Trust Me, I'm An Expert: what the huge HILDA survey reveals about your economic well-being, health and family life
- Written by Sunanda Creagh, Head of Digital Storytelling
On today’s episode of the podcast, we’re talking about what one of Australia’s biggest longitudinal surveys and richest data sets, released today, says about how the nation is changing. And some of the trends may surprise you.
The Household, Income and Labour Dynamics in Australia (HILDA) survey tells the stories of the same group of Australians over the course of their lives. Starting in 2001, the survey now tracks more than 17,500 people in 9,500 households, asking about their economic well-being, health and family life.
So what does this year’s report tell us about the country Australia has become?
Here to break it all down for us today is Roger Wilkins from the University of Melbourne, lead author of the report.
Wilkins said he was surprised by what this huge survey showed about Australians’ financial literacy, our energy use, how many of us are putting off getting a driver’s licence, how our economy is changing, and how our attitudes toward marriage and family life are shifting.
The report reveals some insights into where we perhaps need to concentrate our public policy efforts to boost Australia’s economic well-being.
What does it all mean for you and me? Listening to Roger Wilkins explain it all may just inspire you to rethink your own financial future.
Roger Wilkins spoke to The Conversation’s deputy politics and society editor Justin Bergman. We’ve included an edited transcript below.
What is HILDA and why does it matter?
Justin Bergman: Roger, in a nutshell I’d love to start out by just hearing what the HILDA survey is and why this matters to people.
Roger Wilkins: The HILDA survey is Australia’s nationally representative longitudinal study of Australians. It started in 2001 and it’s a bit like the ABS (Australian Bureau of Statistics) cross-sectional surveys we’re familiar with, where we get information on people’s employment, their family life, their incomes, their health and well-being. But what distinguishes HILDA is that we’re following the same people year in, year out.
So we are getting a moving picture of people’s lives rather than the cross-sectional snapshot or photograph that the ABS surveys give us. So that’s really what’s unique about HILDA.
We’re now entering our 18th year. So we’re getting a really rich picture of how people’s lives evolve over time, and it allows us to answer all sorts of questions that we couldn’t do with cross-sectional data.
Things like: if someone is poor in one year, how likely are they to be poor the next year? You can’t answer that with cross-sectional data but with our data you can see how long, whether it’s the same people who are poor year in, year out, or whether it tends to be a temporary affair.
And moreover, you can look at: well, who are the people who managed to get themselves out of poverty? And who are the people who don’t?
This gives us incredibly useful information for policymakers about who are the people who are persistently struggling, for example, and therefore we should be thinking more about from a policy perspective. And that’s just one example of many in terms of the value of the HILDA Survey.
Some of the findings in this year’s HILDA report.Energy spending is falling
Justin Bergman: I realise it’s probably a great deal of data to pore through and lots of interesting findings we’re going to get into in this podcast. Were there any that you found particularly surprising or interesting, just off the top?
Roger Wilkins: Well, we have been tracking people’s household expenditure since 2005 and that includes their expenditure on home energy. Things like electricity and gas. So, we thought, well, there’s been a lot of attention recently to rising prices for electricity and gas. So we thought, well let’s have a look at what’s been happening to household expenditure. That’s different to the price because your expenditure depends on not only the price but how much of the energy you use.
And one thing that surprised me was that the HILDA data is showing that people’s expenditure actually peaked in around 2014. So since then people have actually been decreasing their expenditure, in real terms at least, adjusting for inflation.
So that was something that I wasn’t expecting because there’s been a lot of recent media about prices continuing to rise since 2014 and yet expenditure hasn’t been rising since 2014.
What it seems is that people are have been adapting to these higher prices and doing things like buying energy-efficient appliances, insulating their homes, installing solar panels, perhaps heating fewer rooms in the house in winter. That sort of thing seems to have been going on.
So, as I said, the total expenditure on home energy has actually declined slightly since 2014.
Cognitive ability and decline
Justin Bergman: Great. And one of the interesting chapters that we thought was quite surprising was the one about measuring cognitive ability. And I wanted to ask you, starting off, what are the factors that you looked at in this chapter, when it comes to what contributes to cognitive decline?
