Let's have an honest debate on "welfare dependency"
I still feel quite new to Australia - so when I hear any organisation say its time to have an honest debate on something, I’ll take them at face value.
In that regard the Institute of Public Affairs (IPA) said the following on Twitter:
“The growing culture of welfare and dependency in Australia is alarming and our leaders need to start having an honest, grown-up conversation about it.”
Which is weird to me, since the opposite is happening. So lets have a proper grown-up conversation, where we try to understand what is really happening - rather than using the term as an some type of appeal to personal virtue.
tldr: The IPA analysis exaggerates the change by double counting, looking at total receipt rather than duration, and starting in 2013 (just before a bunch of payments were shifted into NewStart). The fact is that “dependence” has been falling - implying that the weight of evidence suggests that we should be asking whether we are now excluding people we shouldn’t be from receiving support, rather than saying we need to clamp down more on benefit receipt.
Introducing dependency
It is fine to ask about long-term benefit dependence - contrary to the response we received on Twitter about it. People’s values about this differ, and “dependency” isn’t necessarily bad, but it is a term people understand as “relies on income support”.
If you think people aren’t already using the term, and it isn’t part of the debate that needs to be tackled head on then you live in a bubble. And I’m sure the IPA report which perhaps wasn’t in such good faith would have been a life shock to you that made you reflect on your purity testing of language in our micro note … nonetheless.
Note: For my sarcasm, I do dislike the dependence term and am a closest universal basic income as a social dividend type of guy. But as an economist I want to make sure I’m informing people based on their values - not my own. I also detest people who purity test rather than trying to work to incrementally improve peoples lives - and I’d gladly die on that hill.
Why isn’t dependency bad - mate as a society the vast majority of us want to help out people who are unable to support themselves, or have come into a long period of bad luck. That support is by definition something people receive for a long period of time, and depend on to support themselves - and I damned well hope that everyone who “depends” on this receives it!
Instead, the concern the IPA is raising is that there may be people getting access to payments who could support themselves - but are free-riding on the benefit system.
How do the IPA numbers reflect their concern?
Any concern about free-riding is pretty difficult to make solely from a number of recipients - as we could state that the system itself is set up to support those specific people. If this was an issue that required urgent action - the claim they want to make - we would need to show that the number of recipients had been rising without a legitimate driver (i.e. growing disability, rising unemployment).
And this is the claim they intend to make.
The numbers are given in their report here - and these numbers show a sharp increase in the number of benefits paid. So where do the numbers come from - it doesn’t say. But if we take the monthly DSS data here we can find similar numbers for the JSP, DSP, and YA. For the NDIS we can look at participant numbers here, and and the numbers look similar to the “ALL” category for those aged 15+.
If these are the numbers, which they appear to be, then a few issues crop up:
They are not counting the number of individuals - they are counting the number of payments. So someone who receives the NDIS and DSP (commonly around 75% of participants) at the same time is counted twice.
The inclusion of the NDIS is a bit random as it isn’t income support, it is a payment for services - why don’t we include the childcare subsidy, or small business tax exemptions?
They are looking at all payments - but the concept of “dependence” is the need to stay on the payment for a prolonged period of time.
Starting in 2013 is a little problematic - as the e61 piece below will make clear.
As we’ll see, these assumptions completely change the story.
Quick NDIS note
Why do so many people receive both NDIS and DSP? Why are they getting two separate benefits!
Well, they are actually quite different things. The DSP is an income support payment to prevent poverty among those who are permanently unable to work. The NDIS is an insurance system that will turn around and subsidise expenses related to the specific disability faced by the individual.
The purpose of the NDIS is an insurance scheme to pay for costs specifically related to the condition people face - the DSP is there so an individual in a difficult situation can eat and pay rent.
In this way NDIS is not income support unless we’re going to start counting childcare subsides etc as income support as well.
Other numbers - Antipoverty Centre
On the other side of advocacy is the Antipoverty Centre chatting about benefit numbers.
They do a pretty good job noting some of the complexity that comes in over at their substack - in terms of benefit age changes, partial capacity to work, partial benefit receipt etc.
Their goal is to state that the base number is too high. However, as we’ve noted in the argument above this isn’t really the claim the IPA is making - they aren’t focused so much on the number of the people, more the fact it is higher and they don’t believe it should be.
As a result, a reasonable commentator may look at this and say “APC has some good points about the number - but the trend remains, more people are relying on support and something must be done”. But is this true?
Other numbers - e61
Due to pure circumstance, researchers from e61 (including myself) had asked about this question a few weeks ago. I have blogged about it here, but let me discuss ;)
When looking at this we did not include the NDIS, carer payment, or any supplementary payments - as the introduction of the NDIS makes this complex as noted above.
However, unlike these other pieces we did add in parenting payments - as there were significant reforms switching eligibility between Paid Parenting Payments and what we view as core unemployment benefits. In that way, for a like for like comparison it is good to include both.
Furthermore, we track individuals through time to understand if they have a persistent period of an income support payment - even if they switch between payments. We also allow for short gaps between payments when calculating these persistent periods.
What did we find? We found red line.
Or in other words that the proportion of 22-60 year olds (chosen to avoid the changes to the Youth Allowance and Age Pension over this time - things we have separate work coming on) who are receiving income support for over two years is at historically low levels.
So given this, what is the grown-up conversation we should have now?
Instead of trying to generate a new crisis around benefit numbers, why don’t we try to understand how the system - as it stands - supports those in difficult situations, and helps individuals in hard times get back on their feet. I can’t understand why we would be interested in understanding much else.