BAILII [Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]

United Kingdom VAT & Duties Tribunals Decisions


You are here: BAILII >> Databases >> United Kingdom VAT & Duties Tribunals Decisions >> Megtian Ltd v Revenue & Customs [2008] UKVAT V20894 (11 December 2008)
URL: http://www.bailii.org/uk/cases/UKVAT/2008/V20894.html
Cite as: [2008] UKVAT V20894

[New search] [Context] [View without highlighting] [Printable RTF version] [Help]


Megtian Ltd v Revenue & Customs [2008] UKVAT V20894 (11/12/2008)
    20894

    VAT – input tax – transactions connected with fraud? – Appellant knew or ought to have known? – yes – appeal dismissed.

    LONDON TRIBUNAL CENTRE

    MEGTIAN LIMITED

    (In administration) Appellant

    -and-

    HER MAJESTY'S COMMISSIONERS OF
    REVENUE AND CUSTOMS

    Respondents

    Tribunal: Richard Barlow (Chairman)

    Praful Davda FCA (Member)

    Sitting in public in London on 16, 17, 18, 21, 22, 23, 25, 28, 29, 31 July and 1 August 2008.

    Mrs Penny Hamilton of counsel instructed by the Khan Partnership LLP, solicitors, for the Appellant

    Mark Cunningham QC and Daniel Margolin counsel, instructed by Messrs Howes Percival, solicitors for the Respondents

    © CROWN COPYRIGHT 2008


     

    DECISION

  1. In this appeal the appellant disputes two decisions given by the respondents by which they disallowed input tax claims it had made giving rise to claims for repayments. The decisions were conveyed by letters dated 21 May 2007 and 10 July 2007 and the total amount of input tax for which credit was refused is £5,909,067.50. The disallowed input tax was claimed in respect of two one month prescribed accounting periods namely April 2006 (£2,799,737.50) and May 2006 (£3,109,330).
  2. The appellant was incorporated on 22 July 2002 and commenced trading in mobile phones in December 2003 having previously traded in herbal products and toothbrushes.
  3. There are three shareholders of the appellant, Mr Anthony Andreou (as he was the principal manager of the company we will refer to him as Mr Andreou), Mr Christopher Andreou and Mr Nicholas Nicolau. Mr Andreou was the sole director at material times and Mr Christopher Andreou was the company secretary.
  4. The respondents refused the claims for input tax credit on the basis that the transactions in question, of which there were 15 in each of the two periods concerned, were connected with fraudulent evasion of VAT. The case is one in which the respondents allege "MTIC" fraud has occurred and that the appellant should have known that each of the transactions in question was connected with fraud. The respondents did not make a specific assertion that the appellant knew of such fraud but did invite the tribunal to make such a finding if the evidence justifies it. The respondents do assert a positive case that each of the relevant transactions was connected with fraud and that the appellant should have known that was the case.
  5. In outline MTIC fraud typically operates in the following way. Where goods are bought by a trader (T) registered for VAT in the UK from a trader (E) registered for VAT in another EU country E does not have to charge T VAT on his purchase but when the goods are brought to the UK T has to account for acquisition tax of 17.5%. However, T is also entitled to deduct that amount as input tax and so at that stage the goods are in effect VAT free or tax neutral. If T sells the goods to another trader registered in the UK (T2) he must charge and account for VAT on the sale. If, having collected that VAT from T2 he then absconds without accounting to the authorities for that VAT, T will have made a dishonest profit of 17.5% of the value of the goods. T2 will have deducted the tax he paid to T as input tax and has paid T the tax inclusive price for the goods. T2 will recoup the VAT paid to T from his customers when he sells the goods and, although the authorities have lost the output tax due to them, the direct source of T's illegitimate gain is the money paid to him by T2 and indirectly by T2's customers. Assuming he acted dishonestly that is a straightforward fraud by T.
  6. However, such a fraud depends upon T being able to find a purchaser who wants to sell the goods in the UK because otherwise the purchaser will have paid the VAT inclusive price without being able to recoup it by onward sale to customers.
  7. MTIC fraud depends upon T2, rather than selling the goods in the UK, exporting them to another country which may or may not be in the EU. In that case T2 makes a zero rated sale to the overseas buyer. He still recovers the tax he paid to T as input tax but does not have to collect and account for any tax from his customer and so not only do the authorities lose the tax that T should have paid when he sold the goods to T2 they are also the source of the money T has absconded with. That is because T2 can afford to pay T the tax inclusive price knowing that he will recover the tax from the authorities as input tax. T's opportunity to deal repeatedly with T2 is therefore increased because there is no need for T2 to find customers in the UK in order to be able to account for output tax after paying T the tax inclusive price.
  8. The fraud still depends upon finding customers in the overseas markets but, if whoever is behind the fraud can manipulate that, then the opportunity for repeated fraud can be achieved. Typically, additional UK traders may be interposed between T and T2 to disguise the fact that the fraud is occurring.
  9. None of the above means that T2 is necessarily involved in the fraud because he may well not be aware that his overseas customers or T are in any way involved in fraud and both he and any interposed traders may all be innocents who have unknowingly been manipulated by whoever is behind the fraud.
  10. In the terminology adopted in such cases T is known as the missing or defaulting trader T2 is known as the broker and any interposed parties are known as buffers.
  11. We have decided that the appeal is to be dismissed and that is the decision of both members of the tribunal. Before giving our reasons for that, we would like to record that all the professional persons involved in this case on both sides have played an admirable part in ensuring that the evidence and papers (contained in no less than 41 lever arch files) were presented to us in ideal fashion and that the conduct of the oral hearing was efficient and economical of court time with the result that the hearing was concluded in 11 days. We also spent three days during the hearing reading the witness statements of the witnesses whose evidence in chief was given by reference to those statements. The parties dealt fully with examples only of the transactions in question and although we fully accept that was sensible in terms of saving hearing time and costs the extra work involved in preparing this decision by having to consider the evidence relating to all the transactions has unavoidably delayed it somewhat.
  12. We think it appropriate to commend Mr Matthew Quinn, officer of the respondents and their main witness, who had prepared the case and gave his evidence in exemplary fashion and showed an accurate knowledge of the details of the evidence beyond what could normally be reasonably expected.
  13. Arguments were presented to us about a number of legal issues and we will record our holdings about those issues before turning to the facts.
  14. Legal issues

