How to Read a Pre-IPO SPV, and Who Funds the Frontier

Tool · 17 September 2026

Public

SpaceX (SPCX) · NVIDIA (NVDA) · Tesla (TSLA) · Palantir (PLTR)

Private

3 private positions, held through vehicles whose terms bar naming the company — recused

Commitment

sha256 fc5792fdae164aadbd7280cacf15d828808eb91fe2b593f426bcc1ec2025a593 · positions record version 2026-10-09

I currently invest through private vehicles of the kind this piece analyzes. Restricted positions are listed on the disclosure page.

Intent, 90d    none

How positions are declared, why some are not named, what a recusal means, and how the sealed record works: disclosure. Where a position I may not name sits in a company this piece would otherwise forecast, I publish no forecast on it and say so.

The tool

If someone offers you shares in a private company through a fund, you pay more than the price they quote. This works out how much more.

This is a Tool: a piece outside the numbered issues, and Tools carry no register entry. It answers one question: what price you actually pay when a wrapper — any vehicle that holds the shares on your behalf, whether an SPV (special purpose vehicle, a fund formed to buy one company), a fund unit, or a feeder into someone else's fund — offers you a private company "at the round price," and how much the company has to grow before you are whole.

Six lines. Illustrative figures, invented for this piece and rounded, no sponsor and no company. The structure is the generic soft-circle SPV anatomy as the public platform pages describe it; the arithmetic takes ten minutes on any term sheet.

1. Subscription: $50,000. 2. Deductions before any stock is bought: management fee, 2% a year for four years, prepaid ($4,000); expense reserve, 1% ($500); placement fee, 2% ($1,000); intermediate-vehicle markup embedded in the acquisition cost, 1.5% ($750). Total −$6,250 (12.5%). 3. Net invested, the basis for carry — the sponsor's cut of the gains: $43,750. 4. Stock at the round price: $43,750. 5. Effective entry price versus the round price: 1.143× (a 14.3% premium). 6. Breakeven underlying multiple, after 10% carry on gains above line 3: 1.16×.

Line 6 is the number the offering card never prints. The company has to be worth 16% more than the round before this investor has his money back, and that is with the stock bought at the round price and a modest carry. Raise the carry to 20% and it is 1.18×. Let the vehicle buy at 10% over the round, which is common, and it is 1.27×.

What it decides: for an investor, whether the wrapper is the route in or the reason to find another one — the asset can be fine and the deal bad, and those call for opposite responses. For a corporate treasury, whether "strategic exposure" bought through a feeder is exposure to the company or exposure to the fee schedule.

Who this changes a decision for

Investors. Anyone who has been offered a pre-IPO name through an SPV, a fund unit, or a feeder into someone else's fund in the last eighteen months, which is now most accredited investors. The decision is not whether to own the company. It is whether the breakeven multiple on the specific document in front of you is 1.03× or 1.27×, because the same asset behind those two numbers is two different investments.

Corporate treasuries and strategy groups. A minority stake in a launch, reactor or fusion company taken "for strategic reasons" through an intermediated vehicle carries the same load as a retail ticket, and usually a worse information right. If the purpose is a seat at the table, buy the seat directly from the company in a primary round; if the purpose is a financial return, run the six lines and compare the breakeven with what a direct secondary would cost.

The argument

The capital stack, and where you sit in it. Space, nuclear and fusion are funded by the same three pools in the same order. Primary rounds, led by venture and growth funds and, in the largest names, sovereign and strategic money, buy preferred stock with information rights and a liquidation preference. Secondary buyers — the same funds, plus dedicated secondary firms — buy existing shares from employees and early investors, at a negotiated price that may sit above or below the last round. Everyone else arrives through a wrapper.

