THERE ARE TOO MANY WITCHDOCTORS IN OUR TAX COURTS: IS THERE A BETTER WAY?1 Justin Dabner * The traditional methodology employed by the judiciary in tax cases leads to narrow definitional debates separate from a consideration of policy and principle. With each new case we expect a resolution of fundamental issues and some clarity. But this is rarely achieved and the uncertainty and ambiguity associated with our tax laws compounds. We need to depart from the traditional methodology and embrace an approach in which the judiciary expressly acknowledges policy considerations when deciding hard tax cases. Thus our legal witchdoctors need to embrace a new methodology. Such an approach can be demonstrated by reference to recent general antiavoidance cases. Introduction An old man lives in a village on Tanna, a remote island of Vanuatu. He suffers a debilitating illness and is very unwell. He hears of a great witchdoctor on the other side of the island. He summons all his energy and walks the length of the island, skirting the hissing volcano of Mt Yasur, until he finally arrives at the witchdoctor's village. The witchdoctor is the most revered in all Vanuatu. His knowledge is the accumulation of wisdom handed down to him from his ancestors. He gives the old man a special blend of kava to drink. He smears his body with the sap of the pandamus tree. He puts the old man in a hypnotic spell. He dances the healing dance. He chants the magic words. After all these remedies have worked for centuries. But the old man's condition deteriorates further and he dies never seeing his beloved village again. * 1 Associate Professor Justin Dabner, Law School, James Cook University. This article is derived from a paper presented at the Australasian Tax Teachers Conference, Wellington, New Zealand, January 2005. 36 JUDICIAL METHODOLOGY IN TAX CASES What did the witchdoctor do wrong? He tried all the traditional remedies and he was the most gifted exponent of his craft in the country. The man died of leukemia, a legacy of the French Nuclear testing in the Pacific. Nothing the witchdoctor could have done could have saved him. He simply did not have the technology. Times had changed and the new types of illnesses faced by his people did not respond to traditional remedies. Thousands of kilometers across the Pacific the Courts of Australia and New Zealand sit on tax cases. The witchdoctors in their costumes and with their traditions apply established rules of interpretation to the law to attempt to resolve the complex cases before them. But the law is dying. It has pneumonia. It is drowning in its own detail. Everyday it becomes more complex and perplexing. It is now well beyond the understanding of the average citizen and even some tax academics. The witchdoctors in the courts simply do not have the right technology to prevent this slow death. My argument is that the traditional way tax judges approach cases is not promoting a healthy tax system but rather contributing to its demise. The overly legalistic approach to the interpretation of tax statutes naively assumes that the answer is always to be found in the words of Parliament. This is a fallacy. We are in denial unless we accept that language is an inherently imprecise and flawed means of conveying an idea. Rather a more policy orientated approach to deciding tax cases is to be preferred over semantic analysis. But there is nothing new in this thesis. It has been articulated much more eloquently by others for years.2 However these commentators have typically been marginalized 2 N Brooks, `The responsibility of judges in interpreting tax legislation' in Tax Avoidance and the Rule of Law edited by GS Cooper, IBFD Publications BV 1997, 93; Y Grbich, `Is economics any use to tax lawyers' (1980) 12 Melbourne University Law Review 340; Y Grbich, `New modalities in tax decisionmaking: applying European experience to Australia' (2004) 2 Ejournal of Tax Research 125. Contrast RA McLeod, `An economic approach to taxation avoidance' (1996) 2 New Zealand Journal of Taxation Law and Policy 171 who adopts an economic analysis but proposes a less activist role for the judiciary. 37 (2005) 15 REVENUE LJ as too extreme. What I have attempted to do is to find a compromise that would provide a framework for these more extreme views to work in practice.3 GAARs do not work My starting point is that the majority of difficult tax cases end up involving the general antiavoidance provision (the `GAAR'). Just as people confined by an illness or operation to hospital often are inflicted with secondary infections, this is exactly the effect that I see the GAAR having on our tax system. GAARs add another layer of complexity and uncertainty onto our laws the judicial consideration of which emasculates the real issues at hand. So I believe these provisions are flawed. As proof of this proposition consider some of the difficulties raised by Australia's GAAR, Part IVA.4 Whilst many of the procedural and minor interpretative difficulties have been addressed fundamental defects remain and can be summarized as:5 The Part assumes that there is a `correct' view of the law and a way of doing things. Thus whilst a taxpayer might think