International Law

Winner's Curse? Seth Godin on Auctions

Image via Wikipedia

Seth Godin has wondered if the the winner of auctions usually bid irrationally high or too much. 

"The only thing worse than losing a big-time auction is winning one.

If you win, you feel like a chump because everyone else in the world dropped out before you.

Which is why, if you ever sell something big at auction, you need to bend over backwards to pleasantly surprise the purchaser.

It's the only way to overcome auction-buyer's remorse."

The economics behind action-buyer's was researched almost 25 years ago. In 1992, Richard Thaler re-published a series of articles about anomalies in economics. (Thaler published his article entitled the "Winner's Curse" in 1988.)

In brief, Thaler's point can be understood as follows.

Why does buyer's regret exist? Why do the winner's of auctions or mergers feel that they have overpaid in the end?

Thaler had an interesting observation.

Consider the following card game.

Player A is dealt one card, from ten cards -the ace to the ten. Player B must make one bid for A's card.

The value to B of winning A's card is $10 x 150% of the face value of the card.

The value to A is just $10 x the face value of the card.

So, for example, a winning bid for "card 6" gives B a return of $90 and A a return of $60. Finally, if A shows his card to B, B may take the card without paying for it.

What should B on average bid for A's card? Well, you might think that since on average the return is between $15 and $150, then B should bid around $60 or $65.

However, Thaler points out that as the game is set up, B should never bid! Why? Well, any bid by B that is accepted by A must be for more than the card is worth, otherwise A would not accept the bid.

For example, suppose A accepts a bid of $30, then his card must be a 1 or 2, which on average produces ($15 + $30)/2 return, or $22.50, to B.

Eventually, B should go broke playing the game. This is a simple two person demonstration of the winner's curse, and the same result holds for larger auctions.

There is a vey nice applet by Mike Shor which demontrates this at gametheory.net

Thaler argues that rational individuals would not play this auction.

But I believe that Thaler is wrong and this is a completely unsatisfactory analysis, as far as it goes.

The possible joint gain is always positive not matter what A is dealt. For example, suppose A is dealt the 4, then A could get $50 = $40 + $10 and B , if B could be convinced to bid exactly 4 and split the surplus equally, $60 - $40, with A.

B can achieve $10 by coordinating with A. Can they ever reach this coordinated result? If they don't, A will get $40 and B nothing.

I don't know if they can, but I do know that social and economic life is as much about coordination as it is about competition, the reason we have four way stops.

So if you were A holding the 4, how would you respond to a bid by B for $60?

Take the money and run, or point out that the "price tag" on the card is only $50 and do the deal at the lower price? If there is a price tag, then both parties have the chance of reaching the cooridinated outcome, otherwise they may well miss out out on the joint gain.

I suspect that there is a huge overall advantage to an economy which largely employs the take it or leave pricing strategy, over the let's bid or auction on everything at different prices.

Is this a clash of economic civilizations?

From the WSJ, on June 2, 2009 about fish market auctions.

"White shoppers would fare better in a fish market with fixed and publicly displayed prices, Graddy and Hall conclude.

Here "the inelastic white customers pay lower prices, purchase larger quantities of fish, and make more frequent purchases." On the flip side, the more elastic Asian buyers would end up paying more and buying less.

So how could New York's fish shoppers be in the dark for so long over this? Easy, the report says, they rarely mix with each other."

Full post as published by Misleading Advertising Law on June 09, 2009 (boomark / email).