MOIC Explained: Multiple on Invested Capital and Why the Number You See Is Probably Inflated

MOIC Explained: Multiple on Invested Capital and Why the Number You See Is Probably Inflated

TL;DR: The gross MOIC a GP quotes in a pitch deck is not the net MOIC you receive as an LP. Standard 2-and-20 fee structures cut gross MOIC by 15–30%. A "3.0x fund" can deliver 2.0x–2.2x to your account after management fees and carry — and that's before survivorship bias inflates the benchmark you're comparing it against.

According to Cambridge Associates' Q4 2025 PE Index — drawn from a database covering 2,400+ fund managers, 9,900+ funds, and 93,000+ underlying investments , the global private equity buyout average sits at approximately 1.7x net MOIC, with top-quartile funds clearing a 2.3x threshold. That benchmark, however, conceals a range running from 0.3x (CalPERS' Clean Energy & Technology Fund, -18.5% net IRR) to 4.1x+ among CalPERS' best active partnerships. It also represents net figures after the standard fee structure has already removed a significant slice of gross returns. Before you accept any fund's quoted multiple at face value, you need to understand exactly what MOIC measures, what it hides, and where the number gets manipulated between gross and net.

What MOIC Actually Measures

The formula is simple. MOIC equals total value returned divided by total capital invested. At the deal level: gross MOIC = (realized distributions + unrealized NAV) / capital deployed into that deal, calculated before management fees, fund expenses, and carried interest. At the fund level, the net equivalent is called TVPI , Total Value to Paid-In , which equals (net distributions to LPs + residual NAV after carry) / total LP paid-in capital.

The distinction between the denominator in MOIC (capital deployed into deals) and the denominator in TVPI (all LP paid-in capital, including management fees drawn) creates the first gap. A fund that called $100M total but deployed $85M into companies uses $85M in the MOIC denominator and $100M in the TVPI denominator. The MOIC looks better. The TVPI is what you actually earned.

The second gap sits inside the numerator. MOIC combines two very different things: DPI (Distributions to Paid-In, cash already returned to you) and RVPI (Residual Value to Paid-In, paper NAV still inside the fund). A 2.5x TVPI with 2.2x DPI and 0.3x RVPI is a fundamentally different animal from a 2.5x TVPI with 0.2x DPI and 2.3x RVPI. The first fund has returned most of its capital in cash. The second has returned almost nothing and is sitting on valuations that may or may not survive to exit. Always decompose the number.

MOIC vs. IRR: Why You Need Both, and What Each Hides

MOIC tells you how much you made. IRR tells you how fast you made it. Neither one tells the full story alone.

A 3.0x MOIC over 5 years equals roughly 25% IRR. The same 3.0x MOIC over 12 years equals roughly 9.6% IRR , below the historical S&P 500 return. The multiple is identical. The economic outcome is not. This is where investors who focus exclusively on MOIC get burned on long-dated venture funds.

IRR has its own blind spots. GPs who use subscription credit lines , borrowing against uncalled LP capital to fund early deals before calling LP money , can inflate reported IRR by 25 to 250 basis points without changing MOIC in any meaningful way. The reason: IRR is a time-weighted return, so delaying the start date of capital deployment makes the clock start later and the annualized rate look higher. BlackRock data cited by Canterbury Consulting shows an average IRR boost of 0.5 percentage points from subscription lines, with MOIC reduced by only -0.02x. ILPA's 2017 guidelines recommend GPs report IRR both with and without subscription-line impact. Most don't.

MOIC also ignores deal size. A 50% IRR on a $500,000 angel check is less valuable to your portfolio than a 20% IRR on a $5M fund commitment. Use IRR to compare efficiency across time horizons. Use MOIC to measure total wealth creation. Use both together to make a real decision.

What Good MOIC Looks Like: Benchmarks by Asset Class

Benchmarks vary sharply by strategy and vintage. Here is where the numbers actually sit, based on primary data as of Q4 2025 and H1 2025.

