Consumer Valuation of Network Convenience: Evidence from the Banking Industry

This paper develops and estimates a structural demand model which incorporates consumer preferences for geographical convenience of retailers’ networks. Unlike the previous literature, which assumes consumers only care about retailer’s outlet located closest to their home locations, our model explicitly allows consumers to consider two outlets of a retailer located nearest their home and work locations, respectively. We apply our model to the U.S. retail banking industry. To estimate our model, we use the U.S. Journey to Work dataset to obtain consumers’ home and work locations in the population, and combine it with a dataset that details each branch’s deposit and location of 132 isolated cities. The results show that consumers value the proximity of outlets to home and work locations roughly equally. In the counterfactual experiments, we evaluate the impact of various Work From Home policies on how consumers value a bank’s branch network. In the baseline counterfactual experiment, where everyone is assumed to work from home, the market shares of banks specializing in serving work locations drop by 0.7 percentage point (which is 8.6% of the mean market share in the data).