Aria Technology Ltd v The Commissioners for HM Revenue and Customs: [2018] UKUT 0363 (TCC) [2018] UKUT 0363 (TCC)

UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
[2018] UKUT 0363 (TCC)Case No UT/2016/0225
Between
Aria Technology LtdAppellantThe Commissioners for HM Revenue and CustomsRespondent
Before
MR JUSTICE ROTHJUDGE JONATHAN RICHARDSMichael Firth (instructed by direct access) for AppellantJames Puzey (instructed by the General Counsel and Solicitor to HM Revenue and Customs) for RespondentsDate 2 November 2018Category: Tax
[1]With permission granted by Judge Roger Berner following an oral hearing, Aria Technology Limited (“Aria”) appeals against the decision of the First-tier Tribunal (Tax Chamber) (“FTT”) (Judge Jennifer Dean and Ms Susan Stott). The FTT dismissed Aria’s appeal against the denial by the Respondents (“HMRC”) of a deduction in respect of input tax and a VAT assessment for the relevant accounting period (period 07/06), on the basis that Aria knew, or should have known, that the transactions in question in the appeal were connected with the fraudulent evasion of VAT.[2]This is yet another case of so-called “missing trader” or “MTIC” fraud on the VAT system. The characteristics of MTIC fraud have been described on many occasions, and it is sufficient to refer to the description in the judgment of the FTT (the “Judgment”) at [5]. In this case, as in many others, the products involved were components for computers. Aria accepted in its appeal before the FTT that there was a tax loss to HMRC which resulted from fraudulent evasion of VAT, and that its transactions were connected with that fraudulent evasion. However, it contended that it did not know nor should it have known that its transactions were so connected. The critical issue before the FTT was accordingly whether Aria knew or should have known that its transactions were connected to an MTIC fraud.[3]Aria is a company based in Manchester engaged in both the retail and wholesale of computer components and peripherals. In the year ended 31 July 2005, it declared gross profit of £3.3 million; for the 18 months ended 31 January 2007, a gross profit of £4.6 million; and for the subsequent year ended 31 January 2008, a gross profit of £2.3 million. The managing director and sole shareholder is Mr Aria Taheri, who exercised overall control of the business. At the material times, the finance director was Mr Frank Harasiwka and the purchasing manager was Mr Eddy McFadden. The retail side of the business employed some 120 people, but the wholesale side was essentially run by Mr Taheri and Mr McFadden, with some involvement from Mr Harasiwka regarding the finances. Aria’s wholesale deals involved purchasing central processing units (CPUs) and flat screen monitors from UK suppliers for resale to customers abroad.[4]The appeal was concerned with 11 transactions (the “disputed transactions”), all carried out in a short period between 17 May 2006 and 1 August 2006. Because Aria also placed reliance on another wholesale deal carried out in that period which was not traced to any tax loss, the 11 disputed transactions were numbered chronologically for the purpose of the hearing below 1-2 and 4-12, with the unobjectionable transaction referred to as deal 3.[5]Deals 1-2, made on 17 and 19 May 2006, involved the purchase of Giga CPUs from Supreme Distribution Ltd (“Supreme”) and their resale to Mitz International FZE (“Mitz”) based in Canada. Deals 4-9, in the period 8 June to 10 July 2006, involved the purchase of Intel Pentium 4 CPUs from Supreme and their resale to a purchaser 3 referred to simply as Mona in Luxembourg. Deals 10-12, in the period 21 July to 1 August 2006, involved the purchase of, again, Intel Pentium CPUs from Ashtec Distribution Ltd (“Ashtec”) and their resale to Silver Pound Trading LDA (“Silver Pound”) in Portugal.[6]It appears that the eight deals with Supreme had a value of some £2.8 million (Judgment, [142]) and the three transactions with Ashtec a value of some £2.3 million (see Judgment at [141] and [171]). The Judgment[7]The hearing before the FTT lasted 15 days and the parties submitted extensive written closings and further written submissions after the conclusion of the hearing. As well as significant documentation, the FTT heard substantial oral evidence, including from Mr Taheri, Mr Harasiwka and Mr McFadden who were all cross-examined.[8]The Judgment comprises 100 pages and 374 paragraphs. In view of some of the criticism directed at it on behalf of Aria, it is relevant to refer to its structure. After an introductory section describing the characteristics of MTIC fraud and the governing law, the FTT first deals with a preliminary issue regarding whether the decision of HMRC amounts to an assessment, to which we return below, and then sets out the substantive issue for determination. After summarising some background facts concerning Aria and describing the 11 disputed transactions and the companies involved in the transaction chains demonstrating the connection to fraudulent losses, the next section of the Judgment is headed:
“Did the Appellant know, or should it have known that the transactions in this appeal were connected to fraud?”
There follows a full discussion of the evidence, including some extensive quotations from the oral evidence, covering 155 paragraphs and broken down under various sub-headings. The Judgment then summarises the submissions of the two sides, and the next section is headed, “The Decision”. That includes a sub-section called “Findings of fact on whether the Appellant knew or should have known, that its transactions were connected to fraud”, which covers 41 paragraphs, leading to the “Conclusion” that the FTT was satisfied that knowledge or means of knowledge was made out.[9]On this appeal, Mr Firth submitted that only the sub-section entitled “Findings of fact…” was relevant in considering the basis on which the FTT reached its conclusion, and then subjected the 41 paragraphs in that sub-section to detailed criticism as an inadequate factual foundation for the findings recorded. He contended that the long, earlier section of the Judgment was simply a recitation of the evidence, divorced from any assessment by the FTT, and so served only as background.[10]We reject that contention as fundamentally misconceived. There is no one right way of structuring a judgment. Where the evidence is voluminous, the manner in which a judge may discuss the evidence and justify his or her factual findings is not to be put into a straitjacket. Here, the FTT set out, in considerable detail, those aspects of the evidence which it regarded as relevant for its approach to the question it had to decide. The relatively short, “Findings of fact” subsection is not to be read in isolation. On the 4 contrary, it is to be read with the earlier section giving a much fuller account of the evidence, which provides the basis for the findings. Nor is that long section discussing the evidence, as Mr Firth suggested, entirely neutral. For example, the FTT there criticises an aspect of Mr Harasiwka’s evidence as “untrue”: see [174].[11]In the light of those considerations, we turn to the grounds on which Aria appealed. The Appeal[12]The core of Aria’s appeal involved a challenge to most of the FTT’s findings of fact. Aria challenged numerous specific findings of fact on the basis that:(1) They were unsupported by the evidence or contrary to the evidence.(2) The FTT took into account irrelevant considerations in making its findings or failed to take into account relevant considerations.(3) They were not supported by adequate reasons. Further, Aria advanced what it described as “general criticisms” of certain findings on the basis that:(4) They were not adequately pleaded or put to Aria’s witnesses.(5) The FTT was not entitled to make findings as to how a “reasonable businessman” would have acted without expert evidence. Mr Firth submitted that once such errors were established, or at least errors of the kind set out at (1) to (3) above, the Upper Tribunal must remit the matter to the FTT unless it is satisfied that the decision would “inevitably” be the same.[13]For HMRC, Mr Puzey argued that Aria’s points at para 12(1) to 12(3) above amounted to an Edwards v Bairstow challenge to the FTT’s findings of fact, which as regards each such finding fell to be dismissed according to that standard. He also argued that HMRC’s case had been properly pleaded and put to Aria’s witnesses and that there was no need for expert evidence as to how a “reasonable businessman” would act.[14]Accordingly, we will first consider the legal basis on which Aria can seek to set aside the FTT’s decision on the grounds set out at para 12(1) to 12(3) above. Having done so, we then analyse Aria’s specific criticisms of those findings. Thirdly, we address the “general criticisms” set out at para 12(4)-(5) above.[15]The second aspect of Aria’s appeal involved the question whether HMRC had made an “assessment” for VAT purposes. We deal with that in the final section of this decision. CHALLENGING THE FTT’S FINDINGS OF FACT : THE LAW[16]An appeal to the Upper Tribunal is restricted to a point of law: s11(1) of the Tribunal Courts and Enforcement Act 2007 (“TCEA”). It is not for the Upper Tribunal to conduct its own analysis of the facts to see if it would have reached the same conclusion, and if 5 not, to substitute its own findings. We therefore assess the basis on which factual findings may be challenged as a point of law. The Context[17]It is important to set this assessment in context by reference to the question being addressed: whether Aria knew or should have known that its transactions were connected to the fraudulent evasion of VAT. That test derives from the decision of the European Court of Justice in Cases C-439 & 440 Kittel v Belgium [2006] ECR I-6161, which was analysed by Moses LJ in his judgment in the Court of Appeal in Mobilx Ltd v HMRC [2010] EWCA Civ 517. At [41] Moses LJ stated:
“Kittel … enlarged the category of participants to those who had no intention of committing fraud but who, by virtue of the fact that they knew or should have known that the transaction was connected with fraud, were to be treated as participants. Once such traders were treated as participants their transactions did not meet the objective criteria determining the scope of the right to deduct.”
And Moses LJ explained the correct approach as follows:
“[59] The test in Kittel is simple and should not be over-refined. It embraces not only those who know of the connection but those who “should have known”
. Thus it includes those who should have known from the circumstances which surround their transactions that they were connected to fraudulent evasion. If a trader should have known that the only reasonable explanation for the transaction in which he was involved was that it was connected with fraud and if it turns out that the transaction was connected with fraudulent evasion of VAT then he should have known of that fact. He may properly be regarded as a participant for the reasons explained in Kittel. [60] The true principle to be derived from Kittel does not extend to circumstances in which a taxable person should have known that by his purchase it was more likely than not that his transaction was connected with fraudulent evasion. But a trader may be regarded as a participant where he should have known that the only reasonable explanation for the circumstances in which his purchase took place was that it was a transaction connected with such fraudulent evasion.” Authorities[18]The classic articulation of grounds on which findings of fact, including an inference from the facts, may be challenged on appeal as an error of law is in Edwards v Bairstow [1956] AC 14. That was a tax appeal by the Revenue, against the finding by the General Commissioners that particular transactions by the respondent taxpayer in purchasing and later reselling certain plant and machinery did not amount to an adventure in the nature of trade, so that it was not subject to income tax. The governing statute contained no definition of “trade”. In allowing the appeal, the House of Lords considered the basis on which such a conclusion based on the facts could be disturbed. 6[19]Viscount Simonds (with whom Lord Tucker agreed) said, at [29]:
“… though it is a pure finding of fact, it may be set aside on grounds which have been stated in various ways but are, I think, fairly summarized by saying that the court should take that course if it appears that the commissioners have acted without any evidence or on a view of the facts which could not reasonably be entertained.”
Lord Radcliffe (with whom Lords Tucker and Somervell agreed)1 explained, at [33]:
“… there are many combinations of circumstances in which it could not be said to be wrong to arrive at a conclusion one way or the other. If the facts of any particular case are fairly capable of being so described, it seems to me that it necessarily follows that the determination of the Commissioners, Special or General, to the effect that a trade does or does not exist is not “erroneous in point of law”; and, if a determination cannot be shown to be erroneous in point of law, the statue does not admit of its being upset by the court of appeal. I except the occasions when the commissioners, although dealing with a set of facts which would warrant a decision either way, show by some reason they give or statement they make in the body of the case that they have misunderstood the law in some relevant particular.”
[20]Lord Radcliffe proceeded to consider the test for when an appellate court can intervene, at [36]:
“I do not think that inferences drawn from other facts are incapable of being themselves findings of fact, although there is value in the distinction between primary facts and inferences drawn from them. When the case comes before the court, it is its duty to examine the determination having regard to its knowledge of the relevant law. If the case contains anything ex facie which is bad law and which bears upon the determination, it is, obviously, erroneous in point of law. But without any such misconception appearing ex facie, it may be that the facts found are such that no person acting judicially and properly instructed as to the relevant law could have come to the determination under appeal. In those circumstances, too, the court must intervene. It has no option but to assume that there has been some misconception of the law and that this has been responsible for the determination. So there, too, there has been error in point of law. I do not think that it much matters whether the state of affairs is described as one in which there is no evidence to support the determination or as one in which the evidence is inconsistent with and contradictory of the determination, or as one in which the true and only reasonable conclusion contradicts the determination. Rightly understood, each phrase propounds the same test. For my part, I prefer the last of the three, since I think that it is rather misleading to speak of there being no evidence to support a conclusion when in cases such as these many of the facts are likely to be neutral in themselves, and only 1 The appeal was heard by an Appellate Committee of four Law Lords. 7 to take their colour from the combination of circumstances in which they are found to occur.”
[21]Furthermore, it is appropriate to refer to the elaboration on Edwards v Bairstow by Evans LJ, in a judgment with which Savill and Morritt LJJ agreed, in Georgiou (t/a Mario’s Chippery) v C&E Comrs [1996] STC 463. After referring to the speeches in Edwards v Bairstow, Evans LJ said at [476]:
“It is right, in my judgment, to strike two cautionary notes at this stage. There is a well-recognised need for caution in permitting challenges to findings of fact on the ground that they raise this kind of question of law. That is well seen in arbitration cases and in many others. It is all too easy for a so-called question of law to become no more than disguised attack on findings of fact which must be accepted by the courts. As this case demonstrates, it is all too easy for the appeals procedure to the High Court to be misused in this way. Secondly, the nature of the factual inquiry which an appellate court can and does undertake in a proper case is essentially different from the decision-making process which is undertaken by the tribunal of fact. The question is not, has the party upon whom rests the burden of proof established on the balance of probabilities the facts upon which he relies, but, was there evidence before the tribunal which was sufficient to support the finding which it made? In other words, was the finding one which the tribunal was entitled to make? Clearly, if there was no evidence or the evidence was to the contrary effect, the tribunal was not so entitled. It follows, in my judgment, that for a question of law to arise in the circumstances, the appellant must first identify the finding which is challenged; secondly, show that it is significant in relation to the conclusion; thirdly, identify the evidence, if any, which was relevant to the finding; and, fourthly, show that that finding, on the basis of that evidence, was one which the tribunal was not entitled to make. What is not permitted, in my view, is a roving selection of evidence coupled with a general assertation that the tribunal’s conclusion was against weight of the evidence and was therefore wrong.”
[22]Mr Firth of course accepted the Edwards v Bairstow test and it is indeed the basis of the first ground of appeal as set out in para 12 above. But he submitted it is not an exhaustive test for determination whether a factual conclusion can be challenged as an error of law, and that the grounds set out in para 12(2) and (3) above constituted additional bases on which such a conclusion may be set aside on appeal.[23]Mr Firth relied on a number of authorities in support of his contention. The two most significant appear to be HMRC v Pacific Computers Ltd [2016] UKUT 350, and CCA Distribution Ltd v HMRC [2017] EWCA Civ 1899. Both were cases concerning MTIC fraud, in which the critical question was similarly whether the taxpayer knew or ought to have known that its transactions were connected to fraud. In both cases the FTT had held that this state of knowledge (actual or constructive) was not established, and that finding was challenged by HMRC on appeal.[24]In Pacific Computers, two of the main grounds of appeal were that the FTT had refused to give weight to the evidence of witnesses for HMRC who had not been cross- 8 examined, and that the FTT failed to give proper reasons for its decision. As to the former, since the evidence of those witnesses was not in dispute, this was a clear error of law. As to the latter, in rejecting one of the main arguments advanced by HMRC the FTT stated that in addition to the facts set out in the decision, it also found “such facts as are necessary” to support the rejection of that argument. The Upper Tribunal (Mann J and

Judge Berner) unsurprisingly held that this was “an impermissible short cut.” They stated, at [44]:

“The FTT’s support for inferences and other findings by reference to unspecified further facts is not a proper exercise of the duty to give reasons and must be regarded as an error of law.”
[25]The Upper Tribunal further found that the FTT had failed to appreciate the inferences which HMRC were inviting it to make from the orchestrated and contrived nature of the fraud and the presence of fraudulent companies in the deal chains under consideration. It stated, at [81]-[83]:
“The FTT was keenly aware, it appears, of the need to consider whether inferences could be drawn from the evidence. But in the FTT’s decision that awareness manifests itself, not in a proper consideration of whether inferences could be drawn, weighing the evidence on both sides and reaching a reasoned conclusion, but in a number of statements by the FTT of a general nature that there was no evidence on which to found any inference. … Where there is evidence, and it is evidence from which the tribunal is invited to make an inference, the tribunal must address that question and explain its reasons either for drawing an inference or refusing to do so. It is not simply sufficient simply to say there was no evidence. The failure by the FTT properly to address the submissions of HMRC by reference to the available evidence was an error of law. Although, we accept … that what is required in a decision is not a compendious analysis of every piece of evidence, but a statement of the reasons why the tribunal has reached its factual conclusions, we have concluded that, despite its length, the FTT’s decision regrettably did not meet the basic test.”
