The Australian child support reforms: a critical evaluation.

Smyth, Bruce ; Rodgers, Bryan ; Son, Vu 等

Introduction

The Australian Child Support Scheme aims to ensure that children continue to be supported financially should their parents separate or never live together. It currently affects around 1.4 million separated parents and 1.2 million children across Australia. The Scheme--involving administrative (formulaic) assessment, and the collection and enforcement of child support by the Child Support Agency (1) or by private transfer--was introduced in the late 1980s at a time of rapidly increased divorce rates and ex-nuptial births, when many non-resident fathers provided little or no financial support to their children (Edwards et al. 2001). Prior court-based discretionary assessment of child support typically produced low child maintenance amounts that did not adjust for inflation, leading to high levels of child poverty and consequent high costs to the public purse through sole parent pensions.

The original Scheme sought to ensure, among other things, 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 limited to the minimum needed to attain (a) and (b) (Joint Select Committee 1994). In doing so it was hoped that child poverty in sole parent households would be reduced, particularly in the context of government fiscal constraint; there would be a degree of predictability about the amount and timing of payments for sole parents; and the working relationship between parents would be improved by reducing the stress, anger or fear arising from bargaining over money (Carberry 1990).

Not surprisingly perhaps, given the 'intricate economy of love and money in the family' (Millman 1990: 9-10), child support policy is an area fraught with high personal emotion. It is typically tempered by a litany of stakeholders, interest groups, anecdotes, and competing interpretations of what's going on. For many separated parents, child support acts as a 'lightning rod' for much pent-up anger, grief and disappointment surrounding relationship breakdown--including court outcomes--and the loss of everyday family life (Joint Select Committee 1994: 1). Value judgments about what constitutes 'fairness', highly technical legislation and policy rules, and complex interactions between child support and other policies --particularly family payments--add additional layers of complexity. It is this mix of technical complexity, raw emotion, and disparate competing interests that makes child support one of the most contested areas of public policy.

Paying parents--mostly fathers--and receiving parents--mostly mothers differ markedly in their criticism of the Scheme. The most common complaints by payers are that they are paying too much, they shouldn't have to pay child support when they are denied access to their children (i.e., no-see, no-pay), and that child support is essentially a form of alimony. By contrast, the most common complaints by parents eligible for child support are that payments do not cover the costs of raising children, payments do not occur or are irregular, old debts are not pursued, and the system can be manipulated to minimise or avoid child support obligations altogether (Smyth & Weston 2005). These somewhat gendered competing views are likely to be grounded in several realities for separated parents: (a) the money that supported one family is usually insufficient to meet the costs of two newly formed households, at least one of which includes children for much of the time; (b) many separated parents in the child support system are comparatively not well off (Commonwealth of Australia 2014: 6); (c) the costs of supporting children and those of the parent with whom children live for the majority of time cannot easily be distinguished (Weiss & Willis 1985: 270); and (d) differentiating 'capacity to pay' from 'commitment to pay' is not always straightforward.

The fundamental problem for child support policy is that a series of interlocking conundrums pertain to balancing the complex and competing needs of children, resident parents, non-resident parents, and the State (Blumberg 1999; Parkinson 2007). Striking the right balance between adequacy of financial support for children, on the one hand, and fairness for mothers, fathers, and the State on the other, is no easy task. Tweaking one aspect of policy often inadvertently undermines another, making policy development and change 'somewhat akin to the behaviour of an unsolvable Rubik's Cube, but with potentially serious real-world consequences for families' depending on its configuration (Smyth & Henman 2010: 6).

Revising the Australian Child Support Scheme

Australian society has changed in many ways since the Scheme began in the late nineteen eighties: the Australian population has grown, as has the number of separated parents; cohabitation is on the rise; women's workforce participation has increased; so too has men's involvement in their children's lives. Government family related cash benefits have also increased markedly (de Vaus et al. 2014).

In a bid to modernise the Scheme, and flowing out of broader family law reform encouraging shared parenting (Commonwealth of Australia 2005), sweeping changes recommended by a Ministerial Taskforce (2) were implemented between 2006 and 2008. The total package of changes became fully operational on 1 July 2008 when a dramatically different formula for estimating child support liability for all cases registered with the then Child Support Agency came into effect. Key changes to the formula warrant brief mention (for instructive summaries of the changes, see Fehlberg et al. 2015; Parkinson 2008).

