Ball UK Holdings Ltd v The Commissioners for HM Revenue and Customs: [2018] UKUT 0407 (TCC) [2018] UKUT 0407 (TCC)

UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
[2018] UKUT 0407 (TCC)Case No UT/2017/0160
Between
Ball UK Holdings LtdAppellantThe Commissioners for HM Revenue and CustomsRespondent
Before
MRS JUSTICE FALKJUDGE JONATHAN CANNANNicola Shaw QC (instructed by Hogan Lovells International LLP) for AppellantJames Henderson, counsel (instructed by the General Counsel and Solicitor to HM Revenue and Customs) for RespondentsDate 10 December 2018Category: Tax
[1]This appeal relates to a short, but important, point of law. It raises the question of the proper role of the Tribunal in disputes over the correct interpretation of accounting standards.[2]The background facts can be stated very briefly. The appellant Ball UK Holdings Ltd (“Ball UK”) is a UK intermediate holding company indirectly owned by a US corporation, Ball Corporation (“Ball US”). It was set up in 2002 to hold certain UK subsidiaries acquired by Ball US as part of an acquisition of another group. Ball UK’s material activities were limited to holding the shares in its subsidiaries, and borrowing from and making loans to group companies, including its subsidiaries. In 2005, prompted by changes in US tax law, it also paid a large dividend to its parent. At material times Ball UK undertook all its activities (and in particular its lending and borrowing activities) in sterling, with the exception of the derivative mentioned below. The interest rates on its borrowing and lending transactions were determined by reference to sterling LIBOR or other UK prime rates.[3]For periods up to 31 December 2005 Ball UK’s accounts were prepared in sterling in accordance with Statement of Standard Accounting Practice 20 (“SSAP 20”). Near the end of 2006, on the advice of PwC, a transaction was entered into with the intention of changing the “functional currency” used in preparing the accounts from sterling to US dollars, by triggering the right to move from SSAP 20 to Financial Reporting Standard 23 (“FRS 23”). Like SSAP 20, FRS 23 is a UK accounting standard, but it adopts the 2003 version of International Accounting Standard 21 (“IAS 21”).[4]Under the transaction Ball UK entered into a derivative contract with a group member, worth about $30,000. This gave Ball UK the ability to choose to apply fair value accounting, which it did. The exercise of that choice in turn triggered a requirement to comply with FRS 26 and also with FRS 23, which determines functional currency. The derivative itself was not relevant to the determination of functional currency.[5]Based on PwC’s advice, the Ball group concluded that the effect of FRS 23 was to change the functional currency of Ball UK to US dollars. The result of this was to trigger a foreign exchange (“FX”) loss in its accounts for the year ended 31 December 2006. This arose because its assets and liabilities were revalued in the new currency. The accounts were audited by PwC, who confirmed that in their view they gave a true and fair view in accordance with UK generally accepted accounting principles (“GAAP”).[6]The loss shown in Ball UK’s accounts included a significant loss on its sterling denominated borrowings. Ball UK sought tax relief for this loss for corporation tax purposes under the “loan relationship” rules contained in what was then Chapter 2 of Part 4 of the Finance Act 1996 (“FA 1996”). The loss claimed for tax purposes was about £24.6m. 3[7]In summary, the basis on which the Ball group concluded that a dollar functional currency was appropriate was because all material decisions in relation to Ball UK were taken by employees of Ball US. Ball UK’s board had no real input into its decision-making process and was largely passive, confining themselves to checking that what they were being asked to do was in accordance with applicable law, and signing off on pre-typed resolutions. The First-tier Tribunal (“FTT”) found at paragraph 37 that although decisions made were legally resolved upon by Ball UK’s board, they were in fact decided by persons acting on behalf of its ultimate parent, Ball US.[8]HMRC disagreed with the conclusion that Ball UK had a dollar functional currency. HMRC’s case is that Ball UK’s functional currency was sterling and that the accounts should have been prepared on that basis.[9]The only question in dispute is whether Ball UK’s accounts were prepared in accordance with UK GAAP for the purposes of what was then s 85A Finance Act 1996. This provided:
“(1) Subject to the provisions of this Chapter…, the amounts to be brought into account by a company for any period for the purposes of this Chapter are those that, in accordance with generally accepted accounting practice, are recognised in determining the company's profit or loss for the period.”
FRS 23[10]It is convenient to set out here relevant extracts from FRS 23. These comprise four introductory (unnumbered) paragraphs, extracts from the introduction to IAS 21 (set out in full in FRS 23, numbered with the prefix “IN”), paragraphs 1, 2, 8 to 13 and 17 from the standard itself, and certain paragraphs from the “Basis of Conclusions” that accompanied IAS 21. “FRS 23 (IAS 21) The effects of Changes in Foreign Exchange Rates Financial Reporting Standard 23 embodies IAS 21 “The effects of Changes in Foreign Exchange Rates” and some amendments to that standard adopted for entities subject to UK accounting standards. The Statement of Standard Accounting Practice in FRS 23 is set out in paragraphs 1-62 and the appendix. All the paragraphs have equal authority. Paragraphs in bold type state the main principles. Accompanying the Statement of Standard Accounting Practice is the basis for the conclusions reached in the Statement. This does not form part of the Statement. The Statement of Standard Accounting Practice should be read in the context of its objective as stated in paragraphs 1-2, the Basis for Conclusions set out in paragraphs BC1-BC32, and the Accounting 4 Standards Board's1 “Foreword to Accounting Standards” and “Statement of Principles for Financial Reporting”. …. Introduction … IN3 For IAS 21 the Board’s2 main objective was to provide additional guidance on the translation method and on determining the functional and presentation currencies. The Board did not reconsider the fundamental approach to accounting for the effects of changes in foreign exchange rates contained in IAS 21. The Main Changes IN4 The main changes from the previous version of IAS 21 are described below. … IN6 The notion of “reporting currency” has been replaced with two notions:• functional currency, i.e. the currency of the primary economic environment in which the entity operates. The term “functional currency” is used in place of “measurement currency” (the term used in SIC-193) because it is the more commonly used term, but with essentially the same meaning.• presentation currency, i.e. the currency in which financial statements are presented. IN7 When a reporting entity prepares financial statements, the Standard requires each individual entity included in the reporting entity – whether it is a stand-alone entity, an entity with foreign operations (such as a parent) or a foreign operation (such as a subsidiary or branch) – to determine its functional currency and measure its results and financial position in that currency. The new material on functional currency incorporates some of the guidance previously included in SIC-19 on how to determine a measurement currency. However, the Standard gives greater emphasis than SIC-19 gave to the currency of the economy that determines the pricing of transactions, as opposed to the currency in which transactions are denominated. IN8 As a result of these changes and the incorporation of guidance previously in SIC-19…an entity (whether a stand-alone entity or a foreign operation) does not have a free choice of functional currency… 1 The “ASB”. 2 The International Accounting Standards Board or IASB. 3 An interpretation document published by the Standards Interpretation Committee of the IASB entitled “Reporting Currency – Measurement and Presentation of Financial Statements under IAS 21 and IAS 29”. 5 IN9 The Standard revises the requirements in the previous version of IAS 21 for distinguishing between foreign operations that are integral to the operations of the reporting entity (referred to below as “integral foreign operations”) and foreign entities. The requirements are now among the indicators of an entity’s functional currency. As a result…there is no distinction between integral foreign operations and foreign entities. Rather, an entity that was previously classified as an integral foreign operation will have the same functional currency as the reporting entity… Financial Reporting Standard 23 (IAS 21) The Effects of Changes in Foreign Exchange Rates Objective 1 An entity may carry on foreign activities in two ways. It may have transactions in foreign currencies or it may have foreign operations. In addition, an entity may present its financial statements in a foreign currency. The objective of this Standard is to prescribe how to include foreign currency transactions and foreign operations in the financial statements of an entity and how to translate financial statements into a presentation currency. 