[26]In CCA, the two members of the FTT were divided and the conclusion that the taxpayer did not have the requisite state of knowledge was decided by the casting vote of the judge, who himself said that it was a borderline case. The Upper Tribunal allowed HMRC’s appeal on a number of grounds and CCA’s appeal to the Court of Appeal was dismissed. One ground was that in finding that CCA and its guiding mind, a Mr Trees, did not have actual knowledge that its transactions were connected to fraud the FTT took into account the fact that the criminal investigation into the relevant frauds had not involved CCA or Mr Trees, and that this was an irrelevant consideration. The Court of Appeal upheld the acceptance of this ground by the Upper Tribunal, quoting (at [23]) the following passages from the Upper Tribunal’s decision: “81 ….The Judge was required to determine on the evidence before the F-tT whether it had been demonstrated on the balance of probabilities that Mr Trees knew that the CCA transactions were connected with 9 fraud. What other people thought at an earlier time, probably by reference to material which was different from the evidence before the F-tT, was irrelevant.[82]We therefore conclude that HMRC has established that the Judge took into account an irrelevant consideration. The question as to Mr Trees' knowledge was one of the central questions to be determined by the Judge. At [387], the F-tT stated that the case was a borderline case. In determining a central question in a borderline case, the Judge took into account, in favour of Mr Trees and CCA, an irrelevant matter.[83]HMRC submitted that this irrelevant matter was considered by the Judge to be highly material. While CCA accepted that the Judge regarded this matter as relevant, it submitted that it had not been demonstrated that the Judge regarded it as highly material. We have no way of knowing what precise weight the Judge gave to this matter. He plainly gave it some weight. In a borderline case, a matter which has some weight is capable of affecting the outcome.”[27]A further successful ground of appeal before the Court of Appeal was that the judge in the FTT failed properly to consider and explain the basis on which rejected HMRC’s submissions based on an analysis of the money flows in the chains of transactions in which CCA had been involved. This analysis formed a significant part of HMRC’s case on knowledge. David Richards LJ stated (in a judgment with which Arden LJ agreed), at [66]:
“… given that the Judge was rejecting those submissions, it was necessary for him to summarise their essential elements and explain why he was rejecting them.”
Discussion[28]In our judgment, while a failure to take into account relevant considerations or taking into account irrelevant considerations may be an error of law, it is not a free-standing basis of challenge to a factual conclusion falling outside the scope of Edwards v Bairstow. In so saying, we recognise that there is a danger in being over-concerned with the definition of conceptual categories: it is the governing approach to factual conclusions which is important. But it was clear in the present case that Aria was advancing this as a separate basis of challenge in an effort to avoid the high threshold for a successful Edwards v Bairstow appeal.[29]In reality, almost every overall conclusion of fact, such as here that Aria knew or should have known that the relevant transactions were linked to fraud, is based on an evaluation of an assemblage of findings of primary fact. The decision of what primary facts are relevant for the purpose of reaching the overall factual conclusion is essentially one for the fact-finding tribunal: it is part of the fact-finding process. Therefore, if an appellate body were simply to substitute its own assessment of what primary facts should be taken into account, it would itself be engaged in fact-finding and not restricting the appeal to a question of law. It is only if the tribunal failed to take into account a matter which no tribunal properly instructed would have left out of account, or conversely took into account a matter which no tribunal properly instructed would 10 have taken into account, that there is an error of law. Indeed, the reference to taking into account irrelevant considerations or failing to take account of relevant considerations reflects the classic formulation of the test for judicial review expounded by Lord Greene MR in Associated Provincial Picture Houses Ltd v Wednesbury Corp [1948] 1 KB 223, 229, and it is significant that Lord Greene there regarded these, along with a conclusion that was absurd, as aspects of unreasonableness.[30]We do not see anything in CCA that is inconsistent with this view. In that case, the Upper Tribunal expressly regarded the points of law (other than the ‘reasons’ challenge) being raised as coming within the sphere of Edwards v Bairstow and referred to the cautionary words of Evans LJ in the Georgiou case that we also have quoted above: see [2015] UKUT 0513 (TCC) at [48]. The Court of Appeal did not criticise that approach.[31]We do not gain assistance in this regard from the other authorities cited by Mr Firth. In HMRC v Glyn [2015] UKUT 551 (TCC), the question was whether an individual taxpayer had ceased to be resident in the UK for a particular tax year. The Upper Tribunal (David Richards J as he then was) overturned the finding of the FTT that he had become non-resident. While the FTT had found that the dominant reason for Mr Glyn retaining his house in London was to live there when he eventually returned to England and had little to do with his interim use of the house in the meantime, the judge looked at all the evidence relied on and found that there was simply no proper basis to support this finding. In so doing, he expressly applied the test for challenging a finding of fact set out by Evans LJ in Georgiou. Secondly, in considering the continuing use which Mr Glyn made of his house in London, the FTT took into account irrelevant considerations and failed to have regard to relevant factors: the tribunal had focused on Mr Glyn’s purpose or dominant purpose in retaining the house, whereas it should have looked at the way in which and regularity with which the house was used. The tribunal thus applied the wrong test, failing to focus on the nature and quality of the use. This was therefore a case of the FTT misapplying the law on residence, not a simple challenge to its factual conclusion on the evidence.[32]As for Davis v Wiggett [2016] UKUT 358 (TCC), there the Upper Tribunal found such contradictions on a crucial point between the actual decision of the FTT and its further decision refusing permission to appeal that “it [is] impossible to see the basis of the decision” (at [31]). This was accordingly a rather special case and, in effect, a ‘reasons’ challenge. Wright v HMRC [2013] UKUT 0481 (TCC) involved an appeal against the decision of the FTT to refuse an adjournment. That was an exercise of discretion by the FTT and the Upper Tribunal addressed the basis on which such a discretionary decision could be impugned. That is far removed from the question whether a factual conclusion can be challenged as amounting to an error of law.[33]Turning to the ground in para 12(3) above, we accept that a failure to give adequate reasons is an independent ground of appeal and constitutes an error of law. This is well-established at common law: see Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377. Moreover, rule 35 of the Tribunal Procedure (First-tier Tribunal)(Tax Chamber) Rules 2009, although not expressly relied on before us, requires the giving of reasons by the FTT for its decision. Pacific Computers is a clear illustration of an appeal on 11 this ground and we consider that the further ground in CCA concerning the banking evidence comes into this category.[34]However, in the first place, we think it is clear that the question of reasons is to be addressed by considering the judgment as a whole. As the Court of Appeal observed in CCA, an appellate tribunal is not confined to looking at the particular passage leading up to the expression of a conclusion that is being criticised but can look elsewhere in the judgment under appeal for fuller reasoning: see at [53]. Moreover, the appellate tribunal should take a realistic approach to the way the judgment is expressed: “…, it is entirely proper for an appellate tribunal to seek to give meaning to the judgment under appeal when the judge has not made clear the point he is making” (at [25]).[35]Secondly, and more fundamentally, the requirement for reasons, while undoubtedly significant, must not be set too high. As Jacobs LJ stated in Revenue & Customs Commrs v Procter & Gamble UK [2009] EWCA Civ 407 at [19], in a passage quoted in Pacific Computers:
“ …It was not incumbent on the Tribunal in making its multifactorial assessment not only to identify each and every aspect of similarity and dissimilarity (as this Tribunal so meticulously did) but to go on and spell out item by item how each was weighed as if it were using a real scientist's balance. In the end it was a matter of overall impression. All that is required is that “the judgment must enable the appellate court to understand why the judge reached his decision” ( per Lord Phillips MR in English v Emery [2002] EWCA Civ 605, [2002] 1 WLR 2409 at 19]) and that the decision “must contain … a summary of the Tribunal's basic factual conclusion and statement of the reasons which have led them to reach the conclusion which they do on those basic facts” ( per Thomas Bingham MR in Meek v Birmingham City Council [1987] IRLR 250 )”
[36]To that we would add a further passage from the judgment of the Court of Appeal in English v Emery, quoted by the Upper Tribunal in CCA:
“[21] When giving reasons a judge will often need to refer to a piece of evidence or to a submission which he has accepted or rejected. Provided that the reference is clear, it may be unnecessary to detail, or even summarise, the evidence or submission in question. The essential requirement is that the terms of the judgment should enable the parties and the appellate tribunal readily to analyse the reasoning that was essential to the judge’s decision.”
Consequence of an error[37]Irrespective of whether a failure to have regard to relevant considerations/having regard to irrelevant considerations is a separate ground of challenge or, as we think, falls within a broader Edwards v Bairstow category, there remains the question of the consequence if an error (or errors) in the decision under appeal is established.[38]Mr Firth submitted that where the FTT had committed an error, then unless the appellate tribunal is satisfied that the decision below would inevitably have been the same, the decision could not stand. He contended that unless the appellate tribunal was 12 able to substitute its own decision, which he said was not the position here, the case should be remitted to the FTT for reconsideration.[39]For this proposition Mr Firth relied on the decision of the Court of Appeal in John Dee Ltd v Customs & Excise Commrs [1995] STC 941. That was an appeal against the decision of the Commissioners to require security for the payment of VAT from the appellant company, based on its apparent link to a group of companies which had gone into receivership owing substantial arrears of VAT. The Commissioners’ decision was the exercise of a statutory power in the VAT Act 1983:
“Where it appears to the Commissioners requisite to do so for the protection of the revenue they may require a taxable person …to give security, or further security, of such amount and in such manner as they may determine….”
The basis of the appeal was that before making such a decision, the Commissioners should have asked the company for information about its financial position. The first instance tribunal (there the VAT Tribunal) found that in failing to have regard to the possibility of seeking relevant financial information from the company, given that this was not a straightforward situation of a ‘phoenix’ company (i.e., resurrecting a previously defunct enterprise), the Commissioners had failed to take into account a relevant consideration. This finding was not challenged. But the VAT Tribunal proceeded to hold that it was “most likely” that if the Commissioners had obtained that information their concern to protect the revenue by a requirement for security would have been fortified, so that their decision would have been the same. The appeal was therefore dismissed.[40]The Court of Appeal held that this was an erroneous approach. The VAT Tribunal was exercising an appellate function and could not substitute its own discretion for that of the Commissioners. As Neill LJ stated (at [952]):
“The protection of the revenue is not a responsibility of the tribunal or of a court.”
The Court accepted, as was conceded by the appellant, that “where it was shown that, had the additional material been taken into account, the decision would inevitably have been the same, a tribunal can dismiss an appeal.” However, in that case the finding by the VAT Tribunal that “it is most likely” that the decision would have been the same did not amount to a finding of inevitability, so the decision had to be set aside.[41]Mr Firth submitted that this test of ‘inevitability’ is supported by CCA, where once it was found that the decision of the judge in the FTT was flawed through taking into account an irrelevant consideration in his conclusion regarding the taxpayer’s knowledge, the decision was set aside since it was impossible to know what his conclusion would otherwise have been: see at [27].[42]Of direct relevance to this question is the decision of the Upper Tribunal (Hildyard J) in Edgeskill Ltd v HMRC [2014] STC 1174, on which Mr Puzey principally relied. That was another case of MTIC fraud where the issue was whether the appellant had the requisite knowledge so as to render it a participant under the Kittel test. As part of its appeal, the appellant challenged a large number of the findings of fact by the FTT 13 as amounting to an error of law. Before examining those criticisms in detail, Hildyard J explained the approach to be taken as follows:
“[204] In support of Ground 9, the appellant has assembled some 16 alleged examples, elaborated over the course of some 11 pages, of findings of fact against the appellant which are said to amount to an error of law. [205] The error of law is submitted to be the making of findings of fact that on the evidence available to the FTT it was not reasonably open to the FTT to make (i.e. that were perverse). The appellant relies in that regard on Edwards (Inspector of Taxes) v Bairstow (1955) 36 TC 207, [1956] AC 14 at 36. [206] In my view, the appellant appears to have misunderstood or misapplied that case. The appellant appears to me to have proceeded on the basis that the numerical aggregation of alleged errors of fact will amount to an error of law; but that is not, in my view, the correct approach. [207] What must be demonstrated, on my reading of the case, is that once the erroneous findings of fact are identified, shown to be such that they were not reasonably open to the tribunal to make, and corrected, the only reasonable conclusion from the true facts contradicts the determination made; or put another way, the error in the finding(s) of fact must be unequivocal (in the sense that no reasonable tribunal acting judicially could have made the finding(s)) and the erroneous findings must have vitiated the ultimate determination (or in other words, so infected the ultimate determination as to render it perverse or, at the least, unsafe). [208] If that is correct, it is necessary to asses, first, whether the errors of fact alleged were plainly findings that the FTT, acting judicially and properly instructed, could not have made; and secondly, if unequivocally erroneous findings are demonstrated, whether it has then also been demonstrated that they infected the ultimate determination so as to render it perverse or unsafe. A negative answer at either stage negates any error of law.”
[43]Mr Firth submitted that insofar as Edgeskill appears to set a higher test for an appeal than asking whether the decision would inevitably have been the same, it is inconsistent with the approach of the Court of Appeal and should not be followed.[44]Our consideration of this issue is governed by the statutory framework setting out the appellate function of the Upper Tribunal. Section 12 of TCEA provides, insofar as material: “(1) Subsection (2) applies if the Upper Tribunal, in deciding an appeal under section 11, finds that the making of the decision concerned involved the making of an error on point of law. (2) The Upper Tribunal -- may (but need not) set aside the decision of the First-tier Tribunal, and if it does, must either – 14 remit the case to the First-tier Tribunal with directions for its reconsideration, or re-make the decision.” [our emphasis][45]In that context, we think that there is no real distinction between the various authorities, properly analysed. Edwards v Bairstow was a case where the scope and accuracy of the primary facts found by the first instance tribunal (there, the General Commissioners) was not in issue. But their overall factual conclusion – that the transactions did not amount to an adventure in the nature of a trade – was one which no tribunal, properly instructed, could arrive at. Accordingly, that was an error of law. And as that error lay in the overall conclusion itself, the decision had to be set aside.[46]In CCA, the error was at the level of primary facts, in that in arriving at its overall factual conclusion as to lack of knowledge the tribunal relied on a matter (the non-involvement of Mr Trees and the company in criminal proceedings) which, properly considered, was irrelevant. In so holding, the Upper Tribunal and then the Court of Appeal were therefore finding that no tribunal, properly instructed, would have given that factor any weight. Thus, this was an error of law in itself. Given that the tribunal had described this as a borderline case, it was impossible to say that this was not a material consideration in the tribunal reaching the overall factual conclusion on knowledge, and this meant that the decision had to be set aside and remitted.[47]In Pacific Computers, the failure to give weight to the evidence of witnesses who were not challenged was a clear error of law, as the respondent indeed accepted. But the Upper Tribunal then considered whether that was material to the overall conclusion on knowledge. There, the Upper Tribunal held that this, along with the other errors found in the FTT decision, was material to the conclusion reached: para [85]. In so doing, they expressly followed the approach in Merthyr Tydfill Car Auction Ltd v Thomas [2013] EWCA Civ 815: see at [24]-[25]. That was a case where the trial court had made a similar error, but in contrast the Court of Appeal found that the effect of this error was not material since giving due weight to that evidence could not realistically have led to a different outcome.[48]In Edgeskill, there were a series of criticisms of findings of subsidiary facts. As well as rejecting those criticisms in that the findings were not perverse, the judge found that most of them were in any event not material: they did not go to the root of the overall factual conclusion as to knowledge. Thus they did not amount to an error of law and provided no basis for setting aside the FTT’s overall conclusion.[49]John Dee was, in our view, a very different kind of case. The failure by the Commissioners to consider whether they should seek financial information from the company was an error of law. But the case concerned the exercise of a statutory discretion by the Commissioners, where the court could not itself exercise that discretion in the Commissioners’ place. Thus, although the court was exercising an appellate, not a supervisory, jurisdiction, the case bears similarities to a judicial review, and the fact that obtaining that information might have led the Commissioners to exercise their discretion the other way provided sufficient grounds to overturn the 15 decision. We think that is very different from a challenge on appeal to an overall factual conclusion based on a wide range of primary facts.[50]We consider that application of the proper approach on an appeal such as this is critical. If each paragraph of a long judgment, involving the analysis of a wide range of facts following substantial evidence, is examined with a fine toothcomb, it may often be possible to find some factual error or lack of clarity, or to identify particular facts which were arguably relevant but are not referred to or assessed. If such flaws meant that the decision must be quashed as erroneous in law, little would be left of the cautionary words of Evans LJ in Georgiou. In our judgment, having regard to the wording of s12(2) of TCEA which we have emphasised and to the authorities we have discussed, even if we find some errors regarding any factual matter considered, or which should have been considered, in the decision below, such that any tribunal properly instructed would have taken that matter into account, or left it out of account, only if that matter is material to the overall factual conclusion is that a basis for setting the decision aside. The same approach applies if the reasoning expressed in support of a particular factual finding is unclear or deficient. ARIA’S SPECIFIC CRITICISMS OF THE FTT’S FINDINGS[51]Aria made some 362 specific criticisms of the FTT’s factual findings. We address those criticisms by reference to the numbering used in Mr Firth’s skeleton argument (though we will not deal with them in the same order). We have grouped our analysis by reference to theme. Aria’s awareness of MTIC fraud (Point 5 of Mr Firth’s skeleton argument)[52]At [338], the FTT said:
“We did not accept that Mr Taheri, who is clearly an intelligent man with significant professional and business experience, did not have a full understanding of MTIC fraud, particularly when viewed against the fact that he had in the past had input tax denied as a result of problems in the chain of supply. We also did not accept that Mr McFadden and Mr Harasiwka were not aware of MTIC fraud; again both were experienced in the industry and they had been present at a meeting when HMRC had outlined the problems with fraud.”
