Returnolio
Illustration: a launch pad at dawn seen from a distance across flat scrub, a rocket standing on it, and in the foreground a wooden lectern with a thick bound prospectus open on it

Stock analysis · · 6 min read

The SpaceX IPO already happened. Here is what the filings say you bought

SpaceX listed on 12 June 2026 at $135 and trades at $153. Its first quarterly report as a public company is on the SEC's site: $7.8 billion of revenue in three months, a $541 million loss, $28 billion of spending on equipment in six months, and a rocket company that now owns xAI and X. What the numbers say, from the filing.

By Thomas

Most people searching for "SpaceX IPO" are three months late. It happened on 12 June 2026: Class A shares at $135 on Nasdaq under SPCX, reported as the largest initial public offering ever done (CNBC and Wikipedia, read 22 September 2026). The filing itself puts the money raised at $85.7 billion. The interesting part is what came next, and almost nobody has read it.

When a company lists, it starts filing. SpaceX's first quarterly report as a public company went to the SEC on 4 August 2026, and everything below is from it. A quarterly report is unaudited, and it is still the thing the pre-IPO speculation could never give you: the company's own numbers, on the public record, free.

What the price has done

$135 at the offer on 12 June. The lowest close since then was $108.27, the highest $185, and on 18 September it closed at $153. So anyone who bought at the offer is up about 13% in three months, and anyone who bought on the first excited morning may well be down. That is normal for a new listing and it is the first thing to know about buying one: the IPO price is not a floor.

Our price table has had SPCX since the first day of trading, which is how those numbers are dated.

The three months in the filing

For the quarter to 30 June 2026:

Line$bnAgainst a year earlier
Revenue7.814.07, up 92%
Cost of revenue3.502.28
Research and development3.551.96
Operating lossminus 0.14minus 0.97
Net lossminus 0.54minus 1.01
Loss per shareminus $0.09minus $0.34

Source: Space Exploration Technologies Corp., Form 10-Q for the quarter ended 30 June 2026, filed 4 August 2026 (accession 0001628280-26-052535), via the SEC's XBRL data.

Revenue nearly doubled in a year and the loss shrank by half. The operating line is almost break-even: minus $143 million on $7.8 billion of revenue, which is a rounding error away from a company that pays for itself. Below the operating line sits $629 million of interest in the quarter, which is what turns a small operating loss into a real one.

Read the research line before you read anything else. $3.55 billion in one quarter, 45% of revenue, nearly double last year. This is a company choosing to spend its gross profit on building the next thing. You can call that visionary or reckless; what you cannot call it is hidden.

The balance sheet after the listing

At 30 June 2026$bn
Cash93.5
Property, plant and equipment (net)65.7
Goodwill11.6
Total assets192.8
Debt including leases39.4
Money customers have paid in advance14.3
Total liabilities65.5
Equity127.2

Source: the same 10-Q. Equity was $2.6 billion at 31 December 2025.

Cash went from $24.7 billion at the end of December to $93.5 billion, and the IPO brought in $85.7 billion of that. Equity went from $2.6 billion in December to $34.5 billion in March, before the listing, and $127.2 billion in June: a recapitalisation first, then the offering. Either way, a company with nearly a hundred billion in the bank to spend on rockets.

Two lines are worth pausing on. $14.3 billion of money customers have already paid for launches and service not yet delivered, up $2.2 billion in six months: a customer prepaying is the best kind of liability, and a growing one is a booking book you can see. And $47.5 billion of contracted revenue not yet recognised, the backlog, disclosed in the same filing. That is six quarters of revenue already signed.

Against that, $28.5 billion spent on property and equipment in six months against $3.5 billion of cash from operations. SpaceX is spending roughly eight dollars on new capacity for every dollar the business generates, and the IPO cash is what pays for the difference. That is the trade the shareholder made: dilution now, capacity later. Whether it works is a question about launch demand in 2029, not about this quarter.

The thing nobody mentions

Buried in the same filing: on 2 February 2026 SpaceX completed the acquisition of X.AI Holdings, so xAI is now a wholly-owned subsidiary, and xAI had itself acquired X (the former Twitter) in March 2025. The filing treats it as a merger of businesses already under common control, which means it created no purchase price: the $11.6 billion of goodwill was already on the December balance sheet, carried over from xAI's own purchase of X, and the February merger added $3 million. It also means last year's comparison figures above were restated to include xAI, so the 92% is like for like.

So SPCX is not a pure rocket company. It is a rocket and satellite business that also owns a frontier AI lab and a social network, and the litigation section of the 10-Q carries the lawsuits of all three. If you bought the ticker for Starship, you also bought that. This is exactly the kind of thing a prospectus tells you and a headline does not, and it took me ten minutes to find.

What this says about the next one

OpenAI is the name people ask about now. As of this week it is still private: it confirmed a confidential S-1 in June 2026, completed a $7 billion employee tender at a reported $852 billion valuation in August (CNBC, TechCrunch and Quartz, read 22 September 2026), and is reported to be considering waiting until 2027. Until it lists there is no quarterly report to read, and the secondary platforms quoting a price per share are quoting a market of a few thousand accredited buyers, not a company's numbers. I wrote about that shape in the Anthropic post; it applies here unchanged.

Which is the useful lesson from SpaceX. The years of "how do I buy it before the IPO" produced nothing a normal investor could act on. The week it listed produced a quarterly report anyone can open, a backlog, a real price, and a chart that has already been 20% below the offer and 37% above it. The waiting was not the opportunity. The filing is.

Reading a newly listed company, in order

  1. Price against the offer. Where is it now, what was the range, and how much of the float is still locked up? A lock-up expiry is a known supply of shares arriving on a known date.
  2. Revenue and its growth, then the loss. A shrinking loss on doubling revenue is a different animal from a steady loss on flat revenue.
  3. Cash against cash burn. SpaceX has $93.5 billion and spent $28.5 billion on equipment in six months. Divide: this is the number that tells you whether the next raise is years away or quarters.
  4. What came with the ticker. Subsidiaries, acquisitions, dual share classes, who controls the votes. All in the first filing.
  5. Then a price. Not before.

Every company we cover, listed last week or listed in 1962, gets the same page: the accounts from the filings, and the price against what those numbers would be worth at its industry's multiple and its own. The Stock Deep Dive shows you that page for a company you own.

Educational and proprietary. This explains what our research does, not the exact formula behind it, and it is not personalised investment advice. See the full disclaimer.

Read next