Evans, Chris --- "Curing Affluenza?: A Critique of Recent Changes to the Taxation of Capital Gains in Australia" [2000] UNSWLawJl 40; (2000) 23(2) UNSW Law Journal 299

	Home \| Databases \| WorldLII \| Search \| Feedback University of New South Wales Law Journal

It has long been accepted that taxing consumption, property and labour can be much more efficient than taxing more mobile factors such as capital. As a result, revenue authorities have increasingly built significant parts of their tax systems upon such tax bases, and often tended to downplay the more difficult area of taxing gains from capital. It is therefore somewhat surprising to find that the Capital Gains Tax (“CGT”) is, after the virtually ubiquitous Value Added Tax (or Goods and Services Tax as it has become known in Australasia and Canada), the tax that has been most widely introduced in developed economies in recent decades. Australia, in 1985, was one of the last of the Organisation for Economic Development and Co-operation (“OECD”) countries to introduce a CGT regime.^[1] Generally, arguments about the inequity of not taxing capital gains have outweighed the potential inefficiencies and complexities that inevitably accompany the introduction of such a tax.

There has, however, been a pronounced shift in the way that capital gains have been taxed in a number of countries over the last twenty years or so. The tax-cutting reforms of the late 1970s and 1980s were generally characterised by a trend in developed economies towards taxing capital gains more or less on the same basis as other income (and at considerably lower rates for both income and capital gains than had been the norm in earlier decades), with the equity issues of the taxing of capital gains very much to the fore. More recently, however, efficiency considerations have begun to predominate, and the preferential treatment given to capital gains has significantly increased.

Many countries have recently re-introduced more sharply differentiated treatment of capital gains compared to other income. For example, the UK has introduced and extended taper relief in two recent budgets (1998 and 2000), under which the amount of the taxable capital gain diminishes the longer the asset is held. In a similar vein, the USA has re-introduced separate treatment of short and long term gains (1997) (with significantly preferential treatment for the latter), and Canada has extended its CGT exclusion from one quarter to one third in its most recent Federal budget (2000).

Australia has also been part of this shift that has seen the “headline” rate of CGT cut, compared to other income. With effect from September 1999, and as a result of Treasury acceptance of the recommendations of the Review of Business Taxation (“Ralph Review”), a 50 per cent CGT discount has been introduced for individuals and trusts^[2] (with a smaller, one third, CGT discount for qualifying superannuation entities). As a result, for many individuals and trusts, the effective rate of CGT has been nearly halved.^[3] At the same time the CGT small business reliefs have been considerably expanded, such that it would be an extremely ill-advised or unfortunate small business taxpayer that faced any CGT liability in respect of the disposal of active business assets. And the introduction of the scrip for scrip roll-over relief from 10 December 1999 affords those taxpayers who would have otherwise encountered a CGT liability when their shares were exchanged for those of another entity (for example, in a takeover), the opportunity for indefinite postponement of that liability.

This paper argues that many of these recent changes to the CGT regime do nothing for the integrity of the Australian tax system. They flaunt many of the essential principles upon which good tax design should be based, and are unlikely to satisfy the criterion of revenue neutrality that was supposed to be a key feature of the Ralph Review reforms. Above all, they destroy one of the key features that has given strength to the Australian tax system in the period since 1985 - the notion that (give or take a few concessions) a dollar of income would be taxed on more or less the same basis no matter what its source - whether derived as salary from employment, or received as investment income from capital, or realised as capital gain from the disposition of an asset. In attempting to treat some ill-defined symptoms of affluenza in the wealthier sectors of society, the government may have inadvertently reintroduced the prospect of an outbreak of the far worse plague that infected the Australian tax system in the period before the introduction of the CGT in 1985.

II. EQUITY, EFFICIENCY AND SIMPLICITY

There are aspects of the recent and proposed CGT reforms that do represent sound policy initiatives and which will do much to reduce the burden of compliance costs, help business growth and otherwise stimulate sensible economic activity. The removal of indexation and averaging, aspects of the rationalisation of the small business CGT reliefs and perhaps the introduction of the scrip for scrip roll-over relief may be placed in that category. But these positive impacts are far outweighed by the negative policy implications that will arise from the central feature of the CGT changes - the elimination of 50 per cent of the charge to tax when individuals make capital gains. This preferential treatment can be criticised by reference to the usual criteria for assessing tax policy initiatives - equity, efficiency and simplicity.

