Beyond BLS briefly summarizes articles, reports, working papers, and other works published outside BLS on broad topics of interest to MLR readers.
Over the last two decades, the U.S. retail landscape has been shaped by the steady encroachment of e-commerce on traditional retail, a process largely propelled by technological innovation. At the turn of the 21st century, e-commerce accounted for less than 1 percent of total retail sales in the United States, but by 2021, that share had reached nearly 15 percent. Given this marked shift in a major sector of the U.S. economy, researchers have increasingly turned their attention to examining the impacts of e-commerce on the economic behavior of brick-and-mortar retail establishments and the labor market experiences of their workers.
This task is taken up in a recent article, “Creative destruction? Impact of e-commerce on the retail sector” (National Bureau of Economic Research, Working Paper 30077, May 2022), by Sudheer Chava, Alexander Oettl, Manpreet Singh, and Linghang Zeng. The authors surmise that, in coping with e-commerce competition, traditional retailers may adopt two starkly different, yet both rational, strategies. In one scenario, these retailers may choose to hire more workers, at higher pay, in an effort to improve the shopping experience of their customers and lure more of them through their doors. In another, less-labor-friendly scenario, traditional retailers may adopt the opposite strategy, choosing to cut their operating costs by laying off workers, reducing their hours and compensation, or closing up shop.
To assess how these possibilities hold up empirically, Chava et al. focus on a large U.S. e-commerce retailer, using its geographically diversified rollout of fulfillment centers (FCs) as a measure of e-commerce at the local level. This information, which goes back to 2000, is analyzed in conjunction with credit-bureau payroll data for 2.6 million retail workers, retail establishment sales data from the National Establishments Time-Series (NETS) database, and county-level employment data from the U.S. Bureau of Labor Statistics Quarterly Census of Employment and Wages (QCEW).
The authors’ results point to traditional retail in retreat. With respect to labor impacts, the analysis indicates that the introduction of an FC in a county has a sizable negative effect on wages, leading to a 2.4-percent drop in the income of those employed in local retail, primarily through employer cuts to work hours. However, this effect is geographically bounded, disappearing at distances larger than 100 miles from FCs, and stronger among part-time hourly employees and those with less experience on the job. The authors’ analysis of QCEW data also shows that the presence of FCs in a county reduces that county’s employment growth in traditional retail by nearly 3 percent, and this decline is not offset by a compensating reallocation of labor to FCs or other retail-related sectors, such as transportation.
Similar e-commerce impacts transpire at the establishment level. Using NETS data, Chava et al. find that, after the introduction of an FC in a county, local retail stores suffer an average sales loss of 4 percent, a revenue hit that forces them to cut payroll by about 2 percent. Moreover, the probability of an establishment going out of business in the presence of FCs rises by an estimated 22 percent, with that risk being most pronounced among small and newly opened stores. The authors find no evidence that these results are an artifact of negative selection in the choice of FC locations; rather, the decline in traditional retail in areas affected by e-commerce seems to be driven mostly by the delivery efficiencies offered by FC shipping networks.