Josh Frydenberg says welfare dependency was at a 30-year low heading into the pandemic. Is that correct?
The Coalition Government has reassured Australians that the nation is well placed to emerge strongly from the coronavirus crisis in large part because of its handling of the economy in the years leading up to the pandemic.
Treasurer Josh Frydenberg recently pointed to the Government’s success in cutting the unemployment rate before the pandemic, claiming this had produced the lowest welfare dependency for three decades.
“And let’s not forget that going into this crisis, we saw welfare dependency at a 30-year low because we’d got that unemployment rate down,” .
Is it correct that welfare dependency was at its lowest level in 30 years before the coronavirus crisis sent the economy into reverse? And can this be attributed to lower unemployment? RMIT ABC Fact Check investigates.
Yes, the data shows welfare dependency is at a low, but there is more to the story than Mr Frydenberg’s claim suggests.
Fact Check relied on two measures of “welfare dependency” when assessing Mr Frydenberg’s claim: the proportion of the working-age population being paid a major welfare benefit and the proportion of households getting at least half of their income in the form of government benefits.
In the case of the former, about 14 per cent of working-age Australians were receiving a major welfare benefit in June 2019 — the lowest rate since the early 1990s.
And, in the case of the latter, 22.5 per cent of households relied on government benefits for half or more of their income in 2017-18 — the lowest since at least 1994-95, when the current statistical series was started.
Although falling slightly short of the time-frame referred to by Mr Frydenberg, both measures lend weight to the claim that welfare dependency had reached the lowest level for 30 years before the pandemic struck in early 2020.
However, experts consulted by Fact Check noted that lower levels of unemployment had not been the only reason for this shift.
Government policy changes have also played an important role. These have included the introduction of tougher eligibility rules and obligations for payments such as unemployment benefits and the Disability Support Pension; the abolition of some payments; and, declining payment rates relative to wages for some types of income support.
Moreover, in the years leading up to 2020, there was a significant increase in the number of working-age people in Australia not entitled to social security.
This growing cohort includes international students, working holiday visa holders and skilled migrants and an unknown number of New Zealanders whose social security rights were reduced from the early 2000s.
It also includes recently arrived permanent residents, who now face a four-year wait before they can receive government benefits.
Although difficult to quantify precisely, this increase has had the effect of lowering welfare dependency relative to the labour market.
What does it mean to be dependent on welfare?
There is some subjectivity surrounding the notion of “welfare dependency”.
Roger Wilkins, deputy director of the Melbourne Institute of Applied Economic and Social Research, pointed to a simple definition. He said an individual getting an income support payment from the Government could reasonably be regarded as “welfare dependent”.
“There are a few different ways of defining welfare dependency, but receiving an income support payment — the main ones being JobSeeker, Disability Support Pension, Parenting Payment, Youth Allowance and Carer Payment — is a satisfactory definition of ‘welfare dependent’,” Professor Wilkins told Fact Check.
The Department of Social Security publishes data showing the number of income support payments for various types of welfare.
Peter Whiteford, a professor at the Australian National University’s Crawford School of Public Policy, said individuals can only legally get one major income support payment, ruling out double-counting for major payment types.
“Income support payments are mutually exclusive, so you can’t legally receive more than one at a time,” Professor Whiteford told Fact Check.
“So these numbers are for income support payments only and will be correct at a point in time, usually the end of June each year.”
The level of “welfare dependency” is often expressed as a proportion of the working-age population (16 to 64 years old). This accounts for changes to the size of the labour market over time as a result of population growth (or other demographic changes).
Following advice from both Professor Whiteford and Professor Wilkins, Fact Check combined Department of Social Services income support payments data with Australian Bureau of Statistics (ABS) demographic data to measure the proportion of the working-age population receiving an income support payment at the end of June each year.
Welfare dependency can also be measured using data from the , published by the ABS.
Among other things, the data includes the proportion of households relying on the Government for at least half their income. This offers a second measure of welfare dependency.
Mr Frydenberg referred to welfare dependency as “going into this crisis”. Australia’s first coronavirus case was recorded in late January 2020.
In assessing Mr Fydenberg’s claim, Fact Check has relied on both data sources, bearing in mind that neither covers the time frame he referred to precisely.