The essential reason for introducing regimes for taxing capital gains is one of equity. Put simply, the gain arising from the disposition of a capital asset is income, and deserves to be taxed as income. In the sentiments of the Canadian Carter Report^[4] a “buck is a buck is a buck”, and the fact that an individual or other entity derives an increase in economic power as a result of holding property rather than, say, through operating a business or working in an office, is immaterial. The introduction of the 50 per cent CGT discount savagely offends both the horizontal and the vertical aspects of equity.

For example, how does the concept of horizontal equity sit with the employee who earns $1 million over 15 years and sees nearly half of it disappear in tax, compared to the shrewd property or share market investor who realises a $1 million gain over the same period, but who only pays less than a quarter of that amount in tax? Or how does the principle of vertical equity match up with the small business taxpayer who pays no tax on a $1 million capital gain with the taxpayer who pays nearly $5,000 tax on a gain of $10,000 made as the result of one shrewd (and economically efficient) investment in Telstra shares over an 11 month period? Moreover, despite assurances that we are increasingly becoming a share owning democracy, we need to remember that most of the population who own shares own very few shares. The capacity to make the largest savings of tax on the largest capital gains is reserved for a very small minority of the population - as any survey of Taxation Statistics will rapidly reveal. Cutting tax rates sounds inherently attractive. But when the vast bulk of the benefit is reserved for, or primarily claimed by, just one small sector of society, the appeal is less resounding.

Efficiency is claimed to be the primary rationale for the introduction of the discount. With capital gains taxed at marginal income tax rates, the economic incentives for investment are destroyed, it is claimed, and the market cannot operate in an efficient fashion. Reducing the effective rate of capital gains taxation by introducing the CGT discount will, on this analysis, free the market from its chains and permit taxpayers to unlock themselves from inefficient investments and chase the best dollar.

But this supposed freeing up of the market and removal of economic distortion itself gives rise to a number of questions. Is what is proposed efficient (or more efficient than it was)? What does it tell us about the need to transform highly taxed income into preferentially taxed capital gains? Will these proposals lead to distortionary investments and economic inefficiencies? Will lock in and bunching disappear, or do we have to look at the one year trigger^[5] and the 15 year trigger^[6] very carefully? What will it do for speculative investment and negative gearing? Is the tax tail going to wag the commercial dog all over again as we seek to obtain the best arbitrage outcomes possible as a result of the increased wedge driven between the rates at which we tax income and those at which we tax capital? It is difficult to believe that the introduction of the CGT discount can lead to a more efficient outcome than was the case when all forms of income were taxed at roughly the same rate.

So far as the criterion of simplicity is concerned, it is certainly questionable that the CGT changes will do very much to produce a simpler CGT regime than the one we already have. But simplicity is not usually a benchmark criterion for CGT, as the Asprey Committee long since recognised, when it concluded that a capital gains tax could not be justified on the grounds of simplicity. It noted:

Perhaps, rather than simplicity, we should at least be striving for greater certainty. Sadly, it is difficult to see that these proposals make the CGT regime more certain than was the case before.

III. REVENUE NEUTRALITY

One of the cornerstones upon which the proposed changes to the CGT regime (as well as other aspects of tax reform) are being sold to the Australian community is the broad revenue neutrality of the package. It is suggested that despite there being some equity issues, it is necessary to look at the “big picture” of the package as a whole. This section of the paper challenges the assumption that the CGT changes are self-financing, and provides alternative figures which show that there could be an adverse revenue impact from the proposed CGT changes.

The Ralph Review suggests that, overall, the CGT proposals taken as a package will generate additional revenue of $350 million over the five year period from 2000-01 to 2004-05.^[8] This comprises a net revenue inflow of $420 million in total in the first four years, and a net revenue outflow of $70 million in the final year for which a forecast is made. The reliability of these figures is open to question.

• the positive impact on revenue received by the Government as a result of the removal of averaging and the freezing of indexation, which totals roughly $5 billion over the five years;

• the revenue leakage that will occur as a result of taxpayers converting ordinary income to capital gains. This is estimated to total $500 million over the five years; and

• the impact on revenues of excluding half of the capital gains realised by individuals and one third of the capital gain realised by superannuation funds (which is a static cost), netted off against the dynamic impact of increased revenues as a result of extra realisations brought about by the cut in CGT rates. This accounts for a (net) revenue loss of just over $4 billion over the five years.

The first element - the estimates of increased revenue as a result of removing averaging and freezing indexation - appears to be a reasonable figure. The second and third elements (the conversion of income to capital and the impact of increased realisations) are more problematic, as they involve measurement of behavioural responses to future tax changes which cannot easily be quantified.