The revised formula (a) uses the incomes of both parents after deduction of the same self-support amount for each parent (i.e., an 'income shares' approach) not just the income of the non-resident parent, typically the father; (b) is based on updated research on the net costs of children (i.e., gross costs minus the government contribution through Family Tax Benefit [FTB] (3)), and recognises that parents generally spend progressively more on their children as their income rises, but less as a proportion of overall income; (c) takes into account children's ages --with teenagers allocated higher costs than younger children; (d) caps the cost of children at three children; (e) treats children from first and second families more equally; and (f) reduces the income cap, thereby lowering the amount of child support paid by high-income parents (Commonwealth of Australia 2005).

In addition, under the revised formula, 'regular' contact (where children spend 14 to 34 per cent of nights in the care of the paying parent) is now recognised by a reduction in child support payments to help with the infrastructure costs of regular overnight stays (or eliminating payments altogether where payers are on low incomes but have any of their children between 14 and 34 per cent of the time). The threshold at which a parent with minority care can claim a proportion of FTB was changed from 10 per cent of parenting time to 35 per cent, making the threshold for child support parenting time adjustments and FTB the same.

Furthermore, under the revised formula, payers assessed to pay the minimum payment are required to pay this sum to each child support family in the case of children with different biological parents, up to a maximum of three--in the past, the minimum payment was divided among the families. A 'fixed assessment' measure was also introduced in the form of a fixed annual rate per child ($21.50 per week in 2008 and indexed each year) for payers who reported very low incomes but did not report receiving government income support.

In addition to the new formula, extra resources were provided to the Child Support Agency to ensure that child support is paid in full and on time (ANAO 2009; Smyth, Vnuk, Rodgers & Son 2014). The formula changes and compliance measures--the two central planks of the reforms--are interrelated insofar as a fundamental tenet of child support policy is that a system perceived to be fair is also likely to lead to improved compliance. Until recently there have been little data with which to explore the overall impacts of the child support changes on separated families.

The Child Support Reform Study

The Child Support Reform Study is a large study based on a pre-/post-reform design. It was developed specifically to evaluate the impacts of the Australian child support reforms on separated families. Longitudinal and sequential samples were drawn from the Child Support Agency administrative caseload, which currently represents the best available sampling frame of separated parents with a dependent child in Australia. An initial cross-sectional random sample was selected from separated parents registered with the Child Support Agency before the change in formula for calculating child support came into effect on 1 July 2008. This sample was stratified by (a) time since separation (existing clients separated prior to 1 July 2006 vs those recently separated in the second half of 20 064); (b) level of care (sole care vs shared-time (5)); and (c) method of collection (CSA Collect vs Private Collect). (6) This pre-reform baseline (Time 0) sample yielded 5,046 separated parents (2,809 mothers, 2,237 fathers) who were interviewed using computer-assisted telephone interviews (CATI) of about 25 minutes' duration. Between 20 and 24 months later (Time 1: 2009-10), 3,958 of these respondents were re-interviewed; and of these respondents, 2,927 (1,671 mothers, 1,256 fathers) were interviewed once again after a further 18-25 months (Time 2: 2011). These three points of data collection provide a means of assessing longitudinal change.

A second cross-sectional sample of recently separated parents was similarly selected at Time 1 drawn from those on the Child Support Agency register who had separated in the second half of 2008 (i.e., post-reform). Successful CATI interviews were achieved for 1,000 parents from this Time 1 Supplementary Sample (553 mothers, 447 fathers). This sample was also followed up for re-interview at Time 2 when 799 respondents (427 mothers, 372 fathers) were re-interviewed in 2011, three years post-reform.

A third cross-sectional sample of recently separated parents (those who had separated in the second half of 2009) was drawn at Time 2 (2011). Successful interviews with this Time 2 Supplementary Sample comprised 1,040 separated parents (559 mothers, 481 fathers). The three cross-sectional samples of recently separated families provide a sequential means of assessing change. The timing of various data collections is depicted in Figure 1. (7)

[FIGURE 1 OMITTED]

The Australian child support reforms in many ways constitute a 'natural experiment' insofar as changes in different aspects of post-separation family life (e.g., financial living standards, child support payments, compliance) that occur soon after the reforms might plausibly be linked to those reforms. Of course, other factors or explanations, such as the impact of the global financial crisis, can never be ruled out. The Child Support Reform Study's cross-sequential design is well-suited to exploring 'natural experiments' because longitudinal and sequential data allow changes over time to be examined, while cross-sequential data allow potential policy effects to be identified. (8)

