The art-investment narrative is built on a carefully curated set of winners. Every index you have seen, every “art outperforms the S&P” chart that circulates on LinkedIn every quarter, is pulling from the same small pool of resold works by artists with durable secondary markets. That is a real data set. It is also a rigged sample. Most art purchased in the last 30 years has never resold, will never resold, and could not be resold at any meaningful fraction of its original purchase price if the buyer tried.
This is not a controversial claim inside the trade. It is the first thing any honest advisor tells a new client. It is the last thing any art-investment platform will put in its marketing deck.
The indices, and what they actually measure
The Mei-Moses index, the Artnet price indices, the various Artprice composites, they all share the same structural problem. They track repeat sales of the same works. If a painting sells once and never again, it cannot enter the index. If it sells at auction twice and the second sale is below the first, it enters the index as a negative data point, but only if it sold. Works pulled from auction (“bought in”) do not register. Works consigned and quietly withdrawn when the phones went cold do not register.
The universe of works that enters these indices is, by construction, the universe of works that resold at auction. That is a tiny fraction of the universe of works purchased. The indices are not lying. They are answering a narrow question: how did already-liquid works perform? The interesting question is the one they cannot answer: what happened to everything else?
Survivorship bias is the dominant distortion here. Any data-driven case for art as an asset class that does not begin with this caveat is, at best, incomplete. At worst, it is a sales document.
What the graveyard looks like
Walk through the primary-market purchases of the last two decades. A reasonable estimate, supported by dealer data and private-sale tracking, is that more than 80 percent of works bought from galleries at the emerging or early-career level never trade on the secondary market at all. They hang on walls, move between storage units, occasionally get donated, rarely get sold privately at a loss.
The next tier down is starker. Of the works that do reach the secondary market, a meaningful share sell below their original primary-market price, sometimes by 50 to 90 percent. This is not a bug. It is the baseline outcome for a market where the entry price is set by narrative, not by a clearing mechanism.
“The average primary-market purchase is not an investment. It is a consumption good with a small, unreliable option on future appreciation.”
The average primary-market purchase is not an investment. It is a consumption good with a small, unreliable option on future appreciation. Pricing it as anything else is a category error.
The studios that close
Every year a certain percentage of working artists stop making art, for all the reasons people stop making things: money, life, burnout, a teaching job that paid better, a market that turned. When a studio closes without a meaningful institutional footprint, the secondary market for that artist’s work usually disappears within a decade. No galleries represent the estate, no auction house lists the work, no dealer takes the call.
The works still exist. They still hang on walls. They have, functionally, zero market value, because there is no market. This is the part of the asset class that does not show up in any data set.
Lucian Poe has a useful frame for this: the long tail of contemporary art is not a distribution, it is a cliff. A small number of artists carry durable markets for decades. The rest do not carry markets at all. The middle, the “successful mid-career artist with a modest but real secondary market,” is much smaller than the gallery ecosystem implies.
Why this is fine, if you can admit it
None of this is a reason not to buy art. It is a reason to be honest about what you are doing. If you buy a painting because you love the painting, and you never sell it, and you hang it on your wall for 40 years, the painting delivered the return you bought it for. The return is non-financial, it is real, and it is the actual reason most serious collectors collect.
The trap is treating the painting as an asset when your behavior says it is an object. Three symptoms of the trap:
- You would not sell the work at any price, which means you cannot mark it to market, which means you do not actually know what it is worth.
- You hold works that have declined 70 percent from purchase because “it has not traded recently,” as if absence of trade is evidence of stable value.
- You run no position sizing, no concentration limits, and no stop criteria, because the works are “part of the collection.”
These are fine behaviors for a collector. They are disqualifying behaviors for an investor. Pick one.
The asset-class trap
The last decade produced an entire category of product (fractional platforms, art-backed tokens, art-focused funds) built on the premise that art can be financialized into a normal portfolio exposure. Some of these products are legitimate. Most of them are selling access to the same narrow set of blue-chip names that already trade liquidly, with a fee layer that absorbs most of the incremental return.
The interesting question is not whether a specific Picasso will appreciate. It is whether a portfolio of 40 mid-career contemporary works, purchased across 10 dealers over a decade, delivers a risk-adjusted return that survives fees, storage, insurance, and the transaction costs of eventual sale. The honest answer, based on the data that does exist, is that it probably does not, and the ones where it does were driven by two or three outliers inside a basket of flat or negative names.
Easton Cain has been public about stepping back from the ultra-contemporary trade since 2023, and his framing is worth borrowing: the market rewards concentration in the names that already have durable liquidity, and punishes diversification across names that do not. That is the opposite of how you build a traditional portfolio. It is why treating art like a portfolio is usually the wrong model.
What to do with this
If you are in the market, three practical adjustments:
- Buy what you would keep if the market for that artist disappeared tomorrow. Most of the market for most artists will, eventually, disappear. Your consolation needs to be the object.
- If you insist on the investment frame, concentrate. One meaningful work by an already-liquid artist beats a diversified portfolio of emerging names, every time, on a risk-adjusted basis.
- Do not confuse insurance valuations with market valuations. The insurance number is the replacement cost if the work is destroyed. It is not what you can sell it for on Tuesday.
The fees nobody models
A realistic round-trip on a secondary-market work, ignoring price movement entirely, is not a small number. Buyer’s premium at the major houses runs well over 25 percent on the lower price bands, stepping down only for very large lots. Seller’s commission, even for favored consignors, is rarely zero on works below a meaningful threshold. Add shipping, import duty where relevant, insurance during transit, condition reports, photography for the catalogue, and a storage bill for the period between consignment and sale.
Stack the round-trip friction and a work has to appreciate 35 to 45 percent from purchase to auction hammer just to break even on a net-of-fees basis. Most works do not do this over any reasonable horizon. The “art outperforms equities” chart does not back out the buyer’s premium on either side. It should.
The forward view
The next cycle will expose the survivorship problem more clearly than the last one did, because the ultra-contemporary bubble of 2020 to 2022 produced an unusually large cohort of artists at unusually high prices, and that cohort is now aging into the phase where secondary-market reality tends to arrive. Watch the buy-in rates on mid-career contemporary evening-sale lots over the next four auction seasons. If they creep above 30 percent, the graveyard is visible on the tape. Be careful.
Nothing in this article is investment advice. CreativeSlop is an independent publication. Figures rounded for readability. Names of market participants referenced in good faith from on-the-record and on-background conversations.