Can we afford the welfare system?

18 November 2015

More information

Peter Whiteford is a Professor in the Crawford School. He works on child poverty, family assistance policies, welfare reform, and other aspects of social policy, particularly ways of supporting the balance between work and family life. He has published extensively on various aspects of the Australian and New Zealand systems of income support. He teaches Social Policy, Society and Change (POGO8024) and Social Policy Analysis (POGO8025).

You might also like

The new social services minister Christian Porter has indicated that he is on the “hunt for savings” in his portfolio, hinting spending cuts on carer and disability payments may be needed to get the budget back to surplus.

The Daily Telegraph has likened the welfare system to “a ticking time bomb” and quotes the Minister as saying the outdated social welfare system was in “urgent need of reform”. The Telegraph article notes that “Government modelling has revealed taxpayer-funded welfare spending in today’s dollars by 2026 will be $81 billion more than current tax revenue.”

In an interview earlier this month on Sky News the minister said “in every single category of the very large spend in social services, the growth is, in any rational observation, unsustainable if it were to go on the way it’s gone on over the last ten years. In all areas, things like the disability support pension and a range of other payments – and there’s many of them in the portfolio – they are growing at a rate greater than the ability of the tax base to sustain them.”

Some of these remarks are similar to those of his predecessor Scott Morrison that “the social services budget could swamp the federal Budget” when he argued that already “eight out of 10 taxpayers work every day to pay our $150b welfare bill”. They are also similar to remarks by the previous minister Kevin Andrews in 2014, who described the level of welfare as “unsustainable” and “relentless”.

How can these claims be assessed? In the first instance it can be pointed out that the disparity between spending and taxing raised in the Daily Telegraph article is essentially meaningless as it compares spending in 11 years’ time with tax levels now, when what is relevant is how much tax will be collected in 11 years’ time. Moreover, as with the argument that eight out of 10 taxpayers work every day to pay for the welfare system, this appears to only refer to personal income taxes and not general tax revenue, which is what actually finances social security spending.

Analysing trends in social security spending

So how should the affordability of the welfare system be assessed? As I have argued previously, to understand changes in welfare spending, we need to consider a range of explanatory factors.
By definition, in any year, the total amount of money spent on a social security programme is equal to the number of people receiving the payment multiplied by the average amount of money they are paid. Using this simple arithmetic, it is then possible to look at the factors that influence the number of people receiving benefits and separately identify influences on the amount they are paid. (This approach has been used previously in Australia by Peter Saunders of the Social Policy Research Centre at the University of New South Wales, as well as more recently by researchers in Ireland.)

The number of people receiving payments is a result of the interactions between Australia’s growing population, changes in the age composition of the population, trends in the labour market and in family structure, and the impact of government policy decisions about who is eligible for payments, as well as changes in other parts of the welfare system.

Part of the explanation is also likely to involve the way individuals respond to changing incentives within the welfare system.

The average level of payments will mainly reflect government policy decisions about benefit levels and income tests, but it should be noted that because in Australia we income-test benefits the average level of payments will nearly always be lower than the basic rate of entitlements.

One of the more important government policy decisions is about how payments should be indexed to reflect changes in community living standards. Currently, a number of major payments – Age Pension, Disability Support Pension and Carer’s Payment – are indexed to wages, while most other income support payments and family payments are indexed to prices.

It is no surprise that payments indexed to earnings will rise in real terms, so long as real wages are rising. But it also should be no surprise that the overall cost of payments indexed to prices will also rise in real terms, so long as the population is growing. If the size of the population is growing then the only way to avoid the cost of the programme rising faster than inflation is to either cut the proportion of the population receiving the payment or to cut average benefits in real terms.

Putting trends in context

In thinking about the future of the welfare system we need to look at the system as a whole and not just at isolated parts of it.
The need to look at the system as a whole is because we have a categorical system of income support payments. This means that in order to be eligible for a payment an individual needs to fall into a defined group – either being over the age of 65, having a disability or caring for someone with disability, being unemployed or sick, being a student, or caring for children, for example. We even have a payment – Special Benefit – for low income people who do not satisfy the eligibility criteria for any of the other categories.

