previous
next

I. Terms of Reference and their Interpretation

3.2. Throughout and repeatedly in the terms of reference the phrase ‘taxation system’ is used. This way of regarding a collection of administratively distinct taxes is of fundamental importance. In a complex modern economy where government expenditure is at a high level it is impossible to raise all the revenue needed from any single tax. Each tax will have its own distinct merits and defects when judged by the various criteria commonly applied to taxation. When several taxes are used they have to be seen as supplementing each other and their interactions—and sometimes their conflicts—have to be reckoned with. Whatever their individual characteristics it is their combined impact that must primarily concern the policy-maker. The complete set has therefore to be looked at as an integrated whole, even though before this can be done it is necessary to examine the parts that have to be linked together.

3.3. The Committee is directed to carry out its review of the existing taxation system ‘in the light of the need to ensure a flow of revenue sufficient to meet the revenue requirements of the Commonwealth’. It was with this phrase in mind that the Committee, while drawing attention in paragraph 1.8 to the fiscal importance of the division between taxation and borrowing in the finance of government investment, refrained from offering views of its own on the merits of existing policies about this matter. It would certainly have lightened the load of the Committee's work if this proviso could have been taken to debar any discussion whatever of public expenditure. However, for at least two reasons this is impracticable.

3.4. Firstly, where tax stops and expenditure starts is often unclear. A tax concession to a particular area of spending in the private sector can as well be looked upon as an expenditure of revenue as a failure to collect it, and it is often an issue of importance to tax policy whether such concealed subsidies should not better be given overtly. Still more important is the point that cash transfers to individuals, the whole class of social service payments of every kind, are inextricably bound up with the equity of the taxation system. The Committee certainly does not regard itself as qualified to advise upon the details of the social services, and is aware of other inquiries at work in this area. But some consideration of cash grants, taxable or otherwise, is essential in the design of an optimal tax system. They constitute a fiscal technique to which some attention will be given later, especially in chapters 12 and 13.

3.5. These problems aside, however, once revenue requirements are set there is no scope at all for reducing total taxation: the two are simply the same thing. By this phrase in its terms of reference the Committee is prohibited from suggesting any general set of measures that would necessarily reduce total taxation below revenue needs. The Committee is uninformed, and could not have been informed, of exact future requirements. Any proposal that, explicitly or implicitly, entails a reduction of


  ― 12 ―
present taxation in any particular area has to be matched with proposals that, explicitly or implicitly, approve increases in others. Though the great bulk of the submissions received by the Committee very naturally contain suggestions for lowering present taxation in particular areas, it has been the Committee's task in assessing the arguments offered to ask where the taxation forgone could be recouped more fairly, more simply and more efficiently. Only in Chapter 6, in the context of inflation control, has the Committee sought to refer to tax cuts in a somewhat wider setting.

3.6. The first of the more positive commands in the terms of reference bids the Committee to consider the effects of the taxation system ‘upon the social, economic and business organisation of the community’. This is a phrase with multiple connotations. It is probably helpful to separate these out and attach them to other rather more specific injunctions which follow. Thus, the Committee is to consider the effects of the system upon the ‘economic and efficient use of the resources of Australia’, the desirability that there should be a ‘fair distribution of the burden of taxation’, and that revenue-raising be ‘by means that are not unduly complex and do not involve the public or the administration in undue difficulty, inconvenience or expense’. For brevity, these aims may be referred to as efficiency, fairness and simplicity. However, each of the three when one seeks to define it closely proves to embrace several distinct qualities, and these qualities may conflict with each other in particular applications. Furthermore, each of the ‘big three’ criteria will often, in some respects, conflict with the others. In deciding the best overall tax system and in deciding between alternative provisions in particular taxes, the policy-maker comes repeatedly up against choices between simplicity and efficiency, or fairness and simplicity or fairness and efficiency. These are perhaps the hardest choices he has to make, or invite others to make. Hence it is convenient to give here fuller discussion of the meanings of these simple-seeming terms.

