Professor Nick Bloom.

The Chair: Good afternoon, everybody. Welcome to this House of Lords Home-based Working Committee. I am very pleased to welcome our witness today, Professor Nick Bloom, Professor in Economics at Stanford University. Thank you very much for joining us. This session is being broadcast live. We will send you a transcript of the evidence within a few days to check for accuracy, and you can provide any supplementary evidence—if you feel it necessary—after the meeting. Let us go straight into the first question from Baroness Freeman.

Q126 Baroness Freeman of Steventon: We know you have been studying this internationally for quite a while. I wonder if you could give us an overview of the prevalence of home-based working internationally and how that variation between countries might relate to the demographics of the workforces and the balance of, say, services and manufacturing within the different economies.

Professor Nick Bloom: Great question. There is a group of us—including Dr Cevat Aksoy—who have collected data; we have done four waves across 40 countries. When you collect data, you are never quite sure how to put it together: should it be the north versus the south, rich versus poor, et cetera? Looking at the data, probably the best way to put it together is that there is a block of what I would call English-speaking countries—the US, Canada, UK, Australia and New Zealand—that are pretty high, typically for graduates who work from home about two days a week.

I was born here and left the UK when I was 30. I now live in the US, so I do not know exactly what my nationality is any more. But in both the UK and the US, Tuesday, Wednesday and Thursday in the office and Monday and Friday at home would typically be a very standard practice. If you then go into Europe, northern Europe is a little lower—Sweden, Norway, et cetera—and it gets lower into southern Europe. By the time you are in southern Europe, it is maybe one and a half days. It is lower still in Latin America and Africa. Lowest of all is East Asia; Japan and South Korea are really quite low level, more like 1.5 days.

Why do you get this variation? We have only 40 datapoints, so this is not going to be very sophisticated, but we looked at five things, heavily overlapping with what you suggested. One was just level of economic development, and that is definitely a factor; richer countries have more working from home, partly because there are more graduates, they have better connectivity, et cetera.

Industrial structure also matters; you are right. It is more amenable for services, less for manufacturing and agriculture. Length of the lockdown, interestingly; we had a long lockdown. People were forced to work from home for more and more months, got used to it, businesses adapted and it partly stuck. We looked at density: countries that have more density tend to have lower levels of work from home; there is less space.

Finally—the one that was by far the most important—I am not sure it is the best explanation, but it is what we call cultural factors. There is something called power distance or ability to decentralise. Take Japan and the UK, which have very similar industrial structures, wealth levels, population density—I cannot remember the other factors. They look pretty similar on many things. The one big difference is this cultural dimension.

In the UK and US, bosses are much more likely to let their employees make decisions, but then if they make a mistake, tell them, get tough and maybe fire them. In Japan, there is much more face time; you are expected to be in the office. It is not the most satisfactory answer, but it looks like culture drives a lot of it.

Interestingly, when you look at foreign businesses—let us say US or UK multinationals in Tokyo—they look a lot like they do back in the US. I spoke to Google Japan, which lets its employees in Tokyo work from home three days a week. Culture is a big factor in the cross-country dimension.

Baroness Freeman of Steventon: In the data that you are collecting, you have mentioned your surveys. Do you also have data from other sources? Which countries have the best data on this and what can the UK learn in terms of getting good data?

Professor Nick Bloom: That is a fantastic question. In 2020, it turns out everyone was caught—I do not know what the correct, polite word is—in the wrong place, put it that way, collecting data on this because work from home was just not really a thing. I had been working on it since 2005, and I was trying to persuade people it was interesting. In 2019, you would talk to people and their eyelids would close; they would just get bored about it.

I did a TEDx talk and actually met Barack Obama in 2013. There were a few events, but it was really quiet. No one really measured it, and one of the problems was, to the extent it was measured, it was binary, which is pretty problematic. To give you a sense, in the UK, US and northern Europe, 60% of people never work from home. For example, the police at the front office, the Uber driver who brought me here, the workers when I go and get lunch and nurses in the hospital just do not work from home and are not likely to in the near future. So most people actually never work from home at all.

Then, 30% of people, which is graduates—like my undergraduates when they get their first job—are hybrid, as are probably most people here, most of our world; and then there are 10% who are fully remote. Nowadays, most people are in this hybrid bucket, so a question that asks, “Do you work from home? Yes or no”, is problematic. What do I say if I am working from home for two days and come into the office three days? Most of the surveys pre-pandemic were not even fit for purpose.

The other best international data that I have seen is actually things like office occupancy. JLL, a massive real estate company founded in the UK, has office use benchmarked to probably around February 2019. If you look at their data—the most recent stuff I have seen is for the end of 2024—it is the same pattern. Asia is basically back to roughly pre-pandemic levels. North America, UK and Australia are back to about 50-60% occupancy, and it reflects the fact that folks are only coming in three days a week.

Baroness Freeman of Steventon: Is that data publicly available or do you have to ask for it?

Professor Nick Bloom: It is interesting: the office industry has some great data because they measure swipes. The three best sources of data are, first, survey data; I have some really good monthly US survey data as well. The second is phone data. The amazing thing—I say this in the US all the time—is that Americans are never very far from their phones. The same is true for Brits. I am sure everyone’s phone is within six feet of them, so if you can track a phone, you can track a person. So there is phone tracking data, which is really good.

The third is office swipe data, because you have to swipe in and out in most large offices. The problem is that the office swipe data shows that offices in the UK and US are only about 60% full versus pre-pandemic, and the industry does not really want to share that data; it is very circumspect about sharing it. The best data I have seen is from what is called Castle Security. It is not actually the owner of the office data; it runs the security systems and is number five or six in the US. I do not know if anyone has seen it, but it has day-by-day data, so it is really high-frequency. It runs about 2,500 buildings across North America.

On the plus side, putting that data in the public has generated tonnes of media; the Wall Street Journal, the New York Times, et cetera, cover it all the time. The downside is that I know a bunch of folks in the corporate real estate industry are really unhappy with Castle for sharing that data. It is still sharing it but JLL has the data. I had it a couple of times because a contact sent it to me but when I asked about sharing, it said, “No, please don’t share it”. So I do not put it out in public because it would not be suitable.

Baroness Freeman of Steventon: So if we wanted to get a hold of UK data—you mentioned the phone data; is that available in the UK?

