In this episode, we’ll discuss a study suggested through LinkedIn by Bridget/Brigid? one of our listeners. Published by a research and consultancy/advocacy organization called Autonomy, the paper is titled, “The Results Are In: The UK's Four-Day Week Pilot”. Published in February 2023, this is a non-peer-reviewed paper and the research methodology is a bit skewed as a “pilot” instead of a controlled trial, but still contains valuable results that we’ll discuss.
This report details the full findings of the world’s largest four-day working week trial to date, comprising 61 companies and around 2,900 workers, that took place in the UK from June to December 2022. The design of the trial involved two months of preparation for participants, with workshops, coaching, mentoring and peer support, drawing on the experience of companies who had already moved to a shorter working week, as well as leading research and consultancy organisations. The report results draw on administrative data from companies, survey data from employees, alongside a range of interviews conducted over the pilot period, providing measurement points at the beginning, middle, and end of the trial.
Discussion Points:
Quotes:
“It’s important to note that this is a pre-Covid idea, this isn’t a response to Covid.” - Dr. Drew
“...there's a reason why we like to do controlled trials. That reason is that things change in any company over six months.” - Drew
“ …a lot of the qualitative data sample is very tiny. Only a third of the companies got spoken to, and only one senior representative who was already motivated to participate in the trial, would like to think that anything that their company does is successful.” - David
“I'm pretty sure if you picked any company, you're taking into account things like government subsidies for Covid, grants, and things like that. Everyone had very different business in 2021-2022.” - Drew
“We're not trying to accelerate the pace of work, we're trying to remove all of the unnecessary work.” - Drew
“I think people who plan the battle don't battle the plan. I like collaborative decision-making in general, but I really like it in relation to goal setting and how to achieve those goals.” - David
Resources:
The Harwood Experiment Episode
The Safety of Work on LinkedIn
David: You're listening to the Safety of Work Podcast episode 108. Today, we're asking the question, could a four-day working week lead to happier, healthier, and safer workers? Let's get started.
Hey everybody, my name is David Provan, and I'm here with Drew Rae. We're from the Safety Science Innovation Lab at Griffith University in Australia. Welcome to The Safety of Work Podcast. In each episode, we ask an important question in relation to the safety of work or the work of safety and examine the evidence surrounding it.
Today's question comes from one of our listeners, Bridget. Thank you for that. It's via LinkedIn. She sent us a link to this paper that was released recently, and we're going to have a look at it. We're always looking for ideas, so if you have a topic, feel free to reach out to us on LinkedIn or our feedback email address at the end of each episode.
Sometimes there just isn't high-quality research that we can base an episode on. If it takes Drew's or my interest, we might go and try to find a paper. Maybe we'll do the background first, Drew. Do you want to give us a bit of background on the five-day working week and what we've got today?
Drew: Thanks, David. We started off with this paper and talking about the four-day working week. We thought we ought to first explain why we've got a five-day working week. That took me down a little bit of a rabbit hole. We've got time limited episodes, so I'll try not to go right back to the agrarian revolution. Let's just talk about the industrial revolution.
I think to contextualize this topic, we do have to start with the fact that it's a weird thing to sell your own labor for money. Gradually over history, that's become more and more formalized. It starts off with the idea of peasants and quite informal arrangements. But then once we get to factories and mines, that's when we start having shifts. Once we've got shifts, we've got very sharp lines between when you're working and when you're not working. When the machines are running, you got to have people there. They turn on at a certain time, they turn off at a certain time, and you get paid for the time that you're there.
In those early factories, people are working a hell of a lot. Your best estimates are that prior to that, most people were working pretty long days, but only about half the days in the year. Even during those long days, they're taking lots of breaks. You imagine your typical farm worker has to get up early, has to do jobs, and then they take a break. They do more jobs than they take a break. They're not working constantly that whole time.
When you're in one of these 19th century factories, pretty much every week of the year is a work week. Early on, we didn't even have this notion of holidays. You're working most weeks every year. You're working around 70 hours a week. That's typically long days for six out of the seven days.
Ever since we had that, we've had people pushing back. The labor movement started about the same time as the factories started. One of the earliest positions was this idea of let's make it eight hours work, eight hours rest, eight hours recreation, and let's make it five days a week. You have the ends of the week, Saturday and Sunday off.
That was never scientific. It was never formal. It was just an asymmetrical way or position to advocate for. It gradually became more and more standardized, more and more popular.
David, I don't know about you, but I had heard growing up that Henry Ford was an innovator in this 40-day workweek. He was actually one of the latest adopters. Like a lot of things, Henry Ford managed to get famous for copying off other people's ideas. Doing this in 1926 was actually just before we started having laws and making it mandatory for everyone. There had been a century of advocacy up to that point.
David: I think Ford did become famous for paying above award wage. I guess he and the organization were of the view that if we pay our people more money, then more people can pay off Model T cars, so we'll get that money back anyway.
Drew: I think that's what Ford was really interested in. He tied this to remembering that workers are consumers as well, and that if you're going to want people to buy your own cars, then you've got to pay them enough to be able to buy that car. You want people to have holidays so they can consume products. He saw the whole cycle of capitalism.
I think most people have got fairly more humanitarian ideas about why we should have the 40-hour week as if people will deserve it, not people will buy our products. The 40 hours became the position, and then it's gradually being negotiated down. In Australia, the standard is around 38 hours. In France, it's 35 hours. In the US and Japan, it's still 40 hours.
