Competing Against Simulated Equilibrium Price Dispersions: An Experiment On Internet-Assisted Search Markets

In a four-treatment experiment, we test some of the hypotheses in García-Gallego et al. (2004) concerning competition among a number of firms of which some (or all) are indexed by a price-comparison engine facilitating buyers’ search process. In this paper, we isolate individual behavior from noise due to other players’ actions and learning, facing each subject with simulated rivals whose prices are extracted from mixed strategy equilibrium distributions. We find systematic deviations from both theoretical distributions and previous data obtained in sessions where all players were human. Specifically, departures of experimental data from the corresponding theoretical predictions are enhanced in this setting as compared to our previous research in which all agents were represented by human players. This suggests that the divergence between theoretical and observed price reported there should not be attributed to noisy learning and strategic uncertainty due to subjects’ interaction with other players. Furthermore, economic tests on players’ risk attitudes organize pricing behavior in meaningful, although not always compatible, ways.