Abstract
A series of five first price sealed bid auctions of a single unit were conducted using experimental techniques. The results of the experiments were analyzed to determine the effect on bidding behavior due to the change in market size and from the reporting of additional market information to the participants. The analysis indicates that the percentage of resale value bid increases as market size increases. The type of additional information employed in these experiments had no discernible effect on bidding behavior. The individuals in the markets appeared to be bidding in a way consistent with risk averse preferences. Although theoretically the level of risk aversion should be constant across different market sizes, the analysis shows that in fact, the measure for risk aversion produced dramatically different estimates across the different market sizes. There is some indication that there is adjustment in bidding behavior occurring over time. It appears that this is due simply to the passage of time and is not occurring because of information acquisition. The experiments were conducted by using a technique in which each subject entered a bid in two markets of different sizes each period. The dual market technique seems to work rather well except that there is some indication that the smaller sub-markets are not being maintained as independent markets even though theoretically they should be. Although the evidence is not overpowering, the indication is that the dual market technique may need to be revised.
Kogut, Carl A. (1984). Individual and market bidding behavior in vickrey first price auctions. Texas A&M University. Texas A&M University. Libraries. Available electronically from
https : / /hdl .handle .net /1969 .1 /DISSERTATIONS -592374.