Professor Nick Bloom: Placer.ai almost certainly has the UK; it is basically phone-tracking data. The way it will do it is, it will take a building. The Lords would be a weird building, so you would not do this. In the US, what they do is look at office buildings where there is at most one floor non-office. For example, you can have a ground floor which is Starbucks or a gym or something, but the rest of the floors from two upwards have to be office. So it rules out hotels and mixed use—there are a bunch of weird buildings. But it would be at most one floor non-office, then you benchmark it to a level.

So Placer.ai in the US benchmarks it to February 2019, and you can see as it is going along that it is wiggling, because there is some seasonality. It then plummets—it was 100%—to 15% in April 2020 because of the lockdown. Who was going into an office building then? Security and janitors, and that was about it. It grinds back up again and by spring 2023 it is about 60%, and then it just moves forward. What it has told you is that you had some snapback because of hybrid. We are not back at 100% of pre-pandemic levels; we are at 60%, and particularly Mondays and Fridays are quite empty. So that data is great, and I am sure you can get it for the UK; somebody will have it.

Baroness Freeman of Steventon: We want to be able to look at that data by demographic, business type and all those kinds of things. Is survey data the best way to get at that kind of thing?

Professor Nick Bloom: It can tell you business type, because you know the location and postcode, so you can look up the business if you want. For example, Placer.ai breaks it down by retail, office and hotel. By demographic is harder because it depends on what they were legally allowed to merge in. The phone company obviously knows your name and address and can pull out your credit rating, so it actually knows a lot about you, but whether it is allowed to merge it in is less clear. It actually knows an enormous amount. If that were not a problem, it could tell you the ratio of men and women; it can guess, but generally it does not do that. Mostly demographic stuff comes from surveys, where you collect them at the same time.

Baroness Freeman of Steventon: Is the best survey data basically yours?

Professor Nick Bloom: I was thinking in advance in preparation for this. We were collecting survey data for the UK, on the US basis, with Paul Mason at King’s, but I do not know whether that is still going. The ONS is great; it is an amazing organisation and is collecting some data. Did it have a Business Insights and Conditions Survey? I am trying to remember. Some surveys started during the pandemic because there was enormous interest, but then the interest waned, their budgets were limited and they moved off it. We just have a monthly index for the US, which is definitely the best data for it. I do not know as much about the UK; somebody in the ONS would be the person to ask.

Q127 Baroness Watkins of Tavistock: It is a pleasure to meet you and thanks for coming. I just wanted to pick up and very much follow on from the previous questions. Do you have any data on the impact of both remote and hybrid on recruitment and retention, and in particular in relation to different demographic groups? You have already mentioned that a lot of this is the graduate workforce working in this way.

Professor Nick Bloom: Again, great question. I will break it down and focus first on hybrid and then talk about remote. Again, with the view that mostly people—certainly for hybrid—are typically college graduates; not entirely, but the large majority are. For hybrid, one number to bear in mind is that people value it about the same as an 8% pay increase. How do we get that number? We have surveyed people repeatedly and used relatively sophisticated survey questions, and they say about 8%. Another way to get at that number is just time use. If you think of somebody who is working, let us say, 45 or 50 hours a week in an office and has a 45-minute additional commute, if they get to work from home two or three days a week, they are saving about 8% of their total time.

There is a famous science paper by Daniel Kahneman from 2002 that values what people like and dislike per hour. The second most disliked activity is work, and I always say that is why you have to pay people to do work, because they do not like it. But their most disliked activity is commuting. So if you take those two together—people dislike commuting more than they dislike work—then you should probably count somebody’s day as the minute they leave the door in the morning until the minute they get back. If you let them work from home two to three days a week, that saves about 8% of that time. So one number is 8% but it is maybe 5% or 10%; it depends on how much you hate your commute and other things, but that is one way to look at it.

Another way—the one I know best and is the most recent and relevant—is the randomised control trial we did out in China with trip.com. Trip.com is a big, global multinational. There are three big travel multinationals: Expedia, Booking and Trip. Trip is worth about £50 billion and is Nasdaq-listed. I was working with Trip because its founding CEO and chairman, James Liang, was in my class 15 years ago, so I know him from Stanford. It was a very Stanford thing; I had no idea initially, but the guy was already worth half a billion dollars when he was sitting in the back of my class. It does not happen that often, but it was not the first time either.

His current CEO is Jane Sun, and James and I worked with Jane to do this randomised control trial. We took 1,600 people, all graduates—one third were postgraduates—people like us, very well educated. They worked in finance, accounting, marketing and IT, as in computer engineering, and they randomised them by even and odd birthday. James actually drew a ping-pong ball from an urn and it said “odd”, so if you were born on the first, third, fifth, seventh, et cetera of the month, you got to work from home on Wednesday and Friday and come in the other three days. If you were born on the second, fourth, sixth, eighth, et cetera of the month, you had to come in all five days. They held this condition for six months, and then we collected another 18 months of data after that. We published this in Nature a year ago, so this was a perfect randomised control trial on hybrid.

They found two things. One was on productivity; there was no effect at all, neither positive nor negative. The reason is that when you talk to people, there is clearly a big advantage to going in in person. They said, “I like mentoring, I like building culture, I like connecting up”, and, “Look, by the time I’ve got three days a week in person, it’s not clear whether days four and five are additionally useful”. They said, “You have to remember, days four and five are quiet if I’m at home, so I can do more deep work, and I’m saving about an hour and a half so I can be more rested”.

So the first thing is we looked at lines of code written, performance reviews, promotions, innovation scores and leadership scores up to 24 months, and just did not see anything. So it is not better or worse; you are flat. The other thing—to your point—is you looked at quit rates and they were down by 35%.

Baroness Watkins of Tavistock: Quit rates were down with both groups?

Professor Nick Bloom: No, just for the hybrid folks because they did not have to come in two days a week. For the company, Trip.com, the C-suite’s executive leadership said, “This is a total no-brainer because every person that quits costs us $50,000 because we’ve got to re-advertise, re-interview, re-train, re-get up to speed”, and they said hybrid is phenomenally profitable. So on hybrid, the nutshell is it is very profitable for businesses because it saves them a lot of money and does not affect productivity, and that is why it has become totally dominant.