  15. In principle, a trader who is registered for VAT and who buys goods for the purposes of his business on which his supplier charges VAT is usually entitled to deduct input tax from any output tax he may have to account for and if the input tax exceeds the output tax he is entitled to a payment from HMRC. That is not in dispute but this case concerns whether the circumstances are such that the appellant has lost that entitlement.
  16. The issue arises as a consequence of a judgement of the Court of Justice of the European Communities (the ECJ) in Axel Kittel –v- The Belgian State and Belgian State –v- Recolta Recycling [2008] STC 1537. The judgement related to two unconnected joined cases. Mr Kittel was alleged to be involved in a carousel fraud and Recolta were alleged to have sold goods in Belgium without accounting for VAT, having falsely claimed the goods were exported from that country. The judgement deals in general terms with circumstances related to fraud in which credit for input tax may be refused by the authorities by way of an exception to the general entitlement to claim it. As that case is the basis for much of what follows we here set out the relevant passages from the judgement:
  17. "50      … as the referring court observed, it is settled case-law that the principle of fiscal neutrality prevents any general distinction between lawful and unlawful transactions. Consequently, the mere fact that conduct amounts to an offence does not entail exemption from tax; that exemption applies only in specific circumstances where, owing to the special characteristics of certain goods or services, any competition between a lawful economic sector and an unlawful sector is precluded (see, inter alia, Case C-158/98 Coffeeshop 'Siberië' [1999] ECR I-3971, paragraphs 14 and 21, and Case C-455/98 Salumets and Others [2000] ECR I-4993, paragraph 19). It is common ground, however, that that is not the case with either the computer components or the vehicles at issue in the main proceedings.
    51      In the light of the foregoing, it is apparent that traders who take every precaution which could reasonably be required of them to ensure that their transactions are not connected with fraud, be it the fraudulent evasion of VAT or other fraud, must be able to rely on the legality of those transactions without the risk of losing their right to deduct the input VAT (see, to that effect, Case C-384/04 Federation of Technological Industries and Others [2006] STC 1483, paragraph 33).
    52      It follows that, where a recipient of a supply of goods is a taxable person who did not and could not know that the transaction concerned was connected with a fraud committed by the seller, Article 17 of the Sixth Directive must be interpreted as meaning that it precludes a rule of national law under which the fact that the contract of sale is void, by reason of a civil law provision which renders that contract incurably void as contrary to public policy for unlawful basis of the contract attributable to the seller, causes that taxable person to lose the right to deduct the VAT he has paid. It is irrelevant in this respect whether the fact that the contract is void is due to fraudulent evasion of VAT or to other fraud.
    53      By contrast, the objective criteria which form the basis of the concepts of 'supply of goods effected by a taxable person acting as such' and 'economic activity' are not met where tax is evaded by the taxable person himself (see Case C-255/02 Halifax and Others [2006] STC 919, paragraph 59).
    54      As the Court has already observed, preventing tax evasion, avoidance and abuse is an objective recognised and encouraged by the Sixth Directive (see Joined Cases C-487/01 and C-7/02 Gemeente Leusden and Holin Groep [2004] ECR I-5337, paragraph 76). Community law cannot be relied on for abusive or fraudulent ends (see, inter alia, Case C-367/96 Kefalas and Others [1998] ECR I-2843, paragraph 20; Case C-373/97 Diamantis [2000] ECR I-1705, paragraph 33; and Case C-32/03 Fini H [2005] ECR I-1599, paragraph 32).
    55      Where the tax authorities find that the right to deduct has been exercised fraudulently, they are permitted to claim repayment of the deducted sums retroactively (see, inter alia, Case 268/83 Rompelman [1985] ECR 655, paragraph 24; Case C-110/94 INZO [1996] ECR I-857, paragraph 24; and Gabalfrisa, paragraph 46). It is a matter for the national court to refuse to allow the right to deduct where it is established, on the basis of objective evidence, that that right is being relied on for fraudulent ends (see Fini H, paragraph 34).
    56      In the same way, a taxable person who knew or should have known that, by his purchase, he was taking part in a transaction connected with fraudulent evasion of VAT must, for the purposes of the Sixth Directive, be regarded as a participant in that fraud, irrespective of whether or not he profited by the resale of the goods.
    57      That is because in such a situation the taxable person aids the perpetrators of the fraud and becomes their accomplice.
    58      In addition, such an interpretation, by making it more difficult to carry out fraudulent transactions, is apt to prevent them.
    59      Therefore, it is for the referring court to refuse entitlement to the right to deduct where it is ascertained, having regard to objective factors, that the taxable person knew or should have known that, by his purchase, he was participating in a transaction connected with fraudulent evasion of VAT, and to do so even where the transaction in question meets the objective criteria which form the basis of the concepts of 'supply of goods effected by a taxable person acting as such' and 'economic activity'.
    60      It follows from the foregoing that the answer to the questions must be that where a recipient of a supply of goods is a taxable person who did not and could not know that the transaction concerned was connected with a fraud committed by the seller, Article 17 of the Sixth Directive must be interpreted as meaning that it precludes a rule of national law under which the fact that the contract of sale is void – by reason of a civil law provision which renders that contract incurably void as contrary to public policy for unlawful basis of the contract attributable to the seller – causes that taxable person to lose the right to deduct the VAT he has paid. It is irrelevant in this respect whether the fact that the contract is void is due to fraudulent evasion of VAT or to other fraud.
    61      By contrast, where it is ascertained, having regard to objective factors, that the supply is to a taxable person who knew or should have known that, by his purchase, he was participating in a transaction connected with fraudulent evasion of VAT, it is for the national court to refuse that taxable person entitlement to the right to deduct."
  18. Mrs Hamilton raised several issues relating to Kittel.
  19. She contended that Kittel should be understood to have decided that only a trader whose immediate supplier was fraudulent can lose the right to claim input tax where he knew or should have known of the fraud (the narrow construction of Kittel). Mr Cunningham contended that a trader can lose the right to claim input tax wherever the fraud occurred in a chain of transactions as long as the transaction entered into by the trader in question was connected with that fraud and he knew or should have known it was connected with it (the wide construction).
  20. In support of her argument Mrs Hamilton pointed out that the original language of the case was French and that the vital phrase "il participait à une opération impliquée dans une fraude à la TVA" is translated as "he was participating in a transaction connected with fraud" but she contended that the French version implies a particularly close connection with the fraud. "Impliquer" can mean implicate, involve or entangle according to Cassell's Compact French English Dictionary and we do not agree that any particularly close connection is suggested by the choice of that word, so far as we understand it, and we would only be prepared to draw the conclusion contended for if a qualified interpreter had given evidence to that effect, though even then we doubt if the consequence would be to alter our understanding of Kittel.
  21. The Kittel case was decided as it was because, the ECJ having noted that a person who actually perpetrates a fraud can be deprived of the right to claim input tax (paragraphs 55 and 56), then recognised that transactions carried out to aid the perpetrators of a fraud are carried out by a person who thereby becomes the accomplice of the fraudulent persons (paragraph 57) and so should also be deprived of the right to claim. The ECJ justified that conclusion, inter alia, by the practical argument that denying the right of the accomplice to claim input tax will help to prevent fraud (paragraph 58 and see also paragraph 54).
  22. It would be contrary to the express policy behind the ECJ's conclusions in Kittel if a person who actually knew that his transaction would assist in the perpetration of a fraud was not deprived of the right to claim input tax. Mrs Hamilton did not need to go so far as to suggest that a person who actually knew of the fraud would only be deprived of his input tax if the fraud related to his direct supplier because her client denied that he did know that any of his transactions had that characteristic. But the ECJ did not draw a distinction between the consequences for a trader who knew of fraud and one who should have known of it and concluded that the same rule applies to both. No doubt the underlying policy of reducing the opportunity for fraud and combating fraud lay behind the ECJ's extension of the refusal of the right to deduct beyond the actual fraudulent persons to their knowing accomplices and to those who should have known they were becoming accomplices. The wide construction of Kittel is both inherent in the Court's conclusion and a consequence of its reasoning and we are satisfied that the Court did mean that to be the applicable rule.
