Key issues

• The Australian welfare state is facing significant demographic, economic and political challenges, which are likely to require a reconsideration of the structure and funding of Australian welfare programs to ensure they are meeting their objectives.
• Growing expenditure on universal programs has seen calls for increased targeting of funding, but governments are also looking to expand access to support to assist families with cost pressures.

Introduction

The Australian welfare state is facing significant demographic, economic and political challenges. An ageing population will require more support from health and aged care services. The need for disability supports and the cost of services delivered through the National Disability Insurance Scheme (NDIS) has exceeded initial projections (pp. 152–153). The Intergenerational report 2023 found that structural changes in the economy will narrow the tax base used to fund these services (p. xv). Spending reductions on social security payments will not offset increases in other areas of social spending (p. 147). Many developed countries face similar pressures.

These challenges mean that key features of Australia’s welfare state could be reconsidered – particularly the way that services and supports are targeted. Growing expenditure on universal systems such as the NDIS has resulted in calls for increased targeting, particularly through means testing. However, increasing the extent of means testing in the Australian welfare state is likely to raise further complications. Cost of living issues could even see calls for an expansion in access to government support through universal programs.

Targeting welfare

While there are different academic attempts at classifying the design of welfare states, Australian academic Peter Whiteford has suggested there are 3 main approaches:

• contributory – support is distributed based on individuals’ contributions
• universal – benefits are a right of citizenship
• targeted/selective – benefits are provided based on need.

Most countries combine these approaches. For example, in Australia, income support is targeted to those in need through means testing, mandatory superannuation has been legislated to provide retirement incomes based on contributions, and the NDIS and Medicare are universally available to citizens and permanent residents. Within these broad approaches there may be a further blending of approaches: the NDIS may be considered ‘universal’ in that there is no means test, but it is targeted to specific categories of need through disability eligibility criteria. Whiteford writes that it is the balance between these 3 approaches that characterises welfare regimes.

Contributory schemes are often framed as a form of insurance against risks to an individual’s income such as disability, unemployment and old age. Targeted schemes are often framed as a means of poverty reduction or a safety net. Within the insurance frame, income tax and contributions may be considered as premiums. In the safety net frame, tax revenue is not considered in direct connection to the welfare programs it funds.

Australia relies much more heavily on means testing than other developed countries (OECD, p. 13; Whiteford). Most other OECD countries have c ontributory social insurance systems alongside targeted benefits for those who cannot access the contributory schemes. OECD data shows that some countries, such as the Nordic states, make greater use of universal benefits alongside contributory social insurance.

Targeting in the Australian welfare state

A long history of means testing

Means testing as a way of targeting welfare has been a feature of the Australia welfare system since the first social security payment, the Old-Age Pension, was created in 1908. This payment had eligibility conditions related to age, residency and character, but the decision to include a means test was contentious. Opponents to means testing argued that it was a ‘charity dole’ and stigmatised recipients as paupers, while proponents argued that it should be considered a retirement allowance earned through contributions to the community during the recipient’s working life. Cost was the main reason for not implementing a universal scheme (p. 7). At the time, the Constitution limited the federal government’s avenues for raising revenue (pp. 66–67).

Governments have subsequently proposed versions of contributory social insurance schemes or specific tax contributions to fund social services (pp. 8–13). The pension means test was eased in the period from 1948 to 1976, and was briefly abolished for those aged over 70. However, this trend has since reversed, with ongoing concern over growing expenditure on an ageing population (pp. 30–31).

Means testing was applied to other payment categories as the Australian welfare state expanded. Family payments were initially universal payments, but have gradually become subject to targeting. Now, all the main Australian Government payments are subject to means testing – even those intended to support workers, such as Child Care Subsidy and Parental Leave Pay.

Figure 1 sets out the array of payments and social services and whether they are targeted through means testing. Each program has a different means test containing multiple thresholds, taper rates and payment rates. The ways in which means are assessed can also vary, with different definitions of income used for social security and family payments.

