Sydney Law Review
Ours is a nation in which the family unit is the cornerstone of our lives and where it is the family’s responsibility to raise children to be strong, responsible and independent individuals.
Prime Minister Howard
I ... argue that we leave our aspirations for the traditional family form behind and reimagine what should be our core family connection.
Professor Martha Fineman
The family has lately taken on a new significance in tax law and policy discourse in Australia. It has been a primary focus of all participants in the debate about the tax reform proposed by Prime Minister Howard in the package A New Tax System, which became law on 8 July 1999. The Government’s pitch for the tax reform package is tied to the consequences for families of the Goods and Services Tax (GST) and of the accompanying income tax cuts and transfer reforms, which are aimed both at compensating families for the GST and at improving the economic position of lower- and middle-income families. The Opposition and commentators have also focused on families in their criticism and analysis of the package. Along a different, but related, trajectory, there has recently been a strong push for social and legal recognition as families by lesbian, gay and transgender activists. This has resulted in proposals for reform of tax and transfer laws to extend the meaning of family to cover same-sex relationships, the most significant of which is contained in the Sexuality Discrimination Bill 1996 (Cth), introduced by the Democrats. This essay examines the meaning of family in Australian income tax and transfer laws and policy. I discuss the discursive construction of a ‘regulatory ideal’ family through tax and transfer laws – a married nuclear family with dad as the breadwinner/head of household and mum as homemaker. I then analyse how this ideology of the family has domesticated reform proposals concerning the family in tax and transfer law, including gay and lesbian proposals.
I include as tax and transfer laws the Income Tax Assessment Act (ITA), and A New Tax System (Family Assistance) Act 1999 (the Family Assistance Act), the last of which establishes a statutory regime of family benefits separate from but linked to the ITA and SSA. These are public laws through which the state seeks to extract, allocate and (re)distribute income in society in particular ways. They do not directly regulate the family, but participate in the production of specific relationships as normal or legitimate, to the exclusion and detriment of other relations. They have real economic consequences for people and they may (intentionally or unintentionally) have economic incentive or disincentive effects that tend to modify the behaviour of taxpayers and transfer payment recipients. Tax and transfer laws may also have an excessive symbolic effect, seen in the creation, for example, of the stigmatised category of welfare mother, and in the rhetorical and political significance contained in the very notion of the taxpayer. As Lisa Philipps has observed, ‘lower income people are constructed as non-taxpayers or minimal taxpayers, whose views therefore must not be representative of the true tax mandate set by mainstream society’. This is one reason why it is important to consider tax and transfer systems together; another is that the role of transfer laws in producing a regulatory ideal family has been carefully theorised by feminist scholars and activists, who have also engaged with the state to produce significant changes in the meaning of family in transfer laws. The ideal family constituted through the tax system and the ideal family constituted in the transfer system are not identical, but they are inextricably linked.
The discursive production of the ideal family in both tax and transfer laws takes place through the allocation of benefits and burdens based on implicit presumptions of the meaning of family, or specific relationship categories built on such presumptions. The ideal family is produced and reinforced through the ‘normal’ operation of the tax law as a revenue raising instrument and is thereby implicated in the distribution of the tax burden. In addition, it has long been understood that the tax system does not merely raise revenue but ‘is also a massive spending program’: tax concessions can be considered as functionally equivalent in many respects to transfers, that is, as expenditures. Consequently, the tax system also produces the ideal family through tax expenditures aimed at families. Transfer law is generally understood to be a mechanism for government spending. As important, however, is the role played by transfer law in restraining government costs through the privatisation of care. The ideal family produced through the transfer law is crucial to this less obvious project.
Existing tax and transfer law and policy discourses have domesticated both queer and feminist tax reform proposals around the family. By domestication, I mean a subtle political process by which law reform and critique is contained and controlled. In discussing domestication of law reform specifically pertaining to the family, I am consciously invoking the different meanings of domestication, including:
[T]he relegation of women to the domestic sphere, a private place that can facilitate being dominated and inhibit collective action. It also signals that one’s potential has been circumscribed to the service of another ... [it] can be used to describe what has occurred when the views of the dominant culture – in this case the legal culture – are so internalised that they are considered common sense.
The current tax-transfer law and policy framework effectively confines reform proposals to the ideal family model, foreclosing alternative visions of family. This effect is obscured by the technical nature of the law and policy discourse and by its widely accepted claim to neutrality.
In Part 2, I explain my theoretical approach to the law’s role in the construction of the ideal family and my use of the terms discourse and ideology. Parts 3 and 4 describe and analyse the categories and classifications that produce the family in tax and transfer laws, including A New Tax System. I do not analyse the GST itself but focus on the ‘compensatory’ income tax and transfer reforms. In Part 5, I describe and critique two recent law reform proposals that would incorporate gay and lesbian couples into the definition of family for the purposes of tax and transfer laws. These proposals, while extending equal treatment to lesbian and gay couples, leave the disciplinary framework of the ideal family intact and ignore feminist critiques of the family tax unit. However, they highlight the heterosexist assumptions that limit most feminist critiques of tax and transfer policy. I conclude with some suggestions for revisioning dependency, caregiving and family in tax and transfer laws, taking inspiration from the work of Martha Fineman on the family in United States welfare law.
The Family Court shall ... have regard to:
the need to preserve and protect the institution of marriage as the union of a man and a woman to the exclusion of all others voluntarily entered into for life; the need to give the widest possible protection and assistance to the family as the natural and fundamental group unit of society, particularly while it is responsible for the care and education of dependent children. s43 Family Law Act 1975 (Cth).
There are certain institutions in our community that provide it with bulwarks and stability, and marriage is one of them ... attempts to give the same legal status to homosexual relations as marriage are not things I would support. Prime Minister Howard
Law is a discourse, by which I mean a body of specialised knowledge, language and method. Power operates through discourses such as law, to produce knowledge or truth; thus, truth is an effect of power and does not stand ‘outside’ power, and power is productive and creative, not merely repressive or negative. I use the notion of ideology posited by Terry Eagleton as a ‘set of effects within discourses’. Ideological effects within discourses include ‘closure’, by which ‘certain forms of signification are silently excluded and certain signifiers ‘fixed’ in a commanding position’. Ideology is linked to discourses through the exercise of power. To illustrate, the Family Law Act 1975 (Cth) posits the married nuclear family as ‘natural’ and ‘fundamental’ to society. This ‘truth’ about the family is an ideology, produced and sustained in this instance through legal discourse (the same ‘truth’ is also produced through discourses such as biology, religion, anthropology and economics). The law does not merely regulate a pre-existing ‘natural’ family but by normative reiteration through different acters and mechanisms, participates in the production of the regulatory ideal:
legal norms give behaviour a significance not inherently present in that behaviour. Through its power of signification, the law participates in the formulation of notions of “good” and “bad”, “normal” and “other”, “black” and “white”.
The ideal family in legal discourse is an ‘assumed institution’ with ‘a well-defined, socially constructed form complete with complementary roles – husband/head of household, wife/helpmate, child’. In producing this ideal family, law necessarily excludes other relationships, for example, lesbian or gay relationships, young unmarried mothers raising their children on welfare, or extended Aboriginal households, from qualifying as family and in this process, constructs them as dysfunctional, failed, abnormal or deviant.
To say that law is a discourse does not mean that it is unified; indeed, it ‘operates with conflicting principles and contradictory effects at every level’. Tax and transfer laws are sub-categories of legal discourse, but they differ from other laws in their close association with economics and public finance theory, technical discourses that makes powerful claims to knowledge and truth, ‘asserting their own neutrality while at the same time advancing particular norms about how society should be ordered’. Tax and transfer laws build on the ideal family constructed through other laws. However, their role in allocating and distributing income means that they reproduce the ideology of the breadwinner-homemaker family in a specific way that may cement economic inequalities. Tax and transfer laws may also sometimes operate to contradict other legal discourses around family, and may themselves conflict and be internally inconsistent.
