Nicholas Bloom, economics professor at Stanford University, has been studying remote work and hybrid (a mix of remote and onsite) work for years. Then the pandemic made these modes widespread and lasting. He says as more organizations turn to hybrid work, they face difficult logistical, strategic, and managerial challenges. Bloom shares a guideline to implementing hybrid work plans, and helps managers think through these arrangements while balancing fairness to employees and organizational needs. Bloom is the author of the HBR article “Don’t Let Employees Pick Their WFH Days.”
CURT NICKISCH: Welcome to the HBR IdeaCast from Harvard Business Review. I’m Curt Nickisch.
To say the last year has changed us would be an understatement. The pandemic forced many people to work and manage much more remotely and differently. Now, as a growing number of places return to more in-person work, it’s a good time to explore the issues that organizations and managers face as they reset and reorganize their working lives.
Throughout the next few months, HBR IdeaCast will look at these challenges and today we’re starting with implementing hybrid work across an organization. That’s a mix of remote work and onsite activity. Our guest today has studied remote work for nearly 20 years and he’s here to explain how leaders can think through hybrid work arrangements so that they’re fair to employees and effective for the organization, and not just short term.
Nicholas Bloom is a professor of economics at Stanford University and the author of the HBR article, “Don’t Let Employees Pick Their Work From Home days.” Nick, thanks so much for coming on the show.
NICHOLAS BLOOM: Curt, lovely to be here, thanks.
CURT NICKISCH: So I have a lot of nitty gritty questions for you about this, but just to get us going here, why is it important for managers and leaders to have a strategy in place when it comes to work from home and hybrid work? Why cant you just figure it out as you go along?
NICHOLAS BLOOM: Oh, that’s, in a sense the easy question. Firms and CEOs, in particular, CHRs, are under enormous pressure right now to tell their employees what they’re going to do in future, because these employees need to plan their lives around this. What most firms have decided is to go for hybrid. The typical plan will be something like three days in the office, two at home. So huge numbers of firms have announced this, Google, Microsoft, Apple, Citi Bank, HSBC, blah, blah, blah. Around 70 to 80% of firms are going to hybrid, the other 20 to 30% have mostly not decided.
So why is that? The reason is, hybrid represents a great trade-off between the benefits of being in-person, which is better innovation, typically better creativity, and the ability to build office culture, versus the benefits of not having to go into the office, which enables you to work quietly at home and avoid the painful commute. It looks like from the research firms are going to be about 5% more productive doing that. And employees on average are significantly happier.
CURT NICKISCH: Nick, in a recent HBR article, you argued that workers shouldn’t all get to choose what days they have off. Why is prescribing the days in and out of the office a good policy?
NICHOLAS BLOOM: Choices may be the most controversial and the most difficult issue, I think for managers I talk to right now. So the issue is, imagine you, as in the huge majority of firms have decided you’re going to do hybrid post pandemic. Now the question is how to execute the scheme and in particular, who gets to choose which days of the week they work from home. And also how many days do they get to work from home? So one extreme is full decentralization. Every employee gets to choose which days and how many days. So if you want to work from home four days a week and only come in on Wednesdays, you can, if you want to work from home zero days a week, you can. That’s one extreme. The middle position, I think, is team by team. So as a corporation, you say each team managers decides and he or she basically polls the people in their team, maybe polls other teams that they want to work closely with and makes the decision.
And then the final choice is fully centralized, which is the CEO, CHR, basically the people running the company, they decide, and they tell each team, you will work from home on these days. And importantly, you will not work from home on these other days. So for example, take Apple. Apple has announced you’re going to come in Monday, Tuesday, Thursday, work from home Wednesday, Friday. What that means is Monday, Tuesday, Thursday, you have all of your in-person meetings, big events, training sessions, leaving events, lunches, et cetera. It’s like three exhaustingly social days. And then Wednesday, Friday, you leave report writing, reading presentations, emails. And I have to say, in survey data I’ve seen, and from talking to companies, firms are pretty equally split between those three different positions.
