The Impact of a System of Intelligent Agents on a Market at Equilibrium

Bandwidth gains achieved in telecommunication technology have been followed by a recent proliferation of Intelligent Agents on the internet. Intelligent agents have emerged as mechanisms that aid both buyers and sellers in very different ways. It has been observed by several researchers that electronic search mechanisms promote "frictionless" competitive markets that lower buyers' search costs and seller profits (e.g., see Bakos 1987, 1991, 1997; Malone, Benjamin and Yates 1991). While some research has been done on buyer search for sellers, there are very few formal models of seller search for buyers. Intelligent agents (IAs) facilitate seller search for buyers. IA's also work for buyers and reduce the aggregate disutility of a purchase decision (price, search and comparison related costs). These IA's are called Decision Agents [see Do, O. et al]. Demand agents on the other hand, work for sellers and perform two important functions; first they infer buyer preferences and estimate buyer reservation prices and second, they supply decision agents with product information. Demand agents infer buyer preferences based on the characteristics of the buyers' context such as site demographics, buyer demographics and by employing interactive information collation techniques such as collaborative filtering, observed buyer behavior modification and analysis of self reported data [see Do, O. et al].