Examination of witnesses

Raj Patel, Grant Fitzner and Lorna Landells.

Q25 The Chair: Good afternoon, everybody. Welcome to this House of Lords Select Committee on Home-based Working in the UK. I am very pleased to welcome a panel of three experts to give us evidence today. Rather than introduce you now, it would be great if you could introduce yourselves when you speak for the first time. Again, thank you very much for giving us your time. To kick us off with the first question, I call Baroness Manzoor.

Baroness Manzoor: A very good afternoon and a very warm welcome. My question is in two parts and I will put both parts to all three of you. It would be helpful if you could outline what sources of data you have on remote and hybrid working. What gaps do you think exist within this data—for instance, the coverage of demographic groups or the number of days per week that people are working at home? The second part of my question is: does the data address other effects of remote and hybrid working, such as the impact on housing prices, office rents, crime, transport, a reported increase in domestic violence and, perhaps, the environment. It is a wide-ranging question.

Raj Patel: Good afternoon, everyone. I am the Associate Director of Policy and Impact Fellow at Understanding Society, which is the UK Household Longitudinal Study (UKHLS). Essentially, what that means, if you are not familiar with it, is that we track a panel of people over time, surveying them every year. The study started in 2009, and it builds on something called the British Household Panel Survey that has been going on since 1991. We collect a vast volume of data.

Interestingly, the way to think about the question about the data is probably to divide it between pre pandemic, what happened during the pandemic and then post pandemic, because that probably reflects how the world of data collection itself changed. In the study, pre pandemic we were always interested in where people worked from. It did not answer the question about how many days people worked at home but it asked whether you worked at home, predominantly at employers’ premises or somewhere else—for example, if you are a sales rep or someone who is on the road all the time. You also were asked questions, in some cases, about whether you were self-employed or an employee because the self-employed have a very different pattern of responses. We have data going back 33 years on simply where people worked from, and you can see that prior to the pandemic there was a gradual increase of people responding saying that they worked from home.

When Covid hit the scenario changed. Certainly, when Understanding Society started, we were very interested in flexible working, so we asked a whole bunch of questions about different types of flexible working. We asked about a wide range of choices, whether people might be doing part-time working, work in term only, job sharing, flexitime, etc.

I appreciate that the inquiry is not looking specifically at flexible working, but in the broader context, one of the choices was, “Do you work from home regularly?” There is data on that question going back to 2009 and one could look at differences by age, occupation and regions of the country, because the Understanding Society study is representative of each region, including the home nations. We know that a lot of patterns vary by type of industry. The service industries offer greater choice and options for working from home.

When the pandemic hit, like other studies, we had to switch to online data collection. We were beginning to do that anyway well before the pandemic, but we were able to ask questions about people’s personal circumstances in the context of the pandemic. That is when some of the newer questions were introduced like, “How often do you work from home?” rather than whether you work from home or not. Members could say “always” or “sometimes” and so on. During the pandemic, we also asked questions about whether people were able to get more work done or less work done when they were working from home. There is some data that can tell you a little bit more about personal productivity in that new work setting that was the case.

Post pandemic, we introduced additional questions to reflect the fact that hybrid working was now far more predominant. We now ask a question about exactly how many days per week people work from home, whether it is one day, two days, three days, four days, or less often than that, never and so on. We can get into the data about hybrid working now better than we were able to do before then. I hope that gives you a very quick canter through the data we collect.

You also mentioned the question about how we could get information on wider impacts. It is worth mentioning that Understanding Society is a multi-topic study. By that I mean that we are interested in the broad dimensions of people’s lives. We focus on five core areas with a lot of questions. The areas are education, employment, health and well-being, income and wealth, and family life.

Then there are other questions where we go into less detail but they are equally interesting and relevant to this committee, things such as commuting behaviours and transport and environmental attitudes. We ask about cognitive factors and leisure, and so on; there are certainly all those sorts of things. I suspect that the kinds of things you could examine are what is happening to family relationships, for example, parenting styles and whether people eat dinners together more often or not as a result of being able to work from home.

One can look at gender equality: what is happening to pay progression; what is the difference between men and women, and the options of being able to work in this way. We ask people whether their employer offers particular types of flexible working, including working from home and secondly, whether they take it up. It may be the case that an employer offers it but people may choose that it is not right for them, so there may sometimes be a question of mismatch.

