The Commissioners for HM Revenue and Customs v Pacific Computers Limited: [2016] UKUT 0350 (TCC) [2016] UKUT 0350 (TCC)

UPPER TRIBUNAL
TAX AND CHANCERY CHAMBER
[2016] UKUT 0350 (TCC)Case No UT/2015/0052
Between
The Commissioners for HM Revenue and CustomsAppellantPacific Computers LimitedRespondent
Before
MR JUSTICE MANNJUDGE ROGER BERNERDate 28 July 2016Category: Tax
[1]This is the appeal of the Commissioners for Her Majesty’s Revenue and Customs (“HMRC”) from the decision of the First-tier Tribunal 5 (“FTT”) (Judge Adrian Shipwright and Ms Susan Lousada) released on 20 January 2015, by which the FTT allowed the appeal of Pacific Computers Limited (“PCL”) from a decision of HMRC denying recovery of input VAT in the VAT accounting period 09/06.[2]The case is one that falls into the category, by now well known, of MTIC, or 10 missing trader intra-community fraud. The basis for HMRC’s decision to deny PCL recovery of input VAT was that, so HMRC asserted, PCL’s transactions in computer processing units (“CPUs”) and iPods in the relevant period formed part of an overall scheme to defraud the Revenue, and that PCL knew or should have known that was the case. 15 3. By the time of the hearing before the FTT, PCL had accepted that HMRC had proved that in respect of each relevant transaction there had been a VAT loss, that such a loss had been fraudulent and that each of PCL’s relevant transactions was connected with the fraudulent evasion of VAT. Accordingly, the only issue before the FTT was whether HMRC had proved, on the balance of probabilities, that PCL 20 either knew, or alternatively should have known, of the connection to fraud.[4]The FTT allowed PCL’s appeal. It held, at [263], that HMRC had failed to show, first, actual knowledge of or involvement in fraud by PCL, and secondly that PCL had the means of knowledge or ought to have known that the only reasonable explanation of PCL’s involvement in the transactions was fraud. 25 HMRC’s appeal[5]It is from that decision that HMRC now appeal. HMRC say that the FTT erred in law in a number of material respects. It has eight specific grounds of appeal for which permission to appeal was given in this Tribunal by Judge Herrington. To some extent those individual grounds overlap and are symbiotic on one another. They are 30 helpfully rationalised by HMRC into four broad complaints.[6]The first broad complaint is that the FTT treated the evidence in relation to the overall scheme to defraud the Revenue as incapable of being probative of PCL’s state of knowledge as to the impugned transactions. In doing so, HMRC submit that the FTT erred in law. The first consequence of the FTT’s erroneous treatment of the 35 probative value of this evidence was that it found, repeatedly, that there was “no evidence” to suggest that PCL knew or should have known that the impugned transactions were connected with fraud. These conclusions, say HMRC, were perverse. The second consequence was that the FTT deprived itself of meaningful evidence against which to judge the credibility of PCL’s witnesses. 40 7. HMRC’s second broad complaint relates to the FTT’s treatment of the agreed evidence of several witnesses who were not required for cross-examination by PCL. 4 In short, the FTT refused to give significant weight to their evidence because they had not been cross-examined. The error of law, submit HMRC, is immediate and obvious. This unorthodox and unheralded approach by the FTT had serious ramifications for HMRC’s case, not least in respect of the witness Mr. Clarke, who evidenced the production of banking records from the bank through 5 which the money in the impugned transaction chains passed and on whose evidence the Commissioners relied as proving parts of the transaction chains. The FTT’s erroneous approach to what weight to give to the evidence of witnesses who had neither been called nor cross-examined was adopted of its own motion. PCL’s attack on the evidence of Mr. 10 Clarke had been limited to criticisms of failure to produce underlying documents that PCL had not itself sought. HMRC submit that PCL’s approach was erroneous but cannot be the source of the FTT’s fundamentally erroneous approach to the relevant witnesses. It is notable, they say, that at no point did counsel for PCL endorse the FTT’s fundamental error. 15 8. HMRC’s third broad complaint relates to the FTT’s failure to give proper reasons for its decision. The FTT stated in its decision that in addition to the facts set out in the decision it had also found whatever other facts it needed to justify its decision. HMRC submit that plainly a tribunal should first find the facts in order to reach a decision, rather than reach a decision and seek to find facts to support it. 20 Neither HMRC, nor any appellate tribunal, can know what these facts were upon which the FTT apparently based its decision. Based on this expressly flawed approach, making a decision and seeking to find such facts to support it, it is argued that an appellate tribunal can have no confidence even where the FTT has set out its reasoning. HMRC say that the FTT’s approach was so flawed in giving manifestly 25 inadequate reasoning that the decision cannot be upheld.[9]The fourth broad complaint relates to the FTT’s treatment of the issue of adverse inferences. In short, say HMRC, the FTT refused to draw any adverse inferences against PCL for its failure to provide witness statements from relevant witnesses, again failing to give proper reasons, and sought to justify its decision by a 30 suggestion that the Commissioners, whose plain position was that the relevant witnesses were engaged in the fraud, could have called one of them as their own witness. The FTT’s error is, again, submitted to be obvious. It is argued that it is a fundamental principle that a party cannot call a witness simply to impugn him. The FTT’s approach to the issue of adverse inferences, taken with its approach to 35 unchallenged evidence suggests that either the FTT had misunderstood the adverse inference being argued for, or was being deliberately sarcastic in its decision by characterizing the adverse inference as if it could be deployed against the Commissioners.[10]It is a recurring theme of HMRC’s case that what was required to answer the 40 questions before the FTT was a consideration of all the circumstances of the transaction and the totality of the evidence of those circumstances, and for the FTT to consider what inferences ought properly to be drawn from that evidence. Such an approach is well-established. In the particular context of MTIC fraud cases, it has been endorsed by Moses LJ in the Court of Appeal in Mobilx Ltd (in administration) v 45 Revenue and Customs Commissioners [2010] STC 1436 where he approved what 5 Christopher Clarke J had said in Red 12 Trading Ltd v Revenue and Customs Commissioners [2010] STC 589, at [109] – [111]. It is sufficient if we quote only from [109]: “Examining individual transactions on their merits does not, however, require them to be regarded in isolation without 5 regard to their attendant circumstances and context. Nor does it require the tribunal to ignore compelling similarities between one transaction and another or preclude the drawing of inferences, where appropriate, from a pattern of transactions of which the individual transaction in question forms 10 part, as to its true nature eg that it is part of a fraudulent scheme. The character of an individual transaction may be discerned from material other than the bare facts of the transaction itself, including circumstantial and 'similar fact' evidence. That is not to alter its character by reference to earlier or later transactions but to discern it.” 15 11. The FTT was keenly aware of these principles, which were uncontroversial. It referred to them by quoting extensively, at [202], from the summary provided by Arnold J in Else Refining and Recycling Ltd v Revenue and Customs Commissioners v Revenue and Customs Commissioners [2014] STC 1401, at [6] – [12].[12]The FTT stated, at [204], that it had used the case law cited to it, in particular 20 that derived from the Court of Appeal in Mobilx, as a guide in seeking to ask the right questions. Among that case law, to which the FTT made no express reference, was Regent Commodities Ltd v Revenue and Customs Commissioners [2011] STC 1964 where, in dealing with two particular types of evidence, namely documentation that proved an overall contra-trading scheme and evidence of circularity of funds within 25 the First Curaçao International Bank (“FCIB”), Newey J said, at [46] (with explanation supplied):
“… I should have thought, moreover, that, in the circumstances of the present case, the evidence given by Mr Humphries [contra-trading] and Mr Mendes [circularity of funds within FCIB] … would of itself 30 have sufficed to entitle the tribunal to make a finding of actual knowledge. As already mentioned, the tribunal considered (with justification, in my judgment) that that evidence indicated that Regent knew to whom it was supposed to sell.”
