Amid the Covid-19 global pandemic, remote work (or telework) has gone from an optional perk to an essential capability. While remote work’s short-term benefits are obvious, there is also a long-term case to be made for it that does not depend merely on its role in reducing the spread of disease. Remote work should not necessarily be set aside once the pandemic subsides.
Last May, the payments-processing firm Stripe announced that it would staff its fifth engineering hub entirely with remote workers. The company justified the decision by noting that “the substrate of collaboration has gotten shockingly good over the last decade” and that “while we did not initially plan to make hiring remotes a huge part of our engineering efforts, our remote employees have outperformed all expectations.” Stripe isn’t the only company impressed by off-site workers. In a two-year internal study, Google concluded that the performance of remote and co-located teams showed no significant differences. In February 2020, Twitter CEO Jack Dorsey declared: “Our concentration in San Francisco is not serving us any longer, and we will strive to be a far more distributed workforce, which we will use to improve our execution.” A 2019 study at the U.S. Patent and Trademark Office found remote patent examiners more productive than co-located ones. And a study of Portuguese research-and-development firms from 2011 to 2016 found that enabling remote work increased labor productivity.
Change is clearly under way. Between 1960 and 2000, the share of non-farmworkers working full-time from home never topped 3.5 percent. Since 2000, their share has risen steadily, to 5 percent in 2017—and the gain is almost entirely explained by remote employees, not self-employed entrepreneurs working out of their homes. Strikingly, the share of remotes is even greater for higher income brackets: 7.9 percent of those earning more than $75,000 yearly work off-site. As economist Adam Ozimek argues, moreover, this is a lower estimate on the number of remotes, since it excludes workers in satellite offices, co-working spaces, coffee shops, and other alternatives. A 2019 survey commissioned by Upwork and the Freelancers Union found that including such alternative work sites raises the share of remotes to roughly one in ten; including those who work remotely some of the time increases the percentage to about one in three.
These trends suggest that it may be time to take another look at the “death of distance”—the notion, influential during the 1990s, that the Internet renders geographical location increasingly irrelevant. This idea lost traction during the 2000s, as it became obvious that the productivity benefits of cities were growing instead of declining. Innovation and other “knowledge” work are collaborative phenomena, drawing on social dynamics that have benefited from physical proximity. As the value of knowledge work rose, so, too, did the importance of being near one’s current and potential collaborators; hence, we see the flourishing of big American cities like New York. But that development now seems to be flagging. While the share of Americans living in cities of more than 50,000 people has continued to rise—from 68 percent in 2000 to 71 percent in 2016—America’s largest cities (New York, Los Angeles, and Chicago) have started to lose population over the last few years.
What’s driving the long-delayed emergence of remote work? Most obviously, the technology supporting it has improved. Perhaps equally important: we’re getting better at using the Internet, inventing new social and cultural methods of interacting on it. Comfort with working online is, unsurprisingly, strongest among the young. An Upwork survey found that 40 percent of 18- to 34-year-old small-business principals planned to hire full-time remote workers, compared with just 10 percent of principals aged 50 and older. Younger workers are also more likely to express interest in remote work. As Ozimek observes, even without further technological advances, remote work seems set to expand, simply for demographic reasons.
Co-locating employees for “intangible” or knowledge work has traditionally made sense for firms. Such work requires extensive internal communication, and face-to-face exchanges have made that easier. Face-to-face meetings enable one employee to transmit information (by speaking), while the other party provides simultaneous nonverbal cues on how the information is being received. The message recipient can nod to indicate comprehension, for example, wrinkle his forehead to signal confusion, frown to show displeasure, or interrupt with a question. In-person communication is unparalleled in adjusting messages on the fly, based on audience reaction, and this becomes especially important when complex ideas are under discussion.
Yet is face-to-face communication so superior to high-quality video calls? For a long time, video calls were low-resolution, laggy, and unreliable, making nonverbal cues hard to pick up and verbal interruptions awkward. Nowadays, cameras and microphones are getting better, broadband Internet is widely available, and screens are larger and high-definition. High-quality video streaming is accessible across multiple platforms. What does direct interaction offer that can’t also be had with video calls?
Well, for one, less friction to communicate. The challenge with video communication is the many minor interactions that go on in a typical workday. When everyone is situated in the same office, lots of low-cost opportunities exist for the exchange of low-value information—you can ask someone in the hallway a quick question, or bounce an idea off a person, or pop your head into a colleague’s office to get advice. Initiating a video call for similar discussions seems awkward. While video communication largely solves the problem of communicating complex high-value information, the absence of low-stakes interactions can lead to gaps in information—and those can add up. Moreover, low-stakes conversations, which may not even relate directly to work, are probably very important in building trust among employees.
Increasingly sophisticated videochat technology on smartphones, cleverly utilized, could provide a way around this limitation. At Google, some distributed teams schedule time to share lunch over videochat, hoping to create opportunities for low-stakes talk. Other teams at the company leave videochat windows open all the time, seeking to simulate working in the same space. Meantime, startup funder Pioneer and Emergent Ventures, a grant program of the Mercatus Center, are looking to fund research on videoconferencing interactions and what makes them different. “Given the rise of remote work,” writes Tyler Cowen, “the economic impact of this research could be Nobel-worthy.”
