THE COMMISSIONER OF INLAND REVENUE V BNZ INVESTMENTS LIMITED AND ORS CA CA683/2008

THE COMMISSIONER OF INLAND REVENUE V BNZ INVESTMENTS LIMITED AND ORS CA CA683/2008 4 March 2009IN THE COURT OF APPEAL OF NEW ZEALAND CA683/2008 [2009] NZCA 47BETWEEN THE COMMISSIONER OF INLAND REVENUE Appellant AND BNZ INVESTMENTS LIMITED First Respondent AND BNZ INTERNATIONAL LIMITED Second Respondent AND BNZI SECURITIES NO.1 LIMITED Third Respondent AND BNZI SECURITIES NO.2 LIMITED Fourth Respondent AND BANK OF NEW ZEALAND Fifth Respondent AND BNZ CORPORATION LIMITED Sixth Respondent AND CUSTOM FLEET (NZ) LIMITED Seventh Respondent AND NATIONAL AUSTRALIA LIMITED Eighth Respondent AND NATIONAL NOMINEES LIMITED Ninth Respondent AND BNZ EQUIPMENT LIMITED Tenth Respondent AND NATIONAL AUSTRALIA GROUP (NZ) LIMITED Eleventh Respondent AND BNZ BRANCH PROPERTIES LIMITED Twelfth RespondentHearing: 28 January 2009 Court: Chambers, O'Regan and Robertson JJ Counsel: D J White QC and R J Ellis for Appellant A R Galbraith QC and J F Anderson for Respondents Judgment: 4 March 2009 at 11.30 amJUDGMENT OF THE COURT A The appeal is dismissed. B We direct that the deadline for the appellant to file and serve a witness statement in reply to those of Mr Simcock and Mr McLeod be deferred until those witness statements have been reviewed by the trial Judge or until the trial Judge rules otherwise. C The appellant must pay the respondent costs for a complex appeal on a Band A basis, including second counsel, reduced by 50 percent, plus usual disbursements. REASONS OF THE COURT(Given by O'Regan J)Introduction[1] This is an appeal by the Commissioner of Inland Revenue against a judgment of MacKenzie J in which he dismissed the Commissioner's interlocutory application for orders that two witness statements not be received as evidence in the High Court trial of the respondents' challenge to assessments made by the Commissioner: BNZ Investments Ltd v Commissioner of Inland Revenue (2008) 24 NZTC 23,059. [2] The proposed witness statements are by two tax experts, Mr D K Simcock, a tax lawyer who is a partner in Bell Gully, and Mr R A McLeod, who is an accountant (but also has a law degree) and is the managing partner of Ernst & Young New Zealand. [3] The Commissioner's assessments are made on the basis that certain transactions entered into by the respondents were part of arrangements constituting tax avoidance, and the issue at trial will be whether that contention is correct. The transactions and the broad issues between the parties were described by this Court in an earlier appeal, BNZ Investments Ltd v Commissioner of Inland Revenue [2008] 1 NZLR 598 at [7] – [9], and we will not repeat the description here.Issue on appeal[4] The issue on the appeal is whether the Judge ought to have ruled the proposed evidence of Mr Simcock and Mr McLeod inadmissible at the pre-trial stage. In addition, it is necessary for us to address whether the Court has jurisdiction to deal with the appeal and, if so, whether it should make definitive rulings in advance of the trial.The High Court decision[5] The Judge had before him a witness statement of Mr Simcock, but no witness statement for Mr McLeod was available. The impediment to Mr McLeod's availability to give evidence at that time has now been resolved and a witnessstatement was provided to us for the purposes of the appeal. In the High Court, MacKenzie J had a copy of Mr McLeod's witness statement for the trial of a similar claim, Deutsche Finance New Zealand Ltd v Commissioner of Inland Revenue(2008) 23 NZTC 21,758 (HC). It was broadly similar to the witness statement now proposed for the present litigation. [6] MacKenzie J is not the proposed trial Judge. Like us, therefore, he had limited background knowledge of the proceedings and the material on the High Court file, including the other witness statements filed by the parties. The trial Judge is Wild J. [7] The High Court decision in this case pre-dates the decisions of the Supreme Court in two appeals dealing with tax avoidance issues, Ben Nevis Forestry Ventures Ltd v Commissioner of Inland Revenue [2008] NZSC 115 and Glenharrow Holdings Ltd v Commissioner of Inland Revenue [2008] NZSC 116. Those decisions have clarified the test to be applied by the courts in dealing with the application of the general anti-avoidance provisions in tax legislation. [8] MacKenzie J dealt with the issues in three broad categories: relevance, the appropriateness of opinion evidence and the independence of the experts. [9] In relation to relevance, MacKenzie J: (a) Rejected the argument that the approach to relevance for the purposes of admissibility should be the same as the approach to relevance for the purposes of discovery; (b) Rejected the argument that evidence as to the scheme and purpose of, and background to, the relevant