Celltec Computers Ltd v Revenue & Customs [2013] UKFTT 155 (TC)

FTT-Tax
Celltec Computers Ltd v Revenue & Customs
[2013] UKFTT 155 (TC) · 2013-02-25
[1]This was a Missing Trader or MTIC appeal in which the Appellant appealed against HMRC’s refusal to repay its claims for refunds of VAT input tax in respect of seven transactions. The transactions were referred to as Deals 1 to 7. Deals 1 to 6 were effected in the Appellant’s VAT period ending 30 September 2006, and Deal 7 in the following VAT period ending 31 December 2006. The total VAT reclaimed in respect of all seven transactions was £275,381.41, by far the largest single amount being claimed for Deal 3, namely £190,456.88.[2]Until very shortly before the commencement of the hearing, the assumption had been that the Appellant was disputing not only the “knowledge or means of knowledge” aspect of the Kittel test, but also the facts in relation to HMRC’s claim that the seven transactions had been connected to fraudulent losses of VAT. When, immediately prior to the commencement of the hearing, Andrew Young was appointed to represent the Appellant, the Appellant conceded the connection to tax losses initially in relation to all seven deals, so that it appeared that the only question that we would have to decide would be the “knowledge or means of knowledge” issue as laid down in the Kittel case, and in Mobilx v. HMRC [2010] EWCA Civ 517 .[3]Matters then became slightly more complicated because, while HMRC’s decision letter had essentially denied the input tax in relation to all seven deals on Kittel lines, it had been mentioned in the decision letter that in Deal 1 the goods had never in fact been supplied to the Appellant at all, and in HMRC’s Statement of Case and Skeleton Argument, it had become clear that the principal ground for denying the refund of input tax in respect of Deal 1 was the non- Kittel point that there had simply been no supply by the intended supplier, and therefore no input tax to be reclaimed by the Appellant. As a fallback, HMRC was asserting that if that contention was wrong, then the refund should still be denied on Kittel principles.[4]HMRC’s “no supply” contention led to a number of different contentions by the Appellant. Firstly it was contended that the original decision to refuse refunds of input tax for all seven deals had all been based on Kittel principles and that it was not open to HMRC to seek to sustain the refusal on quite different grounds. As to the fallback contention in relation to Deal 1, based indeed on Kittel grounds, it was contended that since HMRC had not sought to trace the chain of supply in relation to Deal 1 at all, and had not established that the Appellant’s Deal 1 transaction was connected to any demonstrated VAT fraud, the fallback Kittel contention should also fail. Moreover the initial concession to the effect that all seven deals had been traced to frauds was amended, and this concession was said to apply only to Deals 2 to 7.[5]The result of this was that the seven deals fell into three relatively separate categories.[6]The first category included just Deal 1, where we had to consider the issues just referred to, and that we will expand on in paragraph 9 below.[7]The second category included just Deal 3, in which by far the Appellant’s major transaction involved the purchase of Intel CPUs from a company, MVS Digital Ltd (“MVS”), owned and operated by the brother-in-law of Mr. Rajnikant Patel (“Mr. Patel”), the owner and operator of the Appellant, and the supply of those CPUs to a company called Alpha Enterprises (“Alpha”). The unique features of this deal were that the evidence in relation to the payments through First Curacao International Bank (“FCIB”) indicated that:-(i) the payments, through 12 companies, had been circular;(ii) all of the 12 payments had been made within 84 minutes;(iii) every single computer payment instruction (with the exception of the instruction by MVS to pay its supplier, but significantly including the instructions in relation to the payment to and the payment by the Appellant) had been made from a computer or router with one and the same IP address; and finally that(iv) the first company in the chain to commence the circular payment flow in the Appellant’s transaction had in fact been involved in similar circular payment flows involving many other companies and many other deals.[8]The third category of deal included all the remaining deals, namely Deals 2,4,5,6 and 7, some of the attributes of which we will summarise in paragraph 13 below.[9]It is worth summarising the decisions that we need to reach in relation to Deal 1, because Deal 1 became rather complicated. We might indeed add that our Decision has been considerably delayed by the fact that counsels’ final written further contentions on some of these issues were not provided until January 2013. The relevant decisions that we must reach are as follows. We need to decide: · whether HMRC’s decision and the reasons for that decision in relation to Deal 1 had included the contention that there had been no supply, such that HMRC could plainly defend the appeal against the decision on the ground that there had indeed been no supply; · whether, if the reasons for the original decision had been based exclusively on Kittel principles, it could nevertheless now be established that the decision itself was of wider import, and that the original decision might now be justified on quite different “non-supply” grounds. This approach would require the legal position to be that a more widely-phrased decision arrived at for one reason could later be sustained for different reasons, and we would also have to decide whether clear reliance on the “non-supply” contention in HMRC’s Statement of Case and Skeleton Argument now enabled HMRC to defend their decision on those different grounds; · which party bears the burden of proof in establishing any issue relevant to the Respondents’ claim that there had been no supply, and we must then give our findings of fact in relation to that issue; · what the law is in relation to there being “no supply”; · if HMRC could not seek to dispute the Appellant’s appeal in relation to Deal 1 on “non-supply” lines, or the conclusion was that there had in any event been a supply, whether HMRC could nevertheless succeed on their fallback contention on Kittel lines.[10]We will deal with each of those points below. For present purposes it is sufficient to say that we concluded that: · it was open to the Respondents to defend their decision in relation to Deal 1 on the basis that there had been no supply; · the right conclusion, as a matter of both fact and law, was that there had indeed been no supply; but that · the Respondents’ fallback contention, that if there was rightly analysed to have been a supply, the Appellant’s claim to recover input tax was properly denied on Kittel grounds, was also sustained.[11]Before indicating in outline the crucial significance of Deal 3, we must mention that in many ways Mr. Patel’s evidence, that being the only evidence on behalf of the Appellant, left us with a considerable dilemma. His evidence revealed the fact that he appeared to have forgotten countless matters that we found it difficult to believe that anyone could forget. He claimed in meetings with HMRC that he had been alert to detect fraud, but at the same time exhibited the most extraordinary attitude to due diligence, where he appeared to think that he had no need barely to read replies sent to him, and that the whole purpose of the exercise was just to collect bits of paper and file them in case HMRC ever wished to see them. His description of the trading terms, where payment terms were generally not referred to in the documents, but everyone was said to be content to deal with the Appellant on the flexible basis that the Appellant would pay whenever cash flow permitted, were also quite extraordinary. As we said, this left us with the dilemma of having to decide whether Mr. Patel was simply “way out of his depth” and someone who might quite plausibly have been duped, or whether (as HMRC contended) Mr. Patel had periodically been lying and regularly feigning the appearance of forgetfulness and ignorance.[12]The facts in relation to Deal 3 led us to conclude, however, that the explanation that Mr. Patel gave for the payment mechanics of Deal 3 cannot possibly have involved the “whole truth”, and that indeed he must have deliberately lied in relation to that deal, and known that he was a party in a pre-arranged chain transaction. He must also have lied when responding to the question of why he had failed in his first Witness Statement to describe how it happened that all payment instructions, bar one, had been made from a computer or computers using the same IP address. We simply cannot accept the proposition that he had just “forgotten this detail”, until the evidence in relation to the IP address emerged during the hearing, and jogged his memory. We will in due course explain in full how we arrive at our conclusion in relation to Deal 3 that Mr. Patel knew of the connection of that deal to a fraudulent VAT loss. For present purposes, we simply say that the conclusions that we have reached to the effect that Mr. Patel lied in very material respects in relation to the crucial Deal 3, leave us reconciling the dilemma that we mentioned in the previous paragraph in favour of the proposition that we could not trust Mr. Patel’s evidence and rejected the possibility that he had repeatedly been duped.[13]The remaining deals had certain common characteristics. Two of these characteristics were that FCIB evidence was of no particular assistance in relation to them, and that the deals had been effected in the sort of vague and flexible, and implausible, manner summarised in paragraph 11 above. None of the third category of deal contained, however, what Mr. Young had referred to as “the smoking-gun point”, in other words the single feature that immediately led us to conclude that Mr. Patel must have known of the connection of these deals to fraud. There were common characteristics in that the customer in all remaining five deals was a Dutch company referred to as Zaanstrait, and the suppliers were either a company referred to as Plazadome or one referred to as Unique. We accept, however, that notwithstanding evidence about those two suppliers advanced by HMRC that struck us as almost wholly irrelevant, there was still no “smoking-gun” that made our decision clear cut. Our decision is nevertheless that the Appellant must have known that the remaining five transactions were all connected to the admitted VAT frauds. This is because:- · we have concluded in relation to Deal 3 that Mr. Patel must have lied, and that in that Deal, the Appellant must have known of the connection to fraud; · given those two conclusions, and having rejected the claim that he was duped in Deal 3, it becomes manifestly improbable that he might then have been duped in the other deals, and no longer clear that we should accept Mr. Patel’s explanations and claims as wholly truthful; and · the vague, flexible, un-commercial and frankly bewildering way in which the remaining deals had been effected made them totally inconsistent with anyone’s understanding of how genuine grey market deals would have been structured. We expand on this below.[14]Our conclusions are therefore summarised below in relation to the points that need to be addressed in relation to Deal 1, and our decision in relation to all seven deals (relevant in relation to Deal 1 only if the Respondents’ fallback contention on that deal is in point) is that the Appellant must have known that all seven deals were connected to VAT fraud. Were that conclusion one that was held on appeal to be unreasonable, we certainly conclude that the Appellant ought to have known of that connection. The evidence[15]With the exception of the evidence about the content of a particular CD, which we will deal with in paragraphs 17 to 24 below, we have incorporated the evidence given by the Respondents’ various witnesses, and the evidence given by Mr. Patel, who alone gave evidence on behalf of the Appellant, in giving the facts in more detail below.[16]Evidence on behalf of the Respondents was given by Officer Shahid Rajput, the case officer; by Officers Peter Birchfield and Andrew Monk in relation to the FCIB evidence and its particular relevance to Deal 3; by Officer Andrew Letherby in relation to the inferences that can sensibly be drawn from the feature that various companies have all logged onto the internet using one and the same IP address, and by Officer Esther Jelenke in relation to certain facts in relation to Maystar, the asserted supplier to the Appellant in Deal 1. We considered all these witnesses to be straightforward and we see no need to address specifically any elements of their evidence. The CD evidence[17]Evidence in relation to information provided on a CD by HMRC, said to be significant by the Respondents, was given by Officer Michael Downer.[18]We will give a short summary of this evidence now. We consider it inappropriate to give it in the course of giving the full facts because we concluded that the evidence was virtually wholly irrelevant. We only feel compelled to give some short summary because of the extensive argument advanced in relation to this evidence and the amount of time spent in relation to it.[19]The evidence in question consisted of a CD that had apparently been found behind a microwave by the police in the course of a murder investigation. At some point the police presumably recognised that it might be of more significance to HMRC than to the police and it was handed to HMRC. The alleged significance of the evidence was that it contained a vast compilation of information and documentation about MTIC transactions effected about 9 months prior to the Deals involving the Appellant, in some of which three companies referred to as Plazadome, Unique and Masterpiece that were involved as buffers in the Appellant’s transactions in the third category mentioned in paragraphs 8 and 13 above, had been parties, either as buffers or brokers.[20]We barely looked at any of the evidence itself, though we did understand that in a rather odd way the information on the CD only provided details of about half the assumed steps of the earlier transactions, and we also gathered that the information was extremely detailed, relating to such matters as Redhill checks, and including many e-mails.[21]There was considerable argument as to whether the information on the CD constituted planning information or subsequent compilations of transactions that had been effected, the Appellant’s counsel even suggesting that HMRC themselves rather than some mastermind behind the planning of the transactions might have been the source of the information.