You could look at health and well-being. We collect a lot of data on things such as sleep, loneliness, nutrition and mental health. I have already mentioned transport and commuting—how long does it take someone to commute to work—and we ask about both time and distance. We also ask about mode of travel. One could look at wider impacts on climate change if one wanted to.

We collect data on energy use, so you can see what is happening as a result of people working from home and how their energy use is changing. Of course, when one looks at extending the working age now in an ageing society, people may have additional responsibilities that start to emerge later on in life. We ask about whether people have caring responsibilities or not and, indeed, whether they have any kinds of disability or not, for instance, which may not be necessarily age-related.

One of the last couple of things is housing. We can observe where people move because we track people across their life course.[1] However, one of the questions we ask when people move house is, “Did you move because you wanted more office space in your house or not?” There are some relevant aspects, but they do not quite get to the house price question that you raised. Lastly, there are questions about life satisfaction so one could see whether some of these things improve someone’s life satisfaction or not.

That is a pretty quick canter through a wide range of things that the study collects.

Q26 Baroness Manzoor: Could you identify any gaps, Lorna?

Lorna Landells: First, an apology because I have a cough that comes and goes. I will try not to cough into the microphone.

I can give you an idea of what we have done at Remit. I am a partner at Remit Consulting LLP and we deal primarily with the real estate sector. We conduct a lot of research into returning or not returning to the office. Since May 2021, we have been collecting daily data on office attendance. We started out with about 150 multi-let offices in nine cities around the UK. That comes out at surveying about 200,000 people, so it is not a bad sample to extrapolate from.

We have been collecting that data since May 2021. We have a time series that goes back quite a long way. An important gap to note at that point is that we tried to get pre-Covid data when we started because we wanted that baseline. However, nobody had it or they might have had it but they did not know where it was or they could not be bothered to provide it. They probably had other things to worry about at the time. We have had to go with what the accepted norm was, which was that pre-Covid, occupancy varied between 60% and 80%. You would get the odd high-quality office that was occupied perhaps nearly 100% of the time, but by and large, taking into account holidays, summertime and big events, occupancy averaged out at 60%. I am fairly sure that anybody who has worked in an office will remember that Fridays have never been fully populated.

That is one of the areas that we have done research in. We also conduct an office worker survey. That survey goes out not just to property people but to anyone who works in an office. The survey has shown that on average people between the ages of 18 and 24 really want to go into an office. They really want to get out of their houses. We can always make suppositions about why that is. It could be because they are in shared accommodation and do not have their own office at home. They may be working on the kitchen table and so forth. Equally, people over 50 like going into the office, which is interesting. The group that sits between those age groups prefer to keep it to a minimum. However, overall 95% of the people we surveyed like going into an office for at least one day a week. That is the other piece of work we do and we can provide more detail when we send you our written submission.

The final piece of work we did was with the IPF, the Investment Property Forum. It was into the productivity of the office worker. I would love to be able to tell you that I have something that you can use but the reality is that people do not know how to measure the productivity of an office worker. Does it impact the bottom line? That is a good question, but does it? Most of the research comes out as being subjective and people say, “I have had a really productive day”, whereas they have just ticked four things off their list. We have no means of measuring how productive that work really was.

Within that report, we came up with a workable definition of how an office worker could be measured. The most important aspect—and you could possibly influence this—is that employers need to give measurable, clear objectives that have a direct impact at some level on the bottom line. Then you can start measuring it. If employers can tick off those objectives, we really tell if employees are being more productive.

To answer your second point about the impact, I can only give you anecdotal and peripheral research that I have looked at about the impact on people’s home lives and what goes on if they are working from home more. Physically disabled people benefit hugely from being able to work from home and more of them have been able to go back into the working environment. Oddly, people who perhaps struggle with social interaction are feeling more lonely, as a rule, because they are not getting accidental interaction with people. I believe that domestic violence is on the rise.

There is a big issue around childcare that the Government could address. I believe, and please correct me if I am wrong, that you have to be earning a minimum of £8,000 a year before you are entitled to get the 30 hours of childcare. But what are you supposed to do with your child while you earn that £8,000 a year? Things like that are all vitally important in considering people working from home.

Grant Fitzner: I am the chief economist and director of macroeconomics and environmental statistics and analysis at the Office for National Statistics, which is a bit of a mouthful but means that I am responsible for about half the UK’s economic statistics or have oversight of them.