Fairness

3.7. As a quality of a tax or a tax system everyone demands fairness, or equity (the terms will be used interchangeably). But, in tax matters as in law and ethics, it is an ideal exceedingly difficult to define and harder still to measure. It is customary to distinguish the two dimensions of ‘horizontal’ and ‘vertical’ equity: the notions that it is fair that persons in the same situation should be equally treated, and those in different situations differently treated, with those more favourably placed being required to pay more. Both expressions will have to be frequently employed in this report. They reflect the ‘ability to pay’ principle and as such tend to embody the idea that taxation is no more nor less than a sacrifice. As the Committee will record later, this is an idea that needs qualification if it is not to mislead. An alternative principle used in much recent academic discussion is that of ‘benefit’, which relates taxes to the benefits individuals are estimated to receive from government-provided goods and services. Except in a small number of cases where taxes take the form of fees or prices for the direct use of publicly-provided services by particular individuals (e.g. postal facilities), this principle is prohibitively hard to apply. In any case it abstracts altogether from notions of fairness, or implicitly embodies an interpretation of ability to pay that may not be socially acceptable. Hence the Committee argues in this report in terms of ability to pay, though the benefit principle has its place in the discussion of the Australian-origin income of non-residents (Chapter 17) and the taxation of goods and services (Chapter 27).

3.8. When we say that persons in equal situations should pay the same tax we probably say so because we think of the tax as a sacrifice levied upon some kind of private


  ― 13 ―
‘economic well-being’. But the ‘economic well-being’ is a sequence of barely describable psychological states of a thoroughly immeasurable kind. For purposes of practical discussion and decision-making, both the ‘sacrifice’ by way of tax and the ‘well-being’ upon which the tax is levied have to be measured in money terms. Many of the most difficult questions in tax policy stem from the arbitrariness and convention that must be accepted when making this leap from the immeasurable to the measurable, from levels of ‘well-being’ to the choice and exact definition of the tax base.

3.9. It is usually taken for granted that the best available measure of an individual's ‘well-being’ is his income. The ‘burden of taxation’ is thought of primarily in terms of the proportion of a man's income that goes in paying taxes, whether they be taxes levied formally on that income or indirectly by elements of tax in the price of the goods and services he buys. Horizontal equity is then taken to require that two persons with the same income pay the same taxes (at least in the first place and ‘other things being equal’), while vertical equity would require that, of two individuals with different incomes, the one with the larger should pay more by some correct amount.

3.10. Even when income is so regarded, there remain very great difficulties in finding an exact and workable definition of it for tax purposes, as the length of the Income Tax Assessment Act and its frequent amendment testify. The most important of these difficulties are surveyed in later chapters. There is evidently some discrepancy between the legal approach to this problem, which seeks to extend and refine the everyday meaning of ‘income’, and the more abstract approach adopted by many economists which generates a very much wider meaning.

3.11. A further problem arises over the question of the appropriate unit for tax purposes. Many would argue, for example, that in a family the ‘economic well-being’ of individual members is likely to be measured better either by the average or by the total income of the family members than by their separate incomes. Whether or not this is so is an extremely vexed question and is discussed in Chapter 10.

3.12. Irrespective of whether individual or family income is accepted as the appropriate starting-point, it is recognised that this cannot be the end of the matter. Further argument is required before it can be concluded that two individuals should pay identical taxes simply because their incomes are the same. They will certainly be dissimilar in a great many other respects. Some of these differences would be considered irrelevant for tax purposes by almost everyone (for example, sex, race, religion, and many other personal characteristics), but others (such as health and size of family) are widely felt to be very relevant indeed. But to decide which differences are which, and how much allowance for such things should be given in calculating tax liabilities must involve much nice judgment and many decisions of an arbitrary kind. The main problems are examined in Chapter 12.

3.13. Yet other problems arise because taxes are, in most cases, calculated and levied on the basis of annual magnitudes. This is of course necessary for the administration of the public sector, but the period of a year is arbitrary from the viewpoint of fairness. Those with incomes that vary from year to year will over several years pay more tax under a progressive system than others earning the same total income in an even stream. This aspect of a very fundamental issue is widely recognised, and opportunities for ‘averaging’ are already provided for some groups for whom fluctuations are especially marked and unavoidable. But the deeper issues that arise here are not often thought about in explicit terms. In the Committee's view most people would agree on reflection that fairness mainly requires that taxation be the same for


  ― 14 ―
individuals whose total ‘well-beings’ are likely to be the same over their whole lifetimes and that too much importance should not be attached to temporal differences in the way ‘well-being’ is distributed, by choice or necessity, over adult years. Indeed, more than a single lifetime is relevant when the fairness of taxation upon an individual's capacity to do his duty to his heirs is considered. This lifetime perspective is not, of course, a practicable basis for taxation, but it has its implications for the structure of an inevitably annual system.

3.14. The custom of using income as the tax base is not inconsistent with this lifetime view of things. Over any short period, say a year, an individual's consumption undoubtedly reflects his ‘economic well-being’ better than does income. A man can, by consuming out of past savings or by borrowing against future income, achieve levels of ‘economic well-being’ much greater than his current income. For many, over a lifetime, total income and total consumption will be the same. In this sense, lifetime income will be a fairly good measure of an individual's ‘well-being’, whichever base is felt to be the fairer.