Baroness Watkins of Tavistock: I really want to understand two more things. In that randomised control trial, was everybody paid the same or were the people who were only coming in three days a week paid 8% less?

Professor Nick Bloom: They were all paid the same.

Baroness Watkins of Tavistock: Is there any data on paying people 8% less?

Professor Nick Bloom: The reason I put two numbers is that we surveyed people who said it is worth about 8%. There are other ways to get at that number. There were experiments done in Brazil and the US. I will tell you about the US one because it is probably more relevant. Alex Mas at Berkeley and Amanda Pallais at Harvard did a randomised control trial in 2017. What they did was post a lot of job ads—they were actively hiring people—and they randomised salary, hours, a bunch of stuff, including whether you got to work from home. They posted hundreds of these job ads then looked at who applied, and it turned out that if you had all that data, you could basically estimate. They randomised the salary; some people at 2% more, some 2% less, some 4%, and that came up with about 8% as well. So on average, across these hundreds of job ads, if one has 8% more, it pulls about as many postings as one that lets you work from home for two days a week.

Baroness Watkins of Tavistock: So we are pretty clear that hybrid helps retention and recruitment.

Professor Nick Bloom: Yes, I would say hybrid helps retention and recruitment and if it is well organised, it is about net zero on productivity; it is not worse or better.

Baroness Watkins of Tavistock: Which is amazing.

Professor Nick Bloom: And which is why—just as a fact—80% of Fortune 500 companies have hybrid for professionals and managers, because it is profitable.

Q128 Baroness Watkins of Tavistock: I have one question, which was not actually on my original list, but because you are so interesting I just want to push it.

Professor Nick Bloom: You are letting me talk; it is a bad thing to let professors talk.

Baroness Watkins of Tavistock: At the very beginning, you mentioned nursing as something that you have to come in for. I am a nurse, but I am not practising at the moment. There is growing unrest globally between groups of graduates who have to go to work because their work requires them to perhaps do something with their hands, such as nursing. Do you know anybody who has any data on whether people who have to go to work are beginning to get better pay rises, or has nobody done anything on that yet, to the best of your knowledge?

Professor Nick Bloom: Both my sisters and my dad work in the NHS and my grandma was a nurse, so there is some history. There is a paper by Autor, Dube and McGrew. I will take a step back because it is really quite interesting. This is similar to what I teach in graduate labour at PhD. In the US—the UK is a very similar phenomenon—inequality started rising in the 1980s. It was rising and rising and rising, until it got to pretty epic, terrible levels in about 2010, at least in the US data. It moved sideways, stayed incredibly high and definitely did not go down. The inequality was rising; this is where “occupy Wall Street”, “we are the 99%” and all this anger came from.

From 2020 to 2024, you see that about one quarter of that rise in inequality fell when you define it by university versus non-university—college versus non-college in the US terminology. What on earth is going on? Living out there in Silicon Valley, in 2019, if you were a Google employee or a techie, you could get a job; 10 people would be offering you one, and they were desperate to give you the job. Now, you go to our local shopping centre, and McDonald’s, Chipotle, American Girl and Abercrombie & Fitch all have postings: “Help Wanted”.

What has happened is that there has been a big tightening of the labour market at the bottom end, and the top end has become looser. One reason is working from home. Why is that driving this? It is basically because if you can work from home, it is really quite nice, so the labour supply of graduates into work-from-home jobs has gone up, and the labour supply into jobs where you have to go in every day has not gone up. In equilibrium, that seems to have pushed down that relative wage by about 8%. That is somewhat potentially a coincidence with the 8% number I mentioned earlier, but if you look, since 2019 a job working in McDonald’s has had an 8% faster pay rise than a Google employee has.

Baroness Watkins of Tavistock: That is really helpful.

Professor Nick Bloom: That is a labour market result. My understanding from talking to Paul Johnson at the IFS is something similar has happened in the UK. I do not know the exact data, but no one is formally making that adjustment.

Baroness Watkins of Tavistock: I am going to stop because I have taken more than my fair share of time.

The Chair: Yes, we need to press on. This is very engaging but we have only done two questions.

Professor Nick Bloom: I talk to tonnes of executives and this topic comes up over and over again in their company, whereby they say, “In my company, I have a manufacturing plant and they are already annoyed because they are having to come in every day and the executives are not”. Typically, the way they have solved it is that the pay rates have gone up more for the manufacturing plants, on average by about 8%, and it happens to coincide.

The Chair: If you have anything you could send to us afterwards, that would be brilliant because this equity point is something that is exercising the committee.

Q129 I am going to take the next question myself, which is on the subject of management. Something that has been put to us in evidence is that in the UK, management techniques have not really kept pace with this new virtual and hybrid world, and I wondered what your thoughts are on that. From the perspective of the United States, do you think it is better or is it a work in progress? What are your observations?

Professor Nick Bloom: Certainly in the data we have collected over the years, British management practices are pretty good on a global scale; they are towards the top and align where you would expect. Britain's productivity is pretty high; it is just that the Americans are somewhat better managed. They have more business schools and are more cut and thrust and somewhat more aggressive. It has been a longstanding thing.

There is the Dunning report from the 1950s—written post-World War II—which was part of the bigger Marshall Plan and talks about the Americans being more productive because they have better management practices. In fact, one of my students had an enormous study on multiple papers looking at the Marshall Plan. The Marshall Plan was partly aid but partly technical assistance, and American firms were giving European firms technical assistance—even in the 1950s—to improve management.

The reason it is relevant to work from home is this. I actually interviewed Marissa Mayer. I do not know if anyone remembers it here, but she banned work from home at Yahoo in 2013 and there was this big media storm, at least in the US. It got very contentious, and I remember being involved in it. I interviewed her about a year ago and said, “Now the dust has settled, what really happened?” And she said when she took over as the CEO of Yahoo in 2012, she relatively quickly discovered this group of fully remote workers, and she said, “Well, how do I evaluate them? How do I know that they’re doing their job? Because the folks that are coming into the office, I can see they are at their desks typing away. It’s not perfect, but I can kind of walk by them and are they typing away hour after hour. When I look at the screen, is it, for example, Word or Excel, or maybe they’re watching Netflix or the Champions League? You can kind of tell.”

