Report to the Treasurer and Minister of revenue - by a committee of experts on tax compliance

Bill Birch Treasurer

Report to the Treasurer and Minister of revenue

By a committee of experts on tax compliance

Chapter 6 -
Tax Mitigation, Avoidance and Evasion


6.1 Two general points can be made about tax avoidance. First, tax avoidance occurs across the tax spectrum and is not peculiar to any tax type. Secondly, legislation that addresses avoidance must necessarily be imprecise. No prescriptive set of rules exists for determining when a particular arrangement amounts to tax avoidance. This lack of precision creates uncertainty and adds compliance costs.

6.2 This chapter focuses primarily on income tax avoidance. However, this focus does not imply that the same concerns do not apply in the context of other taxes. Indeed, section 76 of the Goods and Services Act 1985 provides an anti-avoidance provision that was drafted broadly along the lines of section 99 of the Income Tax Act 1976.

6.3 The committee did not have the time or resources to consider issues of tax avoidance outside those provisions applying to income tax. However, where the corresponding anti-avoidance provisions that relate to other taxes draw on similar principles, it may be taken that the committee's views are likely to extend to that context. This statement, however, should not be understood to mean that the committee has reviewed section 76 of the Goods and Services Tax Act 1985, nor that it has reached any conclusions on its current drafting.

Income tax avoidance

6.4 In the Income Tax Act 1994, tax avoidance is an imprecise and difficult concept. This chapter focuses on aspects of that concept, the way it is interpreted, and the way it is enforced.

6.5 The first part of the chapter describes the distinctions drawn by the courts between tax mitigation, avoidance and evasion. The committee then presents a policy framework for considering issues relating to tax avoidance, and proposes some amendments clarifying the operation of the general anti-avoidance rule in sections BG 1 and GB 1 of the Income Tax Act 1994. The committee discusses the department's policy statement (published in 1990) on the general anti-avoidance provision, and then turns to a more general consideration of departmental public statements, focusing on the exposure draft on form and substance in taxation law. In the next section, the committee looks at the use of the loss attributing qualifying companies rules for tax planning purposes. The intention of the rules was to eliminate tax considerations as an element in determining the appropriate structure for closely held businesses, but their use appears to have been extended beyond this purpose. A particular type of scheme is then examined involving contrived depreciation deductions, which can take advantage of the loss attributing qualifying companies rules. The committee evaluates the scheme as it provides an excellent example of the general points made in the report.

Definitions of Tax Mitigation Avoidance and Evasion

6.5 It is impossible to express a precise test as to whether taxpayers have avoided, evaded or merely mitigated their tax obligations. Tax avoidance is described in the Income Tax Act 1994 by reference to its intended fruits, but giving meaning to the terms of the description is ultimately a matter of judgment for the courts. As Baragwanath J said in Miller v CIR; McDougall v CIR:

What is legitimate 'mitigation' and what is illegitimate 'avoidance' is in the end to be decided by the Commissioner, the Taxation Review Authority and ultimately the courts, as a matter of judgment.

Tax mitigation

T6.6 axpayers are entitled to mitigate their liability to tax and will not be vulnerable to the general anti-avoidance rule in section BG 1 if they do so. A description of tax mitigation was given by Lord Templeman in CIR v Challenge Corporate Ltd:

6.7 Income tax is mitigated by a taxpayer who reduces his income or incurs expenditure in circumstances which reduce his assessable income or entitle him to reduction in his tax liability. Section 99 [section BG 1] does not apply to tax mitigation because the taxpayer's tax advantage is not derived from an 'arrangement' but from the reduction of income which he accepts or the expenditure which he incurs.97

6.8 Tax mitigation is, therefore, behaviour which, without amounting to tax avoidance, serves to attract less liability than otherwise might have arisen.

Tax avoidance

6.9 The general anti-avoidance rule in section BG 1 provides that a tax avoidance arrangement is void for income tax purposes. Tax avoidance, as Lord Templeman has pointed out, is not mere mitigation.98

6.10 The term is described in section OB 1 as, directly or indirectly:

  1. altering the incidence of any income tax,
  2. relieving any person from liability to pay income tax,
  3. avoiding, reducing or postponing any liability to income tax.

On an excessively literal interpretation, this approach could conceivably apply to mere mitigation, for example, to an individual's decision not to work overtime, because the additional income would attract a higher rate of tax. However, a better way of approaching tax avoidance is to regard it as an arrangement that, unlike mitigation, yields results that Parliament did not intend, unless nullified by section BG 1.99

6.11 In Challenge Corporation Ltd v CIR, Cooke J described the effect of the general anti-avoidance rules in sections BG 1 and GB 1 in these terms:

[It] nullifies against the Commissioner for income tax purposes any arrangement to the extent that it has a purpose or effect of tax avoidance, unless that purpose or effect is merely incidental... Where an arrangement is void... the Commissioner is given power to adjust the assessable income of any person affected by it, so as to counteract any tax advantage obtained by that person.

6.12 As Baragwanath J noted in the Miller case, 'Parliament has deliberately left [section BG 1] open textured'100. Woodhouse J commented on the breadth of the general anti-avoidance rule in the Challenge Corporation case, noting that Parliament had taken:

The deliberate decision that because the problem of definition in this elusive field can not be met by expressly spelling out a series of detailed specifications in the statute itself, the interstices must be left for attention by the judges.101

Tax evasion

6.13Mitigation and avoidance are concepts concerned with whether or not a tax liability has arisen. With evasion, the starting point is always that a liability has arisen. The question is whether that liability has been illegitimately, even criminally, left unsatisfied. In CIR v Challenge Corporation Ltd, Lord Templeman said:

Evasion occurs when the Commissioner is not informed of all the facts relevant to an assessment of tax. Innocent evasion may lead to a re-assessment. Fraudulent evasion may lead to a criminal prosecution as well as re-assessment.102

6.14 The elements which can attract the criminal label to evasion were elaborated by Dickson J in Denver Chemical Manufacturing v Commissioner of Taxation (New South Wales):

An intention to withhold information lest the Commissioner should consider the taxpayer liable to a greater extent than the taxpayer is prepared to concede, is conduct which if the result is to avoid tax would justify finding evasion.103

6.15 Not all evasion is fraudulent. It becomes fraudulent if it involves a deliberate attempt to cheat the revenue. On the other hand, evasion may exist, but may not be fraudulent, if it is the result of a genuine mistake. In order to prove the offence of evasion, the Commissioner must show an intent to evade by the taxpayer. As with other offences, this intent may be inferred from the circumstances of the particular case.104

6.16 Tax avoidance and tax mitigation are mutually exclusive. Tax avoidance and tax evasion are not: they may both arise out of the same situation. For example, a taxpayer files a tax return based on the effectiveness of a transaction which is known to be void against the Commissioner as a tax avoidance arrangement.

