Articles

What is a good DPI for a private equity fund?

Why DPI has to be read against the fund's age and its own TVPI, and how three funds with the same IRR can show very different DPIs at year eight.

Julian R. Sterling · from Private Equity Investor Relations

A good DPI depends on the fund's age, so judge it against the path the fund's own TVPI implies, not against a single number. On an illustrative buyout fund heading for 1.81x and 12.0 per cent net, DPI is 0.14x in year 4, 0.52x in year 6, 1.04x in year 8 and 1.54x in year 10. These are illustrative model outputs, not market benchmarks: a fund of that quality still below 1.0x after year 8 is returning cash later than its marks suggest, and the investor should ask why.

Worked in full in Private Equity Investor Relations by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →

DPI, distributions over paid-in capital, has become the figure limited partners quote first, because it is the only one that is entirely cash. That makes it easy to misread in both directions: a young fund with a low DPI may be perfectly healthy, and an older fund with a high DPI may simply have sold early and cheaply.

The model

Per 100 of commitment, the fund calls 25, 25, 20, 15 and 7 over five years, 92 in all. Its portfolio compounds at 12 per cent a year net, and it distributes a rising share of value from year 3: 5 per cent of value, then 10, 15, 20, 25, 30, 40, 50, 70 and finally all of it in year 12. Nothing here is market data; it is a deliberately plain fund so that the shape of the DPI curve is visible.

DPI year by year

Illustrative fund, per 100 committed. Net IRR 12.0 per cent to date in every year from year 2.

Year	Paid in	Cumulative distributions	NAV	DPI	TVPI	DPI as % of TVPI
3	70	3.0	76.4	0.04x	1.13x	4%
4	85	11.5	92.0	0.14x	1.22x	11%
5	92	27.0	94.6	0.29x	1.32x	22%
6	92	48.2	84.7	0.52x	1.44x	36%
7	92	71.9	71.2	0.78x	1.56x	50%
8	92	95.8	55.8	1.04x	1.65x	63%
9	92	120.8	37.5	1.31x	1.72x	76%
10	92	141.8	21.0	1.54x	1.77x	87%
12	92	166.2	0.0	1.81x	1.81x	100%

Reading the result

Three markers come out of the table. Distributions start to matter in year 5, when DPI reaches 0.29x. The fund returns its paid-in capital in year 8, at 1.04x. And by year 10 most of the value is cash: 87 per cent of TVPI has been distributed. On this illustrative fund, "good" is roughly half the capital back by year 6, all of it by year 8 and 1.5x by year 10. These are the shape of one model, not peer-group statistics; where a fund sits against funds of its own vintage is a separate exercise, worked in how private equity quartile rankings are calculated.

The more useful test is the last column. DPI as a share of TVPI says how much of the reported value has been converted into cash, and it can be compared across funds of different quality. A year-8 fund showing 1.9x TVPI and 0.6x DPI has realised 32 per cent of its value against 63 per cent here: its marks are doing much more of the work, and the investor should ask what is holding the exits back. The three ratios themselves are defined and computed in how to calculate TVPI, DPI and RVPI.

What if the exits come at a different pace?

DPI by year under four exit patterns, per 100 committed.

Pattern	Year 4	Year 6	Year 8	Year 10	Final TVPI	Net IRR	Capital back in year
Base case	0.14x	0.52x	1.04x	1.54x	1.81x	12.0%	8
Exits two years later	0.00x	0.21x	0.70x	1.31x	2.17x	12.0%	9
Early exits, quick flips	0.31x	0.78x	1.21x	1.48x	1.57x	12.0%	7
Lower growth, 8% a year	0.13x	0.47x	0.90x	1.29x	1.48x	8.0%	9

The first three funds earn exactly the same 12.0 per cent a year, yet their year-8 DPI runs from 0.70x to 1.21x. The quick seller looks best on DPI at every age to year 8 and ends with the lowest multiple of the three. The patient holder looks worst for eight years and ends highest, at 2.17x. DPI at a given age measures pace, not quality; quality shows up in TVPI and the IRR, and the honest question is whether the marks behind the TVPI will convert.