She said, “Look, it’s five out of 10 management; management are walking around, but it is better than nothing”. She said, “When these folks go home, I can’t see what they’re doing. So rather than measure inputs, I need to measure outputs, as in performance evaluation.” And she said, “For this group of fully remote workers I inherited at Yahoo in 2012, there was no performance evaluation. I had no idea what they were doing. The only thing we had was the core login data, as in how long they had been working on their computers.” She said, “Well, what does it show?” Apparently, no one had even looked at that. So she said, “A group went away and came back two weeks later and said, ‘We have all the data; we have analysed it’. It turns out, on average, we discovered that many people have not been logging into their laptops for two or three”—I thought she was about to say, “hours at a time”; she said, “two or three weeks”. They have another job. And she said, “That point’s come up over and over again. If people are working from home, you need performance evaluation systems”.

For example, if Dr Aksoy is managing me at home, he does not want to be breathing down my neck every minute, checking if I reply to email and if I am on Slack, and using monitoring tracking software, because it is horrible and neither of us wants to be in that situation. So he needs to say, “Nick, your objectives—teaching, research, university, service: get that done. If you get that done, you’re good. If not, we’re going to have to discuss it. There’ll be sanctions, performance reviews, I don’t know, maybe we’ll force you into the office as a downside”.

For me, it turns out to be much better too because when I pick my kid up from school or go to the dentist or something, I can do it and then make up for work in the evening. This relates very much to work from home in the government sector. One of the essential things to make work from home work—even hybrid, even one day a week—is some kind of performance evaluation system. Because if you cannot have any performance evaluation, all you can do is measure inputs, which is desk watching, and you cannot do that when people are at home.

The Chair: Do you think that is better done in the United States? Are there lessons that we could learn about management practices that are focused on outcomes and not how many clicks you have of your mouse or whatever?

Professor Nick Bloom: The UK private sector is maybe a little behind the US but not massively; I do not get that sense. I worked in the UK Treasury 20 years ago for a bit. I would not say the UK Government are necessarily well managed, but certainly the US Federal Government are a disaster, if you have been following what has been going on. Whenever I talk to public sector or federal employers, they say, “This seems like an industry that should allow some hybrid”. I am not particularly supporting fully remote, which we can come to. We have not really talked about it; it is a more extreme version.

The key thing is performance evaluations. If you are managing me, you need some way to make sure that I am doing my job, and it cannot be desk watching. As it happens, a call centre is a dream job for fully remote because there is a headset that sits on you. I do not need to see anyone to know exactly what they are doing because you randomly audit 1% of calls, you have a quality metric, and you can see the number of calls; you can see everything. So call centres are probably never going to be in the office ever again. But if you are, say, in marketing, or when I was in the Treasury, first, you could probably not do that effectively fully remotely, and secondly, it is hard to evaluate it.

Q130 Lord Parker of Minsmere: Thanks again for coming. It is really good to have you. Could you talk to us for a minute about what you have seen in the US on the return to office phenomenon, and what the curve looks like since the pandemic: if you see it changing in any particular way, maybe compared to here if you know the position here, and what the drivers are for employers?

Professor Nick Bloom: This is a hugely contentious topic. Every so often I post more data on LinkedIn and it seems to go crazy. In some ways, it is very boring. I used to make the joke that it is like Manchester United winning the Premier League; they just win it year after year, and at some point, people are like, “Okay, let’s just focus on something else now”.

Lord Parker of Minsmere: Not lately.

Lord Monks: A long time ago.

Professor Nick Bloom: My team sits below them in the league, so there are not many choices for who that will be.

The situation in the US is, we have several different indices: surveys—our survey of 8,000 folks a month, which we direct to the current population, so it is pretty representative—cell phone mobility data, Placer.ai, and then office swipe data. They all basically show the same thing. They all show that if you benchmark 100% pre-pandemic, it drops to about 15% or 20% in March and April 2020 because of lockdowns. It then gradually rises, gets to about 50% or 60% in spring of 2023, and then it moves sideways. I am aware that the media is full of these stories of everyone returning to the office, and every time I speak to people, they say everyone has returned to the office. The data does not show that.

Why is there the disconnect? I talk to tonnes of journalists, and it is like incentives 101. So if you are a journalist, you get rewarded for writing a piece that is going to be clicked, commented on, has views, et cetera; they are assessed on that internally. If you write a piece saying, “Goliath National Bank cancels work from home and sends everyone back into the office”, people furiously click on it, comment, send it around, and there is a huge debate; it does really well. If you write one saying, “Goliath National Bank extends work from home for another year”, it sinks. It is like shark attack; it is a completely asymmetric incentive to report one type of story and not the other.

With the Atlanta Fed, we have been asking a panel of about 500 US firms about their policies on a regular basis, looking back and looking forwards. Our most recent survey is from February 2025, and what you see is this histogram of changes of days in the office, and it is centred about zero. So there are a bunch of firms that are reducing remote days because they are doing a return to the office, but there are other firms that are increasing it; it is just that the first lot gets picked up in the media and the second lot does not.

Who is the second lot? The second lot is people whose leases are expiring because the average lease length is 12 years. Every lease that expires now was taken out pre-pandemic, and it is very easy to say, “Well, now I can renew the lease, why not halve the space?” There are a lot of smaller, younger firms. So you just do not see this half.

The other, milder thing that is a bit more complicated is misreporting. I know Zoom reasonably well. I met Eric Yuan—the CEO of Zoom—two or three times and talked to him and had lunch with him back in September 2023. In Zoom, on Mondays and Wednesdays, the sales team comes in, on Tuesdays and Thursdays, the engineering team comes in. Zoom got absolutely castigated in the media in September 2023 when it announced that; the headline was, “Zoom cancels work from home”. It did not cancel work from home; it actually formalised and enshrined hybrid. When I asked Eric about it, he said, “Well, of course we have to be hybrid because that’s what most of our clients do. We need to eat the lunch that we make”. It is natural and good for the people to be in; they just do not have enough office space to have them at work. But if you read the news, you would think Zoom was fully back.