The most obvious way in which law produces the ideal family is through the category marriage, although clearly the married family of legal discourse does not uniformly or consistently reflect everyday social existence, as the ordering of family relations ‘is reworked by ordinary people to meet the contradictory demands of everyday life.’ In 1997, Australia recorded its lowest marriage rate this century, while long-term trends show divorces and de facto cohabitation are increasing. Twenty eight per cent of families with dependent children or students are one parent families and one in four Australian children live in single parent or ‘step’ or ‘blended’ families. The law has responded significantly to these changes; for example, the Sex Discrimination Act 1984 (Cth) forbids discrimination on the basis of marital status and most Australian states and territories have legislated to recognise de facto relationships between men and women for various purposes.
Nonetheless, marriage remains of central discursive significance. Marriage has been interpreted as limited to the union of a man and a woman and consequently the ideal family is, as Martha Fineman describes it, the ‘sexual family’, which is ‘tenaciously organised around a sexual affiliation between a man and a woman’. This has been made clear in all the legal definitions of de facto relationships, in terms derived from and controlled by the meaning of marriage; for example, in New South Wales:
De facto relationship means the relationship between de facto partners, being the relationship of living or having lived together as husband and wife on a bona fide domestic basis although not married to each other.
The legal definitions of marriage and de facto relationships perpetuate the view that lesbians and gay men are inimical to the family, marriage and procreation. Lesbians are usually ‘surrounded by silence’ in law, but become visible within the category mother and their characterisation as dangerous to family life is apparent in many child custody cases. However, there has been a strong push in the last few years by lesbian, gay and transgender groups for legal relationship recognition and social acceptance that ‘lesbians and gay men have families too!’. In a recent breakthrough, on 1 June 1999, the New South Wales parliament enacted the Property (Relationships) Legislation Amendment Act 1999, which follows the pioneering lead of the Domestic Relations Act 1999 (ACT) in recognising samesex
relationships. There is also a complex racial history sedimented in the ideal family. Historically, the definition of marriage was used explicitly to construct and maintain the racial character of the family. Aboriginal people and ‘Asiatic’ immigrants were denied marriage, motherhood and fatherhood. Aboriginal Australians today are struggling with a legacy of decades of ‘stolen children’ and community and family upheaval. More subtly, assertions of the universality of marriage and the ideal family have the effect of erasing other forms of relationship and community organisation and structure. For example, a recommendation by the Australian Law Reform Commission that functional recognition should be given to diverse Aboriginal marriage practices, including plural marriages, has not been implemented. It has also been argued that Australian marriage law fails to recognise familial structures from different cultures, thereby failing to take multiculturalism seriously.
These axes of heterosexed and raced exclusion intersect with axes of gender and class in the ideal family. As has been exhaustively demonstrated by feminists and others, the nuclear family form was constructed in the nineteenth century as the central bourgeois institution and a necessary element in the growth of the capitalist, industrialised nation-state. Legal regulation, grounded in much older liberal political philosophy, constitutes ‘two distinct social realms’: the ‘public realm is presented as that of state, market and politics, and is the world of men; the private realm, associated primarily with women, is the world of family’. Public and private are mutually constitutive, but the private is subordinated to the public. This gendered public/private dichotomy is clearly overly simplistic in the 1990s, yet it remains significant. It is necessary to map more carefully the ‘cartography’ of public and private:
The traditional public/private division between work and family, and between state intervention and the family, appears to have been challenged ... [but] the 1990s have revealed the persistence of inequality of women compared with men in various spheres, as well as the inequalities of some groups of women in relation to others.
Tax and transfer laws rely fundamentally on a distinction between private and public. However, the division of public/private is not clear-cut or uniform. Transfer law in particular is ambiguous, as it seeks to characterise women both as ‘private’ carers in the home and as ‘public’ workers in the market and requires surveillance of ‘private’ family finances. Even in tax law, however, the ideal family is not always a site that is private in the sense of being unregulated; indeed, it may be subject to considerable investigation, although not so intrusive as the surveillance of recipients in the transfer system. Paradoxically, tax and transfer laws may exert greater normative or disciplinary power over those who could conform to the ideal family, such as heterosexual couples, while limited freedom from regulation – some privacy – may be allowed, through the very silence that surrounds them, to lesbian, gay and other non-normative families.
In Parts 3 and 4, I show how tax and transfer laws are built upon and creatively reproduce the ideal family through statutory provisions, categories and definitions, case law and administrative and political interpretations. I also demonstrate where and how tax and transfer laws actually subvert or contest the ideal family, by looking at their application to lesbian or gay, or other non-normative families.
[W]hen we restructure the tax system, families, dependent children, will be big winners. ... it has long been one of my policy goals in public life to improve the tax system to give greater recognition to Australian parents. Prime Minister Howard
The Australian federal income tax, introduced in 1915, has always been imposed on the individual taxpayer unit. Income tax is imposed on individuals under a progressive marginal rate structure with several tax brackets. In 1975, the Asprey Committee affirmed Australia’s individual tax unit, stating that ‘the right to be taxed as an individual has always been accorded in Australia’. However, in spite of this significant rhetoric, from the beginning in the Australian system, a family tax unit, subject to a lower effective tax rate, has also been important. The ITA takes account of a carefully defined family both in its ‘normal’ or fundamental structure and in tax expenditures directed at families. Court cases and administrative rulings have produced a gloss on the meaning of family enabling income splitting for some taxpayers.
The definition of income in the tax law inherently depends on a private/public demarcation. The traditional definition of income is the sum of a taxpayer’s consumption and accretions to wealth. The structure of the income tax requires a nexus between the earning of assessable income and deductible expenses and thus produces the hallowed rule that ‘private or domestic’ expenses are not deductible (s8–1(2)(b) ITA97); otherwise, taxpayers might be able to deduct personal consumption outlays, rendering the income tax a tax solely on savings. What is less clear is that the use of observed income as a measure of capacity to pay is itself premised on a public/private dichotomy because it ignores the value of what economists have termed leisure (to distinguish it from market work), which includes all non-market production, sometimes termed imputed income. The definition of taxable income thus reflects the dichotomy of the family and the market, as caregiving and homemaking within a family is not recognised as income. As Patricia Apps explains, this favours the breadwinner-homemaker family in which the wife can specialise in child rearing and caring, supported by the husband’s market income, in contrast to a family in which both spouses work in the market, both are taxed and they must pay for childcare and homemaking expenses (which are non-deductible).
The mutual exclusivity of the domains of the ‘private or domestic’ life and expenses of the taxpayer and his or her ‘public’ work or income-earning activities is illustrated by a judicial statement that "it must be a rare case where an outgoing incurred in gaining assessable income is also an outgoing of a private nature. In most cases the categories would seem to be exclusive". While the meaning of ‘private or domestic’ has rarely been defined by the courts (which have tended to focus on the particular facts in each case), family expenses of the taxpayer are inevitably private or domestic, being:
Losses or outgoings which relate solely to the house, home, or family organisation, of the person incurring them, eg, expenses paid by a person to a domestic to enable the former to carry on his or her own employment.  The courts have defined childcare expenses as essentially private or domestic even where incurred to enable the taxpayer to work or otherwise earn assessable income.51
The prohibition on deducting ‘private or domestic’ expenses is reinforced by a specific rule against deductions for payments for maintenance of a taxpayer’s family, defined as spouse and children (s26–40 ITA97). The same rule generally also applies to maintenance payments in respect of children of divorced or separated spouses; these are exempt from tax in the hands of the recipient and a deduction is denied to the payer (s51–50 ITA97).
The ITA defines family to include only spouse and children for these and other statutory provisions, with few exceptions. A ‘child’ includes an adopted, stepchild or ex-nuptial child, but administrative rulings make it clear that if there is no spousal relationship, only a biological child will suffice to produce a family. The meaning of ‘spouse’ in the ITA implicitly includes married couples and extends to heterosexual de facto couples (s995–1 ITA97):
‘spouse’, in relation to a person, includes another person who, although not legally married to the person, lives with the person on a genuine domestic basis as the person’s husband or wife.
