The ticker had four letters and no history. SPCX opened on the Nasdaq on the morning of June 12 at $150 a share, already eleven percent above the price SpaceX's bankers had settled on the night before. It closed at $160.95. That single session's arithmetic valued Elon Musk's rocket company at more than two trillion dollars — a hair above Tesla, the other firm he runs between meetings — and made the offering the largest the world has recorded. Saudi Aramco, the prior record holder, had raised less than half as much. By the following Friday the shares were changing hands near $185, as though the first day had been an act of restraint.
Forty minutes up the commuter line, in Stamford, Connecticut, the mood was not celebratory. Charter Communications — the cable company that sells Spectrum internet to roughly thirty million American homes — had spent the spring watching its own number move the other way. In the first quarter it lost 120,000 broadband subscribers. A year earlier the figure was 59,000. Analysts had braced for something near 100,000 and still flinched when the report landed. The stock fell twenty-five percent in a day and touched its lowest point in more than a decade.
One company owns no copper, strings no coaxial line, and leases almost no cell towers. The other owns little else. The market has begun pricing the gap between them.
The Loudest Bell on the Nasdaq
The flotation was built to be noticed. SpaceX set aside close to a third of the public shares for retail buyers, routing the allocation through Robinhood, Fidelity, Schwab, SoFi and E*TRADE, so that the people who had watched the boosters come back and land upright could own a sliver of the company that brought them down. The raise — somewhere in the neighborhood of seventy-five billion dollars — dwarfed every debut before it by a wide margin. What the buyers were actually purchasing was not rockets. The reusable boosters are the spectacle. The cash machine sits overhead.
Starlink, the satellite-internet arm, now counts 10.3 million subscribers across 155 countries. In the first quarter of 2026 it booked $3.26 billion in revenue and $1.19 billion in operating income, and it has quietly become the majority of what SpaceX earns. Quilty Space, which tracks the industry, expects the subscriber base to approach seventeen million by year's end. The launches get the documentaries. The dishes pay the bills.
A Carrier Without Towers
The part of the business that unsettles telecom executives is newer, and stranger. It is called direct-to-cell, and it does away with the dish. The satellites talk to an ordinary phone — the one already in a pocket, unmodified, no antenna bolted on — by behaving like a cell tower that happens to be moving at seventeen thousand miles an hour. T-Mobile switched on the commercial version, branded T-Satellite, on July 23 of last year. It began as a way to send a text from a dead zone. By October it carried data. A hiker at the bottom of a canyon can now reach WhatsApp, Google Maps, AccuWeather, AllTrails, X. More than 650 of those satellites are already in orbit. Measured by the ground it covers, Starlink runs the largest 4G network on Earth, live in twenty-two countries, most of it strung over terrain no carrier ever found it worth the cost to wire.
For a man who has spent ten years selling connectivity into the world's empty corners — the mid-ocean ship, the offshore platform, the island that waits years for a submarine cable to reach it — the implication is not subtle. The expensive trench and the chartered cable ship were always the price of distance. Distance is the thing the satellite erases.
Musk's stated ambition runs well past filling dead zones. SpaceX has told regulators and partners that it intends to stand up a fourth American carrier — a peer to AT&T, Verizon and T-Mobile — that owns no towers at all. Its engineers describe a coming generation of satellites, lofted by Starship, carrying antennas many times larger than today's and flying in a lower orbit, the better to push a strong signal straight down through a roof rather than sideways through a city. The figure circulating among investors is a single fifty-dollar plan covering a phone and a home, anywhere — a desert, a deck in the middle of a lake, a pass in the mountains — at speeds that would embarrass most rural DSL. Whether the plan ever prints at fifty dollars, no one knows. That it is being modeled at all is the point.
As long as Starlink rode T-Mobile's frequencies, it was a feature. With its own licenses, it can become the network.
What the Spectrum Bought
None of it works without airwaves, and until last autumn SpaceX did not hold the right kind. It borrowed T-Mobile's. That changed in September, when a debt-laden satellite-television company called EchoStar — long run by the cable veteran Charlie Ergen, and by then under federal pressure for sitting on spectrum it had never fully used — agreed to sell SpaceX a band of licensed frequencies. The AWS-4 and H-block airwaves occupy the mid-band sweet spot: low enough to slip through walls, high enough to carry real data. The price was roughly seventeen billion dollars, split evenly between cash and SpaceX stock. In November the two enlarged the deal, adding another slice of licenses for about 2.6 billion in stock, along with an arrangement under which SpaceX would cover some two billion dollars of interest on EchoStar's debt through 2027.
Owning spectrum is what separates a guest from a host. As long as Starlink rode T-Mobile's frequencies, it was a feature — a supplement bolted onto another company's network, useful in the gaps. With licenses of its own it can become the network. EchoStar's Boost Mobile customers are to be folded in as the first subscribers of a carrier whose infrastructure mostly orbits the planet. The deal pulled EchoStar back from the edge of insolvency. It handed SpaceX the one asset that cash alone cannot conjure overnight: a legal claim to a stretch of the radio spectrum over the United States.
