TRUSTPOWER LIMITED v COMMISSIONER OF INLAND REVENUE [2016] NZSC 91

TRUSTPOWER LIMITED v COMMISSIONER OF INLAND REVENUE [2016] NZSC 91 [27 July 2016]IN THE SUPREME COURT OF NEW ZEALANDSC 74/2015[2016] NZSC 91BETWEEN TRUSTPOWER LIMITEDAppellantAND COMMISSIONER OF INLANDREVENUERespondentHearing: 8, 9 and 10 March 2016Court: Elias CJ, William Young, Glazebrook, Arnold and OʼRegan JJCounsel: G J Harley, S A Armstrong, F M Ward and V V Kumar forAppellantD H McLellan QC, R L Roff, C M Kern and R N Park forRespondentJudgment: 27 July 2016JUDGMENT OF THE COURTA The appeal is dismissed.B Trustpower is to pay the Commissioner costs of $45,000 and reasonable disbursements to be fixed by the Registrar.____________________________________________________________________REASONS(Given by William Young J)Table of ContentsIntroduction [1]Overview of the case [4]Preliminary points [14]The facts in more detail [18]The legislation in more detail [24]The history of the dispute [31]The Commissioner's initial position [31]The Adjudication Unit [32]The Commissioner's response [35]The High Court challenge [36]The Court of Appeal [40]Evaluation [46]Our general approach [46]The statutory text [49]The relevant case law [51]Accountancy treatment [64]Policy considerations [66]On what side of the line does the expenditure fall? [71]Disposition [73]Introduction[1] Trustpower Ltd (Trustpower) derives its income from retail sales of electricity. It generates about half of the electricity it sells and buys the rest from other generators. During the 2006, 2007 and 2008 tax years, it incurred costs totalling approximately $17.7m applying for and obtaining resource consents under the Resource Management Act 1991 (RMA) in relation to four proposed electricity generation projects. We will refer to these by reference to their locations as Arnold, Wairau, Kaiwera Downs and Mahinerangi. The first two were in respect of proposed hydro schemes and the other two were for wind farms. Part of the Mahinerangi project (referred to as stage one) has been completed but Trustpower has not yet decided whether to proceed with the balance of that project or with the other three consented projects. In issue is whether the expenditure incurred in obtaining the resource consents for the four projects1 is on revenue account and therefore deductible, as the High Court held2 or, instead, is on capital account and therefore not deductible, as the Court of Appeal concluded.3[2] The costs for which deductibility is sought have been described as "feasibility expenditure"; this on the basis that the terms on which the resource consents wereobtained were, and remain, material to Trustpower's assessment of the feasibility ofthe projects and thus whether they should be completed. We accept that rational decisions to proceed could not be made prior to the consents being obtained. But,1 This includes all the Mahinerangi expenditure incurred in the three tax years in question. No part of that expenditure is reflected in the capitalised value of the now established wind-farm.2 Trustpower Ltd v Commissioner of Inland Revenue [2013] NZHC 2970, [2014] 2 NZLR 502 (Andrews J) [Trustpower (HC)].3 Commissioner of Inland Revenue v Trustpower Ltd [2015] NZCA 253, [2015] 3 NZLR 658 (Ellen France P, White and Miller JJ) [Trustpower (CA)].that said, the expression "feasibility expenditure" does not fully capture thesignificance of resource consents and thus the costs incurred in obtaining them. Securing the consents amounted to tangible progress towards eventual completion of the projects (which could not be built without them).[3] There is a sense in which the case comes down to the relative importance of the absence of a decision to complete the developments at the time the expenditure was incurred4 (as stressed by Trustpower) and the tangible progress towards their completion which the resource consents represent (as stressed by the Commissioner). To express the same idea in a slightly different way, the case turns on whether intermediate steps (namely the obtaining of resource consents) towards the completion of the capital projects (being the generation projects) are appropriately treated as referable to, and as advancing, those projects, even though (a) when those steps were taken, Trustpower had not decided whether it would complete the projects and (b) any decisions to do so after the obtaining of the consents would be based on extensive additional investigations and market conditions.Overview of the case[4] The case falls to be determined under the Income Tax Act 2004 (the 2004 Act) although some reference will also have to be made to the Income Tax Act 2007 (the 2007 Act). Central to the case is the application of ss DA 1 and DA 2 of the 2004 Act:DA 1 General PermissionNexus with income(1) A person is allowed a deduction for an amount of expenditure or loss (including an amount of depreciation loss) to the extent to which the expenditure or loss is–(a) incurred by them in deriving–(i) their assessable income; or4 This includes the expenditure in relation stage one of Mahinerangi as Trustpower did not decide to proceed until 2010.(b) incurred by them in the course of carrying on a business for the purpose of deriving–(i) their assessable income; orGeneral permission(2) Subsection (1) is called the general permission.DA 2 General limitationsCapital limitation(1) A person is denied a deduction for an amount of expenditure or loss to the extent to which it is of a capital nature. This rule is called the capital limitation.[5] The effect of these provisions is that a taxpayer is entitled to a deduction if:(a) the expenditure or loss was incurred in deriving income or in the course of carrying on business for that purpose, so that s DA 1 (the general permission) is satisfied; and(b) the expenditure or loss is not of a capital nature, so that s DA 2 (the capital limitation) is not engaged.[6] In this way, the statute brings into play the long-standing distinction between capital and revenue. Some assistance with the application of this distinction is provided by specific provisions of the 2004 Act which supplement or modify what would otherwise be the application of that distinction. One of these modifications which we should mention at this point is s DB 13B of the 2004 Act which permits a deduction in respect of costs incurred in respect of unsuccessful or withdrawn resource consent applications. We will discuss these provisions in some detail later.5[7] Although it is not referred to in the judgments of the High Court and Court ofAppeal, the Inland Revenue Department's Interpretation Statement: Deductibility of5 See below at [26]–[30].Feasibility Expenditure very much dictated the shape of the argument presented to those Courts.6[8] The Interpretation Statement discusses deductibility under the general permission (s DA 1) for pre-business commencement feasibility expenditure in this way:12. The element of commitment is also critical. To conclude that a business or an income-earning activity has commenced, it must be shown that a decision has been made to enter into that business or activity. If expenditure relates to activities undertaken in order to decide whether to enter into a particular business or income-earning activity, that expenditure will lack the required nexus and will be non-deductible.