  23. We would add that the words used by the ECJ indicate the wide construction. The reference in paragraph 57 to the perpetrators of the fraud are at least apt to include all those fraudulently engaged in a chain of transactions rather than just the one who evaded output tax at the start of the chain. The phrase "he was taking part in a transaction connected with fraudulent evasion of VAT" in paragraph 56 could very easily have been replaced by a phrase such as "dealt with a fraudulent person" if it was intended that only a direct connection with a fraudulent person had been intended to lead to the loss of the right to claim input tax. The un-circumscribed phrase "connected with" also indicates the wide construction and is apt to cover transactions at one or more remove from the actual fraud.
  24. Whilst it is true, as Mrs Hamilton pointed out, that the Court did refer in paragraph 56 of the judgement, to the trader becoming involved in fraud "by his purchase" and that could suggest that only a direct connection with a fraudulent person was intended to lead to the loss of the right to deduct, we do not agree that is the only interpretation that can be put on those words. In that case the Court was dealing with cases where there was such a direct connection. The Court's conclusions are expressed in more general terms in paragraphs 56 and 61 and we have no doubt that the wider construction is correct. That was also the conclusion of the tribunal in Calltell Telecom Ltd and another –v- The Commissioners (VTD 20266) at paragraph 46 and Dragon Futures Ltd –v- The Commissioners (VTD 19831) at paragraph 67.
  25. Mrs Hamilton argued that R (on the application of Just Fabulous) –v- The Commissioners [2007] EWHC 521 (Admin) should not be taken as authority for the wider construction because that was an application for judicial review on assumed facts and Burton J did not expressly make any such holding. However, he quoted paragraphs 56 and 61 of the judgement in Kittel without expressing any disagreement with them and accepted the Commissioners' argument that the Kittel principles could apply in the case of a contra-trade where there is no likelihood of the trader whose input tax is denied being directly involved, in the way Mrs Hamilton argued was necessary, in the transaction by which the fraudulent loss to the Revenue was caused. That case is therefore binding authority for the wider construction.
  26. Mrs Hamilton also argued in support of the narrow construction on the basis of the introduction in France of a law, the effect of which appears to enshrine in statutory form the narrow construction of Kittel, albeit coupled with a separate provision for joint and several liability potentially affecting a trader who fell outside the narrow reach of the first provision. We do not regard that as relevant even if that is the law adopted in France.
  27. So far as concerns the alleged contra-trading deals with which the appellant was involved, Mrs Hamilton argued that it was inherent in the words used in paragraph 60 of Kittel i.e. that the taxable person "knew or should have known that, by his purchase, he was participating in a transaction connected with fraudulent evasion of VAT" that a person who was involved only in the so-called clean chain could not be denied his input tax because the fraud and the tax loss arose from so-called dirty chain in which he did not participate. We do not agree. The ECJ did not attempt to define what it meant by "connected with" in this context and that is not in the least surprising because the connection between two transactions could take any number of different forms. However, it is entirely consistent with the words used by the ECJ that a person who assists in perpetrating a fraud, either knowingly or where he should have known, should suffer the same consequences whether his assistance is direct, indirect or event tangential. It is the assistance which leads to the loss of the right to deduct not the manner of the assistance. As we have already pointed out Burton J in Just Fabulous agreed that participation in a contra-trade could, in principle, lead to the loss of a right to input tax.
  28. What amounts to the necessary connection is to be judged largely as a matter of fact but it is clear that the connection must be one that somehow assists with the fraud. That is because the fundamental justification for the loss of the right to deduct is based on the need to combat fraud and, unless an act or omission somehow assists the fraudulent scheme, imposing adverse consequences for that act or omission would not help to combat the fraud.
  29. It is likely to be more difficult to prove a connection between a clean chain transaction and the dirty chain transaction in an alleged contra-trading case but the applicable principles are the same.
  30. An issue was raised in this appeal about what the position might be if the Commissioners have recovered the output tax evaded by a missing trader or a "hi-jacker" or anyone else in a chain of transactions leading to, for example, a "broker" which is what the appellant is alleged to be. In that case it might be argued that the broker should not be denied his input tax because that would amount to a double recovery by the Commissioners
  31. Mr Cunningham said that in such a case the Commissioners would set matters right by not denying the input tax. We note that the situation would be difficult to reverse if, for example, the person denied the input tax has been made bankrupt or if it is a company that has been put into liquidation as a result of the input tax being denied, and then later the output tax is recovered from the defaulting trader. We do not agree that the law can be interpreted in light of concessions made by one of the parties to litigation to the effect that they will put things right and so, if there is anything in Mrs Hamilton's point, Mr Cunningham's response is not an answer to it.
  32. There must always be a possibility, however remote, that the defaulting trader will be traced and the lost output tax will be recovered and so, if Mrs Hamilton is right in principle, her argument goes beyond what she submitted and it would not even be enough to deny input tax recovery if the Commissioners show that the appellant knew about the precise output tax fraud because there would always be the possibility that the money might, at least in theory, still be recovered from the defaulting trader.
  33. For our part we see no objection, even in principle, to the input tax recovery being denied where the output tax may still be or even has been collected belatedly from the defaulting trader. The logic of the Kittel case is that the right to recover input tax is lost by the trader concerned by his becoming involved in the fraud and the policy underlying the rule is that the input tax recovery should be refused the better to combat fraud. That reasoning is not affected by how successful the fraud is in the long run or how many participants have avoided detection or recovery action.
  34. However, Burton J appears to have agreed with the Commissioners in Just Fabulous that a practical solution would be found by the Commissioners or the tribunal if the double recovery situation arose and, given that there is no indication that any double recovery has or is likely to occur in this case, we propose to say no more about it except to say that we do not agree with Mrs Hamilton that the theoretical possibility of double recovery says anything about what the appellant must be shown to have known or what it should have known.
  35. In both Calltell at paragraph 51 and Honeyfone –v- The Commissioners (VTD 20667) at paragraph 56 the Tribunal held that the precise identity of the defaulting trader need not be known to the broker or other person concerned in the chain of transactions for their input tax recovery to be denied and we hold that it is sufficient for that person to be shown to have known or to be a person who should have known that by his transaction he was entering into a transaction connected with fraud, whether or not he knew the precise details of that fraud or the identity of those taking part in it.
  36. A transaction can be connected with a fraud that has not yet fully materialised. An obvious example is a case where the defaulting trader (perhaps at that stage it is more accurate to say a trader who intends to default) is part way through a three month tax period when goods are bought and the goods are exported by a broker during a one month tax period that ends before that three month period ends. In such a case the broker does not need to be shown to have known or that he should have known about the tax periods in question and it is enough that the evidence shows that he knew or should have known that a fraud either had occurred or would occur and that his transaction was or would become connected to it. However, we agree with the Tribunal's comments in Calltell that it would not be enough if the fraud was only decided upon by the defaulting trader after the appellant's transaction had occurred.
  37. The parties made submissions about what "knew or should have known" means in this context. Obviously what a person knows is purely a question of fact except that we emphasise that a person can only know something if it is a fact. A person can believe something and may do so sincerely however far fetched it is and without having any objective evidence on which to justify the belief. A person can suspect something to be the case whether it is the case or not, though then it is appropriate to say he suspects it only if there is some basis for the suspicion. A person can reasonably suspect something if there is enough evidence to make it reasonable to arrive at the suspicion.
