On September 13, the Financial Times reported that Vy Capital holds approximately $40 billion worth of SpaceX shares — roughly 3.4% of the company. Bloomberg's data places the firm as SpaceX's fifth-largest disclosed shareholder, ahead of Sequoia Capital and Andreessen Horowitz.
The rest of the numbers are equally spare. Total internal rate of return since its 2014 founding: 41%. Capital distributed to investors: $4.6 billion. Assets under management: $27 billion at the close of last year, $50 billion by June of this year. Core investment team: four people. New external capital accepted since last year: none.
I have spent most of my professional life inside markets that reprice every eight seconds. My 2020 DeFi portfolio was marked continuously against Aave and Compound reserve data; my 2022 liquidity containment plan was executed against thresholds that moved in real time. So the thing that arrests me in the Vy Capital disclosure is not the return. It is the absence of a price.
The ledger remembers what the market forgets.
The firm's entry point matters more than its size. Vy Capital first bought SpaceX in 2016, when the company was valued near $15 billion. The purchase landed shortly after the Falcon 9 launch failure — the moment when headline risk on the asset was at its maximum and the number of willing buyers at its minimum. That is not luck. That is a sourcing model.
The operator of that model is John Hering, an investor with a direct and long-standing relationship to Elon Musk. Since 2019 he has worked inside Starlink's early business development — hiring personnel, constructing financial models — and has held a SpaceX employee badge. He sits as a director of The Boring Company, where Vy Capital participated in the recent $3 billion financing round. The firm is the largest external investor in both The Boring Company and Neuralink. In 2022 it committed $700 million to the Twitter acquisition.
Read that list as a portfolio and it looks diversified: launch, satellite broadband, tunneling, brain-computer interfaces, social media. Read it as a risk register and it is a single line item. Every position underwrites one operator's judgment, one operator's capital allocation, one operator's regulatory and political exposure. The sector labels are decoration.
Vy Capital stopped accepting external investors last year. Its assets under management moved from $27 billion to $50 billion in six months without new subscriptions. There is only one mechanism that produces that result.
In private markets there is no mark. There is a negotiation.
Start with an arithmetic problem the reporting left unresolved. If SpaceX is worth an estimated $1.75 trillion and Vy Capital owns 3.4% of it, the stake is worth roughly $60 billion. The reported figure is $40 billion. Working backwards from $40 billion at 3.4%, the implied valuation is about $1.18 trillion — a gap of close to half against the estimate cited in the same news cycle.
That gap is not a rounding error. It is the entire asset class.
The same equity carries at least three prices simultaneously: the last primary round, the most recent secondary tender, and whatever a fund's auditor accepts at year-end. On-chain, that condition is a broken oracle — every lending market built on it would be liquidated or insolvent on a technicality, and the arbitrage would close within minutes once a sophisticated actor deployed a bot. Off-chain, it is a negotiation between a four-person investment team and the people writing the letter to investors.
This is the first thing crypto people get wrong when they look at private equity. They assume opacity is a bug that tokenization will fix. Based on my 2017 work auditing ICO smart contracts — two hundred of them, fifteen carrying exploitable re-entrancy paths that would have cost investors roughly $4 million — I would frame it differently. An unreconciled price is not an error state. It is a feature that a specific class of capital pays a premium to access. The $200 billion discrepancy between the two SpaceX figures is not a scandal. It is the spread a firm with the right relationship captures by sitting inside it for eight years.
AUM that grows by revaluation is not capital formation. It is accounting.
The jump from $27 billion to $50 billion with no new subscriptions is a mark-up. Crypto has a word for this, and it is not a flattering one. When a lending protocol counts unrealized gains as collateral and lends against them, we call it over-leverage and we wait for the cascade. The private-market version wears a better suit and carries a longer fuse, but the mechanism is identical: the ability to commit capital — $700 million toward a social media acquisition, a slice of a $3 billion tunneling round — is a function of the mark, not of the cash.
If the mark is wrong, the dry powder is wrong. And the commitments are already written.
This is where the macro reading turns uncomfortable. A concentration of this scale in a single privately held operator is not a diversified allocation. It is a leveraged bet on one man's execution cadence, priced on a curve nobody is required to publish. Every time a fund like this signs a commitment off an upward revision, it transmits that valuation into a second and third balance sheet. That is how a private mark becomes a systemic number.
Measure the concentration honestly. Assume the $10 trillion projection holds and arrives in five to seven years. From an estimated $1.75 trillion, that is a compound annual growth rate of roughly 33%. Vy Capital's position, on today's share count, would be worth somewhere north of $200 billion. On paper, the firm's IRR since 2014 would look less like venture capital and more like the best decade a sovereign wealth fund ever had.
Now ask the question the projection avoids: how does that value exit? There is no order book, no continuous market, no redemption window. The gain is real only if SpaceX lists or a secondary buyer pays the mark. Until then, the return is an internal rate of return computed on an assumption. I have written this down before and it bears repeating to anyone pricing private exposure: the ledger does not care about your model; it cares about your settlement. A 41% IRR that cannot be converted to cash on a chosen date is a forecast, not a result.
The Musk complex is one risk factor wearing four tickers.
SpaceX, The Boring Company, Neuralink, and — through the 2022 commitment — the social platform. Same key-man. Same capital source. Same narrative gravity. Anyone who has run a book through a real drawdown knows this is not diversification. It is single-factor exposure dressed in sector costumes.