Of course, this has been an issue of some interest to the ONS, and not just since the pandemic but prior to that too. We have had questions on the Labour Force Survey and Annual Population Survey for some time, although slightly different to the questions that we currently use. They ask whether you have mostly worked from home in the last week or have spent at least one hour working from home in the past week. Prior to the pandemic, 6% of people said that they mostly worked from home and 12% said they had worked for at least one hour in the past week from home. They were low but not insignificant numbers.

Those numbers increased enormously during the pandemic and something new called hybrid working emerged. If you look at Google Trends, hybrid working was not a phrase that anyone used prior to 2020. Now, of course, it is a very common working practice. We have been using surveys that we introduced in 2020, during the pandemic, for the very reason of capturing some things for which there were data gaps.

There have been two surveys. One is the Business Insights and Conditions Survey, which we initially set up to get insights into the pandemic and its impacts on business. Over time, that has evolved to become a broader business survey covering a range of issues, including working from home. The second survey was our Opinions and Lifestyle Survey, which we produced every week or fortnight during the pandemic. It covered a wide range of questions, including whether people were working from home or were engaged in hybrid working. As you would expect, those numbers went up enormously during 2020 but clearly not everyone was able to work from home, particularly during lockdowns.

The numbers have come down somewhat since then and stabilised at roughly about 15% of employees saying that they are working remotely, about 26%, about one-quarter, of people saying they work hybrid, and about 40% of people saying that they travel to work only and never work from home. Those numbers seem to have been relatively stable over the last 12 to 18 months.

We have other surveys that give us some additional insight, including the Time Use survey. We have regular management surveys and a Public Sector Management Practice Survey, which ask about hybrid working among a range of other questions. We have regular questions about this in the BICS and the OPN business and household surveys.

There are some data gaps. I suspect some members of this committee are familiar with the work of Nick Bloom in the US on working from home. He has done a great deal of work. Some of the measures that he produces regularly in the US include the number of days that people are working from home in quite a disaggregated way, which I do not think is regularly available in the UK, that kind of spread from fully in the office or in the workplace to fully at home and everything in between. That is a useful gap to plug.

Another gap, which is quite an interesting question, is asking employees what the desired number of hours worked is, as distinct from the actual number of days that they are working from home. As you would expect, there is a bit of a gap between what employees would prefer, on average, and their actual practice.

On the range of other issues—house prices, office rents, crime, transport, and the environment—of course we publish statistics on them all. We have not done that much secondary analysis of those datasets to unpack some of those questions, although they certainly are interesting questions. Some of those datasets are sufficiently granular that they lend themselves to those kinds of research possibilities. Of course, we make a lot of that data available to academics through the Secure Research Service, if they are so minded to use that data to look into them. They are certainly important questions.

Anecdotally, one that I can point to is that for some time post pandemic, we published what we called the Pret index, which compares the level of sales in Pret A Manger stores across the country to what they were pre pandemic. It is no surprise that in central business districts we saw big slumps in the number of people going into stores. Interestingly, however, we saw a significant increase in the outer suburbs, reflecting the fact that people were spending more time working from home, going to the local Prets in the outer suburbs of the larger cities and buying there. That is somewhat anecdotal, but it points to the shifting pattern of behaviour by workers. Of course, that has also changed people’s consumption patterns. They are spending more in their local suburbs, less in cities, doing less shopping when they go into the office and so forth. Obviously, that has had an impact, both positive and good.

The Chair: I need to move us on. These answers have all been so interesting but we have quite a lot to get through. I think we need to get some concision now. Is there such a word? Concise answers. But thank you all so much.

Q27 Baroness Freeman of Steventon: I will follow on from that and maybe draw you out a bit more, because we have quite a broad remit here. We are trying to look at the pros and cons of home-based working for employers, employees, society and across many different disciplines.

Could you draw out some of the research that has been done on the data that you have been describing? There are quite a lot of data sources. You might know about some of the research that has been done using that data that can help us get at some of these effects. I am particularly interested in the split by different demographics—who really benefits, who really disbenefits—and looking at an employer level, as well as an employee level, and maybe at a societal level too, looking at some of the crime statistics, if you have that. It is a big question again, but a chance to talk about some of the research that you know about.