  38. We make that point because we consider that both parties placed an inappropriate emphasis, in different ways, on Public Notice 726 which the Commissioners issued in August 2003 following the introduction of the joint and several liability provisions in section 77A of the VAT Act 1994 which had effect from 10 April 2003.
  39. Those provisions operate where a person who knew or had reasonable grounds to suspect that VAT would go unpaid in respect of a transaction is given a notice of joint and several liability. How far that means that the person concerned had to be intellectually capable of forming a view about the tax going unpaid and how far he could be said to have the grounds to suspect even if they had not become part of his conscious thinking may yet remain to be decided (in other words whether the test is subjective or objective).
  40. What is clear is that having reasonable grounds to suspect something is not the same as being in a position where it should be known though there may well be a good deal of overlap between the two ideas.
  41. Mr Cunningham submitted that a trader is under a positive duty to take precautions to ensure that he does not become involved in fraudulent transactions. He cited paragraph 65 of the judgement of the ECJ in the case of R (on the application of Teleos and others) –v- The Commissioners Case C-409/04 in which the Court said that it would not be contrary to Community law to require the supplier of goods to take every step (such as those mentioned in Public Notice 726) which could reasonably be required of him to satisfy himself that the transaction he is effecting does not result in his participation in tax evasion. Mr Cunningham also referred in this context to the passages in Kittel (paragraph 51) and in The Commissioners –v- Federation of Technological Industries Case C-384/04 (paragraph 33) which say that a person who takes every precaution open to him should not be deprived of his right to deduct input tax.
  42. We do not agree that those passages amount to imposing a positive duty on anyone to take precautions to ensure (which might as well mean to guarantee) that he does not become involved in fraudulent transactions. No such law has been enacted as that envisaged in Teleos. The Court may have been referring to the "should have known" concept in Kittel, but if that is so then it is not necessary to enlarge the concept of a person who has taken such steps as might reasonably be required of him by interpreting that phrase to mean that a person is under a positive duty to take all steps to ensure or guarantee he does not become involved in fraud. What could reasonably be required of a person may fall short of steps that would ensure or guarantee that that person does not become involved in fraud and the same is true of the concept of what he should have known. In fact we consider the "every step that could reasonably be required" and the "should have known" concepts to be closely related but neither amount to taking all steps that would ensure or guarantee that no fraud was involved.
  43. Mr Cunningham's submission was that trade in mobile phones outside the manufacturer and the network of dealers appointed by the manufacturers had become so corrupted by fraud by 2004 that the only safe course would be to refuse to deal in them at all. Although there is a good deal of reason to think that comes close to the truth in practice, we do not agree that it had become a legal rule that anyone who dealt in mobile phones was required to ensure that there was no fraud in the chain of transactions.
  44. It was put to Mr Andreou that as he could not verify the integrity of every transaction in the chain of transactions leading to his sale to an overseas buyer and as he knew that there was a good deal of fraud in his sector of the economy he should not have dealt with the phones in question at all. We do not agree that the law places such a requirement on a trader. If he does everything that is open to him and proportionate to the risks involved, paragraph 51 of the Kittel judgement makes it clear that a trader is still entitled to input tax recovery. However, if as a matter of fact, he either knew or ought to have known that his transaction was connected with fraud despite taking steps that might otherwise have satisfied him that the transaction was legitimate he will lose that right.
  45. On the other hand we do not agree with the appellant that Notice 726 either by giving examples of what might be appropriate by way of due diligence enquiries when trading with others or by failing to give an exhaustive or definitive list of what is appropriate defines whether the appellant should have known about fraud in the transactions in question. In judging whether the appellant should have known it was involving itself in transactions connected with fraud it is relevant to take into account what the appellant did by way of due diligence, what it did not do and what else it could have done; but compliance or non-compliance with the steps suggested in Notice 726 is only part of the overall picture. The appellant did know about that Notice and it forms part of the overall picture in this case. We do not agree that it should be taken to form the whole of the picture.
  46. Submissions were made about the burden of proof. We preface our consideration of those submissions by remarking that it is only in rare cases that issues of the burden of proof decide any case where, as here, both parties give evidence.
  47. The persuasive or legal burden of proof rests on one or other party in respect of what that party has to prove in order to succeed in making out its case. It is often said that in VAT cases the Commissioners do not have to prove anything and that the legal burden of proof rests on the taxpayer. But that says nothing about what the taxpayer has to prove. The legal burden is satisfied by producing evidence of a prima facie case i.e. credible evidence on which the Tribunal could find in favour of the taxpayer if it is not refuted.
  48. The appellant's case is fundamentally that it is entitled to claim input tax in the sum of £5,909,067.50 and all it has to do to make out that claim, in principle, is to produce evidence that it was registered for VAT and bought goods in the course of a business on which VAT of that amount was charged and that it used those goods to make taxable supplies by exporting the goods so that they were zero rated. None of those facts are in dispute and the appellant would easily satisfy the legal burden of proof in respect of them.
  49. The respondents' case is that despite that, the right to claim input tax was lost because the transactions were connected with fraud and the appellant should have known they were. The legal burden of proving those facts, again by way of a prima facie case, rests on the respondents.
  50. Once the appellant has produced its prima facie case about the matters for which the legal burden rests on it therefore the respondents have to respond and produce their prima facie case about the matters for which the legal burden rests on them. The legal burden of proof does not alter throughout the proceedings. However, the evidential burden shifts. Once a party has produced enough evidence to satisfy the legal burden the other party is obliged, not because of any rule of law but in order to succeed in the appeal, to produce evidence to refute the other party's case so far as possible.
  51. The legal burden of proving that there was fraud or that the appellant should have known of it does not lie on the respondents because of a concession on their part, it is laid upon them because, if they do not produce evidence of that, the appellant will succeed in the appeal by proving the facts necessary to establish its right to input tax in principle.
  52. The relevant degree of proof is the civil standard with the usual caveat about the need for cogent evidence where an allegation of fraud is being made.
  53. Mr Cunningham submitted that the Commissioners' evidence showed that 95% of transactions of the sort in issue in the sector of the mobile phone industry in which the transactions in this appeal occurred were connected with fraud and so a prima facie case for fraud was thereby made out in respect of each transaction. The evidence for the 95% was apparently only statements made to a Parliamentary Committee by Customs witnesses and the anecdotal evidence of Customs Officers who had been assigned to cases in which suspicion had already arisen. No precise definition was given about what types of transaction that 95% related to. Of course, even if that statistical approach were to form a basis for decision in this appeal it would be necessary to identify either one or two of the 30 transactions that fell within the 5% of transactions apparently not affected by fraud. We have no doubt that we have to consider each transaction separately and by examining the evidence about each one and that the appeal cannot be decided upon statistical grounds.
  54. With respect to the Tribunal hearing the case of Mobilx Limited –v- The Commissioners (VTD 20867) we do not agree with the following formulation of the issue, in a passage at paragraph 108 of the Decision which Mr Cunningham relied upon:
  55. "… The essential question is a simple one: was it, or should it have been apparent to Mobilx, by the beginning of April 2006, that if it continued to deal with CPU's as it had been doing for the last two years, its transactions were more likely than not to be connected with fraud"?

