- One ticker, three companies: SPCX bundles a profitable launch-plus-Starlink cash engine, a grounded Starship moonshot, and the freshly merged xAI, priced as if all three already work.
- The price is a promise: less than 7% of the ~$1.75 trillion valuation is backed by current profit; the rest is a bet on future growth.
- Wall Street cannot agree: published targets span roughly $63 to $227, Damodaran and Chanos on one side, ARK and the bulls on the other.
- The xAI question dominates: a ~$1.25 trillion AI merger sits inside the valuation, while a dual-class structure leaves Musk with more than 80% of the votes.
- The India footnote: days before the IPO, New Delhi froze Starlink’s commercial clearance over security concerns.
- On 12 June 2026, SpaceX became the largest stock-market debut in history. It priced on the Nasdaq as SPCX at $135 a share, raised about $75 billion, comfortably beating Saudi Aramco’s 2019 record, and promptly jumped more than 30% before closing up 19%. Within three days it traded near $178 and was, briefly, worth more than $2 trillion. The headlines wrote themselves: the rocket company is now worth more than most countries.
- But the interesting story is not the size of the number. It is what the number is actually paying for. Two details tell you to look closer. First, only about 4% of the company was floated, and a pop on a thin sliver of stock is easy to manufacture and tells you little about fair value. Second, by PitchBook’s reading of the prospectus, less than 7% of that $1.75 trillion is backed by current profits. The other 93% is a bet on the future. To understand the bet, it helps to see what most of the coverage glosses over: SpaceX is not one company. It is three very different businesses wearing a single ticker, and each comes with its own truth.
Company one: the cash machine that already works
The first company is the one that justifies taking SpaceX seriously at all. It has turned orbital launch into something close to a monopoly: it flew a record 165 rockets in 2025 and lifted more than 80% of all the mass humanity put into orbit that year, several times the rest of the world combined. The reason is reusability. Where rivals still throw their rockets away, SpaceX flies the same Falcon 9 boosters thirty-odd times, dragging the cost of reaching orbit down toward $2,700 a kilogram, a fraction of what anyone else charges. Bolted on top of that launch advantage is Starlink, the satellite-internet network that is the real money-spinner. Starlink brought in $11.4 billion in 2025, up nearly 50% in a year, and threw off about $4.4 billion of operating profit; by early 2026 it had passed 10 million subscribers across more than 100 countries. It is, today, the only part of SpaceX that actually makes money. Stop here and the bull case is easy: a near-monopoly launcher feeding a high-margin, recurring-revenue utility. The trouble is that you cannot stop here, because this is not mostly what you are paying for.
Company two: the moonshot still on the launch pad
The second company is the one the whole valuation leans on, and it is not finished. Starship Version 3, which flew for the first time on 22 May 2026, is the largest and most powerful rocket ever built, about 124 metres tall, with 33 engines producing roughly twice the thrust of the Saturn V that went to the Moon. It is also, right now, grounded. On that maiden flight the upper stage performed beautifully, but the giant booster lost control after separation and slammed into the Gulf at around 1,450 kilometres an hour; the US regulator declared a mishap and halted further flights pending an investigation. This matters because almost everything in the bull case runs through Starship: cheaper Starlink launches (a single Starship can carry around twenty times the bandwidth of a Falcon 9), NASA’s Artemis Moon landing, which the agency’s own inspector general has warned could slip further on spacesuit and lander delays, even as NASA still publicly targets 2028, and the Mars ambition that Musk still frames as the entire point of the company. The hardware is genuinely revolutionary. It is also still a test programme that, by SpaceX’s own record, fails about as often as it succeeds. You are being asked to pay today as though the hardest part is already behind it.
Company three: the AI bet bolted on at the last minute
The third company barely existed inside SpaceX a year ago. In February 2026, weeks before the listing, SpaceX merged with xAI, Musk’s artificial-intelligence venture, in an all-stock deal valuing the combined group at around $1.25 trillion. On paper it transforms the story: SpaceX’s prospectus pitches a total addressable market of $28.5 trillion, roughly the size of the entire US economy, of which 93% is not launch or satellite internet at all, but AI. Even sympathetic analysts choked on it. The valuation expert Aswath Damodaran said the figure “borders on fantasy”; one space economist called it “completely off-track”; another described the whole pitch as “an AI valuation wearing a rocket suit.” The reality underneath is heavier. xAI lost close to $5 billion in 2025 and burns more than $12 billion a year, and SpaceX took on a $20 billion bridge loan to refinance its new partner’s debt. The merger folded a cash-incinerator into a cash-generator just in time for the IPO. Whether that is visionary integration or financial engineering is, fittingly, the most contested question in the deal, and a smaller one nags at it: when Musk posted details of an AI-compute lease that appeared to contradict the prospectus, a Columbia law professor noted the choices were that either “the S-1 is materially misleading” or Musk was “up to his old hijinx.”
The number, and the people fighting over it
Put the three companies together and you get one of the widest disagreements on Wall Street. The bulls are buying the trajectory: Cathie Wood’s ARK sees $2.5 trillion by 2030, Morgan Stanley, an underwriter, so read it accordingly, models $3.4 trillion of revenue by 2040, and the veteran investor Ron Baron put another $1 billion in at the IPO, convinced he is holding a future ten-trillion-dollar company. The bears look at the same numbers and recoil. Damodaran pegs fair enterprise value near $1.3 trillion; Morningstar puts it at $780 billion, implying the stock could roughly halve; the short-seller Jim Chanos says it is priced on “hopes and dreams” at about ninety times sales; and the “Big Short” investor Steve Eisman simply called the valuation “silly.” Published price targets run from $63 to $227, a fourfold spread that is itself the message: nobody really knows. Two facts cut through the noise. The company is raising Starlink prices even as its average revenue per user has quietly fallen, to roughly $66 a month from about $86 a year earlier, a sign the growth is increasingly coming from cheaper customers in poorer markets. And whatever happens to the share price, Musk will control the company: the dual-class structure leaves him with more than 80% of the votes while owning around 40% of the equity, and ordinary shareholders even sign away the right to a jury trial. On this rocket, you are a passenger, not a co-pilot.
The India footnote that is bigger than it looks
For Indian investors there is a twist the global coverage skips. Starlink has spent two years inching toward India, it won a GMPCS licence and an IN-SPACe authorisation in 2025, was provisionally granted spectrum, and signed distribution deals with both Reliance Jio and Bharti Airtel after Prime Minister Modi met Musk. And yet, days before the IPO, India’s home ministry reportedly froze the final clearances, citing SpaceX’s cross-holding structure and security concerns, including Starlink’s alleged use during internet shutdowns abroad. The prize is real, India’s satellite-broadband market is forecast to reach roughly $1.9 billion by 2030, with hundreds of millions of rural Indians still offline, but the freeze is a useful reminder: SpaceX’s biggest growth markets are exactly the places where a single foreign-controlled network beaming connectivity from orbit causes the most sovereign discomfort. The same dominance the bulls celebrate is what makes governments hesitate.
The takeaway
None of this makes SpaceX a bad company. It is, by some distance, the most important space company on Earth, and the launch-plus-Starlink core is a genuinely formidable business. But a $1.75 trillion price tag is not a verdict on the company; it is a bet on which of the three companies inside it you think you are buying. Pay for the cash machine alone, and the number looks stretched. Pay for the moonshot and the AI dream, and you are buying a future that is still, quite literally, grounded, and trusting one man, holding more than 80% of the votes, to deliver it. The rocket is real. The question every investor now has to answer is how much of the price is rocket and how much is story. The next few Starship flights, and the next few quarters of Starlink cash, will start to tell us.
