Abstract:
The theory of competition as a discovery procedure argues that the results of this procedure cannot be predicted because of its very nature as a discovery procedure. If this were true, this would imply the impossibility of testing whether competition actually works as predicted. This article rejects this view. We argue that outcomes of the discovery procedure can be predicted if a pattern rather than its specific realization is the subject of the prediction. We de-scribe a double oral auction as a classroom experiment. and show that competition is likely to discover the market equilibrium as well as the wealth maximizing number of transactions.