Distributional Impacts of the Changing Retail Landscape

The last decade has seen a transformation of the U.S. retail landscape, with the press energetically asserting that exits of large retail chains constitute a “retail apocalypse.” This raises the possibility that some households are now facing substantially diminished access to retail opportunities. While much has been written about the transition from manufacturing jobs to new economy jobs having “left behind” a cohort of dislocated workers, there has been less focus on how the transition from traditional brick and mortar retail to the current physical/online retail mix may have left behind a cohort of dislocated shoppers.

This paper begins by documenting new facts on the changing retail landscape in the general merchandise sector during the 2010-2019 period. Despite claims of a retail apocalypse, we show that the number of general merchandise stores has actually increased over the decade from 2010 to 2019 and the increase has been concentrated among large national chains. In particular, this growth has been driven by the near-doubling of dollar store outlets such as Dollar General, Family Dollar, and Dollar Tree. In addition, we find substantial growth among discount department stores and a concurrent decline of traditional department stores and regional and independent retailers.

We use smartphone geolocation data for over four million smartphone devices to estimate a model of consumer shopping behavior that allows us to disentangle preference effects from proximity effects and we quantify the welfare impact of the change in the general merchandise sector over the 2010-2019 period. Combining our demand estimates with historical data on retail locations, we estimate the welfare accruing to consumers of different incomes from the change in general merchandise retail chains over the last decade. The welfare calculation allow us to decompose the welfare changes due to consumer migration, national chain entry and exit, and the entry and exit of small regional chains. Our results suggest that, if all consumers remained at their 2010 locations, the lowest income consumers benefited slightly more than other in- come groups from the changes in general merchandise opportunities over the decade. The welfare gain attributable to the expansion of discount department stores, super centers, and dollar stores outweigh the welfare losses attributable to the exit of smaller regional chains and traditional department stores. Given recent findings that welfare improvements from access to online retail have been substantially concentrated in higher-income consumers, our welfare results alleviate some concerns of widening income consumption inequality in the era of e-commerce. The positive impacts of the changing retail environment are counteracted by migration, as both the lowest and highest income consumers tended to move farther from retail opportunities over the course of the decade.