A war on the middle class? Middle class welfare and the 2011–12 Budget

Adam Stebbing, Macquarie University

The Coalition argues that the cuts to middle class welfare that appeared in the 2011–12 Budget amounted to an attempt by Labor to ‘fan class welfare’ (Franklin 2011). In contrast, claiming it has delivered a ‘Labor’ Budget, the Gillard Government contended that its first Budget took tough decisions to increase support for those who need it most and curb the benefits received by the well off. But, did the 2011–12 Budget really do more than close a few loopholes that skirt at the surface of the burgeoning system of middle class welfare?

Middle class welfare is understood here as benefits received by those who are not poor (Goodin & Le Grand 1987). In Australia, middle class welfare mostly takes the form of tax expenditures. Rather than providing a cash benefit or access to a service, tax expenditures offer selective tax breaks to taxpayers who undertake particular actions or belong to certain social groups. Because those on higher incomes pay more tax and have greater means to purchase the private services or assets being subsidised, tax expenditures tend to benefit the well off while offering little to low-income earners. The cost of these tax measures continues to grow, with the Treasury estimating that about 350 tax expenditures reduced revenue by $113 billion or 8.8 per cent of Gross Domestic Product (GDP) in 2009–10 (Department of Treasury 2010, p. vii).

Cutbacks in the Budget represent modest first steps to reforming middle class welfare.

Yet, the government’s budget savings are rarely considered with reference to this large and expanding system of tax expenditures. When viewed in this broader context, these spending cuts are much more moderate than either Labor or the Coalition would have us believe. Still, the Budget’s cutbacks to tax expenditures do represent modest first steps to reforming middle class welfare.

There was, however, potential for the government to undertake broader reform of tax expenditures in this year’s Budget and, in doing so, enhance both the equity and efficiency of public spending. The recent Henry Tax Review provides a useful starting point—but by no means the end point—of what shape more ambitious reform might take.

Savings from Middle Class Welfare in the 2011–12 Budget

For better or for worse, the Gillard Government has tied its economic credentials to the delivery of a Budget surplus in 2012–13. The government declares that it is on track to meet this goal and that this year’s Budget demonstrates the necessary fiscal discipline. It supports this claim with $22 billion of savings over the forward estimates, which cover the next four years until 2014–15. Of these savings, $4.5 billion or 25 per cent are to be achieved by reform to four programs that benefit the middle class (see Table 1). The Budget also introduces a temporary flood levy that aims to raise $1.7 billion in revenue from the middle class.

The government’s proposal to trim $2 billion from family payments over the next four years is the largest and most controversial of these savings. Rather than directly cutting the income threshold to which family payments are paid, the government decided to keep the threshold at its current level of $150,000 but not raise it in line with inflation until 2014–15 (Daniels 2011, p. 160). This decision will affect a range of family-related payments, including the Family Tax Benefits and the new Paid Parental Leave Scheme. Most of these payments have income tests that do not ‘taper’—that is, the full rate of the benefit is paid up to the means-test threshold, and households with incomes over that threshold receive no payment (Daniels 2011, p. 160). This means that those households that currently have incomes below but close to $150,000 will lose family payments even if their income exceeds this amount because of wage increases that merely match the rate of inflation.

Table 1. Major Savings from Middle Class Welfare in the 2011–12 Budget
  2011–12 2012–13 2013–14 2014–15 Total
Program $m $m $m $m $m
Family Payments 76.7 410.5 743.2 774.8 2,005.2
Fringe Benefits Tax 26.4 135.4 331.2 455.9 948.9
Dependent Spouse Tax Offset 60.0 220.0 230.0 245.0 755.0
Low Income Tax Offset on minors’
unearned income
0.0 240.0 250.0 250.0 740.0
Total 163.1 1,005.9 1,554.4 1,725.7 4,454.0

Source: Calculated from Commonwealth Government of Australia (2011a, p. 45).

The government also proposes to target or remove three tax expenditures that benefit the middle class. It estimates that re-adjusting the formula for the Fringe Benefits Tax on company cars will save $949 million over the next four years. The current formulas for this tax offset that increase with the distance travelled will be replaced with a flat rate valuation of the car (20 per cent annually) to remove the disincentive to overuse company cars. The government also estimates that removing the Dependent Spouse Tax Offset and the Low Income Tax Offset for the investment income of minors will save $755 million and $740 million, respectively.