Remember that this standard week isn't actually how long people work for. In a lot of places, it's just how we negotiate overtime. A lot of people are still working 70-hour weeks, it's just that the 40-hour week means that they get paid extra for those extra 30 hours. Sometimes it works as a limit on work. Sometimes it's just a fairly arbitrary negotiating position.
David: Drew, we have what we have. It's evolved, like you said, it's been pretty static for a hundred years or so, but it evolved the way it is. Probably most of the last decade, it definitely accelerated by necessity through Covid, more and more of this idea that in some occupations that are pushed for much more flexibility around work.
I guess this is a form of flexibility that's actually changing the amount of work we do. Lots of other flexibility ideas look at the ways of working and shaping the way that people work in terms of when, where, and how. This is really saying, no, we overall are going to work a four-day workweek, which is actually work 20% less time.
I guess this is fairly arbitrary. It's just that people can have three days off instead of four days, although that's not the approach that every organization takes to the four-day workweek. I guess this start is just an idea. It doesn't start on the back of any scientific theory that looks at what the human body should and can tolerate or anything like that, or how long we can be productive for.
Drew: It is a bit like The 40-Hour Workweek, the four-day week is a fairly radical proposal initially. The idea is just to make that radical proposal seem more and more normal over time. It's important to acknowledge that this is a pre-Covid idea, this isn't a response to Covid. This is, if anything, something that was gaining momentum in the couple of years before Covid and then had to be put on hold for a couple of years. Now he's coming back very massively.
As companies realize that they need to incentivize people to come back to work, they're competing with labor. They can't offer flexibility because people are currently in maximum flexibility mode. They can't offer more money because the companies can't afford the extra money. This is a way that companies can offer something back to workers that might possibly, and this is a question we'll get to, be cost neutral for the business. The business doesn't actually have to make any sacrifices to give workers something that's good for the workers.
David: That's where the ideas come from. I guess what we're going to do in this episode is focus on two central claims. One is that we can significantly reduce the number of hours worked whilst maintaining the same productivity. Like you said, from an organizational perspective, same pay, same productive output with 20% less time from the worker, and then that this reduction in hours worked will lead to other benefits for the employee, particularly around their well-being and a bunch of other individual related outcomes not associated with with the work they do, but definitely impacted by the amount of work that they do.
Drew: If we can tick both of those boxes, then we have a proposal that should be very, very seriously considered by any business. The business case would be, it's not going to cost us anything in terms of productivity, and it's going to be more attractive to employees. Therefore, we get lower turnover, we meet our obligations to employees.
David, I don't know about you, but I'm not very cynical about CEOs. I think most people who run businesses want to believe that they're the good guys and want to do well by their own workforce. This is if it works, something that genuinely is something that a company can do to better look after its people to show them that it values them.
David: Yeah, I definitely agree. Like you, I think organizations are quite moral. The leaders of organizations today, at least in 2023, are very moral, at least the ones that I interact with.
Drew, I wouldn't mind getting you to introduce this paper because it was something that was published. Link was published on LinkedIn and shared with us for the podcast. Normally we do journal publications, but this is, I guess, more of an industry-wide paper. Do you want to introduce the paper and a bit of the background for the authors and the institutions?
Drew: Okay. This is a little bit of an unusual one to go through. There have been a number of recent high profile trials of the four-day week. There have been three in particular that have been led by roughly the same group of people.
We're going to be focusing on the UK trial. There was also an Australian one. I think the third one was European rather than North American, although I could be wrong about that. The Australian one is not yet published, and the UK one is the one that's getting the most attention.
It's a collaboration between several academic and non academic organizations. The lead on the paper is an independent consultancy called Autonomy. It claims to be a dual research agency in a consultancy, but as far as I can tell, it doesn't do any peer-reviewed published research. It's really a consultancy/advocacy organization.
The paper, if you're looking for it, is called The results are in the UK's four-day week pilot. It was published in February 2023 and was published as a non peer-reviewed white paper. That means it got a lot of the form of an academic paper. It uses a few features that you're not allowed to do in academic publishing like case study boxes, which I think should be in academic publishing, but often aren't. It's got a reference list, it's got footnotes. It's got most of the academic things, except for the formal methods section and the peer review.
The Autonomy authors are a mix of academic and non academic, but their backgrounds are in political philosophy rather than organizational research, which fits in the motives of the paper, but they don't have expertise in doing organizational research. There are teams from Boston College, who did a lot of the quantitative work and University of Cambridge, who did a lot of the qualitative work.
The lead quantitative author is professor Juliette Schor. She has a massively prestigious and long career as an economist. Most recently, she's looked at things like sharing economies and gig employment. The lead qualitative author, Dr. David Frayne, is a sociologist focusing on the anti-worker movement. Hopefully, that gives a bit of indication that the team has got a lot of relevant expertise, but a lot of it is with a certain political slant to it.
The team is very large, so you don't really know how much influence each person might have had. It's often a bit unclear in these cases. Is this large team sort of acting as internal peer review or are they actually having a hand in the design of the work? In this case, I get the feeling that I could be wrong, that the overall design of this thing and most of the authorship of the report comes from the consultancy.