[9] The Interpretation Statement also deals with the application of the capital limitation (s DA 2) in this way:18. Feasibility expenditure incurred principally for the purpose of placing a taxpayer in a position to make an informed decision about the acquisition of an asset (or other enduring advantage) will not generally be expenditure incurred in relation to that particular asset or advantage. However, once the decision has been made to proceed with the acquisition or development of a particular capital asset, any expenditure incurred beyond that point will relate to the acquisition of that asset and will indicate that the expenditure is more likely to be of a capital nature.20. Once a decision has been made to proceed with the acquisition or development of an asset, any expenditure incurred after that time will more readily be treated as being related to the underlying capital project and thereby the profit-making structure of the business, so will not be deductible. For these purposes, it is irrelevant whether the expenditure is successful. In addition, commitment in this context does not necessarily mean a taxpayer will proceed with the acquisition or development regardless of future events (eg, the availability or otherwise of suitable planning consent), only that the taxpayer has made a firm decision to proceed. Similarly, it is considered that the fact the decision to proceed, or aspects of theprocess, may be contingent on factors beyond the taxpayer's controlwill not mean that a taxpayer is not committed to pursuing a certain course of action. Identifying when a decision to proceed has been reached in any particular situation will always be a question of fact and degree, and it is necessary to weigh all the relevant factors to determine whether a commitment has been made to a project.6 Inland Revenue Department Interpretation Statement: Deductibility of Feasibility Expenditure(IS08/02, June 2008) [the Interpretation Statement].[10] In the Courts below, the Commissioner accepted that expenditure in relation to the four projects in issue was on revenue account where it preceded the decision to apply for resource consents. She maintains, however, that the resource consents obtained by Trustpower are identifiable intangible assets and that from the point that Trustpower committed to the resource consent process, the associated expenditure was on capital account. Trustpower, while accepting, of course, the deductibility of expenditure incurred before the decision was made to apply for resource consents, maintains that what is important is not commitment to seek resource consents but rather commitment to proceed with (in the sense of build) the projects. Trustpower contends that such commitments were not made and that all expenditure under consideration is therefore on revenue account.[11] The litigation stance of the Commissioner seems to be a function of the way in which the dispute was addressed by the Adjudication Unit.7 This stance assumes that expenditure cannot be said to be for the development of a project, and thus on capital account, unless the taxpayer is committed to the completion of the project. On this basis, the tangible progress towards the eventual completion of the projects (should they proceed) represented by the resource consents and intended to be secured by the related expenditure did not, in itself, warrant a conclusion that the associated expenditure was on capital account. Accordingly, the Commissioner could only succeed if the resource consents were themselves capital assets, in which case, expenditure incurred after Trustpower had committed itself to applying for the resource consents would be on capital account.[12] The primary arguments of Mr Harley for Trustpower were as follows:(a) He accepted, and indeed propounded, the commitment approach just outlined, that expenditure in relation to capital projects is on revenue account until the taxpayer commits to the completion of the project. He was prepared to accept some qualifications to this, for instance in relation to land acquisitions, but otherwise the proposition was advanced in general terms.7 See below at [32]–[34].(b) The resource consents were not stand-alone capital assets.We will use the expression "commitment approach" to denote Mr Harley's argumentas just summarised. We note that this commitment approach is not necessarily consistent with the Interpretation Statement8 but, since this case falls to be determined by reference by the general law, a detailed analysis of the Interpretation Statement would be beside the point.[13] Our leave decision made it clear that we were going to look generally at the deductibility of feasibility expenditure in connection with capital projects9 and, for reasons we will shortly explain, we consider that we are entitled to decide the case on this basis, unconstrained by stances adopted by the parties at earlier stages of the dispute.10 On this basis, the salient features of the case are as follows:(a) Section DA 1 denies deductibility to feasibility expenditure for a new, or an entirely separate, business venture which is not underway at the time the expenditure is incurred. If activities are undertaken to decide whether or not to enter a business (as against, as in this case, in thecourse of a taxpayer's existing business), the expenditure will lack therequired nexus to a business and s DA 1 will not be satisfied. In determining whether a business has commenced, the commitment (or otherwise) of the taxpayer to that business – or, as the Interpretation Statement puts it, whether "a decision has been made to enter into that business or activity" – is highly material. Paragraph 12 of the8 This is essentially because the argument assumes that the resource consent expenditure is withinwhat was contemplated as "feasibility expenditure" in [18] of the Interpretation Statement,despite resulting in tangible progress towards completion of the projects.9 Trustpower Ltd v Commissioner of Inland Revenue [2015] NZSC 134. One of the approved questions was: "What is the correct approach to determining whether the expenditure of the type at issue in this proceeding has been incurred on revenue or capital account, for the purposes of s DA 2(1) of the Act?"10 See below at [14]–[16].Interpretation Statement cited in [8] above is consistent with this and with the authorities.11(b) As we will see, the Court of Appeal expressed the view that the feasibility expenditure did not satisfy the s DA 1 general permission.12We disagree. The expenditure was incurred by Trustpower in the course of carrying on its business as a generator and retailer of electricity for the purpose of deriving assessable income (see s DA 1(b)(i)). So the general permission was satisfied. This was accepted by Mr McLellan QC for the Commissioner. Indeed, prior to the Court of Appeal judgment, it had never been suggested that the general permission was not satisfied.(c) Although s DA 1 is satisfied, the expenditure is deductible only if not within the s DA 2 capital limitation.(d) In her judgment, Andrews J saw the application of the capital limitation as turning on whether the resource consents were capital in nature.13 This was premised on her acceptance of the commitment approach. She was of the view that the resource consents were not stand-alone assets and, in the alternative, even if they were, they were revenue and not capital in nature.14(e) We do not accept that the capital/revenue issue is controlled by the commitment approach. For this reason, we do not see a conclusion that the resource consents were stand-alone capital assets as a necessary pre-condition to the application of the capital limitation.11 See generally Stevens v Commissioner of Inland Revenue [1989] 1 NZLR 404 (HC); Calkin v Commissioner of Inland Revenue [1984] 1 NZLR 440 (CA); Goodman Fielder Wattie Ltd v Federal Commissioner of Taxation (1991) 29 FCR 376; Softwood Pulp and Paper Ltd v Federal Commissioner of Taxation (1976) 7 ATR 101 (VSC); and Griffin Coal Mining Co Ltd v Commissioner of Taxation (1990) 21 ATR 819 (FCAFC). Reference can also be made to JohnPrebble and Hamish McIntosh "Deducting Expenditure to Assess the Feasibility of ConstructingCapital Assets: Opinions from Inland Revenue, the High Court and the Court of Appeal" 6VUWLRP 24/2016 at 16–19.12 Trustpower (CA), above n 3, at [96].13 Trustpower (HC), above n 2, at [97].14 At [139]–[141].