Lies and auction hammer prices

By Felix Salmon
November 14, 2010
erred in the past, but generally it's good about reporting prices paid at auction: the amount that the winning bidder pays for the item in question is the amount that the item sold for.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

The NYT has erred in the past, but generally it’s good about reporting prices paid at auction: the amount that the winning bidder pays for the item in question is the amount that the item sold for. That amount is then split between the person consigning the item to auction; the auctioneer; and, of course, the taxman. In the case of that Chinese vase, the hammer price was £43 million, the auctioneer charged £8.6 million commission; and the buyer also had to pay 17.5 percent sales tax on the commission, for a total of £53,105,000, or $85.9 million.

Which is why it’s depressing to see the front page of the New York Times today, saying that the vase sold for $69.5 million. (The web story says $70 million in its headline, and leads with the $69.5 million figure; the Guardian, too, gets it wrong, saying the vase sold for £43 million, while the Sun almost gets it right, adding in the commission but forgetting the VAT.)

The point is that the buyer doesn’t care what the hammer price is, only what he has to pay at the end of the day. If you’re willing to pay $86 million for a vase, you’ll pay that regardless of how many different parties get the money, or how that money is divvied up. The auction is mechanism for determining a buyer and a price, it just happens to be full of psychological tricks designed to make the final price as high as possible.

One of those psychological tricks is the hammer price, which is substantially lower than the final price paid: it makes you feel like you’re bidding less than you actually are. There are many other tricks, too, like the flattery and urgency of the auctioneer, the generally feverish atmosphere of the auction room, and the art of getting multi-millionaires with enormous egos to compete with each other to see who has the bigger wad of cash. There’s also the endowment effect, as explained by Don Thompson:

Each bidder starts with a top price in mind. When he momentarily becomes the high bidder, there is an “endowment effect.” He will pay more not to give up the painting, not to lose. Amid the tension of the auction, his reference point has changed to “I should win, this painting should be mine.” He is aware of the regret he would feel at losing what has become “his”.

On top of that, there’s the psychological ratification that comes with the existence of the underbidder and the public nature of the auction price, which is universally considered a benchmark, and is never considered an aberrant rip-off.

In this case, I suspect that other factors conspired to drive up the price, such as the fact that the auction was taking place at a second-tier auction house in the small English town of Ruislip. When would-be Chinese buyers discovered what was being sold there, with an estimate of just £800,000, they thought they were in for a bargain, and they flew in to snap it up.  I’d wager that most of them, being very willing to pay many multiples of the estimate, reckoned that they had a very good chance of walking away with the vase, and got themselves into that psychological state of all but owning it before the auction had even started. Which, of course, is exactly the state that all auctioneers want their bidders to be in — at that point they’re bidding to retain the item, rather than to buy it. And the price they’re paying, psychologically, is their marginal cost — the difference between their new bid and their old bid — rather than the full amount, including that hidden commission.

In any case, when a newspaper reports a hammer prices as though it’s the actual sale price of the artwork in question, it’s giving its imprimatur to a cheap psychological trick. After all, the auctioneer could just as easily say that the bids were for the final price, and that it would then take a percentage of that final price as its own cut. But no auctioneer ever does that, because prices are higher when the total price is opaque.

It’s the job of newspapers, on the other hand, to be transparent, not opaque, about what they’re reporting. So less of this hammer-price nonsense, please. The price paid is the price paid. It’s not — it’s never — the hammer price.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
2 comments so far

Bravo for an accurate summary of the rhetoric of auction behavior!

All that being said, the hammer price is the closest publicly available estimate to what one would yield by selling a similar item. Although still inflated by the seller’s commission (typically 15% to start, but often subject to byzantine formulae), this is the number that is relevant to any reader whose interest is valuation of their own assets.

Posted by earthshiva | Report as abusive

People who have a tough time understanding or get frustrated with the elevation of price after the hammer need to try orther sources for auctions such as penny auctions and reverse auctions, there is no hidden fees when you win a item, you pay the winning price. You should try one I recommend http://www.cra-zee.com they have quality service and products.

Posted by boedean | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/