[13]The FTT also had the benefit of the judgment of the Upper Tribunal in Edgeskill 35 Ltd v Revenue and Customs Commissioners [2014] STC 1174 where Hildyard J considered the relationship between an overall scheme to defraud and actual knowledge that transactions were connected to fraud. At [55], under the heading “Issue (4): was there(a) an overall scheme to defraud(b) to which the appellant was knowingly party?”, Hildyard J said: 40 “The two parts of the fourth, final and most important question are inter-related; but they were, quite correctly, dealt with in turn by the FTT in its decision, since the question whether the appellant participated in an overall scheme to defraud informs, but does not answer, the question whether the appellant knew or should have known 45 that it was participating in such a scheme.” 6 In the same context, at [63], the judge reiterated that the fact that a taxable person’s transactions were found to be part of a wider fraudulent scheme does not mean that the taxable person knew of their connection with the fraudulent scheme. He noted that the FTT in that case had recognised that “[the] Appellant’s transactions must be considered on their own merits, which left open the possibility that 5 the appellant was an innocent dupe.” In Edgeskill itself the FTT had found, and Hildyard J found no basis for upsetting the finding, that the appellant in that case was not a genuine independent trader and it had no rational commercial purpose other than to make huge profits from doing nothing other than submitting VAT returns. 10 14. In addition to the authorities that had been before the FTT, Mr Holland referred us to the very recent judgment of the Court of Appeal in Janan George Harb v HRH Prince Abdul Aziz Bin Fahd Bin Abdul Aziz [2016] EWCA Civ 556 where, in criticising the approach of the judge below to the evidence, the court has provided helpful guidance to the proper approach. In that case, there were issues as to the 15 credibility of certain of the witnesses, including Mrs Harb. The guiding principle identified by the court, at [41], in a case of that kind is that a party is entitled to expect that the judge will engage with the arguments advanced on that party’s behalf and, insofar as the case turns on the facts, deal fully with the evidence and explain how he has come to his conclusions. Having decided that Mrs Harb was a reliable witness, 20 the judge had accepted that she had made out her case in all respects. That was an unacceptable short cut. A failure to address at least the principal grounds of challenge is likely to undermine the fairness of the trial. PCL’s case[15]Mr Webster’s overarching submission on behalf of PCL was that once the 25 FTT’s decision is carefully considered in full, and the way it which it is structured is fully understood, it will be appreciated that it satisfies all the requirements set out in the relevant case law. PCL’s broad case is that HMRC’s grounds of appeal amount to no more than a barely-disguised attack on findings of fact and value judgments of the FTT, which were submitted to be findings that the FTT had been entitled to make and 30 conclusions it was entitled to reach. In this connection Mr Webster reminded the Tribunal of the principles to be applied to such challenges, derived from Edwards v Bairstow 36 TC 207, and explained for example in Megtian Limited v Revenue and Customs Commissioners [2010] STC 840 and Proctor & Gamble UK v Revenue and Customs Commissioners [2009] STC 1990. Those principles are well known, and 35 require little elaboration. There was no dispute on the principles themselves.[16]Thus, so far as HMRC’s appeal relates to the factual findings of the FTT, it is clear that the test is whether the finding was, on the basis of the evidence, one that the tribunal was not entitled to make (Georgiou (t/a Marios Chippery) v Customs and Excise Commissioners [1996] STC 463, per Evans LJ at p 476). This tribunal should 40 also be slow to interfere with a multi-factorial assessment or value judgment of the FTT based on a number of primary facts. That is plain from Proctor & Gamble, per Jacob LJ at [9]. At [10] Jacob LJ referred to what Lord Hoffmann had said in Biogen v Medeva [1997] RPC 1, at p 45 in relation to the need for appellate caution in 7 reversing an evaluation of the facts, which is based on those findings inherently being an incomplete statement of the impression made by the primary evidence.[17]In Proctor & Gamble, at [19], Jacob LJ referred to the task of the tribunal (in that case the VAT & Duties Tribunal) in making a multi-factorial assessment in these 5 terms:
“… It was not incumbent on the tribunal in making its multi-factorial 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 10 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 Reimbold & Strick Ltd [2002] EWCA Civ 605 at [19], [2002] All ER 385 at [19], [2002] 1 WLR 2409) and that the 15 decision 'must contain … a summary of the Tribunal's basic factual conclusion and a 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 City of Birmingham District Council [1987] IRLR 250). It is quite clear how this tribunal reached its decision. In 20 the words of Sir Thomas Bingham in Meek the parties have been told 'why they have won or lost' (see para 8).”
[18]We approach the parties’ respective positions on this appeal with all these principles in mind. The specific grounds of appeal 25 Ground 1: The FTT erred in law by giving insufficient weight to the evidence of some of HMRC’s witnesses whose evidence was agreed[19]As is common in the First-tier Tribunal, directions in this appeal, which were given by Judge Mosedale on 21 May 2012, set out the position with regard to the evidence to be given by way of witness statements, and the challenge to that evidence 30 by way of cross-examination. The following direction was given in that respect:
“At the hearing any party seeking to rely on a witness statement may call that witness to answer supplemental questions (but their statement shall merely be read) and must call that witness to be available for cross-examination by the other party (unless notified in advance by the 35 other party that the witness evidence is not in dispute).”
[20]The provision for notification in advance is a sensible one. If evidence contained in a witness statement is not in dispute, there is no purpose in the tribunal hearing from that witness. That is a proportionate way of dealing with such evidence, in accordance with the FTT’s overriding objective of dealing with cases fairly and 40 justly.[21]In its decision, under the heading “Evidence”, the FTT described, at [37], the witnesses from whom it received oral evidence, by way of cross-examination. It then, 8 at [38], drew attention to the fact that the trial bundle had contained within it a number of other witness statements. It referred to three in particular, those of Officer Michael Clarke, Officer Peter Dean and Officer Michael Downer. Those officers, and others for whom the FTT had only witness statements, did not give oral evidence and were not subject to cross-examination. The FTT said in 5 this regard (at [40]): “… It is hard to decide what weight to give to evidence that has not been tested by cross-examination. Unless corroborated in some way or accepted by the parties we have not generally given such evidence great weight.” 10 22. This is a curious thing for the FTT to have said. It hedges its reservations as to the weight to be given to such evidence by reference to whether that evidence has been accepted by the parties (presumably a reference to PCL, as all the evidence in question was put in by HMRC), but the very reason such evidence was given by witness statement alone, and was not subject to cross-examination, was because it had 15 been accepted by PCL as not in dispute, in accordance with the FTT’s earlier directions. We agree with Mr Holland that what the FTT was effectively saying at [40] was that unless PCL made an additional specific concession beyond simply indicating the witness evidence not in dispute, the FTT would not give any substantial weight to that evidence. That is an error of law, and a significant one in the context of 20 this case.[23]We do not accept the submission of Mr Webster that, in referring at [40] to the weight to be given to the evidence, the FTT was intending to refer to its relevance. It is difficult to see why remarks as to relevance should be confined to evidence that had not been tested by cross-examination. But in any event it is clear from the FTT’s 25 reference to the need for corroboration or agreement that what it was referring to could not be relevance; corroboration or agreement could not render relevant something that was otherwise irrelevant. The FTT cannot be thought not to have understood the difference between relevance and weight.