TC00076 S & I Electronics Plc v Revenue & Customs [2009] UKFTT 108 (TC) (18 May 2009)

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

First-tier Tribunal (Tax)


You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> S & I Electronics Plc v Revenue & Customs [2009] UKFTT 108 (TC) (18 May 2009)
URL: http://www.bailii.org/uk/cases/UKFTT/TC/2009/TC00076.html
Cite as: [2009] SFTD 241, [2009] STI 2862, [2009] UKFTT 108 (TC)

[New search] [Context] [Printable version] [Help]


S & I Electronics Plc v Revenue & Customs [2009] UKFTT 108 (TC) (18 May 2009)
VAT - INPUT TAX
Other
    2009 UKFTT 108 (TC)
    TC00076
    Appeal number: LON/2007/0793
    VAT – Input tax – MTIC Fraud – Incorporation of Kittel into domestic law – Effect of ECHR decision in Bulves v Bulgaria on domestic application of Kittel test – Does ECA 72 prevail over HRA 98? – Whether knowledge of a particular fraud is required – Whether the tax loss is established only if there is an unpaid assessment – Whether all input tax or only that evaded should be denied – Whether discrimination in HMRC's approach is relevant to test
    VAT – Input tax – MTIC Fraud – Whether connection to fraud – Whether fraud – Whether trader knew or should have known of fraud: whether trader took all reasonable precautions – No – Whether trader would have discovered fraud if he had taken all reasonable precaution - Yes
    VAT – Input tax – MTIC Fraud – Contra trade – Need of connection between the clean and dirty chain in the form of some arrangement linking them
    FIRST-TIER TRIBUNAL
    TAX
    S&I ELECTRONICS PLC Appellant
    THE COMMISSIONERS FOR HER MAJESTY'S REVENUE & CUSTOMS Respondents
    (VAT)
    Tribunal: CHARLES HELLIER (Judge)
    CYRIL SHAW FCA
    Sitting in public in London on 13-17, 20, 22-24, 27, 28 October and 11, 13 and 14 November 2008
    Kieran Beal, counsel, instructed by Hassan Khan, for the Appellant
    Michael Davis-White QC and Aidan Robertson, counsel, instructed by Howes Percival, for the Respondents
    © CROWN COPYRIGHT 2009

     
    DECISION
    Introduction
  1. S&I Electronics PLC ("S&I") deals, among other things, in mobile phones. It buys large numbers of mobile phones and sells them on. In many cases it exports the phones. When it exports the phones no VAT is payable on the sale, but if its purchase price for the phones included VAT it will seek credit for that "input VAT". If its total input VAT for a month exceeds its output VAT (the VAT chargeable on its sales within the UK) it will seek repayment of the excess from HMRC.
  2. In the period April to July 2006 S&I sold some 132 lots of mobile phones. Of those 132 sales some 99 were exports. S&I made substantial VAT repayment claims for each of those months.
  3. The jurisprudence of the ECJ indicates that if a person knows or should have known that by his purchase of an item he was participating in a transaction connected with the fraudulent evasion of VAT, then his entitlement to VAT input tax credit may be denied (see Axel Kittel v Belgium 2006 ECR I-6161).
  4. HMRC investigated S&I's sales in the period April to July 2006. They concluded that 89 of the export sales were of phones which could be traced back, along a chain of purchases and sales, to a sale of the same phones by another person who had fraudulently evaded VAT , and that one further chain led back from S&I to a 'contra trader' (see later); that in the circumstances S&I knew or should have known of the connection with fraud; and accordingly that S&I's input tax credit on each of those sales should be denied. They made four decisions in which they communicated this conclusion to S&I in relation to different batches of S&I's sales in those months. S&I appeals against those decisions.
  5. In the period between the making and notification of those decisions and this appeal (and indeed to some extent during the course of the hearing) the Respondents refined their view of the transactions and the amount of the input tax credit they sought to disallow. The total amount of input tax at issue before us was some £4.3m.
  6. In the next section of this decision we discuss the law we must apply. In summary (and we deal with the niceties of the formulation of the issues in that section) we need to determine in relation to each of the 90 batches of mobile phones sold by S&I:-
  7. (i) whether the purchase of the phones can be traced back to a person whom HMRC allege was fraudulently evading VAT. That requires us to evaluate the evidence marshalled by HMRC in relation to each of the 90 sales. That evidence was summarised for each of those sales in a Deal Sheet showing the alleged purchases and sales of the phones in a chain leading to S&I and continuing with S&I's export;
    (ii) whether the person alleged by HMRC fraudulently to have evaded VAT, had done so in relation to its sale; and
    (iii) whether S&I knew or should have known that its purchase was connected to fraudulent evasion.
  8. Deal Chain number 8 was the alleged contra trading chain. We shall deal with that chain separately.
  9. The remainder of this decision is therefore structured in the following way:
  10.   paras
    I. Legal Issues 9 to 89
    II The evidence before us 90 to 98
    III Findings of fact in relation to the Deal Chains 99 to 109
    IV Findings of fact in relation to the alleged fraudulent traders 110
    V Findings in relation to Deal 8, the "contra trader" 111 to117
    VI Findings of fact in relation to the conduct of S&I's business 118 to 197
    VII Conclusions as to the knowledge issue 198 to 220
    VIII Conclusions 221 to 229
    I The Legal Issues
  11. In this section we address the legal issues argued before us in the following order:-
  12. (a) A general description of MTIC fraud
    (b) Kittel
    (c) Incorporation of the Kittel principle into UK law
    (d) Guidance on Kittel in Livewire, Mobilx and Bulves
    (e) Contra trading
    (f) Knowledge of What?
    (g) The nature of connection to fraud in non-contra trade
    (h) A Tax Loss – is assessment reached?
    (i) Denial of all input tax or only the tax lost?
    (j) Discrimination in HMRC's approach?
    (k) Burden of proof.
    10. At the end of the hearing the tribunal indicated that it had some questions on the detailed evidence. It was also perceived to be likely that the High Court would give judgment in the Livewire case before the tribunal released its decision. Accordingly the parties were asked to respond to the tribunal's queries and to make submissions on the subsequent judgment. This they kindly did in the form of Supplemental Closing Submissions which are referred to below.
    (a) A general description of MTIC fraud
  13. The issues before us relate to allegations of MTIC fraud, and contra trading. We adopt with gratitude the following description of these activities from the judgment of Lewison J in HMRC v Livewire [2009] EWHC 15 (Ch):
  14. "The particular form of fraud with which [this appeal] is concerned is known generically as missing trader intra-community fraud or MTIC fraud. This is a description coined by HMRC, but is generally used by those who specialise in this area. Even this generic type of fraud can itself take different forms:
    (i) In its simplest form it is known as an acquisition fraud. A trader imports goods from another Member State. No VAT is payable on the import. He then sells on those goods to a domestic buyer and charges VAT. He dishonestly fails to account for the VAT to HMRC and disappears. The importer is labelled a "missing trader" or "defaulter".
    (ii) The next level of sophistication involves both an import and an export. A trader once again imports goods from another Member State. No VAT is payable on the import. Typically the goods are high value low volume goods, such as computer chips or mobile phones. He then sells on those goods to a domestic buyer and charges VAT. He dishonestly fails to account for the VAT to HMRC and disappears. The domestic buyer sells on to an exporter at a price which includes VAT. The exporter exports the goods to another Member State. The export is zero-rated. So the exporter is, in theory, entitled to deduct the VAT that he paid from what would otherwise be his liability to account to HMRC for VAT on his turnover. If he has no output tax to offset against his entitlement to deduct, he is, in theory, entitled to a payment from HMRC. Thus HMRC directly parts with money. Sometimes the exported goods are re-imported and the process begins again. In this variant the fraud is known as a carousel fraud. There may be many intermediaries between the original importer and the ultimate exporter. These intermediaries are known as "buffers". The ultimate exporter is labelled a "broker". A chain of transactions in which one or more of the transactions is dishonest has conveniently been labelled a "dirty chain". Where HMRC investigate and find a dirty chain they refuse to repay the amount reclaimed by the ultimate exporter.
    (iii) In order to disguise the existence of a dirty chain, fraudsters have become more sophisticated. They have conducted what HMRC call "contra-trading". The trader who would have been the exporter or broker at the end of a dirty chain, with a claim to repayment of input tax, himself imports goods (which may be different kinds of goods) from another Member State. Because this is an import he acquires the goods without having to pay VAT. This is the contra-trade. He sells on the newly acquired goods, charging VAT but this output tax is offset against his input tax, resulting in no payment (or only a small payment) to HMRC. The buyer of the newly acquired goods exports them and reclaims his own input tax from HMRC. Again there may be intermediaries or buffers between the contra-trader and the ultimate exporter. The fraudsters' hope is that if HMRC investigate the chain of transactions culminating in the export, they will find that all VAT has been properly accounted for. This chain of transactions has conveniently been called the "clean chain". Thus the theory is that an investigation of the clean chain will not find out about the dirty chain, with the result that HMRC will pay the reclaim of VAT on the export of the goods which have progressed through the clean chain. I should add that HMRC do not agree with the label "clean chain" because they say that both chains are part of an overall fraudulent scheme."