But in recent years, the problem of low-stakes communication has been solved to some degree by a communications strategy that owes most of its utility to distributed innovation among users: text-based messaging. Text solves a transaction cost associated with videochat, which requires the communicating parties to stop whatever they’re doing and give full attention to the conversation. An e-mail, by contrast, can be responded to when convenient.
But text has its own costs. Norms around polite e-mail usage evolved from letter-writing; and in a professional context, they might require one to write, “Dear so and so, I hope this message finds you well,” before proceeding to the content of the communication—a transaction cost, albeit small, of its own. As linguist Gretchen McCulloch points out in her recent history of online language, Internet users are solving this deficiency. Instant messaging and direct messaging generally take their norms from free-flowing conversation, not formal letters, and typically dispense with formal salutations and other requirements of politeness, lowering the already-low costs to initiate communication. The way text conversations are organized makes it easier to follow—and therefore engage in—long back-and-forth exchanges than is the case with e-mail chains.
As for another weakness of text-based exchanges—the difficulty they have in conveying moment-to-moment emotional cues—web users have been busy inventing digital markers as substitutes. Online, I can now SHOUT, be sarcastic (LOL), and emphasize things. We have adopted the emoji and the GIF to color and supplement the typographical content of messages. Despite the presence of an emoji for almost every conceivable concept, McCulloch reports, the most commonly used remain gestures and faces. And while any movie clip can be turned into a GIF, the most common represent instantly recognizable human reactions.
These norms evolved online in mostly informal ways, but as fluency in them became more extensive, they migrated into the work world. Slack was one of the first firms to realize that informal, message-based communication could be professionalized, but Google, Facebook, and Microsoft all have products in this space. The systems are integrated with videochat services. Not everyone likes these new text-based professional-messaging platforms because they think that they encourage excessive informal communication in the office. But the very fact that some call Slack a productivity killer because it encourages too much communication is a strong indicator that the interactive limitations of remote communication are being overcome.
Complementing text- and video-based communication has been a proliferation of options for sharing digital objects. Google Docs, Dropbox, Amazon web services, GitHub, and other services have made it easy for distributed working teams simultaneously to access and edit documents, code, and other products. Another innovation is the digital whiteboard, used extensively in meetings to solve design and other problems. In a 2016 study of remote collaboration at Google, the lack of a good remote whiteboard technology was seen as one of the major obstacles to distance work; since then, Google and other firms have developed a number of alternatives in this area.
Another piece of the remote-work puzzle is travel. In Google’s study, employees insisted that face-to-face meetings at the outset of a project led to a substantial boost in efficiency. The evidence on whether this belief is well justified is mixed: for example, another 2016 study found that randomly paired undergraduates who met face-to-face and over videochat had no differences in affiliative feelings. But even if face-to-face does matter, that doesn’t mean that everyone must live in the same city. The trust built in a face-to-face meeting is remarkably durable. Remotes may end up making very few daily trips to the main office but more long trips to meet occasionally with distant coworkers. And in the era of ubiquitous wireless connectivity, traveling days no longer need be unproductive days.
One of the primary advantages of living in or of running a business in cities is the presence of thick labor markets. If you have niche skills or niche needs, it’s easier to find a good match in a big market, which pushes firms and employees to cluster together in cities. The increasing ease of remote work erodes this advantage. After all, a remote worker potentially can apply for positions anywhere in the country, and a business can draw on a labor market national in scope. For this to work, though, the employee and employer need to be able to find each other.
Matching typically requires some means of learning about potential partners, beyond a résumé or a job posting, and physical interviews are an obvious way to do that. It has become much easier, though, to learn about possible matches at a distance. Many job seekers use social media (and often have professional websites), for example, so a company can usually get a sense of what a possible hire is like before any personal meeting, or even videochat, takes place. And specialized websites are emerging to provide third-party verification of claims made on professional websites. Upwork and LinkedIn give information about the specific skills of potential workers, validated by others. On the other side, Glassdoor gives job applicants information about employers. These websites aren’t perfect, but they will continue to improve.
Another force that helps workers and firms in big cities match up is social networks. Studies document the importance of connections in hiring decisions. Economists think that this results from friends, acquaintances, and contacts being able to deliver candid information and judgments about potential matches. Would this worker get along with the existing team? Is he or she reliable? Does this company treat employees well? And networks do more than facilitate hiring decisions; they’re also a source of industry-specific knowledge. Remote workers have, in the past, not easily been able to access such networks, which are often built via pickup basketball games, cocktail parties, training workshops, and many other things that depend on physical proximity.
For most, networking opportunities remain stronger for geographically clustered people, but technology is starting to shrink this advantage. Traveling to events or conferences makes it possible for distributed employees to meet face-to-face, and the Internet has greatly improved their ability to keep in touch afterward. The conference-and-event sector has been expanding swiftly and was projected to grow much faster over the next decade, though that will likely change now because of the coronavirus outbreak.
