16/05/2024
Thank you everyone for joining us the day after Budget, we certainly appreciate this is a contested time and thank you to the Economic Society for doing this.
We're going to talk through some of the research we've been doing at CEDA over the past couple of months.
CEDA is a member-based think tank and, as Tom mentioned, our purpose is to achieve sustainable long-term prosperity for all Australians.
The way we're doing this is through advocating policies and good economics which can enable that.
We've been around for 60 years, and we now have over 400 members, which are a mix of corporate, government and NGOs.
Some of you may know us from our events, we have a pretty packed agenda.
We just had a Climate and Energy forum in Brisbane last week and we have our hallmark State of the Nation event in Canberra next month.
Liam and I are part of our growing research team and our research is what we're going to be talking about today.
I'll be talking through our research into working from home, which was a part of our Economic and Policy Outlook report earlier in the year and Liam will be talking about a more recent report, Clean energy precincts and how Australia can seize the green export opportunity.
Pre-pandemic work from home statistics
Remote working in Australia – so while the COVID-19 pandemic really revolutionised remote work, it actually has been around for some time.
There are a few early companies which piloted this in the 2000s, and it was also touted as a way to solve Los Angeles's traffic woes – if more workers could work from home with their laptops, they don't need to commute – so it actually has been around for a while.
But in Australia, there's varying estimates. What we settle on is a pre pandemic rate of working from home of about five per cent and we get to that by looking at the 2016 census journey to work question, which basically asks people ‘How did you commute to work today’.
There's an option there for ‘I did not commute’ and ‘I work from home’.
The problem with this is that it includes a relatively small amount of telecommuting, which is what we know as working from home to be today.
What it captures more of is people whose job site is their home.
This is people like farmers or people who may live at a shop front they run or a small business, their job site is their home, so therefore they work from home.
It's not really the type of working from home we’re accustomed to.
Surge in remote work during COVID-19 pandemic
The pandemic changes everything.
The highest estimate of working from home was about 50 per cent of workers were working from home, when all the states were in lockdown.
We now know from the latest ABS Ways of Working survey that about 37 per cent of Australian workers are working from home regularly.
I think that's a good thing to highlight because I think at least on a personal level, doing this research and presenting it to other professionals and working with other professionals, I always think the level of working from home is much higher than what it is.
But it's only 37 per cent of Australian workers, which is high or low depending how you look on it.
No surprises that the highest rates of working from home are happening with the managers, professionals and clerical administrative workers.
The pandemic occurs, it normalises working from home or telecommuting because we know people have been doing extra work at home for a while, that's not new, the telecommuting is.
The debate around working from home
This debate emerges as we become more comfortable with the risk of COVID-19 and enter this new world, we have all these articles emerge, they're everywhere, but a lot of it presents this picture of worker versus employer.
If I could boil down the views you have on one side, employers saying working from home is not as productive, come back to the office or we’ll make you.
Then on the other side you have employers saying no, working from home is as productive, if not more, the office is dead, let me work from home.
There's a lot of commentary, but not as much evidence in these articles.
When we were looking at what to research for our Economic and Policy Outlook earlier this year, we could see there's a risk with this back and forth between employers and employees – as it intensifies, you might get a couple of outcomes just based on whoever has bargaining power at the time.
You could get employers really wanting to squash this and collectively mandating people to return to the office to kill the idea of working from home.
Or you could get employees more concerned about losing their option to work from home and then pressuring to have working from home embedded as a right in their working agreements.
We were looking at what these possible outcomes were, and we didn't think either of them were that good.
As well as this being an interesting topic, something CEDA internally is working through, it's something our members are interested in.
And of course, CEDA is concerned with growing productivity in Australia as well, it aligns with our purpose.
It's against this backdrop that we launched into researching this.
Worker and employer attitudes towards working from home
The first thing we looked at is when you look at sentiments and attitudes in Australia, you see that debate play out in surveys of workers that show that most people are working from home because they have a flexible working agreement or they're catching up on work.
