DPI (distributions to paid-in capital): formula, calculator and VC benchmarks
What is DPI?
DPI (distributions to paid-in capital) is the cash a fund has paid back to its investors divided by the capital they have paid in: DPI = cumulative distributions ÷ paid-in capital. At 1x, investors have their money back; anything above is profit in hand. Unlike TVPI, DPI counts no unrealized value.
Key facts
- The GIPS standards define the break-even point: "Once the DPI is greater than one, the fund has broken even." (GIPS, 2005)
- The median 2017-vintage US VC fund on Carta has a DPI of 0.31x after about nine years, and fewer than 20% of 2017 and 2018 funds have reached 1x (Q1 2026). (Carta, Q1 2026)
- DPI is "critical" or "most critical" for 54% of LPs, tied with MOIC behind IRR, in McKinsey's 2026 survey. (McKinsey, 2026)
- The 2021 vintage's average DPI of 0.05x is "the lowest five-year DPI multiple this century", according to PitchBook. (PitchBook, Aug 2026)
- Across 71 funds from the 2005–08 vintages, year-5 DPI correlated just 0.22 with later DPI, VenCap's David Clark found. (Clark, 2024)
DPI calculator
Enter what LPs have paid in, what the fund has paid back and what its remaining holdings are worth. The calculator shows DPI next to TVPI and RVPI, and places it against 2017-vintage US VC funds, the most mature vintage in Carta's data.
Tracking DPI across funds? CX Cash is being built to compute DPI, TVPI and IRR per fund from your capital calls and distributions, and to show what is still unrealized.
Get started free (opens in a new tab) See fund metrics in CX Cash
How to calculate DPI
DPI = Cumulative distributions ÷ Paid-in capital
In words: add up everything the fund has paid out to its investors and divide by everything they have paid in.
- Cumulative distributions: Everything the fund has paid to its LPs since inception, after the manager's fees and carried interest.
- Paid-in capital: Everything LPs have contributed so far, management fees included. Not the commitment, which may not all be called yet.
Worked example
The same illustrative $50M fund as on our TVPI page, in its seventh year. LPs have paid in 90% of their commitments, fees included. The results below are computed by the calculator's tested code.
| Input | Value |
|---|---|
| Paid-in capital (90% of $50M called) | $45,000,000 |
| Distributions to LPs so far | $18,000,000 |
| Net asset value of the remaining holdings | $54,000,000 |
DPI: 0.40x TVPI: 1.60xRVPI: 1.20xShare of value already in cash: 25.0%
TVPI = (18,000,000 + 54,000,000) ÷ 45,000,000 = 1.60x = DPI 0.40x + RVPI 1.20x
To reach 1x DPI, the fund has to pay out another $27M, half of its $54M NAV. If it sells that half at its current marks, DPI reaches 1.00x while TVPI stays at 1.60x: selling turns paper into cash without adding value.
In a spreadsheet
- DPI, with distributions in B2 and paid-in capital in B4: =B2/B4
- Distributions still needed to reach 1x: =MAX(0, B4-B2)
- Share of total value already returned, with NAV in B3: =B2/(B2+B3)
How founders and investors read DPI
If you're the founder
- Your exit is your investors' DPI. Until shares are sold, in an acquisition, an IPO or a secondary sale, your company's value shows up only in their TVPI.
- Investors with little DPI find their next fund harder to raise. That is why a VC may want to sell part of a stake in a later round: Hunter Walk described one such sale that "would get them to 1.0 DPI (in combination with some earlier distributions) in their first fund and into the carry" (Hunter Walk, 2024).
- A partial sale need not signal doubt. LPs judge funds on cash returned, and a secondary in a strong round is one of the few ways a young fund can show any.
Questions you'll be asked
- What paths to liquidity do you see in the next three to five years?
- Would you allow early investors to sell part of their stake in your next round?
- Have you had acquisition interest, and at what price?
- Do the preferences on your cap table leave early investors a clean exit?
If you're the investor or LP
- DPI is the one fund multiple that can't be marked up. Read it with the fund's age: after five years, 59% of 2017-vintage funds on Carta had returned any capital, against 39% of 2019 funds (Carta, 2024).
- Ask where the cash came from. ILPA warns that distributions funded by a NAV loan can be a way "to improve DPI to attract LPs to commit to their next fund" (ILPA, 2024).
- Early DPI is a weak predictor on its own. Sapphire Partners has seen "multiple 3x DPI funds that had zero or close to zero DPI by year 5" (Beezer Clarkson, 2024).
