Recent IT research has analyzed how the performance of IT-enabled markets may differ from
conventional markets. This literature has made two unexpected empirical findings. First, ITenabled
markets for commodity goods exhibit significant price dispersion. Second, well-known
retailers in these markets appear to cooperate to set high prices.
This paper presents an analytic model, and confirmatory empirical evidence, that explains this
behavior as a response to the unique characteristics of consumer search in electronic markets. In
conventional markets, consumer search costs are primarily a function of the consumer’s physical
proximity to retailer outlets — and physical proximity is distributed relatively equally across
retailers. In electronic markets, consumer search costs are primarily a function of the consumer’s
mental awareness of different retailers — and this awareness is likely to be concentrated in the
hands of a few retailers.
Based on this model of consumer search, IT-enabled markets for commodity goods exhibit high
price dispersion in equilibrium and a few well-known retailers are able to cooperate to set high
prices. The predictions of the model are shown to be consistent with empirical data for 23,744
books collected from 24 Internet retailers in late 1999. Viewing consumer search in this manner
provides a useful starting point for understanding the likely development of IT-enabled markets,
and for understanding the importance of advertising and first-mover advantage for electronic
market participants.