Asset Class	Average / Median Net MOIC	Top-Quartile Net MOIC	Source
PE Buyout	1.7x (global average)	2.3x	Cambridge Associates Q4 2025
VC (2018 vintage)	1.6x median TVPI	3.1x TVPI	Cambridge Associates / Preqin / Carta via Value Add VC H1 2025
VC (2021 vintage)	1.1x median TVPI	1.6x TVPI	Cambridge Associates / Preqin / Carta via Value Add VC H1 2025
Angel Investing	2.6x over 3.5 years (27% IRR)	Power-law driven. outliers reach 10x–100x	Wiltbank-Boeker, Kauffman Foundation, 1,137 exits

Those 2021 vintage VC numbers deserve a second look. The median fund from that year sits at 1.1x TVPI as of H1 2025. That is barely above return of capital, and much of it is still unrealized. You committed in 2021 at peak valuations, paid two years of management fees, and as of today you are roughly flat on paper with most of your money locked up. That is not a hypothetical risk. It is the actual outcome for the median LP in a median 2021 vintage VC fund.

On the other end of the spectrum, AngelList's data on 2017-vintage AI investments shows a 7.2x MOIC as of the report date. That number is real , but it reflects a narrow category at a specific vintage that rode one of the most powerful secular trends in tech history. It is not a benchmark. It is an outlier.

The CalPERS public disclosures give you the clearest real-world picture. Their PE program since inception shows a 1.5x net multiple and 11.3% net IRR as of September 30, 2025. Inside that aggregate: SL SPV-2 (2019 vintage) at 4.1x / 30.4% IRR. Silver Lake Technology Investors IV (2013 vintage) at 3.5x / 27.3% IRR. Hellman &. Friedman Capital Partners VII (2011 vintage) at 3.4x / 24.6% IRR. Francisco Partners III (2011 vintage) at 3.4x / 22.9% IRR. And the CalPERS Clean Energy &. Technology Fund at 0.3x / -18.5% IRR. Even CalPERS , one of the most sophisticated LPs on the planet with a 20-year annualized PE return of 12.3% , holds a fund that destroyed capital. The losers exist. They just rarely show up in the averages.

The Gross vs. Net Gap: How Fees Eat Your Multiple

Here is the math GPs do not put in the pitch deck. You commit $10M to a fund running standard 2-and-20: a 2% annual management fee on committed capital (tapering after the investment period) and 20% carried interest on profits above the hurdle. Over the fund's life, total management fees typically consume roughly 15% of committed capital. That is $1.5M off the top before a single investment is made.

Now assume the fund generates a 3.0x gross MOIC. Your $10M becomes $30M in gross proceeds at the deal level. The fund subtracts management fees already drawn ($1.5M), then takes 20% carry on the remaining profit above your invested capital. The LP net on a clean 3.0x gross in this structure lands between 2.0x and 2.2x. That is a 15–30% reduction from the headline number , standard 2-and-20 cuts gross MOIC by 15–30% in real-world fund math.

GPs are required by SEC rules to disclose net returns to LPs, but marketing materials, conference presentations, and LP update calls routinely lead with gross MOIC. Blackstone's official disclosure format, as defined in their Q4 2024 earnings release, defines MOIC as carrying value "before management fees, expenses, and Performance Revenues" divided by invested capital. They separately define net IRR as inception-to-date after all fees. Both numbers appear in their filings. In a pitch meeting, you will hear the gross one first.

The rule is simple: never compare your net return to a GP's gross benchmark. When a manager says their fund is "top-quartile," ask which quartile , gross or net , and verify against a net-of-fees benchmark from Cambridge Associates or Preqin.

Survivorship Bias: The Zeros That Do Not Get Counted

This is the most dangerous distortion in private markets data, and it compounds everything else.

Commercial databases , Cambridge Associates, Preqin, Burgiss, PitchBook , depend on voluntary GP reporting. Successful funds report. Mediocre funds report selectively. Failed funds, zombie GPs, and written-off portfolios largely do not report. The result: the "average" return figure in any benchmark database is not the average return across all funds. It is the average across funds that survived and chose to disclose.