6.17 A senior United Kingdom tax official recently referred to this issue:

If an 'avoidance' scheme relies on misrepresentation, deception and concealment of the full facts, then avoidance is a misnomer; the scheme would be more accurately described as fraud, and would fall to be dealt with as such. Where fraud is involved, it cannot be recharacterised as avoidance by cloaking the behaviour with artificial structures, contrived transactions and esoteric arguments as to how the tax law should be applied to the structures and transactions.105

Tax Avoidance in a Policy Framework

6.18 This chapter now turns from the existing legal framework in the context of income tax to a possible policy framework for considering issues relating to tax avoidance generally. The questions considered relevant to a policy analysis of tax avoidance are:

  1. What is tax avoidance?
  2. Under what conditions is tax avoidance possible?
  3. When is tax avoidance a 'policy problem'?
  4. What is a sensible policy response to tax avoidance?
  5. What is the value of, and what are the limitations of, general anti-avoidance rules?

What is tax avoidance?

6.19 Finance literature may offer some guidance to what is meant by tax avoidance in its definition of 'arbitrage'. Arbitrage is a means of profiting from a mismatch in prices. An example is finding and exploiting price differences between New Zealand and Australia in shares in the same listed company. A real value can be found in such arbitrage activity, since it spreads information about prices. Demand for the low-priced goods increases and demand for the high-priced goods decreases, ensuring that goods and resources are put to their best use. Tax arbitrage is, therefore, a form of tax planning. It is an activity directed towards the reduction of tax.

6.20 It is this concept of tax arbitrage that seems to constitute generally accepted notions of what is tax avoidance. Activities such as giving money to charity or investing in tax-preferred sectors, would not fall into this definition of tax arbitrage, and thus would not be tax avoidance even if the action were motivated by tax considerations.

6.21 The committee has noted that financial arbitrage can have a useful economic function. The same may be true of tax arbitrage, presuming that differences in taxation are deliberate government policy furthering economic efficiency. It is possible that tax arbitrage directs resources into activities with low tax rates, as intended by government policy. It is also likely to ensure that investors in tax-preferred areas are those who can benefit most from the tax concessions, namely, those facing the highest marginal tax rates. If government policy objectives are better achieved, tax arbitrage is in accordance with the government's policy intent. Tax avoidance, then, can be viewed as a form of tax arbitrage that is contrary to legislative or policy intent.

What makes tax avoidance possible?

6.22 The basic ingredients of tax arbitrage are the notion of arbitrage, and the possibilities of profiting from differentials that the notion of arbitrage implies. This definition leads to the view that three conditions need to be present for tax avoidance to exist.

  1. A difference in the effective marginal tax rates on economic income is required. For arbitrage to exist, there must be a price differential and, in tax arbitrage, this is a tax differential. Such tax differences can arise because of a variable rate structure, such as a progressive rate scale, or rate differences applying to different taxpayers, such as tax-exempt bodies or tax loss companies. Alternatively it can arise because the tax base is less than comprehensive, for example, because not all economic income is subject to income tax.106
  2. An ability to exploit the difference in tax by converting high-tax activity into low-tax activity is required. If there are differences in tax rates, but no ability to move from high to low-tax, no arbitrage is possible.
  3. Even if these two conditions are met, this does not make tax arbitrage and avoidance possible. The tax system may mix high and low-rate taxpayers. The high-rate taxpayer may be able to divert income to a low-rate taxpayer or convert highly-taxed income into a lowly-taxed form. But this is pointless unless the high-rate taxpayer can be recompensed in a lowly-taxed form for diverting or converting his or her income into a low-tax category. The income must come back in a low-tax form. The benefit must also exceed the transaction costs. This is the third necessary condition for tax arbitrage.

6.23 Since all tax systems have bases that are less than comprehensive because of the impossibility of defining and measuring all economic income, tax arbitrage and avoidance is inherent in tax systems.

Examples of tax arbitrage/avoidance

6.24 The simplest form of arbitrage involves a family unit or a single taxpayer. If that family unit or taxpayer faces differences in tax rates (condition 1 above), and condition 2 above applies, then the third condition automatically holds. This conclusion follows because people can always compensate themselves for converting or diverting income to a low tax rate.

6.25 An example of such simple tax arbitrage involving a family unit is income splitting through, for example, the use of family trust. An example of simple tax arbitrage involving a single taxpayer is a straddle whereby a dealer in financial assets brings forward losses on, say shares, and defers gains while retaining an economic interest in the shares through use of options. Transfer pricing and thin capitalisation practices through which non-residents minimise their New Zealand tax liabilities are more sophisticated examples of the same principles.

6.26 Multi-party arbitrage is more complex; the complexity is made necessary by the need to meet condition 3 above, that is, to ensure a net gain accrues to the high-rate taxpayer. In the simpler cases of multi-party income tax arbitrage, this process normally involves a tax-exempt (or tax-loss or tax-haven) entity and a taxpaying entity. Income is diverted to the tax-exempt entity and expenses are diverted to the taxpaying entity. Finally, the taxpaying entity is compensated for diverting income and assuming expenses by receiving non-taxable income or a non-taxable benefit, such as a capital gain.

6.27 New Zealanders over the years have indulged in numerous examples of such tax arbitrage using elements in the legislation at the time. Examples are finance leasing, non-recourse lending, tax-haven 'investments', redeemable preference shares, assignments of income using section FC 11, and lease inducements.

Is tax arbitrage a policy problem?

6.28 In theory, cases may arise where tax arbitrage is in accordance with the policy intent of tax rules. It might, for example, ensure that the desired high level of investment is diverted to a tax-preferred activity. However, in most cases, it is contrary to the policy intent, which is to provide favourable tax treatment to a specified class of taxpayer. Tax arbitrage gives access to this preference to a much larger class of taxpayer. When taxpayers outside the specified class can access the tax preference in a way that is inconsistent with the policy intent, it is tax avoidance and thus a policy problem.

6.29 Tax incentives, concessions and loopholes in tax legislation create intended and unintended tax consequences for certain types of economic activity. Initially, this result increases the after-tax rates of return from those activities above the 'normal' after-tax rates of return produced by other activities, increases the value of those assets that are used by those concessionally taxed activities, and introduces inequities into the tax system by conferring windfall gains on the existing owners of those assets. However, subsequent investors in concessionally taxed activities will have to pay higher prices for those assets, and this result reduces the after-tax rates of return they earn from those activities.

6.30 This process of arbitrage will continue until the after-tax rate of return produced by the concessionally taxed activity is driven back down to the normal after-tax rates of return available from other activities. It is important to note, however, that this process of arbitrage may take some time to occur, during which taxpayers can earn higher than normal after-tax rates of return.

6.31 In other words, tax arbitrage continues until the value of the tax concession is capitalised into the price of those assets that produce the concessionally taxed income. Eventually, this process removes the initial inequities in the tax system created by those concessions. In so doing, however, it also results in a long-term over-investment in the assets required by the concessionally taxed activity. Tax arbitrage, therefore, results in inefficient patterns of investment. Ideally, anti-avoidance rules should act as a deterrent to tax avoidance arrangements.

What is a sensible policy response to tax avoidance?