    (b) Kittel
  15. The ECJ's judgement in Kittel was given against the background of two different sets of facts. In the first the facts were those of the Recolta case, where that company was innocent of any fraud. In the second the facts were those of the Kittel case where the company had knowingly participated in carousel fraud. It dealt with the first situation in paragraph 51, where it said:
  16. "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, Customs and Excise Comrs v Federation of Technological Industries (Case C-384/04), para 33)."
    13. The ECJ dealt with the second situation in para 61 (echoing para 59) where the Court said:
    "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."
    (c) Incorporation of Kittel into UK law
  17. The ECJ's decisions in Kittel and its related cases set out principles of EU law. The principle at issue is whether, and if so when, a trader's right to repayment of input VAT under the Directive may be denied. There is no specific provision in domestic legislation which implements the injunction of the ECJ. It would be possible for a domestic law to give to a trader a VAT right which was not given by the application of a Directive. It would be possible for domestic legislation to give a trader who fell within the input tax denying words of Kittel the right to retain repayment under domestic law.
  18. Mr Beal did not initially argue before us that the Kittel principle is not incorporated into domestic law, but, in response to a question from the tribunal, the Commissioners helpfully provided their argument that the principle should be treated as part of domestic law. In his supplemental closing submissions however Mr Beal devoted several paragraphs to the issue.
  19. Mr Davis-White and Mr Robertson drew our attention to WHA v HM Revenue & Customs [2008] STC 1695, in which Lord Neuberger considered the argument that the EU abuse of right principle could not be invoked against the appellant because it was only in relation to EU legislation that that principle could apply. He rejected that argument on the basis that the domestic legislation was intended to implement the relevant provisions of the Sixth Directive (see [45] and [47]). Although WHA involved the abuse principle, the ECJ in Kittel noted that "preventing tax evasion, avoidance and abuse is an objective recognised by the Sixth Directive". Thus, he says, the same result would follow.
  20. We agree with the Respondents. It seems to us that the Kittel principle should be read into section 26 VATA in the same manner as the Court of Appeal in Revenue and Customs Commrs v IDT Card Services [2006] STC 1252 found it possible to read into para 3(3) Sch 10A a further disregard - because such a disregard went with the grain of the domestic legislation.
  21. In his supplemental closing submissions Mr Beal says (i) the Kittel case was decided in the context of domestic laws which had the effect that input tax was denied where there was fraud, (ii) the Directive acknowledges that Member States should be able to take derogationary measures to combat fraud and may be authorised (under Art 27) to take special measures, (iii) the right to deduct input tax is fundamental and thus needs special derogation, (iv) no derogation has been authorised for the UK, (v) a Member State may not rely upon its failure to transpose a directive into domestic legislation – thus even if a specific derogation were not needed there is no domestic transposition of the Kittel principle, and (vi) in the absence of a clear basis for denying input tax it would be an infringement of the principle of legal certainty to permit a negligence based approach to liability to be implemented prior to the ECJ's judgment in Kittel, namely 6 July 2006.
  22. We do not agree. First, in our judgment the Kittel principle is not a derogation from the right to input tax but the description of a limitation on that right. In other words as a matter of construction of the Directive the right is flawed or limited (see also para 45 below in relation to Bulves). There is thus no need for any form of derogation to introduce it into the application of the Directive in the UK. Second, for the reasons set out above it seems to us that it is proper to construe the UK domestic legislation so that the right to input tax given by that legislation is similarly limited. And third, if properly construed the input tax right is limited by 'negligence based' considerations then that must always have been the case and there is no need for grandfathering things done before 6 July 2006.
  23. (d) Guidance in Livewire Moblix and Bulves
  24. Various issues were debated before us in relation to the passages from Kittel set out above, but following the hearing the High Court gave judgment in two cases, HMRC v Livewire [2009]EWHC 5 (Ch), and Mobilx Limited v HMRC [2009] EWHC 133 (Ch). Those judgements dealt with a number of those issues, and the parties kindly provided their written submissions on them (together with additional submissions on matters raised by the tribunal after the hearing). We deal first with the relevant conclusions we draw from these two cases.
  25. 21.. First, it seems to us that these cases support the synopsis of the tribunal in Honeyfone [2008] UKVAT 20667 that para 51 of Kittel provides a safe harbour for a trader outside the walls of which he may be sunk, but it does not go so far as to say he will be sunk outside those walls or impose a positive duty. Outside the harbour he will be sunk only if he sails into a storm of which he knew or should have known (i.e., only if, not having taken all reasonable precautions, he finds his transaction connected with fraud of which he should have known, he is not entitled to his input tax).
  26. In Livewire, after a review of the authorities, Lewison J said at [76]:
  27. "I would summarise the state of the jurisprudence of the ECJ on this subject as follows:- …vi) a trader who does take every precaution that could reasonably
    be required of him, and does not realise that he is participating in VAT fraud
    must be entitled to rely on the legality of his own transaction"
    and then, at [88], said:
    "At one stage, by reference to this supposed duty [of taking every reasonable precaution), Mr Anderson seemed to me to be submitting that if a taxable person failed to take every precaution that could reasonably be expected, he would automatically be deemed to be a participant in fraud and would forfeit his right to input tax…[However, … in] my judgement…if a taxable person has not taken every precaution that could reasonably be expected of him, he will still not forfeit his right to deduct input tax in a case where he could not have discovered the connection with fraud even if he had taken those precautions."
  28. In Mobilx, one of the grounds of Mobilx' appeal was that the tribunal appeared to view the test at paragraph 51 of Kittel (which we would call the safe harbour test) as qualified by the test at para 61. Mobilx argued that if the tribunal was satisfied that its conduct satisfied the para 51 test, it was not open to the tribunal to come to an adverse conclusion on para 61 . HMRC on the other hand argued that there was only one test. (see paras 50 to 52 of the judgement). Floyd J said :
  29. " I do not think there is anything in this ground of appeal. The Tribunal rightly focussed ultimately on the question in paragraph 61 of Kittel. For present purposes that is all that matters. In my judgment, it is clear from the Tribunal's decision that it may have viewed paragraph 51 of Kittel as dealing only with precautions in the form of due diligence checks on the supplier, as opposed to the more drastic precaution of ceasing the type of trade in question altogether. That may be what led it to make the criticism of Mr Jones' submission that I have set out above. Whether or not that is so, the manner in which the Tribunal approached the ultimate question is not, in my judgment, seriously open to criticism."
  30. It seems to us that what Floyd J says is not at odds with the judgement of Lewison J in Livewire. Floyd J did not in this paragraph accept that the para 51 test test was qualified by the para 61 test. Instead he says that the tribunal may have viewed the para 51 test as being limited to whether Mobilx had done certain due diligence checks on its supplier rather than as asking whether it had taken every precaution which could reasonably have been expected in the circumstances.
  31. We conclude that we should ask first whether S&I had taken all precautions which could be reasonably expected in the circumstances – recognising that such precautions could be quite drastic. If the answer to that question is that it did then it is entitled to its VAT input tax. If the answer is that it did not, then we must ask the further question whether it knew or should have known of the connection to fraud. And in answering that question, it is not permissible to say that because a trader did not take all reasonable precautions, he should have known of fraud. What has to be shown is that if such precautions had been taken he would have known
  32. The second point which arises is in relation to the question of knowledge. Mr Beal argued before us (and before Lewison J) that the test in para 61 of Kittel, the "knew or should have known" test, was satisfied only if there was an element of dishonesty, and that it should not be regarded as satisfied merely because the trader had been careless or negligent. Lewison J deals with this at paras [84] to [86]:
  33. 84. "Both Mr Scorey (for Livewire) and Mr Beal (for Olympia) submitted that the ECJ was laying down a test of dishonesty. They buttressed this submission by reference to the principle of equivalence, and submitted that the nearest equivalent domestic concept was that of dishonest assistance in a breach of trust, for which dishonesty is a necessary condition. This was not a question that arose in Just Fabulous, because on the assumed facts the taxable persons were dishonest. In addition they submitted that the right to deduct was an essential part of the VAT scheme and that any restriction on that right was a derogation that must be strictly applied.
