude denial forces SpaceX to hit its mobile coverage targets from higher orbits, where the physics of connecting to a cell phone antenna are harder.

The 15,000-satellite authorization is the ceiling SpaceX can now build toward. Whether those spacecraft orbit at 340 km or 540 km determines how much RF engineering brute force is required to close the link to a device in your pocket, and how many engineers SpaceX needs to hire to do it.

The Job Postings Behind the Phone-Connected Constellation

SpaceX's job postings reveal the specific engineering muscle behind the Direct to Cell push, and they read like a shopping list for the EchoStar spectrum and VLEO build-out.

The most telling role is the RF/Microwave Engineer (Direct to Cell), posted for the Redmond, WA satellite development hub. This isn't a general RF position. The job description says the engineer will develop "all aspects of the phased array antennas and modem hardware for the next generation of the Direct to Cell system," the actual payload that turns a Starlink satellite into a cell tower in space. The pay range runs $100,000 to $135,000 base, plus equity. The role is listed under "Starlink Mobile," which is SpaceX's internal name for the D2D product line.

The preferred qualifications make the hardware challenge concrete. SpaceX wants someone who has designed mixed-signal circuit boards from concept through production, including processors, FPGAs, multi-GHz Serdes, DDR4 DRAM interfaces, and analog-to-digital converters. That's the kind of experience you find at Qualcomm or Ericsson, not at a typical satellite shop. The posting also asks for electromagnetic simulation experience and at least two years working on aerospace systems, which narrows the pool to people who've done both terrestrial wireless and space hardware.

A separate RF Engineer (Starlink) posting for the Bastrop, TX manufacturing site covers the ground-side hardware, including phased arrays, gateway antennas, and consumer wireless products. It requires five or more years of professional RF experience with antennas or communications systems, plus hands-on work with spectrum analyzers, vector signal generators, and network analyzers. The Bastrop location matters: that's where SpaceX is scaling Starlink dish production, and the same RF team likely supports the consumer terminals that will eventually talk to D2D-enabled satellites.

Both postings carry standard ITAR restrictions (U.S. citizen or permanent resident only), which is a hard constraint on hiring velocity that competitors like AST SpaceMobile don't face to the same degree.

On the regulatory side, SpaceX has posted a Satellite Policy Associate (Starlink Regulatory Affairs) role that handles the licensing and advocacy work the EchoStar deal demands. The job includes drafting pleadings and advocacy pieces for FCC proceedings, supporting spectrum policy outreach to government officials, and managing the regulatory licensing applications for Starlink's spacecraft and earth stations. That's the position that turns a $17B spectrum purchase into actual operating authority.

Zero G Talent's board shows the company added 97 roles in the past week alone, with openings spanning Hawthorne, Redmond, and Starbase. The Direct to Cell and VLEO build-out isn't a future hiring plan. It's happening now, and it's pulling from the same shallow pool of RF engineers that every other D2D competitor is fishing in.

SpaceX vs. AST SpaceMobile: The Talent War

That hiring pace continued this week, including a Sr. Commercial Counsel for Starlink paying $235,000–$290,000. That salary band signals how aggressively the company is buying the legal and engineering talent needed to lock down its direct-to-device spectrum advantage. The fight for RF engineers and satellite-communications specialists is now zero-sum: every senior hire SpaceX makes for its D2D constellation is a hire AST SpaceMobile or a legacy mobile network operator cannot afford to lose.

What the job boards reveal

SpaceX is recruiting a "Sr. RF Systems Analysis Engineer, Regulatory (Starlink)" in Washington, D.C., a role that merges deep radio-frequency engineering with the regulatory fluency required to defend spectrum filings at the FCC. AST SpaceMobile is chasing the same overlap of skills, posting roles like "Senior RF – GNSS & TT&C Satellite Engineer" and "Senior RF Solution Architect / GPL Lead" out of its Lanham, Maryland office. Both companies want engineers who can own link budgets, manage external antenna vendors, and navigate EMI/EMC testing in dense, multi-transmitter spacecraft environments. The candidate pools overlap almost entirely.

