SpaceX: A Great Business or a Great Narrative?

"When something is important enough, you do it even if the odds are not in your favor."

– Elon Musk, on founding SpaceX

Today.

Today is the day.

The biggest IPO season in the history of capitalism begins.

SpaceX. Ticker: SPCX. Priced last night, first trade this morning on the Nasdaq. Put it on your watchlist. It’ll be changing hands every single day from here on, on every screen, in every headline. And it’s only the opening act. Behind it wait Anthropic (Claude) and OpenAI (ChatGPT), each rumored to follow at around a trillion this summer. Three companies. A combined expected value of roughly $3.75 trillion — more than every single dot-com IPO from 1995 to 2000 put together (chart 1).

The numbers are difficult to even hold in your head. SpaceX is targeting $135 per share, a raise of roughly $75 billion, at a valuation of $1.75 trillion. The previous record IPO — Saudi Aramco in 2019 — raised $29 billion. SpaceX more than doubles it.

Let that valuation sink in for us in Switzerland. $1.75 trillion is more than seven Nestlés. Switzerland’s beloved, most venerable company — founded in 1866, in your kitchen every single morning — fits into this rocket maker roughly seven times over. On Day One.

Or go bigger. Take the entire SMI — Nestlé, Novartis, Roche, UBS, Richemont, all twenty of Switzerland's blue-chip champions, built over a combined centuries of profit. Put together, they're worth about the same as this single, loss-making rocket company. On its first afternoon of trading.

And here’s the part that should stop you cold. That same Nestlé, a company that has been in business for 159 years compared to SpaceX’s 24, consistently generates a net profit of $11 billion annually on revenue of $108 billion, every year, reliably.

SpaceX lost nearly $5 billion last year.

One is a money machine the market values at a seventh of the dream. The other is the dream.

We flagged this moment back in April in The IPO Class of 2026. Now it’s here. And the man at the centre of it owns roughly 42% of the whole thing — yes, do the math 😉 — while controlling 85% of the votes.

Let’s dig into the biggest IPO ever written.

And Elon Musk’s 42% stake.

SpaceX.

Chart 1: Expected IPO value of SpaceX, Anthropic & OpenAI (~$3.75T) vs. all Dot-Com IPOs 1995–2000 (~$3.0T)

SpaceX: What Are You Actually Buying?

So, what is SpaceX, beneath the mythology?

Strip it down and there are really three businesses stitched into one cap table (chart 2):

1. Space — the Falcon and Starship launch business, the most dominant rocket franchise on the planet, flying more often than every other nation and competitor combined.
2. Connectivity — Starlink, roughly 10 million subscribers across 164 countries, beaming internet from low orbit.
3. And since February 2026, there’s AI — xAI and Grok, folded into the company through a merger and presented as if it had always belonged.

Here is the fact the narrative tends to skip: only one of those three makes money.

Starlink generated $11.4 billion of SpaceX’s $18.7 billion in 2025 revenue and threw off around $7 billion in segment EBITDA. It is the cash engine. And almost all of that cash is poured straight into the two businesses that burn it — Starship, which devours capital by design, and AI, which lost over $6 billion in a single year.

The group as a whole?

A net loss of nearly $5 billion in 2025.

Fun factburied in the prospectus of Musk’s new pay package?

Part of it only vests if SpaceX puts one million people on Mars. Not a metaphor — an actual line item in an actual SEC filing, tying the CEO’s pay to founding a city on another planet. Whatever else you make of it, nobody can accuse the man of thinking small.

Make no mistake — the bull case is real.

SpaceX owns roughly two-thirds of all active satellites in orbit, and launches more mass to space than the rest of the planet, governments included, combined. Starlink is adding subscribers north of 50% a year at fat margins. If reusable rockets and orbital internet are the railroads of the next century, SpaceX owns the gauge, the track, and the timetable. This is not Pets.com. This is, plausibly, the most important company of the century.

The bulls aren’t fools.

300 million Starlink subscribers are estimated by 2036 — up from 10 million today — enough, on its own, to justify the entire $1.8 trillion. If they’re right, today’s price is a bargain.

This is the Elon Musk operating system in its purest form: one profitable machine feeding a constellation of moonshots. If you want to understand the man wiring it all together, our summer reading pick — The Book of Elon by Eric Jorgenson — dropped this Wednesday, and it reads like the instruction manual to exactly this kind of empire-building.

But here’s the thing about a man selling a dream this big. The dream has a price. And the goal, always, is to price it as high as humanly possible. How high?

Let’s read the filing.