[22]Having regard to the facts that; · the Appellant itself did not appear at any point on the CD; · the Appellant’s counsel conceded in relation to Deals 2 to 7 that all were connected to fraudulent VAT losses and even conceded (leaving aside the explanation for the Appellant’s participation) that the deals appeared to have been arranged deals; · we ourselves observed that virtually all the margins made by the buffer companies in the Appellant’s Deals 2 to 7 were of small fixed steps that suggested pre-arrangement; · earlier Tribunal decisions had anyway referred to participation of the three mentioned companies in earlier MTIC appeals, and that the present Tribunal Judge had individually dealt with cases involving all of them; · and having regard to the central fact that the only question for us in relation to Deals 2 to 7 was the issue of whether the Appellant knew or ought to have known of the connection of its deals to fraud, we were at an absolute loss to understand how the hotly disputed evidence was of any relevance whatsoever. In other words, insofar as there was any relevance to the fact that the three companies had been involved in MTIC transactions more than 9 months before the Appellant’s transactions, that conclusion was easily to be gleaned from the third and fourth bullet points above, and was virtually totally irrelevant to the only point on which we have to reach a decision.[23]Insofar as we understood the reason why the Respondents had suggested that this evidence was material, the Respondents explained that they were following the direction of Lord Justice Moses in Mobilx to give consideration to “surrounding circumstances”, rather than to over-emphasise due diligence. The answer to this, it seemed to us, was that in the present context the surrounding circumstances to which we should pay regard must be those of which the Appellant was, or might have been aware, and not factors of which the Appellant would almost certainly have been totally ignorant.[24]Accordingly we ignored this evidence. The facts in more detail Background[25]The only witness for the Appellant was Mr. Patel. Mr. Patel explained that he had attended the first year of a course at de Montfort University, Leicester in engineering and electronic engineering after leaving school, but had had to leave the course due to financial constraints.[26]He had then worked as a sales person for a company called Bene Trading Limited for 6 years; then as a technical sales person for Elex Business Systems, and in 2002 he moved to Supreme Distribution Limited (“Supreme”), as a sales manager.[27]He admitted that he had “entrepreneurial ambitions” and said that in 2000 he and two others formed the Appellant. We heard nothing of the two others, and simply gathered that Mr. Patel was only able to operate very occasionally for the Appellant. The trading that we understood the Appellant to undertake initially was retail trading involving the purchase of relatively large lots of computer hardware, and then the sale of smaller lots of such components to various UK retail buyers at very modest profits. It was immaterial to learn what actual manufacturing or assembly functions Mr. Patel had performed, but he did refer to having “made computers”.[28]Until the accounting period that ended on 31 March 2006, the Appellant’s turnover had been very modest, fluctuating around the amount of £40,000 per annum. In the period ending 31 March 2006, the turnover had risen to roughly £1.8 million, but the profit was still modest at roughly £12,000.[29]Mr. Patel said that in December 2005, a Mr. Chandeep Singh (the brother of the managing director of Supreme) became a director of the Appellant. Mr. Patel said that he could no longer attend to all the business on his own, and that Mr. Singh joined in order to assist him. It appears that it was at about this time that the Appellant started to undertake back-to-back wholesale trading, and that this development largely accounted for why the turnover had increased significantly by the end of the March 2006 accounting period. We were given no details of any trading in any period prior to the first in contention in this Appeal (that for the period covering July, August and September 2006) though we might mention a rather strange exchange of remarks that arose in the course of his cross-examination in relation to Deal 1 (effected on 13 and 14 July) and involving the supplier referred to as Maystar. The odd remarks in question related to whether or not two earlier deals had been undertaken by the Appellant with Maystar as the supplier back in March. Certainly we had, amongst the documents, a copy of an Introductory Letter from Maystar to the Appellant dated January 2006, and it was clearly also the case that the HMRC record of a meeting between Mr. Patel and Officer Rajput and a colleague in March 2007 (obviously after all the Deals and in the course of extended verification) referred to the fact that Mr. Patel said that two deals had been done in March with Maystar. In his oral evidence before us, however, Mr. Patel said that he could not recall whether deals had or had not been done with Maystar prior to Deal 1, i.e. the one presently in contention. He said that doubtless his documentation would resolve the doubt, but that he personally did not know now whether the deals had been done.[30]Lapses of memory of the type just recorded, particularly on points that would have appeared to have been difficult to forget, were very regular; indeed almost the invariable rule, rather than isolated incidents.[31]Although we were told that Mr. Chandeep Singh became a director, he gave no evidence and Mr. Patel asserted that he had little involvement with the Appellant’s business, apparently because he frequently travelled to India. Mr. Patel’s sister was the company secretary, but the Appellant had no other employees and Mr. Patel effectively conducted all the activities.[32]HMRC claimed that both Mr. Patel and Mr. Chandeep Singh had had other employments (Mr. Patel’s at Supreme) where the employing company had been involved in MTIC activity, but it was not asserted that either man had been directly involved in the transactions in question.[33]In many MTIC appeals, appellant companies often fund the VAT-gap, in other words the initial excess of the VAT-inclusive amount that they are liable to pay their supplier over the VAT-exclusive amount that they are to be paid by their foreign customer, by loans on extraordinary terms from unrelated entities. We are satisfied that in the present case, there were no such extraordinary loans. There were loans but as Mr. Patel periodically mentioned (and we accept this evidence) the Appellant funded its relevant VAT gaps by borrowing from family members. He mentioned that £25,000 had been loaned by his father, £60,000 by his eldest brother and £55,000 by his second brother, and we were also told that nothing was lent by his brother-in-law, the individual who owned MVS Digital, the supplier in Deal 3. The facts in relation to Deal 1[34]Deal 1 involved the unsuccessful endeavour to purchase from Maystar 1260 Intel CPUs and to sell those CPUs to a Swiss company referred to as Dipro. The VAT-inclusive price payable to Maystar was £102,228, and the VAT-exclusive price chargeable to Dipro was £91,980, such that if the Appellant’s repayment claim had been accepted by HMRC, the Appellant’s profit on the deal would have been £4,977. The deal documentation was dated either 13 or 14 July 2006.[35]It is necessary to summarise the facts in relation to Deal 1 relatively fully, partly because it will save repeating some points in relation to later deals, but principally because it is only in relation to Deal 1 that we will have to decide whether there was a supply, whether the facts indicate connection to a fraudulent tax loss, and whether the Appellant must have known of any such connection. Obviously Deals 2 to 7 raise only the third of those points. The origin of the deal[36]There was very little evidence as to how the deal had been negotiated, and how the Appellant came to deal with the two counter-parties. We have mentioned above that it is possible that the Appellant had undertaken two earlier deals in March in which Maystar had been the supplier and Mr. Patel also said that Maystar was quite well known. Dipro, in contrast, was a Swiss company previously unknown to Mr. Patel. It had initiated the deal by contacting him, presumably having seen one of various free advertisements that the Appellant had placed on trading websites. Due diligence[37]Mr. Patel undertook a very limited amount of due diligence in relation to both Maystar and Dipro. To save repeating these points in relation to every transaction, it is fair to say that the Appellant treated all due diligence questions in a very similar way. Mr. Patel admitted that he never took up any references even where trading counter-parties had mentioned the names of referees (often admittedly without giving contact details). He never troubled about poor credit reports because his buyer would normally pay him before the free release of goods to the buyer, and no credit was given to the supplier. Thus Maystar’s credit score on one Creditsafe report at £500 was immaterial, and it rather seemed that he had not even detected where in the credit report the suggested maximum figure of credit to allow Maystar was mentioned. The fact that much of the material sent by Dipro was written in a foreign language was irrelevant because it was “probably German” and it was quite easy to guess at what it said. This struck us as a particularly surprising suggestion since Mr. Patel again and again confessed that he had not considered replies, or communications from HMRC or trading counter-parties written in English “in depth”. We ended up concluding that Mr. Patel paid scant regard to all responses received in the due diligence exercise and we concluded his failure to consider “various things in depth” was in reality a total failure to consider virtually anything at all. He also seemed to regard the purpose of the exercise of due diligence to be to get material and put it on file, so that if needs be it could be produced to HMRC. Notwithstanding the claim that he was alert to discover whether there had been any fraud, he seemed to ignore the feature that he might be meant to read and consider the material and consider whether it had any possible bearing on fraud at some point in the chain.[38]On 7 July, the Appellant sought “Redhill verification” of Maystar’s VAT registration, and did likewise as regards Dipro on 13 July. We will mention in the next paragraph some odd points that emerged at the hearing in relation to Maystar’s VAT registration, but the more relevant point is that it was not until 2 August (obviously well after the deal had proceeded) that HMRC confirmed the registration of Maystar, and on 21 July they indicated that as Dipro was not registered within the EU they could give no information as regards the Swiss company.[39]The odd points to which we just referred were that back in April 2006 HMRC claimed to have sought to notify the Appellant that Maystar had been de-registered. We say “sought to notify” because Mr. Patel said that he had not received this notification, and since it had been sent by HMRC to the wrong address, it seems entirely credible that the Appellant had not received this notification. Quite what had happened in terms of reinstating Maystar’s registration or possibly clarifying that the relevant notification in an unsigned letter might not have been sent in the first place (perhaps because HMRC concluded that de-registration was not appropriate) we do not know. The two facts that are clear however are that the price receivable by Maystar for the single sale to the Appellant (leaving aside the point that it may not have occurred at all) was beyond the registration threshold such that Maystar should have been registered in respect of that single sale on its own, and secondly on 2 August, Redhill confirmed that Maystar’s registration was then valid. It is of course the case, not that we attach great significance to this point, that the Appellant proceeded with Deal 1 prior to receiving the response from HMRC at Redhill in relation to either trading counter-party. The deal paperwork[40]Turning now to “terms of trade” and the detail of purchase orders and invoices, Dipro issued its purchase order on 13 July. This requested the supply of 1260 CPUs and Mr. Patel said that Dipro was particular about the box numbers of the CPUs that it was acquiring. This claim seems somewhat odd since the purchase order assumed that the 1260 CPUs would comprise two trays of CPUs, each of course containing 630 CPUs, which did not tally with the identification and box numbers of the CPUs that Maystar contracted to supply to the Appellant. The purchase order also contained the text “Payment via T/T in advance after IR”. There was considerable discussion about what this meant, and nobody advanced the suggestion that perhaps it meant that the goods would be delivered on a ship to hold basis by the Appellant’s freight forwarder (or whoever arranged their transport); then be inspected by or on behalf of Dipro, whereupon Dipro would make payment and presumably the goods would then be released. If this is what the relevant wording meant, it was certainly not what happened and indeed nobody particularly suggested that this is what it did mean.[41]The other material term on Dipro’s purchase order was “Delivery cond: must be delivered in transit – T1, no customs clearance”. We were again unclear what this meant, but assumed that it meant that the Appellant would arrange for transport. This would have tallied with one of the terms in the Appellant’s invoice mentioned below.[43]The Appellant’s purchase order to Maystar on 13 July referred to the quantity of 1260 CPUs but provided neither box nor tray details or any contractual terms, other than the words “Goods released to our account in Forward Logistics”. 43. Maystar’s invoice, issued on 14 July referred to 4 trays, each containing 315 CPUs, and gave the box numbers. It is not clear that the box numbers were passed on to Dipro, though Mr. Patel claimed that this was done in an e-mail. The only material contractual term on Maystar’s invoice relating to payment said “Payment: Other” , and it is obviously not clear what that meant.[44]The Appellant’s invoice to Dipro, also dated 14 July, made no reference to payment terms at all, though in fairness this perhaps indicated that the Appellant accepted that payment would be governed by the terms already indicated on Dipro’s purchase order. Mr. Patel was unable to suggest precisely what those payment terms meant, when asked to consider them during the hearing, and it seems unlikely to us that he had had any better understanding, or had indeed even spotted the relevant bit of text, when the deal was documented.[45]The Appellant’s invoice contained no reference to title, or title retention until payment. The only two related terms that it did contain were one saying “Goods shipped to via Forward Logistics [the freight forwarder where the goods were understood to be held] to BLG Basle wharehouse (sic) ” followed by the Basle address of the relevant warehouse. The Invoice also said “Manufacturer’s warranty” and “Please insure the goods as Celltec will not be liable for loss in transit”. We took these delivery and insurance terms to mean that Celltec would procure that its freight forwarder would arrange delivery to the Basle address, but that Celltec would not be insuring the goods. Whether this made sense if the supplier was only effecting delivery at the destination warehouse was not considered during the hearing. Payment[46]On 14 July, the Appellant received Dipro’s full payment to its FCIB account. This payment was of course made before the goods had been inspected by Dipro, since at this point (and indeed at all times thereafter) Maystar had not released the CPUs to the Appellant, and the Appellant had not despatched them to Dipro.