TVPI vs DPI, and why the gap matters as a fund ages
TVPI counts both cash and valuation. DPI is what a fund has paid back per dollar investors paid in, almost always as cash, and RVPI is the manager's valuation of what remains. In two illustrative funds, each with $100 million paid in, show a TVPI of 1.60x at year 7, but one has paid back 0.30x and the other 1.15x.
What are DPI, TVPI and RVPI, and how do they relate?
DPI, RVPI and TVPI are three ratios with the same denominator, the capital an investor has paid into a fund. DPI is what the fund has already distributed, almost always as cash, RVPI is the manager’s valuation of what the fund still holds, and TVPI is the two added together. A fund with a DPI of 0.50x and an RVPI of 1.10x has a TVPI of 1.60x, and only the first 0.50x has been paid.
The LP playbook is about judging a private fund by the cash it returns, and these three measures are where it starts. Invest Europe, the European industry body, defines DPI in its glossary this way:
the ratio of the cumulative distributions to LPs to paid-in capital
The same glossary defines RVPI as the residual value attributable to limited partners (LPs), net of carried interest, over paid-in capital, and TVPI as the sum of DPI and RVPI.
| Measure | Formula | What the numerator is | Who decides the numerator |
|---|---|---|---|
| DPI | Cumulative distributions to LPs ÷ paid-in capital | Distributions received, almost always cash | The fund’s record of cash paid |
| RVPI | Residual value, net of carry ÷ paid-in capital | Investments not yet sold | The manager’s valuation |
| TVPI | DPI + RVPI | Cash plus valuation | Both of the above |
Is DPI the same as TVPI? They are equal only when RVPI is zero, which happens when the fund has sold everything and paid out the proceeds. Until then TVPI is the larger figure, and the gap between them is the value the fund still holds, net of its liabilities and carry: mostly the manager’s valuation of unsold investments.
What is MOIC, and are net MOIC and TVPI the same? MOIC is a multiple of invested capital: total value divided by the capital put in. The GIPS Standards Handbook, published by CFA Institute, says the investment multiple, TVPI, is also known as the Multiple of Investment Capital (MOIC), and an SEC staff FAQ footnote names MOIC and TVPI side by side as performance measures. So where “net MOIC” means distributions plus residual value, after fees and carry, over paid-in capital, it is the same ratio as TVPI. In our reading, the two diverge when MOIC is quoted gross of fees, when it is measured against the capital invested in companies rather than the capital paid in, which also covers fees and fund expenses, or when it is for a single investment rather than a whole fund, so ask what sits in the numerator and the denominator and whether the figure is net.
How do DPI, RVPI and TVPI move over a fund’s life?
In a closed-end private equity fund, DPI stays near zero in the early years, RVPI carries almost all of TVPI until investments are sold, and the two cross late in the fund’s life. Bain & Company’s 2026 report gives the timescale for buyout funds: it found that holding periods at exit now hover around seven years, against an average of five to six years from 2010 to 2021 (Bain, February 2026). Most cash comes back as companies are sold, although a dividend recapitalisation, in which a portfolio company borrows to pay its owners, can return some before a sale and leaves the company in the fund (CNBC, October 2026).
The example below is hypothetical. A fund called Fund A has a $100 million commitment from an LP, all called in four years ($30 million, $30 million, $25 million and $15 million), and distributions of $10 million, $20 million, $40 million, $45 million, $45 million and $20 million in years 6 to 11. Every cash flow and NAV is an Unmatched Capital example input, stated net of fees and carry, with flows at each year end.
