There are a great many issues concerning the Child Support Scheme that were the subject of comment or criticism in submissions to the House of Representatives Standing Committee on Family and Community Affairs, in the course of its Inquiry held in 2003, or that have been raised with the Taskforce since it began its work.
Many of these issues are perennial, such as the definition of income for child support purposes and the problem of how to assess the real capacity to pay of people who are self-employed. Others concern the operations of the Child Support Agency (CSA). Many of these perennial issues were considered in great depth by the Joint Select Committee on Certain Family Law Issues, chaired by Roger Price MP, which reported in November 1994. Other issues were dealt with in the House of Representatives Standing Committee on Family and Community Affairs report,
Every Picture Tells a Story (December 2003).
From research in the course of this Review, the following major issues emerged as central to the Terms of Reference of the Taskforce.
6.1 Child support expenditure as a percentage of income
The Scheme assumes that, across the income range, people spend the same proportion of their income on children. This justifies the set percentages of 18% for one child, 27% for two children and upwards, above the exempt amount. Thus the idea is that the person on $30,000 will spend say, 20% of their income on child-related expenditure for one child, and so will the person on $80,000. That justifies using a common percentage for everyone.
However, the preponderance of international research shows that while the higher the household income, the more parents spend on their children, that expenditure declines as a
percentage of their income.
98 This is because as income increases, people may choose to put more into savings, or to use a greater proportion of income for purposes that do not involve expenditure on children. In the Australian context, the decline of expenditure as a percentage of income is also a result of our progressive taxation system, in that, due to marginal tax rates, disposable income does not increase in proportion to increases in taxable income.
This is borne out by the Australian research commissioned for this Review. Figure 6.1 looks at the costs of one child in an intact family as a percentage of total gross household income, based upon the National Centre for Social and Economic Modelling (NATSEM) costs of children research.
Figure 6.1: Estimated gross costs of two children aged 5 to 12 years as a percentage of gross family income
Note: The gross income figures have been rounded to the nearest $1,000.
Taskforce Child Support Model.
This analysis makes the fixed percentages appear problematic. As a result of fi xed percentages, at the higher ends of the income spectrum the current child support liability is well in excess of levels of expenditure on children in comparable intact families, at least for one and two children under 13 years of age. Figures 6.2 and 6.3 illustrate this in relation to one child and two children aged between five and 12 years of age.
Figure 6.2: Estimated gross costs of children and current Child Support Scheme liabilities for a payer with one child aged 5 to 12 years
Note: The gross costs of children are calculated with reference to the gross costs that would have been incurred in an intact family if the payer at each level of private income had a spouse with zero private income.
Taskforce Child Support Model.
[
top ]
Figure 6.3: Estimated gross costs of children and current Child Support Scheme liabilities for a payer with two children aged 5 to 12 years 
Taskforce Child Support Model.
These figures also show that current liabilities for low-income payers are often less than expenditure levels on children in comparable intact families. However, as will be discussed in Chapters 7 and 8, to work out whether child support obligations are adequate in relation to lower income families, it is important to take account of the (often substantial) level of government family benefits, from which households spend on their children.
6.2 The Child Support Scheme and capacity to meet the costs of children
The first objective of the Child Support Scheme states that parents are meant to share in the cost of supporting their children, according to their capacity. The assumptions underlying the existing Scheme are that:
- the percentages applicable after the liable parent’s exempt income is deducted are broadly commensurate with the expenditure that the parent would have been incurring if the two parents were living together, and to this extent represent an appropriate contribution to the costs of raising children;
- the resident parent is making an in-kind contribution proportionate to her or his capacity to pay.
The Child Support Consultative Group (CSCG) took as its starting point the proportion of family income normally devoted to children in a two-parent family.99 It then took account of a range of other factors in arriving at appropriate percentages after deducting the self-support component.100 What is unclear from the methodology is how the CSCG moved from the research findings on total family income to determine the proportion of the taxable income of the parent who is liable to pay child support. Total family income may include the earned income of both parents, together with government benefi ts payable to the intact family to support children. While the issue of the resident parent’s income was dealt with at length101, it is not clear how, if at all, it was factored into the calculation of the percentages of the liable parent’s income payable in child support.