3.15. But a tax on income is not at all the same as a tax on consumption, even in lifetime terms. When a tax is levied on income, it falls on savings as well as consumption; and when income is earned from those savings in later years, that income too is taxed. This means that the effective tax rate imposed on consumption which is postponed is greater than the rate imposed on current consumption: income tax falls more heavily than consumption taxes on those who prefer to save a high proportion of their incomes and do more of their consumption later in their lives. But individuals may save for reasons other than to supplement consumption in later years: because they wish to bequeath to their heirs, or because the process of accumulating wealth yields ‘satisfactions’ which contribute to ‘well-being’ directly (and independently of any income wealth may bring in). Such motives for saving will be more significant at the upper end of the income scale; but certainly at the bottom end people will primarily save in order that they may consume more in future years (specifically in their old age). For the latter persons, taxation based on consumption is probably fairer than income taxation because it does not discriminate between individuals according to how they spread their consumption over their lifetimes. Higher up, horizontal equity may well be held to require not only taxation of income but taxation of capital as well, and this quite apart from any desire to make the system fairer in a vertical sense.

3.16. Questions of equity are complicated by inflation, as is so much else. The discussion of equity starts from the idea that taxation is a sacrifice of ‘real’ private satisfactions (however much this may be offset by the satisfactions that public expenditure may simultaneously create). But taxes have to be levied in money terms. In the case of incomes from current effort, the distinction may not much matter since incomes and prices may go up simultaneously and tax rates can be adjusted to maintain the same underlying rate structure. However, a serious difficulty arises with property incomes. The real value and the real return on assets with variable money returns may indeed be maintained by rises in their money capital value and their money returns. But the real return on fixed-interest assets declines and their real capital values fall; until redemption is near, their money capital values also fall. With an income tax this latter fall is inadequately reflected in the tax liability. There is a consequent inequity of a horizontal kind; and since fixed-interest securities are often the main financial assets of lower income groups, there may be vertical inequities too. These inequities do not arise so conspicuously—if at all—under consumption taxation.




  ― 15 ―

3.17. There are good reasons, therefore, why consumption might be preferred to income as the primary index of ability to pay. Some economists, notably Lord Kaldor, have found the reasons sufficiently compelling to justify abandoning personal income tax altogether and substituting a progressive expenditure tax. The Committee is not prepared to go this far, recognising as it does that an expenditure tax would probably be even more difficult to administer than personal income tax. But the philosophy underlying an expenditure tax has much to recommend it, and the Committee has allowed itself to be influenced by this philosophy in certain of its proposals, including those bearing on the tax treatment of income-protection insurance premiums (Chapter 7), investment income (Chapter 9), and superannuation and life insurance (Chapter 21).

3.18 The issues just mentioned relate primarily to horizontal equity. Problems of an equally intractable kind arise with vertical equity. Indeed, establishing the ‘right’ degree of progressivity by reference to the criterion of equity is so fundamental to tax policy, as regards both the set of taxes to be chosen and their rates, that it will need to be explored at length in the next chapter.

Simplicity

3.19 After equity, simplicity is perhaps the next most universally sought after of qualities in individual taxes and tax systems as a whole: like fairness it is a word that, in this context, points to a complex of ideas.

3.20 Two of these are explicitly stated in the Committee's terms of reference. A tax will be called simple, relatively to others, if for each dollar raised by it the cost of official administration is small, and if the ‘compliance costs’, the costs in money and effort of all kinds to the taxpayer, are also small. These two ideas are of course connected, and add up to much the same as the ancient canon of certainty. Both costs will be the less if assessor and assessed can each establish with certainty what is due: uncertainty entails the costs of consultation with experts and sometimes the yet greater costs of litigation. Both kinds of cost are increased, and certainty is endangered, when a tax, whether in the interests of equity or of efficiency, requires the drawing of fine distinctions between what is and what is not liable, and when these distinctions involve such uncertain ideas as ‘purpose’ or ‘value to the recipient’. Then the legal definitions get longer and longer and beyond the comprehension of those untrained in the law, and the relevant facts in particular cases become more and more disputable.

3.21. Two further aspects of simplicity require specific mention here. First, when (as is often unavoidable) a quite complex operation is needed before the administrators can make the assessment or the taxpayer can ascertain his liability, it is desirable that the tax be such that the taxpayer, for private purposes unconnected with tax, already needs to perform such operations. A tax on company income may be fairly regarded as a simple tax if the company already calculates its income or profits on the same or very similar basis. A tax on personal income is not a simple tax if it be so structured that many taxpayers who would not otherwise wish (or without hired help be able) to keep accounts at all, have to preserve many records and learn sophisticated accounting. The point, though obvious, is often forgotten.