| Fund A, year end | Paid-in | Cumulative distributions | NAV | DPI | RVPI | TVPI | IRR since start |
|---|---|---|---|---|---|---|---|
| Year 3 | $85m | $0 | $88m | 0.00x | 1.04x | 1.04x | 3.3% |
| Year 5 | $100m | $0 | $135m | 0.00x | 1.35x | 1.35x | 11.3% |
| Year 7 | $100m | $30m | $130m | 0.30x | 1.30x | 1.60x | 10.4% |
| Year 9 | $100m | $115m | $62m | 1.15x | 0.62x | 1.77x | 9.7% |
| Year 11 | $100m | $180m | $0 | 1.80x | 0.00x | 1.80x | 9.3% |
These are Unmatched Capital’s calculations; IRR treats the year-end NAV as a final receipt. At year 7 the working is DPI = $30m ÷ $100m = 0.30x, RVPI = $130m ÷ $100m = 1.30x and TVPI = 0.30x + 1.30x = 1.60x. The NAV in that sum is already net of the general partner’s accrued carry, which is what Invest Europe’s definition requires and what a capital account statement should show. At year 3 the fund has called $85 million and its NAV is $88 million, so TVPI is 1.04x, none of it tested by a sale. At year 5 the fund reports a 1.35x TVPI and has paid nothing. The 0.30x of cash at year 7 is under a fifth of the 1.60x TVPI, and by year 11 DPI and TVPI are the same number.
Bain’s analysis of buyout funds with 2000 to 2015 vintages, using MSCI data to the third quarter of 2025, found that median TVPI begins to flatten after year eight. We built the later years of the example to follow that pattern: TVPI moves from 1.77x to 1.80x between years 9 and 11, while DPI climbs from 1.15x to 1.80x. Late in a fund’s life the total hardly changes and the split between paid and valued does, which makes DPI the figure still moving.
Can two funds have the same TVPI and very different DPI?
At year 7 Fund A has paid back 0.30x and a second fund, Fund B, 1.15x, yet both report a TVPI of 1.60x, because TVPI does not say how much has been paid. Fund B is hypothetical, with the same four capital calls as Fund A and distributions of $20 million, $30 million, $25 million and $40 million in years 4 to 7, then $25 million, $15 million and $7 million in years 8 to 10. Its year-7 NAV of $45 million is an example input.
| At year 7, per $100m paid in | Fund A | Fund B |
|---|---|---|
| DPI | 0.30x | 1.15x |
| RVPI | 1.30x | 0.45x |
| TVPI | 1.60x | 1.60x |
| IRR since start | 10.4% | 13.0% |
| TVPI if the NAV is marked down 30% | 1.21x | 1.465x |
| Final TVPI and IRR, when fully paid out | 1.80x and 9.3% | 1.62x and 12.1% |
The 30% mark-down illustrates sensitivity and is not a forecast. It cuts Fund A’s TVPI by 0.39x and Fund B’s by 0.135x, because $1.30 of every $1.60 in Fund A is valuation and $0.45 of every $1.60 in Fund B is.
Fund A ends with the higher multiple, 1.80x against 1.62x, because its manager held assets longer and they grew. Fund B ends with the higher IRR, 12.1% against 9.3%, because it paid its LP back sooner.
The publication’s rule is to measure adjusted DPI, its own measure: the share of TVPI paid out as cash, after taking out any loan-funded or recallable distribution. Read it against the fund’s age. At year 7 adjusted DPI is 19% for Fund A and 72% for Fund B. In the first years it is near zero whatever the manager does, so TVPI is all there is to read. Bain’s seven years is how long the average buyout company is held before it is sold (Bain, February 2026), not the age of a fund. A fund that invests over four years sells companies bought in years 1 to 4 and held for seven years in years 8 to 11, so by year 10 most of them have reached it. By then DPI should be the larger part of TVPI, and Fund A’s is: 1.60x of 1.80x, or 89%, at year 10. If it is not, most of the valuation has not yet been tested by a sale. Distributions funded by a dividend recapitalisation are worth asking the GP about but are not stripped, since only the GP can identify them. The “larger part” threshold is our rule of thumb, not a benchmark from a source. The year-10 point rests on Bain’s buyout figure; for a strategy whose companies are usually held longer or shorter, move it by the difference.
The counterargument is Fund A. Judging by DPI alone would have preferred Fund B at year 7 and been wrong on the final multiple. The rule answers it: at year 7, with companies that have been held for about five years on average, Fund A’s 19% sits inside the window in which it says to wait, and the rule turns a low adjusted DPI into a question for the manager only at about year 10, by which time Fund A’s is 89%. A fund still at 19% then would owe its LPs a list of the sales to come.