Different samples were used for different analyses. For example, the analysis of policy knowledge was based on: (a) a balanced panel of 2,927 separated parents who participated in all three waves of the study; and (b) sequential data from the three samples of recently separated parents (~1,000 at each wave). The analysis of financial living standards was based largely on (a) a balanced panel of 1,560 separated parents. (9) The analysis of child support compliance was based on (a) a balanced panel of 1,278 separated parents who provided complete information and had a dependent child at all three waves of the study; and (b) data from the three sequential samples of recently separated parents (N=706, 728, and 774). (10) More detailed information on the nature and composition of the various analytic samples is described in: Smyth, Rodgers, Allen and Vnuk (2012); Smyth, Chisholm, Rodgers and Son (2014); Smyth, Vnuk, Rodgers and Son (2014); and Son, Rodgers & Smyth (2014).

Key findings

In this section, we briefly summarise key findings from four distinct lines of inquiry examined thus far: (a) strategic bargaining over child support and parenting time; (b) shared-time parenting; (c) pre- and post-reform financial living standards; and (d) child support compliance.

Strategic bargaining over child support and parenting time

In Australia, there is much anecdotal evidence that separated parents frequently structure their parenting arrangements for financial gain from the child support and family benefits systems. Men's groups, for example, often assert that separated mothers deny fathers overnight stays with children to maximise mothers' child support and social security benefits. By contrast, women's groups frequently claim that separated fathers want more time with children to reduce child support liability. There is no doubt that some parents bargain over money and parenting time to make ends meet, while others do so to maximise their own or their children's interests--financial and/or emotional--in the context of a highly acrimonious co-parental relationship. But widespread strategic bargaining presupposes that separated parents who bargain over child support and parenting time have some knowledge of the child support and family benefits rules (Smyth et al. 2012).

We found that separated parents' level of knowledge of parenting time rules was generally low (Smyth et al. 2012). Specifically, less than ten per cent of separated parents knew the number of nights required for a reduction in child support payments, while less than two per cent of separated parents had accurate knowledge of the number of overnight stays required for FTB to be split between parents. This was true both pre- and post-reform, and for both the longitudinal and sequential samples. What surprised us most was that a sizeable group (14-25 per cent across the study period) of parents thought they had accurate knowledge but this self-perceived knowledge was found to be erroneous: parents who thought they knew the rules were more likely to be wrong than right. Thus any strategic bargaining over child support and parenting time is likely to be occurring in the context of misinformation or a knowledge vacuum.

Policy knowledge is an important area of study in its own right because lack of knowledge about child support rules can create misunderstandings between payers and payees and fuel parental conflict. Some non-resident parents, for instance, might aspire for more parenting time but think they cannot afford it because they are unaware of the parenting time thresholds and the associated reductions in payments; there might also be some resident parents who restrict additional overnight parent-child contact out of concern for the financial implications when such concerns may be unfounded. Moreover, disparities in knowledge between former partners can lead to power imbalances in negotiating post-separation parenting arrangements. Finally, as noted by Nam and colleagues (2009), knowledge gaps can reduce or negate the effectiveness of a policy, program or package of reform.

Our findings on policy knowledge have important implications for program design and implementation. Perceptions influence behaviour and may not accurately reflect the law, policies, and regulations. This becomes acute when schemes are as complex and contested as the Child Support Scheme. There may be room for 'intelligent business systems' that can assist customers and staff in the provision of information to particular groups of clients, and in assessment and negotiation processes (see, e.g., ANAO 2005).

Shared-time parenting

In many countries, shared-time parenting--where children spend at least 30 per cent of time with each parent (i.e., 'joint physical custody')--is emerging as a new family form following divorce or separation (e.g., Melli & Brown 2008). In the United States, around 20 per cent of post-divorce parenting arrangements involve shared-time parenting, although this estimate is as high as 45 per cent in some states, such as Wisconsin (Cancian, Meyer, Brown & Cook 2014; Melli & Brown 2008). Estimates range between 11 and 22 per cent in Australia, Canada, the United Kingdom, Denmark, Norway, and the Netherlands, compared with 33 per cent in Sweden and Belgium (Smyth, Chisholm, Rodgers & Son 2014).

In Australia, the child support changes resonated with the family law changes of 2006, which sought to encourage shared parenting--introduction of the 'regular care' parenting time adjustment is a clear instance of this. But administrative data suggest that the family law and child support changes did not lead to more families entering shared-time arrangements. Rather, such arrangements appear to have plateaued at around 16 per cent of children of recently separated parents since the introduction of those changes (see Figure 2).

