For fifteen years the most influential idea in institutional investing was that patience pays. Lock capital away for a decade, accept that you cannot sell when you would like to, and the market will compensate you for the inconvenience. The illiquidity premium was never precisely measured and never guaranteed, but it was very widely believed — and belief was sufficient to move several trillion dollars out of public markets and into private ones.
Key points
- The illiquidity premium was a residual, not a measurement. Much of what was attributed to patience was leverage, sector selection and valuation smoothing.
- Distribution rates have run below historical norms for three consecutive years, converting a paper problem into a cash-flow one.
- Later vintages show negative excess return against public-market equivalents — the first sustained negative run since the series began.
- The premium has not vanished so much as narrowed to where it was always real: genuine operational improvement and genuine scarcity of capital.
That belief is now being tested in the only way that ultimately matters, which is by cash. Distribution rates from private equity funds have run below their historical norms for three consecutive years. Secondary market pricing, once a niche indicator watched by a handful of specialists, has become a mainstream data point precisely because so many institutions have needed it. And the denominator effect — the awkward arithmetic by which a falling public portfolio mechanically inflates your private allocation past its policy limit — has moved from a theoretical risk in an investment committee paper to a line item requiring an actual decision.
None of this constitutes a crisis. Private markets are not breaking. But the terms of the bargain are being renegotiated in front of us, and the institutions that understand what they were actually being paid for will navigate the renegotiation considerably better than those that did not.
What the premium was supposed to pay for
The theoretical case was always clean. An investor who surrenders the right to sell is bearing a genuine cost, and in a competitive market genuine costs earn compensation. If two assets have identical cash flows and identical risk, but one can be sold on Tuesday and the other cannot be sold for eleven years, the second must be cheaper at purchase. That discount, realised over the holding period, is the premium.
The difficulty is that this clean argument was asked to carry an enormous amount of empirical weight it was never designed to bear. In practice the illiquidity premium was not measured directly. It was inferred as a residual: take the realised return of a private portfolio, subtract the return of some public benchmark, and attribute the difference to illiquidity. Everything the benchmark failed to capture ended up inside that residual.
And a great deal failed to be captured. Private equity portfolios carry systematically more leverage than the public indices they are compared against. They are concentrated in particular sectors, and for most of the last cycle those sectors were the ones that worked. They are marked by managers rather than markets, which smooths reported volatility and flatters every risk-adjusted statistic derived from it. Each of these is a real source of return or apparent return. None of them is illiquidity.
The premium was a residual, and residuals absorb whatever the model forgot. For fifteen years the model forgot leverage, sector and smoothing — and called the remainder patience.
Zimba Capital Research
Strip those factors out carefully and the residual attributable to illiquidity alone becomes far smaller than the marketing implied, and considerably less stable across vintages. That is not an argument against private markets. It is an argument against the specific claim that locking money up is, by itself, a reliable source of excess return.
The arithmetic changed underneath the thesis
The thesis was formed in a particular environment and quietly assumed that environment would persist. Between 2010 and 2021, the cost of capital fell almost continuously. In that world, a strategy built on buying an asset, financing it with debt, holding it for five years and selling it into a market where discount rates were lower than when you bought had a powerful tailwind that had nothing to do with operational skill. Multiple expansion did a great deal of the work, and it did that work silently.
When the cost of capital rose, three things happened at once, and they compounded. The exit environment tightened, so holding periods extended. Extended holding periods reduced distributions, so the capital that was supposed to fund next year’s commitments did not arrive. And reduced distributions arrived precisely when institutions had already committed to future vintages on the assumption that it would.
That last point deserves emphasis, because it is the mechanism by which a valuation question became a liquidity question. Private market allocations are not funded from a bank account. They are funded, in large part, from the distributions of earlier funds. The system relies on a circulation that most models treated as an accounting convenience rather than a dependency. When circulation slows, the institution must either sell something else, borrow, or reduce commitments — and each of those choices has consequences that extend well beyond the private book.
Where the repricing is actually showing up
The clearest signal is in the secondary market, because it is the only place where private assets are priced by someone with an incentive to be right rather than an incentive to be reassuring. Discounts to net asset value in secondaries widened sharply and have narrowed only partially. The narrowing is genuine, but it has been concentrated in the highest-quality books; the dispersion between the best and worst funds on offer has widened considerably, which is itself informative. A market that prices everything at a similar discount is expressing a view about liquidity. A market that prices assets very differently is expressing a view about the assets.
