Summer readings: The dismal science does art

hamiltonLast week came news that a reputable economist at the University of Chicago, David Galenson, has devised a quantitative method to measure the importance of 20th century artists. His rankings, which received major section-front coverage in The New York Times, are based on how often paintings or sculptures by a given artist are reproduced in each of 33 art history textbooks published between 1990 and 2005. Science accords merit on the basis of citations in the expert literature. Why not art?

And the winner is… “Les Demoiselles d’Avignon” … followed by … Vladimir Tatlin’s “Monument to the Third International” … followed by … “Spiral Jetty” … followed by … Richard Hamilton’s “Just What is That Makes Today’s Homes So Different, So Appealing?” Huh? The last picture—you know, the collage with the bodybuilder in the living room—came in just a nose ahead of “Guernica.”

Economists are irrepressible when it comes to drilling down to the essence of things. They peel away layer upon layer of history, nuance, and context—so much “noise”—to get to the hidden underlying algorithms of societal and human behavior. But methodology can devolve into mind mush—as in the case of asserting that looking at pictures in art history books can reveal much more than, well, the likelihood of finding certain pictures in certain books.

This exercise in solipsistic reductionism is a bit like mistaking the warped reflection in a fun-house mirror with reality itself. Even that may be giving too much credit to the theory. A fun-house mirror does reflect all that is placed in front of it, whereas the mirror of institutional art history has some conspicuous blind spots.

I am reminded of another quantitative economic study, of the auction market, which started off with eliminating the top 10 and bottom 10 percent of all auction results: A perfectly legitimate and common statistical maneuver to cleanse the data of trend-obfuscating outliers—only one that removed from the study all the data points that most people concerned with art values actually care about. Nonetheless, one has to admire the chutzpah, the sheer rationalist braggadocio of it all.

But let us not knock the dismal science. Sometimes, economists give us something tasty to chew on, and they have the laudable habit of backing up their notions with data—on this point, I am with Prof. Galenson all the way. Clare McAndrew, an economist from Ireland, has applied the tools of her discipline to the art market. Her recent book, The Art Economy: An Investor’s Guide to the Art Market (Liffey Press, 2007), is a sensible overview of art investing, albeit from a somewhat quirky EU/Irish perspective, providing insights into a global art economy that, according to her estimate, accounted for some 50 billion Euros in combined sales in the year 2006.

McAndrew sometimes falls into the irksome habits of her peers. Case in point, her formula for measuring prices of art works: P=f(EC) + UP, or: Prices of Art Works = f(Characteristics of Artist + Characteristics of Work + Characteristics of Sale) + UP. “What is UP?” you may ask, if you are still reading. It is the “Unexplained Premium” that a buyer in an auction room may be willing to pay for a work (p. 100). This UP, according to McAndrew, “seems to derive from the emotional, subjective and often irrational process peculiarly attached to purchasing art.” Quantify that!

The book is most helpful to the lay reader when McAndrew summarizes the hypotheses of her fellow economists about how art values behave in the real world. I invite our panel to judge which of the following “price anomalies”—i.e. effects on art prices—sounds realistic to you. Some may ring familiar; others may deserve to enter the broader art-world lexicon. Clare McAndrew, to her credit, points out that most of the suggested effects have had “another piece of research carried out to prove the opposite effect.” They are listed here with her descriptions (p. 105) and without further commentary:

The Death Effect: A rise in the value of artists’ values immediately preceding, at or immediately after their death.

The Winner’s Curse: Cost of being the highest bidder (negative immediate return) at auction due to the private valuation component of art (evidence in the short-term only).

The Masterpiece Effect: No evidence of masterpieces outperforming the market and evidence of overbidding/above average prices paid for them at auction.

The Afternoon Effect: Sequential price declines at auction: final bid relative to the auctioneer’s estimated price declines throughout the course of an auction.

The Morning Effect: Sequential rise in auction prices relative to their estimates.

Burned Artworks at Auction: Negative effect on the future value of works that are unsold/bought in the first time at auction.

Law of One Price: Prices higher in US vs. UK and higher in Sotheby’s v Christie’s—the law on one price does not hold for art.

