Post-IPO study
SpaceX listed at a $1.77 trillion valuation, the largest IPO in history. Two months later the stock had round tripped back to its offer price. I went through the S-1 and the trading record to work out why.
Overview
I had been watching the SpaceX IPO rumors for years, so when the company filed its S-1 and listed on the Nasdaq as SPCX on June 12, 2026, I wanted to dig into it properly instead of just reading headlines. This was not a normal listing. SpaceX raised about $75 billion at an implied $1.77 trillion valuation, more than three times the size of the previous record.
It is also not really one business anymore. SpaceX launches rockets, runs the Starlink satellite internet network, and after merging with xAI in February 2026 it owns a frontier AI lab and X as well. Buying the stock means buying all three at once.
My argument: the valuation was priced for near flawless execution across three very different businesses. Starlink genuinely supports a large part of the price tag, but the stock's slide since June suggests the market is already discounting the AI segment's losses and Musk's concentrated control more than IPO day enthusiasm implied.
Key points
Priced at $135, opened at $150, peaked at $225.64 four days later, then fell to an all time low of $108.27 after the first earnings report as a public company. It now trades near the original offer price, with a beta above 6, closer to a speculative small cap than to the company that launches most of the world's commercial payloads.
Connectivity brought in $11.4 billion, about 61 percent of revenue, growing near 50 percent a year at a 39 percent operating margin. Subscribers went from 2.3 million in 2023 to over 10 million by early 2026. This is the strongest argument for the valuation.
The AI segment earned $3.2 billion and lost $6.4 billion, while absorbing $12.7 billion of $20.7 billion in total capital spending. The launch business, meanwhile, sits near a mature revenue ceiling unless Starship meaningfully expands capacity.
Dual class shares give Musk roughly 80 percent of the voting power on about 42 percent of the economic ownership, with no sunset provision. Buying the stock is closer to backing one person's judgment indefinitely than to taking a normal ownership stake.
Context
I compared SPCX to Cerebras, an AI chip company that listed a month earlier, and the shapes match: a huge debut pop, then a hard giveback. Cerebras sat about 34.7 percent below its debut level, a slightly deeper pullback than SpaceX's roughly 30 percent off its peak. Across 25 major 2026 IPOs the average return was 18 percent but the median was negative 17 percent, so a few outliers carry the average while the typical new listing is down.
That reframed the chart for me. The drawdown looks less like something specifically wrong with SpaceX and more like the standard pattern for 2026's most hyped, most capital hungry debuts: a scarcity driven pop, then a reality check once real numbers and lock up expirations arrive.
The paper
Disclaimer: This is an independent student research paper prepared for academic purposes. It is not professional financial, legal, or investment advice, and nothing in it is a recommendation to buy, sell, or hold any security. All figures on stock price, revenue, segment performance, and valuation reflect public reporting and company filings available as of August 11, 2026. Markets and disclosures change quickly, so some figures may already be outdated, approximate, or subject to revision. Sources sometimes disagreed, and the most consistent figures across sources were used where possible.