SpaceX 2029 Valuation Case Rests on Starlink and a Decade of Flawless Execution

The SpaceX 2029 valuation case is straightforward to state and genuinely difficult to underwrite: SPCX trades at $116 today, sits on 597 times forecast 2026 earnings, and looks expensive by almost every conventional measure. The counter-argument is that the multiple compresses hard once growth arrives, and the question is whether execution arrives on schedule.

What the IPO Filing Actually Tells You

The June 2026 IPO priced at $135.00 per share, with SpaceX’s 424B4 final prospectus on SEC EDGAR confirming 555,555,555 Class A shares sold for a total offering of approximately $75 billion. Of that, underwriting discounts of $0.90 per share took $500 million off the top, leaving gross proceeds to the company of roughly $74.5 billion. SpaceX had raised only $9 billion of primary equity capital in the entire period before the IPO, per its IPO factsheet filed with the SEC.

Control is a separate issue worth reading carefully. Elon Musk holds approximately 82.4% of voting power following the listing, with Class B shares carrying 10 votes each against one vote for Class A. SpaceX is classified as a controlled company under Nasdaq rules. UK retail investors who participated through the Marex Retail Offer platform, which closed on 10 June 2026, hold Class A shares with effectively no governance influence.

The financials reported in the prospectus give a clearer baseline than the current share price alone. Full-year 2025 revenue was $18.7 billion, up 33%, per the IPO factsheet. The Connectivity segment (Starlink) generated adjusted EBITDA of $7.2 billion in 2025, up from $3.8 billion in 2024. The Space segment generated $0.7 billion after absorbing $3.0 billion in Starship R&D. These are non-GAAP measures. On a GAAP basis, SpaceX’s S-1 prospectus showed a full-year 2025 operating loss of $2.589 billion and Q1 2026 revenue of $4.694 billion alongside a $1.943 billion operating loss.

The SpaceX 2029 Valuation Case: Where the Numbers Live

Consensus forecasts have SpaceX earning $4.69 per share in 2028 and $10.80 by 2030, implying forward P/E ratios of 24.8 times and 10.8 times at $116. On revenues, the stock sits on 38 times 2026 estimates, falling to 4.1 times by 2030. For comparison, the S&P 500 trades at around three times sales. That compression is the investment argument in its entirety.

The factsheet reported approximately 10.3 million Starlink subscribers as of 31 March 2026, across 9,600-plus satellites in orbit and service in 164 countries. The snippet cites 12 million by June, with Quilty Space projecting 16.8 million by December on $11.3 billion of consumer revenue and Payload estimating 18.4 million. The 2030 bull cases run from MoffettNathanson’s 70 million subscribers to Morgan Stanley’s 116 million. Roughly 2.6 billion people still lack reliable broadband, and Starlink has no comparable orbital rival.

Getting from $18.7 billion in 2025 revenue to the $365 billion some analysts model by 2030 requires near-flawless execution. The Starship cost thesis is central: the Space Shuttle put a tonne into orbit for roughly $54 million, Falcon 9 does it for about $2.7 million today, and Starship targets a few hundred thousand. If that cost reduction arrives on schedule, the economics of both launch and satellite deployment change materially. If it slips, the 2030 multiples never arrive.

The corporate structure has also changed since IPO. SpaceX completed its acquisition of Anysphere, the developer behind the Cursor AI coding tool, on 14 August 2026. Under the terms disclosed in SpaceX’s Form 8-K, shareholders of Cursor received 389,289,254 SpaceX Class A shares based on an implied equity value of $60 billion for Cursor. The prospectus financials were also recast to include xAI and X Holdings Corp following acquisitions between entities under common control.

Cerebras: A Faster Compression Curve, Narrower Analyst Coverage

Cerebras (CBRS) at $193 presents a different version of the same bet. Consensus puts earnings at $5.85 per share in 2028 and $16.55 by 2029, bringing it from 32.9 times to 11.6 times. Revenue multiples drop from 49 times to 2.4 times over the same period.

The snippet describes Cerebras as loss-making. The picture from its May 2026 S-1, as reported by TECHi, is more layered: 2025 GAAP net income was $237.827 million, but the GAAP operating loss was $145.862 million, with the net income figure reflecting a fair-value gain rather than operating profitability. Total 2025 revenue was $509.991 million. Separately, Cerebras’s original 2024 IPO registration, which was filed with Citigroup and Barclays as lead managers, was delayed by national-security scrutiny around Abu Dhabi-based G42, which had accounted for $119.1 million, or 87% of Cerebras’s revenue in the first half of 2024.

To meet the 2029 forecasts, Cerebras needs roughly 20-fold revenue growth in four years while competing against Nvidia and AMD in AI inference workloads. Very few analysts model beyond 2028. If execution slips, the 49 times sales multiple never falls.

Both companies are well capitalised. Cerebras holds $2.23 billion of cash against $1.36 billion of debt. SpaceX carries $30.6 billion of debt and $23.7 billion of cash, leaving net debt of under $7 billion on a $1.5 trillion enterprise: manageable at that scale.

Neither stock is cheap at today’s prices. Both are cheap on 2029 numbers, and that gap is the entire bet. The next material test is whether Starship commercial operations and Starlink’s subscriber trajectory through the second half of 2026 print ahead of or behind the growth curve embedded in current consensus.