4 - Experimental Auction Markets and the Walrasian Hypothesis

Summary

This study reports on a block of experimental market sessions designed primarily to provide (1) the severest test yet attempted of the equilibrating forces operating in competitive auction markets and (2) a more rigorously controlled test of the Walrasian hypothesis. Some data are also supplied which show the effect of cash payoffs on the equilibrating behavior of such markets; in particular, the effect of full cash payoffs to all successful trading subjects as against payoffs to a subset of such subjects chosen at random.

EXPERIMENTAL DESIGN AND SUBJECTS

The supply and demand conditions underlying the experimental design in this study were intentionally unconventional. In each experimental session, each of eleven subject buyers could purchase at most one unit of the fictitious commodity per trading period at a price not to exceed the limit price $4.20. Therefore, the demand per unit of time, or trading period, was perfectly elastic at $4.20 up to the maximum demand quantity of eleven units. In each session each subject seller could sell at most one unit of the commodity at any price not below the given minimum reservation price $3.10. There were thirteen such sellers in two experimental sessions, sixteen in two additional experimental sessions, and nineteen in the final two markets. Therefore, the supply per trading period was perfectly elastic up to the maximum supply quantities of thirteen, sixteen, and nineteen units, respectively, in the three experimental treatments.