It is likely that the CSCG assumed that because expenditure on children rises as the family income rises, the inclusion of the resident parent’s income would not reduce the liability of the non-resident parent, except when family income reached a very high level.102 This was valid on the basis of the research evidence at the time. However, given the preponderance of research evidence now that expenditure on children falls as a percentage of total family income as that income rises, to translate the research evidence on total family income to a percentage of one parent’s taxable income, without taking account of the different components of that total family income apart from the liable parent’s income, is problematic.
The CSCG did include the resident parent’s income to a limited extent. Nonetheless, as the formula now stands, only a relatively small percentage of cases exist in which the recipient parent’s income operates to reduce the liable parent’s child support. The Taskforce research indicates that 12% of payees have income that counts in the application of the basic formula103, that is, an income equal to or greater than the level of average earnings for all employees ($39,312 in 2005).
Thus the current approach looks mainly at the liable parent’s capacity to pay, not at the relative capacity to pay of both parents. While the payee is contributing to the support of the child in-kind, and there are significant indirect costs involved in parenting, especially when children are young, it is not necessarily obvious to the liable parent who is struggling to meet child support payments that the cost is being fairly distributed between the parents in accordance with their relative capacities to pay.
The results of a community attitude survey conducted by the Australian Institute of Family Studies (AIFS) in early 2005, and illustrated in Figure 6.4, showed strong support for both parents’ incomes being taken into account.
A further issue is the way in which the resident parent’s income is taken into account in the cases where it exceeds the disregarded amount of the level of average earnings for all employees. Dollars earned above that point reduce the liable parent’s Child Support Income by 50 cents in the dollar, with the consequence that the relevant percentages will apply to a lower amount than would be the case if the resident parent’s income is not factored in. The effect of payee income is capped, in that the liability cannot fall below 25% of the child support liability that would apply if the payee had no income.
[ top ]
Figure 6.4: Do you think the amount of child support that a father pays for his children should depend on how much he earns, how much the mother earns, or both their incomes?
Notes: GP nonsep = general population non-separated sub-sample; CFC sep = Caring for Children after Parental Separation sample comprising separated/divorced parents with at least one child under 18; χ2 (3) = 54.43, p<.001 (based on the two categories of responses: father’s income and both their incomes).
Smyth B. & Weston R., ‘A snapshot of contemporary attitudes to child support’, in Volume 2 of this Report, p. 31.
The effect of including the resident parent’s income can be seen in the Figure 6.5 where the non-resident parent’s income is $32,000.
Figure 6.5: Liability as resident parent income increases
As Figure 6.5 shows, the resident parent’s income has no effect at all until it reaches the disregarded level, even though it is rather higher at this level than the income of the liable parent, but as that income increases above that point, it has the effect of reducing the liable parent’s income quite sharply until the minimum 25% of the liability otherwise applicable, is reached. A more graduated approach to the inclusion of the resident parent’s income may be appropriate, taking account of the relative capacities of each parent to support the children.
6.3 Child support expenditure and the ages of children
Under the current formula, the amount of child support that a payee receives does not vary with the age of the children and therefore is not sensitive to the difference in the costs of children as they grow older. Australian research estimates that expenditure on teenagers is two to three times higher than for young children, and this pattern prevails at every income level.
104 Figure 6.6 illustrates this in terms of the costs of children as a percentage of gross household income, based upon the NATSEM costs of children research—although it is important to note that these figures exclude the costs of childcare. Costs for pre-school age children vary significantly, depending on whether full-time childcare is required to support parental employment, and on the nature of that childcare. For example, other Australian research shows that the costs of a three-year-old child for a two-parent middle-income household varies from $6,500 to $17,620 per year.
105 [
top ]
Figure 6.6: Costs of children as percentage of gross household income, by age
Percival R. & Harding A., ‘The Estimated Costs of Children in Australian Families in 2005–06’, in Volume 2 of this Report, p. 11.
The difference in costs of children as they grow older was also an issue considered by the CSCG in 1988. It decided that to assess child support with reference to the increased costs as children grow older would make the formula unduly complex.
106 It concluded that using a single percentage rate for all the years of a child’s dependence would achieve a reasonable and balanced result.