tax provisions was irrelevant because the parties agreed that the respondents' tax treatment of the transactions was consistent with the black letter law of those provisions;(c) Found that because the line drawing exercise required in tax avoidance cases involved more than just interpretation of the legislation, more extensive evidence on matters such as the scheme and purpose of the relevant provisions than would normally be admitted as an aid to interpretation would be potentially relevant; (d) Found that the jurisdiction to rule in advance of trial that evidence was inadmissible because it was irrelevant should be sparingly exercised and only in the very clearest of cases. In an interlocutory application, it was not appropriate to venture into the issues to the extent which would be necessary to make a ruling as to the issues in the proceeding and the relevance of the evidence to those issues. [10] In relation to opinion evidence, the Judge: (a) Repeated his observation that evidence as to the scheme and purpose of the relevant legislation for the purpose of the line drawing exercise required for tax avoidance cases potentially went far beyond material which could be relied on for an ordinary exercise of statutory interpretation. (b) Found that the way in which specific tax provisions may have influenced how transactions have been structured, and the understanding of the tax laws in the market, were potentially relevant, in a way which would not ordinarily be the case if the question were one of interpretation of the specific tax provisions involved. The purpose and effect of the transactions could not be properly assessed in a vacuum, divorced from their commercial context. (c) Found that if such matters were relevant, then it was appropriate that the evidence on them come from persons with experience and expertise in tax law and practice, and with knowledge of the relevant market circumstances.(d) Found that, in order to rule the opinion evidence inadmissible at the interlocutory stage, it would be necessary to reach the conclusion that there was no reasonable possibility that the evidence might be substantially helpful, in terms of s 25 of the Evidence Act 2006. That test was not met. [11] In relation to the independence of the experts, the Judge: (a) Observed that the qualification to give opinion evidence is expertise, rather than independence; (b) Found that the independence of an expert went to the credibility and weight of the expert's evidence, not to its admissibility; (c) Considered, therefore, that the involvement of Mr Simcock, Mr McLeod and their respective firms in representing banks and counter-parties in relation to the arrangements in issue or similar arrangements may affect the weight of their evidence, but the weighing exercise was one for the trial Judge and was not to be taken at the interlocutory stage.What the witnesses propose to say[12] The High Court Judge summarised the proposed evidence of Mr Simcock at [4] of his judgment in the following terms:Mr Simcock's statement of proposed evidence is described by counsel as purporting to provide independent expert opinion evidence on the evolution of New Zealand tax laws concerning the tax treatment of interest expenditure and dividend income. Mr Simcock summarises his understanding of the six transactions the subject of these proceedings, and his understanding of the New Zealand tax consequences of the transactions. He then addresses a number of basic general tax principles and the taxation of redeemable preference share financing. He suggests that his approach to the taxation of such financing is supported by a number of matters including statements by the Commissioner in policy statements, technical rulings, manuals and revenue alerts, the judgments in Commissioner of Inland Revenue v BNZ Investments Limited (2001) 20 NZTC 17,103, aspects of the legislative history of certain provisions in the Income Tax Act, the report of theconsultative committee on full imputation and the legislative purpose of the introduction of the conduit relief regime, various discussion documents and commentaries and his own interpretation. He reviews the New Zealand position on international tax issues and the legislative changes in respect of certain matters with reference to the work of the Policy Advice division of the IRD, a consultation paper and a Cabinet paper. He comments on the binding rulings system and taxpayers' use of it.