― 177 ―

16. Chapter 13 Personal Income Tax: Social Security

13.1. Social security is one of the major areas of government responsibility. And while many features of Australia's social security system fall outside the Committee's terms of reference and are, in any case, currently the subject of investigation by other government-sponsored bodies, personal income tax and the social security system nevertheless have important implications for each other which it would be inappropriate to ignore. The present chapter discusses some of these implications. In Section I attention is focused on the manner of treating social security payments under personal income tax; in Section II some observations are made on social security contributions.

I. Social Security Payments and the Tax System

13.2. If dependant allowances and other concessional deductions are one confused area of tax policy in which rationalisation and simplification should eventually be possible, social security (or social service) payments are another. Here the issues are not already within, but mostly still outside, tax policy as usually conceived. But they are important in the context of long-term possibilities, and they have been raised in submissions to the Committee.

13.3. A social security payment gives rise to a receipt which, in any general sense, must be thought of as income, even though it be received in a way different from that in which other kinds of income are received. Sickness and unemployment benefits as well as many other social service grants are specifically excluded from the base of the income tax, as were age pensions until recently. Such benefits and grants are now paid after two tests: the first is a test of eligibility by an objective criterion: age, sickness, inability to find work; the second is economic: a means test. Many who know that they would pass the first do not apply, knowing that they would not pass the second. Many who pass the first and apply, receive less than the maximum grant because their resources exceed a stated limit.

13.4. Translated into the language of tax the two operations can be differently described. First there is a general provision that all those who pass the objective test have an entitlement to a ‘taxable grant’ at the maximum rate. Secondly this grant is taxed not under normal income tax but in accordance with the provisions of a special progressive ‘tax’ on income (often family income rather than individual income) and on some classes of assets. This is at a rate that rises to the level of 100 per cent of the grant at and above the point where the grant is refused. In practice both operations are conducted simultaneously in the relevant social service department: only the net payment to the applicant is made, and is exempt from further taxation under the normal income tax.

13.5. What has recently been done to some age pensions can then usefully be represented as, logically, no more than the abolition of the ‘special progressive tax’ and its replacement by the normal income tax. The accompanying decline in ‘tax’ appears as an increase in the net amount of the grant retained by most of those receiving it, and the payment of some net grant to many from whom the whole of it was earlier taxed away.

13.6. Much discussion of the problems of poverty and the future of social services can be represented as discussion of the possibility of extending what has been done to age pensions to some or all other social service payments. Changes in this direction

  ― 178 ―
can be advocated on three grounds: administrative simplification, social advantage, and fairness.

13.7. Firstly, the application of a means test is undoubtedly complicated, and the Australian system contains a variety of such tests. They usually involve collecting information about income and assets and often about the affairs of the whole family unit. Their abolition, even in an area not normally thought of as relevant to taxation, would be in line with the Committee's desire to simplify public finances wherever possible. The additional load on the tried machinery of the normal income tax would be trivial.

13.8. Secondly, it is often stated that the present means tests are regarded by those to whom they are applied as a trespass on their privacy. It is not within the Committee's competence to assess this argument, but it has no reason to question its force.

13.9. Thirdly, changes in this area, whether or not they are accompanied by proposals simultaneously to increase the permissible maximum level of the grants made, are advocated as a way of raising the actual payments to those in receipt of partial payments and bringing into receipt of grant many now excluded by the means test. Social service grants are, as has already been shown, a major instrument for the realisation of society's aims for vertical and horizontal equity. This argument amounts to a proposal to increase the amount of income redistribution towards the lower end of the income scale. (The Committee has already recorded its view that the extent of such redistribution must be a matter of general social and political judgment, and it is aware of other public inquiries now collecting relevant detailed information and making proposals about adjustments to existing policies concerning poverty.)

