Are We Really More Productive Working from Home?

The survey measured the incidence of working from home as the pandemic continued, focusing on how a more permanent shift to remote work might affect not only productivity but also overall employee well-being. It also examined factors including how work from home would affect spending and revenues in major urban centers. In addition to the survey, the researchers drew on informal conversations with dozens of US business executives. They are publishing the results of the survey and related research at wfhresearch.com.

In an analysis of the data collected through March 2021, they find that nearly six out of 10 workers reported being more productive working from home than they expected to be, compared with 14 percent who said they got less done. On average, respondents’ productivity at home was 7 percent higher than they expected. Forty percent of workers reported they were more productive at home during the pandemic than they had been when in the office, and only 15 percent said the opposite was true. The researchers argue that the work-from-home trend is here to stay, and they calculate that these working arrangements will increase overall worker productivity in the US by 5 percent as compared with the pre-pandemic economy.

“Working from home under the pandemic has been far more productive than I or pretty much anyone else predicted,” Bloom says.

No commute, and fewer hours worked

Some workers arguing in favor of flexibility might say they’re more efficient at home away from chatty colleagues and the other distractions of an office, and that may be true. But above all, the increased productivity comes from saving transit time, an effect overlooked by standard productivity calculations. “Three-quarters or more of the productivity gains that we find are coming from a reduction in commuting time,” Davis says. Eliminate commuting as a factor, and the researchers project only a 1 percent productivity boost in the postpandemic work-from-home environment, as compared with before.

It makes sense that standard statistics miss the impact of commutes, Davis explains. Ordinarily, commuting time generally doesn’t shift significantly in the aggregate. But much like rare power outages in Manhattan have made it possible for New Yorkers to suddenly see the nighttime stars, the dramatic work-from-home shift that occurred during the pandemic made it possible to recognize the impact traveling to and from an office had on productivity.

Before the pandemic, US workers were commuting an average of 54 minutes daily, according to Barrero, Bloom, and Davis. In the aggregate, the researchers say, the pandemic-induced shift to remote work meant 62.5 million fewer commuting hours per workday.

People who worked from home spent an average of 35 percent of saved commuting time on their jobs, the researchers find. They devoted the rest to other activities, including household chores, childcare, leisure activities such as watching movies and TV, outdoor exercise, and even second jobs.

Aside from commuting less, remote workers may also be sleeping more efficiently, another phenomenon that could feed into productivity. On days they worked remotely, people rose about 30 minutes later than on-site workers did, according to pre-pandemic research by Sabrina Wulff Pabilonia of the US Bureau of Labor Statistics and SUNY Empire’s Victoria Vernon. Both groups worked the same number of hours and slept about the same amount each night, so it’s most likely that “working from home permits a more comfortable personal sleep schedule,” says Vernon. “Teleworkers who spend less time commuting may be happier and less tired, and therefore more productive,” write the researchers, who analyzed BLS data from 2017 to 2018.

While remote employees gained back commuting time during the pandemic, they also worked fewer hours, note Barrero, Bloom, and Davis. Hours on the job averaged about 32 per week, compared with 36 pre-pandemic, although the work time stretched past traditional office hours. “Respondents may devote a few more minutes in the morning to chores and childcare, while still devoting about a third of their old commuting time slot to their primary job. At the end of the day, they might end somewhat early and turn on the TV. They might interrupt TV time to respond to a late afternoon or early evening work request,” the researchers explain.

This interpretation, they write, is consistent with media reports that employees worked longer hours from home during the pandemic but with the added flexibility to interrupt the working day. Yet, according to the survey, this does not have a negative overall effect on productivity, contradicting one outdated stereotype of a remote worker eating bonbons, watching TV, and getting no work done.

Remote-work technology goes mainstream

The widespread implementation of remote-working technology, a defining feature of the pandemic, is another important factor for productivity. This technology will boost work-from-home productivity by 46 percent by the end of the pandemic, relative to the pre-pandemic situation, according to a model developed by Rutgers’s Morris A. Davis, University of North Carolina’s Andra C. Ghent, and University of Wisconsin’s Jesse M. Gregory. “While many home-office technologies have been around for a while, the technologies become much more useful after widespread adoption,” the researchers note.

There are significant costs to leaving the office, Rutgers’s Davis says, pointing to the loss of face-to-face interaction, among other things. “Working at home is always less productive than working at the office. Always,” he said on a June episode of the Freakonomics podcast.

One reason, he says, has to do with the function of cities as business centers. “Cities exist because, we think, the crowding of employment makes everyone more productive,” he explains. “This idea also applies to firms: a firm puts all workers on the same floor of a building, or all in the same suite rather than spread throughout a building, for reasons of efficiency. It is easier to communicate and share ideas with office mates, which leads to more productive outcomes.” While some employees are more productive at home, that’s not the case overall, according to the model, which after calibration “implies that the average high-skill worker is less productive at home than at the office, even postpandemic,” he says.