One of the few exceptions is Amazon. Amazon has fully required people back and is a bit of an outlier. First, it has not completely executed on it. It has had problems with lack of office space, car parking, all kinds of stuff. People are non-compliant. I know people there and there’s non-compliance because it’s very hard to comply. Because—let’s say you’re my manager, I am a star employee and I have said, “You know what? If you make me come in five days a week, I’m quitting.” You care about your team performance because that affects your bonus and promotion. So you say, “I’ll tell you what, Nick, you can come in three days a week. The other two, I’ll say you’re travelling for business, seeing clients, going out to conferences”. How is HR going to know? It will not, so I would just remain hybrid for six months.

The other issue with Amazon is it was primarily aimed at reducing headcount. So if you look at Andy Jassy’s memo when he put it out in September 2024, he talks about a full return to the office and about being too big. He says, “We need to de-layer our management and be more nimble”. Interestingly, basically no other big firm has copied it. So that stood alone; it has done it and no one has really followed up.

The only other high-profile replication is the US Federal Government for the same motivation; they both want to reduce headcount. So Jassy basically was hoping a bunch of people would quit because Amazon over-hired in 2023, and the US Federal Government want a bunch of people to quit. But outside some exceptions, hybrid is here to stay, therefore office use is pretty flat and stable now and has been for the last couple of years. So at this point, it is very dull. Every time I put one of these things, the line is still flat and people seem amazed, but it has not really changed.

Q131 Lord Parker of Minsmere: Just one quick question: is the settled position in the US that, when you talk to them privately, the CEOs actually want this, or is it an equilibrium of interests between what workers want and what employers can push on them?

Professor Nick Bloom: I would say it is more of the latter. It is a labour market, so if a CEO were to force folks in five days a week, they could probably do that; they just have to pay employees, say, 8% more to avoid attrition. Given productivity is unchanged, it is therefore not profitable. If I am a CEO and I do something that reduces stock price, I am at risk of getting kicked out and replaced, so it is just not in my interest to do it.

You also notice the CEOs in the public domain speaking out vocally about it and there is very few of them: Elon Musk, Jamie Dimon and David Solomon from Goldman Sachs. Jamie Dimon was at Stanford about six weeks ago and he talked at an event I was at. He said some people are in fully and others are hybrid; he was much more moderate.

I have spoken to David Solomon directly for a while. Since that aberration comment, he has stayed out of the debate, so there are one or two shouty CEOs. Musk says crazy stuff all over the place, so if you set that aside, you are looking at Jamie Dimon, who generally is really great. I am a big fan of his, but his work from home stuff is moderate. They are not typically pushing employees back fully but are mostly against fully remote, which has been a longstanding thing.

Hybrid is where it has settled, including for us. At Stanford, we have 20,000 employees: 10,000 come in every day; they have to—security, janitorial, food service—8,000 faculty and staff are hybrid; and then we have 2,000 fully remote, who are data entry, call centres—back-office stuff that, to be honest, is so much cheaper being fully remote. It could be 20% less efficient, but if it is 40% cheaper, it is worth doing it. Generally, the view is that fully remote is probably less efficient, but it is so much cheaper that it is worth doing.

Q132 Lord Farmer: Thanks for coming; it is great. The UK Government want growth, right? We need growth in the economy. What is the productivity impact of remote and hybrid working at an organisational level, and what other metrics are useful in this context, such as profitability and efficiency?

Professor Nick Bloom: My view in general is, I would not regulate what firms do; I would just let them get on with it. If you look at what they are getting on with, the typical Fortune 500, FTSE 100 firm has managers and professionals on hybrid often coming in three days a week; four and two are also common. Outside tech, two days in the office is rare, actually; most people are coming in three or four.

I worked at McKinsey in London for two years. At McKinsey, you partly do what your client does; so if your client is working from home, you will probably work from home, but you would probably be in the office three or four days a week. The back-office folks—the folks who are in the windowless rooms in the middle of offices—maybe just do not need to go in. Why does that make sense for McKinsey? Because you can shrink office space and hire them in north Wales or Bangalore or something and it is cost-effective for you. For companies, the long run is 60% of the body count will primarily be fully in person, but for executives you are going to see hybrid, and for a few back-office support staff you will be fully remote.

For UK growth, I think on average this is net positive mainly because of labour supply effects. In economics, we have this classic thing: output equals productivity times the number of people working. I do not think working from home has a net positive or negative effect on productivity overall; it is pretty marginal. The real big benefit is on “L”, the number of people working. We have some work showing in particular folks with a disability have a much higher preference to work from home if they can.

In the US, there are about a quarter of a million people working who have a disability. If you look at who has a disability, it rises sharply with age. So with an ageing population, there are things like bad back, vision issues, ambulatory issues, colostomy bags, stuff like that. It is not like people like this would never go into the office; it is at the margin. I interviewed someone—this was the most striking case—who said he was disabled from the neck down once after a car accident. He told me it took him three hours to get into the office because his carer had to come and get him out of bed, wash him, clothe him, take him to his car, and his dad then drove him in. He was on hybrid in marketing, and he said, “Look, I’ll do that two days a week as long as I can work from home three days a week, because it takes me 20 minutes to get out of bed. If I had to come in all five days, I probably wouldn’t work”.

So my best guess for the biggest margin on UK growth in general would be increased labour supply, because people would delay retirement for a couple of years with a disability. Folks on maternity leave would maybe go back a bit sooner. I could see it increasing labour supply by 1% or 2%, which is a lot if you spread that.

Lord Farmer: Talking about profitability and efficiency, BlackRock ordered all its worldwide managing directors back full-time last week. Do you have data on profitability and efficiency? Are such companies more efficient? Are they making more profits at less cost?

Professor Nick Bloom: There is correlational evidence. In general, if you look at the correlation, the flex stuff has data in about 1,000 S&P 1500 companies. In the cross-section, what you tend to see is that firms with more work-from-home days are more profitable and growing faster. But I do not want to claim that is causal. Why? Because tech firms tend to grow faster and be more work from home, so I actually think that is misleadingly pro work from home.

I spoke to David Solomon at Goldman Sachs and Peter Orszag at Lazard, who I know reasonably well. I can get why BlackRock is doing it; it is a mentorship model, and you need managers in to teach the newbies. Whether they need to be in five or four days is less obvious. I often hear that they say five because they want four.

Lord Farmer: Mentorship is an important point.