  56. That formulation risks eliding two issues. In respect of each transaction the Tribunal must decide if the transaction was (i.e. was in fact) connected with fraud and secondly that the appellant knew that to be the case or should have known it. That conclusion must be based on evidence and the degree of proof is the balance of probabilities (with the caveat referred to) but it should not be forgotten that what is in issue is whether the transaction was actually connected with fraud and whether the appellant knew or should have known that to be so; not whether there was a probability that a transaction was connected with fraud. The Tribunal must reach a positive conclusion one way or the other. Having said that, the preponderance of fraud is certainly relevant to what the appellant should have known and the precautions it should have taken and we would add that we have no reason to think the Mobilx case was wrongly decided.
  57. The evidence and the facts – connection to fraud

  58. The questions of fact that we have to decide are the following. Has there been a tax loss to the authorities? Was it caused by fraud? If it has occurred the loss in question will be the non-payment of output tax by the defaulting trader but the allegation is that it is financed, in the sense we have described above, by the input tax claimed by the broker. Is that sufficient to form a connection of the sort referred to in Kittel? Did the appellant know or should it have known that the transaction in question, being either its purchase of the goods from a buffer trader or the export of the goods by way of a zero rated supply (or both), was connected with fraud?
  59. In the cases where the appellant is alleged to have been involved in contra trade deals the questions are essentially the same except that the alleged fraud is in a separate chain of transactions and so the additional question arises as to whether the two chains are connected with each other in such a way that the chain of transactions in which the appellant was involved assisted the fraud in the other chain in some way.
  60. We will deal first with the chains which are not alleged to be contra-trading chains.
  61. There are 27 transactions of this type and they are shown on the schedule annexed to this decision as numbers A1 to A15 and M1 to M15 with the exception of A10 to A12 which are the alleged contra-trading transactions. A refers to April 2006 and M refers to May 2006.
  62. All 27 transactions share the following characteristics. In each case there are at least two alleged buffer transactions between the defaulting trader or traders and Megtian. The goods in question were all mobile phones. In each case phones originating with a particular defaulting trader were sold by Megtian to a particular overseas customer and no consignment bought by Megtian was split between two or more customers, though some of the transactions are made up of two separate sales to the same customer on the same date where Megtian had acquired two consignments on that date. All the transactions in each chain took place on a single day though that does not mean that the negotiations between the parties did not take place over a period of time before the day the deals were transacted. Although Megtian invoiced its customers on the same day as all the other transactions in the chain occurred the actual export may have been shortly afterwards. The mark up achieved by Megtian always exceeded that achieved by any of the alleged buffer traders individually and it always exceeded that achieved by the alleged buffer traders taken collectively. Megtian always made a significant gross profit on each transaction. Each of the defaulting traders owes the respondents a huge sum of money, at least according to unpaid assessed amounts, ranging between approximately £3 million and £77 million. Those sums do not relate only to transactions where Megtian were the exporter.
  63. The following additional characteristics apply to some but not all of the chains of transactions.
  64. Although transactions A1, A4, A6 and A9 involve two or more parcels of phones as far as the sale by the defaulting trader to the first buffer trader is concerned the two parcels were sold to the same customer by Megtian. In the case of A1, A4 and A6 the separate parcels of phones passed through the hands of different buffers before reaching Megtian. Only transaction A7 involves two defaulting traders at the start of the chain and although Megtian sold the two parcels of phones by separate invoices they were sold to the same customer on the same day.
  65. The mark up achieved by the buffer traders collectively in all the chains starting with the defaulting trader CT Co was exactly £1.20 per phone regardless of the type or number of phones or the identity of the buffer traders involved. All the chains starting with CT Co had three buffer traders. The collective mark up per phone for all the buffer traders involved in all the deals ranged from 90p to £2.25 in the chains of transactions in question. In both the transactions in which Oracle featured as a defaulting trader the collective mark up for the buffers was £2.25. The variation in collective mark ups achieved by buffers was also within a narrower range than 90p to £2.25 if analysed by individual defaulters. For example all buffers involved in chains starting with Stockmart achieved collective mark ups between £1.10 and £1.40. Every Stockmart transaction involved three buffers and in total thirteen different buffers were involved.
  66. There was an apparent degree of consistency in the ultimate overseas destination of the phones and the original defaulting trader. All phones starting with Worldwide went to the United Arab Emirates as did all Stockmart phones. All 3D phones went to France as did the two lots of Oracle phones. All except one consignment of CT Co phones went to the United Arab Emirates.
  67. Full documentation for the information on the Schedule was produced by HMRC and it was not challenged by the appellant. Of course, the appellant would not necessarily be in a position to challenge much of it because its dealings were with the last buffer trader and it is the appellant's case that it was not in a position to make itself aware of the chains of transactions in the way that HMRC have been able to do.
  68. The evidence we have summarised above clearly indicates that the chains of transactions in question are contrived as the respondents allege. No normal pattern of trade could result in such coincidences. Also the volume of sales leading quickly from an importation to an exportation of consumer goods, as opposed perhaps to commodities in the usual sense e.g. crude oil, seems unlikely to be a result of normal trading patterns. We will deal later with the evidence that shows that at least some of these phones were sold into markets where they would not have been readily saleable but when that fact is added to the equation these transactions become even less like normal trading.
  69. That the transactions have those characteristics does not in itself prove that the appellant knew or even that it ought to have known that its transactions were connected to fraud but it is clear evidence that the chains did involve fraud.
  70. Indeed Mrs Hamilton accepted in respect of the transactions beginning with CT Co and Stockmart that there were tax losses in those chains and that they were caused by fraud.
  71. We will next consider the evidence alleged to prove that the other alleged defaulters were fraudulent and that they caused tax losses in the relevant chains.
  72. The evidence that the two Oracle chains (A8 and part of A7) involved a tax loss is irrefutable. Following an officer's visit on 18 May 2006 the company was assessed for output tax of £21,833,536 in its final period of trading which was from 1 February 2006 to 6 April 2006 in addition to an earlier assessment for £1,185,533 and the transactions in question related to mobile phones or CPU's. The sales on which those assessments are based included the phones in transactions A8 and part of A7 in this case. The assessments have not been paid.
  73. Oracle's sales in the period from 1 February to 6 April represent approximately £154,000,000 of tax inclusive turnover which had increased from less than £7,088,049 in the previous tax period and from £337,073 in the period before that. (The £7million figure had included some of the £154 million until the Commissioners gave notice of a change to the accounting periods which had the effect that some turnover was transferred to the new period). Oracle had previously dealt in various soft furnishings and alcoholic drinks but had no history of dealing in phones and apart from one unsuccessful attempt it had no previous history in computer products.
  74. The rapid rise in turnover to a level which is itself barely conceivable is indicative of, at the very least, the possibility of fraud. In addition, a company that had received a general enquiry from Oracle about supplying goods had received a request from Oracle to disclose its VAT registration number so that it could be added to an invoice from that company to Oracle but the company had not supplied goods to Oracle and when the supposed invoice was shown to the company it proved to be a forgery. Oracle neither appealed against the assessments nor against its deregistration with effect from 6 April 2006. The fact that the two Oracle chains, which both involved four buffers, gave rise to identical levels of mark up for those buffers collectively even though one of the buffers was different in each chain and even though the quantity of phones was different adds to the evidence that the default that arose was the result of fraud, not simply unfortunate insolvency.
  75. We find that Oracle or someone manipulating it was fraudulent and that the tax loss was caused by that fraud.
  76. C&B Trading had traded as a car valet service provider until it wrote to the Commissioners to tell them that it would be trading in chemicals and car components but without mentioning phones. The letter was undated but other evidence relating to phone calls to the Commissioners suggest that the letter must have been sent between 7 March 2006 and 20 March 2006. An officer attempted to visit the company on 26 April 2006 but was unable to find the premises and, when he telephoned to seek directions, he was told that it was not possible for him to visit on that day as they were busy. Subsequent attempts by an officer to visit were thwarted by the company rearranging a visit and then simply not replying when he rang the intercom on arriving for the visit. The company was assessed for a total of £84,368,780 representing tax inclusive turnover of approximately £566million in the company's last tax period (having been negligible in the immediately preceding periods) and it was deregistered but did not appeal against the assessments or the deregistration. The assessments have not been paid. The assessment included the transactions which were the sales to the first buffer in respect of the part of Megtian's deal A7 that did not involve Oracle.
  77. We find that C&B's trading was fraudulent and caused a tax loss. The evidence of deliberately avoiding contact with HMRC, the disappearance of the owners and managers of the company and the extraordinary increase in trade are all evidence of fraud.
  78. Worldwide, the alleged defaulting trader in respect of deals A13, A14 and A15 had traded in men's clothing but during 2005 various returns and assessments were returned marked gone away. On 25 August 2005 the company notified HMRC of a new address in Bolton but correspondence to that address was also returned and then on 15 March 2006, by which time the company had been de-registered for VAT, it notified an address in Blackburn which had been its address before the Bolton address. In April 2006 an officer tried to visit the directors at the home addresses they had given but found that both the addresses were those of demolished properties. Using information from input tax claims by its known customers the Commissioners assessed Worldwide for £36,591,462.49 for its trade for the period following the February 2006 return which was the last one it filed (representing turnover of approximately £241million). The assessed sum was markedly more than the £131,000, £23,000 and nil sums declared respectively in the three periods ending with the February 2006 period. The company has not responded to or paid the £36 million assessment or resisted its liquidation. The relevant transactions between Worldwide and the first buffer in the supply chain to Megtian are included in the assessment.