Many factors are taken into account in determining the extended meaning of ‘spouse’, including living together, financial interdependence and joint raising of children.
In 1915, in spite of the individual taxpayer unit, a family unit was central for most wage earning taxpayers. The individual income tax must be placed in the context of the ‘family wage’ laid down by Higgins J in the Harvester case, to be paid to a male worker to support himself, his wife and children. The general exemption (or tax-free threshold) in the 1915 income tax was set by reference to the living wage for a single man and increased for a man with a wife plus extra for each dependent child. Most workers with families did not pay income tax as a result of the combined family exemptions and so the effective tax unit for most Australians was the ‘family man’, as head of household, rather than the individual. This statutory effective family unit was whittled away as the income tax became a mass tax. In 1997, the Liberal/National Family Tax Assistance election package re-established a family tax unit for a modest amount of wage earner income. However, this short lived reform is replaced by reforms in A New Tax System, discussed in Part 4.A below.
Although private expenses are not deductible, the ITA takes into account a taxpayer’s family expenses in part through tax rebates or offsets, introduced in their present form in 1975. The tax rebates consist of an amount deducted directly from tax payable which is invariant to marginal tax rate. However, the rebates are nonrefundable, so that if a taxpayer does not owe sufficient tax to claim the entire rebate, the taxpayer loses the benefit of the excess amount of the rebate. Furthermore, the rebates are not limited with reference to income of the taxpayer but (assuming other conditions are satisfied), are available even to taxpayers subject to the highest marginal tax rate. A key feature of all the rebates is that the benefit – the tax reduction – goes to the breadwinner taxpayer, thereby maintaining the breadwinner’s control of income and status as head of household. Family tax rebates are becoming less significant as family expenditures are shifted to the transfer system, a trend continued in the Family Assistance Act.
Transfer payments in the dependant rebate includes the spouse, child-housekeeper, parent maintenance and invalid relative rebates. There is a maximum rebate, reduced with reference to the ‘separate net income’ of the dependant(s). For example, to be eligible for the spouse rebate, the taxpayer must contribute to maintenance of his or her spouse, who is entitled to earn up to $5,578 separate net income in 1998. The child-housekeeper rebate requires that the child must be wholly engaged in housekeeping for the taxpayer. The parent maintenance rebate requires the taxpayer to contribute to the maintenance of the taxpayer’s parents or parents of the taxpayer’s spouse. The invalid relative rebate has a slightly broader compass and is available for taxpayers who support a child, brother or sister who is unable to work or is in receipt of a disability support pension or rehabilitation allowance (s159J(6) ITA36). There is a separate sole parent rebate for a parent who has the sole care of dependent children under age 16 or full-time students under 25 (s159K ITA36). Dependent spouse with child and sole parent rebates are replaced by the Family Assistance Act.
In contrast to the prohibition on deduction of child care expenses in calculating taxable income, costs for child care incurred by a dependant in order to work are allowed in determining the dependant’s separate net income for the purpose of calculating a dependant rebate for a taxpayer. The law effectively allows the earning of a small amount of income by a dependent spouse, but only on condition that she remain directly responsible, in this case financially rather than in person, for care of the children. It thus perpetuates the gendered notion of a sphere of family responsibility wholly separate from the breadwinner’s sphere of work. The tax rebates are carefully targeted to the ideal family. Even the invalid relative rebate does not recognise an extended family beyond parents, parents-inlaw or invalid brothers and sisters. This strict definition contrasts with the definition of family in anti-avoidance provisions, such as the specific antiavoidance provision which enables the Commissioner to disallow a deduction for payments to ‘related entities’ that are not ‘reasonable’ (s26–35 ITA97). A ‘related entity’ for this purpose includes a ‘relative’, defined broadly to include spouse and parent, child, grandparent, siblings, uncle, aunt, nephew, niece, their children and spouses and lineal descendants (s995–1 ITA97).
Only the housekeeper rebate does not require kin status and is not reduced with respect to separate income of the housekeeper (s159L(1) ITA36). However, the housekeeper must be wholly engaged in and have full responsibility for keeping house for the taxpayer, and in caring for the taxpayer’s child, invalid relative or disabled spouse. The spouse, child-housekeeper and housekeeper rebates are mutually exclusive; indeed, they all relate to the existence of a homemaker for the breadwinner taxpayer.
Other tax rebates also take into account a breadwinner-homemaker family and are effectively ‘paid’ to the breadwinner spouse. A taxpayer is entitled to a limited rebate for superannuation contributions made on behalf of a low-income or nonworking spouse. Taxpayers with spouses and children have a higher combined taxable income ceiling for calculation of the private health insurance tax offset (s61–305 ITA97). Medical expenses paid in respect of a spouse may be taken into account by a taxpayer in calculating the medical expenses rebate (s159P ITA36) and the existence of a dependent spouse is taken into account in determining the rate of the pensioner rebates (s160AAA ITA36).
Clearly, alternative or non-normative families will not qualify as family for the tax rebates: a gay man could not claim the spouse rebate for his dependent partner or the parent-in-law rebate for maintenance of his partner’s parents. However, a different result would be produced if a gay man supported dependent children. His categorisation, not as a spouse but as a parent, would then entitle him to the sole parent rebate, even if his gay lover contributed to their support. He could also claim the housekeeper rebate, by this means producing domestic circumstances that mimic the ideal family.
As stated above, the Howard government introduced in 1996 Family Tax Assistance, which will apply only until 30 June 2000. Family Tax Assistance reduces tax rates for families through a higher tax-free threshold for qualifying taxpayers with dependent children. Like the tax rebates, the higher threshold is only available to one spouse in a couple; in consequence, only that person controls the benefit of the extra income available because of the reduced tax. Family Tax Assistance Part A increases the tax-free threshold on the basis of the number of dependent children and eligibility is measured based on ‘family income’, defined as the sum of the taxable income of the taxpayer and his or her spouse, up to $70,000 for one child. Family Tax Assistance Part B provides for a single additional flat exemption increase for a taxpayer supporting at least one dependent child under age five. Where a taxpayer has a spouse, the Part B benefit is only available if the spouse’s taxable income is less than an income ceiling.
The eligibility tests for both Part A and Part B Family Tax Assistance turn on the existence of a parent-child relation, not a spouse relation, and both are available for sole parents as well as for families comprised of married or de facto spouses and children. No other caring relationship is recognised. Once the parent condition is satisfied, the income of the taxpayer and his or her heterosexual spouse (if any) will determine eligibility; it is presumed that a sole parent is a single income family. The complete invisibility of any other form of family relationship, combined with across-the-board recognition of sole parents has results that are quite subversive of the ideal family. A parent in a lesbian or gay relationship is eligible for both Part A and B benefits based solely on his or her own taxable income, so a lesbian mother could earn up to $65,000 and obtain both benefits in respect of her child under age five, while her same-sex partner contributes an unlimited amount of income to the household. To generalise, a household in which at least one person is defined as a parent of dependent children, in which financial support is provided by any number of persons who are not ‘spouses’, can maximise the benefit from Family Tax Assistance.
In contrast, Family Tax Assistance tends to confine heterosexual married or de facto spouses to a traditional family structure. The Part A benefit is limited by the joint ‘family income’ test (including taxable income of each spouse). If both spouses are earning, the $70,000 threshold will be much more easily reached than if only one spouse is earning. While this income ceiling is fairly high, so dual earner heterosexual families earning average or below-average wages will benefit, it nonetheless tends to increase the attractiveness of having one spouse stay at home and look after the children, saving childcare costs. Eligibility for the valuable Part B benefit explicitly requires conformity with the breadwinner-homemaker model, as a result of the strict spouse income test. As this test refers to taxable income (not ‘separate net income’ as in the dependent spouse rebate), childcare costs cannot be deducted in determining the spouse’s income: it appears that she or he must really be a homemaker for the breadwinner to be eligible.