2 The principal issues are which exchange rate(s) to use and how to report the effects of changes in exchange rates in the financial statements. … Definitions 8 The following terms are used in this Standard with the meanings specified: …. Foreign currency is a currency other than the functional currency of the entity. Foreign operation is an entity that is a subsidiary, associate, joint venture or branch of a reporting entity, the activities of which are based or conducted in a country or currency other than those of the reporting entity. Functional currency is the currency of the primary economic environment in which the entity operates. … Presentation currency is the currency in which the financial statements are presented. Elaboration on the Definitions Functional Currency 9 The primary economic environment in which an entity operates is normally the one in which it primarily generates and expends cash. An entity considers the following factors in determining its functional currency: 6(a) the currency: (i) that mainly influences sales prices for goods and service (this will often be the currency in which sales prices for its goods and services are denominated and settled); and (ii) of the country whose competitive forces and regulations mainly determines the sales prices of its goods and services.(b) the currency that mainly influences labour, material and other costs of providing goods or services (this will often be the currency in which such costs are denominated and settled). 10 The following factors may also provide evidence of an entity's functional currency: (a) the currency in which funds from financing activities (i.e. issuing debt and equity instruments) are generated. (b) the currency in which receipts from operating activities are usually retained. 11 The following additional factors are considered in determining the functional currency of a foreign operation, and whether its functional currency is the same as that of the reporting entity (the reporting entity, in this context, being the entity that has the foreign operation as its subsidiary, branch, associate or joint venture): (a) whether the activities of the foreign operation are carried out as an extension of the reporting entity, rather than being carried out with a significant degree of autonomy. An example of the former is when the foreign operation only sells goods imported from the reporting entity and remits the proceeds to it. An example of the latter is when the operation accumulates cash and other monetary items, incurs expenses, generates income and arranges borrowings, all substantially in its local currency. (b) whether transactions with the reporting entity are a high or a low proportion of the foreign operation's activities.(c) whether cash flows from the activities of the foreign operation directly affect the cash flows of the reporting entity and are readily available for remittance to it.(d) whether cash flows from the activities of the foreign operation are sufficient to service existing and normally expected debt obligations without funds being made available by the reporting entity. 12 When the above indicators are mixed and the functional currency is not obvious, management uses its judgement to determine the functional currency that most faithfully represents the economic effects of the underlying transactions, events and conditions. As part of this approach, management gives priority to the primary indicators in paragraph 9 before considering the indicators in paragraphs 10 and 11, which are designed to provide additional supporting evidence to determine an entity's functional currency. 13 An entity's functional currency reflects the underlying transactions, events and conditions that are relevant to it. Accordingly, once 7 determined, the functional currency is not changed unless there is a change in those underlying transactions, events and conditions. 14 [Not relevant.] … Summary of the approach required by this standard 17 In preparing financial statements, each entity – whether a stand-alone entity, an entity with foreign operations (such as a parent) or a foreign operation (such as a subsidiary or branch) – determines its functional currency in accordance with paragraphs 9-14… … Basis for conclusions … This Basis for Conclusions accompanies, but is not part of, IAS 21. Introduction BC1 This Basis for Conclusions summarises the International Accounting Standards Board's considerations in reaching its conclusions on revising IAS 21... … BC3 Because the Board's intention was not to reconsider the fundamental approach to accounting for the effects of changes in foreign exchange rates established by IAS 21, this Basis for Conclusions does not discuss requirements in IAS 21 that the Board has not reconsidered. … BC5 The Board noted a concern that the guidance in SIC-19 on determining a measurement currency could permit entities to choose one of several currencies, or to select an inappropriate currency. In particular, some believed that SIC-19 placed too much emphasis on the currency in which transactions are denominated and too little emphasis on the underlying economy that determines the pricing of those transactions. To meet these concerns, the Board defined functional currency as “the currency of the primary economic environment in which the entity operates”. The Board also provided guidance on how to determine the functional currency (see paragraphs 9-14 of the Standard). This guidance draws heavily on SIC-19 and equivalent guidance in UK and other national standards, but also reflects the Board's decision that some factors merit greater emphasis than others. BC6 The Board also discussed whether a foreign operation that is integral to the reporting entity (as described in the previous version of IAS 21) could have a functional currency that is different from that of its “parent”. The Board decided that the functional currencies will always be the same, because it would be contradictory for an integral foreign operation that “carries on business as if it were an extension of the reporting enterprise's operations”[†] to operate in a primary economic environment different from its parent. 8 † IAS 21 (revised 1993) paragraph 24 … BC7 It follows that it is not necessary to translate the results and financial position of an integral foreign operation when incorporating them into the financial statements of the parent – they will already be measured in the parent's functional currency. Furthermore, it is not necessary to distinguish between an integral foreign operation and a foreign entity. When a foreign operation's functional currency is different from that of its parent, it is a foreign entity… BC8 The Board also decided that the principles in the previous version of IAS 21 for distinguishing an integral foreign operation from a foreign entity are relevant in determining an operation's functional currency. Hence it incorporated these principles into the Standard in that context. BC9 The Board agreed that the indicators in paragraph 9 are the primary indicators for determining the functional currency and that paragraphs 10 and 11 are secondary. This is because the indicators in paragraphs 10 and 11 are not linked to the primary economic environment in which the entity operates but provide additional supporting evidence to determine an entity's functional currency.” The FTT’s decision[11]The FTT heard expert evidence from three witnesses, Mr Chandler and Mr Kraehnke (both partners in KPMG LLP and called by the appellant) and Mr Chopping, a partner in Moore Stephens LLP, who was called by HMRC. Mr Chandler and Mr Chopping provided expert evidence on UK GAAP and IFRS (International Financial Reporting Standards). Mr Kraehnke’s expertise was in US GAAP and US accounting standards (“FAS”), together with IFRS. There was also factual evidence from four Ball group executives, but the real dispute was between the experts.[12]The FTT first considered whether the correct question was whether Ball UK had proved that it had correctly applied FRS 23, or whether in the light of Versteegh Ltd & others v Revenue & Customs Commissioners [2013] UKFTT 642 (TC) it simply had to prove that it had applied a reasonable interpretation of it. The FTT concluded that the former was correct and that there must be a reasonable application of a correctly interpreted standard (paragraphs 13 and 16 of the decision). The FTT did however accept at paragraph 17 that FRS 23 should be interpreted as it would reasonably be understood by accounting professionals, and not as if it was a statute.