[24]It is abundantly clear that, where evidence is not in dispute, it must be accorded 30 full weight. To do otherwise on the basis that the evidence has been untested by cross-examination is an error of law. That was the position in Merthyr Tydfil Car Auction Limited v Thomas [2013] EWCA Civ 815, where a statement strikingly similar to that made by the FTT at [40] was made by the judge at first instance in that case. The Court of Appeal held, at [14], that that was an error of law. It was likewise 35 in this case an error of law, and we did not understand Mr Webster to resist that conclusion.[25]The Court of Appeal in Merthyr Tydfil went on to say, at [15], that it was necessary to consider the effect the error of law may have had on the judge’s reasoning and conclusions in that case. It was necessary to consider whether the 40 judge’s decision to attach little weight to the category of evidence in question had been material to the outcome and, in particular, whether that outcome might have been different had the judge considered that evidence on the correct basis. 9[26]Apart from the evidence of Officer Clarke, which we describe under Ground 2, the submissions of HMRC in relation to the evidence of those witnesses who provided witness statements that were not disputed, and who were accordingly not crossexamined, can be summarised as follows:(a) Officer Downer. Officer Downer evidenced 5 the involvement of three companies, Doktor Ring Telecom GmbH, Masterpiece Technology Limited and Plazadome Limited, in transactions that had been preordained as part of an MTIC fraud in 2005. Plazadome was PCL’s immediate supplier in all the transactions that were the subject of the 10 appeal and Masterpiece was Plazadome’s supplier. Doktor-Ring appears at various places in the chains of supply; in Deal 17 it is a supplier at the start of the chain and at the end of it, it is at the end of Deal 18 and at the start of Deals 19-26.(b) Officer Dean. Officer Dean’s evidence was to the effect that CPUs 15 sold by PCL were part of a wider, commercially inexplicable and tightlycontrolled circularity of CPUs. Officer Dean’s evidence went to the orchestrated nature of the overall scheme to defraud the Revenue for which HMRC contended.(c) Officer Paul Cole. Officer Cole evidenced the fraudulent nature of 20 the VAT loss occasioned by the defaulter JM Technical Systems Ltd.(d) Officer Jan Baltruschat. Officer Baltruschat evidenced the trading pattern of PCL’s supplier, Plazadome, in particular that 429 of 431 of Plazadome’s purchases in April and May 2006 had been traced to fraudulent tax losses. 25(e) Officer Nicholas Parker. Officer Parker evidenced the appearance of a company associated with PCL, Taran Microsystems Ltd, in a previous MTIC fraud that led to successful criminal prosecutions.[27]Essentially, the evidence in question dealt with the alleged fraudulent nature of the deal chains and was relied on by HMRC in respect of its case based on 30 orchestration, contrivance and circularity. We shall consider the way in which the FTT dealt with this evidence, and its materiality in the context of the FTT’s decision as a whole, after we have considered Ground 2 of HMRC’s grounds of appeal, which refers specifically to the evidence of Officer Clarke, one of the officers who provided a witness statement but who was not required for cross-examination, and Ground 3. 35 Ground 2: The FTT erred in law by disregarding the evidence of Officer Clarke and a Schedule of evidence based upon his unchallenged statement[28]The evidence of Officer Clarke was the subject of a specific case management direction, this time by Judge Khan, in directions issued on 30 July 2012. In Judge Mosedale’s directions provision had been made for HMRC to apply for the admission 40 of evidence of three witnesses, including Mr Clarke. That application came before Judge Khan at a hearing on 26 July 2012. In his directions, he stated “The evidence of Mr M Clarke is not objected to and therefore admissible.” 10[29]The witness statement of Officer Clarke recorded his position as the case officer for Operation Tangelo 1, a criminal investigation into the activities of the Universal Mercantile Building Society (“UMBS”) online banking facility. UMBS existed, it was said, through a number of entities, in Sweden, Panama and New Zealand. Officer Clarke’s evidence was that, for the purpose of certain criminal trials, 5 he had analysed the original data of the account database obtained from UMBS, and that he had provided another HMRC officer, Officer Cumberbatch (who gave evidence to the FTT and was cross-examined), with access to the information contained within the schedule of transactions (“the Clarke schedule”) he had compiled on the basis of this 10 original information.[30]From that information, Officer Cumberbatch had extracted relevant entries in relation to the circumstances of PCL’s relevant transactions. It was submitted by HMRC before the FTT that the Clarke schedule and the schedule produced by Officer Cumberbatch (identified by the reference VAC/2) demonstrated the circularity of 15 money flows and consequently the scheme of overarching fraud for which UMBS was used and to provide evidence to support the transaction chains.[31]No application was made by PCL for disclosure of the documents which underlie the Clarke schedule or VAC/2. Officer Clarke was not included in the list produced by PCL on 20 January 2012 of HMRC’s witnesses required to give 20 evidence under cross-examination.[32]In the written opening submissions on behalf of PCL, served 10 days before the FTT hearing, the following submission was made (at para 49):
“Subject to errors and omissions, the Taxpayer [PCL] does not dispute the broad picture painted by the schedules as to the sequence of trades 25 within the UK, to the extent that they are properly evidenced by admissible exhibits …”
A footnote read:
“This concession does not extend to the nature of the transactions which the Commissioners seek to evidence by means of VAC2, the schedule prepared by Clarke. No underlying 30 documentation has been presented.”
[33]It may be noted that there is some confusion here between the Clarke schedule and the schedule, VAC/2, derived by Officer Cumberbatch from the Clarke schedule. However, in the written closing submissions for PCL, the reference is to the Clarke schedule (para 2.12 – 2.17): 35 “2.12 The evidential value of the chain tracing based upon the Clarke Schedule. The state of the evidence is unsatisfactory. Whilst the strict rules of evidence do not apply, there must be some critical evaluation of the evidence presented by the Commissioners. The Commissioners seem to assume that the ability of the Tribunal to regulate the 40 admission of evidence means that they are at liberty to dispense with what one would normally see as absolutely basic rules of evidence. They seek to persuade the Tribunal to rely upon this evidence upon the bases that: 11 a) It had been served, but the Taxpayer had not sought the underlying material; and b) The Taxpayer had not sought to cross-examine Clarke upon it. 2.13 As to the first ground, it is a bold submission to suggest that, where evidence is patently defective, it is for 5 the other party to seek information by which the party which seeks to adduce it can then remedy it. It is for the party which seeks to put the evidence in to justify its admission. No burden should be put upon the other party. The Commissioners had ample time to disclose the relevant material / 10 serve properly admissible evidence, should they wish to have done so. The onus was upon them. This is an adversarial system. 2.14 As to the second ground, as Cumberbatch was forced to admit, he could not see how the Taxpayer could scrutinize the figures and assess their reliability based upon the material disclosed. It is, in the 15 circumstances, somewhat unfair to suggest that the Taxpayer was in a position to cross-examine in any meaningful way 2.15 It is a normal part of litigation that where HMRC wish to seek admissions as to the accuracy of schedules, it serves the underlying material. What it cannot do is to produce a schedule without any 20 underlying material and then ask the parties and the Tribunal simply to accept it. 2.16 Appendix 1 to the Opening helpfully identifies those parts of the trading chains which rely solely upon information said to be derived from UMBS. The evidential chain appears to be: 25 i) UMBS provide details of many bank accounts; ii) An officer called Clarke examines the bank data, which is not produced in this case, and the [sic] prepares schedules based upon them; iii) Cumberbatch adopts Clarke’s schedule without ever having seen 30 the original material or checked its accuracy. 2.17 The Tribunal should reject this evidence as a matter of basic fairness. As pointed out above, when Cumberbatch was asked how anyone could check the accuracy of what was being put forward, he had to concede that he had “No idea.” (22/1/14, p. 17). This is relevant 35 to allegations of circularity. In any event, it cannot be suggested that the Taxpayer knew, or could have known, of these transactions.”