The arrangement carries an awkwardness that neither party advertises. T-Mobile is the partner that let Starlink touch a phone at all; its towers and its airwaves turned a science project into a service that could text from the backcountry. The EchoStar spectrum loosens that grip. A carrier able to reach any handset on frequencies it owns outright no longer needs a host, and a host, given enough time, starts to look like a target. Musk has not said when the supplement is meant to become the substitute. He has simply bought himself the option to find out.
The Cable Math Stops Working
Which brings the story back to Stamford. Charter's bleed is not yet a satellite story. Most of its departing customers are leaving for fiber, or for the fixed-wireless home internet that Verizon and T-Mobile now beam off their 5G towers. The orbital threat is the next wave, not this one. The direction is the problem. Broadband was supposed to be the durable franchise — the fat pipe everyone needed and no rival could cheaply duplicate. In 2025, Comcast and Charter together shed more than 1.1 million broadband subscribers. Charter's first-quarter revenue then slipped one percent, to $13.6 billion. Its video business continues to melt; that line fell more than nine percent as households kept cutting the cord. Earnings came in at $9.17 a share against the $10.63 Wall Street wanted. The company built to be unkillable is being nibbled from four directions at once.
Chris Winfrey, Charter's chief executive, has an answer, and he repeats it like a man who needs it to be true. The future, he argues, is convergence — internet and mobile sold as one bundle, "better, faster, and lower cost," across the whole of Charter's footprint. He has begun telling investors that the number to watch is no longer broadband subscribers but "converged ARPU," the revenue from a customer who buys both, "and that is growing." Returning to broadband growth, he said this spring, is the company's "priority one, two, and three." The mobile half of the bundle, Spectrum Mobile, is in fact flourishing: 368,000 new lines in the quarter, more than twelve million in total, up seventeen percent over the year.
There is a catch folded into that success. Charter does not own a mobile network. Spectrum Mobile is a reseller, riding on Verizon's towers under a wholesale agreement. And Verizon, hedging against the same dead-zone problem that gave Starlink its opening, has tied itself to a satellite startup to beam coverage to its own customers from space. Charter's lifeline runs straight through a competitor that is busy buying insurance from orbit.
The Rivals Overhead
SpaceX does not have the sky to itself. A Texas company, AST SpaceMobile, has spent years building what amount to cell towers in space — vast satellites, the BlueBird line, that unfold phased-array antennas the size of tennis courts. Where Musk favors a swarm of thousands of small craft, AST is betting on a handful of giants. The bet carries serious backing: AT&T and Verizon have both invested, alongside FirstNet, the federal authority that handles emergency responders, and in April the FCC cleared AST to operate a 248-satellite constellation and sell direct-to-cell service. The American sky is sorting itself into two camps. T-Mobile's customers will reach space through Starlink; AT&T's and Verizon's will reach it through AST. The incumbents, it turns out, are not all standing still to be killed. Some are trying to own the weapon. AST sits well behind Starlink in orbit — it has launched a fraction of the satellites and means to add another forty-five to sixty of its next generation by year's end — yet its financiers are precisely the carriers Musk intends to unseat. The fight above the atmosphere is, in good part, a proxy for the one being waged on the ground.
The Math Problem at Two Trillion
The number that opened this story is also its loudest skeptic. More than two trillion dollars works out to north of seventy times the revenue SpaceX is expected to earn this year — a multiple that presumes the direct-to-cell dream arrives roughly on time and roughly as advertised. Analysts who have run the figures, among them several quoted by Benzinga in the weeks before the listing, called the valuation hard to defend on anything but faith in the next decade. Physics is less generous than a pitch deck. A single satellite beam blankets an enormous patch of ground, and everyone beneath it shares the same finite capacity. Over an empty desert that is a marvel. Over a crowded city it is a bottleneck. Pushing a text or a map tile to a phone is one task. Replacing the gigabit line in a downtown apartment, where one terrestrial tower already buckles under thousands of users, is a far heavier one. Where there are no towers, Starlink has no rival. Where there are many, the old arithmetic still holds.
The sober case for the stock is not that satellites replace the ground network. It is that they peel off its worst parts — the lonely rural homes, the empty interstates, the open ocean — the routes that never earned back the cost of reaching them, and leave the incumbents to brawl over the dense middle on thinner and thinner margins. That is not a clean kill. It is a slow squeeze, applied from above.
Charter, for its part, is finishing what Winfrey calls its "generational" investments — a rural buildout, a network upgrade — projects drawn up in an era when the only way to reach a far-off house was to dig a trench to its door. The trenches are nearly done. Above them, more than 650 satellites already pass overhead every ninety minutes, and SpaceX is launching more of them each week. The trench and the orbit are now racing to the same remote house. Only one of them gets cheaper every year.