13.10 The Committee accepts that the future of social services is a matter largely outside its terms of reference. But there is a possible conflict between the trend of reform in the tax structure that the Committee is proposing and the trend of social service reform just described which it feels it can usefully elucidate. The cost of any shift from the means-testing of grants to their taxation as normal income may be large even when calculated by reference to present rates of income tax: the increase in social service payments, net of tax, may entail a substantial increase in tax for the rest of the population. This cost may be judged acceptable and desirable. But if, as the Committee would wish, income tax rates were gradually lowered, the cost would of course become greater still since receipts from tax on the gross benefits would be reduced. Indeed if income tax were effectively removed altogether from low-income earners, the social service payments would move from the position of being theoretically taxed at a rate more severe than present income tax to becoming entirely free of any tax. This might be felt to be altogether too costly. There is apparently a dilemma: to retain complex and unpopular means tests in order that the income tax might be lowered and simplified, or to retain the income tax unchanged in order to permit the abolition of the tests.

13.11 One possible resolution of this difficulty may be worth detailed investigation. The means tests might be abolished, but it could simultaneously be provided that some classes of social service receipts, separately itemised in the income tax return, be taxed on a separate scale superimposed upon the standard scale at some appropriate point. Given any desired level of gross social service payments, by adjustment of the point of entry and the steepness of the special scale, any desired level of recoupment could be achieved. The net cost of social service payments to the rest of the community would then become fully adjustable without either means tests or alterations to the rate structure of income tax. Such an arrangement would of course be a

  ― 179 ―
complicating element in the administration of the normal tax system, but it would allow a simplification in the public finances as a whole by the abolition of the existing complicated means tests.

13.12. Such a scheme would seem particularly suitable in the cases of those social service payments that are normally of a short-term character, especially sickness, retraining and unemployment benefits. Here needs may be urgent and substantial, the poverty very great but essentially temporary: the recipient may be well out of it again before long. Over a year in which he had months of acute need, fully meriting assistance on equity grounds, he may prove retrospectively to have had overall an acceptable total income. It is a defect of a system of tax-exempt grants that it cannot handle this kind of situation, even though in general society is content to consider the broad issues of equity on an annual assessment of relative ability to pay (or need to receive). The community might well be prepared to give temporary assistance more promptly and with a less sparing hand if, when the recipient is subsequently restored to financial comfort, he repays via the tax system some or even all the help. What is at present an outright grant could be increased and in effect become, to some extent, a temporary loan. Though this arrangement might be applied, for example, to students’ living allowances, it may not be suitable for benefits, such as age pensions, which relieve needs that are long-lasting, and these could be taxed in the normal fashion.

13.13. Were all or most of the means tests now separately administered to be removed, and social service grants taxed within the taxation system proper, a large measure of rationalisation and simplification in public finance would have been achieved. The objective tests of eligibility for grants, the decision that a person was aged, or unable to find work, or too sick to work, would still of course be administered by officials outside the tax administration, such tests calling for skills other than those in which tax assessors are trained. Grants would still be paid outside the tax system. The grants, in combination with their tax treatment, could properly be described as providing a guaranteed income to those who had passed the objective tests. Moreover, since these would be established to cover the main situations in which society recognises that economic need is likely to arise, it might be concluded that another ideal of social reformers had been achieved: that of ensuring by the supplementation of its private income that every family had a minimum annual income suitable to its size and composition.

13.14. The removal of the separately administered means tests carries with it the need for a simpler method by which the payment of social service benefits can be readily linked with their inclusion in the recipient's income tax return. One procedure by which this may be achieved in most cases would be for each of the individual tax and social service records to show both the income tax and social service reference numbers. In the United States and Canada statement of a taxpayer's social security or insurance number is required on income tax documents. A similar linking in Australia would not weaken in any way the fundamental principle that the normal details shown in an income tax return shall not be divulged by the taxation authorities except in the few special instances now permitted by law. It would probably serve to strengthen both administrations.

