Report to the Treasurer and Minister of revenue - by a committee of experts on tax complianceBill Birch Treasurer
Report to the Treasurer and Minister of revenue
By a committee of experts on tax compliance
Chapter 2 -
The Rewrite Project continued
2.35 If a more liberal approach, less constrained by the language used, is considered desirable, section AA 3(1) should refer to factors such as historical context and apparent policy. In fact, as explained below, the remainder of section AA 3(1) may limit rather than amplify 'context', by referring to 'core provisions' and 'the way in which the Act is organised'. The third element, the reference to 'purpose provisions', will probably be neutral, except where relatively narrow purpose provisions are enacted for particular sections or groups of sections. In this latter case, the effect of 'context' will be amplified.
2.36 Officials advised the committee that the government had an additional, though related, objective in drafting section AA 3(1). This objective was to persuade courts to interpret provisions according to their purpose as revealed by their locality within the Act. For example, an anti-avoidance provision, located in a subpart of the Act reserved for anti-avoidance provisions, should be interpreted as an anti-avoidance provision only, and not as a provision that enables taxpayers to contrive transactions with structures that entitle people to claim deductions that would not otherwise be allowed (which is sometimes the fate of anti-avoidance provisions).
2.37 Another example flows from a canon of statutory interpretation often employed by the courts: to try to interpret crucial statutory terms in a way that ensures that the terms bear the same meaning wherever they are used in the Act. CIR v Alcan New Zealand Ltd25 may be considered as a case where the court observed this canon, choosing an interpretation that allowed 'company' and 'group' to keep consistent meanings throughout the Act. Maintaining this consistency entailed rejecting the Commissioner's submission.
2.38 The committee agrees with officials that the problem identified in para 2.34 is serious and a worthwhile objective for the rewrite process to address. The best result would be for the rewrite process to achieve a statute where the kinds of interpretation problems that the above examples illustrate will not occur: the outcome of the rewrite would be a statute where sections are not only clear in themselves, but also do not have inadvertent effects beyond the drafter's intention. However, no-one can be sure that such a result could be achieved, and later amendments would always put it at risk. Accordingly, the committee agrees with officials that a back-up interpretation rule is a sensible precaution. The question is, does section AA 3(1) as drafted fulfil this backup role?
2.39 The committee believes that it does not. It is a general principle of constructing not only statutes but all documents that a document should be interpreted as a whole. Accordingly, when a court looks to context to interpret a statutory provision, it will inevitably be influenced by the rest of the Act. Moreover, as has been explained, 'there is no equity about a tax'26. That is, courts approach rules in tax statutes neutrally, without preconceptions: if a rule appears at first sight to operate in one context, that alone is no reason to conclude that it does not also operate in a different context.
2.40The committee returns to the example mentioned above, an anti-avoidance provision that is inadvertently framed in a way that enables taxpayers to use the provision to qualify for deductions that the legislature had not intended. In the committee's opinion, if the anti-avoidance rule were placed in a local context of a series of other anti-avoidance rules the court would take that location into account. But by itself, the placement of the provision within such a local context would not be enough to frustrate a taxpayer who wished to rely on the rule for some other purpose if the language of the rule supported that reliance.
2.41 The committee recommends that drafters should revisit section AA 3(1) and make explicit that provisions that are intended to operate only in a particular manner or in a particular context, should not apply in other contexts.
2.42 The 1994 Act is organised in fourteen parts, A to O, plus part Y, which contains amendments, repeals, savings, and transitional provisions. Each part after A is organised in subparts, for example DA, DB, DC, and so on. In each part, the first subpart is reserved for purpose provisions: BA, CA, DA, EA, and so on. So far, only subpart BA has been drafted and inserted into the Act.
2.43 As mentioned in para 2.25, the committee agrees that it is worthwhile attempting to draft section AA 1 to express the fundamental purposes of the Act. However, the committee has reservations about the formulation adopted. There is both ambiguity and superfluity in the current terminology of section AA 1. As to ambiguity, the word 'income' in its ordinary sense, does not capture all of the intended tax base, for example, as the committee points out in para 2.48, the Act taxes certain capital gains. As to superfluity, paragraphs (b) and (c) of section AA 1 do not add to paragraph (a).
2.44 One possibility is to redraft section AA 1 to say that the main purpose of the Act is to impose a tax on income as that term is used in the statute. The drawback of this draft is that it makes section AA 1 clearly only a mere description of the Act.
2.45 The difficulties in drafting section AA 1 do not markedly diminish when considering drafting purposes for individual parts of the Act. In fact, the Inland Revenue Department advised that the rewrite process may move its focus from part-based purpose provisions to purpose provisions that relate to subparts. Even that exercise may prove unrewarding, depending on the relative heterogeneity of the subpart in question. However, where it is helpful, the committee certainly sees value in purpose provisions for specific sections or groups of sections that have a common subject matter.
2.46 The challenge of drafting meaningful purpose provisions for whole parts of the Act may be illustrated by examining part C, Income Further Defined and part M, Tax Payments, by way of examples.
Purpose provisions and part C
2.47 Part C, Income Further Defined, brings into the income tax net most of the multifarious and miscellaneous kinds of income that one could imagine, together with a number of categories of receipt that are treated as income but that would ordinarily be regarded as receipts of a capital nature. At one extreme is the generality of section CD 5, which captures all items that are income according to the ordinary meaning of the word, though it uses the unnecessarily tortured prose of 'any amount that is included in gross income under ordinary concepts' to do so.
2.48 Among the provisions that capture as income items that would be capital according to the ordinary usage of the term are section CD 1(f) (in some circumstances, profits27 on land held as a capital asset that is subdivided and sold); section CE 1(e) (premiums paid for leases); section CF 2(1)(f) (certain issues of bonus shares by companies); and section CD 4 first limb (profits28 on the sale of personal property held on capital or private account where the taxpayer has a business of dealing in property of a similar kind). The committee has not overlooked that in a line of cases beginning with Hazeldine v CIR29,the courts drew the teeth of the predecessor to this provision so that the rule does not have the effect that the drafter appears to have intended. Hazeldine holds that the first limb rule applies only to property that the taxpayer has committed to his or her business. Wilson J came to this conclusion notwithstanding that profits on the sale of a property that has been committed to a business are taxable as business income in any event.
2.49 Examples of particular provisions in part C that capture particular kinds of income include subpart G (income attributed to New Zealand taxpayers from controlled foreign entities and foreign investment funds); subpart I (fringe benefits); and subpart J (income from minerals, from films; and from certain special kinds of transactions that relate to petroleum mining). Part C also includes rules about life insurance, primary producer cooperatives, international sea freight and renting films. To climax the miscellany, subpart CB contains 15 sections that, in contrast to the rest of part C, are about exempt income.
2.50 Part C also illustrates the difficulty of composing purpose provisions for subparts. For example, referring to some provisions mentioned in the previous paragraph, subpart CJ covers a wide and miscellaneous range of kinds of income from: minerals, timber, flax, films, and four or five different kinds of income from petroleum mining and associated activities. Subpart CN is similar, embracing outward-bound sea carriage, non-resident film renters, and several modes of carrying on life insurance or underwriting businesses.
Purpose provisions and part M
2.51 Part M, Tax Payments, exemplifies a different kind of mixing of categories. Subpart MB deals with provisional tax, which is essentially a collection mechanism. On the other hand, subpart MF deals with branch equivalent tax accounts, which are the mechanism for calculating people's income tax in respect of their interests in controlled foreign companies. Subpart MF does, it is true, involve payments of tax. But it is essentially a fasciculus of substantive rules, albeit framed in a mechanical form, that constitutes a large fraction of New Zealand's controlled foreign company regime. Even in form, subpart MF has little similarity to the provisional tax rules of subpart MB. The other subparts of part M deal with other matters again, some of which have some similarity to one or the other of subparts MB and MF.
2.52 These lists of the components of part M, and of the components of part C discussed above, illustrate the difficulty, and probably the impossibility, of drafting any but the most generalised rules that could capture the purposes of the whole of each of these parts in a way that will help interpretation. Examinations of other parts of the Act lead to similar conclusions. In fact, the Inland Revenue Department drafting team advised the committee that it is coming to the same conclusion: in the team's view it may be that some, at least, of the proposed purpose provisions will never be drafted.
2.53 The second factor that section AA 3(1) directs to be taken into account in interpreting the Act is the core provisions, that is, part B. The function of the core provisions has changed as the rewriting process has progressed. Originally30, the core provisions were to collect the major rules of the Act: the rule imposing income tax, the rule taxing business income, the basic rule that allows deductions, the general anti-avoidance rule, and so on. The reorganised 1994 Act, as originally enacted in December 1994, (before the rewrite of the core provisions by the Taxation (Core Provisions) Act 1996) had the core provisions as shown in Table 1 on page 30.
2.54 The role of the core provisions has steadily developed and now, in a paraphrase of the words of section BA 1, it is essentially to impose income tax and to explain the scheme of the Act and the relationship between its different parts. For example, part B explains that to arrive at taxable income the major steps are: to add all gross income, to subtract allowable deductions, leaving net income. From net income one subtracts any losses carried forward from earlier years, finally to reach taxable income. Each of these steps refers the reader to other parts of the Act. For example, in calculating gross income, one goes to any and all provisions in parts C to I that relate to the taxpayer in question.
Table 1: Core Provisions
(before 1996 No 67)
Corresponding 1976 Act
Income tax imposed
Rates to be fixed by annual taxing Act
Liability to tax of income derived from New Zealand and abroad
Items included in assessable income
65(a) excluding the proviso
65(2)(e excluding the proviso)
Amounts remitted to be taken into account in computing income
No deductions unless expressly provided
Expenditure or loss incurred in production of assessable income
Certain deductions not permitted
Agreements purporting to alter incidence of tax to be void
Rebates to be deducted from income tax
Arrangements for relief from double taxation, and exchange of information
2.55 One of the purposes of the original core provisions was to give the reader a snapshot overview of the Act. In one sense this objective is achieved; that is, the sense in which one gets an overview of a book from a rather terse table of contents. But the new core provisions do not afford an overview in the sense of a summary of the major highlights of the Act.
