Improving Work Incentives and Incomes for Parents: The National and Geographic Impact of Liberalising the Family Tax Benefit Income Test*
This research was supported by a Linkage Grant from the Australian Research Council (LP775396), with the research partners on this grant being the NSW Department of Community Services; the Australian Bureau of Statistics (ABS); the ACT Chief Minister’s Department; the Queensland Department of Premier and Cabinet; Queensland Treasury; and the Victorian Departments of Planning and Community Development and Education and Early Childhood. The authors gratefully acknowledge the support provided by these agencies. However, the authors are solely responsible for the conclusions reported in this paper and the results cannot be interpreted as construing endorsement by New South Wales, Victoria, Queensland, Australian Captial Territory or ABS of any policy views, the methodology or results. The authors also thank two anonymous referees for valuable comments on an earlier draft of this paper.
Abstract
The effective tax rates and possible work disincentives created by Australia’s tax and welfare systems have been receiving extensive policy attention in recent years. Family Tax Benefit-Part A (FTB-A) is one of the key causes of high effective marginal tax rates for many families. This study uses national and spatial microsimulation models to evaluate the national and local impacts of a possible FTB-A reform option, which involves reducing the income test withdrawal rate associated with the FTB-A income test. The modelling suggests that the option would be an effective way to reduce high effective marginal tax rates for around 415,000 parents of FTB-A children, would benefit around 850,000 families, and would deliver additional assistance to middle income families living on the outskirts of our cities.
I Introduction
Like many other industrialised countries, Australia is facing structural population ageing. While today five Australians of working age are available to support every Australian aged 65 years and over, by 2047 that ratio will fall sharply, to only 2.4 Australians of working age for each older Australian (Treasury, 2007a, p. 17). Population ageing is expected to have major consequences for our society and economy – and to create ‘fiscal gap’ pressures for both Federal and State government budgets (Productivity Commission, 2005; Treasury, 2007a). Prompted by these concerns, like other countries, Australia is now focussing on more active labour market policies and on whether its welfare systems need to be restructured to improve work incentives – with Treasury being explicitly asked to consider the impact of high effective marginal tax rates within its current 2008–2009 review of the taxation system (Cotis, 2003; Lim, 2008).
The decision to join the labour force or to increase hours of work is affected by the returns from work. An effective marginal tax rate (EMTR) shows how much of an additional dollar of income is kept by individuals and families, after the payment of income tax and the withdrawal of any means-tested cash payments from the government (such as age pension, Family Tax Benefit (FTB) or Newstart Allowance). For example, an EMTR of 70 per cent means that only 30 cents is retained ‘in the hand’ after a $1 increase in private income.1 A high EMTR means that it may not be worth a person outside the labour force taking a job or, for those who are already in the labour force, increasing their hours of work – particularly for the population sub-groups of mothers whose labour supply is most responsive to high effective tax rates and work disincentives (Apps, 2006; Dandie & Mercante, 2007; Kalb & Thoresen, 2007). Many therefore argue that one way to encourage labour force participation or to increase hours of work is by reducing the level of marginal tax rates (Lateral Economics, 2006).
Over the decade to 2006–2007, there were significant changes in the EMTRs of Australians. Despite income tax cuts, 7.1 per cent of working-age Australians faced EMTRs of 50 per cent or more in 2006–2007, up from 4.8 per cent in 1996–1997 (Harding et al., 2006a). The proportion of working-age parents living in families having couple with children facing such high EMTRs tripled. Sole parents were similarly affected, with one in every five sole parents in 2006–2007 set to keep less than half of their next dollar of wages (Harding et al., 2006a). One of the most important causes of these higher EMTRs lay in the significant liberalisation and expansion of FTB over the decade.