The two academic teams did the quantitative running of the surveys and the running of the qualitative work. That doesn't prevent the work potentially being slightly spun in the final report. I don't want to slight the authors from the academic side and say that they've done anything dodgy, where it may just be other people reporting on what they've done.
David: I think it's important that we also keep in mind, as much as we don't know, these researchers and research consultants. From the outset, though, it may not be a neutral evaluation that this team and the people who make up the team are already advocating for a four-day week at the start of the program. This research is really about supporting that existing position. I guess it doesn't automatically mean that this isn't good academic work, but we need to keep that in mind, particularly when we're looking at what the data says and then what the conclusions that are drawn from that data are.
Drew: Just like when we look at who writes a paper, we don't use that to disqualify it. Very junior people can do really good work. We use it to work out, but we need to be more careful about looking at, for example, how they represent a particular field of study. In the case of people who might be motivated, we need to look at how they spin certain statistics, go back and look at the statistics ourselves, and think, what does the statistics say versus how they're being spun?
David: Drew, I might just describe a little bit about what the researchers did and then get your thoughts on the research design component. Is that all right?
Drew: Yeah, let's do that.
David: I guess it's called a trial, even in the heading it says the trial, but I guess it's not really a controlled trial. You've got in the notes more of a pilot scheme. It's going to find companies who are prepared to do this and give them a bit of a frame for how to do it and what we're going to measure. It's not really a trial at all.
Sixty-one companies in the UK and the research phase ran from June until December 2022, the end of last year. The companies didn't actually implement standardized four-day weeks. They got to choose for themselves what they did within certain parameters. We'll share that through this episode. There was self-reported company data at the beginning and in the end, representing the six months before and the six months after the trial.
Employee survey data at the beginning and end of the trial for 44 out of the 61 companies, then a different employee survey in the middle, and then there were some interviews with some senior representatives of 23 out of the 61 companies before and after the trial. As far as research design goes based on the research question or the claims that we're trying to understand, what are your thoughts?
Drew: For what they're trying to do, it makes sense to run a pilot scheme where you get as many people involved as possible. You give them flexibility to implement it in a way that suits their company. But the trouble is that there's a reason why we like to do controlled trials. That reason is that things change in any company over six months.
If you consider particularly the second half of 2022, every company in the world is going to be different at the end of that six months when it comes to things like employee well-being, employee work-life balance, how much time workers are spending at home, and company revenue, because every single company is coming out of Covid.
That particularly matters when it comes to just doing it over six months because we have seasonal effects as well. Most companies have a different performance in the first six months of the year to the second six months of the year. Now they've done a little bit of work, but actually quite uneven work to try to balance that, to try to take that into account. Because the fact is, the only way to properly take it into account is to have a controlled trial, is to have at least as many companies not doing this and see what happens to those companies, rather than just comparing the companies within the trial.
That's the first big methodological issue, which is going to cover all of these results. The overall design is tailor-made to get a positive result. You will always, if you just do before and after intervention, see a positive result, particularly if you measure lots of different variables, and you get to pick and choose which ones you use.
The other big problem is, and we're getting to the mix of companies, that these are just anyone who decided to take part in the trial. It's a massive range of different businesses. When they're reporting the results, they lump the results from these businesses all together, which hides any effects that might go up or down for individual companies in different circumstances. If you were trying to tailor-make a pilot that would be a success and that you claim that this is a success, this is the way to do it. If you try and do a very fair evaluation, then the overall structure just doesn't work.
David: Drew, I think that's a great point. Particularly, a lot of the qualitative data sample is very tiny. Only a third of the company's got spoken to, and only one senior representative who was already motivated to participate in the trial, would like to think that anything that their company does is successful. I know that's balanced out some survey data, but I guess these data sets aren't really in the shape that we'd like them to be, notwithstanding the overarching methodological issues.
Drew: Right from the very start, we can see some of the spin start to creep in. There are three separate numbers here. The first number is 70 companies, which is the companies who agreed to do the trial. Then we've got 61 companies, the ones who actually successfully went ahead with the implementation. And then we've got 44 companies, which are the ones that actually gave data back.
The study likes to switch between these three numbers depending on when it best suits. Really, if you've only got data from 44 companies, then this is a trial of 44 companies, not 61. But it keeps switching back to 61 every time they had their figures and show how good it was.
David: Drew, let's talk about these companies and what they did because one of the first challenges is that on one hand, it's a wide range of industries. You got eight marketing firms, seven professional services, five charities, five training, five insurance. But if we think about the representative sample of UK industry, you've got no operational organizations here. You've got no logistics businesses, no production businesses, no what we call physical, operational, high hazard type of industries. These are white collar jobs that can be done from home and aren't really time sensitive to support critical physical operations.
Drew: One or two of these companies even need to maintain a five-day office to deal with the public and clients. Most of them are not even in that position where they've got an outward-facing part of the company. It's a very broad range. They've got a lot of different businesses, but it's not particularly representative.
David: Drew, half the companies have 25 or fewer staff, and only five of the companies have more than a hundred employees. Even though we've got 2900 or nearly 3000 staff involved across all of these companies, we need to remember that more than half of those come from just the five big ones. This again is depending on how these numbers get reported. We could be talking half our sample coming from five organizations instead of 61.
Drew: Okay. Good that we've got a diversity of companies, but it can hide some real differences between those companies.
