INDUCING RISK-NEUTRAL PREFERENCES - FURTHER ANALYSIS OF THE DATA

The lottery payoff procedure does not successfully induce risk-neutral bidding behavior in first-price, sealed-bid auctions. This conclusion follows from both ordinary-least-squares estimation with natural data and least-absolute-deviation estimation with transformed data from nu merous experimental designs. Lottery payoffs do not succeed in inducin g behavior predicted from standard expected utility theory assumptions or from assumed utility from winning and/or income thresholds. In con trast, first-price auction experiments with monetary payoffs yield res ults that are consistent with general models of bidding in the indepen dent private values information environment.