II. The Method of Taxing

23.21. As already noted, despite the theoretical merit in taxing capital gains as they accrue, the tax can only feasibly be levied upon realisation. To tax gains on an accruals basis would lead to unacceptable problems in the periodical valuation of assets and would generate severe liquidity difficulties for taxpayers. Furthermore, assets fluctuate in value and an asset that eventually gives rise to no gain may nonetheless have given rise to payments and refunds of tax during the period of ownership. Deferring the tax until realisation does, however, have a number of undesirable consequences. It can cause what is known as ‘lock-in’ whereby the taxpayer, unwilling to pay any tax before he has to, defers realising an asset for as long as possible. Moreover, the asset-holder may, by judiciously arranging to realise his gains and losses in the years in which they will secure him maximum tax advantage, be placed in a more favourable position than other taxpayers who are not able to adjust their income between years. Nonetheless the Committee recommends that the tax be levied only upon the occasion of the realisation of the asset, or upon certain occasions (discussed below) on which there is a notional or, as it is usually termed, deemed realisation.

23.22. The role of inflation in generating illusory capital gains cannot be ignored. Inflation is the factor which leads to the greatest difficulty in devising an equitable and workable capital gains tax. It was suggested in several of the submissions to the Committee that the problems associated with levying capital gains tax in inflationary conditions can be obviated by applying a suitable index to the cost price of an asset to

  ― 417 ―
allow for inflation over the period between purchase and sale, thus separating out the purely inflationary element of the gain. The idea is attractive, but there are two important objections:

  • (a) Indexation of assets will distort the measurement of gains and losses unless indexation is also applied to reduce the real value of monetary liabilities. Indeed, in the extreme example of a person who purchases an asset entirely with borrowed funds and the asset increases in value at a rate less than the rate of inflation, the application of an index will produce a capital loss whereas in reality there has been a considerable gain because the debt has declined in real terms. The application of an index to all liabilities as well as to all assets was considered by the Committee but rejected as impracticable.
  • (b) The choice of an appropriate index presents great problems and its application would involve much administrative labour.

After considering at length various forms of indexation, the Committee has come to the conclusion that such a device cannot be recommended as a general solution to the problem posed by inflation. The merits and defects of indexation should however be accorded further study, particularly in the light of the level of inflation currently being experienced in Australia.

23.23. An alternative method of allowing for inflation in the inclusion of a variable proportion of the gain in assessable income, the proportion to be included declining the longer the asset has been held. This will, in general, not make the desired correction for inflation and will usually give the opposite result to that sought. Consider, for instance, an asset purchased for $1,000 which increases in money value at a constant annual rate of 10 per cent during a period when inflation is running at 5 per cent a year. As Table 23.A shows, the net gain (after the cost price has been adjusted for inflation) becomes an increasing proportion of the nominal gain as the period of holding increases. Where an asset has increased in value at exactly the same rate as the rate of inflation, thus giving rise to a gain in money terms but no gain in real terms, the effect of the declining-proportion method would be to impose a capital levy by taxing a gain that has not in reality occurred. The rate of capital levy would not necessarily decrease as the proportion of the nominal gain that was taxed decreased but it might rise before eventually declining to zero. Where an asset has increased in value at a rate less than that of inflation, giving rise to a gain in money terms but a loss in real terms, the declining-proportion method would again impose a capital levy in addition to the capital loss sustained through inflation. As in the preceding case the rate of capital levy would eventually decline to zero, though it might rise for a time before it starts to fall.


Period of holding Years   Sale price   Nominal gain   Net gain(a)   Net gain as percentage of nominal gain  
Per cent 
1,100  100  50  50.0 
1,611  611  335  54.8 
10  2,594  1,594  965  60.5 
20  6,727  5,727  4,074  71.1 
50  117,391  116,391  105,924  91.0 

  ― 418 ―

23.24. A point in favour of the declining-proportion method of taxing capital gains is that it is less harsh than some other methods in the case of assets which have generated either real losses or no real gains, but as a general method of allowing for inflation it has no merit. In addition there would be unacceptable difficulties in devising an equitable and workable method of dealing with capital losses.