[13]The FTT considered the terms of FRS 23 and accompanying material, and the expert evidence, in detail. It concluded that paragraph 8 indicated that the test was about the company’s economic environment (paragraph 61 of the decision) and that the first sentence of paragraph 9 was relevant, rather than irrelevant as Ball UK had contended (paragraph 63). The FTT also concluded (and this was not disputed between the experts) that each of paragraphs 10(a) and 11(b) to (d) indicated a sterling functional currency. 9[14]A key focus of the decision, and a primary area of dispute, was about the meaning of the test in paragraph 11(a) and, in particular, whether the concept of autonomy should be interpreted as referring only to control over decision-making, or whether it had a wider meaning of “economic or functional independence” from the parent. The FTT agreed with Mr Chopping that it meant the latter, and that applying that meaning the paragraph indicated that Ball UK’s functional currency was sterling (paragraphs 64 to 81). The FTT also rejected an argument that paragraph 11(a) had a “trumping” effect over paragraphs 10(a) or 11(b) to (d), such that if paragraph 11(a) indicated a dollar functional currency that would outweigh the fact that those other paragraphs indicated sterling (paragraphs 82 to 102).[15]The FTT also considered the “spirit and purpose” of FRS 23. It found that the functional currency was intended to be a currency which minimised the FX risk in the presentation of the accounts, by being the currency to which the company had greatest economic exposure. Ball US was not exposed to FX risk on Ball UK’s sterling investments and liabilities, but only on its net investment. Mr Chopping’s interpretation was most in keeping with the spirit and purpose because it reduced the risk of the entity’s financial position being obscured by currency movements (paragraphs 105 to 115).[16]The FTT went on to consider the Basis of Conclusions, and agreed with HMRC that they supported HMRC’s case. In the FTT’s view they supported the conclusion that control of decision-making was not relevant. The cross reference back in BC6 to the 1993 version of IAS 21 was said to support this. The reference in BC9 to the indicators in paragraphs 10 and 11 not being linked to the primary economic environment was dismissed as a badly worded explanation of the first sentence of BC9, because the definition of functional currency was the currency of the primary economic environment in which the entity operates (paragraphs 116 to 132).[17]The FTT also considered certain introductory paragraphs to IAS 21 (which are also incorporated in FRS 23), the Accounting Standard Board’s notes, and aspects of SSAP 20 and the 1993 version of IAS 21. It concluded that autonomy was not used in the 1993 version in the sense of control of decision-making, and that there was no fundamental change between the 1993 version and the 2003 version that was the basis of FRS 23 (paragraphs 133 to 158). Any major change would have been identified in the notes to the new version. The FTT also considered guidance on FRS 23 provided by three of the “Big 4” accountancy firms (Ernst & Young, PwC and Deloitte) and by HMRC, and found that none of it supported Ball UK’s case (paragraphs 159 to 205 and 209 to 216).[18]In conclusion, the FTT decided that Mr Chandler’s approach to FRS 23 was flawed, elevating what was admittedly the main dictionary definition of autonomy to make it a paramount test, taking the term out of context and ignoring its antonym in paragraph 11(a). The main principle was the identification of the primary economic environment. In the FTT’s view Mr Chandler’s approach was not only wrong but was not even a reasonable or possible interpretation of FRS 23 (paragraphs 223 to 231). 10 Grounds of appeal[19]Ball UK contended that there were four errors of law in the FTT’s decision:(1) the FTT incorrectly held that there can only be one permissible interpretation of an accounting standard;(2) it misinterpreted the word “autonomy” in paragraph 11(a) of FRS 23;(3) it incorrectly held that paragraph 11(a) did not outweigh the other indicators in FRS 23 on Ball UK’s facts; and(4) it failed to have proper regard to the evidence as to the interpretation of FRS 23 adopted by accounting professionals in practice. The parties’ submissions Ball UK’s submissions[20]Ms Shaw’s primary submission on behalf of Ball UK was that the FTT went wrong in its approach to the interpretation of FRS 23. The question of its correct interpretation was one of law, and the approach to take was to consider how the words would be understood by the reasonable reader, in this context the reasonable accounting professional. The concept of primary economic environment was not a free-standing concept defined in paragraph 8 of FRS 23, but was established by applying the indicators set out at paragraphs 9 to 14. The FTT’s erroneous approach to paragraph 8 infected its interpretation of the other paragraphs. Paragraph 9 applied only to operating companies, and paragraph 10(a) was of reduced weight in the case of a company whose activities are carried on as an extension of its parent. In concluding that autonomy meant economic or functional independence rather than encompassing its main meaning (the ability to make decisions), the FTT relied on its erroneous approach to paragraph 8 and also failed to identify that the examples in paragraph 11(a) were simply non-exhaustive illustrations rather than explanations of what was said, and were not inconsistent with the main meaning. The examples also focused on operating companies.[21]The FTT incorrectly identified the purpose of FRS 23 as being to minimise FX risk by presenting accounts in the currency to which the company had greatest economic exposure. There was no basis for that conclusion. The fact that Ball US was not exposed to FX risk on Ball UK’s investments was immaterial. The FTT also failed to appreciate the context provided by BC9, the terms of which undermined the FTT’s conclusion. Both BC5 and BC9 indicated that some factors have greater significance than others. IN7 indicated a shift of focus from the currency in which transactions are denominated to the currency of the economy that determines their pricing. The FTT had also relied on notes in the Standard that concerned the preparation of consolidated accounts, and which were irrelevant to the interpretation and application of paragraphs 8 to 14. The 1993 version of IAS 21 could not shed proper light on FRS 23, which incorporated a later version of IAS 21, and the cross-references in BC3 and BC6 did not justify that. 11[22]Ms Shaw also submitted (in line with the first ground of appeal) that there was no reason why the wording of a particular accounting standard could not encompass two possible meanings, and in that case the question was whether the interpretation applied was a reasonable one. There was no real difference between the possibility of different accounting professionals reaching different conclusions by means of a different application of accounting standards, and doing so by applying different interpretations.[23]Ms Shaw submitted that the FTT erred in dismissing Mr Chandler’s approach that the indicators in paragraph 11(b) to (d) and 10(a) carried different weighting where an entity had no autonomy of decision-making, because the outcome of applying those other indicators was dictated by the parent. The FTT also wrongly considered that paragraphs 12 and 13 did not allow autonomy of decision-making to be taken into account.[24]It was appropriate to take account of the guidance in the “Big 4” manuals, and the FTT had failed to appreciate their significance in showing that the fact that an intermediate holding company is not autonomous (in a decision-making sense) will carry significant weight. Submissions for HMRC[25]Mr Henderson, for HMRC, submitted that the FTT heard detailed expert evidence about the correct interpretation and application of FRS 23. Interpreting a document was not always a question of law, and the case law showed that the Tribunal should only interfere with the FTT’s findings about the accounting treatment in the event that it was convinced that the FTT had reached conclusions that were not open to it. It was wrong to suggest that the meaning of the relevant parts of FRS 23 was simply a question of law on which the Tribunal could rule afresh.[26]The test was not how the word autonomy would be understood by the reasonable accounting professional, or whether the interpretation adopted was a reasonable one, but simply what was the correct meaning of FRS 23 and how it should be applied. There could be more than one possible application in a borderline case, but only one correct interpretation. This was also not a borderline case. The FTT took the correct approach to paragraphs 8 and 9, which made no reference to the location of the decision-makers and indicated that FRS 23 is concerned with the economic circumstances of the company. The same applied to paragraph 10. Paragraph 11(a) (and the definition of foreign operation in paragraph 8) focus only on the activities of the foreign operation, and the examples supported the FTT’s interpretation. The FTT also took the correct approach to paragraphs 12 and 13.