DISCLAIMER: This article has been scanned from a printed source. It has been proof-read but may still contain errors or inconsistencies. Please refer to a printed version for complete accuracy when quoting from this document.


In December 1989 the Minister of Inland Revenue released a discussion paper containing proposals to reform various sections of the Income Tax Act 1976 (the Act). Among its suggestions was the possible reintroduction of land into the definition of ”trading stock” as currently stipulated by section 85 of the Act. Such a step was seen as integral to the Inland Revenue's broader objective of producing a more coherent scheme to calculate ”income” for taxation purposes. A consultative committee, chaired by Arthur Valabh, was appointed to consider the proposed reforms. [1] The Committee concluded in favour of treating land as trading stock, though subject to constraints on the method of valuation.

The aim of the present article is not to criticise that conclusion. There is certainly need for reform. Without a trading stock provision embracing land, the structure of the Act would ostensibly appear to permit a full deduction of expenditure at the time the land is acquired, with the proceeds of sale being assessable only upon disposal. In this article I outline the present nature of the Act, in order to highlight the effect upon land transactions of the decoupling of assessable income provisions from those permitting deductions. I proceed to note the suggested reforms, bearing in mind the possibility of a unified, coherent scheme. I then analyse a rather startling development in the case law that seems to have achieved the result proposed without statutory reform. I address the appropriateness of that development, particularly in the light of the difficulty that it purports to overcome, and offer certain conclusions.

II. The Scheme of the Act, and the Need for a Trading Stock Provision

Early New Zealand legislation, in the form of the Land and Income Assessment Acts of 1900 and 1908, was predicated on the British ”net/schedular system” where net income is taxed on a source by source basis. Under the British scheme, and corresponding to ordinary principles of accountancy and commercial practice, any revenue derived from a particular source was set off against the expenditure needed to raise that revenue in the first place. The net profit or gain from that source would then be subject to tax at a rate determined by the particular schedule applying to that source.

In 1916, the Land and Income Tax Act altered the structure of our taxation system by introducing a statutory test for deductibility. The 1916 Act and its successors prohibited any deduction for expenditure or loss unless expressly authorised by statute. [2] Within the present Act the separate deductibility provision is mandated by section 101, which prima facie denies deductibility for outgoings save where those outgoings satisfy the statutory test (enunciated primarily in sections 104 and 106).

The impact has been considerable. In CIR v Farmers' Trading Co Ltd ,[3] Richardson J stated the opinion of the Court of Appeal that the general deductibility provision does not now import the application of accounting principles and commercial practices. [4] As a result of legislative intervention, the calculation of assessable income in New Zealand has assumed a ”gross” character. This means that the gross revenue figure is independently aggregated from the various sources ( per the assessing provision, section 65) and then offset against the total of only those expenditures which have been statutorily approved. Correspondingly, perhaps the most important effect of New Zealand's statutory structure is the loss of the matching principle - the principle that no deductions be taken without reference to a corresponding head of income assessment in any income year - which is inherent in a net approach to the calculation of assessable income. A useful statement of the significance of the matching principle is noted by the Valabh Committee:

[A] true reflex of a person's income over a year is achieved only if expenses incurred in a year are matched with the revenue they produced so that such expenditure is dealt with in the period to which they relate. This recognises that expenditure can clearly relate to a different period from that when the expenditure is incurred. [5]

In its stead, the Act ( per section 104) explicitly provides for the deduction of expenditure or loss in the year that it is incurred , provided that it is incurred in producing assessable income for any income year. This provision apparently removes any requirement for revenue and expenditure to be matched in the calculation of assessable income.

1. The Treatment of Trading Stock in General

Section 85 defines and provides for the valuation of trading stock. It requires that trading stock on hand must be accounted for in the assessment of income. In effect, the section substitutes an artificial matching regime which in specific cases ameliorates the weakness of the gross approach to tax accounting. It does this by providing in section 85(6) for the value of trading stock at the close of the income year to be included in the taxpayer's assessable income for that year, while permitting (in section 85(7)) the value of trading stock at the beginning of the income year to be deducted from assessable income for that year. Thus the section counterbalances the deduction of expenditure upon acquisition, allowed under section 104, and so prevents a cashflow advantage being derived from premature deduction. The regime operates independently of the core assessment and deduction provisions: section 85 makes no specific provision for the treatment of either sale proceeds or purchase expenditures in respect of trading stocks. It is presumed that by implication these items fall to be dealt with under sections 65(2)(a) and 104 respectively - a proposition impliedly accepted by the Court of Appeal in Farmers'.[6]

2. Trading Stock and Land

Generally speaking, revenue from transactions involving land is assessable if the transactions fall within the ambit either of section 65(2)(a) - being transactions in the course of a business of dealing in land - or of section 67(4) - being certain specified (and otherwise capital [7]) transactions involving land, including cases where the land is acquired with the purpose or intention of resale.