  79. The evidence that the company had deliberately evaded contact with HMRC and that its directors had given false addresses as well as the incredible increase in trade leads us to conclude that its trading in the period covering April 2006 was fraudulent and caused a tax loss.
  80. Finally, amongst the non-contra chains leading to Megtian i.e. deals M12, M13, M14 and M15, the alleged defaulter is 3D Animations.
  81. 3D Animations registered for VAT on 3 May 2006 stating that it intended to trade in "design multimedia and animation graphics" and described its business category as "speciality design activities". In fact within about three weeks from that date it had started to trade on an enormous scale in mobile phones. It was de-registered with effect from 7 June 2006 and has received assessments totalling £124,259,499 (approximately £820million turnover) based on its customers' records and those assessments are unpaid. An officer visited the company's stated principal place of business on 7 June 2006 and it turned out to be a residential address which had been turned into offices. The only evidence that 3D Animations was based there was a handwritten note on a piece of paper taped to the door. The officer was unable to gain access or any reply at the premises. The company has never made a VAT return. The 3D Animation deals which lead to an export by Megtian all involved several buffers and were all dated on the same date and the output tax payable by 3D Animations is included in that company's unpaid assessments. In each case Megtian's overseas customer was in France.
  82. We have no hesitation in concluding that the sales by 3D Animations were fraudulent transactions. The company had misrepresented the nature of its trade and its office was clearly not that of a company with a turnover that could have given rise to such assessments and there is evidence that it deliberately avoided contact with the officer who visited.
  83. We have therefore concluded, either on the evidence or by Mrs Hamilton's concession, that all of Megtian's non-contra deals were in respect of goods that were originally sold by fraudulent defaulting traders to intermediaries before they were sold to Megtian and that a tax loss has occurred in respect of each one of them.
  84. We turn now to consider whether the contra-trading transactions were "connected with" fraudulent transactions in the relevant sense.
  85. These transactions are alleged to have the following characteristics. Trader A imported goods and sold them to B who sold them to Megtian who exported them. A accounted for output tax on the sale to B who accounted for output tax on the sale to Megtian. Megtian exported the goods and claimed back the input tax from its purchase from B but did not have to account to HMRC for any output tax because the sale was by way of an export. Such a transaction is called a clean chain in the jargon adopted in cases where MTIC fraud is alleged.
  86. However, in the same tax period A was involved in other transactions in which it had exported goods and claimed input tax which off-set in whole or in part the output tax due on the Megtian transactions and those exports by A were connected to fraud because in those chains of transactions there was a fraudulent transaction of much the same sort as those already described in the preceding paragraphs. The allegation is that the apparently legitimate transactions were undertaken to disguise A's involvement in the fraudulent chains because it would either not have to claim an actual repayment from HMRC and risk drawing particular attention to itself or if it did make such a claim it would have apparently legitimate transactions to draw to the attention of the authorities to attempt to demonstrate its bona fides. The chains containing fraudulent transactions are called "dirty chain" transactions in the jargon adopted in MTIC fraud cases.
  87. We have already held that the connection with fraud that Kittel decided would entitle the authorities to refuse to repay input tax may be indirect but there is of course an issue of fact as to whether the connection exists and whether Megtian knew or should have known that the transactions it did enter into were connected to fraud.
  88. There are three transactions dated 12 April 2006 in which Megtian sold mobile phones to an overseas buyer in each case being the same customer in Cyprus. These transactions are clean chain transactions but HMRC allege that the original importer in these clean chain transactions had acted as an exporter in other transactions in the same tax period where there was a fraudulent importer and which were therefore dirty chain transactions.
  89. In respect of Megtian's deal number A11 a company called @tomic had acquired 5000 Nokia 9500 phones from a German supplier and had sold them to a company called Shelford Trading which sold them to Megtian and Megtian sold them to Lavina Trading Ltd in Cyprus. All those transactions were dated 12 April.
  90. During its April 2006 tax period @tomic had acted as "broker" i.e. had exported phones on at least 22 occasions and HMRC allege that on 20 of those occasions the original importer had defaulted. There were three importers involved in the 20 chains. On behalf of the appellants Mrs Hamilton conceded that the evidence was sufficient to show that two of those importers had acted dishonestly and that concession related to 12 of the 20 chains alleged by HMRC to have included fraudulent defaults. We do not think it necessary to deal with the evidence relating to the remaining 8 chains as a connection between the 12 dirty chains and Megtian's single clean chain would be sufficient to deny Megtian's input tax claim provided Megtian knew or ought to have known that its own transaction was connected with fraud, whether that connection relates to all 20 chains or only to 12 of them. We do record that HMRC alleges that all 20 were chains involving dishonest transactions.
  91. @tomic had been involved in transactions connected to fraud on at least 12 occasions in that month. It does not follow that @tomic knew or ought to have known that its transaction were connected with fraud but we find that on the balance of probabilities it did know.
  92. The evidence includes the following. @tomic's turnover rose from nil in the year ending May 2003 to £234million in the year ending May 2006 and that was nearly all from sales of mobile phones. The original business plan prepared by the directors when they acquired the shares from another owner and which was shown to HMRC in March 2004, when the company applied to make monthly VAT returns, claimed that the business would be mainly the importation and sale of pizza ovens though there is no evidence it ever traded in that way. The company received 49 warnings that others with whom it was dealing were suspected of dealing in MTIC fraud. @tomic has been unable to produce evidence, when requested to do so by HMRC, that it insured the goods it was dealing with despite their enormous value. It did produce some evidence for insurance of the April 2006 goods but on examination that turned out to have been given retrospectively and only in response to the request from HMRC. @tomic's solicitors attempted to persuade HMRC not to mark the packing cases dealt with by @tomic (not the cases in which the phones themselves were shipped) with marks to show that they had been inspected at import or export and claimed that these marks devalued the goods. On other occasions such marks had been removed from the cases. Such marks could not reasonably be expected to affect the value of the goods though they would make detection of fraud by HMRC easier.
  93. On examining all the evidence including that mentioned above we have concluded that @tomic did know that at least the 12 transactions mentioned were connected with fraud.
  94. In respect to Megtian's deal A11 @tomic was, as already mentioned, the importer of the goods and @tomic's output tax on the sale to Shelford Trading, which sold the goods to Megtian, was accounted for. That output tax was £262,500 and forms part of the output tax declared by @tomic in the sum of £1,085,875 for April 2006. @tomic did not have to pay that output tax to HMRC because its input tax for that period was £3,650,625 much of it made up of claims in respect of the 20 transactions in respect of which we have found that at least 12 involved dishonest fraudulent trading of which @tomic knew and the amount involved in those 12 chains far exceeds the output tax accounted for in respect of the single clean chain transaction in which Megtian was involved. It follows that the apparently honest trading, the clean chain, was connected to the dishonest chains albeit indirectly because it served to provide @tomic with an apparently honest deal to suggest that it did not know about the dishonest ones and also to disguise the dishonesty itself.
  95. The situation concerning Megtian's deals A10 and A12, which concern Primeline, is similar to that involving @tomic. Primeline was the importer in respect of two transactions where Megtian was the exporter and in which Primeline had accounted for output tax on its sale to the single other UK trader in the chain, namely Shelford Trading as in the @tomic chain.
  96. Mrs Hamilton admitted on behalf of the appellant that the evidence was sufficient to prove that one importer, which was involved in nine chains in April 2006 in which Primeline acted as exporter, was a dishonest trader which had caused a tax loss. Primeline had also been the importer in respect of seven chains in April 2006 including the two in which Megtian were the ultimate exporter (A10 and A11).
  97. In total, Primeline was the exporter in 25 chains in April 2006 and although Mrs Hamilton's concession related to only 9 of those chains we see no need to examine the evidence relating to the others for the same reason as we have given in respect of the @tomic transaction.
  98. The director of Primeline had received numerous warnings about the prevalence of fraud in the trade he was dealing in. The company obtained a credit check about a trader in February 2006 which recommended that it should only give that trader £4,000 credit but then traded to the tune of over £12million in the month of April 2006 and in March 2006 it received a credit check on another trader recommending a £300 limit but traded with it to the tune of £5million in April 2006. Primeline's turnover grew from £13,851 in the year ending February 2005 to £40,773,840 in the year ending February 2006 and to £152,282,500 in the three months ending May 2006. There is evidence that the director of Primeline asserted that he had made enquiries of HMRC about another trader with which he wished to deal and had been given clearance when that was not the case. The company had falsely represented the nature of its intended trade to HMRC when it applied for registration. The director had also failed to register for VAT as a sole proprietor despite having a liability to do so.
  99. That and the other evidence we have reviewed leads us to find that Primeline was dishonest in its dealings in the relevant chains and that it had used the two transactions with which Megtian was concerned to facilitate fraud by means of contra trading.
  100. It therefore follows that we have found that all of the 30 transactions in which Megtian was involved were connected with fraud in the sense required by Kittel and so we move on to consider whether Megtian knew or should have known that the transactions were connected to fraud.
  101. Evidence and facts – did or should Megtian have known?