Raj Patel: I can start. I will preface my comments by saying that, for us, it might be a bit too soon to start talking about good research on the impact emerging through the Understanding Society study, mainly because the study covers data over two years. The data for 2022 and 2023, which is the post-pandemic period, is data that we collected, cleaned, and released in November 2024, only four months ago. In fact, from an academic research point of view, which is robust and trying to understand causal links, it is probably too soon to say that there is research, but I am sure that someone is thinking about what they could do with this new dataset that has been released.

There was a lot of research previous to the pandemic, which looked at flexible working rather than working from home, and one could easily draw on that. I know that a couple of academics who have been active users of our data are presenting evidence to this committee and they will probably be able to do more justice than I can to their findings. Broadly, however, the picture that emerged from flexible working was that there were benefits for men and women.

For some reason, the benefits were often higher for men than for women, and that may be to do with the choice that they had or the fact that there are all sorts of gender norms and issues tied up into working from home or flexible working, in that you might be working flexibly but you end up doing more of the domestic chores as a result and not sharing them. It does not show up in the well-being data, because there has been no change in the broader culture and norms about how that works. That is one thing. Broadly speaking, it is true that the results on health and well-being look positive.

I think it is worth also briefly mentioning productivity because it is right that it is actually very hard to measure productivity. You could look at subjective measures but, more importantly, one study that came from Understanding Society looked at who went back into work and who did not and looked at whether people felt that they were more productive or not. Broadly, their conclusion was that the people who felt they were not very productive started to go back to work sooner as compared with people who felt they were far more productive while working at home. People will say, for instance, that they are far more efficient at getting things done and are spending less time doing meetings. It is plausible, therefore, that for some people, working from home is more productive. However, as we know, it is very difficult to measure, and it is a subjective assessment.

Lorna Landells: The only demographic I can talk about comes from one of our surveys, and that is that younger people are far more inclined to want to go back, partly because they learn by osmosis, which they just do not get from looking at a screen all the time. We also use the Pret index, ONS statistics, Google and TfL data, but most of those have fallen away now, which is disappointing. We keep going, but the surrounding research is not quite as prevalent as it was.

Lord Fuller: When you say fallen away, is it just that they are not being taken any more or are they not being published?

Lorna Landells: They are not being published. They might be being taken, but we cannot access them. We are trying, but it is possible that they have ended.

Grant Fitzner: I will try to be succinct. My general observation as to where we are now is that the rates of hybrid and home working seem to have stabilised. We certainly have not seen any major changes. However, there are big differences between groups. Among working parents, there is not surprisingly a higher rate of hybrid working, 35% compared with non-parents.

There are big differences by occupation and qualification. Among the professional and senior occupations, professional and managerial and so on, the rate is 45% compared with a 20% average for everyone, and just 3% for elementary occupations. Very similarly, by qualification—for those with degrees, 42% of employees are likely to have used hybrid working versus just 4% of those with other qualifications. Clearly, a lot of that reflects the different industries that people work in and the different nature of their jobs, but it is a 10:1 discrepancy across those occupations and qualifications so that is really quite stark.

Looking at well-being and productivity, and just to touch on well-being briefly, it came across as a major drive across both the employer and the household surveys; 78% of employees who were engaging in hybrid working said that it improved their well-being. Interestingly, the single highest factor that employers who were offering home working gave as to why they were doing it was to enhance staff well-being. That 76-78% is a fairly consistent message coming through. Of course, employers had a range of other reasons why they favoured or were offering home working, including productivity and reduced overheads, and the ability to recruit from a wider geographical pool as well.

Again, there were no real surprises there, but clearly a range of factors come into play when employers are making those decisions. However, I would point out that only 19% of UK businesses said that they were currently offering home working or intending to provide it. That still leaves four-fifths of businesses that do not offer it.

Q28 Lord Fuller: We have heard a lot about data, but I have been reading the newspapers lately and have read about concerns, particularly from the ONS, about the response rates to some of these surveys. I am concerned about what reliance we can place on the surveys on the basis that we are not reaching the right people. Also, what pressure can we apply, say, to TfL to continue to publish their data, for example, just to fill in some of the gaps? These are big, important issues of how the economy works, and if we cannot rely on the data, because it is only the less busy people who are responding, are we getting the right picture?