And, the government plans to further wind back middle class welfare by means testing the Private Health Insurance Rebate (which subsidises private health insurance) and raising the Medicare Levy Surcharge (an additional tax paid by high income earners who do not have private health cover). Although these changes were announced in the 2009–10 Budget, two previous attempts to pass them have failed (Zappone 2011). The government plans to reintroduce the legislation at the end of 2011 (Zappone 2011). If the government had been able to pass this legislation for the reform in 2009 or 2010, as initially planned, it estimates that spending on middle class welfare would have been reduced by a further $1.9 billion by 2012–13 (Commonwealth Government of Australia 2009, p. 31). While it will not save the original amount, it will still reduce spending on middle class welfare (if successful).

Tax expenditures tend to be inequitable and are often regressive.

The major savings from middle class welfare announced in the Budget are projected to have little impact initially, but this is set to grow over the forward estimates. As Table 1 shows, these four measures are estimated to reduce government spending by $163 million in 2011–12 and cumulatively increase to $1.7 billion by 2014. The delayed impact of these reforms is likely to have boosted the political appeal of these measures especially in the context of expressed public concerns about the high cost of living and perceptions that the government needs to rein in spending.

Despite their modesty, the savings measures do give some credibility to the Treasurer’s claim to have delivered a ‘Labor’ Budget (Swan 2011, p. 1). The savings measures from middle class welfare share a similar emphasis to the Hawke and Keating Governments on cutting the benefits received by the middle class and increasing the benefits received by lower income earners (Daniels 2011, p. 160). These measures also continue the policy direction of the Rudd Government in reducing support for the middle class, which underpinned the proposal to means test the Private Health Insurance Rebate. And, more broadly, the Budget shares with those of earlier Labor governments an emphasis on job creation and targeting support.

Nonetheless, the modesty of the 2011–12 Budget measures confirms the government’s hesitation to take politically unpopular moves. The absence of a more ambitious reform agenda in the Budget invites unfavourable comparison with previous Labor governments. With more sweeping reform to the system of tax expenditures that has grown unchecked in recent decades, the government could have bolstered both its commitments to both fiscal sustainability and social democratic values such as equity and fairness.


Estimated to lower spending on middle class welfare by $163 million in 2011–12, the Budget measures do little to offset the $117 billion that Treasury (2011, p. 4) projects will be allocated through tax expenditures next financial year. Most of these benefits will flow to the middle class, as tax expenditures typically assist those with means. Moreover, developments relating to tax expenditures raise questions about the sustainability of public policy not only because of their size, but because of their rapid growth in recent decades (Department of Treasury 2011, p. 4). The virtual absence of tax expenditures from most debates about middle class welfare is thus both staggering and of real concern.

The super tax concessions cost at least ten times more than they are projected to save.

But, as public policies that cut taxes, do tax expenditures really amount to public spending? The answer is a resounding ‘yes’. The neglect of tax expenditures does not stem from controversy over their definition, but from their low profile and lack of regular formal oversight (Spies-Butcher & Stebbing 2009, p. 9). Tax expenditures are public spending programs because they confer resources selectively to certain taxpayers. The most common kinds of tax expenditures include: tax rebates or offsets, tax concessions, tax exemptions and tax deductions. The tax revenue that the state does not collect through tax expenditures is functionally equivalent to the revenue spent on services and income transfers—both types of programs allocate resources to their recipients and must be funded by increasing revenue, cutting other forms of spending or running deficits (Surrey & McDaniel 1985). Nonetheless, tax expenditures are not identical to direct spending, being measured as revenue forgone—the amount by which they reduce tax revenue.

Figure 1. Tax Expenditures as a Proportion of GDP and Public Spending

Source: Treasury (1986–2011) Tax Expenditure Statement; Australian Bureau of Statistics
(1986-2001) National Accounts.

Tax expenditures tend to have inequitable and sometimes regressive benefits compared to other forms of public spending. This is because common design structures—such as flat rate rebates, concessional tax rates and tax exemptions—benefit those with greater purchasing power and may invert the progressive income tax scales (Surry & McDaniel 1985, p. 103; Stebbing & Spies-Butcher 2009). The concessional rate of tax for superannuation contributions is one case in point. This concession sets the maximum rate payable on super contributions at 15 per cent, meaning that the rate of the tax discount grows with the amount of income earned. So, while those earning half average weekly earnings receive a discount of 1.5 per cent on average, those on the highest incomes above $180,000 get an average discount of 31.5 per cent (Henry 2010a, p. 35).

