Three loss ratios from one twelve-deal record, realised against unrealised, and why a threshold measure jumps when a large deal sits near 1.0x.
The capital loss ratio of a private equity track record is the sum of cost minus value on every deal worth less than it cost, divided by total capital invested. On the illustrative twelve-deal record below, $59.5m lost on $465.0m invested gives a loss ratio of 12.8 per cent, even though a third of the deals lost money and nearly a quarter of the capital went into them.
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 →
Allocators ask for the loss ratio in almost every due diligence questionnaire, because a gross multiple can hide a pattern of losses behind one outsized winner. The difficulty is that "loss ratio" is used for at least three different calculations, and each gives a different figure from the same record. An investor relations team should present all three, labelled, before an allocator computes them and asks why only one was shown.
| Deal | Status | Invested | Total value | Multiple | Loss |
|---|---|---|---|---|---|
| A | Realised | 40.0 | 128.0 | 3.20x | |
| B | Realised | 35.0 | 0.0 | 0.00x | 35.0 |
| C | Realised | 50.0 | 112.5 | 2.25x | |
| D | Realised | 25.0 | 17.5 | 0.70x | 7.5 |
| E | Realised | 45.0 | 99.0 | 2.20x | |
| F | Realised | 30.0 | 54.0 | 1.80x | |
| G | Realised | 20.0 | 6.0 | 0.30x | 14.0 |
| H | Unrealised | 55.0 | 93.5 | 1.70x | |
| I | Unrealised | 40.0 | 60.0 | 1.50x | |
| J | Unrealised | 30.0 | 27.0 | 0.90x | 3.0 |
| K | Unrealised | 60.0 | 66.0 | 1.10x | |
| L | Unrealised | 35.0 | 59.5 | 1.70x | |
| Total | 465.0 | 723.0 | 1.55x | 59.5 |
Capital loss ratio = Σ (cost − value) on deals below cost ÷ total invested
Loss-making capital ratio = Σ cost of deals below cost ÷ total invested
Loss rate by count = number of deals below cost ÷ number of deals
In Excel, with invested in C2:C13 and value in D2:D13:
=SUMPRODUCT((D2:D13<C2:C13)*(C2:C13-D2:D13))/SUM(C2:C13)
| Measure | Numerator | Denominator | Result |
|---|---|---|---|
| Capital loss ratio | 59.5 | 465.0 | 12.8% |
| Loss-making capital ratio | 110.0 | 465.0 | 23.7% |
| Loss rate by count | 4 deals | 12 deals | 33.3% |
All three are correct and they answer different questions. The capital loss ratio measures how much money was destroyed: 12.8 cents per dollar invested. The loss-making capital ratio measures how much money was exposed to bad decisions: 23.7 cents per dollar went into deals that came back below cost, and on average recovered 45.9 per cent of it. The count measures decision quality without regard to size: one deal in three lost money.
A related figure allocators like is the loss as a share of the gains: $59.5m of losses against $317.5m of gains on the winners, 18.7 per cent. It says how much of the winners' work the losers undid.
The same measures split by status tell a sharper story.
| Portfolio | Invested | Multiple | Capital loss ratio | Loss rate by count |
|---|---|---|---|---|
| Realised, 7 deals | 245.0 | 1.70x | 23.1% | 42.9% |
| Unrealised, 5 deals | 220.0 | 1.39x | 1.4% | 20.0% |
| Whole record | 465.0 | 1.55x | 12.8% | 33.3% |
On realised deals the loss ratio is 23.1 per cent; on unrealised deals it is 1.4 per cent. Some of that gap is real, because the weakest deals are often exited first, written off or sold to cut losses. Some of it is the valuation: an unrealised deal held at 1.10x has not yet been tested by a buyer. The allocator will read the unrealised figure as a floor, not an estimate.
| Scenario | Multiple | Capital loss ratio | Loss-making capital | Loss rate by count |
|---|---|---|---|---|
| Base case | 1.55x | 12.8% | 23.7% | 33.3% |
| Deal K marked from 1.10x to 0.80x | 1.52x | 15.4% | 36.6% | 41.7% |
| Deal B recovers 0.50x instead of nothing | 1.59x | 9.0% | 23.7% | 33.3% |
Marking the largest unrealised deal down by $18.0m barely moves the gross multiple, from 1.55x to 1.52x, but it moves the loss-making capital ratio from 23.7 to 36.6 per cent, because the whole $60.0m of cost crosses the line. Measures based on a threshold jump when a large deal sits near 1.0x. The multiple does not show that fragility; the loss ratios do, which is why allocators ask for them.
Quoting the smallest number without its definition. A deck that says "loss ratio 12.8 per cent" next to a record where a third of the deals lost money will be recomputed. State which measure, gross or net, on realised deals or the whole record, at which valuation date, and show the other two. The cost of disclosure is low; the cost of an allocator finding the 33.3 per cent on its own is the meeting.
Two other points come up in every review. Losses are measured deal by deal, never netted inside a fund before dividing, otherwise a winner in the same fund erases them. And the denominator is invested capital, not commitments, which would understate every ratio.
Report the capital loss ratio, the loss-making capital ratio and the count, split between realised and unrealised deals, with the definition beside each. Here that is 12.8, 23.7 and 33.3 per cent, with the realised book at 23.1 per cent. The track record schedules an allocator expects, and the measures that sit beside them, are in the free workbook for this case. For the fund-level multiples, see how to calculate TVPI, DPI and RVPI; for how few write-offs undo a programme's gains, how many co-investments one write-off erases.
There is no single benchmark, and any market figure should be sourced from the allocator's own data. What matters is the definition and the split. In the worked case the capital loss ratio is 12.8 per cent on the whole record but 23.1 per cent on realised deals, and allocators usually weigh the realised figure more heavily because it has been tested by a buyer.
Usually gross, at deal level, because fees and carry are charged at fund level and cannot be attributed cleanly to individual deals. State it either way. The $59.5m of losses in the worked case are gross deal losses against $465.0m invested; a net figure would need the fund's fees allocated across deals first.
Because it separates judgement from sizing. A manager can have a low capital loss ratio by sizing losers small, while still losing on many deals. In the worked case one deal in three, 33.3 per cent, came back below cost, while the capital lost was 12.8 per cent of invested.
This article is one calculation from Private Equity Investor Relations. 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 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.