[47]Matters then went off the rails.[48]Mr. Patel claimed that he then pressed Maystar to release the CPUs to him, so that he could instruct Forward Logistics to deliver them to Dipro, but Maystar apparently said that they could not do so until they had received payment. Accordingly on 19 July, the Appellant paid Maystar from its FCIB account the full VAT-inclusive price owed to Maystar. Mr. Patel confirmed in cross-examination that he had not sought by this point to verify the existence of the CPUs.[49]Having received payment, Maystar failed to release the CPUs to the Appellant and failed thereafter to respond to telephone calls from Mr. Patel. Following enquiries made by Mr. Patel, Forward Logistics suggested (perhaps on a number of occasions) that title to the CPUs had not passed to Maystar, which was why Maystar could not release the CPUs to the Appellant. This information was never confirmed but it was on the basis of this information that it was asserted by the Appellant’s counsel that it was reasonably clear that the CPUs in question (probably to be identified by the box numbers given on Maystar’s invoice) did exist. Subsequent events[50]Not surprisingly Dipro contacted the Appellant complaining that it had not received the CPUs, and Mr. Patel explained that this was because he had not received them from Maystar. Thereafter, there was no subsequent communication between the Appellant and Dipro. In other words, Dipro never wrote to the Appellant or sued for the recovery of their payment, and they never confirmed that in some curious manner they had ultimately received the CPUs.[51]Mr. Patel said that at some point he asked his accountant how he should report the Deal 1 transaction for VAT purposes and apparently his accountant suggested that he should still claim a deduction for the input tax. He therefore did so. We were not informed when Mr. Patel first learnt that Maystar had been liquidated, but we do know that on 12 December 2006 Mr. Patel wrote to the Insolvency Service (possibly at the suggestion of his accountant) asking them to investigate the non-payment by Maystar, and asking the Insolvency Service to deal with the enquiry by contacting his accountant. We did not see any reply to this letter and it was certainly the case that the enquiry failed to lead to any action or the provision of further information in relation to Maystar. In May 2007 the Appellant appeared to have been notified by the Insolvency Service that Maystar had been compulsorily liquidated on 20 September 2006 in response to a petition filed on 27 July 2006.[52]HMRC’s exhibits included a letter from the Insolvency Service, sent on 20 May 2009. This recorded the facts that the former director of Maystar had been disqualified from being a director for a 12-year period. The letter referred to the way in which Maystar’s activities had occasioned risks that HMRC would suffer VAT defaults, had undertaken mobile phone trading for the amount of roughly £225 million; had made third party payments, and it specifically identified Maystar’s failure in July 2006 to provide the four boxes of CPUs for which it had received payment. It was not mentioned that that transaction was the Appellant’s transaction but since the amount of the payment tallied almost exactly with the amount that the Appellant had paid it seemed clear that the transaction referred to was the Appellant’s Deal 1 transaction.[53]Maystar did not submit a VAT return for the period that covered its July transaction with the Appellant, presumably because the petition to wind it up had been presented. We understand (a fact with some possible relevance to the “supply” issue) that HMRC issued an assessment on Maystar for the VAT in respect of that deal. Implicitly no appeal can have been made in respect of that assessment by Maystar, so that presumably the debt in respect of the assessment became final. HMRC’s evidence was that the assessment was made before the facts in relation to the possible “non-supply” became clear, though it is fair to say that after those facts emerged, HMRC did not withdraw or vary the assessment. The facts in relation to Deal 2[54]The facts in relation to the Appellant's second deal were extremely simple.[55]Deal 2 involved the purchase of 1,440 AMD Dual 3800 processing units from Plazadome Limited ("Plazadome") on 1 August 2006 for a VAT inclusive price of £106,596 and the immediate release on the same day of the units to the Dutch company referred to as Zaanstrait for the VAT exclusive price of £95,040. Zaanstrait's purchase order, the Appellant's pro forma invoice and invoice to Zaanstrait, the Appellant's purchase order to Plazadome and Plazadome's invoice were all dated 1 August. The limited terms in the documentation[56]The terms included in those documents were minimal in that beyond mentioning the brief description of the goods (to which we will revert) and the prices, only two other terms were even mentioned. One indicated that Plazadome was to release the goods to the Appellant at Forward Logistics and the Appellant was to release them to Zaanstrait's identified transport firm. The other was that in Plazadome's invoice, the box referring to "Terms" contained the detail "TT ADV" which it was assumed meant that the Appellant would pay Plazadome prior to the delivery or final release of the goods. We were told however that the reference to "TT ADV" was wrong because it had been agreed over the phone by Plazadome that some form of undefined credit would be given. Nothing indicated in the invoice issued by the Appellant to Zaanstrait when the price was to be paid, or whether any title retention provision applied to protect the various suppliers pending payment. The release of the goods prior to the receipt of payments[57]Plazadome's instruction to Forward Logistics, the freight forwarder that held the goods, to release them to the Appellant, and the Appellant's instruction to release the goods to the transport firm that was to collect them for Zaanstrait were both issued on 1 August.[58]The price was in fact not paid immediately. On 11 August 2006, Zaanstrait paid the Appellant its full invoice amount, the payment flowing from an ABN Amro account of the former to the Appellant's FCIB account. The Appellant paid a virtually identical amount to Plazadome's FCIB account on the same day. The balance of the Appellant's payment to Plazadome, £11,569, was paid from the Appellant's NatWest account on 15 September, in other words more than a month later and 6 weeks after the various supplies of the goods.[59]The Appellant's percentage profit, assuming a recovery of the claimed VAT, in relation to Deal 2 was 4.77%. We should also mention that Plazadome made an unusually large mark-up for a buffer, being £2.25 and 3.7% per unit, whilst the previous buffers had only made £0.15 and 0.25% per unit. Discussion about risks[60]Some significant evidence emerged in the cross-examination of Mr. Patel in relation to Deal 2.[61]The Respondents' counsel first asked Mr. Patel whether the experience in relation to Deal 1, of having a supplier fail to deliver goods to the Appellant, led to the Appellant tightening up any of its procedures. To this Mr. Patel replied: "The business was - these were the basic rules of our business, so we hadn't changed anything specific. We do the Redhill checks, whatever, you know, the rest of the form fillings. It was the generality of our business, basically to do this this way. We never changed it".[62]There was then considerable criticism of the Appellant's due diligence. It appeared that there had been earlier deals with Plazadome, and that any due diligence in relation to Plazadome had only been done after those deals, albeit in advance of the presently relevant Deal 2. Mr. Patel said that although he had done a few deals with Plazadome, and their office was only about 30 miles away from the Appellant's office, he had never had time to visit Plazadome.[63]The exchanges addressing due diligence in relation to Zaanstrait focused on the fact that many of Zaanstrait's responses were in Dutch. It also emerged that although the Appellant had little knowledge of Zaanstrait itself, Mr. Patel said that his co-director had known the man, Andy Bugden, who was heavily involved in Zaanstrait's trading and married to the owner of Zaanstrait. Mr. Singh had known Andy Bugden at the latter’s previous employer. Discussion about credit terms[64]The Respondents' counsel then drew Mr. Patel's attention to the fact that in HMRC’s visit report of a visit conducted by Officer Rajput in conducting extensive verification in relation to the September and December 2006 claims, Mr. Patel had indicated, in response to questions about trading on credit, that he neither gave nor received credit. When asked how this tallied with the credit received from Plazadome, Mr. Patel explained (rather implausibly) that he had thought that the line of questioning in the earlier interview had related to his retail dealing in small quantities of computer components and not in relation to the large wholesale deals.[65]So far as Deal 2 was concerned, Mr. Patel clarified (to the apparent surprise of the Respondents’ counsel, though not to us) that Plazadome's release to the Appellant and the Appellant's release to Zaanstrait both passed not only control and possession of the goods, but title to the respective purchasers. There were then major and very material exchanges as to why Plazadome was prepared to supply goods to the Appellant with the benefit of 11 days' credit for the element of price that was in fact only paid when Zaanstrait had itself paid the full price, and 6 weeks' credit for the VAT-gap element of the price. Similarly, having just been involved in Deal 1 where a supplier had failed to deliver promised goods, the Appellant was giving 11 days’ undefined credit to a small Dutch company of which the Appellant had little knowledge.[66]In relation to the agreement by Plazadome to grant credit, Mr. Patel said: "It was not a written agreement, it was over the phone. And it was only on those agreements, if they were happy with that, that I will do business with them because, as I said, we had difficulties in making the payments straight away because we had to go to the banks, do the CHAPS transfers, they arrived two or three days later and they were quite happy to agree with that, so I was able to trade with them; otherwise, there would be no trading with Plazadome, if they were not happy with that particular".[67]Again in response to why there were delays on the part of the Appellant in making payment for 11 days and 6 weeks, Mr. Patel replied as follows:
"My priority was to pay as much of my suppliers and whenever the funds were available I paid them. So obviously why the delay or why the reason was that the funds had taken so long to put together, it's obviously due to the banking, and the time gaps that were available at the time to cause these delays." 68 When asked by the Respondents' counsel what made the Appellant such a good customer able to secure such good credit terms, Mr. Patel said: "
Well, we were very good paymasters. We paid every bill we had, every invoice, whether it was a week late or two weeks late but the bills were paid by us. Why they offered the credit facility, I have no idea but our terms were strictly with them that we can't do this business if you wanted payment straight away but when we are commit we know we have the money in some of our accounts."[69]In view of the fact that Mr. Patel explained on a number of occasions that the delays in making payments, and in particular the delay for six weeks in paying the residue of the purchase price to Plazadome, were all due to banking mechanics, we should mention that we were shown a copy of the Appellant's NatWest account, and a copy of the instruction to pay the final £11,569 to Plazadome. There may perhaps have been other accounts that we were not shown, but so far as the NatWest account was concerned, it was not until immediately before the payment was made to Plazadome that there was anything like the required amount in that account to make the payment. Furthermore the bank made the payment on the very day on which the payment instruction was sent by the Appellant. Other minor points[70]The final point that we will mention in relation to Deal 2 was that there appeared to be some doubt about the product description of the goods sold. Various different invoices in the supply chain, release notes and inspection reports described the goods in different ways, containing more or less detail and sometimes discrepancies. This point was never finally resolved though the Respondents' counsel thought that it appeared that the goods supplied to the Appellant and by the Appellant to Zaanstrait had been dual-core processors, whilst it was suggested that Zaanstrait wanted single-core processors. As we have said, this point was never resolved and the most that we will say was that Mr. Patel was unable to clarify any confusion that there was. The facts in relation to Deal 3[71]Deal 3 was by far the most important deal, both from the perspective of the Appellant because the deal was by far the largest, and from the perspective of this Appeal because Deal 3 provided the most illuminating facts.[72]Deal 3 involved the purchase of 15,750 Intel Pentium CPUs from MVS (the company owned by Mr. Patel's brother-in-law) for a VAT inclusive price of £1,278,781.88 and the sale of those CPUs to a company registered in Spain, namely Alpha, for £1,153,687.50. While Alpha’s purchase order was dated 7 August 2006, the remaining deal documentation, including the release notes were all dated 10 August. The origin of the deal[73]According to Mr. Patel's evidence, he had been hoping that the Appellant would do a deal with MVS, and as far back as January 2006 he had undertaken some due diligence in relation to MVS. He had had no opportunity to effect a deal with MVS, however, and the origin of Deal 3 was not an offered supply by MVS but an approach, out of the blue, by Alpha, a company with which he was not familiar. Mr. Patel said that he understood that Alpha had located the Appellant on the free Alibaba website. Due diligence and Alpha’s referees[74]Dealing first with the limited due diligence in relation to MVS, we were shown a Redhill confirmation of MVS's valid VAT number addressed to the Appellant on 2 February. We were shown a Creditsafe report into MVS which gave a credit limit of £500. In the familiar manner, when asked about his reaction to this credit limit and Mr. Patel's readiness to do a substantial deal with MVS, Mr. Patel replied:
"I haven't been through them, as I said yesterday. I have not read into depth of the Creditsafe reports. If you want to ask something, I don't mind, but I haven't read them. It wasn't something that I had delved deep into to see every figure and what it all meant in there."
