DPI Is the Only Private Equity Metric That Actually Matters

TL;DR: Distributions as a share of NAV hit 11% in 2024—a record low. Bain's 2026 report confirms the industry is sitting on $3.2 trillion in unrealized value across 29,000 unsold companies. Yet fund managers still lead with TVPI—Total Value to Paid-In—which counts paper gains alongside actual cash. LPs have stopped listening. Seventy-four percent now rank DPI (Distributions to Paid-In Capital) as their primary criterion for deciding whether to re-up. Cash in the bank beats marks on a spreadsheet. Every time.

TVPI Is Losing Credibility

Here's the disconnect: A fund manager pitches you a 1.8x TVPI. Sounds great. But if the DPI is 0.3x—meaning you've gotten back only 30 cents per dollar invested in actual distributions—the headline number is fantasy. The remaining 1.5x is RVPI: Residual Value, Paid-In Capital. Unrealized. Marked by the GP. Subject to sentiment swings, continued holding periods, and the whims of exit multiples.

The math is simple: TVPI = DPI + RVPI. If TVPI is high and DPI is low, you're betting on a spreadsheet.

This distinction was academic five years ago. Today it's existential. Bain's 2025 report showed that US and Western European buyout funds from the 2018 vintage are sitting at 0.6x DPI—20 percentage points below historical expectations of 0.8x at this stage of the fund lifecycle. Yet those same funds report TVPI in the 1.6x to 1.8x range. Translation: 70 to 80 percent of their reported value has not left the portfolio.

Cambridge Associates put it plainly: "DPI and TVPI multiples provide valuable insight into cash returns and cannot be 'locked in' or managed like the IRR can." IRR is a time-weighted rate that GPs can manipulate with subscription credit facilities, NAV loans, and timing tricks. DPI cannot be gamed. It is cold distributions divided by cold capital paid in.

What DPI Actually Is (And Why It Matters)

DPI = Cumulative Distributions / Paid-In Capital. If you put in $100 million and have received $60 million back in cash, your DPI is 0.6x. When you hit 1.0x, you've recovered your initial investment. That's the fundamental checkpoint. Everything above 1.0x is profit.

Unlike TVPI, which floats with portfolio valuations, DPI is binary. The money either left the fund or it didn't. The ILPA (Institutional Limited Partners Association) Performance Template, released January 2025 and mandatory for funds commencing January 1, 2026, standardizes this calculation—no more GP proprietary formulas, no more room for accounting tricks. This alone is reshaping the PE landscape. For the first time, LPs can compare apples to apples across fund managers.

But here is the hard truth: DPI on its own tells you what you've received. It does not tell you what you should have received by now. That's where the J-curve comes in.

The J-Curve Red Flag: When DPI Falls Behind

In the first few years of a fund's life (years 0-3), DPI will be near zero. That's normal. Fees drag NAV down while portfolio companies are being built. The "valley of tears." Investors expect this.

By year 6-8, a healthy buyout fund should have DPI of 1.0x to 1.5x. Mature funds (2012-2015 vintages) typically show DPI of 1.4x to 1.7x by year 12-13. These are historical baselines. Cambridge Associates benchmarks confirm these ranges across their database of 1,700+ US buyout funds.

When a 2018-vintage fund sits at 0.6x DPI in 2026—six years in—something is wrong. Either the portfolio is genuinely underperforming, or, more likely, the exit environment has seized up. Both outcomes matter to an LP deciding whether to commit fresh capital to the next fund.

CalPERS and the Named-Fund Reality

CalPERS publishes fund-level performance data for every PE holding—a rare window into real institutional LP experience. The divergences are instructive.

Cedar Street Partners (2021 vintage): 0.10x DPI, 2.11x TVPI. That fund has distributed only 10 cents per dollar LP investors sent in. The reported 2.11x? Paper. Ninety-five percent unrealized. No LP should have a restful night holding that position.

By contrast, Hellman & Friedman VII (2011 vintage): 3.30x DPI, 3.38x TVPI. This fund has returned $3.30 in cash for every dollar invested and still has another $0.08 per dollar in remaining value. That's the LP narrative a GP dreams of pitching.

The median for mature (2012-2015 vintage) funds in the CalPERS portfolio shows buyout funds hitting 1.4x to 1.7x DPI by year 10-13. Growth and tech-focused funds often exceed 2.0x DPI. The benchmarks exist. The question is why so many 2018-2020 vintage funds are nowhere close.

The SoftBank Vision Fund Case Study: When TVPI Collapses

SoftBank Vision Fund 1 raised $98.6 billion in 2017. By March 2026, it had distributed $38.7 billion—implying a DPI of roughly 0.39x. Yet SoftBank reports TVPI of 1.25x. The $0.86 per dollar in RVPI represents marks on Didi, WeWork, Oyo, and other portfolio companies that peaked at $1.7x TVPI in 2021 before collapsing.