    85. "I do not accept these submissions. First, the very nature of the fraud under consideration is an intra-community fraud (i.e. it is a fraud in which the crossing of national frontiers is inextricably involved). It is very unlikely that the ECJ was laying down a principle that might be differently applied in different Member States to the same chain of transactions. Second, it seems to me that the ECJ was at pains to stress that the test was not one of dishonesty. This comes out most clearly from the ECJ's answer to the question posed in Teleos, where good faith is not enough on its own (§ 68). The supplier must, in addition to being of good faith, have taken every reasonable measure to ensure that his supply was not participating in VAT evasion. But it is also evident from the court's statement that a Member State may lawfully impose a requirement on suppliers to take all reasonable precautions in order to preserve their right to deduct (Teleos § 65). A requirement to take all reasonable precautions (or to act with all due diligence and care (Netto A-G § 45)) is incompatible with a simple test of dishonesty. The policy is that an honest and careful trader should not be liable for the frauds of others (Teleos A-G § 77 fn). In addition, it seems to me that the proposition that whether a person knew or should have known is to be tested by objective facts or factors (Kittel § 59) is also inconsistent with a simple test of dishonesty. Moreover, if dishonesty were the only test, it is difficult to see why the court in Kittel would have thought that their interpretation would make it more difficult to carry out fraudulent transactions (Kittel § 58). Third, the objective of preventing VAT evasion is positively encouraged by the Sixth Directive, and consequently measures taken in that respect are not a derogation. As Burton J put it in Just Fabulous:
    "The principle of legal certainty must be trumped by the 'objective recognised and encouraged by the Sixth Directive'."
    "86 In so far as a domestic analogy is appropriate, I agree with Mr Anderson that the appropriate analogy is that of constructive knowledge or constructive notice. This was described by Denning J in Nelson v Larholt [1948] 1 KB 339, 343 as follows:
    "He must, I think, be taken to have known what a reasonable man would have known. If, therefore, he knew or is to be taken to have known of the want of authority, as, for instance, if the circumstances were such as to put a reasonable man on inquiry, and he made none, or if he was put off by an answer that would not have satisfied a reasonable man, or, in other words, if he was negligent in not perceiving the want of authority, then he is taken to have notice of it."
  34. In these passages Lewison J clearly holds that dishonesty is not a prerequisite. But neither in these passages nor in any earlier passage does he expressly equate the determination of what a trader knew or should have known with the results of taking all reasonable precautions. We discuss this later below: it seems to us that that conclusion is implicit in later passages.
  35. 28. The third point follows from the domestic analogy. This arguably leaves open a further question. That question is whether the notional reasonable man is either (i) one having the skill and experience of the taxpayer, or (ii) one having (at least) the general knowledge, skill and experience that may reasonably be expected of a person carrying out the taxpayer's functions. In Olympia and Honeyfone the tribunal adopted (ii).
  36. In Livewire Lewison J noted that the test applied by the tribunal (in the Olympia appeal) was "whether a person with the knowledge, skill and experience of the director concerned would have known that the transactions were connected with fraud." Lewison J said at [123]:
  37. "…in applying the test of what ought to have been known by a director with the knowledge, skill and experience of the particular director concerned the Tribunal, in my judgment, fell into a legal error. To the extent that a domestic analogy is appropriate, the Tribunal applied a lower standard than that which would have been appropriate to support a finding of constructive knowledge…[125]…the Tribunal expressly adopted a legal test that required fewer precautions (or a lower level of understanding) than would have been required of a director of ordinary competence."
  38. In other words any inexperience or lack of skill of S&I is to be ignored. Our reasonable man is to be taken as of at least ordinary competence (i.e. equipped with ordinary skill and experience).
  39. Fourth, we note para 122 in Livewire:
  40. "The Tribunal (in my judgment rightly) rejected Mr Beal's submission that the test of "ought to have known" required HMRC to prove that the taxable person turned a blind eye, which involved knowing that there was something there to see (§ 15)."
  41. Fifth, we noted at para 27 above that Lewison J does not provide in the passages quoted any express guidance on the test to be applied in the determination of what trader should have known, though he makes clear that an honest and careful trader, one acting with all due diligence and care, one who takes all reasonable precautions should retain his right to deduct. He provides a domestic analogy "insofar as it is it is appropriate", but does not clearly indicate when that may be.
  42. From the domestic analogy, and the other indications discussed above, we would, in the absence of further guidance, have applied the following test in relation to the "should have known" question: namely whether a reasonable man with ordinary competence in the position of S&I, and knowing what S&I knew, (a) would have taken any additional steps, and (b) would have come to the conclusion, on the basis of what he knew and had found out, that it was more likely than not that the transaction concerned was connected to fraud.
  43. However, in Olympia the tribunal described a similar test but said that it was "not the same as taking every precaution reasonably required" (see para 122). Lewison J rejected this as watering down. He said, at [123]:
  44. "I consider that the Tribunal was wrong to water down the requirement that the taxable person must take every precaution reasonably required. The test does not require the taxable person to take every possible precaution: merely every precaution reasonably required. This test gives the Tribunal sufficient flexibility to decide, on particular facts, that a suggested precaution would have gone beyond what could reasonably have been expected."
  45. It seems to us that implicitly Lewison J is equating "should have known" with "would have known had the trader taken every precaution reasonably required", but also that the formulation we suggest above is merely an expansion of "every precaution which could reasonably have been required". What is reasonably required is that the trader evaluates what he knows, makes all such further investigations as his evaluations reasonably dictate, and in the light of those reasonable precautions determines whether there is connection to fraud. That is the essence of reasonable precautions. If such precautions have been taken, then the trader is entitled to his VAT; if they have not he is entitled to his VAT only if, had he taken those precautions, he would not have come to the conclusion that his transaction was connected with fraud.
  46. In the formulation in para 33 we say "more likely than not": it seems to us that no higher test is indicated. That approach is echoed at para [96] of Livewire.
  47. Bulves
  48. In his supplemental closing submissions Mr Beale also drew our attention to the decision of the European Court of Human Rights in Bulves v Bulgaria (A/3991/03) [2009] ECHR 143. Judgment in this case was published in January after the end of our hearing.
  49. In Bulves the ECHR considered the position of a trader who had incurred input VAT for which he sought credit. The Bulgarian authorities denied that credit because the trader's supplier had not properly accounted for the VAT on the sale to the trader (although the VAT had eventually been paid). That denial was in accordance with express provision of the Bulgarian VAT Act which made the input VAT creditable only if the supplier had properly declared its output VAT. At the relevant time Bulgaria was not a member of the EU and thus the right to input tax credit established in cases such as Optigen was not relevant, but the ECHR referred to the fraud doctrine developed in Kittel.
  50. The ECHR held that in denying the trader his input tax credit, the Bulgarian state had violated Article 1 of Protocol No.1 of the Convention: the refusal was an interference with the trader's property and that interference was not justified. In reaching its conclusion in relation to justification the ECHR's reasoning included a number of statements of potential relevance to the application of the Kittel doctrine:-
  51. (i) a fair balance had to be struck in tax matters between the general interest of the community and the individual's rights (para 62);
    (ii) the State enjoyed a wide margin of appreciation in setting this balance and the Court would not interfere unless the State's action was devoid of reasonable formulation (para 63);
    (iii) the Court accepted that attempts to abuse the VAT system needed to be curbed and that "it may be reasonable for domestic legislation to require special diligence by VAT registered persons in order to prevent such abuse" (para 65);
    (iv) in Bulves' case the supplier eventually complied with its VAT obligations and paid the VAT on its supply. Thus there was no negative effect on the state budget: to the contrary as a result of assessing Mr Bulves the State "in fact received two payments of VAT for the same supply". Accordingly, the refusal to allow the appellant to deduct the input tax did not seem justified by the need to secure payment of taxes (paras 67 and 68);
    (v) lastly the Court accepted that the State may take appropriate measures to prevent fraudulent abuse of the VAT system, but if a fully compliant trader who did not know of the abuse, was not involved in the abuse and had no means of monitoring or securing compliance of a fraudulent supplier, was denied input tax credit the state would be "going beyond what is reasonable and … upsetting the fair balance which must be maintained …" (para 70).
    There are three aspects of this decision to which that judgment may be relevant. The first is in relation to the amount of the input tax which may be denied. This is discussed in section I(i), para 76 below: the refusal of input tax may not be justified where the tax has already been received by the state..
  52. The second relates to the test of knowledge or means of knowledge. Mr Beal says in that regard that the Appellant "clearly had no means of monitoring or securing compliance by the alleged fraudsters" and thus that it would be disproportionate to deny input tax. It seems to us however that it can only be said that a person should have known of a fraud if he had the means of knowing of it, and that if he had the means of knowing of the fraud then it may fairly be said that he had the means of monitoring compliance by the fraudster – not in the sense that he was able to see exactly what the fraudster did when but in the sense that he would or should have been possessed of sufficient information - whether direct or indirect - to be able to conclude that there as a fraud.
  53. If we are wrong in this then a difficult question arises. As a result of section 2(4) of the European Communities Act 1972 domestic legislation must be interpreted in accordance with EU Directives, and it is for that reason that we have found (see para 17) above) that the Kittel principle should be treated as part of or as effecting the right to input tax recovery given by the VATA 1994.However section 3 of the Human Rights Act 1998 adopts the same formula:
  54. "(1) So far as it is possible to do so primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights."