At a specific period of time, however, the categories are mutually exclusive. For example, it is clearly possible that an individual could simultaneously be over the age of 65 and have a disability that means they could not engage in paid work; similarly an individual could be a lone parent and looking for full-time work, or be a lone parent caring for someone with disability. Nevertheless, you can only receive one of the categorical income support payments even if you are potentially eligible for more than one.
The categorical nature of our system means that it may be possible to “game” the system – to claim the specific payment that has the most favourable conditions. But it also means that when government policy changes and a payment is either abolished or phased out over time or eligibility condition are tightened, that individuals legitimately may be able to claim and receive one of the other payments. This also applied to groups of people at different times – following a policy reform, a class of people who previously might have been able to claim one type of payment might still be eligible for another payment.

In fact, some policy reforms explicitly move groups of people from one payment to another in a very short time frame. If we only analyse one payment at a time we overlook this potential substitution and can potentially get a very limited view of what is actually going on in the welfare system.

In addition, in my earlier assessment I pointed out that it is possible to take a longer-term perspective, using Department of Social Services statistical reports going back to 1991 and data collected by its predecessor departments since the 1960s. Does this picture change if we look only at the last ten or twenty years?

Because the Australian population has grown quite significantly over time and the age composition of the population has changed, the best way to track trends is to look at the number of people receiving payments either for all recipients as a percentage of the total population, or the percentage over 65 years receiving payments and separately as a percentage of the working-age population – people aged between sixteen and sixty-four.

In this context, Chart 1 shows trends in the percentage of people aged 65 years and over receiving income support of different forms between 1995 and 2014. The proportion receiving an Age Pension either from the Department of Social Services or the Department of Veteran’s Affairs has increased from around 64 per cent of the population in 1995 to close to 70 per cent in 2014. The proportion of this population receiving a DVA Service Pension, in combination with an Income Support Supplement, however, has fallen from around 19 per cent to 6 per cent of the population, while the proportion receiving other payments – mainly Disability Support Pension, Carer’s Payment or Special Benefit – has increased from 1.7 to 2.6 per cent. Overall, the proportion receiving one or other of these effectively income-related payments has fallen from around 85 to 78 per cent.

The total number of people receiving one of these payments has actually gone up by close to 700 thousand, but this is simply because the number of people over 65 has increased by 1.36 million

Chart 1: Trends in per cent of older population receiving income support payments, Australia, 1995–2014

Chart 1Source: Calculated from Department of Social Services, Income Support Customers, A Statistical Overview, various years; Department of Social Services, DSS Payment Demographic Data, June 2014 and ABS, Australian Demographic Statistics, 2014.

The decline in the share of the older population receiving a Service Pension plus Income Support Supplement largely reflects the movement of the larger age cohorts of people who had service in World War II into this payment between the 1970s and 1980s and their subsequent “exit” through death, and the lack of as significant an experience of war in subsequent age groups. But as discussed earlier, Age Pensions (either from DSS or DVA) are substitutes or alternatives to Service Pensions.
The fact that the increase in the share receiving Age Pensions was only about half the size of the decline in the share receiving Service Pensions suggests that potential new entrants are better-off than previous cohorts of people turning 65. Moreover, as the effects of the introduction of the Superannuation Guarantee in 1992 start to lead to an increase in older people’s resources in future years, it is likely that the share of older people receiving an income-tested payment will continue to decline, although the full effect will not be seen until after 2030 when people who retire will have had the opportunity to contribute for their full working lives.

Trends among people of working age differ significantly from those among people over the age of 65.

Chart 2 shows the proportion of people aged 16 to 64 years receiving income-support payments between 1995 and 2014. After a small jump from 1995, from 1996 onwards there was a long steady decline, with a more modest rise and fall since 2008. The chart shows two figures – one labelled “working age” and the other labelled “adjusted working age rate”.

The “adjusted working age rate” has been calculated to show the impact of one of the most important policy changes in social security in the last 20 years. In 1995 the Keating government started to change the eligibility age for women to receive age pension from 60 to 65 years. This was phased-in between 1995 and 2013, with the number of women aged 60 to 64 years receiving an Age Pension falling from 211 thousand in 1995 to zero in 2014.
The unadjusted rate is calculated by including women aged 60 to 64 in the working age population but counting this group of age pensioners in the pensioner population over 65, while the adjusted rate (correctly) involves counting this group in the working age recipient population. The adjusted figures also involves counting a smaller – but growing - group of people over 65 years of age who are actually receiving “working-age” payments as in the aged population.