AST SpaceMobile currently lists roughly 317 open positions on Glassdoor, spanning everything from satellite manufacturing engineers at its Midland, Texas facility to RF system integration engineers. The volume shows a company scaling fast to build its own D2D network. But SpaceX's recruitment engine is larger and its compensation ceiling is higher. When SpaceX posts a Starlink commercial counsel role at nearly $300,000, it sets a benchmark that pulls experienced spectrum-policy professionals out of legacy telecom and defense contractors, the same people AST SpaceMobile needs to argue its own case before regulators.

Where the engineers come from

The D2D talent war draws from three established labor pools:

The national average for a satellite communications engineer sits around $96,000 to $103,000, with top earners reaching roughly $164,000. Both SpaceX and AST SpaceMobile are pushing past those ceilings for candidates who combine satellite payload experience with cellular-network knowledge, a hybrid skill set that barely existed five years ago. LinkedIn job postings in the Maryland and D.C. corridors show SpaceX, AST SpaceMobile, Lockheed Martin, and Johns Hopkins Applied Physics Laboratory all competing for the same RF and satellite-communications engineers within a 30-mile radius.

SpaceX's asymmetric edge

SpaceX holds two cards that AST SpaceMobile and legacy MNOs cannot match. First, launch cadence: engineers want to see their hardware fly, and SpaceX launches Starlink satellites weekly. AST SpaceMobile must buy rides from others. Second, capital: the EchoStar spectrum deal signals to recruits that SpaceX's D2D push is funded for the long haul, reducing the startup-risk discount that might otherwise push candidates toward incumbents.

AST SpaceMobile's careers page asks recruits to "be part of a historic transformation," the kind of mission pitch that works when you cannot match cash compensation. It will need that pitch to land consistently, because SpaceX is writing checks that legacy satellite operators and mobile carriers simply will not match for RF and regulatory talent.

Why 10 Million Subscribers Are Just the Start

Starlink crossed 10 million active subscribers in February 2026, doubling from 4.5 million barely a year earlier. That number alone would make it the largest satellite communications business in history. But the investors sizing up SpaceX's IPO aren't pricing the company on 10 million users. They're pricing it on what happens when direct-to-cell turns every smartphone in a carrier partner's footprint into a potential Starlink terminal.

The revenue math is shifting fast. Starlink generated $11.8 billion in 2025: $7.5 billion from consumer broadband, $1.3 billion from hardware, and $3.0 billion from U.S. government contracts. Quilty Space projects $20 billion for 2026, a 69% jump driven by consumer growth, a Starshield government breakout, and a brand-new mobile segment that didn't exist a year earlier. Payload Space's more conservative model puts 2026 at $18.7 billion. The $1.3 billion gap between the two forecasts sits almost entirely on how fast Starlink's direct-to-cell service ramps.

That mobile segment is where the EchoStar spectrum deal stops being a regulatory story and becomes a revenue story. Quilty models Starlink Mobile reaching 25 million monthly active users by year-end 2026 through carrier partners (T-Mobile in the U.S., Rogers in Canada, KDDI in Japan), with roughly 18,000 new users joining daily at the time of the forecast. The EchoStar purchase gives Starlink the licensed spectrum to make that service reliable rather than experimental, and the long-term commercial agreement ties EchoStar's Boost Mobile subscribers into the network. This is genuinely new revenue: wholesale carrier fees on top of, not replacing, the residential broadband base.

The operating margins make the case stick. Starlink posted $4.42 billion in operating income for 2025, roughly 37% margins, up from $469 million in 2023. Traditional satellite operators trade at 8 to 12x EBITDA. High-growth SaaS companies rarely sustain above 40x. At a $2 trillion enterprise value, the market is pricing Starlink at roughly 143x EBITDA, which only works if every segment in Quilty's model performs and the mobile ramp doesn't stall.

The risk is real. Starlink's network capacity supports about 20 million broadband subscribers today. Pushing past that requires the mass deployment of V2 Mini satellites and next-generation hardware. Amazon's Project Kuiper, starting with 700 satellites against Starlink's 7,000-plus, is a distant threat for now but a real one by 2028. And the $3.2 billion Starshield government projection depends on Congressional procurement timelines that frequently slip, a one-year delay there alone could knock $1 to $2 billion off the 2026 number.