    And the section continues
    (2) This section – (a) applies to primary legislation and …"
    42. As Arden LJ said in R (on the application of IDT Card Services) v HMRC [2006] STC 1252 at [75]:
    "The approach of the English courts when interpreting United Kingdom legislation designed to give effect to Community legislation is to construe the English legislation so far as possible so as to make it compatible with the Community legislation. This is the approach that the English courts adopt to legislation implementing international treaties generally. In addition, when Parliament recently incorporated the European Convention on Human Rights (ECHR) into domestic law, it took the same formula and used it to impose an obligation on English courts to interpret domestic statute law, so far as possible, compatibly with human rights …"
  55. The difficult question which thus arises is this: if the Convention requires that a right to input VAT should not be denied to a compliant taxpayer ignorant of fraud who has no means of monitoring or securing compliance of a supplier, but the Directive, properly interpreted, denies that right in wider circumstances, then should there be read into domestic legislation the broader denial required by the Directive or the narrower one required by the Convention (on the assumption that both are possible)Does the later enactment of the Human Rights Act mean that its provisions take precedence over the European Communication Act?
  56. This issue was not argued before us – indeed we had before us only that part of Mr Beal's Supplemental Closing Submissions adverting to Bulves without any reply from the Commissioners. Given the time this case has already taken we did not think it right to ask the parties to return to debate this issue – particularly given our conclusions both as regards the principle and by reference to our findings in relation to the knowledge or means of knowledge issue ( see para 218 below). However, given that this is in effect a decision in principle we have given leave for the parties to seek a relevant reference.
  57. The third issue aring from Bulves relates to the comments in Moblix that a reasonable response to intimations of fraud would have been for Moblix to cease trading in this area. Those statements says Mr Beale must be read in light of the ECHR's judgment in Bulves: if a trader had no means of monitoring or securing compliance by a supplier to deny it input tax on its legitimate trade would go beyond what was reasonable. We do not read Floyd's judgment in Moblix as indicating that failure to cease to trade in suspicious circumstances was itself enough to conclude that the trader should have known of fraud, rather we read it as indicating that the trader should, in determining what he should have known, be treated as having taken all reasonable precautions, including those, the insistence on which vis-ΰ-vis his supplier, could have resulted in the relevant trade not taking place. In other words a precaution may be reasonable even if insisting upon it is inimical to the deal. That view of Mobilx does not seem to us to be at variance with the ECHR in Bulves.
  58. There was one other aspect of the Bulves judgment which we should note. We noted above that the ECHR found that the refusal of the VAT credit was an interference with the trader's property. It came to that conclusion by rejecting an argument that there was no property because the relevant statute provided a right to input VAT only if the trader's supplier had been compliant – in other words effectively rejecting the argument that the right was a flawed right of the type we have found the UK right to input tax credit to be (see (c) above and (j) below). It does not however seem to us that the ECHR's finding in that respect casts doubt on the conclusions we reach. That is because the question it was answering was whether or not there was a possession within the meaning of Article 1 of the Protocol and it found that such a right arose by reason of its legitimate expectations – see para 57 of the judgment. That is a different question unrelated to the issue of the nature of the right to repayment which we address in this decision.
  59. (e) Contra trading
    (i) knowledge
    46A. In Livewire at [102] ff Lewison J said:
    "102. In my judgment in a case of alleged contra-trading, where the taxable person claiming repayment of input tax is not himself a dishonest co- conspirator, there are two potential frauds:
    (i) The dishonest failure to account for VAT by the defaulter or missing trader in the dirty chain; and
    (ii) The dishonest cover-up of that fraud by the contra-trader.
    103. Thus it must be established that the taxable person knew or should have known of a connection between his own transaction and at least one of those frauds. I do not consider that it is necessary that he knew or should have known of a connection between his own transaction and both of these frauds. If he knows or should have known that the contra-trader is engaging in fraudulent conduct and deals with him, he takes the risk of participating in a fraud, the precise details of which he does not and cannot know…
    105. In other words, if the taxable person knew of the fraudulent purpose of the contra-trader, whether he had knowledge of the dirty chain does not matter.
    106. However, if the contra-trader is not himself dishonest, then there will only have been one fraud, namely the dishonest failure to account for VAT by the defaulter in the dirty chain. In that situation, the taxable person will not, in my judgment, be deprived of his right to reclaim input tax unless he knew or should have known of that fraud. But if the taxable person knew or ought to have known of that fraud, then he will be deprived of his right to reclaim input tax, even if the contra-trader is wholly innocent (as, for instance, where the missing trader and the taxable person between them dishonestly orchestrate a sale to and purchase from an innocent intermediary, which appears to have happened in Recolta).
  60. We note that in relation to knowledge about the contra trader, the question to be asked is whether it was known (or should have been known) that he was dishonest, not whether he knew or should have known that his transaction in the dirty chain was connected with fraud. It is thus a higher test than that relevant to the simple application of Kitttel to the contra trader in relation to the test of whether he is entitled to his input tax on the dirty chain.
  61. (ii) connection
  62. The trader must know or should have known of a connection with fraud. There must be a connection. What is necessary for there to be a connection for this purpose: suppose A, a fraudster, sells goods "X" to B, who sells goods "Y" to C, who sells goods "Z" to D, is that chain sufficient connection between D's purchase and A's fraud?
  63. In Just Fabulous it was part of the assumed facts (see [16]) that the contra-trader persuaded its clean chain counterparts to enter into the covering trade; and Burton J at [25] accepts that "such trading is put in place for the express purpose of providing a diversion or smokescreen". There is an assumption that there is connection via the orchestration or structuring of the trades.
  64. In Livewire at [107] Lewison J noted that the implication of HMRC's case that an accounting connection was sufficient would mean that every taxable person could be connected with every other. He then referred to the tribunal's comment that if the contra traders were not involved in covering up the missing traders' defaults, there was no logical connection between the clean and dirty chains. Lewison J found (at [109]) that there was no flaw in this, and, at [101], that the cover up was part of an over arching scheme. We conclude that HMRC are required to prove that the clean chain and the dirty chain were part of an overall scheme to defraud the revenue.
  65. We conclude that there is a connection in a contra trade only if there is some arrangement which encompasses the contra trader's participation in the clean and the dirty chain and pursuant to which its sale in the clean chain takes place.
  66. (f) Knowledge of what?
  67. The test requires that the trader knew or should have known of a connection to fraud. In our opinion that means the knowledge (or means of knowledge) required is that there is some VAT fraud connected with the transactions (ie that someone in the chain of supply was fraudulent) which in some way the transactions assist, rather than knowledge of a particular fraud or of an identified specific person's fraudulent intent.
  68. Mr Beal referred us to para [91] in Livewire:-
  69. "In cases (such as the present) where the taxable person is not a co-conspirator in an overall scheme to defraud the revenue by the fraudulent evasion of VAT it is, I think, necessary to identify the fraud with which the taxable person's transaction is alleged to be connected, and of which he should have known … I accept that the honest trader need not know the identity of the missing trader but unless he knows or should have known that there was (or was likely to be) a missing trader somewhere in the dirty chain, I do not see how it can be said that he knew or should have known that his transaction was connected with fraud. In fact, in the case of a "straight" MTIC fraud the missing trader will always be the importer of the goods, so his position in the chain (if not his identity) will be a fact which can be known …"
    In paragraphs 30.4 and 30.5 of his Supplemental Submissions Mr Beal, drawing on Livewire, says:
    "In the absence of any properly pleaded allegation … that the Appellant was … dishonestly participating in an orchestrated fraud, it falls to the Commissioners to show that the Appellant knew or should have known of the particular fraud in the transaction chain which has been identified;
    In each case bar … the Commissioners' evidence does not come close to establishing that the Appellant knew or should have known of that fraud …"
  70. We do not agree with Mr Beal that the test propounded by Lewison J is that the Appellant must have known of the particular fraud. In the extract quoted all Lewison J requires is that the trader must have known that there was a missing trader (ie fraudulent trader) "somewhere" in the chain. We are not dissuaded from our conclusion in para 52 above.
  71. (g) The nature of connection to fraud in a non contra-trade
  72. Where a defaulting trader, A, imports phones which he sells to B, who sells them to C, who sells them to Exporter E, there is an obvious connection between Exporter's acquisition of the phones and A's fraud. The connection lies in the phones sold by A (who evades the VAT on its sale).
  73. In the case of a contra trade there is no such simple connection, and in the passages quoted above Lewison J makes it clear that there must be some scheme which links the two parts of the contra trade.
  74. A similar issue arose in our examination of the non contra trade Deal Chains. In some chains there were steps in which the phones eventually obtained by S&I might have been sourced at some stage in the chain from someone other than the alleged defaulting trader. Thus, for example, a defaulting trader might sell 1000 phones of a particular sort to B who, at about the same time, might also purchase 1000 of the same type of phones from X, about whom nothing is known. B might then sell 1000 phones to the Exporter; but it cannot be determined whether those were the phones from the defaulting trader or those from X. In these circumstances it seemed to us that there must be some orchestration or linkage of the traders – some arrangement pursuant to which they are all made – before it can be said that there is a connection to fraud. We have applied that principle in our consideration of the chains.