107 Whether a fair result is achieved in assessing child support over the course of childhood may depend, however, on the time of separation. If the parents never lived together, and therefore the child support liability began soon after the child’s birth, then the effect of having a fixed percentage of income throughout childhood is that the liable parent’s contribution to the costs of the child, relative to his or her income, would average out over time, being a higher proportion of the cost when the child is younger than when he or she is a teenager. However, if the children are 13 and 11 when the parents separate, the payee will not have had the benefit of the higher payments relative to cost when the children were younger and so will not benefit from this averaging.
Furthermore, the approach of averaging the costs of children over the entire age range means that child support payments are likely to be inadequate at the time that the costs of children are at their highest. Since people tend to spend their income when they have it, and savings among lower income families are not very high, it is unlikely in most cases that some child support paid in earlier years will have been saved to cope with the costs of raising teenagers. The consequence of averaging may be that separated parents caring for teenagers suffer relative disadvantage compared to those with younger children.
It is difficult to achieve the right balance between simplicity and fairness. The approach of averaging the costs of children across childhood may have been the right decision at the time, but now the system of administrative assessment of child support is well established, it may be that a different balance between simplicity and fairness can be found.
The results of the 2005 community attitude survey conducted by AIFS, in Figure 6.7, showed very strong support for child support payments being related to age.
This view was expressed by over 80% of non-separated men and women in the general population sample, and by 70 to 75% of resident and non-resident parents in the
Caring for Children sample.
Figure 6.7: Do you think the amount of child support should depend on the children’s ages? 
Notes: GP nonsep = general population non-separated sub-sample; CFC sep =
Caring for Children after Parental Separation sample comprising separated/divorced parents with at least one child under 18; χ2 (3) = 17.91,
p<.001.
Smyth B. & Weston R., ‘A snapshot of contemporary attitudes to child support’, in Volume 2 of this Report, p. 35.
[
top ]
6.4 The Child Support Scheme and the costs of contact
A parent has the same child support liability whether he or she has no contact with the children or has the children to stay overnight for 29% of nights per year. The Scheme therefore does not take proper account of the in-kind child support that is provided when children are staying with the non-resident parent.
This is not to say that regular contact was ignored when the Scheme was fi rst established. The CSCG indicated that the percentages chosen ‘recognise that while a non-custodial parent may not have high costs of access…he or she may have some costs of access.’
108 A similar approach was adopted in the United States in jurisdictions with a Scheme similar to Australia’s, for example, in Wisconsin.
109 However, because the great majority of payers pay the same amount whether or not they have the children staying regularly overnight, it is not at all clear to people that the costs of contact are recognised in the Scheme. Recognition in the formula only occurs once the level of overnight stays reaches 30% or 110 nights.
Taking account of the costs of contact is, however, problematic. If the liable parent has regular contact with the children, then the total family expenditure related to the children is necessarily much higher than it would be if the relationship had not broken down. There are duplicated infrastructure costs from having two households suitable for children to stay in, and there are transportation costs involved in seeing the children. The costs incurred by one parent are not necessarily offset by savings in the other household, other than in relation to food and entertainment costs during contact visits.
Research conducted for the Taskforce using the Budget Standards approach identified that the costs incurred by a middle-income non-resident parent having 20% contact represent 38% of the costs of raising the child by the sole parent if no contact were occurring or similar. However, the converse is that the resident parent with 80% contact incurs 99% of the cost borne had he or she been caring for the child 100% of the time.
110
6.5 Regular contact and Family Tax Benefit
The way in which contact and shared care arrangements affect entitlement to Family Tax Benefit (FTB) is quite different from the affects on child support payments under the Child Support Scheme.
Introduced in 1999, the provision for splitting FTB allows the parents to split FTB Parts A and B on the basis of the number of hours of care provided by each parent, subject only to the proviso that FTB cannot be paid to a parent who provides less than 10% of the care. By contrast, the Child Support Scheme formula only provides for a reduction in child support liability if the child spends 110 nights (which is 30%) or more each year with the paying parent.
111 Whatever the logic might be underpinning these differences, it is not apparent to the parties concerned in family matters.
In the FTB system, the FTB is shared in direct proportion to the amount of time spent in the care of each parent, without recognition that expenditure on children is not typically proportionate to the hours of care provided, except in relation to food and other such day-to-day expenses.
A consistent approach is required that minimises conflicts over money in making contact arrangements, and that operates fairly to both parents.
