Why Are 1990s Satellite Flops Suddenly Worth Billions?
Iridium and Globalstar were the defining bankruptcies of the dot-com era. Ten months and $37 billion later, they anchor Big Tech's land grab for the airwaves between space and your phone — and the FCC just turned those licenses into moats.
On Friday afternoon, while the market was busy arguing about AI capex, Amazon quietly filed an application with the FCC for 5,105 new satellites. Not for Leo, its broadband constellation that already numbers in the thousands. A second, separate system — orbiting at 510 to 580 kilometers, designed to do one thing: talk directly to the phone in your pocket.
To understand why that filing matters, you have to start with two of the most famous corporate failures of the dot-com era.
The most valuable bankruptcies in America
Iridium spent roughly $5 billion building the world's first satellite phone network in the 1990s. Nine months after launching service, it filed for bankruptcy — its assets sold in 2000 for about $25 million, roughly half a cent on the dollar. Globalstar, its rival, followed into Chapter 11 in 2002. For two decades, both companies survived as niche operators serving sat-phones, ships, and remote industrial sensors. Business schools taught them as cautionary tales about technology arriving before its market.
Now look at the last ten months:
- September 2025: SpaceX buys EchoStar's AWS-4 and H-block spectrum licenses for approximately $17 billion — half cash, half SpaceX stock — plus a commitment to cover roughly $2 billion of EchoStar's interest payments through 2027.
- April 2026: Amazon agrees to acquire Globalstar for approximately $11.6 billion — $90 per share in cash and Amazon stock, expected to close in 2027.
- June 2026: Rocket Lab announces an $8 billion acquisition of Iridium.
That is nearly $37 billion committed to satellite communications companies and their frequencies in under a year — most of it aimed at businesses that, on current revenue, would struggle to justify a fraction of those prices. AST SpaceMobile, the pure-play in the space, didn't buy a company; it signed 80-year usage rights to up to 45 MHz of Ligado's L-band spectrum in mid-2025.
The buyers are not paying for satellites. Satellites depreciate — Amazon's own filing gives its new fleet a design life of six to eight years. They are paying for the one asset in the space economy that cannot be manufactured, launched, or replaced: licensed spectrum that works with an unmodified smartphone.
The FCC just turned licenses into moats
The physics is unforgiving. A phone's antenna is tiny and transmits at a fraction of a watt. Only a narrow set of frequency bands — low enough to penetrate, licensed cleanly enough to avoid interference — can close a link between a handset on a sidewalk and a satellite moving 27,000 km/h overhead. The companies that hold clean, exclusive, global mobile-satellite licenses can be counted on one hand. Most of them got those licenses in the 1990s, when nobody else wanted them.
Then, on April 23, the FCC did something the market has still not fully priced. In a broad order on mobile-satellite spectrum, it reaffirmed Globalstar's and Iridium's exclusive rights to the so-called Big LEO band — and dismissed petitions from SpaceX, Kepler, Sateliot, and AST SpaceMobile asking to share access to portions of it. Chairman Brendan Carr framed the decision as positioning the U.S. to lead in direct-to-device services. Its practical effect was simpler: it converted legacy licenses into regulatory moats, and made buying the incumbent the only way in.
Amazon's July 24 filing is the first hard look at what that logic builds. The 5,105 satellites would run on Globalstar's 1.6 and 2.4 GHz spectrum — the band the FCC just fenced off — delivering mobile data, messaging, emergency communications, in-vehicle connectivity, and IoT links where cell towers don't reach.
And there is one more detail that turns this from an industry story into a Big Tech story: Apple has spent years and more than a billion dollars — including a $400 million purchase of a 20% equity stake — turning Globalstar into the invisible backbone of the iPhone's Emergency SOS feature. Under the acquisition, Amazon buys Apple's stake out. Apple keeps its satellite features, but its critical infrastructure will now be owned by a direct competitor, with Apple as the anchor tenant.
The land grab is nearly complete. The question that matters for the next eighteen months is who actually collects — and where the prices haven't caught up to the April order.
The rest of this briefing is for paid members: the full ownership map of every satellite-to-phone band in the U.S. market, the merger-arb structure sitting inside Globalstar's $90-a-share deal, what the April FCC order did to AST SpaceMobile's bull and bear cases, the Amazon capex question heading into this week's earnings, and the EchoStar playbook for whoever sells next.
AlphaBriefing Paid gets you every investment thesis, scenario framework, and catalyst brief we publish — the analysis private intel clients pay four figures for, at a fraction of that.