It is clear from Chart 2 that the adjustments to more accurately divide the payment population by age means that in 1995-96 a higher proportion of the working-age population were receiving payments but that this percentage declined at a somewhat faster rate. Overall, in 1996 at its peak nearly 25 per cent of the working age population were receiving basic social security payments but the most recent figures for 2014 is that 16.8 per cent were, a decline of around eight percentage points, or close to one third.

While numbers on payments are available for 2015, ABS data on the age composition of the population are not, but it can be noted that the number of working age people on payments rose by around 1.9 per cent in the year to June 2015, but the total Australian population rose by 1.4 per cent in the year to March 2015. Given that the population under 16 years of age has been rising at a slower rate than any other group in the population, it seems unlikely that the rate of receipt of payments has changed to any large extent, although it is possible that it has gone up slightly.

Chart 2: Trends in per cent of working age population receiving income support payments, Australia, 1995–2014

Chart 2Note: Working age is defined as the population aged 16 to 64 years. Source: Calculated from Department of Social Services, Income Support Customers, A Statistical Overview, various years; Department of Social Services, DSS Payment Demographic Data, June 2014 and ABS, Australian Demographic Statistics, 2014.

What explains these fluctuations? The number of working-age people receiving welfare payments at any one time is strongly related to the state of the labour market. Not surprisingly, it increases significantly in periods of recession. Unemployment was generally below 2 per cent of the labour force up until the 1970s, but doubled to 4.6 per cent between 1974 and 1975, and then rose to over 6 per cent in 1978. The recession of the 1980s saw a peak unemployment rate of 9.9 per cent in 1983, followed by a decline through to 1989. In the midst of another recession, it increased to 11 per cent in 1993. Unemployment declined after 1993, falling to 4 per cent, the lowest level since 1974, in February 2008. So part of the explanation for the long-term decline is Australia’s very strong employment performance, particularly up to the time of the GFC.

But other factors are at work as well, including the dynamics of different categories of payments. Changes in the number of lone parents receiving benefits, for example, partly reflect shifts in family formation. People who are unemployed for lengthy periods and experience a disability may drop out of the labour market and end up on the DSP, for example, and unemployment can lead to family breakdown and growing lone parenthood.

Nor does the number of people receiving welfare payments necessarily fall as rapidly as the unemployment rate. The most commonly cited reason is that long-term unemployment leads to a deterioration in skill levels and morale, which reduces the intensity of an individual’s search for a job. This can lead to higher wage pressures at a given rate of unemployment, lifting the rate of unemployment towards which the economy tends. Between the 1970s and the 1990s, unemployment in Australia “ratcheted-up” after recessions but didn’t return to its pre-recession levels afterwards – and the people who gained jobs during those recoveries were not necessarily the people who had lost jobs during the recessions.

Policy changes are also a major cause of changes in the number of welfare recipients. In periods when benefits are more generous or easier to access, the number of recipients tends to grow, while periods of tighter eligibility or entitlement conditions have the opposite impact.

Reflecting these and other factors, the proportion of people receiving welfare payments peaked at nearly one in four of the working-age population in 1996, before falling to one in six in 2008, just before the global financial crisis. After the GFC, the (adjusted) proportion of working-age people receiving benefits rose to 17.5 per cent in 2010, then started to fall again, reaching 16.7 per cent in 2014 – not quite back to the 2008 level, but the second lowest level in the past two decades.

Chart 3 breaks down trends since 1995, showing what has happened to the share of the working-age population receiving the DSP, the share receiving unemployment-related payments, the share receiving Carer Payment, and the percentage receiving any other form of working-age income support, including parents, the sick, wives, widows and partners and recipients of student assistance.

