In Defense of Auction Houses

Felix Salmon submits:

A most peculiar column appeared on the WSJ's editorial page Wednesday, by Daniel Grant. In the context of an art market which is utterly broken, he took aim at the one tiny sliver of transparency therein: the auction houses.

Grant complains of auction-house "secrecy" – reserve prices aren't public, nor are the names of winning bidders – and then comes out with this:

The lack of transparency raises the question of whether prices are artificially high. To what degree is the art market, reaching new heights in the volume of sales and prices, particularly for modern and contemporary art, buoyed on factors that make it noncompetitive?

I simply have no idea what Grant is talking about, here. An "artificially high" price would be one which isn't real: for instance, if a buyer was secretly paid by the auction house to raise his bid. But that isn't going on – if a painting sells for a certain amount at auction, you know that the buyer is paying the whole sum, in cash. And given that the value of a painting is simply whatever someone is willing to pay for it, it's hard to see how any such transaction can be considered "artificially high".

Grant's also quite exercised about the fact that bidders don't know the reserve price, which allows auctioneers to "take bids off the chandelier" below that price in an attempt to create a bit of sale momentum. Again, I see this as a case of no harm, no foul. If you're willing to pay a certain amount then you should bid that much; it really doesn't matter whether you're bidding against a real person or a chandelier, since if you don't bid that much you won't get the painting either way. (OK, it's conceivable that if the painting is passed and reverts to the consignor, you'll be able to do a private deal at below the reserve price. But that's a minor issue, and happens very rarely.)

So what is the problem that all these economists have? Grant quotes three of them, including Nobel laureate George Akerlof, who talks about what happens "if buyers are willing to pay more than a picture is worth" – a concept which assumes some kind of criterion of worth which is separate from what buyers are willing to pay. But I have no idea what that criterion would be.

Grant says that "artificially high prices also make artworks less accessible," which is ridiculous: it's high prices which make artworks less accessible, and whether those prices are "artificial" or not is entirely irrelevant. And he goes on to undercut his own theory when he points out that "more information leads to higher prices overall, as consumers will pay more when they have greater confidence borne of more knowledge". But he then discards that idea, deciding that the art market is better modeled using "the theories around behavioral finance, such as herd behavior, market anchors, greater fools theory, overconfidence, etc."

That's probably true. But even if it is, that doesn't mean the auction houses are doing anything wrong.