[27]The question in this case was whether Ball UK’s accounts were in accordance with UK GAAP. The overriding objective of financial statements is to give a true and fair view of an entity’s state of affairs in a way that reflects economic reality, and the determination of functional currency is fundamental to the objectives of FRS 23. The FTT correctly considered FRS 23 as a whole in the light of its objective and its “spirit and reasoning” (which the Foreword to Accounting Standards made clear that it was 12 necessary to consider). Ball UK’s 2006 accounts did not achieve this overriding objective, and the FTT understood this. Ball UK was exposed solely to the UK economy and there was no justification for identifying US dollars as its functional currency.[28]The FTT’s conclusions were also consistent with the background material, including the Basis of Conclusions. BC6 was particularly important. The 1993 version of IAS 21 was also relevant because it showed the origins of paragraph 11(a) and made it clear that what it was concerned with was functional inter-dependence. The distinction between a foreign operation which was or was not a functional extension of its parent was carried through to FRS 23. The FTT also correctly identified that there was nothing in the introduction to FRS 23 or the ASB notes indicating that it was intended to introduce a new test of strategic decision-making.[29]The FTT also correctly dismissed Mr Chandler’s “trumping” approach to paragraph 11(a), and correctly concluded that the “Big 4” accounting manuals could not override FRS 23. In any event they did not support Mr Chandler’s approach. Discussion Interpretation of accounting standards: law or fact?[30]At the heart of this appeal is the question whether the correct interpretation of accounting standards, and specifically FRS 23, can be characterised as a matter of law or fact. If it is a question of law then this Tribunal may consider the question afresh. If it is a question of fact then, since an appeal may only be made on a point of law4, the role of this Tribunal is limited to determining whether the FTT’s conclusions were unsupported (or not sufficiently supported) by the evidence, such that the findings were not ones that it was entitled to make. See Georgiou (t/a Marios Chippery) v Customs and Excise Commissioners [1996] S.T.C. 463 at 476, discussing the principles established by Edwards v Bairstow [1956] AC 14.[31]Somewhat surprisingly, there is virtually no direct authority on this point. Ms Shaw submitted that the correct interpretation of accounting standards was a question of law. She relied by analogy on cases in other areas, in particular First Secretary of State v Sainsbury’s Supermarkets Ltd [2005] EWCA Civ 520 at [16] (which related to the interpretation of a planning policy in respect of out-of-town supermarket developments), R (oao Davies and others) v HMRC [2011] UKSC 47 and R (oao Veolia ES Landfill) v. HMRC [2016] EWHC 1880 (Admin) at [110] and [111] (both of which were judicial review applications considering the interpretation of Revenue publications). In Veolia, for example, Mr Justice Nugee applied the test of construing the document objectively, asking how it would have been understood by a reasonable reader of it in the industry to which it was directed, taking account of the background context reasonably available to those in that industry. He commented as follows at [111]: 4 Section 11(1) Tribunals, Courts & Enforcement Act 2007. 13 “Questions of construction of a document are as a matter of general law regarded as questions of law for the Court, not questions of fact to be determined on evidence as to what the document was intended by the draftsman, or understood, to mean..”[32]Ms Shaw pointed out that Nugee J’s approach chimed with comments made about the approach to the interpretation of contracts by Lord Lloyd in Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896 at page 902 and Lord Hoffmann at pages 912 to 913, where Lord Hoffmann referred to the meaning which the document would convey to a reasonable person having all the background knowledge which would reasonably have been available. Here the reasonable reader was an accountant.[33]Mr Henderson relied on the approach taken to accountancy issues in other tax appeals, referring to Smith v HMRC [2011] STC 1724, in particular at [70], Degorce v HMRC [2016] STC 542 at [128] to [131], Greene King v HMRC [2014] STC 2439 at [38] and [56] to [58], and GDF Suez Teesside v Revenue and Customs [2017] STC 1622 at [60] and [61] (UT) and [2018] EWCA Civ 2075 at [51] to [56] and [101]. Rather than expressing the point in Edwards v Bairstow terms as such, his formulation was that the FTT heard detailed factual and expert evidence, and this Tribunal should act cautiously and only interfere with the FTT’s conclusions on the correct accounting treatment if it is “convinced” that the FTT reached a conclusion that was not open to it.[34]Smith was an appeal to the Upper Tribunal against a decision of the FTT that the taxpayer’s accounts had not been prepared in accordance with GAAP. The appeal was brought on an Edwards v Bairstow basis so there was no argument about the correct approach to apply. Nevertheless, both Mr Henderson and Ms Shaw agreed that part of the ratio of Arnold J’s decision was his conclusion at paragraph [70] that accounting standards are not law, and that the interpretation of FRS 5 (the standard in issue in that case) was a matter for the FTT unless its interpretation “was one that no reasonable tribunal could hold”.[35]The other cases relied on by Mr Henderson relate more clearly to the application, rather than interpretation, of accounting standards. For example, the issue in Greene King was what was required to comply with FRS 5, not its interpretation. As Mann J commented at [57], which side was correct on that point needed to be determined based on expert evidence, and (at [58]) that the FTT’s conclusion could only be challenged on appeal if it was not open to it. In GDF Suez the question was, again, not one of interpretation of accounting standards but whether the accounting treatment adopted complied with GAAP on the facts, and whether the accounts contravened a provision of the Companies Act 1985.[36]At first sight, Ms Shaw’s submissions seemed convincing. The interpretation of documents, in the manner referred to by Nugee J in Veolia, is generally said to be a matter of law and not fact. However, we are persuaded that this is not the correct approach to take to accounting standards. 14[37]Accounting standards are not legal documents. They are not statutes or contracts. This is also not a case where public law concepts, such as the doctrine of legitimate expectation, are engaged in a way that means that the document in question may form the basis of a legal right (as in Veolia and Davies, for example). So it is not necessary to construe them to determine any legal effect to be given to them. They are documents written by accountants for accountants, and are intended to identify proper accounting practice, not law. No accountant would consider turning to a lawyer for assistance in their interpretation, and nor should they. Where appropriate, interpretations are provided by the relevant accounting standards board itself, not only in notes, the “Basis of Conclusions” and other accompanying material published with standards, but also via committees whose function is to provide interpretations (SIC-19, for example, was published by such a committee).[38]Where accountancy clearly does meet up with legal principles is in the Companies Act, and in particular in the requirement now in s 393(1) Companies Act 2006 that accounts must give a “true and fair view of the assets, liabilities, financial position and profit or loss…of the company”. The interaction between this requirement and accounting principles has been the subject of careful consideration, including highly influential opinions written by the then Leonard Hoffman QC and Miss Mary Arden, originally in 1983 and subsequently updated. We do not need to consider this advice, save to note that its thrust, as noted by the Upper Tribunal in GDF Suez at [56], was that “the courts will treat compliance with accepted accounting principles as prima facie evidence that the accounts are true and fair”.[39]The ASB’s Foreword to Accounting Standards5 explains the position as follows: “16. Accounting standards are authoritative statements of how particular types of transaction and other events should be reflected in financial statements and accordingly compliance with accounting standards will normally be necessary for financial statements to give a true and fair view.[17]In applying accounting standards it is important to be guided by the spirit and reasoning behind them. The spirit and reasoning are set out in the individual FRSs and are based on the Board’s Statement of Principles for Financial Reporting.