13.15. It is logically permissible to think of grants as ‘negative taxes’; and if these grants be taxable under the income tax, a system incorporating them may be described as a ‘negative tax system’, though perhaps, in the case so far discussed, ‘taxable grant system’ would be a clearer phrase. Many schemes, varying greatly in detail, have been propounded under the former title. Some approach from the income

  ― 180 ―
tax end and arrive at proposals for a guaranteed income via the suggestion that tax rebates or tax credits that exceed tax on assessable income be made refundable. Others start from the idea that the guaranteed minimum incomes for particular family sizes and compositions should actually be paid in cash and then subsequently subjected to tax with other income, a break-even point being reached when the tax recoupment brings the income down to the guaranteed minimum. This is indeed a problem that can be presented in a bewildering number of ways.

13.16. The Committee's discussion is offered not as leading to recommendations, but rather as pointing to issues in an area where decisions are perhaps not immediately called for. Equally it does not attempt recommendations on the further problems that would have to be faced were fully-fledged negative income tax schemes seriously contemplated. But it concludes by drawing attention to a point of difference between variants of these schemes that is often overlooked but which in its view is of the greatest significance; that of the primary test of eligibility for receipt of grant. In the kind of evolution of the social service system discussed above, and in many schemes using the term negative income tax, the primary tests of age, unemployment, sickness—tests recognising particular causes of need—are assumed still to be applied. But in others these are discarded, and indeed the whole system of social services to assist the needy is discarded with them. It is replaced by the sole tests of income and family size, with the tax authorities as the sole administrative body. As a necessary corollary, the guaranteed income is paid irrespective of whether the recipients have genuinely sought work and been unable to find it or are genuinely sick or disabled or have any other valid reason for not earning their own livings. In the Committee's view, such schemes seem likely to have consequences for incentives to work and save which make it impossible to consider them seriously.

II. Social Security Contributions

13.17. In the brief comparison between Australia's taxation system and those of other OECD countries in Chapter 2, it was noted that this country is unique among the countries represented in Table 2.E in having no levies entitled to be classified as ‘social security contributions’. The OECD definition of this category confines it to non-voluntary contributions made to ‘general government’ (widely defined) specifically for expenditures within the ‘social security’ area. Apart from Australia only three member countries had taxes thus classified which amounted to less than 10 per cent of total tax revenues over the years 1965–71 (Canada, Denmark, Ireland), while six had them to the extent of between 30 and 40 per cent (France, Germany, Italy, Luxembourg, Netherlands and Spain). The figure for the United Kingdom was 14 per cent, and for the United States 18 per cent.

13.18. As an indication of the comparative level of the social services in these countries, such statistics are, needless to say, quite valueless. They merely reflect the fact that in Australia government has chosen to finance social services directly out of general revenue and indirectly by concessional deductions, such as those for contributions to medical benefit funds, without tying particular taxes to those services. It is simply a matter of form and presentation.

13.19. An arrangement not unlike those of OECD countries was once used in Australia. A separately named social services contribution existed from 1945 to 1950, in administrative substance an addendum to personal income tax but formally separate and formally not a tax. Furthermore its proceeds were earmarked to the National Welfare Fund for expenditure in the area of social security to which the proceeds of

  ― 181 ―
payroll tax, until 1952, were also paid. Though the tax was eventually abolished and the earmarking of payroll tax rescinded, the National Welfare Fund itself still exists, though since it is now wholly financed from Consolidated Revenue Fund and from interest on its unspent balances, it is very little more than a paper archway with a well-sounding name through which tax revenue is channelled to individuals, via appropriations for the Departments of Social Security, Health, and Housing and Construction.

13.20. So long as what was called in Chapter 3 the ‘burden’ fallacy forms part of the attitude of us all to the payment of taxes, there is some truth and good sense in this presentation. Much public expenditure such as that on defence, general administration, law and order and research confers benefits on the community which cannot, in any but an arbitrary way, be traced to individuals quantitatively. But expenditure on, say, hospitals can and so of course can cash transfers to individuals. Moreover, in this area all are entitled to receive them when in need, as defined, and though the definition of means will exclude many people, no one can be certain that he will be excluded all his life.