  75. (h) A Tax Loss
  76. In section IV of this decision we consider in relation to each company alleged by HMRC to be a defaulter whether or not that company fraudulently evaded VAT in relation to its sale of the phones which are alleged to have passed down the chain to S&I. In the parties' submissions this issue was addressed in two stages: (1) whether there was a relevant tax loss to the Revenue and (2) whether that tax loss was caused by the fraud of the defaulting trader.
  77. In relation to the first of those stages Mr Beal drew our attention to a number of deals in which it appeared that no assessment had been made by HMRC in respect of the tax alleged to have been evaded. Mr Davis-White says that whether or not HMRC have raised an assessment is not relevant to the question of whether there has been a tax loss. Mr Beal disagrees.
  78. Mr Beal says that without an assessment under section 73 VATA (or determination of a debt due to the Crown in respect of an invoice under Schedule 11 paragraph 5), no recoverable right to the VAT is established. The right to the tax had not crystallised in the form of an enforceable debt. Only if the debt was recoverable could lack of recovery amount to a loss to the Revenue.
  79. We do not agree. The issue is whether there is, in the words of paragraph [59] of Kittel, "fraudulent evasion of VAT". It seems to us that this will be the case where, as the result of fraud, the state does not receive the VAT it ought to have received had the relevant legislation been complied with by the trader. The question of whether or not an assessment has been made is irrelevant.
  80. Article 10 of the Sixth Directive indicates that the tax becomes chargeable when the tax authority becomes entitled to claim the tax from the person liable to pay. In that context there is fraudulent evasion where the person who is liable to pay, because the relevant chargeable event (the delivery of the goods) has occurred, defeats the entitlement of the state by fraudulent means; that entitlement exists not by virtue of administrative action but by reason of the occurrence of the chargeable event. The ECJ said in Societι Financiθre d'Investissements v Belgium [2000] STC 164 at [23], that Article 10 "enables the date on which the tax debt arises to be determined". What is at stake in our view is the fraudulent evasion of the payment of that debt, not of a later assessment.
  81. (i) Should an effective participant in the fraud be deprived of all his input tax or only so much of it as is equal to the tax lost by the fraud?
  82. In Honeyfone the tribunal held that only the tax lost in the fraud should be denied. The Commissioners argue that this was incorrect.
  83. There was an aspect of this question which arose after the hearing as a result of the tribunal's consideration of the evidence offered that traders at the start of the alleged deal chains were fraudulent. In the case of some of these alleged defaulting traders, the tribunal came to the provisional conclusion that that trader had fraudulently defaulted on the VAT in respect of its sale but was uncertain whether the evidence showed that the trader had imported the phones. If that defaulting trader had not imported the phones there would be a VAT input tax credit attributable to its sale, and (in the absence of, and of the requisite knowledge of, an earlier fraud) the VAT lost -fraudulently evaded – would be only the VAT attributable to its margin on the sale.
  84. At the time of seeking the parties' comments on Livewire and Mobilx, the tribunal also sought the parties' comments on this issue, seeking references to the evidence of import, and asking whether if all that was proved was fraud in relation to the first trader's margin, all S&I's input tax should be disallowed. Unfortunately the way in which the tribunal phrased the second question was found to be ambiguous by the Respondents who addressed a slightly wider question than that intended by the tribunal. Happily, however, the Respondents' submissions covered the question concerning the tribunal. In summary the Respondents also submitted:-
  85. (i) if the first trader in the alleged chain was not the importer, but there was an importer prior to the first trader, then the fraud of that importer would be relevant. We agree, but the factual position we were concerned about was that in which no vendor before the first trader had been identified on the deal sheets and where it was not proved that there was an earlier fraudulent importer;
    (ii) that the denial of input tax is not contingent upon intermediate members of the chain between S&I and a fraudster knowing of their connection to fraud. We agree.
  86. We asked whether, if a trader were found to have evaded the VAT on his margin rather than the whole of the VAT on his sale (because he had a valid claim to input tax) that smaller fraud is sufficient to support a denial of all S&I's input tax in a case where S&I had the requisite knowledge of the smaller fraud. The Respondents submit that if S&I should have known of the fraud of a non-importing missing trader which is connected to S&I's purchase, then S&I's VAT should be denied in toto. The Respondents say that the precise nature of the fraud we postulated was unclear, but if the smaller fraud was simply that of a trader who fails to declare his VAT and never intended to, "that on the face of things is not MTIC fraud".
  87. We fear that we fail to understand the last point. Fraud is fraud. If there is fraud – whether "MTIC" or otherwise – and the other Kittel conditions are satisfied,then input VAT must be denied. Kittel is not limited to "MTIC" fraud. Neither did we understand the Commissioners case to be limited to an assertion that the import and export had to be "Inter-Community" (IC). But the Respondents' submission was otherwise clear: if any part of the chain was tainted by fraud of which S&I knew then S&I's VAT should be denied in toto.
  88. In the extreme case the Commissioners would, we believe, therefore say that if (i) importer A imports phones and pays £1,000,000 VAT on his sale of those phones to B; (ii) B sells the phones to C making a total profit of £1 but with fraudulent intent evades the payment of 17.5p attributable to his sale; and (iii) C should have known of B's fraud, then C, having paid £1,000,000.175 as part of its purchase price, should be denied all that £1,000,000.175 of input tax, and HMRC should profit by £1,000,000. It is in our view no defence for HMRC to say "we would not pursue the case" because C's rights depend upon the law, not HMRC's action (see "Discrimination" below).
  89. In Red 12 VATD 20900, the tribunal at [76] found that there was "an orchestration of the chain in order to separate the exporter from the importer, or defaulter where they are not one and the same." It continued: "in Kittel the court did not indicate that the right to deduct is lost only to the extent of the tax lost … it may be inferred from Kittel that a trader with the requisite knowledge … forfeits the entire right to deduct regardless of the measure of the tax lost, a desirable outcome as a means of discouraging fraud. We adopt the reasoning of the tribunal in Calltell:
  90. "If the Commissioners are right … the objective of the fraud is to extract from the Commissioners the very input tax which is the subject of the appeal … it cannot be said that preventing the participation from achieving that objective is tantamount to the imposition of a penalty."
  91. The Respondents in their skeleton argument said that the Kittel principle is not that input tax should be disallowed to compensate for the VAT evaded, but that input tax should be disallowed save for the extent that such disallowance amounts to an impermissible penalty. Where there are increments in the VAT as phones pass along a chain, they are relatively minor. Looked at as a whole the 'accomplice' – the exporter – is funding the whole deal chain and the intermediate input tax arises only because of and in the context of the original defaulter's default. They say that the discussion by Burton J in Just Fabulous was in the context of multiple recovery of substantially the entirety of the VAT rather than relatively minor 'over-recovery' arising from incremental increase in price (and so VAT) through the chain.
  92. The Respondents say that the Livewire judgment is consistent with their approach. In the context of a taxpayer who is not a co-conspirator in the fraud but someone who knew or should have known of the fraud, the Respondents point to paragraph [76] of Livewire:-
  93. "… (ii) This objective [of prevention) evasion] precludies the recovery
    of tax when the tax is evaded by the taxable person himself (Kittel 53). In such cases where the right to deduct has been examined fraudulently the deduction may be retrospectively disallowed (Kittel 55);
    …
    (iii) A person who knew or should have known that by his purchase he was taking part in a transaction connected with the fraudulent evasion of VAT is to be treated in the same way as person who fraudulently exercises the right to deduct (Kittel 55,56).
    Discussion
  94. Whilst we indicate our suspicions that many of the chains were orchestrated we were not asked to, and do not make such a finding. We are thus not in the same position as the tribunal in Red 12.
  95. In Calltell the tribunal's conclusions (quoted in Red 12 above) were premised on the assumption ("[if] the Commissioners are right") that the objective of the fraud was to extract the VAT from the Commissioners in the exporter's input tax claim. In Livewire at [91] Lewison J discusses the Respondents' submissions that the fraud encompassed the extraction of money from the Exchequer by the exporter. He has difficulty with this approach in the case of a trader who is not a dishonest co-conspirator (see also [96]: the fraud is that of the defaulter). If we find that S&I was not a knowing conspirator, it seems to us that the only fraud was that of an earlier trader who failed to pay the VAT due to the Commissioners. Unless one can find that each chain was orchestrated the monies to provide the defaulting importer could equally well come from a sale to a UK retailer as from HMRC. It does not seem to us therefore that the premise adopted by the tribunal in the extract from Calltell is relevant in this case, or that the Respondents' reliance in their submissions on the funding of the chain is persuasive.
  96. In the summary Lewison J gives in Livewire quoted above, he concludes that the 'knowing' exporter should be treated in the same way as the trader who fraudulently exercises the right to deduct. But in relation to such a fraudulent trader the right to deduct is lost in relation to the "tax … evaded". It is not clear to us that Lewison J is dealing with components of the input tax which have not been fraudulently evaded. We find the quoted passages consistent with both denial of all the input tax and with denial of only the tax evaded.