Chart 3: Trends in per cent of working age population receiving selected income-support payments, Australia, 1995-2014

Chart 3

Note: Working age is defined as the population aged 16 to 64 years. These figures are adjusted to count women aged 60 to 64 years on Age Pensions as being included in the working age payment population and counting people 65 years and over and not on Age Pensions as being included in the retirement age payment population. Source: Calculated from Department of Social Services, Income Support Customers, A Statistical Overview, various years; Department of Social Services, DSS Payment Demographic Data, June 2014 and ABS, Australian Demographic Statistics, 2014

A fairly steady rise in the share of people on the DSP is apparent from 3.9 per cent in 1995 to a peak of 5.4 per cent of the working age population in 2011 to about 5.2 per cent in 2014. The number of people of working age on Disability Support Pension fell further from 790 thousand in 2014 to 778 thousand in 2015. It is of interest that the number of people 65 years and over receiving Disability Support Pension rose from around 4 thousand in 1995 to nearly 40 thousand in 2014, presumably because they are not residentially qualified for an Age Pension, but acquired a disability after they settled in Australia.

The long decline in the share of the unemployed in the working age population until the GFC is also apparent, with an increase in 2008–09 and a sharper increase between 2012 and 2013. The proportion of people on Carer Payment rose from a negligible level in 1995 (0.2 per cent of the working age population) to 1.4 per cent in 2014.

What is most striking, however, is the trend in the number receiving other payments; this peaked at 13.6 per cent of the population in 1996 but continued to fall to 5.1 per cent by 2014. In numerical terms, the number of people receiving other social security income support benefits has fallen from 1.6 million in 1996 to 781 thousand in 2014 (and further to 758 thousand in 2015).

While the improvement in labour market conditions between 1996 and 2008 is likely to have contributed to the decline in the share of working age people on these other payments, policy changes appear to be the most important factor behind these falls in rates of welfare receipt.

But in particular if we only look at the programmes where numbers have been going up – Disability Support Pension and Carer’s Payment and more recently unemployment payments then we will have a very partial view of overall trends and miss the contribution of policy changes in other parts of the system.

Policy change – the need to look at the system as a whole

Why have the percentage of the working age population receiving DSP and Carer Payment been rising since 1995 and the share on unemployment payments since 2008, while the share receiving other payments has declined significantly?

The state of the labour market is part of the answer, and so are two other factors that have had a major impact among people of working age: the ageing of the baby boom generation and the major social security policy reforms introduced by Australian governments over the past two decades.

Chart 4 shows the age profile of the main social security payments for people of working age in 2014 (the most recent year for which ABS data on the age structure of the population are available). The figure excludes student payments which primarily go to people under 25 and age pensions going to people over 65 – around 9 per cent of people under 25 receive student assistance and 72 per cent of people over 65 receive an age pension or another DSS payment.

Chart 4: % of age group receiving income support payments by payment type, 2014 Chart 4

Source: Calculated from Department of Social Services, Income Support Customers, A Statistical Overview, various years; Department of Social Services, DSS Payment Demographic Data, June 2014 and ABS, Australian Demographic Statistics, 2014

It is clearly apparent that the proportion of people receiving social security benefits increases with age – doubling from 10.5 per cent of those aged under 25 years to 21 per cent of those aged 55 to 64 years, although for those between these two age groups rates of receipt only range between 14 and 15 per cent.

The types of payments that people receive also vary with age. The partial exceptions are Newstart and payments for unemployed youth where around 5 per cent of each age group are receiving these payments. In contrast, the proportion of people receiving either Parenting Payment Single or Parenting Payment Partnered peaks at around 5 per cent of those aged 25 to 34 years and then declines.

For the other major payment categories, rates of receiving payments rise significantly with age - particularly for people receiving Disability Support Pension - from around 2 per cent of those under 35 years, to 4 per cent of those aged 35 to 44 years, 7 per cent of those aged 45 to 54 years and close to 12 per cent of those aged 55 to 64 years. The proportion of people receiving carer’s payment also rises from less than 1 per cent of those aged 25 to 34 to 2.6 per cent of those aged 55 to 64 years. People receiving Wives, Widows or Partner Payments are predominantly aged between 55 and 64 years.

So why do rates of receipt of these payments rise so strongly with age? The obvious answer is that your likelihood of acquiring a disability increases with age, due to lifestyle related diseases and natural wear and tear, particularly for those with manual jobs. The US Social Security Administration has estimated that more than 25 per cent of current 20 year olds will acquire a disability before the age of 65. It is also obvious that the number of people who need to care for relatives with disability will rise as the structure of the population shifts.