Figure 1           Targeting Australian Government social spending through means tests

Source: Parliamentary Library

Narrowing the target

While the means test for the Age Pension is intended to target those without adequate means to support themselves, a majority of those who meet the age criteria (currently 67 years or older) receive the payment. In 2022–23, around 64% of the resident pension age population received the Age Pension and 71% received some form of income support (p. 168 and chart data for Chart 7.20). Around 44% of those above pension age receiving a government payment received the maximum payment rate. These percentages have remained fairly steady over the last 30 years. However, Treasury expects that over the next 40 years, as retirees with higher superannuation balances encounter the means test, the proportion of this age group receiving a pension will decline to around 57%, with only 21% of recipients receiving the maximum rate (Figure 2).

Figure 2           Persons of Age Pension eligible age by pension recipient category

Note: Includes receipt of any government income support pension including Age Pension, Disability Support Pension, Carer Payment and Service Pension.

Source: Australian Government, 2023 Intergenerational Report, (Canberra: The Treasury, 2023), 168 and linked chart data for Chart 7.20.

Tightened means testing has resulted in a reduction of the proportion of children being supported through family payments. What was once an almost universal payment – with around 80.5% of children aged under 16 attracting Family Tax Benefit Part A in 2003–04 – is now paid only to lower income families, with an estimated 44.7% of children under 16 being supported by the payment in 2021–22 (Figure 3).

Figure 3           Family Tax Benefit Part A children as a proportion of the population by age group

Note: Unique count of children attracting Family Tax Benefit Part A in the entitlement year, age at 30 June.

*Data for most years as at quarter 8 (2 years) after entitlement year to allow for reconciliation, however the year 2021–22 is as at quarter 6 and is not yet mature. ABS 3101.0 National, state and territory population, Table 59. Estimated resident population at 30 June each year.

Source: Economic Inclusion Advisory Committee, 2024 Report to Government, (Canberra: Department of Social Services, 2024), 93. Chart data provided by the Department of Social Services.

Recent decisions: funding the NDIS and aged care

There have been various proposals over the last 15 years to address the growing need for, and cost of, disability and aged care services.

The 2011 Productivity Commission report that guided the establishment of the NDIS stated that it would be ‘inequitable and inefficient’ to apply a means test – inefficient because it would exclude high-income people from the most efficient form of insurance against a severe disability, and inequitable because higher income people could not access a scheme they had helped fund (p. 262). The commission argued that ‘the well-off would pay a much larger contribution towards the NDIS than medium and low income earners’ under its preferred funding model: Australia’s progressive tax system (p. 262). The commission also argued that means testing would create disincentives for work and savings.

However, the commission argued that disability care accessed after Age Pension age should require a means-tested co-contribution, in the same way as aged care services (p. 17). In this depiction, the universal disability scheme provides an efficient public insurance model while the targeted scheme promotes private savings to cover the inevitable care costs in old age.

The NDIS co-contribution for older participants was never implemented, but those older than pension age cannot apply for NDIS supports.

Recent changes to the NDIS aimed at reducing expenditure growth have involved targeting the range and level of supports offered: by further specifying funded and excluded services, and changing the structure of participants’ budgets.

The 2021 Aged Care Royal Commission reconsidered the co-contribution model and recommended establishing an income tax levy to fund universal aged care (pp. 301, 309). However, the Morrison and Albanese governments decided against the proposed levy and universal model.

The 2024 Aged Care Taskforce report provided a rationale for not implementing a specific tax increase to fund aged care:

There are substantial intergenerational equity issues in asking the working age population, which is becoming proportionally smaller to pay for these services. Moreover, superannuation has been designed to support people to grow their wealth and fund the costs associated with retirement including aged care. (p. 20)

The taskforce supported increasing co-contributions through a means test to help fund growing aged care costs (p. 20), but did not consider alternatives, such as changes to tax concessions for superannuation, so that wealthy older people could indirectly contribute more to funding aged care.

Current debates – universal child care

In 2023, the Albanese Government tasked the Productivity Commission with analysing a shift to universal early childhood education and care ‘in the great tradition of universal Medicare and universal superannuation’ (p. v).