There was an initial spike in the proportion of children in shared-time arrangements in new cases after the new child support formula was introduced in 2008. However, this increase disappeared one year post-reform--perhaps reflecting a disappearing 'nudge' effect experienced in many areas of policy reform, where a policy change produces a brief behavioural effect, but then things return to the status quo (Smyth, Chisholm, Rodgers & Son 2014).

Shared-time parenting post-separation tends to be adopted by mainly well-educated, dual-income, cooperative parents with primary school-aged children. These parents typically live near each other, have flexible parenting and work arrangements, and adequate economic resources on which to draw. In the context of the global financial crisis, the tight job market in Australia, and the tendency for poorer families to be disproportionately represented among the separated parent population in Australia, demographic and economic factors may limit the number of families able to manage shared-time arrangements--even where policy is intended to encourage such arrangements (Smyth, Chisholm, Rodgers & Son 2014).

Pre- and post-reform financial living standards

Parental separation is both a leading cause and correlate of child poverty. In Australia, as elsewhere, most studies of the economic consequences of divorce have found that mothers--and their children--are worse off financially than fathers following divorce (e.g., Bianchi et al. 1999; de Vaus et al. 2009; Smyth & Weston 2000). In particular, sole-parent families headed by mothers, and older divorced women living alone typically experience a marked decline in financial living standards after separation (Smyth & Weston 2000).

Other Australian research, however, paints a slightly more complex picture. Kelly and Harding (2005) found that both women and men experienced a drop in living standards after divorce: women can become 'asset rich but income poor', gaining the matrimonial home but struggling with the day-to-day costs of raising children. By contrast, men typically lose the family home and take on greater debt to get by.

But child support is not just about money. Perceptions of fairness are fundamental to the success of any child support system because perceived inequities can undermine compliance. Did the financial living standards of separated mothers, fathers, and their children change following the introduction of the revised child support formula? What about perceptions of fairness towards the Scheme?

Our data suggest that the child support reforms generally led to lower child support payments to separated mothers. (11) Payments decreased, on average, by $20-27 per week after the reforms, but then increased slightly two years later. (Some of the initial reduction, however, was likely to have been offset by increased FTB. (12)) Three years after the reforms, fathers reported paying slightly less child support than they were paying under the old formula.

Not surprisingly perhaps, given the above, separated fathers believed their arrangements for child support were slightly fairer under the new formula than under the old formula, whereas mothers believed child support was slightly less fair under the new formula. But by three years post-reform, separated mothers' and fathers' views of fairness converged: both gave an average rating of 4.7 on a 0-10 scale (with average scores ranging from 4.3-5.0), which does not depict an especially high level of perceived fairness out of 10 (where 0 = 'Totally Unfair', and 10 = 'Totally Fair'). Indeed, perceptions of fairness by separated mothers and fathers were, on average, not especially high at any point.

Consistent with most studies of the economic consequences of divorce, separated mothers were more likely to be worse off financially than separated fathers--both pre- and post-reform. The level of income disadvantage (operationalised as equivalised household income below 60 per cent of the median) experienced by separated fathers (15-17 per cent) remained relatively static pre- and post-reform. By contrast, the prevalence of income disadvantage among separated mothers in the longitudinal sample increased after the introduction of the new formula in 2008 (pre-reform: 20 per cent; one year post-reform: 26 per cent, three years post-reform 24 per cent). Separated mothers were poorer than separated fathers at all three waves.

Curiously, changes to the child support formula appeared to have had little impact over time on separated mothers' subjective financial wellbeing and their experience of financial hardship (15 per cent reported being 'poor' or 'very poor' pre-reform compared with 13 per cent three years post-reform). (13) By contrast, following the child support reforms, the clear overall trend for separated fathers was that they perceived their financial circumstances to be improving over time (19 per cent reported being 'poor' or 'very poor' pre-reform compared with 13 per cent three years post-reform), and they reported that their experience of financial hardship declined by the 12-18 month follow-up (31 per cent reported financial hardship pre-reform, compared with 22 per cent three years post-reform). Earning capacity is one of the most important personal resources on separation, and men's usually higher earning capacity might thus allow them to recover faster than women from the initial economic fallout of separation.