Notwithstanding that revenue therefrom may be assessable income, land is expressly excepted from the section 85(1) definition of trading stock. [8] It is thus not immediately clear how the taxation of land transactions is integrated with the core assessment provisions in section 65 - in particular, whether the provisions in section 67 which capture revenue from certain land transactions are linked to the assessment of revenue from other sources by a gross revenue calculation, or are subject to a separate, schedular net calculation. The lack of clarity arises because section 65(2)(f) explicitly catalogues amongst assessable forms of income any profits or gains from the sale of land which fall within the ambit of section 67(4). By contrast, section 104 does not make any such reference to the specific regime in section 67 (for either net losses or gross deductions).

Prima facie , given the present relationship between the core provisions and the specific regime affecting land, expenditure is seemingly deductible once it is incurred in the production of assessable income without reference to the provision under which income arising from the expense is assessed. Without the saving provisions of section 85, this means that the outgoings in relation to that land may be deductible prior to the assessment of any profits or gains which arise from that land transaction. Indeed, they may be deductible whether or not any revenues arise. In the context of land, there appears to be no necessary match between expenditure and revenue.


In order to avoid the persistent mismatch of deductions against assessable income, the Inland Revenue Department has in practice adopted a net/schedular approach to the calculation of assessable income wherever the Act is silent as to the timing of expense recognition, and in particular with respect to section 67(4). The Department's justification for doing so is that the language of section 67(4) indicates that only net profits or gains arising from those transactions are subject to taxation. The practical outcome of the net approach is that nothing is deductible or assessable prior to the calculation of the net profit or gain, where the time for calculating net profit or gain is at the moment of disposal. I shall return later to consider the case law, but it is worth noting here that, although prima facie inconsistent with the approach stated in Farmers', the Revenue Department's analysis is compatible with ( inter alia ) the view taken in Lowe v CIR ,[9] supporting the suspension of revenues and expenditures pending the disposal of an asset; though Richardson J did not there explore the point, as counsel had concurred that a net approach was appropriate.

The stance that section 67(4) should be subject to a net/schedular approach is adopted also by John Prebble, who argues that, in cases where the Act taxes profits or gains that are not ordinary income, standard principles of deductibility do not apply:

In ”capital” cases these principles cannot apply. The correct approach must be to wait till the land in question is sold and to calculate the profits at that point, taking into account then all deductible items from the year of sale and previous years. Logic and practicality dictate this conclusion and it seems to have been accepted without discussion in relevant reported cases.

The taxpayer must wait until he disposes of the land before he can claim the deductions because until disposal, the land has not been committed to an income producing activity. It is the disposal in circumstances falling within one of the paragraphs of s 67(4) that triggers the tax liability. There is no preliminary action by the taxpayer that renders the profits taxable. Consequently, it is impossible to say that any particular cost, when incurred, relates to assessable profits from the future sale of the land. [10]

The Valabh Committee, on the other hand, has proposed to resolve the question of accounting for assets such as land that is purchased for resale by legislating for a separate ”matching” provision within the Act. [11] Such a provision would apply to what might be termed ”revenue” assets, being assets whose proceeds from sale - unlike capital assets - are fully assessable (the provision would except certain financial arrangements and trading stock already covered by section 85). The nominated section 85D diverges from the ordinary trading stock regime in that it expressly limits the valuation of assets covered to their historical cost. Section 85D also differs from section 85 in that the assessment of sale proceeds and deduction of acquisition costs would also be regulated by that section.

It is important to recognise that, while the practical upshot of the new provision is a mechanism in some ways similar to a net regime, formally what is suggested is better termed a ”revenue” approach to the calculation of assessable income arising from land transactions. It is this alternative that Richardson J was prevented from considering in Lowe:

Because of that concession [that the net method was appropriate], and in the absence of any argument on the point, I shall not explore the alternative view that in such a case all assets engaged are held on revenue account with the deduction provisions applying in the ordinary way to the outlays all of which are on revenue account and that until a sale occurs the land involved stands in the books at cost for tax purposes, thus matching the outlays on the acquisition, holding and development of the land in the revenue account as at the particular balance date. [12]

On the revenue approach posited by section 85D, the acquisition cost of land would be deductible ”in the ordinary way” under section 104; but would be offset by the closing stock figure, attributable to assessable income under the proposed section. By contrast, under the net method no entry is made in the books for tax purposes until revenue is derived from disposal.