[18]The requirement to give a true and fair view may in special circumstances require a departure from accounting standards. However, because accounting standards are formulated with the objective of ensuring that the information resulting from their application faithfully represents the underlying commercial activity, the Board envisages that only in exceptional circumstances will departure from the requirements of an accounting standard be necessary in order for financial statements to give a true and fair view.”[40]In our view the question of what is generally accepted accounting practice, as well as the question whether a particular set of accounts are prepared in accordance with it, 5 We were referred to the June 1993 version of this document, which we understand was in force at the relevant time. 15 is a question of fact to be determined with the assistance of expert evidence. Professional accountants are best placed to understand accounting statements in their context, and in particular their “spirit and reasoning”. For example, and relevant to this case, the purpose and significance of the concept of functional currency is not clear without an understanding of accounting concepts and the wider context. We agree therefore with the approach of Arnold J in Smith.[41]What is a matter for a court or tribunal, however, is the proper assessment of expert evidence. Clearly a judge may prefer the evidence of one expert to that of another, but this should be fully reasoned and the judge should not simply “develop his own theory” (see for example Devoran Joinery Co Lt v Perkins (No. 2) [2003] EWCA Civ 1241 at [24]).[42]Whilst the remainder of this decision does engage with the detailed submissions put to us about the meaning and effect of FRS 23, it should be read with the comments already made in mind. The role of this Tribunal is to determine whether the conclusions reached by the FTT were ones that could properly be reached on the evidence. Significance of functional currency[43]It is helpful to start with a brief explanation of the significance of functional currency, and how the concept came into being.[44]Once an entity’s functional currency is determined, any other currency is a “foreign currency” (paragraph 8 of FRS 23). Gains and losses arising from currency movements between a foreign currency and the functional currency in respect of foreign currency transactions, assets and liabilities are generally recognised in the profit and loss account. This is intended properly to reflect the entity’s economic exposure to the foreign currency.[45]Particular considerations arise where an entity conducts business through an entity (such as a subsidiary) in a different country, referred to in FRS 23 as a “foreign operation”. This is dealt with in paragraph 11 of FRS 23. In such a case it is necessary to determine whether the parent entity (referred to as the “reporting entity”) and the foreign operation have the same functional currency. If they do, then both the foreign operation and the reporting entity will reflect FX gains and losses on the foreign operation’s transactions in a similar way in profit and loss. If they do not, then the reporting entity’s exposure is reflected only in movements in its “net investment” in the foreign operation. These movements are reflected in the consolidated balance sheet of the reporting entity as a movement on reserves (balancing the change in value of the asset, being the net investment), and not in the profit and loss account.[46]Applied to the facts of this case, Ball UK is the foreign operation and Ball US is the reporting entity. Ms Shaw submitted that the FTT wrongly identified the intention of FRS 23 as being to minimise FX risk in the presentation of accounts by presenting the accounts in the currency to which Ball UK had greatest economic exposure (paragraph 109). She also submitted that the fact that Ball US was not exposed to any 16 FX risk on Ball UK’s assets and liabilities (except through its net investment) was immaterial to determining Ball UK’s functional currency.[47]Whilst we would not necessarily agree with the FTT’s assessment of the intention behind FRS 23 as being to “minimise” FX risk (as opposed to properly reflecting FX movements in the accounts such that they present a true and fair view), we do prefer Mr Henderson’s submission that understanding the relevance of the position of the reporting entity is an important key to understanding FRS 23 as it applies to foreign operations.[48]The factors that are now listed in paragraph 11 of FRS 23 are substantially derived from the 1993 version of IAS 21. In that version their function was not to determine the functional currency of the foreign entity, but instead to determine how the reporting entity should prepare its consolidated accounts. That turned on whether the foreign operation was or was not “integral” to the operations of the reporting entity, and the factors were listed as indicators of that. If it was not integral it was referred to as a “foreign entity”. Paragraph 24 of the 1993 version explained that a foreign operation that is integral to the operations of the reporting enterprise “carries on its business as if it were an extension of the reporting enterprise’s operations”. It provided the first example now reflected in paragraph 11(a) of FRS 23, and stated that in such cases “a change in the exchange rate between the reporting currency and the currency in the country of foreign operation has an almost immediate effect on the reporting enterprise’s cash flow from operations”, such that the change in rate affects individual monetary items rather than the reporting enterprise’s net investment. In contrast, in what is now the second example in paragraph 11(a), paragraph 25 stated that changes in exchange rate have “little or no direct effect on the present and future cash flows from operations of either the foreign entity or the reporting enterprise”, and affect the net investment only. Paragraph 26 went on to list “indications” that a foreign operation was a foreign entity rather than an integral foreign operation. The list included factors similar to those that now appear in paragraphs 11(b) to (d) of FRS 23, together with the question whether the activities were carried on with a significant degree of autonomy (now of course part of paragraph 11(a)), and factors which had similarities to some of those now listed in paragraphs 9 and 10. (See also BC8, which expressly states that the IASB incorporated the principles for distinguishing an integral foreign operation from a foreign entity into the revised IAS 21, and a similar comment in IN9.)[49]When the revised version of IAS 21 was developed, the term “reporting currency” was replaced with the term functional currency, and a new concept of presentation currency was developed to allow accounts to be presented in a different currency from the functional currency (see IN6, above). As IN9 explains, what were previously indicators to determine whether foreign operations were integral are now among the indicators of an entity’s functional currency, and an entity that would previously have been classified as an integral foreign operation will have the same functional currency as the reporting entity.[50]An understanding of the origins of what is now paragraph 11 of FRS 23 is important, not only to appreciate the introduction of the concept of “reporting entity” 17 in paragraph 11 (and in the definition of foreign operation in paragraph 8), but also to understand that what paragraph 11 requires accountants to address is whether the functional currency of any foreign operation is or is not the same as that of its parent, and that this is a question that is potentially relevant not only to the foreign operation itself but also to the parent (if it adopts FRS 23 or IAS 21). The underlying concept is that if the foreign operation is so closely linked to the reporting entity that it can properly be seen as an “extension” of it, then profits and losses of each of them should be measured in the functional currency of the parent.[51]Ms Shaw submitted that the FTT wrongly relied on the 1993 version of IAS 21 at paragraphs 149 to 158 of the decision, noting that it was not referred to in either Mr Chandler’s or Mr Chopping’s report, and that it dealt with a concept (integral foreign operations) which had been abolished. However, the 1993 version was clearly covered in oral evidence and we consider that it is helpful in reaching an overall understanding of what the relevant part of FRS 23 is seeking to achieve. We do not agree with Ms Shaw that the fact that the 1993 version considered the matter from the parent’s perspective makes it irrelevant to the consideration of the position from the foreign operation’s perspective. The incorporation of the same examples and factors from the 1993 version is a strong indication that the tests are not intended to be fundamentally different from one perspective rather than the other. It would also make little sense if they were. The aim is to ensure that a foreign operation adopts the same functional currency as its “parent” where it is so tied to its parent that it is appropriate to regard it as operating in the same “primary economic environment” as the parent. In such a case, the parent’s accounts will (if prepared in accordance with FRS 23 or IAS 21) also reflect that fact by recording individual items and transactions undertaken by the foreign operation, rather than simply a movement on its net investment and a balancing movement on reserves.