And yet Davis, Ghent, and Gregory’s model projects that after the pandemic winds down, highly skilled, college-educated workers will spend 30 percent of their time working from home, as opposed to 10 percent in prior times. While physical proximity may be superior, working from home is far more productive than it used to be. Had the pandemic hit in 1990, it would not have produced this rise in relative productivity, per the researchers’ model, because the technology available at the time was not sufficient to support remote work.

A June article in the MIT Technology Review by Stanford’s Erik Brynjolfsson and MIT postdoctoral scholar Georgios Petropoulos corroborates this view. Citing the 5.4 percent increase in US labor productivity in the first quarter of 2021, as reported by the BLS, the researchers attribute at least some of this to the rise of work-from-home technologies. The pandemic, they write, has “compressed a decade’s worth of digital innovation in areas like remote work into less than a year.” The biggest productivity impact of the pandemic will be realized in the longer run, as the work-from-home trend continues, they argue.

Lost ideas, longer hours?

Not all the research supports the idea that remote work increases productivity and decreases the number of hours workers spend on the job. Chicago Booth’s Michael Gibbs and University of Essex’s Friederike Mengel and Christoph Siemroth find contradictory evidence from a study of 10,000 high-skilled workers at a large Asian IT-services company.

The researchers used personnel and analytics data from before and during the coronavirus work-from-home period. The company provided a rich data set for these 10,000 employees, who moved to 100 percent work from home in March 2020 and began returning to the office in late October.

Total hours worked during that time increased by approximately 30 percent, including an 18 percent rise in working beyond normal business hours, the researchers find. At the same time, however, average output—as measured by the company through setting work goals and tracking progress toward them—declined slightly. Time spent on coordination activities and meetings also increased, while uninterrupted work hours shrank. Additionally, employees spent less time networking and had fewer one-on-one meetings with their supervisors, find the researchers, adding that the increase in hours worked and the decline in productivity were more significant for employees with children at home. Weighing output against hours worked, the researchers conclude that productivity decreased by about 20 percent. They estimate that, even after accounting for the loss of commuting time, employees worked about a third of an hour per day more than they did at the office. “Of course, that time was spent in productive work instead of sitting in traffic, which is beneficial,” they acknowledge.

Overall, though, do workers with more flexibility work fewer hours (as Barrero, Bloom, and Davis find) or more (as at the Asian IT-services company)? It could take more data to answer this question. “I suspect that a high fraction of employees of all types, across the globe, value the flexibility, lack of a commute, and other aspects of work from home. This might bias survey respondents toward giving more positive answers to questions about their productivity,” says Gibbs.

The findings of his research do not entirely contradict those of Barrero, Bloom, and Davis, however. For one, Gibbs, Mengel, and Siemroth acknowledge that their study doesn’t necessarily reflect the remote-work model as it might look in postpandemic times, when employees are relieved of the weight of a massive global crisis. “While the average effect of working from home on productivity is negative in our study, this does not rule out that a ‘targeted working from home’ regime might be desirable,” they write.

Additionally, the research data are derived from a single company and may not be representative of the wider economy, although Gibbs notes that the IT company is one that should be able to optimize remote work. Most employees worked on company laptops, “and IT-related industries and occupations are usually at the top of lists of those areas most likely to be able to do WFH effectively.” Thus, he says, the findings may represent a cautionary note that remote work has costs and complexities worth addressing.

As he, Mengel, and Siemroth write, some predictions of work-from-home success may be overly optimistic, “perhaps because professionals engage in many tasks that require collaboration, communication, and innovation, which are more difficult to achieve with virtual, scheduled interactions.”

Attracting top talent

The focus on IT employees’ productivity, however, excludes issues such as worker morale and retention, Booth’s Davis notes. More generally, “the producer has to attract workers . . . and if workers really want to commute less, and they can save time on their end, and employers can figure out some way to accommodate that, they’re going to have more success with workers at a given wage cost.”

Companies that offer more flexibility in work arrangements may have the best chance of attracting top talent at the best price. The data from Barrero, Bloom, and Davis reveal that some workers are willing to take a sizable pay cut in exchange for the opportunity to work remotely two or three days a week. This may give threats from CEOs such as Morgan Stanley’s James Gorman—who said at the company’s US Financials, Payments & CRE conference in June, “If you want to get paid New York rates, you work in New York”—a bit less bite. Meanwhile, Duke PhD student John W. Barry, Cornell’s Murillo Campello, Duke’s John R. Graham, and Chicago Booth’s Yueran Ma find that companies offering flexibility are the ones most poised to grow.

Working policies may be shaped by employees’ preferences. Some workers still prefer working from the office; others prefer to stay working remotely; many would opt for a hybrid model, with some days in the office and some at home (as Amazon and other companies have introduced). As countries emerge from the pandemic and employers recalibrate, companies could bring back some employees and allow others to work from home. This should ultimately boost productivity, Booth’s Davis says.