that they have a choice as to how to structure a transaction, in fact, there might be only one acceptable alternative from a tax perspective. This may not be expressly stated in the legislation but those taxpayers and their advisers `in the know' will appreciate what is acceptable. It is the `smell' or `vibe' of the thing. That is, Part IVA delineates boundaries in an unprincipled way. The provisions cannot be sensibly reconciled with the Duke of Westminster principle6 or a business purpose exception. The minimization of expenses such as tax is clearly a business end. But if legitimate business transactions are caught by the Part then how are taxpayers to ascertain what is permissible? The focus on identifying a taxpayer's dominant purpose typically degenerates into an attempt to force a distinction between primary and secondary level purposes or motive and purpose onto the facts. Furthermore, it is only permissible to draw this distinction from the objective evidence, much of which was initially created by the taxpayer. 3 4 5 6 See, in particular, `In Search of a Purpose to our Tax Laws' (2003) 6 Journal of Australian Taxation 32. Income Tax Assessment Act 1936. For further analysis see my article referred to in fn 3 and also `The Spin of a Coin in search of a workable GAAR' (2000) 3 Journal of Australian Taxation 232. [1935] All ER Rep 259; [1936] AC 1. 38 JUDICIAL METHODOLOGY IN TAX CASES The application of Part IVA requires a comparison between the arrangement entered into and an `alternative hypothesis'. However the courts are currently tackling the dilemma that if the alternative hypothesis is equally tax advantageous then the Part would seem to have no application. The nature of the relevant `scheme' is indeterminable. Even the principle, though endorsed in FCT v Peabody,7 that the arrangement identified must be one not `robbed of all practical meaning' presented little certainty in its application. Now it appears from Hart v FCT8 that even this prescription might not be available. In turn, the mandate to select a narrow set of facts as the relevant scheme can eschew the requirement that the tax avoidance purpose be dominant.9 The view that a tax adviser's purpose is relevant almost guarantees a finding that the dominant purpose of the arrangement is to achieve a tax benefit where a specialist tax adviser was employed. Are we now seeing tax experts rebadging themselves? The technical detail of the provision results in decisions that focus on the minutiae. The real focus as to whether something is unacceptable tax avoidance, in the sense that it is economically and socially undesirable, is emasculated in technical analysis and semantics. If you need convincing that the provision is a mess I invite you to read the High Court's latest offering on the subject in Hart and Hill J's attempt in Macquarie Finance Ltd v FCT10 to make sense of it all. His Honour had great difficulty identifying a precedent on the meaning of `scheme' from the High Court decision. Fortunately for His Honour, he was able to avoid having to reach a decided view on the scope of the scheme on the facts. Ultimately, whilst concluding with reluctance that Part IVA, as interpreted in Hart, applied, Hill J expressed doubt as to whether this would have been Parliament's intention when the Part was enacted. His Honour's judgment is a reflection of the reality that there is a great degree of uncertainty as to the scope of the provision. 7 (1994) 181 CLR 359. 8 (2004) HCA 26; 2004 ATC 4599. 9 Hill J in Macquarie Finance Ltd v FCT [2004] FCA 1170 sought to avoid this result by suggesting that, notwithstanding a narrow identification of the relevant scheme, the broader circumstances in which it occurs may be taken into account. Branson QC has argued that such an approach is inconsistent with the legislation: `Hart's Case: what may constitute a scheme' (2004) 39 Taxation in Australia 315, 318. He concludes that the Part is so broad that quite benign arrangements are within its scope. 10 [2004] FCA 1170. 39 (2005) 15 REVENUE LJ Such is the state of play after almost a quarter of the century. The legislation is clearly inadequate.11 Whilst it is understood that the ATO is content with the decisions they are currently achieving it needs to bear in mind the phenomenon that the judiciary often react to broad interpretations by subsequently narrowly applying them. Furthermore, as the guardian of the administration of the tax laws, the ATO should not be satisfied with the current uncertainty for taxpayers that Part IVA generates. It is not sufficient to explain this away by saying that the Part operates on a case-by-case basis. This intimates that we are not to know whether it applies until the highest court has ruled on it. Hardly an effective tax system. Whilst the primary focus of this article is on Australia, similar observations can be made about the New Zealand GAAR. It is contained in ss BB 3 (1), BG 1, GB 1, with important definitions in S OB 1.12 It is well known that it draws heavily from the former Australian GAAR in s 260 of the Income Tax Assessment Act 1936. For the purposes of my argument this is significant in two respects. Firstly, it means that the New