2.56 That original objective may have been over-ambitious. Now, second and third purposes of the core provisions have come to the fore, and the original objective has been abandoned. The second purpose is to gather in one place the normative links that bind together the other parts of the Act. A third purpose is to add more normative links so as to ensure that each link is made explicit, and that no step of reasoning relies on necessary implication. Providing these additional links may well be a useful function, though it has to be said that much income tax legislation seems to operate well enough without such rules. For example, the former section 65(2)(a) of the Income Tax Act 1976 provided that business profits were assessable as income. It was not thought necessary to have in addition an express rule equivalent to the current section BD 1(1), which ensures that this form of gross income, now caught by section CD 3, is added into a taxpayer's gross income calculation.
2.57 Thus, the core provisions serve as a quick response should, for example, intrepid taxpayers one day be tempted to argue that just because the Act labels an item as gross income it does not necessarily follow that taxpayers must take this item into account when calculating how much tax to pay. But it is not clear to the committee how the current core provisions are likely to help future courts or officials to interpret the Act, as provided by section AA 3(1). At the same time, neither does it seem likely that the core provisions will obscure or hamper the interpretation process: they seem to be neutral as far as interpretation goes.
2.58 Neutrality is not a good enough reason to leave the core provisions as one of the elements that contributes to interpreting the Income Tax Act. Despite the bland appearance of the core provisions, it is impossible to predict whether counsel in some future case may be able to seize on one or other of those rules to support an otherwise questionable argument. The committee recommends that the government should reconsider whether the reference to the core provisions in section AA 3(1) should stay.
The Organisation of the Act
2.59 The third factor that section AA 3(1) directs to be taken into consideration for interpretation purposes is 'the way in which the Act is organised'. The organisational scheme of the Act is one of formal function rather than of regime. This functional scheme entails that provisions that operate in the same manner are grouped together in the same part of the Act. Thus, for instance, provisions framed as deductions are grouped in part D, rules framed as anti-avoidance provisions are in part G, and anything that involves a credit is in part L.
2.60 One result is that rules that deal with substantively quite different kinds of factual situations are found cheek by jowl. For instance, within part F, Apportionment and Recharacterised Transactions, Matrimonial Transfers in subpart FF follows Amalgamation (of companies) in part FE and precedes Apportionment of Interest Costs in subpart FG. Another result is that regimes that in the 1976 Act were collected together are scattered in different parts of the 1994 Act. For example, the rules that apply to controlled foreign companies are mostly in subparts CG, MF, and OD.
2.61 A third result is that the location of a rule depends on its form rather than on its substance. Take, for instance, the rule that taxes the profits on certain land sales when the profit is largely the result of a zoning change. Most of this regime is found in section CD 1, particularly section CD 1(2)(e). The bite of this regime tapers by 10 per cent each year until after the taxpayer has owned the land in question for ten years there is no tax to pay at all.
2.62 It happens that when the rules about zoning were inserted in 1975, Parliamentary Counsel chose to frame the tapering rule as a deduction, first capturing the whole of relevant profits, and then allowing a deduction that grows at ten per cent annually until by year 10 of land ownership, there is no profit left to tax. Because it allows a deduction, the tapering rule is separated from the rest of the regime, and appears in the Act as section DJ 14. The rule could just as well, and more consistently with readers' intuition, have been framed positively, to tax the appropriate percentage of profit for each year of ownership: 90 per cent for one year, 80 per cent for two years, and so on.
Implications of the statute's form-based organisation
2.63 The formal, rather than substantive, nature of the Act's organisation that the committee describes above is a fundamental and pervasive aspect of the rewrite. A thorough understanding of the Act's organisational principles is important to an understanding of the organisation itself. An analogy helps. Consider a criminal code. Criminal codes are ordinarily organised in the manner of the New Zealand Crimes Act 1961. That is, offences of similar kinds are grouped together: fraud, theft, assaults, offences against public order, and so on. As an alternative, it would be possible to organise a criminal code according to penalties or procedures: offences carrying a fine, offences carrying less than three months' imprisonment, offences carrying less than seven years' imprisonment, very serious offences, and capital offences would be a possible categorisation, perhaps overlain by procedural aspects of offences: offences triable by information, by indictment, or by either, and with or without the right of trial by jury.
2.64 In fact, as far as the committee knows, no-one has ever organised a criminal code in the manner just described, but such an essentially formal or procedurally based approach affords a reasonably close analogy to the organisation of the rewritten Income Tax Act.
2.65 The discussion above suggests that the place of a rule in the organisational scheme of the Act will ordinarily shed no light on the way in which the rule is to be interpreted. Indeed, and subject to what is said in the next section of this report, as a general principle, a rule's place in the Act's scheme will not influence the interpretation of the rule. At best, it would be a factor to consider in some cases. The reason is that if a court calls on the scheme of a statute for help in interpreting it, the court does so because it is trying to work out the substance of Parliament's intention in respect of that rule. The place of the rule in the statute and its relationship with other rules may help the court in this task. But if the location of the rule is purely a matter of form and of the way in which the rule functions, and not a matter of the relationship between the rule and other rules that together with the rule form a coherent regime, then generally speaking the location of the rule vis-Ã -vis other rules can shed little light on interpretation problems. Subject to what this reports says earlier about the concept of the local context of a rule within an Act31, at best the location will be irrelevant and at worst misleading.
Organisational scheme and local context
2.66 The situation described in the preceding paragraph is in a sense the opposite of what officials intended. Officials explained to the committee that one hoped-for benefit from the Act's functional organisation was that courts would be more ready to confine rules within their own appropriate local contexts, and would be less inclined to permit, say, an anti-avoidance rule to have the effect of allowing an unintended deduction. The report addresses this issue under the heading 'Local context' above, see paras 2.36 to 2.41.
2.67 The result is that, on one hand, the organisational scheme isolates provisions from other sections of the regimes of which they form part, so that courts lose the benefit of being able to interpret sections in the light of other nearby parts of their regimes. Further, because regimes are scattered it is not possible to have over-arching regime-specific purpose provisions to inform the interpretation process.
2.68 On the other hand, if provisions are clearly identified by function, it may be that carefully drafted interpretation rules could ensure that courts would not allow them to influence the interpretation of provisions with other functions.
2.69 In this last context, in theory the functional organisational scheme has the merit of enabling the courts and other readers to identify the purported function of any particular provision. However, that identification process could be achieved in ways that would not require the various regimes and sub-regimes of the Act to be disaggregated and spread through the statute.
2.70 For example, it would be possible within a part of the Act devoted to, say, the imputation regime or trusts or controlled foreign companies to set aside subparts for anti-avoidance rules, for rules that relate to accounts that must be kept, and so on. In principle, such smaller, regime-specific subparts would serve drafters' purposes better than purely functional subparts, because the operational effect of sections would be limited by both function and regime, and not by function alone, as is the case with the Act's present organisational scheme.
2.71 The committee recommends that the government should review the way in which the Act's interpretation provisions together with the organisational scheme operate in relation to what the committee has called the 'local context' of a statutory rule. The committee endorses the intention of the rewrite process that the Act should be interpreted in a manner that prevents rules from operating outside the context where Parliament intended them to operate, but is of the opinion that the present interpretation provisions do not achieve that goal.
Reference to organisational scheme in section AA 3(1)
2.72 As mentioned, section AA 3(1)'s reference to the Income Tax Act's organisational scheme cannot logically refer to substantive, regime-based, relationships between sections because the scheme does not have a substantive basis. What effect might there be from the direction in section AA 3(1) to courts to take into account 'the way in which the Act is organised'?
2.73 The committee suspects that it is unlikely that courts will be misled. The organisational scheme of the Act is so clearly unrelated to substance that it is hard to see a court being influenced by that organisational scheme in the process of statutory interpretation, notwithstanding the clear direction in section AA 3(1). However, it is hard to predict what might happen in marginal cases, particularly tax cases, which often come before High Court judges who are not familiar with the idiosyncrasies of tax law. Statutory interpretation arguments based on the Act's organisational structure and authorised by section AA 3(1), but having no substantive merit, could lead to perverse results. The effect of this recommendation, combined with the recommendations in paras 2.41, 2.58 and 2.71, is that the committee recommends that section AA 3(1) should be reviewed as a whole.
2.74 If the government accepts the committee's recommendation to review the Act's interpretation provisions insofar as they relate to its organisational scheme these concerns will no doubt be addressed. If that recommendation is not adopted, the committee alternatively recommends that the government should consider whether section AA 3(1) should omit any reference to the way in which the Act is organised, in order to avoid the possibility of perverse interpretations.
2.75 The immediately preceding sections of this report are concerned with the implications that the Act's formal organisational structure has for interpretation purposes. Elsewhere in this report the committee discusses the advantages that those who established the architecture of the Act have sought from the method of organisation that they specified, see paras 2.126 to 2.147.
Relationship between the Acts Interpretation Act 1924 and the Income Tax Act 1994
2.76 Sections AA 1 and AA 3(1) of the Income Tax Act 1994, taken together, probably do not conflict with section 5(j) of the Acts Interpretation Act 1924. Rather, they may be seen as a particular application of the principles of section 5(j), which is probably the intention of the drafter. However, one member of the committee finds the position uncertain, and foresees possible submissions to courts in future cases that sections AA 3(1) and AA 1 together oust section 5(j) and any other relevant general interpretation provisions from application to the Income Tax Act. The committee believes it unlikely that drafters of the Income Tax Act had this intention. However, if sections AA 1 and AA 3 are to remain in the statute, in order to put the matter beyond doubt, the committee recommends that the Income Tax Act should state that its interpretation provisions do not oust any statutory generally-applicable rules of interpretation unless the former are clearly inconsistent with the latter.