The economic effect of Family Tax Assistance is to shift some of the tax burden away from qualifying families to other taxpayers. The ideological or discursive effect is to strengthen the ideal family, but because of the validation of taxpayerparents even where they are not in a spouse relationship, Family Tax Assistance has perhaps unintended consequences, enabling queer and non-normative families or social groups supporting children, although unrecognised by the law, to potentially obtain maximum financial benefit.
The family unit in the income tax is not limited to the tax statute. It is necessary to consider the less visible court cases and administrative rulings to discover the widespread existence of an effective family unit arising from practices of income splitting between spouses and other family members with property or commercial income. Income splitting essentially involves the transfer of income from a taxpayer with a high marginal tax rate to another person with a nil or low marginal rate. Overall, the income is subject to lower tax and this benefit can be shared between the high and low rate taxpayers. The reasons why income splitting is problematic are nicely put by Murphy J:
The Act is based on graduation of tax for individual taxpayers and the evident legislative policy is to require those who are more financially able (as measured by their income) to pay at a higher rate than those less able .... [The Act] should not be construed to impute an intention to Parliament to have its main purpose defeated by an income splitting device which tends to defeat its policy of graduated tax rates, is available only to the professional and commercial classes, and allows the burden of taxation to be shifted increasingly from those taxpayers most able to afford it to those less able.
However, certain kinds of income splitting have been consistently upheld by the courts and administration. Income splitting may take place through the disposition of income-producing property directly to another person, or to a trust for their benefit, or through the sharing of business or professional income.
(i) Splitting of Property Income
The existence of an ideal family – a private sphere set apart from the world of commerce – will save a transaction that splits income from property from classification as tax avoidance. As expressed by Lord Denning in Newton, an arrangement will not amount to tax avoidance if the transactions ‘are capable of explanation by reference to ordinary business or family dealing’. An ‘ordinary family dealing’ has been consistently interpreted by the courts and administration to mean income splitting of a family provider’s income to his wife and children, where the provider’s obligations to the family and purpose for the income transfer are defined by ‘both legal obligation and natural love and affection’. The goal of this formulation is to encourage provision of financial support for the taxpayer’s family, but this does not extend to financial independence for family members. A primary narrative in the cases and rulings on income splitting has been to ensure that control of the assigned income or property stays with the provider, ensuring continued dependence of wife and children.
Splitting of income from property was established as acceptable as early as 1920, when the High Court upheld an arrangement in which a taxpayer declared a trust of the beneficial interest in farming property for himself, his wife and his daughter equally, while retaining full powers over the property as trustee. His continued control of disposition of the income did not render the arrangement tax avoidance in the eyes of the Court, which justified its decision on the basis of strict property law principles. The Court has since refined the technical property analysis to protect the donor of the income-producing assets, enabling him to assign everfiner divisions of property without divesting control. This was taken to an extreme in Everett. The taxpayer, a solicitor in partnership, assigned by gift sixthirteenths of his share in the partnership to his wife, including the right to receive profits; the deed of assignment stated that she would not become a member of the partnership. The Court, with Murphy J (quoted above) in lone dissent, upheld a division of the partnership rights between the (property) right to profits and the (personal) right to act as a partner, and held that the former could be independently assigned. Parliament did little to counter Everett and the administration acquiesced.
The ideal family in the context of property income splitting is the traditional private sphere represented by the husband as head of household, in which the state declines to intervene. Judith Grbich has shown how the notion of ‘ordinary family dealings’ always constitutes the taxpayer and his spouse in a hierarchical and gendered fashion, and in particular, how the female subject is inevitably defined by her wifely status. A wife’s agency (however actively she may participate in the plan) is constrained by the structural requirements for successful income splitting set by the tax law. She must be financially dependent for her husband to obtain most benefit out of the income splitting. The taxpayer husband’s control of the income (for non-tax purposes) is generally assured by convention and operation of law. Settlers of family discretionary trusts usually ensure control by appointing a solicitor or associate as trustee. A trust deed will often spell out consequences if the family arrangement (for instance, the marriage) enabling the income splitting were to break down, for example, by ensuring that trust property is returned to the settler-husband following divorce. Such arrangements have been accepted by the ATO:
[I]t is in the nature of family-inspired arrangements that the head of the household may build into the divestment arrangement some sort of variation provision should things not work out, in family terms, as he or she would wish.
Since 1980, some law reforms aimed to prevent the more speculative or obvious of these arrangements. In 1980, Division 6A ITA36 was inserted, stating that an assignment of a right to income is ineffective if it is for a period of less than seven years. The income is taxed to the donor, obviating the income splitting.
Assignments of income to minor children were effectively outlawed by Division 6AA ITA36 (inserted in 1981), which ensures that a child’s income over a minimal amount is taxed at the highest marginal rate. At the same time, however, then- Treasurer Howard confirmed that the new anti-avoidance regime, Part IVA ITA36, would not impede assignments of income-producing assets by a parent to spouse or adult children. The introduction of capital gains tax by the Labor government in 1985 should have ensured that the value of property transferred is taxed to the assignor when assigned. This change has probably discouraged such assignments, although pre-1985 transfers were fully protected from taxation and many issues arise, not least determining the value of the asset assigned. However, the use of discretionary trusts has increased over the last few years.
It is interesting to speculate whether these techniques of property income splitting are available to non-normative families. It seems almost certain that an arrangement established by a de facto couple would be upheld. It also seems possible that persons not usually part of a normative family may be joined in an existing income splitting structure, for example, friends or lovers of children (gay or straight) may be made additional beneficiaries of a family trust; this is the model of the patriarch providing for his household and may be upheld if no tax avoidance purpose was apparent. However, whether an assignment of income-producing property by a lesbian or gay man for the benefit of a lover and other dependents could qualify as an ‘ordinary family dealing’ (and hence not tax avoidance) is by no means certain, especially in view of the requirements of a legal obligation for support and ‘natural love and affection’. Same-sex love has scarcely been conceived of by the law as ‘natural’ and, as noted in Part 2, it is only very recently that changes to state laws have begun to recognise the possibility of support and maintenance duties in the context of queer or other ‘domestic relationships’.
The lack of legal recognition of same-sex relationships has meant that legal dealings between gay or lesbian partners and their children have been construed in other contexts as if they took place between ‘strangers’ and not as family dealings. In W v G, W made an equitable claim for part of an inheritance of G, to support W’s two children conceived by her through artificial insemination. W and G had separated, but had previously been partners for several years and G had participated as co-mother to the children. The legal claim was detrimental reliance by W on a promise by G, made preceding the insemination, to support the children. Jenni Millbank suggests that, largely as a result of the limits of the equitable doctrine on which the parties were forced to rely, the relationship of W and G ‘was conceptualised by the court not as one of ‘natural love and affection’, where the law should hesitate to involve itself, but as a somewhat more impersonal, ‘unprivate’, even commercial connection’. The court did not discuss G’s relationship with the children, although whether she should support those children was the central issue:
A functional family relationship between Grace and the children thus becomes irrelevant to the legal framework which determines her liability to support them. ... Equity is here concerned with a promise, not a relationship between two women or between one of them and two children.
In the result, G was required to pay $151,125.00 for annuities to support the children until age 18. Had G made this arrangement to support W and the children and also split income from property, I suggest that it would have been at risk of being found to be tax avoidance. In contrast, in the event of family breakdown for heterosexual spouses, s106–5 ITA97 enables a tax-free assignment of property from one spouse to another and s102AGA ITA36 allows the splitting of a taxpayer’s income with children through a child maintenance trust. The latter option would not be open to G even though the ATO interprets ‘family breakdown’ to include the breakdown of ‘other relationships’ as well as de facto or de jure marriages. A child maintenance trust is only available for support of a child of an ‘other relationship’– defined to mean a relationship where the parents ‘never live together as spouses on a genuine domestic basis’ – if the child is the biological child of both parents. G cannot qualify as either a spouse or a parent for these provisions of the tax law.