[53]Mr Firth criticised this factual conclusion for the following two broad reasons:(1) The basis for the conclusion is untrue: Mr Taheri had not previously had input tax denied; rather a claim for input tax credit had been subjected to extended verification (but was ultimately granted).(2) Mr Taheri’s evidence was that he knew generally about MTIC fraud, but did not understand the “carousel” nature of that fraud (which involves goods 2 They were there numbered (1)-(35) in Mr Firth’s skeleton argument, but two were numbered (11). 16 repeatedly entering and leaving the UK)3. That evidence was not challenged in cross-examination. Therefore, the finding that his understanding of MTIC fraud was “full” was irrational, against the weight of the evidence, had not been put to Mr Taheri and was not available to the FTT.[54]However, the relevant consideration is the extent of Aria’s knowledge of MTIC fraud and not merely that of Mr Taheri. There was ample evidence before the FTT, reflected in the Judgment, to support a finding that Aria was aware that the products it was dealing in were particularly susceptible to MTIC fraud and that HMRC expected taxpayers trading in such goods to make checks of both their suppliers and customers. At [124] and [125] of the Judgment, the FTT referred to letters sent to Aria about MTIC fraud, including a letter dated 2 December 2003, which expressly referred to checks on both suppliers and customers.[55]No doubt as a purely linguistic matter, very few people will have a “full” understanding of MTIC fraud. Even those most familiar with the fraud might not know its full extent or precisely how it is perpetrated in all circumstances. Moreover, HMRC accept that Mr Taheri’s previous input tax claim had not been denied but had merely been delayed while further checks were pursued. However, that does not undermine the basic point made by the FTT that, because Mr Taheri had experienced some difficulties in the past with input tax recovery and given his professional experience in an industry that was susceptible to MTIC fraud, he could be expected to have a good understanding that MTIC fraud was a live issue in his (and Aria’s) chosen line of business. Aria’s awareness of MTIC fraud generally was plainly relevant to the FTT’s assessment of whether Aria knew or should have known that the disputed transactions were connected to fraud.[56]Mr Taheri had said in cross-examination that he did not understand the “carousel” nature of the fraud. However, the finding that the FTT made did not require it to reject that evidence. There was ample basis for the FTT to conclude that that Aria had a sufficient understanding of MTIC fraud to appreciate the importance of making checks on both suppliers and customers4. Due diligence checks (Points 7, 8, 9,10, 115, 13 and 35 of Mr Firth’s skeleton argument)[57]At paragraphs [340], [341], [343] and [344] the FTT set out conclusions that it had drawn from the extent of Aria’s due diligence on both customers and suppliers as follows: 3 Moreover, being unaware of the “carousel” nature of the fraud, Mr Firth argued that Mr Taheri could not have been aware of the importance of performing due diligence on Aria’s customers, as well as its suppliers. 4 As noted at para 54, there was ample evidence that Mr Taheri realised that HMRC attached significance to checks on customers as well as suppliers, even though he evidently thought that checks on suppliers were more important. 5 i.e. the second point 11: see note 2 above. 17 “340. The Appellant had not traded with Supreme since 2004 and never to such a high value yet the Appellant was willing to enter into deals on the basis of trust. We found the evidence lacked credibility; whilst knowledge or trust in a trader may provide some comfort this must be viewed in the context of the sudden increase in the value of trade, the limited due diligence on the trader, the profit achieved by the Appellant and the fact that a customer appeared “out of the blue” expressing interest in exactly the type and quantity of goods Supreme was selling. The only due diligence carried out on Supreme was a VAT number check and Creditsafe report; the latter of which in our view would have caused any reasonable businessman to make further inquiries given the zero credit rating and negative equity report. The Appellant’s argument that it did not give credit to Supreme only raised further questions as to why the report was obtained if it was to be ignored. We concluded on the evidence before us that the relationship with Supreme amounted to little more than it being a trader with whom the Appellant had dealt in the past. The evidence as to what the Appellant did to satisfy itself that the deals in 2006 was vague and we concluded that this indicated knowledge on the part of the Appellant; the circumstances of the deals were quite obviously too good to be true and we inferred that the Appellant’s failure to query this indicated that it knew the deals were contrived.[341]Similar issues were present in the Creditsafe report on Ashtec which we found would have caused any reasonable trader seeking to satisfy himself that the deals were legitimate to have undertaken more meaningful checks on his supplier particularly in circumstances where Ashtec had not traded for 6 years. We were satisfied that the visit to Ashtec by one of the Appellant’s employees was no more than window-dressing; if the Appellant trusted Ashtec to the extent it purported to there was no need for the visit other than to satisfy HMRC. Furthermore the documents obtained by the employee such as a copy of Mr Afzalnia’s driving licence did nothing to assist the Appellant in assessing the veracity of the deals and the evidence that she “had a quick look around his office” provided no detail as to what she had looked for, what she had found or why this was necessary. The Appellant relied on Mr Taheri’s pre-existing relationship with Mr Afzalnia as the basis for trust. However this simply did not withstand scrutiny in cross-examination; the evidence as to what inquiries Mr Taheri had made as to the veracity of the deals was vague, uninformative and amounted to no more than possibly having asked about the VAT issue and taking Mr Afzalnia’s word that the deals were genuine at face value. We found that this was wholly insufficient for any reasonable businessman seeking to protect himself from fraud. …[343]Given that the Appellant was reliant on its customers for payment we found that the lack of any meaningful due diligence on those customers was indicative of the Appellant’s knowledge of contrivance as it knew there was no credit risk associated with those customers being unable to pay. The consequences to the Appellant of such a risk were significant as it would be left in a position where it could not honour its 18 obligations to its suppliers which could have given rise to a claim against it. The fact that the Appellant carried out no due diligence to assess the extent of this risk could only lead to the conclusion that it was aware of the overall scheme to defraud the Revenue. The due diligence was wholly inadequate for its purpose of assisting a trader to avoid connection to fraud. In our view a reasonable businessman would have found the information worthless and we concluded that the only purpose of carrying out due diligence was to satisfy HMRC. Despite the witnesses for the Appellant stressing that Mr Taheri had emphasised the importance of such checks, in reality very little was done and little or no independent inquiries were made.[344]We noted the Appellant’s submission that there is no legal requirement to conduct due diligence and we did not focus unduly on this feature. However the relevance was not whether the Appellant had any obligation to carry out due diligence but rather what reasonable steps the company took to check the integrity of its deals in an industry rife with fraud. Any reasonable businessman seeking to protect himself from fraud would see due diligence and verification of the integrity of its deals as a commercial necessity. We concluded from the inferences we have drawn from the Appellant’s lack of commercial checks that its lack of due diligence was indicative of knowledge on its part.”[58]Also, at [368], the FTT described Aria’s due diligence as “woefully inadequate” for the purpose of assessing the integrity of the transactions that Aria had undertaken, as part of its conclusion that, even if Aria did not actually know that its transactions were connected with fraud, it should have known.[59]Mr Firth criticised the FTT’s conclusions for the following broad reasons:(1) Aria had last dealt with Supreme in January 2006, so the FTT’s conclusion in the first sentence of paragraph [340] was wrong.(2) The FTT reached an unreasonable factual conclusion and/or took into account an irrelevant consideration at [343] in concluding that Aria’s witnesses had “emphasised” the importance of checks on customers in their evidence when they had done no such thing.(3) The process of reasoning that led the FTT to conclude, at [340], that Aria’s dealings with Supreme were “too good to be true” was flawed, irrational and involved the FTT taking into account irrelevant considerations. For example, the FTT stated that Aria’s customers appeared “out of the blue” and that this was relevant to its assessment of the quality of Aria’s due diligence. This bald statement, Mr Firth argued, demonstrated that the FTT had ignored Aria’s evidence as to the extensive groundwork put in to produce a large network of suppliers and customers. Moreover, the FTT’s conclusions were inadequately reasoned: it should have explained what it meant by deals being “too good to be true” and explained what aspects of Aria’s “profit achieved” were suspicious.(4) In Mr Firth’s submission, by focusing, at [344], on the fact that there was no legal requirement to perform due diligence, the FTT demonstrated 19 that it had failed to address the core of Aria’s submissions on the significance of due diligence. More specifically, in reaching its conclusion, the FTT ignored potentially benign reasons why Aria might perform limited due diligence on suppliers (for example the fact that Aria already knew and trusted its suppliers, and its policy of not undertaking “bulk deals” with a supplier with whom it had not traded for a full year). The FTT also ignored evidence that Aria’s policy was real and had resulted in it turning down brokerage transactions in the past.(5) HMRC’s pleaded case at paragraph 40.5 of their Statement of Case was that Aria had “no commercial risk” in entering into its transactions since it received payments from its customers before making payment to suppliers. Therefore, in finding at [343] that Aria faced “significant” risk if its customers did not pay, the FTT was making an irrational finding and, moreover, a finding that had not been adequately pleaded or put to Aria’s witnesses.(6) On no view could an absence of meaningful due diligence on customers “only” lead to the conclusion that Aria was aware of fraud particularly since Aria’s due diligence on customers in deal 3 (which was not connected with fraud) was no different from its due diligence on customers in the disputed transactions that were connected with fraud. Mr Firth argued that the FTT’s use of “unsustainable hyperbole” at [343] demonstrated that the FTT had closed its mind to Aria’s case and submissions.(7) As a matter of pure logic, Mr Firth submitted that absence of due diligence could not be indicative of connection to fraud: a trader who knew that its deals were connected to fraud would have every incentive to “paper” its transactions with seemingly extensive due diligence in an attempt to mislead HMRC into thinking it was a legitimate trader. Therefore, it followed that the FTT’s overall conclusion, that lack of due diligence on suppliers indicated that Aria knew its transactions were connected with fraud, was irrational, took into account irrelevant considerations and ignored relevant considerations.(8) At [341], the FTT deduced from Mr Taheri’s inability in cross-examination to recall details of a conversation about VAT from Mr Afzalnia (the managing director of Ashtec) that took place many years ago, that his evidence on the issue was vague and uninformative. In doing so, the FTT failed to take into account the highly relevant consideration that the passage of time would inevitably have dulled Mr Taheri’s memory.[60]Mr Firth is correct to say that the FTT was mistaken in concluding that Aria last dealt with Supreme in 2004. In fact, Aria had bought £12,000 worth of computer monitors and other goods from Supreme in January 2006. However, in our view this error is clearly not material. The fundamental point was that the disputed transactions were out of all proportion to any previous dealings with these suppliers, and Mr Taheri accepted that Aria had never traded to such a high level with Supreme (or Ashtec) before: see at [143]. 20[61]Much of the cross-examination of Mr Taheri on due diligence dealt with due diligence on suppliers. However, due diligence on customers was also covered. Mr Taheri said it was an “exaggeration” to say that he did no due diligence on customers. He accepted that he knew that part of his due diligence should involve checks on customers (though he maintained that the bulk of the checks needed to be on suppliers). He asserted in terms that, while Aria’s emphasis was on performing due diligence on its suppliers, “We did check customers”. When asked what details Aria would be interested in checking, Mr Taheri referred to information on creditworthiness of customers saying:
“So, for example, if we are selling to that company can we give them credit. And how many years have they been in business.”
[62]Further, Mr Taheri described checking the address and VAT number of Mona as “very, very, very important standard procedure”, and said that he would have instructed Aria’s staff to make sure they did this. He put into evidence a “due diligence checklist” that Aria used for its brokerage deals which required information on both customers and suppliers. However, no completed checklists for any of the disputed transactions were produced as evidence before the FTT.[63]Overall, we are satisfied that the FTT was entitled to reach the conclusion that Mr Taheri “emphasised” the importance of due diligence on customers and we do not regard that conclusion as in any way ignoring his evidence that he thought due diligence on suppliers was more important. Moreover, despite recognising the importance of due diligence on customers, Aria performed little such due diligence.[64]We do not agree that the FTT ignored Aria’s evidence as to how it set about producing a network of suppliers and customers6. The FTT referred to that evidence at [184] and [349] and it was not obliged to set it out in full. At [137], the FTT recorded Mr Taheri’s acceptance that Mona, Mitz and Silver Pound, the customers in the disputed transactions, were all new in the relevant period. In cross-examination, Mr Taheri had accepted that he had had no prior contact with them. The mere act of sending a spreadsheet to contacts from whom Aria obtained business cards at trade fairs resulted in customers coming forward to purchase the precise types and quantities of stock that Aria was offering. The FTT was fully entitled to conclude that, even taking into account the “groundwork” performed in putting together the network of contacts, those customers appeared “out of the blue”. Nor do we accept that the FTT failed to give adequate reasons in explaining what it meant by customers appearing “out of the blue” or what specific “profit achieved” it was referring to since its overall conclusion was clear.[65]We should add that at the hearing below, Aria attempted to support its case concerning the way these transactions came about by reference to handwritten notes 6 Very broadly, Aria’s evidence was that Mr Taheri and others would regularly attend trade fairs and would collect business cards from potential customers and suppliers. It used contact details on those cards to create a network on MSN that enabled it to receive offers of purchase and sale from potential counterparties. It distilled all offers to sell products that it received in a particular period into a single spreadsheet, added the mark-up it hoped to obtain and sent that spreadsheet out to potential purchasers. 21 said to have been made at the time by Mr Taheri and Mr McFadden. After considerable cross-examination regarding the circumstances of these notes and their late discovery, the FTT declined to place any weight on them because of its concerns about their provenance: Judgment at [358]-[366]. We discuss the question of these handwritten notes further below. But the FTT’s conclusion as to the entry into the disputed transactions has to be assessed on the basis of the sparse evidence on which it felt it could rely.[66]There is no force in the argument that the FTT ignored Aria’s case that there were benign reasons for its approach to due diligence on suppliers. It clearly recognised that part of Aria’s case was that it did not need to do extensive due diligence because it knew and trusted its suppliers. The FTT referred to this argument at [155] and, at [149], the FTT quoted an extract from Mr Taheri’s cross-examination on the basis of his trust in Ashtec. The fact that the Judgment does not refer to the evidence of Aria rejecting transactions with suppliers with whom it did not have a 12-month trading history does not mean that evidence was “ignored”. The FTT clearly focused on the nature and extent of the prior contact and relationship between Aria and the two suppliers (Supreme and Ashtec) in the 11 disputed transactions. We consider that the FTT was fully entitled to regard that aspect as much more relevant to its assessment than evidence of rejected deals: there is nothing unreasonable or irrational about its approach.[67]The FTT was also entitled to look behind Aria’s assertion that it trusted its suppliers and consider the basis of that trust. The FTT had found, and was entitled to find, that Aria had a good understanding of MTIC fraud (see para 56 above). The passage of cross-examination quoted at [149] clearly suggested to the FTT that the reassurance that Aria received from Ashtec that VAT had been paid on earlier supplies of goods that were particularly susceptible to MTIC fraud amounted at most to a mere assertion by Mr Afzalnia in a conversation that may not have even mentioned VAT at all. When that is placed alongside the fact that Ashtec had been dormant between 2000 and mid-2006 (i.e. virtually until the time Aria made these purchases from it), the FTT was entitled to conclude that the foundation of any trust in Ashtec in particular was flimsy.[68]Standing back and reading paragraphs [340], [341] and [344] in the context of the Judgment as a whole, the FTT’s reasoning on supplier due diligence is clear. As the FTT noted at [344], Aria was operating in an industry sector that was “rife with fraud”. The FTT had concluded that Aria had a good awareness of that fraud. There can be no sensible criticism of the FTT’s statement at [344] that a reasonable businessman would see due diligence as a “commercial necessity”7. Yet Aria performed “woefully inadequate” due diligence and the FTT asked itself why it did so. Taking into account the nature of the goods, the nature of Aria’s business, its awareness of MTIC fraud and the scale and circumstances of the disputed transactions, the FTT rejected Aria’s explanation that trust in its suppliers explained the inadequate due diligence on them Having rejected the proffered explanation, the FTT was entitled to conclude that Aria’s 7 This was one of the few factual findings of which Aria made no specific criticism and, as noted below, we reject Aria’s general argument that the FTT should not have made findings about the actions of a “reasonable businessman” in the absence of expert evidence. 22 failure to perform adequate due diligence pointed towards a conclusion that it knew its transactions were connected with fraud. Other inferences might have been possible (for example that Aria’s business procedures were simply slipshod, or that it was so busy trying to make money that it cut corners with due diligence), but the negative inference that the FTT drew was, in our judgement, available to it and sufficiently explained. In stating the basis of its conclusion, the FTT was not bound to set out and explain why it rejected each alternative hypothesis. Hence Mr Firth’s submission that a trader who knew its transactions were connected to fraud would have an incentive to “paper” its transactions with apparently plausible due diligence is nothing more than speculation as to how people might be supposed to act; it was not evidence and the absence of discussion of such a speculative submission is not an error of law.[69]In paragraph 40.5 of their Statement of Case, in a section headed “Further Evidence of Knowledge or Means of Knowledge”, HMRC referred to Aria having “no commercial risk” since it received all payments before suppliers were paid. Read in context, that was an assertion that Aria’s business model in the disputed transactions was uncommercial and contrived. We do not think it can be read as suggesting that HMRC were acknowledging that Aria’s failure to perform significant due diligence on customers was understandable because customers would pay in advance.[70]At [343], the FTT was identifying a “risk”, namely that if Aria’s customer did not pay, Aria would be exposed as it still had to pay its supplier and it would be little comfort to Aria that it was entitled to obtain the goods. The FTT was not saying that the risk had actually materialised and so caused Aria loss. Mr Taheri had testified that Aria had previously cancelled orders from suppliers because customers cancelled their order and that Aria had not suffered any significant financial loss as the supplier had simply sold to someone else. However, it was not clear whether Mr Taheri’s evidence related to orders being cancelled before, or after, a binding contract of sale was in place. Moreover, even if various suppliers had been understanding in the past, there was always the risk that these two particular suppliers would not be so accommodating. Furthermore, Aria’s actions in deals 10 and 12, in which the goods had been released to the customer prior to full payment, operated to increase that risk (although Aria said that these were mistakes: see para 124 below). But all this demonstrates, in our view, that the FTT was fully entitled to identify a “significant” commercial “risk” even if that risk had not actually resulted in a loss to Aria.[71]Having identified the “risk”, the question for the FTT was what, if any, inference should be drawn from the fact that Aria performed little due diligence on its customers. Conceptually this was similar to the issue, described at para 68 above , that the FTT was addressing when considering why Aria performed limited due diligence on suppliers. Aria was inviting the FTT to draw no adverse inference (on the basis that it expected customers to pay in advance and no adverse consequences had arisen in the past when customers had pulled out of deals). Other more negative inferences were possible: for example, that Aria realised that it was not running the ordinary commercial risks of non-payment since the transactions were part of an orchestrated VAT fraud. HMRC put to Mr Taheri an allegation that all of his talk of due diligence was “hollow” because he knew that, irrespective of when the goods were despatched, Aria would get paid. 23[72]As a purely linguistic matter, the FTT did overstate matters when it said at [343] that the absence of customer due diligence pointed “only” to an awareness of fraud. In a multi-factorial assessment of competing inferences that should be drawn from a large volume of facts, it will seldom be the case that a single piece of evidence points “only” towards one conclusion. However, we do not accept Mr Firth’s submission that this in itself demonstrates that the FTT had closed its mind to Aria’s case or had refused to take into account relevant considerations that might indicate a conclusion that Aria was unaware of connection to fraud. The FTT had pointed to the facts that Mr Taheri had emphasised that in general due diligence on customers, and not only on suppliers, was important; and that the risks involved in these transactions in particular were significant. When the Judgment is read as a whole, it is clear that the lack of due diligence on customers in the disputed transactions was considered alongside the wholly inadequate due diligence on the suppliers in leading to a conclusion of knowledge.[73]The absence of checks as regards deal 3 does not vitiate this conclusion. As the FTT noted, at [354], once regard is had to the details of that transaction it was very different. The supplier, Digimate Ltd, was a long-standing trading partner of Aria and the goods purchased were Digimate monitors, i.e. the seller was an associated company of the manufacturer.[74]Mr Firth criticises the FTT for ignoring a relevant consideration (namely the inevitable difficulty that Mr Taheri would have had in remembering conversations that took place a long time ago). That criticism is misplaced. The FTT could not have failed to be aware that all events relevant to the appeal took place a long time ago. The FTT heard the witnesses give evidence whereas we can only read a transcript of that evidence. It was pre-eminently for the FTT to assess what was said by each individual witness based on the totality of his evidence. Thus, we note that on another point the FTT recorded that Mr Taheri (like Mr McFadden) claimed to recall important information when they gave oral evidence which was absent from their written statements and that it found their evidence on that point was untrue: see at [350]. It would be wrong for us to substitute a different inference on the implication of the passage of time from that of the FTT.[75]Accordingly, we consider that the FTT’s factual conclusions from the lack of meaningful due diligence, including customer due diligence, were fully open to it. Further, we note that the FTT also relied on this for the alternative finding that Aria should have known of the connection to fraud: [368]. Even if, contrary to our view, the FTT’s approach is flawed as alleged in justifying a finding of actual knowledge, it seems to us that it can manifestly support the conclusion as to means of knowledge. Treatment of box numbers (point 12 of Mr Firth’s skeleton argument)[76]The Judgment also contained findings on a different aspect of due diligence, namely the way that Aria recorded and dealt with the unique numbers that appeared on boxes of goods it purchased.[77]At [173] - [177], the FTT referred to evidence that indicated that Aria recorded, or asked its agents, to record details of box numbers. However, having recorded those 24 details, it did nothing with them. The FTT drew the following conclusions on the treatment of box numbers, at [345]:
“345. We reached a similar point on the issue of box numbers which were retained by the Appellant but were put to no use. Mr McFadden’s oral evidence on the matter was poor; having initially asserted that the box numbers were checked against each other he ultimately accepted that he was “just performing a function” and “checking they were all there…I don’t recall doing anything else with them.”