The lower-growth fund is the one to watch. At year 8 its 0.90x DPI looks only slightly behind, but its TVPI, 1.40x, and its 8.0 per cent IRR show a weaker fund, not a slower one. DPI alone cannot tell the two apart.

The common mistakes

• Comparing DPI across vintages. A 2016 fund at 1.3x and a 2020 fund at 0.3x may be identical funds observed at different ages.
• Reading a high early DPI as skill. Quick flips raise DPI and lower the final multiple.
• Dividing by commitments. DPI uses paid-in capital. In year 8 here, 95.8 over 92 is 1.04x; over 100 committed it would read 0.96x.
• Ignoring recallable distributions. Cash that can be called back flatters DPI until the recall right lapses.
• Ignoring how the cash was raised. A distribution funded by borrowing against NAV lifts DPI without an exit, as worked in what a NAV loan distribution does to DPI.

Takeaway

Judge DPI against the fund's age and against its own TVPI. As an illustrative working shape for a fund of this quality, not a market benchmark: about 0.5x by year 6, 1.0x by year 8, 1.5x by year 10, and a realised share of value that climbs past half by year 7. The cash-flow curve behind it is in how the private equity J-curve is calculated, and DPI, TVPI and both rates can be computed on your own fund in the free workbook for this case.

Questions readers ask

When should a private equity fund reach 1.0x DPI?

In an illustrative model fund heading for 1.81x and 12.0 per cent net, not a market benchmark, paid-in capital comes back in year 8, at 1.04x. Exit timing moves this by a year or more without changing the return: a fund that sells two years later reaches it in year 9, one that sells early in year 7. A fund still well below 1.0x after year 9 deserves questions.

Is DPI more important than TVPI?

Neither is enough alone. DPI is cash and cannot be marked; TVPI includes the firm's own valuations. In the worked case three funds earning the same 12.0 per cent show year-8 DPIs from 0.70x to 1.21x, while the best final multiple, 2.17x, belongs to the fund with the lowest DPI. Read DPI as a share of TVPI.

What does DPI divided by TVPI show?

The share of reported value already returned in cash. In the worked case it is 36 per cent in year 6, 63 per cent in year 8 and 87 per cent in year 10. A year-8 fund at 1.9x TVPI and 0.6x DPI has realised only 32 per cent, so its marks carry much more of the reported performance.

Also on this site

• How do you calculate TVPI, DPI and RVPI for a private fund?Divide by paid-in capital: 791 distributed and 143 of NAV on 564 called give DPI 1.4025x, RVPI 0.2535x and TVPI 1.6560x. Worked case.
• Six ways to state the same fund's return, all of them trueOne fund, one set of cash flows: DPI 1.216x, TVPI 1.784x, net IRR 18.04% or 7.12%, PME 1.323x. The eleven-point gap is the valuation, and it is not cash.
• How do you calculate a private equity J-curve and its trough?Cumulate net cash flows (distributions minus calls) year by year. An illustrative 1.74x fund bottoms at 58 per cent of commitment and breaks even in year 7.
• What does a NAV loan distribution do to DPI and TVPI?It lifts DPI at once and lowers TVPI over time: $147M distributed from a $150M NAV loan takes DPI from 0.500 to 0.663 and costs 0.047 of TVPI over three years.
• How are private equity quartile rankings calculated?Against the peer upper quartile, median and lower quartile. A 17.9 per cent IRR is top quartile at a 17.80 threshold and second at 18.20, on the same peers.
• How do you calculate a loss ratio for a private equity track record?Losses on deals below cost over total invested. $59.5m lost on $465.0m is a 12.8 per cent loss ratio, though a third of the deals lost money.

Reading guide: private equity and private markets → · All 453 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.