6.32 The most sensible way to reduce tax avoidance is to target the conditions that make tax arbitrage possible. This approach means broadening the tax base and lowering the variability of tax rates. A number of policy considerations may make variability of rates a continuing feature of aspects of New Zealand's tax system. As the committee noted earlier, all taxes have less than comprehensive bases. The approach that has been adopted in New Zealand has been to move as far as possible towards a broad-base, low-rate system, in particular, targeting areas in which tax arbitrage is most evident.

6.33 Examples of such measures and the tax avoidance activity they were designed to counter have been:

  • Thin capitalisation rules - non-residents minimising their New Zealand tax liabilities through thin capitalisation practices.
  • Transfer pricing rules - non-residents minimising their New Zealand tax liabilities through transfer pricing practices.
  • Specified lease rules - finance leasing.
  • Accrual rules - pre-paid interest schemes.
  • Controlled foreign company and foreign investment fund rules - tax haven 'investments'.
  • Inter-corporate dividend exemption (removed) - redeemable preference share schemes.
  • Section FC 11 (repealed) - assignments of income using section FC 11 in conjunction with tax-exempt entities.
  • Measures relating to films and petroleum mining in the Taxation (Accrual Rules and Other Remedial Matters) Bill introduced on 17 November 1998 - film and petroleum mining investors receiving two tax deductions for, in effect, one amount of expenditure.

6.34 This legislation is a relatively detailed response to tax arbitrage. An alternative response could be to rely instead on the general anti-avoidance provisions, such as sections BG 1 and GB 1 of the Income Tax Act 1994 and section 76 of the Goods and Services Tax Act 1985. The New Zealand approach has been to rely upon the general anti-avoidance rules only as a backstop to the substantive legislation. Some commentators have argued that there is no place for general anti-avoidance provisions.

The value and limitations of general anti-avoidance rules

6.35 The argument sometimes advanced against adopting general anti-avoidance rules is that, because tax bases are less than comprehensive, tax law does not apply to theoretical concepts, but applies to detailed prescriptions of what is taxable and what is not. Income tax, for example, is not applied to 'income' in either an accounting or economic sense, but to what the Income Tax Act 1994 sets out to be income. If the Act does not bring something into taxable income, deriving non-taxable income cannot be said to be tax avoidance.

6.36 While this argument has a theoretical attractiveness about it, it assumes that tax policy makers can identify and deal with all the various arbitrage opportunities inherent in the tax system or live with the results of such opportunities. When this issue was previously considered107, the response was to acknowledge the practical need for anti-avoidance provisions.

6.37 This approach is consistent with the government's strategy of broadening the tax base and reducing the variability of rates. The committee recommends that the government should continue to restrict the conditions that make tax avoidance possible by continuing its broad-base, low-rate tax policy.

Income Tax: the General Anti-Avoidance Rule

6.38 In this part of the chapter, the committee proposes several amendments to the general anti-avoidance rule in sections BG 1 and GB 1 of the Income Tax Act 1994. The amendments generally clarify only and do not make any substantive changes to the ambit of the general anti-avoidance rule.

Nature of the general rule

6.39 It is uncertain whether the general anti-avoidance rule in sections BG 1 and GB 1 can be regarded as a code, so ousting common law anti-avoidance rules, such as the fiscal nullity doctrine. This doctrine has developed in the courts in the United Kingdom over the last two decades. In broad terms, it provides that any steps inserted in a related series of transactions for the purpose of avoiding tax can be disregarded by the revenue authorities, and the related transactions can be viewed as a whole.

6.40 The doctrine was first expounded in the judgment of Lord Wilberforce in Ramsay v Commissioners of Inland Revenue108. It is an example of the court applying a purposive approach to construing tax legislation, and allows the court and the Inland Revenue Department to 'see through' a preordained series of transactions. In Ramsay, Lord Wilberforce identified three key features of avoidance schemes which earmarked them as such: their self-cancelling structure, their non-commerciality, and the expectation that all the consecutive steps in the exercise would be performed even though there was no contract stipulating that they would be.

6.41 In a later case, Inland Revenue Commissioners v Burmah Oil Co Ltd, Lord Diplock said:

It would be disingenuous to suggest, and dangerous on the part of those who advise on elaborate tax avoidance schemes to assume, that Ramsay's case did not mark a significant change in the approach adopted by this House in its judicial role to a preordained series of transactions (whether or not they include the achievement of a legitimate commercial end) into which there are inserted steps which have no commercial purpose apart from the avoidance of a liability to tax which in the absence of those particular steps would have been payable.109

6.42 The committee considers that the general anti-avoidance rule in sections BG 1 and GB 1 should not have the effect of ousting common law anti-avoidance rules, such as the fiscal nullity doctrine. The committee recommends an amendment to provide that the general anti-avoidance rule in sections BG 1 and GB 1 does not affect the application of any principles of common law. The amendment would, therefore, ensure that the courts in New Zealand would not be precluded from applying common law anti-avoidance rules, such as the fiscal nullity doctrine. A precedent for this approach is contained in section 75 of the Defamation Act 1992 which provides that nothing in the provisions dealing with absolute privilege affects any other rule of law relating to absolute privilege.

Application of the rule

6.43 The committee considers that the general anti-avoidance rule should apply automatically to any arrangement involving tax avoidance. In other words, the application of the rule should not depend on the Commissioner invoking it. The wording of the provisions makes this clear, and the judicial authority for this view is the Privy Council decision in Newton v FCT110.

6.44 Nevertheless, it is a common belief among tax advisers that section BG 1 is not self-actuating, and operates only if and when the Commissioner invokes it. In the committee's opinion, people should be disabused of this belief. Accordingly, the committee recommends an amendment to make the position absolutely clear. 111.

6.45 The automatic application of the general anti-avoidance rule is also consistent with self-assessment of tax obligations. It would provide an effective deterrent to taxpayers entering into tax avoidance arrangements and therefore help to preserve the robustness of the tax system.


6.46 The committee takes the view that when the reconstruction provisions in section GB 1 are applied by the Commissioner, the effect of the section is that the reconstruction applies from the date of the original transaction that is void against the Commissioner for income tax purposes under section BG 1. The committee again considers that this current position should be made clear in the legislation, and recommends that an amendment to the legislation should be made.

Scope of the rule

6.47 If a general anti-avoidance provision is to be effective, it cannot be precise. Although this feature of an anti-avoidance provision means less certainty for taxpayers, the committee believes that this cost is outweighed by the benefit provided by the flexible wording of the general anti-avoidance rule, allowing the court to address new and different types of tax avoidance arrangements. Again, this helps to preserve the robustness of the tax system.

6.48 There is a danger, then, in being overly precise in defining the term 'tax avoidance', as this precision could restrict the flexibility of the courts in addressing particular tax avoidance arrangements.

