Photography has spent a decade being the asset class nobody wanted to talk about. After the 2008 to 2014 peak, with Gursky’s Rhein II at $4.3M and the Sherman and Prince markets trading at full painting-market multiples, prices softened, editions traded below their historic highs, and the collector base aged without obvious replacement. The category became, by 2019, a backwater. That is now changing, and the catalysts look structural rather than sentimental. For a portfolio allocating to art, this is the cleanest undervalued-asset-class setup in the contemporary market today.
The analog: photography in 2026 looks a great deal like photography in 2004. A legacy market with real scarcity, a new collector base forming underneath the old one, and a price curve that has compressed to the point where convexity is the actual trade.
The drawdown nobody remembers correctly
Between the 2013 to 2014 highs and the 2021 to 2023 trough, the top tier of the photography market, meaning Gursky, Wall, Sherman, Prince, Struth, and the vintage Diane Arbus, Irving Penn, and Richard Avedon markets, drew down by 30 to 50 percent on a repeat-sales basis. Editioned works from the 1990s and 2000s with nominal top estimates of $800K to $1.5M in the prior decade were clearing in 2022 at $400K to $700K, when they cleared at all.
The drawdown was not a referendum on the work. It was a generational rotation. The collectors who built the photography market in the 1990s aged into estate mode. The collectors who might have replaced them got distracted by NFTs in 2021 and by AI in 2023, and by a general preference for painting. Volume thinned. Prices followed.
What has actually changed
Three things, none of them sentimental.
First, vintage print scarcity has tightened materially. Every year that passes removes a piece of the available supply of vintage gelatin silvers (Arbus, Penn, Avedon, Frank, Evans, Model, Abbott, and the photojournalism canon below them) because the best of those prints are now in institutional collections that will not deaccession. The trade-able pool for genuine vintage material has contracted roughly 20 to 30 percent over the last decade, depending on whose estimate you use. Scarcity compounds.
Second, the Gursky-Wall-Sherman-Prince tier is no longer making new work at the 1990s pace. Gursky’s production has slowed. Wall’s output was never fast. Sherman’s most market-active bodies are settled. Prince has moved into phases the market has not embraced. Top-tier supply has turned into a closed universe. That is the condition a category needs to rerate.
Third, and most important, the new collector base is visible. It is younger (median age in the low forties rather than the mid sixties), more female than the old photography buyer class, and more institutionally aligned (meaning buyers who sit on museum boards and photography committees, who buy with institutional placement in mind). Phillips’s photography sales have been the cleanest read on this; their 2024 and 2025 new-registrant mix looks materially different from 2019.
These three factors compound. Scarcity alone is not a thesis; buyers alone are not a thesis; closed-universe production alone is not a thesis. Together they are the setup.
The undervaluation math
Compare a few pairs.
- A top-tier Gursky trades in the $2M to $5M range. A top-tier Richter abstract from the same years trades at $15M to $30M.
- A strong Cindy Sherman Untitled Film Still in a small edition trades in the low-to-mid six figures. A single Christopher Wool word painting from the same era trades in the low-to-mid seven figures.
- A vintage Diane Arbus print trades in the $200K to $600K range. A single Philip Guston drawing from the same period trades at a meaningful multiple of that.
These comparisons are not apples-to-apples. Photography is editioned; painting is unique. That is the usual objection, and it is the wrong one at this tier. Top-edition Gurskys, Sherman’s Untitled Film Stills, and Wall’s lightboxes are effectively unique for market purposes, because the edition sizes are small, the best prints are in institutions, and what trades is a very short list of available examples. The “editioned so discount forever” argument is lazy.
“Photography is priced like a category that is still losing collectors. It isn’t. It’s gaining a different kind of collector, and the price curve hasn’t caught up.”
The specific trade
The strategist’s position here is threefold.
Position one: accumulate top-tier vintage where condition and provenance are clean. The edge is in paying up for A-condition material rather than bargain-hunting for B material. In a rerating, A material moves first and moves most. The spread between A and B will widen.
Position two: selectively buy the Gursky-Wall-Sherman-Prince tier on the secondary market, specifically works that have traded once or twice and are sitting with dealers at 2019 to 2022 pricing. Hold period three to seven years. The convex part of the trade is that downside is capped by institutional demand at current levels while upside is a return to 2014 peaks plus the intervening inflation.
Position three: the younger tier that sits adjacent to this opening. Wolfgang Tillmans (whose market is structurally undersized versus his cultural footprint), Catherine Opie, An-My Lê, LaToya Ruby Frazier, Deana Lawson, Tyler Mitchell. Not all of them will work as investments. The thesis is that institutional demand at this tier has already rerated while the auction tape has not, which is the setup for the tape to catch up.
Within each of the three positions, condition discipline is the operating constraint. This is a market where an A-minus print can sit below the bid for a year and an A-plus example of the same image sells in a week. A portfolio that ignores condition ends up with optionality on paper and illiquidity in practice.
What could invalidate this
Easton Cain, who has been public about stepping back from the ultra-contemporary trade since 2023, has argued the opposite case to us privately, which is worth addressing. His view: the new photography collector base is real but smaller in capital terms than the rerating thesis assumes, and the category requires painting-market depth to sustain painting-market multiples, which it will not get.
That is a serious argument. The trigger that would invalidate our thesis is a soft 2026 and 2027 at the top of the photography market, specifically a Gursky or Sherman offered at a reasonable level that fails to clear, and a Phillips photography sale (which has been the category’s cleanest barometer) where sell-through breaks below 70 percent by lot. If those two things happen, the rerating is premature and the position is a multi-year hold at best.
If those two things do not happen, the trade works. The path dependency is simpler than it looks.
One other risk worth flagging: the generational rotation cuts both ways. If the new buyer base turns out to be smaller than the cohort it is replacing rather than merely differently composed, the rerating stalls regardless of scarcity. That is the quieter version of the Cain argument and it deserves the same seriousness.
The forward view
Photography is where contemporary painting was in the middle of the 2000s. A category with genuine scarcity at the top, a collector base that is forming rather than decaying, and a price curve that has bottomed after a long drawdown but has not yet begun to reprice upward.
Over a five-year horizon, the base case is that the Gursky-Wall-Sherman-Prince tier retests and modestly exceeds its 2014 highs in nominal terms, which after inflation is a recovery rather than a new peak. The bull case is a full rerating toward painting-adjacent multiples at the top, which would imply a two-to-three-times move from current levels on top examples. The bear case is that the rerating stalls for another cycle and photography remains a structurally cheap category for longer than the thesis allows.
The risk-reward is asymmetric. That is the actual argument. Size positions accordingly, accept a three-to-seven-year hold, concentrate on A-condition material, and watch Phillips’s photography sell-through as the single best real-time indicator. If that number holds above 75 percent through 2026, the thesis is on track. If it breaks below 70, revisit. That is the trigger.
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.