[34]The FTT accepted PCL’s submissions in this regard. It dealt with the Clarke schedule at [163] – [166]: “Clarke Schedule 40 163. The evidential value of the chain tracing based upon the Clarke Schedule is uncertain. The state of the evidence is unsatisfactory.[164]As Mr Webster submitted “Whilst the strict rules of evidence do not apply, there must be some critical evaluation of the evidence presented by the Respondents. They seek to persuade the Tribunal to 45 rely upon this evidence upon the bases that: 12 164.1. It had been served, but the Appellant had not sought the underlying material; and 164.2. The Appellant had not sought to cross-examine Clarke upon it”.[165]He continued “What it [HMRC] cannot 5 do is to produce a schedule without any underlying material and then ask the parties and the Tribunal simply to accept it”.[166]We agree with Mr Webster and do not accept the Clarke Schedule. We do not consider at any rate that it would have added 10 much to what was before the Tribunal.”[35]Furthermore, at [250], when addressing a submission for HMRC that the circular nature of transactions and the contrived nature of the deal chains should lead to the inference that PCL must have been told when and what to buy and sell (and from and to whom), the FTT said: 15 “We did not find that the Clarke schedule added much to the position. If it were to be given significant weight it would have needed to have been proved by cogent evidence including working documents and a witness who could be cross examined as to how it was produced. This was not the case. Accordingly, it is hard to give it to any significant 20 weight.”[36]Mr Holland submitted that, by not disputing the witness statement of Officer Clarke, both the Clarke schedule and the means by which it had been created had been accepted by PCL. There can be no requirement for a party to further prove an agreed document. It is open to a party to agree a schedule without requiring the underlying 25 documents. It is not open to that party subsequently, in a skeleton argument produced just before the hearing, and in written closing submissions after the hearing, to resile from that agreement. In ruling as it did on the acceptance of the Clarke schedule and by according it no significant weight in the absence of a cross-examination of a witness not required by PCL to be called and the absence of the underlying materials 30 for which disclosure had not been sought, Mr Holland argued that the FTT erred in law.[37]Mr Webster submitted that the Clarke schedule was hearsay evidence, and that the FTT made no error in not admitting it, or in giving it no substantial weight. In support of that he referred us to his cross-examination of Officer Cumberbatch on the 35 first day of the hearing when the officer had confirmed that the only UMBS banking record he had seen had been what he described as the “master spreadsheet” compiled by Criminal Investigation. He had not made any checks as to the accuracy of the Clarke schedule, and he was unable to assist when asked how he could verify that schedule. 40 38. The relevant evidence of Officer Cumberbatch was of course derived from the Clarke schedule. It cannot, however, properly be argued that such evidence should have been thereby excluded from admission by the FTT. It is not a case of one officer seeking to give evidence of something done by, or known to, another officer who is not presented as a witness. In this case, the direct evidence of Officer Clarke, 13 including the Clarke schedule itself and Officer Clarke’s own evidence as to its provenance and the materials on which it was based, was before the FTT in the form of an unchallenged statement. We agree with Mr Holland that the FTT erred in law in refusing in these circumstances to accept the Clarke schedule, and in failing to give it significant weight in the absence of underlying material, when 5 Officer Clarke’s evidence, which exhibited the statement, was unchallenged.[39]There remains, as with Ground 1, and on the basis of Merthyr Tydfil, the question whether this error affected the FTT’s reasoning and its conclusions. We shall come to that question when we have addressed Ground 3. 10 Ground 3: The FTT erred in law by purporting to find whatever unspecified facts were necessary to be found in order to justify the decision. This reversed the proper approach of finding specified facts and then reaching a decision based upon them, with transparent reasoning, capable of review by the parties and any appellate tribunal. 15 40. HMRC point to two instances where, they submit, the FTT fell into error in this respect. The first is at [247], where the FTT, having described (at [243], [244] and [246]) as assertions by HMRC first that because of PCL’s integral role in the deal chains and significant share of profits it followed that PCL knew or ought to have known of the fraud, secondly that because of the uniformity of many of the deal 20 chains PCL must have been told when and what to buy and sell, and thirdly that PCL knew of sales by its Dutch customer to another Dutch company, rejected those arguments and said this:
“We do so because there was no evidence before us to support these assertions and nothing on which to find any such inference. To the 25 extent that we have not already done so we find such [f]acts [sic] as a [sic] necessary to support the rejection of these assertions and the invitation to make such inferences. To the extent possible we find these matters as matters of primary fact.”
[41]The second instance is at [253], in a section headed Overview. The FTT said: 30 “We find that HMRC have not shown that the only reasonable explanation for PCL's involvement in its transaction(s) was fraud within the meaning set out by the Court of Appeal in Mobilix [sic]. The evidence did not show this to be the case and there was no foundation laid before us from which any such inference could be drawn. To the 35 extent that we have not already done so we find such facts as are necessary to support the rejection of these assertions and the invitation to make such inferences.”[42]Mr Holland submitted that the FTT’s decision displayed a fundamentally flawed approach. Rather than find facts and reach a decision based on those facts, the 40 passages we have quoted suggest, he argues, that the FTT’s view was that it should reach a decision and invite a reader, including an appellate tribunal, to find its own facts to support the decision. Alternatively, submitted Mr Holland, what is 14 demonstrated is the FTT seeking to justify a decision rather than setting out reasons for it.[43]Mr Holland referred us to Flannery v Halifax Estate Agencies Ltd [2000] 1 WLR 377, in the Court of Appeal, where at p 381-382, giving the judgment of the court, Henry LJ commented on the 5 duty to give reasons:
“(1) The duty is a function of due process, and therefore of justice. Its rationale has two principal aspects. The first is that fairness surely requires that the parties especially the losing party should be left in no doubt why they have won or lost. This is especially so since without 10 reasons the losing party will not know (as was said in Ex parte Dave) whether the court has misdirected itself, and thus whether he may have an available appeal on the substance of the case. The second is that a requirement to give reasons concentrates the mind; if it is fulfilled, the resulting decision is much more likely to be soundly based on the 15 evidence than if it is not. (2) The first of these aspects implies that want of reasons may be a good self-standing ground of appeal. Where because no reasons are given it is impossible to tell whether the judge has gone wrong on the law or the facts, the losing party would be altogether deprived of his 20 chance of an appeal unless the court entertains an appeal based on the lack of reasons itself. (3) The extent of the duty, or rather the reach of what is required to fulfil it, depends on the subject matter. Where there is a straightforward factual dispute whose resolution depends simply on which witness is 25 telling the truth about events which he claims 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. But where the dispute involves something in the nature of an intellectual exchange, with reasons and analysis advanced 30 on either side, the judge must enter into the issues canvassed before him and explain why he prefers one case over the other. This is likely to apply particularly in litigation where as here there is disputed expert evidence; but it is not necessarily limited to such cases. (4) This is not to suggest that there is one rule for cases concerning the 35 witnesses truthfulness or recall of events, and another for cases where the issue depends on reasoning or analysis (with experts or otherwise). The rule is the same: the judge must explain why he has reached his decision. The question is always, what is required of the judge to do so; and that will differ from case to case. Transparency should be the 40 watchword.”