[52]Ms Shaw also criticised the FTT’s reference to two notes to FRS 23 which are not reproduced above, N26 and N27 (paragraphs 139 to 144 of the decision). These relate to the treatment of an entity preparing consolidated financial statements – in other words they are looking at it from the perspective of the reporting entity – and in the FTT’s view they supported its conclusion. For the same reasons, we think this criticism was unjustified. The notes helpfully describe a foreign entity the affairs of which are so closely interlinked with those of the investing company that its results are regarded as being “more dependent on the economic environment of the investing company’s currency than on that of its own currency”, and include a footnote explaining that where a foreign operation is not so dependent the exposure that the investing company has relates to its net investment, rather than the individual assets and liabilities. Intermediate holding companies and paragraphs 8 to 14[53]Central to Ball UK’s case is the submission that, as an intermediate holding company, significant parts of paragraphs 9 to 14 of FRS 23 are not applicable, or (on the facts of the case) have reduced weight, and that the FTT placed inappropriate weight on the definition in paragraph 8. In particular, Ms Shaw submitted that the definition of functional currency in paragraph 8 is most obviously applicable to 18 operating companies, and that paragraph 9 has no application at all to companies that are not operating companies. Whether paragraph 9 is engaged is important because paragraph 12 requires priority to be given to the “primary indicators” in paragraph 9 before considering paragraphs 10 and 11. In contrast, HMRC’s position is that the first sentence of paragraph 9 is relevant, and this was the FTT’s conclusion at paragraphs 62 and 63 of the decision. Both parties agreed that paragraph 10(b) is not relevant. As regards paragraph 11, Ball UK’s position is that where a company is not autonomous in a decision-making sense, paragraphs 11(b) to (d) have very little weight because they relate to matters decided upon by the reporting entity.[54]We find it difficult to see why the first sentence of paragraph 9 should not be capable of being relevant to investment or holding companies as well as to operating (in the sense of trading) companies, in line with Mr Chopping’s evidence and paragraphs 62 and 63 of the FTT decision. A company that receives cash in the form of interest income or dividends and spends it, for example in servicing borrowings, both generates and expends cash, just as a manufacturing or other operating company might do so. Indeed, in appropriate circumstances borrowing and lending activities can of course themselves amount to a trade, and it is hard to see why the question whether they do or not should make a difference to the application of paragraph 9. It does not seem to us to be fatal that the remainder of paragraph 9 focuses on operating companies.[55]In substance Ms Shaw’s submission on paragraph 8 was that the definition of functional currency in that paragraph should be accorded less weight where a company is an intermediate holding company, and that it should instead be interpreted in the light of paragraph 11(a) as referring to the currency of the economy that influences or controls that company’s transactions. In principle, we do not see why it should make a difference to the significance of the definition whether an entity is an intermediate holding company or is conducting some other activity. The status of an entity and the nature of its activities are clearly relevant to the application of FRS 23 to determine what a particular entity’s functional currency is. The relevance or otherwise of some of the indicators will be affected by those activities and (in the case of paragraph 11) by whether the entity is a foreign operation. However, there is no indication that we can see in FRS 23 that the underlying concept of functional currency is affected by these different factual scenarios.[56]Ball UK did have some operations, in the sense of activities, and it is not impossible to apply the terms of the definition to it (that is, the “currency of the primary economic environment in which the entity operates”). Paragraph 8 is also in bold, and the second introductory paragraph at the start of FRS 23 states that all paragraphs have equal authority but that paragraphs in bold type “state the main principles”. As indicated by the subheading that follows paragraph 8, paragraphs 9 to 14 are an “Elaboration on the Definitions”, setting out the factors that need to be applied to determine the functional currency (see also paragraph 17, which refers to determining the functional currency “in accordance with paragraphs 9-14”). The fact that those factors are set out does not mean that the terms of the definition itself, and in particular the reference to the primary economic environment in which the entity operates, are irrelevant. The definition itself sets out what the concept is. Paragraphs 9 19 to 14 elaborate on this by setting out the factors to be used to identify the functional currency in any particular case. Not all of them will be relevant in all cases. Notably, paragraphs 9 and 10 are capable of general application whereas paragraph 11 is specific to foreign operations.[57]In the circumstances we consider that the FTT was entitled to conclude, in accordance with Mr Chopping’s evidence, that paragraphs 9 to 14 should be interpreted with the wording of the definition in paragraph 8 in mind (paragraphs 58 to 61 of the decision).[58]Paragraph 9 states that the primary economic environment in which an entity operates is “normally” the one in which it primarily generates and expends cash. This clearly envisages that there will be some circumstances where this primary indicator is outweighed by other indicators, set out in paragraphs 10 and 11. Ms Shaw suggested that paragraphs 10 and 11 do not address “primary economic environment” at all, and that this is confirmed by paragraphs 12 and BC9. We do not agree. Paragraph 12 makes it clear that management must use its judgement where indicators are mixed, and that paragraph 9 contains the primary indicators. Notably, it states that what management is using its judgement to determine is the currency that “most faithfully represent the economic effects of the underlying transactions, events and conditions”. Whilst paragraph 12 does indicate that weighting may be appropriate, it does not provide support for Ms Shaw’s submission and indicates that the test is about economic effects.[59]As the FTT indicated, BC9 is not very well worded. The second sentence appears at first sight to make little sense, given that functional currency is defined in paragraph 8 by reference to the primary economic environment. It might have been better if the second sentence had referred to the indicators in paragraphs 10 and 11 not being “directly” linked to the primary economic environment, or something of that nature. Alternatively, it may have been a poor shorthand for the primary indicators in paragraph 9. But reading BC9 as Ball UK does to support a proposition that paragraphs 10 and 11 have nothing to do with the primary economic environment cannot be right. It disregards the fact that paragraph 10 and 11(b) to (d) each contain economic indicators, or at least indicators that focus on the transactions undertaken. It disregards the fact that paragraph 8 defines functional currency and is stated to be part of the main principles. It also elevates BC9 to a significance that in our view it cannot bear, bearing in mind that the third introductory paragraph to FRS 23, set out above, states that the basis for the conclusions do not form part of the Statement. In our view Ball UK’s approach is contrary to the wording of paragraphs 8 to 14 overall.[60]Ms Shaw also submitted that the reference to “underlying transactions, events and conditions” in paragraph 12 (and 13) shows that the tests are not only concerned with economic factors: autonomy of decision-making was a “condition”. We agree that this may be correct, but do not agree that this demonstrates that the test is not at its heart an economic one when applied to intermediate holding companies. As paragraph 12 states, the aim is to identify the currency that most faithfully represents the “economic effects” of those transactions, events and conditions. As already indicated, we see no basis for drawing a distinction as a matter of principle between intermediate holding 20 companies and operating companies, such that economic factors are determinative for the latter and not for the former.