[13] Mr McLeod's evidence is conveniently summarised in his witness statement as follows:3.7 In my evidence, I will address the Commissioner's allegations concerning the transactions (in his statement of defence), and the related opinions expressed by certain of the Commissioner's witnesses, which I have just summarised. I will explain that the characteristics they rely on to support the allegation of tax avoidance were present in redeemable preference share (or "RPS") transactions, that were commonplace in New Zealand throughout the late 1980s and early 1990s, and were accepted by Inland Revenue. I will explain that the two key tax attributes underlying the RPS transactions (receipt of tax free dividends, and deductibility of associated costs) are the same tax attributes that support the transactions in issue here, albeit that the type of relief in these transactions arises under either the foreign tax credit regime (in the case of one transaction) or the conduit regime (in the case of the other five transactions), rather than the inter-company dividend exemption, as was applicable to the earlier RPS transactions. 3.9 I will explain the background to the conduit regime, and also the measures accompanying it that limit deductibility of expenditure. In my opinion, this background is critical in order to evaluate the Commissioner's claim that the BNZ transactions involved "artificially" generated deductions, and were not "normal" commercial transactions. The difficulty with such claims is that much of our tax system is inherently artificial, and heavily influences commercial parties in the transactions they undertake. To properly and objectively evaluate the Commissioner's allegations of "artificiality" and absence of commerciality in this case requires an understanding of the relationship between tax policy and commercial behaviour. That is a significant focus of my evidence. 3.10 My evidence is organised under the following subject headings: (a) first, the concepts of "debt" and "equity". I provide an overview of these concepts, and then comment on the associated commercial and tax characteristics from the perspective of issuers and investors (section 4 of my evidence); (b) secondly, the history, in the New Zealand context, of redeemable preference share (or "RPS") financing, and thecommercial and tax characteristics of these transactions (section 5 of my evidence); (c) thirdly, the background to the introduction of the conduit tax relief regime (section 6 of my evidence); (d) fourthly, approaches New Zealand has taken to interest deductibility generally (section 7 of my evidence); (e) fifthly, I address the significance of a transaction having a "pre-tax negative" outcome, and explain how asymmetrical outcomes (whereby interest expenditure is deductible but the related income or gain is non-taxable) are unremarkable in the New Zealand tax system. I provide some common examples where that is the case (section 8 of my evidence); (f) seventhly, I comment on the binding rulings regime and taxpayer attitudes to it (section 9 of my evidence); and (g) finally, I summarise my opinions on these matters, and comment on aspects of the evidence to be given by the Commissioner's witnesses with which I disagree (section 10 of my evidence).[14] We will concentrate on the evidence of Mr McLeod, as opposed to Mr Simcock because Mr McLeod's evidence covers almost all the ground covered in Mr Simcock's witness statement. Also, because Mr McLeod's statement was filed later, it seeks to respond to the Commissioner's witnesses' statements. We pointed out to Mr Galbraith QC that Mr McLeod's witness statement contained a great deal of material which repeated what was in Mr Simcock's brief. He accepted this and said that it was possible only one of the two would actually be called to give evidence at the trial. We commend any effort to avoid repetitive evidence being given.Jurisdiction[15] The parties addressed us only briefly on the issue of jurisdiction. For the Commissioner, Mr White QC argued there was jurisdiction to hear and determine this appeal under s 66 of the Judicature Act 1908. He accepted the somewhat fluid boundary between rulings which are part of the trial conduct or management process (not normally susceptible to appeal) and those which have some substantive effect on rights and liabilities in issue (in respect of which appeals are permitted), as identified by this Court in Association of Dispensing Opticians of New Zealand Inc v OpticiansBoard [2000] 1 NZLR 158. He argued that the following features supported the proposition that there was jurisdiction in this case: (a) The issues of wider principles that are at stake; (b) The potential impact of the appeal on the scope, length and cost to the parties of the trial; (c) Material developments since the hearing in the High Court (particularly the delivery of the Supreme Court judgments in Ben Nevis and Glenharrow); (d) The principles underlying the new Evidence Act 2006. [16] For the respondents, Mr Galbraith accepted that there arguably was jurisdiction for the appeal, and did not take the matter further. Rather, he argued that the appropriate course was for the Court to dismiss the