Roger Wilkins: Yes, so we have now in two years - in 2012 and 2016 - administered these tests which are called “cognitive ability tasks”. They ask the respondents to perform various activities which allow us to produce measures of their cognitive functioning or their cognitive ability.
And because we have, as I said before, we’re following the same people year in, year out, we can actually look at how these measures of cognitive ability changed between 2012 and 2016.
And we do indeed find that, particularly at the older end of the age spectrum, that there is considerable cognitive decline; that people’s performance on these tests does decline, particularly once you sort of get over the age of 70 - 75. That’s when we really start to see that decline becoming quite sizeable.
So one of the things that we did in this year’s report is looked at whether there were things other than age that were predictive of cognitive decline. And we were particularly interested in whether there were various cognitive activities or other activities that you might engage in that could protect against cognitive decline.
So we looked at things like how often you do puzzles, things like crosswords, how often you read, how often you write, whether you use a computer regularly, whether you do any volunteering, whether you are actually doing any paid employment, how often you look after grandchildren. These sorts of activities, the basis that perhaps the more stimulated you are cognitively, the less decline you’d experience.
And the overriding result we found is that very little seems to protect against cognitive decline. We find some evidence in favour of doing puzzles regularly, things like crosswords, where on one of the measures of cognitive ability it did seem to reduce the extent of decline.
But broadly speaking, most of these cognitive activities didn’t seem to impact on the extent of decline.
Justin Bergman: But doing puzzles was one that you saw that did have an impact. Any idea why that might have been?
Roger Wilkins: Well, I mean, the logic is that it’s sort of the “use it or lose it” argument; that if you’re using your brain, in the same way as if you were exercising a muscle, it keeps it in better condition. That’s sort of the logic. But for some reason we don’t, for example, find that with writing regularly.
That probably is suggestive that doing your crosswords or Sudoku or the like is perhaps not a bad idea, particularly if you enjoy doing them, because it might be having this beneficial side effect.
We also looked at perhaps what you think of as behaviours that might be adverse to cognitive functioning. So, in particular, things like smoking and drinking. And there is some evidence that heavy consumption of alcohol does accelerate cognitive decline but we don’t find any effects of smoking.
Justin Bergman: Very interesting. So do your puzzles and try to avoid alcohol as much as possible.
Roger Wilkins: Sounds like common sense, doesn’t it?
More young people are delaying getting a driver’s license
Justin Bergman: So, going to the chapter about people driving in Australia, what did you notice about the data on driver’s licences?
Roger Wilkins: Yes, well, I mean, people would not be surprised to learn that most people do have a driver’s licence. Although a surprising - well, for me, at least - quite a surprisingly high proportion of young people in the 18-24 range don’t have a driver’s licence.
So while most people — over 90% — eventually get their licence, for many of them it’s not until their late 20s or even their 30s when they do get their licence. So, for example, in the 18-19 range, over a third of people in that age range don’t have a driver’s licence.
And something that we see in just the four-year period between 2012 and 2016: when we asked people whether they have a driver’s licence, even over that short period, we have seen a decline in the proportion of people who have a driver’s licence in that age range.
So whether that’s because the requirements in order to pass the test have been tending to ramp up in most states, I’m not sure. Certainly, there obviously have always been significant costs for obtaining a licence which might be a barrier for young people but I’m not sure that, you know, the extent to which those costs have increased. For example, requiring logbooks with a certain number of hours of driving, I’m not sure exactly the timing of when those increases in requirements have occurred.
But certainly this data is showing an increasing proportion of young people without a licence.
Many men hang on to their driver’s licence until later in life
Justin Bergman: Right, and on the opposite end of the spectrum, we also noticed that a large number of people in the older generations have driver’s licences. But specifically, you found 74.6% of men born in the 1920s still held a licence in 2016. So what does the data tell us about older people with driving?
Roger Wilkins: I think certainly we see that loss of licence - whether it’s relinquished or having it cancelled - is very much concentrated amongst older people.