6.49 The committee does not favour adopting the 1992 Valabh committee recommendations112 on the general anti-avoidance rule, which sought to define its scope more precisely. The committee considers that these recommendations would make the general anti-avoidance rule less effective and less robust. In particular, the committee considers it preferable to keep the current objective nature of the general anti-avoidance rule, rather than making it more subjective as contemplated by the Valabh committee. A subjective test would make the provision too difficult for the Commissioner to apply. The committee does not endorse the benchmark criteria suggested by the Valabh committee for testing whether an arrangement constitutes tax avoidance, because this proposal could result in the courts having less flexibility to deal with particular cases, even if non-exclusive language were used in the enactment.

6.50 The committee notes that the 'incidental' exception in the general anti-avoidance rule did not mean minor in a quantum sense.

6.51 The committee did not identify any readily apparent way of improving the drafting of the general anti-avoidance rule apart from the clarification recommended earlier in this chapter. As discussed in para 6.34, the general anti-avoidance rule is a necessary, but not a sufficient protection of the tax base. It will continue to be necessary to counter particular types of tax avoidance by enacting legislation to address the underlying conditions that make tax avoidance possible. For example, the repeal of the inter-corporate dividend exemption in 1992 put an end to schemes involving redeemable preference shares, which had effectively allowed the trading of company tax losses between unrelated parties.

6.52 Other than amendments to clarify aspects of the present legislation, the committee favours keeping the drafting of the general anti-avoidance in substantially its present form. The committee acknowledges a trade-off in a lack of certainty for taxpayers in knowing when the boundary of acceptable tax behaviour has been crossed. However, the committee notes that this lack of certainty is partly addressed by the binding rulings system, which provides taxpayers the opportunity to obtain certainty.

6.53 In summary, therefore, the committee recommends that first, the general anti-avoidance rule in sections BG 1 and GB 1 should be clarified to ensure that it is not interpreted to preclude the application of common law anti-avoidance rules, such as the fiscal nullity doctrine. Secondly, for the avoidance of doubt, the general anti-avoidance rule should be clarified to ensure that it applies automatically, and does not depend on action by the Commissioner. Finally, an amendment should be made to clarify that any reconstruction under section GB 1 applies from the date of the original arrangement.

Interpretation Statements, Public Rulings, and Product Rulings


6.54 The general anti-avoidance rule, among others in the Income Tax Act 1994, is broadly framed. As the committee has noted in para 6.47, it is a necessary feature of the rule that its precise scope is not clear. However, the Inland Revenue Department must apply the legislation, and to this end, from time to time, the Commissioner issues interpretation statements, interpretation guidelines, and public rulings. Incidentally to its deliberations on another topic, the committee had occasion to consider a group of several such statements of law, one of which is as yet only an exposure draft. The statements considered by the committee related broadly speaking to avoidance matters. The committee was concerned about the quality of the analysis in the statements, analysis that no doubt has an impact on decisions made within the Inland Revenue Department. The committee explains its concerns about the statements that it examined in the paragraphs that follow.

Policy statement on section 99 of the Income Tax Act 1976

6.55 The committee begins with a departmental policy statement published in 1990 on the application of section 99 of the Income Tax Act 1976, the general anti-avoidance section. This statement seems to have governed much of the department's approach to tax avoidance since it was published.

6.56 Section 99 was last amended as to substance in 1974. The 1974 wording was carried forward into the 1976 Act. The section was disaggregated into several components in the 1994 Act. It is now found in section BG 1, section GB 1, and in a number of definitions in section OB 1. However the essential terminology remains as it was in 1974, and the 1990 statement continues to apply to it.

6.57 The committee understands that the department now intends to withdraw the statement. However, withdrawal depends on a substitute statement being completed; until that is done the statement remains extant.

Objective/subjective test

6.58 The committee was concerned about a number of passages in the policy statement, in particular passages that have the effect of imposing burdens on Inland Revenue officers who try to deploy section 99 where those burdens are not inherent in the text of the section itself. Correctly, the statement recognises that the section 99 test is objective. The test is: 'Does the impugned arrangement avoid tax?' not, 'Did the parties try to avoid tax?' However, some of the text of the statement is framed in terms that could be interpreted as employing a subjective test. An example is, 'The evaluation will be with a view to concluding whether one can predicate whether the arrangement was implemented in its particular way so as to achieve an income tax advantage.' This passage is apt to take investigators' attention away from arrangements themselves, and to encourage them to seek a tax avoidance purpose entertained by the people responsible for implementation. However, the anti-avoidance section does not require the Commissioner to establish such a purpose.

Four-step analysis

6.59 Secondly, the statement sets out a four-step analytical framework that is said to be required by 'the Commissioner's approach' to section 99 cases. One problem with the four steps is that their phraseology seems to accept that the Commissioner has the burden of proof, which is not so. A second problem is that step (d) asks whether an arrangement that may already have been determined to involve more than merely incidental tax avoidance at step (c) 'frustrates the underlying scheme and purpose of the legislation'.

6.60 The underlying scheme and purpose of the legislation is not mentioned in section 99. That is not to say that underlying scheme and purpose are irrelevant to a section 99 inquiry. In fact, whether an arrangement frustrates the underlying scheme and purpose of the legislation can be relevant to whether the arrangement entails tax avoidance, which is a question that the statement poses at an earlier step in its analytical framework. Tax inspectors accept too heavy a burden if they are required to elevate underlying scheme and purpose to an independent test, that is, to establish that an arrangement involves both tax avoidance and frustration of the Act's scheme and purpose.

Practical example

6.61 Thirdly, the policy statement contains an annex that gives several examples of arrangements that might be avoidance, and blesses some of them. The committee will mention only one example of the arrangements that the statement accepts as not involving avoidance. This is example 4, an arbitrage scheme that contrives to grant a tax preference to foreigners when acquiring shares in petroleum mining companies, even though section 160A of the 1976 Act in terms awarded the preference to New Zealand residents only. Alternatively, depending on the price at which New Zealand residents sell their shares to foreigners, the scheme involves not arbitrage but New Zealand residents obtaining a preference for investing in petroleum mining shares that they own for only two weeks.

6.62 On the basis of a formalistic analysis, the policy statement determines that the scheme in example 4 is not vulnerable to attack under section 99. The Commissioner may be correct, but the committee is doubtful. At any rate, if the example is correct it follows that section 160A did not achieve its intended effect. If so, the section should have been amended. If section 160A was deficient in the manner that the Commissioner explains in the statement then, of course, the Commissioner had no option but to assess people on the basis of section 160A's correct, though deficient, meaning. However, the committee doubts the merits of publishing a statement that highlights this assumed deficiency and that invites taxpayers to exploit it.

6.63 Since it was published, the 1990 policy statement seems to have had considerable influence. It may be part of the explanation for the infrequency with which the Commissioner has invoked section 99/BG 1 from the time that it adopted its present terms in 1974. However, the committee is pleased to note the increased use of section 99/BG 1 that is recorded in para 13.51.

