Private and public markets belong in one portfolio view

An investor with a securities depot, a handful of fund commitments and a few direct stakes usually keeps two sets of numbers: the bank's report for the liquid part, a spreadsheet for the rest. That separation hides what matters. Private holdings are largely equity risk that is reported late and in smoothed form, they draw on the same cash as everything else, and their share of the whole can only be judged against the liquid part. How much of a portfolio may be private follows from the liquidity the investor needs, not from an illiquidity premium that may or may not be there.

Is there an illiquidity premium?

The illiquidity premium is the extra return an investor is supposed to earn for locking money up for ten years. It is the standard argument for private markets, and the evidence for it is mixed.

On the supportive side, Harris, Jenkinson and Kaplan studied nearly 1,400 US funds and found that buyout funds had on average outperformed the S&P 500 by more than 3 per cent a year. In the same data, venture funds outperformed public shares in the 1990s and underperformed them in the 2000s (NBER working paper 17874, later published in the Journal of Finance). On the sceptical side, Ilmanen, Chandra and McQuinn of AQR found that private equity's realised and expected edge over cheaper public counterparts shows a decreasing trend over time. They suggest that investors may accept a smaller premium because they value the smoother reported returns of illiquid assets (Demystifying Illiquid Assets, Journal of Alternative Investments).

Both studies describe averages of institutional portfolios in the United States. A private investor with five funds does not hold the average, pays the same fees or higher ones, and may not get into the funds that lifted it. We would treat the premium as an open question. Private markets have a place in a portfolio for other reasons: access to companies that are not listed, managers who can change a business, and for angels the work itself. A plan that only adds up if private assets return three points more than shares rests on something nobody can promise.

Why do private holdings look calmer than they are?

A share is priced every second. A fund stake is valued four times a year by the manager, on the basis of comparable companies and recent financing rounds, and the report arrives weeks after the quarter ends. A direct stake in a start-up is often carried at the price of the last round until the next one.

The consequence is that private holdings appear to move little and to move independently of the stock market. Neither is a property of the companies. A software company owned by a buyout fund is exposed to the same economy and the same interest rates as its listed competitor. Its valuation just follows later and in smaller steps, because valuers average and wait. The AQR authors put it plainly: smoothed returns understate the economic risk.

Seen separately, this produces two errors. In a falling market, the investor looks at the depot, sees the loss, and takes comfort from the private part that "held up". Two quarters later the fund reports follow. And in an allocation spreadsheet, low measured volatility and low correlation make private assets look like a diversifier, which pushes their share up. We wrote about the first effect after the tariff shock in April 2025.

Seen together, the honest working assumption is simple: count private equity and venture as equity when you ask how much equity risk you carry.

What do the two sides have to do with each other in cash terms?

The private side makes claims on the liquid side. An unfunded commitment is a promise to pay at short notice, often within ten business days, and the money for it sits in the depot or the bank account. Distributions flow the other way, at times nobody controls.

An example with invented numbers. A portfolio of EUR 5 million consists of EUR 2 million in fund stakes and direct holdings at their last reported values, EUR 2 million in shares and EUR 1 million in bonds and cash. Unfunded commitments are EUR 800,000. The private share is 40 per cent, and the open commitments are 27 per cent of the liquid assets.

Shares fall by 35 per cent. The liquid part is now EUR 2.3 million. The private part is still reported at 2 million, so its share has risen to 47 per cent without a single purchase. The unfunded commitments are now 35 per cent of liquid assets, and in such phases distributions tend to dry up while calls continue. If the investor also needs EUR 150,000 a year to live on, the EUR 1 million in bonds and cash is spoken for by the open commitments and one year of spending. After that, shares have to be sold at depressed prices.

None of this can be read from the bank's report or the fund spreadsheet alone. It appears when both are in one table with the commitments beside them.

What do fees look like side by side?

A typical fund charges a management fee of around 2 per cent a year on the commitment during the investment period and carried interest of 20 per cent of the profit above a hurdle. On a commitment of EUR 250,000, five years of management fee come to EUR 25,000, a tenth of the commitment, before any carry. An index fund with costs of, say, 0.2 per cent charges EUR 2,500 on the same amount over the same time.

That gap is not an argument against funds. It is the bar the manager has to clear before the investor is better off than with the index, and it is the reason to compare net figures with net figures: the fund's net IRR and TVPI after all fees against what the same cash flows would have earned in a share index. Only a portfolio that holds both sides can make that comparison.

What share of private assets is defensible, and for whom?

There is no number that fits everyone, and we are not giving investment advice. But the question can be turned into three tests that anyone can run on their own figures.

The first is the spending test: what do you need from the portfolio over the next five years, including taxes and known purchases? That amount should be liquid and not in shares. The second is the call test: after a fall of 35 to 40 per cent in the share portfolio, can all unfunded commitments still be paid without selling shares at the low? The third is the patience test: private holdings return money on their own schedule, so the amount in them should be money you can leave for ten years or more.

An entrepreneur who has sold a company, lives on a fraction of the proceeds and understands the sector can pass all three with half the portfolio in private holdings. Someone who lives on the portfolio's income, or whose wealth is mostly in one property, may fail the call test at 15 per cent. For the first investor the limit comes from concentration in one kind of risk; for the second from liquidity. Our piece on rebalancing with illiquid assets describes how to steer within such limits when half the portfolio cannot be traded.

The practical step is to put both sides into one table once: depots at current prices, fund stakes at the last reported NAV with its date, direct holdings at their last round, unfunded commitments in their own column. We built Valued to keep that record from the documents themselves, but the first version can be a spreadsheet and an afternoon. The dates beside the private values are the point of the exercise: they tell you how old the calm part of your portfolio is.