Although EMTRs are important to examine in their own right, the other side of the equation is their impact upon disposable (post-income-tax, post-transfer) income. Recent research in Australia has developed a method to calculate the local- or small- area effects of policy change on incomes and poverty rates (Tanton et al., 2008, 2009b). Small area disadvantage and poverty is becoming recognised in Australia as an important aspect of disadvantage and social exclusion, and there are now many small-area indexes of disadvantage in Australia (Vinson, 2007; ABS, 2008; Tanton et al., 2009a). A locational approach to disadvantage is now one of the Federal Government’s (2009) social inclusion priorities. A key interest from the Federal Government in small area disadvantage is in ensuring that cash transfers and services are directed to the right places – and the method outlined in this study allows researchers and policy-makers to see where the proposed policy change has the greatest effect. This then allows policy-makers to target areas of disadvantage or, alternatively, areas of concern to policy-makers (such as neighbourhoods where young families or unemployed families are concentrated).
This study examines the impact of liberalising the income test for FTB-A on the effective marginal tax rates and disposable incomes of families, using microsimulation models to estimate both the national and local effects of policy change. The next section explains the data and methodology used; Section III reports the results at a national level; Section IV presents the spatial results and Section V provides some policy implications and concludes the paper.
II Data and Methodology
(i) FTB Programme
FTB-Part A (FTB-A) is a Federal Government income-tested programme of cash assistance to families with children which is generally paid to the mother. High-income families are precluded from receipt of the payment, with the income test varying with the number and age of children. As an example, for the financial year 2008–2009, a family with three children aged less than 12 years old will not receive any FTB-A if family income is more than $122,263 per year. The maximum rate of FTB-A is received when family income is below $42,559 a year, and the payment is progressively reduced once the income exceeds that level.
There are currently two income test tapers, a 20 per cent taper which applies above the bottom threshold of $42,559 and a 30 per cent taper which applies above a second higher threshold of $94,316 (plus $3796 for each child after the first). As noted before, when the family income is less than $42,559, the maximum rate of FTB-A is paid. The payment then reduces by 20 cents for each dollar above that limit until the payment reaches what is termed the ‘base rate’ of FTB-A (at $82,855 in this three children example). The payment then stays at this rate until the family income reaches $94,316 per year (plus $3796 for each FTB child after the first), when the 30 per cent taper starts. After that, the payment reduces by 30 cents for every dollar of family income over the second threshold until the payment is extinguished. (The first threshold of $42,559 does not vary with the number and age of children, whereas the second threshold does vary with the number of children eligible for FTB-A).2 About 1.9 million families with children are expected to receive FTB-A in 2008–2009. Around two-thirds of all families with children receive FTB-A, with the remaining one-third being excluded by the income tests or other eligibility conditions.
In the policy reform option modelled in this simulation, the first income taper is reduced from 20 to 10 per cent and the second taper is left untouched at 30 per cent. This policy reform is summarised in Figure 1.
Illustration of the Current Family Tax Benefit-Part A System and the Reform Option for a Couple with Two Children, 2008–2009
Note: The figure compares the weekly amount of assistance (excluding supplement amount) received by couples with two children aged 4 and 10 years at different levels of private income.
(ii) National Results
The national simulations contained in this analysis are produced using NATSEM’s (National Centre for Social and Economic Modeling) Static Incomes Model (STINMOD). This model was first developed in 1994 and has been continuously updated since by NATSEM. STINMOD is Australia’s best-known static microsimulation model and is used by commonwealth departments for their analysis of the impact of policy reforms (Bremner, 2005; Treasury, 2007b). NATSEM is also regularly commissioned by government agencies and other policy actors to undertake research using STINMOD (Beer, 2003; Harding et al., 2005, 2006b; Lloyd, 2007). A user-friendly version of the model is publicly available. The model is used to calculate the simulated impact of major Australian Federal Government cash transfers, income tax and the Medicare levy. The model estimates the aggregate first-round fiscal impact of a change in tax and/or transfer policy on revenue or government expenditure. The model also estimates the first-round distributional impacts of policy change at the income unit level, for groups of people and individuals, that is, who wins, who loses and by how much.