The rewrite's objectives as to interpretation provisions
2.77 Redrafting the income tax legislation is one thing. The way in which people, especially judges, interpret the redraft is another. Those responsible for the rewrite have clearly given considerable thought to the interpretation question. One result is that the new Act contains a number of interpretation provisions of novel forms. Of these provisions, this report has discussed sections AA 1 and AA 3(1) at some length. However, in the end the committee has been left in a state of some uncertainty as to just what it is that the rewrite process is attempting to achieve as far as interpretation is concerned. There are five explanations, of increasing levels of intensity:
- There is no intention to change the way in which tax statutes are interpreted at all.
- There is no intention to change the way in which tax statutes are interpreted, but drafters have added or will add some instructional material (notably sections AA 1 and AA 3(2) and the proposed purpose provisions) for the benefit of readers without legal education, who are not familiar with the principles of statutory interpretation.
- While there is no intention to change the way in which tax statutes are interpreted, drafters hope the new interpretation provisions will change the result in particular cases.
- There is no intention to change the general way in which tax statutes are interpreted, but there is an intention to emphasise a principle that rules should be interpreted according to their local context in the Act.
- There is an intention to make a reasonably major change, towards a more purposive style of interpretation.
2.78 Officials' explanations to the committee support variously explanations 1 to 4. Officials disavow explanation 5. The committee deals with these possible objectives below, though in a different order.
2.79 To the extent that the rewrite's interpretation provisions try to instruct lay people in methods of statutory interpretation they are misconceived. The committee hopes that it has made the point in earlier parts of this chapter that statutory interpretation is an inexact, multi-faceted exercise, replete with guidelines that confuse the uninitiated by (a) often appearing to be inconsistent with one another, and (b) masquerading in the form of rules. It is hard to see that choosing two or three of these guidelines (as section AA 3(1) does) and stating them in a very concise form, will help lay people to interpret the Act.
2.80 Take, for instance, section AA 3(1)'s reference to 'the way in which the Act is organised'. It is certainly true that this passage reflects the 'organisation and scheme' approach to statutory interpretation that is often enjoined by modern judges. But that approach does not take us very far in difficult cases. For example, the Court of Appeal in CIR v Alcan New Zealand Ltd32, quoted Richardson J in Challenge Corporation Ltd v CIR33,to the effect that the 'scheme of the legislation' is one of the 'twin pillars' of statutory construction (the other being the 'relevant objectives' of the legislation). But in most cases one can look at the scheme of the relevant legislation at different levels of generality.
2.81 The Alcan case was no exception. One possibility, urged by counsel for the Commissioner, was that the scheme of the Income Tax Act insofar as it applied to corporate groups was that all companies that were members of a group had to be resident in New Zealand. Another possibility, ultimately adopted by the court, was that the word 'company' retained the same meaning throughout the Act, a meaning that included non-resident companies. The court adopted this second approach because it fitted best with the literal words of the Act. But it is hard to see how lay people could be assisted towards a similar result in respect of either the 1976 Act or the rewritten Act by knowing that they are expected to interpret the legislation by taking into account 'the way in which the Act is organised'. If anything, lay people taking the instructions in section AA 3(1) at all seriously risk being misled into undue confidence in their interpretations of statutory ambiguities.
The local context rule
2.82 The committee has discussed the rewrite team's ideas about interpretation according to local context in para 2.52 of this report. It suffices to repeat that while the committee sees some merit in the team's overall policy of, for example, taking precautions to ensure that anti-avoidance rules cannot be exploited to permit contrived deductions, in the committee's opinion the measures taken in the rewritten Act will not achieve that objective.
Change of result
2.83 Officials explained that while the rewrite was not calculated to change the courts' approach to statutory interpretation, it was intended to change the result in particular cases. Two examples were given: the one just mentioned (anti-avoidance rules being exploited to permit deductions) and CIR v Alcan New Zealand Ltd, where the Court of Appeal decided the case by following the literal words of the statute. The result was to sanction a loophole that led to an asymmetry in tax treatment to the advantage of the taxpayer and to the disadvantage of the Commissioner.
Change of approach
2.84 The committee considers that, if the courts were to change their methods of statutory interpretation to a degree that was sufficient to achieve changes of result in cases similar to those just described one could not describe the change as merely a change of emphasis that would lead to different results in particular cases. On the contrary, the changes would mean that the courts are henceforth to take an avowedly purposive approach to interpreting the Act, bearing in mind that its overriding and fundamental purpose is, in the words of section AA 1(a), 'to impose tax on income'. This change moves away from both the pro-taxpayer tilt seen in such cases as in Plimmer v CIR34,and the according of equal status to the literal and the purposive approaches to statutory interpretation, seen in CIR v Alcan New Zealand Ltd. 35
2.85 The committee would see these developments as a sea change in statutory interpretation. They would entail interpreting the Income Tax Act 1994 in a different manner from other statutes. Officials disagree: they do not see the interpretation provisions as having such a profound effect. Further, it was suggested that the examples of particular cases where results would change had perhaps not been well chosen. Probably, the results in those cases would not change, but results in other, presumably closer, cases would be expected to change, though no examples were suggested.
Interpretation provisions: what should be done?
2.86 The committee concludes from the foregoing discussion that more consideration should be given to the interpretation provisions of the rewritten Income Tax Act, and to the objectives that the interpretation provisions are expected to serve.
2.87 Approaches to statutory interpretation do not fit neatly into two categories, strict and purposive. Rather, there is a continuum, leading from the very strict to the very purposive. Generally, common law jurisdictions tend to be stricter in statutory interpretation than civil law jurisdictions. Among common law jurisdictions, New Zealand and other countries that follow the English approach are more literalist than countries that follow the United States approach.
2.88 Some believe that New Zealand would be better off, at least in tax cases, to move along the continuum towards the more purposive approach of the United States. The committee mentioned in para 2.34 the relatively common pattern in tax cases where interpretation is finely balanced: a narrow, literal interpretation helps the taxpayer, whereas a broader, more purposive interpretation helps the Commissioner. The argument is that a shift along the continuum towards a more purposive approach would achieve a better balance between the taxpayer and the Commissioner.
2.89 Some members of the committee would support a change of this nature. Others do not, noting that a possible cost is that the law might become less certain. Most see at least some case for interpreting tax legislation differently from other legislation, though at this stage the case is not yet fully made out. Be that as it may, it appears that the rewrite process has not yet included a thorough study of the courts' interpretation of income tax legislation with a view to determining whether Parliament should be satisfied with present practices and emphasis, or whether the rewrite should mandate a change. The committee recommends that the government should consider this issue and, if there appears to be a case for change, proposals should be formulated for study pursuant to the generic tax policy process.
2.90 Secondly, the committee returns to sections AA 1 and AA 3(1). To the committee, these sections promise more than they achieve. It may be that no official has ever expected that they would do more than explain to lay people how the Act should be interpreted (and the committee has explained that such an expectation is misguided). But such is not the public perception of sections AA 1 and AA (3) 1, nor (in the committee's experience) is it the general expectation in the operations sections of the Income Revenue Department. It behoves the government to decide what the interpretation provisions should achieve, and to have them drafted in a manner that achieves that result.
2.91 An external factor that contributes to the need for this exercise is the possibility that the Acts Interpretation Bill that is currently undergoing study will be enacted. It is at least arguable that the counterpart in the Bill to the existing section 5(j) of the Acts Interpretation Act 1924 mandates a slightly narrower, more literal approach to statutory interpretation than does the text of section 5(j). In particular, the Bill does not deem all legislation to be 'remedial', and uses much more restrained language than section 5 (j)'s familiar 'fair, large, and liberal'. If the Bill indeed mandated stricter approach than the 1924 Act, the government needs to be aware that the Bill will apply to all legislation including income tax law.
Plain language drafting
2.92 The rewrite attempts to redraft the text of the Act in plain language, so that the probable users can follow it. Broadly speaking, the committee agrees that this level is appropriate, and, so far as redrafting has gone, the committee is of the opinion that the drafting team has achieved the language level that was planned. In this area of language, however, the committee has several reservations.
2.93 The first is that the government should be careful not to raise too far people's expectations about how easily they will be able to understand the rewritten Act. The reason is that several factors cause difficulties for people who try to understand statutory rules, and unfamiliar or long words and elaborate sentence constructions are only one of those factors.
2.94 Equally important is the factor that all disciplines use terminology that carries a great deal of freight. This usage is inevitable if people are to avoid explaining fundamental, or even higher level, concepts whenever they make a statement or frame a rule within their discipline. Tax law is no different. Take, for example, section BC 3(2), which reads:
If a taxpayer has one type of schedular gross income for an income year, the taxpayer's schedular income tax liability for the year is the amount that would be the taxpayer's income tax liability for the income year if the taxpayer's only gross income for the year were that schedular gross income.
2.95 Apart from 'schedular' (which has no particular meaning in this context apart from acting as a kind of label) each word in this passage is relatively plain and simple. Yet it is unlikely that many people could understand the passage without some study. Plain language alone does not immediately make a statute clear.
2.96 A further reservation as to comprehensibility of the legislation arises from the use of interpretation provisions. Interpretation provisions can contribute greatly to the accessibility of the core meaning of statutory rules. But if too much of a rule is moved to the interpretation section, the rule becomes hard to understand and to reconstruct. This appears to have happened to the general anti-avoidance provision, which has been disassembled from its former incarnation as a single rule, section 99 of the 1976 Act, and now appears partly in each of sections BG 1, GB 1, and OB 1, the definition section.
