TVPI Explained: What Total Value to Paid-In Capital Tells You About a Private Fund

TVPI Explained: What Total Value to Paid-In Capital Tells You About a Private Fund

TL​DR: CalPERS's disclosed fund data shows private equity TVPI ranging from 0.3x (near-total loss) to 4.2x within the same portfolio. A fund sitting at 1.8x TVPI with only 0.3x in actual distributions has not proved anything. It has made a promise.

According to the CalPERS Private Equity Program Fund Performance Review (as of September 30, 2025), California's pension program has returned a net investment multiple of 1.5x and an 11.3% net IRR since inception. The fund-level breakdown is more instructive. Hellman and Friedman Capital Partners VII returned 3.4x on CalPERS's capital with a 24.6% net IRR. Clearlake Capital Partners III returned 2.9x at a 40.7% net IRR. CalPERS's own Clean Energy and Technology Fund from the same era produced just 0.3x, a near-total loss at -18.5% net IRR. That is the reality of private equity TVPI. When you understand what the number actually measures, the spread between those outcomes stops being surprising.

What TVPI Actually Measures and What It Does Not

TVPI stands for Total Value to Paid-In Capital. The formula is straightforward: add every dollar distributed back to investors (called DPI, or Distributions to Paid-In Capital) to the current estimated value of everything still inside the fund (called RVPI, or Residual Value to Paid-In Capital), then divide by every dollar investors have actually sent to the fund. TVPI = DPI + RVPI.

A 2.0x TVPI means the fund has returned or is currently estimated to be worth twice what you put in. A 1.0x means you are breaking even. Anything below 1.0x means you have lost principal. The CalPERS Clean Energy and Technology Fund's 0.3x TVPI translates directly: investors got back about thirty cents for every dollar committed.

The "net" label in net TVPI is critical. Gross TVPI is calculated before management fees and carried interest (the GP's profit share, typically 20% of gains). Net TVPI is what you as a limited partner (LP) actually keep. Always confirm which figure you are looking at. A fund advertising 2.5x gross TVPI may deliver 1.8x net to LPs after fees.

Here is what TVPI does not tell you: it says nothing about time. A 2.0x net TVPI delivered in five years is an entirely different outcome than the same 2.0x returned over fifteen years. The time dimension requires IRR, which stands for internal rate of return and accounts for the pace of both capital calls and distributions. Insight Venture Partners IX shows 4.0x TVPI alongside a 23.1% net IRR in CalPERS's data. The TVPI shows magnitude. The IRR shows efficiency. You need both numbers to evaluate a fund.

TVPI, DPI, and IRR: How the Three Metrics Work Together

I use this table with every LP I work with. It stops the confusion fast.

Metric	Full Name	What It Measures	Key Limitation
TVPI	Total Value to Paid-In Capital	Total return multiple (realized plus unrealized) relative to capital invested	Ignores time. Includes GP-estimated unrealized value.
DPI	Distributions to Paid-In Capital	Cash actually returned to LPs relative to capital invested	Can be low even in a strong fund that has not yet exited positions
RVPI	Residual Value to Paid-In Capital	Estimated NAV of remaining unrealized positions relative to capital invested	Self-reported by the GP. Not market-verified until exit.
IRR	Internal Rate of Return	Annualized return accounting for timing of cash flows	Can be distorted by subscription lines. Not directly comparable across fund sizes.
PME	Public Market Equivalent	How the fund performed versus investing the same cash flows in a public index	Benchmark-dependent. Less commonly disclosed by GPs.

The relationship between TVPI and DPI is the most important split in private fund analysis right now. A fund with 1.8x TVPI and 1.6x DPI has largely proved its value. Distributions are real. Francisco Partners III illustrates this: 3.4x TVPI with a DPI of 3.16x. Nearly all of that value has been returned to CalPERS in cash. Clearlake Capital Partners III shows the same pattern at 2.9x TVPI and 2.86x DPI. Almost fully distributed.

Now take Tiger Global Private Investment Partners XV, also in CalPERS's data. TVPI: 0.8x. Net IRR: -5.9%. No narrative changes those numbers. Capital has been destroyed. That is the other end of the spectrum, existing inside the same institutional portfolio as the funds returning 3x and 4x.

What Good TVPI Looks Like: Real Benchmarks From Cambridge Associates and Preqin

Context is everything here. A 1.5x TVPI for a 2021-vintage buyout fund in 2025 is roughly at the median. The same 1.5x for a 2012-vintage fund still mostly unrealized after thirteen years is a serious red flag.

Cambridge Associates and Preqin, which together cover more than 140,000 private capital benchmarks across 50-plus years of data, report PE buyout median TVPI by vintage year as of Q4 2024. The 2015 vintage sits at 1.91x. The 2016 at 1.85x. The 2017 at 1.78x. The 2018 at 1.72x. The 2019 at 1.80x. The 2020 at 1.58x. The 2021 at 1.34x. These are medians. Half of funds in each vintage sit below these numbers.

Top-quartile PE buyout funds from 2015 through 2019 vintages cleared between 2.35x and 2.61x TVPI. That is the performance bar you are trying to cross when you write a check to a top-tier manager. The Cambridge Associates Private Investment Benchmarks portal also shows the US Private Equity Index generating a 10-year annualized return of 15.25% through Q3 2024, outperforming the Russell 3000 by 219 basis points on a PME basis.

Venture capital shows wider dispersion. Carta's platform data shows the 2018 vintage VC median TVPI fell from 1.55x to 1.37x over four consecutive quarters through Q1 2024. Only 12% of 2019-vintage VC funds had reached 2.0x TVPI as of 2024. Top-quartile VC from 2015 through 2019 clears 3.0x. Top-decile funds from strong vintage years hit 3.5x to 5.0x or higher. The downside is also sharper. The worst VC vintages on record produced median TVPI below 1.0x, which is permanent capital impairment for the average LP in those cohorts.