A smaller, but still significant portion of respondents are saying the main reason they work from home is to care for family members.
On the flip side of this is the attitudes of directors.
This is a survey done by the Australian Institute of Company Directors and it shows the concerns which are played out in the media.
It shows that directors recognise the value of working from home for its attraction and retention and wellbeing benefits, but they're a little bit more sceptical about the productivity and cultural impacts as well as innovation.
Pretty in line with what you see written up in various news outlets.
One of the reasons we think that there is this difference between employer and employees is that the perceived productivity benefits mostly accrue to workers rather than the firm.
If you're a worker who commutes one hour each way to work, you can do the same eight-hour day in eight hours, rather than the 10 hours it would have taken you before.
You feel more productive, but the manager doesn't see this. They only see the same eight hours done on teams, so they have a different perception of the productivity.
Research on the productivity of working from home
We wanted to jump into the literature on this topic and the first thing to say is that a majority of the literature and what gets reported on is survey data.
Most of these surveys find that staff report higher productivity, higher wellbeing, and while we appreciate this, it's also worth highlighting that it's very difficult to eliminate bias from these studies.
If I'm a worker and I want to work from home, I'm more likely to perceive that I'm more productive and report that.
We wanted to really focus on studies which were quantitatively measuring outputs and were also randomising the working from home as a treatment so that we could be more confident in saying yes, working from home is more productive or it's not productive.
Key studies on working from home productivity
I’ll just quickly talk about two of these studies and the first and most significant study in this area is by a professor at Stanford, Nick Bloom, who's emerged as the pioneer of working from home research.
He was researching it before the pandemic, so he's now very in demand and he basically takes workers at trip.com in China and randomises who gets to work from home.
What they observe is that people who work from home work longer with the time they've saved from their commute, and they also answer more calls per hour, so that's the 13 per cent productivity gain
Then a second paper, which is done by Natalia Emmanuel, who's a research economist at the Federal Reserve in New York.
She looks at call centre workers in the US and what she observes is a similar improvement in the productivity of workers who start on site and then transition to remote.
But the workers who were hired into fully remote positions are less productive than their on-site colleagues.
She presents this very interesting selection and signalling problem where I'm an on-site worker and I get the option to remote work, but I may choose not to do so because I don't want to be lumped in with this cohort of lower productivity workers, I want to be seen to be more productive and working at the office.
Limitations of working from home productivity research
There are other studies which show productivity decreases, and I haven't highlighted them in the slide because a lot of them are lower in quality or have some flaws.
But it's a good segue into the limitations of this literature as a whole.
It's still quite new, there's not a host of studies in this.
There are three studies which have been done in the US and the rest have been done in India and China.
This is partly because the researchers want to run field experiments, and it's a lot cheaper to run field experiments in these countries.
The problem with this is that how working from home productivity impacts those workers in India and China might have something to do with their working from home arrangements and setups, and this is probably very different from the way we work from home in Australia, so it's not exactly clear how we can draw those results across.
A lot of these studies have also been done during COVID-19, so it's again impossible to separate the anxiety-inducing effects of the pandemic on worker productivity and the fact that children are home from school, so people have to balance their caring responsibilities.
And the third is more of an existential consideration in that, because we're focusing on workers with very measurable outputs, like patent examiners, coders, call centre workers, data entry workers, it's not exactly clear how taking these outputs and applying it to other knowledge workers, such as economists, really translates.
It's not clear how the impact of economists working together, and that collaboration generates returns to our productivity.
This is a broader problem with measuring productivity of knowledge workers in general.
Takeaways from working from home productivity research
There are a few takeaways from the literature which I’ve highlighted here.
The first being that it's not entirely clear what the productivity effects are, but it's also probably a very narrow focus and not the whole story.
The narrow focus of, ‘do people take more calls when they work from home?’ is one element for managers to consider, but there are also other things that make well-functioning, productive teams.
One of those is the wellbeing of your staff, how you can retain your staff, the office costs, what type of tasks you're doing.