Questions to ask a GP
- How much of the DPI came from exits, secondary sales or NAV-loan proceeds?
- Are any distributions recallable?
- What DPI do you expect by year 8, and which companies will drive it?
- How does the DPI compare with funds of the same vintage?
- How much of the remaining value could you sell in the next two years?
What is a good DPI?
DPI depends heavily on age: most VC funds return little cash in their first five years. Compare a fund only with its own vintage. The figures below are after fees and carry.
Show the numbers
| Vintage | Median | Top quartile |
|---|---|---|
| 2017 | 0.31x | 0.60x |
| 2018 | 0.15x | 0.42x |
| 2019 | 0.02x | 0.16x |
| 2020 | 0.01x | 0.13x |
| 2021 | 0.00x | 0.02x |
| 2022 | 0.00x | 0.00x |
| 2023 | 0.00x | 0.00x |
| 2024 | 0.00x | 0.00x |
Benchmarks
| Segment | Measure | Typical | Top quartile | As of | Sample | Source |
|---|---|---|---|---|---|---|
| US VC funds (Carta), DPI · The data behind the chart above. Carta-administered US funds, most of them under $100M. | ||||||
| 2017 vintage | DPI | 0.31x | 0.60x | Q1 2026 | 131 funds | Carta, Q1 2026 |
| 2018 vintage | DPI | 0.15x | 0.42x | Q1 2026 | 191 funds | Carta, Q1 2026 |
| 2019 vintage | DPI | 0.02x | 0.16x | Q1 2026 | 229 funds | Carta, Q1 2026 |
| 2020 vintage | DPI | 0.01x | 0.13x | Q1 2026 | 253 funds | Carta, Q1 2026 |
| 2021 vintage | DPI | 0.00x | 0.02x | Q1 2026 | 545 funds | Carta, Q1 2026 |
| 2022 vintage | DPI | 0.00x | 0.00x | Q1 2026 | 526 funds | Carta, Q1 2026 |
| Emerging-manager funds (AngelList), DPI · A different universe from Carta, so don't compare across the two blocks. | ||||||
| 2017 vintage | DPI | 0.34x | 0.85x | 1 Jul 2026 | Not disclosed | AngelList, Jul 2026 |
| 2018 vintage | DPI | 0.38x | 0.70x | 1 Jul 2026 | Not disclosed | AngelList, Jul 2026 |
| 2019 vintage | DPI | 0.15x | 0.53x | 1 Jul 2026 | Not disclosed | AngelList, Jul 2026 |
| 2021 vintage | DPI | 0.01x | 0.06x | 1 Jul 2026 | Not disclosed | AngelList, Jul 2026 |
| How long until LPs get their money back | ||||||
| VC funds in Sapphire Partners' database | Time to 1x DPI | Year 8 on average | — | Sep 2024 stale | Funds that have returned capital | Clarkson, 2024 |
| Private equity funds, 1981–93 | Time to return invested capital | A little under 7 years | — | 2003 stale | 73 funds of one large LP | Ljungqvist & Richardson, 2003 |
| 2017-vintage US VC funds | Share with any DPI after 5 years | 59% | — | Oct 2024 stale | Funds on Carta | Carta, Oct 2024 |
| 2019-vintage US VC funds | Share with any DPI after 5 years | 39% | — | Oct 2024 stale | Funds on Carta | Carta, Oct 2024 |
Few funds reach 1x. On Carta, fewer than 20% of 2017 and 2018 funds had a DPI of 1x or more by Q1 2026; the top decile starts at 1.18x and 1.12x.
Late cash is normal. Sapphire Partners finds funds reach 1x DPI by year 8 on average, and Ljungqvist and Richardson found the average fund in their sample took a little under seven years to return invested capital.
Compiled Oct 2, 2026. Next review: December 2026, after Carta's Q3 2026 release. Left out: Cambridge Associates (Its licence bars reuse of its benchmark figures without written permission); PitchBook benchmarks (Available to PitchBook clients only; we cite only figures PitchBook published in articles). Download these benchmarks as CSV.