For this analysis, STINMOD/07 was used, with some modifications to incorporate the tax cuts and other changes announced in the May 2008 federal budget. In this article, the year under study is the 2008–2009 financial year, with tax and social security parameters averaged over the 2008–2009 financial year.
The methodology used for estimating the EMTRs of working-age individuals aged 15–64 years is described in more detail in Harding et al. (2006a, pp. 22–3). In summary, the private incomes of all family heads were increased by $1 and the subsequent change in the family’s disposable income was used to calculate the EMTR facing the individual.3 The same methodology was then applied to any working-age spouse, where present. Thus, the results show the EMTRs resulting for working-age individuals when their private incomes increase by $1, taking full account of the impact upon their family’s disposable (post-income-tax, post-transfer) income.
This methodology is required because it is possible for an individual to appear to have a personal EMTR of zero, yet for the disposable income of their family to be affected if there is $1 increase in their private income. For example, let us take a low-income family with a couple and two children where only the father is in paid work and the mother is initially not working. If the mother finds a very small amount of low-paid part-time work, her income may still be below the relevant tax and social security thresholds, so that her personal EMTR may at first glance appear to be zero. But if the family is already in the income zone where FTB-A is being reduced as a result of the father’s earnings, then the actual EMTR faced by the family and thus the mother when her private income increases could be 20 per cent or more.4
(iii) Small-Area Results
In recent years, NATSEM has been attempting to move beyond the national- and state-level results produced using STINMOD, by using spatial microsimulation to show the effect of policy change at the small-area (or neighbourhood) level (Chin et al., 2005, 2006a,b; Chin & Harding, 2006, 2007; McNamara et al., 2007; Tanton et al., 2008). When the ABS issues the microdata files from its national sample surveys, it attaches a ‘weight’ to the record of every household within the sample. For example, the weight attached to the first household within this sample file represents the number of comparable households within Australia that the ABS believes are similar to that particular household. These weights are the mechanisms used to ‘gross up’ from the sample survey results to estimates for the whole of Australia. In a series of research projects, NATSEM has refined the technology to weight the ABS sample survey files to small-area targets derived from the census (Lymer et al., 2008). This then creates a synthetic household microdata file for each Statistical Local Area (SLA) in Australia. In essence, the technique creates a set of synthetic households who replicate, as closely as possible, the characteristics of the real households living within each small area in Australia.
This paper utilises SpatialMSM/08A. This version of the modelling has derived results for each of the 1435 SLA across Australia by reweighting the microdata from the 2002–2003 and 2003–2004 Surveys of Income and Housing to SLA benchmarks from the 2006 Australian Census data. These two ABS income surveys are also the two surveys used as the basefiles for STINMOD, so the weights from our spatial microsimulation model can be fused to simulations from our STINMOD model to derive the small-area effects of changes to social security and tax policies. The results show what proportion of families gain from the policy change, and by how much, on average, each SLA would gain in terms of incomes from the policy option. SpatialMSM/08A thus blends together the 2002–2003 and 2003–2004 ABS Survey of Income and Housing Costs unit record files, the 2006 Census and a specially modified version of STINMOD/07. Earlier validation of the results of our spatial microsimulation techniques has suggested sufficient reliability for the results to be used in policy analysis (Chin et al., 2005, 2006b).
It should be noted that, in both the national- and small-area results, the simulations only show the first round effects of the policy change, before any Australians change their behaviour in response to the policy change. Thus, the simulations show the ‘morning-after’ impact of the new policy, before some parents are attracted by the higher returns from working to increase their labour supply and before others feel that now that the family is receiving increased cash assistance they can afford to cut back on their working hours and spend more time with the children. A further qualification is that, in assessing the policy reform, no attempt is made to offset the additional government expenditure through increased taxes or reductions in other outlays.