Chart 2: SpaceX 2025 revenue & operating profit by segment ($bn)

SpaceX’s Big Dreams and Big Expectations

Here is how you price an IPO: you don’t sell what the company is. You sell what it could become.

And SpaceX has written the most ambitious “could become” in the history of the SEC.

Depending on who you read, this IPO will either end Western civilization or usher in a global utopia. The truth almost surely lies in between.

Its filing claims a total addressable market of $28.5 trillion. Larger than the entire GDP of the United States. The breakdown: Space at $370 billion, Connectivity (Starlink) at $1.6 trillion, and AI at $26.5 trillion — of which $22.7 trillion is labelled “enterprise applications” (chart 3).

Read that again.

93% of the addressable market SpaceX is selling you is AI software it does not make, does not sell, and does not own. This is a rocket-and-satellite company with two large — and cancellable — AI compute contracts (Google’s Alphabet and Anthropic).

Let’s be honest, every IPO inflates its TAM. SpaceX just set the all-time record.

And the future markets, copy-pasted straight from the “Future Markets” section of the prospectus, read like a teenager’s notebook:

• space tourism
• in-orbit manufacturing
• passenger and cargo transport to the Moon and Mars
• energy production on the Moon and Mars
• point-to-point terrestrial travel
• and — yes — asteroid mining

This is a real document. Filed with the SEC. When a company asks roughly 90 times sales today on the premise that asteroid mining will one day fund the cash flows, it is not selling a stock.

It is selling a worldview.

Consider the AI line alone. To justify $26.5 trillion, SpaceX is effectively asking you to believe Grok becomes one of the most valuable software franchises on Earth — on the back of two compute contracts that either side can cancel with 90 days’ notice.

In our view at arvy, that is not a moat. That is a handshake.

That worldview might even turn out to be right. The question — the only question — is whether you want to buy it on Day One, at this price.

And here lies the real risk — not the liquidity swirl around the listing, but who holds the controls. Elon Musk owns roughly 42% of the equity, yet commands 85% of the votes and the entire board. According to basic ESG criteria, officials called it “unprecedentedly bad.” An investment in SpaceX is, quite literally, an investment in one man’s judgment — his vision, his timelines, his whims. Tesla shareholders know exactly what that feels like. We unpacked the wider scramble in our notes on SpaceX at 90x sales — who’s buying when the smartest investors are selling.

A great company and “Good Story”? Possibly. A great investment, today, at this price, in a company where one person answers to no one?

That’s a very different question.

Time for the “Good Chart”.

Chart 3: SpaceX’s estimated TAM by segment — $28.5T total

SpaceX’s Good Chart That Doesn’t Exist Yet

Here’s the problem with SPCX, and it’s one no analysis can solve.

There is no chart.

At arvy, we have three rules for any IPO, and we laid them out in full in The IPO Class of 2026:

1. Never buy on Day One
2. Wait for the «Good Chart» to form — enough price history to read the trend, the support, the real level
3. Ask whether you’d buy it at this price if it were already public.

For SpaceX, all three point the same way. There is no base. No support. No trend. Just a fixed price, a roadshow, and a story so big it needs a telescope.

So instead of a real chart, look at how these stories usually go (chart 4).

The pattern repeats across almost every hyped debut. An initial pop as retail piles in. Then the slow, grinding hype unwind as expectations meet reality — often down 50% or more. Then the part nobody has the patience for: dead money. One to four years of going nowhere while the company quietly grows into its valuation and builds, at long last, a base.

And then — only then — the real move.

It happened to Alibaba. To Meta, which traded below its IPO price for over a year before becoming a trillion-dollar company. To Airbnb which is still at the same level after six years. To the entire Class of 2021. Even Amazon itself fell nearly 90% after listing before it became Amazon. The opportunity was real. But the timing was everything.

Great companies at great-narrative prices have humbled a generation of investors before. They will do it again.

We’ve watched this movie. We even wrote the review: the IPO Class of 2021 lost investors up to 97% from their Day-One highs. Robinhood (-92%). Coinbase (-92%). Rivian (-95%). Oatly (-97%). All companies everyone knew, everyone used, everyone wanted to own.

A great business and a great investment are not the same thing. The price you pay decides which one you get.

SpaceX may well be the most important company of the next fifty years.

That is precisely why there’s no rush to overpay for it on the first afternoon. You don’t have to buy all the stock you want on Day One.

Ask us again after the first lock-up expires. After the first earnings print. After the chart has eighteen months of price behind it. That’s when we’ll have something worth saying.

The rocket launches today. The Good Chart launches later.

We’ll be waiting for it.

Chart 4: How hyped IPOs typically trade — pop, unwind, dead money, then lift-off (illustrative, not a prophecy)