The results of the AIFS survey, as shown in Figure 6.8, indicated strong support for contact being taken into account in setting child support payments. All groups most commonly believed that overnight stays should be taken into account in setting child support liability. Non-resident fathers were the most likely to feel this way (82%), but close to three quarters of non-separated men and women and 62% of resident mothers in the
Caring for Children sample endorsed this view.
Figure 6.8: When children often stay overnight with their father, should this be taken into account when calculating his child support payments?
Notes: GP nonsep = general population non-separated sub-sample; CFC sep =
Caring for Children after Parental Separation sample comprising separated/divorced parents with at least one child under 18; χ2 (9) =55.78,
p<.001.
Smyth B. & Weston R., ‘A snapshot of contemporary attitudes to child support’, in Volume 2 of this Report, p. 53.
[
top ]
6.6 Numbers of payers who have a minimum assessment
Almost 41% (288,057) of all payers in the Child Support Scheme are paying the current minimum rate of $260 per year or less as at 30 June 2004. Only 138,725 (or about half) were in receipt of a Centrelink benefit (excluding those who receive FTB only). Only 9% of all payers had a minimum assessment and were in receipt of a Newstart Allowance. Unemployment is therefore not a major explanation for the substantial percentage of all payers with a minimum liability.
The CSA on behalf of the Taskforce conducted further analysis of the low-income group. It reported that of all minimum assessment cases:
- 22.4% of parents were in receipt of Newstart Allowance;
- 12.2% were receiving Parenting Payment (Single);
- 10.7% were in receipt of Disability Pension;
- 0.9% were receiving Carer Pension;
- 2.3% were receiving Partner Allowance; and
- 0.6% were receiving Age Pension.
A further 17,771 parents had relevant dependent children, meaning second families. In these cases the exempt income was a minimum of $24,842, reducing the child support liability to the minimum amount. Another 10,184 parents had major or shared care of their children, and in this situation are exempt from the minimum payment. There are also some other payers who have a minimum assessment by agreement with the other parent, or whose low income is explicable for other reasons.
112 For example, a person may have no income because he or she is in prison.
However, this still leaves a substantial proportion of all payers who have a minimum assessment of $260, and for whom there is neither a ready explanation for their low income (generally less than about $13,000) nor an indication of how they manage to support themselves in the absence of Centrelink benefits or any other form of government help, such as Rent Assistance. This is probably around 15% of all payers. Some of these are payers who lodged a tax return showing either a negative income, or an income so low that they fell below the self-support exempt amount, even though they were not on any form of income support. A substantial proportion of payers with a minimum assessment (23.9% of all those with a minimum assessment) did not lodge a tax return at all and do not fall into any of the categories listed above which would provide a ready explanation for their minimum liability.
There are many ways in which people may legally organise their financial affairs so as to minimise their taxable income. They should not also be exempt from paying all but a minimal sum towards the support of their own children. There are also ways in which people may organise their affairs so that they operate illegally, either partially or wholly, in the cash economy.
The evidence that there is a significant proportion of all payers who have a minimum liability without any obvious explanation for their low income or their means of self-support, suggests that tax minimisation and avoidance are both signifi cant problems for the Child Support Scheme. There may be many more payers who have a liability above the minimum, but whose reported taxable income does not reflect their real financial capacity to support their children.
6.7 Non-lodgment of tax returns
The problem of working out a person’s taxable income for the purposes of assessing child support payments is exacerbated by the large number of payers who either do not file tax returns at all, or do so irregularly, in breach of their legal obligation to do so.
As has been seen above, nearly one quarter (23.9%) of all those with a minimum assessment and who were not on Centrelink benefits are people who did not lodge a tax return in the most recent tax year. This is 68,770 payers.
While payee parents with incomes below the tax threshold, or with income purely from Centrelink benefits, are not required by law to lodge a tax return, parents with a child support liability (payer parents) are specifically required to lodge a tax return, irrespective of the benefit or source of income.
113 Figures provided to the Taskforce by the CSA provide an indication of the proportion of payers who are in breach of this legal obligation. Data from the CSA as at 3 December 2004 indicates that nearly 20% of all payers had not lodged a tax return in the last four years. The figure provided for ‘multi-clients’ represents those who either have more than one child support assessment or who are both payers and payees.