The propensity to convert income to capital will be a function of two important factors: the incentive to convert and the capacity to convert. A reduction of 50 per cent in the rate at which capital gains are taxed represents a very strong incentive to convert, notwithstanding the removal of averaging and the freezing of indexation. Individual taxpayers on the highest marginal rates will clearly have the greatest incentive, but it can be expected that all individuals will seek to take advantage of halved rates. Ironically, the abolition of averaging will cause asset wealthy family units where only one family member is earning income to seek to ensure that both the earning spouse and the non-earning spouse take full advantage of capital gains - previously the incentive was only there for the non-earning spouse.

The capacity to convert income to capital will clearly depend upon the response of the tax planning profession to the new opportunities afforded by the significant preference now to be given to capital gains, and the ability of the revenue authority to challenge such strategies. The pre-1985 Australian experience suggests that the profession will quickly identify a number of such opportunities, and the recent proliferation of salary packaging involving shares and share options is likely to significantly increase. Limiting the types of capital gains that can attract the halved rates (which appears to be one proposed response to the new threats to the revenue base) may lead, once again, to the sorts of boundary disputes over capital/income that the Ralph Review is so keen to remove.

It seems highly contentious that the arbitrage effect related to the conversion of income to capital will be as low as only $500 million over five years, which is the figure given in the Ralph Review. There is very little detail in the Ralph Review to show how the figures of only $20 million leakage in the first year, rising to $180 million in the fifth year, were arrived at. The only reference to this figure seems to be a note to the effect that:

In addition to the impact upon dividend receipts, there is also likely to be an impact upon Pay-As-You-Earn (“PAYE”) receipts, as some directors and employees seek to take salary and salary increases in a form that attracts capital gains taxation rather than income taxation (for example, through receiving shares and options over shares rather than salary).

The Commissioner's Report for 1998-99 (Table 3.1) shows gross revenue collections in 1998-99 through the PAYE system of $71.8 billion. Table 1 of this paper shows the impact on PAYE tax receipts of various levels of conversion of income to capital. It shows that if individual taxpayers were successful in having just 0.5 per cent of their income currently subject to PAYE treated as concessionally taxed capital gains, the revenue leakage in one year (at 1998-99 levels) would be $179 million dollars. Over five years the loss would be $895 million dollars. In a worst case scenario where individual taxpayers succeeded in transforming 2.5 per cent of income currently taxed at full rates to capital gains taxed at half rates, the annual threat to the revenue could be $897 million, or nearly $4.5 billion in the five year period.

Table 1: Impact on PAYE Receipts of Converting Income to Capital^[10]

Per cent of PAYE receipts converted from income to capital	0.5%	1.0%	1.5%	2.0%	2.5%
Impact (negative) upon PAYE receipts ($m): one year	-179	-359	-538	-718	-897
Impact (negative) upon PAYE receipts ($m): five years	-895	-1,795	-2,690	-3,590	-4,485

In addition to the impact of conversion of income to capital, there is also the effect on revenue of increased realisations (the dynamic benefit) weighed against the static cost of lost revenue, both as a result of the effective cut in the CGT rates. The problematic area here is the estimation of enhanced revenue as a result of changed taxpayer behaviour induced by the cut in rates. The lower rates will, it is argued by proponents of the cuts, lead to an increase in the realisation of assets.

In the literature this is normally expressed in terms of elasticity - in this context the responsiveness of capital gains realisations to changes in the CGT rate. An elasticity of zero would imply that taxpayer behaviour is unaffected by taxes. An elasticity of minus 1 would imply that the loss of revenue as a result of the cut in the tax rate would be matched by the increase in revenue as a result of the extra realisations. And an elasticity above minus 1 would imply that realisations would increase more than enough to make the additional tax revenue on the extra realisations outweigh the direct effect of the reduction in rates.

In the Ralph Review the assumption is made that the elasticity would be minus 1.7 in the first two years, and minus 0.9 in the next three years. The research upon which these elasticities are based is a study commissioned from Alan Reynolds of the US Hudson Institute in July 1999 by the Australian Stock Exchange Ltd for the Review of Business Taxation. The contentions of Reynolds and the Hudson Institute in this area have been hotly contested by a number of commentators, including economists from the US Treasury and the Brookings Institution. Leonard Burman, a deputy assistant secretary for tax analysis at the US Department of the Treasury, notes in a recent publication that:

In October 1999 the Senate referred aspects of the Business Tax Reform proposals to the Senate Finance and Public Administration References Committee. That committee heard expert evidence from a number of witnesses from Australia and overseas which confirmed that the Ralph Review's suggested elasticities were highly optimistic and improbable. In particular, two prominent US economists^[12] suggested that the short term elasticity would be less than 1 and longer term only between 0.2 and 0.25.