Such patchwork reform might perhaps be criticised on the ground that no attempt is made to achieve a uniform pattern of assessment throughout the Act. And indeed, there is ostensible force in the submission that - absent uniformity - the potential for non-neutrality between sources of income is increased, especially as to the timing and scope of deductible expenses; while at the same time the administrative simplicity and efficiency of the Act are diminished. On this matter, however, the Committee was unpersuaded that specific regimes ought to be placed on a single net or gross basis throughout the Act. [13] In the Committee's view, the courts have evinced an ability to achieve sensible results on the question of linkages between core provisions and the net or gross approaches of specific regimes. In its report the Committee endorsed the views expressed by Gallen J in Gunson:

While there is some strength in Mr Patterson's contention, I think it proceeds on a basic fallacy, that being the suggestion that the Income Tax Act is to be regarded as a coherent whole based upon a consistent pattern, the logic of which illuminates every part and that the validity of separate aspects of the taxing provision depends upon its coherency and consistency with the general pattern. The fact of the matter is that taxing legislation is invariably pragmatic. While it may start from a particular approach or indeed evidence a particular philosophy, taxation legislation is used as a vehicle for a number of aspects of Government policy. As particular ends are sought to be met, particular measures are engrafted on to what already exists. It is hardly surprising if the very complexity of the legislation leads to situations which reflect the differing approaches adopted to meet particular ends. I accept that the general pattern of the present statute may reveal a global approach rather than a schedular approach, but that is not to say that Parliament will not for specific purposes deal with particular matters by other approaches. It might have been possible in this case for the end which sec 188A seeks to achieve to have been done by defining and limiting deductions, but Parliament has chosen to separate out a particular activity defined as a specific activity and to create in effect a code which deals with the assessment of income where such an activity is part of the business operations of the taxpayer. [14]

Decisively, perhaps, uniformity is in any event an unrealistic ideal. There are already particular regimes which deviate from the gross approach: section 65(2)(eb), for example, relating to foreign investment fund income, is calculated on a net basis and accounted for separately under its own schedule. More recently Richardson J remarked, in Hill v CIR , that:

Tax legislation is not a seamless web. It is not totally consistent in its terminology and reflects different approaches taken in different amendments over the years. Even within a section it is not wise to expect an all pervading logic and symmetry. [15]

Given that taxation systems are driven by policy objectives which constantly mutate, it is perhaps as well that no attempt is being made to achieve uniform linkages between the core provisions and the various specific regimes. Even were it possible to achieve uniformity within the current Act, one may be certain that over time such perfection would once again be lost.

Most recently, however, the High Court has apparently achieved the same result without need for legislative amendment to the Act - and indeed, without the need for a piecemeal approach to distinct regimes such as that contained in section 67(4). It is to this development that I now turn.

IV. Murray Darnill Ltd v Taxation Review Authority [16]

In March of this year, the High Court heard an application for judicial review regarding the taxation accounting for land bought for the purpose of resale. The plaintiffs were two private companies constituting a specified group of companies for the purposes of section 191 of the Act. The first plaintiff purchased a piece of land in Queenstown with the purpose of reselling it. The land was sold during the 1990 tax year at a loss, and the first plaintiff claimed a deduction for the whole cost of the land in the year of acquisition. The amount claimed as a deduction left the first plaintiff with an alleged tax loss for the year ended 31 March 1987 - since, although the company claimed the expense of purchasing the land as a deduction, it did not credit the land as trading stock in its accounts for the year ended 31 March 1987. The first plaintiff claimed to set off its loss in 1987 against a profit made by the second plaintiff. The Commissioner disallowed the first plaintiff's claim for a deduction of the purchase price and related expenses, and reduced the loss available to be transferred to the second plaintiff under the specified group rules. The plaintiffs' objections were disallowed and a case was stated to the Taxation Review Authority. The Authority upheld the Commissioner's approach, determining that it had not been shown to be incorrect. [17] The plaintiffs sought to challenge the Authority's decision by means of judicial review. [18] While finding that judicial review was not the appropriate guise under which to appeal a substantive decision of the Taxation Review Authority, the Court nevertheless determined the appeal also on the question of substance.