3 thoughts on “Summer readings: The dismal science does art

  1. Personally, I love the two “time of day” effects, and there probably is some credence to them, if only because the specialists I’ve spoken to seem to take it seriously. More interesting, though, are the momentum speculations that the specialists make in ordering the appearance of works within their sales, though I haven’t seen any economist address this phenomenon in any systematic way–i.e. is the spot immediately following, say, the heavily promoted catalogue cover lot always doomed to under perform?

    One thing that makes the dismal science so much more dismal when it comes to the art market its numbers appear so completely arbitrary. How McAndrews came to her 50 billion Euros number is never explained, or even given a margin of error. Let’s face it, not until someone with on-the-ground knowledge and contacts within the art market decides to go back and get a Ph.D. in Economics will we ever get close to the kind of refined scholarship that might actually tell us something we don’t already know.

  2. In addition to the Morning and Afternoon Effects cited above, how about the “Morning After Effect”, which I would (uneconomically) define as delayed sticker shock, akin to “I can’t believe I ate the whole thing”. Awakening the day after ingesting massive quantities at auction, the plutocrat evinces a bit of indigestion, coupled with the rueful misgiving that he possibly spent too much.

    The obvious question: it there a pill?

  3. Jerome Weeks of artandseek.org, and former book critic of the Dallas Morning News, fills in some gaps about Galenson’s theory:

    What Galenson employed is called “historiometry,” and as crude or ludicrous as it might seem – the reputation of artists is to be determined by the cost of book illustrations? – it has a 139-year-old tradition. It’s generally taken to have been invented (or actually, popularized) by British eugenicist Francis Galton with his 1869 book, Hereditary Genius. The link between Galton’s racialist ideas and his promotion of a certain kind of white, European achievement continues to dog historiometry.

    For example, one of the better known recent applications came out in 2003: Human Accomplishment: The Pursuit of Excellence in the Arts and Sciences, 800 B.C. to 1950 by Charles Murray – he of the infamous Bell Curve. In it, Murray set out to compile humanity’s “resume” by determining all of our “greats” in human endeavors: the Einsteins, the Michelangelos, the Lao Tzus. Murray compiled his list of 4,002 “significant figures” from standard encyclopedias and art histories.

    And he did this by, yes, counting the number of lines each genius received in each book. Then he counted the color plates in each book.

    In short, Murray’s greats are those people who have already been called great by others. The solipsism of this, as you note, is plain. One test of any such scientific system, though, is consistency, and all historiometrists buttress their work with databases and statistical analyses. Yet Murray flatly declared, “I am choosing one type of expertise and rejecting another, allying myself with the classic aesthetic tradition and rejecting” what we might term ‘relativistic modernism.’ Conservative critics loved this, but it did tend to undermine any claims toward objectivity or consistent results; what Murray was making was clearly a political argument, not a scientific case.

    Another test of such a system – especially for cultural critics – is the fineness of the system’s distinctions. Having compiled far too many year-end “top 10” lists, I know that it’s always easy to pick the top two or three spots. Who’s going to argue against Picasso being superior to Leroy Nieman? Or Dante over Carl Sandburg? But it’s the choices toward the bottom that tell you more about the values in play, when it’s often a muddle among three dozen writers kinda-sorta like Sandburg, and you have to pick one.

    In his book, Murray makes an arbitrary 50 percent cut off because there were so many figures cited by only a single source. As beloved as they might be, he argues, James Thurber and Dorothy Parker aren’t up there with John Steinbeck or Theodore Dreiser.

    Perhaps. But this is a difference in genre, not genius. Thurber and Parker were humorists. They weren’t even trying to write major novels. Ergo, historiometry – like the Oscar for best picture – has a definite bias toward the big as being more “profound.” Ditto when it comes to the epic or tragic over the comic. Mahler is valued, Milhaud isn’t. Michelangelo, not Miro.

    This approach also presupposes that all of these artists were, in effect, in one big, revolutionary competition with each other, creating works intentionally to supplant one another in the history books instead of, as with Michelangelo and Miro, working in entirely different directions, in very different traditions, with different aims. Murray would probably have been horrified if he’d realized his process was more or less a pseudo-scientific application of Harold Bloom’s “anxiety of influence” ideas.

Leave a Reply