This does not imply that Burman, Gravelle and Auerbach are right and Reynolds is wrong. This particular debate has not been satisfactorily resolved over the last 20 years, and probably never will be. But it does seem dangerous to base predicted outcomes on one perception and to ignore others. Even a small change in the elasticity will throw the figures out significantly.

Table 2 illustrates the sensitivity of the Treasury estimates to changes in the assumptions of elasticity. In the first place, the elasticities have been varied to minus 0.85 for the first two years and minus 0.45 for years three to five. This is half the elasticity rates adopted in the Ralph Review, and is referred to in the Table as the “Burman option”, as it more closely reflects his perception of the likely behavioural responses. As an alternative, the second option (referred to as the “Midway option”) measures the outcomes on the basis of adopting elasticities that are the average of the Burman and Ralph Review elasticities. For completeness, the figures shown in the Ralph Review are included in the Table, as the “Ralph option”.

Table 2: Impact on Revenue of Increased Realisations with Alternative Elasticities

	2000-01 to 2004-05 $m
Burman option (elasticities of minus 0.85 in years 1 and 2, and minus 0.45 in years 3 to 5)	1,340
Midway option (elasticities of minus 1.275 in years 1 and 2, and minus 0.675 in years 3 to 5)	2,010
Ralph option (elasticities of minus 1.7 in years 1 and 2, and minus 0.9 in years 3 to 5)	2,680

Under the Ralph option the positive revenue impact of extra realisations as a result of the proposed CGT changes amounts to $2.68 billion over five years. Under the Burman option the positive impact is reduced to $1.34 billion, whilst the midway option produces additional revenue of just over $2 billion.

Table 3 combines the information in Table 1 with that in Table 2. It shows that over the five year period the impact on CGT revenues could be significantly different from that shown in the Ralph Review, depending on which assumptions are used.

Table 3: Overall CGT Revenue Impact (Aggregate over Five Years)

Income/capital conversion assumption	Aggregate revenue impact of CGT proposals $m
	Burman elasticity option	Midway elasticity option	Ralph elasticity option
Ralph	-990	-320	350
0.5%	-1,885	-1,215	-545
1.0%	-2,785	-2,115	-1,445
1.5%	-3,680	-3,010	-2,340
2.0%	-4,580	-3,910	-3,240
2.5%	-5,475	-4,805	-4,135

The message from Table 3 is clear. Varying the assumptions on the propensity for converting income to capital, and/or varying the assumptions about the likely responsiveness of CGT realisations to cuts in the rate of tax, will have significant revenue effects. In a worst case scenario, rather than being revenue positive to the extent of $350 million as suggested by the Ralph Review, there could be a revenue loss of up to $5.475 billion. In one of the best case scenarios shown in Table 3, where income/capital conversions only represent 0.5 per cent of gross PAYE receipts and elasticities are assumed to be midway between the Ralph Review and the Burman estimates, there is revenue leakage of over $1.2 billion over the five year period. And even if the Ralph Review estimates of income/capital conversions were accepted, there would still be significant revenue leakage of nearly $1 billion if the Burman prognosis on elasticities does prove to be correct.

IV. CONCLUSIONS

The Government has gone further in dismantling the CGT regime in Australia than the Ralph Review initially anticipated. The early Ralph Review discussion papers raised the possibility of some concessions for CGT, such as the introduction of taper relief on the UK model, the introduction of an annual CGT exempt amount (again as used in the UK), or the capping of the CGT rate at 30 per cent. The ultimate proposal of exempting half of the gain came as a surprise to most, and many suspect that the decision owed more to political considerations than to sound policy. Sadly - and ironically - the introduction of the CGT discount has undermined one of the very few aspects of the Australian CGT regime that added integrity to the tax system - the fact that all forms of income, including capital gains - were taxed at more or less the same rate of tax.

There is no doubt that in many circumstances Australia's CGT was - before 21 September 1999 - more onerous than that of many international competitors. Modelling^[13] showed that an unmarried individual on twice average annual earnings who made a capital gain selling US$40,000 worth of shares in August 1998 would face an effective tax rate of 47 per cent (ignoring Medicare) if the gain occurred in Australia, compared to 38 per cent in Canada, 20 per cent in both the UK and USA, 19 per cent in Ireland and 0 per cent in New Zealand. The tax payable on that capital gain would amount to US$15,801 in Australia, compared to an average of only US$8,069 for all six countries mentioned. Clearly there is cause for concern if one looked at only the stark figures shown by these comparisons.