Professor Nick Bloom: I totally agree. I am not promoting or pushing fully remote. Fully remote works for some things, such as call centres. We have a study that shows that, even for fully remote in call centres, it is best to have them in the office for the first six months. But that is only 5% to 10% of the economy.

My two oldest kids are 19 and 21. I am advising them, “When you get a job, be in the office three or four days a week”. It is not obvious that fifth day is helpful. I do not know whether BlackRock is going to formally enforce Friday. My view is that if they are doing that, they probably think it is profitable.

If you look at Fortune 500, the firms are mostly hybrid but there is a mix. There are some such as Airbnb and State Farm that are fully remote, and there are others like BlackRock, but the mean—where the mass is—is three or four days a week. You are right: mentorship is probably the most important, and the other is innovation. If this were on Zoom, we would not have been chatting outside beforehand; half of you would have been doing email, and I might have even been doing email. So I totally agree; I am not against in-person.

Q133 Lord Monks: In a sense, there is still a degree of novelty in people’s minds about hybrid working in particular. It is several years old, but it is not well entrenched. I have heard it said by friends and acquaintances of mine, “We’ll get them back in the office when we can”, almost waiting for opportunities to round them up and get them in, et cetera. Does that ring any bells with you at all? I doubt it from what you have said so far.

Professor Nick Bloom: A lot of people have made that sentiment but I hear that less and less now. Ultimately, I am an economist, so I mainly talk to businesses. I do an enormous amount of executive education and talking to companies. I am in favour of work from home because it is good for workers and employers in general, but that is not the way you persuade executives to do it. The way you persuade executives is to figure out what is profitable for them, and it turns out that hybrid is profitable.

If you were to take BlackRock and make it fully remote, it would probably be a disaster so it would never do that. If BlackRock went from four days to five days though—I do not know if it has gone for five days, and whether it is going to get five is a different question. But generally, most profitable companies have something like three or four days in the office. That fifth day in particular—it just annoys people to force them in on Friday and does not seem to improve productivity.

I poll my undergrads and ask them, “How many days do you want to work from home?” They mostly say, “one”. They say, “Look, it’s social in the office. I want to get mentored, and where the heck am I going to work? I’m sharing a flat with, like, four other people; where am I going to go?” One or two of my MBAs said, “I want to be fully remote”, but they do not want to be at home; they want to be on a beach in Mexico. It is a totally different plan.

I do not think it is changing, because of capitalism. Ultimately, it is profitable to be hybrid at this point. For big companies, it is more complicated and depends on the task. It is profitable for 60% of people to come in all the time because they cannot do it otherwise; but, typically, for about 30% to be hybrid, coming in three, maybe four days a week, and for probably about 10% to be fully remote. Why is that profitable? Because even if they are 10% less productive, if they are 30% cheaper, say, because you are hiring them in north Wales rather than central London, then net, you are up.

I interviewed a guy in a taco restaurant who said he runs a whole chain of taco restaurants down in LA, and he said, “I love working from home”. I was surprised that he would say that, and he said, “Everyone who cooks, cleans, and serves food has to come in every day, no exceptions. But my back office, I have just sent them all remote, and they are basically all based in Alabama”. Alabama has a much lower wage rate than Los Angeles, and he said, “They are about 80% as efficient. I am paying for no offices, and my wage bills are 70% of the hourly rate. So net, I’m way up”. That is a classic example.

Q134 Baroness Nye: You have explained why remote and hybrid working can influence economic growth—which is obviously the holy grail of all Governments—by broadening participation of the economically inactive or disabled groups, et cetera. But it seems like the US is better than the UK at doing that. What are the reasons for that, and are there any initiatives the US does that we should replicate over here, or that we can recommend in our report, that would help broaden that participation through remote and hybrid?

Professor Nick Bloom: It is a good question. It is not clear to me that the US is any better. All the data and evidence I have seen in the US and the UK looks similar. I have talked to a bunch of folks in Australia, and it is similar. The obvious thing is broadband connectivity. It is a no-brainer; you just need to make sure the whole country—including rural places—has really strong connectivity.

I am not in favour of regulating to make businesses improve work from home; they will do it because it is profitable. I know some countries have done that, and it is just an excessive burden. I would facilitate it—as in support it—but no more. If I were the Government or the UK Civil Service, I probably would not be that in favour of fully remote, but I might have people work from home one or two days a week. Management practices are not always the best, and I worry about people having two jobs, et cetera. Connectivity is the main thing.

The other thing the UK does not have is the tax nightmare the US has. The US has this awful tax issue. I do not know how much people know, but there is a state, federal and local tax. It is really weird, but New York taxes on where you live and California on where you work, so if you are a remote worker, you can end up paying tax twice, and there are all kinds of complexities. I do not think the UK has that. It is the legal and tax thing, just supporting it, but I would not force anything. I do not think the UK is bad on this; it actually looks fine, has relatively high levels and is getting it roughly right.

Q135 Lord Fink: Thank you very much for coming in. The question I was given has largely been covered, which was how have spending and property usage changed in US cities since the pandemic and how might this manifest differently in the UK cities, given the different structures, which you have covered. So I guess we are down to what you think is the Government's role in this in supporting economic growth. You said, do nothing for private companies, so again the whole thing seems to be covered. So if I can switch to a slightly derived question: do you see any profound effects on property markets, both commercial and residential, and spending patterns in the hospitality industry changing dramatically as a result of the hybrid working?

Professor Nick Bloom: On the growth thing, it is pro-growth through labour supply. If I were the Bank of England, I would think this was a positive thing.

The Chair: Why?

Professor Nick Bloom: It may increase labour supply by 1%. It is not enormous amounts, but UK growth is so anaemic that a 1% spread over five years is a major thing. So that is the big positive.

In terms of cities, we have a paper in the National Academy of Sciences—which is global, and we have some UK data because we have MasterCard data—showing activity shifting from city centres out to suburbs. We call it the doughnut effect. It is like an American doughnut: stuff drops in the centre and goes up in the suburbs. That seems to be permanent. I spoke to the board of Panera Bread, which is a bit like Pret-a-Manger, and it just said, “Sales and business are fine, but we have just seen people are no longer buying as many sandwiches in city centres, and they are shifting out to the suburbs”. It also said catering at offices is down on Mondays and Fridays and way up on Tuesdays, Wednesdays and Thursdays because people are trying to bribe folks with food to come in.