Lorna Landells: You will always have a problem with people responding to surveys. There is an argument that the only people who respond to surveys are the people who like responding to surveys, so you are looking at a skewed sample anyway. However, you have to start somewhere and if you can make the surveys meaningful and short, the better the response rate. We have found that the shorter we can make the survey and tell people, “This will take you no more than seven minutes to complete”, or three minutes or whatever, that gets a better response rate. Certainly, it would be massively helpful to get TfL’s figures back, so if there is any way of encouraging TfL to bring them back out into the market, please do.

Lord Fuller: You mentioned TfL, but what about National Rail and other agencies, say the Manchester network?

Lorna Landells: For a while, Network Rail was giving us figures while we were working with them but that has also fallen away a bit. According to National Rail, the numbers of people on trains were massively down, particularly during Covid but except at weekends, oddly. People were still travelling on the train at weekends but not so much during the week. According to one of our surveys—we asked about attitudes—the major reason for not going into an office was because of public transport. That data is more skewed towards London and Manchester—big metropolitan areas—because outside those areas people drive more, but certainly public transport and how you get there is vitally important to getting people back to the office, if that is what we want.

Grant Fitzner: If this was just what surveys with perhaps relatively low response rates were telling us, I would be concerned. However, whether we ask employers or employees, whether we are using the Labour Force Survey or the Opinions and Lifestyle Survey, we are getting fairly consistent and quite similar results. That suggests that there is some veracity to those numbers.

Other administrative data, such as travel to work, are telling you much the same story about the changes that we have seen since 2020. If you were to pull all of that data together, you would get a reasonably consistent story. Maybe the exact percentages would be slightly different from one data source to another but because they are all more or less telling us the same story about what has happened and people’s behaviour, I think we can place some reliance on them.

Lord Fuller: Finally, Chair, because I know you want me to declare my interest, I worked in the fertiliser and agricultural business and we were exempted from staying at home during Covid so we always went in. You mentioned occupational qualification, but are there some particular industries in your slicing and dicing such as, say, farming and housebuilding, where there is very little penetration?

Grant Fitzner: Certainly, agriculture is one. In manufacturing, it is only 9% and there are a number of other industries where it is quite low. When you think through what is required in those jobs and workplaces, it is not surprising. However, there is a big variation across industries. In information and communication, it is 47%. Frankly, I was surprised it was not higher because what are the other 50% of firms doing if they are not offering hybrid working for mainly digital-based work? It is also reasonably high in some of the other service sectors but, again, not sectors such as accommodation and food services, where you would expect that people need to be there physically. That overlaps with the occupational mix that I mentioned earlier.

Q29 Baroness Watkins of Tavistock: A lot of my question has been answered in your iteration but I want to pick up, particularly perhaps from you, Grant, whether there is any significant issue around the elementary occupations that you talked about earlier—that is, on employment status and career progression—and whether there is any evidence on wages. There seems to be a general feeling that the more senior you are in an office role, the less you might go in, which might be because your home is larger.

Also, what about sickness absence? People keep telling us their well-being is really good working at home. I need to declare my other interest here, which is that I am a fellow of the Royal College of Nursing. There is also clear evidence that obesity is increasing for people who work at home and there are musculoskeletal problems. When people talk about well-being, are they just ticking that box rather than actually thinking about their physical and mental health?

Grant Fitzner: There are lots of good questions there. Going first with your last point about obesity, last year we ran a time use survey and quite a few respondents said that they were spending more time on exercise, sports and well-being and that that was facilitated by them working from home. They could go for a walk at lunchtime, or they could go to the gym or whatever, more easily than if they were in the office five days a week. However, that is for some people working from home. It is obviously not for all. There may be others who essentially never go anywhere and are maybe getting less exercise initially than when they had to travel to the bus stop or the train station and walk to the office. We have not looked at this in detail but it probably is a mixed picture.

Your question about career progression is important. This is not an area that the ONS has researched but there is other work internationally, and I think possibly some in the UK, that suggests that particularly younger workers—people relatively new to working in an office, say, and who perhaps did their university degrees all on screens and are used to that as a way of working—do not know what they are missing out on because they have never worked in an office and understood all the information and other things that you pick up just working with and sitting next to your colleagues.

At the ONS, we encourage younger workers to spend time getting to know their line manager and the other members of their team, and spend time in the office sitting with other people, because they can learn a huge amount that way—that tacit knowledge. However, that is a challenge. Some US research has shown that the younger workers who spent more time working at home did not have the same level of learning development opportunities and they did not progress their careers as fast as those who spent more time in the office. That is just one or two studies in the US. It is an area that needs further research.

