Research report no.13 2005

A snapshot of contemporary attitudes to child support

by Bruce Smyth and Ruth Weston

1. Why study attitudes to child support?

A core concern for most developed countries continues to be how to foster the ongoing support of children following parental separation or divorce. Over two decades ago, as divorce rates and ex-nuptial births rose, child support became a pressing issue. Most non-resident parents (usually fathers) were providing little, if any, financial support to their children, with consequent high levels of child poverty and high costs to the public purse (Harrison and Tucker 1986; McDonald and Weston 1986).

In Australia, the Child Support Scheme was introduced in the late 1980s to tackle child poverty and to shift the onus of responsibility for financial support back onto parents. Central to the Scheme is the administrative assessment of child support liability via the application of a child support formula, removing the need for parents to have recourse to court-based discretionary assessment, which typically produced low child maintenance amounts that did not adjust for inflation.1

The fundamental shift brought about by the Scheme was that the registration of a child support liability with the Child Support Agency converted 'a personal obligation into a debt owed to the Commonwealth' (Harrison 1994: 178) that was collected by the Child Support Agency. Through the elimination of the need for private enforcement action, the Scheme sought to improve the working relationship between parents by reducing the stress, anger or fear often associated with 'bargaining' over money (Carberry 1990, 1992). It also sought to offer a degree of predictability and certainty about payments with respect to their amount, regularity, and timing, and to promote parent-child contact in the hope that 'where money goes, the heart goes'.

Specifically, the Scheme was designed to ensure that: (a) children of separated or divorced parents receive adequate financial support; (b) both parents contribute to the cost of supporting their children according to their respective capacities to do so; and (c) government expenditure is restricted to the minimum necessary to attain these objectives (Joint Select Committee 1994). The Scheme also seeks to avoid work disincentives for parents, and to be 'simple, flexible, efficient' and nonintrusive in its operation (Child Support Evaluation Advisory Group 1992: 67-68).

In support of these objectives, at least six carefully considered principles underpin the Scheme.

First and foremost, each natural (or adoptive) parent has a duty to support his or her own children. Where possible, this responsibility should thus not be passed onto other adults or the government.

Second, all non-resident parents regardless of their income should pay something towards the financial support of their children.2

Third, children whose parents have separated should themselves have a standard of living that reflects that of both their parents, not just the one with whom they usually live.

Fourth, the amounts assessed as owed should seem 'reasonable' (for instance, given economies of scale, simply doubling the payment for each additional child is unlikely to be perceived as 'fair').

Fifth, no contingent link should be made between the occurrence of child support payments and parent-child contact because such a link is unlikely to be in children's best interests.3 Thus payment of child support should occur regardless of whether parent-child contact is possible, and vice versa.

Finally, where parents enter into new relationships and have new family responsibilities, children of the past relationship ('first children') should be given special policy consideration. This is because family dynamics are such that (new) children in a household typically receive the lion's share of income flowing into that household while children living elsewhere may be 'out-of-sight; out-of-mind' (Joint Select Committee 1994).4

These and other principles also underpin child support schemes elsewhere, although with varied emphasis - sometimes markedly - within and across countries. (The United States is a policy microcosm in this regard, with different states having different policies.)

Recent policy challenges

The fundamental problem for child support policy world-wide is that a series of interlocking conundrums exist in relation to balancing the complex and competing needs of children, resident parents, non-resident parents, and the State (see Blumberg 1999 for a good precis of these competing interests). While most child support systems are now well established, they nonetheless continue to be reworked in a bid to improve the balance between 'adequacy' and 'equity', and to take into account changing social trends.5

In recent years, much of the policy refinement has centred on equity issues through the inclusion or refinement of 'special factors' - most notably financial adjustments for shared care, second family responsibilities, and low or high income - or the broadening of departure provisions (Venohr and Williams 1999). These refinements seek to make formulaic assessment flexible enough to take account of the ever-increasing complexity of social reality.

But tweaking has its limits. Indeed, there is mounting criticism - in the United States context at least - that first-generation child support models (tweaked or not) contain fundamental flaws that can result in child support payments being unjustifiably low or high (see, for example, Harris 1999; Ellman 2004a; McNeely and McNeely 2004).

This criticism has led to the development of second-generation child support models, of which the American Law Institute's (2003) Principles is at the vanguard. Unlike first-generation models which typically have competing values and interests embedded deep within their conceptual framework, proposed second-generation models take as their starting point the explicit identification of these interests and then attempt to develop a framework for harmonising them (Blumberg 1999; Harris 1999).6 In doing so, these models purport to be able to strike a more defensible balance between adequacy of financial support for children, on the one hand, and fairness for mothers, fathers, and the State on the other (Harris 1999).