Zealand GAAR is similar to Part IVA in primarily focusing on the taxpayer's (not merely incidental) purpose.13 Here lies a major limitation that renders the provision potentially subject to the deficiencies in Part IVA referred to above. In particular, it is noteworthy that the Internal Revenue Department (`IRD') adopts the view that a step in a wider arrangement can be considered separately for the purposes of determining whether a tax avoidance purpose is not merely incidental and that an arrangement will nevertheless be caught, even though it is capable of explanation by reference to ordinary business or family dealings.14 It also means that that the legislative prescription is very vague and open to judicial interpretation. In a similar way that judicial glosses came to characterize the interpretation of s 260 such would also appear to be the New Zealand experience. 11 For the same conclusion see the views expressed by Sackville J in `Avoiding tax avoidance: the primacy of Part IVA' (2004) 39 Taxation in Australia 295. His Honour takes the view that a broader judicial approach to interpretation would be preferable to the application of a GAAR. For another view that Part IVA is inadequate see P Donovan, `The aftermath of Hart's Case ­ a case for reform of Part IVA?' (2004) 39 Taxation in Australia 253. However that commentator, in contrast to my views and those of Sackville J, believes that the answer lies in redrafting the provision. 12 Income Tax Act 2004. 13 Although reference can also be made to the `effect' of the arrangement. This means that any tax adviser's involvement is made relevant. 14 See Appendix C to the Tax Information Bulletin Vol 1, No 8, February 1990. With a view to clarifying the application of the provisions, the IRD issued an exposure draft on the interpretation of the GAAR in September 2004. This exposure draft arguably takes a stricter approach than the 1990 Bulletin on these matters. Notably, Australian cases are cited in support of the positions taken. 40 JUDICIAL METHODOLOGY IN TAX CASES Most notable has been the judiciary's endorsement of a predication test, flirtation with a distinction between tax mitigation and tax avoidance (the former apparently not being caught by the GAAR), further flirtation with an impropriety test and an ongoing debate as to the relevance or otherwise of a choice doctrine.15 Meanwhile government enquiries in the 1990s either proposed amendments along the lines of Part IVA or rejected the need for any substantial amendment.16 All this time the IRD has effectively ignored the wording of the provision and applied its own (relatively conservative) approach. Notably, this approach gives predominance to the purpose of the primary legislation and applies the GAAR in such a way as to ensure that a literal reading does not frustrate this purpose.17 My perspective on the New Zealand experience is that, whilst the GAAR has presented significant interpretational difficulties when visited by the judiciary,18 the opportunity for this has largely been avoided by the IRD's approach to its application. This approach, which seemingly elevates the importance of the purpose of the primary legislation over the purpose of the taxpayer, might be referable to the mandate to adopt a purposive approach in the Interpretation Act 1999 and its predecessors.19 Given this approach it is, in my view, arguable that the New Zealand GAAR adds nothing to the analysis other than the potential to obscure and draw the focus away from the fundamental issues in an avoidance case.20 15 See the discussion in J Dabner, `The spin of a coin ­ in search of a workable GAAR' (2000) 3 Journal of Australian Taxation 232, 2434. 16 Contrast the recommendations of the Valabh Committee (Consultative committee on the taxation of income from capital, final report October 1992) with those of the Committee of Experts (Report to the Treasurer and Minister of Revenue by a Committee of Experts on tax compliance December 1998). 17 See Appendix C to the Tax Information Bulletin Vol 1, No 8, February 1990 and the recent exposure draft. 18 For a highly critical review of the judicial interpretation of the NZ GAAR (especially by the Privy Council) see Thomas J in Commissioner of Inland Revenue v BNZ Investments Limited (2001) 20 NZTC 17,103. 19 Discussed below. 20 In contrast to the Australian Part IVA, the New Zealand GAAR catches arrangements which have either the purpose or `effect' of tax avoidance. The 1999 Australian Ralph Review of Business Taxation (http://www.rbt.treasury.gov.au/publications/paper4/index.htm) had rejected a proposal to extend Part IVA through reference to `effect' (which would make it consistent with the Australian GST GAAR ­ Division 165 of the A New Tax System (Goods and Services Tax) Act 1999). Whilst this suggests that the NZ GAAR is broader the approach adopted by the IRD in adopting a purposive interpretation to the provisions as its final step in the application of the GAAR generates some balance. In fact, it could be said that the combination of these 41 (2005) 15 REVENUE LJ An alternative approach ­ accepting the reality that language is imperfect I propose that the GAAR be repealed21 and replaced