← Insights
Portfolio

The J-curve in private equity, and why a young fund's IRR tells you so little

Almost every fund starts below zero. A worked ten-year example shows the two J-curves an LP actually lives with, one in returns and one in cash, why they bottom out years apart, and what to report while a fund is still in the dip.

In short: a private equity or venture fund almost always shows a negative return in its first years. Fees and set-up costs are charged from day one, while the investments they paid for are still held at cost. Only when holdings are marked up and exits return cash does the line turn upward. Plotted over time, that is the J-curve.

Most explanations stop there. For someone who has to report to a family or plan liquidity, the useful part comes after: there are two J-curves, not one, and they bottom out years apart.

A fund, year by year

Take a €10 million commitment to a buyout fund. The figures below are invented but typical in shape: capital is called over six years, the first distribution arrives in year four, and the fund is fully wound up after ten years, having returned 1.82 times what was paid in.

Calls and distributions are assumed to fall in the middle of each year, and the NAV is the manager's mark at year end. The IRR for each year is computed as if the position had been sold at that NAV.

YearPaid in (cum.)Distributed (cum.)NAVNet cashDPITVPIIRR
12.00.01.7−2.00.000.85−27.8%
24.50.04.0−4.50.000.89−11.9%
36.50.06.3−6.50.000.97−2.1%
48.00.58.5−7.50.061.125.7%
58.82.09.6−6.80.231.3210.3%
69.25.08.6−4.20.541.4812.0%
79.28.56.5−0.70.921.6313.0%
89.212.53.5+3.31.361.7413.5%
99.215.51.0+6.31.681.7913.8%
109.216.70.0+7.51.821.8213.9%
Invented figures for a €10m commitment, € million. The IRR treats each year-end NAV as if it were cash.

Three things in that table are worth reading slowly.

The IRR is at its worst in year one, at −27.8%, and it is the least meaningful number in the table. One year of fees against a portfolio still held at cost, annualised, produces a large negative figure from a small loss: TVPI at that point is 0.85, so fifteen cents in the euro.

The IRR turns positive in year four. That is the moment the fund stops looking like a mistake in a report. Nothing dramatic happened in the portfolio. The holdings bought in years one and two were revalued, and the first small exit came in.

The cash bottom comes in year four, not year one, at €7.5 million out of pocket, and the position only gets back to zero in year eight. That is the second J-curve, and it is the one that matters for liquidity.

Two curves, years apart

  1. Year 1Return low pointIRR −27.8%, TVPI 0.85. Fees against a portfolio at cost.
  2. Year 4IRR turns positive, cash low pointIRR 5.7%, but €7.5m net out of pocket
  3. Year 7DPI near 1.0x0.92x: almost all paid-in capital is back
  4. Year 8Cash break-even+€3.3m net. From here, the fund is returning profit.
The same fund, read two ways. The return curve bottoms out first; the cash curve later and deeper.

The return curve is what IRR and TVPI show. It bottoms out early, because the early losses are small in absolute terms and only look large once annualised. It recovers as soon as the manager marks the portfolio up, which can happen long before any money comes back.

The cash curve is what your account sees: everything called, minus everything distributed. It keeps falling for as long as calls outweigh distributions, so its low point sits at the end of the investment period, not at the start. In this example the IRR was already positive when the family had the most money tied up.

For liquidity planning, only the second one counts. The calls still to come are covered in planning for unfunded commitments, and the cash low point is that plan's worst quarter.

What changes the shape

Fees on commitment. In the investment period the management fee is usually charged on the full commitment, not on what has been invested. That is most of the early dip. A fund that charges on invested capital has a shallower curve, without being a better fund.

Subscription lines. Many managers bridge investments with a credit facility and call capital months later. The LP's money is out for a shorter time, so the IRR rises and the early dip can almost disappear. The multiple does not improve; it usually falls slightly, because the interest is paid by the fund. A young fund with an unusually smooth start is worth asking about this before it is compared with another.

Strategy. Venture funds typically have a longer and deeper dip than buyout funds, because exits come later and early rounds are often held at cost for years. Secondaries funds buy positions that are already several years old, so their curve is shallow or barely visible. Comparing a three-year-old venture fund with a three-year-old secondaries fund on IRR mostly compares where they sit on their curves.

Valuation policy. How quickly a manager marks holdings up decides when the return curve turns. Two funds with identical cashflows can cross zero in different years. DPI is the only figure in the table that no valuation can move.

Reporting a fund that is still in the dip

A family member who sees −27.8% next to a fund in a quarterly report will ask whether something went wrong. Usually nothing has. Three habits make the report honest without making it alarming.

Lead with paid-in, DPI and TVPI, not IRR, for the first three or four years. TVPI at 0.85 says "fees and costs, not losses" much more plainly than an annualised rate. The arithmetic is in IRR, TVPI and DPI, and why your three sources disagree.

Show the vintage year next to every return. Comparing a fund with others that started in the same year removes most of the J-curve from the comparison, because they are all in the same place on it.

Look at the portfolio, not the fund. A family that commits to new funds every year has some funds in the dip and others distributing. Spread across vintages, the curves overlap and the portfolio's cash curve becomes much flatter. That is the main argument for steady commitment pacing, rather than committing to several funds in one good year.

What it does not excuse

The J-curve explains a negative return in years one to three. It does not explain a TVPI still below 1.0x in year six, or calls that keep coming after the investment period has ended. At that point the fund is not in the dip; it is underperforming, and it should be compared with its vintage peers like any other.

To check where a fund stands on its own cashflows rather than the manager's quarterly report, the free IRR, TVPI and DPI calculator works in the browser without uploading anything.

EN
EnglishDeutschEspañol