appeal without substantive determination, deferring the question of admissibility to trial. [17] In support of that submission, Mr Galbraith argued that there was still no clear definition of the issues at trial, the parties had differing interpretations of the Supreme Court judgment in Ben Nevis and the witness statements of Mr Simcock and Mr McLeod were two of a total of 45 witness statements. (Mr White clarified this: there are nine witnesses for the Commissioner and 22 (including those of Mr Simcock and Mr McLeod) for the respondent. The figure of 45 witness statements includes reply briefs). We do not have copies of these witness statements before us, and Mr Galbraith argued that this meant that we could not fairly consider the issues arising on the appeal, as the trial Judge will be able to do at trial. He pointed out that, if the Simcock/McLeod witness statements are ruled out, there would be no real possibility of challenging that finding, and if it ultimately turned out that the exclusion of the evidence was incorrect, that would occur after the trial and lead to the need for a rehearing.[18] Mr Galbraith also argued that the additional time and cost involved in dealing with the McLeod/Simcock evidence would not be significant in the overall context of the case, pointing out that the Commissioner had filed a reply brief in relation to Mr McLeod's evidence in the Deutsche proceedings which did not challenge the substance of the description by Mr McLeod of the applicable regimes. [19] We see considerable force in Mr Galbraith's submission that it is not appropriate for this Court to make a definitive determination on the issues at stake in this appeal. In particular: (a) The issues appear to us to be more nuanced than the Commissioner suggests, and not susceptible to an "all in or all out" solution. Mr White submitted that, if MacKenzie J had considered that not all of the proposed evidence was inadmissible, he should have embarked on a paragraph by paragraph evaluation and required that the briefs be edited accordingly. That may be correct, but it is not practical for this Court to engage in such an exercise on appeal. Rather, it is a task which is much better suited for the trial Judge, who will have a good understanding of the issues which will arise at the trial. (b) The argument in this Court about the relevance of some of the evidence was focused on the differing views as to the test to be applied in tax avoidance cases, as set down in Ben Nevis. Mr Galbraith argued that the Commissioner's summary of theBen Nevis test was not accurate. We have to say that the formulation set out in the judgment of Tipping, McGrath and Gault JJ in Ben Nevis does not appear to us to be susceptible to more than one interpretation, but we see some difficulties in our expressing a view on the way in which that test should be applied in the present litigation in advance of the trial, given that: (i) we do not have a full understanding of the issues which will arise at the trial or the evidence that will be presented;(ii) we have not heard full argument on the Ben Nevis test; and (iii) there is a high degree of likelihood that the outcome in the High Court will be subject to a subsequent appeal to this Court. (c) Mr Galbraith told us that at least some of the Simcock/McLeod evidence was necessary to reply to evidence in the witness statements of the Commissioner's witnesses. Without access to those witness statements, we cannot test that proposition in the way that the trial Judge will be able to test it. [20] The one issue on which we can express a concluded view is that relating to the independence of Mr McLeod and Mr Simcock. We will deal with that issue first. In relation to the issues arising from the Commissioner's other objections to the proposed evidence, we intend to set out indications of our views in the hope that these will provide some assistance to the trial Judge in the exercise of evaluating the admissibility of the witness statements.Independence[21] Mr White submitted that Mr Simcock and Mr McLeod were not sufficiently independent to give expert evidence at the trial. This was said to be because of the association that their respective firms had with the respondents (or associated entities) and with other banks which are parties to similar tax avoidance litigation. He argued that the High Court should have excluded evidence on this basis alone, and that it was appropriate for a decision of that kind to be made prior to the trial, relying on the English practice as outlined in R (Factortame Ltd) v Transport Secretary (No 8) [2003] QB 381 at 410 (CA). [22] In the New Zealand context, the question for decision is whether the proposed expert evidence will meet the tests set out in ss 7 and 25 of the Evidence Act. To the extent the Simcock/McLeod evidence is