Vision Fund 2 is worse: TVPI 1.03x, IRR 0.2 percent. Flat for investors. These were the world's largest PE/VC funds. They demonstrate the thesis perfectly: headline TVPI hides a distribution drought. LPs have received back less than 40 percent of capital in one fund and essentially zero real returns in the other.

This is what the PE industry's current $3.2 trillion NAV backlog looks like in practice.

What LPs Are Demanding Now

Four converging pressures are forcing the shift to DPI-first evaluation:

1. Cash flow reality. Pension funds and endowments spend real cash. Paper gains do not fund the pension. A 1.5x TVPI fund that has a 0.4x DPI is not available for spending—it's trapped in the fund waiting for an exit.

2. The NAV loan shadow. ILPA's 2024 guidance on NAV facilities explicitly warns that debt-funded distributions artificially inflate DPI and IRR. Smart LPs now ask: Is your distribution cash from operations, or is it a loan you'll call back when performance stalls? One counts. The other is a mirage.

3. The exit drought. Holding periods have stretched to 6.7 years on average—up from 5.7 years historically. The 2019-2021 vintage wave raised capital at peak multiples into a zero-rate environment. Those companies are now aging in portfolio. DPI progress is stalled. LPs are tired of waiting.

4. Standardized reporting. The ILPA Performance Template eliminates excuse-making. Starting January 1, 2026, every new fund must report gross IRR, net IRR, TVPI, DPI, and RVPI using a standardized formula. Comparison shopping just became possible. Mediocre GPs have nowhere to hide.

The data backs this. ILPA's 2024 survey shows 74 percent of LPs now rank DPI as their primary re-up criterion, up from 52 percent five years earlier. This is not a preference shift. This is a thesis rejection of TVPI as a meaningful metric.

How to Use DPI in Your Due Diligence

If you're evaluating a fund—whether for initial commitment or re-up—ask five questions:

1. What is the DPI relative to fund age? A 2018-vintage fund should be at 0.6x minimum; 0.5x or below is a warning sign. Compare to Cambridge Associates or Preqin benchmarks for that vintage and strategy.

2. What is the DPI/TVPI gap? If DPI is 0.5x and TVPI is 1.8x, the fund is holding $1.3x in unrealized NAV. Ask the GP: When does that portfolio exit? What is the downside case if exit multiples fall from current marks?

3. How much DPI came from operational distributions versus NAV loans? Push back on managers who rely on debt facilities to hit distribution targets. Clean DPI—from actual portfolio exits and dividend recaps—is worth more than borrowed cash.

4. What does the GP's track record show across fund generations? Did Fund III hit 1.5x DPI by year 8? Did Fund IV? If not, why should you trust Fund VI will be different?

5. What is the remaining portfolio's exit runway? A fund at 0.8x DPI with 60 percent of NAV still invested is in the harvest phase. Ask for exit timelines. If the GP says "2-3 years" for a $4 billion residual NAV, that is 2-3 years of capital tied up with no distributions.

Cambridge Associates noted that in H1 2025, PE managers distributed $78.9 billion against $67.6 billion in capital calls—a slight positive after a three-year drought. The distribution faucet is starting to turn on. But distribution activity remains below historical averages. Patience is finite.

The Bottom Line

TVPI is not a lie. It's a leading indicator. Early in a fund's life, TVPI can signal strong portfolio appreciation even before DPI materializes. But the moment a fund reaches year 6 and beyond, TVPI becomes secondary. DPI is the verdict.

The PE industry built a generation of performance reporting around TVPI and IRR because those metrics allowed funds to tell optimistic stories. In 2026, institutional LPs have stopped buying the narrative. They are writing contracts that mandate DPI disclosure. They are walking away from funds that cannot show cash returns. They are asking hard questions about the $3.2 trillion in unrealized NAV sitting in private equity portfolios.

If you are evaluating a fund manager, remember: There is no penalty for asking about DPI. There is a massive penalty for ignoring it.

Internal Links

• How to Evaluate a Private Equity Fund: A Detailed Checklist
• Preferred Returns and Hurdle Rates in Private Equity: What LPs Need to Know
• Carried Interest Explained: How Private Equity Fund Managers Get Paid
• The Secondary Market for Private Equity: How LPs Create Liquidity
• Fund of Funds Explained: A Diversification Strategy for Accredited Investors

External Citations

• CalPERS Private Equity Fund Performance (Named Fund Data)
• Bain Global Private Equity Report 2026
• Cambridge Associates US PE/VC Benchmark Commentary: First Half 2025
• ILPA Performance Template 2025: Standardized Guidance and Granular Methodology
• SoftBank Vision Fund Investor Briefing Q4 FY2025
• ILPA NAV Facilities Guidance 2024
• Cambridge Associates Private Investment Benchmarks

Author Disclosure: Jeff Barnes, MBA has no personal position in any company, fund, or platform named in this article. Angel Investors Network has no current commercial relationship with any party mentioned. AIN provides marketing and education services, not investment advice. Past performance does not guarantee future results. All investments involve risk, including loss of principal.