The quantified impact is stark. Source Scrub's analysis of PE database methodology found that calculating performance using only still-active funds produces an average return of approximately 11%. When you include all observations , closed funds, failed GPs, liquidated portfolios , the average drops to approximately 3%. That is a 73% overstatement from survivorship bias alone. The same effect documented in mutual fund databases boosted reported returns by 1.3 percentage points per year over a 10-year period when defunct funds were excluded.

The same bias runs through angel investing data. The Wiltbank-Boeker study , the largest angel dataset at its time, covering 539 investors, 86 groups, and 1,137 exits , produced the widely cited 2.6x average MOIC in 3.5 years. That study captures group angels only. Solo angels do not report. Angel groups that shut down do not report. The 2.6x figure reflects the subset of organized angel investors who bothered to track and disclose their outcomes. The actual population-level return is almost certainly lower.

What the distribution actually looks like: most angel investments return 0x to 1x. A small fraction return 5x–20x. A tiny fraction return 50x–100x. Those outliers pull the average above 2x while the median sits well below it. The AngelList fund benchmarks confirm this power-law structure across 1,808 pre-Series C investments , a small number of deals drive most portfolio returns. Angel Capital Association data shows that portfolios with a board member present at exit achieved 12.8x MOIC and 43% IRR. without a board member, the same data set shows 0.9x MOIC and -1% IRR. The average masks everything.

You cannot opt out of this bias when reading benchmark data. What you can do is acknowledge it explicitly, apply a conservative haircut to any "average" MOIC figure you see cited, and ask GPs what their inclusion criteria are when they claim top-quartile status.

How to Use MOIC Intelligently When Evaluating a Fund or Deal

MOIC is not broken. It is just one number, and one number is never enough.

First, always demand net MOIC. If a GP cannot produce audited net figures, that is a due diligence failure, not a negotiating point. Net figures are the only ones that represent money you actually receive.

Second, decompose TVPI into DPI and RVPI before forming any opinion. A 2.5x TVPI with 0.2x DPI means 92% of the multiple is paper. The median VC fund from the 2019–2020 vintages had returned less than 0.3x DPI as of Q3 2024, despite TVPI multiples that looked acceptable in quarterly reports. As of that same quarter, distribution rates across VC averaged single-digit percentages of NAV for eight consecutive quarters , far below the decade average of 16.8%. TVPI without DPI is a promise, not a result.

Third, pair MOIC with IRR and hold period together. A 2.5x MOIC over 5 years equals roughly 20% IRR. A 2.5x MOIC over 10 years equals roughly 9.6% IRR , close to public equity long-term averages, with far less liquidity and higher fees. The multiple alone tells you nothing about whether the opportunity cost was worth it.

Fourth, ask about subscription lines. Request IRR both with and without credit facility impact. The ILPA 2017 standard recommends this disclosure. Any GP that refuses to provide it is hiding an IRR boost that does not reflect actual investment skill.

Fifth, benchmark against net, vintage-matched data. A 2018-vintage VC fund at 2.8x net TVPI is performing near the top-quartile threshold of 3.1x for that year. The same 2.8x from a 2021-vintage fund clears the top-quartile of 1.6x by a wide margin , but that top-quartile threshold itself is low because most 2021 vintage funds entered at peak valuations. Context is everything.

Sixth, size your portfolio to survive the power law. Angel investing's 2.6x average requires you to stay in the game long enough to catch the outliers. ACA data shows portfolios of 15 to 25 or more companies achieve dramatically higher median IRRs than portfolios of one to five. If you put all your angel capital into three deals, the average MOIC statistic is irrelevant to your actual outcome.

The number GPs lead with is the number that makes their fund look best. Your job as an LP or angel investor is to find the number that tells you what you actually earned, on what timeline, net of everything you paid, compared against a benchmark that includes the funds that did not make it. That number is almost always smaller than the headline. Knowing that before you commit is the only edge that cannot be marketed away.

Disclosure: This article is for informational and educational purposes only. It does not constitute investment advice, a solicitation, or an offer to buy or sell any security or fund interest. All performance data cited is sourced from publicly available third-party sources and reflects historical results. Past performance does not guarantee future results. Private equity, venture capital, and angel investing involve substantial risk, including the possible loss of all invested capital. Consult a licensed financial advisor before making any investment decision.