[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. A statement that the tribunal finds such facts as are necessary to support other findings or determinations is not itself a finding of fact at all and 45 therefore contravenes the principles in Flannery. It may have been in part the result of the fact that the period between written final submissions and the release of the decision was some seven months, but that does not affect the principle. Nor do we 15 accept, as Mr Webster urged upon us, that this can in any way be affected by what Jacob LJ said in Proctor & Gamble, at [19] and [61]. Although we accept that in making a multi-factorial assessment it is not incumbent on a tribunal to “spell out item by item how each was weighed as if it were using a real scientist’s balance”, it remains necessary for the decision to contain “a summary of the 5 Tribunal’s basic factual conclusion and a 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 City of Birmingham District Council [1987] IRLR 250). The FTT adopted an impermissible short-cut. 10 45. Nonetheless, if such attempts at findings are discounted, the question remains whether, on the basis of the facts actually found by the FTT, and the evidence before it, the FTT’s reasoning and findings fall to be impugned. We shall, accordingly, now turn to consider that question.[46]We do so at this stage for two reasons. First, of the remaining grounds, Ground 15 4 submits there are errors of law in the FTT’s analysis of the evidence relating to the detail of the overarching fraud, Ground 6 is a cumulative ground on the basis of what are submitted to be errors of the FTT in finding there was a lack of evidence for certain propositions advanced by HMRC, Ground 7 is also a cumulative ground in relation to a failure to set out findings of fact and give reasons and Ground 8 is a 20 cumulative ground based on a submission that the effect of Grounds 1 – 7 is that the FTT did not have proper regard to matters relevant to its assessment of the credibility of PCL’s case and its witnesses. Those are all matters that require consideration of the FTT’s decision as a whole, and we can therefore address those grounds compendiously. Secondly, given the errors of law we have found in relation to 25 Grounds 1 – 2 in particular, it is appropriate that we consider whether those errors were, taking the decision as a whole, material to the outcome and, in particular, whether that outcome might have been different had the FTT considered the evidence in question on the correct basis. We shall address the only remaining ground, Ground 5 (adverse inference from the failure to seek evidence from witnesses capable of 30 supporting PCL at a time when it believed they were so capable), after we have reviewed the decision as a whole. The FTT’s decision[47]We summarise below the FTT’s decision. We have done so at some length and by reference to both elements that are not subject to criticism as well as those which 35 are for two reasons. The first is that, as Merthyr Tydfil makes clear, the errors of law we have identified must be considered in the context of the whole of the FTT’s reasoning and conclusions. The second is that PCL’s case, as advanced by Mr Webster, was very much focused on a detailed analysis of the structure and content of the FTT’s decision as a whole, and his submissions, particularly those he advanced in 40 oral argument, were based on a step-by-step review of that decision. In the same vein, whilst emphasising what he submitted were specific errors of law in the FTT’s decision, Mr Holland, for HMRC, argued that the approach taken by the FTT to the evidence was, also having regard to the decision as a whole, flawed as a general matter. 16[48]The only issues before the FTT were whether PCL knew, or alternatively should have known, that its transactions were connected with the fraudulent evasion of VAT. As the FTT correctly directed itself, in relation to the question whether PCL should have known of the relevant connection, the question, as described by Moses LJ in Mobilx, at [59], is whether PCL should have known that 5 the only reasonable explanation for the transaction in which it was involved was that it was connected with fraud.[49]Although HMRC point to numerous examples within the FTT’s decision where they submit the FTT failed to have regard to the evidence, the overriding submission 10 is that the FTT failed to have proper regard to the evidence of contrivance in the deal chains, including through the fraudulent conduct of participants in the chains, the circularity of fund flows and carouselling of products, and the control which HMRC argued the orchestrators of the fraud would need to have exerted over all the transactions in the deal chain. The FTT, say HMRC, either refused to place any or 15 any proper weight on certain of the evidence or discounted it as relevant only to the facts of fraud and connection to fraud, which were accepted by PCL, and did not have regard to that relevant evidence when assessing the evidence of PCL, including the credibility of its witnesses.[50]In its decision, the FTT placed considerable emphasis on its assessment of the 20 credibility of PCL’s witnesses. The FTT placed its assessment in the context of certain doubts it expressed, at [42], as to the relevance of some of the topics on which those witnesses had been cross-examined, and the nature of that cross-examination. The FTT said, at [43] – [45]: “43. This meant we were able to observe the witnesses closely when 25 they were under challenge. The effect of that was that we were in a very good position carefully to consider their evidence and how it was given and what weight to give to it. We have carefully considered these aspects amongst others in reaching our conclusions. The Appellant's witnesses came through this very well which served to enhance their 30 credibility and give greater credence to what they said. The way HMRC dealt with them enhanced this which was possibly not HMRC's intention.[44]We find, having seen them giving evidence, the Taxpayer's witnesses to be credible and reliable and honest in the evidence they 35 gave. We accept their evidence and in the case of any conflict with other evidence we prefer it. It was not discredited in the course of cross examination. Such inconsistencies as emerged only went to reinforce our view on credibility and reliability. This is hard to put into clear words but essentially comes from having the benefit of both of us 40 having seen and heard them particularly when being cross examined. We consider that they were not delivering a story made up after the event to account for what went on but trying to help the Tribunal by telling the truth. In any case of a conflict of evidence, as noted above, we prefer the evidence of the Appellant's witnesses. 45 45. We have carefully considered this position and what to put in this decision before doing so in order to be clear as to what we made of the 17 witnesses. It is done on the basis of careful and sober reflection and consideration to assist the parties and any higher Court or Tribunal should this go any further. Any such body should be only too aware of the impression we formed of the Appellant's witnesses and why we believed their evidence. They came across, albeit 5 with different personalities, as people who were trying to run their business sensibly and commercially and during this hearing trying to assist the Tribunal to the best of their ability. They did so in a calm, reflective and polite way whilst displaying realism, honesty and acceptance of the need to 10 assist the Tribunal even when pushed extremely hard in cross examination. As noted above we found the Taxpayer's witnesses credible and reliable and honest in the evidence they gave.”[51]Those are important findings. As one of the appellate tribunals which the FTT clearly had in mind when making those remarks, we pay them due regard. It is the 15 FTT, and not this tribunal, which has had the benefit of seeing the witnesses give evidence, and of being subject to cross-examination. Of note, however, is the FTT’s broad conclusion that, in any case of conflict with other evidence, it had decided to prefer the evidence of PCL’s witnesses. That is, of course, a conclusion the FTT would have been entitled to reach, on a proper analysis of the competing evidence. 20 What HMRC say, on the other hand, is that the FTT failed to analyse the evidence in that way, and its conclusions are accordingly flawed as a matter of law.[52]The FTT painted a clear contrast between the evidence of PCL’s witnesses and those of HMRC, saying (at [46]) that HMRC’s witnesses had not assisted much. It reserved special criticism for Officer Cumberbatch in relation to the terms in which 25 the original decision to deny deduction of input tax had been expressed, and continued, at [46]: “… The rest of his evidence was aimed more at tax loss in the chain caused by fraud. This was not a matter of dispute but common ground between the parties. He showed little commercial awareness and 30 knowledge which meant his comments on such matters whilst interesting could not carry the weight of an acknowledged and informed expert. It was thus hard to give his evidence any considerable weight. It was not put forward as expert evidence.[47]The other HMRC witnesses were of a different calibre. However, 35 they were concerned with matters at a much higher level and in Dr Findlay's case with information not available at the time of the transactions and which, as he said, surprised him when it came to light.”[53]In making the remarks it did at [47], the FTT was not referring to HMRC’s 40 evidence as a whole, but only to the witnesses it had seen give oral evidence. Those witnesses, as the FTT described at [37], were Officer Deborah Toynbee, who gave evidence of a visit to PCL’s premises in November 2006 (which the FTT appears, at [104], to have found not to be significant), Mr Roderick Stone, whose evidence concerned MTIC fraud generally, and Dr Kevin Findlay, who provided evidence, as 45 an expert, on the grey market element of the distribution market in electronic components. 18[54]The FTT made a number of findings of fact. These included;(a) The ownership and governance structure of PCL ([53] – [54]).(b) The knowledge of PCL concerning a company, Taran Limited, of which a Mr Andrew Miles, a shareholder and former director of PCL, was a director and shareholder ([56] – [63]). The FTT 5 found that, although PCL traded with Taran, it did not know that in November 2003 HMRC officers had executed a search warrant at Taran and that others, not including Taran or its officers, had been convicted of conspiracy to cheat the Revenue. 10(c) The effect on PCL’s turnover of the absence of Mr Marc Roach, the managing director, from the business between May and September 2005 ([66] – [68]). The FTT rejected HMRC’s case that PCL had been a business in decline prior to the transactions at issue in the appeal.(d) The circumstances in which Plazadome, which was PCL’s supplier 15 in each of the transactions in question, had become such a supplier ([69] – [71]) and [105] – [113]). The FTT accepted that, having been contacted in February 2006 by a long-standing contact, one Ms Theresa Ching of Zaanstrait BV, a Dutch company which was PCL’s customer in eight of the deals in question, concerning the possibility of PCL making wholesale 20 supplies to Zaanstrait of computer and electronic goods, Mr Richard Donaldson, a director and shareholder of PCL, had contacted Taran for assistance on a number of occasions. It also accepted that Mr Miles of Taran had considered that this was disruptive to Taran’s business and that he had, for that reason, provided Mr Roach with the name of Plazadome 25 as a potential supplier.