CURT NICKISCH: And you recommend the harder prescription, right?
NICHOLAS BLOOM: My views have also changed to start off with, so the fully decentralized, the full choice version was what I was actually suggesting a year ago. And the justification for this is there is a huge variation in what employees want. If we look at survey data, around 20% of employees never want to work from home, they want to come in five days a week. They tend to be young singles or empty nesters. And they’re frankly fed up with working at home and they find it isolating. At the other extreme, 30% of people never want to come into work again. They tend to be middle-aged with kids and family, they live far away, or it’s expensive to commute. And then the remaining 50% want anything between one to four days a week. So given this huge variety, why not let people choose?
And in fact, one very senior manager I spoke to perfectly encapsulated this when she said, “Look for my team, I have great performance metrics and I’ve told them, I’m going to treat you adults. America is a free country. If you perform your job well, you can choose which days and how many days you come in and it’s up to you. Of course, if you don’t perform your job, I’m going to get serious with you and maybe haul you back into the office. But as long as you are performing, it’s entirely your choice.” That’s the defense for the full choice model. My view has changed because I see three problems with the full choice model.
CURT NICKISCH: Can I guess that one problem with this is that if it’s a complete mix of people every day with who’s in and who’s out, then some people start to think what’s the point of coming in if other people that I need to work with today are remote anyway, I might as well just stay home.
NICHOLAS BLOOM: Yes, exactly. So one issue is what I’d call the challenges of mixed mode. Mixed mode is the name that’s being given when some are in the office and some are at home, turns out that really doesn’t work that well. So imagine trying to take Zoom or a Teams call with six people of which three are in the office, three are at home. You can imagine what happens. The three in the office will get in a conference room. They’re like three small heads in one screen. The other three are big heads on the other three screens. A, you can’t really see their expressions. And even worse, you can see there’s whispering –
CURT NICKISCH: Some companies try to solve this by making everybody join remotely, even when they’re at the office.
NICHOLAS BLOOM: Exactly. So that’s the next step along this evolution is, right –
CURT NICKISCH: So let’s make this equitable. We’re all talking to each other over a screen, even though we’re next to each other.
NICHOLAS BLOOM: Exactly. And so a lot of firms are going to set up these cubicles, I hear there’s soundproofing issues that people shout. They try their best. The problem with having people, join on their laptops in the office is A, as you said, why come in, in the first place, and B, more seriously, any controversial meeting, you know exactly what’s going to happen. If you’re one of the three from home, as soon as the meeting ends, the three folks in the office close their laptops, walk out of the cubicle and talk to each other, probably go grab a coffee.
So you cannot escape really this in-group out-group without some kind of Orwellian setup, some managers said, well, we’re just going to ban employees after meetings from talking about it. It’s like, can you imagine how you implement that? That’s like the secret police. Mixed mode is honestly really problematic and you can have better or worse mixed mode, but even the best execution runs into problems.
CURT NICKISCH: What other downsides are there to letting workers choose when to work from home?
NICHOLAS BLOOM: Imagine people were asked, if you get to work from home two days a week, which two days would you think they typically choose to work from home?
CURT NICKISCH: Monday and Friday, so they have long weekends.
NICHOLAS BLOOM: Yes, exactly. In the survey data, what we find is we asked, we’ve done this two waves, we’ve asked 10,000 people. So it’s pretty good data at this point, which days you would choose. And you find that only 18% percent of people choose Wednesday and 64% choose Friday. So if you reverse that around, you can see the problem. So if you think about how many people are going to come in, 82% of employees are going to come in on a Wednesday and you’re only going to have 36% come in Friday. What that means is if you have enough desks and office space to fully accommodate everyone on Wednesday, more than half of that is empty on the Friday and a Monday. So Wednesday is crammed, people are uncomfortable. There’s packed elevators, kitchens, doorways. Post COVID, people are very nervous about dense congestion. Monday, Friday in particular, it’s like tumbleweed blowing down the office.