David: To participate in the trial, like we say, companies have to commit to a meaningful reduction in work time and maintain current employees pay. There are a few different examples. One company shut down entirely for an extra day a week having Friday as a genuine weekend, where offices were closed and the company all shut down at the same time. The company works a five-day week, but like you said, Drew, each person only works four days to maintain coverage.
Staff would work on average 32 hours a week, but this would go up and down as there were demands over the six month period for the trial. I guess we only know what 44 of the companies did because only 44 companies said what they were actually doing.
Within that, Drew, it looks like more than half really did implement a genuine four-day week with either the company completely shutting down or each worker having a designated extra day off. But if we remember, that's only 22 out of 44, out of 61, out of 70. We're now talking about 22 companies out of our initial pool of 70, where we know that they genuinely went to a four-day week.
Drew: David, I don't know about you, but I was actually surprised by how high that number was, given how flexible their definition was that they started with everyone just has to have a meaningful reduction, it doesn't have to be a four-day week. Actually, they did have a solid base of 22 companies who went all in. I wish they had done the report on those 22 companies. These are the ones who'd fully embraced the concept of let's either shut down the company or let's have a rolling shutdown, where you have the employees take it off on Monday, the other half take off on Wednesday, or something like that.
A lot of companies did actually do a fairly radical trial, but they're hidden in this swamp of other companies. We don't know, for example, whether the five big companies are included in those 22 or not. That aggregated everything.
David: And also to control other variables because of how much time having people who physically shut down their system access, their email access, and everything like that at 5:00 PM and didn't turn it back on until 9:00 PM the next morning. We actually knew that people were genuinely working reduced hours, not just being at work for reduced hours, but still working the same. We'll get to some of that in the results as well.
Let's maybe talk about the results because the paper title is the results are in. I guess the bulk of our listeners are going to want to know what were the results and what do we think about those. Like you said earlier, 70 companies signed up, 9 abandoned the idea. We got 61 companies that actually implemented something.
At the end of the research, let's call it that. Five said they weren't continuing with the arrangements and 18 said they've decided to make the policy permanent. What we've actually got is a huge number of companies in the middle, almost 50 companies that actually haven't decided whether they're going to continue or not. If these results were as compelling as they are, I would have thought we'd have more than 18 companies at the end who have said yes, we're going to continue with this as a permanent policy.
Drew: That's fair enough. A six month trial, no matter how good it might seem, I would not be committing immediately as a company too. I'm going to make this a policy. To be fair, a lot of the companies were doing their own data collection. They might need time to do their own analysis before they make the decision.
I would suspect that most of them, if they've got some idea, would just want the trial to run a little bit longer. They'd want it to run over a full year, so they saw how things worked over the whole cycle of operations for a year, maybe see how things emerged out of Covid, how things are developing before they make any commitment.
It doesn't surprise me that most people are non committal. I love the fact that their headline statistics, 56 out of 61, are continuing with the four-day week, even though only 18 of those actually said that they were making a deliberate committed decision to continue with it rather than just, aren't we haven't stopped at the moment the trials stopped?
David: Yeah. I guess I'm a little bit more cynical than you about what the actual outcomes were in some of these organizations because I think management makes big decisions with less than six months of research trials inside their business.
Drew: David, just before we move on, I just want to give the most pessimistic cynical view. Remember, only 22 of the companies have gone full-on in the four-day week. I would love to know the crossed hands on of those 22, whether they fall in the continuing, not continuing, versus have abandoned altogether partway through.
David: If it's 18 out of those 22, then that is a very different thing if none of those 22 are in the 18.
Drew: Exactly.
David: I guess there are other things in the aggregate that we haven't got their board data for, and revenue is one of those things. We don't have individual company data, so we just don't have the breakdown. I guess white papers often don't do that because if you could imagine this paper being sent to Drew or something to do some peer review, so we have to see the data.
We don't have enough data reported to actually know whether the claims can be made the way they are or not. Even revenue data, they say they weighted the revenue data by company size and provided the weighted result, but we don't know what data relates to which company, relates to which pilot program.
Drew: There are lots of ways of doing this, which preserve the anonymity of the companies. The very least they could have done something or said, over the 56 companies, 20 went up in terms of revenue, 20 went down of the ones that went up. Some went up by this much, some went up by that much. It's easy to do, but instead they've given this weighted average that we've got no idea about. It could be dominated by one of the five big companies. It could have done really, really well. Everyone else could have done really, really badly.
You get the same average as if everyone had gone up by a little bit. I really wish they'd done that for their own sake because if you look carefully at the revenue data, it looks really, really bad. For 23 companies, they've got data for the six months before and the six months after. They say there was a revenue increase of 1.4%, which sounds positive, revenues gone up over a period when inflation in the UK was 10%. On average, everyone went badly backwards compared to the UK.
David: Again, revenue is not the right thing because if you're looking at wages and input costs going up by 10% during the period, then the weighted margin of all of these organizations would be significantly down.
Drew: Yeah. The reason why they're using revenue is to try to get some comparison and to have companies that are willing to share the data. That's one of the real difficulties with this multi company trial. It really depends on people being very, very open about their experiences and none of these companies want to look bad. None of them are willing to share nitty gritty operational details. That may be something that's forced on the researchers to use just the best indicator of productivity they've got. Your revenue is not perfect, but it might just be all that they had.
David: Yeah. Drew, they did a comparative period. Obviously, this certain time of the year is different for different businesses. What you would usually do here is take the second half of 2022 when the study was going and compare that to the second half of 2021. They reported this 35% increase in revenue from the same comparable prior period.