Attitudes as a window into 'fairness'

Regardless of which model - or generation of models - is adopted, three axioms hold: first, difficult choices and 'trade-offs' will always need to be made in relation to balancing the various competing interests in any model (Bassi and Barnow 1993; Betson, Evenhouse, Reilly and Smolensky 1992); second, assessment of a model's adequacy and fairness will always require some level of subjective assessment of 'what's fair'7; and third, a child support system is unlikely to be effective unless it is perceived to be reasonably 'fair' for everyone since laws applying to parenting require reasonably wide acceptance for their compliance and success (Garrison 1998; Lin 2000; Minow 1998).8 These axioms highlight the importance of considering community perceptions, values and expectations (including those of separated parents) in thinking about child support policy.

At the same time, there is a danger in placing too much emphasis on community attitudes in the shaping of policy, particularly in cases where the rationale behind policy decisions entails a complex set of issues that may not be well understood. The area of child support may be a case in point. In addition, there is an emerging literature suggesting that attitudes about what is fair are often influenced by self-protective concerns and/or concerns about friends or kin, or groups with which one identifies, rather than about other groups or the community at large (see Clayton and Opotow 2003). Finally, history is replete with examples where group behaviour consistent with community attitudes and values has led to morally suspect outcomes (for example, cutting off the hands of thieves).

Nonetheless, attitudes are a lens to understanding a community's sense of fairness. Critically, child support policy raises fundamental issues about 'fairness' (Garrison 1998; Minow 1998). A perceived lack of fairness may suggest a need either to change policy or to make the rationale behind the policy more transparent and well known. Either way, tapping community attitudes and values about fairness is an important aspect of policy development.

Prior work

Studies of attitudes to child support have been undertaken in a number of countries. In the United States, where most of the early work has been conducted, complex multi-factorial vignette designs have mainly been used.9 This approach has sought to assess how much child support should be paid, and the extent to which different family-related circumstances influence these estimates (see, for example, Bergmann and Wetchler 1995; Corbett, Garfinkel and Schaefer 1992).

Schaeffer (1990), for example, found that respondents favoured: the level of child support being based on capacity to pay; the use of both parents' income in estimating child support liability; and a reduction in payment liability where resident mothers had repartnered. Moreover, Ganong and Coleman (1999) found that beliefs about 'fairness' guided respondents' thinking on child support. However, respondents differed on whose fairness should be considered - resident mothers' and children's, non-resident fathers', or all family members'.

In the United Kingdom, two major studies of attitudes to child support have recently been conducted.10 White (2002) analysed data from several large British longitudinal surveys in an attempt to benchmark public attitudes to child support. Among other things, she found considerable community support for the principles that: (a) the main responsibility for child support should lie with parents not government; (b) non-resident parents (male or female; rich or poor) should always pay child support; (c) payments should reflect capacity to pay; and (d) most mothers on government income support should be able to keep some or all of the child support received.

However, White also found that respondents tended to be evenly divided about whether there should be an upper ceiling on the amount of child support paid by high-income fathers, and whether overnight stays should be taken into account in child support. Nonetheless, while most women and men in general maintained that a father's new family responsibilities (involving children born of the new relationship) should not affect their child support liabilities for first-family children, this was more commonly expressed by the women. Furthermore, non-resident parents were more likely than resident parents to believe that the denial of contact should reduce child support payments.

Building on White's (2002) work, Peacey and Rainford (2004) conducted a study of attitudes to child support using essentially the same set of questions and research design. Much the same pattern of results emerged. These results were interpreted as offering broad support for the aims and principles of the UK Child Support Agency.

In Australia, the Child Support Agency has conducted several attitudinal surveys.11 In 1995, the Agency investigated clients' perceptions of its service. It repeated this survey in 1996 and 1997 (Child Support Agency 1997). Around 2,300 clients were interviewed each year by telephone. The study found, among other things, that issues related to natural justice ('treating both sides equally') appeared to be of greatest concern to payers (mostly fathers), whereas enforcement issues were highly salient to payees (mostly mothers) (Child Support Agency 1997: 5-7).

In 1998, the Agency surveyed the general Australian community about its perceptions of the Agency and the Child Support Scheme. It repeated this survey in 1999 and 2000 (Child Support Agency 2001a). The Agency found that virtually all respondents agreed with the idea that both parents should support their children financially after separation; and most respondents believed that all non-resident parents should pay at least $5 a week to support their children (see also Child Support Agency 2001b). It also noted that males were more likely than females to be critical of the formula and administrative issues.

The Australian Institute of Family Studies also found widespread public support for the idea that both parents should provide financial support for their children after separation (Funder and Smyth 1996), although one study suggested that divorced men were more likely than divorced women and their children to believe that child support should cease once children turn 18 (Smyth 2000). In a more recent survey, over half (55 per cent) of the resident mothers reported being dissatisfied with the amount of child support paid, compared with around one third (36 per cent) of nonresident fathers. Those who were most dissatisfied were: resident mothers who reported father-child contact occurring but without any child support being paid (79 per cent of resident mothers in this group reported dissatisfaction); and non-resident fathers who reported paying child support but who had no contact with children (68 per cent of fathers in this group reported dissatisfaction). Not surprisingly, those who were the least likely to express dissatisfaction with child support payments were separated parents who reported the co-occurrence of father-child contact and child support (46 per cent of resident mothers and 28 per cent of non-resident fathers).12

In summary, it would seem that there is considerable overlap between the views of the general population in the United States, the United Kingdom and Australia on several key principles on which each scheme rests.