Interpretation guideline exposure draft on form and substance in taxation law

6.64 Issues of form and substance are different from questions of avoidance, but in practice they are often closely related. The committee discussed the Inland Revenue Department's 1998 exposure draft of an interpretation guideline, Form and Substance in Taxation Law. As the draft explains:

Interpretation guidelines are intended to clarify general points of interpretation that are causing, or may cause, difficulty for practitioners, taxpayers, and Inland Revenue. An interpretation guideline is Inland Revenue's opinion as to the better view of the law. That view is developed from an appreciation and assessment of the law on a particular topic, as gleaned from the cases.113

6.65 The draft finishes with this warning:

Draft items produced by the Adjudications and Rulings Business Group represent the preliminary, though considered, views of the Commissioner of Inland Revenue.

In draft form these items may not be relied on by taxation officers, taxpayers, and (sic, or?) practitioners. Only finalised items represent authoritative statements by Inland Revenue of its stance on the particular issues covered.

6.66 Although the document is an exposure draft with the qualifications outlined above, the committee understands that the document describes the manner in which the department is interpreting the law. For the reasons given in paras 6.72 to 6.88, the committee disagrees with the approach that the exposure draft reveals.

6.67 The exposure draft suffers from a shortcoming in that it employs an analytical framework that is not refined enough for the purposes of its subject matter. It says, 'The [courts'] only significant departure from [a formalistic] approach is when the essential genuineness of a transaction is challenged [by alleging that the transaction is a sham].' However, when a transaction is challenged as a sham and not genuine, or as being in substance something different from what its form suggests, the courts have essentially one response. This response is to seek the true legal obligations and rights that the transaction imposes or confers on the parties to it.

6.68 Courts often explain this exercise by adopting one of two bi-partite frameworks. The first is the form/legal substance dichotomy. Where the form of a transaction, or the label that the parties give to the transaction, is different from the true legal substance of the transaction, then the courts construe the transaction according to the true legal rights and obligations that it creates, that is, according to its true legal substance. An example is Ensign Tankers (Leasing) Ltd v Stokes114. In that case, the House of Lords held that a transaction that was constructed as a non-recourse loan was in legal substance a partnership, and that it should be treated as a partnership for tax purposes.

6.69 The second bipartite framework entails distinguishing between a transaction's true legal substance and its economic effect. A recent example is Wattie v CIR115. In that case, the Privy Council held that a payment that the Commissioner of Inland Revenue argued was a rent subsidy was in legal substance a premium paid by a landlord to attract a tenant, notwithstanding that in economic effect the payment was just the same as a rent subsidy. Being a premium, the payment was a non-taxable capital receipt, whereas a rent subsidy would have been a revenue item.

6.70 Two features of these alternative bipartite frameworks require noting. First, the concept of legal substance is common to each framework. Secondly, whichever framework is appropriate to the case at hand, the correct answer is always the same: the court must analyse the transaction according to its legal substance and the true legal rights and obligations that it creates. If the first framework is used, the courts reject form in favour of true legal substance. If the second framework is used, the courts reject economic substance in favour of legal substance.

6.71 The result is that a conspectus of both types of case, that is, cases like Ensign Tankers and cases like Wattie, reveal three relevant categories: form, legal substance, and economic substance. The courts reject the first and third in favour of the second. This principle does not mean that the first and third categories are never correct. Rather, they are correct only if they happen to coincide with legal substance, with the parties' true rights and obligations. Thus, in Wattie legal form (a premium) did in fact coincide with legal substance (a premium) and the Privy Council rejected analysis according to economic substance that would have led to classifying the payment as a rent subsidy. In Ensign Tankers, economic substance (a partnership) did in fact coincide with legal substance (a partnership) and the House of Lords rejected form (a non-recourse loan).

6.72 It is an important shortcoming of the exposure draft that its analytical framework fails to identify the three categories of form, legal substance, and economic substance, and instead relies on a formulaic division between form and substance. In the context of a particular case, that distinction can sometimes be enough. But in the context of an interpretation guideline about form and substance that attempts a comprehensive coverage of the relevant elements of the whole field this failure is misleading.

Discussion of judicial dicta in the exposure draft

6.73 The exposure draft quotes two judicial statements of principle that are notoriously difficult to reconcile, without apparently recognising their inconsistency. The first is the 'no taxation on substantial or economic or business character of what was done' statement from CIR v Europa Oil (No 1)116, and the second is Dixon J's emphasis that tax analysis 'depends on what the expenditure is calculated to effect from a practical and business point of view, rather than upon the juristic classification of the legal rights, if any, secured, employed, or exhausted in the process' in Hallstroms Pty Ltd v Federal Commissioner of Taxation117.

6.74 The draft cites Attorney-General v Barker Bros Ltd and other cases on contract formation (that is, on the question of whether or not a contract exists) as helpful cases for analysing contracts that, ex hypothesi, do exist. The committee is not aware of tax cases, apart from cases of alleged shams, where there was any issue as to whether the contract in question existed. It is hard to see how the Barker Bros118 case can shed any light on the apparent conflict between Europa Oil and Hallstroms.

6.75 The conflict between Europa Oil and Dixon J in Hallstroms is resolved, however, when one remembers that there are three relevant categories. When Dixon J rejects a 'juristic classification of legal rights' in favour of 'what the expenditure is calculated to effect from a practical and business point of view' he must not be interpreted as embracing taxation by economic substance. He must be taken as rejecting the pure juristic form of a transaction if this form does not reflect the true legal rights of the parties. He is not saying, as Europa Oil says he must not, that people should be taxed according to the economic substance of their transactions.

Committee's reservations

6.76 The committee hesitated before recording its reservations about a document that is, as yet, only an exposure draft. At the same time, the draft has now been extant since 4 June 1997 and there have been only five submissions on it, none of which has made the points made by the committee. It may be unrealistic for the department to rely on voluntary public comment to correct this kind of document. The committee suspects that, had the draft not come to its attention, the points that it makes would never have come to light; nor are they the committee's only reservations about the draft.119

6.77 The committee is concerned about the likely use of the draft. The document is in essence a text on how to approach transactions by using a form/substance analytical framework. But in status, and in some of its language, the document is a draft statement of the law as the Commissioner understands it. There are several problems here. They stem from the fact that the form or substance question is more a tool of argument or analysis than it is a statement of law. Moreover, it is a deceptive tool, in that judges often state it in firm, almost dogmatic terms, whereas in fact it is infinitely flexible and elusive. In this respect, it has some similarity to the rules of statutory interpretation, which the committee discusses earlier in this report.120

Assessment of the interpretation guideline

6.78 The matters discussed in the last paragraph give cause for concern about the users of the interpretation statement. If people are knowledgeable about tax law, they will understand that, despite its form, the statement can in only a limited sense function as a statement of the Commissioner's view of the law. But these people will already know enough about tax law that they will either not need to use the statement, or, worse, they will be able to use it against the Commissioner, picking on passages that can be deployed against him.

6.79 If people's knowledge of tax law is such that they need the statement to inform them about the form/substance distinction there is a risk that they may be misled. In difficult cases, the form/substance distinction is a matter of shadings of grey. The draft statement does not paint a picture that is purely black and white, but it does give an impression of much more certainty and logic than in fact exists. An Inland Revenue officer relying on the statement to help in analysing and categorising a difficult transaction could well come to the wrong conclusion.