III National-Level Results
The national first round impact of the proposed policy reform is estimated to increase the cost of FTB-A to $14 billion, up by $1.7 billion on the estimated cost for the existing system in 2008–2009 of $12.3 billion. As the change only applies to the income test taper associated with FTB-A, the cost of the related FTB-B programme (which provides additional assistance to single-income families) is not affected.
When contemplating any such policy change, policy-makers are always keen to assess where any winning (or losing) families are within the income distribution. Column 2 in Table 1 reports the number of families who gain from the proposed policy by their decile of equivalised disposable income.5 To derive the deciles, all Australians were ranked by the equivalent disposable income of their STINMOD income unit (colloquially described here as ‘families’). The bottom decile thus consists of the 10 per cent of Australians with the lowest incomes (and thus includes, e.g. older Australians and single people who will not benefit from the FTB policy change). To give a sense of how these deciles translate from equivalent to more easily grasped non-equivalent income measures, those FTB-A families whose equivalent disposable incomes are sufficiently high to propel them to the top decile of the national income distribution have a weekly gross income of around $4500 – equivalent to just under $250,000 a year. These high-income levels explain why there are only a handful of ‘winning’ families at the top end of the income spectrum as, for most, their incomes preclude them from receipt of FTB-A.
| Income decile | Number of gaining families | Proportion of parents of FTB-A children facing high EMTRs | Number of parents of FTB-A children facing high EMTRs | |||
|---|---|---|---|---|---|---|
| Current system | Reform option | Current system | Reform option | Change | ||
| No. (‘000) | % | % | No. (‘000) | No. (‘000) | No. (‘000) | |
| 1 | 0 | 1.0 | 1.0 | 0.9 | 0.9 | 0 |
| 2 | 0 | 1.6 | 1.6 | 4.2 | 4.2 | 0 |
| 3 | 4.8 | 11.9 | 11.5 | 52.6 | 50.7 | −1.9 |
| 4 | 82.0 | 21.2 | 11.7 | 128.9 | 69.8 | −59.1 |
| 5 | 186.2 | 26.5 | 8.5 | 166.4 | 51.6 | −114.8 |
| 6 | 193.1 | 21.8 | 7.8 | 133.3 | 48.9 | −84.4 |
| 7 | 171.6 | 14.2 | 3.5 | 88.2 | 22.1 | −66.1 |
| 8 | 141.5 | 21.0 | 7.4 | 125.8 | 44.8 | −81.0 |
| 9 | 74.0 | 5.4 | 3.5 | 24.8 | 16.6 | −8.2 |
| 10 | 2.3 | 0.7 | 0.8 | 2.6 | 2.9 | 0.3 |
| All | 855.6 | 15.6 | 6.7 | 727.8 | 312.6 | −415.2 |
- Notes: The first column examines the number of families. The shaded columns shift the focus to considering individuals rather than families, also, looking at the extent of change in EMTRs for all FTB-A working-age parents, not just those who are winners under the policy option. It should also be noted that the term ‘parents’ is used colloquially here. That is, for example, if the mother of an FTB-A child has re-partnered with a new person and is sharing accommodation with him, then the new (biologically unrelated) man is counted here as a ‘parent’. Thus, the term ‘parent’ here captures living arrangements rather than biological parenthood.
- Source: STINMOD/07.
Returning again to Column 2 in Table 1, which shows the number of winning families within each income decile, it can be seen that the proposed policy reform essentially benefits only those families who fall within deciles 3–10. Those families at the very bottom of the income distribution do not benefit at all from the proposed changes. This is partly because the incomes of any families with children at the bottom of the income distribution are not high enough to be affected by the proposed changes to the FTB income test. (In other words, such families already receive maximum FTB-A, and so are not affected by the liberalisation of the income test). It is also because families with children now tend not to be clustered in the bottom deciles of the national income distribution. Approximately 856,000 families are expected to gain from the increase in FTB-A simulated under the reform option.