Drew: Just to be clear, in 23 companies, they did six months before six months after. For another 24, they decided that wasn't a fair comparison. They got a comparable period and used that instead. That's what I said, they did some attempt to fix this seasonal thing.
The trouble is that there is no fair comparison for the second half of 2022 because even if you take the same period from the year before, that's right in the middle of Covid. Of course for all of these companies, their revenue went up from 2021-2022. I'm pretty sure if you picked any company, you're taking into account things like government subsidies for Covid, grants, and things like that. Everyone had very different business in 2021-2022.
David: Drew, let's talk about the results from the staff. We see we've got some challenges with the company level data. This was the outcome at a company level. We don't have the insights that we'd like to have knowing which company went in which direction, over which, and what was their baseline or what's a comparable point for them.
On the staff level, we supposedly got 2900 staff in these companies. Roughly 2500 actually got sent the first survey, or 2548, and only 77% of those completed it. If you do the math, that's less than 2000 or about 2000 or so completed it. Only 70% of those completed the second survey as well, so that 2900 staff probably comes down to about 1500 by the time we take how many of those staff did both surveys.
Drew: You nailed that. It's pretty good there, David. It works out between 1300 and 1400, depending on how you count it. A bit less than half of the participants, which in terms of company surveys, it's actually pretty good.
The trouble is that we don't really know where those staff are from in that set of companies. We don't know if some companies were really good at sending stuff back. It is possible that those half entirely come from the big companies. It's possible that that ignores a couple of the big companies, we just don't know.
David: We could have 40 small companies with a 10% response rate or something like that.
Drew: Yeah. Let's put that caveat on it. But then for the rest of it, let's just assume that these are representative because unless we make that representative, there's just no point in having that in the rest of the conversation. Let's say that this is the general experience of people across these companies. First thing to note is that our sample is very female, very white, and very educated. I think that's a result of the type of companies that participated rather than skewing the sample itself.
I think these tend to be professional services type companies, charities. They just don't represent your average manufacturing, construction shop floor type company. That's the business that we're talking about. Maybe that's you. Maybe you are an office work type company, or at least you're thinking about the four-day week for your office staff. This population is probably fairly representative of office staff.
Weirdly, more than 10% of them don't live in the UK. I'm still trying to work out where that one comes from, whether they had some problem that the survey was a little bit too accessible, or whether these companies just broadcast the survey including their overseas staff. Anyway, more than 10% of the people who responded don't actually live in the UK.
Let's ignore the sampling problems and just get on to the direct results. What's happening to people who are experiencing the four-day week? A great idea to start with is that they're working less hours. That means that at least we're getting some intervention.
The official change has actually resulted in them working less. On average, it's shifting from 38-34, so not actually creating a full day off, creating about half a day off. If you ask the question a little bit differently and ask them how many days they're working, it changes from around 4.9-4.5 days, which is less than half a day off, which is interesting, which says people don't actually know how much they work. But this is what you'd expect, which is that the radical change in policy has created a real change in less work, but it's not as big of a change as the radical policy.
People are still coming in on that day that they've supposedly gone off. They're still doing that a heck of a lot of the time. I honestly don't know how that compares to how often people currently work on Saturdays. I'm used to the academic environment, where pretty much everyone works at least one day on the weekend. How about your business?
David: This is what's driving behind this study. You don't have to go too far in the mainstream media to read about burnout, hours of work, and things like that. I think these samples represent them. Even 30% of people reported no change in their working hours or that they were working more. I guess 70%, 71% of people have reported that they were actually working less. I think we've got to take that at face value that 70% of those people said, yes, we're working less than we were.
Drew: Maybe that is a positive thing to come out of this study. It really does take a radical change in policy to achieve a small reduction in actual hours worked. If we try small things to reduce hours of work, maybe they just have no difference at all. To get people to take half a day or a third of a day less work, we've got to tell them to take an entire day off.
David: I think, even if you remember that 22 companies actually fully embraced this and shut down for a day or rostered someone for a full day off. This idea that on average, working hours went from 38-34 or half a day, if you take those 22 companies out, and again, we don't know how many people came from which company who actually maybe did get a genuine eight hour type reduction, there's a lot of organizations that did a radical policy change for maybe one hour difference or less, or slightly more.
Drew: To look at it more charitably, maybe for the 22 companies, it did work really well. They're actually underselling their own results by trying to beef up the number of companies. Maybe if you strip down to just those 22 companies, it actually looks a lot better.
David: I was going to say just some other outcome measures for individuals. On average, they reported a small reduction in stress, I guess some more stress and some less stress, but 50% of people said their level of stress was unchanged. Burnout improved for most people, so three quarters of people that their perceived level of burnout improved. And job satisfaction increased for about half of the people.
We've got half of the people no changing stress and half of the people no change in job satisfaction. That surprises me if 70% report working less than having more time for other things and less work impacts. Either stress and that is not actually coming from the work, but I don't know. What's your take on those results?
Drew: This doesn't surprise me, given that overall, we've only achieved a small reduction in how much people are working. What actually surprised me, if people are working just a few hours extra a week, suddenly everyone was reporting less stress, less burnout, mostly more positive emotions, that would be rather suspicious.