So there has just been a relocation of activity. Some folks say, “Oh, isn’t that terrible? City centre property prices are down”. I do not think it is terrible because in the UK—as in the US—we had this problem that people could not afford to live in the centre of cities. Who do I most want to live in the centre of London? It is not actually techies and bankers, but people in essential services. My brother-in-law works in the Met and needs to go in in person; he cannot work from home. He is living in High Barnet, but somewhere like that is going to price out. If city centres become a bit cheaper because some techies and bankers move out to the suburbs, that is actually good.

So there are changes, but I do not see it as particularly problematic. Effectively, it has made it easier to access space in our suburbs, so people have moved out, which has actually reduced the demand for space in city centres. So that is a permanent effect.

Interestingly, the UK is much better adapted to that than the US. Something I did not realise until I went to the US is that US cities are very small in terms of their fiscal footprint. San Francisco, for example—which used to be under London Breed but is now under Daniel Lurie—is 50 square miles and does not include the Golden Gate Bridge or the airport. You think, “What the heck?” It is so small, and Candlestick Park—where the 49ers play—is not even in the city. The reason it is relevant is that if people move out of the city, they do not have to go very far to move out of the city tax base, so they stop paying tax to the city of San Francisco and pay it to some suburb. So all the suburban cities around San Francisco have done great because all this money is flowing in, but the city itself has done really badly. Whereas if you look at London, it is 500 square miles. If you go from Zone 2 to 4 or 2 to 5, you are still paying money into London.

So it turns out fiscally having a bigger area actually makes you more robust, and it has increased the size of what you think of as a natural labour market. To the extent there is a policy out of that, you do not do what Americans have done in these really tiny cities because it makes it really hard to deal with this. New York—as a result of some historical accident—has these five boroughs that merge voluntarily, but most US cities are really small and so they have seen a huge tax drop in the centre. Boston was in the news; it has lost $1 billion of tax revenue and is thinking of taxing residents, but there are not so many residents left. In one case, the UK structure is actually better, having these bigger city structures.

Q136 Baroness Featherstone: This is on externalities, and some are a bit left field in terms of the impacts that remote and hybrid working might have on things like fertility. I note you said we might get an increase in population, which, actually, given the drop, would be very good. Where are the gaps in the research on burglaries and the environment? Where do we find the data?

Professor Nick Bloom: Fertility is a hugely important, really interesting question.

Baroness Featherstone: I had not even thought about it.

Professor Nick Bloom: You saw there was a bit of a fertility bump. Again, I know the US data better than the UK, but there was a bit of a fertility bump during the pandemic. People certainly thought it would go down and it went up. We have been collecting data in the US and globally, asking people how many kids they have had since 2015, which year they were born, and how many they were planning to have, along with work from home data, collecting wave after wave.

What you see is, on average, if somebody works from home one or more days a week, the total number of children they plan to have by the end of the cycle is about 0.2 more, and if their spouse works from home, it is about 0.2 more as well, and it is gender symmetric. If you take a couple such as my wife and I—we were both fully in person when we were in London and are now both hybrid—on average that would be 0.4. Those numbers are not causal— just to caveat as a researcher. We have run regressions and tried to control for stuff, and who knows whether it is causal right now. But if those numbers are true, 0.4 is just a huge number; that is actually a massive difference. The UK’s replacement rate is 1.6 or something.

Baroness Featherstone: We are dying out.

Professor Nick Bloom: I do not know what it is. The US is 1.7; 0.4 up takes us to 2.1, which is Indonesia; 0.4 down takes us to 1.3, which is Italy; so you can imagine those. For south-east Asia, it is a much bigger issue; a place such as Korea where it is 0.9—is in rapid population collapse. So the fertility thing is enormous, and the reason why is it is just easier to look after kids.

Baroness Featherstone: Who is doing the research on that apart from you?

Professor Nick Bloom: Academically, it is hard to do because of the timing. Coming back to the discussion we had earlier, it was not obvious that work from home was here to stay until 2023-24. So everyone was still saying, “We’re all going back to the office, et cetera”. You can see it is flat, but the media is full of it. So you are a couple, and you say, “Well, I’m going to get called back by Goliath National Bank; let’s not make child plans on it”. So by 202425, you are maybe convinced that you are going to stick around; you then maybe plan to have another kid or two, and the date becomes another year. So you are looking at research that is going to happen in 2028-29. We have been trying to collect data now to kickstart this, but I actually think, when we look back 20 years from now, that may be one of the biggest benefits because—for Europe and Asia in particular—fertility rates are really down.

I was actually giving a seminar yesterday in Northwestern University and one of the people there had a really fascinating paper. There was an extension of maternity leave in India, and she was talking about having a very radical mandatory extension from 12 to 26 weeks. Initially it sounds great. What she found was there was a big drop of employment of young women because firms then said, “Well, this woman may have another baby, therefore we’re not going to hire her”. So that has all these horrible, unintended consequences as a maternity policy. If you have work from home, there are much fewer of these unintended consequences. I see it as an instrument of supporting birth rates.

Baroness Featherstone: Do you have anything on the other things, like burglaries?

Professor Nick Bloom: Jesse Matheson at the University of Sheffield has a fantastic study on burglary; it is just entertaining. I will tell you about it. It turns out criminals have noticed too. He basically gets data from the Met and Yorkshire Police and looks at areas with more or less work from home rates, and it turns out that if you see an increase in work from home, house burglaries go way down. Partly it is because no one wants to break in when there is somebody sitting, working away in their pyjamas in the living room. He actually calls the paper, Eyes on the Street, because you are working in your living room, you look out and say, “Who’s that with a crowbar at the Perkins’ house? What are they doing?” So apparently, burglaries are way down, but some crime has shifted to city centres, because exactly the same people who are phoning the police are now at home and not in city centres. So criminals have even noticed this and moved where they work.

We have another study—actually it was on Panorama, which also had UK data in it—which is golf. Golf has exploded. This comes back to some discussions about performance evaluation. Baroness Scott is my boss. As long as she evaluates what I do, it does not really matter if I am a golf fanatic and disappear out to play golf for two hours in the morning as long as I make it up in the evening by watching less TV. So I get my job done, and am actually happy to get to play golf when it is empty, in the morning.

