The SpaceX IPO: We Have Liftoff

A Different IPO Era: Why the Game Has Changed

The landscape that produced history's greatest IPO returns no longer exists. In the late 1990s, roughly 8,000 companies traded on U.S. exchanges. Today that number has been cut nearly in half to approximately 3,900. This is a function of rising regulatory burdens, abundant private capital, and a wave of merger & acquisition that has steadily consolidated the public market.

The consequence is that companies now stay private far longer. The venture ecosystem, flush with capital, can fund businesses well past the point where they would historically have gone public. By the time a company does list, the early-stage growth story is largely played out and the valuation reflects it. A slight majority of IPOs are underwater after three years, and roughly 65% underperform their public peers. The reasoning is intuitive: strong companies tend to IPO when it is most advantageous to be a seller, near peak valuation.

There are, of course, spectacular exceptions. Amazon, Nvidia, and Tesla generated returns north of 100x for early public shareholders. But each of those companies IPO’ed when their market cap was a tiny fraction of the largest publicly traded firms at the time. Amazon went public at roughly $400 million. Nvidia at under $700 million. Tesla at around $1.7 billion. The runway ahead of them was immense.

SpaceX listed on June 12 at a valuation of approximately $1.8 trillion, already among the five largest companies in the world. Anthropic and OpenAI are similarly valued in private markets near or above the trillion-dollar mark. These are extraordinary businesses, but the math is fundamentally different at this scale. A 100x return from a $1.8 trillion starting point would imply an $180 trillion company. Roughly double the current value of every stock in the S&P 500 combined. The opportunity is real, but investors should frame expectations accordingly.

Three Businesses, One Ticker:

SPCX, is three distinct businesses at very different stages of maturity, bundled under a single ticker.

Starlink (“Financial engine”)

• The satellite broadband business generated $11.4 billion in revenue in 2025, 61% of the company’s and $4.4 billion in operating income.
• Explosive subscriber growth: 10.3 million customers across 164 countries as of Q1 2026, up from 5.0 million a year earlier, with 750,000 to 1.5 million new subscribers being added per month.
• Starlink is currently the only segment generating meaningful profit at scale.

Space and Launch (“Company identity”)

• Remains the company’s identity but not its profit center.
• Generated $4.1 billion in revenue in 2025, with Falcon 9 commanding an estimated 90% share of global commercial launches by mass to orbit.
• This business operates at a loss, weighed down by approximately $3 billion in annual Starship R&D spending. But this is the business that creates everything else. Falcon 9’s reusability made Starlink economically viable.
• Starship, if it delivers on its promise of dramatically lower cost-per-kilogram to orbit would unlock an entirely new category of opportunities: Orbital data centers, deep-space logistics, and infrastructure deployments that are simply not feasible on any other rocket.
• The launch business doesn't generate the profit; it generates the optionality.

xAI (“Wildcard”)

• Following the February 2026 merger: the AI segment posted $3.2 billion in revenue and a $6.4 billion operating loss in 2025.
• Its Grok models lag peers in capability and commercial traction, yet the segment has become the source of the company’s most surprising recent revenue catalyst.
• xAI is renting compute infrastructure to others.

In late May, Anthropic signed an agreement to pay SpaceX $1.25 billion per month through May 2029 for exclusive access to the Colossus 1 data center in Memphis, roughly 220,000 NVIDIA GPUs. On June 5, Google signed a separate agreement at $920 million per month through June 2029 for approximately 110,000 GPUs, describing the deal as "bridge capacity" for surging Gemini Enterprise demand.

Combined, these two contracts represent approximately $26 billion in annualized compute revenue. This is a reflection of extraordinarily tight GPU supply in a compute-starved market, not necessarily a validation of xAI's own AI capabilities. SpaceX has effectively become a neo-cloud at a premium to market. That is a powerful near-term revenue story, but it is a supply-scarcity story, and investors should understand the distinction.

