Why the same-day netting of a call and a distribution moves every multiple and no rate, which way it pushes DPI and TVPI, and how to rebuild the series gross.
Netting a distribution against a capital call on the same day leaves the IRR exactly where it was, but it changes every multiple, because the same amount comes off both the numerator and the denominator. On an illustrative fund that calls 100 and distributes 165, netting the 30 that overlap in years 3 to 6 lifts the final TVPI from 1.65x to 1.93x while the IRR stays at 11.83 per cent and the profit at 65. In mid-life the same netting pushes DPI the other way, from 0.80x to 0.71x.
Worked in full in Private Markets Performance by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
When a fund sells an investment in the same quarter as it needs money for a new one, the GP often sends a single notice: the investor receives the distribution minus the call, or pays the call minus the distribution. The investor's bank sees one net amount. The IRR is computed on net cash flows by date, so it cannot tell the difference. TVPI and DPI are ratios of two totals, distributions over paid-in capital, and both totals shrink by the netted amount. Subtracting the same number X from the top and bottom of a ratio pushes it away from 1: (D − X) / (C − X) is larger than D / C when D exceeds C, and smaller when it does not. The standard multiples assume both legs are recorded gross.
| Year | Calls | Distributions | Amount that can be netted | Net cash flow |
|---|---|---|---|---|
| 0 | 20 | 0 | 0 | -20 |
| 1 | 25 | 0 | 0 | -25 |
| 2 | 20 | 0 | 0 | -20 |
| 3 | 15 | 10 | 10 | -5 |
| 4 | 10 | 15 | 10 | 5 |
| 5 | 5 | 25 | 5 | 20 |
| 6 | 5 | 30 | 5 | 25 |
| 7 to 10 | 0 | 85 | 0 | 85 |
| Total | 100 | 165 | 30 | 65 |
Distributions in years 7 to 10 are 30, 25, 20 and 10. The overlap in each year is the smaller of the call and the distribution.
Netted amount in a period = MIN(call, distribution)
Gross TVPI = Σ distributions ÷ Σ calls Netted TVPI = (Σ distributions − Σ netted) ÷ (Σ calls − Σ netted)
In Excel: =SUM(C2:C12)/SUM(B2:B12) on the gross columns, and =SUMPRODUCT(--(C2:C12>B2:B12),C2:C12-B2:B12) for the distributions the investor actually saw net
Step 1: gross multiples. 165 ÷ 100 = 1.65x. The IRR on the net cash flow column is 11.83 per cent.
Step 2: what the investor's ledger shows if every overlap is netted. The calls it paid are 20, 25, 20 and 5, a total of 70. The distributions it received are 5, 20, 25 and then 85 from years 7 to 10, a total of 135. TVPI on the ledger is 135 ÷ 70 = 1.93x.
Step 3: the IRR. The net cash flow in every year is unchanged, so the IRR is still 11.83 per cent and the profit is still 165 − 100 = 135 − 70 = 65. Nothing economic has happened.
Step 4: mid-life. At the end of year 6, with a NAV of 75.0, the gross record shows paid-in of 100, distributions of 80, a DPI of 0.80x, RVPI of 0.75x and TVPI of 1.55x. The netted ledger shows paid-in of 70 and distributions of 50: DPI 0.71x, RVPI 1.07x and TVPI 1.79x. DPI is below 1, so netting lowers it; TVPI is above 1, so netting raises it. The interim IRR, 13.28 per cent, is identical on both.
| Amount netted | Final TVPI | DPI at year 6 |
|---|---|---|
| 0 | 1.65x | 0.80x |
| 10 | 1.72x | 0.78x |
| 20 | 1.81x | 0.75x |
| 30 | 1.93x | 0.71x |
Netting only the two notices in years 3 and 4, 20 in all, gives 80 of calls and 145 of distributions, 1.81x. A fund that recycles heavily, buying new companies with the proceeds of old ones, can overstate its TVPI by a quarter of a turn without a single wrong number in its cash ledger. The gap grows with the fund's success and with how much of its capital it turns over.
Netting on the same date is harmless to the rate. Withholding a distribution and using it for a later call is not. If the fund keeps the 10 it would have paid in year 3 and applies it to the year-4 call, calls fall to 90 and distributions to 155, TVPI reads 1.72x, and the IRR falls to 11.60 per cent, 23 basis points lower, because the investor's money is now held for a year without earning anything. That is recycling by retention, and the LPA, not the reporting convention, decides whether it is allowed.
The mistake is comparing a fund whose multiples come from the investor's cash ledger with one whose multiples come from the GP's gross records. The first fund in this example would rank above a rival reporting a true 1.80x, although its true multiple, 1.65x, is lower. The fix is to record both legs of every netted notice: the call at its gross amount as a contribution and the distribution at its gross amount as a distribution, with the net figure only as the cash that moved. The same discipline applies to fees, which belong in paid-in capital for a net multiple, as the paid-in capital question shows.
A quick test for a reported series: if the sum of calls is well below the commitment while the fund has made more investments than its size allows, look for netting.
Netting changes the multiples, never the same-date IRR. Here it turns 1.65x into 1.93x and a year-6 DPI of 0.80x into 0.71x, with 11.83 per cent unchanged. Rebuild the series gross before comparing funds. The cash flow series, with its contribution and distribution traps and the quarterly reconciliation test, is set out in the free workbooks and working documents for the book.
Not if both happen on the same date: the IRR is computed on the net cash flow of each date, which netting does not change. On the illustrative fund the IRR is 11.83 per cent gross and netted. It does change if a distribution is withheld and used for a later call: retaining 10 for one year cuts the IRR to 11.60 per cent.
Taking the same amount off the numerator and the denominator pushes a ratio away from 1. A mid-life DPI of 80 over 100, 0.80x, becomes 50 over 70, 0.71x, after netting 30. A final TVPI of 165 over 100, 1.65x, becomes 135 over 70, 1.93x. Ratios below 1 fall and ratios above 1 rise.
Record both legs gross: the call as a contribution and the distribution as a distribution, at their full amounts, with the net figure kept only as the cash that moved. On the illustrative fund that preserves 100 of paid-in and 165 of distributions, and a TVPI of 1.65x that compares with other funds.
This article is one calculation from Private Markets Performance. 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 1682 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.