2.97 Anyone coming fresh to section BG 1 would have some difficulty in working out what the section is about without referring to section OB 1, which defines 'tax avoidance' and 'tax avoidance arrangement'. Some definitions are truly definitions. Others form parts of their rules, parts that have been split out in order to simplify the drafting of the rule. The two categories are not precisely distinct. But definitions that fairly clearly fall into the second category should not be moved to definition sections simply because they are drafted in the form of definitions. They make the rules to which they belong more understandable if they remain with those rules. When such a definition is used also in other rules, the definition section can direct users to the section where the term in question is defined. Other examples include definitions that more helpfully fit within the trust regime than in section OB 1, such as 'beneficiary income' and 'qualifying trust'.
2.98 A second problem with definitions has been endemic to income tax legislation for decades, but is exacerbated in the progress of the rewrite so far. This problem is the re-phrasing of rules as definitions. An example is the provision that captures as income the value of employer-provided accommodation. This rule is now part of the definition of 'monetary remuneration', in section OB 1. Another example, already present in the legislation before the rewrite, are the rules that tax certain company distributions that are not strictly speaking dividends in the hands of shareholders.
2.99 Generally speaking, where a rule is to provide that a particular receipt is to be taxed, it is better to frame the provision as a rule, rather than to define the receipt as something that it is not, followed by another rule that taxes the item that is defined. The committee recommends that these considerations should be borne in mind in the rewriting process.
Broad-brush factual tests and concepts
2.100 The overall impression given by income tax legislation is of a huge fasciculus of one complex rule after another. There is considerable fact behind this impression, but it masks another fact: that within the statutes there are a good many concepts and tests that are framed in very general terms. Examples include 'capital', 'tax avoidance arrangement', 'business' and 'reasonable' Section BD 2(1)(b)(ii) permits the deduction of an expenditure or loss to the extent that it is 'necessarily incurred by the taxpayer in the course of carrying on a business for the purpose of deriving the taxpayer's gross income.' Section OB 1 defines 'business,' for most purposes in the Act, as including ' any profession, trade, manufacture, or undertaking carried on for pecuniary profit.'
2.101 Section BD 2(2)(e) prohibits the deduction of expenditure or loss that is 'of a capital nature' (unless it is specifically permitted by another section). The words 'capital' and 'capital nature' are not defined.
2.102'Tax avoidance arrangement' is defined in section OB 1, in terms of the 'purpose or effect' of an arrangement. The definition of 'tax avoidance' itself is:
'Tax avoidance', in sections BG 1, EH 1, GB 1, and GC 12, includes -
- Directly or indirectly altering the incidence of any income tax:
- Directly or indirectly relieving any person from liability to pay income tax:
- Directly or indirectly avoiding, reducing, or postponing any liability to income tax:'
2.103 Section 144(4) of the Tax Administration Act 1994 provides:
(4)No person is to be convicted of an offence under subsection (3) for not presenting an instrument for stamping, if the person can show reasonable cause for the person's failure to present an instrument for stamping within the time specified by the Commissioner under section 53 of the Stamp and Cheque Duties Act 1971.
2.104 Section 141A provides:
- A taxpayer is liable to pay a shortfall penalty if the taxpayer does not take reasonable care in taking [a taxpayer's tax position] (referred to as 'not taking reasonable care') and the taking of that tax position by that taxpayer results in a tax shortfall.
- The penalty payable for not taking reasonable care is 20 per cent of the resulting tax shortfall.
- A taxpayer who, in taking a taxpayer's tax position, has used an acceptable interpretation of the tax law is also a taxpayer who has taken reasonable care in taking the taxpayer's tax position.
2.105 The committee considered a suggestion that these kinds of expression should be expanded and made more specific. Just what is meant by 'capital', 'tax avoidance arrangement', 'business', or 'reasonable'? Some of these concepts are defined in the Act, but the definitions are fairly general. The suggestion was that they should be made more specific.
2.106 The committee did not adopt this suggestion. The problem that the suggestion identifies is one that is found throughout law. It is that law often needs to refer to a category of fact or facts by an omnibus term. The core meaning of the term is ordinarily clear enough. The difficulty is to define its boundaries. For example, most activities that might be called 'businesses' clearly are businesses. But there are some activities at the edge of the concept of business where people may reasonably disagree: are they businesses or not? Examples from the cases include carrying on the profession of peripatetic evangelist36 and farming blocks of land of uneconomic size.37
2.107 To define 'business' precisely, it would be necessary to define various categories of activity that amounted to businesses, and perhaps to nominate a number of activities that either are or are not businesses. But this kind of drafting serves only to move the boundaries of the exercise from one point to another. It would not solve generic problems about whether particular activities amount to businesses. In many cases it would make the decision more difficult, as particular sets of facts were tested not against a generic concept but against detailed lists.
2.108 The short answer is that when a rule must refer to a category that is extremely broad and that must embrace an unlimited number of factual situations it is ordinarily unrewarding to try to refine the definition of the category beyond a fairly limited point. Moreover, it is almost always bad practice to try to compose a list of definitions that is exhaustive. The result of that exercise is inevitably to leave out of account examples that would have been included had one thought of them in advance. This is particularly true for tax legislation, which is subject to continuous scrutiny by advisers who try to discover routes around rules and definitions. For these reasons, the committee does not accept the suggestion that broad factual concepts within the Act should be made more specific, much less that they should be redrafted to be exhaustive.
2.109 Later in this chapter, at para 2.160 to 2.164, the committee discusses the codification of judicially formulated rules. While in other areas of the law it may make sense to adopt judicial statements, tax law is not so amenable, carrying with it as it does elements of fiscal policy. The committee considers this distinction important, and for this reason cautions against the practice of moving judicial pronouncements directly into tax legislation.
The gross income approach
2.110 The Valabh committee 38identified two approaches to consider in the design of tax legislation: whether the Act should proceed on a 'gross' or a 'net' basis, and whether it should follow a 'global' or a 'schedular' pattern. The 'gross' basis independently calculates gross revenues and gross expenses and offsets the two to produce net income. The net basis calculates net income in accordance with commercial and accounting principles and then modifies the result for tax purposes. The starting point for the statutory calculation is the essential difference. The global pattern calculates either gross revenues or net income (according to which basis is chosen) from all sources, while a schedular pattern does so on a source by source basis. In its Final Report39, the committee recommended the adoption of a global/gross approach.
2.111 The Working Party on the Reorganisation of the Income Tax Act considered the possibility of retaining some special regimes that would result in net amounts being fed into the calculation of income. However it rejected the idea as being contrary to the general global approach of the Act40.
2.112 The consistency described above has not been achieved without cost. Almost by definition, some kinds of income are by nature net, and cannot fit neatly into a legislative plan that is composed on the basis of starting always with gross receipts. Examples are the income of non-resident shipowners and film renters. The drafters' solution has been to create a concept of 'schedular income tax liability', which is reached by several steps that begin with 'schedular gross income', or even with several types of schedular gross income. This elaboration is inevitable, considering the one-size-fits-all basis of the gross approach.
2.113 It is not possible to conclude whether the benefits of the standardised gross/global approach outweigh the costs. To do so, the committee would have needed to go through the provisions of the previous legislation and identify all cases where the term 'income' was used, sometimes in one sense and sometimes in another. The second step would be to inquire which of these different uses caused problems, the third to decide which problems were real and which perceived. Finally, the committee would need to compare problems that arise from inconsistency with problems that arise from the need to legislate around the flanks of the consistency that is now the basis of the legislation. Which problems are worse?
2.114 The exercise described above was beyond the time and resources available, and the committee refrains from judging the merits of the gross approach. On the other hand, the committee is concerned that the switch to the comprehensive gross/global approach was made without the kind of pragmatic testing, described in the last paragraph, that the committee believes was necessary to determine whether the change would be justified. Officials advise that the drafting process and the forthcoming exposure draft of parts C, D, and E of the Income Tax Act will perform this function. The committee is surprised by this news. It entails accepting that one cannot test whether the gross/global approach has more benefits than costs without putting the whole country through the exercise of getting to grips with the core provisions revisions of 1996 and the proposed 1999 revisions of parts C, D, and E.
2.115 Does the foregoing imply that after all this drafting has been done, it is possible that there may be a return to the pragmatic, though occasionally inconsistent, drafting that prevailed before in the 1976 Act? The committee is advised that in fact this idea is not seriously contemplated. The result then is that a rigorous theory of uniformity is being imposed on legislation without prior testing, even though income tax law, the subject matter of the exercise, is by its nature illogical, to a degree unprincipled, and unlikely to be amenable to such an approach.
Gross income and principles of accounting
2.116 The change to gross income as the first step in computing taxable income has entailed moving from business 'profits' as the tax base for business income to 'any amount derived from any business'41. The former terminology had been glossed by a great many court judgments, and, in particular, by a number of judgments that explain the extent to which general principles of accounting should determine when income is derived or expenses are incurred for purposes of calculating assessable income.
2.117 As income tax legislation has developed over the decades, the ambit of these judge-declared rules has become gradually more constrained. For example, the introduction in 1986 of the qualified accruals rules to govern financial arrangements brought a large area of timing rules under specific statutory provision. However, there remain areas where timing questions are a matter of judicial interpretation. The leading example is probably profit-recognition in long-term contracts, where the rules were established for New Zealand purposes in HW Coyle Ltd v CIR42 and Horizon Homes Ltd v CIR43. The problem facing taxpayers is that they are uncertain whether the rules declared in these cases have survived the changes wrought by the 'gross income' approach of the rewrite.
2.118 The committee recommends that the government should resolve this uncertainty, though it has not been able in the time available to settle on a firm recommendation as to how the resolution should be achieved. There are four possibilities.
- Enacting a general rule to say that pre-1994 judge-declared timing rules remain in effect unless they are clearly inconsistent with specific timing rules in the Act.
- Doing nothing, and leaving it to the courts to resolve the matter again.
- Identifying specific areas where this kind of problem exists and enacting area-by-area timing rules, leaving unidentified areas to manifest themselves in due course.