with eight strategies that attempt to delineate between acceptable and unacceptable transactions in a way that creates greater certainty, or at least recognizes that where certainty cannot be achieved this is a cost that has to be shared by the community. Legislative purposive rule First, there should be an express mandate in the tax legislation requiring the courts to interpret the legislation to give effect to its underlying purpose.22 Currently, such a mandate exists in the Australian Acts Interpretation Act 1901.23 but the judiciary exercises the right to choose between this methodology and more literal interpretation rules. A consistent approach should be required. Furthermore, the existing purposive approach to interpretation does not allow the judiciary to read words into a provision to further its purpose. A broader purposive approach should be mandated that gives the judiciary such power. Purposive legislation and objects clauses To assist the judiciary to ascertain the purpose of legislation a drafting technique that abandons detailed legislation in favour of statements of broader principle and the use of objects clauses should be adopted.24 Much research has been conducted in relation to this proposal in recent years and comments last year from the Australian 21 22 23 24 two factors brings the NZ GAAR closer to the ideal, as the taxpayer's purpose is ultimately largely irrelevant and the purpose of the legislation predominates. This is supported by Mr Justice Sackville, who recently suggested that if a completely new tax system could be created it would be unlikely to include Part IVA as its central anti avoidance provision but rather ought rely more on a different approach to statutory interpretation: `Avoiding tax avoidance: the primacy of Part IVA' (2004) 39 Taxation in Australia 295, 300. As had been the case in New Zealand by virtue of s AA3(1) Taxation (Core Provisions) Act 1996, not since replicated in the 2004 rewrite apparently because it had since been included in the Interpretation Act 1999. Section 15AA. Also see the New Zealand Interpretation Act 1999 and, previously, s 5(j) of the Acts Interpretation Act 1924. For a discussion of the New Zealand experience using objects clauses see `In Search of a Purpose to our Tax Laws' (2003) 6 Journal of Australian Taxation 32. 42 JUDICIAL METHODOLOGY IN TAX CASES Government suggest that Australia is likely to experiment with this approach in tax legislation in the near future.25 As an aside, it is notable that there is almost universal acknowledgement by tax academics, at least, that the way to most effectively rid the world of tax avoidance is through the structural design of a tax system. For example, qualifications to general principles should be minimised and tax expenditure provisions avoided. Whilst an ideal, it is, however, recognised that politically this is unachievable.26 Unfettered reference to extrinsic material Currently the judiciary is restricted by the need to find ambiguity in its mandate to refer to extrinsic material in order to make sense of legislation. The type of material that may be sighted is also restricted. A new approach is required. The rule that Parliament's purpose is to be found solely in the words of a provision should be abandoned. Thus, the words should no longer take precedence but be simply the starting point for the judiciary in its quest to identify parliament's purpose. Furthermore, the judiciary should be trusted to refer to whatever material they consider useful, including oral testimony, and give it the appropriate weight. 25 See the Assistant Treasurer Senator Helen Coonan's address to the Australian Institute of Company Directors, Perth, 12 July 2004. 26 Ralph, Review of Business Taxation, Commonwealth of Australia July 1999 Recommendations 6.1 to 6.5 ; R Kreever, `The Ghost of the Duke of Westminster laid to rest in Australia?' (1997) 45 Canadian Tax Journal 122, 130; V Krishna, `The Tension between the Westminster Principle and Abusive Tax Avoidance' (1997) 7 Canadian Current Tax 65; 1997 CCT LEXIS 15 and `The Illusive Interest Deduction' (1998) 8 Canadian Current Tax 89; 1998 CCT LEXIS 19; RA McLeod, `An Economic Approach to Taxation Avoidance' (1996) 2 New Zealand Journal of Taxation Law and Policy 171; NF Orow, `Part IVA Seriously flawed in principle' (1998) 1 Journal of Australian Taxation 57; `The future of Australia's general antiavoidance provision' (1995) 1 New Zealand Journal of Taxation Law & Policy 225; and `Towards a conceptually coherent theory of tax avoidance Parts 1 & 2' (1995) 1 New Zealand Law of Taxation Law & Policy 288 and 307; L Muten, `The Swedish Experiment with a General AntiAvoidance Rule' 307; J Waincymer, `The Australian Tax Avoidance Experience and Responses: A Critical Review' 247; CH Gustafson, `The politics and practicalities of checking tax avoidance in the United States' 349; M Brooks & J Head, `Tax Avoidance: In Economics, Law & Public Choice' 53; M Gammie, `Tax Avoidance and the Rule of Law: The Experience of the UK', 181; I Richardson, `Reducing Tax Avoidance by Changing Structures, Processes & Drafting' 327 all in Tax Avoidance & the Rule of Law (GS Cooper (ed), IBFD Publications 1997). 