(a) and “autonomy”

[61]Mr Chandler’s approach was to interpret the reference to “autonomy” in paragraph 11(a) as a reference to what both parties accepted was its main dictionary definition, being the ability to make decisions without being controlled by another entity. In this case it was clear that Ball UK was not an independent decision maker, and simply did what it was told by Ball US. Mr Chopping’s approach was very different. His evidence was that the context, and the fact that the reference to carrying out activities with autonomy was intended to be the opposite of acting as an “extension” of the reporting entity, meant that a wider meaning was intended, more akin to economic or functional independence from the reporting entity (see paragraphs 64 to 68 and 72 of the FTT decision).[62]We consider that the FTT was entitled to prefer Mr Chopping’s approach. Overall the focus of paragraphs 9 to 14 is on economic factors. The term “functional” currency itself gives something of a clue of the need to concentrate on what the entity actually does. Paragraph 11(a) refers to whether the foreign operation’s activities are carried on as an extension of the parent, rather than on whether those activities are carried on with a significant degree of autonomy. The examples that follow, although the first at least is written with trading companies in mind, are of course non-exhaustive illustrations, but they are relevant because they indicate that the focus is indeed on the nature of the entity’s activities, and its relationship with the reporting entity. There is no indication that the determining factor is who makes decisions about those activities. Clearly that may be relevant as part of the overall picture – and certainly an entity that functions as no more than an extension of its parent (what Mr Henderson described as a conduit) is less likely to carry on any independent decision making than one that does not – but it is not determinative. The FTT was entitled to reject Mr Chandler’s suggestion that it was implicit in the first example that the company had its decisions made for it, but in the second that the company reached its own decisions, and to prefer Mr Chopping’s evidence to both that of Mr Chandler and Mr Kraehnke on this point (paragraphs 74 to 79).[63]Ms Shaw suggested that the very fact that paragraph 11 is aimed at group entities rather than stand-alone entities must mean that the concept refers to autonomy of decision-making. That cannot be right. At one level no subsidiary has full autonomy of decision-making from a parent: there are simply degrees of independence. But more significantly there is nothing to indicate why autonomy of decision-making, as compared to economic factors, should not only be relevant but have particular significance, and do so more in the case of an intermediate holding company than it would in the case of a trading company, where the primary indicator in paragraph 9 would, Ms Shaw accepts, be in point.[64]Ms Shaw also submitted that the FTT’s approach would make paragraphs 11(b) to (d) redundant. We disagree. Those paragraphs seem to us to be complementary to paragraph 11(a) as construed by the FTT. Given the origins of these factors in the 21 1993 version of IAS 21, as factors indicating whether a foreign operation is or is not carried out as an extension of the reporting enterprise’s operations (and so is “integral”), that is not surprising.[65]The FTT found support for its conclusions about the meaning of autonomy in the fact that the relevance of decision making was not flagged as one of the main changes in the IASB’s introductory notes to the revised IAS 21 (paragraphs 133 and 134 of the decision) or in the ASB’s notes (paragraphs 136 to 138). Ms Shaw submitted that this failed to appreciate the shift in emphasis described in IN7 towards the currency of the economy that determines the pricing of transactions as opposed to the currency in which transactions are denominated, and the fact that paragraph 11(a) was only of significance to foreign operations, and was also a secondary indicator as compared to paragraph 9: the real significance was for intermediate holding companies, a very limited category, so it was understandable that the notes did not address the point.[66]We consider that the FTT was entitled to conclude that this omission supported its conclusion that the revised IAS 21 did not introduce a new test of autonomy of decision-making. Intermediate holding companies, let alone foreign operations, are far from unusual beasts. We do not consider that the reference in IN7 is anything like clear enough to encompass such a change. On its terms, the currency of the economy that determines the pricing of transactions for a company such as Ball UK which borrows and lends in sterling using sterling rates, is pounds sterling, not the currency of the economy where its parent is based, even if its parent is making the decisions. BC3 makes it clear that the IASB’s intention was not to reconsider the fundamental approach established by the previous version of IAS 21. BC6 states that it would be contradictory for an entity carrying on business as an extension of the reporting enterprise to “operate in a primary economic environment different from its parent”. This also indicates that the FTT was right to prefer Mr Chopping’s evidence. The focus is on the economic environment in which the entity operates, not the location of the decision-making. The point being made is that if an entity in fact operates as an extension of another entity, it is operating in that other entity’s economic environment. This only makes sense if the concept of “extension” is used in a functional or economic sense, rather than (as suggested by Ms Shaw) as the opposite of autonomous decision-making. More than one meaning?[67]In the alternative, and in line with ground (1) of the appeal, Ms Shaw submitted that there could be more than one permissible interpretation of an accounting standard, in which case the question would be whether the interpretation applied was a reasonable one. She drew an analogy with Expert Witness Institute v Customs and Excise Commissioners [2001] EWCA Civ 1882 and submitted that there was no real difference between the possibility of different accounting professionals reaching different but permissible conclusions by virtue of a different application of an accounting standard (which the FTT had agreed in Versteegh6 could be possible in 6 Referred to in paragraph 12 above (Versteegh Ltd & others v Revenue & Customs Commissioners [2013] UKFTT 642 (TC)). 22 borderline cases) and the possibility of them reaching different conclusions as to interpretation.[68]One of the points in Versteegh related to the correct application of GAAP and specifically FRS 5. In that case HMRC failed to establish the taxpayer’s accounting treatment was not a permissible application of GAAP, and that HMRC’s preferred approach was the only permissible approach. The FTT commented at [67] that the expert evidence in that case showed that there was scope for accountants to disagree on the result of an application of FRS 5 to a given set of circumstances, and concluded that the taxpayer’s approach was not “outside what was permitted by the application of FRS 5”. Although the case was the subject of an appeal, HMRC did not pursue this aspect further.[69]Expert Witness was a VAT case that raised the question of the meaning of “aims of a civic nature” in the Sixth Directive, in the context of the scope of exemptions from VAT. The Court of Appeal considered dictionary definitions together with well-known principles under which VAT exemptions are construed strictly but fairly, and concluded that “civic” encompassed not only municipal matters but also matters concerning the relationship between citizens and the state, including the administration of justice.[70]Whilst we do not consider the dividing line between interpretation and application to be straightforward, we do not agree with Ms Shaw’s submission. It is clear from FRS 23 that an entity’s functional currency is not a matter of choice: see IN8. We think it follows from this that the standard must be intended to be interpreted in a uniform way, rather than as a matter of choice or discretion. Whilst accountants might disagree as to the result of its application to a particular set of facts, it seems to us that in principle they must start with a “common rule book” rather than a menu of choices.[71]This does not mean that a particular term cannot encompass or apply to a spectrum of different things, as was the case in Expert Witness. But on the facts of this case we consider that the FTT was entitled to conclude, in the context of FRS 23, that the word autonomy is used primarily in the sense of functional independence, as an antonym of an entity the functions of which are such that in economic terms it operates as an extension of the parent. Independence of decision-making may be a clue as to which side of the line an entity falls, but it does not describe the essence of the concept. Weighting[72]Mr Chandler’s approach depended not only on his interpretation of autonomy as referring to decision-making, but also on concluding that paragraphs 10(a) and 11(b) to (d) had very little weight for a non-autonomous intermediate holding company. This was referred to as the “trumping” argument in the FTT decision. Ms Shaw criticised the FTT for misunderstanding Mr Chandler’s approach, which was not to accord no weight to those other factors and treat paragraph 11 as some kind of flowchart, but instead to give them little weight in the specific situation of a non-autonomous intermediate holding company. 