(e) Discussions with Zaanstrait ([80] – [86]). The FTT found that the initial contact by Zaanstrait had been by phone, that Mr Donaldson and Mr Roach had travelled to the Netherlands to discuss the supply of CPUs to Zaanstrait and that there had been subsequent negotiations over the 30 telephone.(f) The funding of PCL ([87] – [93]). The FTT found that the funding arrangements for PCL showed a degree of commercial and financial prudence. Its initial funding for the new business was provided by HSBC. The FTT rejected the criticism of HMRC in relation to the difference 35 between the business plan produced to obtain bank funding and what had actually happened. It accepted as commercial the decision of PCL to use VAT repayments as funding for subsequent trading.(g) The significance of “checklists” produced by Mr Andrew Hall, a director of PCL who was responsible for management of financial 40 accounts and related matters ([96] – [98]). The FTT accepted that these were simply working documents; they were not intended to be complete records. The FTT rejected HMRC’s argument that the checklists were a “diversion”. 19(h) Bank accounts ([99] – [102]). The FTT found that PCL had an account with HSBC and that, although PCL had considered opening an account with UMBS in order to speed up receipt of payments, it had not done so.(i) Trading and documentation ([117] – [122]). 5 The FTT found that PCL received enquiries about the supply of goods and then considered whether it could match that demand. If PCL found a match, it would send a purchase order to the supplier. In each case in question, the supplier was Plazadome because it was Plazadome which could supply the goods at a 10 price which made it economic for PCL to deal. The FTT remarked, at [118], that in its view this did not make what was done by PCL uncommercial or fraudulent. PCL would then send the sale documents to the buyer. Terms and conditions were included in the documentation. Title to the goods was to pass on payment, but goods could be shipped 15 “on hold”. The FTT found that the risk of shipping on hold was not a disproportionate one in a business context.(j) Insurance ([123]). The FTT found that PCL had insurance in relation to all consignments of goods.(k) Inspection ([124] – [129]). The FTT found that Mr Leighton 20 Birtchnell, the logistics manager of PCL, and other members of the warehouse team had carried out, under the direction of Mr Roach and Mr Hall, visual inspections of a number of consignments of CPUs. It rejected any adverse inferences it had been invited by HMRC to draw from the way in which the inspections were carried out. 25 55. Having described, at [131], the deal chains, and noting, at [132], the consistent nature of the participants in those chains, the FTT went on to find, at [134], that the deals were “back to back”, in the sense that they were arranged essentially to take place on the same day. However, drawing an analogy with residential conveyancing, the FTT declined to find that this demonstrated fraud or means of knowledge of fraud 30 on the part of PCL. It rejected, at [135], the submissions of HMRC that the requirements for the products could be “instantly matched”, and found that even if that had been the case “it would not necessarily mean that the deals were artificially contrived on [PCL’s] part or that [PCL] should have known of this”.[56]At [136] – [137], the FTT described the decision letter of HMRC from which 35 the appeal had been made (and which the FTT had, when discussing the evidence of Officer Cumberbatch, criticised). It is not clear to what extent the decision letter, and the view which the FTT took of it, played in the FTT’s overall findings, but it seems to us that a letter of that nature could at best be of historical interest only. The FTT’s concern should have been focused less on analysis of the basis on which HMRC had 40 decided at the outset to deny input tax recoverability, and more on the way the case was put on the appeal, and the evidence said to support that case.[57]After referring to the common ground that there was a tax loss caused by fraud in the chains, the FTT rejected the case that PCL knew that its transactions were connected with fraud. At [138] – [141] it said: 20 “138. It is common ground that there was tax loss caused by fraud in the chains. There is nothing before us to show that the Taxpayer was aware of this when it entered into its transactions and we so find.[139]We find as a fact and to the extent possible as a primary fact that the Taxpayer had no actual knowledge of 5 fraud in the chains.[140]We have carefully considered all the evidence before us in reaching this conclusion and do so both subjectively and objectively from the perspective of the Taxpayer and its officers and that of an officious bystander. We make this finding on the balance of 10 probabilities and because there was no evidence before us to show otherwise and no evidence laying a foundation from which an inference could be drawn.[141]We reject HMRC's assertion that PCL had actual knowledge of the fraud at the relevant times as there was nothing before us to support 15 HMRC's assertion.”[58]This then is the first example of what Mr Holland argued, on the basis of Ground 6, was an erroneous finding by the FTT that there was “no evidence” before the FTT which could support the case made by HMRC.[59]At [142], the FTT said that, given its findings as to actual knowledge, the 20 question whether PCL “should have known” that fraud was the only reasonable explanation for PCL’s transactions remained open, but it went on to anticipate its eventual conclusion that this had not been shown to be the case.[60]The FTT then made certain further factual findings;(a) Record of serial numbers ([143] – [147]). The FTT accepted the 25 rationale for the goods, whether iPods or CPUs, not being capable of being electronically scanned but instead being randomly tested. It accepted that made commercial sense.(b) At [148] – [151], the FTT essentially repeated its rejection of any suggestion of impropriety on the part of PCL in its entering into back to 30 back deals. It found, at [150], that there was no evidence of impropriety. The FTT placed reliance in this respect on having heard and seen the witnesses.(c) Terms and conditions ([152] – [153]). The FTT noted the existence of standard terms and conditions on the reverse of PCL’s order forms, as 35 well as certain implied terms.(d) Due diligence ([154] – [162]). The FTT described the elements of the due diligence which it found had been performed by PCL. It declined, at [160], to find or infer that the due diligence had been done with the objective of demonstrating compliance with HMRC examples, but 40 considered, by its finding at [161] – [162], that the transactions were commercial transactions on PCL’s part, that there was no evidence to support a contrary inference and that the due diligence, though imperfect, was both commercial and sensible. 21(e) At [163] – [166] the FTT dealt with the Clarke schedule in the way we have described above. Essentially the schedule was rejected in the absence of the underlying material but in any event considered relatively insignificant.(f) Counterfeit software ([167]). The FTT found that, 5 although it was accepted by PCL that it had bought counterfeit software, this had been done unwittingly. The FTT rejected any suggestion that this would have, or should have, made PCL suspicious of the transactions in question.[61]At that stage, and we consider by way of summary of what the FTT had 10 concluded on the basis of the findings it had recorded and its subsequent discussion of the parties’ submissions, the FTT said (at [168] – [170]): “168. In broad terms, we would summarise our findings as follows: 168.1. The Taxpayer's witnesses came across as honest and reliable and we accept their evidence having seen them pushed hard; 15 168.2. PCL and its officers whilst aware of MTIC in general terms had no actual knowledge of fraud in the chains at the relevant times; 168.3. The deals in question were carried out commercially and in an honest manner as far as PCL was concerned; 168.4. The Decision Letter did not survive any sensible scrutiny; 20 168.5. HMRC failed to show anything or to show anything from which it could be inferred that PCL ought to have known or had the means of knowledge objectively or subjectively of fraud in the chains at the relevant time(s); 168.6. HMRC failed (inter alia because they were unable to show 25 the matters set out in the preceding subparagraph) to show that the only reasonable explanation for PCL's transactions was connection with fraud.[169]HMRC failed to discharge the onus of proof of showing either: 169.1. That PCL had actual knowledge of connection to fraud of 30 PCL's transactions; or 169.2. That PCL ought to have known of the connection to fraud in its transactions or that fraud was the only explanation for PCL's transactions within the approach in Mobilix [sic].[170]Insofar as we have not already found these matters as matters of 35 fact we do and to the extent possible we find them as matters of primary fact.” (We note that the FTT appended a footnote to [168.5]: “This is not to suggest that this was the test.”)[62]Between [171] and [197], the FTT summarised the submissions of the parties. 40 For PCL, the principal submissions identified by the FTT were, first, that PCL did not have actual knowledge of the fraud in the chains, and secondly that it was not the case that the only reasonable explanation for PCL’s involvement was fraud. The 22 involvement of an innocent party was a reasonable explanation and had the merit, it was said, of being true in this case. For HMRC, the essence of the case was summarised as being that the transactions were part of an overall MTIC fraud scheme involving a web of companies and a chain of “transactions” the sole aim of which was to defraud the Revenue. The transactions, it was argued, were 5 orchestrated and contrived for such a purpose and had no ordinary commerciality to them. The CPU transactions were part of a wider circularity of CPUs designed to facilitate an MTIC fraud. The same entities participated on numerous occasions. PCL must have known of the connection to fraud to have played such an integral role and taken such a 10 significant share of the profits. If PCL did not actually know, then by virtue of the cumulative circumstances presented to it, it should have known of the connection. The only reasonable explanation was that of connection to fraud.[63]The FTT summarised the law, before concluding, at [205]:
“… on the basis of Mobilix [sic] in the light of the relevant factual 15 circumstances, the Tribunal must be satisfied on the basis of cogent evidence, that at the time PCL entered its deal it either knew that there was a connection between those transactions and the fraud or that the only reasonable explanation for the transactions in question was that they were connected with fraud. The threshold is a high one and 20 deliberately set by the Court of Appeal, if input tax recovery is to be denied. The onus is on HMRC on the balance of probabilities.”