It is also important to examine the impact of these proposals in terms of EMTRs. Table 1 also reports the proportion of working-age parents who live in a family containing at least one FTB-A child and who face an EMTR of more than 50 per cent. Such parents are ranked into national deciles of equivalent disposable income and the table compares the current system with the policy reform. An EMTR of at least 50 per cent can be considered to be high, as this is above the top rate of income tax plus Medicare levy (Harding et al., 2006a). Looking at the middle of the income distribution first, for those FTB-A parents whose family incomes place them in decile 5, Table 1 suggests that, under the existing system, 26.5 per cent of such parents face high EMTRs of more than 50 per cent. Under the reform scenario, this falls to 8.5 per cent of such parents. Similarly, for those whose family incomes are marginally higher, so that they are placed in decile 6 of the national income distribution, 21.8 per cent of the FTB-A parents currently are set to lose more than half of the next dollar of income they earn in welfare cuts and tax increases. Under the reform option modelled, this falls to 7.8 per cent.
Table 1 also examines the number of working-age parents of FTB-A children facing EMTRs of more than 50 per cent. It indicates that some 199,200 parents in deciles 5 and 6 currently facing EMTRs above 50 per cent would no longer face such high EMTRs if the reform option was adopted. Similarly, 66,100 FTB-A parents in decile 7 families would see their EMTRs fall below 50 per cent if the reform option was implemented – and another 81,000 parents in decile 8 families would benefit similarly. Overall, the reform option would mean that only 6.7 per cent of all FTB-A parents would face high EMTRs, down from the 15.6 per cent of such parents facing such high EMTRs under the existing system.
Moving from the proportion of parents affected to the actual number, an estimated 313,000 parents of FTB-A children would still face high EMTRs if the reform option was implemented. However, this is a substantial improvement on the 727,800 FTB-A parents who currently face such high EMTRs (Table 1). By ‘liberating’ about 415,000 FTB-A parents from high effective tax rates, the reform option would significantly reduce the number of working-age Australians affected by high effective tax rates. To put this reduction in perspective, our calculations using STINMOD show that an estimated 900,000 working-age Australians faced high effective tax rates in 2008–2009. The fall by 415,000 in the number of working-age parents facing high EMTRs resulting from the policy reform would thus ‘liberate’ nearly half of all the working-age Australians facing high EMTRs in 2008–2009. Alternatively, focusing just on the parents of FTB-A children rather than upon all working-age Australians, the policy reform option would more than halve the number of FTB-A parents facing high EMTRs of more than 50 per cent.
At first glance this appears to be a significant benefit of the proposed change in policy. But what are the outcomes by gender and by labour force status, given that earlier research suggests that the labour supply of married men is relatively inelastic, whereas that of mothers is relatively elastic (Breunig et al., 2005; Dandie & Mercante, 2007)? Table 2 reports the proportion of working-age parents, classified by gender and labour force status, who live in a family containing at least one FTB-A child and who face an EMTR of more than 50 per cent, both for the current system and after the policy reform option. It suggests that, under the existing system, 11.4 per cent of female FTB-A parents face high EMTRs of more than 50 per cent. That proportion is almost double for male FTB-A parents, at 20.5 per cent. Under the reform scenario, the proportions reduce significantly, to around 6.8 per cent for both genders. Moving to the numbers affected, the proposed reform measure would release more than 123,000 women and more than 291,000 men from facing high EMTRs, totalling to 415,000 people.
| Labour force status | Current system | After policy change | ||||||
|---|---|---|---|---|---|---|---|---|
| Woman | Man | Woman | Man | |||||
| % | No. (‘000) | % | No. (‘000) | % | No. (‘000) | % | No. (‘000) | |
| Full-time employed | 18.1 | 125.8 | 21.3 | 392.7 | 8.1 | 56.3 | 6.1 | 111.6 |
| Part-time employed | 18.6 | 155.2 | 30.8 | 33.2 | 12.2 | 101.8 | 22.0 | 23.7 |
| Unemployed | 0.3 | 0.2 | 2.8 | 0.9 | 0.3 | 0.2 | 2.8 | 0.9 |
| Not in labour force | 1.1 | 10.2 | 6.6 | 9.6 | 1.0 | 9.7 | 5.8 | 8.5 |
| Total | 11.4 | 291.4 | 20.5 | 436.3 | 6.6 | 168.0 | 6.8 | 144.6 |
- Source: STINMOD/07.
