This paper evaluates empirically the importance of auction format for bidding behavior and seller revenue, focusing in particular on differences in performance under uniform-price and discriminatory-price formats. The analysis is based on a standard benchmark model from which empirically testable hypotheses are derived on the optimal amount of bid shading that generates revenue equivalence between the two formats. More specifically, we apply Vickrey’s (1961) model of shading, following closely the presentation of McAfee and McMillan (1987). Statistically significant shading in excess of the theoretically derived optimum under the discriminatory format suggests greater seller revenue under the uniform-price format.
The model is applied to a data set based on IMF press releases issued after each of the IMF’s gold auctions in 1976-80, These auctions, which were run using both uniform- and discriminatory-price formats, represent a distinct experiment that has apparently escaped rigorous study in the literature. Thirty-five of the 45 auctions--and 10 of the first 20--followed a discriminatory-price format, while the other 10 followed a uniform-price format.
The appropriate choice of auction format is a matter of great practical concern. For example, in the United States, the Treasury currently sells some government securities in uniform-price auctions and others in discriminatory-price auctions, in an attempt to determine which technique provides higher revenues (lower interest costs) to the U.S. Government in auctioning its debt. Mexico has apparently shared those concerns over the comparative performance of discriminatory- and uniform-price formats, as evidenced by a switch from a discriminatory-price to a uniform-price approach in 1990 for its treasury bill auctions, and then, in 1993, by a return to the discriminatory approach. More generally, the uncertainty with regard to the “best” auction technique is readily apparent in the prevalence of both uniform- and discriminatory-price setups for auctioning similar items; practical advice on auction choice from empirical study is rather limited.
From the paper’s findings, it is concluded that Vickrey’s benchmark model offers useful and empirically valid insights into bidding behavior. What is particularly relevant and apparent in this regard is that auction participants do, in fact, shade their bids under a discriminatory-price format, as the basic model would suggest. At least as important, the paper also provides statistically significant evidence that the extent of this bid shading is, if anything, even larger than this model would indicate, pointing to the superior revenue-generating properties of uniform-price over discriminatory-price auctions.