Although surprisingly, for me at least, a surprisingly high proportion of young people do seem to lose their licence over a four-year period. So, you know, at least sort of 2-3% of people in their 20s and 30s reported that they had a licence in 2012 and they didn’t in 2016.
The extent to which that is because they had traffic violations that resulted in suspension or cancellation, I’m not sure. We didn’t ask why they didn’t have a licence but that did surprise me.
Certainly, the rates of loss of licence are much higher amongst the older age groups. But, as you said, nearly three-quarters of men born in the 1920s - so they’re all, I guess, at least 86 years old in 2016 - so at least three-quarters or nearly three-quarters of men aged 86 and over still held a driver’s licence.
And that does certainly seem quite high and it’s certainly a lot higher than amongst women. So it does seem that men hang on to their licences a lot longer.
Now, that doesn’t necessarily mean that they’re driving. You can have a licence without driving but it’s probably strongly correlated with driving. So it does suggest that, you know, particularly men are able to hang on to their licence longer than or are willing to hold on to them longer than women.
The share of workers who describe themselves as ‘self-employed’ is falling
Justin Bergman: Moving on, can you tell us a little bit about what the data showed us about self-employed workers that you found interesting or surprising in the survey?
Roger Wilkins: Yes, certainly what is interesting is that self-employment has been declining this century. So at least the proportion of people in the labour market who describe themselves as self-employed has been declining for the last 16 plus years. And that’s especially concentrated on people who employ others, so we’re really talking here about a decline in small business.
And this is, I guess, rather at odds with a lot of the rhetoric we see amongst politicians about how small business is the engine of the economy and where growth in jobs comes from. And the evidence is quite to the contrary: that in fact our employment growth has really not come from small business or from self-employment. It’s really been coming from larger employers, be they medium or large businesses.
And I think given what we’re observing in the trends, I think that’s something that over coming years will continue. That it is probably, to my view, a bit wrong-headed to think that that stimulating employment in small business is the way to generate jobs in the community more broadly. I think the changing structure of the economy is actually moving us further away from that than towards it.
And in that context it’s also interesting that for all the talk about the rise of the gig economy, these are these jobs that are I guess facilitated by apps on phones and things like Uber and Deliveroo and things like that. We don’t see evidence in the data of much growth in employment of this kind.
So as I said, self-employment has actually been declining. You might think well, maybe many of these gig-type jobs are secondary jobs. So that while people in their main job are employed, perhaps in a second job they’re a gig worker doing some Uber driving on the side. But we haven’t seen a growth in multiple job-holding either.
So it seems so far that most of these gig jobs have really been about transforming jobs that already existed, so taxi drivers have always tended to be self-employed. Uber drivers are self-employed. We’ve got sort of a compositional change, a bit of a shift away from traditional taxis towards these Uber drivers.
It’s the same with, I guess, food delivery. Casual observation would suggest that there has been a rise in these gig-type jobs but I think it’s easy for us to overestimate how significant a phenomenon this is.
On marriage and housework, our attitudes are changing faster than our behaviour
Justin Bergman: It is really interesting, actually. We were also quite surprised by some of the things you see in the attitudes towards marriage and family changing. And I was curious what you’re seeing in terms of attitudes in Australian society becoming more progressive on this front.
Roger Wilkins: Well, you’ve essentially summed up what we find. For a long time now, we’ve been tracking people’s attitudes to marriage and family and to parenting and paid work.
This allows us to produce measures of the extent to which people have what we might call “progressive views”, which, in very loose terms, is the extent to which people are in favour of men taking a more active role in raising children and women taking a greater role in bringing in the household income, so being more engaged in the labour market. There are other dimensions to these measures of traditional views versus progressive views. But that’s probably the most important dimension and we’re certainly seeing that views are becoming considerably more progressive. There’s been quite substantial change over the course of this century.
What, to me, was interesting is that when you actually look then at how behaviour is changing then it seems that these changes in attitudes aren’t really translating so far into much change in how people behave.