Is DPI better than IRR?
Fund A’s IRR is 11.3% at year 5 with a DPI of 0.00x, and falls to 9.3% by year 11 while its TVPI rises from 1.35x to 1.80x. The two measures answer different questions: DPI measures cash back and ignores timing, while IRR is a rate that depends on when cash moved and, for anything unsold, on the manager’s valuation. When they disagree, trust DPI on whether the return exists and IRR on how fast it came.
In the same Bain analysis of buyout funds with 2000 to 2015 vintages, IRR starts to stagnate around year seven and declines after that (Bain, February 2026), earlier than TVPI flattens. A fund can look worse on IRR and better on multiple in the same year, which is what Fund A does after year 5.
IRR is also the figure most open to adjustment. A fund that delays calling investors’ money by borrowing against the commitments shows a higher IRR and a lower multiple, as the subscription lines page works through. The SEC’s staff said in a February 2024 FAQ that presenting only a net IRR that includes the effect of fund-level subscription facilities could mislead investors (SEC Marketing Compliance FAQ). That answer concerns IRR. The same page mentions MOIC and TVPI only in a footnote, which treats them, like IRR, as performance measures outside its guidance on portfolio characteristics.
Where IRR is high and DPI is near zero, treat the IRR as a statement about the valuation, not about cash.
What is a good DPI for a private equity fund, and is 2x TVPI good?
A good DPI is one at or above the median of funds from the same vintage year and strategy at the same age, taken from a named benchmark provider, because no single figure is good for every fund. The one fixed marker is 1.00x, at which an LP has been paid back all it paid in.
Published medians by vintage come from data providers such as MSCI, whose figures Bain uses. Bain’s 2026 report compared DPI for 2017 to 2021 vintages with a benchmark defined as the average of the median DPI for 2010 to 2017 vintages, using data to the third quarter of 2025, and found that GPs underdelivered against it for every vintage in the cohort. Bain’s press release adds that distributions as a percentage of NAV have been below 15% for four consecutive years, at 14% for 2025 (Bain, February 2026). Benchmark medians differ by provider and are struck at a data date such as the third quarter of 2025, so any DPI comparison should name the provider, the strategy and the data date.
Is 2x TVPI good? The answer depends on timing and cash. A 2.00x multiple returned in one sum after 10 years on money paid in at the start is an IRR of 7.2% a year, and returned after 5 years it is 14.9% (our calculation, from 2 raised to the power of one tenth and one fifth). In the example, Fund A’s final 1.80x is 9.3% a year and Fund B’s 1.62x is 12.1%, so the smaller multiple has the higher annual rate. Fund B leaves the LP with more by year 11 only if the cash from both funds can be reinvested at more than about 4.7% a year (our calculation). A 2.00x TVPI at year 5 with a DPI of zero is also a valuation, where the same figure at year 12 with a DPI of 2.00x is cash.
Bain also found a link between cash and fund-raising: after the global financial crisis, the failure rate among buyout fund series closed in 2003 to 2008 that tried to raise a successor was much higher for funds in the third and fourth quartiles of DPI. In Bain’s words:
distributing cash to investors on time was a big predictor of who reraised
Can DPI and TVPI be flattered?
A 10% higher NAV for Fund A at year 7 would add 0.13x to TVPI and nothing to DPI, which is how TVPI can be flattered through valuations. DPI can be flattered through how distributions are funded, and an LP who reads either alone can be misled. The evidence below shows where the openings are, and in the study cited below the effect was concentrated among small, young and low-reputation firms.