CURT NICKISCH: You also wrote in your article about a diversity crisis, explain that.
NICHOLAS BLOOM: Yes. So the third point was something that, to be honest, came up looking at the data and talking to managers. So there are two facts you need to know. Fact one is who chooses to come into the office and how many days is not random. So for example, in our surveys, we found that if you look at college graduates, so that’s just over half the labor force folks with one plus year of college and you look at college grads with young kids, so kids under the age of 12, you see women are almost 50% more likely to want to work from home five days a week than men. It’s also the case, if you look at people who are disabled, who live far from the office, who are lower income, so can’t afford a car, these groups have a higher preference to work from home more days than other groups. And there’s nothing wrong with that, too, that’s completely as you’d expect, and it makes a lot of sense, but just to say that the choice if left open, it’s not going to be random.
Fact two, where this causes the problem is if you have teams where some people are working from home and others are coming in the office, those working from home face a very large promotion penalty. So to give you a number on that, I did a big randomized control trial on working from home in China, back in from 2010 to 2015. And what you found is people that work from home, their promotion rates were 50% of those that were coming in everyday. That is an enormous effect. And you can see why, they’re not around, they’re getting forgotten about, maybe they’re not developing managerial skills, et cetera. So if you put that together, you can see if you allow choice, you could easily find a situation five, 10 years from now whereby young single men are promoted up the firm, married people with young kids, particularly women, disabled people, people living further away don’t come in and they fall behind. And there’s both a diversity and a huge legal issue around them.
CURT NICKISCH: All right. So I’m a manager and I’ve decided to have a bit more structure when it comes to who works from home and when. How do I approach that decision? How do I decide that in an equitable way and communicate it and have it be an effective policy? What do you recommend?
NICHOLAS BLOOM: I think there are really two choices that I would advise. One is the team-based and one is the company-based and it depends what matters to firms. So you can say, look as a company, we would like some choice, but not complete choice. So we’re going to allow each team to decide it within the team, you’ve all got to come in on those days and all stay at home on the same days. And you may want to say, look, we advise teams to have three days in the office and two working from home. If you set it up at a team level, you immediately address the mixed mode problem for most meetings, because most meetings, at least if companies are set up on a team or a project basis or within team, within project, it’s not perfect, but you deal with most of it.
And you address most of the diversity concerns because promotions are normally considered against other folks in your team, that’s often the benchmark. You obviously don’t address the office use question. So at the team level you give more choice, but you’re going to have an inefficient use of office because most teams you’re going to discover are going to choose at least one of, if not both, Monday, Friday to work from home. The other version is just to centralize it. And coming back to the discussion earlier is say, we’re going to start off centralizing it. And we’re going to see how it plays out. Deciding upon hybrid now and announcing it’s a very good idea. And that’s what a huge number of companies have done, but announcing the full set of all the details and precisely how this will work is probably not best done now because it’s hard to predict and it’s going to change. And if you notice in the press, most companies have not announced the details precisely for those reasons.
CURT NICKISCH: There are also some people who have to come in more because of the nature of their work. When you have that variation, you are going to have differences and that can breed some issues of resentment and perceptions of unfairness. How do you deal with that?
NICHOLAS BLOOM: Multiple managers have said to me, roughly half or a third or two thirds of our employees can work from home, they’re very happy with this hybrid plan. The other one third or half, whatever it is, can’t and they’re feeling angry. They’ve said that they’ve had the whole pandemic having to come in everyday. Some of them have faced infection risk, been infected in fact, and now they’re not going to get this nice perk post-pandemic. First thing is a genuine issue. It is a real issue. That’s there’s no getting around it. And just to put numbers on it, in our survey, people report that working from home two days a week is equivalent to something like a 6% pay increase. They report it’s a pretty valuable thing to have.
