Long before Andy Warhol, Théodore Duret established the principle that prices provide evidence of artistic success. What is disappointing, however, is how poor the quality of the art world’s economic discourse remains even in the post-Warhol era. It continues to be fashionable among many critics and scholars to claim that art markets are irrational, and that prices have no value as indicators of artistic importance. These claims are both ignorant and foolish. Art scholars must overcome their distaste for economics (emphasis mine), and become more sophisticated in examining how changes in art markets have influenced artists’ attitudes and behavior.
This is from a new NBER paper by David Galenson.
The problem with valuing art is that art is not for sale: it's non-rivarlous and non-excludable. Art relates purely to the idea, not the material. Any half-decent colour printer can churn out an exact copy of the Mona Lisa (let alone the Warhol images), and an exact replica of the Aphrodite of Milos will cost you next to nothing compared to what the Louvre would ask for the original. And I won't even go to Duchamp's urinal.
So what do collectors pay for? Mainly, they pay for symbolism. And symbolism is politics, not art. An object's price is a very weak predictor of its artistic value.
This is not to say that art and economics are orthogonal to each other. Prices dramatically 'influence the attitudes and behaviors of artists' - and thus the art of the present and of the future - and deserve to be taken as seriously by art scholars as by the artists' themselves.
Anti-materialistic art scholars 1, Galenson 1.