The fact that a substantial minority of people achieved positive results and most people were unchanged, that I'd read is overall positive. The warning sign is we've got a substantial minority that are positive and a smaller, but not insignificant number of people who have a negative effect. I'd really like to know, are these different groups of people? Who is it working for? Who's it not working for?
Why are a lot of people not experiencing a change? Is it because they didn't really get the policy? The policy didn't actually give them the reduction? Or is something else going on? Do they have mixed feelings?
David: Not much else to talk about in the individual results other than overall respondents were pretty happy. Of those 1300-1400 survey responses, again, not knowing where they were, 90% said they wanted to continue with their current working arrangements. 96% had a preference for four-day weeks over five day weeks. Again, that didn't really surprise me that much. In some ways, I'm surprised that 4% of people actually want five-day workweeks instead of four-day workweeks for the same pay. That 4% surprises me, but the 96% and the 90% doesn't surprise me at all.
Drew: David, if you ask people on a survey how blue is the sky, never blue, sometimes blue, et cetera, then you'll get 3%-4% of people who say that the sky is green all of the time. Asking people all else being equal, would you prefer to work for four days a week or five days a week, that's as close to the is the sky blue question. And 96% of people say, yes, the sky is blue. Give me a choice, and I'd rather work less for the same pay.
I think the significant figure is actually that 90% said they want to continue their current arrangements under the trial. Most people are actually happy with what their particular company is doing. That is a fairly very positive result, even if all of the more objective measures about how much it's actually helping people are very mixed, people like it.
David: Do we know that, though? Do we know the question and how the question was worded because 90% said they wanted to continue? If they said, do you want to continue with what we're doing or go back to 40 hours, or do we know whether they said, I am satisfied with the current arrangements, were they given alternatives or other things? They might be satisfied with this compared with the alternative, but they might not necessarily be that happy with the arrangements at all?
Drew: David, I believe the question was worded in, do you want to continue with the current arrangements or go back to how things were?
David: That 90% doesn't really surprise me. Again, 10% surprise me that they want to go back and work more again.
Drew: Remember, we've got something like 15% of people saying that they're now working more than they were before. That 10% is less than the number of people who are more stressed, working more hours, experiencing more burnout. We've got 22 people, burnout is worse. Half of those people are saying they want to go back.
David: I guess we're talking a lot about numbers and a lot about data, but this is a very broad research paper that's aggregated down into something like a white paper that's designed to be very compelling. You can see that we've got literally tens, if not hundreds of questions that we don't know the answers to, that we would like to know the answer to to know the veracity of the claims that this report is making.
Drew: Hopefully, the readers get the overall impression, which is that these are very mixed results from a very uncertain method that are being spun as being unquestionably positive. They're not, they're mixed.
David: The qualitative results section, like we said, there were 23 interviews before and after with the executives. I guess it was called perspectives from the shop floor, which is interesting because it was mainly conversations with CEOs. It seems like there was a lack of frontline participation in any aspect of the qualitative research data.
Drew: David, just a small correction. As well as the CEOs, they interviewed a small group in the middle of the trial of workers. We're talking about 15 interviews, and they didn't say whether these are all from one company or which companies these were from. What we do know is that they definitely weren't from the shop floor because there was no retail participation in the trial. These perspectives from the shop floor are a mix of CEO and office staff.
David: Got it.
Drew: I still think this is really interesting qualitative data. Let's go into some of the things that they did ask them and what they found.
David: When they asked why CEOs wanted to participate in the trial, there were some reactions to making changes post-Covid, it being a difficult labor market, trying to be a bit more attractive to employees, they want a competitive edge in that space for staff retention, staff attraction, and they thought this might be an alternative to raising salaries or just letting people work more from home. If we can actually create the same productive output for the same cost, it means we may not have to create the same productive output for rising costs if we have to raise salaries and things like that—typical things that you'd want a CEO to reason they may be interested in doing something like this.
Drew: David, I didn't include in that list, they've got some juicy quotes of CEOs making themselves out to be absolute saints caring for the workforce wanting to be humanitarian. I actually think that some of the people who went into this trial did it out of genuine humanitarian concern for their staff. I just don't think you can read much into that as data and what CEOs say are their reasons for doing things.
David: There's a general process and leading that the companies followed for how to do this, how to create the policy around this, how to get a pilot up and going, how to communicate with their staff, what opt-in agreements look like, and how to devise some productivity initiatives to try to balance the reduced working hours with reduced workload and introducing software.
Let's talk through some of these points because the first idea is if we're going to reduce the number of hours people are working, which is the supply side of the equation, we almost also need to reduce the demand side of the equation, or obviously things just break. I guess it's an opportunity to reflect on the amount of time wasting that goes on at work. Is 20% of people's time a week maybe genuinely unproductive that we can afford to lose without the company losing any productive output? Do you want to share some insights around that?
Drew: That is basically the overall theory behind this four-day workweek proposal. How is it possible to go from five days to four days? The assumption is that by making this work for five days, we're just spending a lot of unproductive time, a lot of make-work, a lot of being at work because we have to. That's why people think it might be possible. We're not trying to accelerate the pace of work, we're trying to remove all of the unnecessary work.
It's quite similar to something that the Vice Chancellor of our university said when we were going through our Covid related staff cuts. She just said, clearly, we're losing a chunk of staff. The only way we can reasonably do this is if as an organization, we do less stuff. As part of this process, we're not just reducing costs, we're looking at how we can do less stuff.
