Andrew Podger. Australia’s ‘welfare system’: Family assistance and tax elements.Jun 19, 2015
While it is important to consider our tax and transfer arrangements as a single integrated system, there are various (overlapping) parts to it: retirement incomes (including superannuation tax arrangements and the age pension), the core welfare system (pensions and benefits for people not able or not expected to work, including the aged, disabled and sole parents with young children, as well as the aged) and family assistance (family payments, childcare subsidies and related personal income tax arrangements). This article examines family assistance and the personal income tax system; an earlier article addressed retirement incomes and a separate article focuses on inequality and Australia’s welfare system.
Family assistance serves two core objectives:
- horizontal equity, to ensure capacity to pay tax (and means for the purposes of social security means tests) takes due account of individuals’ family responsibilities and the costs involved; and
- Vertical equity, to support most those families with limited resources to meet the costs of children and their care.
These objectives are supplemented by other important principles including:
- Efficiency in its widest sense including minimizing any negative impact on incentives to work and save;
- Administrative efficiency, including the costs of compliance for those affected by the system as well as the costs of government administration;
- Simplicity and certainty, including being easy to understand and presenting no ambiguity over entitlements and obligations; and
- Social cohesion, promoting mutual obligations in society which make for harmonious relations and a shared sense of democracy and social values.
And of course the tax system must raise the revenue the government needs.
The unique nature of the Australian approach to family assistance, like the rest of the ‘welfare system’, is its emphasis on vertical equity through means tests. We have one of the most targeted systems in the world – around two-thirds of family assistance is received by the poorest 40% of the overall population (ABS, 2013). . This has its strengths: in particular, support for most lower income families has increased substantially over the last 30 years, with the real value of Family Tax Benefit Part A per child more than doubling from around $40 per week in 1975 to around $90 per week in 2010 (in 2010 dollars) (Whiteford, Redmond and Adamson, 2011) . But it also leaves horizontal equity problems, efficiency costs due to high effective marginal tax rates and administrative costs and complexities. Yet most political commentators keep calling for even more targeting to reduce ‘middle class welfare’.
Looking at family assistance on its own inevitably draws attention to its vertical distribution. Looking at it, as we should, as part of a wider tax and transfer system and in particular as part of the way we treat families in the personal income tax system, leads to consideration also of horizontal equity and broader notions of capacity to pay tax, as well as of the principles of efficiency and simplicity. Debate will continue about the balance between these issues but at least this way they all get put onto the table.
Indeed, if we want an approach that preserves vertical equity but gives more emphasis to efficiency and giving mothers in particular more choice including to return to the workforce and to pursue their careers, we need to reduce the emphasis on means tests and increase the emphasis on horizontal equity. There is also a case for reconsidering the current emphasis on child care to give more weight to assistance that allows more choice about the form of care families want for young children.
The personal income tax scale also delivers vertical equity (though not as much as transfers). This is primarily through the impact of the tax threshold rather than the higher marginal tax rates applying further up the scale. Some commentators have been questioning the need for such a high threshold but it is an essential component of a coherent tax and transfers system; if anything, it should be increased further. Not only is it the primary means of achieving progressivity in the tax system but, if it is set above the level of pensions and benefits, it ensures that few people fully reliant on welfare, nor anyone else on the same low income, would need to pay personal income tax, . For too long tax relief for low income people has been achieved artificially by complex ‘pensioner tax offsets’ and ‘low income tax offsets’ rather than having a simple and transparent tax threshold. Wayne Swan took some action towards fixing this but there is still further to go as the Henry Report recommended. The costs can be met by applying a standard tax rate of around 35% above the threshold rather than the lower rates now applying initially (again as recommended by the Henry Report and, indeed, as actually applied back under Malcolm Fraser’s 1978 reforms).
With progressivity in the personal income tax scale achieved more clearly and simply in this way, the next issue is how best to recognize different capacities to pay tax given family circumstances ie horizontal equity. Around the world, tax systems grapple with this issue. Some, like the Australian system, are based primarily on individuals and then make adjustments for family circumstances; others tax family units and then make adjustments for the impact on those without families or living alone. Such adjustments are not normally considered separately or as ‘welfare’ but as part of the very structure of the tax system, optimising revenue collection by ensuring tax reflects accurately capacity to pay.