According to Table 6.9, less than 60% of payers lodged in all four years between 2000 and 2003.
114 Thus, while most payers do file tax returns, there is a substantial number who are either in serious default of their obligation or who do not file on a regular basis, even when late tax returns are included in the statistics.
Table 6.9: Number of tax returns clients have lodged over the past four years (as at 3 December 2004)
| |
Number of years lodged returns from 1999–00 to 2002–03 |
Total who lodged at least one return |
Total clients with active cases as at 03/12/04 |
% Lodged a return in last four years |
| Client |
1 |
2 |
3 |
4 |
| Payee |
54,454 |
54,959 |
56,157 |
211,011 |
376,581 |
526,323 |
71.5 |
| Payer |
36,608 |
37,521 |
53,820 |
299,133 |
427,082 |
531,562 |
80.3 |
| Multi-client |
20,790 |
19,555 |
19,948 |
78,600 |
138,893 |
204,439 |
67.9 |
| Total |
111,852 |
112,035 |
129,925 |
588,744 |
942,556 |
1,262,324 |
74.7 |
Child Support Agency.
[
top ]
To calculate the child support liability of payers who have not lodged a tax return for the last full financial year the CSA seeks to determine a derived income from other sources. This income is calculated by accessing a taxable income figure for an earlier year and then using an uplift factor account for inflation. It may also be derived by accessing Centrelink data or through information from other Australian Taxation Offi ce systems, or employers. This is the primary default method used by the CSA.
Where a prior year’s taxable income is not available, and there is no other information on which to base an assessment, the CSA will use a default income related to the median income of child support payers.
115 However, that default income may be reduced to zero if the liable parent fails to make any payments.
The Guide, the CSA’s online law and policy guide, provides that:
If a payer has made no payments (either to CSA or to the payee) by the end of a child support period for which the assessment was based on a median default income, CSA will consider amending the assessment to a nil default income. CSA will not amend a default income to nil without contacting the payee and giving them the opportunity to provide any relevant information. If the payee has no information CSA will reduce the income to nil.116
The consequence of this policy is that where a payer both fails year after year to lodge a tax return in breach of his or her legal obligation, then further fails to make any child support payments based upon the income that he or she is deemed to have, the CSA will deem them to have no income at all. The position is then reviewed every six months.
The effect of the current policy is that a payer can avoid the obligation to pay child support by first failing to lodge tax returns year after year, and secondly failing to make any child support payments based upon the CSA’s default assessment. Of the 68,770 child support payers who had a minimum assessment at 30 June 2004 and were: (a) not on Centrelink benefits (b) had not lodged a tax return in the most recent tax year, 13,159 were treated as having an income of nil. This was based on a range of methods of determination, some more reliable than others.
117 One outcome of this policy is that the CSA’s statistical record of success in child support compliance and debt recovery may not accurately reflect (and indeed overstates) the real levels of success in enforcing child support obligations, because this group of non-compliant parents is treated as having no income—and therefore debt of only the minimum liability—in relation to that child support period.
6.8 Second families
The CSCG that recommended the formula for child support in 1988 had a clear view of the relationship between first and second families. It wrote:
118
The fundamental precept of the Consultative Group is that all children of a parent share equally in that parent’s income.
However, that has not been the outcome of the Scheme as it currently operates. The way in which children of second families are taken into account under the current formula (increasing substantially the liable parent’s exempt income before the relevant percentage is applied) is of great benefit to low-income liable parents, but does not provide as much proportionate reduction to those on higher incomes who have new children to support. This is because the flat-rate nature of the increase in exempt income due to a second family represents a much higher proportion of the income of a low-income payer than of a high-income payer.
This can be illustrated by comparing the position of a payer with one new child to a payer without new dependants, both paying child support for two children of a previous relationship (as seen in Table 6.10). The figures given are percentages of before-tax (taxable) and after-tax (net) income.
As a proportion of net income, payers on the higher incomes get between 4.5 and 6 percentage points reduction in their child support for the new dependant, but they are still paying between 12.5% and 15.5% of their net income for each child for whom they are paying child support. By way of contrast, the liable parent with an income of $32,600 has an 11 percentage points reduction in the level of their net income paid out in child support and is paying only 4.5% child support for each child.