The commission’s final report presented a universal system as one where all children have access to high-quality services (p. 4). The proposed system retained a means-tested subsidy with higher levels of support to low-income families, removed activity-testing and contained measures to improve the supply of child care places (pp. 5–7). The commission found non-means tested options would involve greater increases in government expenditure, with most of the benefits and increased demand for child care going to higher income families (pp. 33–34). However, the non-means tested options would see a stronger (though small) increase in labour market activity from parents, compared to its preferred option (p. 37).

International responses to pressure on welfare systems

Contributory pension schemes in Europe, which rely on the contributions of current workers to fund the pensions of retirees, are under increased financial pressure due to smaller ratios of workers to retirees. Most OECD countries have increased pension ages and shifted occupational pension schemes from defined benefit to defined contribution models (pp. 42–45; 47).

The UK’s welfare system, which combines contributory, means-tested and universal elements, has seen the gradual decline of contributory benefits. Contributory payments to working-age people have declined from around a third of welfare benefits (including tax credits) in the late 1970s to around 8% in 2019–20. Benefits for retirees were following the same trend, but policy changes in the 2000s and 2010s have restored the basic state pension as a flat-rate, non-means tested payment for most retirees, which can be topped up with means-tested additional support or private savings.

Other countries are implementing universal programs. In 2021, the Canadian Government announced a plan for a universal child care system with average out-of-pocket costs at $10 per day and 250,000 new places by 2026. Some academics have questioned whether the goal can be achieved and whether there will be enough additional places to meet increased demand.

How social spending compares

Australia’s use of targeting has kept its public social spending, including on health, close to the OECD average – at 20.5% of GDP in 2019 (the most recent comparable data). However, when private social spending (such as superannuation contributions) and a variety of tax concessions are included – what the OECD describes as ‘net social spending’ – Australia is one of the biggest social spenders at 24.9% of GDP (ranked sixth) (Figure 4). These tax concessions often benefit higher income earners (because they have higher tax liabilities) and low-income workers often miss out on private social benefits.

Figure 4           Public and net total social expenditure, % of GDP 2019

Notes: ‘Gross public expenditure’ refers to total direct public spending on benefits and services. ‘Net total social expenditure’ includes social benefits delivered through the private sector such as superannuation, and tax expenditures which deliver social supports such as tax credits for dependent children. The numbers in brackets refer to country rankings in terms of gross public and net total social expenditure, with number one being the highest spender. For example, the US ranks 23rd in the OECD in term of gross public social expenditure and second in terms of net total social expenditure.

Source: OECD Social Expenditure database 2023

Future directions

Countries around the world are responding to economic, demographic and environmental challenges to their welfare states in a variety of ways – through greater targeting, through universalism, and through hybrid approaches.

In Australia’s case, some commentators and international bodies have argued that budget pressures create the need for more private spending and means testing of social supports (p. 31). While this can improve the government’s fiscal balance, it shifts costs onto individuals and families and will see some miss out on the support they need.

Expanding social spending and creating more universal supports would require more tax revenue or government debt. Universal systems provide broader coverage against risks to wellbeing and income, are simpler to navigate and administer, and remove the stigma attached to means-tested welfare. However, universal systems have finite resources – supports may be rationed (through waiting lists or eligibility criteria) and broader coverage can limit the level of support provided. Additional support to high-need groups is often necessary.

In meeting the challenges, policymakers will need to consider:

• What are the social objectives of welfare policy – an investment in the wellbeing of Australians; insurance against risks to income and wellbeing; reducing poverty or inequality; as a safety net of last resort; or a combination of these?
• How should resources be redistributed to fund welfare policies – from the rich to the poor; across generations; from workers to those unable to work?
• What are the most effective and efficient mechanisms for achieving the objectives and redistributing resources?

Further reading

• Don Arthur, Myths of Entitlement:  A History of the Age Pension and the National Welfare Fund, Research paper series, 2021–22, (Canberra: Parliamentary Library, 2021).
• Ben Spies-Butcher, Politics, Inequality and the Australian Welfare State after Liberalisation (London: Anthem Press, 2023).
• Dimitri Gugushvili and Tijs Laenen, ‘Two Decades after Korpi and Palme’s “Paradox of Redistribution”: What Have We Learned So Far and Where Do We Take it From Here?’, Journal of International and Comparative Social Policy 37, no. 2 (July 2021): 112–127.