In this paper, we investigate the demand response of a firm’s existing customers to retail store entry and which aspects of the response are demand-expanding and which cannibalize online sales. The empirical goal of the paper is to separately identify the effects of customer-to-store distance on purchase frequency, channel choice, and expenditure per purchase, using transaction-level data from a multichannel apparel brand. Our identification strategy exploits within-customer variation in distance resulting from store entry during a period of rapid retail expansion. We establish retail distance effects using descriptive regressions before developing a unified structural model that affords rich counterfactual analyses and further controls for product category preferences. We find material effects of decreasing retail store distance on purchase frequency and retail channel choice but not expenditures per purchase. Our structural model ascribes mechanisms to these effects in the form of increased brand consideration, higher retail utility from nonmonetary factors (e.g., reduced driving times), and infungibility of monetary transportation costs with budgets for apparel. Our estimates imply that a 10% reduction in retail store distance increases retail channel expenditures by 1.9% and decreases online channel expenditures by 1.2%, resulting in a 0.4% increase in total expenditures. Through counterfactual experiments, we demonstrate that retail expansion can ultimately limit the ability of the firm to price discriminate across channels because reducing transportation costs weakens the firm’s ability to enforce channel-based segmentation schemes. This paper was accepted by Matthew Shum, marketing. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mnsc.2022.4308 .
Read full abstract