  97. In Honeyfone the tribunal said:
  98. "121. Paragraphs 33-39 and 46 to 54 of Burton J's judgment in Just Fabulous touch on multiple recovery both in the sense discussed in paragraph 15 above, and in relation to the £0.875 discussed above, but give no clear steer (understandably – because the issue was not before him on the judicial review application) as to whether the full £19.25 or the lost £18.375 should be denied: at paragraph 47 he records an argument for HMRC, which he accepts at paragraph 54, and which seems to relate mainly to the paragraph 15 Multiple Recovery, that if the Revenue sought such recovery "no doubt ways can be found by the Tribunal or the Courts to prevent double or multiple recovery" …
    "123. Mr Foulkes refers us to the argument in Just Fabulous that the recovery of more than is lost could amount to a penalty. He reminds us of Burton J's extreme example in which £2m of input VAT was denied where the original default was only £1m, and that Burton J notes that the Revenue did not accept that "recoupment of a sum more than the loss caused by the original defaulter, simply by virtue of the mark-ups on each buffer transaction along the chain, constitutes a penalty either." Mr Foulkes notes that Burton J then says at paragraph 50 that he was "wholly persuaded by the submissions" of the Revenue. The possible increase in the amount of tax claimed beyond the original default in this case falls, Mr Foulkes says, far below the level that Burton J envisaged as possibly giving rise to issues such as penalty and multiple recovery. …
    "126. We do not accept Mr Foulkes' inference that Burton J was saying that multiple recovery was acceptable. Burton J was "wholly satisfied" by submissions which included one that ways could be found by the Tribunal or the Court to prevent multiple recovery.
    "127. We agree with the tribunal in Calltell that the language of the Court in Kittel is of the right to deduct being lost rather than lost in part. But the Court was dealing with the position of the purchaser from the fraudulent trader not a trader at several removes. The reasoning of the Court is that denial of that trader's right to deduct is because the knowing trader is a participant in the fraud and is relying on the right for fraudulent ends, and that denial is apt to prevent fraud. But both these criteria would be fulfilled if only the tax lost were denied as input tax: it is only that tax in respect of which a claim is being relied upon for fraudulent ends, and the denial of a deduction for that tax would also be a deterrent. It would not be perhaps as much of a deterrent as denial of all the input tax, but to the extent the denial exceeds the tax lost the denial has more of the flavour of a penalty.
    "129. In our view: (a) the decision of the Court in Kittel leaves open the possibility that the denial should be only of the tax which was evaded; (b) denial of the evaded tax only is a denial of the use of the right to deduct to extent it is used for fraudulent ends; (c) such (limited) denial is apt to discourage further fraud; (d) if and to the extent that the denial extended beyond the tax lost it is possible that the denial could be regarded as a penalty; and (e) that last result is avoided and the neutrality of the Common System of VAT maintained by denying only the tax evaded.
    "130. We therefore find in this case that tax actually paid by the buffer traders should not be denied as input tax to Honeyfone.
    "131. We are comforted in this decision by the language of the Court in Kittel: that it is for the national Court to deny the input tax, and also by the jurisdiction given to the tribunal under section 83(c) VATA – to hear an appeal in relation to the amount of any input tax credit rather than an appeal against a decision of HMRC. That jurisdiction seems well suited to exercising the flexibility of determination which helped convince Burton J that denial of a trader's input tax would not be a penalty."
  99. We see no reason to depart from that conclusion in this case. We also find the approach taken by the ECHR in Bulves (see para 39 above) points in this direction: a refusal to allow the taxpayer to deduct input tax which had been accounted for did not seem to be justified by the need to secure payment of taxes.
  100. In the circumstances where we find a fraud at the start of one of the Deal Chains, but cannot conclude either that that trader was an importer or that there was an earlier fraudulent importer and where, as a result, the only fraud we identify is in relation to that trader's margin, it seems to us that to deny the whole of S&I's input tax rather than just that relating to that trader's margin would be tantamount to a penalty. In such cases therefore we find that only the VAT on that trader's margin should be denied.
  101. In cases where the trader at the start of the chain is a fraudulent importer, the penalty issue is not as clear, but for the reasons given in the extract from Honeyfone above, we find that only the VAT evaded by that fraudulent importer should be denied.
  102. (j) Discrimination in HMRC's approach?
  103. Mr Beal says that the Kittel principle is a derogation from the entitlement to deduct. It applies only where prima facie a right to deduct has arisen. The principle is that that right can be denied if the tax authorities establish the relevant knowledge. It applies where the tax authorities act. It is thus subject to the tax authorities acting lawfully. He says that the Commissioners should be barred from contesting input tax entitlement if their approach is tainted by a discriminatory policy vis a vis exporters for which there is no objective justification. He says that the jurisdiction of this tribunal under section 83(c) VATA extends to considering that question. He notes that in section 16 FA 1994 the powers of the tribunal are "confined" to a review of the reasonableness of HMRC's actions and refers us to para [38] and[39] of Gora and others v Customs and Excise Commissioners [2003] EWCA Civ 525. In section 83 there is by contrast no limitation, thus the powers thereby conferred must include such a review.
  104. The Respondents say that the issue is whether it is a breach of the EU non-discrimination principle to subject repayment traders to extended verification. They say it is not: rather it is an application of the EU law principle in a manner which makes for effective fiscal supervision. The measure is not arbitrary discrimination or a disguised restriction on trade. Extended verification is imposed, not because the trader is an exporter, but because it is a repayment trader. Mr Davis White says that the jurisdiction of the tribunal in this case does not extend to a review: section 17 FA 1994 provides that jurisdiction ; section 83(c) relates simply to the determination of an amount of input tax.
  105. Discussion
  106. If the way in which the Kittel principle were incorporated into domestic legislation was that domestic legislation should be treated as granting a right to input tax credit except in a case where HMRC, acting in accordance with EU law, seeks to deny the right and it is established that the trader knew or ought to have known of fraud, then the way in which HMRC acted would be relevant to the determination of this appeal. Mr Beal points to the language of para [55] of the Court's judgement in Kittel: "Where the tax authorities find that right to deduct has been exercised fraudulently, they are permitted to claim the repayment of the deducted sums retroactively." He says that is the recognition that the power can be exercised by the authorities. The national court gets involved only if the authority has exercised its entitlement. We do not think it is right to read the judgement in Kittel in this way – almost as if it were a statute. In that paragraph the Court is merely describing how denial is enforced, not setting out the only way in which it may be take place. There is no suggestion of any discretion conferred on the national authority; if there were one would expect at least some indication of the conditions for its exercise. There is no suggestion that if the authority does nothing then the fraudulent trader gets home scot free. That would not discourage evasion.
  107. It seems to us that the effect of the Kittel principle is to limit a trader's right to repayment of VAT: it makes it a flawed right rather than an absolute one. It does not confer upon anyone a discretion as to whether or not to repay input tax – the principle is that there is no such right in the specified circumstances. As a result the effect of construing UK domestic legislation in a way which conforms with EU law is to limit the right of a trader under domestic law, and that is a limitation of his right rather than a discretion conferred upon the domestic authorities to deny the right (see also para 45 above in relation to Bulves).
  108. That is in principle the same as any other provision which denies a right to input tax – for example where an input is used to make an exempt supply repayment of input tax is not available. The denial is automatic: it is a flaw in the general right to recover input tax. Where inputs are so used the authorities can deny the right to input tax if wrongly claimed and can do so retrospectively. Just as the exempt trader should not claim his input tax, so the trader who knows or should have known of fraud, should not do so. And if either trader does so the tax authority may deny his claim, or recover the wrongly claimed input tax retrospectively.
  109. Neither section 83(c) nor any other provision of section 83 provides a general duty or ability for the tribunal to review the actions of HMRC. Section 83(c) deals with an appeal against the amount of tax. We do not believe that this encompasses a duty or ability for the tribunal to review HMRC's action, but even if it did, the effect of such a review would have to be relevant to the determination of the proper amount of tax at stake before the tribunal cold commence such a review. On our view of the law HMRC's action cannot affect the proper amount of tax in this case. As a result, the actions of the administration in applying that law and in selecting the cases in which they seek to apply it are irrelevant to us. If S&I say that HMRC's actions are, in this case, or more generally, contrary to EU law, then that cannot be a matter for this tribunal for we have no general jurisdiction to review their actions: S&I must seek relief in a different forum.
  110. We are comforted in our decision on this matter by the recent decision of the tribunal in 4 Distribution Limited (VVATD 20931). The tribunal in 4 Distribution left open the possibility of a reference to the ECJ if at the substantive hearing it found that the requirement for equal treatment of domestic transactions and intra Community transactions had been breached. In his supplemental submissions Mr Beal suggests that a reference to the ECJ would be necessary in this case if we concluded that the Commissioners had intentionally taken enforcement action against the Appellant but declined to proceed against buffers more proximate to the fraud. The Respondents dispute the suggestion in 4 Distribution that discrimination might become relevant on the merits and say that in any event there was no evidence upon which we could find that there was discrimination.
