Australia's Future Tax System

Final Report: Detailed Analysis

Chapter F: The transfer system

F1. Income support payments

F1–1 Striking the balance between payment adequacy and incentives to work

The primary focus of the income support system is poverty alleviation through the provision of a minimum adequate standard of living to people unable to support themselves through work, savings or other means.

Societies can choose to meet this objective in a number of ways. They include negative income taxes, social insurance and category-based mean-tested approaches. A negative income tax system could be based on a single basic level of income paid to everybody and taxed at a single rate. Alternatively there could be a number of category-based minimum-income payments with different tax rates. A pure negative income tax system with an adequate basic income would involve a very high tax rate as every individual would be paid a transfer. This would be unlikely to gain public acceptance in the face of perceptions that people are being paid regardless of need, obligations or behaviour. To date, no country has instituted a pure negative income tax based scheme.

Most countries provide assistance to people's category-based life circumstances or contingencies such as old age, disability, sickness and unemployment. The two main models are social insurance and general revenue-financed means tested assistance payments. Many countries combine both approaches; however, Australia has opted for category-based means-tested assistance payments supplemented by private superannuation (including a mandatory component — the superannuation guarantee), industrial entitlements (including sick leave) and State government workers' compensation schemes.

Social insurance is based on income-related contributions and pays income-related benefits. History has shown that there is unlikely to be support for a major shift in Australia to social insurance as there is an underlying social preference for payments targeted to poverty alleviation rather than to income maintenance. Also, the transition costs of moving to a social insurance scheme could be particularly high. This would be the case even if payment levels were not increased because more people would receive a payment, or an increase in payment, which would have to be funded by higher taxes on those already in the workforce. While history and revealed preferences should not rule out major policy changes, current approaches are widely supported. For those who wish to make additional provision for possible risks, insurance arrangements are generally available.

While there is currently no strong case for a shift away from a categorical approach, the number and structure of categories — including differences in rates of payment, means tests and activity requirements — can and should reflect changing social expectations such as attitudes to work (and retirement), parenthood and disability.

Targeting support to categories of need

Categorical distinctions — such as single parenthood, disability or unemployment — assist in targeting support to those with varying need and capacity to support themselves. Such distinctions also give effect to various social judgements about who should receive assistance.

Within a categorical means tested system of income support, the most important design elements — aside from the eligibility criteria for different payments — are the adequacy or rate of the payment, means tests (how assistance is withdrawn as income rises and need falls), and participation or activity requirements. Participation requirements primarily define who is and who is not expected to undertake an approved activity in return for assistance. Requirements are based on social expectations of particular groups to undertake paid work in order to support themselves. For unemployed adults, the primary activity requirement is to look for full-time work, while primary carers are subject to part-time activity requirements when their youngest child is aged six. Disability Support Pension and Carer Payment are not subject to participation requirements, as recipients are either unable or unavailable to participate in significant part-time work in the open labour market.

Another argument in favour of categorisation is that different groups of people respond differently to the incentives they face, particularly in relation to work. Total costs and disincentives to work may be minimised if rates of payment and means tests reflect group differences (Akerlof 1978). However, there can also be significant differences within groups, and this may impact on the effectiveness of incentives targeted to a particular group (Viard 2001).The practical effect of varying the incentives faced by different groups of people is to maximise the number of hours worked in the economy — maximising full-time work by those able and available to work full-time and maximising part-time work by those who are not able, not available or not expected to work full-time (Saez 2001; Kaplow 2006).

A disadvantage of categorical income support systems is that they are complex. There is also a tendency to increase complexity by creating new categories of payment and additional means tests in an attempt to target the payments ever more tightly to need, while ensuring that no deserving group is left out. There is certainly an optimal level of categorisation2, though determining this would be difficult in practice. The level would in any event vary with social, economic and labour market conditions. However, one factor to take into account in determining the effectiveness of existing categories is the extent to which differences in rates and conditions of payment give people an incentive to change their category.

To the extent that it is possible for the same person to be eligible for different payment categories, or to experience large changes in assistance without commensurate change in circumstance, there will inevitably be concerns with horizontal equity (that is, treating people in similar economic circumstances similarly). For example, a person with disability who is assessed as being able to work 14 hours a week rather than 15 hours a week can receive a higher level of assistance and lower rate of withdrawal of income support. Another example occurs for single parents on their youngest child's eighth birthday. Even if the single parent was already working 20 hours a week, their disposable income would fall significantly on the birthday of their child.


Income support payment categories should be limited in number and reflect current social expectations about the responsibilities that individuals in different circumstances have to look for work.

Differences in the treatment of individuals in similar circumstances on different payments should be limited to minimise horizontal inequities and disincentives to work created by categorical differences. This means that minor changes in personal or family circumstances should lead to only minor changes in payment conditions.

Payment adequacy, poverty and incentives to work

There is no single agreed definition of income poverty, nor is there an agreed way to measure the adequacy of support rates. What would have been seen as an adequate level of payment according to 'community standards' in 1950 might not be seen as adequate today, and what might be seen as adequate today might not be seen as adequate in 2020. Aside from changes in living standards and the distribution of work and income, views about the best way to avoid or alleviate poverty can also change.

The level of payment has to be balanced with incentives to work. It also needs to be seen as affordable, sustainable and fair by the community. These considerations would support a lower level of payment for those expected to work full-time compared to those not expected to work and likely to be reliant on income support for a long time. For all those not expected to work, maintaining incentives to work is not as crucial. Social expectations that individuals would be able to share resources such as income or accommodation — with parents or a partner — may also justify different rates of payment.