(f) On a purist view of the capital/revenue distinction, any expenditure (feasibility in nature or otherwise) addressed to a capital project (as the generating projects in this case are) is necessarily on capital account. On this approach – which has been espoused by Professor John Prebble QC and Hamish McIntosh15 – the feasibility expenditure in issue was necessarily not deductible.(g) The approach which we adopt is broadly similar to that proposed by Professor Prebble and Mr McIntosh but, for reasons which we explain, allows for some flexibility, for instance, in respect of initial stages of feasibility work.(h) As is apparent, we consider that some feasibility expenditure referable to proposed capital projects might sometimes be deducted. We do not, however, see such deductibility as extending to external costs incurred in respects which do, or were intended to, materially advance the capital project in question.Preliminary points[14] In two respects, the approach adopted by the Court of Appeal deviated from the case advanced by the Commissioner: first as to the non-satisfaction of s DA 1; and second as to the non-applicability of the commitment approach in relation to s DA 2.[15] In relation to the first deviation, Mr Harley raised the preliminary point that because the statements of positions of the parties had not put s DA 1 in issue, it was not open to the Court of Appeal to have done so given s 138G of the Tax Administration Act 1994. We accept that s DA 1 had not been put in issue in the statements of position. For this reason, it would not have been open to the Commissioner to argue that s DA 1 had not been satisfied unless the leave of the hearing authority (in this case the High Court and, on appeal, the Court of Appeal) under s 138G(2) had been obtained. Such leave was not sought. As well, given the15 Prebble and McIntosh, above n 11, at 6–9.strictness of the s 138G criteria, it is difficult to see how leave could have been obtained if sought. So far we are with Mr Harley on this point. However, what is not quite so clear is whether s 138G precludes a hearing authority, without invitation from the taxpayer or Commissioner, deciding a challenge on a basis which goes outside the propositions advanced in the statements of position. As it turns out, however, we do not need to determine this issue because, as already noted, we are well-satisfied that s DA 1 was satisfied.[16] The application of s DA 2 has always been in issue in the case. For this reason it would have been rather more difficult to raise a s 138G point in respect of the second deviation and Mr Harley did not do so. Instead he addressed directly whether the commitment approach was correct and, in this way, responded to the way in which our leave decision had identified the issues we intended to address.16[17] We also note that Mr Harley argued that the Court of Appeal's judgment waspremised on factual findings or assumptions which differed from the findings of fact made by Andrews J and were not consistent with the evidence given at trial. As will be apparent from our discussion at the beginning of our reasons, there is scope for different descriptions of the same facts. More generally, we are of the view that, to the extent to which the facts are material to the approach which we adopt, they are not in dispute. It is to these facts which we now turn.The facts in more detail[18] Trustpower maintains what it describes as a "development pipeline" whichconsists of over 200 possible new generation projects. Not all of these projects will be carried through to completion.[19] Projects which are completed will have gone through a three step feasibility analysis:(a) The first step consists of all assessments made prior to the obtaining of resource consents.16 See above at n 9.(b) The second step occurs after resource consents are obtained, and involves, among other things, engineering and geo-technical assessments, the working up of detailed costings, working out how the scheme might be connected to the national grid or local networks and a judgement as to the net present value of the project if completed.(c) The third and final stage involves the preparation of a business case for the board and a decision by the board whether or not to proceed.[20] In respect of the four projects, Trustpower spent a total of $21,149,149 during the 2006–2008 tax years. Of this $279,543 related to land acquisition and it has been agreed between the parties that this is not deductible. The Commissioner has allowed deductions of $2,662,527 for preliminary investigative expenditure and a further deduction (under s DB 13B) of $534,789 for work on a consent application (associated with the Mahinerangi scheme) which was withdrawn. This leaves in dispute the deductibility of expenditure totalling $17,672,290.[21] As a result of this expenditure, Trustpower has obtained resource consents (land use consents, water permits and discharge permits) in respect of all four projects in issue. Apart from some of the land use consents which are not limited as to time, the consents are for fixed periods, commencing from 2008 to 2011 (generally 10, 15 or 35 years).[22] The Mahinerangi project has been partly completed in that stage one, an operating wind farm, has been established. The balance of that project and the other three projects are in limbo. Trustpower has not agreed to sell any of them and it has likewise not committed in any sense to their construction. It has, however, not abandoned any of them. It is reasonable to suppose that the bundle of rights which Trustpower has accumulated (consisting of the consents, land acquired, land access agreements and the associated technical data) in relation to each of the uncompleted projects (including the uncompleted stage or stages of Mahinerangi) is marketable and indeed Trustpower has been approached about its willingness to sell two of them.1717 Specifically, Kaiwera Downs and Mahinerangi.[23] With the exception of stage one of the Mahinerangi project, step two of Trustpower's three step analysis referred to in [19] has yet to be completed. So the only completed step in relation to those projects is step one. That comment, however, underplays the significance of the progress which has been made. For instance, as of August 2013, only nine of the 200 odd projects in the development pipeline had reached the consent application stage. The consents for the four projects are of strategic significance as they give Trustpower the option to move reasonably quickly towards construction and, as noted, they might be thought to be already marketable. At this point we should make it clear that marketability or otherwise of the projects in their present state is not in itself fundamental. It does, however, emphasise the tangible nature of the progress which had been made.The legislation in more detail[24] The s OB 1 definition of "property" expressly encompasses resourceconsents.18 Resource consents are within the s EE 53 meaning of "depreciable intangible property"19 and are "depreciable property" for the purposes of s EE 6when they are time limited unless the costs of obtaining them are deductible.20 This means that the question whether a time limited resource consent is depreciable or not turns on the application of the capital/revenue distinction.21 We accept therefore that s EE 53 does not resolve the case in favour of the Commissioner.