12 July 2001




To the Social Services Committee

on the Child Support Amendment Bill 2001




1.      This submission is from Stuart Birks, Director, Centre for Public Policy Evaluation, Massey University, Private Bag 11222, Palmerston North


2.      I wish to appear before the committee to speak to my submission. I can be contacted at: 06-350-5799 X2660


3.      The objective of the Centre for Public Policy Evaluation is “To facilitate the achievement of excellence in research in priority areas and to develop its domestic and international links. The focus will be on economic aspects of policies in a multidisciplinary context.” The work of the Centre has included, among other things, research and publications in the areas of family law and law and economics.


General comments


4.      This Bill presents a piecemeal change to legislation which has larger problems. There appears to be little regard to background information, suggesting that the Bill has been poorly thought through. Any changes should be considered with a view to the broader social implications and the signals given, and in relation to the current review of guardianship, custody and access. The Government appears to be inconsistent in its approach when it calls for a comprehensive review of the latter legislation, while rushing through changes to relationship property and child support legislation.


5.      There are major problems with the Child Support Act which are not addressed by this Bill. In particular, a basic flaw is that the objects of the Act are inconsistent with the formulae. This is discussed in the August 2000 issue of the New Zealand Law Journal. An extract from a version of the article is appended to this submission. Any piecemeal change to the Act is likely to exacerbate these problems.


Raising the minimum level


6.      There may be some justification for this, if the original level is justified. That should be considered in the context of the Child Support Act in its entirety, taking account of both parents’ direct costs and living circumstances, and the role of government policies in determining the amount of contact each parent has with the children. Failing that, the increased payments may be of little or no financial benefit to the affected children, while they may have a significant impact on a liable parent’s ability to have effective contact with the children.


Raising the ceiling


7.      In the Bill, the increase in the ceiling for child support income is explained on the basis that, “The increase will require those liable parents with higher incomes to make a more appropriate contribution towards the costs of raising their children”. No further explanation is given, and it is therefore difficult to know what is intended by the term “more appropriate”. Confusion on the same matter has been evident for some time, suggesting a lack of supporting logic or analysis, as discussed below.


8.      On p.11 of Child Support Review 1994: Consultative Document, it states that:


"The maximum liable income amount is twice the average ordinary time weekly wage. It reflects the maximum cost of keeping a child."


9.      This quote is puzzling because child support is assessed on income less living allowance. Two liable parents, both on the maximum income level, could be paying different amounts of child support because one has dependants and the other does not. In other words, it suggests that the maximum cost to the recipient parent of keeping a child varies according to the liable parent's living circumstances.


10.  The quote is also very worrying. It suggests that a person paying the maximum is paying the whole cost of the child(ren), rather than the cost being shared by both parents. There is also no consideration of the financial burden to the liable parent and savings to the recipient parent associated with the time that the child(ren) may be living with the liable parent. If the maximum liable income amount does represent the maximum cost of keeping a child, then the Act is clearly in breach of several of its objects. Moreover, the proposed increase in maximum liable income would have to be taken to mean that the liable parent in the extended income range would be paying more than the maximum cost of the child(ren). How can it then be claimed that the increased liability is “more appropriate”?


11.  A contradictory description of the relationship between child support liabilities and costs of children is given on p.29 of the same consultative document:


"It is not intended that expenditure be reflected in child support liabilities. Nevertheless, international research shows that expenditure on children remains roughly proportional up to a very high income." (p.29)


12.  If expenditure is proportional to income in a household, it indicates that any increase in the recipient household’s income through higher child support received will be split in a fairly constant proportion between all the children and the adults. One could ask why a liable parent should be subsidising a whole household when only one member of that household might be related. This is particularly relevant in relation to object (c) of the Act - "financial support in respect of these children".


13.  The consultative document’s unsupported claim of roughly proportional expenditure may be incorrect. Expenditure on children as a share of gross family income appears to decline as income rises according to figures 1 and 2 of Harding A and Percival R (1999) “The private costs of children in 1993-94”, Family Matters, 54 (Spring/Summer), pp.82-87.