  111. On the evidence before us it was fairly clear that HMRC did not seek to deny input tax credit to the buffers in the chains. It was also fairly obvious that it would not have been worth their while to do so as it seemed likely that the buffers would not have had the resources to pay. It seems to us that the relevant question is whether or not the actions taken or not taken had the effect of discriminating, not whether or not HMRC intended to discriminate, and that the rather thin nature some of the buffers might be objective justification for not pursuing them. However, whether or not there was discrimination, we are sufficiently clear in our conclusion as to the scope of the Kittel principle not to consider a reference to the ECJ necessary in this case.
  112. (k) Burden and standard of proof
  113. Before us the Commissioners accepted that the burden of proof in relation to showing on the balance of probabilities that there was fraud giving rise to a loss of tax, and that the fraud was connected to the Appellant's transactions. The Commissioners reserved their position should the matter go further.
  114. As regards the question of knowledge or means of knowledge the Commissioners accepted for the purposes of the hearing before us (but again reserving their position should the matter go further) that they had an evidential burden of raising a case, but submitted that once that evidential burden was satisfied the legal burden lay on the Appellant. For the reasons expressed by the tribunals in Calltel [2007] UKVAT 20266 and Honeyfone, we agree that that is the proper approach.
  115. As regards the standard of proof, the normal civil standard of the balance of probabilities is appropriate taking into account, as the Respondents accept, the probability of fraud in context.
  116. II The evidence before us
  117. We had 52 lever arch files of copy documentation: including documents relating to the deal chains, the alleged fraudulent traders and the conduct of S&I's business.
  118. We heard oral evidence from the following who also provided witness statements:-
  119. (1) Mr Ashraf Mohammed, a director of, and a shareholder in, the Appellant; and Mr Tariq Mohammed, a director and manager of S&I. Tariq Mohammed is Ashraf Mohammed's nephew.
    (2) John Frazer Holmes, a VAT adviser, who from September 2005 up until mid 2006 provided advice for S&I on its VAT compliance and its 'due diligence' procedures. In this context 'due diligence' was used before us (and we use it in this decision) as a term encompassing the checks for authenticity, existence and probity (and the record keeping of records in relation thereto) which a trader might, or might be advised to, conduct in relation to its counterparties or otherwise to ensure the lack of connection of its suppliers to fraud.
    (3) Robert Ross, the HMRC officer who had been principally responsible for the tracing of S&I's purchases of phones back along chains of purchases and sales to alleged defaulting traders, and who was primarily responsible for preparation of the Deal Sheets. Mr Ross had also been the officer who had had responsibility for S&I (together with others) in the years before 2006 and who made the decisions against which this appeal is made.
    (4) Nigel Saunders, Claire Badminton, Phillip Bennett, Kathleen Bushby, Matthew Bycroft, Peter Cameron-Watson, Paul Cole, Peter Dean, Simon Devine, Jennifer Davis, Martin Evans, Joanne Flober, Kastur Hirani, Jane Humphrey, Janice L'Argent, Jonathan Laing, Romaine Lewis, Gerard Marescaux, Phyliss Mee, Michael McBrine, Terence Mendes, Susan Okolu, Vivien Parsons, Barry Patterson, Christine Quinn, Jonathan Read, Timothy Reardon, Gary Saul, Brian Selwood, Gordon Smith, Alistair Strachan and Malgorzarta Wanat. Each of these witnesses were HMRC officers who gave evidence relating to the alleged defaulters in the Deal Chains. Some also gave evidence relating to documents supporting the Deal Sheets.
  120. We also saw a witness statement given by Roderick Stone, one of HMRC's officers. Mr Stone did not give oral evidence before us. We note below any respect in which we have taken account of that statement.
  121. In their Supplemental Closing Submissions addressing issues raised by the tribunal on the evidence before it, the Respondents sought leave to introduce additional evidence to support one of the deal chains. We declined to give leave to introduce that evidence. There had been a long period of preparation for the hearing and a long hearing. Enough was enough.
  122. Redactions
  123. A significant number of documents in the Respondents' documents and witnesses' exhibits had had passages redacted. Some of those documents were, during the course of the hearing, provided to the Appellants without the redactions but many were so provided after the cross examination of the witness who had exhibited them.
  124. Mr Beal says that no application was made by the Commissioners to withhold disclosure on the grounds of privilege, that the redactions impeded the proper cross-examination of the Respondents' witnesses, and that the tribunal was being asked to give a decision on the basis of potentially misleading documents (he notes in this context the passages from the reports of HMRC's officers on the allegedly fraudulent traders, which, when the documents were provided without redaction, showed views expressed which were not unfavourable to the trader concerned). The Respondents merely said that it is doubtful that anything relevant arises from the redactions.
  125. The Tribunal rules do not require documentary evidence to be provided without redaction nor did any specific direction in the pre-trial stage of this appeal so require. It seems to us however that redaction from documents provided to the tribunal is generally to be discouraged because it lowers the quality of the evidence available to the tribunal. It lowers the quality because the redaction gives rise to the question as to whether the redacted passages cast doubt upon or relevantly enhance the evidential quality of the rest of the document. It is a practice we would prefer to see pursued only with the express consent of the tribunal. In any event unredacted versions shold have been provided to the tribunal together with the redacted version.
  126. Where a passage in a document relevant to our consideration had been redacted and we had not later been provided with an unredacted copy we asked ourselves whether the place and context of the redacted passage (in circumstances where we were not provided with the unredacted document) could have cast doubt on the value of the remainder of the document. Where we concluded it may have done we placed little weight on the unredacted passages.
  127. Other matters
  128. We wish to record our gratitude for the organisation of the documentary evidence, and our particular admiration and gratitude for the summaries of the evidence and transcripts provided by those representing each party. The collation and cross-referencing in the parties' final submissions was most helpful in producing our decision.
  129. III Findings of Fact in relation to the Deal Chains
    Connection
  130. In this section we set out our conclusion on the Commissioners' contentions in relation to the 90 chains: insofar as those contentions relate to whether or not the Appellant's specified transaction was connected with the sale by a vendor whom the Commissioners contend to have been fraudulent in relation to the sale. In section IV of this decision we consider in relation to each of the alleged fraudulent vendors, whether it was fraudulent in relation to that sale.
  131. The Commissioners produced Deal Sheets which showed the transactions alleged between each member of the chain from the fraudulent trader to S&I. In some cases they included details of transactions before the import of the phones into the UK or after the Appellant's sale. Save as expressly noted we have not considered the evidence in relation to such earlier or later transactions.
  132. The evidence before us in relation to the chains consisted principally of the evidence of Mr Ross, the officer who compiled the Deal Sheets and, in relation to each chain, a bundle of copy documents. These documents included:-
  133. (i) invoices
    (ii) purchase orders
    (iii) 'release notes': instructions by a vendor of mobile phones to a freight forwarder to release phones currently held to the vendor's order to the purchaser;
    (iv) spreadsheets showing, in relation to a particular vendor, the purchases of consignments of mobile phones made in a particular period and the person to whom those phones were sold; these generally detailed the number and type of phone, the dates of sale and purchase, and the relevant prices. These spreadsheets we were told were prepared by other HMRC officers from the records of particular taxpayers, or had been prepared by the vendor itself.
    (v) documents showing the flow of monies.
    There was additional evidence in relation to the alleged contra trading chain, Deal 8.
  134. Mr Beale advanced in relation to many of the chains particular criticisms of the proof offered by HMRC. We shall deal with those criticisms individually in relation to each chain. He also advanced a number of more general criticisms which he applied in relation to all or almost all of the alleged chains before us. We shall deal with these criticisms before we deal with the individual chains.
  135. We should start by giving our impression of Mr Ross' evidence. In relation to each final version of the Deal Sheets (the Deal Sheets were amended and updated in the course of the hearing) Mr Ross' evidence was that the Appellant's relevant transaction had been traced back as shown on the relevant Deal Sheet. We found Mr Ross a conscientious and frank witness. He did not shy away from admitting an error or a deficiency. He gave us the impression of having worked methodically when preparing the deal sheets from the information available to him. There were occasions when he asserted that a chain would lead back to a defaulter – because his experience had been that a particular vendor's sales always had done so – without documentary proof, but there was no attempt to hide from us the difference between such an assertion (where the Deal Sheet would typically show a gap) and a Deal Sheet based on more particular evidence.
  136. In reaching our conclusion in relation to each Deal as to whether or not S&I's acquisition of phones could be traced back to a defaulter we approached the evidence available to us in the following way. First, in each case where Mr Beal highlighted some particular deficiency with the documentary evidence we examined that evidence and also all the documentary evidence produced in relation to that particular Deal Sheet. Second, we noted Mr Ross' evidence that S&I's acquisition had been traced back as indicated on the Deal Sheets. Third, we looked in detail at the documentary evidence in relation to a number of randomly selected deals and also in relation to deal sheets where there were oddities in the alleged chain (for example changes in the numbers of phones passing down the chain). Our more detailed examination of the documentary evidence covered more than 76 of the 90 deals. That examination gave us confidence in Mr Ross' evidence regarding the remaining 14 (or so) in relation to which we generally accepted Mr Ross' evidence without a detailed trawl through the available documentation.