3.22. A second observation is perhaps even more obvious and even more frequently forgotten. The fewer, per million dollars raised, are the individuals or organisations from whom tax is collected the simpler is a taxation system. The sheikdom that can raise all the revenue it requires (and maybe much more) from a single tax on a single oil company has what is unquestionably the simplest tax system of all.




  ― 16 ―

Efficiency

3.23. The ‘economic and efficient’ use of national resources is of course a longstanding and by now almost conventional objective of public policy: the phrase was long central to the terms of reference of the Tariff Board and is now in those of the Industries Assistance Commission.

3.24. Narrowly interpreted, efficiency requires that the resources available for public use be as nearly as possible equal to the resources withdrawn from the private sector: that is, that the process by which resources are transferred involve minimal ‘waste’. One example of such waste has already been mentioned in the context of simplicity objectives: the cost of administering and complying with the tax law is a ‘deadweight’ cost to the community and ought to be minimised. Waste can however also arise where the tax system is such as to encourage individuals to substitute things they value less for things they value more, or business to continue to use or to substitute productive processes that are technically less efficient for productive processes that are more efficient. In so far as it can be presumed that, left to their own devices, individuals will spend their incomes wisely, and business will choose the most efficient means of production, the minimisation of waste requires that the tax system should not influence individual and business choices. This is the requirement that the tax system should be neutral. Thus the tax system should not interfere with the manner in which an individual spends his income by changing the relative prices of the goods he buys; it should not alter the relative rewards of the different types of work between which he has to choose, or the relative attractions of work and leisure, or the relative returns from different modes of investment; it should not alter the relative attractiveness of different types of business organisation, or the relative prices of productive resources; and it should not discriminate between different types of production.

3.25. But it cannot always be presumed that consumers will spend their incomes wisely; and even when they do, consumers and producers alike may fail to take adequate account of the effects that their activities have on others, of what economists have come to call ‘externalities’. The efficiency of the use of available resources can sometimes be improved by departures from neutrality, by government interventions to alter what would otherwise have been the outcome of private market operations. There are many such interventions by government through, for instance, tariffs, subsidies, monetary control and marketing organisations. But the tax system is an alternative instrument. Tax concessions may aim at increasing the output or consumption of goods and services the government wants to encourage. Similarly, taxes can be used to discourage the output, or increase the prices to consumers, of commodities whose unrestricted production or consumption might otherwise have harmful effects, for example on health or urban development. Taxes can also be employed to charge the users of facilities, such as roads, on the use of which it is otherwise difficult to impose a price. Discriminatory taxes of these kinds will here be called ‘efficiency taxes’.

3.26. The Committee is persuaded that neutrality should be the general aim when efficiency is under consideration. Departures should be made only in a deliberate and explicit way for proven, explicit and quantified purposes and after it had been shown that other approaches (such as regulation and subsidy) are likely to be less effective for the end chosen. In the Committee's view, when there are circumstances warranting encouragement for a field of activity by means of a reduction in tax, any such


  ― 17 ―
reductions should be granted only for a specified purpose and a limited period, thereby ensuring periodical review.

Other Objectives

3.27. Equity, simplicity and efficiency seem to the Committee the three dominant tests of merit for individual taxes and for the tax system as a whole. There are others, however. In particular, flexibility in the taxation system is a characteristic of obvious importance to economic management. Economic management or ‘stabilisation’ requires, firstly, that there be at least some taxes in the total package the rates of which can be easily raised or lowered in the light of short-run fluctuations in the level of economic activity. Secondly, it requires that these taxes be such as to operate very quickly in altering revenue yields and influencing individuals’ and firms' behaviour. Thirdly, in so far as smaller rate changes are politically more acceptable than larger ones (particularly when rate increases are called for), the taxes available should have as large an impact as possible on the level of economic activity per dollar of revenue change.

3.28. Economic growth is another objective that, in the view of some, should be deliberately and distinctly pursued in taxation policy. As a general purpose of public economic policy it has, of course, been long accepted as important, though its interpretation can be the subject of much controversy. In the context of taxation its encouragement is often taken as implying that the overall level of taxation should be kept lower than it would be otherwise—a prescription upon which its terms of reference make the Committee unable to comment. It is also however taken to indicate that, in the interests of greater investment, savings should certainly not be discouraged under the tax system and perhaps even be taxed less than considerations of equity and neutrality alone would suggest. This is an argument that will need mention in the next chapter in connection with the general question of progressivity.

previous
next