The Chair: I want to move us on, Lord Fuller. Did you want to ask your question?

Lord Fuller: The question that I had prepared has been answered.

Q30 Lord Fink: My question is primarily for Lorna Landells. What data do organisations use when they make decisions about policies on remote, hybrid or office-based working, and what factors can influence those decisions? I will also ask a supplementary. Do any of the companies going to hybrid or remote do manager training on how to manage staff in a different way from before? That may not be within your evidence. It may be anecdotal.

Lorna Landells: No, that supplementary point is not in my area, but I hope that they do different management training. However I do not think manager training gets the focus that it needs anyway, regardless of that circumstance.

Generally speaking, the main reason why people go into an office is for face-to-face meetings. We find that businesses are encouraging people to come in for that collaboration. They are making better use of the space, refurbishing offices and so forth, so that people want to go in and get the benefit of the better facilities.

We also find, which is a sort of answer to you as well, that the more senior people are going into the office. As I mentioned earlier, the over-50s, the CEOs and COOs are going into the office more. We have also heard that a lot of banks are tying bonuses to office attendance, which links back to a wages question that was covered earlier.

How are people being encouraged back into the office? Some are using brute force, although we have seen evidence to suggest that a mandate results in a large proportion—more than half of the workforce—considering leaving their business if they are forced to go back. People like to feel that they have a choice in it.

There are things like putting more money into AI and technology. That was another reason that people go into the office as, with so many people working at home, broadband can be less reliable, and we rely so heavily on Teams and Zoom and so on when at home, so it makes more sense to be in the office.

That was a bit of a vague answer, but I would say it is mainly decisions around how we can attract and retain talent. Young people need to develop, as you were saying, and rightly or wrongly, they perceive their career is negatively impacted if they do not show their faces from time to time.

Q31 Baroness Manzoor: Could I pick up a bit on technology and AI? Are you seeing any evidence—or is there any data—showing that technology is being used to look at activity-based management by employers, and particularly managers, or is it in its early infancy?

Lorna Landells: We are looking quite a lot into AI and how it is being used, and particularly at what levels of jobs are being impacted. Most senior people believe that their jobs will be fine and that the impact will be at a lower level, but I cannot talk about this area particularly knowledgeably. However, I am more than happy to include a bit more about it in our written submission because we have quite a lot of data. I just do not have it in my head, I am afraid.

Baroness Manzoor: I was thinking not so much about the data but more about whether employers were using AI and technology as a way to manage people working from home and the activities they do at home—to better enforce performance management, culture, bonuses, and so on and pay parity.

Lorna Landells: Some are. You can monitor employees at home but that is more about people doing data input, that kind of thing. You can actively monitor the use of laptops and keyboard. In the kind of work that I do, for example, how much I use the keyboard is almost irrelevant to how productive I am. Yes, it is being used but it would only be useful in specific types of job.

The Chair: Before you move on, can you say a bit more in answer to the first question when you talked about office occupancy? Is it too soon to say whether all the media stuff we see about everybody going back to the office is reflected in what is happening in office space?

Lorna Landells: It is, but it is a really gradual increase, a very slow increase. It is not even a whole percentage point a month but it is going up. It has been steadily going up since about October last year.

The Chair: Do either of the other two want to come in on that?

Raj Patel: On the technology question, it is true that larger firms were far better at getting systems up and running and managing home working or remote working, and certainly the firms that struggled were often smaller firms. We certainly ask—it is a data question rather than what is actually happening—whether they use a range of technologies in their work. That could include computers, but it does not have to be just computers; it could be mobile devices, a range of devices, that a person may have to use in different settings. You could get to the question about whether some businesses are possibly interested in remote working but do not necessarily have the investment or expertise to make these things work, while larger businesses clearly do.

The Chair: Anyone else on that topic? No? Then I call Lord Farmer.

Q32 Lord Farmer: Mr Patel said earlier that it was very difficult to measure productivity, yet the purpose of work obviously is to be productive, so it is very important that it should be measured. I think from our point of view it is important that we can have access to data that gives us some steer as to what is productive and what is not. How do the three of you say that productivity could be measured so that the impact of remote and hybrid working is understood? What other data relating to employer interests is collected, such as the take-up of working policies?