In DeFi, correlated risk is visible early. When utilization on a lending pool spikes, every participant sees it within minutes; when a stablecoin's reserve composition drifts, the data is on-chain before the price is. That transparency is not a moral virtue. It is a latency advantage that lets a disciplined operator cut exposure before the crowd. My 2022 containment work — reducing a fund's crypto exposure from 60% to 10% inside 72 hours after the Terra collapse, preserving about $12 million through the FTX contagion — was possible because the risk indicators were observable in real time, not because I was smarter than the market.
Private concentration strips out that latency. The failure mode arrives as a legal notice, a delayed audit, or a cap-table entry nobody sees. By the time the information reaches the allocator, the exit has already been priced away.
Institutional capital did not buy the protocol. It bought the wrapper.
I spent the first half of 2024 building a compliance framework for a DC asset manager ahead of the spot Bitcoin ETF approval — custody standardization, reporting automation, a 25% reduction in client onboarding time. The lesson from that work was unambiguous, and it should have been humbling to everyone who spent a decade arguing about decentralization.
The buyers did not want the protocol. They wanted the wrapper. They wanted a ticker, a custodian, a settlement cycle, and a report they could file. The Bitcoin inside the wrapper was almost incidental to the decision. Nothing in the Vy Capital disclosure suggests the next wave behaves differently: if a compliant tokenized SpaceX share existed tomorrow, the same capital would buy it, and the fact that settlement ran on a chain rather than through a transfer agent would be a footnote in the offering memorandum.
Which reframes the entire real-world-asset narrative. Tokenized treasuries, tokenized private credit, tokenized equity — these are not crypto's victory over traditional finance. They are crypto agreeing to become the distribution channel for the asset class that already dominates it. We do not build on hype; we build on consensus. And the consensus among allocators is that they want the exposure, the wrapper, and the relationship — not the permissionless part.
Concentration is the product. Fragmentation was the marketing.
For three years the industry has held conferences about liquidity fragmentation — bridges, rollup liquidity, unified order books, intent-based routing. From where I sit, having run a $5 million book across Aave and Compound with dynamic hedging and zero impermanent loss, the diagnosis was always wrong. Liquidity does not fragment because the technology is immature. It fragments because issuers prefer it that way, and because a fragmented market is easier to sell a product into.
Now look at the strongest position in the story you are reading. Vy Capital is not fragmented. It is perfectly concentrated, deliberately illiquid, and closed to new capital. That structure produced a 41% IRR. It is the exact opposite of every product the DeFi industry has shipped over the same decade, and the exact opposite of the pitch that fragmentation is a problem requiring a token.
The rollup wars taught the same lesson at a smaller scale. OP Stack and ZK Stack differ far less in their proving systems than in which team convinced more projects to deploy first. Distribution beat cryptography, as it usually does. Vy Capital won its SpaceX allocation by being present at the launch failure, not by possessing a better model. Access, not analysis, is the scarce input.
A valuation is only as good as the cash flow that services it.
Here the SpaceX case is stronger than most crypto assets, and I will not pretend otherwise. Launch cadence, Starlink subscriptions, government contracts — these are revenue lines, auditable, recurring, growing. That is a fee market in the only sense that matters.
Which is precisely why Bitcoin's experience should be studied alongside it. For years the security budget rested on a subsidy with an expiry date, and the honest analysis was that the model did not close without a fee base that did not yet exist. Ordinals and inscription activity, whatever one thinks of the JPEGs, generated that fee base. They injected a real revenue line into a network that had been running on emissions and sentiment. The lesson generalizes: a valuation without a fee market is a narrative with a deadline.
Apply that standard to the $10 trillion projection. The number requires Starship to reach operational cadence, Starlink to defend its subscriber base against terrestrial competition and sovereign alternatives, and the regulatory environment to remain permissive across multiple jurisdictions for six consecutive years. Each condition is individually plausible. Compounded, the projection is a claim that one capital-intensive company will grow faster than any capital-intensive company in industrial history has managed, for longer, without a systemic interruption.
Maybe. The Falcon 9 landed. I am not in the business of betting against execution.
But I am in the business of pricing it. And the price of a claim like that should be a discount rate, not a headline.
The ledger remembers what the market forgets.
Here is the part the crypto industry will not want to read, and the reason this story belongs on a blockchain desk rather than an equity one.
The prevailing thesis is that tokenization will open private markets — that putting SpaceX, or a treasury, or a private credit fund on a chain will democratize access and create continuous price discovery. The Vy Capital disclosure is evidence for the opposite conclusion. The returns that matter are being generated in structures that are intentionally illiquid, intentionally concentrated, and intentionally opaque. Continuous price discovery would destroy the mechanism. A public mark would force the mark-downs, invite the redemption pressure, and expose the correlation the structure exists to hide.
So crypto faces a fork it has not acknowledged. It can keep building markets that genuinely price things — transparent, liquid, occasionally brutal — and accept that the largest pools of capital will route around it for the comfort of a negotiated mark. Or it can build compliant wrappers for other people's concentration, in which case the settlement layer is decentralized and the economics are not.
The industry will almost certainly choose the second and call it adoption. The evidence is already in the ETF flows. The question is whether anyone will say so out loud.
Watch three signals from here. The next SpaceX tender — or the listing itself — will be the first real mark, and the distance between that number and every fund's year-end carry will tell you how much of the reported private-market return was negotiation. If tokenized SpaceX exposure appears before the IPO, watch its structure: permissioned, KYC'd, custodied, with the DeFi label applied cosmetically to a compliant wrapper. And watch the benchmark. If a 41% IRR earned by four people becomes the number against which liquid crypto strategies are finally measured, the honest answer for most of those strategies is that they have spent a decade underwriting beta to a man in a hurry, and charging fees for the privilege.