evidence of fact, it must be relevant, and to the extent it is expression of an opinion, it must be substantiallyhelpful to the fact finder. We accept that there may be cases in which the position of the proposed expert is so lacking in independence as to make it obvious that an opinion he or she expresses in evidence will not be able to be substantially helpful, and in those circumstances it may be appropriate to rule out the evidence at the pre- trial stage in order to avoid the costs which may otherwise be incurred in responding to it. [23] However, we do not see this case as being in that category. Rather, we see the position of Mr Simcock and Mr McLeod as advisers to (or as members of firms which have provided advice to) taxpayers facing similar litigation as a factor affecting the weight to be attributed to their evidence (to the extent it is admissible). This assessment is properly to be made by the trial Judge, in his role as fact finder. [24] Rule 9.43(2) of the recently enacted High Court Rules requires expert witnesses to comply with the code of conduct set out in Schedule 4. That code states that expert witnesses have "an overriding duty to assist the court impartially" and are not to act as advocates for the party that has engaged them. We do not see anything in this case which presently leads us to conclude that Mr Simcock and Mr McLeod will not conduct themselves in accordance with these requirements. We accept that much of their evidence could fairly be classified as legal submission, which arguably means they are acting as advocates for the respondents (and are not complying with the code of conduct), but we see the solution to that as being the removal of the material which is, in essence, submission, rather than the disqualification of them as expert witnesses. [25] Overall, we agree with the way in which MacKenzie J dealt with this aspect of the case. Like him, we see the independence issue in the present case as one which goes to the weight to be attributed to their evidence at trial, rather than as a basis for excluding it altogether at the pre-trial stage.Comments on the Commissioner's objections to the proposed evidence[26] We now comment briefly on the objections raised by the Commissioner.Evidence of legal propositions[27] Mr White argued that much of the proposed evidence amounts to legal propositions and legal "facts" that are not susceptible of proof. In addition, he pointed to the paragraphs in Mr McLeod's evidence dealing with the creation of the controlled foreign company (CFC) and foreign investment funds (FIF) regimes and the subsequent development of the conduit tax relief regime in which Mr McLeod had a significant personal involvement. Mr White suggested that this went beyond the permissible scope of extrinsic materials available for the purposes of statutory interpretation. [28] We agree that both witness statements contain much in the nature of submissions on legal issues. The difficulty which this poses is that it invites a response from the Commissioner from another tax expert essentially providing counter submissions. We do not see it as helpful to the Court to have the roles of counsel and expert witnesses intermingled in this way. [29] Submissions on the interpretation of the black letter provisions (to the extent this is relevant, given the consensus in the present case that the respondents complied with the black letter provisions) can be informed by submissions dealing with, and referring to, discussion papers which were publicly circulated, select committee consideration of legislation and the like. Mr Galbraith pointed us to the discussion of the permissible limits in this area in Burrows Statute Law in New Zealand (3ed 2003), but there is nothing in that discussion to indicate that any of the approach he described involves the giving of evidence by experts in the field, let alone evidence by participants in the law reform process as to their actual role within it. [30] Mr Galbraith argued that Mr McLeod's evidence was akin to referring the Court to a Law Commission report, or a discussion paper prepared for the Law Commission in the context of a law reform proposal. We disagree. It is more akin to calling a member of the Law Commission, or the writer of a report to the Law Commission, to give evidence as to what was intended in relation to a particular reform proposal. That would invite a response from the Commissioner calling astatutory drafter, or a departmental official who advised the select committee, which is clearly well outside the bounds of what is permitted in this area.Opinion evidence[31] Mr White's second point was that, to the extent the evidence constitutes opinion evidence, it does not meet the threshold for admissibility provided by s 25 of the Evidence Act. [32] Mr White referred us to the comment by Simon France J in Commissioner of Inland Revenue v Rabobank New Zealand Ltd HC WN CIV-2006-485-2907 28 November 2008:How a commentator thinks the case should be decided seems to me to be a classic example of irrelevant evidence. I consider there are considerable dangers, and potentially a great deal of wasted cost, in accepting the proposition that it is open to a party to file opinion evidence on the domestic law applicable to a case let alone how that law should be applied to the case.