Raj Patel: I can say what we have and there may possibly be a way through. It is quite a complicated problem to crack but certainly at a very basic level—and it is not a measure of productivity—we ask people, for instance, how many hours they worked in the last week.[2] From that, you could see whether people were working more hours or not. One of the issues with working from home is that your work and domestic life tend to bleed into each other. You could easily end up doing another hour, that you might have spent commuting, by carrying on working. Clearly one could track and see whether the number of hours people are working is changing in the UK or not, and you could compare people who work from home with people who mostly do not work from home.

You could look at people’s wage per hour, which is again a complicated factor. For instance, we ask people about their wages and about how many hours they work. If you divide the two, you could get some kind of factor of wage per hour worked, and see whether that is different for different groups in society. However, the bigger question about productivity is that productivity is measured at the level of the firm, and so you are trying to combine two very different sources of data. You trying to combine data from employees with data from employers, and in the UK we do not have any good way to link employer data and employee data. If you could potentially combine something like the Interdepartmental Business Register data with our wide-ranging population survey data and look at what is happening to the business as much as what is happening within the workplace to employees, you might begin to start unpacking the question of productivity.

Lord Farmer: What about a company’s balance sheet, its profitability before or after it introduced hybrid working compared to other companies, maybe in a similar business, that did not do the same or did more of something? Is that not data that can guide us along a certain path as to what is profitable and what is not, what is productive and what is not?

Raj Patel: Yes, you could use that sort of thing. The question again would be that you are using indirect measures of productivity. You are looking at profits, which could have a range of drivers, or you would have to look at sales figures, which are the two easiest ones to extract, I guess, from financial data. That might work for companies that have to file accounts. It would exclude a lot of self-employed or companies that do not have to file accounts, but certainly for companies that have to file accounts you could extract that kind of data.

The Chair: Baroness Watkins, is your question for Mr Patel?

Q33 Baroness Watkins of Tavistock: Yes, if I may. Do you have any evidence that some employers have conducted studies of their own staff, their well-being and whether they like home-based working, vis-à-vis what managers and leaders in large companies think about how things are working for productivity? I accept your argument. If you are making widgets and widgets go down in price across the world, you might still be being very productive but not making a lot of money. It strikes me that maybe we need a meta analysis of what firms themselves think from both employers’ and employees’ perspectives. What do you think?

Raj Patel: That is probably the only way to get to the problem because I think that if you go down the route of measurement and data, it is quite a complicated problem to tackle. Essentially, you are triangulating very different sources of data and trying to make the best assessment of what these different ranges of survey information or administrative data are telling you. The most precise or scientific way to do it would be to link data but linkage is very complicated in this situation. That is the problem.

Grant Fitzner: Can I come in on that point? It is a complex methodological research problem because there are lots of associations in the data, but proving that greater use of hybrid working or home working produces more productive employees is a tough nut to crack. Most of the studies tend to look at a particular company, often a call centre or a digital firm, and randomise working from home, which I think is quite difficult to do now when many people with those kinds of occupations already do hybrid working.

To get a sense of the impact across the UK, you do indeed need something like a linked employer-employee dataset where you have the characteristics of the employees and of the different firms, and you track them over time. We are looking at developing that kind of dataset at the ONS but at the moment we do not have that capability.

Having said that, we recently sponsored research that was published by the Economic Statistics Centre of Excellence late last year. I will send a link to that research after this hearing. It found that businesses with good management practices that tend to be more productive and more profitable, as you would expect, maintained higher rates of home working and hybrid working between 2020 and 2023. There is certainly an association, but I do not think you can say that home working drove better productivity. You can say that there is an interesting association that we need to unpack.

Lord Parker of Minsmere: Grant, may I follow that up? I like the word “macroeconomic” in your long job title. You just commented on the 2020-23 period. Push the brackets out wider. Is there anything you can tell us, at the macroeconomic level for the UK, about productivity from 2019 or earlier compared with the present? We skip entirely the Covid period but want the before and the after, because what we are very interested in is anything that we can draw as learning from how changing patterns of work have some sort of correlation with productivity at a time when, of course, we are all intently concerned with the UK’s economic growth and international competitiveness.

Grant Fitzner: That is a big question and we have only a few minutes. Let us take a really big picture. If you go back to before the global financial crisis of 2008, labour productivity growth was sort of humming along at about 1% or 1.5% a year. Since the global financial crisis, it has been quite sluggish but there has been some growth.