Table 6.10: Effect of exempt income amount for second families
| |
No new dependants |
One new dependant |
| Taxable Income |
% of taxable |
% of net |
% of taxable |
% of net |
| $32,600 |
16.3% |
19.9% |
7.2% |
8.8% |
| $52,600 |
20.4% |
26.3% |
14.7% |
19% |
| $72,600 |
22.2% |
30.5% |
18.1% |
24.9% |
| $102,600 |
23.6% |
35.3% |
20.7% |
30.9% |
[
top ]
The proportionate benefit of the allowance for a new dependent child is thus much greater for the lower income liable parent and, in comparison, higher income liable parents don’t see a substantial decrease in their child support liability. This can be perceived as treating the children from the different families unequally. The effect of the allowance for a new dependent child of low-income liable parents is that it takes precedence over the children from the first relationship, whereas the converse is true for higher income liable parents.
The results of the AIFS survey of community attitudes showed people had mixed views about whether a second family should reduce a liable parent’s child support obligation to the first family. As shown in Figure 6.11, the majority in all groups, except the non-resident father group, maintained that fathers should not be permitted to pay less child support if he has a child with another partner (64–68%). Most non-resident fathers, on the other hand, felt that an allowance should be made for such children (62%).
Figure 6.11: Do you think that if the father has another child with a new partner (not step-children), he should be allowed to pay less child support for the children he does not live with?
Smyth B. & Weston R., ‘A snapshot of contemporary attitudes to child support’, in Volume 2 of this Report, p. 46.
Another issue related to second families is the issue of support of step-children. Generally, the Child Support Scheme only takes new biological or adopted children into account in determining a liable
parent’s child support obligation to a child living elsewhere. In some circumstances, courts have been willing to make orders under S.66M of the Family Law Act, providing that a person has a legal obligation to support step-children, but these orders are not common.
119 The Child Support Scheme is predicated on the view that all parents should support their biological children. If the non-resident parent is paying child support, money will be coming into the family where there is a step-parent, and there is no need to take account of the step-parent’s financial support of those children in a way that reduces his or her financial responsibility to his or her own biological children.
As shown in Figure 6.12, the AIFS survey of community attitudes did not show strong support for the inclusion of step-children as relevant dependants for the purposes of reducing a liable parent’s child support obligation. Again, with the exception of non-resident fathers, most respondents in the various groups rejected the notion that a non-resident father should be allowed to pay less child support if he is living with step-children. Half the non-resident fathers believed that step-children should be taken into account, while 41–42% disagreed.
Figure 6.12: If the father has re-partnered and now has step-children to support, should he be allowed to pay less child support for the children he does not live with? 
Notes: GP nonsep = general population non-separated sub-sample; CFC sep =
Caring for Children after Parental Separation sample comprising separated/divorced parents with at least one child under 18; χ2 (6) = 68.38,
p<.001.
Smyth B. & Weston R., ‘A snapshot of contemporary attitudes to child support’, in Volume 2 of this Report, p. 45.
[
top ]
6.9 The need for major change
The identification of these problems does not imply that the original design of the Scheme was deficient. To a great extent, the Child Support Scheme has achieved the objectives that successive governments have given for it over the last 15 years. The Scheme has also been successful in promoting community acceptance of the idea of child support obligations. Indeed for a ‘first generation’ scheme it has proved remarkably durable.
The changes that have been made over the years, although significant, have not involved major alterations to the fundamental design elements of the Scheme. The Scheme has, to a substantial extent, fulfilled its purpose in ensuring that, where possible, levels of child support are paid that provide children with an adequate standard of living and that allow them to benefit from the earning capacity of higher-income non-resident parents. Other aspects of the Scheme, including the administrative system for assessment of child support and for changes of assessment, and the measures put in place for collection of child support, have created a much better system than existed when courts were responsible both for the assessment and enforcement of child maintenance liabilities.
Nonetheless, the issues are significant. In particular, the problem that expenditure on children is not a consistent percentage of before-tax income across the income range is fundamental. While adjustments to deal with these issues could be made to the existing system, the level of change needed is considerable. Consequently, the Taskforce has concluded that redesign of the formula underlying the Child Support Scheme is a better option than making piecemeal changes to the existing model.