(ii) Splitting of Other Forms of Income
The ideal family is also central to tax cases concerning splitting of business or professional income, but it appears in this context as a tool for the revenue not for the taxpayer. It is generally accepted that income from a taxpayer’s own personal exertion or services cannot be assigned: ‘a family man ... cannot achieve taxation immunity by the simple expedient of assigning his earnings to his wife and family’. However, the establishment of husband-wife partnerships and employment of family members in a taxpayer’s business and operation through trusts have been common techniques of splitting business income between family members. Such arrangements are often challenged by the ATO on the basis that they are mere disguises for the splitting of one taxpayer’s personal exertion income.
To ensure the arrangement is upheld, the taxpayer’s goal is not to emphasise that there is an ‘ordinary family dealing’, but to elide the family nature of the deal altogether, so as to establish its authenticity as a business transaction. However, this can be difficult to prove. The cases and rules often reveal a gendered assumption that the husband is head of the private household, so the wife is presumed not to be a genuine business partner: ‘the traditional presumption of female dependency and the extent of domestic services which a wife would normally render to her husband make it ‘difficult’ to draw the line between her truly private services as a wife and her public position as partner contributing business labour’. In contrast, the lack of family status of gay and lesbian or other non-normative families – their ‘stranger’ status – should make it easier for them to prove the existence of a true business partnership, thus enabling sharing of profits and reduction of the tax burden.
Income splitting is, in the end, about the distribution of the tax burden, as it lowers the effective tax rate on those families who accomplish it and (by default) increases it on other taxpayers. In the past, Prime Minister Howard has advocated the introduction of a statutory family tax unit (which would deem income to be split between spouses), and the short-lived Family Tax Assistance reform extends the ability to split income in a limited way to wage earners and their families. Statutory tax provisions regarding the family and the cases and rulings allowing income splitting both rely on and reinforce the ideal family. However, the increased emphasis on parenting, especially the inclusion of sole parents, can work to the advantage of non-normative families with (biological) children.
The ideal family is most significant in producing ‘ordinary family dealings’ that enable splitting of property income. In this context, the income tax seems both to require and to privilege a traditional family which is relatively wealthy and functions in a hierarchical and gendered fashion, with a taxpayer-provider in control of income distribution to a dependent wife and children. The link between the ideal family and the tax law’s role in practices of distribution of wealth and property is not a coincidence. Grbich has observed that the position of women in revenue discourse ‘follows the logic of capital, [as] she appears in a gendered and familial form where the discourse of capital takes her’. My analysis suggests that the ideal family, heterosexed, gendered, classed and raced, is itself embedded in the ‘logic of capital’. It is produced and maintained through the tax law in a way that maintains the poverty of women and, more generally, inequality in distribution of the tax burden and hence of income throughout society.
Given the important social role of the family as the primary caregiving institution, it seems logical that our core or essential family connection should not be the sexual or reproductive affiliation exemplified by marriage, but rather the nurturing, caretaking relationship between a mother and a child.
I turn now to consider the ideology of family in transfer laws. It has been observed that transfer payments fall into two main categories: those relating to family responsibilities and those relating to work. The public/private distinction is, as in the tax system, inscribed in the very structure of the transfer system, but the relations of dependence and support in which the payment recipient is engaged with the state, market and family are complex. Feminists have shown that family payments in the welfare state at first aimed to build Australia through a pro-natalist discourse that constructed white women as breeders of the ‘Australian race’. Programmes such as the 1912 Baby Bonus propagated a notion of domestic maternal citizenship for white women, from which Aboriginal and non-white mothers were excluded. This maternal citizenship ideally took place in marriage, although the general aim of encouraging white women to breed took precedence. Child endowment support, introduced in 1941, provided benefits to women with children on the basis of need, but was still denied to unmarried mothers with one child. Benefits were rarely paid to Aboriginal women because they were required to be effectively assimilated.
In the 1960s and 1970s, major reforms cemented the notion of need-based entitlement for all Australians which is still acknowledged as the ‘cornerstone of the Australian social protection system’. In 1972, child care for working mothers became an object of federal funding for the first time, with the introduction of the Child Care Act 1972. Reforms also initiated changes to the meaning of family in the welfare state, most significantly by the introduction of a need-based single mothers’ pension, a response to findings by the 1970 Henderson Inquiry that many families headed by single mothers were living in poverty. In 1977 this was extended to supporting fathers. Today, however, women still make up the vast majority of recipients and high levels of poverty still exist among sole mother families.
The Hawke/Keating Labor governments in the 1980s and 1990s increased some rates of assistance but tightened means testing and other entitlement rules, emphasising the obligation of recipients, including single mothers, to work in the market through the Jobs, Education and Training (JET) program. At the same time, Labor continued to modify the meaning of family in the transfer system through frequent reforms to the structure of payments. These changes were often lobbied for by feminist activists, or pushed through by feminist bureaucrats. Provisions increasingly referred to the parent role and not the spouse relationship. In 1992, Labor introduced Family Payment, an allowance that was explicitly required to be paid to the female member of a couple, ‘consistent with the Government’s social justice strategy and the national agenda for women’. In 1994, Labor replaced various benefits with a means tested Parenting Allowance for the primary caregiver, stating:
[L]ow income families will have a greater degree of choice about whether one partner will stay at home and care for their children ... in the International Year of the Family, the parenting allowance initiative will give the partner caring for the children an independent source of income, recognising financially the valuable work of the carer. 
The payment of this and other benefits directly to the caregiver was a significant victory for feminists, who argued for it on the basis of studies showing that the underlying policy assumption that the couple shared household income and assets was empirically inaccurate. Until this time, benefits for wives and children were paid to the recipient of a work-related benefit, generally the husband, implicitly maintaining his role as breadwinner-head of household. These changes emphasised the carer’s financial independence from her spouse and provided her with control of income, significantly challenging the dependence implicit in the ideal family. At the same time, supporting her choice to ‘stay at home and care for the children’ in the context of a gendered labour market implicitly resurrects the heterosexual breadwinner-homemaker dyad. However, women’s market work was supported by federal childcare funding, which gradually gained legitimacy at this time. Publicly funded childcare began to be seen as a broad economic issue, not a welfare measure, and was subject to innumerable reviews and reorganisations. Cash assistance for childcare was introduced in the mid-1980s, but exploded in the early 1990s when the Keating government expanded places dramatically and introduced the means-tested Childcare Cash Rebate, subsidising childcare fees up to 30% and extending assistance to informal care.
Changes in the early 1990s instigated a paradigmatic change in the defining family relationship in the transfer system from that of spouse to that of parent-child, which bears a striking resemblance to the call by Martha Fineman (in the context of the United States welfare system) to value the mother-child bond, instead of the ‘sexual family’, as the core family connection.
The Howard government has continued the move towards supporting a genderneutral concept of parenting. In 1996 the government introduced Family Tax Payment for families whose taxable income is too low to take advantage of Family Tax Assistance (see Part 3 above). Unlike the tax break, Family Tax Payment is paid to the primary carer. In March 1998, the government consolidated the Parenting Allowance and the sole parent pension as Parenting Payment (Part 2.10 SSA):
The new payment simplifies income support arrangements for parents, providing support for people with child rearing responsibilities, regardless of whether they are partnered or not. It also aims to reduce the social stigma which is sometimes associated with being a recipient of [sole parent pension].
The Family Assistance Act and related legislation make significant changes to the structure and administration of family assistance, and substantially increase the value of family assistance benefits, as this is a crucial part of the compensatory measures for the GST. Three new benefits for families with children have effect from 1 July 2000. Family Tax Benefit Part A (FTA) and Family Tax Benefit Part B (FTB) are cash benefits for parents supporting children in their care. In spite of the nomenclature, they are not provided through the tax system as rebates or increases in the tax-free threshold but are transfer payments, although recipients will have a choice of ‘delivery’ by means of reduced salary tax installment deductions, or a lump sum payment at the end of the year. The new legislation abolishes dependent spouse (with children) and sole parent tax rebates and Family Tax Assistance in the ITA and Family Tax Payment and basic Parenting Payment in the SSA. Child Care Benefit (CCB) replaces the child care cash rebate and child care assistance (repealing the Childcare Rebate Act 1993 and the new Childcare Payments Act 1997, which had not yet taken effect). CCB is a cash subsidy for work- or study-related childcare fees paid to approved child care centres or registered carers in the informal sector, available as a periodic installment paid to the centres or as a lump sum paid to the eligible parent at the end of the year.
