Art experts enjoy claiming that markets for fine art have no logic. So for example in 1978 Robert Hughes, of Time magazine, declared that "The price of a work of art is an index of pure, irrational desire." Similarly, in 2006 Sotheby's chief auctioneer, Tobias Meyer, claimed that the market's valuation of art could only be described as "magical." In 2005 Richard Benson, dean of the Yale School of Art, denied that the market's pricing of an artist's work was in any way related to the artist's importance, as he told the New York Times that "We don't consider success in the marketplace has anything to do with being a successful artist." In 1998 Robert Storr, then a curator at New York's Museum of Modern Art, denied the possibility of gauging artistic importance in any systematic way, declaring that artistic success "is completely unquantifiable." This implies that studying market outcomes is misguided; and this was directly stated by Michael Rooks, a curator at Chicago's Museum of Contemporary Art, who explained that "There's a real sense that when you start quantifying artistic output in dollars and cents, those things are tangents to what we really should be talking about." Even worse, some art experts believe that anyone who analyzes market evidence is dishonest: thus Robert Rosenblum, a curator at the Guggenheim Museum, told the Wall Street Journal in 1998 that "I immediately distrust anybody trying to detect patterns ... in art, especially in terms of economics."
In the face of these vehement denunciations by high priests of the art establishment, perhaps only a very brave -- or perhaps foolhardy -- individual would persevere in an attempt to find logic in the art market. Yet the very certainty of these assertions might raise suspicions. The statements I've cited here all are claims about economic and statistical analysis. Yet not one of the six people who made these statements has any formal qualifications to discuss either economics or statistics. So let us not be unduly intimidated by the strident tone of these claims; let us simply consider whether they are true.
Are art markets in fact irrational? We obviously can't answer this question without knowing what a rational market looks like. There is no immediate formula for this, because this is not a standard concept in economics. Yet it is not difficult to point to some elements that might reasonably be associated with this term.
One of these involves the investment value of art. Many of the claims of irrationality appear to be motivated by the belief that art prices have in some sense become unreasonably high. To say that the price of any asset is excessively high presumably implies that it is a bad investment -- i.e., that its price will increase little in future, or perhaps even fall, yielding a low, even negative, rate of return. But a series of empirical studies that have used auction prices to calculate rates of return to paintings over time have in fact tended to show that art is a good long-term investment. So for example in one study based on repeat sales of paintings at auction, Michael Moses and Jianping Mei found that during 1950-2000, American paintings yielded an average annual nominal return of 12.2 percent, only slightly below the average return realized by the S&P 500 during the same period, of 12.6 percent. Impressionist and Old Master paintings yielded moderately lower average rates of 11.6 percent and 11.2 percent, respectively.
Now we would of course expect rates of return to art to be lower than those on financial assets, because unlike stocks, which yield only pecuniary benefits, paintings have additional advantages for their owners. Some collectors enjoy being known for their sophistication: as Thorstein Veblen famously wrote in 1899, "Conspicuous consumption of valuable goods is a means of reputability to the gentleman of leisure." Collectors can of course also simply enjoy looking at their art, and this enjoyment and that of public reputation are not mutually exclusive: Veblen pointed this out, for he observed that "very many if not most of the most highly prized works of fine art are intrinsically beautiful," and consequently "their exclusive enjoyment gratifies the possessor's sense of pecuniary superiority at the same time that their contemplation gratifies his sense of beauty." What studies like that of Moses and Mei tell us is that both of these types of benefits can be gained at very low cost: owning paintings instead of stocks may reduce an investor's expected return only slightly.
If paintings are a good investment, then the prices collectors pay for art are not irrational. But this is not the only evidence of rationality that emerges from auction outcomes. Specifically, the relative prices of paintings produce strong evidence of rationality, in two distinct respects.
