Profiting from Asymmetrically Dominated Alternatives: The Case of Online Diamond Pricing

Nov 24 2015

Download Here

  • Revise and Resubmit at Journal of Marketing Research.
  • With Koray Cosguner


The asymmetric dominance effect, first documented by Huber et al. (1982), has been robustly documented in various lab experiments in the literature. However, the practical validity and significance in a real marketplace has never been verified. In this paper, we empirically test the existence of this effect and, more importantly, quantify its magnitude using a unique panel data set from a major online jewelry market. We estimate a proportional hazard model by carefully incorporating the underlying dominance relationships among every diamond pair in our data set. Our model estimates suggest that in general consumers have low awareness about the existence of the decoys in the market place; however once they discover, the decoy diamonds significantly increase sales likelihoods of dominant diamonds, and this effect becomes stronger for diamonds in higher price ranges. We further quantify the overall profit impact of selling decoys and decompose this impact into two effects: the information effect that comes from selling to uninformed (about the existence of decoys) consumers; and the decoy effect that comes from selling to informed consumers.We find that the existing decoy pricing in the diamond market improves the retailer's overall gross profit by 11.85%, compared to a uniform pricing strategy. This profit contribution is driven largely by the decoy effect rather than by the information effect. Finally, we explore various strategies that the retailer can adopt to improve its profitability through further utilizing the asymmetric dominance effect. We find that there is a great potential for the retailer to gain additional profits.

Keywords: Asymmetric Dominance Effect, Proportional Hazard Model, Diamond Pricing, Context Dependent Choice