Mr Taheri, who had included box numbers on his checklist of due diligence stated that nothing was done with the numbers. We found that this was indicative of the Appellant’s attitude to due diligence generally; that it was a box-ticking exercise of no substance. No real assessment was made from the information collated which in our view supported knowledge on the part of the Appellant that the deals were contrived; clearly there would be not be any need for a knowing participant in the fraud to undertake any effective or thorough due diligence.”[78]Mr Firth criticises the FTT’s reasoning in the following respects:(1) The evidence demonstrated that Aria was not aware of any need to maintain a database of box numbers and check whether any of its transactions involved it purchasing goods it had purchased before. HMRC had asked Aria to keep a record of box numbers and Aria did so. Therefore, at most Aria’s conduct demonstrated that it did not understand why HMRC made their request; it could not support an inference that it knew its transactions were connected with fraud.(2) HMRC had not pleaded, nor put to Aria’s witnesses, that its failure to perform checks using the box numbers it had recorded indicated it knew its transactions were connected to fraud. In those circumstances it was not open to the FTT to make the finding it did.(3) HMRC had only argued that the way Aria dealt with box numbers indicated that it was “indifferent” to whether it was connected with fraud. The FTT had gone further and had argued it indicated actual knowledge of fraud.[79]Overall, we reject Mr Firth’s criticisms. As a matter of pleading, at paragraph 31.6 of HMRC’s Statement of Case, HMRC had set out their case that a business trading in computer chips should have kept a record of box and lot numbers. In cross-examination, Mr Puzey had put it to Mr Taheri that the purpose of recording box numbers was to enable a trader to check whether it was dealing in goods that had been “carouselled”. Therefore, the significance of the reason for recording box numbers was put to Mr Taheri.[80]The evidence therefore indicated that Aria was recording information and doing nothing with that information. The relevant question for the FTT was what, if any, inference should be drawn from this apparently pointless action. One possible conclusion was that no inference should be drawn: Aria recorded box numbers simply because HMRC asked it to do so. Another possible conclusion was that, because Aria 25 knew its transactions were connected with fraud, it realised it needed to present HMRC with superficially thorough due diligence to obtain its input tax credit, which explained why it was prepared to perform such a pointless “box-ticking” exercise without query or complaint but knew that there was no point making meaningful use of the box numbers. This was relied on by Mr Puzey as one of the ten factors that he submitted were relevant to knowledge.[81]Mr Firth’s criticisms effectively amount to an argument that the FTT should have drawn the neutral inference rather than the negative inference. That is pre-eminently a matter of evaluative assessment for the FTT. There is nothing irrational in the FTT drawing the negative inference in all the circumstances. Nor is the FTT bound by the way submissions on the part of HMRC are framed. It is entitled, in making its assessment of the evidence, to draw its own conclusions.[82]Overall, we are satisfied that the FTT was entitled to the inference it drew. The issue is very similar to that of due diligence discussed at para 68 above. In different circumstances, the FTT might well have concluded that maintaining a pointless record of information demonstrated nothing more than a desire to do what HMRC asked. However, in the light of all the evidence, the FTT concluded that the appropriate inference was more negative and that conclusion was open to it. Absence of commerciality (Points 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 30, 31, 32, 33 and 35 of Mr Firth’s skeleton argument) The FTT’s findings of a lack of commerciality[83]The FTT made several findings that Aria’s business lacked commerciality. At [347], the FTT commented on the rapid growth of Aria’s turnover saying:
“347. We concluded that the Appellants’ turnover figures from the broker deals were, on any view, significant. In our view any reasonable businessman would be fully aware of his turnover, at least in general terms, and we were satisfied that the rapid growth would have been obvious to Mr Taheri. When viewed in the context of the company’s trade as a whole, the contrast with the manner of retail trading, the apparent ease with which deals took place and the short period over which such significant sums were achieved we took the view that the reasonable businessman would have wished to understand why such profits could be made in circumstances that were too good to be true. In our view the mere ability of the Appellant to export goods does not explain the significant and consistent profits achieved. We found that the only reasonable explanation for Appellant’s sudden success in brokerage deals that reaped such large rewards simply as a result of trusted relationships was not opportunity but contrivance and we found that the circumstances were such that the Appellant must have known of this fact.”
The FTT also referred to Aria’s “consistently high profits” at [368]: 26 “368. The factors identified above, from which we inferred the Appellant’s actual knowledge, would in our view also support a finding of means of knowledge. The deals at the time were quite clearly “too good to be true”; something which we were satisfied would have been obvious to Mr Taheri, Mr Harasiwka and Mr McFadden for example the consistently high profits, the little amount of work done to earn those profits, the casual way the business was conducted without contractual terms and the woefully inadequate due diligence which provided no assistance in assessing the integrity of the transactions. In our view a reasonable businessman would have undoubtedly queried these features of the broker deals and would not have reached the conclusion as asserted on behalf of the Appellant that the deals were simply opportunistic.”[84]There had been disputed factual evidence (see [236] - [240] of the Judgment) as to whether Aria knew anything about the “Gigapro 4074” that it was buying and selling in Deals 1 and 2. The FTT said at [350]:
“350. We found Mr McFadden’s oral evidence regarding the product unpersuasive; he was vague as to where he had found out about the product citing the passage of time as the reason yet he was able to provide information about the product which was consistent with that given by Mr Taheri one month earlier. We found unconvincing that neither witness had made reference to this information in their written statements but recalled the information in oral evidence so many years later. We rejected the evidence as untrue; more about which we will say in due course.”
And further, at [351], in the context of discussion about inspection reports (see at para 108 below), the FTT said:
“… the Appellant on Mr Taheri’s admission, knew little or nothing about the goods it was trading, ….”
[85]There was competing evidence before the FTT as to whether, when CPUs were described as “OEM”, that meant that they were in “tray” form or in “box” form8. Dr Findlay’s evidence was that the “OEM” description meant only that the CPU in question was being sold by an “original equipment manufacturer” (such as Intel) but the CPUs being sold could be in either “tray” or “box” form. Mr Taheri’s evidence was that the “OEM” designation referred specifically to CPUs by an original equipment manufacturer in “tray” form. The FTT evidently preferred Dr Findlay’s evidence and concluded that an inadequate description of goods in documentation indicated a lack of commerciality, at [357]:
“357 We did not accept Mr Firth’s submission that Dr Findlay’s evidence was misleading and we did not doubt the truthfulness of it. Our conclusions from the evidence were that the description of the goods on the documents, which we accept could constitute the contractual documentation, taken together with the lack of any other record of 8 “Tray”
CPUs are sold in bulk format in the form of a tray and are designed to be used in a system manufacturer’s production line. By contrast “box” CPUs are aimed at retail purchasers and tend to include a single CPU in a box. 27 agreement of the precise specification, was not indicative of normal commercial trading and gave rise to significant risk. We concluded from this that any reasonable trader would seek to minimise that risk by ensuring that its documents, particularly those said to form the basis of the contract between the parties, contained sufficient information to correctly identify and match the goods. In the absence of such important descriptions, taken together with the nature of trade we were satisfied that a reasonable businessman would have concluded that the trade was not normal arbitrage trading nor` opportunistic. We were satisfied that the Appellant was not concerned about the documentation or contents thereof as it was aware that the deals did not fall within the legitimate grey market”[86]At [349], the FTT found that the apparent absence of negotiations was also indicative of a lack of commerciality as follows:
“349. The evidence as to important terms when entering into transactions, for example regarding dates of payments, prices or return of faulty goods was vague and there was a notable absence of any record of agreements, verbal or otherwise or features such as offers, counter-offers or discussions about payment which we would expect to see in commercial arms length transactions of such high value. The circumstances in which the deals came about and were undertaken was vague and we noted that the Appellant had not carried out this type of deal for a number of years. Despite the extensive evidence we were unable to form a clear picture as to the way in which deals were actually negotiated on a day-to-day basis in the period in question; there were no details of the individual negotiations that might have been expected to have been conducted with suppliers and customers, only general assertions that there were telephone conversations messages and many MSN messages. We found that the manner of trading lacked commerciality; in our view any trader would ensure that the details of his purchase and sale were recorded in order to guard against receiving goods or having goods returned as incorrect. We concluded that the absence of such details between the parties was indicative of knowledge that the deals were contrived.”