The need for an up-to-date Australian snapshot

Obtaining a contemporary picture of attitudes to child support has particular relevance in Australia at present. Flowing out of its 2003 inquiry into 50/50 shared care after separation, the Standing Committee on Family and Community Affairsconcluded that the Australian Child Support Scheme 'has serious flaws and produces inequities for a high number of payees and payers' (Commonwealth of Australia 2003: 174-175). The Australian Government recently established a Ministerial Taskforce on Child Support to conduct a comprehensive re-evaluation of the Scheme (Anthony 2004). The Institute's Attitudes to Child Support Study, reported here, was conducted to help inform the Taskforce's review. An awareness of attitudes can also provide a national benchmark of attitudes to child support against which comparisons can be made in subsequent years in the event of any major policy reform.

This report examines attitudes to several pressing child support policy issues. Should parents on low income be exempted from paying any child support? Should there be a penalty (such as a fine) for non-payment? Should the $5 per week minimum amount of child support be increased? Should parents whose children are taken overseas have to keep paying child support? Should (wealthy) parents have a cap on how much child support they pay? Should child support take account of parents' set-up costs for housing? Should child support be based on gross or net income? Should non-resident parents with new obligations to second families have their child support payments reduced?

Structure of the report

Following a brief overview of the study's design (Chapter 2), attitudes to some of the broad principles of the Scheme are presented in Chapter 3. Chapter 4 then examines the views of respondents about the extent to which second families and parent-child contact should be taken into account in child support calculations. In Chapter 5, respondents' comments about changes they would like to see to the Scheme are summarised. The final chapter synthesises and discusses key findings, and sets out several policy implications arising from the data.


  1. The formula is currently expressed as a percentage of the non-resident parent's gross taxable income after a 'self-support' component has been deducted. The self-support component is increased where the non-resident parent has any natural or adopted children in his or her care. The formula is based on the number of dependent children under the daily care of the other parent (18 per cent of taxable income for one child, 27 per cent for two children, 32 per cent for three, 34 per cent for four, and 36 per cent for five or more children). Currently, all nonresident parents are expected to pay at least $260 per year (that is, around $5 per week). There is a ceiling for the non-resident parent's child support income which is set at 2.5 times average weekly total earnings for persons in Australia. Further, the income of the resident parent is taken into account when it equals or exceeds an aggregate of average weekly earnings plus a supplementary amount related to caring for children. The resident parent's income is also taken into account where each parent has the children for at least 30 per cent of nights.
  2. This principle was added to the Scheme relatively recently (1 July 2001). A $260 annual minimum assessment (that is, $5 per week) currently exists.
  3. However, the 'high' costs of substantial contact involving overnight stays (30 per cent of nights or more per year) reduce child support liability.
  4. See footnote 1 for a description of how these principles are operationalised under the Scheme.
  5. The notions of 'adequacy' of support for children versus 'equity' between households are discussed by Eekelaar (1988, 2000).
  6. The extent to which these different models can achieve this goal remains to be seen since none has yet been adopted.
  7. A basic axiom of family economics is that there is no single absolute 'cost' of a child: 'children's 'needs' above subsistence level vary with the income and values of the parents' (Harris 1999: 717).
  8. The basic models include: percentage-of-obligor-income (in which only the income of the non-resident parent is taken into account); income shares (in which both parents' incomes are taken into account); Melson/Delaware (which is a form of income shares but includes a selfsupport component for parents); income equalization (in which the main aim of child support is to equalise the financial living standards of both parents); and cost shares (in which the costs of children are estimated and then pro-rated according to both parents' earnings). Hybrid models also exist, of which the Australian Scheme is a good example (see footnote 1).
  9. At least 37 public opinion polls have been conducted in the US over the past 30 years (see also Garrison 1998: 99, note 247). These polls typically comprise a single question focusing on child support enforcement strategies (Dialog - POLL database, searched 9 January 2005).
  10. Earlier studies include those by Burgoyne and Miller (1994) and Kiernan (1992).
  11. While several parliamentary inquiries have received numerous submissions in relation to child support (for example, Joint Select Committee 1994; Commonwealth of Australia 2003), it is unclear to what extent these views are indicative of broader community sentiment.
  12. These results, derived specifically for the present discussion, are based on Wave 1 of the Australian Institute of Family Studies' Caring for Children after Parental Separation Project, conducted in 2003.

Research Report 13: Contents | Next |Previous