An important issue is the labour force status of those parents released from facing high EMTRs under the policy reform. Table 2 indicates that 18 per cent of female parents (or 125,800 mothers) who work full-time currently face high EMTRs – and that proportion would fall to around 8 per cent (56,300 mothers) after the policy reform. The shift is even more dramatic for male parents, from 21.3 to 6.1 per cent, corresponding to 392,000 fathers facing high EMTRs under the current system, falling to only 111,600 fathers after the reform. The reform policy, therefore, would release a total of 350,000 parents working full-time from facing high EMTRs.
Nevertheless, a more important implication from the results in Table 2 is the number of mothers who work part-time and who would be released from facing high EMTRs by the proposed reform policy. This is because previous research has shown the labour supply of this group to be particularly responsive to work incentives. Under the current system, 155,200 mothers holding down part-time paid jobs face high EMTRs. Under the reform policy, this figure would reduce to around 102,000, releasing some 50,000 mothers working part-time from high EMTRs. The lower number of fathers in part-time work means that this change has less impact on fathers.
IV Small-Area Results
Although the national averages are useful to see the effect of this policy change across Australia, they mask the highly pronounced differences for small areas. Understanding these differences gives a greater understanding of how the policy affects certain groups of people – and also gives policy-makers within the government some idea of where the effect of the policy change is the greatest.
Figure 2 illustrates the small-area impact of the liberalisation of the lower FTB-A income test taper on the weekly household disposable incomes of households with dependent children. The average gain in each SLA has been calculated by summing the total increases in the disposable incomes of households with dependent children in that SLA, and then dividing by the number of households with dependent children in that SLA. The map thus shows the average weekly gain in the disposable income of households with dependent children in each SLA from the policy shock.
Estimated Change in Weekly Disposable Incomes of Households with Dependent Children Under Reform Policy, by Statistical Local Area, 2008–2009
Source: SpatialMSM/08A applied to STINMOD/07.
The pale areas on the map are those neighbourhoods showing low gains from the policy reform, with gains averaging less than $8.50 per household per week. (It should be emphasised here that only households with dependent children are considered within this map, so it is not the case that neighbourhoods show low gains simply because they contain very few households with children.)
It can be seen from Figure 2 that the greatest effect of this policy change is in the newer, middle- to low-income suburbs, on the outskirts of the capital cities. This is not surprising as the change in the taper rate (illustrated in Figure 1) shows that for families with either very low incomes and very high incomes there is no change to the benefit paid (this is because either the maximum is paid; or nothing is paid). The change to the taper rate means that those on middle incomes (earning from $1000 to $2000 per week) benefit the most.
These areas tend to be in the newer areas on the outskirts of the cities. So in Canberra, for example, such trends are evident at Ngunnawal (which is in the new area of Gungahlin in the north) and at Gordon in Tuggeranong in the south. In Melbourne and Sydney, it is those suburbs on the outer fringe of the city, like Liverpool in Sydney; and Cardinia-Packenham in Melbourne. Low-income areas like Dandenong in Melbourne show only small increases in income after the policy reform, because the change does not affect families on low incomes.
In Hobart, the simulated increases are again reasonably low, because incomes in Hobart tend to be low. This story is repeated in remote Queensland and New South Wales. There are some remote areas with a large change, and these may be mining areas, where the typical family incomes are from medium to high. These are also very large SLAs, so there may be some averaging of extreme high incomes in mining towns and extreme low incomes in the surrounding Indigenous communities, so the results for these areas may be affected by this heterogeneity.
In Brisbane, high-income areas like Pinkeba/Eagle Farm showed small increases, because of the quarantining of the gains to middle-income families. In contrast, middle-income areas like Ipswich showed high increases in disposable incomes from the policy change.
