Do Shopbots and Lower Search Costs improve the Efficiency of Electronic Markets? An Agent-based Approach

Buyers' search behaviors on electronic markets are characterized by two distinctive features: i) diminishing search costs and ii) the use of informational intermediaries such as price comparison agents (shopbots). We build a simple agent-based model that captures these two features, and we use this agent-based modeling to explore how the co-evolution of buyers and sellers (i.e. the joint dynamics of learning) may affect market's efficiency. First, we show the ambiguous role of shopbots on the efficiency of such markets: although the use of shopbots is frequently assumed to induce positive effects ­- by enlarging buyers' information space and so enhance price comparisons - they have a destabilizing role on buyers' and sellers' learning co-evolution. This may result either in the emergence of a dispersed distribution of prices (commonly noticed by empirical studies) or in some market' crashes. Second, we study the impact of a decrease in search costs. We also point to the existence of a non deterministic relationship between market's efficiency and search costs. This suggests that the use of a priori more efficient matching technologies do not necessarily lead to more efficient outcomes.