The Family Assistance Act, in line with previous reforms, has as its basic eligibility test the existence of the parent-child caregiving relationship. Like the SSA, the Act provides for transfer payments based not only on legal obligations of support or existing understandings of family, but also on actual care of a child by the recipient adult. The focus on care of children is highlighted by the novel inclusion of the CCB in the same legislation as assistance for private caring of children. An adult caring for a child may be eligible for FTA, FTB and/or CCB where the adult either has legal responsibility for the care of the child; cares for the child under a family law order or parenting plan under the Family Law Act 1975; or where the child is in the actual care of the adult and not in the care of anyone else with legal responsibility for the child’s care. This means, for example, that a lesbian co-mother who actually cares for the biological children of her ex-partner, where they are not in the care of the ex-partner, will be entitled to FTA even though she has no legal responsibility for care of these children. The requirements for a child to be ‘in the care of’ an adult are not spelled out in the legislation (is it enough that the child lives with the adult or must s/he be the primary caregiver, however that is determined?) but these issues have been considered by the administration in applying the SSA for some years now.
Ironically, the emphasis on parents and children in the Family Assistance Act tends to reinforce the generally narrow view of the meaning of family. The Act does not include benefits for any caregiving activity other than caring for children. The narrowly drawn parent and invalid relative tax rebates remain in the ITA, where they are not means-tested and are of no assistance to the lowest income people who pay insufficient income tax to claim them. Why were these rebates not brought under the rubric of family assistance and modified to make them more relevant to contemporary realities of caring? The government has missed (or perhaps avoided) an opportunity to broaden the scope of support for other kinds of caregiving. In spite of the emphasis on parenting, however, the sexual family, in the form of the ‘couple’, remains important in means-testing for family assistance benefits. Rather than being a category that enables or increases transfer payments (although this is sometimes true), the ‘couple’ in this new paradigm operates to restrict entitlement to benefits and to enable increased regulatory surveillance of payment recipients.
Being a member of a couple can affect entitlement and rates of transfer payments in the Family Assistance Act and the SSA. The ‘couple’ relationship is extensively defined in the SSA, to which the Family Assistance Act refers, and has been addressed in many tribunal and court decisions. Relevant factors include financial aspects such as joint ownership of property, joint liabilities, pooling of financial resources and sharing of day-to-day household expenses; living arrangements, distribution of housework and care of children; social aspects including a sexual relationship and whether people hold themselves out as married; and the length and nature of the commitment. While ‘couple’ appears broader than a married or de facto spouse, it has been held that ‘it is only the name, and not the essential concept, which has changed’ and so marriage is still a benchmark. Members of a couple must be of the opposite sex. However, as sexual relations are not required, the definition of ‘couple’ appears broad enough to encompass long-term friendships between members of the opposite sex, especially involving shared household or financial arrangements.
The ‘couple’ relationship affects entitlement to transfer payments through modifications to income and assets tests. Feminists have been active in seeking the ‘uncoupling’ of income and asset tests and have to some extent succeeded. Since July 1995, the law has applied a personal income test and a partner income test for most payments in the SSA (it should be noted that ‘income’ in the transfer system is not taxable income, but a much broader measure). A basic rate of Parenting Payment is tested solely on the primary carer’s income; however, eligibility for additional payments depends on the couple’s combined income and assets or on the amount of maintenance or child support that a sole parent receives.
The existence of a ‘couple’ also modifies pension rates. The age pension and work-related payments are paid at a ‘partnered’ rate, intended to provide for the cost of supporting two persons (s1064(4) SSA). It is higher than the individual rate but is approximately 84% of the total pension that would be received by two eligible single individuals. The policy is again premised on the assumption that the couple shares the transfer payment and expenses, producing economies of scale. Under the Family Assistance Act, if the applicant is a member of a couple, the benefits FTA and CCB are tested based on joint adjusted taxable income of the couple, which may affect the rate of payment and taper rates, hence reintroducing the family unit which applied in the ITA and SSA family benefits. FTB tapers out based on the lower of the taxable income of the applicant or his or her partner. While FTB is not income tested at all for single parents, additional Parenting Payment (which is a significant element of cash transfers to single parents and caregiver partners of transfer recipients) remains in the SSA and is still subject to income and asset tests.
The continuing importance of the ‘couple’ relationship and its role in enabling bureaucratic regulation and surveillance is clearest when determining entitlement of sole parents to additional Parenting Payment. If a recipient has shared a home with a person of the opposite sex for at least eight weeks and they have at any time in the past been members of a ‘couple’, or shared a home, or they share assets or liabilities (the latter as low as $1000), FACS must decide if there is a ‘marriagelike relationship’ on the weight of the evidence; this is known as the ‘cohabitation rule’ (s4(4) SSA). If the situation is ambiguous, there is an effective presumption that they are a ‘couple’ and FACS must take into account the other person’s income and assets, consequently reducing the amount of Parenting Payment that may be payable (regardless of actual support). Sole parents are reviewed at 4, 8 and 12 weeks after grant of the pension and every 12 weeks thereafter and may be required to provide information about their domestic circumstances (s501E SSA). A woman who bears a child more than nine months after grant of a pension is automatically reviewed to ensure she is not living in a marriage-like relationship.
As in the tax law, the position of non-normative families in the transfer system is somewhat contradictory. The recent focus on support for parents will aid lesbian or gay parents caring for children, even non-biological children. The lack of recognition of same-sex relationships as a ‘couple’ is an advantage when compared with heterosexual families: none of the income or asset tests will reach income and assets of a gay or lesbian partner, or of any other person who is not in a ‘couple’ relationship with the recipient. A person could not receive the partnered rate of a pension payment for her same-sex dependent partner, but if she and her partner each qualify, they would each receive the individual rate, providing them with greater combined income than a heterosexual couple. A lesbian mother could claim Parenting Payment without any concern about her same-sex lover’s income or assets: as long as she never lives with a man, she cannot fall foul of the cohabitation rule.
The child support scheme aims to enforce maintenance and child support agreements and orders made in the family law system, for example, following orders of the Family Court. The muscle of the ATO is brought to bear in collecting payments, including deduction of child support from tax refunds and collection through salary tax installment deductions.
The goal of the child support scheme is expressed entirely in terms of parental obligation and child poverty: the principal object ‘is to ensure that children receive a proper level of financial support from their parents’. As has been observed by others, however, a vital reason for introduction of the scheme was to ensure that children are paid for privately, thereby relieving the taxpayer of the burden of supporting single parent families. This was previously ensured by a legal obligation for claimants of the pension to seek ‘reasonable’ maintenance. From March 1998, this legal requirement has been repealed, but it remains FACS policy. Currently, around 43% of single recipients of Parenting Payment receive some child support, while more than 90% are seeking such maintenance. The reduction in social security outlays due to the scheme is estimated by the government to have risen from $35.5 million in 1989–90 to $378.3 million in 1997–98. While the primary carer is required to seek support, it may provide her with little additional benefit, either because the father is poor and payments are nil or low, or because the support reduces her Parenting Payment in any event. The defining category in the child support scheme is that of parent not spouse, but in creating an obligation on the primary carer to seek support (or lose her pension), the scheme ‘perpetuates the existence of an ‘eternal biological family’,’ (re)producing the heterosexual family in which the primary carer is again dependent on a breadwinner. Encouraging support of their children by absent parents (usually men) is laudable, but as currently structured, the scheme produces financial dependence of the caregiver on the absent parent. The biological tie to the father is severed only in the case of a child who is conceived through artificial insemination by a woman who is not a married or de facto spouse: In the matter of B and J (Artificial Insemination) (1996) No. ML 4677 Family Court. The case concerned a lesbian couple, J and her lover, and their male friend, B, who donated sperm so that J could bear two children. J was in receipt of a sole parent pension and had been informed by the DSS that her pension would cease unless she sought child support from B. Fogarty J held that B was not a ‘parent’ from whom child support could be claimed. This analysis produces the right result for the donor; what is missing is any discussion of the position of J’s lesbian partner. She cannot have child support obligations and J’s sole parent pension cannot be reduced as a result of their relationship.