Since the pandemic of 2020, labour productivity has fallen, but that is probably due to a range of factors—trade disruption, low investment and the fact that we have had not one but two recessions since 2020. I could give you a long list of potential factors but attributing which were the main drivers is quite hard to do. Of course, there are institutes like the ESRC-funded Productivity Institute, which looks at this very question, and we, from time to time, publish research on this, but it is fair to say that there is a number of factors that probably have contributed to the weak productivity performance in recent years.

Lord Parker of Minsmere: Take a punt.

Grant Fitzner: What is interesting is that we are not the only country where productivity has slowed since 2008, so that would suggest it is not just domestic factors. I know some people would say, “Well, it is because of Brexit, obviously”, but the fact is that if European countries are seeing a similar pattern and the US is seeing a similar pattern, there are clearly some global forces at work. However, the fact that we have relatively low levels of business investment and automation in the UK does not help.

Flexible labour markets have many advantages. They lead to lower unemployment rates than many comparable European countries, but conversely they can also lead to lower levels of productivity growth.

Q34 Baroness Freeman of Steventon: One more question, which is for any of you but probably Grant, about whether you have data on where workers are in the country and whether that tells us about any geographical redistribution of work and workers around the UK. If so, what kind of workers, what demographics and how has it affected you?

Grant Fitzner: For the take-up of home working, you mean?

Baroness Freeman of Steventon: No, just generally. Are people moving out of cities because they can do more home working? Does that reflect in wage distribution across the country, and how does that affect different demographics living in different places?

Grant Fitzner: Anecdotally, there was a significant relocation of people out of the major cities after the pandemic, and that was partly because people had more flexible working options. I think an interesting question, which we have not looked at but probably should, is the pattern of behaviour of working parents. In the past, if you had to pick up your kids at 3 pm from the school gate, you were quite restricted in your range of employment options.

But let us say that in a previous career you were a lawyer, and you could still work part-time with a London-based firm, for example—that opens up a wider range of employment options. How that has affected the geographical distribution is something that certainly would be worth looking at, but I do not have anything at my fingertips to be able to answer that question. We can come back to you on that if you like.

Raj Patel: From our data, you could look at whether people’s commuting distances have increased or not. That is a slightly better measure than whether they have moved because people do not move all that often. It is a good indicator, maybe in the medium term.

The second thing is we ask people if they are prepared to tell us who their employer is and give us a postcode. If there was a clever researcher out there, they could start linking employer data that we are asking from our population panel with separate business data and make some broad correlations, possibly not causal links but even that would be a good start, I think.

Q35 The Chair: Thank you. You have given us some helpful clues as to where we, as a committee, need to be looking—data on transport and so on. You have also suggested some areas where you and organisations like yours need to do more work to understand what is going on.

As we come to the end of our session, would you each like to take the chance to come up with one more thing that you would like to be able to do yourself, would like someone else to do or that you would like the committee to look at?

Raj Patel: The ultimate prize would be the ability to link data. That is a long shot and it might be quite technically complicated to do. We know what is happening to productivity in the short term, but we still have a huge challenge with the long term. If we can invest and try to come up with better ways to measure certain things, that certainly would be a good case.

It is true that worker well-being and skills are drivers for productivity. The Productivity Institute has put the evidence together for that. How that plays out, which is the data that you can get from population-level surveys, will be very useful.

Lorna Landells: This is a bit of a unicorn, but I would like to be able to measure the productivity of the office worker properly. Maybe drilling it down into things that were task-based, categorising different types of tasks for the office worker, and linking their objectives so that they directly have a causal effect on the bottom line of businesses would give you the data because it would all be regularised. That would be my wish.

Grant Fitzner: I am going to give a plug for the linked employer-employee dataset, which is quite niche, but excites a lot of researchers because it opens up a wider range of possibilities for unpacking what is causing what. Particularly if you build a time series, you can see how different firms change their patterns of hiring, or maybe introduce home-based working, and what happens to that firm before and after. That may take a little while to pull together, but I think that would help to start to answer those questions that we were not able to answer today.

The Chair: Thank you very much indeed, all three of you, as I bring this session to a close.

[2] Note by the witness: To clarify, the study question referred to “a normal week” rather than “the last week”.