[25] At this point, we should introduce the problem of expenditure on projects which do not proceed. If such expenditure is on capital account, the taxpayer might be thought to be prejudiced. Because the project never eventuates, there is no corresponding asset or permanent advantage accruing and, therefore, nothing to depreciate. Further, because the wasted expenditure is on capital account, there is no available deduction. Such expenditure is sometimes said to have disappeared into a"black hole".2218 Income Tax Act 2004, s OB 1, definition of property, para (b), which is applicable to subpart EE.19 Income Tax Act 2004, s EE 53 and Schedule 17, cl 9.20 Income Tax Act 2004, ss EE 6(3) and EE 7(j).21 Trustpower (CA), above n 3, at [49]. By way of example, resource consents may form part of the stock-in-trade of a land-developer/speculator and, if so, the costs of obtaining them are obviously on revenue account.22 Craig Elliffe "The Problem with 'Black Hole' and Feasibility Expenditure: Some Suggestionsfor Reform" (2011) 17 NZJTLP 67 at 67 and n 1.[26] In relation to resource consents, the "black hole" expenditure problem is in part addressed by the 2004 Act in s DB 13B which provided:DB 13B Expenses of failed or withdrawn application for resource consentDeduction(1) A person who applies for the grant of a resource consent under the Resource Management Act 1991 and is refused the grant or withdraws the application is allowed a deduction for expenditure –(a) that the person incurs in relation to the application; and(b) that would have been part of the cost of depreciable property, or otherwise a deduction if the application had been granted; and(c) for which the person is not allowed a deduction under another provision.Timing of deduction(2) The deduction is allocated to the income year in which the grant is refused or the application is withdrawn.Link with subpart DA(3) This section overrides the capital limitation. The general permission and other general limitations still apply.[27] The explanatory note to the introduction of s DB 13B stated:23Patent and resource management application costsAn amendment is proposed that allows costs associated with patent and resource-management-consent applications to be deducted, although the applications are not granted or are withdrawn. Costs for such applications cannot currently be claimed under the general deductibility rules as they are a capital expense. Nor can they be depreciated as there is no depreciable asset. Under the proposed change, the deductible expenditure consists of those costs that would have been depreciable if a patent or resource management consent had been granted.23 Taxation (Annual Rates, Venture Capital and Miscellaneous Provisions) Bill 2004 (110-1) (explanatory note) at 5.[28] The Commentary on the Bill stated:24Patents and certain consents issued under the [RMA] are depreciable intangible property. To the extent expenditure incurred in applying for a patent or resource management consent results in an application being granted, the costs must be capitalised and depreciated. However, if an application is unsuccessful or is withdrawn, any costs incurred up to that point are not depreciable as there is no depreciable asset. Nor can this expenditure be expensed under the general deductibility rules because it is capital in nature.[29] The Court of Appeal was of the view that the explanatory note and commentary were inaccurate.25 We agree with the Court of Appeal on this to the extent that we accept that there are circumstances in which expenditure associated with applications for resource consents are on revenue account. But both the explanatory note and the commentary are suggestive of a legislative understanding that such expenditure will usually be on capital account; thus the need for s DB 13B.[30] Section DB 13B is not engaged in the present case as the resource consent applications were successful. But there is now also a deduction available under the 2007 Act for expenditure incurred in respect of obtaining resource consents which lapse or are surrendered. This is pursuant to s DB 19. This, however, is applicable only in the case of resource consents which are depreciable property (that is, are time limited). This section was not in force during the tax years in issue in this case. The current position, however, is that s DB 19 of the 2007 Act now addresses, albeit notcompletely, the "black hole" problem with expenditure on resource consents and, should Trustpower surrender the resource consents, a deduction in relation to those which are time limited will be available.The history of the disputeThe Commissioner's initial position[31] The Commissioner accepted that the deductions claimed were within the general permission under s DA 1 but disallowed them on the basis of the capital limitation under s DA 2.24 Policy Advice Division of the Inland Revenue Department "Taxation (Annual Rates, VentureCapital and Miscellaneous Provisions) Bill 2004; Commentary on the Bill" (Inland Revenue,Wellington, 2004) at 9.25 Trustpower (CA), above n 3, at [45].The Adjudication Unit[32] The resulting dispute was addressed first by the Adjudication Unit of the Inland Revenue Department.26 In its report the Unit took as its starting point theInterpretation Statement to which we have referred and began its assessment by noting:The point at which the expenditure alters from revenue to capital is when a commitment, or decision, has been made to proceed with the acquisition or development of a capital asset. Commitment does not require a legal or other form of binding decision that is final and irrevocable. Rather commitment requires a decision to proceed. This is in contrast to a taxpayer continuing to weigh up whether or not to proceed. A commitment can still be made despite recognising that whether the development or acquisition ultimately goes ahead may be contingent on particular factors.[33] The Adjudication Unit concluded that Trustpower had not committed tocomplete any of the projects. It was, however, of the view that "generally" theresource consents should be regarded as capital assets (with the corollary that the associated expenditure was on capital account and therefore no deduction was available). The Unit left open the possibility that some of the consents might berevenue in nature and it remitted the dispute to the Inland Revenue Department'sService Delivery Group to calculate the necessary adjustments.