How Much Did Store Closures Boost Online Sales During COVID-19?

This paper examines the impact of retail store closures on omnichannel sales and consumer shopping behavior during the COVID-19 pandemic. To explain the likelihood of store closure, we develop a novel instrumental variable motivated by the varying geopolitical response across the US to the pandemic. Using data from a luxury fashion retailer, we find that when a store is closed, the volume of online orders originating from its location increases by 29%. Furthermore, omnichannel total sales (offline + online) decrease by 5.5% if the retailer closes 10% of the stores. We find that the online channel enables the retailer to salvage 25% of the offline sales that would have been lost due to store closure. We also show that new e-shoppers exhibit different online behaviors as compared to existing e-shoppers. First, they are more likely to order popular product models to mitigate the higher mismatch risk they face. The likelihood of ordering a popular model is 72% for a new e-shopper, whereas it is 45% for an existing online consumer. Second, the conservative behavior of favoring popular models negatively affects the likelihood of a return by new e-shoppers.