6.80 Issuing the guideline in a form consistent with the committee's views may not be the answer. The guideline is not really an interpretation of a difficult or ambiguous rule of tax law, and hardly qualifies to be called an 'interpretation'. As mentioned earlier, it is more in the nature of instruction in the use of a particular analytical technique. Because the technique is a tool, in close cases it can be used to argue for either side. Publishing an explanation of such a technique as a formal statement of the Commissioner's view of the law can inadvertently give tax advisers in the private sector an argument that, in substance, does not exist. Further, it can cause officers to reach incorrect conclusions.

6.81 The committee bears in mind that there are many tax professionals who do not have a formal legal training. For them, guidelines can serve a very useful purpose. Indeed, the draft statement under discussion was developed with the encouragement of the accounting profession. If the requirements of such people are to be met, then there is a need for finalising and issuing the draft guideline in the respects outlined.

Transactions that balance one another

6.82 An example discussed in the exposure draft is a case in point. The example involves a pair of transactions that balance one another as to subject matter and parties. The example is:

P purchases some assets from M for $100,000. P and M then enter into a simultaneous put and call option agreement under which:

  • M has the right to buy the assets back for for $110,000 (call option);
  • P has the right to sell the assets to M, also for $110,000 (put option).

6.83 The draft states that these options must necessarily be treated as separate transactions, even though they are part of a single agreement. It is possible that there are circumstances where the opinion in the draft would be correct. However, there are not enough secondary facts in the example to decide whether a court would consider each transaction separately and give each transaction full effect, or whether it would conflate the transactions into a single, self-cancelling matrix. Indeed, there are no secondary facts given in the example.

6.84Inland Revenue officers faced with cases where the primary facts match the balancing transactions in the example could be forgiven for following the analysis in the draft and assuming that the law will inevitably require each transaction to be given separate effect.

The Magnum case and balancing transactions

6.85 Interestingly, some people seem to have adopted that approach in analysing the Magnum scheme in the Winebox papers. It may be recalled that the core transactions in the Magnum scheme were an agreement to sell a promissory note, and another agreement to buy the same note. The Magnum scheme's pair of transactions were, if anything, less closely interrelated than the put and call options in the exposure draft's example. The differences are first, that the Magnum transactions were not formally part of a single agreement, and secondly that, while one party was the same in both Magnum transactions, the second parties to the two transactions were not identical but were related companies in the same group.

6.86 On their facts, and to put the matter at its lowest, there is a tenable argument that the Magnum transactions were self-cancelling and did not have the effect that was purported by their authors. In fact, in European Pacific Banking Corporation v Television New Zealand121, the Court of Appeal went further, and, taking into account the secondary facts of the Magnum scheme, held that Television New Zealand had established a seriously arguable case that the whole scheme was iniquitous. In the recent case of Peters v Davison122, the Court of Appeal confirmed its earlier opinion that the Magnum promissory note transaction could be ineffectual because of cancellation of one leg of the transaction by the other.

Difficulties of form/substance analysis

6.87 Comparing the exposure draft's example with the Magnum scheme illustrates that form/substance analysis is much more subtle, elusive and impressionistic than would appear to be the case to a reader of the draft. The committee is concerned that, although the draft purports to be no more than a draft, and is subject to correction, the ordinary course of events would not necessarily see the necessary corrections made. If there is reliance on the period of exposure of the draft to provoke professional comment that would identify errors, that reliance may well be misplaced. Frequently, the view that the draft espouses will suit the taxpayer rather than the Commissioner. It would be a most altruistic practitioner from the private sector who would seek to correct the draft.

6.88A further cause of the committee's concern is that, although the draft is subject to revision after exposure, at the moment it states the Commissioner's current 'considered views'. Have those views affected any private binding rulings that have been issued in the last few years? Have they influenced decisions about completed transactions that have come to the attention of Inland Revenue investigators? The committee cannot answer these questions, because private rulings are not published, and because decisions about individual taxpayers are confidential.

Interpretation guideline on shams

6.89 Like issues as to form and substance, questions of sham are in practice closely related to questions of avoidance, often arising in the same case. The committee considered an interpretation guideline entitled Sham - Meaning of the Term, that was published in 1997 in the Tax Information Bulletin123, and that remains in force. The guideline is an item of three or four pages much along the lines of a short expository article that one might find in a professional or scholarly journal or as part of a chapter in a text book.

6.90 The guideline mentions relevant law, draws certain conclusions, and gives some examples. However, it suffers from the same analytical shortcoming as the exposure draft on form and substance that the committee has discussed, in that the discussion takes place within an analytical framework that is not adequate. The guideline adopts a simplistic form/substance dichotomy. It does not make the point that in order to discover the true legal rights and obligations that a transaction creates courts may ask two, separate, questions, each apt for a different kind of case: form/legal substance, relevant in cases like Ensign Tankers Ltd v Stokes124, and legal substance/economic substance, relevant in cases like Wattie v CIR125.

'No half-way house'

6.91 A second difficulty is that the guideline keeps to the framework, 'There is no half-way house between a sham and an effective transaction.' There are plenty of dicta in the cases that appear to support this principle, and it is accurate as far as it goes. But the principle must be understood within a wider context. That context is that, above and beyond the doctrine of sham, the courts do in fact decline to accord to certain categories of genuine, non-sham, transactions the effect that those transactions purport to have. In strict logic, these impugnable transactions are sub-categories of genuine transactions. However, their legal effect is such that they are no more effective in achieving their hoped-for tax result than if they were shams. For practical purposes, these transactions do constitute a quasi-half-way house between sham and genuine transactions.

6.92 The main inhabitants of this quasi-half-way house are: mislabelled transactions, self-cancelling transactions, and transactions that, when interpreted in context, have an effect different from the initial impression that the reader gains from one or more of the documents.126

6.93The difficulty with the department's interpretation guideline is that most readers would take it to be comprehensive in scope, (in the sense of covering or referring to the whole relevant field, rather than in the sense of being a fully detailed analysis). The guideline reinforces this impression by quoting the 'no half-way house' principle, which has misled a good many readers of judicial judgments in the past. The problem is compounded by the fact that the guideline appears to be a general, authoritative statement. In contrast, reported judgments can be misleading enough, but at least most readers of law reports appreciate that statements of principle in judgments can be taken as generally authoritative only within limits.

Interpretation guidelines about legal reasoning

6.94 The committee concludes its discussion of the section 99 statement, the draft interpretation guideline on form and substance, and the guideline on shams, by reflecting on the purpose of guidelines and statements that set out not interpretations of law but, in effect, instructions or information on how to go about methods of legal reasoning. To the extent that users are Inland Revenue Department staff, the purpose is commendable and necessary: it is most important for staff to be educated in methods of legal reasoning. But interpretation statements are an awkward vehicle for such education.

6.95If interpretation statements and guidelines are to fulfil the function that their name implies, they must be reasonably concise and dogmatic. Legal reasoning has many characteristics, but concise dogmatism is not one of them. Where it is a question of education or instruction of officials on methods of legal reasoning, the conventional vehicles of text books, articles, and class instruction may be more appropriate. However, as the committee discusses below, resource constraints make this strategy not altogether practical.