Still, the floor is high. Even the bear-case subscriber projection from analyst Tim Farrar puts Starlink above 35 million users by 2030. The EchoStar spectrum and the 15,000-satellite VLEO filing aren't speculative bets. They're the infrastructure for a business that's already the majority of SpaceX's revenue and its only profitable division. The IPO roadshow won't be about whether Starlink grows. It's about whether growth can outrun the valuation.

The Regulatory Moat Nobody Is Talking About

The FCC's May 2026 approval didn't just hand SpaceX 65 megahertz of spectrum. It handed SpaceX something more durable than any satellite or rocket: a regulatory position that rivals can't replicate on any realistic timeline.

Here's what actually changed. Before the EchoStar deal, every direct-to-device service in the U.S. ran through a carrier partnership. T-Mobile and SpaceX. Verizon and AST SpaceMobile. AT&T and AST. The terrestrial carrier always held the spectrum, and the satellite operator was a vendor. Light Reading's Roger Entner put it bluntly after the FCC ruling: "SpaceX is no longer a vendor to terrestrial carriers in the D2D market. It is a spectrum-holding network operator."

That shift is now locked in at the regulatory level. SpaceX acquired 15 megahertz of AWS-3, 40 megahertz of AWS-4, and 10 megahertz of H-block, 65 megahertz total of exclusive, contiguous, nationwide spectrum. The FCC confirmed the "exclusive nature of D2D spectrum bands, including the spectrum SpaceX is acquiring from EchoStar." Exclusive is the operative word. AST SpaceMobile's recently approved 248-satellite system operates under a different and more constrained arrangement. No other satellite operator holds exclusive nationwide U.S. mid-band spectrum for direct-to-device service, and acquiring that position from here requires either a similar secondary-market transaction or a similar target.

That's the moat's second layer. EchoStar's spectrum was the product of merger conditions from the 2020 T-Mobile/Sprint deal and years of buildout obligations. It's not replicable through an FCC auction because no comparable swath of mid-band MSS spectrum is coming to market. The FCC's own press release committed to releasing roughly 300 megahertz of low- and mid-band spectrum by the end of 2027, but most of that is terrestrial, and none of it carries the same D2D-exclusive designation. A rival constellation operator trying to enter the U.S. D2D market would need to negotiate with SpaceX itself, or wait for a regulatory window that may not open for years.

The FCC also granted SpaceX a set of "pathbreaking waivers" that allow flexible terrestrial, space-based, or hybrid use of the spectrum, subject to first-of-its-kind performance obligations. These waivers break down the regulatory silos between satellite and wireless frameworks. That creates a template, but one the FCC shaped around SpaceX's specific position, launch cadence, and technical capacity. Whether the agency extends similar terms to AST SpaceMobile or future applicants is an open question, not a given.

The commercial consequence follows directly. Boost Mobile, as part of the transaction, gets access to Starlink's next-generation Direct to Cell service under a long-term commercial agreement. That gives SpaceX an immediate distribution channel without having to sign carrier-by-carrier deals for every market. Every other satellite operator in the D2D space still needs to negotiate those relationships individually.

The moat has a global dimension too. SpaceX acquired not just U.S. spectrum but also global MSS licenses bundled into the EchoStar transaction. International spectrum coordination runs through the ITU, and a company that already holds national allocations and operates a deployed constellation has structural advantages in cross-border frequency coordination that a new entrant must build from zero.

Scotiabank analysts wrote that SpaceX is "using the Federal Communications Commission process to transform spectrum rights, service approvals, and satellite rulemaking into a regulatory moat around Starlink." That's not hype. It's a description of what happened in the May 12 ruling. The question for rivals, AST SpaceMobile, Amazon Kuiper, and every MNO evaluating whether to build or buy into D2D, is whether they can afford to wait for the next spectrum window. At SpaceX's current launch cadence, that window may close before they can put a single satellite on orbit.

Working in space? Zero G Talent tracks the openings: browse space jobs, openings at SpaceX, and the people building the field.