Finally, there was an apparent mismatch between separated parents' objective financial circumstances and subjective financial disadvantage and financial hardship. Despite separated fathers being better off financially than separated mothers (as measured by equivalised household income), they were more likely than separated mothers to report being 'poor' or 'very poor' before the new formula was introduced (19 per cent vs 14 per cent, p v.01), and more likely to report experiencing financial hardship (31 per cent vs 23 per cent, p v.01). This mismatch disappeared post-reform. It is unclear to what extent perceived inequities in the original formula drove this perception, along with the drop in living standards and increased debt for men that typically accompanies separation.

In summary, while the new Scheme initially led to lower child support payments to separated mothers and their children, within three years child support payments almost returned to their pre-reform level. Consistent with most studies of the economic consequences of divorce, separated mothers were more likely to be worse off financially than separated fathers pre- and post-reform. Yet pre-reform, fathers were more likely than mothers to report subjective financial disadvantage and the experience of financial hardship. Fathers' financial circumstances appeared to improve in the medium-term, unlike mothers' circumstances. By three years post-reform, separated fathers' and mothers' perceptions of fairness converged--suggesting a policy gain in terms of perceived equity--but were not especially high.

Child support compliance

A working hypothesis of the Ministerial Taskforce on Child Support (Commonwealth of Australia 2005) was that addressing formula-related inequities for some groups of separated parents would improve perceptions of fairness, which would in turn result in improvements in voluntary compliance. Improving compliance was also intended to increase payees' perceptions of fairness. Of course, changing a formula may be easier than changing (compliance) behaviour, particularly where separated parents are mired in acrimony.

Despite a major investment by government post-reform to ensure that child support is paid in full and on time (ANAO 2009), our data suggest that the child support changes have not led to any significant increases in child support compliance. Specifically, pre-reform, 83 per cent of male payers and 55 per cent of female payees reported that child support was paid in full and on time. Similar rates of compliance were reported post-reform.

Compliance research rarely examines whether changes in compliance are largely due to increases in voluntary payments or due to stronger enforcement--such as routine income withholding and interception of tax refunds, which typically limit non-compliance. We explored this issue by comparing compliance rates between Private Collect cases--where child support is directly transferred between parents without government involvement--and Agency Collect cases, where the DHS-CSP undertakes the collection on behalf of the children and the State. As expected, compliance among Private Collect cases was found to be generally higher than among Agency Collect cases (e.g., female payees' reports for 2008-11: 66-69 per cent vs 42-48 per cent). That said, the child support reforms did not lead to any significant increases in child support compliance among Private Collect or Agency Collect cases.

Caveats

Before concluding, several caveats warrant brief mention. First, disentangling the impacts of the child support reforms from other economic and policy changes occurring around the same time (e.g., the global financial crisis) is not straightforward (Son et al. 2014). Second, a four-year time frame is a relatively short period in policy terms in which to identify change. New policy needs time to bed down. Third, the use of computer-assisted telephone interviews is an especially crude method for the collection of financial data, including child support. Finally, not all separated parents are registered with the Child Support Scheme, and not all parents are accessible by telephone. Our results need to be interpreted with these caveats in mind.

Conclusion

Sweeping changes to the Australian Child Support Scheme were introduced between 2006 and 2008, featuring a dramatically different system for the calculation of child support. Extra resources were also provided to the Child Support Agency to ensure that child support is paid in full and on time.

We found that although the new formula initially led to lower child support payments, and an increase in the proportion of separated mothers experiencing income disadvantage, payments increased slightly two years later. It is curious that a reduction in payments and an increased likelihood of financial disadvantage did not translate into changes in subjective financial wellbeing for separated mothers.

More broadly, the new scheme appears to have resulted in little change in separated parents' policy knowledge, parenting arrangements, perceptions of fairness, and child support compliance. Taken together, these findings suggest that Australia may not have made as much progress as it would have liked in this thorny area of social policy--especially in relation to compliance and perceptions of fairness.

There is currently yet another major inquiry into child support by the Australian Government. (14) The House of Representatives Standing Committee on Social and Legal Affairs is due to table its final report in the second half of 2015. It will be interesting to see whether any new problems are identified beyond the perennial chestnut issues that have dogged the Scheme since its inception, and to what extent the Committee has made use of the wealth of new data available. But of greater interest will be the Committee's recommendations. Parliamentary inquiries raise expectations and offer hope for positive change. The risk in a politically complex and tight fiscal environment--where there is little money and political appetite for major change on controversial social issues--is that problems are not solved and those most adversely affected by a policy become disillusioned. The latter, of course, have real-world flow on effects for children insofar as they are inevitably affected by the wellbeing and circumstances of their parents.