Because rates of receipt of payments rise with age, the number of people receiving social security is influenced not only by changes in the size of the population but also by changes in the demographic profile of the population, particularly by the share of the population in the age groups over 50 years of age

Chart 5 shows changes in the size of the population in different age groups in Australia between 1995 and 2014, separating out the changes since 2005 from those over the longer period.

Chart 5: Changes (%) in the size of the population by age group, Australia, 1995 to 2014

Chart 5Source: Calculated from ABS, Australian Demographic Statistics, 2014

It can be seen that between 1995 and 2014 the total Australian population increased by close to 31 per cent and the adult population over 16 years of age increased by just over 35 per cent. It is important to understand the implications of this: if all income support payments were only indexed to prices and the share of the population over 16 remained constant, and the percentage of the population receiving benefits had also remained constant, then we would expect that real social security spending would still have increased by 35 per cent since 1995.

It is also apparent that different age groups have grown at very different rates. The number of children in the population has increased by only 14 per cent since 1995 or less than half the rate of the increase in the total Australian population, while the proportion of the population between 16 and 49 years has grown by around 20 per cent. In contrast, the proportion of the population aged 65 years and over increased at more than twice the rate of the total population. Even more strikingly, the population aged between 50 and 64 years – those working-age groups where receipt of social security is highest - increased by nearly 72 per cent.

However, trends in the second half of this period differ in subtle but significant ways. The total Australian population increased at about the same rate as in the total period, but the share of children rose at a faster rate. The rate of increase in the population aged 50 to 64 years fell by close to 70 per cent, but the number of people aged 65 years and over grew marginally faster in the second period than it did over the whole period.

These trends reflect divergent demographic trends, primarily the ageing of the baby boom generation and to a lesser extent after 2000 the increase in total fertility rates. The “baby boom” generation were born between 1946 and 1964 and again by a simple matter of arithmetic started to turn 50 in 1996 and 65 in 2011. The movement of this generation through the population first started to increase the percentage of the population in the age groups likely to receive Disability Support Pension and Carer’s Payment, and more recently those in the age range potentially entitled to Age Pension.

Up until 1996, demographer Natalie Jackson has shown that changes in the age structure of the Australian population acted to slow the growth of the DSP. But people born in 1946 started to turn fifty in 1996, reversing that slowdown.

The effect of the ageing of the baby boom generation is illustrated in Chart 6, which shows the change in the number of persons by years of age between 1996 and 2014 (after taking account of deaths and migration). The largest population increase is among fifty to sixty seven year-olds, with each year group being around 100,000 larger than the comparable group in 1996. Cumulatively, there were 1.9 million more people aged fifty to sixty-four years – the age at which rates of receipt of the DSP start to rise significantly – in 2014 than in 1996.

Chart 6: Change in number of persons sixteen and over by age, Australia, 1996–2014

Chart 6Source: Calculated from ABS, Australian Demographic Statistics, June 2014.

Between 1996 and 2012 the proportion of people of working age receiving the DSP rose from 4.3 per cent to 5.6 per cent. If the age structure of the population were held constant at 1996 shares, then the figure would have been 5.0 per cent – in other words, nearly half of the total increase is unrelated to changes in the labour market, the incidence of disability or individual behaviour, but simply reflects changes in the age structure of the population.

It is also apparent that from 2015 onwards for a number of years, the increase in the size of each year cohort will decline significantly, and then increase again, before declining again. For example, while there were 93 thousand more 50 year olds in 2014 than in 1996, there were only 42 thousand more 49 year olds (and 15 thousand more 35 and 36 year olds).

These shifting patterns suggest that the number of potential inflows into DSP and other payments predominantly claimed by people over 50 will shift over time.

More importantly, a series of policy changes from the late 1980s and mid-1990s also had a major impact on the number of people receiving the DSP and other payments.

As noted above, one of the most important of these was the increase in the age pension qualifying age for women from sixty to sixty-five after 1995. Previously, women receiving the DSP were required to shift to the age pension once they turned sixty, and women who became disabled after turning sixty weren’t able to claim the DSP unless they had lived in Australia for less than the ten years needed to qualify for an age pension.