In the High Court, counsel for the plaintiffs (as might be expected) utilised a gross approach to tax accounting in order to justify a preliminary deduction under section 104. Counsel conceded that which Tipping J expressed succinctly:

It is well settled that for income tax purposes the computation of profits/gains and losses must proceed on ordinary commercial principles unless some specific provisions of the statute require a departure from such principles. [19]

Nevertheless counsel sought to contend that ordinary principles (and a net approach) were impliedly excluded in respect of land by section 85; and that therefore, while the purchase price might be deductible under section 104, the land need not be recorded as closing stock in the calculation of assessable income. Tipping J, however, rejected such an analysis:

In my view the fact that s 85 is a self contained code for the matters it covers does not mean that a similar approach should not be taken to land in circumstances when it is properly classified as trading stock. [20]

This is not yet to say that land must be accounted for in the manner of other closing stock; merely that the plaintiffs had failed to convince that the exclusion of land from section 85 mandates that it must not be included in trading stock figures when calculating assessable income. What must now be considered is whether Tipping J was correct then to find that land should be so accounted for.

I begin with Tipping J's own analysis. His Honour concluded first that land captured by the terms of section 67(4) is a revenue, rather than capital, asset; [21] as such (and by contrast with capital assets), the proceeds of its sale are fully assessable for income tax. This proposition, though once debatable, may now be regarded as relatively uncontroversial following the Court of Appeal's decision in CIR v Inglis ;[22] it is outside the aims of the present article to test the finding. [23]

His second proposition is more questionable. Tipping J's conclusion, that a (closing stock) figure equal to the purchase price of the land should be included in the first plaintiff's assessable income in 1987, depends upon his finding that trading stocks of land subject to the regime in section 67(4) are to be recognised in the taxpayer's assessable income; notwithstanding the non-applicability of section 85. His Honour held that the trading stock regime applies, in effect, to all revenue assets (and, given the first proposition, therefore to land under section 67(4)). The ground for so holding was expressed to be that the inclusion of closing and opening stock figures are integral to any correct or even realistic evaluation of assessable income:

To omit year end stock-in-trade in the present case would be to give a false result. It would suggest that the company's initial outlay was no longer represented by a corresponding trading asset. I do not regard the self contained provisions of s 85 as requiring such a commercially unreal, indeed false, reflection of the taxpayer's revenue account. [24]

His Honour buttressed the reasoning with a second point. Counsel for the plaintiffs had objected that closing stock figures could only be included in assessable income if opening stock figures were deductible. Yet section 101, which establishes a code for deductions, prohibits deductions unless expressly allowed - and while section 85(7) permits deductions for other types of opening stock, there is no provision permitting deductions for opening stocks of land. This argument was rejected, according to Tipping J, because:

In so far as opening trading stock must be deducted from closing trading stock to compute the profit or loss for the year, I do not consider that this is the sort of deduction at which s 101 is directed. It is not in that sense a deduction. It is simply part of the method of determining the profit or loss gained by a taxpayer from transactions involving trading stock. This point can be put in another way. In accountancy and assessment terms the same result is achieved by avoiding the concept of deduction altogether. The question can equally be framed in this way: by how much does closing stock exceed opening stock or vice versa? The result is your profit or loss. [25]

An interposition is warranted. With respect to his Honour, this second point cannot be regarded as a strong one. Were it valid, those who drafted section 85(7) would no doubt be astonished, for that section would consequently be unnecessary. And there are other difficulties with Tipping J's interpretation. If the gain or loss in trading stock were nothing but a net figure to which section 101 were inapplicable, then so also would be the net profit made on sales of stock bought during the year. There are two ways in which the sale of some item can be sourced: from purchase of the relevant item during the year, or from opening stock. If depletion of the latter resource is not to be treated as a deduction, then nor presumably is the former. This is surely an absurd result. Profit, it is often said, is a net concept; [26] costs incurred in deriving income are deducted from revenues before arriving at the assessable income. But they are not deducted without reference to section 101.

Further, suppose that in a given year a taxpayer's opening stock were to exceed its closing stock. The net diminution, like any other loss, could only constitute a deduction from assessable income. Yet, if not under the statutory deductions regime (or section 85(7)), how could this be achieved? Certainly not by means of section 65, which refers only to ”profits or gains”.

Thus we are left, in essence, with only one argument: that it is commercially realistic to regard the land purchased by the plaintiffs as trading stock. Qua net profit in accounting terms, this may well be so; but is such a treatment consistent with the structure of the Act?

