Four identical deals held once through a fund and once as co-investments: the interim IRR, the cash trough, and what a quarter of exposure in co-investments does to a programme.
Co-investment removes most of the J-curve in returns, because it carries no management fee and no carry, but very little of the J-curve in cash. On four illustrative $20M deals that each return 2.0x, the fund's since-inception IRR after year one is minus 18.2 per cent; the same deals held as co-investments show 0.0 per cent, and turn positive a year earlier. The deepest cash outflow and the cash breakeven date barely move: 6.3 years for the fund, 6.0 for the co-investments.
Worked in full in The Co-Investment Practitioner by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
A fund's early returns are negative for two separate reasons. Fees start on the whole commitment from day one, while the capital they are charged on has not been invested yet. And new investments are held at cost for a year or two before any value creation is recognised. A co-investment has the second problem but not the first. Separating the two tells an LP how much of the J-curve a co-investment programme can actually remove, and the fund J-curve calculation on its own does not show that split.
| Input | Value |
|---|---|
| Deals, one a year from year 0 | 4 × 20 |
| Gross multiple on each deal, exit after 5 years | 2.00x |
| Marks: held at cost for the first 2 years, then rising evenly to exit | |
| Fund management fee, years 0 to 4 | 2.0 a year |
| Fund management fee, later years | 1.0 a year |
| Carried interest over an 8% preferred return, full catch-up | 20% |
| Co-investment fee and carry | none |
1. Fund contributions: 80 of deals + 13.0 of fees = 93.0 called.
2. Fund distributions: 160 of gross proceeds less 13.40 of carry (20% of the 67.0 profit) = 146.6 to investors, a net TVPI of 1.58x and an IRR of 10.2 per cent.
3. Co-investment: 80 in, 160 out, 2.00x and an IRR of 14.9 per cent.
4. Interim IRR at each year end: treat the NAV as if it were distributed on that date, through the fund's waterfall for the fund, so that accrued carry is deducted, and directly for the co-investment. In Excel, one column per year: =IRR(CashFlowsToDate + NAV in the last cell).
| Year | NAV | Fund TVPI | Fund IRR | Co-invest TVPI | Co-invest IRR |
|---|---|---|---|---|---|
| 1 | 40.0 | 0.91 | −18.2% | 1.00 | 0.0% |
| 2 | 60.0 | 0.91 | −9.4% | 1.00 | 0.0% |
| 3 | 86.7 | 0.98 | −1.0% | 1.08 | 5.4% |
| 4 | 100.0 | 1.11 | 4.4% | 1.25 | 9.1% |
| 5 | 80.0 | 1.31 | 8.0% | 1.50 | 12.0% |
| 6 | 60.0 | 1.42 | 8.9% | 1.75 | 14.0% |
| 7 | 33.3 | 1.52 | 9.7% | 1.92 | 14.7% |
| 8 | 0.0 | 1.58 | 10.2% | 2.00 | 14.9% |
The co-investment never shows a loss. It sits at cost, a 0.0 per cent IRR, for two years while the marks wait, then rises. The fund, holding exactly the same companies, is at minus 18.2 per cent after a year, still negative after three, and positive only in year four. All of that difference is fees: the assets are identical. That is the part of the J-curve co-investment removes.
The cash J-curve is another matter. The fund's cumulative net cash flow bottoms at minus 90.0 in year four, 97 per cent of what it called; the co-investments bottom at minus 80.0, all of what they invested. Both wait for the same exits, so cash breakeven comes after 6.3 years for the fund and 6.0 for the co-investments. An LP that needs liquidity sooner gets very little from co-investing; secondaries, which buy assets already part-way to exit, do much more, as worked in how much secondaries mitigate the J-curve.
| Co-investment share | Year 1 | Year 2 | Year 3 | Year 4 | Final | Final multiple |
|---|---|---|---|---|---|---|
| 0% | −18.2% | −9.4% | −1.0% | 4.4% | 10.2% | 1.58x |
| 25% | −14.0% | −7.1% | 0.5% | 5.5% | 11.3% | 1.67x |
| 50% | −9.5% | −4.8% | 2.1% | 6.7% | 12.5% | 1.77x |
| 100% | 0.0% | 0.0% | 5.4% | 9.1% | 14.9% | 2.00x |
A quarter of exposure in fee-free co-investments lifts the year-one figure from minus 18.2 to minus 14.0 per cent and brings the first positive year forward from four to three. The improvement is roughly proportional to the share, which makes it easy to forecast: decide the co-investment share, and the shape of the programme's early reporting follows.
Real co-investments are rarely fee-free in every case, and some carry transaction or monitoring fees of their own. Each of those puts a small dip back into the curve; what they cost is worked in what transaction and monitoring fees cost a co-investor.
Reading the early gap as selection skill. After two years this co-investment book shows 0.0 per cent against minus 9.4 for the fund, and after three, 5.4 against minus 1.0. A committee looking at those numbers may conclude that the co-investment team picks better deals. It picked the same deals. Compare co-investments with the fund's gross performance on the same companies, or with the fund's net figure only once both are mature. The opposite error is promising the board that co-investment will fix a liquidity problem: it shortens the wait for cash by a fraction of a year, not by years.
Co-investment removes the fee-driven J-curve: minus 18.2 per cent becomes 0.0 in year one on identical deals, and returns turn positive a year sooner. It does not remove the cash J-curve, which depends on exits: breakeven at 6.0 years against 6.3. The programme-level arithmetic of the book, from fee savings to the cost of a follow-on, is in the free companion files for this book.
Fees are charged from the first year while investments are held at cost, so early net value sits below contributions. In this illustrative fund, 2.0 a year of fees on top of 20.0 of investment leaves a TVPI of 0.91 after two years and a since-inception IRR of minus 18.2 per cent after one, even though every deal ends at 2.0x.
Only a little. Co-investment capital is invested at once and pays no fees, but it still waits for exits. Here the co-investments break even on cash after 6.0 years against 6.3 for the fund, and the deepest cumulative outflow is 80.0 against 90.0, about the same share of the capital each deploys.
The effect is roughly proportional. With a quarter of exposure in fee-free co-investments, the year-one IRR improves from minus 18.2 to minus 14.0 per cent and turns positive in year three instead of four; with half, it is minus 9.5 per cent, and the final multiple rises from 1.58 to 1.77.
This article is one calculation from The Co-Investment Practitioner. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: private equity and private markets → · All 1340 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.