The criticisms of the FTT’s findings on commerciality[87]Mr Firth criticised the FTT’s conclusions at [347] for the following broad reasons:(1) The evidence before the FTT demonstrated that Aria’s brokerage business was seasonal with “ups and downs”. The finding of a “sudden increase in turnover” in the 07/06 period failed to take this evidence into account and was nothing more than a statement that, in the 07/06 period, Aria undertook brokerage transactions. The findings related to the profits that Aria was able to enjoy were inadequately reasoned because the FTT did not specify what profit figures it was referring to.(2) The evidence before the FTT demonstrated the disputed transactions were not “easy” to put together and much could go wrong in the implementation of those deals. The FTT’s conclusion to the contrary, 28 including its conclusion that there was little work done, was therefore inadequately reasoned, irrational and against the weight of the evidence.(3) HMRC did not plead that an increase in turnover or “sudden success” were relevant to Aria’s knowledge of fraud or put that point to Aria’s witnesses.(4) The FTT did not explain what it meant by “consistently high profits”. There was no evidence before the FTT that Aria’s profits were suspiciously high. Dr Findlay had stated that brokers in the “grey market” could expect to make gross margins of some 25% and Aria’s margins were much lower.(5) The reference to profit being received over a “short period” failed to address Aria’s evidence that, because brokerage deals offered a low margin, they had to be done in large volume to generate a profit.(6) Aria had never argued that the “mere ability to export goods” enabled it to carry on its brokerage business. Rather, as noted at footnote 6 to para 64 above, it relied on its large network of contacts. The FTT’s reference to an assertion that Aria had never made demonstrated that it was ignoring Aria’s case on this issue.(7) The evidence before the FTT demonstrated that it had achieved no “sudden success” in its brokerage deals. It carried on a successful brokerage business before the 07/06 VAT period and continued to do so afterwards.(8) The FTT failed adequately to explain what contrast it had identified between Aria’s retail business and its wholesale business. HMRC’s officer, Mr Bailey, had accepted in cross-examination that the two businesses were not readily comparable and Dr Findlay had accepted that brokerage businesses are run differently from retail businesses. Therefore, the FTT’s conclusion as to the significance of the contrast was against the weight of the evidence.[88]Mr Firth made the following criticisms of the points the FTT made at [349] on pre-contract negotiations:(1) There was no “vagueness” about the way that the deals came about. Aria had explained in its evidence that it had a large network of contacts and periodically sent a spreadsheet to its contacts indicating goods that Aria could sell (after Aria had first purchased those goods from its network of suppliers). The FTT ignored this evidence.(2) HMRC had not pleaded, or put to Aria’s witnesses, any assertion that the absence of details of negotiations indicated that it knew its transactions were connected with fraud.(3) Aria had explained in its evidence its practice of giving sales quotes to potential customers as part of the process of negotiating deals. If a deal resulted from those negotiations, the sales quote would disappear from Aria’s computer system (and be replaced with an invoice). The FTT ignored this evidence and simply made a general conclusion to the effect that there was “vague” evidence about matters such as price negotiations. 29(4) Similarly, the FTT ignored Aria’s evidence that discussions about price largely took place over the telephone which explained why there was relatively little documentary evidence as to how price was agreed.(5) Aria had explained in unchallenged evidence that it had never come across a single instance of a CPU in which it was trading being faulty and, in that evidence, had explained what would have happened if a CPU were faulty. The FTT was wrong to conclude that evidence relating to faulty goods was vague and, in doing so, took into account an irrelevant consideration.[89]Mr Firth’s criticisms of the findings relating to Aria’s knowledge of the goods were as follows:(1) Mr Taheri had only accepted in cross-examination that he was not aware of one specific product (the “Gigapro 7074” computer controller) before Supreme had offered it for sale in deals 1 and 2. All other products in which Aria dealt in deals 4 to 11 were well-known Intel processors of which Aria was fully aware and the contrary was never suggested to any of Aria’s witnesses in cross-examination.(2) The FTT did not explain adequately what it meant by the descriptions of the goods in the contractual documentation and why it felt those descriptions were not indicative of ordinary commercial trading. This point was neither put to Aria’s witnesses nor pleaded.(3) It was not put to Mr Taheri in cross-examination that describing CPUs as “OEM” was imprecise. The FTT did not give adequate reasons for preferring Dr Findlay’s evidence (that the term “OEM” was imprecise) to that of Mr Taheri. In particular, the FTT did not refer to the fact that, in cross-examination, a document was put to Dr Findlay which, in Mr Firth’s submission, indicated that Intel agreed with Mr Taheri on the issue and, that Dr Findlay ultimately accepted that the designation “OEM” referred to “tray” processors.(4) In any event, the question was not what Dr Findlay thought the description of the products meant, but whether it was adequate for Aria’s purpose. Even if Aria had used a description of the goods that differed from that used by others within the industry, that could not sustain a conclusion that Aria “was not concerned about the documentation”. Therefore, it followed that, at [357] of the Judgment, the FTT had ignored relevant evidence and had reached an irrational conclusion. Having done so, in determining that the description of goods used in contemporaneous documents pointed in favour of an inference that Aria knew its transactions were connected with fraud, the FTT had taken into account irrelevant considerations. The FTT’s findings at [347] relating to Aria’s turnover[90]The FTT was clearly aware that Aria considered its brokerage business to be seasonal: see at [184]. In any event, the fallacy in Mr Firth’s contention that it was not 30 open to the FTT on the evidence to find that there was a “sudden increase” in turnover is demonstrated by a graph produced in Mr Taheri’s witness statement: this showed clearly that turnover in 07/06 increased sharply, not just in comparison with the previous VAT period but in comparison with all of Aria’s VAT periods since 10/03.[91]Mr Taheri had given unchallenged evidence to the effect that much could go wrong when putting together brokerage deals: for example, suppliers or customers could pull out of the deal or customers could fail to pay. The FTT did not expressly refer to this evidence, but it was not obliged to refer to all Aria’s evidence as otherwise its decision would have been unmanageably long. However, at [183] - [185], the FTT referred to Mr Taheri’s evidence that the brokerage side of the business did not require extensive human resources. Only three individuals spent any time on brokerage deals and, as noted at para 121 below, none of those individuals claimed to spend a large amount of their working time on them. At [180], the FTT had referred to Mr Bailey’s evidence that half of Aria’s turnover in 07/06 came from brokerage deals and it was entitled to conclude that there was a relevant contrast between the brokerage business and the retail business that needed several hundred employees to generate the same amount of turnover in the period We consider that there was manifestly sufficient basis for the FTT reasonably to find that these brokerage deals involved relatively little work. This finding involves no error of law.[92]Although HMRC did not, in their Statement of Case make reference to Aria’s increased turnover, or what they saw as Aria’s “sudden success” in its brokerage business, as we noted above this was apparent from Mr Taheri’s own witness statement. Further, the sudden and very significant increase in Aria’s turnover resulting from the 11 disputed transactions was specifically put to Mr Taheri in cross-examination, where he confirmed that these transactions contributed some £3.4 million to Aria’s turnover of £9.9 million in period 07/06. Mr Taheri was also cross-examined about how his company was suddenly purchasing these large quantities of goods from Supreme (from whom Aria had only made much smaller purchases) and from Ashtec (from whom Aria had never previously purchased) and selling them to three customers (Mona, Mitz and Silver Pound) with whom it had had no prior contact. Moreover, it was specifically put to Mr Taheri that in all the circumstances deals 1 and 2 were “too good to be true” although he did not accept that (saying only that he “might have been pleasantly surprised” and that it was an “unusual occurrence”).[93]As to profit margin, the analogy that Mr Firth seeks to draw with the profits of grey market brokers is a false one. Dr Findlay’s evidence, which the FTT clearly accepted, was that Aria was not pursuing any of the “legitimate” grey market opportunities identified at [250] of the Judgment. There was, therefore, no reason to expect it would obtain the margins that operators in the legitimate grey market could obtain. Accordingly, the fact that operators in the “grey market” achieved higher margins cannot call into question the FTT’s conclusion, in essence, that the disputed transactions involved Aria obtaining relatively large sums of money for relatively little effort.[94]The fact that Aria had never suggested that its ability to export goods was the reason it could carry out its brokerage business does not demonstrate that the FTT ignored Aria’s evidence. The FTT was not obliged to set out the profits to which it was referring 31 or resolve the difference of opinion between Mr Taheri and Mr Bailey referred to at [210] of the Judgment as to whether Aria’s profit margins were “consistent” or not. The FTT had found that turnover increased significantly in 07/06, that Aria received large sums of money in connection with just 11 transactions that took place with customers it did not really know and with suppliers who were trading in very much larger volumes than they had previously. In our view, that can fairly be described as involving both the receipt of profit over a short period and “sudden success”.[95]Altogether we think the submissions under this head are yet another example of failing to see the wood for the trees and to read the Judgment as a whole. The FTT was fully entitled to find on the evidence that the disputed transactions, objectively viewed, were “too good to be true” and that a reasonable businessman would have sought to understand how that could be achieved. Looking at all the deals together and the explanation being offered, the FTT was entitled to reach the conclusions it expressed at [347]. The FTT’s findings at [349] relating to negotiations[96]We do not consider that HMRC failed adequately to put their case such that the FTT was not entitled to rely on the lack of negotiations prior to these contracts. At paragraphs 40.5, 40.9 and 40.10 of their Statement of Case, HMRC made it clear that they regarded the fact that Aria was able to buy and sell large quantities of goods on a “back-to-back” basis without formal written contracts as indicators that Aria was involved in a contrived trade. Aria could have been in no doubt that HMRC were arguing that it knew or should have known it was party to contrived transactions, rather than genuine commercial trading. Indeed, it is significant that Aria sought as a means of rebutting this contention to put in evidence, in the form of Mr Taheri’s notes, that there were substantive negotiations before these transactions were concluded. HMRC put to Mr Taheri an assertion that these notes could not be relied be relied on as they did not accurately record what had happened. Thus, we think it is clear that the question of how the deals were concluded was fairly in play, and Aria was well aware of the issue.[97]The FTT devoted 18 paragraphs of the Judgment to discussion of the notes (see [358]-[366]), in which it expressly held that Mr Taheri’s explanation for failing to mention the notes earlier was untruthful, and that the evidence as to the circumstances in which the notes came to light was “vague and varying.” In consequence, there was no reliable evidence of any pre-contract negotiations. It might perhaps have been preferable if the Judgment had referred in [349] to the fact that Mr Taheri had produced notes which sought to describe negotiations but that they could not be relied on, with a cross-reference to the later discussion of that evidence. But reading the Judgment as a whole, the view the FTT reached about pre-contract negotiations and its reasons for doing so are sufficiently clear.[98]In paragraph [349], the FTT is not simply commenting on the detail of the procedure by which Aria’s sales quotes became purchase invoices, or its procedure for dealing with faulty CPUs. The FTT’s core finding, which is clear from reading paragraph [349] as a whole, is that the way in which Aria carried out the 11 disputed transactions was uncommercial and that this pointed in favour of an inference that Aria knew those 32 transactions were connected with fraud. The finding is an important one. If we considered that the FTT had ignored Aria’s evidence in reaching it, or had taken into account considerations which any reasonable tribunal would have regarded as irrelevant, that would amount to an error of law. However, for the reasons set out below, we are satisfied that there was no such error of law and the conclusion which the FTT expressed at [349] was available to it.[99]First, the fact that the FTT did not refer expressly to the way in which sales quotes turned into purchase orders, or the evidence as to the absence of faulty CPUs, does not mean that the FTT “ignored” that evidence. A tribunal is not obliged to refer to every piece of evidence in setting out its reasoning., Nor did the FTT “ignore” evidence that Aria largely conducted its brokerage business over the telephone: that point is mentioned at [349]; or Aria’s evidence about its network of contacts: see both [184] and [349]. Mr Firth’s real complaint is that the FTT did not conclude from this, and other evidence, that these transactions were part of a normal commercial business.[100]Apart from the lack of reliable evidence about negotiations, the point made by the FTT at [349] is that if a trader was entering deals of this scale on a commercial basis, it would expect to make a record of the essential terms, both of the purchases and the sales. That view is strengthened, not weakened, by the fact that the contracts may have been oral. We might add that it applies with still greater force when the trader is buying from suppliers with whom it had never dealt on remotely this scale (and in the case of Ashtec, not for several years) and selling to customers with whom it had never dealt before. We were taken at the hearing of the appeal to the contemporary documents concerning some of these transactions. There were no e-mails confirming the arrangements, only purchase orders, invoices, instructions to the freight forwarders to release the goods, and inspection reports (as to which see paras 107-113 below).[101]Aria needed few members of staff to conduct this business and those members of staff who did conduct it seemed unsure as to fundamental questions associated with the business such as when customers were supposed pay for goods. For example, Aria’s invoice to Mona, its customer in deal 4, stated that payment was to be due “by courier inspection”. Having read the relevant parts of the transcript, we consider Mr Taheri’s evidence as to what this term meant, for whom this inspection would be carried out, and how Aria would know when Mona was due to pay so that goods could be released, can fairly be described as “vague”. Mr Harasiwka also did not know what “payment by courier inspection” meant: see at [226]. Mr Firth argued that questions such as this should have been put to Mr McFadden who was responsible for putting the deals together. However, Mr Harasiwka told the FTT that “the specifics of the deals” were the preserve of Mr Taheri as well as Mr McFadden: [223]. We consider that the FTT was entitled to regard the fact that Aria’s managing director was not able to give a clear answer to questions as fundamental as when Aria expected to be paid in these very substantial transactions, in particular when it was dependent on such payment in order to pay its suppliers, as an indicator that the disputed transactions were lacking in commerciality.[102]The FTT also had available to it the expert evidence of Dr Findlay (referred to at [248] - [258]). The FTT clearly considered, and was entitled to consider, that his 33 evidence pointed against the conclusion that Aria’s brokerage business formed part of the legitimate “grey market” in CPUs. The FTT’s findings at [350] and [351] relating to Aria’s knowledge of the goods and the description of goods in contractual documents[103]We do not accept that there was any obscurity about what the FTT meant in referring to the description of the goods in the documents. As [357] makes clear, this was based on Dr Findlay’s evidence, and the FTT summarised, and indeed quoted from, Dr Findlay’s evidence in this regard earlier in the Judgment: see at [259]-[260]. Anyone reading the Judgment as a whole would be in no doubt as to the criticisms of the description to which the FTT was here referring. It is of course correct that Aria challenged Dr Findlay’s evidence as to the ambiguity in the description “OEM” in the contractual documents, and that Mr Firth put to him a document from Intel which suggested that Intel thought “OEM” and “tray” CPUs were the same thing. However, having read the transcript of Dr Findlay’s cross-examination on this point, we do not see that Dr Findlay ultimately came to accept this proposition. On the contrary, as we understand his evidence, Dr Findlay said that the Intel document divided CPUs into two categories: “boxed” CPUs (which were sold with fans to retail purchasers) and “tray” CPUs (which were sold wholesale to system manufacturers). Dr Findlay then explained that there was a third category of CPUs which were sold wholesale to “mid-market assemblers” in boxes that did not contain fans. That third category of CPU was not a “tray” CPU but, since it was sold wholesale by OEMs could still be described as “OEM”. As we understand it, this was why, notwithstanding the Intel document, Dr Findlay maintained that “tray” CPUs and “OEM” CPUs were not simply synonyms for each other. It was of course for the FTT to assess that evidence and decide what weight to place upon it (and we note that Aria’s expert, Dr Billing, did not address this point), and we reject the suggestion that the FTT misunderstood the evidence.[104]As for the assertion that this was not put to Aria’s witnesses or pleaded, the question is whether there was any unfairness to Aria. In our view, there clearly was not. It is manifest from HMRC’s pleaded case that they contended that the disputed transactions were not part of a legitimate “grey market”. Dr Findlay’s written report expanded on this and set out in detail (at para 5.10) the features which he considered were the “minimum requirements” in order to identify a CPU, including whether it was in box or tray. That report is dated 12 October 2011, some three years before the hearing in the FTT. Aria would therefore have been well aware of this point, and indeed obviously obtained the Intel document so as to challenge Dr Findlay. Nonetheless, Aria did not address the point in its own written evidence in chief (Mr Taheri’s consolidated statement is dated April 2014). Moreover, when Mr Taheri gave oral evidence, he was expressly given an opportunity to comment on the point raised by Dr Findlay about “OEM” and boxed CPUs.[105]Having decided that Dr Findlay was correct and that the term “OEM” was imprecise, the FTT needed to decide what inference to draw. Mr Firth’s complaint, referred to at para 89(4) above, is effectively that the FTT should have drawn a neutral inference (that the description was good enough for Aria’s purposes, even if it was imprecise) over the negative inference (that Aria’s use of an imprecise term suggested 34 that it knew it did not matter how it described the goods in contractual documentation). It was a matter for the FTT to evaluate what inference to draw and there is no basis for us to interfere with its conclusion.[106]Turning to Aria’s knowledge of the products it was selling, we agree that Mr Taheri had not accepted that he had little knowledge about the CPUs that Aria sold in deals 4 to 119. To that extent, the statement in [351] that Aria “knew little or nothing about the goods it was selling” is erroneous. It should have been qualified as referring only to the “Gigapro 7074”, the subject of deals 1-2. Although that amounts to an error , we do not regard it as material. Moreover, as regards the Gigapro 7074, the FTT rejected the evidence Mr McFadden gave as untrue. That was a serious finding. Despite the numerous other challenges to factual findings which Aria advanced in this appeal, this was not challenged. Inspection reports (Point 25 of Mr Firth’s skeleton argument)[107]Inspection reports were an issue in the appeal in two distinct, but similar, respects:(1) Aria produced written reports of inspections undertaken by Imex Logistics addressed to it in relation to deals 4 to 11. On deals 1 and 2, Aria could not produce a copy of an inspection report addressed to it. However, it did have copies of reports prepared by Alpha International Freight Forwarders Ltd (“Alpha”) that were addressed to 4A Developments Ltd. At [240] and [241], the FTT recorded the explanation Aria’s witnesses gave: Mr Taheri speculated that Aria must have relied on a verbal inspection report. Mr McFadden said that he had either received a written report or had relied on a verbal report.(2) At [226], the FTT referred to evidence that for some of the disputed transactions goods were released before receipt of the written inspection report. Mr McFadden’s evidence was that Aria must, in deciding to release goods, have relied on a verbal inspection report.[108]At [351], the FTT rejected Mr Taheri’s evidence as mere speculation and Mr McFadden’s evidence as unconvincing:
“351. As regards the inspection reports we found that the content provided scant detail as to the goods. This was striking given that the Appellant on Mr Taheri’s admission knew little or nothing about the goods in which it was trading, it never took physical possession of the goods and the inspections were purportedly carried out by freight forwarders upon which the Appellant conducted no due diligence. We found the evidence of Mr McFadden that verbal reports would have been obtained prior to a written report unconvincing and we found the 9 At paragraph 168 of his witness statement, Mr Taheri referred to other products as “popular products at the time” and was apparently not cross-examined on that point. However, in paragraph 169 of his witness statement, Mr Taheri acknowledged that he had not heard of the Gigapro 7074 until it was offered for sale to Aria by Supreme. 35 evidence of Mr Taheri on the issue amounted to no more than speculation and assumption.”
[109]Mr Firth’s criticisms of these sections can be summarised as follows:(1) The Judgment was inadequately reasoned. It should have spelled out in detail why it was rejecting the evidence of Aria’s witnesses or why it considered there were “scant details” of the goods.(2) The FTT’s conclusion that there were “scant details” of goods in the Appellant’s inspection reports was unreasonable since companies performing the inspections described goods by reference to “stepping codes” rather than full product descriptions.(3) It was not suggested to Aria’s witnesses in cross-examination that a failure to perform due diligence on freight forwarders indicated knowledge or means of knowledge of fraud.[110]We reject Mr Firth’s criticism that the FTT gave insufficient reasons for rejecting the evidence of witnesses. The FTT summarised some of the evidence on inspection reports by Mr Taheri at [240] and quoted extensively from the evidence of Mr McFadden at [241]. In Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377, quoted in Pacific Computers at [43], Henry LJ stated:
“Where there is a straightforward dispute whose resolution depends simply on which witness is telling the truth about events which he claimed to recall, it is likely to be enough for the judge (having no doubt summarised the evidence) to indicate simply that he believes X rather than Y; indeed there may be nothing else to say.”