[75]Due diligence in relation to Alpha was very limited, consisting basically of a Redhill check to which the positive response returned on 4 July, a few documents in Spanish, a sheet indicating that Alpha’s main bank was FCIB and a sheet giving Alpha’s two referees, namely MVS and Supreme. As we have already indicated Mr. Patel said that he had not reviewed these in depth, and had not spotted that one of the referees was the company that ended up 5 or 6 weeks later supplying the stock to the Appellant that the Appellant would supply to Alpha and that the other referee was his own former employer. The limited terms in the documentation[76]Alpha’s purchase order, dated 7 August, contained no terms about payment or title, and the only indicated terms were "Manufacturers' warranty" and that the Appellant would pay the freight charges. The delivery address was given, namely that of a freight-forwarder in the Netherlands.[77]The Appellant's pro forma invoice and invoice were both issued on 10 August. The former indicated that payment should be made to the Appellant's FCIB account and gave bank details, and it also indicated that the buyer should insure the goods during transit, since the Appellant would not do so. Whether this made sense in view of the fact that the Appellant was paying for the freight carriage and was liable to deliver the goods at the customer’s warehouse was not explored. The actual invoice made no mention of the bank details or the point about insurance. Whilst the Respondents' counsel considered this to be of great significance, we will ignore this relative detail.[78]The Appellant's purchase order to MVS made no reference to payment or title, but did mention that the goods were to be released at the relevant freight-forwarder's warehouse in Feltham. MVS's invoice gave its FCIB bank details for payment, and said that "Goods remain the property of MVS Digital Ltd until paid for in full". The way in which payment was made by the Appellant[79]All the contention in relation to Deal 3, save for the significance of the fact that Mr. Patel failed to spot that MVS was one of Alpha’s referees, occurred around 5.00 p.m. BST on 10 August.[80]As we indicated in the Introduction, all payments between all 12 companies involved in what emerged from the FCIB evidence to be a circle, were made in 84 minutes. The first payment was made by the foreign company referred to as Komidex. Although the payment times on the Paris FCIB server indicate the timing gap between each payment with certainty, it is fairly well known that nobody is certain which time zone was used on this server when recording the time at which FCIB payments were made. Since both the Dutch and Paris servers are in the same time zone, one hour ahead (in August) of BST, and since facts in relation to releases in the UK in this case tend to confirm the likelihood that the FCIB timings were based on continental European times (again CEST in August), we will make the tentative assumption that that is the case. It is not central to any of our conclusions in this case that that assumption is right, but it seems a sensible assumption to make.[81]As we also indicated in the Introduction, the far more relevant fact in relation to the payments in Deal 3 is that all of them, including the Appellant's payment to MVS, but with the sole exception of MVS's payment to its supplier, were made from a computer with the same IP address.[82]It was only at the start of the hearing that Mr. Patel and the Appellant's advisers spotted this point about the common IP address. When this was spotted, Mr. Patel provided a further Witness Statement, which sought to explain what had led to the Appellant’s payment being initiated by someone logging on to the internet from the common IP address. He explained that when he had provided his initial Witness Statement he had forgotten the following details and that he had only recalled them once the Respondents' documents had drawn to his and his advisers' attention the feature that his payment, and virtually all the other payments had been made from a computer using the same IP address.[83]The further Witness Statement and later oral evidence in the course of cross-examination maintained that there had been absolutely no discussion about payment mechanics between the owner of MVS, his brother-in-law (Mr. Anil Patel, who we will refer to as "AP") and himself prior to the moment (at somewhere around 5.00 p.m. BST on 10 th August) when Mr. Patel was in his car, and AP rang him on his mobile phone. The claim by Mr. Patel was that AP said to him that he needed to obtain the Appellant's payment in respect of Deal 3 (or at least the element of it matched by the Appellant's receipt from Alpha). Mr. Patel said that he fortunately had in his wallet a folded sheet of A4 paper that gave the log-in details, Mr. Patel's user ID for his FCIB account and the individual test key required for making the next payment from that FCIB account. Since Mr. Patel fully trusted AP, since the significance of the test key was that anybody being given the three relevant pieces of information could only make the one single next payment, and since that next payment was owing to AP's company anyway, he was perfectly happy to give AP these three key requirements that would enable AP to make the Appellant's payment to MVS. He accordingly gave the three details to AP during this phone call.[84]Not surprisingly, if Mr. Patel was indeed in his car and receiving a call on a mobile phone, Mr. Patel said that at the point he received this phone call, he did not know whether the Appellant had indeed received its payment from Alpha. The FCIB evidence indeed makes it clear that whoever logged on to the FCIB website to make the Appellant's payment to MVS logged on before the Appellant's account had been credited with that receipt, but that the Appellant's payment to MVS was then made within 3 minutes of the person logging on to the website. FCIB would only have made that payment to MVS once the Appellant's account was in funds to make it, so that it is clear that within the space of 3 minutes, someone logged onto the FCIB website to make the Appellant's payment, the funds arrived from Alpha and were credited to the Appellant's account and the matching payment was made to MVS. According to the FCIB timings given on the Paris server, Alpha’s payment was made at 5.30 p.m. and the Appellant’s matching payment at 5.33 p.m. Making the tentative assumption that the Paris server timings were based on CEST, the payments would have been received and made at 4.30 and 4.33 p.m. BST respectively.[85]While dealing with payments, we will mention that the balance of the Appellant's payment to MVS, the element matching the VAT gap, namely the quite significant amount on this deal of £125,163.14, was paid by the Appellant, using on this occasion a different IP address from which to log on to the FCIB web-site, on 16 August.[86]Before leaving the facts in relation to payments, we must mention that the Respondents' evidence demonstrated that the company Komidex that commenced the payment flow in the Appellant's chain and received the money back again 84 minutes later was the initiator of circular payment flows in numerous other payment chains, involving payments flowing through roughly 130 companies (some of them admittedly being involved with payments in the various different chains). We do not of course know that all the payments in the other circular chains were made with quite the 84-minute efficiency of the ones in the Appellant's case, but it was obvious that all of these chains had been pre-arranged and that Komidex had been the central company in all the chains. The release of the goods by MVS and by the Appellant[87]Two other things occurred around 5.00 p.m. GMT on 10 August. First, assuming that the time settings on MVS's fax machine were set at the correct time, at 4.56 p.m. MVS sent a fax to the freight forwarder, confirming that the goods could be released to the Appellant. We do not know for certain that at this point MVS had received the Appellant's major payment, namely the one triggered by the mobile phone conversation to Mr. Patel in his car, but it seems reasonable to suppose that it would have done. After all its payment terms actually indicated that title would be retained until it had received full payment so that it seems unlikely that it would have instructed the freight-forwarder to release the goods to the Appellant without any Ship to Hold reservation if it had not received the major payment of £1,153,600. Indeed, these timings do seem to confirm our assumption in relation to the time zones used on the Paris FCIB server, because if our tentative assumption is correct then MVS received the major payment at 4.33 p.m. GMT and sent the release instruction at 4.56 p.m. GMT.[88]The release document just referred to was followed by a release document faxed by the Appellant at 4.59 p.m. (according to the timings on the Appellant's fax machine). Another fax, referring to the first and changing one box number, because one was said to have been wrong in the first fax from the Appellant, was also sent, according to the Appellant’s fax timer, at 4.59 p.m.[89]We will need in due course to make a number of comments and findings in relation to all these matters concerning the payment mechanics, and the timing points, but will defer doing this until explaining our decision. Minor points[90]The Respondents' counsel finally raised a number of points in relation to the fact that whilst the goods were held by the Appellant, they were not insured, and seemed probably not to have been covered by any insurance held by the freight forwarder. There was also discussion about poor inspection reports. We will not summarise this detail because we consider it to be relatively immaterial in relation to the points that even the Appellant's counsel conceded constituted the "possible smoking-gun” in this Appeal. The facts in relation to Deals 4 – 7[91]We can deal with the facts in relation to Deals 4 – 7 quite shortly, largely because the deals were very similar. The deal documentation[92]All four deals involved sales of 4G Apple iPods to the Dutch company Zaanstrait. Deals 4,5 and 7 involved purchases from the company that we have referred to as Unique, whilst Deal 6 involved a purchase from Plazadome.[93]All the relevant documents in relation to deals 4, 5 and 6 were dated 29 September. By "all the documents" we mean, the respective parties' purchase orders and invoices, and the release instructions by both the Appellant's supplier to the Appellant and by the Appellant to Zaanstrait. The only irrelevant qualification to the common dates on the documents was that Zaanstrait's purchase order for Deal 6 was missing.[94]Addressing the terms of the documents, the only single reference in the documents for all four deals to payment terms or retention of title was the one invoice from Plazadome. As with Plazadome’s invoice for Deal 2, this gave the payment term as TT/ADV, and said that Plazadome would retain title until it had been paid in full. No other document contained any term whatsoever. Whether or not this was in conflict with Plazadome's invoice term just mentioned or not, the release facts in relation to the three deals documented on 29 September were that all parties released them on that date, the Appellant to an identified driver employed by Zaanstrait's Dutch transport firm who collected them in the UK. At this point, nobody had paid any of the respective purchase prices. Payment details[95]The actual payments for Deals 4 to 6 were made as follows. The Appellant paid Plazadome and Unique the modest amounts representing the VAT gap on 4 October. Then on 10 October, Zaanstrait paid the full price in respect of the three deals, and on the same date, the Appellant on-paid the amounts relevant to each of the three deals to the relevant supplier.[96]The dates and payment dates in relation to Deal 7 were different, though in reality they raised the same points. The documents were dated 20 October 2006, save that oddly the purchase order was dated 24 October. The releases were also dated 20 October, the driver of Zaanstrait's forwarder again being nominated but a different man from the one who drove the truck on 29 September. Zaanstrait's payments were not made until 9 and 28 November, and we do not have the payment date or dates for the Appellant's payments to Unique. Consistent profit margins[97]We should mention two points in relation to profit margins and percentage profits. The Respondents stressed the oddity that the Appellant always made a profit, assuming recovery of VAT, of 4%. In relation to this, Mr. Patel said that he was more interested in the margin as the cash amount of profit rather than as a percentage, though he also said that his normal aim was to make a 6% profit but that he was often haggled down to a lower figure. The more relevant point, and one where we immediately concede that there was no reason to suppose that Mr. Patel would have known about this, is that in Deals 6 and 7, the Appellant's immediate suppliers' figures indicated that they made quite significant losses.[98]There is no particular relevance to why those suppliers made such losses. One possibility just worth mentioning is that the Respondents' CD evidence that we have otherwise found virtually irrelevant does suggest that for a long period Plazadome and Unique had been heavily involved in MTIC transactions. It appears possible therefore that the losses that each recorded in respect of Deals 6 and 7 might have been pre-arranged losses that enabled the iPods to circulate again at realistic prices, when some previous exporter's profit margin had increased the price for the product above a realistic looking level. And were this so, and were this procured by the mastermind behind the transactions, it would have been perfectly easy for the two suppliers to be recompensed in some way to reverse the losses. Whatever the explanation it has no great relevance in this case.[99]Whilst the facts in relation to the losses, and the unknown explanation for them, are irrelevant, the comments that Mr. Patel made when the Respondents' counsel asked him whether he could account for the losses were in fact quite revealing. The exchange was a follows: Counsel:
"Look at deal 6 for a moment. …. Can you explain why Plazadome agreed to sell to you at a loss?"
Mr. Patel:
"I don't know why they made a loss. I wouldn't have expected people in this game would make a loss but if these figures are correct and they made a loss, I can't explain why they made a loss."
Counsel:
"You make the biggest margin in deal 7, as we have seen. Can you explain why Unique Technology sell to you at a loss?"
Mr. Patel:
"Again I have no idea why they made a loss. I would assume that they would have made a profit because they would normally not need to make a loss. They are just buffers."
Whilst there is no relevance to the fact that Mr. Patel had no idea why both Plazadome and Unique had made losses in Deals 6 and 7, we do find the references to "this game", "buffers" and the quite clear assumption as to what the buffers were actually doing "in this game" all to be fairly revealing.[100]During the cross-examination in relation to Deals 4 to 7, Mr. Patel was asked about the margins made in all the deals, and the respective levels of margin that buffers and exporters might expect to make. Again we preface the quotation that we will record very shortly by the proposition that in genuine grey market trading, one might expect quite varied profit margins and losses, with countless factors potentially occasioning opportunities to make profit and risks of incurring losses. In responding to the question on the relatively stable margins made by the Appellant, and the difference in profit margins in UK to UK trading and export trading, we find the following extract again to be revealing: Counsel:
"You did some UK to UK trading?"
Mr. Patel:
"I never done any UK trading."
Counsel:
"You knew the most profitable place to be in these supply chains was as the exporter?"
Mr. Patel:
"As far as profit is concerned, yes, I knew that anyway from being in the industry: that to be an exporter gave more profit than to be a buffer, as we put it."
Counsel:
"So anyone who was in this market knew that the best, most profitable place to be, if they wanted to make money, was as the exporter?"