Private credit operates under different math. Closed-end direct lending funds typically show TVPI in the 1.1x to 1.35x range for fully realized vehicles because returns come from coupon income, not capital appreciation. Comparing a private credit fund's 1.2x TVPI to a buyout fund's 1.8x TVPI and concluding the buyout fund is better is an error. The Cliffwater Direct Lending Index, which covers approximately $549 billion in US middle market loans, uses time-weighted returns rather than TVPI because direct lending vehicles are structured differently from closed-end PE funds. The CDLI has returned 9.5% annualized over twenty years. That is a competitive result. It just does not express well in TVPI terms.

What TVPI Hides: Three Structural Problems Every LP Should Know

TVPI has three structural blind spots that are especially dangerous in the current market.

The NAV subjectivity problem. The RVPI component of TVPI is based on the GP's own fair value estimate under ASC 820 accounting guidelines. The GP marks the portfolio. The GP reports the TVPI. Academic research, including work by Brown, Gredil, and Kaplan published in 2019, has documented that GP-reported NAVs are inflated by approximately 10% to 15% during active fundraising windows. Secondary market pricing provides an independent check. In 2024, LP-led secondary transactions cleared at an average 9% to 10% discount to reported NAV. That discount is the market's verdict on what GP-reported RVPI is actually worth on a liquidation basis. If your fund reports 1.8x TVPI but carries minimal DPI and the secondary market would price it at a 10% haircut, your real liquidation TVPI is closer to 1.65x to 1.70x.

The subscription line facility problem. Many GPs use revolving credit lines, called subscription lines or capital call facilities, to fund investments before drawing capital from LPs. This delays capital calls, reducing the paid-in denominator and inflating TVPI by 10% to 20% versus the unfacilitated figure. This effect is most pronounced early in a fund's life. The trap here is evaluating a Year 3 TVPI without knowing how much of the apparent outperformance comes from timing the denominator rather than creating actual value.

The GP-led secondary problem. The secondary market hit a record $162 billion in transaction volume in 2024. A growing portion of that was GP-led, meaning the general partner moved assets from an older fund into a new continuation vehicle and set the transfer price in the process. When the GP prices the transfer, the TVPI of the continuation vehicle is anchored to a price the GP itself determined. That is not independent price discovery. A 2.0x TVPI on a continuation vehicle seeded at a GP-controlled price is a weaker signal than a 2.0x TVPI where every position has been sold to an independent third party.

The DPI drought makes all three problems worse. According to data summarizing Bain's 2025 Global Private Equity Report, distributions as a share of NAV fell to 11% in 2024, the lowest in over a decade. Approximately 28,000 PE-backed companies remained unsold globally as of late 2024, up from 19,000 in 2019. That overhang represents roughly $3.2 trillion in unrealized value that has not been tested at the exit window. 74% of institutional LPs now rank DPI as their primary re-up criterion, up from 52% five years ago per the ILPA's 2024 survey. Cash returned is verified. Everything else is a mark on a spreadsheet.

How to Use TVPI When Evaluating a Fund

When I review a fund, I run four checks before taking any TVPI figure at face value.

Check the DPI-to-TVPI split first. A fund with 1.8x TVPI and 0.3x DPI is a very different proposition from a fund with 1.8x TVPI and 1.6x DPI. In the first case, 83% of the reported value is still inside the fund, marked by the GP, untested by any buyer. In the second case, the majority of returns have already cleared the exit. Always ask: what percentage of reported TVPI has actually been realized in cash?

Adjust for vintage year maturity. A 2021-vintage fund in 2025 is only four years old and deep in the J-curve, which is the early period when capital has been deployed but exits have not yet occurred. The Cambridge Associates data shows 2021 PE buyout median TVPI at 1.34x as of Q4 2024. That is not underperformance. It is normal for the age of the fund. Compare TVPI only against same-vintage benchmarks from Cambridge Associates or the Preqin Q4 2024 private markets performance report. A 2015-vintage fund with 1.34x TVPI in 2025 is ten years in and well below median. That is a fundamentally different picture.

Pair every TVPI with its IRR. The CalPERS data is instructive. Hellman and Friedman Capital Partners VII and Francisco Partners III both show 3.4x TVPI, but Francisco Partners carries a 22.9% net IRR against Hellman and Friedman's 24.6%. Both are exceptional outcomes. The slight IRR difference reflects the timing of cash flows, specifically how quickly capital was called and how quickly distributions came back. If you only saw the 3.4x TVPI figures without the IRR, you would miss that dimension entirely.

Know where the benchmark data comes from. CalPERS, the Washington State Investment Board, and other public pension funds disclose fund-level performance via public records. Preqin's benchmark database is largely built from FOIA requests to those same institutions. Cambridge Associates draws on voluntary reporting from 2,400-plus managers across 9,900-plus funds. None of these databases is free from survivorship bias. Funds that shut down or stop reporting tend to fall out of the dataset, which pulls up median figures for those that remain. The true median for every LP dollar committed to the strategy is lower than what published benchmarks show.

TVPI is not a broken metric. It is an incomplete one. Used correctly, alongside DPI, IRR, vintage-year benchmarks, and an honest assessment of how much RVPI has been independently verified, TVPI gives you a fast and comparable measure of where a fund stands relative to peers. Used in isolation, it flatters managers at the worst possible moment: when exit markets are closed, distributions have stalled, and the score on the board has not been confirmed by anyone except the team keeping score.

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.