A most important point to notice in this debate is that there is no commercial unreality in the long run . The revenue from disposal of the land will enter the accounts, in the normal way, in the year of disposal. What is at issue here is solely whether there is a timing mismatch, such that the deduction for costs of acquisition may enter the accounts earlier to advantage the taxpayer. Were a net approach taken to transactions captured by section 67(4), the costs of acquisition would correspondingly only be recorded for tax purposes in that year of subsequent disposal. Tipping J, however, found that a net approach to that section may not be adopted. One may doubt that finding, and it may safely be said that the cases on this point are far from unanimous in their support for his Honour. But let us for now defer the question of a net approach, and ask Tipping J's own question: should a ”revenue” approach be taken instead? According to this method, which was eventually the one adopted in Murray Darnill , deductions for acquisition costs are available with respect to revenue assets in the year incurred (akin to the gross, rather than net, view); but the assets acquired thereby are themselves to be recorded in the accounts as trading stock until disposed of - thereby ensuring that, in effect, the benefit of tax deductibility for costs of acquisition is deferred until the year of disposition. By way of a practical illustration, the land in Murray Darnill is be accounted for in the following manner if it be regarded as trading stock:

1986 (i) Deduct cost of purchase, by virtue of section 104;

(ii) Include closing value in assessable income.

1990 (i) Include sale proceeds in assessable income under sections 65 and 67;

(ii) Deduct opening value of trading stock.

The features of the revenue approach are, thus, that the trading stock device now universally overcomes the loss of matching when the net approach is abandoned. Further, while section 104 (permitting deductions for costs in the year incurred) is applicable only to revenue assets, it is now also applicable whenever the asset is revenue rather than capital in nature, where the finding that the proceeds of some asset constitute assessable income is taken to imply that the asset is a revenue rather than capital asset. Finally, not only are all items of trading stock also revenue (rather than capital) assets, but now all revenue assets are also, in Tipping J's deeming, trading stock.


With respect to Tipping J, his analysis is open to a number of objections. First, the result appears to render section 85 in important respects redundant. Second, it ascribes too much importance to commercial principles in the ascertainment of assessable income. Third, it is inconsistent with extant case law. These criticisms will be elaborated below.

1. A Gratuitous Section?

Opening stock is deductible because of section 85(7). Had Parliament intended all revenue assets to be accounted for as trading stock, surely it would not have enacted in section 85 a regime that expressly excludes land? Tipping J's explanation of this point is that the Legislature by omitting land sought to give ”a clear indication, on the valuation front, that land must be carried forward at cost”. [27] That is, Parliament means land to be treated as trading stock but valued only at cost rather than according to the various valuation options available under section 85. Yet how clear an indication of that design do we have, when the Legislature - rather than expressly excluding land altogether - could so very easily have specified a distinct method of valuation for that form of asset?

Moreover, it should be noticed that section 85 was expressly enacted as a device required to overcome the general difficulty of lack of matching, created by the gross structure of the Act. Under Tipping J's revenue analysis, there is of course no problem at all - and the provisions of section 85 are largely otiose. It would surprise if the Legislature's intention had been to create such a redundant regime.

A further point may be made. There are, as has been mentioned, no explicit provisions by which trading stocks of land may be included in and deducted from assessable income. Tipping J's analysis, hence, relies on an extra-statutory concept of trading stock - a general concept based upon ordinary commercial understandings and which can be accounted for independently of the specific regime.

Logical space exists for such a concept. The definition of trading stock in section 85 is given only for the purpose of that section; [28] there is, expressly, no attempt to define ”trading stock” for the Act as a whole. But the ordinary principles vaunted by his Honour conspire against the particular facts of Murray Darnill . The case law indicates that for property to be ordinarily regarded as trading stock, it must be the asset of a taxpayer who is in the business of trading in that type of property. [29] This being so, then on the facts of the present case it is questionable whether the land in question would fall within that ordinary meaning. While Tipping J termed the transaction a business activity, because the plaintiff had purchased the land for resale, [30] he did not find that the plaintiffs (who were primarily grocers) were in the business of dealing in land. Indeed, the report of the case discloses only one transaction concerning land on the part of the taxpayer. His Honour cited FC of T v St Hubert's Island Pty Ltd (In Liq) [31] in support of the proposition that land may in appropriate circumstances be classified as trading stock, but missed the constraint upon which circumstances count as appropriate. Mason J had averred that land can be classified as trading stock where it conforms to the ordinary meaning of that concept; that is, where it is part of a business of dealing in land.