On valuation, Brad Barber and Ayako Yasuda studied the interim performance of more than 800 private equity funds raised between 1993 and 2009 and found that the size and frequency of NAV markdowns increase in the period after a fundraising (Barber and Yasuda, 2014 working paper). Their reading is that managers are reluctant to mark down while raising a successor fund. In the published 2017 version, fundraising responded most to interim returns backed by realised exits, and the peaks that faded after fundraising belonged to low-reputation managers with few realisations (Barber and Yasuda, Journal of Financial Economics, 2017). The working paper also found that post-fundraising performance declined for buyout funds only among small, young and low-reputation firms, so the pattern is not a finding about every manager. It does mean RVPI is the part of TVPI that is best read with the fundraising calendar in view, and that returns backed by realised exits are the ones the fundraising market rewards most.
On distributions, DPI counts what the fund has paid out, whatever funded it. Invest Europe’s glossary counts cash and the value of any distribution in specie, and leaves out amounts returned from temporary, bridging or aborted investments and amounts later clawed back. Invest Europe defines NAV as net of liabilities, so a distribution paid from a loan against the portfolio lifts DPI and lowers RVPI by the same amount at that date. In Fund A at year 7, a $15 million loan-funded distribution takes DPI from 0.30x to 0.45x and RVPI from 1.30x to 1.15x, while TVPI stays at 1.60x. The loan’s interest then reduces later returns. ILPA, the institutional LP association, recommended in July 2024 that GPs seek the approval of the limited partner advisory committee (LPAC) before using a NAV-based facility for distributions (ILPA, July 2024).
An LP can test for this by asking the GP for DPI both with and without distributions funded by borrowing, and for the amount that can be recalled. Only the GP holds that split, so the request belongs in the due diligence questionnaire or a side letter. In the loan example the adjusted DPI is 0.30x and the reported one is 0.45x.
Other routes to cash are legitimate and worth identifying. Bain reported that GP-led continuation vehicles grew 62% year over year in 2025, and that more than half of the GPs in the 2026 StepStone/Bain Private Equity GP Survey said a primary reason for launching one was generating liquidity to cash out LPs. The question for an LP is whether the price was fair and whether the distribution can be recalled: a recall adds to paid-in capital and cuts DPI, as the recallable distributions page shows, in an example where DPI falls from 0.75x to 0.50x.
Who sets the rules for reporting TVPI and DPI in the US, Europe and elsewhere?
Industry bodies and a global standard that firms opt into by claiming compliance set the definitions of DPI, RVPI and TVPI. The regulatory rules in the table concern how performance is presented, not how these ratios are defined.
| Body or rule | Where | What it covers | Status |
|---|---|---|---|
| GIPS Standards for Firms, 2020 edition (CFA Institute) | Global: adopted by organisations from 54 markets | For composites of portfolios with committed capital, a firm must present the PIC multiple (paid-in capital to committed capital), DPI, RVPI and TVPI (Provision 5.A.4) | Apply to firms that choose to claim compliance; the Handbook says TVPI is also known as MOIC |
| IPEV Valuation Guidelines, 2025 edition | Global | A principles-based framework for determining the fair value of private capital investments, the valuation behind RVPI where a manager follows it | Published 11 December 2025; endorsed by private equity and venture capital associations; not law |
| Invest Europe Investor Reporting Guidelines, 2024 edition | Europe | Defines DPI, RVPI (net of carry) and TVPI | Published 25 January 2024; GPs claiming compliance must adopt it for the quarter or year ending 31 December 2024 |
| ILPA Performance Template | International | Gross and net IRR and TVPI/MOIC, with and without subscription facilities | Recommended for funds commencing operations on or after 1 January 2026; Alston & Bird wrote in February 2025 that it is not mandatory |
| SEC Marketing Rule (Rule 206(4)-1) and staff FAQs | United States | How investment advisers present gross and net performance in advertising | The rule is in the Code of Federal Regulations; the FAQs are staff views, not Commission rules; FAQ page last updated 15 January 2026 |
| SEC Private Fund Adviser Rules | United States | Standardised fund performance reporting | Vacated by the Fifth Circuit in June 2024, per Proskauer (August 2026) |
Gross and net TVPI differ in what has been taken off, and both the SEC rule and the ILPA template address the difference. The Marketing Rule defines gross performance as performance before the deduction of all fees and expenses and net performance as performance after it, and allows an adviser to show gross performance only with net performance at least as prominent (17 CFR 275.206(4)-1, Cornell Law School). Invest Europe’s RVPI is net of carried interest by definition, so a European-format RVPI and a US gross figure are not comparable until the carry is taken off. The GIPS Handbook recommends that a firm present both gross-of-fees and net-of-fees since-inception money-weighted returns; the gap between a gross and a net TVPI is likewise fees, fund expenses and carry, and it is the net figure that reaches the LP. Ask for both, and which one a headline number is.