14.  A recent study in Australia attempted to assess the costs of access for liable parents (Henman P and Mitchell K, “Estimating the costs of contact for non-resident parents: a budget standards approach”, forthcoming in the Journal of Social Policy). Quoting from the abstract: “Costs of contact are found to be high. Where contact is with one child for 20 per cent of the year, the cost of this contact represents about 40 per cent of the total yearly costs of raising that same child in an intact couple household with a medium income, and more than half of the total yearly costs of that child in a household with a low income.”


15.  The Child Support formula does not take these costs into account. While there is some discretion to review an assessment due to high costs of “enabling” access, there is no scope to allow for costs of “enjoyment of” access. This can affect the provision and “enjoyment” of access. As a result, the Act runs counter to the second of Mr Maharey’s aims of the liable parent providing financial and emotional support for children (media statement, 21 June 2001).


16.  Others have suggested that there may be a bias against fathers in relation to child support policy:


“It would help if the application of the child support formula could take better account of the remarriage family situation, including the costs of access visits of non-custodial children, which can be considerable. Talking to some policy analysts about this, I have detected an implicitly punitive attitude towards non-custodial fathers, which is expressed in remarks such as “if a man leaves his wife he deserves what he gets” or “that man’s problem is that he has too many wives”. (Fleming R and Atkinson T (1999) Families of a different kind, Waikanae: Families Remarriage Project, p.159)


Weakness in implementation


17.  There is no guarantee that the Child Support Act is being properly implemented. For example, lawyers are engaged as review officers under the review process. This process requires them to consider claimed income and expenditure patterns by the liable parent and the recipient parent. A proper assessment would involve, for example, determining if the claimed expenditures are realistic, and if benefit entitlements are accurately stated. As the Minister of Revenue stated in the House on 30 May 2000, child support review officers do not receive specialized training to enable them to critically assess claimed income and expenditure levels. Nor is it a requirement for child support review officers to be familiar with benefit entitlements. In other words, these reviews are being undertaken by people who may be lacking the basic knowledge required. From the information earlier in this submission, it appears that policymakers are presenting confused signals as to appropriate information. Inconsistencies between the objects and details of the Act create further problems.


Emotional support


18.  Paragraph 13 above refers to Mr Maharey’s use of the term “emotional support”. A recent US study (see http://www.umich.edu/~urecord/0001/May21_01/5.htm for a brief summary) assessed the time parents were “engaged or accessible” to their children. It found that, in two parent families, fathers provided over 42 per cent of the time. Continuity of pre-separation care patterns would therefore suggest an involvement by both parents of over 40 per cent of nights, the Act’s threshold for shared parenting. The failure of the Act to make any allowance for contact under this threshold level may serve as a major barrier to such arrangements. The study also found that women sole parents were available for an average of 21 hours per week, which is considerably less than the close to 31 hours of mothers in two parent families, and lower than the nearly 23 hours by fathers in two parent families. This suggests that the Act, based primarily on a sole parenting model and not considering liable parents’ costs, may well be serving to limit the emotional support of children whose parents live apart.


Broader social implications


19.  The Bill proposes to raise the ceiling child support income to approximately $85,000. Consider a family man earning an income approaching this level. He risks losing home and family at the whim of his partner. Under the recently passed Property Relationships Act he would get far less than half the matrimonial property due to his earning capacity.


20.  Of his take-home pay, if he had a student loan, he could be paying over 45 per cent in child support, leaving him in the position of someone on an income of about $37,000, but with the heartache of being distanced from his children. Out of an extra dollar earned he might see less than 15 cents following deductions. Child support payments could continue for 19 years. He has no say in how the money is actually used, but is expected to believe that it is for his children. For one child, he could be paying $1100 per month from his after tax income. This is about $256 per week, or 16 per cent more than the amount given for both a child and a parent on the Domestic Purposes Benefit. The amount is not affected by the mother’s income, or by his time with the child, even up to 145 nights a year.


21.  Meanwhile he can expect no support from the government in his attempts to maintain a relationship with his children. What message is this giving our sons about study, hard work, building a career, and long term planning? What does it lead our daughters to expect from men? What basis is this for our children to form relationships and create an environment in which to bring up their children?