Mr. Patel: "Well, in this assumption that, yes, if you wanted to make a lot of money in this business, you would need to be an exporter, yes”. Summary of facts relevant to all transactions, and to the Appellant's general mindset[101]Having given the facts in relation to the seven Deals, and a few examples of the points that we will now mention, it is simplest to summarise generally the Appellant's or rather Mr. Patel's attitude to: · due diligence; · product description; · inspections of stock by freight forwarders; and · insurance.[102]The significance of these points is not so much that we regard any of them as particularly significant in themselves, but rather because Mr. Patel's attitude to each of them was significant. Due diligence[103]Mr. Patel's attitude to due diligence was particularly significant. We regard the emphasis that HMRC themselves placed on due diligence to be somewhat regrettable since it almost invited brokers or exporters to emphasise the lack of anything damaging known about their immediate trading partners, and to emphasise the difficulty of obtaining any information about parties further removed up or down the supply chain, so appearing to insulate brokers from any knowledge of the fraudulent losses. Mr. Patel's attitude to due diligence was, however, somewhat novel in that, having claimed as a general matter in interviews with HMRC officers that he was alert in looking for indicators of fraud, he treated due diligence as an exercise to collect a few bits of paper and information for HMRC, in case they wished to see them later, but not as an exercise where anyone expected him to consider the information gained. Creditsafe reports were ignored, partially explained by the traditional claim that the Appellant was providing no credit to its suppliers, but more generally in this case because we believe that Mr. Patel barely glanced at them. They were just something to collect. Referees[104]Similarly, having asked trading partners to provide referees, Mr. Patel never gave a moment's though to taking up references, but he appears never even to have glanced at who the proffered referees happened to be. This omission was particularly significant in relation to Deal 3. Product description[105]In relation to "product description", there was a particular dispute about the CPUs that featured in Deal 2, and we were indeed far from clear that Mr. Patel knew much about the different types of CPU, and the distinction between single-core and double-core processors. Again the impression was that some item would flow down and along a chain of suppliers and so long as he described it in the same way on his purchase order and his invoice as his supplier and customer had done that was "job done". The issue that some chosen description might in fact be far too general actually to identify the required product or the offered product seemed not to matter. Of course if the description was just a tag to attach to some product that was to pass along a chain of companies, this attitude was understandable, but if a bona fide trader was genuinely seeking to identify a precise product required, or offered, the descriptions were sometimes lacking. Inspection reports[106]The same approach applied to product inspections and the reports provided by freight forwarders. Wherever a report drew attention to some defect in the products, Mr. Patel never appeared to give thought to whether the defect was material to his customer, and never sought to obtain further information about any possible problem drawn to his attention. He appeared to suggest that his customers would always have seen the inspection reports, and provided that they were not troubled by any minor defects mentioned, there was no need for him to enquire about the possible defects. We were, however, far from clear that the Appellant's inspection reports were in fact passed to the Appellant's customers, and believe generally that they were not. Mr. Patel finally suggested that as no customer had ever complained about any defects when product had been delivered, this confirmed that his relaxed approach to defects mentioned in inspection reports had been justified. Insurance[107]Most brokers in MTIC transactions took out insurance for product when it was owned by them and when it was transported at their risk. We make no particular criticism of the fact that Mr. Patel generally notified his customers, even where he was responsible for carriage, that the Appellant would not be insuring the goods, but that customers should do so. We never touched on whether this made sense; i.e. as to whether the customer had an insurable interest in product still owned by the Appellant, being transported by the Appellant at its risk, and whether the Appellant would have still been liable to provide equivalent product if uninsured product being transported at its risk was lost or stolen. Since none of this was explored, we will ignore it.[108]The fact about insurance that we did note was Mr. Patel’s claim that product owned by it held in the freight forwarder's warehouse was believed by him to be insured on a general policy held by the freight forwarder, when the full Terms and Conditions provided by the freight forwarder appeared to undermine this expectation. The difficult question 109 Again therefore the approach shown by Mr. Patel was one of neither knowing nor caring. Whether this general attitude was the result, to be blunt, of very extreme naïvety or stupidity, or on the other hand a realistic but not particularly tactful reaction to the fact that known product was just flowing very quickly along a chain, with no likelihood of anybody complaining about anything is the difficult question that featured during this Appeal. The contentions on behalf of the Appellant[110]The Appellant’s main contentions were as follows, namely that: HMRC’s decision had been a Kittel decision on Deal 1, just as on all the other transactions, and it was not open to HMRC to seek to defend that decision by contending that there had been no supply; if the first contention was wrong, it was nevertheless still for the Respondents to have the burden of proof in relation to their claim that there had been no supply in Deal 1; as a legal matter, and for the various reasons given in paragraph 136 below in dealing with our decision in relation to Deal 1, there had in any event been a supply for VAT purposes; on the Respondents’ fallback Kittel contention in relation to Deal 1, the Respondents had failed to demonstrate any connection to a fraudulent loss of VAT because they had failed to trace the transaction back to any identified defaulter, and so far as Maystar was concerned, it was entirely possibly that their default was occasioned by insolvency rather than fraud; the recent ECJ decisions of Mahagében (C80/11) and David (C142/11) had clarified the Kittel test and indicated that the notion that an appellant ought to have known of the connection of its transactions to fraudulent VAT losses was tantamount to a need for the Respondents to demonstrate that the appellant must have known of that contention, or at least that the explanation for absence of actual knowledge resulted from a deliberate aim to remain ignorant, rather than from genuine innocence and ignorance; and that it was insufficient for the Respondents (in seeking to demonstrate knowledge or means of knowledge) to list characteristics of the Appellant’s trading, most or all of which could subsist in legitimate grey market back-to-back trading, for instead the Respondents had to produce “objective evidence” that established the knowledge or means of knowledge to the standard of the balance of probabilities. The contentions on behalf of the Respondents[111]It was contended on behalf of the Respondents that: the Decision Letter had refused the claim for input tax in Deal 1 on the basis that there had been no supply; in any event the decision on Deal 1 could be justified on reasoning that differed from the reasons originally given for the denial; the burden of proof in relation to the non-supply contention was nothing to do with fraud, and so the burden of proof had to be on the Appellant to demonstrate that there had in fact been a supply, which plainly the Appellant could not, and did not, demonstrate; the denial of input tax in relation to Deal 1 could be justified in the alternative (if there had been a supply), on Kittel grounds, on the reasoning that the Appellant’s transaction had been connected to fraudulent VAT losses and that the Appellant knew or ought to have known of that fact; the recent decisions of Mahageben (C80/11) and David (C142/11) had not refined the Kittel test. Had they sought to do this it would have seemed that this would have been made absolutely clear. The two recent cases should be read in the context of the precise points referred to the ECJ, namely points concerning whether national taxing authorities could deny input claims by the recipient of a supply because of failings in documentation remote from the provision of the VAT invoice itself; in this Appeal, it was specifically claimed that Mr. Patel had lied and that it was clear from the surrounding circumstances that the Appellant knew, or must have known (the equivalent of actually knowing) that its transactions were connected to VAT frauds; but that in any event, on the fallback approach, the Appellant certainly ought to have known of that connection, and it was not right to treat the claim that the Appellant ought to have known of the connection to fraud as being equivalent to a claim that he “must have known” of the connection. Our decision Deal 1[112]Deal 1 raises the five separate issues listed in paragraph 9 above, and we will deal with each in turn. We will deal with all the issues in case our decision is the subject of an appeal, which might render issues that we consider irrelevant in fact to be relevant.[113]The first two issues posed in paragraph 9 above were worded carefully to reflect the fact that the decision of Mr Justice Blackburne in CEC v Alzitrans SL (2003) All ER 288 (para 38-39), the decision in BUPA Purchasing Ltd & Others v CEC (2) (2008) STC 101 , and the decision of the First-tier Tribunal in Plazadome Limited v. HMRC [2009] UKFTT 229 (TC) all indicate that a distinction must be made between “the decision” and “the reasons for the decision”. According to those authorities, there is firstly no compulsion on HMRC to indicate reasons for their decision, and secondly if they do give reasons for a decision, they can subsequently advance quite different reasons to support the decision, provided that as a matter of litigation practice it is fair for them to do so. The first issue - the reasons, in the decision letter, for the disallowance of input tax[114]Whilst the second issue posed in paragraph 9 is more important than the first if we are right to follow the legal distinction just mentioned, the Respondents did contend that not only the decision, but even the expressed reasons for the decision given in HMRC’s decision letter of 12 March 2010, included the non-supply point. We must obviously thus decide whether this contention is right because if it is, then it is obvious that the Respondents can now seek to sustain their decision by reference to the non-supply contention.[115]In the manner that has become conventional in MTIC appeals, the decision letter commenced with two general paragraphs, the first of which referred to the various cases on the doctrine of “abuse”, and the Kittel principles. The second paragraph then merely made the point that connection to VAT fraud on Kittel lines could be traced through a contra-trader.[116]The letter then contained the following two paragraphs:- “The Commissioners are satisfied that the transactions set out in the attached appendix form part of an overall scheme to defraud the Revenue. The Commissioners are also satisfied that there are features of those transactions, and conduct on the part of Celltec Computers Ltd, which demonstrate that you knew or should have known that this was the case. Accordingly your right to deduct the input tax claimed in respect of these transactions is denied”.[117]The next paragraph then said that “This decision affects input tax…” , and it gave the disallowed figures and identified the VAT periods 09/06 and 12/06 that were affected by the decision.[118]The paragraph after that commenced with the relevant wording “In the making of this decision the Commissioners have taken into account the following information and features……” and most of the rest of the letter then listed factors that had been taken into account. Those factors were listed in relatively familiar form, much of it obviously drawn from a template simply adapted to refer to the specific facts in relation to the Appellant and the seven deals. Paragraphs (a) to (u) which ran to three pages itemised factors that were said to have supported the proposition that the Appellant “knew or ought to have known that the transactions were artificially contrived”. Paragraph (e) was then the one and the only one that referred specifically to the “non-supply” point, saying the following:- “(e) In the case of the deal with Maystar Enterprises Ltd, you confirmed that no goods were supplied even though you had documents indicating that payments were made to Maystar Enterprises and received from your customer Dipro AG, Switzerland. You paid £102,228 for the goods without checking their existence and your customer also paid for them without ensuring what he was purchasing. However, you still attempted to recover the input tax on goods that you knew were not supplied”.[119]Our conclusion in relation to the issue of whether “non-supply” was treated as the, or a, reason for the disallowance of the claimed input tax is that it was not. The whole tenor of the letter was along traditional Kittel lines, and the paragraph just quoted that referred to the fact of non-supply was amongst a list of factors clearly enumerated to establish why HMRC claimed that the Appellant knew, or ought to have known, of the connection of its transactions to fraud. Even the middle sentence in that paragraph was concentrating on the artificiality of the Appellant’s dealing, and not on the simple fact of “non-supply”. We accept that the last sentence in the passage just quoted referred specifically to the fact that the Appellant “knew that the goods were not supplied”, but we still consider that the fair interpretation of the paragraph, appearing in a list of factors said to demonstrate “knowledge or means of knowledge”, was that it was emphasising the Appellant’s obvious awareness of the artificiality of the refund claim, and not the simple fact of non-supply. The second issue - Whether HMRC can sustain a decision by advancing different grounds from those initially given[120]This is a more difficult issue and it requires us to distinguish between the decision itself, and the reasons for the decision.[121]We should start by saying that we accept the contention of the Appellant’s counsel that a disallowance of a claim for input tax based on a successful Kittel claim is different in nature from one based on the proposition that there has been no supply at all. The consequence of the “non-supply” approach is to eliminate the supply, the output liability on the part of the claimant’s counter-party (i.e. the non-supplier, however described), and the very existence, in the hands of the claimant, of input tax. By contrast, the consequence of the Kittel contention is that there has still been a supply; the supplier should still have accounted for input tax, and may have done so, and the sole point is that by virtue of ranking as a participant in the fraud, the claimant forfeits its right to the deduction for the input tax under European law.[122]We accept the line of authority referred to in paragraph 113 above that HMRC can advance a different reason to support their original decision. In distinguishing between the decision and the original reasoning, the crucial paragraphs of the HMRC letter of 12 March 2010 are the paragraphs that we quoted in paragraph 116 above. The first of those two paragraphs was clearly referring to the supporting reasons for the decision, and we conclude that the decision itself was that given in the second paragraph. In other words the decision was contained in the words:
“Accordingly your right to deduct the input tax claimed in respect of these transactions is denied.”
