Workers flock to cities in pursuit of higher wages, a phenomenon economists call the urban wage premium (UWP). As economist (and City Journal contributing editor) Edward L. Glaeser has explained, cities offer agglomeration effects that boost productivity, and thereby wages, as people and ideas collide. But changes spurred by the pandemic—especially remote work—threaten to undermine this dynamic.
A new paper by Sitian Liu and Yichen Su, “The Effect of Working from Home on the Agglomeration Economies of Cities,” examines employment trends from 2018 through the first five months of 2022, finding that the pandemic-induced switch to remote work has reduced the urban wage premium. Liu and Su analyze wage data from Burning Glass, a company that tracks online job postings and captures about 70 percent of vacancies in the U.S. labor market. Only one-fifth of those postings include wage information, however, so the analysis covers around 14 percent of U.S. job vacancies. The paper also breaks the labor market into segments based on their friendliness to remote work.
The findings are striking. For jobs with high remote-work friendliness, the UWP fell from 6.23 percent before the pandemic to just 3.56 percent in its aftermath. That is, a worker in a big city who was paid 6.23 percent more than a worker doing a similar job elsewhere is now paid only 3.56 percent more—a 43-percent premium reduction.
Changes vary across the labor market. High-skill jobs that can be performed remotely saw a steeper decline than did other jobs. In computer and mathematical fields, for example, the wage premium dropped from 6.58 percent to 3.80 percent, while in food service it fell less, from 4.11 percent to 2.46 percent. In health care, the premium rose from 1.77 percent to 2.49 percent. Many jobs that saw smaller reductions or increases have in-person requirements.
Two factors seem to explain the variation. First, technological advances and the permissibility of remote work have expanded the supply of laptop labor for big-city firms to include people who don’t live in those cities, putting downward pressure on wages. The authors consider this an important explanation for why the drop has been meaningful only for certain jobs. A worker can code from his bedroom, but he can’t roll a burrito or repair a torn labrum there.
The authors also identify a less intuitive but more troubling mechanism. The drop in the wage premium for high-skill, remote-friendly jobs might result not only from expanded labor competition but also from diminishing agglomeration effects and an accompanying productivity slowdown. “We find that the decline in the urban wage premium for jobs with high WFH adoption is led by the decline in the urban wage premiums of relationship-building skills and other skills that are compatible with interactive activities with co-workers, customers, clients, and other professionals, based on intuition,” write Liu and Su. In other words, “the agglomeration economies of large cities were weakened by the adoption of WFH.”
The knowledge spillovers that propel productivity are less common when workers aren’t mingling. High-skill jobs hitherto have delivered outsize economic returns to workers in cities—the 6.58 percent pre-pandemic premium for computer and math jobs far outstripped the figure for any of Liu and Su’s other examples—and have also delivered wider economic growth. Industry clusters, such as in Silicon Valley and Shenzhen for tech and New York and London for finance, are economic engines that drive growth elsewhere. While remote work may help some smaller cities attract white-collar employment (another, more hopeful interpretation of the authors’ findings is simply that those places are getting richer), the potential relative decline of high-productivity clusters nevertheless carries economic risks. Diminishing agglomeration in big cities can reverse the virtuous urbanization circles, leading to further residential and employment exoduses.
To thrive in a remote-work era, cities need somehow to maintain the benefits of agglomeration and curtail the costs and other negatives, such as high crime. If they can do so, we’ll all benefit.
Photo by Gary Hershorn/Getty Images