The second signal is the growth of continuation vehicles. There are perfectly respectable reasons to move an asset into a new structure with fresh capital and a longer horizon — a genuinely good business can need more time than the original fund has left. But a structure that allows a manager to avoid establishing a market price for an asset, while continuing to charge fees against a value that the manager themselves has determined, deserves scrutiny in proportion to how convenient it is. The volume of such transactions has grown considerably faster than the number of situations that plausibly justify them.
The third signal is the quiet one: the changing composition of who is committing new capital. As traditional institutional allocators have slowed, the industry has turned decisively toward wealth channels and retail-adjacent structures. This is often presented as democratisation. It can also be read as a search for the marginal buyer — and the identity of the marginal buyer usually tells you something about where a market is in its cycle.
| Cohort | 2015 | 2025 | Change |
|---|---|---|---|
| Large university endowments | 31.0% | 42.5% | +11.5 |
| Public pension plans | 9.4% | 16.8% | +7.4 |
| Sovereign wealth funds | 12.1% | 19.2% | +7.1 |
| Corporate defined benefit | 7.8% | 9.1% | +1.3 |
| Insurance general accounts | 4.2% | 3.6% | −0.6 |
The table is worth sitting with. The cohorts that increased their allocations most aggressively are precisely those with the longest stated horizons and the least tolerance for a liquidity event — a combination that is comfortable in every year except the one where it is not.
What patience is still worth
It would be a mistake to read the above as a case against private markets, and we do not intend it as one. The premium has not disappeared. It has narrowed to the places where it was always real, and those places are more specific and more demanding than the last cycle required anyone to be.
The first is genuine operational improvement. A manager who buys a business, changes how it works and sells it materially better than they found it has produced a return that has nothing to do with liquidity, leverage or multiple expansion. This has always been the honourable core of the asset class, and it remains fully intact. It is simply a much smaller share of the industry than the industry’s aggregate size implies, and it is difficult to identify in advance.
The second is genuine scarcity of capital. Where a business needs financing that public markets structurally cannot supply — because the time horizon is wrong, the disclosure requirements are impossible, or the asset is too complex to price quickly — the private provider holds real bargaining power, and bargaining power is the most reliable source of excess return there is. Much of the more interesting activity in private credit and certain kinds of infrastructure sits here.
The third is behavioural, and it is the one most often dismissed. An institution that cannot sell in a panic does not sell in a panic. The lock-up that looks like a cost in a liquidity squeeze is the same mechanism that prevented the institution from crystallising losses at the bottom in every previous one. This is a real benefit. It is worth paying something for. It is not, however, worth paying the fee load the industry currently charges for it, and it should be valued as governance rather than as alpha.
What would change our mind
We hold this view with the confidence the evidence supports, which is moderate rather than high, and we would revise it under several conditions.
If distribution rates return to their historical range over the next four to six quarters while secondary discounts continue to narrow across the quality spectrum rather than only at the top, the cash-flow interpretation weakens substantially and the recent period looks more like an exit-window problem than a repricing. If the 2020 and 2022 vintages recover materially as they mature — and both are young enough that they could — the negative excess return in Figure 2 will read as a timing artefact rather than a structural break. And if fee structures compress meaningfully in response to competitive pressure, the arithmetic changes in favour of the asset class without anything else needing to.
What we would not accept as evidence is a recovery in reported net asset values unaccompanied by cash. The distinguishing feature of this period is precisely that marks and money have diverged, and any assessment that resolves the question using marks alone is answering an easier question than the one being asked.
The allocation question
For an institution setting policy today, the useful frame is not how much to allocate to private markets. It is what specifically the institution believes it is being paid for, and whether the portfolio it actually owns delivers that thing.
If the answer is operational improvement, the portfolio should be concentrated in a small number of managers with a demonstrable record of it, and the institution should be comfortable that concentration is the price of the thesis. If the answer is scarcity of capital, the portfolio should be tilted toward situations where that scarcity is structural rather than cyclical, because cyclical scarcity closes exactly when everyone else notices it. And if the honest answer is that the institution allocated to private markets because its peers did and the reported volatility was flattering, then the last three years have delivered an expensive but genuinely useful piece of information, and the correct response is to act on it rather than to wait for the marks to make the problem go away.
Patience remains a virtue in investing. It was never, on its own, a strategy. The institutions that treated it as one are now discovering what it cost.