[111]Here, as regards the inspection reports it was not even a question of truthfulness but of the reliability of the evidence of Aria’s witnesses. We see no basis to find that the FTT had to go further in explaining why it found Mr Taheri’s evidence speculative and Mr McFadden’s unconvincing. Indeed, in addition to what is set out in the Judgment, we were taken to further passages in the transcripts of the cross-examination of Mr Taheri and Mr McFadden, and while it was primarily a matter for assessment by the FTT, we share their view of the evidence.[112]We think that Mr Firth is mis-reading the reference to the inspection reports providing “scant details” as to the goods. The FTT is not simply making a point about the way the goods were described in inspection reports. Rather, it is making a point as to the quality and nature of inspections that took place as a whole, and the way the inspection reports were used. The evidence before the FTT indicated that the inspections were largely “closed box” inspections which did not involve any of the goods being removed and examined individually. Therefore, the inspection reports apparently involved simply a finding that the right number of boxes with the correct serial numbers were present, and in some cases, but by no means all, whether the boxes were in good condition and whether they contained customs stamps. In our view, given the sums involved, the FTT was fully entitled to its conclusion that the details in these reports were scant, given that the obligation to pay was in some cases dependent on the inspection report. 36[113]We also think that Mr Firth has overstated the significance of the FTT’s finding that Aria performed no due diligence on freight forwarders. The FTT was not saying that this failure on its own indicated knowledge of connection to fraud. Rather, the FTT’s conclusion was that it was surprising that Aria was prepared to trust freight forwarders, whom it did not know, to tell it whether valuable goods that it had purchased were in order or not. Therefore, the point here related to the commerciality or otherwise of these transactions. Significance of “back to back trading” (Point 6 of Mr Firth’s skeleton argument)[114]At [339], the FTT drew the following inference from the “back to back” nature of Aria’s brokerage business: “The transactions took place on a back to back basis and there was no documentary evidence to show that the Appellant was ever left with unsold stock. We did not conclude from this feature of trading alone that the Appellant knew or should have known that the transactions were contrived. However it was a factor that added to the larger picture of the Appellant’s trading as a whole, bearing in mind the questions posed by Moses LJ in Mobilx at [72]: "(1) Why was…a relatively small company with comparatively little history of dealing in mobile phones, approached with offers to buy and sell very substantial quantities of such phones? (2) How likely in ordinary commercial circumstances would it be for a company in [the Appellant’s] position to be requested to supply large quantities of particular types of mobile phone and to be able to find without difficulty a supplier able to provide exactly that type and quantity of phone? (3) Was [the supplier] already making supplies direct to other EC countries? If so, he could have asked why [the supplier] was not making supplies direct, rather than selling to UK traders who in turn would sell to such other countries. (4) Why are various people encouraging [the Appellant] to become involved in these transactions? What benefit might they be deriving by persuading [the Appellant] to do so? Why should they be inviting [the Appellant] to join in when they could do so instead and take the profit for themselves?"[115]Mr Firth criticised this aspect of the FTT’s reasoning on the following grounds:(1) There was no dispute that Aria engaged in “back to back trading”: it purchased goods from a vendor and sold those exact goods, in identical quantities, to a purchaser. Aria was not saying that there was documentary evidence demonstrating that it could have been left with unsold stock and, by suggesting otherwise in the quoted passage, the FTT demonstrated that it had misunderstood Aria’s case. 37(2) Dr Findlay’s expert evidence demonstrated that traders in the legitimate grey market engaged in “back to back” trading and this business model was not, therefore, in any way suspicious.(3) The four questions that the FTT quotes from the Court of Appeal’s decision in Mobilx were not relevant to the evaluation of the significance of “back to back trading”.(4) In short, therefore, the fact that Aria had no unsold stock was an entirely neutral factor that was nothing more than a consequence of its chosen business model. The FTT had taken into account an irrelevant consideration in determining that the absence of unsold stock was an indication that Aria knew its transactions were connected with fraudulent evasion of VAT.(5) HMRC did not put to Aria’s witnesses in cross-examination the proposition that back to back trading was relevant to Aria’s knowledge or means of knowledge of fraud.[116]Mr Firth’s arguments involve a significant misreading of [339] of the Judgment. It is of course correct that the absence of unsold stock is a necessary consequence of carrying on a business that involves back to back trading. However, the FTT is not suggesting that a business that trades on a “back to back” basis should be expected to have unsold stock, or that Aria was arguing that there was any possibility of unsold stock. Rather, reading the paragraph as a whole, with the reference to the four Mobilx questions, the FTT is making a point about Aria’s ability to execute these very substantial transactions on a “back to back” basis in the first place and the circumstances in which these transactions took place. The FTT’s point, therefore, relates to the commerciality of the disputed transactions and the similar concerns articulated at [349]. Once that is appreciated, the references to the four questions posed by Moses LJ in Mobilx become clearer. At [339], the FTT is noting that Aria was able, in each of the disputed transactions, to sell the precise quantity of goods that it bought in short order to a single purchaser. At [137], the FTT had noted that Aria’s suppliers in the disputed transactions were trading in greater volume than before and that it did not know its customers. Moreover, Aria entered into the disputed transactions as part of lengthy deal chains of the kind Dr Findlay had said in his evidence (see [251] of the Judgment) he would not expect to see if Aria was dealing in the legitimate “grey market”10. Those were the sort of issues to which the Court of Appeal was referring in at least the first three of its questions. As for the submission that Aria had, in effect, not been given a fair opportunity to respond to this point, that ignores the fact that HMRC had pleaded in their Statement of Case, at paras 34-35, that Aria’s ability to enter into trades on a “back to back” basis in the context of long deal chains indicated it was engaged in “contrived” trading. . 10 Therefore, Mr Firth’s argument that “back to back” trading could feature in legitimate grey market transactions was beside the point given that the FTT had evidently accepted Dr Findlay’s evidence that the disputed transactions were not typical legitimate grey market transactions. 38[117]The FTT could, perhaps, have explained in more detail how the four questions set out in Mobilx related to the specifics of Aria’s disputed transactions. For example, it could usefully have noted that Aria had a track record of dealing in CPUs and was not the kind of “start-up” operation that Moses LJ was referring to in the first of his questions but that the size of the disputed transactions over a very short period was different from its recent trading history. Similarly, it could usefully have explained in more detail how, if at all, it thought the fourth of Moses LJ’s questions arises. However, that minor criticism cannot invalidate the FTT’s reasoning at [339]. Earlier in the Judgment, the FTT made it clear that it realised that Aria was an established business and not a “start-up” operation that generated significant turnover in just a few months after starting trading: see e.g. at [59]-[60], [184]-[185][118]Overall, therefore, we consider that the FTT was entitled to have regard to the way in which Aria’s “back to back” trading took place in assessing the commerciality or otherwise of the disputed transactions. As the FTT there stated, it was “a factor” to be considered in evaluating “the larger picture”. The FTT’s findings that there were “contradictions” in Aria’s case (points 1, 2 and 3 of Mr Firth’s skeleton argument)[119]At [336], the FTT referred to what it regarded as “contradictions” in Aria’s case:
“336. We also found the Appellant’s case was contradictory in parts. By way of example, it was suggested that the deals did not happen “overnight” and that groundwork had been put in to effect the deals. Aside from the fact that we saw no evidence to support this assertion and the oral evidence on the topic was vague, we noted that the witnesses all claimed that the broker deals formed little part of their daily roles. The evidence that Mr Taheri oversaw all of the deals was at odds with his own evidence from which we formed the conclusion that he had little or no involvement in the practical side of carrying out the deals. When we viewed this against the evidence of Mr Harasiwka and Mr McFadden who also gave the impression of knowing little about the detail of deals whilst simultaneously claiming responsibility for them we formed the view that the transactions were carried out with minimal effort and little work involved.”
[120]Mr Firth criticised the FTT’s findings in the following respects:(1) At [367] of the Judgment, the FTT had found that Mr Taheri had the “final say” on transactions and “authorised the deals” but at [336], the FTT was saying something different. Therefore, there was a contradiction in the FTT’s analysis, not in the Appellant’s case. Similarly, while at [336], the FTT had suggested that Mr McFadden and Mr Taheri had little knowledge of the deals, at [367], the FTT concluded that they had a sufficient knowledge and awareness of the transactions to know that they were connected with fraud.(2) In addition, the FTT’s reasoning set out a flawed decision-making process. When the FTT said that there were “contradictions” in Aria’s case, 39 it was engaging in a wholesale rejection of all of Aria’s evidence. It should have given better reasons rather than relying on allegations of “contradiction”.[121]The paragraph could, perhaps, have been expressed more clearly, but read in the context of the Judgment as a whole, it is reasonably clear what the FTT is saying. Paragraph [336] is not primarily concerned with the extent of Mr Taheri’s involvement in the day to day running of the business. Rather the contradiction that the FTT identified was that, on one hand, Aria was asserting that brokerage deals took time and effort to put together but, on the other hand, none of Aria’s witnesses claimed to spend much time on the brokerage business. It is entirely misleading to describe this paragraph as a “wholesale rejection” of Aria’s witness evidence. It says that Aria’s evidence was “contradictory in parts” and this observation is expressly made in addition to the finding at [335] that Aria’s oral evidence was unconvincing because it was vague and speculative. It was for the FTT to weigh the significance of the point made in [336] and its finding involves no error of law. Release of goods without payment (point 26 of Mr Firth’s skeleton argument)[122]At [226] to [228], the FTT referred to evidence that, in two deals, Aria had released goods that it was selling without receiving payment. It was common ground on this appeal that the transactions effected were deals 10 and 12, not deals 5 and 12 as the FTT stated, and that insofar as the FTT was suggesting that no payment had been made that was incorrect as, prior to the release of the goods, Aria’s customer had paid half the purchase price that was due. At [352] the FTT drew the following conclusions from this episode:
“352. We considered the evidence regarding the deals in which goods were released prior to full payment being received by the Appellant; we found the oral evidence of Mr Taheri, Mr McFadden and Mr Harasiwka was unsatisfactory. Mr Taheri claimed not to have been aware of the fact whilst Mr McFadden stated he relied on Mr Harasiwka and his team to be told when payment was made yet Mr Harasiwka could provide no explanation. In deals of such high value where the Appellant’s main comfort was said to come from the lack of risk by receiving payment prior to releasing the goods we concluded that the only explanation for these events was that the inadequacy of the inspection documents, time of payment and release of the goods were irrelevant as the Appellant was fully aware that the deals were contrived.”
[123]Mr Firth criticised the FTT’s findings on the following broad grounds:(1) Aria’s witnesses had, in cross-examination, put forward an innocent explanation of why goods had inadvertently been released before full payment had been made. The FTT was wrong to reject that explanation.(2) It could not be correct that the “only explanation” for early release of the goods was that Aria knew that its transactions were connected with fraud. By engaging in such hyperbole, the FTT demonstrated that it had 40 closed its mind to the innocent explanation for the oversight that Aria had put forward.[124]Both Mr McFadden and Mr Harasiwka were cross-examined on this incident. Mr McFadden said that he had not been aware of the oversight at the time. In summary, his evidence was that he did not have access to Aria’s banking information and so relied on Mr Harasiwka and his team to tell him whether payment had been received. Therefore, he must have received such a confirmation before the goods were released. Mr Harasiwka accepted in that he had no ready explanation for the early release of the goods but noted that it “may have been an error”. He thought that it was possible that someone in his team had misread the account and thought full payment had been made when it had not.[125]The question that the FTT had to address was not simply why goods were released prematurely. The witness evidence was clear that a mistake had been made and that Aria’s normal practice would be to wait until full payment was received before releasing goods. Rather, the FTT needed to consider what, if any, inferences should be drawn from the mistake. One possible conclusion was that it was, as Mr Firth neatly submitted, like a routine mistake made while driving a car that happens simply because of oversight and is of no consequence unless a collision occurs as a result. However, another possible conclusion was that the mistake indicated that Aria knew that, since its transactions were part of contrived chains linked to fraudulent evasion of VAT, the ordinary commercial imperative to safeguard payment was not present. There was clearly evidence before the FTT to justify the adverse evidence which the FTT drew. Having seen all the evidence that Aria was putting forward, it was for the FTT to decide whether it should draw the adverse inference or not.[126]The submission about the FTT’s finding that the “only explanation” was a knowledge of connection to fraud is very similar to that Mr Firth made regarding the FTT’s reasoning in respect of due diligence towards customers: see para 59(6) above. We accept that the word “only” could suggest that no other explanation was possible, which is self-evidently incorrect. But we think that, in context, the FTT meant that having found the explanation offered by the three relevant witnesses from Aria to be “unsatisfactory”, the only satisfactory explanation was that Aria knew the chains were contrived. Accordingly, we do not discern any error of law in the FTT’s approach. The absence of “set procedures” for brokerage transactions (point 4 of Mr Firth’s skeleton argument)[127]At [335] the FTT commented on the absence of set procedures in Aria’s brokerage business:
“335. The oral evidence on behalf of the Appellant was not convincing. The evidence of each of the witnesses was vague and littered with what they “would have done” or “would have been told” yet the clear impression we formed from each of the witnesses, and on Mr Taheri’s own admission, was that there were no set procedures in place for the brokerage deals: 41 “…if we were doing [the deals] often, day in day out, I suppose there would have been procedures in place…” (transcript 15 August 2014 page 35). In those circumstances we found the witnesses’ references to what would have been done was not supported by any evidence of usual practice and was wholly unconvincing.”
[128]Mr Firth criticised this paragraph in the following respects:(1) Mr Taheri made no “admission” that there were no set procedures in place for brokerage deals. He had accepted, in cross-examination, that there were no procedures for doing due diligence on freight forwarders, but the FTT was wrong to extrapolate from this and draw a conclusion that there were no procedures at all. Moreover, all witnesses had referred in their evidence to procedures that Aria followed (for example, dealing only with trusted suppliers, not releasing goods without payment and seeking validation of VAT registration details with HMRC’s Redhill office).(2) By overplaying the significance of the evidence, the FTT had failed to perform a balanced assessment of it and had therefore made an error of law of the kind referred to in HMRC v Glyn [2015] UKUT 551.[129]Mr Firth is correct in saying that Mr Taheri did not admit that there were no procedures at all in place in Aria’s brokerage business. Indeed, we find it difficult to conceive how any business could function without having at least some procedures. However, the FTT’s finding was that Aria had no set procedures in place. This is not a mere verbal quibble. The clear conclusion that the FTT reached is that, despite claiming that it had fixed procedures, Aria frequently departed from those procedures. There was ample evidence available to support that conclusion. For example:(1) As noted at [226] to [228] of the Judgment, Aria’s stated procedure was not to release goods until its customer had paid for them. Mr Taheri had described it as a “deal-breaker” if its customer was not prepared to pay cash in advance: see at [198]. Nevertheless, in two out of the 11 disputed transactions, Aria released goods without having received full payment.(2) Aria said that it required inspection reports before it would pay its suppliers. However, the FTT referred to an instance where goods were released before an inspection report was received: see at [226]. Moreover, the FTT found that, in two out of the 11 transactions, the FTT had no inspection report addressed to it: see at [232].(3) Aria had put in evidence a pro-forma checklist that sought to capture details of suppliers and customers. However, there were no completed checklists relating to any of the disputed transactions: see at [139].[130]Therefore, the FTT did not, as Mr Firth submitted, “overplay” the evidence and reach a flawed conclusion that Mr Taheri accepted there were no procedures at all. It was a matter for the FTT to decide what inferences it would draw from the apparent fluidity in Aria’s procedures. There is no basis, in an appeal on a point of law, to substitute a different inference from that made on the evidence by the FTT. 42 The lack of rigour in Aria’s approach to contractual documentation (points 28 and 34 of Mr Firth’s skeleton argument)[131]At [183] and [186] and [244] to [247] of the Judgment, the FTT referred to evidence which suggested that Aria had not paid attention to what the terms of its contracts might be on the important issues of potential liability and transfer of title.[132]Before the FTT, at least part of Aria’s case was that no particular inference could be drawn from any such lack of thought as many businesses fail to formulate their terms and conditions with the precision that a lawyer might hope for. The FTT expressed its conclusion at [355]:
“355.We considered the submissions of Mr Firth as regards legal title and the intention between traders. However, the issue for us to determine is what was known to the Appellant at the time, not the various legal possibilities. It was quite clear from the evidence that the holding and passing of title was not a matter to which the Appellant had given any thought. Building on the other evidence before us which indicated the lack of concern on the part of Aria as to the commercial risks in its business, we found that this lack of thought on the part of Aria as to the position of legal title was another issue which bore on the inference of knowledge that the deals were not genuine business transactions but were part of a fraudulent scheme.”