22.  That there be no change to the Child Support formula at this time.


23.  That the Child Support Act be reviewed in conjunction with other aspects of family law.


24.  That special attention be paid to the broader social implications of the legislation and, in particular, the signals that are being given to the nation’s children.




Extract from: Objects and Fairness in Family Law


By Stuart Birks


Centre for Public Policy Evaluation, Massey University, Palmerston North


8 June 2000




Equity is a nice term. Outcomes are more likely to be acceptable if they are considered equitable. It is a powerful term, but its precise meaning is hard to specify.


Economists refer to horizontal and vertical equity. Horizontal equity relates to like circumstances, under which equitable outcomes require like treatment. Vertical equity refers to differing circumstances, for which an "appropriate difference" in treatment is required.


"Like circumstances" depend on the degree of detail considered - should one look at current income, or lifetime income, or how the income is earned, should one also consider the number of dependants, their ages and specific needs? If there are problems with the concept of horizontal equity, there are even greater problems with vertical equity. "Appropriate differences" in treatment depend on the variables used and the values placed on them. Should time with children be considered? How should it be measured - hours, days, nights? What costs should be included? What effect would specific differences have on treatment?


Discussion of disadvantage and discrimination hinge on the same points - what variables are selected to draw comparisons, what are "appropriate differences" and what values are assigned in terms of specifying the significance of a difference and whether it is considered beneficial or detrimental to a particular group. Policy debate frequently revolves around differences in choice of and interpretation of variables. For example, current proposals for unequal splitting of matrimonial property are based on the monocular consideration of caregivers' assumed sacrifice of earning capacity, disregarding numerous other dimensions. Resulting policy, often specified in legislation, can be confused and inappropriate.


Child Support


Section 4 of the Child Support Act 1991 lists its objects. It is stated on the IRD Child Support web page that: "Child Support is governed by the objectives set out in the Child Support Act 1991".[1]


Child support is paid by a liable parent to a custodial parent. Both parties are assessed in a shared parenting situation, with a net payment going from one to the other. I shall only consider the formula under sole custody. Key aspects of the basic child support formula in a sole custody situation are as follows:


        Child support income is equal to taxable income up to a maximum level, now $68,436.00.


        A living allowance is deducted from this as follows:


Description Living Allowance
Single with no dependent children


Married or defacto with no dependent children


Single, married or defacto with one child living with the paying parent


Single, married or defacto with two children living with the paying parent


Single, married or defacto with three children living with the paying parent


Single, married or defacto with four or more children living with the paying parent



Child support is equal to the balance times a percentage rate, 18 percent for one child, 24 for two, 27 for three and 30 for four or more children.


        The custodial parent's income is not considered.


        The liable parent's time with the children is not considered unless it includes at least 40% of nights.


        There is some scope to deviate from the formula through an administrative review or hearing, but most applications by liable parents are declined.


        If a custodial parent is on the DBP, child support payments go first to offset that payment. Any additional child support is passed on to the custodial parent.


When the objects of the Act are compared with the formula, several inconsistencies can be observed. Neither the custodial parent's income nor the liable parent's time with the children are taken into account. Vertical equity cannot generally be achieved. If the formula assessment is correct for one level of custodial parent's income, then it is incorrect for others, and similarly for time with children. There are also problems with payments not being linked to expenditure on children, and with both parents potentially being caregivers for some of the time. To consider specific objects in the Act:


        Object (a) is To affirm the right of children to be maintained by their parents, but the act does not ensure that payments are made by both parents, nor does it ensure that payments actually benefit the child(ren).

        Object (b) is To affirm the obligation of parents to maintain their children, but the act focuses only on "liable parents".

        Object (c) is To affirm the right of caregivers of children to receive financial support in respect of those children from non-custodial parents of the children, but non-custodial parents could be the caregivers for up to 40 percent of nights with no effect on child support obligations.

        Object (d) is To provide that the level of financial support to be provided by parents for their children is to be determined according to their capacity to provide financial support, but only liable parents are required to provide, and the other parent's circumstances are generally not considered.

        Object (e) is To ensure that parents with a like capacity to provide financial support for their children should provide like amounts of financial support, but under the formula this only applies for comparisons between liable parents. The other parents' circumstances are generally ignored.