- Attempting to construct a statutory general timing rule that would cover all the areas that are now left to the courts. Though theoretically attractive, this approach appears to be impractical.
2.119 The committee leans in favour of the third option. The committee's preference requires evaluative work because it is a matter of judgment and intuition, rather than being based on an in-depth examination of the various possibilities.
The core provisions
2.120 This report explains in para 2.53 that the core provisions as now constituted are not at all as they were proposed in the 1990 report of the Valabh committee44. The original proposal was for the most basic and important rules in the Act to be gathered together as a core from which readers could obtain an overview of the Act's substance.
2.121 This objective is not consistent with the role for the core provisions that has developed during the rewrite process, to provide a series of normative links between the several parts of the Act. The two objectives cannot be pursued together. However, it does not necessarily follow that the first objective should be abandoned. It may be possible to pursue each objective separately, with one part of the Act containing the new-model core provisions, and another part containing the most important rules of the Act.
2.122 The rewrite process has not pursued the strategy described in the last paragraph. Instead, the important rules that were formerly gathered together as core provisions have been relocated where they seem to fit best in the substantive parts of the Act. The committee does not quarrel with this decision. It has the demerit that the overview envisaged by the Valabh committee is not possible. On the other hand, it avoids having to decide whether an important rule is important enough to be a core provision, or just fails to make the grade. Decisions like that could never be made on a wholly consistent basis. The decision also has the merit of allowing important provisions to join rules of similar kinds, in their appropriate parts and subparts of the Act, though the consequent reordering that is appropriate, to place important provisions first in subparts, has yet to be done.
Alphanumeric section numbering
2.123 Following the recommendation of the Working Party on the Reorganisation of the Income Tax Act 1976, the 1994 Act renumbered the legislation using an alphanumeric system. This reference system compares with that used in the 197645 Act as shown in Table 2.
Table 2: Section Numbering
1976 Act reference
1994 Act reference
* (often not needed)
2.124 It was expected that the new system would enable a ready identification of the location and function of sections as users became familiar with the new structure of the Act. Part C would deal with income; all sections in the part would begin with a 'C', and so on. It was also expected that the new system would permit the insertion of additional material into the Act without breaking the alphanumeric sequence. This pattern will be an improvement on an Act's standard single series of numbers, which requires either the renumbering of groups of unamended sections to preserve sequence where new matter is inserted, or gives rise to such numbers as 394ZZZJ to cope with the cumulative effect of relentless amendment. It was recognised that this system had a cost, in that it was not consistent with the numbering of the rest of the statute book. The Working Party believed that the bulk and frequency of amendment of tax legislation placed it, in this respect, in a category of its own where the benefits of the new system outweighed the costs of that inconsistency.
2.125 While the committee considers that the new alphanumeric system is worthwhile, already some symptoms of the old system have re-emerged. The Valabh committee hoped that it would never be necessary to insert sections into subparts, but that they would be added sequentially. However, sections and paragraphs with modifiers 'A' and so on are being added, for example, section DJ 13A, and sections DK 3A to DK 3E, and within sections, paragraphs (1A) and so on. It is not necessary, of course, to insert whole parts between sections, and it is unlikely that the situation will ever get as bad as the previous system, but the impact of the alphanumeric system is not as positive as the committee had hoped. Inserting provisions into subparts and sections will continue to be a necessary part of the amendment process. Balancing that, the system takes users very little time to become accustomed to it, and will pay increasing dividends over the years as the Act is amended and amplified. The alpha-numeric system would have been worthwhile whether it had been decided to retain a regime-based organisational system, or whether the Act adopted its functional organisational scheme, as it has done.
2.126 The committee's report describes the Act's functional organisation in paras 2.59 to 2.62. The decision to move from the regime-by-regime structure that New Zealand and all other jurisdictions with which the committee is familiar to a functional organisation was both major and brave. There is a very heavy cost, one that can be justified only if there are commensurate benefits.
2.127 The cost is the obvious one: if taxpayers want to discover how a particular regime applies to them it makes sense for them to be able to find the regime set out in one place in the Act. This remark applies equally to regimes that are defined by reference to business form, such as dividend imputation or controlled foreign companies, and to regimes defined by reference to industry segment, such as life insurance or forestry. Indeed, for two or three years now it has become a clichÃ© among tax advisers that there is a captive and increasingly impatient market waiting for the commercial publisher who will dismember the 1994 Act and reassemble it in a thematic form.
2.128 The committee has no doubt that, other things being equal, a regime-by-regime structure is superior to a functional structure from the point of view of comprehensibility by, and ease of use for, users of the Act. The committee envisages that a regime-by-regime structure would begin, after the core provisions, with a part or parts of the Act that would set out the substantive rules that apply to all forms of income in the absence of special considerations. Later parts would contain regimes that are activity-specific (such as forestry or life assurance), or that are specific to certain investment or trading structures (such as companies or foreign entities).
Claimed advantage of functional structure
2.129 As far as the committee has been able to discover, the only advantage claimed for the functional structure over a regime-based structure is that the former allows to be brought together in one place provisions from different regimes that, within their several dispensations, operate similarly to one another. There are said to be four benefits.
- If several rules turn out to contain almost the same language they can be consolidated into an omnibus rule that can be made to apply within a number of separate areas of the Act.
Even if there cannot be consolidation the rules can be rewritten using standardised phraseology.
A functional scheme reduces repetition.
In the view of officials most importantly, a functional scheme will help future policy makers and drafters.
2.130 The committee does not find these benefits compelling. The second can be achieved whether the rules in question are gathered together or scattered. It is simply a matter of applying a common drafting template to rules that operate in a similar manner to one another wherever they may be found in the Act. It is true that common drafting templates have not been applied consistently in the past, but that is more a result of fashions of parliamentary drafting coming and going, rather that of any great difficulty in the task itself.
2.131 At first sight, the first suggested benefit carries more weight. Consolidation of numbers of similarly-phrased provisions into a single omnibus rule has attractions. However, it is not clear to the committee just how extensive this consolidation will be able to be. There are two or three provisions in subpart GC, being specific anti-avoidance rules, that might be able to be consolidated. Probably, at least some of the associated parties rules in subpart OB could also be consolidated. No doubt there are other examples. However, the committee suspects that, like the examples that it has given, most of the rules that might benefit from consolidation would be discovered to be ancillary rather than substantive provisions. Further, the committee suspects that the total of such provisions would turn out to be a rather small fraction of the whole Act, too small to justify the formalistic rebuilding of the statute in the functional scheme that has taken place.
2.132 Thirdly, although a functional scheme stands to reduce one kind of repetition it increases another. The reduction may come about as a result of the consolidation of rules that have similar language, which was described above. As explained by drafters, the increase comes from the need in each subpart to reintroduce, at least to some extent, the topic of a rule that, in a functional scheme, is isolated from the other rules that are part of the rule's regime.
Help for policy makers and drafters
2.133 As it was explained to the committee, the functional scheme will help future policy makers and drafters when the Act is amended or when new regimes are added. The segregation into statutory parts of rules that relate to deductions, timing, apportionment, avoidance, and so on is hoped to have the effect of ensuring that each rule of a new regime is placed in its proper context. The statute's parts and subparts will in effect act as guidelines for future drafters, and it will not be necessary to rely on institutional memory. Segregating rules should ensure that inadvertent looseness of drafting will not cause rules to have unintended effects, because, for example, an apportionment rule will find itself in the apportionment part of the Act, and taxpayers will not be able to argue that the rule gives them, say, a timing advantage.
2.134 Elsewhere, the committee's report comments on this 'local context' canon of statutory construction that officials advise is part of the prescribed drafting policy46. If it is to be effective, it needs to be made explicit. There is also the desirability of subjecting this and other changes to existing statutory interpretation practices to the rigours of the generic tax policy process.47
2.135 A second benefit is that when drafters add, say, an anti-avoidance provision to the Act they will be able to check on the Act's other anti-avoidance provisions, conveniently gathered in one place, to ensure that the language and approach of the new provision are consistent with what already exists.
2.136 The committee is not persuaded that the help that the Act's organisation is expected to afford to future drafters justifies employing a functional scheme rather than a scheme that is based on substance. Similar guidelines for identifying and segregating rules according to function could operate within regimes, with subparts of regimes dedicated to particular functions. The same idea of having subparts within regimes dedicated to rules of particular kinds could be the basis of an interpretation principle that attempts to confine rules to their own contexts.
Ease of use of the Act
2.137 The most notable feature of the reasons that are advanced to justify a functional rather than regime-based scheme for the Income Tax Act is that all the reasons relate more to the processes of policy-making and of drafting than to the question of ease of comprehension by the user. This feature is particularly true of the last reason.
2.138 One suggestion from officials was that concern about ease of comprehension and about questions of users being sure that they could find all rules relevant to a particular regime that interests them is misplaced, considering people's increasing use of electronic forms of the legislation that are machine-searchable.
2.139 The committee is uneasy about giving much weight to this last suggestion. For a start, the suggestion in a sense confirms the committee's impression that the functional scheme is not friendly to readers, and needs repackaging before it can be used easily. Secondly, while it is true that increasing numbers of tax professionals use electronic versions of the Act, many general practitioners who refer to the Act from time to time do not. Thirdly, the committee has the impression that even those who do use electronic versions of the Act tend to do so for references to particular, short passages, but that they turn to paper-based versions to get, for example, an overall understanding of a regime, or to read passages longer than a page or so. Finally, the committee notes that it is currently policy to draft a paper-based statute, not an electronic database.
2.140 The committee understands and sympathises with the concerns of the government as to the policy making and drafting process and understands the perceived benefits that are thought to flow from a functional scheme. However, to put the matter at its lowest, and referring to its terms of reference, the committee is unable to conclude that a functional statutory scheme is likely to make 'tax laws more coherent and understandable' than the more intuitively understandable regime-based alternative.