[33] We agree with that statement.Prolonging the proceedings[34] Mr White argued that the High Court Judge ought to have considered the evidence on a paragraph by paragraph basis and struck out the elements that were inadmissible on the grounds that their probative value was outweighed by the risk that they would needlessly prolong the proceedings: s 8 of the Evidence Act. He pointed to a number of cases where evidence had been admitted in tax cases but was later found to have been of no or very limited relevance, and suggested that this indicated that it was better to take a firm line at the pre-trial stage, rather than allowing the evidence to be admitted and then determining that it is not relevant. [35] As we indicated earlier, it is not feasible for us to do a paragraph by paragraph analysis at the appeal stage, and Mr White did not ask us to do so. In addition, as also mentioned earlier, the relevance assessment which the Judge can make at trial will be much better informed than any assessment we could make now.Irrelevant material[36] Mr White argued that much of the material contained in the briefs was irrelevant and should be excluded on the basis that it does not meet the primary test for admissibility in s 7 of the Evidence Act. In this category he referred to the evidence which both witnesses proposed to give about the rulings system operated by the Inland Revenue Department. He argued this evidence was irrelevant, given that the respondents' pleading based on the Commissioner's favourable rulings for other, similar, transactions had been struck out in BNZ Investments Ltd v Commissioner of Inland Revenue (No 5) (2008) 23 NZTC 21,821 (HC) and the respondents had not appealed. In addition, he pointed out that rulings made by the Commissioner in respect of other taxpayers for similar transactions had been found not to be discoverable in the present proceeding in Commissioner of Inland Revenue v BNZ Investments Ltd (2008) 23 NZTC 21,992 (CA). [37] Mr Galbraith suggested that there remains an issue of relevance in relation to the rulings system, which meant that the proposed evidence remained relevant. He said the evidence showed that rulings were relied on by tax practitioners and commercial parties in relation to transactions that were not covered by the ruling. That relevance seemed to be a reply to an anticipated submission by the Commissioner that the fact that the respondents did not get a ruling in relation to the transactions in issue in this litigation could be held against them. We cannot see how evidence about rulings and how the BNZ or others relied on rulings, even for transactions not covered by them can inform the Court on the core issue, which is whether the arrangements, of which the transactions at issue in this case formed a part, constituted tax avoidance. But again, we leave the final decision on that to the trial Judge, who will have a better appreciation of the potential validity of the point made by Mr Galbraith.Scheme and purpose analysis[38] Mr White argued that the High Court decision was based, at least in part, on the Judge's view that the analysis of the scheme and purpose of the black letterprovisions (to determine whether they had been used in a manner which was beyond the contemplation of Parliament) involved a broader inquiry than that undertaken in other exercises of statutory interpretation. He said that such an inquiry was contrary to the position outlined by Richardson J in Challenge Corporation Ltd v CIR [1986] 2 NZLR 513 at 549 (CA) that the reconciliation of the different and conflicting objectives of the black letter provisions and the anti-avoidance provision is "a matter of statutory construction" that requires a purposive interpretation based on "the scheme of the legislation and the relevant objectives of the legislation". He also pointed to the decision of the Supreme Court in Ben Nevis in support of that submission, particularly the statement at [104] in the judgment of Tipping, McGrath and Gault JJ that ascertaining when an arrangement crosses the line between the legitimate and the illegitimate "should be firmly grounded in the statutory language of the provisions themselves". [39] That submission was mainly aimed at the evidence proposed to be given by Mr Simcock and Mr McLeod on the interaction between changes