[64]At [206] the FTT concluded that HMRC had come nowhere near even satisfying a low threshold in this case.[65]In relation to actual knowledge, the FTT held, at [207], that there was no 25 evidence that showed actual knowledge of fraud on the part of PCL or any of PCL’s witnesses. At [208], the FTT went on to say that HMRC had not pointed to any evidence of actual knowledge or matters sufficient to found the drawing of such an inference from all the circumstances.[66]In relation to HMRC’s arguments based on orchestration and contrivance, the 30 FTT found, at [211], that the most likely explanation was that PCL was an innocent party who knew nothing of the fraud. That can be seen as a finding in relation to actual knowledge. The FTT went on to find that such an explanation was not unreasonable, and that accordingly it had not been shown that the only explanation for the transactions was connection to fraud. 35 67. Referring to HMRC’s submissions in relation to orchestration, the FTT said, at [212], that these assumed that PCL knew at the relevant times that the transactions were orchestrated by fraudsters as part of an overall scheme to defraud the Revenue. The FTT found that there was no evidence that PCL had such knowledge.[68]The FTT then dealt more specifically with HMRC’s submissions regarding the 40 wider fraudulent scheme involving Doktor-Ring, Masterpiece and Plazadome in 2005. The FTT found, at [215], that there was no evidence that PCL knew of such a scheme 23 at any relevant time. CDs obtained by Officer Downer in February 2008 were dismissed as not showing any connection to PCL. At [218] – [219], the FTT said: “… PCL is a company not an individual and we were taken to no evidence that PCL information was on the discs or that it was in any way involved. Accordingly, we do not see what 5 weight it has other than to show certain companies in the chain may have been involved in fraud. PCL had already agreed that this was the case. It does not go to PCL's actual knowledge or that the only reasonable explanation for PCL's transactions was fraud. It may explain fraud in the chain and 10 related matters but of itself it does not go to PCL's transactions.[219]Whilst this may be interesting information we do not see that it helps us answer the question was the only reasonable explanation for the transactions PCL took part in fraud? It does not go to what PCL did or did not do or know.” 1569. The FTT adopted a similar approach to the submission by HMRC concerning the significance of the goods being carouselled. The FTT recorded, at [221], HMRC’s submission in this respect as being: “… There is no commercial explanation for companies at both ends of the chains dealing in the same goods again in such a short period of 20 time. Crucially, to ensure the completion of the carousels, PCL had to sell the goods to the right company. The Respondents submit that over so many transaction chains the carousel pattern cannot be attributed to coincidence, and that PCL must have been told from whom to purchase, to whom to sell, at what prices and when”. 2570. At [222] the FTT held that there was no evidence that PCL knew of the carousels or was in receipt of instructions as to what to do. Likewise, at [223], the FTT found that there was an insufficient evidentiary base from which such an inference could be made. It said (at [224] – [226]): “224. We are concerned with PCL's position here and not the chain of 30 transactions and whether or not that was fraudulent and involved tax loss. In this case the onus of proof is on HMRC in respect of PCL not the other persons in the chain. HMRC seem at times to merge the chain and PCL. It is common ground that there is fraud in the chain. Our task is to consider whether or not the only reasonable explanation for PCL's 35 transactions was fraud.[225]The carousel argument does not assist in that task particularly when PCL has accepted that there was fraud in the chain which caused a tax loss. HMRC were taking a holistic view of the chain rather than a more individualistic view of PCL and its transactions. This does not 40 mean that all the circumstances do not need to be considered but rather they have to be considered in the context of the precise question the Tribunal has to answer in respect of the taxpayer who is making the appeal.[226]Merely because PCL appears in a chain of transactions which 45 involved tax loss through fraud does not mean that PCL knew or ought to have known this was the case or in more correct language the only 24 reasonable explanation for PCL's transactions was fraud. The onus of proof is on HMRC to show this was the case in respect of PCL.” The FTT found, at [227], that HMRC had failed to discharge that burden of proof.[71]The FTT made further findings in relation to HMRC’s submissions of contrivance at [236] – [241]. It rejected the reliance by HMRC 5 on the contrived nature of the deal chains, describing it, somewhat puzzlingly, as a “bootstraps” argument, since it looked at the chain rather than the position of PCL. It concluded, at [240], that HMRC had not established contrivance or knowledge of contrivance on PCL’s part. 10 72. The distinction between the chains and the position of PCL was further emphasised by the FTT at [242] – [251] under the general heading of:
“Failure to look at PCL’s transactions rather than the chain as a whole”
. Describing it, again rather curiously, as a regulatory approach, the FTT referred to HMRC’s approach as being one of looking at the whole of a chain and asserting that because there was a loss of 15 tax caused by fraud in the chain it therefore followed that PCL’s transactions could only be explained by fraud. The FTT rejected such an approach, saying (at [243] and [244]) that whether PCL knew or should have known of the fraud did not necessarily follow either from PCL’s integral role and significant share of profits (which the FTT found had not been proved) or from the fact that the same company had been at both 20 ends of a chain. There was, the FTT found at [245], no evidence to show that PCL knew the positions of other parties in the chains, and nothing on which to base such an inference. The FTT also rejected arguments as to the uncommerciality of the profits achieved by PCL. It gave little weight to the Clarke schedule. Discussion 25 73. The decision of the FTT is a lengthy one, running to 29 pages and some 266 paragraphs. It sets out the common ground between the parties, noting that the focus was on the question of actual and constructive knowledge, states the FTT’s views on the evidence presented for each party and in particular on the credibility of PCL’s witnesses, makes findings of fact, recites a summary of the parties’ respective 30 submissions, summarises the relevant law by reference to key authorities, and reaches its conclusions by reference to the case put by HMRC and the FTT’s own view of the evidence.[74]Nonetheless, as we have decided, the FTT made a number of errors of law in its approach to the evidence. The question for us now is whether, having regard to those 35 errors and in any event, the approach of the FTT was flawed such that its decision must be set aside.[75]Having considered the case that HMRC put to the FTT, and the way in which the FTT’s decision sets out how it reached its conclusions, we have reached the view that the FTT failed properly to examine the evidence before it. That failure, in our 40 judgment, can be attributed to a number of factors. First, the FTT had effectively closed its mind to a material part of the evidence put forward by HMRC which was unchallenged; secondly, the FTT misunderstood the case as put by HMRC, and 25 accordingly asked itself the wrong question in relation to the evidence of orchestration and contrivance; and thirdly, in considering the evidence put forward by PCL, the FTT failed to test that evidence by reference to the surrounding circumstances, including in particular the orchestrated and contrived nature of the fraud with which PCL’s transactions 5 were connected.[76]HMRC’s closing submissions invited the FTT to find that the evidence showed that the level of orchestration in the deal chains was very high. It was then submitted that two questions arose: first, how did the orchestrators of the fraud manage it so well, and secondly how likely was it that an orchestrator of such a fraud would 10 involve an unknowing party and why? The submission was that the only way in which the orchestrators of such a fraud could ensure a carousel pattern and speed was to tell each party from whom to purchase, to whom to sell and at what price. It was argued that the carousel, circularity and timings that occurred simply could not have happened without that level of instruction. It was further submitted that, because a 15 fraudster would wish to retain control of the component parts of such a fraud, it was highly improbable that an orchestrator of such a fraud would involve an unknowing party.[77]Although the FTT’s summary of HMRC’s submissions hints at these arguments, when it comes to the discussion of the submissions the FTT falls into 20 error. Thus, at [242], the FTT said:
“HMRC approached this case by looking at the whole of a chain and asserting that because there was a tax loss caused by fraud in the chain it therefore followed that the Taxpayer's transactions could only be explained by fraud. Such an overall regulatory approach may be 25 appropriate in some circumstances but does not address the issue of the Taxpayer's position which is what the Tribunal is concerned with.”