Push and Pull: The Stock's Supply and Demand Dynamics

It is very difficult to justify the valuation on current fundamentals without taking an extremely optimistic long-term view. And yet, the market has spoken loudly. The IPO raised approximately $80 billion but generated orders in excess of $250 billion, with roughly $100 billion of that demand coming from retail investors alone. Trading volume in the first three days has been massive, with the stock surging from its $135 IPO price to above $200. Elon Musk commands perhaps the most devoted retail following of any executive in public markets today, and that enthusiasm was on full display from the moment the ticker went live.

In the near term, there will be some additional passive flows as SPCX is added to broad market and large cap indexes over the coming weeks. These are mechanical buyers, but not transformational ones on their own. The more significant structural event lies roughly a year out.

The U.S. equity market is valued at over $75 trillion, with roughly half of that sitting in passive strategies. Of that passive universe, approximately half tracks the S&P 500, meaning roughly 25% of the entire U.S. market is indexed to a single benchmark. On top of that, a significant portion of active money is also benchmarked to the S&P 500, meaning underweight positions create their own buying pressure. If SpaceX earns inclusion in the S&P 500, the structural demand would be enormous, but that is at least a year away. The index requires a 12-month listing history and demonstrated Generally Accepted Accounting Principles (GAAP) profitability, among other criteria. SpaceX appears poised to clear the profitability hurdle, in large part because of the recently announced data center compute deals with Anthropic and Google that dramatically improve the consolidated earnings picture.

In the meantime, we expect the stock to be very choppy. On the supply side, only about 4% of total shares are freely tradable today, with the remaining 96% locked and set to unlock in staggered waves over the next 12 months. Roughly 20% of restricted shares unlock after Q2 earnings, with additional tranches releasing at regular intervals through a full 180 day unlock in December 2026. Importantly, Musk owns approximately 50% of the company and his shares are subject to a separate 366-day lockup, but even beyond that restriction, there is little reason to expect him to be a seller. That effectively removes half the share count from the supply equation on a structural basis. Still, as each unlock window arrives, new supply will meet whatever demand exists at that moment, and the small float will amplify price swings in both directions.

The net picture is one of push and pull. Near term retail enthusiasm and modest passive index flows on one side, met by a thin float and systematic supply unlocks on the other. A year from now, if SpaceX secures S&P 500 inclusion, the demand story changes materially. Until then, investors should expect volatility and size positions accordingly.

Source: CI GAM Research and Space Exploration Technologies - S-1

How We’re Positioned

We participated in the SpaceX IPO allocation in CI Global Artificial Intelligence Fund (ETF ticker: CIAI). Our ask was heavily paired back consistent with the offering being approximately 3x oversubscribed with demand reportedly exceeding $250 billion. We received shares and are treating the position with the careful, measured approach that the analysis above would suggest.

The business is genuinely extraordinary. Starlink is a globally scaled, high-margin infrastructure platform with no close peer. The launch business commands a near-monopoly on commercial access to orbit. The xAI compute deals, while driven by market scarcity rather than AI model superiority, represent real contracted revenue at scale. The long-term optionality: orbital data centers, Starship commercialization, direct-to-cell connectivity, is large and difficult to value using traditional frameworks.

Ultimately, this is not a typical IPO, and we are not treating it like one. The valuation on current fundamentals is difficult to justify by conventional metrics. The stock supply dynamics are complex and will play out over months, and as our broader IPO analysis underscores, even the best companies can be challenging investments when purchased at peak market enthusiasm. The opportunity is interesting and one to be treated carefully. 

To open this article in a shareable format, click here

GLOSSARY:

Correlation: A statistical measure of how two securities move in relation to one another. Positive correlation indicates similar movements, up or down, while negative correlation indicates opposite movements (when one rises, the other falls).

Volatility: Measures how much the price of a security, derivative, or index fluctuates. The most commonly used measure of volatility when it comes to investment funds is standard deviation.