Commitment pacing: how much to commit each year, and to how many funds

A private LP who wants a fixed share of the portfolio in private funds reaches it by committing a similar amount every year, not by committing the target once. In the model below, a steady EUR 200,000 a year builds to about EUR 1.07 million of fund value after nine years and stays there. A single EUR 1 million commitment never reaches EUR 1 million of value and is gone again after ten years. Commitment pacing is the name for working out that yearly amount.

Why a commitment is not an allocation

A commitment is a promise to pay on request. The fund calls it in portions over four or five years, and it starts paying back before the last portion is called. So the amount actually at work, the NAV, is at every moment smaller than the sum you signed, apart from a year or two at the peak if the fund does well.

Two consequences follow. To hold a given amount of NAV, the commitments to funds that are still alive must add up to considerably more than that amount. This is what over-commitment means, and it is arithmetic, not bravado. And a programme that stops committing shrinks by itself, because old funds pay out and nothing replaces them.

A pacing model you can rebuild in a spreadsheet

Institutional investors use cash flow models for this, the best known being the one Dean Takahashi and Seth Alexander of the Yale Investments Office published in the Journal of Portfolio Management in 2002. A private LP does not need its parameters. A fixed pattern per fund is enough to see the mechanics.

The assumptions, invented for illustration and deliberately unspectacular: each fund calls 25, 25, 20, 15 and 10 per cent of the commitment in its first five years and never calls the last 5 per cent. What is invested grows by 8 per cent a year after fees. From year four the fund pays out a rising share of its value each year and is wound up in year ten. Over its life such a fund returns about 1.5 times the paid-in capital. For EUR 100,000 committed, its life looks like this:

The fund's value peaks at about 92 per cent of the commitment in year five and is above 75 per cent for only three of the ten years. Averaged over its life, the NAV is a little over half of what was committed.

What a steady EUR 200,000 a year builds

Now commit EUR 200,000 to this kind of fund every year. The private portfolio is the sum of funds of different ages:

• After three years the NAV is about EUR 320,000.
• After five years, about EUR 690,000.
• After seven years, about EUR 950,000.
• From year nine it stays at about EUR 1.07 million, because each year one fund is wound up and one is added.

In the steady state the NAV is about 5.4 times the yearly commitment. That number is the whole pacing rule in this model: divide the NAV you want by 5.4. Someone with a EUR 4 million portfolio and a target of a quarter in private funds wants EUR 1 million and should commit a little under EUR 190,000 a year.

Two more figures come out of the same sheet. The unfunded commitment settles at about twice the yearly amount, here EUR 400,000. And at any time ten funds are alive with EUR 2 million of original commitments between them, against EUR 1.07 million of NAV. The over-commitment ratio that sounds alarming in the abstract is just the table above, added up.

What it costs in cash before it carries itself

The uncomfortable part of the model is the first seven years. Calls exceed distributions by EUR 50,000 in year one, EUR 100,000 in year two, EUR 140,000 in year three and EUR 150,000 in year four. Then the gap narrows, and from year eight the programme pays for itself: distributions of roughly EUR 280,000 a year against calls of EUR 190,000. Until then, about EUR 700,000 has to come from the rest of the portfolio.

That money must be available when called, and the EUR 400,000 of unfunded commitment must be coverable in a bad year in which distributions stop while calls continue. We covered how much to hold liquid in the article on planning liquidity for capital calls. Pacing decides the size of that problem years in advance.

Where the standard advice goes wrong

The shortcut is tempting: the target is EUR 1 million, so commit EUR 1 million this year across three or four funds and be done. In the model this produces a NAV of EUR 260,000 after one year, a peak of about EUR 920,000 in year five, and nothing after year ten. The investor is under target for four years, briefly near it, then under it again. And every fund has the same vintage, which is the one risk in private funds that picking good managers does not remove, as we argued in the piece on the J-curve and vintages.

The opposite mistake is to commit by feel: a lot in the year after a large distribution, nothing in the year after the stock market has fallen. This buys the vintages that follow good years and skips the ones that follow bad years, which historically has been the wrong way round often enough to make a rule worthwhile. The rule is dull. The same amount each year, adjusted slowly.

Adjust for three things. If the whole portfolio grows, the target grows with it, so the yearly amount rises by about the same rate. If distributions dry up, as they did from 2022, NAV stays in old funds longer and the private share overshoots without any new commitment; then reduce the yearly amount for a year or two, do not stop. And re-run the sheet with your own funds' actual calls once a year, because real funds call faster or slower than any pattern. Valued does that last step from the booked cash flows: it projects expected calls and distributions per quarter, fund by fund.

How many funds does a private LP need?

Fewer per year than is usually suggested, and for more years. With EUR 200,000 a year and typical minimum tickets, one or two funds a year is what is possible. After nine years that is nine to eighteen funds across nine vintages. We think this is enough. The spread across years does most of the work, and each additional fund in the same year adds less.

What we would not do is go below one commitment a year to reach a manager with a high minimum. Three large commitments in ten years is a concentrated bet on three vintages, however good the managers are. If the yearly amount is too small for direct commitments, a fund of funds or a feeder vehicle buys the spread at the price of a second fee layer, and that trade is often worth making for the first few years.

The first step is not choosing a fund. Write down the target NAV, divide by five to get a first yearly figure, list what the existing funds can still call, and check that the next four years of net calls can be paid from money you will not need. If that fails, the target is too high, and it is cheaper to learn it from a spreadsheet than from a default notice.