Illiquidity is supposed to pay you extra. Sometimes it does. Here is how to check whether the premium you were promised actually reached you.
14 min read · Level 4
The liquidity premium is real, and it is one of the few sources of return in investing that does not require you to be cleverer than anyone else - you are being paid for a constraint you accepted, not for an insight. That should make it the most reliable premium available. In practice it is among the least reliable, and the reason is structural rather than mischievous: the premium is earned at the asset level and consumed at every layer between the asset and you. Higher fees are one layer, uncalled capital is another, an extension that stretches the same multiple over more years is a third, and none of them is hidden - each appears in the documents, and each is individually defensible. The investor's task is therefore not to detect a deception but to do arithmetic that nobody else in the chain has any reason to do for them. Which leads to the uncomfortable conclusion: if you cannot state, in one number, the premium you expect to receive net of every layer over the real worst-case term, then you are not being paid for illiquidity. You are simply illiquid, and hoping.
The liquidity premium is the extra expected return an investor should require for holding something they cannot readily sell.
The logic is sound and it runs in both directions. A seller of an illiquid asset must accept an illiquidity discount to find a buyer, and that discount is the buyer's higher expected return. In a liquid market the same trade costs only the bid-ask spread, which is small precisely because there are many buyers.
So far this is a description of a market, not a promise. The premium is a requirement, not a guarantee: it says what an investor ought to demand, not what any particular vehicle will deliver. Three things routinely stand between the two, and all three are measurable in advance.
It may never have been there. A large multiple over a long period can be an ordinary annual rate, and multiples are what illiquid vehicles report. It may be consumed by costs. Illiquid vehicles typically carry higher fees, and the premium is quoted gross. It may be diluted by capital that is not working. You commit the whole amount; only part of it is deployed, and the rest earns cash-like returns inside a commitment made for something better.
Worth naming one more thing that is not a premium at all. An illiquid asset is valued rather than priced - its net asset value is produced by a process, often with the manager's involvement - and a value that does not move looks like low volatility. It is not low volatility. It is unobserved volatility, and mistaking the two makes an illiquid holding appear safer than it is.
Because illiquid assets are sold on the premise that the premium is automatic, and the arithmetic to check it is arithmetic you already have.
Every figure below uses compounding and a multiple - nothing more. Which means this is one of the few claims in investing that a reader can verify before committing rather than discover afterwards, and the verification takes about five minutes.
And because getting it wrong is expensive in a specific way: you cannot correct it. A liquid holding that disappoints can be sold next week. An illiquid one that turns out never to have carried a premium has to be held for years while it does not carry one.
This is educational information, not investment, securities or financial advice. Rules differ by country and change — get advice from a qualified professional in your jurisdiction before acting.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Sources: U.S. SEC — Investor.gov; FINRA — investor information; UK FCA — information for consumers; IOSCO — global securities regulators
Three checks, in order of how often they matter.
Check one: convert the multiple into a rate.
A vehicle returns 2.0x over a ten-year life. Doubling your money sounds substantial. As an annual rate:
2^(1/10) - 1 = 7.18% a year
Now the liquid comparison. 100,000 compounding at 8.00% a year for ten years:
100,000 x 1.08^10 = 215,892.50
The boring liquid alternative ends 15,892.50 ahead of the 2.0x, having been sellable on any day of the decade. To merely MATCH 8.00% over ten years, the illiquid vehicle needs 1.08^10 = 2.16x. To deliver a genuine three-percentage-point premium - 11.00% a year - it needs 1.11^10 = 2.84x.
So the question to ask about any ten-year illiquid vehicle is not 'is a 2x good'. It is 'is it above 2.16x', and a multiple below that is a liquidity premium of less than zero.
Check two: ask whether the number is gross or net.
The premium is the compensation for illiquidity and the fee is the price of access, and they are quoted in the same units, which makes them easy to confuse. A gross return carrying a three-point premium and a fee structure costing three points has handed the entire premium to the manager. The fee drag lesson earlier in this subject is the tool; the only new step here is noticing that illiquid vehicles typically charge more, so the fee deserves the same scrutiny as the return.
Check three: measure the return on what you committed, not on what was deployed.
This is the one that is almost never done, and it is often the largest of the three.
Suppose a vehicle genuinely earns 11.00% a year - a real three-point premium over the liquid 8.00%. You commit 100,000. But capital is drawn down over years and returned before the end, so across the ten years only part of the commitment is at work. Take a simplified version: 65,000 is deployed for the full ten years at 11.00%, and the remaining 35,000 sits available for calls, earning 2.00%.