2.141 The logic of the committee's position is to recommend a change to a regime-based structure. At the present stage of the rewrite that change would be significant shift of direction in terms of design. The committee notes however that the implementation of the original design, in terms of actual redrafting, still has a great distance to go, with drafters still working on the first exposure draft of parts C, D and E at the time of this report. The committee has not enjoyed the time or the resources to examine the Act or the progress of the rewrite process in enough detail to be confident in recommending such a major change. There needs to be an assessment of the value of the benefits to be obtained from continuing on the present course together with an assessment of the work involved in reorganising the statute on a regime by regime basis. The committee recommends that the government consider having those assessments done, with a view to deciding whether to persist with the functional organisation of the Act. The committee would be concerned if this recommendation would cause the project to be significantly deferred. An alternative recommendation would be for the government to bear the committee's comments and recommendations in mind, and to accommodate them as much as possible.
2.142 In making its recommendation, the committee bears in mind that officials advise that the whole rewrite process remains to some degree experimental even at this late stage. It is not clear just which drafting design principles remain experimental, but if the functional structure is one of those principles the committee urges that it should be abandoned.
Ordering of sections
2.143 When the former legislation was split up and sorted into the functional classification that the rewrite follows, sections were often put into appropriate parts and subparts without much concern for logical order within subparts. Subpart CD is a good example. This approach was undertaken wittingly, in order to make progress, and in the knowledge that there would be an opportunity to improve the order later in the rewrite process. In some subparts, logic was further eroded when part B became the more formal structure that it is today, and the substantive rules of the original core provisions were moved to other parts of the Act. The plan is that ordering of sections into logical sequences should occur part by part as rewriting goes on.
2.144 As the committee prepares this part of our report in November 1998 a good deal of the Act remains ordered in a haphazard manner. The committee does not criticise the present ordering, because it appreciates the reasons that have led to the current situation. However, one result is that in evaluating the rewrite the committee cannot comment on the ordering of sections because that exercise is still to be done, except for part B.
2.145 The committee notes with approval, however, that, ordinarily, subparts within parts, and sections within subparts, will start with the important or the general and proceed to the less important or the particular.48
2.146 The committee notes that part B, which has been ordered, does not always follow the pattern of more important or general before less important or particular. For example, section BC 2 (non-filing taxpayers) precedes section BC 3 (taxpayer with schedular gross income) which in turn precedes section BC 4 (taxpayers with annual gross income). No doubt the reason was to clear non-filing taxpayers and schedular gross income out of the way before proceeding straight from the annual gross income in sections BC 4 to BC 5 (deductions), BC 6 (net income) and BC 7 (taxable income). The drafters' ordering is appropriate.
2.147 The committee understands that the rewrite and drafting teams agree that, ordinarily, the general should precede the particular and the important the less important, and that the teams already approach their task from this point of view. The committee mentions the ordering question from an abundance of caution, because it is aware of criticisms of the present sequencing. These criticisms fail to take account of the fact that the present order is a temporary parking order, awaiting the next stage of the rewrite.
Principles of drafting
2.148 Legislative drafting should follow a number of reasonably commonsense principles. For example, generally speaking drafters should express the essence of a provision first, separately from limitations or expansions, which follow.
2.149 The committee has not evaluated the rewrite for compliance with this kind of principle. The reason is that so far only the core provisions have been released, and they are unlikely to be typical of the drafting of the Act as a whole. Nevertheless, the Inland Revenue Department has engaged competent staff, one with very long legislative drafting experience, to compose the rewritten statute. The committee has spoken to several members of the drafting team, and is confident that the drafting process is in good hands.
The general income provision
2.150 It is a sensible precaution for drafters to include in income tax legislation an omnibus provision that captures simply 'income' according to the ordinary meaning of the word. In Australia, this provision has historically been section 25 of the Income Tax Assessment Act 1936, which simply says that taxpayers' assessable income includes their gross income, without defining the term. The Australian section 25 may be thought of as an umbrella provision: underneath the umbrella, and often extending outside its coverage, later sections capture many other kinds of receipt as assessable income. But section 25 ensures that nothing that is 'income' according to the ordinary meaning of the word escapes tax.
2.151 New Zealand has used the opposite structure: a safety net, rather than an umbrella. In the Income Tax Act 1976 section 65(2) there appeared a list lettered from (a) to (ka) of receipts that were deemed to be included in assessable income. The final item in the list was '(l) Income derived from any other source whatsoever.'49 That is, if any receipt escaped all of paragraphs (a) to (ka), paragraph (l) would catch it if it was income according to the ordinary meaning of the word. In the rewritten statute, the former section 65(2)(l) is replaced by section CD 5. The committee comments on both the drafting and the position of section CD 5.
Drafting of section CD 5
2.152 Section CD 5 reads: 'The gross income of a person includes any amount that is included in gross income under ordinary concepts.' This locution appears to have its origin in a dictum of Jordan CJ in Scott v CT (NSW)50, where the Chief Justice said: 'The word 'income' is not a term of art, and what forms of receipts are comprehended within it must be determined in accordance with the ordinary concepts and usages of mankind ' That is, Jordan CJ was explaining how 'income' is to be interpreted in a tax statute.
2.153 It is supererogatory for part of Jordan CJ's explanation to be imported into section CD 5: the purpose of the section is better served by using the word 'income' unadorned. In fact, if the adornment has any effect, it must be to qualify rather than to amplify the meaning of 'income' when the term is used by itself. That is, logically, the addition of the words 'under ordinary concepts' mean that 'income' cannot extend to any meanings that it might bear over and above meanings that are 'under ordinary concepts'.
2.154 A second problem is the word 'under'. In the English language something can be 'under' a rule, but it cannot be 'under' a concept. No doubt, 'under' supplanted the more usual 'according to' by way of an exercise in plain language drafting. Curiously, the title to the section uses 'according to', as though the editor was not willing to follow the drafter into the uncharted territory of being under a concept. Good intentions have led the drafting astray.
2.155 For these reasons, the committee recommends that 'under ordinary concepts' should be eliminated, and section CD 5 should be rephrased using the term 'income' without qualification.
Position of section CD 5
2.156 In some senses, section CD 5 is the most important provision in the Act. It is, after all, the section that captures income in the most general sense of the term. It is for this reason that the committee has devoted a long explanation to its recommendation about the drafting of section CD 5.
2.157 For the position of section CD 5 in the statute, the committee favours the umbrella structure that is employed in Australia over the traditional New Zealand safety net formula. There are two reasons. First, it seems to the committee to be more intuitively reasonable to start with a general taxing provision and then to amplify it with specific rules, rather than to proceed from the particular to the general.
2.158 Secondly, giving section CD 5 primacy of place as the most prominent and most general charging rule should help to minimise the number of occasions when the Commissioner loses a case as a result of relying on the wrong charging rule and discovering the mistake too late to be able to correct it. VH Farnsworth Ltd v CIR51 is an example. If the Act's charging provisions begin with a general umbrella rule the Commissioner's usual practice should become to rely first on this general rule and additionally on any relevant specific rule. The chances of taxable profits escaping through being charged with an incorrect, narrowly focused rule should be minimised.
2.159 As it is currently placed in subpart CD, section CD 5 falls between being an umbrella and being a safety net. That is, some charging provisions precede it, and some (rather more) follow. The committee recommends that section CD 5 should be the first charging provision that places 'income' within 'gross income' and, as the calculations are worked through, 'net income' and 'taxable income'. Another possibility would be to weave section CD 5 into section BC 4, where it could act as the core provision that initially captures gross income as annual gross income.
Codification of judicially-formulated rules
2.160 It is a common practice in the codification of laws for Parliament to adopt judicial formulations of rules and to arrange them into a systematic matrix. This approach has carried over into income tax legislation. For example, for many years New Zealand's test for corporate residence followed passages in The American Thread Co v Joyce52 and De Beers Consolidated Mines Ltd v Howe.53 Even now, as section OE 2 of the Income Tax Act 1994, the test continues to exhibit traces of that early influence. Another example is what is now section CD 4 third limb (the taxation of profit-making schemes), which follows Ruhamah Property Co Ltd v FCT54.
2.161 When lawyers' law on topics like the sale of goods or the law of partnership is codified, it often makes sense to adopt judicial statements from the cases. The reason is that judges have heard arguments on the relevant issues, and are in a good position to formulate legal principles that take account both of other relevant rules and of competing economic interests.
2.162 These considerations carry over only imperfectly, if at all, to the tax area. Judges are ill-equipped to make fiscal policy, and disavow trying to do so. As a result, judicial statements of tax law will typically be either interpretations of existing statutory rules, or an effort to make sense of what judges believe the law to be. Either way, judicial statements may or may not reflect the law that tax policy makers would formulate if they started from a basis of trying to put into effect the most appropriate fiscal policy for the transaction in question.
2.163 It follows that tax policy makers should not uncritically adopt judicial statements of law as embodying appropriate fiscal policy55. In the opinion of the committee, government tax policy makers and tax law drafters should first determine what the law should be from the point of view of economic policy. If drafters then want help in constructing an elegant formulation it makes sense to turn to judgments to see whether they help. But judgments should not be looked on as a source of economic policy.
2.164 There are sometimes suggestions that New Zealand should continue its former practice of moving judicial pronouncements straight into the Income Tax Act with little or no editing. Something of that nature may have happened with section CD 5, discussed in the preceding sections of this report. The committee cautions against that practice.56
2.165 The process of rewriting the Income Tax Act is expected to last some years. The Act in force today is a mixture of rewritten provisions (the core provisions) and provisions yet to be rewritten. Current plans are for the future results of the exercise to be enacted in two bills that are likely to be a year or more apart. During this time it is inevitable that some provisions will be rewritten or moved several times, as the implications and consequences of later decisions and of changes made to later rewritten provisions are worked through the core provisions and on other provisions enacted earlier. This situation involves an obvious cost to users. If the committee's recommendation elsewhere in this report, that schedule 23 should be continually updated in the manner described in para 2.185 is implemented, that would go some way towards addressing that cost.