Baroness Featherstone: Yes—it is if you play golf and watch TV.

Professor Nick Bloom: That is where there is trouble. So you see a big explosion in golf, going to the gym, using hair salons, all this kind of stuff, during the day, which is not necessarily bad as long as you have a performance evaluation system.

Baroness Featherstone: Do you have anything on the environment?

Professor Nick Bloom: The environment is a really interesting question and is not as obvious as it seems. Initially, I thought, “Well, this has to be positive because there’s less commuting; this is a no-brainer”. There is definitely less commuting; you can see less car traffic. The complicated thing is, yes, that is true, so you lose in the short run and have less pollution from transit, but there are some other offsetting effects.

One is HVAC systems. It turns out the offices are heated or air conditioned anyway, whether you are there or not. So if people are at home—this applies probably less in the UK—but in the US temperatures are more extreme, so there is a lot more air conditioning or heating going on. So that is a negative because now you are basically HVACing two buildings.

Secondly, people are now living further away. This is the doughnut effect, so they are commuting less but their distances are going up. Thirdly, when you live further away, you tend to live in a larger residence, which is less environmentally friendly. Everyone says San Francisco is the most environmentally friendly city because it is so close in and the temperature is so moderate, whereas living in some house way out in the suburbs in Kansas is an environmental disaster.

I have not seen any rigorous studies because there are multiple margins. There was something in the proceedings of the National Academy of Sciences about a year ago that found it was a net good for the environment, but it did not take into account some longer-run effects.

Baroness Featherstone: It is all quite new in terms of data, is it not?

Professor Nick Bloom: Yes, and you have to take into account behaviour responses. Yes, there is less transit, but you have to add in the HVAC stuff, the change in the distance and where people are living. I would not be surprised if it comes out of the wash at about zero, ultimately.

The other angle is business travel; that is probably net positive. I talked to Amex GBT—American Express Global Business Travel—which is again doing fine, and there is less of “I fly to see you to spend an hour in a meeting trying to persuade you to buy whatever piece of software or whatever I am trying to sell you”. Instead, my company—let’s call them Tech Software Inc—has a retreat once a quarter because all the employees claim they are not connected, and they all fly into some place and out again. So business travel between different companies is down, but internal business travel has gone up and as a result has not really changed.

Q137 The Chair: The final question: if we get you back in 10 years’ time and have some of this conversation, where do you think we will be? Do you see a trajectory?

Professor Nick Bloom: I actually think the 10-year forecast is much easier to make. We have data going back to 1965 in the US on work from home levels, and if you plot it in a log scale—growth rates are straight lines—it is basically a straight line up until about 2019. Why? So it is about 0.5% of days for work from home in 1965; it is a few outliers. Hardly anyone was working from home. It rises, rises, rises.

I am one of four kids. Both my mum and dad worked full-time, and I used to talk to them and I said, “What was it like in the ‘80s when one of us was sick?” And my mum was saying, “It was terrible. I had to stay home or your dad would stay home. He’d be carrying piles of paper; phone calls were really expensive. Work from home was awful”. Then they got an Amstrad green screen in the early 1990s; I remember that. I remember the first game was that golf game. That was when I became interested in it; I did not use it otherwise. Then there was the internet in the early 2000s; video calling; Zoom was 2010. Technology has got better and better and better, and that has roughly doubled work from home rates every 15 years, so that is the long run trend up to 2019. It then jumped up, and came back down again. But if you predict 10 years out, there is a bunch of amazing technologies in the pipeline that would just raise it.

I visited Google and it has this thing called Starline, which is this big 3D portal. There is another company, Nora, which had a full, very thin wall, 12 feet by 12 feet. I had a meeting with someone; it was a startup. One of the co-founders was in New York and I was in San Francisco, and you could walk around. If we met on Zoom, you would just see a head, but if you see someone’s whole body and height it is better.

That tech is coming. Some other company—some VC I was talking to—is investing in a company doing holographs, like Jedi Council type stuff. That is not going to happen this year, but 10 years from now, it is clearly going to be up because the technology is a lot better.

Baroness Freeman of Steventon: Looking at the longer term, you have talked about lots of things that are changing: management styles, lifestyles and how services will work through the week versus weekends. Can you just expand a bit on how society might change around that?

Professor Nick Bloom: An interesting fact is, I talked to Sainsbury’s and we see it in our US data that Friday is now the most popular shopping day. I do not know if you are aware of this, but Sainsbury’s used to say Saturday; it is now Friday. Friday is now peak day because of work from home, so there is a whole bunch of shifts that are happening. I do not think any of them are that radical.

At this point we have seen them all, so now it is just this gradual increment, longer fertility, more disability folks working, people working later. My parents may have retired a bit later if they could have worked from home two or three days a week because it was just not as arduous. Generally, the city centre is less expensive and some more essential services can live there. It is mostly pretty positive. This is not some techie, fully remote utopia; it is mostly hybrid. Hybrid is where it is at.

Fully remote is rare. The interesting thing about fully remote is that in 10 years' time it may even be lower in the UK because anything that is fully remote now is at some risk of offshoring or AI. So 10 years from now, we may see mostly hybrid and fully in-person. I do not know if anyone has used an AI call centre, but they are now pretty good actually. I would rather speak to a person, but you have to wait and it is crackly and increasingly they are in the Philippines or something.

Baroness Freeman of Steventon: That brings up that schism in society between those who can and those who cannot work from home and do hybrid, and within a country; you were talking about hybrid in Alabama versus Los Angeles.

Professor Nick Bloom: You are right. That is a real schism, and the labour market is partly offsetting that because the labour supply has gone up. Hybrid is so appealing that labour supply has gone up for hybrid types and, at least in the last three years, that has actually meant that non-hybrid type wages have gone up. For example, McDonald’s salaries have gone up faster than Google’s salaries, which is amazing. We have not seen that for 30 or 40 years. The David Autor paper is really striking. With inequalities like that, it goes sideways and there is a really sharp reversal. The most likely cause is work from home.

The Chair: Thank you very much indeed. I can tell the committee has very much enjoyed your evidence, and we appreciate your coming straight from the airport.

Professor Nick Bloom: Thank you very much for the interest.

The Chair: Thank you very much indeed, and we will bring the session to a close.