[34] As is apparent, the Adjudication Unit adopted the commitment approach which we have summarised at [11]. This was to drive the way in which the case was later argued in the challenge proceedings.The Commissioner's response[35] The Commissioner's response to the Adjudication Unit was to maintain thatall consents were capital assets. There was a reassessment letter of 29 March 2012which affirmed the Commissioner's original position.26 Now called the Disputes Review Unit.The High Court challenge[36] Understandably, given this procedural history Andrews J approached the case on the basis that what was in issue was whether the resource consents obtained were stand-alone assets and, if so, whether they were capital in nature.27[37] In approaching the case in this way she referred to the leading cases on the capital/revenue distinction,28 including BP Australia Ltd v Commissioner of Taxation for the Commonwealth of Australia.29 Andrews J then considered in some detail the juristic nature of resource consents and analysed their treatment in the 2004 Act.30Having done so, she rejected the Commissioner's argument that s DB 13B assisted the Commissioner and more generally saw that section and the other provisions of the Act as being of limited assistance.31[38] Andrews J's judgment proceeded on the basis that the expenditure was onrevenue account unless the resource consents were properly regarded as stand-alone assets and that they were not.32 Although this conclusion was sufficient to resolve the case in favour of Trustpower, Andrews J went on to consider whether the resource consents, if properly regarded as stand-alone assets, were capital or revenue in nature.33 She concluded this aspect of her judgment by saying that even if the resource consents were stand-alone assets, she would categorise them as being revenue and not capital assets.[39] Her approach generally proceeded on the basis that the applications for theconsents were part of Trustpower's general business operations and that the consents were only one of the components of particular project options in the development pipeline. That pipeline is only one of the possible sources of supply of electricity as it is also possible to acquire power from other generators. The resource consents were not sufficient for a decision to be made to take any project through to the next27 Trustpower (HC), above n 2, at [5]–[6].28 At [44]–[56].29 BP Australia Ltd v Commissioner of Taxation for the Commonwealth of Australia [1966] AC 224 (PC) at 264–265.30 Trustpower (HC), above n 2, at [57]–[74].31 At [75]–[79].32 At [97].33 At [98]–[139].stage. While they may have been of value to Trustpower, the consents were not used and no income was generated. The expenditure incurred was, thus, part of the feasibility process and revenue in nature.The Court of Appeal[40] The Court of Appeal agreed with Andrews J that the scheme and provisions of Parts D and E of the 2004 Act were not controlling.34 It did, however, consider that the ability of Trustpower to treat the expenditure as being on capital account andto claim depreciation was "not irrelevant".35 The Court then went on to review the leading decisions on the capital/revenue distinction36 and, in doing so, paid particular attention to cases which concerned the categorisation of expenditure in relation to the exploration for mineral resources,37 assessing the feasibility of additional capital plant,38 and the obtaining of resource or like consents.39[41] The Court then concluded that the transactions were on capital account. This was primarily because:[87] the expenditure was incurred for the purpose of enablingTrustpower to extend or expand its electricity generation business. It was incurred in taking preliminary steps and then applying for and obtaining resource consents for four projects that were in the development pipeline.The "development pipeline" was a means of determining the viability,feasibility, and costs of building new generation capacity. [The] newgeneration capacity related to the acquisition of the means of production by extending the business organisation. From a practical and business point of view, the expenditure was calculated to effect the extension or expansion ofTrustpower's business structure. In the words of Richardson J in McKenzies, the expenditure had the purpose of acquiring assets that would be part of the profit-making structure of the business, namely the addition of four new projects.(citations omitted)34 Trustpower (CA) , above n 3, at [20].35 At [50].36 At [51]–[84].37 Commissioner of Taxation v Ampol Exploration Ltd (1986) 13 FCR 545 (FC); and Esso Australia Resources Ltd v Commissioner of Taxation (1998) 84 FCR 541 (FCAFC).38 Griffin, above n 11.39 ECC Quarries Ltd v Watkis (Inspector of Taxes) [1977] 1 WLR 1386 (Ch); Waste Management New Zealand Ltd v Commissioner of Inland Revenue (1995) 17 NZTC 12,147 (CA); Case T53(1998) 18 NZTC 8,404 (TRA); and Milburn New Zealand Ltd v Commissioner of Inland Revenue (2001) 20 NZTC 17,017 (HC).In the opinion of the Court it was irrelevant that Trustpower had not committed to the projects before the expenditure was incurred.40 Further:[89] Determined objectively, there was a sufficient connection between the expenditure and capital. The object of the expenditure was capital evenif, as a matter of Trustpower's corporate governance, the final decision toapply for the resource consents had not been made and that decision was contingent on what the preliminary work might show.Although of the view that the resource consents were in fact valuable,41 the Court thought it irrelevant to consider whether they were stand-alone assets:[95] Acceptance of Trustpower's arguments that the resource consentswere obtained in the course of taking the respective projects further along the development pipeline confirms that the disputed expenditure was for thepurpose of extending or expanding Trustpower's existing business and wastherefore on capital account.[42] The Court went on to say:[96] the disputed expenditure was not incurred "in carrying on" Trustpower's business or in earning the income of the existing business or inperforming the income-earning operations of the existing business.Trustpower's profit-making enterprise is the generation and retailing of electricity, not the development of its pipeline of possible new projects or the investigations of, and applications for, resource consents for those projects. Possible future projects in its development pipeline are for the purpose of extending, expanding or altering its business structure in the future, not partof the carrying on of Trustpower's ordinary business activities or the takingof steps within that framework, being the generation and retailing of electricity. In terms of s DA 1 the requisite nexus between the incurring of the expenditure and the deriving of the income is not established.[97] This means we do not accept Trustpower's submission that the buildor buy decision was so intimately connected with the revenue side of the business that the disputed expenditure was on revenue account. Thedecision Trustpower faced is no different in principle from that of a manufacturer who must decide whether to buy some input or to invest in the capacity to build it by acquiring land and plant. Firms routinely face build or buy decisions which turn on whether the item concerned can be built more efficiently by the firm itself. By obtaining resource consents, Trustpower invested unequivocally in capacity, whether or not it was committed at that time to proceed with the build. The investment was inherently capital in nature.(emphasis added)40 At [88].41 At [92].[43] The Court concluded:[101] In this context it cannot be decisive whether the applications for consents were ultimately successful or not, as the focus must be on the point in time at which the expenditure is incurred. Consequently, expenditure on an unsuccessful application may still be on capital account.[102] This conclusion is also supported by the scheme of the ITA. As we have already noted, the ITA recognises that certain resource consents are depreciable intangible property and expenditure on them may be depreciated under sub-pt EE.