While workers and employers reap many benefits from work-from-home arrangements, coming into the office also has advantages, particularly for innovation-based occupations where the exchange of ideas and formation of professional networks are important.
These types of jobs are typically concentrated in large cities, where workers’ productivity is the highest. Our research shows that since the start of the pandemic, the reduced presence of in-office workers has reduced the wage premium attached to large cities and may have diminished the productivity edge of these areas.
Forming Networks, Relationships in the Workplace
A large body of research has shown that the interactive activities at a workplace facilitate the exchange of new ideas and accelerate the formation of professional networks and business relationships. The concentration of firms and workers in large cities naturally supports these interactive activities.
The ability to easily exchange ideas and form new relationships is key to why employers in larger cities are relatively more productive and can offer higher wages than their counterparts in smaller communities. This is especially true for firms in which innovation plays an outsized role.
The COVID-19 pandemic forced large numbers of workers to adapt to working from home, emptying out previously dynamic central business districts and industry clusters. As many as half of total hours worked in the United States were expended from home at some point in 2020 after the outbreak.
For individuals, working from home often means reduced commuting costs and increased time spent with family and friends. For employers, research shows that the virtual mode of work has a positive and higher-than-expected effect on workers’ productivity. Employers also can access a much larger talent pool nationwide thanks to remote technology.
However, there is a major disadvantage of working from home. Spontaneous workplace interactions are significantly reduced. “Water cooler talk” doesn’t occur, reducing the intensity of knowledge exchange and slowing the building of new relationships and professional networks, which may be why some employers have started asking employees to return to offices.
Since large cities and industry clusters are places where these interactive activities occurred before the pandemic, work from home likely reduced these locals’ productivity premium the most. We can’t observe labor productivity directly; therefore, we assess it using wages in job postings and find that the wage premium offered in large cities fell during the pandemic due mostly to the rise of working from home.
Chart 1 compares wage offerings by employers in two different occupations: high-work-from-home occupations (finance and computer-related) and low-work-from-home occupations (local service jobs that include food service, personal care, construction and maintenance/repair).
If remote work reduces large cities’ productivity premium, wages for high-work-from-home occupations in large cities should grow less compared with wages offered in small cities. Meanwhile, among low-work-from-home occupations, the relative wages in large and small cities should not change markedly.
Accounting for a Diminished Large-City Wage Premium
Large cities offered a sizable wage premium in 2019 for the finance and computer occupations; that premium was significantly reduced by 2021, as Chart 1 shows. In contrast, local service workers experienced a small but largely symmetric increase in wage offerings across meters of all sizes, leaving the wage premium of large cities largely unchanged.
While the diminished large-city wage premium is consistent with the hypothesis of declining relative productivity, there is an alternative explanation—large-city employers could now hire nationally, with the help of remote-work technologies, without incurring the costs of worker relocation and housing. These firms, drawing on the increased pool of accessible labor, could offer a reduced wage to acquire talent.
How do we know that the decreased large-city wage premium is not primarily driven by the increased labor supply but by a reduced productivity premium?
Presumably, if productivity in large cities did not decrease, the newfound ability to hire remotely should enable employers in large cities to increase hiring in high-work-from-home occupations, resulting in job gains. Such a pattern would sharply contrast with the substantial job losses in large cities’ local service sectors (a byproduct of diminished demand due to outmigration).
However, data show that employment in high-work-from-home occupations declined more in large cities (based on employers’ locations) than it did in small cities during the pandemic (Chart 2).
This means that even with the ability to hire nationally, firms in large cities still hired fewer workers, implying that productivity in large cities must have decreased because of work from home.
How can we tell whether such a relative productivity decline in large cities can be attributed to the reduction in the interactive activities in large cities?
The job posting data provide detailed information on skills required for each job, which allows researchers to dissect each position’s total wage compensation into the wage rewards for the individual skills needed for that job. As a result, we can identify the skills experiencing a large reduction in the wage premium.
Our research finds that the most important factor explaining the drop in large cities’ wage premium among the professions with high-work-from-home adoption is diminished rewards for “relationship-building” and “marketing and public relations” skills. In other words, an important reason for the big decline in finance and computer jobs’ large-city premium is that the wage reward for relationship-building and marketing skills fell much more in large cities relative to small cities.
Productive Interaction Lost with Working from Home
Since relationship-cultivating skills are especially useful in interactive environments, the declining wage return to these skills in large cities suggests fewer interactive activities in firms based in large cities.
Research has shown that such interactive activities at the workplace are the crucial breeding ground for knowledge exchange and network building and are more likely to occur in large cities or industry clusters.
Thus, the decline of these activities and the falling wage reward for relationship-cultivating skills used in these activities in large cities provide evidence that the productivity premium generated by dynamic interactions among workers and innovators is likely in decline due to the widespread adoption of working from home.
About the Authors
The views expressed are those of the authors and should not be attributed to the Federal Reserve Bank of Dallas or the Federal Reserve System.