DPI in the wild: four 2018 funds, eight years in
Four CalPERS private equity commitments from the 2018 vintage, as CalPERS reported them as of 31 March 2026. The cash-back column is DPI, our arithmetic from CalPERS's figures. (CalPERS, Mar 2026)
| Fund (vintage) | Cash in | Cash out | Cash out + remaining value | Net IRR | Multiple | Cash back (DPI) |
|---|---|---|---|---|---|---|
| Carlyle Partners VII, L.P. (2018) | $608.0M | $657.0M | $915.1M | 12.8% | 1.5x | 1.08x |
| Onex Partners V-B LP (2018) | $572.6M | $411.4M | $836.7M | 10.1% | 1.5x | 0.72x |
| The Rise Fund (A), L.P. (2018) | $84.3M | $53.2M | $122.7M | 8.9% | 1.5x | 0.63x |
| Insight Venture Partners X, L.P. (2018) | $261.8M | $308.0M | $718.7M | 20.3% | 2.7x | 1.18x |
The first three share a 1.5x multiple and a vintage. Carlyle Partners VII has paid back 1.08 times CalPERS's money, so CalPERS is in profit in cash; Onex V-B has paid back 0.72 times and The Rise Fund 0.63 times. Insight Venture Partners X has the highest multiple, 2.7x, yet has paid back 1.18 times: more than half of its value is still to be realized.
Cash in and cash out are CalPERS's "actual cash flows", so the cash-back column is not an estimate. The remaining value is the managers' own, and CalPERS notes "there is generally a 2-quarter delay in performance reporting."
DPI vs TVPI, RVPI, IRR, MOIC and PME
| Metric | Answers | Ignores | Use it when | It misleads when |
|---|---|---|---|---|
| DPI | Cash returned per dollar LPs paid in | Value still held, and time | Judging a mature fund, or what LPs can spend | The fund is young, or a NAV loan funded the payouts |
| TVPI | Cash returned plus value still held, per dollar paid in | Time, and how much is still paper | Tracking a fund against its vintage | Marks are stale or optimistic |
| RVPI | Value still held per dollar paid in | Cash already returned | Sizing what is left to realize | Read without DPI beside it |
| IRR | Annual rate of return, timing included | How much money was made | Comparing speed, or testing a required return | Exits come early or a credit line delays calls |
| MOIC | Value per dollar invested in deals, usually gross | Fees, carry and time | Judging a deal or a portfolio | Set against a fund's net DPI |
| PME | Return against a public index over the same dates | Absolute size | Asking whether illiquidity paid off | The index doesn't fit the strategy |
DPI + RVPI = TVPI. As a fund sells its holdings, value moves from RVPI to DPI; when the last one is sold, RVPI is zero and DPI equals TVPI.
Common mistakes: how DPI gets flattered
- Judging a young fund by it. Cash comes late in venture. VenCap's David Clark found the correlation between year-5 DPI and later DPI "was just 0.22" across 71 funds (2024).
- Not asking where the cash came from. Distributions funded by a NAV loan or a dividend recap raise DPI without a sale. ILPA asks LPs to watch for exactly this (ILPA, 2024).
- Selling too early to show DPI. Pressure for cash can push managers to sell winners before they peak. Samir Kaji: "selling prematurely can easily create an adverse impact on long term returns" (LinkedIn, Oct 2024).
- Reading DPI without RVPI. A 1.2x DPI with nothing left to sell is a finished result; a 1.2x DPI with 1.5x still held is a fund mid-way. Look at both before ranking funds.
- Setting a fund's net DPI against a deal's gross realized multiple. A deal's realized multiple is usually before fees and carry; a fund's DPI is after them. The gap between the two is what LPs pay the manager.
- Forgetting that more calls lower it. DPI never falls because of a write-down, but it does fall when the fund calls more capital, because paid-in capital grows (our arithmetic).