(citations omitted)[44] The sentence we have emphasised in the passage cited in [42] was sharply challenged by Mr Harley. As we have noted, the Commissioner has always accepted that the general permission was satisfied. The indication that it was not thus came as something of a surprise to all parties. As already explained, we have no difficulty accepting that the general permission was indeed satisfied.42 The expenditure was incurred by Trustpower in the course of carrying on its existing business as a generator and retailer of electricity, a business which it carries on for the purpose of deriving assessable income. That this is so was accepted by Mr McLellan, for the Commissioner.[45] Mr Harley also criticised other passages in the Court of Appeal judgment set out in [42] above. He suggested that the Court of Appeal's conclusion that theexpenditure was for the purpose of acquiring assets was inconsistent with the finding of Andrews J that the purpose was instead simply to inform the decision whether to build. This comes back to the feasibility expenditure/development expenditure point which we mentioned at the beginning of these reasons.43 We do not see the relevant remarks of the Court of Appeal as necessarily inconsistent with the finding of Andrews J but rather as reflecting a different perspective. Since it would not be practicable for Trustpower to commit to construction until it was known what the terms of the resource consent would be,44 the resource consent expenditure can be treated as informing the ultimate decisions which Trustpower must make. But42 See above at [15].43 See above at [2].44 Considerable other investigative work and assessment exercises would also precede any final decisions to proceed with construction.likewise, if it is accepted that a purpose of acquiring or developing assets can beconditional, the Court of Appeal's description of the expenditure was also correct.EvaluationOur general approach[46] Mr Harley was critical of what he regarded as the propounding by the Courtof Appeal of a "sufficient connection" test. We, however, read the Court of Appeal'sjudgment as proceeding on the basis that where expenditure has been incurred for the purpose of acquiring or developing capital assets, there is "sufficient connection between expenditure and capital" to warrant the conclusion that the expenditure is on capital account. So construed, "sufficient connection" is simply part of theconclusion that the expenditure was on capital account because it was incurred for the purpose of developing capital assets. The real issue in the case is whether such a conclusion can only be reached on the basis of the commitment approach advanced by Mr Harley.[47] We accept that commitment is material to the application of s DA 1 (the general permission) in respect of new business activities.45 We do not, however, see the commitment approach, as advanced by Mr Harley, as helpful when determining whether the s DA 2 capital limitation is engaged. Instead, we are substantially in agreement with the approach espoused by Professor Prebble and Mr McIntosh.46Their proposition is that everything that relates to a possible capital asset is non-deductible, whether or not a capital asset results. However, for reasons which we will explain, we consider that preliminary expenditure on feasibility studies may sometimes be deductible. The legislation does not prescribe any particular test for determining the point at which such expenditure ceases to be deductible and it is not for us to construct one. Rather, we are required to determine whether the expenditure in this case is in the nature of capital. To the extent that this is so, the capital limitation in s DA 2 applies. For reasons which we will explain, our determination is in favour of the Commissioner: the expenditure was in the nature of capital and therefore not deductible.45 The authorities are referred to above at [5](a).46 See Prebble and McIntosh, above n 11.[48] We accept that the proposed projects may or may not proceed. If they do not proceed, the development proposals, including resource consents and land access arrangements may be sold. In the event that (a) the projects do not proceed and (b) the development proposals are not sold, the expenditure will be wasted (that is neither capitalised nor deducted), a consequence which now (although not at the time the expenditure was incurred) is able to be at least in part ameliorated under s DB 19 of the 2007 Act. The possibility that some, or all, of the expenditure may in the end turn out to be wasted does not displace the operation of the capital limitation.The statutory text[49] Such assistance as can be derived for the statutory text is in favour of the Commissioner rather than Trustpower. This is by reason of the specific provisions of the 2004 and 2007 Acts as to (a) resource consents being depreciable property and (b) deductibility of expenses associated with failed or withdrawn resource consent applications and abandoned consents. These provisions suggest a legislative assumption that, in the absence of such provisions, resource consent expenditure is usually on capital account.47[50] We also see some significance in the way in which the capital limitation is expressed. Section DA 2 denies a deduction for expenditure "to the extent to which it is of a capital nature". The phrase "to the extent" suggests that questions of degreemay be involved.The relevant case law[51] There is some Canadian authority to the effect that expenditure on a capital project which does not result in the acquisition of a capital asset is deductible.48 The preponderance of authority, however, is to the effect that expenditure on a capital project is on capital account even though the project is never completed. One of the many authorities to this effect is Milburn New Zealand Ltd v Commissioner of Inland47 See above at [26]–[30].48 Bowater Power Co Ltd v Minister of National Revenue [1971] FC 421; and Wacky Wheatley'sTV & Stereo Ltd v Minister of National Revenue [1987] 2 CTC 2311 (Tax Court).Revenue,49 which involved facts which are reasonably similar to those of the present appeal.[52] In Milburn, the taxpayers required quarries to produce aggregate for concrete plants which they operated. In issue was expenditure they had incurred in seeking the resource consents and licences which were required if they were to use the sites as quarries. The applications for consents and licences in relation to one site failed. There being no equivalent to s DB 13 then in effect, the expenditure in relation to that site was thus wasted. The other applications appear to have been successful.50The taxpayers claimed that the expenditure in relation to all applications wasdeductible. This was rejected by the Commissioner and the taxpayers' appeal to theHigh Court was dismissed by Wild J.[53] The approach of the Commissioner in Milburn was along the same lines as that later formalised in the Interpretation Statement to which we have referred.51The Commissioner refused to accept the deductions in issue because he was of the view that the taxpayer had committed itself to the development of the three sites before the expenditure in question was incurred.52 In contradistinction, the Commissioner allowed the deductibility of expenditure in relation to 48 other sites in respect of which consents were not sought.53 We note in passing that there is someambiguity inherent in the use of the word "commitment" and like words in this context. Obtaining the resource consents was a necessary pre-condition for establishment of the quarries and thus part of their development. It may be that all the Commissioner was asserting was that by deciding to apply for the consents, the taxpayers thereby committed to the development of the quarries, even though final decisions to proceed could not be made until the result of the applications was known.49 Milburn, above n 39.50 There were three sites: Bombay, Fraser and Alpha Creek. It is clear from the judgment that the applications in relation to the Bombay site were successful and in the case of Fraser were unsuccessful: at [10]–[12]. Although the judgment is not explicit as to the fate of the applications in relation to the Alpha Creek site (see at [11]), it would appear that they were successful as the discussion of failed expenditure was in relation to one site only: at [30] and [39].51 Above at [7]–[9].52 Milburn, above n 39 at [23]–[26].53 At [8]–[9].