Start with the size comparison, because it frames everything that follows. Art owned by private individuals is roughly a $1.5 trillion asset class. Venture capital and private equity together are about $3.5 trillion. There are roughly 6,000 firms that help people allocate to venture and PE. There is really only us trying to do it for art at any scale. So when an investor asks whether art or venture is the better diversifier, the honest answer starts one level up: these are two of the most inefficient, illiquid corners of a wealthy person's portfolio, and they earn their place through different mechanisms. Venture pays you for taking enormous, concentrated, skewed risk. Art pays you for owning something durable that the rest of your portfolio largely ignores. We think the interesting question is not which one wins. It is what each one is actually for.

A note on what we mean by "risk return profile." We mean four things you can measure: how the returns are distributed, how much they bounce around, how correlated they are to the rest of your portfolio, and how long your money is locked up. Art and venture sit at almost opposite corners on the first two and look surprisingly similar on the last two. That contrast is the whole article.

How are venture capital returns actually distributed?

Venture is a power law business, and the power law is not a footnote. It is the entire return engine. A 2025 deal-level study found that roughly 6% of venture deals generate something like 60% to 80% of the total upside in a typical early-stage portfolio, with most of the rest clustering between zero and the money invested [1]. Correlation Ventures and Horsley Bridge, two firms that have studied large samples of actual deal outcomes, report the same shape from different angles: the vast majority of investments return less than the capital put in, and a tiny set return more than ten times it. The World Economic Forum's 2026 venture report puts the failure rate as high as 75% and states the case plainly, that a few winners drive returns [4].

This shows up at the fund level, not just the deal level. Capnamic's analysis of historical venture fund cash flows found realized fund IRRs ranging from negative 50% to over 200%, with an average around 6% and a median of negative 6% [3]. Read that median again. More than half of the funds in that dataset lost money for their investors. Cambridge Associates, working over a longer 30 year window, reports a healthier picture for the asset class as a whole, roughly 21% average annual net IRR for early-stage funds and about 12.6% for later-stage funds [2], but those averages are pulled upward by a handful of extraordinary vintages and the funds at the top of the distribution.

So the lived experience of venture depends almost entirely on which funds you are in. The spread between a top-quartile manager and a median one runs well past 20 percentage points of annual IRR, far wider than you would ever see in public equities. That feels like the defining feature of the asset class.

Venture is a game of access and selection. If you can get into the top funds, the returns are exceptional. If you cannot, the median outcome is a loss.

How does art behave by comparison?

Art does not work this way, and that is the point of putting the two side by side. Contemporary art has appreciated at roughly 11.4% a year from 1995 to 2024, according to repeat-sale data we track and the Artprice indices [5]. The broad all-art index, going back to 1871 in the Mei Moses series, has compounded at about 9% a year, against roughly 7% for the S&P 500 over the same very long window [5]. The Artprice 100, a basket of the top artists at auction, has run near 10% a year [5].

Those are index-level numbers, and they smooth a lot. Realized returns on individual resales are choppier and cycle-sensitive: Bank of America's 2026 U.S. art market report found that resold evening-sale lots in New York returned about 4.4% annualized in 2025, 5.3% in 2024, and 8.1% in 2021, with the figure trending down each year as the market normalized from its 2021 peak [6]. We have said for a while that the market has been working through a correction, and those numbers are what that looks like up close.

But the distribution is the real contrast with venture. Art works rarely go to zero. A failed startup is worth nothing. A painting that falls out of fashion still has a floor, because it is a physical object with scarcity, provenance, and some residual collector demand. There are 21 Jackson Pollocks left in private collections. Supply in major artist markets tends to shrink over time as works are donated to museums and leave circulation permanently. The result is a return distribution that is far less skewed than venture's, with much lower odds of total loss and correspondingly lower odds of a 50x outcome.

Where venture is a small number of enormous winners carrying a sea of losses, art is a large number of modest, durable appreciators with a long holding period. Different machine, different output.

What do the volatility and Sharpe numbers say?

Here is where the comparison gets useful for portfolio construction. Early-stage venture carries annual volatility that institutional benchmarks put in the 25% to 35% range, with heavy left-tail risk [1][3]. Public equities sit around 15% to 20%. Art indices come in lower, generally modeled in the 12% to 15% band [5].

A fair caveat: some of art's apparent calm is an artifact of infrequent trading. A painting is marked when it sells, not every day, so the measured volatility understates the true swings. The same smoothing applies to venture marks, which is why 2018 to 2021 vintages looked fine on paper until valuations reset in 2022 and 2023. Neither asset is as smooth as its reported series suggests. We try to be honest about that rather than lean on it.

On a Sharpe ratio basis, which is just appreciation divided by volatility, the picture roughly sorts out like this:

  • Contemporary art: about 11% to 12% return on 12% to 15% volatility, a Sharpe in the neighborhood of 0.6 to 0.8 [5].
  • S&P 500: about 9% to 10% return on 15% to 20% volatility, a Sharpe around 0.4 to 0.6 [5].
  • Venture capital: mid-teens or better returns on 25% to 35% volatility, with a Sharpe that is highly dependent on the vintage and the manager, and not obviously better than public equities once you account for skewness and survivorship [1][2].

We would not over-read those figures. Sharpe ratios for illiquid assets are estimates built on smoothed data, and any number with a decimal point implies more precision than the underlying market offers. But the direction is consistent across Citi and Deloitte's art-finance work: art has delivered roughly equity-like returns with somewhat lower measured volatility and very low correlation. That combination is what improves a portfolio's efficiency, and it is a different value proposition than venture's.