This is very much connected to the arrival of children. So before children arrive, men and women have quite similar-looking division of their labour – the amount of time they spend in employment, the amount of time they spend on housework and so forth is quite similar.
But once the first child arrives, and this is probably not news to anyone who has had kids out there, but there’s a sharp divide that opens up between men and women.
Women withdraw, to a large extent, from the labour market and men, to a large extent, withdraw from the home production - if you like, from the housework and the care.
And what’s really interesting is how this persists. So even once the children age and move through school and even beyond, we still see this division persisting. So the arrival of the child precipitates a change, so even when the care requirements of the children diminish and so forth, we still see this divide between men and women persist.
So, I think there are good economic explanations for this but I also think that there are reasons, from a public policy point of view, for us to be concerned about this.
And really, it relates to the fact that we know probably around one in three marriages will end in divorce - maybe more, maybe a bit less going forward. That, therefore, means that women are much more vulnerable in that post-divorce world than men because they’ve put their careers on hold, their income-earning potential is considerably lower than men’s. And so therefore their economic well-being is likely to be lower than men’s post-divorce.
And then that has flow on effects into their retirement living standard because their superannuation contributions will be lower. So I think while it might make economic sense for men and women to specialise in this way, it is having this undesirable longer-term consequence for women’s well-being and that’s why we do see higher rates of poverty amongst single women, particularly single parent women and elderly single women, than we see for men.
Justin Bergman: Interesting. I don’t know if you’ve tracked these data for same-sex couples as well. Have you noticed any divisions in terms of attitudes toward housework and the divide in other types of marriages?
Roger Wilkins: No, so we haven’t looked at that this year. One of the problems in doing so is that the HILDA survey is a sample survey, so while we have 17,500 people from right across Australia, which gives us a lot of potential to produce reliable estimates on what is happening in the community, when you look at particular demographic subgroups it becomes more difficult to make reliable statements about overall trends.
So while same-sex couples are a significant minority in the community, they are still quite a small part of the HILDA survey sample. So it becomes a bit more difficult to be confident in estimates based on small demographic groups.
Australia has its problems, but society still functions well for most
Justin Bergman: Just wrapping up, looking at the data as a whole, I’m curious what story you think it’s telling about how Australia’s going, how it’s changing, where it’s going at the moment?
Roger Wilkins: Well, I mean for all its problems - and there are, of course, many - the clear picture from HILDA Survey that Australia is a well-functioning society in which most people feel able to pursue fulfilling lives, pursue their aspirations and live the life, or a form of the life, that they aspire to.
So while, of course, there is much to do to make our society work better, I think we risk making some big mistakes, moving forward, if we aren’t cognisant of how much is already working quite well.
I think that’s something that probably gets lost a bit in a lot of public discussion and media. We tend to focus on the negative and that creates, I think, an impression of much greater dysfunction in our society than is actually the case.
Now, you have always got to very quickly follow up such a statement with the caveat that of course there are problems and HILDA certainly identifies many of these problems and concerns that we should be addressing.
But I guess there’s always the risk of of overreacting and therefore damaging things that are good about our community in seeking to solve some other problems.
So, that said, what sort of trends come out of the data that would be of concern? I think decline in home ownership is a very big concern that has a very strong link to growing evidence of intergenerational inequality, so particularly younger people in the age range up to around 40, compared with older people, the baby boomer generation. There’s been a growth in inequality across the generations and it’s very much tied to home ownership.
We also we have this persistent disadvantage among many single parents and I think that’s a continuing priority for policy, in my view. And the other persistent trend that remains a concern is that household incomes are quite stagnant and that’s very much related to the stagnation in wages.
It’s one thing to be concerned about it but it’s less obvious what you do to address it.
Justin Bergman: Roger, thank you so much, very illuminating. We really appreciate you breaking it down for us and taking the time to be with us. Thank you.
Roger Wilkins: Thank you.
Additional audio
Kindergarten by Unkle Ho, from Elefant Traks
The Anthill podcast, episode 27: Confidence, from The Conversation UK.