Though he did not take the point overtly, Tipping J did have some authority on his side. In Inglis, Cooke P at one stage remarked:

In a case such as the present, the taxpayer by buying shares for resale has in substance stamped their cost as circulating capital and the shares themselves as stock in trade. They have become held on revenue account. [32]

It seems unlikely, however, that by this the President intended to herald a new approach to trading stock in the calculation of assessable income, particularly since the categorisation of shares as stock in trade is - unlike land - not precluded by section 85. What he sought to achieve was a parity within the Act: where the profit from shares which have been acquired for resale is assessable, he found that the statute characterises expenditure upon such shares as being upon a revenue rather than capital asset (notwithstanding that there was no connection with a business or trade by the taxpayer), which can therefore be expensed bypassing the prohibitions contained in section 106(1)(a). His depiction of the shares as trading stock was incidental. Indeed, Cooke P expressly disclaimed a trading stock analysis as the basis of his decision, [33] grounding it solely in the characterising of shares as revenue (more specifically, ”circulating capital”) rather than (”fixed”) capital in nature. Nor did McKay J in his judgment characterise the shares as trading stock. [34] It is clear, moreover, from a reading of the case that Cooke P did not contemplate anything other than a net approach to the tax accounting for the shares, according to which ”the cost price of the shares would prima facie be deductible under section 104 in calculating the assessable income in the year of the loss ”.[35]

2. Just How Strong is the Commercial Realities Point?

A further objection is that the force of the ordinary commercial principles point is in fact rather less overwhelming than Tipping J surmised. If ordinary commercial principles justify recording land as trading stock, why must that recording be done at historic cost only? Why, where ordinary commercial realities are thereby reflected, cannot devaluations or revaluations be recorded in the closing stock accounts prior to disposal? That this is sometimes realistic (at least where the taxpayer deals in land) is acknowledged in St Hubert's Island .[36] Had the Legislature meant that option to have been foreclosed, it could - very easily - have done so expressly: that it did not is an indication that there was no need. The delimited importance of ordinary principles was acknowledged by Cooke J (as he then was) in Lowe:

As to ordinary methods, it is a well-settled principle of income tax law that, in the words of 23 Halsbury Laws of England (4th ed) para 258, ”The profits are to be arrived at on ordinary commercial principles, subject to such provisions of the Income Tax Acts as require a departure from such ordinary principles, for example the prohibition of certain deductions”. [37]

The deductions regime is a statutory creature that, as was noted in Farmers',[38] no longer affords room for the intrusion of accounting and commercial practice. If section 104 requires acquisition costs to be deducted in the year incurred, then ordinary principles must yield. In any event, the consistent appropriateness of such principles may be doubted. Richardson J has pointed out that the aims of commercial and taxation principles frequently diverge:

An income tax system is concerned only with the measurement of income whereas accounting principles and practice are directed towards producing financial statements of a business which fairly present the financial position at a point in time as well as the results of operations for the accounting period ending at that time .... Neither perfection nor absolute precision is required. It is sufficient and necessary that the accounting method be best calculated to reflect the income which is subject to tax. [39]

Accounting practice is no more that a subservient tool to assist in the determination of assessable income for the purposes of the Act. To the extent that there is a disparity between assessable and accounting or reported income, accounting principles do not override the Act; for they are designed to arrive at the latter rather than the former statistic.

3. Abandoning the Gross Method

In finding that the revenue approach is more appropriate than the net method, and that ”[w]hichever of these two methods is regarded as preferable, the method adopted by the taxpayers in the present case cannot be correct”, [40] Tipping J expressly relied upon the authority of Inglis and Hill. Most importantly of all, however, Murray Darnill appears in fact to be inconsistent with both Inglis and Hill, as well as with Farmers'.

Inglis has already been discussed: the case is clearly authority for use of a net method, as it was taken for granted by the Court of Appeal that the taxpayer's share acquisition costs [41] may only be brought to account in the year of disposal. The conclusion in Inglis that shares purchased for resale constitute a revenue rather than capital asset does not mandate that they should be accounted for under a revenue rather than gross approach.

