The Effects of Information Costs on Search and Convergence in Experimental Markets

Nearly all markets contain some kind of friction, making it difficult to reach full efficiency. One ubiquitous source of market friction is the cost of obtaining pricing information. We determine how market performance as well as buyer and seller behavior are affected by the introduction of price information costs in experimental posted markets. Rothschild [Journal of Political Economy (1973) 1283; Journal of Political Economy (1974) 689] theorizes that search is negatively related to knowledge of the price dispersion and the cost of search, and positively related to market price dispersion. We find that market knowledge and the cost of search itself affect search, but we find no evidence supporting the role of price dispersion in search decisions. We also find evidence supporting Smith and Plott [Review of Economic Studies (1979) 133] and Walker and Williams (1988), both of which show price convergence comes from below in posted-bid markets and from above in posted-offer markets. High information costs tend to reduce or eliminate convergence. Average prices are above the market-clearing price for posted-offer markets, and below for posted-bid markets, and never cross the market-clearing threshold, also consistent with Walker and Williams. More generally, our results support the notion of symmetry between experimental posted-offer and posted-bid markets, broadening the relevance of experimental search research