And you can imagine why. So the types of solutions, one basic solution is obviously, and this has been discussed a fair bit, is having some make good pay bonus yearly for those that can’t work from home. Say, for example, you say for anyone that is not able to work from home, that’s coming in five days a week, we’re going to give you a 5% pay top up in reflection of the fact that you’re not getting this benefit. That’s mostly going to go to lower income individuals. If you look in the data, that’s slightly more likely to go to women, that’s slightly more likely to go to minorities, because they’re the kinds of roles we see in the data they hold, it tends to be slightly harder to have these working from home. So that’s definitely one policy that is on the table.
CURT NICKISCH: I imagine that some companies are in a little bit of a pickle there too, because they’ve also hired people who were remote first. Now you may have to merge that or you have that same resentment from other people who say, “Well, now these people were hired and they get to stay remote and meanwhile, I have to come in three days a week.”
NICHOLAS BLOOM: What I’ve been hearing a lot from various companies is, particularly actually tack firms, during the pandemic, because they’ve hired so aggressively, they have now way more headcount that they can ever fit in their headquarters. They also, as you say, have a number of employees that have been hired and never been into the office. So I think the way this is going to shake out, there’re some guidelines on this. I personally would advise against having teams where some people are fully remote and others are coming in three days a week because of the promotion costs and the disruption. If you had a team of say six people and you hired a seventh person fully remote in the pandemic, I would probably make them come in.
And if they are extremely resistant to coming in, you face a devil’s dilemma of, do you basically move them, ask them to come in or quit. I’m not sure. I personally would not have a team, where one person is remote and feeling resentful, doesn’t get promoted for three, four years, because I just don’t think that’s a healthy situation. For teams where I would try and unify the team level and that may well mean some people quit and some turnover, I think that’s unavoidable. Then as you say, there are other teams where they’ve mostly been staffed up remote. And for those teams, companies may take the view, look, if I have a team of eight people and they’re all basically remote and they’ve been hard remote, sure it may be better to have had them all else equal hybrid come in three days a week, but most of them are going to quit rather than do that. And I’d probably rather the team working remotely at 90% efficiency, then two thirds of them quit and try and get the other one third in. So I think there’s going to be a lot of legacy effects.
CURT NICKISCH: It’s definitely an interesting business problem. Organizations have smade some short-term decisions to really see themselves through a crisis and then accelerate out of the crisis. And now you have this problem of trying to integrate that into a more sustainable way, but also not lose the advantages that you’ve clawed towards yourself, short term.
NICHOLAS BLOOM: Yes. And another thing just from watching this since the beginning of the pandemic is suddenly entering spring and particularly summer of 2021, the jobs market is completely heated up and now it’s the war for talent is back on. And that has turbocharged the remote work discussion because you see in the data what firms want is typically one to two days a week remote work on average, what employees want is typically two to three as in at least a day to a day and a half more. And employees are getting their way more and more because you’re desperate to hire. And this is meaning the number of days a week remote work is drifting up because it’s, often when I talk to companies that are saying, “Look, we wanted to have people work from home one day a week, we realized we’d never managed to hire or retain anyone. So we’ve moved to two or we wanted two, we’ve moved to three.” The pandemic is having an effect actually through the rapid recovery, forcing remote work, really as a perk in a way to retain employees.
CURT NICKISCH: Let’s talk about the timing of seeing this decision through. This will vary naturally a lot around the world and how the response to the pandemic has gone in different regions, in different countries. In the U.S. it seems like a lot of organizations are eyeing these returns after the Labor Day holiday, which is in early fall, typically when children go back to school. How do you think through the timing of making this change?