65,000 x 1.11^10 = 184,562.36 35,000 x 1.02^10 = 42,664.80 total = 227,227.16
On the full 100,000 you committed, over ten years, that is:
(227,227.16 / 100,000)^(1/10) - 1 = 8.55% a year
An 11.00% return on deployed capital became 8.55% on committed capital. The three-percentage-point premium became 0.55 of a point - and the 0.55 is what actually reached the investor, because the whole 100,000 was theirs and the whole 100,000 was committed.
That is a simplification of a real drawdown schedule, and its purpose is directional: whatever the true pattern, the reported rate is measured on deployed capital and yours is earned on committed capital, and the gap between them belongs in the decision.
Three checks, three ways the premium disappears, and none of them requires any view about the asset itself.
The specific wrong move is comparing an illiquid vehicle's multiple against a liquid alternative's annual rate.
'2.5x versus 8% a year' is not a comparison, because the two figures are measured in different units, and the multiple is the one that sounds larger. Over ten years 2.5x is 9.60% a year, which is a real premium of 1.60 points - a genuine but much less impressive statement than 2.5x against 8%. Over fifteen years the same 2.5x is 6.30% a year and the premium is negative. The multiple alone cannot tell you which of those it is, because it contains no time.
It is tempting because multiples are what illiquid vehicles report, and rates are what liquid ones report, so the only figures readily available are already in the wrong units. Converting is one line and almost nobody does it.
The second mistake is treating a stable reported value as evidence of low risk. An asset valued quarterly by a process cannot produce a jagged line, so it does not - and a portfolio that mixes it with daily-priced holdings will show a calmer aggregate that reflects the measurement interval rather than the risk. A holding that is not observed is not thereby stable.
multiple^(1 / years) - 1. Do it for the target, and do it for the track record you were shown, using each fund's actual life rather than its stated one. This single step reprices most private-market comparisons.
Name the specific thing you would otherwise hold and the rate you are assuming for it. Without a named alternative, the premium has nothing to be a premium over, and 'better than cash' is not the comparison anyone is actually making.
Over N years, matching a liquid rate g needs (1 + g)^N, and a premium of p needs (1 + g + p)^N. Those two numbers are the bar, and they should be written down before any track record is presented rather than after.
All of them: management, performance, and any at a feeder or platform layer. Ask for the net-of-everything figure specifically, and if it is not available, that is itself the answer to the question you asked.
Estimate how much of the commitment will be at work, and for how long, and what the rest earns while it waits. The reported rate is measured on the money the manager was using; you are earning it on the money you set aside.
After all three adjustments, one number remains: the premium you would actually receive for being unable to sell for the real worst-case term. Write it as a single figure and ask directly whether that is a fair price for the option you are giving up. Frequently it is. The point is that it should be a conclusion, not an assumption.
Before committing to any private fund, direct private position, or vehicle with redemption restrictions, and whenever a return is quoted to you as a multiple.
Do not use this to conclude that illiquid assets are not worth holding - some genuinely deliver, and the check is how you find them. And do not apply it where the illiquidity is incidental rather than structural, such as a small direct holding you could sell in a week at close to its quoted value.
Private fund terms commonly permit the life of the vehicle to be extended, often by a year at a time, to allow remaining assets to be sold in better conditions. Each extension is usually justifiable on its own terms and can genuinely improve the final multiple. But the multiple is divided by a longer period, and the annual rate falls even when the total proceeds rise. A 2.5x over ten years is 9.60% a year; the same 2.5x reached after fifteen years is 6.30%.
When a track record is presented as multiples, ask for the actual elapsed life of each vehicle and convert each one yourself. The conversion is one line and it is where the difference between a headline and a result becomes visible.
Standard risk measures are computed from observed price movements. An asset valued quarterly by a process contributes few observations, and those it contributes are smoothed by the valuation method, so its measured variability is low and its measured correlation with everything else is lower still. Adding such a holding therefore reduces a portfolio's reported volatility almost mechanically - an improvement that is a property of the measurement rather than of the assets.
Treat a reported risk figure for a portfolio containing infrequently-valued assets as an understatement, and judge those holdings on what the underlying businesses or properties could actually do, not on the smoothness of the line describing them.
Create a free account to read the rest and apply it to your own investing.