A $3.5 Trillion Merger Run by Committees Musk Cannot Sit On

中文版 Chinese version

What is confirmed, reported, and merely estimated is labeled as such throughout, as of October 22, 2026. No merger agreement exists. Valuations are market estimates, not deal terms. This is analysis, not investment advice.

What actually happened this week

Two earnings calls, forty-eight hours, one story.

On October 20, on SpaceX’s quarterly earnings call, Elon Musk confirmed that special committees of independent directors at both SpaceX and Tesla are now exploring a combination of the two companies. He said he thinks it would be a good idea, that there is a lot of merit to combining them — and that he cannot take part in the talks, because as CEO of both companies he is a conflicted party. He called it, in his own words, an “odd kabuki dance.”

On October 21, Tesla reported its third quarter — and took no merger questions at all. Investor relations said Musk had said everything that should be said the day before. A banned question on a public earnings call is its own kind of confirmation: this is no longer a rumor to be waved away, it is a live corporate process to be managed.

Markets voted immediately. SpaceX shares rose about 8% in extended trading on the announcement (and as much as 10% premarket in some sessions); Tesla rose roughly 2%. Prediction markets, for what they are worth, put the odds of a deal at about 18.5% by the end of 2026 and 64% by the end of 2027 — speculation, not evidence, but a useful gauge of how seriously the idea is now taken.

The numbers that make the pitch

The timing was not accidental. Musk floated the review on the same call where SpaceX posted the loudest quarter in its history:

• SpaceX Q3 2026 revenue: $14.6 billion, up 161% year over year, with net income of $4.4 billion — against a $1.1 billion loss a year ago.
• The standout: SpaceX’s AI segment (Super Intelligence) generated $7.56 billion in the quarter, up roughly 830% — and overtook Starlink as the company’s largest business for the first time. Starlink itself is at about 13.6 million consumer subscribers.
• The combined valuation being discussed is roughly $3.5 to $3.6 trillion — Tesla at about $1.5 trillion, SpaceX at about $2 trillion after its June IPO. At current prices the combined group would employ about 155,000 people and generate roughly $153 billion in 2026 revenue, per analyst estimates compiled by Visible Alpha.

And Tesla’s own quarter explains why the pitch lands now, from the other side:

• Revenue $28.05 billion, roughly flat year over year; adjusted EPS $0.48, a beat against roughly $0.43–0.45 expected.
• But GAAP net income fell 43% to $787 million and operating income fell 56%, as AI and Optimus spending surged and regulatory-credit revenue shrank.
• The bright spots are all future businesses: FSD subscriptions up 67% to 1.74 million active, Optimus production has started in Fremont with external deliveries targeted around the end of 2027, the Austin robotaxi fleet has grown 10x since launch — and meaningful robotaxi revenue is now a 2027 story, by management’s own framing.
• Tesla also cut its 2026 capex guide from “more than $25 billion” to about $20 billion, which flattered free cash flow ($1.53 billion, positive) while quietly admitting the build-out is slower than promised.

Read the two reports together and the merger logic writes itself. Tesla is a car company whose valuation rests on AI, robots, and chips it does not yet sell at scale. SpaceX is suddenly an AI company with rockets attached — its AI segment alone did nearly four times Tesla’s entire profit problem in revenue. They already share the Terafab chip project (first test chips targeted for May–June 2027), personnel, and a $2 billion Tesla investment in SpaceX stock disclosed in Tesla’s Q2 filing. The industrial-overlap argument is genuine.

The part nobody can committee away

Here is the conflict the structure cannot hide. Musk holds roughly 20% of Tesla, where every share gets one vote — he has complained for years that this is not enough control to build AI and robotics there safely. At SpaceX, super-voting shares give him roughly 82–84% of the vote. Barclays analyst Dan Levy’s math, published October 16 and widely cited since, is that an all-stock acquisition of Tesla by SpaceX would leave Musk with about 37% of the combined company’s equity and 56% of its voting power; other estimates put the voting figure at 56–58%.

In other words: the merger does not just combine two companies. It moves Tesla’s shareholders from a structure where Musk can be voted down into one where he effectively cannot — and it does so as a side effect of a deal he proposed, negotiated by committees he appointed directors alongside, for a company he runs on both sides. That is exactly what special committees are for, and the process so far is textbook: independent directors, no terms agreed, Musk recused. But process can make a deal defensible; it cannot make the control transfer free. Tesla shareholders are being asked to pay for it, in governance, and the price has not been named yet.

The hurdles are correspondingly high:

• Tesla needs approval from at least 50% of all outstanding shares — not 50% of votes cast. Musk himself called that “quite a high hurdle.” Every passive holder who does nothing counts as a no.
• It would be the largest merger in history, reviewed everywhere: Washington may be friendlier under the current administration (reports cited by ANSA say Musk wants to close during Trump’s second term), but Europe and China are another matter — Beijing already views SpaceX’s US government contracts with suspicion.
• A buried governance flashpoint is already circulating: reporting this week that Musk personally ordered a slowdown of Starlink service used by Ukraine’s military has been folded into the merger coverage as Exhibit A for what concentrated, unreviewable control means in practice. Fair or not, that is the frame opponents will use.

My take

Strip the personalities out and there are two separate questions being conflated into one vote.

The industrial question — should the West’s largest launch, satellite, EV, robotics, and now AI-infrastructure assets sit in one balance sheet, sharing chips, compute, and talent? — is honestly arguable, and SpaceX’s $7.56 billion AI quarter made it more arguable this week, not less. Terafab, space-based compute, and Optimus are not PowerPoint synergies; they are shared physical inputs.

The governance question — should that combination be executed in the one structure that permanently solves Musk’s personal control problem at Tesla, without a control premium being explicitly paid to the Tesla shareholders giving control up? — is the one the committees actually have to answer. Watch the exchange ratio, not the keynote. If Tesla holders are paid for the control they surrender, this becomes a expensive-but-legible deal. If they are asked to donate it for “synergy,” the 50%-of-all-shares hurdle will do exactly what it was designed to do.

Nothing is agreed, no vote is scheduled, and Tesla’s annual-meeting deadline of December 6 is the next date that matters: a proposal that misses it likely waits six to twelve months. Until terms exist, the honest description is the one Musk himself gave it — a good idea, being danced around by the only people allowed to decide it, while the man who wants it watches from outside the room he built.

Sources: SpaceX and Tesla Q3 2026 earnings calls and reports (Oct 20–21, 2026); Barchart and Reuters/ANSA coverage of the committee announcement and SpaceX results; Barclays analyst note via The Wealth Advisor (Oct 16, 2026); Tesla Q2 2026 10-Q disclosure of its $2B SpaceX stake; Visible Alpha estimates via ANSA. Prediction-market odds are sentiment gauges, not facts, and are labeled as such above.