The B and J case can be contrasted with W v G (Part 3 above), in which equitable principles were applied to require G, the lesbian ex-partner of W, to provide financial support for W’s two biological children. The judge explicitly stated in W v G that the ‘community’ had ‘already paid substantial child support to the plaintiff’ because she was in receipt of a supporting parents pension. The W v G decision thus sought to privatise financial responsibility for the children in a similar way to the child support scheme, with the crucial difference that as G was still not legally recognised as a parent, she would have no parental rights.
Recognition is long overdue that lesbians, gay men, bisexual and transgender people are ... members of families, they are members of society and they deserve to be treated as such, not as outcasts. Democrats Senator Bartlett 
Parts 3 and 4 show that tax and transfer laws are legal discourses that participate in the construction of an ideal family premised on a heterosexual marriage-like relationship and comprising a breadwinner and dependent homemaker spouse and children. At the same time, tax and transfer discourses do not invariably privilege the ideal family; they are ‘at once complex and partial’. This provides opportunities for resistance and change, but also means that reform is complex and may itself reproduce the existing familial ideology. In this Part, I describe recent proposals for recognition of gay and lesbian relationships in tax and transfer laws and then discuss these proposals in the context of tax/transfer policy discourse around the family, considering in particular feminist interventions in these debates.
Recently, reforms have been proposed by the Greens and the Democrats that would include same-sex relationships as equivalent to de facto spouses for the purposes of tax and transfer laws. The proposals have arisen in the context of a strong push by lesbian, gay and transgender groups for legal relationship recognition, discussed in Part 2. Tax and transfer laws are unaffected by recent state law reforms.
The Democrats have introduced the Sexuality Discrimination Bill 1996, originally introduced by Senator Spindler and read a second time in the Senate on 28 May 1998. The Bill was referred to the Senate Legal and Constitutional Committee, which in December 1997 approved the Bill and recommended that a ‘consistent and gender neutral definition of living in a ‘genuine domestic partnership’ or in a ‘bona fide domestic partnership or relationship’ be established’ by government departments and that ‘all couples or personal partnerships achieve legal recognition at the Commonwealth level’. Section 107 of the Bill provides that any law that confers a right or entitlement, imposes an obligation, or applies in a particular way to a person because that person is a de facto spouse, should confer the same rights or entitlements, impose the same obligations, or apply in the same way, to a person who lives together with another person of the same sex on a genuine domestic basis. Section 107 would treat ‘genuinely domestic’ gay and lesbian couples the same as de facto couples for the purpose of tax and transfer laws. The Greens have a sexuality platform in which they affirm ‘the innate rights of gay, lesbian, bisexual and transgendered people to live their lives freely ... with the same rights and responsibilities as other citizens.’ In 1997, Green Senator Margetts introduced an amendment to the ITA to modify the definition of ‘spouse’ to apply to a person who lives with the taxpayer on a bona fide basis in a marriagelike relationship ‘regardless of whether that person is of the same sex or opposite sex to the taxpayer’. While the amendment failed, Senator Margett’s speech is interesting because it illustrates the power of the claim to objectivity and neutrality of the tax law. She said in moving the amendment:
The tax act, like the SSA, looks at realities of society most of the time rather than seeking to preserve the sanctity of marriage. It is a secular thing and recognises that the key element of concern in taxation is the economic relationship between people and the social bonds that create expectations of economic cooperation and mutual support.
Senator Margetts sought with this proposal to draw attention to the realities of caring same-sex relationships, and to wrest control away from tax ‘experts’; in response to Labor Senator Sherry’s comment that the amendment ‘would be a very major change in tax policy’, she said ‘but, no, it is not. It is about changing the way we treat people, changing the way we recognise people in our society.’
However, Senator Margett’s characterisation of tax and transfer laws as realistically (even benignly) reflecting society assumes that they have no power to produce economic and social relations but are essentially neutral and apolitical. This view is not unusual, but tax and transfer laws are not neutral in their treatment of the family. They are, rather, ‘social policy document[s], influenced by notions of just distribution and ideologically-specific understandings of ideal forms of social ordering’.
The focus on attaining equality which is apparent in both the Sexuality Discrimination Bill and the Greens proposed amendments, frames the debate solely in terms of a comparison with the existing norm – the heterosexual family. Shelley Gavigan, writing in the Canadian context about the challenges posed by familial ideology in lesbian and gay claims to family status under the law, observes that ‘[t]he focus on the specific contexts of lesbian and gay relationships has paradoxically decontextualised heterosexual relationships.’ The context of the specific meaning and importance of family in tax and transfer laws is not considered. There appears to be an assumption that the individual tax unit, apart from specific family tax expenditures, applies universally, and that the family is not especially significant in tax and transfer law. The emphasis on equality between gay and straight couples ignores broader social and economic inequalities in the distribution of income in society, to which familial ideology contributes.
The family tax unit policy debate is well-represented in the collection Tax Units and the Tax Rate Scale (1996) edited by John Head and Richard Krever. In brief, the case for an individual tax unit ‘rests on a belief that economic capacity – to consume, to save, or to redirect for others to consume or to save’ should be taxed in an income tax. Individuals should be taxed on the income that they control, whether or not they share it with others. As stated by Murphy J in Everett (Part 3 above), income splitting subverts the rationale that individuals are to be taxed on income that they control. In contrast, the conceptual case for a family tax unit seems to take a view of economic capacity to pay premised on the use of or benefit from income. This relies on assumptions of income sharing between spouses to conclude that the tax burden should also be shared.
In 1975, the Asprey Committee was reluctant to view the tax unit purely as a tax technical matter, finding that ‘at a time when women are playing an ever greater role in the economic and other affairs of society ... social attitudes to the separate status of the sexes, rather than purely economic considerations, are involved’. The choice of tax unit was thus positioned as a social choice, that is, prior to tax policy discourse. More recently, however, there has been a rise in the power of economic discourse; even the family has been theorised in economic terms by Gary Becker (and others). The affiliation of tax policy discourse with economic discourse has become closer and the tax unit debate today takes place predominantly in terms of economic efficiency and equity. Equity refers to equal treatment of taxpayers, while efficiency is premised on the theory of the market: a tax is least inefficient if it does not distort a person’s choices in the market by discouraging particular choices and encouraging others.
Australian feminist tax theorists generally favour individual taxation and oppose the family tax unit on the basis of economic efficiency and equity arguments. As stated by Claire Young, feminists have largely focused ‘on the role that the tax system plays as a disincentive to women’s participation in the paid labour force (the public sphere) and the consequent reinforcement of their primary responsibility for child care and household labour in the heterosexual (private) family’. The family tax unit obscures control of income within the defined family as it assumes either that benefit from income equates with control, or that control is irrelevant, that is, that marriage is an equal economic partnership. However, feminist research has shown that control of monetary income does matter a great deal for women’s independence. The family tax unit in a progressive income tax system privileges the ideal breadwinner-homemaker family twice when compared with a dual-income family – first, in the failure to tax ‘imputed income’ from household production (discussed in Part 3.A) and, second, through deemed sharing of market income. Apps shows that as women are generally secondary earners, they face a higher tax rate at the margin under a family tax unit; this is both inequitable and inefficient because it produces disincentive effects for women’s market work. The effect of the family tax unit is thus to reinforce for women the ideologically sanctioned and financially dependent role of homemaker, and further to reinforce unequal distributions of income.
