The great pandemic work-from-home experiment was a remarkable success
Share this article:
Nicholas Bloom, an economist at Stanford University, had been studying working-from-home arrangements for years before corporate America sent its white-collar employees home in droves in March 2020. In a paper published in 2015, for example, he and three co-authors explored what happened when workers at a Shanghai travel company were randomly selected into a work-from-home program: It turned out they were 13 percent more productive. Then suddenly, last year, his niche academic focus was thrust to the center of every business discussion. After fielding numerous calls from executives and reporters to discuss his older work, he said, "it became clear that it would be helpful to get contemporaneous data." Working with colleagues including José María Barrero, of Mexico's Autonomous Institute of Technology, and Steven J. Davis, of the University of Chicago, he began a fresh set of studies exploring how the pandemic was reshaping working and living arrangements. He discussed that work recently, in a conversation edited for length and clarity.
Q: In May 2020, you started surveying a random, representative sample of 2,500 workers monthly - recently expanding the sample to 5,000. [They are 20 to 64 years old and make at least $20,000.] Something like half of those people don't work from home. But of those who do, 40% report being more efficient at home, 45% say they are about equally efficient, and only 15% say they are less efficient. Those are kind of amazing figures, given all that we've read about the chaos of working from home, especially early on. Do they surprise you?
A: It's amazing. Probably the biggest surprise of the pandemic was that working from home worked so well. We didn't find a 13 percent productivity increase, but we're finding an average of 4% or so. Something like 60 percent of respondents say working from home has worked out "better than I expected." Firms are astounded.
Q: In that survey, you're asking people to estimate their productivity. Can we trust this kind of self-reporting?
A: This is a good question. The Shanghai study was fantastic because we were able to measure productivity directly [by examining phone calls travel workers made, whether sales were finalized and so on]. Whereas the self-assessed stuff is super broad. But there are a few other papers out there that measure productivity gains from working from home, and they come up with a number similar to ours.
Q: You estimate that the increased productivity comes mostly from saved commuting time.
A: Yeah, about two-thirds of it is saved commuting time, and one-third of it is greater efficiency.
Q: Your surveys of workers show that, in general, employees would prefer about 2.5 days of working from home - but employers, although they have moved up a bit from the start of the pandemic, only want to allow a little more than one day. Does that suggest a clash when things return to normal?
A: There are two dimensions to look at there. You have to break it down to those who can and those that can't work from home. Those who can are basically university grads, professionals, managers and so on - about half the workforce. And then there's those that can't, who are front-line manufacturing, retail, health care. That second group definitely wants to work from home, but they're not going to get it. Their jobs just can't be done remotely. So in some ways, it's a clash.
But the other workers - those who worked from home during the pandemic - have won, in a sense. From the start of our survey, they've said they want to work from home two to two and a half days a week. Now employers have edged up to meet what employees want. And that's driven partly by managers becoming more comfortable with the idea of working from home, but it's primarily driven by labor market strength.
Goldman Sachs and JPMorgan are good examples. They both announced [at first] that people were going to come back to the office five days a week. They've basically given up on that. It's impossible. And the reason is Merrill Lynch, Morgan Stanley, Deutsche Bank, Citibank are not doing that. If Goldman Sachs and JPMorgan try to force people back, a bunch of people will say, "Get lost, I'm quitting."
Q: There's some skeptics of working from home, like the economists Ed Glaeser and David Cutler, who think cities are important, offices are important: These interpersonal connections fuel productivity.
A: I'm very aligned with that. I am uniformly advising organizations that the default is hybrid, which means either two or three days in the office. So I am totally on board that there are major benefits from meeting in person. The research evidence for that is honestly a lot lighter - but my experience is guided by literally hundreds of managers who say, over and over and over again, you need face-to-face interaction to really spark innovation and culture. And so I'm totally on board with that.
I just don't think you need five days a week. In a hybrid plan, the team comes in three days. We all agree on the same days. On those three days we have all our meetings, trainings, events, lunches - the hyper-social things. Then the other two days we work from home. So we end up spending as much face time together as we ever did. We just crush it into three days. We reallocate the quiet time we used to have at work to the two days we are at home. It's honestly better time management.
Q: Your survey also covers some cultural issues. It sounds like the handshake is dead - but women and men have different preferences. Does that suggest another kind of culture clash when we return?
A: Pre-pandemic, both men and women, about two-thirds of them, would shake hands. Post-pandemic, men have moved to about a third each: verbal greeting, handshake, fist bump. Women have moved two-thirds to a verbal greeting. So women have made the greatest change - but generally, if you'll be greeting someone and you don't know them, I really wouldn't put your hand out because it's unlikely that's going to be a welcome gesture.
I've heard anecdotes from women, though, who'll say: "I came back from this meeting, and the client was a senior male. He pumped the hand of everyone in the room. And most of us ran out of the meeting as soon as we could and went to the bathroom to cleanse our hands." The unwanted handshake could be seen as a kind of microaggression.
Q. You also have a paper called "The Donut Effect of Covid-19 on Cities," which documents - using post office data and changing housing prices - how people are moving from central business districts to inner suburbs, and from inner suburbs to outer suburbs.
A. There was an extreme doughnut effect in the pandemic. What's more interesting is the post-pandemic question: Will this survive, and how does that affect the big city? I think it will survive. [One piece of evidence - since home prices represent a bet on the future value of properties - is that prices in the outer suburbs in the 12 largest cities are up about 12 percent from pre-pandemic values, while prices in the central business districts are down about 3%.]
So demand and higher property prices get moved out to the suburbs, and the centers get cheaper and a bit emptier. I think that's really great, because the one group of workers that really do need to be in city centers are lower-income workers that are essential service workers - people that have to work every day. It makes much more sense for them to be in the city centers and for investment bankers and techies that can work from home to be out in the suburbs. It's one step toward unraveling the affordability crisis - not a big step, I have to say, but it's one step.
THE WASHINGTON POST