At [368], the FTT referred to the “casual way the business was conducted without contractual terms” as part of its conclusion that, Aria “should have known” that its transactions were connected with fraud.[133]Mr Firth criticised the FTT’s reasoning on the following grounds:(1) The FTT had misunderstood his submissions. He was not asking the FTT to consider various “legal possibilities” as to when title passed; he was making the simple point that lots of businesses show a lack of care in their standard form contractual documentation with the result that no particular adverse inference could be drawn even if Aria had also shown a lack of care.(2) In any event, there was evidence before the FTT that legal title would not pass until payment was made (which the FTT failed to address).(3) The FTT had not seen any evidence as to when title might pass in “legitimate” grey-market transactions. Therefore, it was not open to the FTT to infer that Aria’s approach was unusual.(4) It was irrational for the FTT to conclude that Aria had “no contractual terms”. In any event, the way that Aria chose how to enter into contracts cannot support any inference of knowledge or means of knowledge.(5) It followed, therefore, that the FTT’s conclusion that Aria’s approach to the passing of legal title supported an inference that it knew its transactions were connected to fraud was inadequately reasoned and unsupported by evidence. 43[134]Mr Firth is of course correct to say that Aria had some contractual terms, even if those were not reduced to writing. However, this criticism ignores the core of the FTT’s finding which was that Aria’s contractual terms were uncertain (as they were not set out comprehensively in writing) and did not deal adequately with the passing of legal title. We reject Mr Firth’s assertion that such issues were irrelevant to Aria’s knowledge or means of knowledge. Conceptually, Aria could have adopted lax or uncertain contractual terms for the disputed transactions because it realised many transactions it entered into in its brokerage business would be connected with tax fraud so that Aria did not need the kind of certainty on contractual issues that would normally be important. That was by no means the only possible inference. Indeed, since Aria had carried on a brokerage business for some time and followed a similar approach to its terms and conditions in its business generally, the inference might not be strong. However, the FTT was entitled to consider whether it should make such an inference[135]We do not consider that the FTT failed to understand or consider Aria’s submissions. It was also reasonable for the FTT to take into account the fact that the standard-form documentation of Aria and its suppliers posed something of a legal conundrum. If Aria could not get legal title to goods it was purchasing until it paid its supplier, but was proposing to use money received from its customer on the “back to back" deal to pay its supplier, it might not have legal title to the goods at the point at which it was subject to an unconditional contract to sell them.[136]Our own view is that there may have been force in Mr Firth’s submission that the conundrum referred to at para 135 might be of interest to lawyers but would be of little interest to businessmen. However, the FTT was not being asked to make a determination as to how business persons generally might view the legal question of when title passes. It had to decide what, if any, inference should be drawn from the way in which Aria chose to enter into these high value contracts for purchase of CPUs. Aria traded on a “back to back” basis, which made the question of when legal title passed more relevant than it would be to many businesses. Aria did not know its customers in the disputed transactions but was entering into high value contracts with them. In the light of all the evidence, it was open to the FTT to conclude that Aria realised that it did not need terms and conditions dealing with the passing of legal title because such terms and conditions would never be tested since it was engaged in contrived transactions connected with fraudulent evasion of VAT. Had that been the only basis on which the FTT concluded that Aria knew of connection to fraud, we doubt whether it would have been supportable. However, viewed in conjunction with other evidence before the FTT, and given the FTT’s conclusions on that other evidence, we consider that the FTT was entitled to draw the adverse inference it drew. The significance of Aria’s possession of a letter of introduction from Mitz addressed to Sanche (Point 14 of Mr Firth’s skeleton argument)[137]At [162] - [163], the FTT noted that the among documents that HMRC said that they had obtained from Aria was a letter of introduction sent by Mitz to Sanche Technologies Ltd (“Sanche”). The question that arose was how, if at all, that document came into Aria’s possession. In cross-examination of Mr Bailey, it was suggested that in fact the document had never come into Aria’s possession as Mr Bailey had 44 inadvertently caused documents received from someone else to be added to Aria’s case file. Mr Bailey denied that and, at [346], the FTT made the following findings:
“346. As regards the letter from Mitz addressed to Sanche which contained the fax number for MVS Digital we found the Appellant’s evidence contradictory and wholly unconvincing. An email from Mr Harasiwka to Mr Bailey in 2007 explained that: “the bundle of papers provided to you were per those received by us and clearly Mitz transmitted a letter addressed to Sanche Technologies Limited to us in error.”
Mr Taheri’s first witness statement also explained this as an administrative error. In evidence the Appellant sought to argue that the document may not have been in its possession and that one possible explanation was that HMRC had somehow mixed documents from another trader’s file on the EF with those of the Appellant. We did not accept the Appellant’s assertion and we were satisfied that the document was one in the Appellant’s possession and which had been given to Mr Bailey by Mr Harasiwka. Having reached that conclusion and there being no explanation, credible or otherwise before us as to why the Appellant would have this document in its possession we concluded that there was no legitimate reason for the letter being in the Appellant’s possession other than that the transactions were contrived.”[138]Mr Firth criticised that conclusion on the following grounds:(1) Aria had, in its witness evidence, put forward an explanation – namely that Mitz had simply made a mistake and sent a letter meant for Sanche to Aria. Therefore, it was incorrect for the FTT to say that no explanation “credible or otherwise” had been put forward.(2) It was irrational for the FTT to assume that Aria could put forward an explanation of how Mitz came to mis-address its letter of introduction. Aria was the passive recipient of that letter, so any explanation had to come from Mitz rather than Aria.(3) Had the FTT turned its mind to the fact that Aria had simply received a document intended for someone else, it should have realised that Aria’s possession of the document did not raise any inference that it knew its transactions were connected with fraud. There is no reason to think that administrative errors are more likely in transactions where everyone knows of connection to fraud than they are in ordinary everyday life where communications sometimes go astray.[139]We agree with Mr Firth that, in this respect, the FTT has not adequately explained its reasoning. Aria had advanced an explanation, namely a simple mistake, so the FTT was wrong to say that there was “no explanation, credible or otherwise”. We can infer that the FTT thought it relevant that the fax header contained details of the fax number of MVS Digital. However, the FTT has not explained why Aria’s possession of a mis-addressed fax indicated that it knew the transactions (or even just deals 1-2 involving Mitz) were connected with fraudulent evasion of VAT. 45 The significance of Deal 3 (Point 27 of Mr Firth’s skeleton argument)[140]One of Aria’s transactions in the 07/06 VAT period (deal 3) was not traced back to a fraudulent evasion of VAT. We have addressed above Aria’s argument that because it performed the same level of due diligence in relation to deal 3 as it performed on the disputed transactions, that demonstrated that the level of due diligence on the disputed transactions was irrelevant to the question whether it knew, or should have known, that those transactions were connected with fraud: see para 73 above. Here, we address Aria’s other criticisms about the conclusions that the FTT drew from deal 3.[141]At [354], the FTT said:
“354. We noted the Appellant’s submissions regarding deal 3 which was verified and repaid. In our view the circumstances of this deal are distinguishable from the remainder in which input tax was denied; the supplier to the Appellant was a longstanding trading partner and the goods were Digimate branded monitors which were manufactured by and purchased from the associated company Digimate Hong Kong. We took the view that the circumstances of this deal did not assist us in determining the issues in this appeal.”
[142]In addition to the point already discussed, Mr Firth made the following criticisms:(1) The FTT’s reasoning was contradictory. It had at [340] rejected Aria’s argument that, since it had a long-standing relationship with Ashtec and Supreme, it had a well-founded trust in the integrity and provenance of goods that they were selling. Therefore, the fact that Aria had a long-standing relationship with Digimate Ltd could not logically differentiate deal 3 from the disputed transactions.(2) The FTT was wrong to regard Digimate branded monitors as being less at risk of MTIC fraud than the CPUs in which Aria was dealing. Mr Bailey had accepted in cross-examination that even branded goods being sold by a company affiliated with the manufacturer could be the subject of contra-trading.(3) Overall, therefore, the FTT had failed to take into account the relevant considerations arising out of deal 3. Had it taken into account those considerations, it would have realised that Aria’s approach to deal 3, which was untainted by any connection to fraud, was no different from its approach to the disputed transactions and that, accordingly, Aria neither knew, nor should have known, that the disputed transactions were connected with fraud.[143]We regard these submissions as wholly misconceived. First, the FTT clearly considered deal 3, and the argument that Aria advanced based upon it, and concluded that it did not assist because its circumstances were materially different. Since in deal 3 Aria was buying goods from an existing and long-term supplier, which was an affiliate of the manufacturer, Aria necessarily had much greater awareness of the goods’ provenance than it did when buying from suppliers who were brokers and who were selling to Aria CPUs in far greater quantities than they had previously (and in the case of Ashtec, from whom Aria had not purchased for several years). Therefore, it was 46 manifestly open to the FTT to conclude that the trust that Aria had in the provenance of the goods in deal 3 was greater than the trust it could reasonably have had in its suppliers in the disputed transactions.[144]Secondly, the FTT did not find that Digimate branded monitors were less at risk of MTIC fraud. That is not part of the FTT’s reasoning at [354], and we note that in its decision refusing permission to appeal, the FTT made clear that it had not considered that the goods in deal 3 could not be used in an MTIC fraud: see at [55]. Nor did Mr Bailey accept in cross-examination that a purchase of branded monitors from Digimate, a company affiliated to the monitors’ manufacturer, was at just as much risk of being connected to MTIC fraud as a purchase of CPUs in brokerage transactions. On the contrary, Mr Bailey simply made the point that deal 3 could not be traced directly back to a tax loss; and that in view of the very different characteristics of deal 3 he was satisfied that it was a bona fide deal and that it was unnecessary to pursue inquires to see if the transactions could involve a contra-trader: Treatment of Dr Findlay’s evidence (Point 29 of Mr Firth’s skeleton argument)[145]At [317], the FTT referred to the criticisms which Aria had made of Dr Findlay’s evidence (though it did not set them out). At [334], the FTT observed that it did not doubt the credibility of Dr Findlay’s evidence but found it of limited assistance. At [357], the FTT rejected the assertion that Dr Findlay’s evidence was misleading.[146]In his skeleton argument, Mr Firth submitted that the FTT was wrong to dismiss Aria’s criticisms of Dr Findlay’s evidence without giving an explanation since “the evidence clearly showed Dr Findlay's evidence was misleading”. However, in his written and oral submissions to this Tribunal, Mr Firth did not seek to explain why Dr Findlay’s evidence was misleading or incorrect beyond referring us to his written closing submissions to the FTT. There, he had made several criticisms of Dr Findlay’s expert evidence, describing it as including “serious errors” and “some seriously misleading comments”. The points raised were highly detailed.[147]In view of the way this submission was advanced at the hearing of this appeal, it is neither practicable nor appropriate for us to consider whether there was any substance in Aria’s criticisms of Dr Findlay’s evidence. This was not a case where the FTT was being asked to choose between the evidence of two competing experts, in which circumstances a tribunal would be expected to set out its reasons. We will only say that here, where there was a great deal of evidence and the FTT expressly said that it derived “limited assistance” from the evidence of Dr Findlay, we consider that the FTT’s approach was acceptable and that it was not erroneous in law for the FTT to express an overall view of Dr Findlay’s reliability and truthfulness without explaining why it rejected the various detailed points on which Aria had alleged Dr Findlay was misleading. 47 GENERAL CRITICISMS[148]In addition to the 36 specific criticisms directed at particular paragraphs in the Judgment, Mr Firth also advanced what he described as “general challenges” to the FTT's approach. Lack of expert evidence[149]The FTT in a number of passages contrasted the conduct of Aria with that of "a reasonable businessman seeking to protect himself from fraud". For example, it did so as regards the lack of checks or investigation by Aria of the integrity of the impugned deals (Judgment, para [344]); and as regards Aria's failure to make any record of the details of the purchases and sales involved.[150]Mr Firth argued that the FTT would have no knowledge of how a typical or reasonable IT component trader would conduct itself, and so was not entitled to draw inferences on that basis concerning Aria's conduct, without expert evidence. In support of that submission, he relied on authorities concerning professional negligence and, in the field of VAT, on the decision of the First-tier Tribunal (Tax Chamber) in Perenco Holdings v HMRC [2015] UKFTT 65 (TC).[151]We reject Aria's submission as misconceived. Aria accepted that it was aware of the prevalence of MTIC fraud in the field of computer components and these 11 deals were of significant size involving sellers with whom Aria had not dealt with at anything approaching that scale and purchasers with whom Aria had not previously dealt with at all. The observations made by the FTT regarding the conduct of a reasonable dealer in such circumstances are a far cry from questions of the standard of care in the exercise of professional skill arising in a professional negligence claim. As for Perenco, that was a so-called 'Fleming claim' for unrecovered input tax as regards payments for professional services in connection with various share issues made by the company over 20 years previously. The question at which the FTT's observations were directed was a specialist issue involving accountancy evidence.[152]In CCA [2015] UKUT 513 (TT), a case on which Aria relied (albeit for a different purpose), the Upper Tribunal (Morgan J and Judge Herrington) quoted at [115] from the judgment of another Upper Tribunal (Asplin J as she then was and Judge Walters) in S&I Electrical v HMRC [2015] UKUT 0162 (TCC)at [65]:
"It is true that the FTT was required to invest the reasonable man for these purposes with the characteristic of being a reasonable businessman with ordinary competence, but in our judgment a reasonable businessman with ordinary competence is not so egregious or specialist a variant of the anthropomorphic conception of justice that the FTT needed evidence of the normal characteristics of legitimate trade in the grey mobile phone market, or any other expert evidence, in order fairly and justly to apply the required impersonal standard."
[153]The Upper Tribunal in CCA proceeded to hold, as one of several reasons for overturning the decision of the FTT, that the judge below had been wrong to decline in 48 the absence of expert evidence to form an assessment of what a reasonable trader would have inferred from the various circumstances (judgment at [118]).[154]It is not without significance that Perenco was the only authority in the VAT field on which Mr Firth was able to rely. The question of whether a trader took precautions "which could reasonably be required to ensure that their transactions are not connected with fraud" (the language used in Kittel at para 51), and if not, how far that is relevant in determining whether they knew or should have known that the transactions were connected with fraud, has arisen in a large number of MTIC cases. It forms part of the assessment of the trader's conduct, considered as a whole. If Aria's submission were correct, expert evidence of the practice of traders in the relevant field would be a common feature of these cases. However, the opposite is the case. Here, the particular aspects on which the FTT considered that Aria's conduct differed from that expected of a reasonable trader in the all the circumstances were matters of commercial common sense and, in our view, well within the tribunal's competence. Findings on matters not put to Aria[155]It was alleged that the FTT made findings of fact that were not put to witnesses, and thus contravened "a basic requirement of natural justice." It is of course well-established that a party in effect 'accused' in an MTIC case should have a proper opportunity to answer serious allegations against it: see for example the decision of the High Court in Mobilx v Revenue & Customs Comrs [2009] EWHC 133 (Ch), at [68], [70]11 .[156]However, where HMRC's case as regards many of the individual factors relied on is not that each individual factor demonstrated that the taxpayer knew or should have known of the connection to fraud, but that the accumulation of factors together demonstrated that the transaction was not legitimate such that the taxpayer had this knowledge, it would be a distortion of the case to require HMRC to cross-examine on each specific factor on the basis, in effect:
"You must have known from this that the transaction was fraudulent?"
