A carry table with ten names can be a two-person firm. The Herfindahl index, the unallocated reserve and the deals each person leads show which.
Measure carry concentration with the Herfindahl index: square each holder's share of the carry pool in points out of 100 and add the squares, then divide 10,000 by the result to get the effective number of holders. In an illustrative GP, Fund III scores 2,364, the equivalent of 4.23 equal holders. Fund IV appears to fall to 1,498, but once the 10-point unallocated reserve is attributed to the founders who control it, the index is 1,898 and the founders hold 55 points, not 45, while leading 21.4 per cent of the current portfolio's deals.
Worked in full in Operational Due Diligence in Private Equity by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Carry allocation is where a manager's succession plan shows up in money. A due diligence questionnaire usually returns a list of names and points, and the number of names is the figure that ends up in the memo. That number says little. A firm where nine people share the carry can be a firm where two people own most of it and seven are on retention terms, and the people doing the deals may be among the seven. The test below turns the table into three numbers an operational due diligence reviewer can track from fund to fund.
| Holder | Role | Fund III points | Fund IV points | Deals led |
|---|---|---|---|---|
| A | Founder | 35 | 25 | 2 |
| B | Founder | 30 | 20 | 1 |
| C | Partner | 10 | 12 | 3 |
| D | Partner | 8 | 10 | 3 |
| E | Principal | 6 | 8 | 2 |
| F | Principal | 5 | 6 | 2 |
| G | Vice president | 3 | 4 | 1 |
| H | Vice president | 2 | 3 | 0 |
| I | Associate pool | 1 | 2 | 0 |
| Reserve | Unallocated, held by founders | 0 | 10 | 0 |
| Total | 100 | 100 | 14 |
Deals led is a rough proxy for contribution: it ignores sourcing, board work and the founders' role in fundraising. It is still the proxy LPs can verify, from board seats and deal memos, and it is the one that shows whether the economics follow the work.
HHI = Σ (points held)2, with points out of 100
Effective number of holders = 10,000 ÷ HHI
The index runs from 10,000, one holder with everything, down to 10,000 divided by the number of holders for an equal split. In Excel, with points in C2:C11: =SUMSQ(C2:C11) and =10000/SUMSQ(C2:C11).
Fund III: 352 + 302 + 102 + 82 + 62 + 52 + 32 + 22 + 12 = 2,364. Nine names, but the pool behaves like 4.23 equal holders, and the two founders hold 65 points.
Fund IV has a 10-point reserve, set aside for future hires and promotions. Treated as a tenth holder, the index falls to 1,498 and the effective number rises to 6.68: a sharp move towards the next generation. But the reserve is held by the founders until it is granted, and if it is never granted it stays with them. Attributed to them equally, the index is 1,898 and the effective number 5.27.
| Measure | Fund III | Fund IV, reserve as a holder | Fund IV, reserve with founders |
|---|---|---|---|
| HHI | 2,364 | 1,498 | 1,898 |
| Effective number of holders | 4.23 | 6.68 | 5.27 |
| Founders' points | 65 | 45 | 55 |
| Next generation's points | 35 | 55 | 45 |
The manager's presentation will say the founders went from 65 to 45 points, a 20-point transfer. The economic transfer is 10 points, half of what is claimed, until the reserve is granted.
Carry-to-contribution ratio = share of carry ÷ share of deals led
A ratio above 1 means a holder's carry exceeds their share of the deal work; below 1, the reverse.
The founders lead 3 of the 14 deals, 21.4 per cent, and control 55 per cent of Fund IV carry, a ratio of 2.57. The two partners, C and D, lead 6 deals, 42.9 per cent, and hold 22 points, a ratio of 0.51. B leads one deal and holds 25 points with the reserve share, a ratio of 3.50. The next generation as a whole leads 78.6 per cent of the deals for 45 per cent of the carry, a ratio of 0.57. These are the people the LP is backing to sell the portfolio, and the people a competitor would find easiest to hire away.
| Allocation | HHI | Effective holders | Top two | Founders |
|---|---|---|---|---|
| Fund III, for reference | 2,364 | 4.23 | 65 | 65 |
| Fund IV as allocated | 1,898 | 5.27 | 55 | 55 |
| Reserve granted to C and D | 1,668 | 6.00 | 45 | 45 |
| Plus founders cede 10 points to E, F and G | 1,432 | 6.98 | 37 | 35 |
| Equal split among nine, benchmark only | 1,111 | 9.00 | 22 | 22 |
Granting the reserve to the two partners who lead the most deals would take C to 17 points and D to 15 and bring the effective number to 6.00. That is the concrete request an LP can make: not a vague commitment to succession, but a grant date for named points.
Counting names and reading one fund. Nine or ten carry holders sounds broad in both funds, and the headline founders' share falls by 20 points, so the memo concludes that succession is under way. Attributing each point to whoever controls it shows half that movement, and the contribution ratio shows the gap between who does the work and who is paid for it. The same weighting logic, applied to people who have left rather than those who stay, is in measuring investment team turnover at a GP; and a pool this concentrated feeds straight into the probability of a key person event.
Compute the Herfindahl index and the effective number of holders for each fund, attribute every unallocated point to whoever controls it, and set each holder's share against the deals they lead: here 1,898, 5.27 and a founders' ratio of 2.57. Track the three numbers from fund to fund and ask for dated grants of the reserve. Operational Due Diligence in Private Equity assesses managers beyond the investment case, and the free workbook for the book shows how a rating model can be built so that one structural gap is not averaged away.
There is no single benchmark, so compare the firm with itself and with the work. In the illustrative case the founders lead 21.4 per cent of current portfolio deals and control 55 per cent of Fund IV carry, a ratio of 2.57, while the two partners lead 42.9 per cent of deals for 22 per cent of carry. The trend across funds and the gap to contribution say more than any market figure.
Because points that have not been granted are usually controlled by the founders until they are. In the illustrative Fund IV, a 10-point reserve makes the founders' share look like 45 points when they control 55, and lowers the apparent Herfindahl index from 1,898 to 1,498. Ask who holds the reserve, who decides grants, and what has happened to past reserves.
Divide 10,000 by the Herfindahl index computed on points out of 100. Nine holders with 35, 30, 10, 8, 6, 5, 3, 2 and 1 points give an index of 2,364, so the pool behaves like 4.23 equal holders. An equal split among nine would give 1,111 and an effective number of 9.00.
This article is one calculation from Operational Due Diligence in Private Equity. 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 1614 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.