Document Type


Publication Date



This article offers a brief survey of bidding theory in high price auctions, of experimental studies of behavior in such auctions, and of the interplay between the design and results of the experiments and efforts to further develop the theory. Two new series of experiments are reported. The first applies a convex transformation of payoffs in an attempt to induce a lowering of subject bids "as if" the bidders had become less risk averse. The second applies a method for inducing any prespecified utility function for risky choices on an individual. It is used to induce "as if" risk-neutral behavior. Both series use baseline control to "calibrate" the hypothesized effect of the procedures on "risk-averse" behavior. A close examination of individual bidding suggests that subjects bid less only when the profit potential is below the 50 cent "break-even" level. Above this 50 cent potential profit level, subjects tend to bid relatively higher. This can be interpreted as a type of "satisficing" behavior in which subjects attempt to do at least as well under the quadratic transformation as in the baseline experiments.


Originally published in:

Cox, James C., Vernon L. Smith, and James M. Walker. 1985. “Experimental Development of Sealed-bid Auction Theory; Calibrating Controls for Risk Aversion”. The American Economic Review 75 (2): 160–65.

(c) American Economic Association; posted with the permission of the publisher.

Included in

Economics Commons