The scale is not small. Seraphim's Q2 2026 index is an analyst estimate with a stated methodology. It counts $7.5 billion of venture investment into space companies in the quarter and $23 billion in the trailing twelve months across 620 deals. Series D and later took 47% of the dollars. The Fusion Industry Association's July 2026 survey, self-reported by 56 companies, puts fusion at $4.48 billion raised in the twelve months to July and $14.24 billion cumulative since 2021. Fortune reports estimates of the secondary market for private shares at between $62.5 billion and $120.9 billion in 2025, an estimate whose origin the article does not name and a range wide enough to say that nobody measures it. What is measured is that it exists, and that it is where the retail investor is.

The reason wrappers exist is a rule, not a business model. Section 3(c)(1) of the Investment Company Act exempts a fund from registration only if its securities are held by not more than one hundred persons; 3(c)(7) removes the cap for funds owned exclusively by qualified purchasers, a higher wealth test. A sponsor who wants to raise a few million dollars from a few hundred accredited investors cannot do it in one 3(c)(1) vehicle. So he builds several, or a feeder into a master, or a forward contract on an interest in a fund that owns the shares. FINRA said this plainly in August: in some cases the fund owns the stock indirectly, leaving the investor several layers removed from it. Each layer has a sponsor, and each sponsor has a fee.

The anatomy of the soft-circle SPV. The offering card shows a company and a valuation. The documents show something else, and the six things to find in them are these.

The load is every deduction before stock is bought. Management fees for a fixed-term vehicle are usually prepaid for the full term at subscription, which on a four-year vehicle turns "2%" into 8% on day one. Add the expense reserve, the placement fee if it is borne by investors, and the fee of any intermediate vehicle, which is not a line item at all but a markup buried in the price the SPV paid for its position. This last one is the one the card never shows: the only way to find it is to compare the acquisition cost in the closing statement with the round price.

The carry basis is what the carry is charged above. "Net invested amount" is the usual definition, and it is smaller than the subscription, so the sponsor earns carry on the recovery of his own fees before you have recovered them. In the worked example the carry starts at $43,750, not $50,000; the $6,250 between them is profit to the sponsor and not to you.

"Valuation determined at closing." The price you are subscribing to has not been set. A soft circle commits you first and prices you later, and in a name where the secondary trades above the round, later means higher. The FINRA note has the polite version: the costs of acquiring the interest may require material growth to achieve a gain. The SEC's August enforcement release has the other version: a sponsor who told more than 800 investors across eleven funds that fees were at most 12.5%, and charged prices on average 46% above what he paid. The 46% was the largest cost in those vehicles, and it was never disclosed as a fee.

Extended deadlines are the sponsor's option, not yours. A closing that slips a quarter while your money sits in escrow is a free loan; a closing that slips because the block has not been sourced is the sponsor selling stock he does not yet own.

In-kind distribution rights decide what you hold after a listing. A vehicle that can distribute shares gives you the public stock and the lock-up on your own terms; a vehicle that must sell and distribute cash decides your exit date for you, with a fee on the way out.

The intermediate-vehicle fee is the one to ask about by name. If the SPV bought a unit in another fund rather than shares, the unit's own management fee and carry sit inside your acquisition cost and inside every future distribution, and they appear in none of your documents.

Where the published platforms sit against that, from their own filed disclosures, is the point of comparison a reader needs. These are the platforms' list prices, not an endorsement.

Three things the table shows. The listed brokers charge less than the SPVs that buy from them, because the SPV's fee is layered on top of the broker's; a 4.85% platform commission is the floor of the wrapper's cost, not its total. Every platform disclosure says in writing that fund routes carry additional sponsor fees the platform does not collect, so a "no management fee" line on the platform tells you nothing about the fund inside. And the one route with a breakeven near 1.0× is the direct secondary with a six-figure minimum, which is the route the wrapper exists to replace.