2.166 The benefit of adopting this staged implementation of the rewrite is the earlier availability of portions of the Act showing the benefits of the rewriting. An alternative was for rewrite to proceed in parallel with the ordinary annual amendment process, but be enacted only when it was substantially finished.
2.167 The choice between enactment in stages and enactment only on completion of the whole Act involves a difficult judgment, but the committee believes that the government was probably correct to opt for an ambulatory process.
The tandem simplification/substantive improvement programme
2.168 A distinctive feature of the New Zealand rewrite project, in contrast with approaches taken by concurrent rewrite projects in Australia and the UK, is the decision to include within the scope of the project the making of substantive changes in policy and to the the law, as opposed to limiting the project's mandate to re-expressing the status quo. Relatively minor changes, and the resolution of less contentious ambiguities, are able to be undertaken by an accelerated path, under the overview of an independent advisory panel which reports separately to government. Major changes for which a rewrite bill is expected to be the enactment vehicle are fed through the generic tax policy process in the same manner as other policy initiatives. The broader scope of the New Zealand project, which includes the opportunity to address matters of substance, enhances the chances of the project succeeding in clarifying the legislation. The committee believes this approach is advantageous.
Topics addressed in tandem with the rewrite
2.169 Officials advised the committee that the following substantive topics are being addressed in tandem with rewrite of parts C, D, and E, timing issues, property transactions, death and deceased estates, and self-assessment.
2.170 In addition to matters dealt with as part of the rewrite, the enactment of remedial legislation and government policy initiatives in non-rewrite 'business as usual' bills is expected to continue while the rewrite is underway. Two such bills57 were introduced and Acts passed in 1998, and a third bill58 was introduced. Among topics conspicuous by their absence are apportionment, movement of assets in and out of the tax base, accounting for long-term contracts, and tax avoidance provisions.
2.171 The committee recommends that the government should give a higher priority to these topics in the rewrite programme.
Review of compliance-intensive regimes
2.172 The tandem approach affords an opportunity to evaluate existing regimes to see whether the original reasons for the rules still obtain, and whether base-protecting elements can be removed, streamlined, or replaced with less compliance-intensive alternatives.
2.173 Officials have advised that the rewrite process does in fact focus on this kind of issue. The committee notes with approval that as part of the generic tax policy process, the rewrite exercise will include an element of cost/benefit analysis of the balance between compliance costs and the revenue expected to be raised, in respect of the portions of the rewrite exercise that must pass through the generic tax policy process.
2.174 The committee commends the focus on compliance cost reduction and looks forward to seeing the results when the rewritten legislation is progressively released.
2.175 In its Second Report, the Working Party in the Reorganisation of the Income Tax Act 197659 noted that it had tested the functional structure that it proposed for the Act. However, as reported by the working party, that testing established two things.
2.176 First, every existing provision of the 1976 Act could be fitted into one of the fifteen strategy parts that the working party proposed, leaving no existing provision without a home. Secondly, all possible additions to the Act would be found an appropriate place. Such possible additions included previous regimes that had been repealed, and regimes that are found in foreign systems but not in New Zealand. The fact that all such regimes that the committee knew about could be accommodated gave the working party confidence 'that the parts are both robust and durable'60.
2.177 The committee does not agree with the working party's conclusion as far as it goes. However, the committee has in mind deeper and more extensive testing than the working party reported. To pick up several of the committee's points, the exercise should have tested whether:
It was likely that purpose provisions could usefully redrafted for parts and subparts.
It was feasible to consolidate any significant numbers of similarly worded rules into single omnibus rules.
Eliminating inconsistencies in the use of word 'income' by adopting a uniform gross approach is worth the cost of creating the awkward concepts and rules that are a consequence.
Adoption of drafting policies without testing
2.178 One of the initial approaches to the rewrite was to adopt firm policies and principles and to attempt to follow these principles rigorously. Two such policies in the rewrite of the income tax legislation were to begin each part of the Act with a purpose subpart, and to collect provisions that function in a similar manner (not necessarily that share a similar function) together, with a view to consolidation.
2.179 As mentioned in para 2.45, it has now become apparent that it is unlikely that part-based purpose provisions can be drafted in any useful manner, assuming the functional structure of the legislation is retained. The Inland Revenue Drafting Unit has advised the committee that it has reviewed the original proposals for the use of purpose provisions in the rewrite, in the light of more recent New Zealand and United Kingdom62 discussion of 'purposive drafting', and that it now intends to consider the merits of using such provisions on a case by case basis rather than assuming that a global or systematic application of them is appropriate. This development means that it is unlikely that the systematic function originally envisaged for subpart A of the various parts of the Act will be preserved. Secondly, it is far from clear that it will in the end prove possible to consolidate any significant numbers of provisions, even though the prospect of achieving that goal was the major driving force behind the Act's functional organisation.
2.180 Each of the policies described in the paragraphs above (starting each part of the Act with a purpose provision and gathering functionally similar rules together with a view to consolidation) appears logical and sensible when stated simply as a policy. In most areas of law, it is probably true that these policies could be put into effect when drafting extensive codes. But these policies do not allow for the heterogeneous and often internally inconsistent nature of income tax law.
2.181 The committee believes that before policies such as those described are allowed to govern a major drafting exercise they should be tested empirically. Such testing would require first, the gathering together the component sections of two or three proposed statutory parts (redrafting would not be needed; the sections could be plucked straight from their former contexts). Secondly, drafters could attempt to compose meaningful purpose provisions and effective consolidated sections. Success at these exercises would not establish that they would be successful throughout the Act; nor would failure in respect of any one possible statutory part establish the opposite. The exercise would be more reliable if repeated for several proposed parts. The results would give an indication of whether the proposed policies could usefully be put into effect.
2.182 The committee notes that the drafting unit has advised that it sees the current drafting process as itself testing some of the detail of the drafting policies originally prescribed, and that it expects comments on the proposed exposure draft to provide it with independent views as to the success or otherwise of drafting policies that are adopted. However, at the same time officials advise that the kinds of relatively structural policies that the committee has discussed, such as the functional scheme of the Act, are unlikely to be subject to review. The committee notes that policies of the kind now under discussion did not originate within the Inland Revenue Department, but were approved by Ministers on the recommendation of consultants. If the Inland Revenue Department has a responsibility, it arises from embarking on the policies without testing them and advising Ministers of the results. Had tests been carried out, the difficulties which have been encountered in the rewrite process may have been avoided.
2.183 The committee recommends that statute-wide drafting policies should not be adopted as a matter of principle without reasonably rigorous practical testing. What appear to be significant difficulties in the rewrite process might have been avoided by pre-testing of this nature.
Two separate Acts
2.184 The committee considers that there might be some advantage in splitting the Income Tax Act 1994 into two separate Acts. If feasible, such a division could produce an Act of more manageable size for provisions of general application, and remove some relatively complex groups of provisions that concern only a limited number of taxpayers into a second Act. Such a second Act might be used to carry the provisions relating to cross-border transactions and certain industry specific regimes, such as petroleum mining and life assurance. The committee has not worked out the detailed implications of this proposal, and makes no firm determination as to its ultimate feasibility. The committee recommends that the government should direct officials to evaluate whether such an approach should be taken.
2.185 Schedule 23 of the Income Tax Act 1994 contains a series of tables to convert section numbers between 1970s and 1990s legislation. The schedule is very useful, but its utility is progressively eroded as the rewrite process re-orders section numbers within the 1994 Act. The committee appreciates that as sections are moved from one place to another in the Act it becomes increasingly difficult to compose a table that tracks their history. Designing tables with increasing numbers of columns is only one problem. Nevertheless the committee recommends that schedule 23 should be kept under continuous review, and updated whenever there is renumbering, or, at least, whenever there is a reasonably significant renumbering exercise, such as the currently foreshadowed adoption of reorganised parts C, D, and E of the Income Tax Act 1994. The ambulatory nature of the rewrite process has some advantages from the point of view of drafters, but regular changes make the statute difficult for users to follow. Good, up to date, conversion tables can mitigate the problem to some extent.
2.186 The committee considers that it may also be useful if a schedule were to be added to the Act, that listed all the thresholds contained in various places in tax legislation. An example of the kinds of threshold that the committee has in mind is the threshold below which people qualify as cash basis holders under the qualified accruals rules. It would repeat information found in the tax Acts but its usefulness would lie in the collection of all thresholds in one place. If the information was out of step with the substantive provision for any reason, the latter would prevail.
Repairs and maintenance unit
2.187 The committee regards it as inevitable that despite the care taken by officials, and the close scrutiny by tax professionals and taxpayers, the rewritten parts C, D and E (and subsequently, other parts of the Act) will at some stage reveal textual uncertainties and produce unintended outcomes. The committee believes it would be appropriate for the Inland Revenue Department to establish a special 'repairs and maintenance' unit to address promptly any queries raised as to the effect on established principles of the rewritten Income Tax Act, and to deal with any unintended outcomes. Such a unit would provide an administrative mechanism to ensure both that the general body of taxpayers and tax advisers are informed of issues as they arise, and that remedial legislation is developed and introduced at first opportunity. The committee recommends the establishment of such a repairs and maintenance unit.
2.188 The formalistic design structure proposed for the rewritten Act requires that virtually all functional categories have their own part or subpart. As is apparent from this report, the committee favours the organisation of the Act on substantive lines rather than on formal lines. However, if the government determines to adhere to a formalistic organisational structure, the committee recommends that it should consider allocating a specific part or subpart to rules about valuation, which at present are not gathered together in the manner that is standard for the rewritten Act.