  102. Mr Andreou's evidence was that he traded in phones in what he called back to back deals on "ship on hold" terms. That means that each transaction was arranged so that Megtian only bought phones to satisfy an order from a buyer, in this case always an overseas buyer, when the buyer had paid for the phones. The transactions took place in effect at the same instant in time. Naturally such trading meant that the deals, both the purchase and the sale, had to be agreed in advance and then were put into effect when the payment from the buyer was certified as having been received into Megtian's bank account. In fact the person selling to Megtian was always a buffer trader and that trader bought from another buffer and so on up the chain of transactions to the defaulting trader in each case. There were always at least one more buffer and the defaulting trader in each chain as well as Megtian's supplier and customer and all the transactions took place at the same instant.
  103. HMRC had always understood the appellant's case to be that the transactions, no matter how many traders were involved in the chain, had always been arranged and carried out in a single day. That understanding was based on the fact that the relevant paperwork appeared to suggest it and Mr Andreou had said in his witness statement that he and Mr Nicolau developed a pattern of trading by which at the commencement of "each trading day" they sent out faxes to potential suppliers seeking information about stock availability and then mail shots to potential customers to show what stock Megtian had available "for that day" (which was the stock they had been offered, not stock they yet actually owned) or they received requests for stock from customers and offered them stock they had been told was available. Mr Andreou had also said that the deals had to be arranged quickly as prices would fluctuate within a single day.
  104. Ship on hold terms meant that ownership of the goods would not be transferred until the purchaser had been satisfied that the goods were as described when the agreement to buy was made and that depended on the purchaser receiving an inspection report. At that point and when the purchaser had made the payment the transactions would all take place. On at least three occasions the evidence showed that the goods had been despatched (in one case to Dubai and in two cases to France) before the purchaser had received that assurance and paid for the stock. Mr Andreou explained that by saying that, as the goods were still on ship on hold terms, he could have called them back or sold them to someone else in those countries if the purchaser rejected them. The CMR documents for the two consignments to France did say the goods were "on hold".
  105. The goods themselves stayed at the premises of the freight forwarder where they were kept from the time of importation to the time of exportation and the freight forwarder acted for all parties by holding the goods though we were not told much about the terms on which they acted.
  106. Mr Andreou's evidence was that the deals sometimes took a couple of days to negotiate and while we note that his witness statement suggested otherwise and that that contradicts what he said about the price fluctuating within a single day, so that the deals had to be concluded as quickly as possible, we do not think it particularly relevant to the issues in this case whether the deals were concluded in one day or in two or three days.
  107. It is relevant to have regard to the fact that in all thirty deals Megtian made a substantial profit in deals in which it had no risk of a loss except the relatively minor overheads involved if a deal fell through after some expenses had been incurred. Given Mr Andreou's assertion that Megtian was trading in a grey market where prices fluctuate within a single day the company would have been extremely fortunate to have made the profit in excess of £6,000,000 that he admitted it had made in the ten months to May 2006 or the nearly £2,000,000 made in April and May 2006, in which the transactions in question in this case occurred.
  108. Mr Andreou put this down to hard work over a period of time in which he had built up the connections that enabled him to find buyers and sellers. We regard that evidence as implausible. The sellers were the various buffer traders who, the evidence shows, were in effect men of straw and who were themselves evidently being directed by third parties as is clear from the minimal and in some cases consistent mark ups they always achieved in round figures regardless of the number and make of the phones in question. The buyers were overseas traders who, the evidence shows, were willing to buy large quantities of phones which were at least in some cases not apparently readily saleable in the markets in which they operate, as was demonstrated by their being accompanied by warranties not applicable in those markets and handbooks in incorrect languages. The evidence of Mr Gary Taylor, the respondents' expert witness called to deal with the nature of the grey market in phones, satisfied us that the phones in question were being traded by Megtian with higher profit margins than would normally be achievable in the market and his evidence also satisfied us that at least in some cases the warranties would have been invalid in the markets where the buyers operated.
  109. We attach considerable significance to the evidence about inspection of the phones while in the custody of the freight forwarders.
  110. It is an essential part of Megtian's case that the phones must have been inspected at the freight forwarder's premises before any transaction was put into effect because Megtian would not have known that the stock being offered to it met its requirements and indeed Megtian's customer would also presumably have needed to inspect before making payment. In some cases at least, goods left the freight forwarder before Megtian had received the inspection report though Mr Andreou did say that he would have received an oral report.
  111. The reports were prepared by a company called A1 Inspection Limited. That company was managed by a Mr Hersh Patel who was called as a witness by the appellant but when he gave evidence he refused to answer most of the questions put to him on the grounds that the answers might incriminate him. Mr Andreou had refused to speculate about how long it would have taken A1's staff to inspect each phone but he did assert that he had instructed the company to carry out a "100% open box inspection" which meant that every phone and the contents of the box such as the warranty and instruction booklet should have been physically inspected. Given that the number of phones in question could be as many as 28500 in one day and given that we did not have evidence about how long each inspection took we conclude that although we cannot say how long it would have taken we are quite satisfied that such a full inspection could not have been carried out in the time available and indeed that that should have been apparent to Mr Andreou. That is without taking into account that the buyer would also need to carry out an inspection as indeed should all the buffer traders if they were bona fide though it is true that each of them might have been satisfied with samples rather than 100% inspections.
  112. Mr Andreou stated several times in his evidence that he "paid for the report" from A1 but of course what he was paying for was supposed to be the inspection and we conclude that his answers reveal the truth that he paid for the report purely as window dressing, as Mr Cunningham put it.
  113. Asking ourselves why Mr Andreou would have ordered a 100% inspection and then ignored the fact that it could not have been properly carried out we have concluded that the only purpose was to be in a position to say that he held evidence in the form of inspection reports that formed part of the due diligence procedures in order falsely to satisfy HMRC that the transactions were genuine and untainted by fraud.
  114. We have also concluded that the records kept by Megtian showing the due diligence checks carried out in respect of each deal were deficient in a number of respects. Just by way of examples we note that the Notice of Intention to Trade in respect of deal A7 was not sent to HMRC until after the goods had left the country and there is no evidence that a check on the counterparty's VAT registration had been carried out with HMRC though Mr Andreou said that Megtian might have carried out a check on the Europa website in stead though, if it did, it did not record that fact. Other warnings about counterparties were ignored and some entries were made on the due diligence records kept by the appellant when no such diligence enquiry had been completed. The due diligence was partly at least a sham intended to persuade HMRC falsely that Megtian had carried out proper enquiries when it had not.
  115. Mr Andreou's evidence was that he neither knew nor could know the details of the supply chains that HMRC have traced in terms of the other participants. However, at one point in his evidence he said "At the time of doing the transactions, as far as I was concerned, everyone was still in the supply chain". He tried to explain that by saying that he meant nothing by saying everyone in that context but it was apparent to us that he had realised that he had made a mistake by speaking as he did and that he had given it away that he did know there were chains involving more than just his supplier and purchaser. We should add that although we assume Mr Andreou is of Greek or Cypriot heritage and we do not know if he was raised in this country from birth he does speak English as well as a person brought up speaking it as a first language and that he would be fully familiar with the exact meaning of the phrase "everyone was still in the supply chain" and that that phrase is likely to refer to more than just the two counterparties who must of necessity have been involved and would more naturally be referred to as "both" still being in the chain or some such phrase.
  116. We place significance upon the fact that in some cases Megtian dealt in very substantial sums with companies about which it had received reports from a company called Veracis with various warnings about those companies or statements that the companies were only good for much smaller amounts. It is true that the nature of back to back trading did mean there was little risk to Megtian in trading with those companies despite the credit warning but nonetheless the warnings must have raised alarm bells about the bona fides of the transactions.
  117. HMRC accepted there is no direct evidence that Megtian knew or must have known that the 30 transactions in question were connected with fraud. But we have no hesitation in finding that the cumulative effect of the evidence we have heard over eleven days of hearing and extensive examination of the documents placed before us in 41 lever arch files including the evidence we have summarised above but not limited to that evidence; has satisfied us that Megtian knew that the transactions were connected with fraud. Mr Andreou is a highly intelligent man and a very experienced businessman and so the case for a conclusion that Megtian ought to have known of those facts is strong. It is not therefore necessary for us to determine whether the test of what someone ought to have known is subjective or objective and whether it should take account of the person's experience and ability. Clearly if the correct test is that it should be viewed objectively the answer would be the same in this case.
  118. The appeal is therefore dismissed and Megtian is not entitled to the input tax in question.
  119. If there is to be an application for costs it should be made within 35 days of the release of this decision. That is to say an application for a direction in principle should be made by then we are not directing that a fully quantified application should be made by then.
  120. Richard Barlow
    CHAIRMAN
    Release date: 11 December 2008