Is too much expected of child support dollars, and of child support policy more broadly? For many families, there is simply not enough money to go around after separation: two households are not as cheap to run as one, and parental separation typically duplicates the costs of raising children (two rental payments or mortgages, two beds for children, separate sets of clothes, and so on). The reality is that there are limits as to how far child support dollars can stretch after separation, even when paid in full and on time.

The Australian child support reforms were developed in very different economic and political times. The Australian economy was strong, and the then Howard government was in a strong position to pass--at times bold--legislation. (15) Increases in the real value of payments to families with children over the past decade or so (de Vaus et al. 2014) have taken some of the pressure off child support--especially in low-income families. But in the aftermath of the global financial crisis when governments are under pressure to reduce welfare expenditure, there is a move back to where parents (still together or separated) --rather than the State--are increasingly responsible for the financial support of their children. For instance, if passed, several measures in the 2014-15 Federal Budget--for instance, the freezing of family benefits, and the tightening of eligibility rules for Family Tax Benefit Part B--are likely to impact adversely on many separated families. Child support will be required to do more of the 'heavy lifting' after separation, especially in middle- to higher-income families, if this policy direction continues. As the Ministerial Taskforce on Child Support (Commonwealth of Australia 2005: 265) reminds us:

   At the heart of the currency of the child support system is its
   capacity to respond to social change. Equally, it must respond
   to legislative change. Alterations to social security, tax or other
   legislation impacting on social policy may change the operation
   of the formula, creating undesired outcomes at points where two
   legislative schemes intersect. The formula must be monitored to
   ensure it keeps pace with these changes.

Post-script

While this article was under review, the House of Representatives Standing Committee on Social and Legal Affairs released its final report (Parliament of the Commonwealth of Australia 2015). The Committee received ~11,300 responses to an online questionnaire, 130 written submissions from organisations, experts and members of the public, and oral evidence from 80 witnesses appearing at public and private hearings. The Committee made 25 recommendations for reform. Detailed discussion of the content and recommendations of the Committee's report is beyond the scope of the present article, and will form the basis of a separate article to be published in the near future.

Acknowledgements

This article draws together key findings from a series of articles from the Child Support Reform Study. This research was conducted with financial support from the Australian Research Council (LP0989558), with additional support from the Australian Government Department of Social Services, and the Department of Human Services. We are especially grateful to the respondents who participated in the study, and to Liz Keogh and two anonymous reviewers for valuable feedback on an early draft. Any shortcomings or errors are ours alone. The views expressed in this article may not reflect those of any affiliated organisations involved in this research. This work is dedicated to the memory of Professor Stephen J. Dowrick, who helped to shape some of the early ideas for the study. His presence and friendship is very much missed at the Australian National University.

Endnotes

(1) The Department of Human Services (DHS) now incorporates the Child Support Agency (or Child Support Program: CSP), which administers the DHS-CSP on behalf of the Australian Government.

(2) One of the authors (Smyth) was a member of the Ministerial Taskforce.

(3) Family Tax Benefit (FTB) is the main form of family payment in Australia. It comprises two parts: FTB-A, and FTB-B. FTB-A consists of a graduated payment linked to the number of children for whom a parent is responsible. Receipt of more-than-base-rate FTB-A results in eligibility for Rent Assistance and several Australian Government concessions (e.g., Health Care Card, and the Medicare Safety Net threshold). To limit government expenditure to the minimum necessary for ensuring children's needs are met, FTB-A is reduced by 50 cents for each dollar of child support above a prescribed threshold, usually $1,544 per annum plus $515 for each child after the first. The other FTB payment, FTB-B, provides additional assistance to single parent and two-parent families with one main income. It is based on the age of the youngest child. Unlike FTB-A, it is not paid per child.

(4) To match the supplementary samples, 1,000 recently separated respondents were sampled, compared with 4,000 existing clients.

(5) 'Shared-time' cases (where children spent at least 30 per cent of nights in the care of each parent) and 'recent separation' cases were oversampled to ensure that there were a sufficient number of cases to make analysis of these groups statistically robust. In this article, 'shared-time' refers to parenting arrangements where children spent at least 30 per cent of nights in the care of each parent. This threshold was used because the survey was conducted before 1 July 2008, after which the 'shared care' threshold for shared parenting time adjustments changed to at least 35 per cent of nights.