I think the attitude that everyone went into this trial with is, how can we do less of the things that we think are unproductive? Lots of these organizations in this preparation period, some of them went into quite formal auditing their work processes. Some of them did it more consultatively, some of it did it more just collaboratively. As an organization, let's just do it, let's just find all the things, and get rid of them.
It gives this permission to people to say, oh, if we're going to go to four-days weeks, then we better stop doing this. They gave examples of people who are looking at during the day travel time and how they get rid of that, automating things that could easily be automated, things that have multiple people doing them, just doing it with half as many people. If you have an hour-long meeting and only half as many people are there, then half of those people don't need to be there that day, and you still get the same meeting done.
David: A connection here to lean engineering and where's the waste and inefficiency in the company. This is a forced way of saying, we're going to work less hours, so we've got to find where this sits in our business. I like the way you give permission to say, we're only working four days now, so is it really critical that we do this? Drew, I'll let you pull out the themes from this qualitative data.
Drew: One thing that I thought that's worth talking about is organizations try to say that we are working collaboratively with our workers. We want shared goals, shared values. The four-day workweek is a genuine shared goal for productivity. For a lot of these organizations, it was like, okay, the company wants four days, the work is one four days, let's work together to get there.
It's quite different from normal efficiency initiatives, where often workers will fight back against efficiency if they see that as just an excuse to make them do more work with less resources. The company benefits, but what does the worker get out of things being more efficient except having to work harder? Whereas here, there's a genuine thing that the workers achieve.
It was interesting that a lot of the dissatisfaction came from organizations that did that explicitly. You set up conditional KPIs and said, you can only have the four-day week if you prove that you can achieve the productivity and if you maintain it. If the KPIs drop, we're going back to five days. Those companies have the least buy-in and satisfaction from the qualitative results. That's where most of the complaints came from. Whereas people seem quite to enjoy this collaborative, let's just work together and cut the week down from five days to four.
David: Drew, if we can't reduce the demand like that, one of the other solutions is achieving it by growing the workforce. I think a couple of companies or two organizations said that they were going to recruit some additional people to achieve the four-day workweek. We can't quite get the work down enough, so we'll give those people the same pay, but we're going to bring on some more people.
Two other companies said that if they move to this four-day week permanently, then what they'll do is they'll increase their reliance on subcontractors. They wouldn't be engaging the subcontractors on the same four-day week, five-day pay terms. They would be outsourcing some of the work in traditional working arrangements.
Drew: It's only a small number of companies here. We're talking whether that's 4 out of 22 or 4 out of 44. I think this is the warning sign for more operational organizations who might be thinking of the four-day week. What have we really achieved if out permanent employees have this flexibility, but our solution is given to a subcontractor organization who pay people less, less unionized, less control over their work, less control over working hours, and we just ended up with a whole bunch of employees who are not our employees, who are working long hours, in order for our people to have the four-day week?
That's just eerily similar to what sometimes happens with dangerous jobs and safety. Organizations try to improve their management of hazards by outsourcing it to subcontractors who have no specialist expertise. They're just willing to subject their employees to more risk.
David: I think that's right. There's a few other things in this qualitative data. We'll link the report in the show notes and in the LinkedIn post so you can read through what people actually said about it in some more detail. I thought we'd move on to the relationship between the company, what people did with the extra time, and then move on to some conclusions.
Drew: I know you're trying to excel at this. Let's just quickly pause on the list of ways that the company has improved productivity because I think we could just quickly read out that list as interesting.
We've got companies reforming the norms around meetings. Shorter meetings, less frequent meetings, clear agendas and objectives, improved email etiquette, basically seeing fewer people in emails, analyzing different steps of the manufacturing process, having designated focus periods each day for heads down work with no talk, reorganizing calendars so there's less multitasking, more dedicated time on individual tasks, and more creation of task lists and to-do lists so that people hand over work and start the day more productively.
Some interesting things there that might actually be good for the organization and good for the people. I was really interested that they even asked this question, though, which was about what people did with the extra day. It starts off really wishy-washy about work-life balance. But then they ask the companies what they thought people did on the extra day, and that's where it got really fascinating because the companies were making assumptions about what people would do with the free time.
Some of the companies seem to think that they had a right over what people did with that extra day. Most extreme example of this is, are people allowed to go and do another job on their fifth day? A lot of the companies just said, whoa, hold on a second. That's your day, but it's your day so that you can relax, so that you're productive for me. You're not allowed to exhaust yourself on that fifth day by working for someone else.
David: Even if my inability to afford my rent and cost of living crisis that's going on at the time means that the reason that half of the people have no change to their stress, maybe having eight hours more of their time just gives them eight hours more time to worry about stuff.
Drew: Some of the companies, particularly the charities are thinking, okay, they get an extra day, they can do volunteer work and be good for the community, and their employees are all playing video games instead. Hold on, we want you to use this time productively and positively.
David: I think it's this paternal idea that one of the examples was encouraging staff to share stories and bring pictures of things that they did with their day off on message boards. It's like in safety when we ask people to bring photos of their children. My reason for being safe is much. We just make these assumptions that are just overreach. What organizations should legitimately have is boundaries into people's private lives.
Drew: The research is divided into three categories of disinterest, interest, and direction. Some companies just didn't care. We've given you the time back, it's your time. Other companies wanted to do this paternal—we're giving you this time, but we want you to come back and tell us how great this time was. What do you do with this time? And how are you sharing more time with your family?