It's a bigger picture than just simply do you do more stuff at home.
A few of the stronger findings, which come across quite consistently in the literature, is that remote workers are promoted less than their on-site peers.
This probably comes down to two things, the first being, if you are out of sight of your manager, you're out of mind when it comes to promotion opportunities.
And the other is that we have a suspicion that it's quite difficult for the informal learning and development to occur when workers aren't together.
The key result which supports this is that when call centre workers transition to remote work, they're actually far less likely to take on more challenging calls with more challenging customers and that's one of the markers of how our call centre workers get promoted – they handle more difficult calls, calls get escalated to them.
It seems like when they're not around their managers, they take on less of this, they're more likely to pass it on, therefore they develop less and they're not as competitive when it comes to promotion opportunities.
A similar finding is that junior workers receive less mentoring and feedback when they or their manager are remote working and this is definitely a concern.
Lots of pundits will be worried for the impact on junior staff as they develop and it seems like this may be borne out in the data.
Junior workers are less likely to receive feedback, less likely to receive mentoring, but on the flip side, this may be one of the reasons why senior staff are actually more productive from home because they actually save time not doing that mentoring and that informal teaching – they're actually just able to zero in on their work.
Another pretty consistent finding is that retention is improved and it's unclear how this will change now that working from home isn't novel.
When trip.com were doing their experiment there wasn’t a lot of offering for remote work.
Now that there is, it's not clear how retention will be, but it is pretty clear that workers value remote work, so this stands for something.
Hybrid work – the best of both worlds?
Hybrid work appears to be the best of both worlds, and it appears that two to three days is the right mix for keeping pretty good productivity for hybrid working.
For teams who are concerned about the collaboration impacts of remote work, it's very important that teams coordinate an in-office presence altogether.
The research shows us that even as soon as one member from the team is working from home, the measures of collaboration decreased to a level where all members may as well be remote or virtually calling.
We've all got anecdotal stories of fairly agonising hybrid meetings, but it seems like this is true, you need people all in the room together to get that collaboration benefit.
Impact on labour force participation
This is one element of productivity, but we also wanted to look at whether working from home boosted labour force participation for certain groups.
The benefit here is that if you can boost participation, you can deepen the labour pool and you can get better matching between firms and employees – so there may also be a productivity benefit here.
We've also known for a long time that people with disability have advocated for more flexible work because it makes accessing the labour market easier for them.
We wanted to see that with the widespread adoption of working from home if this meant that more people were able to access work.
We were looking at not just people with disability, but also primary carers, women with young children, these are groups that we perceived would have some kind of barriers to accessing a job site because of their caring responsibilities or health.
Analysing labour force participation data
How did we do it? Well, we went to the gospel of all Australian data – HILDA has been collecting data on working from home since its inception, so shout out to the Melbourne Institute, that's great foresight.
We had all the information on people's demographic background and how much they were working from home, what occupation they worked in, etcetera.
What we're looking for here is participation in working-from-home occupations.
These are occupations we defined as those which saw significant shifts to working from home during the pandemic.
The reason we were looking at these occupations is it excludes occupations which can't work from home - things like hospitality or hands-on medical roles where you need to go to a hospital.
But it also excludes occupations which didn't see a shift, which may do working from home but didn't see a shift during the pandemic.
This is farmers and tradespeople and people that run businesses out of their home – they are technically working from home but what we're really looking at is the impact of telecommuting, so not needing to go to a job site.
This is about 65 per cent of all workers that we focused on in this analysis.
Labour force participation results
The results pretty quickly show a good labour force participation benefit when we compare the change 2019 to 2022.
What we have in this table here is the change in participation both in labour force participation, so of this group how many people are working, and then how many people are working in work-from home-occupations – the denominator in these percentages is the same, it's just the total amount of people in these groups and it's the change from 2019, so pre-pandemic, to post pandemic.
What we see pretty quickly is that it looks like working from home has increased the participation of women with children under four quite a lot – so that's the 8.55 per cent here.