NICHOLAS BLOOM: One way to think about the timing is if you’ve ever been, as probably everyone has, in an airport and your plane has been delayed and it’s been delayed twice, you currently have a departure time. You think the plane may take off then, but you’re also thinking, well, it could be delayed again. Well, that’s what’s happening with the return to work. It’s been on the cards for some time. Currently, as of June, July 2021, it’s penciled in for most organizations for Labor Day, September. A lot of companies are getting a few employees back, a few companies like Goldman Sachs is the most obvious, they’re trying to get all their employees back. But I would say the stages of this for most companies, the best guess is you start to get 10, 20 people back from July, August onwards to test the systems out, blow the cobwebs off. From September, October, you start to get in large numbers of people probably ramping up pretty rapidly. And maybe by the end of this year, by Christmas 2021, you’re pretty much back to what you think is the long run, probably hybrid.
But this is a revolution in the future of work. A change that we’ve never seen before, and it’s going to take some time to play out. Revolutions are not quick. And so I think the whole throughout 2022, probably into 2023, there’s going to be endless turbulence. My advice for companies is the exact reverse of the Silicon Valley mantra of go fast and break things. I would be slow and cautious and boring on working from home, go for hybrid, maybe you decide that the team go for a three two plan, the very typical vanilla flavor decision, because sure, there are other things you can do, but it’s not clear they’ll work and I’d frankly wait for other firms to try it out and learn from their mistakes. So I think a lot of firms will do boring vanilla stuff throughout this year and it will become clearer what works and what doesn’t as we run into ‘22, ‘23.
CURT NICKISCH: Nick, you study management decisions as an economist. You’ve got these big sweeping views of how these very practical changes at the managerial level effect the economy. What’s your advice for a manager or leader anywhere in the world, really for how to go about keeping workers happy with remote options, but also making sure their business runs well. What’s a good mindset to approach this whole conundrum with?
NICHOLAS BLOOM: One other key piece of advice, I’ll call the Marissa Mayer story. So to set the scene, Marissa Mayer was the CEO of Yahoo from 2012 until I think maybe 2016, quite a few years, actually. And when she took over Yahoo, I know this because I spent about 45 minutes interviewing her what, six months ago. And it was fascinating. She said as she was putting in this performance management system, she discovered this whole group of people who worked from home full-time permanently. And when she started to look at what was going on, she pulled up their call logs and discovered a bunch of them had basically never logged in. It was like they hadn’t turned on their computer for over a week. And so she canceled it and it was in the media. She got, I think, quite unreasonably, a bunch of bad press, but the lesson she said she learned is very relevant to managers now.
And so she said, “Look, there’s two ways you can evaluate manager people. There’s what economists would call input based management. You can look at what they’re doing. Do they seem to be at their desk, working hard, furiously typing away, appearing to be productive, or you can use output based management, which is you manage based on what they do. Do they hit their sales targets, do they get their reports written, do they produce new products? Et cetera.” Now, if you’re running input based management, working from home is a nightmare because you can’t see what people are doing. Now, firms that are trying to do that have ended up trying to go for that spooky, horrible surveillance software, where you’re taking screenshots and using cameras to look at employees, faces, et cetera. It’s just awful.
Input based management is kind of okay, I’d never say it was actually particularly good, but it’s workable in the office. It is a disaster for working from home. Output based management, on the other hand is really from the get go has been best practice in the office. You want to reward people on what they achieve, but it turns out to be really the only workable way to run working from home. So what I’ve heard from a lot of managers is the pandemic has really accelerated their push to put in place, better performance management systems. So better data collection, better 360 reviews, more regular performance appraisals, better tracking and monitoring, et cetera, because that turns out to be absolutely essential if you want to have effective working from home.
CURT NICKISCH: Nick, this has been so great to hear about your research and thinking about this really big problem. Thanks for coming on the show to talk about it.
NICHOLAS BLOOM: Thanks very much for having me on. Great to be here.
CURT NICKISCH: That’s Nicholas Bloom, professor of economics at Stanford University. He wrote the recent HBR article, “Don’t Let Employees Pick Their Work From Home Days.” You can find it at HBR.org.
This episode was produced by Mary Dooe. We get technical help from Rob Eckhardt. Adam Buchholz is our audio product manager. Thanks for listening to the HBR IdeaCast. I’m Curt Nickisch.