Tax expenditures have grown unchecked over recent decades to become highly significant in both number and scale. The number of tax expenditures listed in the Tax Expenditures Statement (TES) has more than doubled from 170 in 1984–85 to around 350 in 2009–10. The revenue forgone for tax expenditures grew in nominal terms from $7 to $113 billion between 1984–85 and 2009–10. This amounts to an increase from 3.6 to 8.8 per cent of GDP over the same period (see Figure 1). These policies have also expanded as a proportion of public spending, but their recent growth was slowed—and overshadowed—by stimulus spending in the wake of the Global Financial Crisis. Nonetheless, tax expenditures still comprise more than double the proportion of public spending that they accounted for in the 1980s, climbing from 10.3 per cent in 1984–85 to their current level of 25 per cent (see Figure 1). While the figures from the TES should be read with some caution because its coverage is incomplete and tax expenditures are therefore underestimated, there can be little doubt that such expenditures have substantially grown in recent decades, a trend that has largely escaped attention.

The housing tax exemptions and superannuation tax concessions are currently the largest categories of tax expenditures, estimated at $40 and $31 billion of revenue forgone in 2011–12 respectively (Department of Treasury 2011, p. 4). These policies are also the largest middle class welfare programs, with their benefits going to those who can afford private housing and/or earn higher incomes. The vast scale of these polices is evident in Table 2, which displays the four largest tax expenditures and their projected cost. As the table shows, these four policies alone are estimated to cost about $292 billion of revenue forgone from 2010–11 to 2013–14, a period roughly equivalent to the Budget’s forward estimates. Moreover, the table highlights that these tax expenditures are projected to grow by over $12 billion by 2014, which is nearly three times the Budget savings from middle class welfare (at $4.5 billion) over the same period. This highlights the modesty of these proposals and calls into question whether family payments (that amount to $18 billion per year) are the spending program most need of attention.

Table 2. Estimates of Revenue Forgone for Major Tax Expenditures, 2010–15
  2010–11 2011–12 2012–13 2013–14 Total
Program $m $m $m $m $m
Concessional tax of employer
super contributions
14,300 15,800 16,300 17,900 64,300
Concessional tax of super
entity earnings
12,200 13,600 15,200 17,900 58,900
Capital gains tax main
residence exemption
17,500 19,000 19,000 19,000 74,500
Capital gains tax main residence
exemption – discount
22,500 23,500 24,000 24,000 94,000
Total 66,500 71,900 74,500 78,800 291,700

Source: Calculated from Treasury (2011, p. 120, 154–155).


How to address the middle class welfare provided via tax expenditures? Billed as a ‘root and branch’ appraisal of the Australian tax system, the Henry Tax Review made a strong case for reform of the tax expenditures for superannuation and housing. It also put forward proposals for what these reforms might look like and these make useful starting points.

There are strong grounds to reform the superannuation tax concessions that were introduced as part of the 1915 federal income tax and have grown unchecked since the Hawke and Keating Labor Governments extended private super to the workforce. Besides being highly inequitable, the super tax concessions are also inefficient because they now have a similar cost to the age pension but it seems highly unlikely that these concessions will significantly offset public spending on the pension in the future. The age pension currently costs about $32 billion, while the super tax concessions cost around $28 billion (Commonwealth Government of Australia 2011b; Department of Treasury 2011).

Tax reform would curb the inflationary effects of the subsidies for investment housing.

The Harmer Pension Review estimated that private super would only reduce spending on the age pension by 6 per cent in 2050, which is about 0.2 per cent of GDP (Harmer 2009, pp. 9–10; Spies-Butcher and Stebbing 2010). At $28 billion, the super tax concessions amount to around 2 per cent of GDP. The super tax concessions thus cost at least ten times more than what they are projected to save (in relation to the size of the economy) when the population ages.

With the Superannuation Guarantee Scheme mandating that almost all workers direct 9 per cent of wages to private super, the tax arrangements for super contributions are of particular interest. Currently, individuals earning up to a little over the minimum full time wage ($35,000 per year.) receive a 15 per cent discount on their super contributions whereas those earning higher incomes ($180,000 p.a.) receive a 30 per cent tax discount (see Table 3). Moreover, as men still receive higher incomes and have higher super balances than women on average, these tax arrangements exacerbate gender inequality (Sharp & Austen 2006).