Looked at this way, the tension many see between horizontal and vertical equity in family assistance arrangements is greatly exaggerated – horizontal equity should be achieved as part of the very design of the personal income tax system. Yet there is still some tension because the system as a whole is expected to give particular support to low income families and hence must balance that against all the other elements. But addressing the tension simply by means testing all family payments as the Henry Report suggests is unlikely to optimize the system: a better approach is likely to involve some modest universal assistance for tax equity, adjusting the tax scale (eg by applying the top rate from a lower level) and more tightly means-tested supplementary payments to achieve vertical equity (and to limit the numbers affected by high effective marginal tax rates).
Family responsibilities affect capacity to pay (and need) in two ways – via the direct costs of children and via the costs of their care. The first increases with the number and age of the children, the second is inversely related to the age of the youngest child.
The Henry Report rightly suggested varying Family Tax Benefit A, which addresses the first aspect, according to the age as well as the number of dependent children. That could be achieved with a universal payment complemented by a tightly means-tested supplement.
Currently, the second aspect – the costs of caring for young children – is addressed in a most complex mix of payments and subsidies which fail to address almost any of the principles of a well-designed system. These include Paid Parental Leave (set at the minimum wage, available to birth mothers for 18 weeks), Family Tax Benefit B (means-tested and, aside from sole parents, essentially limited to mothers not in paid work with a child under 18, and with a higher payment if there is a child under 5), parenting payments (means tested, with two levels of payment, the higher one designed for sole parents with children under 8 and the lower one for the carer in two parent families with a child under 6) and childcare subsidies (some universal, most means-tested, subject to various rules on the nature, amount and price of care and the extent of workforce participation by the parent).
The complexity reflects in part that the care of very young children involves additional objectives including the development of children from an early age (evidence suggests investing in development programs for pre-school children aged 3 and 4 yields significant benefits later) and the return to paid work of mothers as children begin to grow in order to sustain and build on their human capital and capacity to contribute; the latter is particularly important for disadvantaged parents in order to provide a path towards sustained independence and self-reliance.
These considerations do not however justify the current complex and contradictory measures. There is a strong case for a more uniform approach to payments based on the age of the youngest child, with highest payments in respect of the care of babies and reasonably generous payments also where the youngest is under school age; some payment where the youngest is at primary school could also be justified given the after-school care costs involved. (It might also help coherence of the system if we returned to a separate scheme for sole parents – not lumbered into family payments – that recognized their limited capacity to work when they have young children and provided a safety net consistent with that for other core welfare groups and with suitable work tests.)
There is a strong case for reducing the emphasis on means tests given the high effective marginal tax rates they impose on so many parents. A trade-off is involved between very high marginal tax rates for a small minority of families or rates around the maximum in the personal income tax scale applying to a much higher proportion of families. My preference is the former approach with modest universal payments to ensure horizontal equity.
Despite the current push to focus more on childcare, there is also a case for giving more emphasis to cash payments that offer more choice, and to supplement these with more targeted approaches to childcare subsidies and a general education/development program for pre-schoolers managed through schools and childcare centres. Much more effort is also needed to ensure employers offer flexible work options suited to help mothers transition back into paid work, and to offer continuing education and training options also.
Very young children require the full-time care by an adult, and there is no particular economic advantage in that adult being paid rather than unpaid. Most mothers choose to drop out of the paid workforce when their youngest is under two, and 90% report they do so not because of lack of childcare but because they want to look after the child themselves. Facilitating these choices while recognizing the costs of caring for young children (whether through direct child care costs or the opportunity costs of stay-at-home mothers) would be better and more simply achieved by generous universal payments where the youngest is under two (for example by extending the PPL to 26 weeks and having a further slightly lower universal payment while the youngest is under two) and a moderate universal payment where the youngest is over two but under five or six. This could replace much of our massive system of child care subsidies. The payments could be complemented by means tested supplements. (Consideration could still be given to targeted child care for mothers needing help to become welfare-independent.
Current arrangements are not just overly complicated: they are inconsistent and incoherent, overly means tested and having insufficient regard for efficiency or the way families and mothers in particular today want to exercise choice in how they care for their children and pursue their careers. The Government’s latest budget proposals would only increase the system’s complexity and its inconsistencies. Its proposals from the last budget, still on the table, would tighten means tests also making the system less coherent. At least its earnings-related PPL is no longer on the agenda, but the series of confusing proposals over the last year or so demonstrate a lack of careful policy analysis. But Labor too could give family assistance closer attention and develop a coherent approach that is fair and better promotes choice and efficiency.
Andrew Podger is Professor of Public Policy at ANU and former Head of several Commonwealth departments.