in tax law and the resulting commercial arrangements that are structured to meet the requirements of the new black letter provisions. [40] We do not consider that there is anything objectionable about the witnesses giving evidence about the nature of the transactions being undertaken in the market in the light of the tax laws then in place, subject only to that being relevant to the present proceeding. The general tenor of the evidence proposed to be given is that redeemable preference share deals have been a common feature in New Zealand financing markets, but have evolved over time as different black letter provisions have been enacted, creating different taxation advantages in the use of redeemable preference shares for what are essentially financing transactions. [41] If the connection between redeemable preference share transactions structured to meet the requirements of historical taxation provisions is relevant to the assessment of whether or not there is tax avoidance in the present case, then we can see no other basis for excluding that evidence. The relevance assessment is best left to the trial Judge. In that regard, we note the observation in Ben Nevis at [109] that, in considering whether the use of black letter provisions is outside the contemplationof Parliament in relation to those provisions and therefore crosses the line into tax avoidance: the Courts are not limited to purely legal considerations. They should also consider the use made of the specific provision in the light of the commercial reality and the economic effect of that use. The ultimate question is whether the impugned arrangement, viewed in a commercially and economically realistic way, makes use of the specific provision in a manner that is consistent with Parliament's purpose.Other matters[42] Mr McLeod's witness statement was prepared after the witness statements of the Commissioner's witnesses had been provided to the respondents and in places his evidence responds to the evidence proposed to be given by the Commissioner's witnesses. We were told that a number of reply briefs have been filed, from which we infer that there has been a direction permitting or requiring the filing of reply briefs in this proceeding. What we say about reply evidence proceeds on that basis. [43] We have not seen what the Commissioner's witnesses intend to say and we therefore do not have any basis for evaluating this reply evidence. However, there is obviously no reason in principle why the respondents should not be entitled to call expert evidence responding to expert evidence given on the Commissioner's behalf, and contesting its correctness. There is also no reason for this evidence to be excluded from a witness statement if reply briefs from other witnesses have been filed. [44] Mr McLeod also has a section in his witness statement which is designed to show that it is not unusual in commercial transactions for a transaction to have a negative pre-tax cash flow but a positive post-tax cash flow. We were told that this is intended to respond to evidence to be given on behalf of the Commissioner that transactions having pre-tax negative cash flows are indicative of tax avoidance. If that is what the Commissioner's witness says, then we can see no reason in principle to exclude evidence contesting that proposition.Conclusion[45] We find that there is no basis for excluding the witness statements on the basis of lack of independence. We decline to rule on the admissibility of the statements, or particular paragraphs of the statements, at this stage in the proceeding. We agree with MacKenzie J that such rulings on admissibility are better made at trial, where the Judge has an understanding of the nature and scope of the issues in the litigation and the evidence tendered by both parties. We uphold the position of MacKenzie J to defer the admissibility assessment until the trial and formally dismiss the appeal. However, given the likelihood that some of the material in the current witness statements will need to be edited out, we direct that the deadline for the Commissioner to file and serve a witness statement in reply to those of Mr Simcock and Mr McLeod be deferred until they have been reviewed by the trial Judge or until the trial Judge rules otherwise.Costs[46] The respondents are entitled to costs, but at a reduced rate to reflect that the Commissioner has had a measure of success even though the appeal has been formally dismissed. We have endorsed many of his criticisms of the respondents' briefs. A 50 percent reduction is appropriate. We classify the appeal as a complex appeal, given the undoubted complexity of the litigation of which this appeal is a component. We certify for second counsel.Solicitors: Crown Law Office, Wellington for Appellant Russell McVeagh, Wellington for Respondents