    LON/2007/0980

    MEGTIAN DEAL CHAINS                                  
            Defaulter   Buffer   Buffer   Buffer   Megtian's Megtian's Megtian's Megtian's Megtian's Megtian's
    Deal Date Phones Qty   Net 1 Net 2 Net 3 Net   Net       Mark
    No 2006       Sales   Sales   Sales   Sales Immediate Sales Export Export Gross Up
              Value   Value   Value   Value Supplier Value Customer Price Profit  
      April       £'000   £'000   £'000   £'000   £'000   £'000 £'000 %
    1 6 Nokia 2000 C T Co 477.60 CityPhones 478.00 MobileMag 479.00     TeamMob 480.00 Ankite Ent 509.0 29.0 6.04
        9300 5000 C T Co 1194.00 Stylez 1195.00 TalkingFonez 1197.50     TeamMob 1200.00   1272.5 72.5 6.04
                                       
    2 6 D 600 1000 C T Co 158.80 Cityphones 159.00 MobileOne 159.50     TeamMob 160.00 Sahil Int 170.0 10.0 6.25
                                       
    3 6 Nokia 500 C T Co 213.40 Cityphones 213.50 TalkingFonez 213.75     TeamMob 214.00 Network 226.5 12.5 5.84
        8800                       Trading      
    4 6 D 600 500 C T Co 77.90 CityPhones 78.00 DigitalComm 78.25     Cybacom 78.50 World 84.0 5.5 7.01
        W 8000i 500 C T Co 92.90 TradeEazy 93.00 DigitalComm 93.25     Cybacom 93.50 Cellular 100.0 6.5 6.95
        Nokia                              
        9300 1500 C T Co 355.20 Stylez 355.50 MobMag 356.25     Cybacom 357.00 World 381.75 24.75 6.93
                                Cellular      
    5 6 W 8000i 1000 C T Co 188.80 Stylez 189.00 MysticMob 189.50     Globelink 190.00 Sahil Int 201.0 11.0 5.79
                                       
    6 6 Nokia 3000 C T Co 791.40 TradeEazy 792.00 Liffy Ltd 793.50     Globelink 795.00 Network 843.0 48.0 6.04
        6280 3500 C T Co 923.30 TradeEazy 924.00 MobileOne 925.75     Globelink 927.50 Trading 983.5 56.0 6.04
                                       
    7 6 Nokia 8801 3000 Oracle 1343.25 Deepend 1344.00 BlueWire (1) 1344.75 RVM (1) 1347.00 InterCom 1374.00 L.P.D.C 1431.0 57.0 4.15
        SEW 9001's 3000 C&BTdg 849.00 HillGrove 850.50 Lone Ent (2) 852.00 Primeform(2) 853.50 InterCom 855.00 L.P.D.C 906.0 51.0 5.96
        SEW 9001's 4000 C&BTdg 1132.00 HillGrove 1134.00 Danum (3) 1136.00 Micropoint(3) 1138.00 InterCom 1140.00 L.P.D.C 1208.0 68.0 5.96
                                       
    8 6 Nokia 8801 2000 Oracle 895.50 Deepend 896.00 Bluewire 896.50 Gcomms 898.00 InterCom 900.00 FranceAffai 954.0 54.0 6.00
                                       
                                       
                                       
    9   Nokia 8800 500 C T Co 211.90 TradeEazy 212.00 Talking Fonez 212.25     Globelink 212.50 Sahil Int 225.0 12.5 5.88
        Nokia N90's 1000 C T Co 291.80 TradeEazy 292.00 Talking Fonez 292.50     Globelink 293.00 Sahil Int 301.5 8.5 2.90
                                       


     

            Defaulter   Buffer   Buffer   Buffer   Megtian's Megtian's Megtian's Megtian's Megtian's Megtian's
    Deal Date Phones Qty   Net 1 Net 2 Net 3 Net   Net       Mark
    No 2006       Sales   Sales   Sales   Sales Immediate Sales Export Export Gross Up
              Value   Value   Value   Value Supplier Value Customer Price Profit  
      April       £'000   £'000   £'000   £'000   £'000   £'000 £'000 %
    10 12 Nokia 8800 5000 AscompTdg 1975.00 Primeline 2000.00         Shelford 2002.50 Lavina Tdg 2142.50 140.00 6.99
                                       
    11 12 Nokia 9500 5000 Atomic Ltd 1500.00             Shelford 1502.50 Lavina Tdg 1607.50 105.00 6.99
                                       
    12 12 Nokia 7380 3000 Prime Line 735.00             Shelford 736.50 Lavina Tdg 787.50 51.00 6.92
                                       
    13 13 Nokia 8800 3000 Worldwide 1223.40 Stylez 1224.00 TalkingFonez 1225.50     Globelink 1227.00 SpaceTelec 1300.50 73.50 5.99
                                       
    14 13 SamsD820 2000 Worldwide 537.60 TradeEaz 538.00 DigitalComm 539.00     Cybacom 540.00 Buy&Sell Tdg 576.00 36.00 6.66
                                       
    15 13 SamsD820 3000 Worldwide 717.00 Stylez 717.00 DigitalComm 718.50     Cybacom 720.00 Buy&Sell Tdg 763.50 43.50 6.04
                                       
      May                                
    1 9 Nokia 8801 4000 StockMart 2063.60 GlobalAcc 2064.00 TalkingFonez 2066.00     TeamMob 2068.00 NetworkTdg 2214.00 146.00 7.06
                                       
    2 9 Nokia 8801 6000 StockMart 3081.60 AaroLtd 3082.50 DigitalComm 3085.50     Cybacom 3090.00 SpaceTelec 3306.00 216.00 6.99
                                       
    3 11 SamsD820 2500 StockMart 572.25 GlobalAcc 572.50 RedroseDist 573.75     GlobeLink 575.00 Sahill Int 615.00 40.00 6.96
                                       
    4 11 Nokia 6280 4500 StockMart 1052.33 AaroLtd 1053.00 MobOne 1055.25     TeamMob 1057.50 AnkitaEnt 1131.75 74.25 7.02
                                       
    5 11 Nokia 8800 5000 StockMart 1999.50 GlobalAcc 2000.00 DigitalComm 2002.50     TeamMob 2005.00 Buy&Sell Tdg 2146.25 141.25 7.04
                                       
    6 11 Nokia 8800 1000 StockMart 400.80 Mopani 401.00 LiffyLtd 401.50     GlobeLink 402.00 Buy&Sell Tdg 429.25 27.25 6.78
                                       
    7 11 Nokia 9300 2000 StockMart 445.70 AaroLtd 446.00 DigitalComm 447.00     Cybacom 448.00 SpaceTelec 479.00 31.00 6.92
                                       
    8 11 Nokia 9300 4000 StockMart 891.20 Mopani 892.00 TalkingFonez 894.00     Globelink 896.00 SpaceTelec 958.00 62.00 6.92
                                       
    9 16 Nokia 8801 3500 StockMart 1780.80 Mopani 1781.50 TalkingFonez 1783.25     Eurospec 1785.00 CellularSol 1909.25 124.25 6.96


     

            Defaulter   Buffer   Buffer   Buffer   Megtian's Megtian's Megtian's Megtian's Megtian's Megtian's
    Deal Date Phones Qty   Net 1 Net 2 Net 3 Net   Net       Mark
    No 2006       Sales   Sales   Sales   Sales Immediate Sales Export Export Gross Up
              Value   Value   Value   Value Supplier Value Customer Price Profit  
      May       £'000   £'000   £'000   £'000   £'000   £'000 £'000 %
    10 16 Nokia 6280 1700 StockMart 394.23 GlobalAcc 394.40 DigitalComm 395.25     Eurospec 396.10 CellularSol 424.15 28.05 7.08
                                       
                                       
    11 19 Nokia 8800 2000 StockMart 737.60 MobMemory 738.00 DigitalComm 739.00     Cybacom 740.00 SynergyOv 792.00 52.00 7.03
                                       
    12 25 Nokia N80 2600 3DAnim 946.66 IHTechn 947.05 ITPlayers 947.70     InterCom 949.00 L.P.D.C 1006.20 57.20 6.03
                                       
    13 25 Nokia E60 3600 3DAnim 894.60 Deepend 895.50 ITPlayers 896.40 MobHeaven 898.20 Shelford 900.00 L.P.D.C 954.00 54.00 6.00
                                       
    14 25 Nokia N80 2000 3DAnim 728.20 IHTechn 728.50 PhonetoPhone 729.00     InterComms 730.00 FranceAffai 774.00 44.00 6.03
        Nokia N80 2400 3DAnim 873.84 IHTechn 874.20 DualiteLtd 874.80     InterComms 876.00 FranceAffai 928.80 52.80 6.03
                                       
    15 25 Nokia E60 3400 3DAnim 844.90 Deepend 845.75 AEResor 846.60 ARComms 848.30 Shelford 850.00 FranceAffai 901.00 51.00 6.00
                                       
          101200   33051.96               33226.10   35366.9 2140.8 6.44
    Megtian's suppliers' mark up 174.15                                
                                       
    Average MU per unit £0.17     £2.12                          


     


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/UKVAT/2008/V20894.html