The Winner's Curse
"The auction winner is the one who most overestimated the prize, and so overpaid."
"The auction winner is the one who most overestimated the prize, and so overpaid."
In a common-value auction under uncertainty, where the item is worth roughly the same to everyone but nobody knows exactly what, the winning bidder tends to overpay. The winner is disproportionately likely to be the bidder who most overestimated the value, so winning is itself bad news about your own estimate. A rational bidder must therefore shade their bid below their raw estimate to correct for this.
Rated Empirical: grounded in studies or measurable evidence. The other 608.
Suppose a single unknown value is common to all bidders, and each forms an independent, error-laden estimate of it. Even if every estimate is unbiased on its own, the highest of several estimates is biased upward: the maximum of noisy guesses tends to overshoot the truth. A first-price auction hands the item to roughly the highest estimator, so the winner's estimate was probably too high. The more bidders, the worse the overshoot. The rational response is to condition on the very event of winning, which lowers the value you should expect, and to bid below your estimate accordingly.
The rational correction: bidding strictly below your raw estimate, by an amount that grows with the number of competitors and the noise in your signal, so that in equilibrium expected profit is not negative.
The curse arises only in common-value auctions, where the item is worth the same to all; in pure private-value auctions, where each bidder has an idiosyncratic valuation, winning conveys no bad news and the curse does not apply.
It is a foundational result of auction theory and a sharp lesson in conditional reasoning: that winning a competitive bid is informative, and usually bad news, about your own valuation. It explains chronic overpayment across oil leases, takeovers, spectrum sales, and sports contracts, and it shaped the design of real auctions, work recognised by the 2020 Nobel in economics.
The curse requires both a common-value component and genuine uncertainty about that value; it does not apply to pure private-value auctions, and it weakens as uncertainty shrinks. Laboratory studies find even experienced bidders fall prey to it — construction executives did about as badly as students, apparently because their field rules of thumb do not transfer to the lab, so whether experience cures it is mixed. And its magnitude in real markets is contested and hard to identify cleanly, being far better documented in the laboratory than in the field.
The winner's curse is not the same as merely overpaying or feeling buyer's remorse. It is specifically the statistical selection effect that winning correlates with having the most over-optimistic estimate. It is not simple irrationality either: a bidder who shades correctly avoids it, and only one who fails to condition on winning suffers it. It does not apply to private-value auctions, where overbidding, when it occurs, is a different phenomenon. And the winner does not always literally lose money; the claim is a tendency toward overpayment on average.
Yes, as far as the evidence goes. The Winner's Curse is rated Empirical here: it rests on studies or measurements rather than on a saying.
Rated from three sources. 609 of the 1,116 entries carry this rating. How the scale works.
The best-known laws are the least likely to rest on measurement.
Described by three petroleum engineers, Capen, Clapp, and Campbell, in 'Competitive Bidding in High-Risk Situations' (Journal of Petroleum Technology, 1971) from oil-lease bidding; formalised in auction theory by Wilson and by Milgrom and Weber, and popularised by Richard Thaler.
Everything above traces to onepetro.org, aeaweb.org and cs.princeton.edu: the publications themselves, not accounts of them. Nothing is written from memory, and the method is written down. Spot an error or a better source? Suggest a fix.