[78]From a review of the case as put by HMRC in their written closing submissions to the FTT, we are satisfied that the FTT was wrong to describe HMRC’s case in this way. The FTT went on, at [225], to say that the carousel argument did not assist in 30 considering whether the only reasonable explanation for PCL’s transactions was fraud “when PCL has accepted that there was fraud in the chain which caused a tax loss”. The FTT was thus saying that the evidence of orchestration and contrivance was relevant only to the existence of the fraud (which had been conceded), and failed to appreciate the relevance of contrivance and the submissions based on it to the 35 question of knowledge.[79]The same error of approach can be detected in the way the FTT dealt with HMRC’s submissions on the previous involvement of Doktor-Ring, Masterpiece and Plazadome in MTIC fraud. HMRC’s case was that those companies were not companies engaged in legitimate trade in an ordinary market, but were fraudulent 40 companies who were told from whom to purchase, to whom to sell, what goods to deal in, at what prices, and when. The inference the FTT was invited to draw was that the presence of those companies in PCL’s transaction chains provided a compelling inference that those chains had been orchestrated by fraudsters as part of an MTIC scheme. 26[80]The FTT failed to appreciate, and thus failed to address, the link that HMRC was seeking to make between the evidence of fraudulent behaviour on the part of the three companies through the submission that the deal chains in which PCL’s transactions had been orchestrated by fraudsters to the submission that all companies involved, including PCL, must have been instructed so as to facilitate 5 the fraud, and PCL must therefore have known, or should have known, of the connection to fraud. It did not connect the evidence and submissions in relation to the three companies with the question of PCL’s knowledge that it had to address. It simply considered whether PCL had been aware of the involvement of the three companies in the 2005 fraud. It 10 said, at [218], that that involvement did not go to PCL’s actual knowledge or that the only reasonable explanation for PCL’s transactions was fraud; and at [219] it remarked, rather caustically, that “whilst this may be interesting information we do not see that it helps us answer the question was the only reasonable explanation for the transactions PCL took part in fraud? It does not go to what PCL did or did not 15 know.”[81]It is regrettable that the FTT failed to appreciate the inferences which HMRC was inviting the FTT to make from the orchestrated and contrived nature of the fraud and the presence of fraudulent companies within the deal chains at issue in the appeal. The FTT was keenly aware, it appears, of the need to consider whether inferences 20 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. 25 82. Although, as for example at [140], when describing its conclusion that PCL had no actual knowledge of fraud in the chains, the FTT stated that it had carefully considered all the evidence before it in reaching that conclusion, and that it had done so “because there was no evidence before us to show otherwise and no evidence laying a foundation from which such an inference could be drawn”, it is evident from 30 how the FTT later addressed the question of orchestration and contrivance that it did not consider, or did not properly address, the evidence before it. 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 sufficient simply to say that there was no 35 evidence. The failure by the FTT properly to address the submissions of HMRC by reference to the available evidence was an error of law.[83]Although we accept, as Proctor & Gamble explains, 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 40 that, despite its length, the FTT’s decision regrettably did not meet that basic test. In our judgment the FTT fell into the trap of taking a short cut, as identified in Harb (at [38]), by reference to its acceptance of the credibility of PCL’s witnesses, and broad statement (at FTT, [44]), that that evidence was to be preferred in any case where it conflicted with other evidence. That, along with the FTT’s erroneous understanding 45 of HMRC’s case and its failure to have proper regard to certain of the evidence, led to 27 the FTT failing properly to analyse the evidence and to take a reasoned approach to the question of credibility. The FTT did not subject the evidence of PCL’s witnesses to scrutiny by reference to the factual evidence produced by HMRC and the inferences which HMRC submitted ought to be made from that evidence as a counterweight to the evidence of those witnesses. The failure to 5 give proper weight to the evidence of the officers was, in our view, part and parcel of this overall failure in relation to the evidence.[84]As an example, the FTT found, at [211], that the most likely explanation for PCL’s transactions was that it had been an innocent party who knew nothing of the 10 fraud. That is stated as a bald conclusion, with no reasoning and no weighing of the competing factors favouring that conclusion or the conclusion that PCL was either a knowing participant in the fraud, or should have known that fraud was the only reasonable explanation. Furthermore, when the FTT said, at [243] and [244], that it did not “necessarily” follow that PCL’s integral role and significant share of profits, 15 or circularity, would lead to the conclusion that PCL knew or should have known of the fraud (or more accurately the connection to fraud) or that PCL must have been told when and what to buy and sell, it mis-stated the argument and led itself to the wrong conclusion. It was not submitted that the conclusion “necessarily” followed from the integral role. It was said that the consistent chains provided evidence from 20 which HMRC’s conclusion could be drawn. The FTT ought to have considered the evidence in that context. Instead, it simply found, at [245], that there was no evidence to support any such inference and that the assertion that such an inference should be made missed the point. That mis-stated the case before it, mis-stated the proper analysis and, as we have found, was an error of law. It obviously contributed 25 significantly to its ultimate decision.[85]We need not refer to other examples where HMRC submitted the FTT failed to provide reasons for rejecting specific elements of HMRC’s case before the FTT. We have concluded that, having regard to the errors of law we have identified, and after considering the FTT’s decision as a whole, first that the errors of law in Grounds 1 – 3 30 were material to the outcome of the appeal and secondly, and in any event, that the FTT erred in law in failing properly to address HMRC’s case on the evidence, and in failing to give proper reasons for certain of its conclusions.[86]The only proper course in those circumstances is for the FTT’s decision to be set aside. Having done so, we do not consider that we, as an appellate tribunal, are in 35 a position to re-make the decision. The case depends on a proper consideration of the evidence, which includes (as the FTT rightly identified) issues of credibility of witnesses. What is required, and where the FTT erred in law in failing to undertake, is a reasoned analysis of all the evidence, according the evidence its appropriate weight, and coming to a fully-reasoned conclusion. It follows that this case must be 40 remitted to the First-tier Tribunal. We understand that Judge Shipwright has retired from his position as a judge of that tribunal, but we would in any event have concluded that the proper course, in view of the fundamental errors in the approach of the FTT, is for this case to be remitted to a new panel to be heard afresh. 28[87]Our conclusion in this regard means that we do not need to address Ground 5 of HMRC’s grounds of appeal, which sought to criticise the failure of the FTT to draw an adverse inference from the failure of PCL to seek evidence from certain witnesses. As that is an issue that may have to be addressed by a new panel of the First-tier Tribunal we shall 5 say nothing about it. Decision[88]For the reasons we have given, we allow this appeal.[89]We set aside the decision of the FTT, and remit the case to the First-tier Tribunal with the direction that it be re-heard in its entirety by a fresh panel. 10 MR JUSTICE MANN UPPER TRIBUNAL JUDGE ROGER BERNER 15 RELEASE DATE: 28 July 2016

Cited in 9 later judgments