6.96 The committee is also concerned about directing interpretation statements of the kind under discussion to taxpayers and practitioners. Non-specialists who do not know how this kind of statement fits into the total context of the law may be misled. Specialists do not need them, and are apt to turn them against the Commissioner.

Merits of interpretation statements and guidelines

6.97 As is apparent from the foregoing paragraphs, the committee's discussion reflected grave misgivings about interpretation statements and guidelines that involve approaches to lines of legal reasoning rather than statements of law. The committee considered a recommendation that the Commissioner should not issue such statements and guidelines. In the end, the committee did not follow this course, for three reasons.

6.98 First, there is the question of Inland Revenue officers. The Commissioner will always have to rely on officers who do not have a deep knowledge of tax law and of legal analysis. It will never be practical or economical wholly to remedy this problem by providing enough education to train all, or even most, staff of the Inland Revenue Department as tax specialists. In the end, the committee was persuaded that non-specialists are better off with guidelines, even over-simplified and sometimes misleading guidelines, than with no guidelines at all.

6.99 Secondly, in modern public administration, New Zealanders' expectations are such if such guidelines are published they must be made available to the public, and not kept within the department, as was the case until the enactment of the Official Information Act in 1982. If guidelines are available, particularly if they contain examples, tax advisers will sometimes rely on them to make arguments detrimental to the tax system.

6.100 Thirdly, tax law, particularly income tax law, will always have areas of uncertainty. This is particularly so for cases that depend on the application of a somewhat flexible form of reasoning rather than on relatively black letter law. Examples of such flexible forms of reasoning include: applying the principles of statutory interpretation; analysing facts according to the form/substance distinction; drawing the line between capital and revenue items; and applying the statutory anti-avoidance rule. Both taxpayers and tax advisers welcome succinct guidelines that try to reduce this uncertainty, even (or especially, depending on one's point of view) if the reduction of uncertainty is at the expense of some erosion of the tax base.

6.101Bearing in mind the difficulties with the interpretation statements and guidelines that it has studied, the committee recommends that the government draws to the Commissioner's attention the committee's view that the department should:

Review existing interpretation statements, interpretation guidelines and public rulings that depend on high-level legal analysis in order to determine whether these statements should be revised.

Immediately withdraw any such statements that are found to be deficient, without waiting until replacement drafts are available.

In particular, immediately withdraw the 1990 policy statement on section 99 of the Income Tax Act 1976.

Where appropriate, and especially for issues involving complex reasoning, seek external expert input into interpretation guidelines and interpretation statements before they are released for public consultation.

Take particular care when including in such statements generic examples that do not incorporate contextual facts, and, in general, incorporate contextual facts in examples so that examples do not inadvertently apply to wider areas than officials intend.

Reconsider and refine the department's apparent view on how form/substance and sham/genuine analysis should be approached.

Loss-Attributing Qualifying Companies


6.102 Having examined the general anti-avoidance rule and the department's approach to problems of interpretation, the committee now turns to the use of some particular rules, the loss attributing qualifying companies rules, for tax planning purposes.

6.103 The qualifying companies rules form a subset of the company tax regime. The rules address a problem faced by small businesses: if they wish to obtain the benefits of limited liability and perpetual succession that are afforded by the corporate form, they lose the tax benefits of sole traders and partnerships. The qualifying companies rules allow closely held companies to be treated more or less as partnerships for tax purposes. That is, profits may be attributed directly to shareholders, without the need to go through the imputation system. In order to prevent avoidance, the qualifying companies rules are hedged around with certain formalities and anti-avoidance rules.

6.104 Loss-attributing qualifying companies are a further subset of qualifying companies. Standard qualifying companies permit profits to pass through directly to shareholders, but carry losses forward within the company. Loss attributing qualifying companies permit losses to be passed into shareholders' personal accounts. Because of the potential for abuse, there are additional formal and anti-avoidance rules associated with loss attributing qualifying companies.

Loss attributing qualifying companies as tax planning vehicles

6.105 Tax planning structures that are marketed to the public tend to take one of a relatively small number of basic forms. One of these constitutes structures that offer clients an opportunity to take advantage of accelerated deductions. Many such structures are efficient only if there can be a number of participants. They therefore require a scheme with two characteristics: first, people can combine in groups for investment purposes; secondly, the combination can pass losses through to individual members of the group. Companies have the first characteristic, but not the second. Partnerships have the second, but, because of joint and several liability, have limited appeal for people who need to combine with strangers to join together for investment.

6.106 In the past, New Zealand taxpayers turned to the special partnership, which offered the best of both worlds: limited liability, and the ability to pass losses through to members. During the 1980s, special partnerships were popular structures for investors in films, livestock, and other investments that appeared to offer opportunities from accelerated deductions. People's ability to use special partnerships in this manner was foreclosed from the start of the 1986-87 income year.127

6.107 Nowadays, partnerships formed of numbers of loss attributing qualifying companies fill the gap left since the demise of special partnerships. This structure can offer investors both limited liability and the pass-through of losses. They can also be large enough to gather together enough investors to obtain sufficient economies of scale to make loss-generating investments worthwhile. Two areas where they are used are forestry and intangible property depreciation schemes. There may be other areas.

6.108 Towards the end of its term, the committee became aware of an aggressive tax shelter scheme that involved the depreciation of intangible property. The marketers of the scheme recommend that customers should invest via loss attributing qualifying companies. This structure is not essential to the scheme's operation, but there is little doubt that investing via a loss attributing qualifying company makes the scheme more attractive to participants.


6.109 At present, in New Zealand, forestry investment enjoys a tax preferred status. That status is a matter of government policy and specific Parliamentary enactment and not an accident. The tax-preferred status is unusual, and perhaps close to unique in the current New Zealand tax system. Loss attributing qualifying companies enable middle-income people to pool funds to invest in forestry, and to take advantage of the tax preference. Without pooling funds, middle income people would find it hard to attain the economies of scale that are needed before one goes into forestry.

6.110 It may be, therefore, that, in the field of forestry, loss attributing qualifying companies are promoting government policy, though if this is so the situation has evolved rather than come about by design. The position is almost certainly otherwise in any other areas where loss attributing qualifying companies are being used to gain access to tax benefits. That is particularly so in respect of schemes that rely on claiming depreciation in respect of intangible property.

The intended role

6.111 Officials were not able to tell the committee the extent of the use of loss attributing qualifying companies for conventional, non-tax shelter purposes. However, the committee is aware that loss attributing qualifying companies do play a useful role, particularly for start-up ventures. In this context, they permit initial year losses to be transferred to business proprietors, which is an example of the reason for establishing this category of taxpayer in the first place.


6.112 In chapter 1 of this report, the committee explained the phenomenon that it has called the 'fiscal paradox'. This paradox is that the more people try to make tax systems equitable, the more they make them complex. Loss attributing qualifying companies are an excellent illustration. Their objective is to eliminate tax considerations as an element in determining the appropriate structure for closely held businesses, whether company, partnership, or sole tradership.