[123]The reference there to a “denial of the right to deduct input tax” appears initially to suggest that the decision was conceding the existence of input tax, and merely denying the right to deduct it. That approach suggests that even the decision was exclusively a Kittel based decision and not one that was even consistent with the alternative proposition that there was actually no input tax, and indeed no supply and no output tax liability on the part of Maystar, the intended supplier. The opening word “Accordingly” also links the decision to the reasoning, and makes it clear that the decision derived originally from Kittel reasoning. In view of the conclusion reached in relation to the first Deal 1 issue (in paragraph 119 above), the feature that the reasons for the original decision were all geared to Kittel principles was indeed obvious.[124]Notwithstanding these points, the essence of the decision was simply to deny input tax in relation to all seven deals. Furthermore although the phrase “your right to deduct input tax is denied” might suggest that HMRC conceded that there was available input tax and that the decision merely refused the right to deduct it (i.e. a decision implying a Kittel approach) the actual decision sentence was referring to “the input tax claimed in respect of these transactions”. The fact that the Appellant had manifestly claimed that there was available input tax that it ought to be able to deduct undermines the initial supposition that the wording of the decision assumes available input tax, and is just confined to denying the right to deduct it on Kittel lines. For whether or not Deal 1 generated input tax for the Appellant or not, the Appellant was certainly claiming that it did and that the input tax was deductible, and it was that “ claimed input tax” the refund of which was denied by the decision.[125]We accept that if the reason for denying relief for input tax had been that there had been no supply, the decision might more obviously have been worded along the lines that since there had been no supply, and no VAT liability, and accordingly no input tax credit for the Appellant, there was nothing that could be deducted. Nevertheless, worded as it was worded in the letter of 12 March 2010, the decision simply denied the claim to deduct input tax. Even if there had been no supply, and no input tax, the Appellant was still claiming that it should be able to deduct the claimed input tax credit, and we conclude that the denial of the right to make such a claim was a sufficiently broad decision to encompass a denial where there had been no such input tax at all. Accordingly we decide that the actual decision, included in that crucial sentence in the second of the paragraphs quoted in paragraph 116 above was sufficiently broad to justify HMRC in now sustaining that decision on different reasoning than that originally indicated.[126]We also decide that since it had been clear for many months that HMRC was seeking to sustain its decision to disallow the claim for input tax in relation to Deal 1 on “non-supply” grounds, this being made particularly clear from HMRC’s Statement of Case and Skeleton Argument, it was legitimate for HMRC to pursue this contention in the Appeal hearing.[127]We note in passing that the approach adopted in the Plazadome case was similar. In that case the original reason for the disallowance had been an abuse contention, and had that been correct, its result would have been to re-analyse the transaction and to eliminate any available input tax to be claimed. The Regulation 14 approach, subsequently advanced to support the disallowance, assumed that the acquirer had received a supply, and that it simply lost its right to deduct input tax because of the failure to meet the terms of the Regulations, requiring the invoice to identify the goods or services supplied. And on those facts, the Tribunal had concluded that the decision could be sustained on grounds that differed from the original grounds and indeed that involved a similar difference in nature between the two different grounds for denying a refund of input tax to the one in this present Appeal. The third issue – which party bears the burden of proof in relation to matters relevant to the question of whether there was a supply, and our findings of fact in relation to that matter, consequent upon the answer to the burden of proof question[128]It was clear that Maystar had not delivered the Intel CPUs to the Appellant, and that the Appellant had thus had no opportunity to deliver the CPUs to its own customer, Dipro. It is equally fair to say that Maystar had contracted, if only orally, to supply the CPUs to the Appellant, had furnished an invoice to the Appellant in respect of that supply, and had received full payment for them. What is not clear is: whether the CPUs ever existed at all; if they did, whether some party higher up the chain of sales might have short-circuited the supply chain and delivered the CPUs straight to Dipro, and whether (this being a mixed matter of fact and law), if such a direct delivery had been made to Dipro, the conclusion should then be that that delivery would have discharged all other contracted movements between the intermediate parties, such that Maystar should be treated as having discharged its liability to supply to the Appellant. The burden of proof issue[129]These major gaps in our understanding of what happened render the issue of who bears the burden of proof a very important question.[130]The claim by the Respondents that there was no supply to the Appellant in relation to Deal 1 has nothing to do with a Kittel challenge. We accept that all matters relevant to a Kittel challenge are ones involving allegations of fraud with the result that the burden of proof in relation to all such matters passes to the Respondents. The claim by the Respondents that there was no supply is in no way dependent on any claim about fraud.[131]We conclude that the ordinary rule, requiring the Appellant to satisfy the burden of proof in establishing its case applies to the supply issue.[132]The Appellant did not seek, and manifestly could not have sought, to show that there had been any actual delivery by Maystar to the Appellant. So far as factual claims were concerned, two points were advanced. One was that the responses by the freight forwarder, to the effect that title had not been passed to Maystar clearly indicated that the goods did exist. This was of possible relevance to one of the legal contentions. The second claim was that the feature that Dipro ceased to complain to the Appellant about the non-delivery of goods that it had paid for made it entirely credible that some other party (most obviously a party in the supply chain above Maystar) might well have delivered the goods directly to Dipro. And particularly if we had reached the opposite conclusion on the burden of proof issue, it might have been very difficult for HMRC to show that there had not been such a direct supply to Dipro.[133]We cannot say with certainty that the CPUs definitely existed, but on the balance of probability we accept that the information from the freight-forwarder to the Appellant that title had not passed to Maystar indicated that the goods existed was credible, and that the probability is that the goods did exist and were held by Forward Logistics to the order of some totally unknown company.[134]By contrast, there is no ground for concluding to the required standard that some unknown party in fact delivered the goods directly to Dipro. We see from the letter from the Insolvency Service that Maystar had been making third party payments, and it is perfectly possible that the money paid by the Appellant to Maystar could have been paid to a third party and reimbursed to Dipro. That would have accounted for Dipro ceasing to pursue claims against the Appellant. In any event this is all pure speculation, as indeed is any suggestion that the CPUs were delivered directly to Dipro, and that this impacts on the Appellant’s claim for input tax. We find that it is not demonstrated that anybody delivered the CPUs to Dipro. 134. We will nevertheless deal below, in relation to the legal contentions, with the possibility that the CPUs were delivered directly to Dipro since the better view seems to us to be that even if that was so, this would still not involve the conclusion that the intermediate supplies (i.e. from Maystar to the Appellant and from the Appellant to Dipro) should be treated as having occurred.[135]We consider that the fact that Dipro ceased to pursue any claims against the Appellant, having made one phone call to complain about non-delivery, has other significance relevant to the fallback Kittel claim by HMRC, and will deal with that below. The fourth issue - T he law in relation to “supply” and “non-supply” Contentions by the Appellant[136]Whilst the Appellant’s principal contention had been that HMRC’s decision had been entirely based on Kittel reasoning and that HMRC could not now seek to sustain that original decision in relation to Deal 1 on “non-supply” grounds, during the hearing the Appellant did advance some arguments in relation to the Respondents’ “non-supply” claim. It was argued: · that the case of C&E Commissioners v. Oliver [198] STC 73 established that there had been a supply in this case; · that the law in relation to supply for VAT purposes was distinct from the ordinary commercial law, had a special and a broader meaning than the ordinary commercial law and that it was clear that for VAT purposes the contract, the payment and the invoice did mean that there had been a supply; · that HMRC’s practice in relation to “shrinkage” indicated that there should be treated as being a supply in this case; · that the fact that HMRC had assessed Maystar for the tax in respect of the supply to the Appellant confirmed the supply analysis; and · that when the provisions in relation to “time of supply” made it clear that a supply was treated as having been made when payment for the supply had been received and an invoice issued, this must involve treating a supply as having been made when those timing rules have been triggered. The lack of further submissions[137]There were some unavoidable, but still unsatisfactory, features about the hearing in this case. Since Andrew Young had been instructed only at the last minute, he had been unable to produce a Skeleton Argument, and none was ever produced that dealt with the various points just summarised. Furthermore the Respondents had had no idea that the Appellant would challenge the claim that in Deal 1 there had been no supply, so that no attention was initially given to that matter.[138]Our purpose in requesting written submissions from the parties at the end of the hearing was largely geared to the hope that both parties would deal with the legal issues in relation to what demonstrated a supply for VAT purposes. Neither did, though the Respondents commented on the likely facts. Our decisions on the five points just recorded Oliver[139]The case of C&E Commissioners v. Oliver established that when a second-hand car dealer sold stolen cars, he could not escape VAT. The VAT Tribunal’s decision had been in favour of Mr. Oliver on the ground that as the sale had been void, and did not comply with the legal requirements of the law of contract, there could be no supply for VAT purposes. Mr. Justice Griffiths’ decision, on appeal, was that there had been supplies. Since possession of the cars had passed to the buyers, with relatively little risk of this ever being challenged because all identifying marks had been erased and replaced, this conclusion seemed hardly surprising. When, however, the test of supply propounded by Mr. Justice Griffiths is considered, the suggestion that the case had any bearing on this case becomes extraordinary. Mr. Justice Griffiths said that “ ‘Supply’ is the passing of possession in goods pursuant to an agreement whereunder the supplier agrees to part and the recipient agrees to take possession. By ‘possession’ is meant in this context control over the goods, in the sense of having the immediate facility for their use. This may or may not involve the physical removal of the goods”.[140]Our decision is that the Oliver case undermines, rather than supports, the case that there had been a supply to the Appellant. There had been no form of release of the goods either by Maystar to the Appellant, or the Appellant to Dipro. There were indeed contractual obligations to make the supplies, but there was no question of the Appellant (or indeed Dipro) taking possession, control or the immediate facility to use the CPUs. The VAT meaning of “supply”[141]On the second point we accept the point that for VAT purposes, the notion of a “supply” is meant to cover supplies of every description, but there is no special definition of the term for VAT purposes. Applying the ordinary rules of commercial law (in relation to which no guidance was given by either counsel), we might well accept that if the owner of goods contracted to sell specific goods at the point when payment was received, then under the contract, title and risk might pass at that point, and it would follow that there would be a “supply”. We do not consider that there had been a supply by Maystar, however, when Maystar had not acquired title to the CPUs; when we have no knowledge of any seller to Maystar or the terms of any such sale; when Maystar’s contract with Appellant made no mention of when the goods would pass, and the payment terms were “Other”. The most likely intended moment for delivery would have been when the goods were released to the Appellant, inevitably after they had been released to Maystar, and since neither release had occurred, we conclude that there had been no supply on general grounds. The treatment of “shrinkage”[142]Nobody explained HMRC’s practice in relation to “shrinkage” to us, though we understood that if A agreed to supply 20 items to B, and one was stolen, HMRC would treat the supply as still being of the 20. We can certainly understand this if risk and title had passed to B by the point of the theft or other loss. For then the reality would have been no different from the case where 20 were delivered and one of the items was subsequently stolen from B. If on the other hand, title and risk remained with A at the point when one item was stolen, we would have expected the analysis to be that A had supplied only 19; A should issue a credit note to B simply reflecting the non-supply of one item, and A should refund the price or pay damages for having failed to honour its contract to supply the 20 th item. Equally, A could properly claim on its insurance assuming that its goods (for at this point, they would be “its” goods) were insured. The assessment on Maystar[143]Addressing the fourth point, we understand that the assessment on Maystar was made before HMRC appreciated the point that the goods might not have been supplied. We also accept that whilst it is obvious that the liquidator of Maystar, with Maystar presumably having no assets, would not have appealed against the assessment, HMRC could (on learning of the non-supply) have withdrawn the assessment, though the result as regards Maystar would have been immaterial. We however accept the Respondents’ argument on this limited point, to the effect that we should pay regard to what actually happened and not be bound by the assessment that was not withdrawn. We entirely understand the requirement for VAT to achieve neutrality. If, however, the legally correct analysis is that there has been no supply, then that analysis should prevail as regards the acquirer, regardless of whether the supplier might fail, in due time, to appeal against a wrong assessment on the wrongly claimed supply. And the same conclusion should prevail if HMRC fail to withdraw what transpires to have been a wrong assessment, particularly where to have withdrawn it would have had no effect on Maystar. The “time of supply” point[144]On the fifth and final point, we consider that the provisions fixing the “time of supply” do just that, but that if on a proper analysis there has been no supply then there is no VAT point to be fixed. This seems to be well illustrated by the example that there could hardly be a supply of goods if A contracted with B to supply goods that did not exist (or that A did not own or possess) and then invoiced B and received payment. And that appears to be precisely what happened in this case. If authority is required for the proposition that the “supply” is central to the supplier’s liability to account for VAT, it seems to us that the ECJ case of Terra Baubedarf-Handel GmbH v. Finanzamt Osterholz-Scharmbeck [2005] STC 525 confirms this. Conclusion on the legal issues[145]We accordingly dismiss all five points advanced in argument by the Appellant and conclude that not only are the Respondents right to seek to defend their decision in relation to Deal 1 on the “non-supply” basis, but that there was indeed no supply of the relevant CPUs from Maystar to the Appellant in Deal 1. The Appellant’s claimed input tax was correctly denied in the decision letter, and was rightly justified on the “non-supply” reasoning. The “short-circuit” supply[146]We indicated in paragraph 134 above that we considered that on the balance of probabilities the Appellant had not established that some party in the supply chain had delivered the CPUs directly to Dipro. Were that finding of fact to be disputed, we add the point that we would still not consider that any such “passing” of the CPUs would necessarily justify the conclusion that Maystar had thereby satisfied its obligation to deliver to the Appellant. In a chain involving contractual liabilities of A to transfer to B, B to C, and C to D, we would expect the analysis to be that each transfer had been effected if each agreed that A’s transfer to D would satisfy the intermediate steps. But if A transfers directly to D, with B and C oblivious to this transfer, and possibly deliberately excluded from involvement (possibly because one of them is in, or shortly to be in, compulsory liquidation) we would not accept that the intermediate transfer liabilities had been discharged. The facts in this present case clearly fall into that latter category.[147]It follows that we dismiss, now for a second reason, any claim that Maystar made a supply to the Appellant, merely because of unfounded speculation that some unknown party might have passed the relevant CPUs to Dipro. The fifth issue - the “fall-back” contention that in any event the disallowance of the input tax in Deal 1 can be justified on Kittel grounds[148]At this point, we will deal only with the issue of whether in Deal 1 HMRC has established that the Appellant’s transaction was connected to a fraudulent VAT loss. We will deal with the “knowledge or means of knowledge” issue in relation to Deal 1 when dealing with that critical issue in relation to all seven deals.[149]We initially considered that we would conclude that HMRC had not established on the balance of probability that the Appellant’s Deal 1 transaction was connected to a fraudulent VAT loss. The reason for this was that while HMRC had not sought to trace any chain of supply back to companies or “buffers” in any supply chain above Maystar, we still assumed that there would almost certainly have been such a chain. It then seemed initially to follow that as: HMRC had manifestly not identified some importer of the relevant CPUs as the fraudulent defaulter that had failed to pay the initial output tax; and it was far from clear that Maystar’s own default (on the alternative analysis that it had made a supply and was liable for VAT on that supply) was fraudulent, we would have to conclude that no fraudulent default had been demonstrated. So far as Maystar is concerned, we add the points that although the letter from the Insolvency Service, and the disqualification of Maystar’s director from being a director for 12 years all indicated that Maystar was heavily involved in MTIC transactions, we still assume that Maystar was intended merely to perform the role of a buffer company. We accept that Maystar could still have been liable for the gross VAT on the supply to the Appellant (on the “supply” analysis), either because in making no return it manifestly failed to claim any input tax, or because it might have been held to suffer joint and several liability for the initial non-paid VAT. Those facts appeared to make it clear that while Maystar had failed to meet certain or possible VAT liabilities, this had quite possibly resulted from Maystar’s compulsory liquidation, and not from any fraudulent plan that Maystar should default.