Default or Dare: The Default Effect on Consumer Choice, Assortment, Pricing, and Estimation

Abstract

The default effect, the tendency of individuals naturally to pay more attention to the default option than alternatives, is commonly utilized in influencing consumer choice behavior. In this paper, we incorporate the default effect into the classic multinomial logit choice model and study the associated default selection, assortment, and pricing problems. We characterize the default effect by multiplying an enhancement factor with the attractiveness of a selected product. Interestingly, setting a default option does not always increase the total expected profit, while optimally setting the default option guarantees an improvement of the total expected profit. We find that the optimal selection of the default option depends on the product markup and attractiveness, as well as the magnitude of the default effect. Moreover, jointly optimizing the default option and assortment shrinks the size of the optimal assortment while jointly optimizing the default option and prices improves the markups of all products, compared to those without considering the default effect. In the extensions, we first consider selecting multiple default options with different enhancement factors and characterize the optimal selection of the default options to maximize the total expected profit. Second, we find that to maximize consumer surplus, the optimal selection of the default options differs from that of maximizing the total expected profit and identify the conditions under which the optimal sequence of default selections is the same for both the firm and consumers. Third, we incorporate product fairness into the default optimization problem and develop an algorithm to efficiently find the optimal default option subject to the fairness constraint. Fourth, we characterize the default effect by an additive enhancement. Finally, we develop a maximum likelihood estimation method and use synthetic data to validate the proposed model and illustrate how to mitigate estimation bias, regardless of whether the default option is observed or not in the data.

Keywords: Default Effect; Assortment Optimization; Pricing; Consumer Surplus; Product Fairness