2.189 In Tax Accounting Issues and in its Final Report63, the Valabh committee addressed the issue of apportionment. Officials advise that they have not yet had the opportunity to address the Valabh committee's recommendations in depth. That lack of opportunity is regrettable, because it is desirable that the intended policy approach to apportionment issues in general should be settled before or at the time of the release of the rewritten parts C, D, and E of the Act. The reason is that apportionment questions figure largely in the matters that are covered by those parts.
2.190 Even without the benefit of a policy review, the question of apportionment raises a number of important issues that will have an impact on the rewriting of parts C, D, and E. The committee has discussed several of these issues with members of the rewrite team. It is not necessary for the committee to form a view as to the most appropriate approach. Officials are aware of the issues in this area.
Determinations under the qualified accruals rules
2.191 'The qualified accruals rules', as they are labelled in the Act, are more colloquially known as the 'financial arrangements' or 'FA' rules. They occupy subpart EH of the Act. Broadly speaking, their function is to prevent people accelerating expenses or deferring receipts in the context of loans and transactions that from an economic point of view may be partly or wholly equivalent to loans.
2.192 There is an infinite variety of transactions and business structures that have an economic similarity to loans. The qualified accruals rules do not attempt to address each of these possible forms of transactions individually. Instead, the rules adopt a single general principle: for tax purposes, income and deductions related to a financial arrangement must be spread across the duration of the arrangement according to the principle of yield to maturity, which is used in banking circles.
2.193 The yield to maturity principle applies readily enough to a simple loan that lasts for a defined period, but it is not immediately obvious how the principle should apply to more complex transactions. For example, how does a New Zealand borrower take account of exchange rate gains and losses in respect of a loan or of a credit sale where the transaction covers several tax years?
2.194 The solution adopted by the Act is for the Commissioner to issue rulings, in this context called 'determinations', that set out how the income and expenditure of identified kinds of financial arrangements may be calculated. For example, for transactions that involve foreign exchange, a determination may sanction one or more of a number of specified exchange rates as allowable for calculation purposes.
2.195 The idea of enacting a general principle in the statute (yield to maturity) and of concretising this general principle in a series of extra-statutory determinations was attractive when the system was established in 1986, and even now retains some attractions. However, it has never worked well. The committee notes that the drafting of determinations has proved a difficult task; the resulting products are often opaque and occasionally almost unintelligible. Most need to be rewritten. The committee's view is that this redrafting should take place as part of the rewrite of the Act itself.
2.196 The committee recommends that in parallel with the rewrite of the Income Tax Act, the department should redraft the existing qualified accruals rules determinations, in an endeavour to publish fresh drafts at the same time as the proposed exposure draft of part E of the Act is published. The committee further recommends that the procedure for issuing determinations should take on the basic features of the generic tax policy process. In particular, proposed new determinations should be made available for public consultation as to both substance and clarity. Where possible, each determination should follow one of a limited number of standard templates.
The annual taxing Act
2.197 New Zealand inherited from the United Kingdom the constitutional rule, developed in the seventeenth century, that the Crown does not have the right of its own prerogative to levy taxes. Tax can be levied only by Parliament. Historically, this rule has been emphasised in two ways. First, Parliament confers taxing rights on the Crown only on a year-by-year basis. Secondly, until relatively recently the annual taxing Act has been a separate, readily identified statute, typically called 'The Income Tax (Annual) Act', which concisely confers power on the Crown to levy taxes pursuant to the continuing rules of the Income Tax Act.
2.198 Under the Income Tax Act 1994 section OB 1, ''annual taxing Act' means the provisions of any Act by which the rates of income tax are determined for any year'. This definition was carried forward from section 2 of the 1976 Act, having been inserted in 1987. Before this amendment a separate Act was required each year to determine income tax rates. It is a sensible definition, in that it goes to the substance of the concept that is defined, and does not turn on a particular name for a particular statute. The definition means that the Crown's annual authority to tax can be conferred simply by including a provision about tax rates for the next twelve months to any tax amendment bill that is before the House, and that there does not need to be a separate bill for this purpose. The committee understands that this amendment was made to save on the additional House time that is required for a separate bill.
2.199 In substance, it is true that so long as Parliament annually passes an Act to empower the Crown to levy tax the name of the statute does not matter, nor does it matter whether the statute deals also with other things. However, the committee regrets that this change has occurred because a separate annual taxing Act reflects the important constitutional principle of parliamentary sovereignty in tax matters.
2.200 The relationship between the Crown and Parliament in respect of taxing powers goes to the heart of New Zealand's constitution. The committee, therefore, recommends that New Zealand should return to separate annual taxing Acts to symbolise this relationship. If there are concerns about the additional pressures on House time that a separate bill may entail, consideration should be given to including the annual taxing Act in a current part of the parliamentary timetable such as the Budget debate. This proposal may require an amendment to Standing Orders.
Relationship between Income Tax Act and rules of criminal law
2.201 There is a belief in some quarters that when taxpayers strictly comply with the requirements of the Income Tax Act, including anti-avoidance provisions, or believe that they have done so, they cannot be described as dishonest, and therefore cannot be guilty of fraud. This belief is mistaken. Strictly speaking, there is no need for a provision in the Income Tax Act to make the point. However, because the belief is so common, and because it leads people astray, the committee recommends that the rewrite should include a provision to state the law as to the relationship between the Income Tax Act and fraud, so that there can be no doubt.
2.202 The statement should include specific reference to the relationship between anti-avoidance provisions and criminal fraud. That is, whether a transaction is void by virtue of an anti-avoidance provision does not necessarily shed light on the question of whether the transaction is fraudulent.
Relationship between Income Tax Act and other legislation
2.203 A number of other statutes affect the operation of the Income Tax Act in one way or another. The Diplomatic Privileges and Immunities Act 1968 is an example. The committee considers that such legislation should be listed in a schedule. The objective is that the proposed schedule should not change the law, but that it should act as a useful tool for users of the Act, along similar lines to schedule 23, which contains conversion tables that compare the numbering of the 1994 Act with the numbering of the 1976 Act. The schedule should begin with a statement to the effect that the omission of any legislation does not mean that the omitted Act does not affect the operation of either that Act or the Income Tax Act.
2.204 An additional schedule could list sections of the Crimes Act 1961 that are potentially relevant to income tax fraud. Again, this schedule would not purport to alter the law. Rather, it would bring home to users of the Income Tax Act that the Act is not an isolated edifice that has no relationship to the rest of New Zealand law. The schedule would help people to realise that the criminal law can be as relevant to the income tax area as it can to other areas of economic life. The schedule should begin with a statement to the effect that the rules that it contains are not an exhaustive list of the forms of civil and criminal liability that may arise in connection with tax matters.
2.205 The committee recommends that schedules that list legislation which affects the operation of the Act, and relevant sections of the Crimes Act 1961 be added to the Income Tax Act 1994.
26Cape Brandy Syndicate v IRC  1 KB 64 at 71, Rowlatt J, discussed above in para 2.12.
27 The Income Tax Act 1994 uses the term 'any amount'.
28 See footnote 28.
29  NZLR 474, Wilson J
30 Consultative Committee on the Taxation of Income from Capital, The Core Provisions of the Income Tax Act 1976, Discussion Paper, September 1990, para 1.3.
31See paras 2.36 to 2.41
32  3 NZLR 439 at 444
33 2 NZLR 513 at 549
34 NZLR 147, Barrowclough CJ, discussed above, para 2.12
35  3 NZLR 439 CA, discussed above, para 2.18
26Graham v CIR  NZLR 994, McCarthy J
37 Golightly v CIR (1972) 1 TRNZ 135, Speight J; Grieve v CIR (1983) 6 TRNZ 461 (CA)
38 Consultative Committee on the Taxation of Income from Capital, The Core Provisions of the Income Tax Act 1976, Discussion Paper, September 1990, para 2.2
39 Consultative Committee on the Taxation of Income from Capital, Final Report, October 1992, page 16
40 Working Party on the Reorganisation of the Income Tax Act 1976, Second Report, Discussion Paper, September 1993, pages 41-45
41 Section CD 3, Income Tax Act 1994
42 (1980) 4 TRNZ 1, Holland J
43(1994) 16 NZTC 11,064, McGechan J
44Consultative Committee on the Taxation of Income from Capital, The Core Provisions of the Income Tax Act 1976, Discussion Paper, September 1990, para 1.3
45Working Party on the Reorganisation of the Income Tax Act 1976, Second Report, Discussion Paper, September 1993, pages 14 -18
46 See para 2.36
47 For an outline of the generic tax policy process see appendix 2
48 Working Party on the Reorganisation of the Income Tax Act 1976, Second Report, Discussion Paper, September 1993, page 4; Parliamentary Counsel Office, Drafting Manual (Draft of 16 May 1997) 5.9; NZLC R35, Legislation Manual Structure and Style, May 1996, para 144 Back
49 (l) is the alphabetical letter that follows the letter k, and not the arabic numeral 1.
50 (1935) 35 SR (NSW) 215 at 219
51 (1982) 5 TRNZ 754; 5 NZTC 61,259 52 (1913) 6 TC 163 at 165 HL
53  AC 455 at 458 HL
54(1928) 41 CLR 148 at 151 Back
55Contrast, again, judicial statements on lawyers' law, which should be respected as likely to be of intrinsic merit.
56Note the discussion earlier in this report at para 2.160
57 The Taxation (Simplification and other Remedial Matters) Bill; the Taxation (Tax Credits, Trading Stock, and other Remedial Matters) Bill
58The Taxation (Accrual Rules and other Remedial Matters) Bill
59Working Party on the Reorganisation of the Income Tax Act 1976, Second Report, Discussion Paper, September 1993, page 10
60 See footnote 59
61 See paras 2.111 to 2.114 of this report.
62 Tax Law Rewrite (UK) Responses to the Second Technical Discussion Document - A Purposive Approach to Rewriting Tax Legislation, August 1998, http://www.open.gov.uk/inrev/rewrite.htm
63 Consultative Committee on the Taxation of Income from Capital, Final Report, October 1992, page16