[150]We nevertheless conclude that on the balance of probability the Appellant’s Deal 1 transaction was connected to a fraudulent VAT loss even though we cannot identify the supposed fraudulent defaulter. The factors that have led us to this conclusion are as follows: We simply cannot believe that Dipro’s intended purchase of the CPUs from the Appellant could have been a legitimate grey market transaction in the light of the fact that Dipro pursued no further action, beyond one phone call, when it failed to receive the CPUs that the Appellant had contracted to supply and that Dipro had paid for. In a bona fide situation, Dipro would rightly have regarded Mr. Patel’s response to that enquiry, that his own supplier had failed to deliver to the Appellant, to be irrelevant to the fact that the Appellant had breached its contract with Dipro and should at the very least either have sourced the product elsewhere or refunded Dipro’s payment. If Dipro’s lack of complaint stemmed from the speculative possibility that some party higher up the supply chain had supplied directly to Dipro, one would perhaps have expected Dipro in a bona fide situation to notify the Appellant that “all was now well”. Additionally the fact of that direct supply would undermine any expectation that the various intermediate sales had all been honest and quite independent transactions. When Mr. Patel was asked why he was content to deal with Maystar, and confident that Maystar could source product that was going to cost approximately £100,000 when Dipro’s credit limit had been put at £500, Mr. Patel explained rather revealingly that he expected Dipro not to need any significant capital because it would only be passing the Appellant’s payment up the line, and implicitly only paying for goods when it itself had been paid. He was not expecting Maystar to have the goods in stock, or to fund their purchase in any way other than one that implied a chain transaction, and a pre-arranged transaction. It seems inconceivable that such expectations and such payment terms would have existed between parties of no credit standing in genuinely independent transactions. The information that we saw from the Insolvency Service indicated that Maystar had been heavily involved in MTIC-type trading from at least September 2005, and had participated in transactions with a total turnover of “at least £225 million”. It had made third party payments to parties remote from any supply chain notwithstanding that it had been warned by HMRC that this practice indicated MTIC fraud. Whether Maystar had been a buffer only dealing between UK parties, or an exporter we do not know, but it was indisputably heavily involved in MTIC trading. In the context of the present question of whether we are satisfied that the Appellant’s transaction was connected to fraudulent VAT losses, the fact that Mr. Patel may have been wholly ignorant of Maystar’s past activities is irrelevant. The point that is decisive is that we consider it inconceivable that Maystar’s transaction with the Appellant, days before Maystar was closed down, could have been an innocent transaction. Whilst this point is a point of general application to all MTIC trading, and one of relevance to the “knowledge or means of knowledge” issue as well as the present issue, we consider the feature that an export transaction of a product that must inherently have been imported into the UK that generated a 4% profit for doing nothing, particularly in a chain where presumably at least one other company was making a profit for doing nothing is inherently indicative of VAT fraud. For if the VAT has been paid on importation, VAT will inherently contribute to neither profit nor loss at the point of export, save possibly for a small element of interest cost where VAT is paid before it is recovered. In contrast the transport, insurance, freight forwarding and other incidental costs of importing and then exporting product are more likely to occasion loss than profit, or to render the transactions non-viable. Regular profits made in futile back-to-back trading, with incredible matched payment terms, such that the export trader only needs to finance the “VAT gap”, and well appreciates that he only needs to finance the VAT gap, are strongly indicative of participation in MTIC fraud.[151]Our decision on HMRC’s fallback contention on Kittel grounds in relation to Deal 1 is that the above four factors justify the conclusion that the Appellant’s Deal 1 transaction was connected to fraudulent VAT losses. We leave aside at this point the subsequent issue in relation to “knowledge or means of knowledge”. The connection between the Appellant’s transactions and the fraudulent VAT losses[152]We now give a formal finding that in all seven Deals, the Appellant’s transactions were connected to fraudulent VAT losses. This finding is based, in the case of Deal 1, on the conclusions recorded earlier, and is only material if the “non-supply” conclusion is overturned on appeal. In relation to Deals 2 to 7, it is appropriate to give a formal finding on the “connection” point, notwithstanding that the Appellant conceded both the connection point in relation to these six deals and the fact that it was obvious that the deals had involved pre-arrangement, even if it was not conceded that the Appellant was a party to that pre-arrangement. Knowledge and means of knowledge The law[153]We will defer giving our short conclusions in relation to the points of dispute between the parties in relation to the correct way to apply the Kittel test until we give our conclusions on the “knowledge and means of knowledge” issue in relation to Deals 1,2 and 4 to 7 below. General considerations 154 . We will deal first with our general understanding of what Mr. Patel appreciated about MTIC fraud and risks.[155]In many MTIC appeals, appellants will have received a number of visits from HMRC officers prior to the disputed deals and they will have been taken through all the features of MTIC trading. Evidence will generally show that they will have been given a copy of HMRC's Notice 726 in relation to joint and several liability and appellants will often confirm that they are familiar with the content of that Notice.[156]It is likely that Mr. Patel was given a copy of Notice 726 but that was back in 2005 at a time when the Appellant was dealing only with domestic retail sales of split lots of hardware product. In that context it was understandable that Mr. Patel would not have read the Notice, albeit that in view of our understanding of the general attention that Mr. Patel paid to all documents, we rather doubt whether he would have read or fully understood the Notice had it been of immediate relevance to the transactions that he was then undertaking.[157]There was some dispute about whether at a later point the Appellant had received a general warning letter from HMRC, also enclosing a copy of Notice 726. The letter was sent to the wrong address, however, and we think it extremely unlikely that the Appellant saw either the letter or the Notice on that occasion.[158]We think it also fair to record something which, while not particularly complimentary, also occasioned some doubt in both our minds that the Appellant knew of the connection of his deals to MTIC fraud, namely that we were far from clear of the degree to which Mr. Patel did actually understand MTIC fraud in any detail. His English was not particularly good, as will have been made clear by even the very few extracts that we have taken from the transcript, and we were in some doubt as to whether Mr. Patel was simply "way out of his depth" in the transactions with which he became involved, or a knowing participant, perhaps feigning some of the ignorance.[159]These doubts were eventually swept away by the facts in relation to Deal 3 to which we will turn below, but at this stage we will list the various factors that led us to conclude that Mr. Patel knew quite enough about the reality of VAT fraud and MTIC risks to have been a very knowing participant in such deals.[160]We note firstly that Mr. Patel was actually surrounded by people, all of whom appear to have been, at least on HMRC's evidence, involved in MTIC trading. His previous employer, Supreme, was said to have been so involved; Mr. Singh who became his co-director at the time the Appellant's trade both switched largely to wholesale export deals from retail business, and at the time when the level of trading increased substantially, had also been employed in a company alleged by HMRC to have been involved in MTIC-type trading. Mr. Patel’s brother-in-law was the owner of MVS, whose role in the present Appeal suggests considerable awareness of MTIC trading. Finally, every one of his counter-parties, both suppliers and customers, appears to have been involved in MTIC trading at quite an extensive level. We have already said that we found HMRC's evidence about the earlier involvement of Plazadome, Masterpiece and Unique in many MTIC deals nine months prior to the deals under appeal to be irrelevant to the Appellant's awareness that its deals were connected to fraud, but certainly that evidence, and the facts in relation to those companies in earlier appeals (at least in one of which the present Tribunal Judge was the judge) confirm a certain awareness of MTIC fraud on the part of those companies. Equally Zaanstrait has featured in an earlier appeal in which again the present Tribunal Judge was the judge.[161]The feature therefore that, surrounded by these parties, all involved to a greater or lesser degree in MTIC transactions, Mr. Patel remained ignorant of the essence of MTIC trading seems highly implausible. While Mr. Patel claimed that he was duped in Deal 3, a claim that we dismiss, the prospect that he was duped in all seven deals and the considerable number that had been undertaken in the previous nine months, seems remote.[162]Mr. Patel was also very familiar with the feature that, "in this game" to quote him, the finance that he required was the modest amount required to fund the VAT-gap, in other words the excess of the VAT that he hoped to recover over the profit margin in each deal assuming such recovery. He periodically made the point that the capital that he and his family could provide (mostly or all by way of loans) entirely governed the level of business that the Appellant could undertake and the constraint was always geared to the need to finance the VAT gap. It was thus implicit that the bulk of the purchase price would not have to be financed at all, either because the supplier would release the goods on credit and wait for payment until the Appellant's customer had paid the Appellant, or because the customer would pay for goods before delivery. The assumption that these were the hallmarks of the trading in which he was to be involved inevitably meant to our mind that that trading was pre-arranged fictitious trading, and not the type of trading and trading terms to be expected in arms' length grey market trading.[163]We also note that Mr. Patel expected his suppliers to have similar trading terms. When asked how Maystar, with a poor credit rating, could be expected to purchase goods for a very considerable price, Mr. Patel had no hesitation in replying that the point was irrelevant because he expected Maystar also to have back-to-back deals and payment terms that would eliminate the need for it to finance purchases outright either. This was the way things worked "in this game".[164]Mr. Patel was certainly also aware that in the type of trading in which he was involved, there were "buffers", "buffers" made only modest profits, and it was the exporter that made much the biggest profit. That was why he always wanted to be the exporter, and indeed why he always was the exporter, albeit of course that this preference limited the level of trading because of the need to finance the VAT gaps.[165]We finally consider that anyone with mere average intelligence and the most elementary knowledge of how the VAT system worked (and having worked at Supreme, Mr. Patel must have had that level of knowledge) would appreciate that there was no regular and legitimate reason why exporters might make significant and often relatively consistent levels of profit in exporting product that had inevitably been imported in the first place. No CPUs or iPods are manufactured in the UK, and so all must have been imported. Since the product purchased was appreciated always to be held in some freight forwarder's warehouse, and to be likely to pass through the hands of various buffers whilst held in that warehouse, it was even implicit that the products were likely to have been imported quite recently. If VAT is properly accounted for on importation, and then recovered on export, the significance of VAT is to create a fairly modest cash flow disadvantage to the import/export transaction. Making further deductions then for the double transport costs, the possible insurance cost and the freight forwarder and inspection costs, honest import/export transactions are rather obviously unlikely ever to be viable, some very special factors of special purchases apart. It is only, of course, VAT fraud that makes such transactions viable, and that has tended to produce the sort of profits expected by exporters in MTIC transactions.[166]The conclusion that we reach, taking all the above points into account, is that while Mr. Patel may not have had the full warnings from HMRC officers that other appellants have often received in advance of the challenged transactions, all these surrounding facts do lead us to conclude that Mr. Patel did know quite enough about the essence of MTIC trading for us to conclude that he was unlikely to be an innocent dupe, and an innocent dupe in numerous transactions. We conclude that he knew the game in which he was playing, and he knew the essence of the frauds that proffered the type of profits that the exporter had come to expect. Deal 3[167]By far the most critical facts in relation to Deal 3 are those relating to the way in which the payments were made. Before we reach the payments, however, we need to address the way in which it was claimed that the deal was initially put together.[168]The Respondents' evidence made it clear that this deal, and indeed all the similar looking Komidex deals, all of which were circular, must have been pre-arranged. The Appellant conceded that the deal must indeed have been pre-arranged, but Mr. Patel claimed that he had been unaware of this at the time. He must have been duped into playing his role.[169]Since Mr. Patel claimed that the origin of the deal commenced with a contact from Alpha that had located the Appellant on the Alibaba website, and Mr. Patel claimed that neither Alpha nor anyone else had intimated to him that he should source the product requested from MVS, we find it rather extraordinary that it just happened by coincidence that Mr. Patel decided to try to make his first ever purchase from MVS, and that MVS happened to have, or be able to acquire, precisely the product and in the quantities that Komidex obvious planned to rotate. Even if Mr. Patel was duped into participating, we consider it highly improbable that the persuasive introduction could have omitted some very clear intimation of how he might source the product.[170]Turning to the claims in relation to the payment mechanics and the mobile phone call and the provision (without any prior mention of payment arrangements) of the FCIB code keys to AP, we do not accept this evidence. It is obvious that Mr. Patel must either have made all the payments by the 11 companies or must have provided his test keys to someone else. It is equally obvious that the latter possibility is infinitely more likely. What is far from obvious, however, and indeed we conclude that it is actually inconceivable, is the extraordinary claim that Mr. Patel just happened to be reached on a mobile phone, and happened to be able to provide from his wallet (virtually immediately before payment was made by the Appellant) the test keys so that AP could make or arrange immediately for someone else to make the Appellant's payment.[171]It seems clear that some mastermind (quite possibly Komidex) managed to so pre-arrange matters that all the payments were in fact made within 84 minutes. It is incredible to suppose that when all the remaining payment steps must have been efficiently pre-arranged that just one step would be left to the pure chance that Mr. Patel could be contacted at the crucial moment, or in fact that Mr. Patel would have his mobile phone on when in the car; that the car would be in an area with mobile coverage, that Mr. Patel would happen to have in his wallet all the information required to enable someone else to make the FCIB payment for the Appellant, and that Mr. Patel would be prepared to supply that information to AP. We do not accept that the mobile phone call explanation is true. It simply must have been the case that prior to the last minute arrangements Mr. Patel would have been party to some properly organised pre-arrangement as to how his payment would be made. That must have been the situation for all 10 other companies and we simply do not believe that one step in the chain, for the making of the swift payments that were achieved, would have been left to pure chance.[172]Mr. Patel had explained when responding to cross-examination that he had probably been in the car at the time of the phone call because it was around the time of his daughter's birthday, and he and his wife were probably taking the daughter out to celebrate her birthday. We have no idea what Mr. Patel's private arrangements were, but since apparently the daughter's birthday was not for at least three days on 10 August, this claim seems somewhat dubious. More significantly, we know that Mr. Patel must have been in his office when he sent the release faxes that we referred to in paragraph 88 above. The feature that MVS's release instruction was given with a fax time of 4.56, and the Appellant's two faxes with fax times of 4.59 rather suggests that the fax timers were reasonably accurate. If that was so and the Paris server timings were based on continental European timings, it very much emerges that Mr. Patel was still in the office , not in his car, sending the faxes to effect releases, after the time at which the claimed mobile phone call was said to have been made.