The three wrapper types. The secondary fund unit is the honest case: no management fee, no carry, a distribution fee at entry. Its breakeven is 1.04× and it keeps 94% of the upside at a 3× outcome. The cost is that a unit in a fund is not a share in a company: no audited NAV unless the fund publishes one, no information right, and a transfer restriction that makes the unit's price whatever the sponsor says it is. That is the trade, and at 4% it is a fair one.

The fund holding common at the preferred price is a different error, and no fee schedule reveals it. Preferred stock from the round carries a liquidation preference; common does not. Paying the preferred price for common means the wrapper is senior to nothing, and in any outcome below the preference stack the common is worth less than the price paid, however cheap the fees. The number to ask for is the company's own 409A valuation of its common, which every US private company with an option plan commissions and none of them volunteers.

The convexity point. Run the worked example at two outcomes. If the company ends at 0.8× the round, the investor holds 0.70× of his subscription — the wrapper turned a 20% loss into a 30% loss. If the company ends at 3×, he holds 2.45× — the wrapper turned a 200% gain into a 145% gain. Downside capture, the investor's loss over the asset's, is 1.50; upside capture is 0.73; the ratio is 0.48. A wrapper that captures more of the downside than of the upside has inverted the asymmetry, and this one captures twice as much. The investor is long the company's losses at 150% and its gains at 73%.

That is the inversion, and it is the whole piece. A venture position is worth owning because the payoff is convex: capped loss, uncapped gain. A wrapper that charges on the way in and on the way out, with carry on a basis below your cost, takes more of the loss than of the gain. In the worked example the downside capture is 1.50 and the upside capture 0.73. Any wrapper with carry on net invested and a prepaid load above 10% shows the same inversion when the six lines are run on its own term sheet.

Definitions

Price versus cost. A published price is one party's number; the cost of holding what it buys is another. In launch, a dollars-per-kilogram figure is a price divided by a maximum payload nobody fills, and the operator's own cost is a different number it rarely publishes. Here the distinction runs the other way: the round price is the company's number, and your cost is the round price plus everything the wrapper adds before and after it.

Investable slice. Of a headline market figure, the part a financial investor can actually buy: not government budgets, not captive internal revenue, not the downstream commerce an analyst counted to make the total large. This piece narrows it once more: of the investable slice, the part reachable through a wrapper whose breakeven is near 1.0× is smaller still, and that is the slice a reader of this brief can actually hold.

What would change my mind

A wrapper, from a sponsor with an audited NAV, with a load under 5% and carry charged on gross — on the subscription, not on net invested — and paid only on realised profit. Run through the six lines that vehicle breaks even near 1.03×, keeps most of the shape, and moves the question from the structure back to the asset, where it belongs. Two of the three platforms above already disclose loads in that range on direct routes; what does not yet exist in any offering I can cite is the sponsor-level version with audited marks. If one appears, the argument here narrows to "read the carry basis" and I will say so. In the other direction: if the SEC's August action is the first of several and the disclosed loads across the market turn out to be the honest ones, then 12.5% was the good case, and the table above understates the problem.

Close

Tools carry no register entry. The register opens with the first issue on 15 October 2026.

A Spanish edition ships with every piece — edición en español, same figures, same table, same worked example. The register is public and append-only: reckoning-register.

Sources

Every real figure above traces to a line in the sources for this piece, with publisher, publication date and access date. The worked example and the three wrapper rows marked illustrative are arithmetic, not reported figures; their inputs were invented for this piece, follow the generic soft-circle SPV structure the platform pages describe, and are labelled as such wherever they appear. One figure — the "under 5% to 18%" access-fee range attributed to Forbes — could not be retrieved from the publisher's page during the run and was verified from the archived copy on 2026-09-14. No confidential offering document was used, quoted or paraphrased; every platform term comes from a filed Form CRS or the platform's own published page.

---

Leer en español →

Email address Subscribe

Each issue in your inbox the morning it publishes, a note the day a forecast is graded, and occasional Reckoning news. Free, in English and Spanish. No list sharing, unsubscribe in one click. Privacy.