Where DPI came from and how it's regarded today
| When | What happened | Source |
|---|---|---|
| Sep 2026 | David Zhou: "DPI (distributions to paid-in capital) is a harsh judge", while "Mark-ups become the mild judge." | Zhou, 2026 |
| Aug 2026 | PitchBook finds the 2021 vintage at the lowest year-5 DPI of any vintage since at least 1997. | PitchBook, Aug 2026 |
| Jul 2026 | Venture LP stakes trade at 79% of NAV in the secondary market, the price of turning paper into DPI early. | Jefferies, Jul 2026 |
| Jun 2026 | McKinsey's LP survey ranks DPI joint second with MOIC, behind IRR. | McKinsey, 2026 |
| Mar 2025 | Carta finds that half of all 2018-vintage funds have not yet distributed any capital. | Carta, Mar 2025 |
| Oct 2024 | Mercury's Aziz Gilani calls DPI "the metric that rules them all". | Carta, Oct 2024 |
| Sep 2024 | Sapphire Partners finds funds reach 1x DPI by year 8 on average. | Clarkson, 2024 |
| Aug 2024 | VenCap's David Clark finds year-5 DPI a weak predictor of later DPI (correlation 0.22). | Clark, 2024 |
| Jul 2024 | ILPA's guidance on NAV loans warns that loan-funded distributions flatter DPI. | ILPA, 2024 |
| Jul 2023 | Bain: "For cash-strapped LPs, DPI (distributed to paid-in capital) is becoming the new IRR (internal rate of return)." | Bain, 2023 |
| 2020 | The 2020 GIPS standards name DPI the realization multiple, required for any pooled fund with committed capital. | GIPS, 2020 |
| Jul 2006 | Howard Marks's memo "You Can't Eat IRR" puts cash returned ahead of reported returns. | Oaktree, 2006 |
| 2005–06 | The GIPS private equity provisions require DPI for each year presented and define DPI above one as break-even. | GIPS, 2005 |
| 2003 | Ljungqvist and Richardson find the average fund takes a little under seven years to return invested capital. | Ljungqvist & Richardson, 2003 |
How it's regarded today
- Cash is scarce. "Since 2022, net cash flow to LPs has been negative $202 billion, even as market value and AUM have continued to increase" (PitchBook, Aug 2026).
- Ranked just behind IRR. McKinsey's 2026 survey: "DPI is now considered “critical” or “most critical” by 54 percent of LPs, tied with MOIC as the second-most-important performance metric" (McKinsey, 2026).
- Paper moves faster than cash. AngelList's mid-2026 refresh: "DPI barely moved across the board", even as median TVPI rose for every vintage from 2019 on (AngelList, Sep 2026).
- Secondaries as a DPI tool. Selling LP stakes early brings DPI forward at a discount: venture stakes traded at 79% of NAV in the first half of 2026, against about 91% for buyout (Jefferies, Jul 2026).
- The Cash Flow, Return and Risk Characteristics of Private Equity (Ljungqvist & Richardson, 2003). The average fund in one large LP's portfolio took a little under seven years to return invested capital.
- Post on year-5 DPI as a predictor (thread) (Clark, 2024). Across 71 funds from 2005–08, year-5 DPI correlated just 0.22 with later DPI.
- Why 2021 vintage funds shouldn't panic yet (PitchBook, Aug 2026). Year-5 DPI isn't a reliable predictor of future distributions; TVPI predicts better at that point.
- Interpretive Guidance for Private Equity (GIPS, 2005). The standard that made DPI a required disclosure and defined DPI above one as break-even.
- NAV-Based Facilities: Guidance for Limited Partners and General Partners (ILPA, 2024). How NAV loans can flatter DPI, and why LPs end up computing a "synthetic" DPI.
What experts say about DPI
“But from my perspective, DPI is the metric that rules them all.”
“We live in a world where DPI (distributions to paid-in capital) is a harsh judge.”
“We leveraged our database of funds that have returned capital, and on average funds get to 1x DPI by year 8.”
“A high internal rate of return does not in and of itself put money in one's pocket.”
Where they disagree
An LP's data and a VC's priorities on how much early DPI should count.
“So we quickly looked at our data to see if DPI in year 5 is a good predictor of final performance. We analysed 71 funds from vintage years 2005-2008. The correlation between year 5 DPI and current DPI was just 0.22.”
“I tell my LPs all the time: We are fully focused on DPI … Because until you realize that liquidity, we have failed.”
DPI FAQ
What is DPI in private equity and venture capital?
DPI stands for distributions to paid-in capital. It is the cash a fund has paid back to its investors divided by the capital they have paid in. A DPI of 0.5x means investors have had half their money back; at 1x they are whole, and every distribution after that is profit.
How do you calculate DPI?
Add up all distributions the fund has made to its LPs and divide by the capital they have paid in, fees included: DPI = cumulative distributions ÷ paid-in capital. A fund that has called $45M and paid back $18M has a DPI of 0.40x.
What is a good DPI?
It depends on the fund's age. After about nine years, the median 2017-vintage US VC fund on Carta has a DPI of 0.31x and the top quartile starts at 0.60x (Q1 2026). Reaching 1x puts a fund ahead of more than 80% of its 2017 and 2018 peers.
What is the difference between DPI and TVPI?
DPI counts only the cash returned; TVPI adds the value of what the fund still holds. TVPI = DPI + RVPI, so the gap between the two is unrealized value. A young fund can have a high TVPI and a DPI of zero.
