Questions Arising over the Taxation of Family Units

10.20. The first question of detail under a family unit system is the rate scale to be used. Under the British system the same scale as for single persons is applied to the

  ― 136 ―
aggregate income of the married couple (though with qualifications in the form of allowances to moderate the extreme severity of the result). If this be called ‘pure aggregation’, then ‘pure averaging’ might be the term for the system introduced in the United States in 1948, when a schedule with tax brackets twice as wide as those for single persons was adopted for those electing family treatment. This, as noticed, gave the same arithmetical result as if each spouse was taxed separately on half their combined income.

10.21. In terms of economic equity between married and single persons, neither of these very simple solutions to a difficult issue commends itself. Pure aggregation produces the same result as the current Australian individual basis in the special case of one spouse having no income (apart from the effect of the modest dependent spouse allowances). But it would increase the tax paid by the pair in almost any case where both spouses have significant incomes. They would then pay more tax than if they had elected to be taxed as individuals, and the amount by which that tax would increase would be the greater the more equal were their separate incomes. An election system with this rate would ensure no elections. Pure averaging, by contrast, would mean that election saved tax for all married couples except those who happened to have equal incomes. The tax each paid would in no way reflect the additional capacity to pay inherent in the notion of shared enjoyment. Everyone would have a financial inducement to elect, including perhaps those whose relationship was more fairly described as, and felt to be, economically quite separate: the tax result would be the same as if all couples achieved complete income-splitting. The loss of revenue would be so great as to entail, if the income tax revenue were to be maintained, a severe increase in tax on single persons (and those married couples who resisted the tax inducement to represent themselves as a sharing unit). It was for this reason that the pure averaging provisions of the United States tax were abandoned in 1969.

10.22. It is inherent in the equity arguments for an elective family unit system that there should be a distinct family rate somewhere between the extremes of pure averaging and pure aggregation. The former is undoubtedly too kind to electing pairs, too unkind to single persons and couples who require individual treatment; the latter undoubtedly too harsh on electing pairs, too kind to single persons and those who accept individual treatment. The choice of family rate relative to the individual rate is therefore one of nice judgment.

10.23. A further question that has to be dealt with under a family system is the treatment of dependent wives. There are here two distinguishable situations: (i) when the wife's dependence is primarily due to her looking after small children rather than going to work; and (ii) when, without children, she simply prefers to take no paid work (or can obtain none). In the present Australian income tax system they are treated alike. As regards (i), married couples who both work often feel that, because their child-minding and their housework involve more strain than for families where the wife is at home, they are unfairly treated at present. Under a family unit system this situation could be handled by an earned income allowance for the second working spouse. As regards (ii), under the present system the family in which the wife stays at home may feel over-taxed relative to the family in which both spouses work and receive the same total income, since the dependent spouse allowance is small and the wife will almost certainly be contributing domestic services of value even though they are unmeasurable. But under a family unit system with a rate near the pure averaging end of the range, the situation might be reversed. Examining this problem, the Canadian Royal Commission on the Status of Women (1970) was tempted to propose the abolition of any allowance for dependent wives without dependent children,

  ― 137 ―
but concluded against this recommendation because it felt there were inadequate employment opportunities for married women of working age in many places and it was concerned also for the position of elderly wives. The future of the dependent spouse allowance in Australia is further discussed in Chapter 12.

10.24. The treatment of children in a family unit system also requires thought. It is implicit in the concept of a family unit that the income of children is part of that of the family as a whole. Certainly when living at home children have full use of the common equipment of the family and may properly be said to share in much parental expenditure. Considerable scope for tax avoidance would remain if any property income they derived, whether the property came to them by parental gift or otherwise, were left to be wholly taxed on an individual basis. This matter is dealt with in Chapter 11. Income being accumulated in a trust estate to which a child is contingently entitled must also be considered. Indeed there might be income so accumulating to which husband or wife is contingently entitled. These matters are taken up in Chapter 15.

10.25. Concessional deductions for dependent children, and the associated issue of child endowment, appear to be unaffected in principle by the possible adoption, or rejection, of a family unit system and their discussion can be left over to Chapter 12. This would not be so, however, if the French version of the system were selected. That variant, summarised in paragraph 10.11, provides, as there remarked, extremely generous treatment for parents with children and seems to be expressly designed to encourage a high birth rate. The Committee presumes that this is not an objective of Australian Government policy and would therefore not recommend a quotient system.