[173]Our conclusion that Mr. Patel's evidence in relation to the phone call to him in the car was untrue is so critical to our decision in relation to Deal 3, and indeed in relation to the whole credibility of Mr. Patel, that we consider it fair to test our conclusion by supposing that the mastermind behind the payment arrangements might have considered it vital that the one innocent party, duped into participating in the circular transaction, might have had to be excluded from the properly arranged prior planning, and his payment might have had to be left to pure chance in the interests of not arousing the suspicion of that innocent party. We consider this possible scenario to be ruled out by the following factors, namely that: · it stretches credulity to suppose that on the initial approach from Alpha Enterprises, Mr. Patel just happened to choose to source product for the first time from the entity that was obviously intended by the mastermind behind the transactions to make the supply; · if the Appellant was to be left out of the pre-planning involving all parties except MVS making their payments through the use of one computer IP address, why would the plan in relation to the Appellant involve this last minute attempt to re-insert the Appellant into that chain, rather than leave the Appellant to pay on the following day if its continued innocence was so critical? · Mr. Patel's innocence was put under considerable scrutiny by the fact that he claimed not to have looked "in depth" at (i.e. to have even read) the disclosure that MVS was one of Alpha Enterprises' referees; · whilst Mr. Patel's claim to have forgotten all about the way in which he gave the FCIB test keys to AP is especially damning in the context of our conclusion that he was actually a party to a properly organised earlier plan that that would be done, it is still highly suspicious that he claimed that he had forgotten, until the IP address evidence emerged, the facts about the phone call to the car, even if that story was true; and · the fact that MVS released product to the Appellant without any reservation when MVS had yet to receive the final £125,163 of the purchase price, contrary to its own term that title would be retained until full payment had been received, and contrary to Mr. Patel's answer to Officer Rajput that the Appellant neither received nor gave credit, again suggests pre-arrangement between MVS and the Appellant.[174]Our conclusion is that these other considerations confirm our conclusion that Mr. Patel must have been party to arrangements made at a far earlier point as to how his payment would be made on the Appellant’s behalf.[175]It follows from that conclusion that Mr. Patel must have appreciated that his involvement in the earlier planning was a vital fact central to his "knowledgeable participation" in Deal 3, and therefore the claim that he had forgotten this crucial fact, when producing his initial Witness Statement, becomes a vital piece of deception rather than an example of merely improbable forgetfulness. In consequence, we simply do not believe that this fact had been forgotten. Until the IP evidence emerged, this coordination of the payment mechanics simply had to be concealed.[176]Our overall conclusion therefore is that much of Mr. Patel's evidence in relation to the crucial facts of Deal 3 was untrue. We conclude that he was a party to the pre-arrangement of the payment mechanics; that he knew that that fact had to be concealed and that it indicated that he must have known that the deal was pre-arranged. In the context of the prevalence of MTIC fraud, and bearing in mind Mr. Patel's basic awareness of the essence of that fraud and the feature that Mr. Patel cannot possibly have regarded Deal 3 as a legitimate grey market trading transaction, there can be no other explanation for the deal than that it was a fraudulent deal in which some party had fraudulently failed to pay VAT. We decide that Mr. Patel must have known that that was so and that the Respondents have amply demonstrated that the Appellant actually knew of the connection of its role in Deal 3 to the fraudulent evasion of VAT. Our decision in relation to Deals 1 and 2 and 4 to 7[177]Having dealt generally with what we believe Mr. Patel must have known about the background to MTIC fraud, we now turn to our decision as regards all Deals, except Deal 3. We have already concluded in relation to Deal 3 that the Respondents have amply satisfied all the Kittel tests, and we have given our decision on all the technical points, but not the Kittel "knowledge" point in relation to Deal 1. We therefore now address the "knowledge" points in relation to all the deals except Deal 3. The legal test[178]We first address the test that we must apply, and deal in other words with the Appellant's counsel's contentions in relation to "subjective and objective evidence", and the claim that the European jurisprudence has now almost reached the point at which the "ought to have known" limb of the Kittel test means that "the appellant must have known of the connection to fraud".[179]On the crucial subject of our needing to reach our conclusion as regards "knowledge" on the basis of objective evidence, we believe that the test that we must apply is as follows. Since the result of a decision against an appellant on Kittel grounds is based on the analysis that the appellant has been "a participator in the fraud”, it seems to us that this is most obviously satisfied where the appellant had actual knowledge of the connection to fraud. Actual knowledge obviously refers to the subjective question of what the appellant actually knew and thought. We entirely accept that we must reach any conclusion about subjective actual knowledge on the basis of cogent evidence and not just speculation. Since, however, we ought to be concentrating on the subjective question of what the appellant did actually know, we find the contrast between subjective and objective evidence to be slightly confusing, and if anything to be misleading. Ignoring at this point the "ought to have known" limb of the Kittel test, the test that we mean to apply in this Appeal is whether we consider there to be cogent evidence that the Appellant actually knew of the connection of its transactions to VAT fraud.[180]We are deciding this case substantially on the basis of actual knowledge so that we are barely concerned with the "ought to have known" element of the Kittel test. Since, however, it seems clear that a person who ought to have known something cannot by definition be said to have known it, and the expression "must have known" plainly implies actual knowledge, we cannot agree with the Appellant's counsel's suggestion that the "ought to have known" and the "must have known" notions are identical. We will shortly mention below the only basis on which, as a fallback conclusion, we say that the Appellant in this case "ought to have known" of the connection to fraud, should our decision on actual knowledge be overturned, but beyond that we will make no further comment on the "ought to have known" limb of the test.[181]We specifically endorse the Respondents’ counsel’s claims about the scope of the two recent ECJ decisions. They essentially dealt with the precise questions addressed to the ECJ by the local courts, and have little or no bearing on the Kittel test. Not that the converse was suggested, we certainly say that the Mahagében case has no bearing on the “supply” or “non-supply” issue. There was no finding as to whether the acacia logs existed or not, and everything depended on the documentation. The two possible analyses[182]There are two possible analyses in this Appeal. The one is that Mr. Patel knew that all his transactions were connected to VAT fraud. He knew all about “the rules of the game”; the role and profitability of “buffers” and “brokers”; the need to finance the VAT gap; the expectation that the vast majority of an exporter's purchase price of product would be matched by an earlier receipt of the customer's payment or by undefined supplier credit, and all about the significance to be attached just to the credibility of the immediate counter-parties, revelling in the inability to know anything about other more remote parties.[183]The other possible analysis is that Mr. Patel was extraordinarily naïve or stupid, and that he effected a number of transactions in the belief that by making a few phone calls, back-to-back deals in exactly matched quantities and descriptions of product could regularly be put together so as to make commercially risk-free, but still honest, profits in return for doing, bluntly, nothing. Insofar as that might be criticised as merely a prejudicial way of expressing the only non-fraudulent way of describing all broker's transactions in MTIC trading, we do not consider that it is a summary and an explanation that we can reject out of hand in this case. While we have reached the very critical conclusion that the Appellant lied in two if not three crucial respects in giving his evidence in relation to Deal 3, at other times we found Mr. Patel's bewildering lack of understanding of so many aspects of what he had done to render it just possible that he did blunder around in ignorance, genuinely thinking that his transactions were honest, without understanding how implausible that belief might be. At one point, indeed, it was the Respondents' counsel who was so bewildered by Mr. Patel's answers to questions that he asked Mr. Patel just to confirm whether or not he had been involved at all with the running of the Appellant's business. The answer was that he alone had run the business, but we were not entirely surprised that the Respondents' counsel eventually aired that question.[184]Our conclusion is that the Appellant in fact knew that all of its transactions were connected to VAT fraud. We justify this conclusion on the following grounds.[185]We have already reached the conclusion just mentioned in relation to Deal 3, and in doing so concluded that Mr. Patel had lied about the last-minute provision of the payment details to AP, lied about having forgotten the information about the payment mechanics, saying that he only remembered this detail when learning about the common IP address, and he must almost certainly have lied about the way in which having received an approach from Alpha, it was entirely his own choice to source the product from the company that had obviously been identified in Komidex's overall planning.[186]The dual facts that Mr. Patel did know of the connection of Deal 3 to fraud, and that he lied about it in very material ways are of course critical to any suggestion that his other deals were "honest and ignorant", and that his evidence is to be believed.[187]In paragraph 160 above, we summarised the way in which Mr. Patel was surrounded by people involved, in one way or another, in MTIC transactions. We find it inconceivable that he did not learn all about the large profits seemingly readily available in connection with VAT fraud perpetrated by others. We are particularly struck at the way in which the Appellant's turnover increased dramatically and its trading pattern changed around the time when Mr. Patel was joined on the board by Mr. Singh, the brother of the main director of Supreme, and a man who HMRC claimed in evidence had been employed by a company also involved in MTIC activity.[188]The payment reality of the Appellant's transactions was that even when one supplier indicated in writing that payment in advance was required, this was ignored and the payment terms, ostensibly orally agreed, were just flexible. The evidence about being 6 weeks late in paying the balance of the Deal 2 purchase price, with the supplier seemingly not complaining, the delay being attributed to "banking delays" was not credible. Mr. Patel periodically referred to "having to go round to the bank", and to the 3-day delays wrongly said to arise when CHAPS payments were made. The truth was that the Appellant could not afford to pay the element of the price matching the customer's payment until the customer had paid, and generally he delayed in paying the balance. He said that he was given these unusual and undefined credit terms because he was a good payer. We reject this claim. In bona fide grey market deals, suppliers would expect to be paid immediately or they would check the credit worthiness of their customer. When the reality was that the customer (i.e. the Appellant) could only have paid the bulk of the price when the Appellant's customer paid, the supplier would not deal without knowing how certain the deal with that customer might be. The perfectly obvious reality in this case is that all parties were operating on the basis that there was a pre-planned chain so that the back-to-back payments could be anticipated with confidence. No other explanation is possible.[189]Our conclusion that the Appellant knew of the connection to fraud in relation to Deal 1 (which is only relevant in relation to our fallback decision in that case) is based both on the Appellant's overall circumstances, applicable to all the other deals, but in particular on the fact that the absence of any further complaint on the part of Dipro, when it failed to receive goods that it had paid for, must have indicated a pre-arranged transaction that somebody else sorted out in some way unknown to us. Had Dipro been an independent and honest company, genuinely wanting particular product that it had paid for, would not the failing supplier have heard more about its total failure to supply than one phone call in which Dipro received an irrelevant response? We would accept that if all other evidence suggested that the Appellant had been honest and that it had only been when Dipro failed to complain that the Appellant suddenly suspected fraud and pre-arrangement that this realisation might well have arisen at too late a point to be relevant. We believe, however, that the Appellant knew of the reality from the start. There was not the slightest suggestion that when Dipro failed to complain it suddenly occurred to Mr. Patel that there might be something suspicious about Deal 1.[190]It is unbelievable that an honest trader, however naïve and ignorant, would have despatched goods in Deal 2, having just encountered what most honest people would have regarded in Deal 1 as a terrible problem, without having received payment for them or made any payment to the Appellant's own supplier.[191]Mr. Patel's bewildering disinterest in relation to product inspections, the condition of product, and the description of product, is easily understood if the reality was that in pre-arranged chains, product would pass down a line of parties, so that the only test to be surpassed in describing it was to ensure that the purchase orders and invoices looked right. We conclude that the reality was that the Appellant knew that the product was something that somebody else had set on its preordained path down the line, and that it, the Appellant, just had to copy the product descriptions from paperwork originating with others.[192]We mentioned Mr. Patel's references to "this game", and the existence and margin expectations of "buffers" in MTIC transactions in paragraph 99 above. We found these remarks to be revealing.[193]We consider that HMRC's excessive emphasis on the significance of due diligence has been a regrettable feature that has probably tended to encourage rather than diminish MTIC transactions. However, as we have already said, Mr. Patel's contempt for due diligence, extending to not reading responses and to claiming that the only purpose of asking the questions was to get and file various bits of information, should HMRC later call for it, has raised disregard for due diligence to new levels. The failure to spot that Alpha gave two such significant companies as its referees was one extraordinary example of ignoring everything. We also believe that Mr. Patel obtained but never read credit reports. This next conclusion is a minor one in our overall thinking, but we actually attribute this failure to consider evidence to two general arguments that MTIC traders will have resolved to rely on after the ECJ's decision in the Optigen case in early 2006. One would be the invariable claim that poor credit scores on the part of suppliers were not relevant because suppliers were not making payments to the exporter, notwithstanding that poor credit scores might throw into question the general standing of such companies. The other would be the general plan to protest total faith in the immediate counter-parties and be able to feign regret that it was obviously impossible to obtain any information about more remote parties.[194]Our conclusion is that Mr. Patel, and thus the Appellant, did know and must have known of the connection of all seven deals to VAT fraud. The “ought to have known” conclusion[195]The only respect on which we say that, on a fallback basis, the Appellant "ought to have known" of the connection to fraud, were our decision on actual knowledge to be challenged, is that the Appellant appears to have traded in a manner designed to know the absolute minimum about every trading counter-party, and to have paid the least regard to every aspect of its trading with a view to turning a blind eye to every possible indicator of fraud. "The less known the better" appeared to have been the Appellant's stance, possibly encouraged by the ECJ's decision in Optigen. On a fallback basis we say that the Appellant ought to have known of the connection of all seven transactions to VAT fraud.[196]Without trying to be frivolous, we can perhaps clarify the different situations in relation to “means of knowledge” by giving three variations on the Nelsonian example.[197]If Nelson has seen the French and Spanish ships with his good eye, and then aimed the telescope in their direction but to his blind eye, that is plainly a case of knowledge.[198]If Nelson knew that a particular course had to be set to engage the enemy; had not seen the ships, but nevertheless scanned the horizon with the telescope to his blind eye, he would still have been deliberately ensuring that he failed to ascertain the relevant facts. That involves the deliberate lack of knowledge that we attribute to Mr. Patel in this case.[199]If Nelson was incompetent and misread the compass and looked with his good eye in the wrong direction, when any more competent admiral would have seen the ships immediately, that is negligence. We make no comment, though other authorities have done, on whether in that situation it would have been apt to conclude that Nelson should have known that the ships were indeed on the other horizon.[200]We simply say that it is our judgment that Mr. Patel deliberately sought to know the minimum about the surrounding circumstances of the transactions and he deliberately refrained from considering and analysing the information that he did have. In answer to the obvious question of whether he would have known of the connection to fraud had he not deliberately sought to know the bare minimum, and had he not shut his mind deliberately to all sensible enquiry, our decision is that he would have done. Costs[201]Both parties applied for their costs if successful, and we award the Respondents their reasonable costs on the standard basis, the quantum to be decided by the appropriate Costs Office in the absence of agreement between the parties. Right of Appeal[202]This document contains full findings of fact and the reasons for our decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) Tax Chamber Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice. HOWARD M. NOWLAN TRIBUNAL JUDGE RELEASED: 25 February 2013