The Henry Tax Review recommended that the 15 per cent concessional tax rate be replaced by a 20 per cent flat rebate. Instead of providing more to high-income earners, this rebate would deliver a 20 per cent tax discount for super contributions made by all tax payers. As Table 3 shows, this proposal is more equitable—for while higher-income earners still receive a larger dollar benefit, this benefit is proportionate to income received. The review was also adamant that the Superannuation Guarantee remain at 9 per cent (and not raised to 12 per cent) to avoid disadvantaging low-income earners by reducing their take home pay (Henry 2010b, p. 109).

Table 3. Proposals for Reforming the Tax Concession for Super Contributions
  Current arrangement Henry Tax Review CPD proposal
Income Marginal
tax rate
($) (%) (%) (%) ($) (%) ($)
35,000 15 15* 20 630 20 630
60,000 30 15 20 1,080 20 1,080
100,000 38 18 20 1,800 0 0
250,000 45 30 20 4,500 0 0
500,000 45 30 20 9,000 0 0

* The Rudd Government introduced a 15 per cent rebate for super contributions
for those earning up to $37,000 in the 2010–11 Budget.

Source: Spies-Butcher & Stebbing (2010, p. 272).

This review’s recommendation is more equitable than the current tax arrangements, but its potential to increase revenue is less clear. Elsewhere, my colleague Ben Spies-Butcher and I have proposed that the 20 per cent rebate apply to only those earning up to $80,000 per year, with a taper that cuts the rebate’s rate by one per cent for every additional thousand dollars of income (see Table 3; Spies-Butcher & Stebbing 2009). This is more likely to significantly reduce the cost of the super tax concessions.

Both the Henry Review and CPD proposals would be likely to reduce the cost of the super tax concessions, but it is difficult to gauge by how much, because these policies avoid many budgetary oversights for direct spending programs and long-term projections of public spending such as the Intergenerational Reports. What is clear, however, is that both these proposals are much cheaper than increasing the Superannuation Guarantee to 12 per cent of income (which the government proposes to do against the advice it received). Such a move will substantially increase super contributions and thus spending on the tax concession that applies to them—and potentially by billions of dollars per year.

There are also strong grounds for reforming the tax expenditures and other subsidies for housing. Along with $5 billion spent on negative gearing (which exempts the interest paid on mortgages when it exceeds the rental income earned from the property), the $40 billion of housing tax expenditures only benefits those who are able to afford private housing (Spies-Butcher & Stebbing 2010, p. 274). In addition to its inequity, the tax exemption for capital gains on the principal residence means that when current homeowners sell their homes they do not pay any tax on profit they have made from their investment, which they are required to do for other investments. This contributes to housing stress because it increases the incentives to retain large and expensive housing assets (as the potential windfall typically increases with the amount of time the assets are held) and puts further pressure on inadequate supply. As well as adding to inflationary pressures, this may also contribute to the volatility of the housing market (Productivity Commission 2004).

The Budget was a missed opportunity to rein in wasteful and inequitable spending.

While acknowledging inadequate supply is the main factor behind the shortage of affordable housing, the Henry Tax Review (2010, p. 414) suggests tax reform to curb the inflationary effects of the subsidies for investment housing. Under current arrangements, the owners of rental property receive a 50 per cent tax discount on capital gains (provided it is held for more than one year) and access to negative gearing. The review advises halving the tax subsidy from negative gearing (to 50 per cent) so that the rate is the same as the capital gains tax (Henry 2010b, pp. 417–418). Although still providing an incentive to purchase rental property, such a measure would reduce the bias in favour of rental properties while increasing revenue.

However, the review refrains from supporting tax reform of owner-occupier housing because it also acts as a ‘lifetime savings vehicle that provides security in retirement’ (Henry 2010b, p. 416). Instead, it recommends that the primary residence be included in the means test for the age pension to reduce the current incentives to occupy larger and more expensive homes than practical in retirement. Considering that the government has explicitly ruled out including the primary residence in the pension’s means test, it could alternatively better target the capital gains tax exemption for the family home on properties sold for more than some benchmark price. Such a move would mean that capital gains for housing would still receive a 50 per cent tax discount so long as it was held for at least a year. Clearly then there was considerable scope for the government to improve the equity of the tax concessions for super and housing in this year’s Budget, while simultaneously reducing inefficient spending.


Reforms to the tax expenditures for super and housing are not only more urgent than the suspension of the indexation arrangements for family payments, but they would have had a greater bearing on the Budget’s long-term fiscal sustainability and enhanced the equity of public policy. Had the government taken up these or similar proposals, claims that the Budget took tough decisions or amounted to a declaration of class warfare may have been more convincing. While the limited nature of the savings from middle class welfare is likely to reflect the need for consensus among a range of parties in conditions of minority government, it is hard not to view the Budget as a missed opportunity to significantly rein in wasteful and inequitable public spending.