6.113 The cost in statutory terms is a whole subpart of the Income Tax Act 1994, subpart HG, which occupies 25 pages in the 1998 reprint, not including definitions. The provisions of subpart HG show an instructive contrast to the Act's next subpart, HH, which covers the taxation of trusts, conceptually a more complex topic, in only 15 pages. From the point of view of simplicity alone, eliminating loss attributing qualifying companies would clearly be a positive step.

6.114 The committee recommends that the government should examine loss attributing qualifying companies to determine:

whether the use of loss attributing qualifying companies as tax avoidance vehicles is a threat to the tax base;

whether the use of loss attributing qualifying companies promotes a government policy of preferring forestry investment, assuming that there is such a policy;

whether, taking into account the factors listed and any other matters that appear to be relevant, the provisions as to loss attributing qualifying companies should be amended in order to prevent these companies being used as vehicles for tax shelters or, if necessary, be repealed.

Contrived Depreciation Schemes

6.115 The committee is aware of a tax shelter scheme that offers generous depreciation deductions to participants for a modest price. The scheme depends for its effect on three elements: claiming depreciation on fixed life intangible property; an asymmetry between participants, being on the one hand ordinary taxpayers and on the other hand a tax-exempt charity; and the use of a syndicate of loss attributing qualifying companies as the investment vehicle. Loss attributing qualifying companies are not essential to the operation of the scheme, but make it more attractive for participants.

6.116 The scheme provides an excellent illustration of several of the general points made in this report. First, it shows how loss attributing qualifying companies, a concept created to promote equity as to tax between sole traders and closely held companies, can be employed to facilitate the marketing of tax shelters.

6.117 Secondly, it illustrates both in this respect and in respect of fixed life intangible property the operation of the fiscal paradox128. In 1993, Parliament amended the Income Tax Act to enable people to claim depreciation allowances for fixed life intangible property in order to promote equity between businesses that use tangible (and thus depreciable, assets) and businesses that use intangible assets, which were formerly non-depreciable. Parliament's well-intentioned measure forms the fulcrum for achieving arbitrage between the charity and scheme participants.

6.118 Thirdly, the scheme illustrates how standard approaches to statutory interpretation allow a measure that is enacted for one purpose (here, to promote horizontal equity between taxpayers) to be used for another purpose (in this scheme, to create a tax shelter). The committee discusses officials' concern about such unintended effects in para 2.38.

6.119 Fourthly, the scheme illustrates the way in which people can take advantage of the tax exemption that is enjoyed by charities129. Indeed, the scheme constitutes a more dramatic illustration of this point than the committee had in mind when writing the part of this report that relates to tax-exempt institutions. The committee is not aware of schemes that turn on arbitrage between charities and taxpayers being marketed in New Zealand in the past, though this activity has occurred in Australia.130

6.120 Fifthly, the scheme may illustrate an unexpected problem with the rewrite's adoption of the gross receipts approach, discussed by the committee in para 2.110 to 2.119. The problem is that section EG 1(1) allows a deduction for depreciation of 'depreciable property'. Section OB 1 relevantly defines 'depreciable property' to include property that people use 'in deriving gross income'.

6.121 Before the gross income approach was adopted, arguably entitlement to deduct depreciation allowances turned on at least an intention to derive net assessable income, even though the intention might have been unrealistic from an objective point of view. Now, it seems that an intention to derive gross income is sufficient, even though the taxpayer never expected the venture in question to result in net income that is taxable.

6.122 Sixthly, the scheme is a good example of the archetypal form of tax arbitrage between taxpaying and tax-exempt entities that is mentioned in para 4.1.

6.123 The committee was informed that the department is aware of arrangements of this type and that they are already under investigation.


96 (1997) 18 NZTC 13,001
97 (1986) 8 NZTC 5,219 at 5,225
98 CIR v Challenge Corporation Ltd (1986) 8 NZTC 5,219
99 (1986) 8 NZTC 5,001 at 5,013
100 (1997) 18 NZTC 13,001
101 (1986) 8 NZTC 5,001 at 5,007
102 (1986) 8 NZTC 5,219 at 5,225
103 (1949) 79 CLR 296
104 As to using inference to prove intent, see paras 13.55 to 13.80.
105 Gribbon J, Taxation, 13 November 1997, page 157
106 Some forms of capital gain may escape, as is the case in New Zealand 107 By the Consultative Committee on the Taxation of Income from Capital in Key Reforms to the Scheme of Tax Legislation, Discussion Paper, October 1991 and Final Report, October 1992
108 [1981] STC 174
109 [1982] STC 30 at 32
110 [1958] AC 450 at 469
111While the general anti-avoidance provisions operate of their own force, at a practical level an ar-rangement will be treated as void only when the Commissioner so determines. In all cases, where the section applies, the arrangement is void for tax purposes from the outset. The Commissioner also needs to decide whether any reconstruction is required to counteract more appropriately the tax avoidance, and to issue an assessment accordingly.
112 Consultative Committee on the Taxation of Income from Capital, Final Report, October 1992, pages 20-33 and Key Reforms to the Scheme of Tax Legislation, Discussion Paper, October 1991, pages 6-53

113 New Zealand Inland Revenue Department document reference IG9703 (1998)
114 [1992] 1 AC 655
115 (1998) 18 NZTC 13,991
116 [1971] NZLR 641 at 648 (PC)
117 (1946) 72 CLR 634 at 648
118 [1976] NZLR 495 (CA)
119 Another is the draft's discussion of a practical example formed by a pair of opposing transactions that balance one another in both legal and economic substance. The draft specifically denies the possibil-ity of considering the transactions together as a self-cancelling matrix even though this is a standard analytical approach. See below paras 6.82 to 6.88.
120 See paras 2.7 to 2.91

121[1994] 3 NZLR 43
122Unreported, Court of Appeal, CA 72/98, 17 November 1998
123 Volume 9, No 11, page 7
124 [1992] 1 AC 655, discussed in para 6.68
125 (1998) 18 NZTC 13,911, discussed in para 6.69
126 See Prebble J, 'Criminal Law, Tax Evasion, Shams, and Tax Avoidance: Part II - Criminal Law Consequences of Categories of Evasion and Avoidance' (1996) 2 New Zealand Journal of Taxation Law and Policy, 59, pages 63-66. The interpretation guideline mentions this third category by impli-cation in its discussion of Richardson J's judgment in Re Securitibank (No 2) [1978] 2 NZLR 136, but it does not explore the implications of the category in a manner that is sufficiently informative.
127 By the Income Tax Amendment Act (No 4) 1986
128 See paras 1.3 to 1.5
129 See para 4.1
130 See, for example, Prebble J, Criminal Law, Tax Evasion, Shams, and Tax Avoidance: Part II-Crimi-nal Law Consequences of Categories of Evasion and Avoidance (1996) 2 NZ Journal of Taxation Law and Policy for a discussion of an Australian scheme.