(6) To deal with the over-samples and attrition, sets of design, response, and attrition weights were developed.

(7) Fieldwork occurred in the following periods: Time 0: February-April 2008; Time 0 follow-up: October 2009-February 2010; Time 1 Supplementary sample: March-April 2010; and Time 1 Supplementary sample follow-up: September-November 2011.

There are many ways to calculate response rates. Following the approach taken by Data Analysis Australia, we define the response rate as: the ratio of total number of valid responses divided by the total number of respondents approached after excluding 'out of scope' or non-contactable respondents (see http://www.daa.com.au). Response rates were as follows: Time 0 (baseline) sample = 62 per cent; Time 1 = 52 per cent; and Time 2 = 48 per cent; Time 1 Supplementary Sample = 61 per cent; Time 2 Supplementary Sample = 55 per cent. These rates, however, fail to depict the proportion of respondents who successfully completed a follow-up survey (i.e., the completion rate). In total, 78 per cent of participants at Time 1 were re-interviewed at Time 1 Follow-up; 83 per cent were re-interviewed at Time 2 follow-up; and 80 per cent of participants at Time 1 Supplementary Sample were re-interviewed at Time 1 Supplementary Sample Follow-up (i.e., attrition rate = 20 per cent).

(8) In the cross-sequential analyses, those who moved from the old to the new formula, and thus experienced the policy change (the pre-/post-reform group), are compared with those who came onto the system after the child support reforms and thus had their assessments made under the same set of rules (the post-reform only--no policy change--group). A policy effect may be present when changes occur for the former group but not for the latter group.

(9) Four groups of respondents were excluded from the initial balanced panel of 2,927 separated parents: (a) 70 respondents who reported that the focal child turned 18 years at Time 1 or Time 2 follow-up; (b) 482 self-employed respondents, and 573 respondents who reported some 'other' main source of income (e.g., gifts from others, superannuation, etc.); (c) 187 respondents with more than three biological/adopted children from the target union; and (d) 55 respondents living with more than one other adult (e.g., group houses, and multigenerational households).

(10) Respondents whose focal child was over 18 years, or who did not provide information on child support liability, or who reported nil liability on child support were also excluded from each of the sequential samples.

(11) Household income was derived using the question: 'What is your household's total annual income from all sources, including child support for all children in the household?'. FTB was presumed to have been included in this global estimate.

(12) We did not collect detailed information on Family Tax Benefit payments and thus were unable to examine the extent to which FTB offset any changes in child support payments (but see Commonwealth of Australia 2010).

(13) 'Financial hardship' was defined as pawning or selling something, going without meals, or seeking help from welfare or community groups because of a shortage of money

(14) See http://www.aph.gov.au (accessed 5 August 2015).

(15) The Family Law Amendment (Shared Parental Responsibility) Act 2006 (Cth), the move to mandatory mediation (with exceptions), and the introduction of 65 Family Relationship Centres across Australia to support and strengthen families, are good examples (Smyth Chisholm, Rodgers, & Son 2014).

References

ANAO (Australian National Audit Office) (2009) Child Support Reforms: Stage One of the child support reforms and improving compliance, Canberra, Australian National Audit Office.

--(2005) The Edge Project, The Auditor-General Audit Report No.40 2004-05, Performance Audit, Canberra, Australian National Audit Office.

Bianchi, S.M., Subaiya, L. & Kahn, J.R. (1999) 'The gender gap in the economic well-being of non-resident fathers and custodial mothers', Demography, 36 (2), 195-203.

Blumberg, G.G. (1999) 'Balancing the interests: The American Law Institute's treatment of child support', Family Law Quarterly, 33, 39-110.

Candan, M., Meyer D.R., Brown P.R. & Cook S.T. (2014) 'Who gets custody now? Dramatic changes in children's living arrangements after divorce', Demography, 51, 1381-1396.

Carberry, F. (1990) The child support scheme: An evaluation of its personal and social impact. Unpublished Masters Thesis, Melbourne, University of Melbourne.

Commonwealth of Australia (2014) Department of Social Services & Department of Human Services: Joint submission to the House of Representatives Standing Committee on Social Policy and Legal Affairs: Inquiry into the Child Support Program, Canberra, DSS & DHS.

--(2010) Changes to the child support population: Actual Transfers Modelling, Canberra, Commonwealth of Australia, https://www.dss.gov.au (accessed 5 August 2015).