Or they could allow some to work from far-flung locales. Harvard’s Prithwiraj Choudhury has long focused his research on working not just from home but “from anywhere.” This goes beyond the idea of employees working from their living room in the same city in which their company is located—instead, if they want to live across the country, or even in another country, they can do so without any concern about being near headquarters.

Research Choudhury conducted with Harvard PhD student Cirrus Foroughi and Northeastern University’s Barbara Larson analyzes a 2012 transition from a work-from-home to a work-from-anywhere model among patent examiners with the United States Patent and Trademark Office. The researchers exploited a natural experiment and estimate that there was a 4.4 percent increase in work output when the examiners transitioned from a work-from-home regime to the work-from-anywhere regime.

“Work from anywhere offers workers geographic flexibility and can help workers relocate to their preferred locations,” Choudhury says. “Workers could gain additional utility by relocating to a cheaper location, moving closer to family, or mitigating frictions around immigration or dual careers.”

He notes as well the potential advantages for companies that allow workers to be located anywhere across the globe. “In addition to benefits to workers and organizations, WFA might also help reverse talent flows from smaller towns to larger cities and from emerging markets,” he says. “This might lead to a more equitable distribution of talent across geographies.”

More data to come

It is still early to draw strong conclusions about the impact of remote work on productivity. People who were sent home to work because of the COVID-19 pandemic may have been more motivated than before to prove they were essential, says Booth’s Ayelet Fishbach, a social psychologist. Additionally, there were fewer distractions from the outside because of the broad shutdowns. “The world helped them stay motivated,” she says, adding that looking at such an atypical year may not tell us as much about the future as performing the same experiment in a typical year would.

Before the pandemic, workers who already knew they performed better in a remote-working lifestyle self-selected into it, if allowed. During the pandemic, shutdowns forced remote work on millions. An experiment that allowed for random selection would likely be more telling. “The work-from-home experience seems to be more positive than what people believed, but we still don’t have great data,” Fishbach says.

Adding to the less optimistic view of a work-from-home future, Booth’s Austan D. Goolsbee says that some long-term trends may challenge remote work. Since the 1980s, as the largest companies have gained market power, corporate profits have risen dramatically while the share of profits going to workers has dropped to record lows. “This divergence between productivity and pay may very well come to pass regarding time,” he told graduating Booth students at their convocation ceremony. Companies may try to claw back time from those who are remote, he says, by expecting employees to work for longer hours or during their off hours.

And author and behavioral scientist Jon Levy argues in the Boston Globe that having some people in the office and others at home runs counter to smooth organizational processes. To this, Bloom offers a potential solution: instead of letting employees pick their own remote workdays, employers should ensure all workers take remote days together and come into the office on the same days. This, he says, could help alleviate the challenges of managing a hybrid team and level the playing field, whereas a looser model could potentially hurt employees who might be more likely to choose working from home (such as mothers with young children) while elevating those who might find it easier to come into the office every day (such as single men).

Gibbs concurs, noting that companies using a hybrid model will have to find ways to make sure employees who should interact will be on campus simultaneously. “Managers may specify that the entire team meets in person every Monday morning, for example,” he says. “R&D groups may need to make sure that researchers are on campus at the same time, to spur unplanned interactions that sometimes lead to new ideas and innovations.”

Sentiments vary by location, industry, and culture. Japanese workers are reportedly still mostly opting to go to the office, even as the government promotes remote work. Among European executives, a whopping 88 percent reportedly disagree with the idea that remote work is as or more productive than working at the office.

Regardless of what research establishes in the long run about productivity, many workers are already demanding flexibility in their schedules. While only about 28 percent of US office workers were back onsite by June 2021, employees who had become used to more flexibility were demanding it remain. A May survey of 1,000 workers by Morning Consult on behalf of Bloomberg News finds that about half of millennial and Gen Z workers, and two-fifths of all workers, would consider quitting if their employers weren’t flexible about work-from-home policies. And additional research from Barrero, Bloom, and Davis finds that four in 10 Americans who currently work from home at least one day a week would look for another job if their employers told them to come back to the office full time. Additionally, most employees would look favorably upon a new job that offered the same pay as their current job along with the option to work from home two to three days a week.

The shift to remote work affects a significant slice of the US workforce. A study by Chicago Booth’s Jonathan Dingel and Brent Neiman finds that while the majority of all jobs in the US require appearing in person, more than a third can potentially be performed entirely remotely. Of these jobs, the majority—including many in engineering, computing, law, and finance—pay more than those that cannot be done at home, such as food service, construction, and building-maintenance jobs.

Barrero, Bloom, and Davis project that, postpandemic, Americans overall will work approximately 20 percent of full workdays from home, four times the pre-pandemic level. This would make remote work less an aberration than a new norm. As the pandemic has demonstrated, many workers can be both productive and get dinner started between meetings.