As the cut-off age started to increase, women with disabilities in this age group increasingly claimed the DSP. As Chart 7 shows, the proportion rose from close to zero to 13.3 per cent by 2013 (age breakdowns by gender are not available for subsequent years).

Chart 7: Change in percentage of women aged 60–64 years receiving DSP, Australia, 1995–2013

Chart 7

Source: Calculated from Department of Social Services, Income Support Customers, A Statistical Overview, various years and Australian Bureau of Statistics, Australian Demographic Statistics, June 2013.

But as the number of women receiving the DSP went up, the number receiving the age pension went down – and, as Chart 8 shows, it went down by much more.

In 1995, only about 650 women aged sixty to sixty-four received the DSP and 212,000 received the age pension. By 2013, 86,000 female DSP recipients were in that age group, but only 27,000 age pensioners (and since 2014 there have been none). The number of female carers has risen from 132 in 1995 to more than 24 thousand in 2013. Where once 67 per cent of women of that age received a pension or other payment, now the figure is 31 per cent. Overall, close to a quarter of the growth in the number of DSP recipients over the past 20 years can be accounted for by the growth in the number of women aged 60 to 64 receiving DSP rather than Age Pensions.

Chart 8: Percentage of women aged 60–64 years receiving DSP, Age Pension or other income support, Australia, 1995–2013

Chart 8Source: Calculated from Department of Social Services, Income Support Customers, A Statistical Overview, various years and Australian Bureau of Statistics, Australian Demographic Statistics, June 2013.

In future, these two major pressures on DSP numbers – the ageing of the baby boomers and the increase in women’s pension age – won’t operate to the same extent. Because the last of the baby boom generation turned fifty in 2013, the pressure on DSP numbers should start to lessen. The younger part of the baby boom generation will not turn sixty-five until 2028, but the rate of increase will slow because the size of the age group is not increasing as rapidly. For similar reasons, in the United States, both the Social Security Administration actuaries and the Congressional Budget Office project that spending on their Disability Insurance programme will fall as a share of GDP in the coming decade as baby boomers convert from DI benefits to retirement benefits and are replaced in the peak disability-receiving ages by smaller cohorts. If anything, the effect is likely to be even stronger in Australia as the increase in the pension age for women was also fully phased in by 2014, so this pressure will also decline.

It is worth noting, however, that the Rudd government announced an increase in the pension age for both men and women from sixty-five to sixty-seven years, to phase in between 2017 and 2023. It is likely that some of the people affected by this change will be entitled to the DSP, leading to an increase in numbers on the DSP after 2017, but past experience suggests that the overall number of people in this age group receiving payments will be reduced.

In addition, the 2010-11 Budget announced revised access procedures for some DSP claimants, to commence from 1 January 2012 and new participation requirements came in from 1 July 2012, affecting current and new DSP recipients under age 35 assessed as having a work capacity of 8 hours or more a week.

As shown in Chart 9, these changes appear to be associated with a very large slowdown in the rate of increase in the number of people being paid DSP, albeit after a relatively large spike in numbers following the GFC (after a lag).

Chart 9: Annual average change (%) in number of people receiving Disability Support Pension, 1995 to 2015

Chart 9

Perhaps as a result, the most recent figures for numbers on the DSP show a fall from around 827,000 in 2012 to 814,000 in 2015; excluding people over sixty-five who receive the DSP this is a fall to around 5.2 per cent of the working-age population, about the level it has been since 2002.

“Dependency” payments

Starting from the 1980s and continuing for nearly 20 years, the government began phasing out a number of other payments or limiting access to new claimants.
For example, access to Widow B Pension was limited in 1987, and then closed to new entrants in 1997. In 1994, the government introduced Partner Allowance to provide support to the partners of beneficiaries who had previously received a “married rate“ of payment, then in 1995 restricted this to older women without recent workforce experience, while introducing Parenting Payment Partnered for partners with dependent children. In addition to the phasing out of these payments, the income test for unemployment payments was changed in 1995 to require both individuals in a couple to claim benefit in their own right, and part of their individual earnings did not affect their partner’s benefit entitlements.