--(2005) In the best interests of children: Reforming the Child Support Scheme Report of the Ministrerial Taskforce on Child Support--Full report, Canberra, Australian Government Publishing Service.

de Vaus, D., Gray, M., Qu, L. 8c Stanton, D. (2009) The financial consequences of divorce for later life, in P. A .Kemp, K. Van den Bosch and L. Smith (eds.), Social Protection in an Ageing World, International Studies on Social Security --Volume 13, Antwerp, Intersentia.

--(2014) 'The economic consequences of divorce in Australia', International Journal of Law, Policy and the Family, 28 (1), 26-47.

Edwards, M., Howard, C. Sc Miller, R. (2001) Social Policy, Public Policy, Sydney, Allen Sc Unwin.

Fehlberg, B., Kaspiew, R., Millbank, J., Kelly, F. Sc Behrens, J. (2015) Australian Family Law: The contemporary context (2nd ed), Melbourne, Oxford Press.

Joint Select Committee on Certain Family Faw Issues (1994) Child Support Scheme: An examination of the operation and effectiveness of the Scheme, Canberra, Australian Government Publishing Service.

Kelly, S. Sc Harding, A. (2005) Fove can hurt, divorce will cost: Financial impact of divorce in Australia, AMP-NATSEM Income and Wealth Report Issue 10, April, Canberra, AMP.

Melli, M.M. Sc Brown, P. (2008) 'Exploring a new family form--the shared time family', International Journal of Law, Policy and the Family, 22, 231-269.

Nam, K., Cancian, M. Sc Meyer, D.R. (2009) 'How program participants learn program rules: Implications for implementation and evaluation', Social Service Review, 83, 53-78.

Millman, M. (1990) Warm Hearts & Cold Cash, New York, The Free Press, Macmillan.

Parkinson, P. (2007) 'Reengineering the Child Support Scheme: An Australian Perspective on the British Government's Proposals', Modern Law Review, 70 (5), 812-836.

--(2008) 'The future of child support', University of Western Australia Law Review, 33 (2), 179-206.

Parliament of the Commonwealth of Australia (2015) From conflict to cooperation: Inquiry into the Child Support Program, Canberra, Parliament of the Commonwealth of Australia, http://www.aph.gov.au (accessed 5 August 2015).

Smyth, B., Chisholm, R., Rodgers, B. 8c Son, V. (2014) 'Legislating for shared-time parenting after parental separation: Insights from Australia?', Journal of Law and Contemporary Problems, 77 (1), 109-149.

Smyth, B. 8c Henman, P. (2010) 'The distribution and financial impacts of the new Australian Child Support Scheme: A 'before and day-after reform' comparison of assessed liability', Journal of Family Studies, 16 (1), 5-32.

Smyth, B., Rodgers, B., Son, V., Allen, L. & Vnuk, M. (2012) 'Separated parents' knowledge of how changes in parenting-time can affect child support payments and Family Tax Benefit splitting in Australia: A pre-/post-reform comparison', Australian Journal of Family Law, 26 (3), 181-213.

Smyth, B., Vnuk, M., Rodgers, B. & Son, V. (2014) 'Can child support compliance be improved by the introduction of a 'fairer' child support formula and more rigorous enforcement? The recent Australian experience', Journal of Family Studies, 20 (3), 204-220.

Smyth, B. & Weston, R. (2005) A snapshot of contemporary attitudes to child support, (Research report no. 13), Melbourne, Australian Institute of Family Studies.

--(2000) Financial living standards after divorce: A recent snapshot, (Research Working Paper 23), Melbourne, Australian Institute of Family Studies.

Son, V., Rodgers, B. & Smyth, B. (2014) 'The impact of child support reform on the financial living standards of separated families in Australia', Australian Journal of Family Law, 28 (2), 193-232.

Weiss, Y. & Willis, R.J. (1985) 'Children as collective goods and divorce settlements', Journal of Labor Economics, 3 (3), 268-292.

Figure 2. Percentage of children in a shared-time
arrangement--DHS administrative data 2002-13

           All cases   New cases

2002-03        6           9
2003-04        7          11
2004-05        8          13
2005-06        8          14
2006-07        9          16
2007-08       10          16
2008-09       10          19
2009-10       11          16
2010-11       11          15
2011-12       11          15
2012-13       11          16

Source: Customised tables supplied by Department
of Human Services-Child Support Program.

Note: Table made from line graph.