One of these concerns prices realized at auction across artists. If auction markets were irrational, there would be no reason for the work of the greatest artists to sell for the highest prices. I have demonstrated elsewhere that we can systematically identify the most important artists by surveying evidence generated by art experts: the greatest artists are featured most prominently in the narratives of art history textbooks, and are featured in exhibitions presented at the greatest museums. Does the art market recognize these artists with its highest prices? The answer is obvious: it does. The living artists whose work has brought the highest prices at auction are Jasper Johns, Jeff Koons, Damien Hirst, Gerhard Richter, and Takashi Murakami. These are obviously among the very most influential artists of their respective generations; all figure prominently in textbooks of art history, and all have been honored with major exhibitions at the greatest museums. The work of lesser artists yields lower prices at auction. So here we see that auction outcomes are closely correlated with the careful and measured judgments of art scholars on the relative importance of different artists. And of course this is hardly surprising. As the critic Peter Schjeldahl conceded during the art boom of the late 1980s, "I must admit that the artistic judgment of current big bucks is better than the average among, say, critics. (Like the prospect of being hanged, shelling out millions may concentrate the mind wonderfully.)"
Another relationship involving relative prices offers perhaps an even more striking demonstration of the rationality of the art market. This concerns the prices of paintings done at different stages of a single artist's career. I have used tens of thousands of auction prices for more than 100 important modern painters to estimate age-price profiles for these artists. I have then compared these profiles to tabulations of each artist's age at the time of execution of the paintings illustrated in textbooks, and of each artist's age at the date of execution of the paintings included in the artist's most recent complete retrospective exhibition.
The results of these comparisons are striking. One example: Cézanne's age-price profile rises to a peak in the final year of his life, at age 67. This is precisely the same age from which American and French textbooks published in the past 30 years most frequently reproduce his paintings, and it is also the same age from which his most recent retrospective, presented at the Musée d'Orsay and the Philadelphia Museum of Art in 1996, included the largest number of his paintings. Another example: Picasso's age-price profile peaks at age 26, then declines thereafter, for the remaining 66 years of his life. This peak age is again precisely the same age from which his work is most frequently reproduced in American and French textbooks, as well as the same age from which his most recent compete retrospective, at New York's Museum of Modern Art in 1980, included the largest number of his works. I could go on: I have done this analysis for more than 100 important modern painters, and the peaks of the age-price profiles almost always correspond closely to the peak ages from the textbook illustrations and the retrospective exhibitions.
In one sense, these results are hardly surprising. It is widely recognized that Cézanne was at his greatest late in his life: Meyer Schapiro, for example, wrote that "the years from 1890 to his death in 1906 are a period of magnificent growth." And for Picasso, age 26 -- the year 1907 -- was when he painted Les Demoiselles d'Avignon, which George Heard Hamilton described as no less than "the watershed between the old pictorial world and the new." The late Cézanne and the early Picasso are so obviously the greatest periods of innovation for these artists that it should come as no surprise to any reasonable person -- certainly to any economist -- that work of this vintage yields higher auction prices than any other periods in these artists' careers. Yet from another vantage point, art experts must consider these results surprising, for the results directly contradict these experts' claims that the art market is irrational.
The Oxford English Dictionary defines irrational as "contrary to or not in accordance with reason; unreasonable, utterly illogical, absurd." Does this apply to the prices of paintings, as Robert Hughes would have it? Quantitative studies of auction records show that the average return to paintings over long periods is only slightly below the return to the S&P 500. Studies of auction results also reveal that it is the greatest artists whose work commands the highest prices. And studies of auction prices clearly demonstrate that it is the most important periods of great artists' careers' that bring the highest prices. Irrational? Hardly. Art experts need to become more sophisticated in their understanding of art markets: it is not the art market, but rather the claims of art scholars and critics about it, that are unreasonable, utterly illogical, and absurd.
[An earlier version of this article was published in The Art Economist. I thank Bruce Helander, Marisa Pascucci, and Andy Ruppanner for the opportunity.]