23.25. Broadly, there are two main systems in use in those countries that levy capital gains tax: the flat-rate method, which imposes a fixed rate of tax on the gain, and the proportional-inclusion method, which includes a proportion of the gain in the taxpayer's income. The former is employed in the United Kingdom while Canada has adopted the latter.

23.26. A flat-rate capital gains tax has many atractions. It is simple, certain and easily understood. Because it is divorced from the income tax system there is no incentive for the taxpayer to realise capital gains in years when his other income is low, whereas this incentive does exist with the proportional-inclusion method. For the same reason there is no way in which the tax can be minimised by diverting capital gains to other members of the taxpayer's family. It avoids the problem of bunching of gains in one income year, particularly bunching arising from a provision for deemed realisation at death, and the necessity for provisions to mitigate this. It is a simple and effective method of taxing capital gains made by non-residents; and it can be argued that it is a more neutral method of taxing gains made by companies and certain types of trusts as against gains made by individual taxpayers. In addition, since the assessment of the tax would be separate from the assessment of income tax it would be easier to calculate its revenue yield and the issue of an assessment for capital gains tax would not be delayed pending the issue of the normal tax assessment.

23.27. On the other hand, it has the drawback of lack of progressivity. If the rate of tax were set at, say, 30 per cent, then a low-income taxpayer who makes a small capital gain would pay more tax than if the gain had simply been added to his other income and taxed accordingly, while the high-income taxpayer would pay less. This inequity is recognised to some extent in the United Kingdom legislation which gives the taxpayer the option, subject to certain limitations, of including half the gain in his income. With such an option, however, much of the simplicity of this method is lost. More importantly, the tax will normally be progressive over a narrower income range than in the case of personal income tax.

23.28. After considering the alternatives, the Committee has come to the conclusion that the less unsatisfactory method is the second of those mentioned: to tax a fixed proportion of the gain by including it in income, subject to the arrangements described in paragraph 23.35–23.37 for spreading the taxable gain over a period of years. It is recognised that the use of a fixed proportion for determining the taxable elements of all capital gains is open to a number of objections. In particular it will be overly harsh in those cases where gains are largely or entirely due to inflation; and it will be excessively lenient in those cases where very large gains have been made in relation to the amount of inflation that has taken place during the period of ownership of the asset, more especially when the ownership was financed from borrowed funds. Nonetheless the Committee believes it to be an approach that should prove satisfactory in the majority of cases, and it has the advantage of ease of understanding and administration.

  ― 419 ―

23.29. The choice of the proportion of the gain to be included in taxable income is not one that can be made without reference to the actual level of inflation being experienced. It should not be regarded as fixed and immutable and should be varied if there is a significant increase or decrease in rates of inflation. For illustrative purposes only, this chapter assumes the inclusion of one-half of a capital gain in income and with capital losses being dealt with in accordance with the recommendations in paragraph 23.71. It is recognised that adjustment to the proportion to be included is a very blunt instrument for dealing with variations in the rate of inflation, but it will at least operate in the right direction. It is to be noted in this respect that the Government's proposals for the introduction of a capital gains tax involve the inclusion of half the gain in income; but the rationale for this has not been made public and in particular there has been no statement as to whether or not this half-inclusion is designed as an implicit allowance for inflation. The Committee believes that half-inclusion in circumstances where inflation is at an annual rate approaching 20 per cent is altogether too harsh. If capital gains tax is to be introduced in such circumstances, the amount of the gain to be included in income should be considerably less than half— possibly even less than a quarter.

23.30. The Committee is conscious of the fact that proposing an allowance for inflation in regard to capital gains raises questions as to similar allowances in regard to income gains, particularly when trading stock and fixed assets subject to depreciation are involved. There is also a question of the fairness of treating the whole of any interest on fixed-money return investments as income. These questions are considered in Chapters 8 and 9.