And those with an impactful condition a lot as well, we see a big jump in labour force participation of 7.4 per cent and then of that, 5.81 per cent is in working-from-home occupations.
This is particularly strong because people with impactful health conditions, because of sorting or because of the correlation there are sorted into occupations which typically can't work from home.
It seems like there's like a barrier to the typical white-collar professions which make up that difference.
If we go to the graphs, I think it shows the story a bit better.
On the left we have women with no children and women with children over four, and you can see there's a pretty similar jump in participation, not really that much of a story there.
Women with young children participation rates
Whereas when we consider that or compare that to women with children under four, we see this really large jump in their participation in working from home occupations, certainly far higher than what their trend was.
It was trending up to about 46 per cent and we see 52 per cent and the caveat here as this comes with a bunch of changes to childcare subsidies and things like that brought about by the recent Labour government in this same time, but we still think this is pretty compelling.
People with disabilities/health conditions participation
On the next slide this is people with an impactful condition which I think is another pretty compelling result.
We see a very flat trending participation in working from home occupations until the pandemic, in that pandemic year we see a slight dip and then we see this five per cent increase on the trend.
Primary carers participation
Likewise for people who are primary carers, we see a declining rate of participation in these occupations until the pandemic and we see now this jump of about four per cent.
We think there's a pretty big win for these groups.
Working from home rates across groups
We have here this is the percentage of workers who are working from home in these occupations.
What we can see is if we compare the average from 2012 and 2019, carers and people with impactful conditions worked from home at a higher rate, they always valued this flexibility.
If we look at our post pandemic world, people are pretty much working from home at a similar rate.
The playing field has really been levelled there, there's not a huge amount of difference between people who always valued that flexible work and their peers.
I just want to give mention to Leonora Risse at the University of Canberra, who published analysis of HILDA which showed a very similar thing.
Her work is in The Conversation, which is great stuff.
Downsides of fully remote work
While this is a really good win, we know from the literature that fully remote work comes with some downsides, one of which is collaboration.
And one of the explanations for this is that it seems like there's actually something specific about the way we respond to screens when we're on video calls, which prevents creative idea generation.
It's far better suited to narrow tasks, like prioritisation, and this finding basically comes out of this study published in nature about virtual communication.
Video meetings impact on collaboration
We know that as we transition to remote work, we have more frequent meetings with more people in those meetings and we have less one-on-one meetings with people.
This is what contributes to that idea of zoom fatigue.
And just to really reemphasize this point again, collaboration decreases in this hybrid setting as soon as one person begins to dial in from home, you really need all team members there present together if you're going to get the most collaboration out of that meeting.
Advice for managers on remote work
What's our advice in all of this? We know a lot of experimentation with this is going to fall to middle managers, and we know that they need buy in from their employees.
Our research found that one-third of managers hadn't really consulted with their employees on working arrangements.
Importance of mental health
The other thing to call out is that there is a risk to mental health.
Our previous research has shown that managers typically don't get a lot of training and mental health awareness, and the cost of poor mental health to the economy is estimated to be between $12 and $39 billion, as estimated by the Productivity Commission.
Fully remote work in particular is not for everyone. It can be lonely, it can blur the boundaries of work and home. It can lead to people end up working longer hours, and these all contribute to some of the poorer mental health outcomes, which have been reported with working from home.
Need for dynamic managers
We know that it's really important that we have effective and quality managers because this is the difference between having productive and unproductive employees in a lot of cases and a lot of what this boils down to is to call out our previous research again is that managers need to be empowered to be dynamic.
Our previous research has highlighted that managers in Australia don't always score well in measures of dynamic capabilities and these are capabilities which are different to their ordinary management capabilities.
They indicate managers ability to take advantage of change whether that's organisational or technological.
As firms navigate working from home, it's increasing complexity in the environment, among many other things, managers need to experiment and work it out as they go.
This is one of the many reasons that managers need these dynamic capabilities if they're going to thrive in this environment.