43 (2005) 15 REVENUE LJ Substance over form Currently there is an inconsistent application by the judiciary of the principle of whether substance should prevail over form. It should be made clear that substance should dominate and the inquiry before the court should be how did parliament intend this economic result to be dealt with. Such a rule should be particularly effective in dealing with complex artificial transactions designed to replicate a legal effect for tax purposes removed from their commercial reality. Indeterminable purpose ­ adopt the best policy solution Opponents of the purposive approach to interpretation argue that often it is impossible to determine parliament's purpose. Alternatively legislation, particularly tax expenditure provisions, may reflect competing purposes. This is all true. However, where the purpose(s) of the legislation is obscure or it does not assist in resolving the issue at hand, then the judiciary should undertake a policy analysis and openly decide the matter by reference to the preferred policy outcome. Then, if the government disagrees it is open to parliament to change the legislation, but with the benefit of access to an express policy consideration by gifted legal minds. Whilst this sounds radical there is a substantial amount of literature supporting a more economic approach to deciding tax cases.27 Anti-avoidance cases are infamous for their judicial debates over semantic niceties. This is a huge waste of public resources and proceeds on a misguided view that words have a `correct' meaning. This is a fallacy. There are only better argued meanings. Thus there needs to be recognition that language is an imperfect tool to communicate ideas. Expanded discretion in quantifying liability The above strategies may do little to improve certainty for taxpayers where the legislation is complex or poorly drafted. Proponents of the literal approach to interpretation argue that it is unfair to apply an impost on citizens other than by reference to express words in the legislation. This may be true but it is also the mantra of tax avoiders. We need a compromise that recognises that some quite innocent taxpayers may have been positively misled by the legislation, especially where the court has adopted a broad purposive approach or policy analysis.28 Therefore, the courts should be 27 See above n 2. 28 As was the case in McDermott Industries (Aust) Pty Ltd v FCT [2004] FCA 1044 where the taxpayer, who had relied on the literal meaning of the words of a treaty, was ambushed by the Court taking a purposive interpretation. See the acknowledgement of the harshness of 44 JUDICIAL METHODOLOGY IN TAX CASES provided with an expanded discretion to quantify liability by taking into account the clarity with which tax policy is expressed and the taxpayer's conduct. This could extend from simply refraining from entering a costs order through to negative penalties or discounts off the tax liability. It is only at this level that the taxpayer's purpose would be relevant. Furthermore, a fund should be established to compensate taxpayers who have incurred expense arising from their interpretation of what has been held by a court or the Tax Policy committee (see below) to be misleading legislation but whose cases did not proceed to litigation. In this way the inherent costs of the imperfections of language could be borne by the community generally rather than by individual taxpayers. Tax policy committee Furthermore, with a view to reducing uncertainty and the prospect of litigation, a tax policy committee should be established with representatives from the revenue authority, Treasury, business and the profession. Its primary role would be to enunciate on the perceived policy behind tax legislation or desirable policy direction where the existing policy is unclear and to monitor whether tax rulings reflect this policy. Applications could be made to the committee by a taxpayer (following a ruling application to the ATO) or by the ATO. The deliberations of the committee would not be binding on courts but would provide highly persuasive evidence of either the existing or desirable policy position and therefore would be unlikely to be contested. A `Norm Campaign' to change community values There is a substantial body of literature to suggest that taxpayers' perceptions of unfairness and excessive taxation may be factors in their decision to avoid tax.29 Hopefully the strategies outlined above would go some way towards improving views as to fairness. But we need to go further. Just as campaigns of the past have changed society's views as to littering, drink driving, exercise, smoking and safe sex we need a campaign to persuade taxpayers that they have a moral obligation to pay tax in accordance with parliament's intention. Tax is the cost of the public goods we all enjoy. this result in M Burton, `The meaning of `permanent establishment' in McDermott Industries (Aust) Pty Ltd v Commissioner of Taxation [2004] FCA 1044 [2004] 39 CCH Tax Week 844. 29 For example, see P Niemirowski, S Baldwin and AJ Wearing, `Tax related behaviours, beliefs, attitudes and values and taxpayer compliance in Australia' (2003) 6 Journal of Australian Taxation 132 and the references cited there. 