        Similarly, object (f), To provide legislatively fixed standards in accordance with which the level of financial support to be provided by parents for their children should be determined, refers to parents, but the formula only refers to payments by liable parents of money received by the state or by custodial parents.

        Object (h) is To ensure that equity exists between custodial and non-custodial parents, in respect of the costs of supporting children, and is commonly not met because only the circumstances of the liable parent are considered.

        Object (j) is To ensure that the costs to the State of providing an adequate level of financial support for children and their custodians are offset by the collection of a fair contribution from non-custodial parents, but can it be a "fair" contribution when there is no change in contribution as the liable parent's time with the children rises from 0 percent to 40 percent of nights?

In other words, the specifics of the law are inconsistent with its objectives.


The law commenced on 18 December 1991. There has since been a review headed by Judge Peter Trapski.[2] Not only was there no mention of these inconsistencies, but the Act also appears designed to achieve other unspecified labour market objectives. On consideration of the custodial parent's income the Consultative Document states: "a strong disincentive to workforce participation could result if every dollar earned by the custodian over a given threshold resulted in a decrease in child support. As 84% of lone parents are women, structural gender based inequities in the labour market could be worsened." (p.24)


The use of nights as a measure is also problematic. It is explained in Butterworths Family Law in New Zealand as follows:


The choice of 40 per cent of nights might at first appear curious but it is probably explained by the fact that most children spend a great part of the day time in school and most of the parental care is later in the day and in the early morning with sleep in between. [3]


Young children are not at school, but for those who are there can be a big difference between custodial and non-custodial parents’ time with them. Non-custodial parents would generally care for children at weekends and school holidays, when contact time and associated expenses can be much greater.


It could be suggested that anomalies can be allowed for through the review process. However review officers are lawyers and may be equally unaware of the problems. In fact some aspects, such as the “costs of enjoyment of access” for the liable parent, are expressly ignored. S 105(2)(b)(i) of the Act refers specifically to the costs of "enabling access" only.


Rather than ignorance, disregard for the inconsistencies may be tacitly accepted because the status quo is considered desirable. Benson describes how, in Commissioner of Inland Revenue v Aspinall [1999] 3 NZLR 87, certain objects were considered and others ignored.[4] Benson is concerned about the decision in the case. In fact he highlights a more serious problem, namely the judiciary's adoption of a flawed line of reasoning. I have referred elsewhere to judges selecting, as convenient, from a “menu of principles” to support of their preferred outcome.[5] Benson describes this process also. The result is a post hoc rationalization for what may simply be the whim of the judge in question. Judge Boshier has actively encouraged lawyers to "push these boundaries".[6] This covert form of judicial activism can be very harmful given the judiciary’s limited information and understanding of the wider social implications of their actions. We see this in operation elsewhere in family law.




[1] http://www.ird.govt.nz/childsupport/csa.htm#legislation

[2]   See Child Support Act Working Party (1994) Child Support Review 1994: A Consultative Document, and Trapski P, et al. (1994) Child Support Review 1994: Report of the Working Party, Wellington: Inland Revenue Department

[3]    Butterworths (1995) Butterworths Family Law in New Zealand, 7th edition, p.294

[4]  Benson S (2000) "Departures from Child Support Assessments", New Zealand Law Journal, May, pp.176-179

[5]   Section 4.2 of Birks S (1998) Gender Analysis and the Women's Access to Justice Project, Issues Paper Number 2, Centre for Public Policy Evaluation, Massey University (http://econ.massey.ac.nz/cppe/papers/waj1.htm); section 2.8 of Birks S (1998) The Family Court: A View from the Outside, Issues Paper No.3 Centre for Public Policy Evaluation, Massey University (http://econ.massey.ac.nz/cppe/papers/cppeip03.htm); section 3.4 of Birks S (1999) "Parenting and the Family Court: An Economists's Perspective", in Birks S and Callister P (eds.) (1999) Perspectives on Fathering II, Issues Paper No.6 Centre for Public Policy Evaluation, Massey University.


[6]  Boshier P (1998) "Developments in Matrimonial Property", Family Law Conference, 31st August – 2nd September 1998, Christchurch, New Zealand Law Society, pp.51-69.