. Thus a criticism that selected elements of HMRC's case, or the FTT's findings, were not put to the taxpayer's witnesses must be examined on a realistic basis, having regard to the nature of the case which HMRC were advancing and the conclusions reached in the Judgment. Moreover, it is frequently possible to suggest with hindsight different or fuller questions that could have been put to a witness, and in our view there is no unfairness just because a point was not put in precisely the terms in which a finding is expressed in a judgment, as long as the taxpayer had a fair opportunity to respond to the essential allegation against it which the tribunal proceeds to accept. See Edgeskill Ltd at [147]-[148].[157]We turn to the nine specific factors relied on in this regard in Mr Firth's skeleton argument. However, several of these repeat specific criticisms levelled at particular paragraphs of the Judgment that we have addressed above: 11 The Court of Appeal (para 17 above) did not disturb this conclusion. 49(1) that Mr Taheri had a "full understanding of MTIC fraud": [338]. We have addressed the challenge to that finding at paras 54 to 56 above. As regards fairness, Mr Taheri set out in his written witness statement what he said was the extent of his knowledge and understanding of MTIC fraud, which he said he knew was taking place in his industry: see the quotation at [130]. In cross-examination, it was put to him that he knew about carousel fraud, which he denied. It was also put to him that he had been warned by HMRC of the risk of MTIC fraud (which he accepted), and that he knew from the HMRC Notice that, to guard against fraud, he should carry out checks not only on suppliers but also on customers. We do not accept that Aria lacked a fair opportunity to respond to this point.(2) that the deals on their face carried a significant commercial risk: [343]. Again, we have addressed the challenge to this as inconsistent with the pleaded case at paras 69 to 71 above. As we there explain, the finding was that Aria was prepared to take what looked like a significant commercial risk because it knew it would get paid and so did not regard the disputed deals as carrying the commercial risk which that would normally arise from the way the transactions were structured. Mr Taheri was clearly aware of the point that such a transaction could involve a risk for Aria if the customer cancelled the order, since he addressed that specifically in his witness statement (saying that Aria's experience was that its suppliers would then let Aria cancel in turn). Mr Taheri was cross-examined about the basis of his trust in the two particular suppliers involved in the disputed transactions. But the significant point was the assertion that here Aria was not concerned about risks because it knew it was going to get paid under these deals in any event. That was put to Mr Taheri in cross-examination, albeit in the context of due diligence (and disputed by him). Looking at the matter in the round, there was no unfairness.(3) that Aria knew it should have done something with box numbers: [345]. We think that this misrepresents the Judgment. The point about box numbers was part of the broader point concerning due diligence: i.e., that a prudent trader involved in transactions in these goods, arising in these circumstances, would have carried out various checks to ensure that the transactions were not connected to fraud. In that context, the box numbers could sensibly have been used to check that the goods were not being "carouselled". That was indeed put to Mr Taheri: see para 79 above.(4) and (5): the significant increase in turnover and "sudden success" in brokerage deals: [347]. See para 92 above, where we note that the sudden and significant increase in Aria's business from just these 11 transactions was put to Mr Taheri. Moreover, [347] of the Judgment has to be read as a whole. The inference which the FTT drew was derived from the increase in turnover, the sudden success in brokerage deals, the relative ease with which the deals came about, and the short period over which they were done: it is that combination of circumstances which led the FTT to find that it was all "too good to be true". That inference was put directly to Mr Taheri. He 50 offered an alternative explanation, which the FTT evidently did not accept, but there was no unfairness. (6) lack of individual negotiation: [349]. We have addressed the allegation that this aspect was not fairly put at para 96 above. (7) lack of knowledge of the goods: [351]. The criticism here is that this was not suggested to Aria as regards the goods traded in deals 4-12. That is correct. The Judgment is mistaken in suggesting that Aria did not know about the goods in those deals: see para 106 above, where we conclude that this error was not material. (8) due diligence on freight forwarders: [351]. Again, this is to extract one aspect of a comprehensive paragraph. The FTT did not find that a lack of due diligence on freight forwarders indicated knowledge of fraud. Manifestly, in isolation, it could not and so obviously that was not put to Aria. The relevant point is the way Aria was prepared to rely on inspection by freight forwarders whom it did not know, given that many of their written reports gave scant information and in some cases Aria received those reports only after paying for the goods. It is not suggested, nor could it be, that Aria's witnesses were not cross-examined about the nature of the inspection reports and the way they relied on them. (9) the description of the goods in the documents: [357]. We have addressed the allegation that this aspect was not fairly put at para 104 above.[158]We should add that Mr Firth also advanced in general terms a broad criticism of the Judgment for failing to take into account the effect of the passage of time on the witnesses' recollection. We have addressed that at para 74 above. Lack of reasoning[159]Mr Firth suggested that, even though the Judgment was long, it was insufficiently reasoned. He argued that the FTT spent too much time (paragraphs [65] to [121] of the Judgment) in making findings on matters that were not in dispute, in trying to summarise evidence (at [122] to [279] of the Judgment) and in providing an inadequate summary of the parties’ submissions (at [280] to [317]). It followed that, in Mr Firth’s submission, only paragraphs [329] to [369] of the Judgment were devoted to “making findings of fact and grappling with the issues” and the decision was inadequately reasoned as a result.[160]We consider that general criticism fails to do justice to the decision. The FTT had been confronted with a large volume of evidence, with the parties having very different perspectives both on the relevance of that evidence and the conclusions that should be drawn from it. In those circumstances, it was perfectly sensible for the FTT first to summarise the evidence that it thought was particularly relevant and, having done so, to make findings as to what conclusions should be drawn from it. Mr Firth is quite wrong to submit, as he did on several occasions, that this process necessarily involved the FTT “ignoring” evidence that was not mentioned either in the summary of evidence or in the FTT’s findings of fact. The mere fact that evidence is not referred to in a 51 decision does not mean that it has been “ignored”. Moreover, as noted at para 10 above, the structure of the Judgment is such that the factual conclusions expressed need to be read together with the FTT’s recital of relevant evidence. OVERALL CONCLUSION ON KNOWLEDGE[161]Mr Firth subjected to detailed textual analysis almost every paragraph in the section of the Judgment where the FTT summarised its “Findings of fact [on knowledge]”, and in some cases almost every sentence within a paragraph. When an appeal is advanced by such a toothcomb approach, we think it is important to bear in mind that a finding of knowledge or means of knowledge of a connection to fraud is an evaluative judgment based on an accumulation of primary factors which, taken together, lead to that conclusion although looked at in isolation each factor may permit of an innocent explanation. The FTT’s approach and the adequacy of its reasoning as regards any individual factor must therefore be assessed in the context of all the other factors that make up the overall picture. As Christopher Clarke J stated in Red 12 Trading Ltd v HMRC [2009] EWHC 2563 (Ch), in a passage significantly cited by the FTT (at [10]):
“…in determining what it was that the taxpayer knew or ought to have known the tribunal is entitled to look at the totality of the deals effected by the taxpayer (and their characteristics), and at what the taxpayer did or omitted to do, and what it could have done, together with the surrounding circumstances in respect of all of them.”
[162]We have gone through all of Mr Firth’s 36 points of criticism of the FTT’s reasoning or findings of fact. We emphasise that this is an appeal on point of law, and that it is not for the Upper Tribunal to assess the evidence and decide whether we would make the same findings, but whether the findings made by the FTT were reasonably open to them and sufficiently reasoned. On that basis, we have upheld just one of the 36 criticisms (as regards the document from Mitz which we discuss at para 139 above), and on which we accept no proper reasons were given by the FTT and which should not, in the light of the evidence, have been regarded as relevant. However, having regard to all the other facts and matters taken into account, we think that it is clear that this in itself was not material to the FTT’s overall conclusion: see e.g. the summary at [368]. If the test, contrary to our view, is whether the FTT would inevitably have reached the same conclusion without regard to the Mitz document, we have no doubt that it would. In our judgment, the fundamental question is then whether, on the basis of all the findings of fact properly made, the overall inference or conclusion that Aria knew, or alternatively should have known, of the connection to fraud was one that a reasonably tribunal could properly reach. Only if the answer to that question is No, is there an error of law vitiating the decision.[163]As regards such a multi-factorial assessment based on a large number of primary facts, we bear in mind Lord Hoffmann’s oft quoted observation in Biogen Inc v Medeva PLC [1996] UKHL 18 at [54], when discussing the issue of obviousness:
“The need for appellate caution in reversing the judge's evaluation of the facts is based upon much more solid grounds than professional courtesy. It is because specific findings of fact, even by the most meticulous judge, 52 are inherently an incomplete statement of the impression which was made upon him by the primary evidence. His expressed findings are always surrounded by a penumbra of imprecision as to emphasis, relative weight, minor qualification and nuance (as Renan said, la vérité est dans la nuance), of which time and language do not permit exact expression, but which may play an important part in the judge's overall evaluation. It would in my view be wrong to treat Benmax as authorising or requiring an appellate court to undertake a de novo evaluation of the facts in all cases in which no question of the credibility of witnesses is involved. When the application of a legal standard such as negligence or obviousness involves no question of principle but is simply a matter of degree, an appellate court should be very cautious in differing from the judge's evaluation.”
[164]Referring to that judgment among others, Lewison LJ stated, in Fage UK Ltd v Chobani UK Ltd [2014] EWCA Civ 5 at [114]:
“Appellate courts have been repeatedly warned, by recent cases at the highest level, not to interfere with findings of fact by trial judges, unless compelled to do so. This applies not only to findings of primary fact, but also to the evaluation of those facts and to inferences to be drawn from them.”
[165]In the present case, when standing back and taking an overall view, we are entirely satisfied that the conclusion that Aria knew or should have known that the 11 disputed transactions were connected to fraud was a conclusion well open to the FTT on the evidence, for reasons that are sufficiently and clearly explained in the Judgment. We accordingly dismiss the main part of Aria’s appeal. WHETHER HMRC MADE AN “ASSESSMENT” Background[166]In its VAT return for the 07/06 accounting period, Aria claimed credit for input tax of £1,513,316.35. After setting off input tax against output tax, Aria claimed a repayment of £445,156.98.[167]In a letter dated 6 October 2008, Mr Bailey of HMRC wrote to Aria explaining that HMRC were denying Aria’s claim for input tax of £758,770.69. That letter informed Aria of its right to appeal against HMRC’s decision within 30 days and stated that:
“A further letter showing the corrected amount of VAT now due in respect of 07/06 is enclosed.”
[168]The “further letter” was also written by Mr Bailey and, perhaps by oversight, was dated 7 October 2008. The letter referred back to Aria’s VAT return for 07/06 and stated, so far as relevant:
“As you have been notified, HM Revenue & Customs consider that the amounts shown should properly be amended as follows: Box 4 Input Tax £754,545.66 53 Box 5 Net tax due to HMRC £313,613.71 The reasons for this are detailed in my letter to you dated 6th October 2008.”
This letter also notified Aria of its right to appeal against HMRC’s decision within 30 days.[169]The parties were not agreed on the legal effect of the correspondence above, but its arithmetic at least is clear. Aria had claimed a repayment of £445,156.98 on the basis that all of its input tax was creditable. When HMRC decided that £758,770.69 of Aria’s input tax was not creditable, the logical corollary was that Aria’s repayment claim should be reduced by £758,770.69 so that Aria owed HMRC £313,613.71.[170]Aria denied that the letters of 6 and 7 October 2008 amounted to “assessments”. In Aria’s submission, HMRC had simply made a decision that was subject to appeal under s83(1)(b) of the Value Added Tax Act 1994 (“VATA”), that it was not entitled to input tax credit for £758,770.69. If Aria is unsuccessful in its appeal against that decision, Aria accepts that it is not entitled to the repayment of £445,156.98 that it claimed and HMRC will not be obliged to pay that sum to Aria. However, since, in Aria’s submission, HMRC did not make an “assessment” for £313,613.71 it follows that, even if Aria is not successful in its appeal, it has no obligation to pay any amount to HMRC.[171]Aria also argued that, in any event, the Tribunal has no jurisdiction to decide whether an assessment has been made. Its jurisdiction, under s83(1)(p) of VATA is limited to situations where an assessment has been made. Aria accordingly argued that, if it is unsuccessful in these proceedings and HMRC take steps to seek to enforce payment of £313,613.71, the proper forum for deciding whether HMRC made an “assessment” will be those enforcement proceedings. Scope of the Tribunal’s jurisdiction[172]The Tribunal’s jurisdiction in this appeal derives from s83 of VATA and that jurisdiction is subject to the qualifications and limitations set out in s84 of VATA.[173]Section 83 provides, so far as relevant: 83 Appeals(1) Subject to section…84, an appeal shall lie to the tribunal with respect to any of the following matters:… (p) an assessment under section 73(1) or(2) in respect of a period for which the appellant has made a return under this Act Section 84 provides, so far as relevant: 84 Further provisions relating to appeals … 54(3) Subject to subsections (3B) and (3C), where the appeal is against a decision with respect to any of the matters in section …[83(1)(p)]…, it shall not be entertained unless the amount which HMRC have determined to be payable as VAT has been paid or deposited with them. (3B) In a case where the amount determined to be payable as VAT or the amount notified by the recovery assessment has not been paid or deposited an appeal shall be entertained if— (a) HMRC are satisfied (on the application of the appellant), or (b) the tribunal decides (HMRC not being so satisfied and on the application of the appellant), that the requirement to pay or deposit the amount determined would cause the appellant to suffer hardship.[174]Mr Firth is correct to observe that the question whether HMRC have issued an assessment is not one of the “matters” set out in the list in s83 of VATA. However, we reject his argument that this means the Tribunal has no jurisdiction in this appeal to consider whether an assessment has been made.[175]First, in Aria’s Notice of Appeal submitted to the VAT Tribunal on 3 November 2008, Aria ticked a box indicating that it was appealing against an “assessment”. In Part C of its Notice of Appeal, Aria ticked a box indicating that it wished to apply for a direction that the appeal be considered without payment or deposit of the tax in dispute (under the “hardship” provisions referred to above).[176]Aria, therefore, was itself indicating that it was appealing against an “assessment” and was asking the VAT Tribunal to make a direction that the appeal could proceed without payment or deposit of the disputed tax (which would only be relevant if the appeal was against an “assessment”). In those circumstances, it is plain that the Tribunal has jurisdiction to consider whether an “assessment” was made. If it could not address that question, there would be no way for it to tell whether the appeal was one that the Tribunal could entertain. Moreover, the Tribunal could not gauge the relevance of the “hardship” application without knowing whether HMRC had made an assessment or not. For example, if the Tribunal refused to accept that Aria would suffer “hardship” by paying or depositing the VAT in dispute, only if Aria was appealing against an “assessment” would this be an obstacle to the Tribunal entertaining the appeal. Whether HMRC made an assessment[177]As Evans LJ noted in Customs and Excise Commissioners v Bassimeh [1997] STC 33, the making and communication of an assessment involves a three-stage process:
“the decision to assess, followed by the assessment, then by the notice given”
. Where there is little evidence as to HMRC’s internal processes, the terms of the notice (the third stage) can be relied on as evidence as to what happened at the first and second stages, but this does not alter the fact that the three stages are still separate.[178]In Courts plc v Customs and Excise Commissioners [2005] STC 27, the Court of Appeal endorsed HMRC’s internal guidance as giving a correct summary of the law when it stated: 55 “An assessment is made when you have finished calculating the amount upon which the assessment is to be based and a final decision to assess that amount has been taken. This will be when the amount has been quantified, documented, checked, signed and dated.”[179]In House (t/a P&J Autos) v Customs and Excise Commissioners [1994] STC 211 at [223], May J (as he then was) commented on the form assessments should take as follows:
“Although the commissioners choose to use printed forms headed ‘Notice of Assessment’, there is in my judgment no magic about such forms. They are not required by statute or regulation which impose no particular formality at all. All that is required is that the commissioners should make an assessment to the best of their judgment and notify it to the taxpayer. There is perhaps an understandable tendency to merge the assessment with the notification and to look only or mainly at a single document if it is called notice of assessment. But there appears to be no reason why notification should not be given by letter, nor any reason why in this case the letter dated 24 May 1990 should not be as, or part of, due notification.”
[180]May J was there referring to the previous statutory provisions, but they were in identical terms to those relevant to this appeal. The judgment of May J was upheld by the Court of Appeal: [1996] STC 154.[181]In the light of these and other authorities, we are satisfied that the FTT correctly summarised the law at [44] of the Judgment as follows:
“The “making” of an assessment refers to the determination that an amount is due; There is no set formula by which an assessment must be made; The processes of assessing and notification of that assessment are separate; The assessment process involves a decision that tax is due and a calculation of that amount; Notification can take any form so long as the terms are clear to the taxpayer.”
[182]We also agree that applying those principles leads to the clear conclusion that HMRC had, in their letters of 6 and 7 October 2008 communicated an “assessment” to Aria. Those letters demonstrate that HMRC had concluded that, because input tax had been disallowed, far from being owed money by HMRC, Aria owed HMRC the sum of £313,613.71. The letters that HMRC sent Aria communicated that conclusion clearly and succinctly.[183]In urging us to a contrary conclusion, Mr Firth relied on the following passage from the Upper Tribunal’s decision in Benridge Care Homes Ltd v HMRC [2012] UKUT 132(TCC) (Judge Gammie QC and Judge Hellier) at [38]: 56 “as an administrative act we consider that the Commissioners, as the assessing body, must believe that they are making an assessment. We do not think that they can assess, so as to speak, “by accident”.” He submitted that, since Mr Bailey had accepted in cross-examination that he did not think he had made an “assessment”, it followed that he had not done so.[184]We do not consider that the Upper Tribunal’s decision in Benridge Care Homes Ltd establishes that the subjective views of a particular HMRC officer on the legal question whether he or she has made an assessment are determinative. In Benridge, the taxpayer made a claim for repayment of VAT. HMRC decided that the repayment was not due as the taxpayer had overstated input tax claims and understated output tax due and they decided to refuse the claim for repayment. Therefore, the Upper Tribunal’s conclusion in the passage quoted was that the “administrative act” which HMRC thought that they were performing, and which they were performing, was a straightforward refusal to pay the taxpayer the sum claimed. That was clearly very different from an “assessment” which involved a decision that the taxpayer was liable to pay HMRC a sum of money.[185]Mr Firth also referred us to Pill LJ’s statement in Courts plc v Customs & Excise Commissioners [2005] STC 27 at [119]:
“What this case has highlighted is the importance of officers of the respondents being clear in their own minds what they are doing at each stage; whether they are making an assessment or a decision to assess or some other exercise.”
In that passage, Pill LJ was not saying that the subjective beliefs of a particular officer can determine whether or not an assessment was made. Rather, he was saying that, because the question of whether, and when, an assessment is made will depend in large part on HMRC’s internal actions, there is a public interest in HMRC officers understanding what they are doing at each stage.[186]Therefore, despite Mr Bailey’s acceptance that he did not think he was making an assessment in law, we consider the FTT was correct to conclude that the letter of 7 October 2018 constituted an assessment. It follows that the FTT had no jurisdiction to consider the appeal against the assessment unless Aria had paid or deposited the tax in dispute or had been excused from this requirement under the “hardship” provisions. At the hearing, Mr Puzey confirmed to us that Aria has not paid the tax in dispute but that HMRC were not taking any point that this meant the appeal could be entertained. Neither party was able to say explicitly that “hardship” had been accepted. However, given Mr Puzey’s statement, we have inferred that HMRC have, at some point in the long history of this appeal, accepted that paying the tax in dispute would cause hardship with the result that the FTT had jurisdiction to consider the appeal against the assessment, and the Upper Tribunal has jurisdiction to hear the appeal from the FTT. 57 DISPOSITION[187]The appeal is accordingly dismissed. MR JUSTICE ROTH JUDGE RICHARDS RELEASE DATE: 2 NOVEMBER 2018

Cited in 4 later judgments