45 (2005) 15 REVENUE LJ How would this work? Let us consider an application of this approach to the facts of Hart. The facts were that the taxpayers took out a wealth optimizer split loan product to enable them to purchase a home and retain their former residence as an investment property. The product was structured such that in the early years of the loan all repayments were directed towards paying off the home loan while allowing the interest on the investment property loan to be capitalized (and compound interest imposed). In essence the progressive shift of the loan balance from the residential side of the account to the investment side permitted interest on the loan funds used to purchase the Harts' home to be claimed as a tax deduction. The High Court held (overruling a unanimous Full Federal Court) that the tax benefit associated with the loan was subject to Part IVA. A key reason for leave to appeal to the High Court being granted was to clarify the operation of the definition of `scheme' for the purposes of Part IVA. As noted above, the decision has done anything but, as the pivotal decision of Callinan J is open to competing interpretations. One interpretation would see the Commissioner's powers under Part IVA expanded, while the alternative interpretation would merely reinforce the status quo that has prevailed since the High Court decision in Peabody. The debate over semantics continues. An alternative approach that the High Court might have adopted would have been to focus on the purpose of s 81, namely to allow a tax deduction for expenses incurred in generating assessable income and not to allow a deduction for personal expenses. Then, looking at the substance of the transaction, it is clear that it attempts to achieve a tax deduction for interest on borrowings used for private purposes. The Court, with an expanded discretion to quantify a tax liability, might then have apportioned the interest deduction to give effect to their decision. Alternatively, the Court might have considered the policy merits or otherwise of permitting the deduction. On one side, the deduction would encourage investment activity particularly in the housing sector with all the attendant benefits for the community. On the other hand, it would create a disparity between taxpayers in similar situations who had not accessed the wealth optimizer product, at least until such time as they could have refinanced. Furthermore, there would clearly be a loss to the Revenue from the additional tax deductions. With these and other relevant considerations expressly stated, the Court could have made a policy decision as to whether the deduction should be available. Such an approach is clearly to be contrasted with the technical, convoluted analysis of the wording of Part IVA that the Court embarked upon. Clearly the traditional 46 JUDICIAL METHODOLOGY IN TAX CASES methodology to approaching the decision has not given the community greater certainty or enhanced the reputation of the law. It is argued that it never will. Another example can be drawn from Macquarie Finance. This case involved a tax effective financing arrangement whereby Macquarie Bank issued stapled securities providing holders with preference shares in Macquarie Finance and a beneficial interest in unsecured notes. Under the terms of the issue, holders of the securities would receive interest payments, unless certain prescribed events occurred at which stage they would be entitled to dividends. The arrangement permitted Macquarie Bank to recognize the capital raised as Tier 1, whilst claiming a tax deduction for the payments to service the securities. In other words, this hybrid arrangement allowed the securities to be counted as equity for prudential purposes but with the tax characteristics of debt. Naturally, the decision was being closely watched by the finance sector and had the potential to seriously erode government tax revenues. Hill J reviewed the deductibility of the `interest' in the context of the composite nature of the transaction, concluding that the `interest' was capital or of a capital nature. The outgoings were concerned not with the cost of maintaining a loan but rather with the cost of raising capital. Accordingly the amounts were not deductible under s 8-1. In the event that he was wrong, his Honour went on to consider the application of Part IVA. As previously observed, his Honour had difficulty in identifying a precedent as to the definition of scheme from the High Court judgments in Hart. In the result he took the view that Hart probably stands for the proposition that the `particular form' of the loan agreement was the scheme. Adopting this view of the relevant scheme he concluded that Part IVA would apply. On the approach advocated here the part of his Honour's judgment dealing with Part IVA would have been unnecessary. His Honour had decided the case in the appropriate manner with reference to s 8-1. His Honour's finding was effectively an acceptance that the substance of the outgoings to the payer were dividends in nature rather than interest and, therefore, not deductible. Had his Honour