Direction is where companies want to try to influence it and say what you could be doing, should be doing with the extra time, or what you're not allowed to be doing with the extra time. It was really interesting, given the philosophy of why we're having the fourth day. We'll get to this point. We've actually got quite an inconsistent view from the companies about what the extra day actually is for.
David: Drew, I want to throw some conclusions from the report and get your views on it. Three conclusions and then some practical takeaways. The first conclusion is that the report claims that there is a business case for the four-day week based on productivity. Thoughts?
Drew: Absolutely not. The way they're gathering the data means that those productivity figures cannot be trusted. Way too many caveats to claim that we know anything about when these companies are more productive, less productive, or the same, or which subset have gone up or down. We have no idea from this.
David: We can't take this report and drop it on the table inside a real organization and say, the business case is clear. The second claim in the report is that there are strong benefits for staff with changing these work arrangements.
Drew: That one's a bit better. There are some caveats. There are mixed results tending to be positive. It's good for some people, it's bad for some people within the sample of people that they surveyed. It's good for more people than it's bad for, but we don't really know which people it's good for and which people it's bad for. That one is probably positive, but you got to be cautious. You might be good, but you got to check, will this be good for us or for our employees?
David: The report raises some really interesting qualitative data about the relationship between a goal like the four-day week and employee engagement in things like decluttering and workplace improvement activities. If you recall, when we did the Hardwood experiments episode from Kurt Lewin in the 1940s, I think that was somewhere in the 90s episode or numbered somewhere in the 90s. It raised similar things about how you can get employee engagement and collaboration around what the targets should be and how to achieve them. Drew, what are your thoughts about that claim?
Drew: I think these researchers are looking to actually publish a couple of academic papers out of this. I'm looking forward to them because I think there's some really insightful stuff. We can't tell from the limited way it's reported here. We need to know more about which organizations and at least some sense of who said what.
I think the reasons organizations adopted it, the ways they went about adopting it, and some of the motivations and experiences of the CEOs, I think they do have some genuine qualitative data there that I'm looking forward to seeing that published properly in fuller form.
David: Yeah, great. Maybe you can get yourself on a peer review and find out which journal. Drew, takeaways. Do you want to share your practical takeaways for people having listened to this episode, maybe going and reading this white paper for themselves?
Drew: Okay. My first one is just my gripe when it comes to papers like this and using them as evidence. If you're talking about employee well-being, never ever use averages. Don't do it as an academic, don't do it in your own organization. If 70% of people are happy and 30% of people are unhappy, don't say on a weighted scale, we moved from 3.5-3.7. Go and find out why 30% of people are unhappy. It's not a good way to use statistics at all.
David: I'll let you keep going, Drew. You're on a roll.
Drew: Okay. The second one is that I think there's a real lesson here about genuinely finding shared goals for people to work together when it comes to safety and well-being. I think the four-day workweek shows that it doesn't have to be patronizing, but it can be a genuine shared goal to make life better for everyone and good for the company. As long as you're not being patronizing about it, and you're not sneakily just trying to do it to be more efficient, I think things like this are genuinely good as just thought leadership of initiatives that we can take. I'm interested in your thoughts though on that one, David?
David: I think people who plan the battle don't battle the plan. I like collaborative decision making in general, but I really like it in relation to goal setting and how to achieve those goals. If an organization were to say, we're committed to making this work and if we can get the same outcome for the organization, workers can get a day off a week, and we've all got to work together to figure out how to do this, I think that's an incredibly productive environment.
Drew: The final takeaway, I think, is really practical, which is just be careful of burden shifting when you're trying to improve safety and well-being, particularly burden shifting to other parts of the organization or to subcontractors. When we try to make life easier for some people, often that does just make life harder for other people. We do need to be really aware of whether we're genuinely helping or whether we're causing the improvement just by putting someone else at risk or making someone else work harder.
David: I think also, the lack of equity in the workplace as well. We see this in organizations who have pursued flexible work arrangements, and they've got operational staff who do those things, like you said, operate equipment, drive trucks, do field work, work genuine shift patterns, they can't work from home, they can't choose which hours they work. It created these tools to speed economies in organizations, where we make things better for other workers. One of the unintended consequences is it can really damage our culture too.
Drew: Particularly, in order for the people to be able to work at home, it's frontline people, then you have to fill out more paperwork to facilitate working from home.
David: The paper claims that the results are in and claims at a success for the first day week. The question we asked in this episode is could a four-day working week lead to happier, healthier, and safer workers? Do we know the answer to that?
Drew: Certainly it could, but the results are definitely not in. We've got a short, uncontrolled trial at an unusual time across an unrepresentative group of companies. Even then, it produced mixed results. We've got this interesting, promising idea of the four-day week.
I personally love the idea, but it requires a lot more rigorous examination before we just say, yes, we can put this in front of our companies, and you expect them to just think that the business case is already made. It's not. I wouldn't put this in front of my own organization and say, look, here's the evidence, we should just do this. It's a good idea that we should be very cautious about promoting it as a good idea.
David: Thanks, Drew. That's it for this week. We hope you found this episode thought provoking and ultimately useful in shaping the safety of work in your own organization. Send any comments, questions, or ideas for future episodes to us at feedback@safetyofwork.com.