When do VC funds reach 1x DPI?
Late. Sapphire Partners finds funds that return capital reach 1x by year 8 on average, and many funds take longer or never get there: fewer than 20% of 2017 and 2018 US VC funds on Carta had reached 1x by early 2026.
Is DPI net of fees?
Yes, for a fund. Distributions reach LPs after management fees, expenses and the manager's carried interest, and paid-in capital includes the fees LPs paid. A deal's realized multiple, by contrast, is usually gross.
Can DPI go down?
Not because of a write-down, since cash already paid stays paid. It falls when the fund calls more capital, because paid-in capital grows, or when distributions are recalled, which some fund agreements allow.
What does DPI stand for in private equity?
Distributions to paid-in capital, sometimes written "distributed to paid-in capital". The GIPS standards also call it the realization multiple.
Related metrics and guides
Sources
Every link was opened and checked. Archived copies guard against links that move or disappear.
Academic
- The Cash Flow, Return and Risk Characteristics of Private Equity. Ljungqvist & Richardson, NBER Working Paper 9454, 2003. Free NBER working paper. Archived copy
Primary data
- VC Fund Performance: Q1 2026. Carta (Peter Walker, Kevin Dowd), 4 Jun 2026. Vendor data from Carta-administered funds. Archived copy
- For venture fund LPs, DPI is ‘the metric that rules them all’. Carta (Kevin Dowd), 10 Oct 2024. Vendor data from Carta-administered funds. Archived copy
- VC Fund Performance: 2024. Carta (Peter Walker, Michael Young, Kevin Dowd), 24 Mar 2025. Vendor data from Carta-administered funds. Archived copy
- Fund Benchmarks: A Mid-Year 2026 Refresh. AngelList (Abe Othman), 4 Sep 2026. Vendor data; full report gated. Archived copy
- Private Equity Program (PEP) Fund Performance Review. CalPERS, As of 31 Mar 2026. Archived copy
- Global Secondary Market Review. Jefferies, Jul 2026. Survey by a secondaries adviser; no archived copy found.
Standard
- Interpretive Guidance for Private Equity. CFA Institute (GIPS), Effective 1 Jan 2005. Archived copy
- Global Investment Performance Standards (GIPS) for Firms. CFA Institute (GIPS), Effective 1 Jan 2020. Archived copy
- NAV-Based Facilities: Guidance for Limited Partners and General Partners. ILPA, Jul 2024. Guidance. Archived copy
Consultancy
- Global Private Markets Report 2026. McKinsey & Company, Jun 2026. Archived copy
- Stuck in Place: Private Equity Midyear Report 2023. Bain & Company, 17 Jul 2023. Archived copy
Practitioner
- Post on year-5 DPI as a predictor (thread). David Clark (X), 19 Aug 2024. X needs a login; read from the archived copy. Archived copy
- Post on early DPI and Carta's fund data. Beezer Clarkson (LinkedIn), 11 Sep 2024. LinkedIn post; practitioner data. Archived copy
- Post on illiquidity and GP liquidity strategy. Samir Kaji (LinkedIn), 22 Oct 2024. Practitioner opinion. Archived copy
- DPI is a Harsh Judge. David Zhou (Cup of Zhou), 18 Sep 2026. Newsletter opinion; no archived copy found.
- What Do VC Returns Actually Look Like? Here's a Screenshot From My AngelList Account. Hunter Walk (blog), 14 Aug 2024. Practitioner opinion. Archived copy
- You Can't Eat IRR. Howard Marks (Oaktree memo), 12 Jul 2006. Archived copy
News
- Venture capital's current recovery is all IRR, no DPI. PitchBook (Kyle Stanford), 4 Aug 2026. Archived copy
- Why 2021 vintage funds shouldn't panic yet. PitchBook (Kyle Stanford), 7 Aug 2026. Summary of PitchBook research; the full note is for clients. Archived copy
Changes to this page
- · Major · Page published. Benchmarks as of Q1 2026 (Carta) and 1 July 2026 (AngelList), compiled 2 October 2026.
Cite this page
Dominique Bouillet, "DPI (distributions to paid-in capital): formula, calculator and VC benchmarks", CX Cash, updated Oct 11, 2026, https://cxcash.com/metrics/dpi
CX Cash builds software for founders and investors. This page is education, not investment advice. Third-party figures link to their source, and our own arithmetic and illustrative examples are labelled as such.