Wife Pension was closed to new entrants in 1995. Partner Allowance and Mature Age Allowance were also both closed to new claimants in 2003, and by 2008 there were no longer any recipients of Mature Age Allowance. Since 2005, new grants of Widow Allowance have been limited to women born on or before 1 July 1955.

Most of these payments had effectively been based on the assumption that women were “dependents” of men, or in the case of widows that they had been dependent and should not be expected to look for work. Even the lower Age Pension for women had been partly based on the assumption that women would want to leave the workforce at roughly the same time as their assumed older husbands.

It has been further argued by Bob Gregory that these changes affected the likelihood of partnered men to claim payments, as their behaviour may change if their partner cannot receive a payment.
These changes had a profound impact not only on the total number of people receiving welfare payments but also on which payments they received. In the mid-1990s, the “closed payments” – mainly for women – were received by around 4 per cent of the working-age population; now, only 1 per cent of the population receive their successor payments.

About 1.3 per cent of the working age population are now receiving Carer Payment. As with the age pension/DSP trade-off for older women, the rise in the number of people on the carer payment is more than offset by the decline in the number of people on these “dependency” payments.

What about the unemployed?

Two main factors have driven the growth in Newstart numbers. The number of people receiving unemployment benefits tracks broader labour-market trends fairly closely, and so the increase in the unemployment rate from around 5.0 per cent to closer to 6.0 per cent since June 2012 could be expected to result in more than 100 thousand additional people on benefits.

Changes to the DSP under the previous government that appear to have slowed its growth may have increased numbers on Newstart, while policies designed to shift people from parenting payments to the lower level of Newstart payments since 2006 and particularly since the beginning of 2013 have added to the total.

The then government forecast that around 10,000 people would lose all entitlement to benefits as a result of the 2013 decision, and that around 75,000 would transfer onto Newstart. (This effect is the same as the substitution between the age pension and the DSP and carer payment and other closed benefits, as described above.)

The monthly statistics on labour force payments show that between December 2012 and February 2013 the number of people on unemployment payments jumped from around 700,000 to 796,000, an increase around four times as great as the corresponding periods for the previous two years. Moreover, around 83 per cent of the increase in the number of recipients was made up of women, reinforcing the impression that this very large jump is mainly explained by parents transferring from parenting payments.

Table 1 shows trends in the number of lone parents receiving either Parenting Payment Single or Newstart between 2006 and 2014. From a negligible number of lone parents receiving Newstart in 2006, the number has grown to around 120 thousand in 2014. Data on the number of lone parents receiving other payments such as Carers Payment or DSP are not publicly available, and it is possible that a sizeable number of people who would otherwise have received Parenting Payment Single would now be receiving one or other of these payments, if they were eligible because they had a disability themselves or were caring for someone with disability. Nevertheless from the figures in Table 1 it can be seen that while the numbers on Newstart have grown significantly, the total number of lone parents receiving one or other of these payments has fallen.

Table 1: Number of lone parents receiving either Parenting Payment Single (PPS) or Newstart (NSA), 2006 to 2014

Table 1

Source: Data provided by ACOSS from answers to Parliamentary Questions on Notice

In Chart 3 it was shown that the share of the working age population receiving Newstart has risen since 2008, with there being about 300 thousand more people on Newstart in 2014 than just before the GFC. As Table 1 shows in 2014 there were about 100 thousand more lone parents on Newstart and 100 thousand fewer on PPS than in 2008, and as suggested earlier a rise in the general unemployment rate by 2 percentage points could have been expected to add around 200 thousand more people to Newstart.

While the situation is more complex than this due to other changes in the population including changes in the number of lone parent families – falling since around 2003 – this suggests that a very significant component in the rise in numbers on Newstart can be explained either by labour market change or deliberate policy change.

What has happened to spending?

Governments interested in attaining Budget balance are not unexpectedly most concerned with programme spending, even if the number of recipients is one of the main drivers of aggregate spending. In the Annual Budget, governments also tend to focus on nominal spending over the forward estimate period.