any doubt as to this, he could have embarked upon a policy analysis as to whether it was desirable for such transactions to proliferate in a tax effective manner. This might have involved obtaining expert evidence on, and considering such matters as, the cost to the Revenue and the desirability of encouraging such complex transactions with their commercial ramifications and establishment costs as against the advantages of enhanced investment activity. The ultimate focus would have been on the economic and social implications for the country. As always, it was 47 (2005) 15 REVENUE LJ open to parliament to subsequently legislate in relation to such arrangements, as indeed it has.30 A final illustration, drawn from New Zealand, where a policy analysis by the courts may have been particularly appropriate, is the case of Challenge Corporation Limited v CIR.31 This decision both applied a GAAR and is New Zealand authority for the traditional purposive approach to interpretation. The case involved the purchase of a loss company during the year in artificial circumstances.32 At the time, the legislation only required that companies share common ownership at the end of the income year in which a loss was to be grouped. The litigation at all stages focused on the application of the GAAR and its relationship with the loss and grouping provisions. The Privy Council, in a problematic decision that has largely been ignored by the New Zealand courts, allowed the Inland Revenue's appeal on the basis that this was unacceptable `tax avoidance' to be contrasted with acceptable `tax mitigation'.33 On the other hand, those judges supporting the taxpayer's case accepted that there was no evidence that Parliament's intent was otherwise than to permit the transaction and the specific provisions took priority over the GAAR. The approach advocated here would have the Court focus solely on the purpose of the loss and grouping provisions and not be distracted by the nuances of the GAAR. If 30 See the debt/equity rules in Division 974 of the Income Tax Assessment Act 1997. 31 [1986] 2 NZLR 513. 32 For example, the price of the shares in the acquired loss companies was related to the potential tax savings that Challenge might achieve and the companies never traded again after acquisition. 33 The New Zealand Court of Appeal has ignored this distinction in favour of applying the Newton v FCT (1958) 98 CLR 1 predication test together with a choice doctrine. See, for example, Miller v CIR (1998) 18 NZTC 13,961. Subsequently, the Privy Council in CIR v Auckland Harbour Board (2001) 20 NZTC 17,008 and O'Neil v CIR (2001) NZTC 17,051 (the appeal from Miller) appears to have accepted that the distinction between mitigation and avoidance is problematic in favour of an approach that explores whether there is a tension between the commercial and juristic character of a transaction, in which case the GAAR applies. Notably, the Privy Council in the Auckland Harbour case adopted a purposive interpretation of the relevant legislation as it then existed allowing them to conclude that the transaction was not contrary to the intent of the legislation, so a specific antiavoidance provision had no application (the comments on the GAAR were obiter). Subsequently, the NZ Parliament, in fact, amended the legislation to prevent such an outcome in the future. The O'Neil case involved an artificial scheme to convert profits into tax free capital receipts that could have been resolved by taking a substance over form approach, without any reference to the GAAR. 48 JUDICIAL METHODOLOGY IN TAX CASES no relevant parliamentary intent could be ascertained from all the publicly available material, or it was ambiguous, then an adjudication based on the most desirable tax policy would be appropriate. Then the extent to which this decision caught the taxpayer by surprise, or it could be concluded the taxpayer's interpretation of the legislation lacked credibility or was unreasonable, would factor into the quantification of liability. Conclusion Our income tax laws are degenerating into complex and confused masses of information. When you add another layer of uncertainty that GAARs generate then certainty for all but the most basic transactions is impossible. This suggestion is an attempt to compromise between the rights of taxpayers to structure transactions in a cost effective manner on the one hand and the need to protect the Government's revenue base and ensure that the tax system applies fairly on the other. Just as the witchdoctor was misguided in his belief that the traditional remedies would cure the old man's illness, our lawmakers labour under a fundamental misconception that parliament can agree on a form of words that can provide an answer for any of the complex transactions that legislation may have to apply to. This is a fallacy. Rather, we need enhanced guidance for the judiciary at the policy level, together with greater trust in them to give effect to this policy in a rational way and the means to deliver this fairly. The traditional technology employed in our tax courts to address our ailing tax system simply provides no cure or, at best, provides only palliative care. 49