The tariff shock and the denominator effect in a private portfolio

Between the US tariff announcement of 2 April 2025 and the close on 8 April, the S&P 500 lost about 12 per cent, while fund stakes and start-up holdings stayed at the values of their last quarterly report. The private share of a mixed portfolio therefore rose without anything being bought. That is the denominator effect, and for a private investor it is arithmetic, not a signal: the figure to check after a week like this is whether the liquid part still covers the capital calls that are coming.

What happened between 2 and 9 April

On Wednesday 2 April the United States announced a baseline tariff of 10 per cent on almost all imports from 5 April, and higher country-specific rates from 9 April, among them 20 per cent for the European Union and 34 per cent for China. The S&P 500 fell 4.8 per cent on Thursday and a further 6.0 per cent on Friday. In Frankfurt the DAX opened on Monday 7 April more than 10 per cent lower and closed the day at 19,789.62 points, down 4.13 per cent, with its gains for the year gone.

Yesterday, 9 April, the country-specific rates were suspended for 90 days for every country except China, whose rate was raised to 125 per cent. The 10 per cent baseline stays. The S&P 500 rose 9.5 per cent in one session to 5,456.90 points. It is still below where it stood on 2 April, and nobody knows what happens when the 90 days are over.

What is the denominator effect?

An allocation is a fraction. The private holdings are the numerator and the whole portfolio is the denominator. When the listed part falls and the private part is carried at an unchanged value, the denominator shrinks and the fraction grows.

Say a portfolio of EUR 5,000,000 consists of EUR 2,500,000 in listed shares, EUR 1,000,000 in bonds and cash, and EUR 1,500,000 in fund stakes and direct holdings at their last reported values. The private share is 30 per cent. If the shares fall by 12 per cent, they are worth EUR 2,200,000 and the portfolio EUR 4,700,000. The private share is now 31.9 per cent. Nothing was called, nothing was bought, and the investor is two points above the target.

The term comes from institutions. A pension fund with a board-approved ceiling of 30 per cent for private assets is in breach at 31.9 and has to explain itself, stop new commitments or sell stakes on the secondary market. A private investor has no board. The ceiling, if there is one, is a note in a spreadsheet, and nobody enforces it. This is an advantage and should be used as one.

Why the private marks did not move, and when they will

A fund values its holdings at a reporting date, normally the end of a quarter, and the report arrives weeks later. The report most LPs hold today shows values as at 31 December 2024. The next one will show 31 March 2025, which is two days before the announcement. The first valuation date after the shock is 30 June, and that report will reach investors in August or September.

So the private half of the ratio will reflect this week five months from now at the earliest, and even then only in part. A manager values an unlisted company from the multiples of comparable listed companies, from recent transactions and from the company's own plan. By 30 June the listed comparables may be lower, the same or higher than today. A start-up that last raised money in 2024 is carried at that round's price until there is a reason to change it, and a tariff schedule that has been suspended for 90 days is not obviously such a reason.

This cuts both ways. If shares recover from here and the funds write their holdings down by 10 per cent at mid-year, the same portfolio shows shares at EUR 2,405,000 (the index is 3.8 per cent below 2 April as of last night), private holdings at EUR 1,350,000 and a private share of 28.4 per cent. The ratio would then be below target because of a markdown that describes April. Anyone who steers by the percentage compares today's prices with last quarter's, every time.

What to check instead of the percentage

Three things are real behind the arithmetic, and all three are about cash.

The first is the unfunded commitment. A fund calls what it was promised whether markets are up or down, with a notice period of ten days or so. In the example, assume EUR 600,000 is still uncalled across all funds. Against EUR 1,000,000 in bonds and cash this is comfortable. Had the plan been to fund calls by selling shares, the investor would have sold on Monday at prices 12 per cent lower, and the question becomes how much of the listed part is truly available at a bad moment. The honest test is to take the liquid assets at this week's lowest prices, not at the year-end statement, and hold them against everything that can still be called.

The second is distributions. A fund distributes when it sells a company, and sales need a buyer who is willing to name a price. In weeks like this one buyers wait. Whoever counted on distributions in 2025 to pay for calls in 2025 should plan the calls without them.

The third is the age of each figure. A portfolio overview that adds a depot value from this morning to a fund NAV from December is adding two different dates. That is unavoidable, and it is fine as long as the date stands next to each number. The private values are not stable because the companies are stable. They are stable because nobody has revalued them yet.

What we would not do

We would not sell a fund stake to repair a ratio. A secondary sale takes months, needs the manager's consent and is priced at a discount to a NAV that the buyer will assume is too high. The ratio moved two points in four days and came most of the way back in one afternoon.

We would not cancel a commitment that was planned for this year because the private share reads 31.9 per cent. A commitment signed now is called over the next four or five years. It should be decided on whether the cash for it will be there in those years, which is the first check above, and on the fund.

And we would not read anything into the calm of the private column. Open the latest quarterly report of each fund, write its valuation date next to the NAV, add up what can still be called, and compare that sum with what is liquid today. That takes an evening and answers the only question this week has actually raised.