Reformers concerned with tax/transfer treatment of gay and lesbian relationships need to take note of the feminist critique of the family tax unit. While it is not clear that feminist insights into the gendered dynamics of marriage can be unproblematically applied to same-sex relationships, it is clear that the extension of the spouse definition in the tax law to cover same-sex couples will benefit only those lesbian and gay couples who fit the ideal family model of breadwinner and homemaker, producing relations of financial dependency, and will subject them to the constraints currently felt by dual-earner heterosexual couples. In tax and transfer systems, lesbian or gay individuals in relationships who now receive benefits could suffer a loss of those benefits and enforced dependence on their same-sex ‘spouse’, particularly if they are full-time parents. The often closeted relationships of lesbian and gay parents may be subject to additional scrutiny, as queer sole parents would effectively be outed to the state and to others, with potentially difficult and even dangerous consequences.
The reforms proposed by the Democrats and Greens overturn a fundamental aspect of the ideal family – its heterosexuality – but in so doing they risk reinscribing existing gendered, classed and raced norms of family. A reform that simply extends the definition of ‘spouse’ to cover same-sex couples is both overinclusive and underinclusive. It fails to recognise the ideological significance of family in tax and transfer systems. Most importantly, it does not open up the category family to recognise new and diverse relations of care and dependency. The NSW Gay and Lesbian Rights Lobby has recognised some of these difficulties with a formal equality approach to family in the tax context, observing that ‘the concept of family can be so liquid or contentious’.
However, these law reform proposals do reveal a problem with the feminist critique of the family tax unit – the assumption of an essentially static and universal ‘sexual family’, centred around the bond of husband and wife. The heterosexual family is unquestioned background in the economic discourse on which feminist tax theorists rely. The discourse focuses the discussion on equity and efficiency in respect of incentives or disincentives for women to enter waged work or to stay at home as full time caregivers. Judith Grbich argues that the discourse effectively reduces the debate to a comparison of two families with equal taxable income but which differ in that one family has a waged wife and the other does not, with the consequence that ‘neo-classical assumptions about equity and efficiency eventually constrain the issue as one of equity between the women of each family’. While Grbich does not discuss queer or alternative families, her analysis shows how the economic discourse produces an unduly narrow feminist critique of the family tax unit which divides women as ‘workers’ or ‘carers’ located within the heterosexual family. Tax policy discourse obscures the processes by which the family in the tax law operates to maintain and support particular distributions of property and wealth. Queer activism has the potential to destabalise the ideal family, enabling a broader perspective on income inequality and caregiving.
Part 4 showed that in recent years there have been significant changes in the meaning of family in the transfer system, from a focus on the ‘sexual family’ to a focus on the parent-child relationship. This interesting shift has taken place while the welfare state is operating primarily within what is termed by Bruce Bradbury the ‘resource paradigm’, in which the state’s role is limited to managing or intervening in markets when they fail due to externalities, and to taking account of social values only to the extent required to reduce poverty and inequality to a level acceptable to society. The resource paradigm is premised on a fundamental public/private distinction, that encompasses the view that ‘the state should not interfere in the private actions of individuals, including their economic relationships within the family, unless there is some pressing need to do so.’
While there is direct government support for caregiving of children in Australia, the privatising discourse of the resource paradigm, inextricably tied to the overarching issue of constraint of public costs, is reflected in the tension between categories of parent and spouse (‘couple’) in the transfer system. The SSA and Family Assistance Act authorise investigation of income and assets within a couple for the purpose of reducing entitlement to benefits; (re)construct the private dependency relationship even in the absence of any actual care or support; and require intrusive surveillance of single parents’ relationship status. At the same time, by counting joint income and assets, the distribution of control of income within the family is ignored, although this omission is rectified to some degree by payment of benefits direct to the primary carer. As a result of their exclusion from the ideal family, lesbian and gay families with children are truly private, but other families that do not fit the ideal because of race or class or lifestyle may suffer heightened regulation as a result of these rules.
Feminist theorists and activists have proposed various approaches to recognise women’s caregiving role, increase support for children and other dependents and alleviate women’s poverty through the transfer system. Throughout this article, I have referred to Martha Fineman’s work regarding the dominance of the ‘sexual family’ and to her call, in the US welfare context, for a ‘reimagining’ of the core family connection, taking as the paradigm the mother-child relation. Fineman’s work is valuable because she seeks to de-centre the heterosexual union in the welfare debate, focusing on rewarding the activity of caregiving itself. Fineman does not mean that the mother-child bond is the only caregiving relation, but that its combination of ‘biological’ or ‘inherent’ dependency of the child and ‘derivative’ dependency of the mother or caregiver is paradigmatic of all caregiving relations.
Marcia Neave is also concerned with dependency of women as primary caregivers but takes a different approach. Her chief concern is with the fact of dependency and its production of poverty in women. She observes that the ideal family and the transfer system both produce financial dependence in caregivers; what is required is support for women’s market work and financial independence, even during caregiving of children or others. Public childcare facilities and subsidies are essential to this projects
The common feature of the feminist reform proposals of Neave and Fineman is that both are directed at creating a culture of ‘collective responsibility for children and other dependents’. This is an essential step in refiguring family relations so as to encompass all kinds of family structure. Gay and lesbian activists should be lobbying for further changes that would recognise all kinds of caring, not for the uncritical assimilation of gay and lesbian couples to a pre-defined category of family.
This article has attempted to trace the meaning and weight of family in tax and transfer laws and policy discourses. It is not always recognised that the ideal family of husband/breadwinner, wife/homemaker and their children continues to play a large role in tax and transfer systems. However, there have also been considerable shifts in both rhetoric and substance in the meaning of family, especially in the transfer system. In particular, provisions that take account of parent-child care, especially actual care and subsidies for paid childcare, provide scope for support of non-normative caregiving relationships including single mothers, gay and lesbian couples looking after children, and other alternative families.
The ideal family retains its strongest significance when fulfilling two different functions in tax and transfer laws and discourse. In the tax system, the ideal family is essential in enabling wealthier taxpayers to split income from property and business, generally providing for a dependent wife and children, but without transferring control of the property or income to the wife. The ideal family is thus a crucial element in the logic of capital and wealth circulation – it enables the accumulation of property untaxed or taxed at lower rates. The tax law participates in shifting the tax burden away from holders of property and business income, effectively transferring the burden to wage earners, and the ideal family is a critical element of this process. In the transfer system, although there has been increasing recognition of the importance of financial independence of primary caregivers, the continued importance of the ‘couple’ relationship ensures that public costs of child support and caregiving are minimised by privatising such care within an ideal family through assets and income tests and the child support scheme.
The ‘active rewriting of family definition and policy’ which is demanded by activists seeking recognition of queer and other non-normative families ‘presents an important opportunity for feminists of all sexual orientations to affect what a ‘family’ means and does’. The Democrats and Greens law reform proposals discussed here raise the issue of the meaning of family, but do not open it up to critique. Instead the formal equality approach taken obscures the problems associated with the heterosexism of the ideal family as it intersects with gender, race and class. It is necessary to think about the role of tax and transfer laws in supporting relations of care and dependence separately from other laws concerning family relations, and to remember the broader context of income inequality between men and women, and throughout society.
I suggest that the individual and the income she or he controls remains the correct focus of the tax system. The individual tax unit should be fully enforced and property and business income splitting eliminated where control remains with the transferor in substance. The recent reforms by the Howard government that place an increased emphasis on actual care should be expanded and should not be tied to narrow and out-dated concepts of family but to evidence of dependence and care. Eligibility should depend on evidence of support, not evidence of a ‘couple’ relation. Our aim should be to move tax and transfer laws away from a focus on the family as we know it and towards support for caregiving of dependents where we find it, with the goal of encouraging independence where this is possible and providing support for diverse caregiving relationships. If we do this, we may really begin to reimagine the family in tax and transfer laws.