23[73]Ms Shaw submitted that it was clear from paragraph 12 of FRS 23, BC5 and BC9 that weighting was appropriate. She suggested that BC6 also supported Mr Chandler’s approach, because an entity that operates as an extension of its parent “will” have the same functional currency, suggesting that other factors do not matter.[74]In addition, Ms Shaw submitted that the way that the factors that now appear in paragraphs 11(b) to (d) appeared in paragraph 26 of the 1993 version of IAS 21, as indicators that a foreign operation was not integral, was an indicator that they are relevant only to an autonomous enterprise, and therefore assisted in demonstrating that those paragraphs carry little weight where an entity is not autonomous in the decision making sense.[75]We do not agree. To a large extent, the point goes hand in hand with the question of the correct interpretation of autonomy. On Mr Chopping’s approach, for example, paragraphs 11(b) to (d) are naturally complementary to paragraph 11(a). On a proper understanding, the 1993 version of IAS 21 does not support Mr Chandler’s approach because it demonstrates that the test is essentially an economic or functional one. In any event we would not agree with Ms Shaw’s particular point about paragraph 26. Whilst it is correct that the factors set out in paragraph 26 were stated to be indications that a foreign operation was a foreign entity, rather than a foreign operation that was integral to the operations of the reporting enterprise, that simply goes to the manner in which they were expressed. For example, what is now paragraph 11(b) was expressed as “transactions with the reporting enterprise are not a high proportion of the foreign operation’s activities”. All the factors listed were relevant to decide whether a foreign operation either was or was not integral, rather than only applying to an operation that was not integral.[76]We accept that the “trumping” and flowchart references might have been differently expressed, but we do not think that there was any material error by the FTT in failing to understand Mr Chandler’s evidence. Mr Henderson actually used the term trumping in cross-examination, and Mr Chandler agreed with the description. The FTT also recorded a lack of clarity about whether Mr Kraehnke agreed with Mr Chandler about this issue (paragraphs 91 and 92 of the decision). “Big 4” manuals and other evidence[77]Ground (4) was that the FTT had failed to have proper regard to the evidence about the interpretation of FRS 23 adopted by accounting professionals in practice, and in particular extracts from the three “Big 4” manuals that cover the subject, namely those of Ernst & Young, Deloitte and PwC. The FTT considered this guidance in some detail at paragraphs 159 to 205 of the decision, and went on to consider a Deloitte report supporting Ball UK’s use of dollars, and HMRC guidance (paragraphs 206 to 216).[78]Ms Shaw criticised the FTT for misunderstanding the reliance placed by Ball UK on the manuals: they were not being used as an aid to interpretation, as the FTT assumed, but as evidence of their application in practice, in the same way as they were referred to by the FTT in Smith and Nephew Overseas Ltd & others v HMRC 24 [2017] UKFTT 0151 (TC) at [30] and [38]7. However, in reality we think that Ball UK was seeking to use the guidance to support its interpretation, and the FTT recognised this.[79]We were taken to the same extracts from the manuals and HMRC guidance that the FTT considered. In summary we do not consider that the conclusions that the FTT drew were not open to it. Although there are some aspects of the “Big 4” manuals that appear to lend support to Mr Chandler’s approach to the significance of decision-making in the context of an intermediate holding company, overall we do not consider that they support the conclusion that Ball UK must be treated as having a dollar functional currency. The passages relied on by Ball UK, when read in context, do not seem to us to bear the weight placed on them, and the overall flavour is of a fact specific enquiry focusing on economic factors, and whether the particular entity operates as a functional extension of its parent.[80]Turning to HMRC guidance, Ball UK relied on a comment included in HMRC’s Corporate Finance manual at CFM64510 that under FRS 23 an investment company’s functional currency is “likely to be that of its immediate parent company”, and that this could create exchange rate volatility where its assets and liabilities are different to its functional currency. It is unclear to us to what extent, if at all, such comments are relevant but we detect no error in the FTT’s approach. Mr Henderson made clear that HMRC do accept that an intermediate holding company may well have its parent’s functional currency, but whether it does depends on the facts.[81]It is clear from the manuals, and indeed from HMRC’s guidance, that in some – and indeed many – cases, an investment company’s functional currency will be the same as that of its parent. But that does not mean that this will always be the case where the strategic decisions are taken by the parent. It will depend on what the entity actually does.[82]Ms Shaw also referred to the fact that Ball UK’s accounting treatment had been the subject of a clean audit report by PwC, as well as being supported by a report from Deloitte. Whilst it was clearly not a “numbers game” that depended on how many accountants supported its approach, Ms Shaw said that the existence of these reports did provide evidence that it was relevant to take into account, and the existence of the audit opinion should have given the FTT pause for thought.[83]We agree that the existence of the PwC audit and Deloitte report is relevant evidence, but it is clear that it did not preclude the FTT from deciding, on the basis of expert evidence (and having considered the evidence in detail), that Ball UK’s accounts did not in fact comply with GAAP. In particular, we consider that the FTT was entitled not to put weight on the Deloitte report, bearing in mind that the person who wrote it was not called to give evidence (paragraphs 206 to 208 of the decision). 7 We note that this decision has, since the hearing before us, been upheld by this Tribunal on appeal ([2018] UKUT 0393 (TCC)). 25 Did the FTT go too far?[84]Whilst we are confident that there was no material error of law in the FTT’s decision, and it was entitled to reach the conclusion that it did, we do wish to make a few points about the proper role and approach of the Tribunal.[85]As we have explained, our conclusion is that the FTT’s findings about the meaning as well as the practical application of FRS 23 are properly matters of fact. A number of the statements made in the FTT decision need to be read with this in mind, so as to avoid an incorrect impression being obtained that the FTT was expressing legal views about the correct accounting approach, rather than making findings of fact.[86]In making findings of fact, the FTT took the proper course of considering the conflicting expert evidence and explaining in detail why it preferred the evidence of Mr Chopping to Mr Chandler (and to some extent Mr Kraehnke). Overall the FTT clearly understood the significance of FRS 23 in ensuring that accounts show a true and fair view, and concluded that a dollar functional currency would not achieve that objective for Ball UK because, in reality, it was economically exposed to sterling rather than to US dollars. But whilst clearly not required to follow the views of either expert slavishly, it is important that the FTT avoids any impression of forming its own views on factual matters that are not clearly supported by evidence. An example here, although not in our view a material error, was the suggestion that functional currency was intended to be the currency that minimised FX risk (paragraphs 107 to 109 of the decision).[87]We should not be taken to endorse the comments made by the FTT at the end of the decision, and in particular paragraphs 227 and 228, to the effect that no accountant could reasonably have read FRS 23 in the manner that Ball UK did. It is not necessary for us to express a view on this because it relates to the alternative approach referred to in ground (1) of the appeal, which we have rejected. Disposition[88]For the reasons set out above the appeal is dismissed. MRS JUSTICE FALK JUDGE JONATHAN CANNAN RELEASE DATE: 10 December 2018

Cited in 4 later judgments