Venture capital returns vary by manager, vintage, strategy and the period being measured. A strong year for the asset class can coexist with a disappointing longer-term comparison, and a valuable portfolio can coexist with limited cash distributions.
Start by identifying the result you want to assess: an individual deal, a fund, an investor's complete venture allocation or an industry benchmark. Those are different portfolios.
What recent benchmark evidence shows
Cambridge Associates' July 2026 report gives the following US venture benchmark figures for periods ending 31 December 2025:
| Period | US VC index | S&P 500 mPME comparison |
|---|---|---|
| One year | 21.1% | 18.0% |
| Five years | 10.2% | 14.8% |
| Ten years | 14.9% | 15.0% |
The private figures are pooled horizon IRRs, net of fees, expenses and carry; multi-year returns are annualised. The public comparison uses Cambridge's modified public-market-equivalent methodology, rather than simply placing an ordinary index return beside a private-fund IRR. Cambridge Associates, calendar-year 2025 report, Figures 1 and notes.
This evidence supports a qualified answer: venture led that comparison over one year but lagged over five years and was close over ten. It does not establish that any available new fund will reproduce the index, or that one benchmark captures every investor's opportunity cost.
Read multiples and annualised returns together
TVPI measures distributions plus residual NAV relative to paid-in capital. DPI measures distributions alone. Both need a common net/gross basis and as-of date. A high TVPI supported mainly by private marks is a different position from the same TVPI supported mainly by distributions.
IRR adds timing. It is the discount rate that makes the specified cash flows, including terminal value where used, have zero net present value. It does not ignore time; its sensitivity to time is precisely why early distributions, delayed capital calls and financing arrangements affect the result.
For a single £1m contribution and £2m payment five years later, the annualised return is about 14.9%. If the payment arrives after ten years, it is about 7.2%. The multiple is 2x in both cases. Real funds require the complete sequence of contributions, distributions and residual values. See TVPI vs DPI for the underlying definitions.
What a public-market equivalent does
A PME asks how a specified set of private-investment cash flows compares with investing in a public benchmark over matching dates. Methods differ, so name the one used.
For the Kaplan–Schoar method, grow each contribution and distribution to the measurement date using the selected total-return index. For each cash flow, multiply its amount by the index value at the measurement date divided by its value on the cash-flow date. Then calculate:
KS-PME = (sum of benchmark-grown distributions + remaining NAV) ÷ sum of benchmark-grown contributions.
Use positive amounts for contributions and distributions. Remaining NAV is measured at the end date, so it needs no further growth adjustment. A result above one indicates outperformance of that cash-flow-matched benchmark on this basis.
Kaplan and Schoar's original paper explains the approach and illustrates it with simple contribution-and-realisation examples. It also discusses the limits of interpreting the comparison as a full risk adjustment. Kaplan and Schoar, 2005, section III.A.
PME works for a single SPV too
Suppose an investor contributes £100,000 to an SPV and receives £180,000 five years later, with no other cash flows or residual value. Over the same period, the selected total-return index grows by 50%.
The benchmark value of the contribution is £150,000. KS-PME is £180,000 ÷ £150,000 = 1.2. The investment outperformed that comparison, subject to the assumed net cash flows and benchmark.
There is no mathematical requirement for multiple deals or a blind pool. A composite of several SPVs can also be measured if the cash flows, valuations, ownership and attribution are specified consistently. Selecting only successful SPVs would make the record misleading.
Ask how the benchmark was built
Check the population, geography, vintages, weighting and source of valuations. Capital-weighted results can be dominated by large funds; a median describes a different experience. Industry data may also reflect voluntary reporting and changing coverage.
Then choose a relevant public comparison. The S&P 500, a small-cap index and a technology-heavy index represent different exposures. Currency, dividends and measurement dates must match the intended question. A benchmark should not be selected after seeing which one makes the manager look best.
Finally, assess the remaining obligations. Undrawn commitments, fees, liquidity restrictions and concentration affect the investor's experience even when a reported performance measure is accurate.
A defensible return assessment includes the full record, a transparent valuation basis and an appropriate comparison. A quoted target multiple is only an ambition until those results exist.
FAQ
Do venture capital funds beat the S&P 500?
The answer varies by fund and measurement period. The Cambridge comparison above shows different relative results across one-, five- and ten-year periods. It is a dated pooled benchmark, not a promise about the next fund or a simple average of every LP's experience.
What is PME in venture capital?
A public-market equivalent compares private-investment cash flows with a chosen public benchmark over matching dates. Methods differ. The KS-PME calculation above includes distributions and remaining NAV relative to benchmark-adjusted contributions; its interpretation depends on the specified cash flows and whether they are net of costs.
What is a good IRR for a venture fund?
Assess it alongside risk, duration, strategy, vintage, multiples and remaining valuation uncertainty. A high IRR with little capital returned can describe a different investor experience from a mature realised result. Use a relevant comparison and the full cash-flow record instead of a universal hurdle quoted without context.
Can a single SPV be compared with a public-market benchmark?
Yes. A PME calculation can use one investment's contributions, distributions and remaining value. Multiple holdings are not a mathematical requirement. Choose the benchmark and measurement basis consistently, include costs as intended and avoid selecting only successful vehicles for a composite record.