Because the spatial analysis shows results for a number of areas, it can give us a better understanding of how the policy change affects families in Australia. What the regional analysis shows is that areas benefiting most from this policy change are also areas on the outskirts of capital cities, where young families with low- to middle-incomes are seen. Areas that do not benefit are those where families with low incomes predominate (such as remote areas) and areas where families with high incomes reside (such as inner city areas).
For policy-makers, the spatial analysis shows that most of the benefit from this policy change will go to middle-income families with children on the outskirts of capital cities. If the government was looking for a policy to assist these middle-income families, then the policy described in this article would achieve this objective.
V Conclusion
The effective tax rates and possible work disincentives created by Australia’s tax and welfare systems have been receiving extensive policy attention in recent years. The imminent mass retirement of the baby boom cohort and structural population ageing have caused many countries in the developed world to review whether their tax-transfer systems are adversely affecting incentives to work.
Earlier research has identified FTB-A as one of the key causes of high effective marginal tax rates for many families. Accordingly, in this article we tested a possible FTB-A reform option, which involved reducing the first income test taper for FTB-A from 20 to 10 per cent. NATSEM’s national microsimulation model of the tax-transfer system, STINMOD, suggested that this option would cost the Federal Government an additional $1.7 billion if implemented in 2008–2009, raising total expenditure on FTB-A by just under 14 per cent. The measure would benefit around 856,000 families and reduce the number of FTB-A parents facing high effective marginal tax rates by around 415,000, almost halving the number of working-age Australians facing EMTRs above 50 per cent.
More detailed analysis of these 415,000 parents by gender and labour force status indicated that the reform policy would release a total of 350,000 full-time parents from facing high EMTRs, although it should be appreciated that the majority of these parents were fathers working full-time, a group whose labour supply might be relatively less responsive to work disincentives. More importantly, more than 50,000 mothers working part-time would cease facing high EMTRs of more than 50 per cent under the reform policy.
The small-area impact of the reform option was assessed using NATSEM’s SpatialMSM/08A model, which has been linked to the STINMOD/07 model. The results showed that neighbourhoods populated by middle-income households with children would benefit from the policy change, with households in many areas expected to enjoy an increase by $11 to $13 in disposable income from the policy change. Such households were typically concentrated in newer suburbs on the outskirts of the cities.
Apart from illustrating key impacts of a possible reform in the FTB-A income test, a second goal of this article was to demonstrate how microsimulation models can be used in the policy reform process to illuminate both the national and local consequences of possible policy changes. In NATSEM’s earlier research, we have used our new spatial microsimulation technology to assess the small-area impact of income tax changes (Chin et al., 2005) and changes in Commonwealth Rent Assistance (McNamara et al., 2007). The research reported in this article is the first attempt in Australia to simulate the small-area impact of changes in FTB. The results suggested that the spatial microsimulation modelling was performing well, with both the national- and small-area models providing similar estimates of the total cost to the Federal Government from the policy changes. (For example, both the national- and small-area versions of the STINMOD model suggested that the existing FTB-A programme would cost the government $12.3 billion in 2008–2009, whereas the national STINMOD model suggested that the reform option would cost an extra $1.7 billion and the small-area STINMOD model suggested that it would cost an extra $1.4 billion).
Although there are clearly some differences between the two modelling approaches, this nonetheless suggests that the small-area modelling could be of great use to policy-makers when they are attempting to explore the local implications of possible policy changes. Thus, for example, this new technology could be used to assess the likely spatial consequences of adjusting income support or FTB-A payments for a proportion of the cost increases facing low-income Australians as a result of price increases triggered by an Emissions Trading Scheme. Similarly, the new spatial modelling infrastructure could be used to examine the distribution among different neighbourhoods of the multibillion dollar stimulus packages introduced in 2008–2009 by the Federal Government in response to the Global Financial Crisis.