[54] In finding for the Commissioner, Wild J observed:[32] I consider the Commissioner's position is the correct one in relation to the expenditure here. I base my view on:[a] The nature of the business of [the taxpayers].[b] The importance of [the sites to the taxpayers' businesses].[c] The amount of the expenditure.[d] Its sustained nature ie the length of time over which the expenditure was incurred.[e] The nature of the expenditure: all on the obtaining of consent necessary before production could begin.[f] [c]-[e] when contrasted with the amount, duration and nature of expenditure on [the] 48 other prospects.These six factors, certainly in combination, indicate to me that the taxpayers, having investigated or evaluated the three sites, had made business decisions to expend money in developing the sites for commercial production.[55] This reasoning has been seen as providing support for the approach taken in the Interpretation Statement. As well, as Mr Harley pointed out, at the time they applied for resource consents the taxpayers in Milburn owned the land on which they proposed to operate quarries. So on obtaining resource consents on terms they sawas acceptable, they were in a position to develop the quarries. Trustpower's situationis, in this respect, different, as it has yet to obtain all the land which must be acquired before the generation projects can be completed.[56] Despite the considerations which we have just mentioned, we regard Milburnas being against Trustpower's argument. At most, Wild J took the Commissioner'sposition as to commitment as a starting point, rather than independently evaluating and adopting it. Further, it is, as we have pointed out, by no means clear from thereport what the Commissioner meant by "commitment". The taxpayers in Milburnhad argued that they could not be taken to have committed to any of the proposed quarries prior to the obtaining of resource consents; this given the possibility that the consents obtained may have been subject to conditions too onerous to allow economic quarrying.54 If the Commissioner was required to establish that the54 At [29].taxpayers had a firm commitment to establish working quarries this argument was logically unassailable. But in fact the taxpayers lost their appeals. This is becausethe Judge concluded that they had "made business decisions to expend money indeveloping the sites for commercial production".55 It is clear that the Judge considered that the money was (a) spent developing the sites for commercial production and (b) on capital account; this despite the absence of business decisions to establish the quarries. For this reason, we see the case as not consistent with the commitment approach espoused by Mr Harley.[57] There are other cases which are distinctly against the proposition that a commitment to complete a capital project is required as a pre-condition to treating the associated expenditure as being capital in nature.[58] The English case ECC Quarries Ltd v Watkis (Inspector of Taxes) involved facts which were reasonably similar to those in Milburn – applications for planning consents to develop quarries.56 These were called in for determination by the Minister who eventually declined them. In doing so the Minister indicated that they might be renewed subject to certain qualifications. There is no suggestion in the judgment that the conclusion that the associated expenditure was on capital account was dependent upon the taxpayer being committed to the quarry developments in the sense proposed by Trustpower.[59] Another English case on point is Sargent (Inspector of Taxes) v Eayrs which concerned the deductibility of expenditure incurred by an English farmer investigating the possibility of extending or developing his business by acquiring a farm in Australia.57 His investigations were never more than preliminary and plainly there was no commitment to the acquisition of any particular farm. Goff J was nonetheless of the view that, if the equivalent of our general permission had been satisfied, the expenditure was on capital account.55 At [32].56 ECC Quarries Ltd, above n 39.57 Sargent (Inspector of Taxes) v Eayrs [1973] 1 WLR 236 (Ch).[60] Finally, we note that in Waste Management Richardson J had no doubt that expenditure associated with the feasibility of a proposed capital project was, in the absence of specific legislative provision to the contrary, on capital account.58[61] One of the Canadian cases relied on before us was Bowater Power Co Ltd v Minister of National Revenue.59 In Bowater, the taxpayer had incurred expenditure in commissioning engineering studies as to the generation of additional electricity from two existing facilities. In the end neither of the proposed projects proceeded. Noel ACJ concluded that the expenditure was on revenue account and therefore deductible. In doing so, he said:60I do not feel that merely because the expenditure was made for thepurpose of determining whether to bring into existence a capital asset, it should always be considered as a capital expenditure and, therefore, not deductible. In distinguishing between a capital payment and a payment on current account, regard must always be had to the business and commercial realities of the matter. While the hydroelectric development, once it becomes a business or commercial [reality] is a capital asset of the business giving rise to it, whatever reasonable means were taken to find out whether it should be created or not may still result from the current operations of the business as part of the every day concern of its officers in conducting the operations of the company in a business-like way. I can, indeed, see no difference in principle between all of these cases.[62] There is a broadly similar passage in the judgment of Davies J in Griffin Coal Mining Co v Federal Commissioner of Taxation.61 In issue was the deductibility of a feasibility study for the proposed construction of an aluminium smelter by a joint venture, of which the taxpayer was a member. A majority of the Court held that the cost was not within what we call the general permission. The Judges in the majority therefore did not have to – and did not – address whether that cost was on capital account. Davies J, in dissent, concluded that the general permission was satisfied and, as well, that the capital limitation did not apply. As to the latter point, this is what he said:62In my opinion, the expenditure at this early stage was relevant and incidentalto Griffin's existing business and was not of a capital nature. It is clear that no capital was contributed to the venture. The time for the contribution of58 Waste Management, above n 39.59 Bowater, above n 48.60 At 837–838.61 Griffin, above n 11.62 At 827.capital to the venture had not arrived. The expenditure was not directed to the acquisition of an asset of an enduring character. In relation to the aluminium project, no asset was acquired. The expenditure was not on items such as roads and drainage works necessary in the event the opening up of the new mines, or on the taking away of the overburden, or on the construction of buildings and the acquisition of plant and equipment. The subject expenditure was expenditure by a coal mining company made out of current revenue directed to ascertaining whether the development of an enhanced market for the coal was feasible, as well as attaining the intangible ends I have mentioned.