13.21. As a matter of logic it is in general defensible to conceive of the money required to pay for these social services each year as the out-goings of an insurance fund financed by the compulsory contributions of the actual and potential beneficiaries. Quite possibly many of the public do not see it this way and think rather that all their taxes disappear for ever into the maw of a distant Leviathan. If so, it is arguable that there would be a real gain if it could be done simply by relabelling some specific quantum of existing taxation social security contributions and passing them through explicit national superannuation, medical or unemployment insurance funds or some agglomeration of such funds. The taxpayer might then better understand what he is getting for his money and perhaps the standard of discussion of future policy would be raised. The idea, at first sight, is especially attractive when large increases in benefits of this kind are under discussion, somewhat ahead of their costing.

13.22. There are, however, conditions to be met before this representation of a slice of what are now taxes could be made entirely convincing:

  • (a) The individual contributions would have to be related to expected benefits. Whether the fund actually existed or was of the ‘notional’ kind, it would have to be demonstrably solvent.
  • (b) Though age, sex and family situation might be relevant variables, income could not be (except in the case of an income-related superannuation scheme of contributions). Premiums progressively or even proportionately related to income would, as premiums, be indefensible if benefits were equal for everybody. In all but name social security contributions would be poll taxes and thus regressive on income.
  • (c) As a corollary of (a), the contributions would have to be confined to beneficiaries. In terms of the insurance principle a general medical fund could only be financed by contributions from individuals, not from business employers.

13.23. Though terms such as ‘fund’ and ‘insurance’ abound in the titles of such schemes overseas, these conditions are never, so far as the Committee is aware, in practice met. In the OECD countries the employer contributions are usually larger than the employee, and are generally payroll taxes of such universality that they are probably passed on in prices. And even when the government is not committed to a direct annual contribution when they start, inflation and changes in the desired levels of benefit end by making them dependent on general revenue. When contributions

  ― 182 ―
cease to be genuine premiums, it is likely that even the most innocent taxpayer will come to think of them as taxes, if indeed he ever thought otherwise.

13.24. Even if they were not so, a deeper difficulty remains. The basic social issue to which Chapter 4 was devoted, that of the proper degree of progressivity in the taxation system, would not go away. It might be deflected into a debate about the right distribution of the ‘true’ burden imposed by the finance of the rest of public expenditure. Or it might take the form of questioning the fairness between persons of the social service contributions themselves, as it was in fact questioned in Parliament when Australia's National Welfare Fund was established in 1943.

13.25. The Committee in no way disputes the importance of appropriate names and of the right formal presentation of public accounts. But in this and other areas of taxation policy it believes that the most genuinely urgent and constructive task is to present the citizen with information about public finance in a way that allows him to understand his own personal transactions with the State and to compare them with those of other people.

13.26. It is doubtless true that at the present time normal Australian taxpayers do not clearly identify any part of the tax they pay with the social service benefits to which they are entitled, and that it would be well if they could. But this is only one gap in their understanding. They may know their own income tax payments, though they are likely to know their marginal rate rather than their average. They will not know what they pay in taxes on the goods and services they consume. They may know what cash social service payments they have recently received, but not what, in a prospective lifetime of average health and length, they are likely to receive. They doubtless know that their taxes pay for much of their children's education; but not how much they cost annually or over their childhoods, nor what taxes contribute, say per week, towards the hospital and medical services they and their families may have made use of. The most normal families do not know these elementary facts about themselves and those in their same income group: still less do they know them about other income groups. Still less again do they know how many are in these various groups and what their total contributions are, or could be, to the average finance of the services of the State. It is small wonder that in this vacuum prejudice and myth flourish.

13.27. In the Committee's view detailed and up-to-date estimates of the transactions with the State of all persons and families in normal situations, at different income levels, together with estimates of the distributions of income and wealth, are the necessary foundation for informed debate on tax policy. That they do not now exist has been a handicap to the Committee, and must be one too to those who have to assess the validity of its analysis. The Committee recommends that, as soon as possible, information of this kind be made widely available. It is needed not only by the experts but by every citizen who thinks about the size and fairness of the taxes he pays.