Australian Bureau of Statistics 1986–2002, National Accounts, Australian Bureau of Statistics, Canberra.

Commonwealth Government of Australia 2009, Budget Overview 2009–10, Commonwealth of Australia, Canberra [Online], Available: http://cache.treasury.gov.au/budget/2009-10/content/overview/download/Budget_Overview.pdf [2011, Jun 3].

Commonwealth Government of Australia 2011a, Budget Overview 2011–12, Commonwealth of Australia, Canberra [Online], Available: http://cache.treasury.gov.au/budget/2011-12/content/download/Overview.pdf [2011, Jun 4].

Commonwealth Government of Australia 2011b, Budget Strategy and Outlook 2011–12, Commonwealth of Australia, Canberra [Online], Available: http://www.budget.gov.au/2011-12/content/bp1/html/bp1_bst6-02.htm [2011, Jun 10].

Daniels, D. 2011, ‘Family assistance retargeted’, in Budget Review 2011–12, Research Paper No. 13, Department of Parliamentary Services, Commonwealth of Australia, Canberra [Online], Available: http://www.aph.gov.au/library/pubs/RP/2010-11/11rp13.pdf [2011, Jun 5].

Department of Treasury 1986–2002, Tax Expenditures Statement, Department of Treasury, Canberra.

Department of Treasury 2011, Tax Expenditures Statement 2010, Department of Treasury, Canberra [Online], Available: http://www.treasury.gov.au/documents/1950/PDF/2010_TES_consolidated.pdf [2011, Jun 5].

Franklin, M. 2011, ‘Julia Gillard defends war on middle-class welfare in budget’ The Australian, 12 May.

Goodin, R.E. & Le Grand, J. (eds) 1987, Not Only the Poor: The Middle Classes and the Welfare State, Allen & Unwin, Sydney.

Harmer, J. 2009, Pension Review Report, Department of Families, Housing, Community Services and Indigenous Affairs, Canberra [Online], Available: http://www.fahcsia.gov.au/about/publicationsarticles/corp/BudgetPAES/budget09_10/pension.pdf [2011, Jun 8].

Henry, K. 2010a, Australia’s Future Tax System: Report to the Treasurer, Part One: Overview, Department of Treasury, Canberra [Online], Available: http://taxreview.treasury.gov.au/content/downloads/final_report_part_1/00_AFTS.pdf [2011, Jun 6].

Henry, K. 2010b, Australia’s Future Tax System: Report to the Treasurer, Part Two: Detailed Analysis, Department of Treasury, Canberra [Online], Available :http://taxreview.treasury.gov.au/content/downloads/final_report_part_2/AFTS.pdf [2011, Jun 6].

Productivity Commission 2004, First Home Ownership, Report No. 41, Productivity Commission, Melbourne.

Sharp, R. & Austen, S. 2007, ‘The 2006 federal budget: A gender analysis of the superannuation taxation concessions’ Australian Journal of Labour Economics, vol. 10, no. 2, pp. 61–77.

Spies-Butcher, B. & Stebbing, A. 2009, Reforming Australia’s hidden welfare state: Tax expenditures as welfare for the rich, Occasional Paper No. 6, Centre for Policy Development [Online], Available: http://cpd.org.au/wp-content/uploads/2009/02/CPD_OP6_Reforming_Australias_Hidden_Welfare_State.pdf [2011, Jun 5].

Spies-Butcher, B. & Stebbing, A. 2010, ‘Getting value for public money’, in More Than Luck: Ideas Australia Needs Now, eds M. Davis & M. Lyons, Centre for Policy Development: Sydney, pp. 267–279.

Surrey, S. & McDaniel, P.R. 1985, Tax Expenditures, Harvard University Press, Cambridge.

Swan, W. 2011, Budget Speech 2011–12, Commonwealth of Australia [Online], Available: http://cache.treasury.gov.au/budget/2011-12/content/download.pdf [2011, May 11].

Zappone, C. 2011, ‘Roxon ploughs on with health insurance means test’, The Sydney Morning Herald, 13 May.

Adam Stebbing is an Associate Lecturer in the Department of Sociology at Macquarie University. His research broadly focuses on the links between public policy and social inequality in contemporary democratic societies.

View other articles by Adam Stebbing