In Hill, the question was not taken for granted but raised expressly in the opening words of the judgment by the Court of Appeal:

The short point arising on this appeal is whether certain revenue and capital expenditure and depreciation allowances in relation to the forestry business of the appellant are deductible in the years in which they were incurred or whether deductibility is deferred to the year or years in which income is derived from the sale of timber. [42]

While it is true that Richardson J acknowledged in Hill that ”a profit is a net concept”, [43] ”net” in this sense means only ”after permitted deductions (especially, those incurred in that year)” - and not ”net” in contrast to gross or revenue. The terms of section 104 are plain. Revenue (rather than capital) expenditures are to be deducted in the year incurred, irrespective of the year in which the corresponding revenues arise. Hill overtly grants that truth, and that the requirement for matching expenses with the revenues to which they relate has been abrogated:

[I]t is readily understandable that Parliament should distinguish between revenue and other specific deductions allowable in the years in which they are incurred and a set off of the capital cost of timber as and when it is sold ... [Section 104] confines deductibility to the year in which the expenditure is incurred. [44]

According to Tipping J, Hill points ”in the direction of the revenue account method rather than the net method”. [45] This is simply erroneous. Hill points directly to the gross method.

Much might be made of remarks that profit is a net figure, and that income is to be ascertained using ordinary accounting principles. Yet it is clear that what is meant by this is only that assessable income is to be taxed net of such statutory deductions as are permitted. Since deductions are not to be assessed using ordinary principles, it follows that income in the sense of profit or gain is not to be ascertained from such ordinary principles; though the gross revenues may be. [46] Reference here need only be made to Richardson J's statement in Farmers', where his Honour stated, with admirable clarity and accuracy:

Notwithstanding the language of s 88(1) [now section 65(2)(a)] what is brought to charge is income in the sense of revenue as ascertained according to accounting and commercial practice (but subject to the guidance provided for in the legislation) less only those deductions allowed for by the statute. [47]

(emphasis added)

By recoupling the assessability and deduction regimes, under the rubrics of commercial reality and an ordinary concept of trading stock, Tipping J does not so much follow the law as revolutionise it.


There was an alternative route available by which to find against the plaintiffs. It is respectfully submitted that a less disruptive analysis - notwithstanding its rejection by Tipping J - is that, even if the structure of the Act as a whole leads to a gross approach to calculation of assessable income, section 67(4) creates a distinct, net/schedular regime. Consequently, as there was neither a profit nor a loss in 1987, the transaction should not be permitted to affect assessable income in that year; and indeed, for taxable income purposes there is nothing to be accounted for until the land is disposed of. Such a view accords both with the practice of the Commissioner and with the remarks, cited earlier, of Professor Prebble. Moreover, it is consistent with at least some of the cases, and in particular with Inglis and Lowe.

But it should immediately be noticed that such a result is no more than a lesser interpretive evil. The analysis does not, at least, sacrifice the structure of the Act on the altar of thematic unity. And there is much to be said for leaving nice questions in the taxation realm to the discretion of the judiciary, expecting the result not to be a ”seamless web”. Yet, in the present instance, a net method would once again be inconsistent with the terms of the Act. Section 67(4) does not provide for the deduction of net losses, only for the assessment of gains. Section 104 is the medium by which such losses would have to be deducted; and that section - if it applies at all - permits the acquisition expenditure to be deducted without awaiting disposal.

In upshot, Murray Darnill arrives at a result that might be regarded as desirable. The decision certainly accords with the reforms suggested by the Valabh Committee, particularly in relation both to the valuation of land held as trading stock, and to the linkage between this specific regime and the core provisions. Nevertheless, one would hope that the proposed section 85D does enter the Act, as the matter would than be dealt with somewhat more happily. Certainly, when we compare the difficulties posed by Tipping J's judgment with the simplification that a well-drafted ”matching provision” might achieve, there is every reason to advocate statutory reform; and that as soon as possible.

What surprises about the reasoning in Murray Darnill is its one-sidedness. Tipping J is prepared to construe extensively so as to find against the plaintiff, in a field of law traditionally interpreted such that the charge to tax must be clear before the taxpayer is required to meet it. The Act sets out, by establishing a distinct regime for deductions, to eschew Tipping J's fabled commercial reality - contra the (presumably, legitimate commercial) interests of taxpayers. Wherefore, when occasionally the Legislature's artificial intrusion backfires, ought a judge to work so hard to assist the Crown? ”There is no equity about a tax”: [48] all the more so when the Legislature has in the first place created a potential mismatch by legislating so as to deny the taxpayer the advantage of ordinary principles of deductibility.

[*] LLB (Hons) (Auckland), MTax (London), Lecturer in Law, University of Waikato.

[1] The ”Valabh Committee” produced five reports in all, including two of direct relevance to the discussion at hand: Discussion Paper (1990) and Tax Accounting Issues (1991).