What about correlation and illiquidity, where they overlap?

This is the part where art and venture look like cousins. Both are uncorrelated to public markets, and both lock up your capital for years.

On correlation, contemporary art has been almost entirely uncorrelated with bonds, housing, the S&P 500, and gold since 1995, with coefficients generally in the 0.0 to 0.2 range [5]. Its highest correlation is with gold, and even that is roughly 0.1 to 0.2. Venture is also driven by its own forces, company outcomes and exit windows and innovation cycles rather than daily market sentiment. True diversification means owning something that is largely indifferent to whatever is moving the rest of your holdings. Both assets clear that bar.

The difference is what drives them. Venture rides the technology cycle and the IPO window. Art rides wealth creation at the very top, what we think of as a call option on the top 0.01%, the Bezos-level buyers writing nine-figure checks. Those two engines do not run on the same clock, which is why owning both can reduce concentration even within the illiquid part of a portfolio.

On illiquidity, they are close to equivalent and worth respecting. A venture fund has a 10 year life with extensions, and average holding periods run about 8 years for early-stage deals [2]. Art is similar: the average holding period for works resold at auction is around 13 years, with round-trip transaction costs (buyer's premium, seller's commission, insurance, storage) that can run 10% to 25% [5][6]. The 2024 to 2025 exit drought made venture's illiquidity especially visible. Secondary stakes were trading near 60 cents on the dollar by the end of 2025 [4]. Neither asset is something you sell in a hurry. You should underwrite both as multi-year, lumpy-exit commitments.

How are HNW investors actually sizing these?

The data here is messy, because private wealth does not report itself cleanly, but the pattern is consistent across the major wealth surveys.

High-net-worth investors hold somewhere around 28% to 31% of their wealth in private and alternative assets, per Long Angle's 2024 study [7]. Ultra-high-net-worth investors and family offices go further, with alternative allocations that can reach 40% to 50% once you count operating businesses, real estate, funds, and luxury assets [8][9]. Inside that bucket, the rough division looks like this: private equity and venture together run about 5% to 15% of total net worth for most families, sometimes 10% to 20% of the portfolio in growth-oriented family office models, while art tends to sit at 1% to 5% for diversified investors and higher for dedicated collectors [9][10].

We think those guardrails are sensible, and they match how we have always talked about art: a small, long-term, illiquid allocation, somewhere in the 1% to 5% range to start, held for 3 to 10 years or longer. The reasoning is not that art will beat venture. It is that art and venture do different jobs. Venture is your high-variance growth engine, the place you accept a power-law gamble in exchange for the chance at a step-change outcome, and where manager selection is the entire ballgame. Art is closer to a store of value with low correlation, a durable asset whose dispersion between a good buyer and a naive one is real but far narrower than the gap between a top-quartile and bottom-quartile venture fund.

Sized correctly, they are not competitors for the same slot. They are two different tools in the illiquid sleeve.

The Bottom Line

Art and venture capital are both inefficient, illiquid, low-correlation assets, and that is most of what they have in common. Venture is a power-law machine where roughly 6% of deals drive 60% to 80% of returns, the median fund has historically lost money, and your outcome depends almost entirely on getting into the right managers. Art is a steadier, more durable appreciator, compounding in the high single digits to low double digits over long horizons, with much lower odds of total loss, lower measured volatility, and a Sharpe ratio that compares favorably to equities. We do not think art beats venture, and we would be skeptical of anyone who claimed it did. We think a long-horizon investor with a real alternatives budget has room for both, sized small, held long, and understood for what each one actually does. Venture for the asymmetric upside. Art for the diversification and durability. That feels like the right way to hold them.

Sources

  1. Financial Models Lab, "Analyzing Returns of Different Venture Capital Deals," 2025. https://financialmodelslab.com/blogs/blog/analyzing-returns-different-venture-capital-deals
  2. Industry Ventures (using Cambridge Associates and PitchBook/NVCA data), "The Venture Capital Risk and Return Matrix." https://www.industryventures.com/insight/the-venture-capital-risk-and-return-matrix/
  3. Capnamic, "The Risk-Return Profile of Venture Capital." https://capnamic.com/post/the-risk-return-profile-of-venture-capital
  4. World Economic Forum, "The Future of Venture Capital," 2026. https://reports.weforum.org/docs/WEF_The_Future_of_Venture_Capital_2026.pdf
  5. Masterworks research summarizing Artprice, Mei Moses, Citi Private Bank, and Deloitte Art & Finance data (contemporary art appreciation 1995 to 2024, all-art index since 1871, correlation and volatility estimates). https://www.artmarket.com
  6. Bank of America / Merrill, "U.S. Art Market Report," 2026 (resold evening-sale lot returns; holding periods; transaction costs). https://mlaem.fs.ml.com/content/dam/ust/articles/pdf/US-Art-Market-Report.pdf
  7. Long Angle, "High-Net-Worth Asset Allocation Study," 2024. https://www.longangle.com/research/high-net-worth-asset-allocation
  8. Altrata, "UHNW Asset Allocation," 2025. https://altrata.com/articles/uhnw-asset-allocation
  9. The Wealth Advisor via The Cashmere Fund, "Alternative Investments Explained," 2024 (UHNW alternatives and PE/VC allocation shares). https://www.thecashmerefund.com/blogs/alternative-investments-explained
  10. CAIS / Mercer, "The State of Alternative Investments in Wealth Management," 2025. https://www.mercer.com/en-us/insights/investments/financial-intermediaries/the-state-of-alternative-investments-in-wealth-management-2025/