Networking can also take place entirely online. This shift is in its infancy, but the wave of canceled academic conferences due to Covid-19 is leading to rapid experimentation with online conferences that facilitate informal interaction. Online gaming, to take another example, could turn out to be a way to maintain and build social networks remotely, mirroring in the digital world the role that pickup basketball or soccer games play in the physical one. According to the latest report by the Entertainment Software Association, 41 percent of American adults play video games with other people online, averaging nearly five hours per week. New cloud-based streaming platforms from Google, Microsoft, and Ubisoft are reducing barriers for new gamers, so the share of people playing online will probably keep rising.
The expansion of web-based education, especially when it uses pedagogy that requires students to interact, is also acclimating a new generation to collaborate online. It’s easy to imagine this development leading to online professional training events that bring participants into contact with new people.
I’ve experienced something like this already on Twitter. Twitter was not designed to be a professional networking site, but among a large group of economists (academic and aspiring) that calls itself #EconTwitter, the platform has enabled a forum not only for sharing and discussing economic ideas but also for simply joking around with people sharing common interests. These links may start casually, but they can lead to deeper relations when #EconTwitter participants arrange in-person connections on the seminar circuit, at conferences, and while traveling. I’m currently working on a paper with someone I met on #EconTwitter.
And Twitter helps solve another problem in overcoming distance. Yet another economic advantage of successful cities is the intangible benefit of knowledge that exists “in the air.” Dense cities create opportunities for encountering new ideas and making unexpected associations, which sparks innovation. But I’ve lived in London and Washington, D.C., and have learned more on Twitter than from the interactions I’ve had in these big cities. In a tweet by Lyman Stone, I first discovered the statistic that I cited on remote workers as a share of the nonfarm labor force. I’ve never met Stone, but I used Twitter to send him a message asking if I could cite the number in this essay. As most universities increase online education, the academic Twitter community has stepped up to share a (frankly overwhelming) number of ideas, tips, and resources.
While I learn a lot from the #EconTwitter community, I also follow many scholars outside of economics, as well as writers, artists, wonks, lawyers, nurses, venture capitalists, socialists, and others. When someone from outside the circle of people I follow makes an interesting point, it often shows up in a retweet by one of them, so it comes onto my radar. Twitter, used well, is a serendipity machine.
Many do not find Twitter so useful—some of its own users call it a “hell site” for the angry denunciations it circulates—but #EconTwitter is far from the only group that has made it something close to an online networking opportunity. Political science Twitter, accounting Twitter, investor Twitter, linguist Twitter—one can find any number of well-functioning virtual communities, expressing ideas and information in friendly ways.
Most economists don’t see the death of distance as imminent, and strong empirical evidence shows agglomeration effects to be large and having grown for decades. But caution is warranted. That research is mostly backward-looking, and the trend toward remote work is very recent. It may well be that steady gains in the productivity of remotes will have a negligible effect on most industries until it crosses a threshold. Once this occurs, though, we may see the remote trend accelerate.
The belief that agglomeration effects are an immutable fact of modern economies naturally leads to policies that accommodate and encourage agglomeration. But for several reasons beyond slowing the growth of Covid-19, we should be encouraging the rise of remotes, as well. First, the change could reduce the economic inequality between cities and rural areas. Physical agglomeration leads to a highly uneven distribution of economic activity. Cities prosper but rural areas decline, and the widening divide spills over into politics, making them more fractious. If rural residents can enjoy the agglomeration benefits of urban life without moving to the city, prosperity can be more widely shared. And it wouldn’t be necessary for everyone living outside major cities to work online. The local presence of highly paid workers would generate demand for additional jobs in other sectors.
Second, an increase in remote employment could reduce the carbon footprint of work. Most Americans commute to work by car, and transportation accounted for 29 percent of U.S. carbon emissions in 2017. (Remote work wouldn’t reduce these emissions to zero, of course—it might bring more frequent plane trips to meet distant colleagues, for instance.)
Remote work could also boost productivity growth. In most economic models of physical agglomeration, clustering people generates economic gains but also leads to costs, including those associated with congestion (longer commutes, traffic clogs, escalating housing prices). Eventually, these costs get so large that they prevent people from moving to cities and further benefiting from agglomeration. But if distance becomes less important, we could enjoy the economic benefits of a gigantic digital city without the associated congestion woes—and this could well boost productivity growth, especially in intangible industries.
Policies to accelerate the end of distance would include expanding broadband in rural areas and supporting online education, which would help people who might want to work remotely improve their skills. More thinking on how to make the Internet a more enjoyable place to hang out and interact with new people would help, too.
Lastly, it’s important to bear in mind that the pivot to remote work due to Covid-19 is being made under extraordinary conditions, rushed and relatively unplanned. Many will be attempting to work remotely while simultaneously providing child care and dealing with other pandemic-related exigencies. Since firms are mostly retaining their office space and paying the same wages, it won’t save them much money. And since workers are still working for the same firms, no one is benefiting from the option of separating the decision of where to live and where to work. The sudden expansion of remote work will feel especially socially isolating, since it is occurring amid general social distancing. In short, this is the worst version of modern remote work. It will get better.
Illustrations by The Heads of State