[63] To the extent that the judgments of Noel ACJ and Davies J proceed on the basis that expenditure on a capital project is on capital account only if an asset is created, we disagree. On the other hand, we can envisage situations in which a judgment call may have to be made in relation to feasibility assessments which areso preliminary in nature that they cannot sensibly be seen as "directed to the acquisition of an asset of an enduring character", to use the language of Davies J.63Accountancy treatment[64] In both ECC Quarries and Milburn the taxpayers relied unsuccessfully on arguments as to the accountancy treatment of the expenditure in issue. Mr Harley maintained, and we accept, that the correct treatment under the relevant accounting standard (NZ IAS 38) was to expense costs associated with the resource consent applications. This accountancy treatment focuses on the point at which an intangibleasset is to be recognised in a firm's accounts and, on the basis of NZ IAS 38, theresource consents were appropriately not so recognised.[65] If the case turned on whether the resource consents were stand-alone assets, we accept that the accounting treatment would be material, although not controlling.64 We do not, however, see it as material on our approach to the case. NZ IAS 38 proceeds on the basis that expenditure which precedes (and thus necessarily expenditure which does not result in) the creation of an asset should be expensed. So expenditure of the kind involved in this case is either capitalised when an asset is recognised or, if not capitalised, expensed. Such a dichotomy is not reflected in tax law and in particular is not in accordance with the cases which establish that expenditure which is wasted (because it does not result in the creation63 At 827.64 Compare BP Australia, above n 29, at 271.of an asset) can nonetheless be on capital account. Although the analyses in ECCQuarries and Milburn (which focused on the appropriateness of accountancy conservatism in identifying profits), are expressed rather differently,65 the underlying premise is similar.Policy considerations[66] Policy considerations are against Trustpower.[67] The cases which we have cited show that expenditure on capital projects which do not proceed (that is wasted expenditure) can nonetheless be on capital account (as in Milburn). But on Trustpower's case, a proportion of the costs of development of a project which is completed may be on revenue account and able to be deducted even though it may always have been probable that the development would proceed. And, if it does proceed (as stage one of the Mahinerangi projecthas), Trustpower's position is that there is no need to reverse the deduction for theassociated resource consent expenditure. Given that the wasted expenditure inMilburn was on capital account, it is difficult to see why the successful expenditure on stage one of Mahinerangi should be on revenue account.66 In this respect we do not see the concept of commitment advanced by Trustpower as providing a logical or principled explanation for the differential tax treatment.[68] In Milburn, the practicality of the taxpayers establishing the proposed quarries was conditional on the obtaining of resource and other consents on satisfactory terms. So any commitment to proceed on the part of the taxpayers was necessarily conditional. More generally it might be thought that development decisions seldom, if ever, involve a commitment to proceed irrespective of the way events pan out or changes in circumstance. So the commitment approach could not be practicably applied on the basis that unconditional commitment is required (as Mr Harley accepted in oral argument). But once conditionality is allowed for, as it must be, the concept of commitment becomes indeterminate.65 ECC Quarries Ltd, above n 39, at 1398C–G and 1402D–G; and Milburn, above n 39, at [64].66 Given that no capital asset was brought into existence in respect of the resource consent application in Milburn which failed whereas stage one of the Mahinerangi project has beencompleted, Trustpower's case in this respect might be thought to be distinctly weaker than the corresponding but still unsuccessful argument in Milburn.[69] We accept that in marginal cases, the application of the capital/revenue distinction may involve indeterminate questions. Such indeterminacy, however, has to be accepted as implicit in the underlying rule. The same cannot be said of the indeterminacy associated with the commitment approach. As is apparent, we do not see this approach as providing a logical and principled explanation for why some expenditure is on capital account (as in Milburn) even though wasted and other expenditure which is identical, other than that it was successful, should be on revenue account (as Trustpower maintains in relation to stage one Mahinerangi). There being no practical utility in the commitment approach, we see the indeterminacy associated with it as introducing complications to this area of the law which serve no useful purpose and are therefore unnecessary.[70] The subjectivity which is implicit in the commitment approach would give rise to some practical problems. This is because a decision whether there was commitment would have to be made in a context in which a taxpayer:(a) can be expected to defer commitment as long as possible to maximise the extent to which expenditure can be deducted; and(b) has control of the contemporaneous records (board minutes and papers for example) which will form the basis of the later assessment whether there was a commitment.On what side of the line does the expenditure fall?[71] The expenditure on obtaining resource consents in this case was directly related to specific projects that would be on capital account if they came to fruition. The projects could not proceed without resource consents. Obtaining the consents thus represented tangible progress towards their completion. The expenditure is thus on capital account and not deductible.[72] We are not required to determine the status of the expenditure which preceded the decisions to apply for resource consent. It may be that the Commissioner could have denied deductibility in relation to at least some of that expenditure. We are, however, also of the view that expenditure associated withearly stage feasibility assessments may be deductible. Such assessments can be seen as a normal incident of business. Treating the associated costs as deductible is consistent with the passages of the judgments of Noel ACJ and Davies J which we have set out.67 It is also consistent with the use of the expression "to the extent" inthe capital limitation, which, as noted, suggests that questions of degree may be involved.68 Expenditure which is not directed towards a specific project or which is so preliminary as not to be directed towards the advancement of such a project is likely to be seen as being on revenue account.Disposition[73] The appeal is dismissed. Trustpower is to pay the Commissioner costs of $45,000 and reasonable disbursements to be fixed by the Registrar.Solicitors:Russell McVeagh, Auckland for AppellantCrown Law Office, Wellington for Respondent67 Above at [61]–[63].68 Above at [50].