However, just as it is preferable analytically to focus on recipients of benefits as a percentage of the population, it is also preferable to look at spending as a percentage of GDP, as the best measure of the affordability of the system. This is because as discussed earlier real spending will grow simply as a result of population growth, even if all other parameters are held constant.

The most consistent and comprehensive time series on social spending as a percentage of GDP can be found in the OECD Social Expenditure database. Overall levels of public social spending can be analysed up to 2014, but details of spending disaggregated by programme are available only to 2012. In addition, social spending figures for Australia aggregate Commonwealth and State Government spending, although it can be noted that virtually all spending on cash benefits is undertaken at the Commonwealth government level.

Chart 10 shows trends in spending on Age and related pensions (mainly DVA payments) from 1995 to 2012. It is apparent that with the exception of the year 2000 spending on age and related pensions has remained between 3.0 and 3.4 per cent of GDP. The spike in spending in 2000 is the result of the compensation package for the introduction of the GST in that year. These pensions are indexed to wages, so the stability of spending as a per cent of GDP is likely to reflect the declining coverage of the Service Pension, as discussed earlier.

Chart 10: Spending on Age and related pensions, Australia, 1995 to 2012.

Chart 10

Source: OECD Social Expenditure database.

Chart 11: Spending on working age cash benefits, Australia, 1995 to 2012.

Chart 11

Source: OECD Social Expenditure database.

Chart 11 shows trends in spending on the other major components of cash transfers from 1995 to 2012. Overall, spending has fallen from 5.5 to 4.9 per cent of GDP over this period. There was a spike in spending on working age payments in 2008, at the time of the stimulus payments made during the GFC. Excluding “Other incapacity” (mainly state government worker’s compensation payments) overall spending fell from 4.7 to 4.3 per cent of GDP.

Spending on the unemployed – which includes spending on parenting payments in the OECD’s definition – fell from 1.2 to 0.6 per cent of GDP. The only component to rise as a percentage of GDP was on disability payments (which include Carer’s Payment) from 1.1 to 1.5 per cent of GDP.

Spending on family cash benefits rose substantially from 2.1 to 2.6 per cent of GDP between 1995 and 2003, but have since fallen back to 1.9 per cent of GDP (although with a temporary spending spike at the time of the GFC).

Overall the spending trends in the OECD data are consistent with the picture derived from considering trends in the number of recipients. Spending has gone up in some categories, notably DSP and Carers, but spending in other social security areas has fallen to more than offset these rises.

Where to from here?

To sum up, the data on trends in the number of welfare recipients show a prolonged fall since the mid-1990s, due to a long period of economic growth, a strong labour market, and the positive impacts of policy changes since the early 1990s. While trends have not been as positive since 2008, they have been mild by the standards of earlier economic downturns in Australia. Trends in spending as a per cent of GDP show broadly similar patterns, with no evidence of major increases after 2008.

This analysis also shows that past trends are not necessarily a reliable guide to the future. The two main pressures on DSP numbers – the ageing of the baby boomers and the increase in women’s pension age – are unlikely to continue to have such a significant impact, suggesting that numbers of recipients and levels of spending relative to GDP are unlikely to grow in the near future, absent some form of economic shock.

While concerns about relentless growth are difficult to substantiate – particularly when the total number of welfare recipients is close to its lowest level in the past twenty years – we should not be complacent.

The evidence shows that our main concern should be to avoid any significant blow-out in unemployment. Previous increases in unemployment in the recessions of the 1980s and 1990s had very long-term consequences, particularly for jobless families with children.

Nor should we be complacent because welfare numbers are at their second-lowest level in the past two decades. Successive governments have argued that the “best form of welfare is work” and it clear that the economic wellbeing of individuals and families is much greater in paid work than outside it.

The problems that some people on welfare face in moving into work require a comprehensive analysis, however. Not all these problems are caused by the welfare system: other barriers to work include labour-market programs that are not equally effective for all, a lack of job opportunities in the regions in which people live, poor public transport, inadequate and expensive child care, mismatched skills, and negative employer attitudes to people disadvantaged in the labour market. Changing incentives in the welfare system through reform of eligibility for specific payments is only part of an effective response to these challenges.

Related documents:
Filed under:

Updated:  24 March 2017/Responsible Officer:  Crawford Engagement/Page Contact:  CAP Web Team