[2] Cf s 86(1)(a); also s 110 of the Land and Income Tax Act 1954.

[3] (1982) NZTC 61, 200 (hereinafter Farmers').

[4] At 61, 207.

[5] Discussion Paper (1990) 31; taken from the Statement of Standard Accounting Practice No. 11 , issued by the New Zealand Society of Accountants (December 1979).

[6] At 61, 208.

[7] Cf Lowe v CIR (1981) 5 NZTC 61, 017, 61, 019 (hereinafter Lowe).

[8] Land is included in two other definitions of ”trading stock”, enunciated by ss 90 and 91: these sections deal with the valuation of trading stock land respectively where sold as an undivided part of a business and where sold at an undervalue. The Act stipulates that the definition of ”trading stock” in each of these sections is solely for the purpose of that particular section.

[9] (1981) 5 NZTC 61, 006, 61, 024.

[10] Prebble, J The Taxation of Property Transactions (1986) 136.

[11] Tax Accounting Issues (1991) 83.

[12] Supra note 9, at 61, 024.

[13] Discussion Paper (1990) 30.

[14] (1987) 9 NZTC 6, 167, 6, 172.

[15] (1994) 16 NZTC 11, 037, 11, 045 (hereinafter Hill).

[16] (1994) 16 NZTC 11, 126.

[17] Reported as Case M 111 (1990) 12 NZTC 2, 712.

[18] Although the plaintiffs had intended appealing direct to the High Court, they failed to observe the time limits prescribed in s 43 of the Inland Revenue Department Act 1974 and lost their right of appeal. Hence they followed the alternative route.

[19] At 11,131; citing the judgment of Cooke J (as he then was) in Lowe v CIR [1981] 1 NZLR 326, 334; (1981) 5 NZTC 61, 006, 61,014; also 23 Halsbury's Laws of England (4th ed) para 258.

[20] At 11, 132.

[21] At 11, 132.

[22] (1992) 14 NZTC 9,180 (hereinafter Inglis).

[23] See, however, J McDermot's critique in Circulating Capital: A Judicial Label (LLM Research Paper, Victoria University of Wellington, 1993).

[24] At 11, 132-11, 133.

[25] At 11, 133.

[26] Cf Farmers', at 61, 207; Hill, at 11, 045; CIR v Grover [1987] 2 NZLR 736, 740.

[27] At 11, 134.

[28] A similar limitation applies in ss 90 and 91.

[29] See, eg, FC of T v St Hubert's Island Pty Ltd (in liq) (1978) 8 ATR 452; Investment and Merchant Finance Corp Pty Ltd v FC of T (1971) 125 CLR 249; 2 ATR 361; Craddock v Zevo Finance Co Ltd (1946) 27 TC 267 (HL); FC of T v Suttons Motors (Chullora) Wholesale Pty Ltd (1985) 157 CLR 277; 16 ATR 567; AGC (Investments) Pty Ltd v FC of T [1991] ATC 4180.

[30] At 11, 128; see also at 11, 132.

[31] (1978) 8 ATR 452.

[32] At 9, 186.

[33] At 9, 186.

[34] Hardie Boys J concurred with both, and did not offer independent enlightenment on this point.

[35] At 9, 185 (emphasis added). Cooke P was here summarising the Commissioner's view as presented in argument. Although he did not accept other aspects of the Commissioner's submission, he raised no objection to the net approach taken either in argument or in the Department's leaflet IR276, Taxation of Proftis from Selling Shares (1989), where it is stated that ”Shares acquired for resale but not sold at the end of the year are not taken into account, until actually sold in a later year”.

[36] At 456.

[37] At 61, 014.

[38] At 61, 207.

[39] Farmers', at 61, 206.

[40] At 11, 133.

[41] Normally a capital expenditure, but subject to assessment (when the shares are bought for the purpose of resale) by virtue of s 65(2)(e).

[42] At 11, 038 ( per Richardson J).

[43] At 11, 045.

[44] At 11, 045.

[45] At 11, 133.

[46] That is to say, whilst the language of ss 101 and 104 prevents any principles of commercial accounting practice from impacting upon calculation of the expenditure component in the overall measurement of assessable income, at the same time calculation of the revenue component may be assisted by those very principles; because, in practice, it is not possible for the core provisions of the Act to provide a complete statutory formula for the measurement of assessable revenues.

[47] At 61, 207.

[48] Cape Brandy Syndicate v IRC [1921] 1 KB 64, 71 (perRowlatt J).