What should an LP do with the three numbers?
An LP should ask for DPI, RVPI and TVPI at the same date, net of fees and carry, and then check how much of TVPI is cash against the fund’s age. This is general information, not investment advice.
- Ask whether each figure is gross or net, whether “MOIC” means the same ratio as TVPI, and whether the RVPI is net of carried interest, as Invest Europe’s definition requires.
- Compare DPI with funds of the same vintage and strategy at the same age, naming the provider and the date.
- Work out the share of TVPI that is DPI, after taking out any distribution funded by borrowing or open to recall, and re-run TVPI with the unsold value marked down. A 30% mark-down was the example’s input, not a norm. Read the share beside DPI itself, because a write-down raises the share without paying anything: Fund A’s 19% at year 7 would be 75% if its $130 million NAV were written down to $10 million.
- Ask whether any distribution came from a continuation vehicle, and how its price was set.
- From about year 10 for a buyout fund that invests over four or five years (the rule above, built on Bain’s buyout holding period), treat a low DPI as a question for the manager: which investments turn the remaining NAV into cash, and when.
For how these figures appear on a statement, see capital account statements, and for how carry timing changes what reaches an LP, see European and American waterfalls.
Sources
- Invest Europe, Glossary, Investor Reporting Guidelines. Accessed 1 November 2026.
- Invest Europe, Investor Reporting Guidelines, 25 January 2024. Accessed 1 November 2026.
- Bain & Company, Global Private Equity Report 2026, 22 February 2026. Accessed 1 November 2026.
- Bain & Company, Press release on the Global Private Equity Report 2026, 23 February 2026. Accessed 1 November 2026.
- Barber and Yasuda, Journal of Financial Economics 124(1), 172 to 194, Interim Fund Performance and Fundraising in Private Equity, 2017. Accessed 1 November 2026.
- Institutional Limited Partners Association (ILPA), NAV-based facilities guidance: news item, 25 July 2024. Accessed 1 November 2026.
- US Securities and Exchange Commission, Division of Investment Management, Marketing Compliance Frequently Asked Questions, 15 January 2026. Accessed 1 November 2026.
- Alston & Bird, ILPA Updated Reporting and Performance Templates, 24 February 2025. Accessed 1 November 2026.
- Institutional Limited Partners Association (ILPA), ILPA Performance Template. Accessed 1 November 2026.
- Proskauer Rose, ILPA's Reporting Templates After PFAR: What Fund Sponsors Should Know, 17 August 2026. Accessed 1 November 2026.
- Barber and Yasuda, Wharton Working Paper 14-18, Interim Fund Performance and Fundraising in Private Equity (working paper), 2014. Accessed 1 November 2026.
- CFA Institute (GIPS Standards), GIPS Standards Handbook for Firms. Accessed 1 November 2026.
- CFA Institute (GIPS Standards), The GIPS standards. Accessed 1 November 2026.
- IPEV (International Private Equity and Venture Capital Valuation Board), Valuation Guidelines, 11 December 2025. Accessed 1 November 2026.
- IPEV, IPEV Board publishes updated International Private Equity and Venture Capital Valuation Guidelines 2025, 11 December 2025. Accessed 1 November 2026.
- Legal Information Institute, Cornell Law School, 17 CFR 275.206(4)-1, Investment adviser marketing. Accessed 1 November 2026.
- CNBC, The IPO drought is driving private equity-owned companies to the debt market to pay investors, 16 October 2026. Accessed 2 November 2026.