But the stark figures mask complications. Australia (largely by the accident of its CGT design) was the only one of those six countries that would have charged the same amount of tax (in the scenarios modelled) regardless of whether the gain was treated as a capital gain or as “ordinary” income. And that was a fundamental principle of which Australia should have felt proud rather than immediately rushing off to reduce the tax payable on capital gains. Until 21 September 1999 it treated all gains - all accretions to net wealth, whether income gains or capital gains - on the same basis.

Of course, it would be better for international competition if Australia's tax rates - on income and capital - were not as high, but to have reduced one and not the other will be to deal a body blow to the major reason that capital gains tax regimes exist: to ensure horizontal equity in the tax system. Australia was one of the last OECD countries to introduce a CGT, and the primary reason for its introduction was to prevent the blatant unfairness that had hitherto existed whereby a taxpayer (usually wealthy) could enjoy tax-free capital gains whilst other taxpayers were fully taxed on their income.

Equity is not the only consideration in the design of the tax system. Efficiency and simplicity are also important criteria. Indeed, the recent clamour for reform was largely driven by those who considered that the pre-existing CGT regime in Australia distorted economic decision making and acted as a disincentive to investment. But the evidence for this was flimsy at best and did not constitute a sufficiently strong case for destroying the integrity of the CGT regime as it existed prior to 21 September 1999.

So what shape should reform have taken, or should it take in the future? The essential features of the reform of capital gains must lie in tackling any inefficiencies that do exist and reducing the complexities of the current regime without impugning its equity. Interfering with capital gains tax rates, which will undermine the equity basis on which the regime is established, will not do this. Capital gains should be charged to tax at prevailing income tax rates, and, in recognition of this, arguably capital losses should be available for set off against all income (subject to certain anti-avoidance provisions designed to prevent taxpayers taking advantage of obvious timing differences), not quarantined against capital gains. There should not be any distinction between short term and long term capital gains (for example, with taper relief) as different treatment of such gains causes difficulties at the margin, encourages the “lock-in” effect, significantly adds to the complexity of the regime and undermines the fundamental principle that all gains are income and should be treated as such.

It is difficult to justify the introduction of the CGT discount and other changes on any tax policy grounds. The changes have considerably reduced the equity of the tax system, are of dubious benefit or value on efficiency grounds, and do nothing for the simplicity of the Australian CGT regime. Moreover, they are likely to cost the public purse considerable sums of money, contrary to claims of revenue neutrality. Cynically viewed, the changes are an exercise in good politics rather than good tax policy. They have been sold (and have been largely accepted) as a sop to the growing numbers of shareholders who fear that the tax system may strip them of their easily won gains from volatile and burgeoning stock markets.^[14] In reality, they provide relatively insignificant relief for the vast bulk of share investors (many of whose gains have proved to be illusory), whilst furnishing significant concessions to the wealthiest sector of society.

^[$] Af-flu-en-za . 1. The bloated, sluggish and unfulfilled feeling that results from efforts to keep up with the Joneses. 2. An epidemic of stress, overwork, waste and indebtedness caused by dogged pursuit of the American dream. 3. An unsustainable addiction to economic growth. Available at <www.pbs.org/kcts/affluenza/index.html>.

^[*] BSc (Hons) Econ (University of London), MA (Leicester University), Associate Professor of Taxation and Associate Director of the Australian Taxation Studies Program ("ATAX"), UNSW.

[2] Though trusts will lose the benefit of this discount for assets acquired on or after 23 December 1999 where the asset is disposed of on or after 1 July 2001: Australia, Review of Business Taxation, A Tax System Redesigned: More Certain, Equitable and Durable, July 1999 (“Ralph Review”).

[3] The freezing of indexation at 30 September 1999, and the removal of averaging from 21 September 1999, has the effect of clawing back part of the tax cut. But, in a time of relative low inflation, the claw back is not significant.

[6] Where a small business taxpayer has held an active asset for at least 15 years, complete exemption of the capital gain may be available.

[11] LE Burman, The Labyrinth of Capital Gains Tax Policy: A Guide for the Perplexed, Brookings Institution Press (1999) p 59.

[12] Dr Jane Gravelle and Professor Alan Auerbach, both cited in Senate Finance and Public Administration References Committee, Inquiry into Business Taxation Reform, November 1999, particularly at 30.

[13] C Evans, C Sandford, “CGT - The Unprincipled Tax” (1999) 5 British Tax Review.

[14] There are grounds for believing that this section of the population may be suffering the delusional symptoms of sudden wealth syndrome - a condition that can rapidly disappear in response to fluctuations on the ASX and NASDAQ.

University of New South Wales Law Journal