Working-age income support
The proportion of the working-age population getting an income support payment from the Government peaked at about 23 per cent in 1996, falling to 16 per cent in 2008. It rose again in the wake of the Global Financial Crisis, before falling from 2014 to a low of just over 14 per cent by June 2019.
This supports Mr Fydenberg’s claim that welfare dependency was at the lowest level in 30 years before the onset of the pandemic.
As discussed, another way to assess “welfare dependency” is using .
The latest figures show that 22.5 per cent of households were receiving at least half of their income from the Government during the 2017-18 financial year. This was the lowest since at least 1994-95 when the current series began.
Although a household that receives at least half of its income from the Government can reasonably be regarded as “welfare-dependent”, it is worth noting the proportion of households that might be regarded as “very welfare-dependent” was not the lowest, according to the latest data.
In 2017-18, 15.2 per cent of households received at least 90 per cent of their income from the Government. This was slightly higher than the figures for 2007-08 (14.3 per cent) and 2013-14 (15.1 per cent).
Lower unemployment: the reason for lower welfare dependency?
Mr Frydenberg claimed the historically low rate of welfare dependency had been achieved “because we’d got that unemployment rate down”.
Experts consulted by Fact Check said lower unemployment was only one part of the explanation.
Professor Whiteford has estimated the level of welfare dependency stretching back for almost five decades. He said that by his calculations welfare dependency in 2019 was in fact at the lowest level since 1980.
However, Professor Whiteford said it was not clear the decline in welfare dependency was entirely due to lower unemployment.
The following graph shows major payment types as a proportion of the working-age population.
It shows the number of individuals receiving unemployment benefits as a proportion of the working-age population was not at the lowest level for three decades in 2019.
On the other hand, there have been significant declines in the proportion of the working-age population getting the parenting payment, youth allowance and, in recent years, the disability support pension.
Both Professor Whiteford and Professor Wilkins pointed out that, while employment growth had made an important contribution to lower welfare dependency, changes to welfare policy — including the tougher rules and the abolition of various payments — had also played a significant part.
Some of these tougher rules forced people from other more generous payments onto unemployment benefits. This partly explains an increase in the proportion of people receiving unemployment benefits.
For example, in 2012 the former Labor government introduced to qualify for the Disability Support Pension, pushing thousands of additional people on to unemployment benefits, paid at a much lower rate.
Professor Wilkins said changes over the years included increased conditionality of payments, especially in the case of unemployment benefits and parenting payments; tightening of eligibility criteria, especially for the Disability Support Pension and Parenting Payment; the elimination of a range of payment types, especially for partnered women and people aged over 45; and a decline in payment rates for some allowances relative to average incomes in the community.
“Commonwealth governments of the last 30 years can certainly claim ‘credit’, but I’m not sure all of the decline is something to be proud of,” Professor Wilkins said.
“Reduced access to benefits will have increased economic hardship for at least some members of the community.”
Growing numbers of workers are ineligible for welfare
Professor Whiteford also pointed out that a complicating factor was a change in the share of resident working-age population not eligible for social security.
In particular, he said, up until 2020 there had been a significant increase in temporary residents without social security entitlements.
This included students, backpackers and skilled migrants, “as well as an unknown number of New Zealanders whose social security rights have been reduced since the early 2000s”.
“These migrants can’t get income support but have boosted the working-age population — and I am pretty sure that this will have had a moderately significant effect on the welfare dependency rate,” Professor Whiteford said.
This argument was supported by Grattan Institute policy analyst Henry Sherrell, an expert in migration policy.
Mr Sherrell said there had been growing numbers of temporary and permanent residents ineligible for welfare.
He estimated that temporary residents ineligible for welfare now made up between 6 and 8 per cent of the labour market.
In addition, permanent arrivals are now made to wait for four years (up from two years previously) before being eligible for welfare.
He said this cohort now made up a further estimated 2 to 3 per cent of Australia’s labour market.
“One of the things that has happened since the mid-1990s is that there has been a growing number of people ineligible for benefits because of changed conditions for visas,” Mr Sherrell said.
Principal researcher: , economics and finance editor