While there is no clear principle for establishing the level of payment for those who are expected to work, the rate once determined should be maintained against some measure of community living standards. One measure that has been used for pensions is a wages benchmark. Alternatively, there could be indexation to some other measure with regular review of whether this was continuing to meet community standards. Indexation of the payment solely to prices can reduce adequacy relative to members of the community who work.


The highest level of payment — pensions — should be paid to those who are not able or expected to work and who are likely to be reliant on income support for a long time. Rates of payment for those who are able and expected to undertake significant levels of work — allowances — should be lower in order to maintain incentives to work.

Relativities between pension and allowance rates of payment should be maintained by indexing all rates by the same measure of community standards.

The higher the level of income support for people of working age, the more likely it is a disincentive to work. The point at which an individual will choose to work will vary with individual characteristics (such as skills, experience and capability reflected in a person's potential wage rate), work preferences (for example, to work full-time, part-time or not at all), and the design of the tax and transfer system (such as income support participation requirements and withdrawal rates as income from work increases).

Participation requirements ensure that income support recipients with the capacity to undertake paid work actively seek such work. Participation requirements also provide assurance to the broader society ('taxpayers') that those experiencing a period without sufficient means are not only provided with an adequate level of income but are obliged to make appropriate efforts to reduce their reliance on government assistance.

Another consideration is the level of income support relative to minimum wages, and minimum wages relative to median earnings, as this will determine the number of people whose workforce incentives may be affected.

These factors interact. For example, where there is a high minimum wage and a moderate payment rate, it is possible to have a high income support withdrawal rate with little impact on incentives to work. This would occur where individuals wish to work full-time and income support is withdrawn well before the minimum wage. This conclusion would be less clear if there were limited labour demand for low-paid full-time work, but some demand for low-paid part-time work.

In addition, some people prefer part-time work to non-employment but prefer non-employment to full-time work, such as some people with caring responsibilities or primary care of young children. Others may not be able to work full-time (due, say, to disability) and would be adversely affected by an incentive structure that favoured full-time work.

If income support is withdrawn beyond the minimum wage there would be limited impact on incentives to work for individuals who prefer part-time work. This is likely to be the case for many primary carers of young children. However, such a design may create disincentives to work if individuals prefer full-time work.

In considering the effect of any incentives to work, it is important to consider both substitution effects and income effects. A substitution effect is where an improvement in the return from work, for example due to either a withdrawal rate or marginal tax rate reduction, leads to an increase in the hours worked, with a reduction in the return from work leading to a fall in the hours worked. Income effects occur when additional income — for example, from an income support rate increase or an effective tax rate reduction — reduces hours worked, as the person can achieve a similar standard of living with less work. Conversely, the income effect could mean that a rate reduction or withdrawal rate increase would lead to an increase in hours worked.

These effects are present together in almost any change to the financial arrangements of income support. It is important to consider both effects at the same time. For example, a reduction in the withdrawal rate of Parenting Payment Single (PPS) may have the effect of increasing the return from working. But it may also increase disposable income for those on PPS, thereby reducing or negating the positive effect of the reduction in the withdrawal rate. Put another way, if people can achieve an adequate standard of living by only a limited amount of work they are more likely to do so. In any given change, which of these effects dominates is a matter of empirical evidence — theory will not tell us which is more important. However, in most circumstances the substitution effect tends to dominate the income effect where income levels are low, in which case higher taxes and tighter means tests will reduce the amount of work by those who face them. This is less of an issue where a person is highly motivated to work and can attain enough work to get off payment altogether.

Ideally, the rate of payment to people expected to take up full-time work should remain well below the minimum wage and cut out before the minimum wage. This would help to ensure that people can see a clear benefit from work. The effectiveness of this minimum wage constraint depends on a number of factors, including how easy it is for people to qualify for higher-rate payments such as the Disability Support Pension or Parenting Payment Single, and the structure of payments for children as parents increase their work effort. It will also depend on the overall need to ensure that recipients take up part-time or intermittent work where it is the only work available to them.

The trade-off between payment adequacy and program affordability or sustainability is outlined in Box F1-1. To reduce disincentives to work without reducing payments for income support recipients would involve either relaxing (or removing) means tests or targeting payments to people according to characteristics such as age or duration on payment. The income support system uses both of these strategies.

Box F1-1: Balancing adequacy, incentives and affordability

With any income support payment, there is an 'iron triangle' associated with means testing (see Chart F1-1). The generosity of the payment (including the breadth of its coverage) needs to be balanced by how much it costs taxpayers, and the incentive for people to get off the payment by earning income. Improving one of these worsens one or both of the others.

Chart F1-1: Iron triangle of means testing

Chart F1-1: Iron triangle of means testing

To improve incentives without reducing the payment level there are only two possible strategies:

  • relax or remove some means tests — apart from raising inevitable questions of 'middle class welfare' and 'churning', this would greatly increase the cost to taxpayers and jeopardise affordability; or
  • reduce or remove the payments to some people on grounds other than income, such as age or an expectation that duration on income support will be short — this would reduce the total income of those people and may compromise adequacy.

Whether the pattern of incentives is changed by adjusting the means test or using a tax instrument such as an earned income tax credit does not affect this trade-off; it simply changes it from a question of welfare design to a question of tax design. Which is the better approach depends on other issues such as administrative practicalities, signalling effects (such as the weight given to work), and tax churning.

2 This can be defined as 'the point where the marginal social benefit equals the marginal social cost in terms of distortions in decisions of what group to belong to' (Moffitt 2008, p. 11).