  137. We should also note our approach to the spreadsheets which showed a particular vendor's transactions. In relation to the cases where Mr Ross said that a spreadsheet had been produced by the trader we accept his evidence that it was so produced and find them good evidence of the sales and purchases recorded and of the connection between sales and purchases so recorded. This conclusion was bolstered by cases in which invoices were exhibited which paralleled entries upon the spreadsheets. In relation to cases where Mr Ross said that the spreadsheet was produced by another HMRC officer we were concerned that we were being asked to accept very second hand evidence: we were being asked to accept that that officer had seen evidence of the transactions sufficient to conclude that they had taken place, and that that officer had accurately recorded that evidence, but without any opportunity to hear the officer and to form view on his accuracy, judgment or truthfulness. In making our overall evaluation in relation to any chain we therefore approached these spreadsheets with more caution but allowed that caution to relax where there was other evidence, such as an invoice in the deal chain in question or another deal chain, which corroborated the collation in the spreadsheet.
  138. Mr Beal advanced two general criticisms of the evidence of connection before us. First, he said that in most cases there was no, or no complete, evidence of the release of the phones by one member in the chain to the next; and second, that there was little or no evidence of the payments by one member of the chain to the next. In a well documented case he said such evidence would have been put before the tribunal.
  139. It did not seem to us that the absence of evidence of release and payment should cause us to doubt the other evidence before us. Our task is to decide on the balance of probabilities and on the basis of the evidence before us whether there was the relevant connection. Comprehensive evidence of release and payment could make it possible to arrive more certainly at a conclusion, but the absence of such comprehensive evidence did not point away from connection.
  140. We set out our conclusion on each Deal Sheet in Appendix I and, in relation to Deal 8 in Section V. Those conclusions are summarised in column 2 of the summary schedule in Appendix III.
  141. Comments on the chains
  142. We were struck by the following features of the deals during our audit of the deal sheets:-
  143. (i) in about 70% of cases the date on which the phones had been invoiced by each member of the UK chain was the same as, or within one day of, S&I's invoiced sale, and in almost all other cases within 6 days. We also concluded that the transactions were negotiated back to back and that generally the phones moved through the chain within the space of at most a few days;
    (ii) in the substantial majority of cases S&I took a profit per phone of between £5 and £20 whereas each "buffer" (other than sometimes the last one or two) would make a profit of 5p to £1.00 per phone;
    (iii) in some cases the documentation indicated that the phones had been exported by S&I, and a few days later had been imported again by another trader;
    (iv) in some chains S&I's supplier would be S&I's supplier's supplier in another chain. At earlier stages in some chains the same happens. However generally there was a small group of traders supplying S&I;
    (v) in some chains TopTelecom and Princeways were suppliers to S&I's suppliers. The daily logs (see below) indicated that both these companies had on occasion offered stock directly to S&I;
    (vi) in some chains we found a carousel to have been proved; and
    (vii) see also our comments on Deal 32.
    IV Findings of Fact in relation to the alleged fraudulent traders
  144. We set out our conclusions on each alleged fraudulent trader in Appendix II. Those conclusions are summarised in column 4 of the summary schedule in Appendix III.
  145. V Findings in relation to Deal 8
  146. In Deal 8 S&I sold 1000 Nokia 9300i phones on 7 April 2006. We accept Mr Ross' evidence that S&I's purchase of these phones can be traced back to an import of 2000 of these phones by SPC Direct on 30 March 2006. There was no allegation that SPC Direct had gone missing or defaulted. This chain was called the clean chain.
  147. Mr Saunders gave evidence of SPC Direct's transactions in the period March to May 2006. He said that the investigation of its return for that period indicated:
  148. (i) 30 deals in which it had been an exporter of goods which could be traced back to a tax loss at the beginning of the UK chain;
    (ii) 157 deals in which the goods could be traced back to a contra trader at the beginning of the UK chain; and
    (iii) 147 deals in which SPC had been an importer.
  149. One of the 30 deals also related to 1000 Nokia 9300i's. SPC had on 21 April 2006, three weeks after its sale in the clean chain, exported these phones. It had acquired the phones on 7 April. Its acquisition had been traced back through a series of deals, all taking place on 7 April, to an import by CT Co UK Ltd. We heard evidence from Mr Patterson that CT Co UK Ltd had defaulted on the VAT on its sale in this chain. We find that this chain traces back to CT Co UK Ltd, and that CT Co UK Ltd defaulted on VAT payments in a fraudulent manner. This chain was called the dirty chain.
  150. HMRC tendered no evidence that SPC knew or should have known that its export of phones in the dirty chain was connected with fraud or that SPC was, by its transaction in the clean chain dishonestly covering up CT Co UK's fraud. The onus of proof in these circumstances must be upon HMRC. We find that SPC did not know, and was not in a position where it should have known, of any connection with fraud. We therefore find that S&I did not know and could not have known of any fraudulent complicity or dishonest cover up by SPC. Thus S&I's input may be denied only if (a) its sale was connected to CT Co UK's fraud, and (b) it knew or should have known of that fraud (see para [106] Livewire).
  151. (a) Connection
  152. Mr Saunders said that he had "allocated" SPC's transaction in the clean chain to a transaction in the dirty chain. He said that he so allocated it because SPC's input tax on the dirty chain (on 1000 of the 2000 phones it had purchased) was almost the same as its output tax on its sale in the clean chain. A number of colleagues who had asked him for allocations in relation to other traders and chains which he had made.on the basis of correspondences between the relevant dirty and clean chains.
  153. SPC's import in the clean chain took place on 30 March. It invoiced the sale of these phones to PDA Stuff on the same day; PDA Stuff invoiced their sale on 6 April, and after two further steps they were invoiced to S&I on 7 April. SPC's purchase in the dirty chain took place 8 days after 30 March on 7 April ( the same day as S&I's export). We can accept that it would be possible to use an earlier clean transaction to mask a later dirty transaction, but the mere approximate correspondence of the amounts of VAT included was not sufficient for us to conclude that SPC or anyone else had so arranged those transactions. (The identical subject matter of the transactions did not seem relevant to this issue.) The only other evidence which might point to some arrangement pursuant to which S&I's purchase (or PDA Stuff's purchase from SPC), and CT Co UK's sale (or SPC's consequent purchase) were entered into was (i) the fact that both chains involved SPC and (ii) that SPC had been involved in a number of deals which traced back to alleged defaulters or contra traders. This was not enough for us to conclude that there was someone or were persons who made any such arrangement. As a result we did not conclude that S&I's purchase was connected with any fraud.
  154. (b) Knowledge or means of knowledge of CT Co UK's fraud
  155. In those circumstances it is not necessary for us to express any view on whether S&I knew or ought to have known of any connection with any fraud in the dirty chain.
  156. VI Findings of Fact in relation to the conduct of the Appellant's business
  157. We start by making finding of fact from the evidence before us. We then draw inferences from those findings. Some of those inferences are drawn at the end of the relevant sections. Where there was indirect or conflicting evidence we deal with that evidence before reaching our conclusions.
  158. Summary
  159. S&I had, in the period 2001 to 2007, a business which included the purchase and sale of large quantities of mobile phones. Most of its sales of mobile phones were exports. These transactions resulted in VAT reclaims. Mr Ross visited S&I to verify its VAT returns on 16 occasions prior to October 2004 other officers also made visits for the same purpose. Prior to the periods relevant to this appeal S&I received from HMRC the VAT it had claimed. In that period S&I was aware of the possibility of fraud in the chains of transactions in which mobile phones were traded. It undertook certain "due diligence" measures in relation to its suppliers and customers and the mobile phones in which it traded. In 2006 S&I received three letters (we call them the Fraudulent Chain letters) from HMRC indicating that a number of the transactions in which it had been involved had been traced back to fraudulent traders. S&I took certain steps following those letters. In the following sections we make detailed findings in relation to each of these areas, but first we need to say something about our impression of Mr Ashraf Mohammed and Mr Tariq Mohammed. Those impressions are relevant to our attitude to their evidence and the initial findings we make, and also in relation to the later inferences we draw.
  160. Mr Ashraf Mohammed
  161. On several occasions Mr Ashraf Mohammed was evasive in answering Mr Davis-White's questions. It seemed to us that on these occasions Mr Ashraf Mohammed's regard for the truth was such that he could not bring himself to lie, but neither could he bring himself to give an answer which was damaging to S&I's cause. We draw three conclusions from this course of conduct: first that Mr Ashraf Mohammed had a respect for the truth, second that in those cases where he avoided giving an answer that the true answer was one which was damaging to S&I's cause, and third that we should consider with caution those of his answers in which we detected any evasion. Our conclusions on the facts reflect these considerations.
  162. There were also occasions when Mr Ashraf Mohammed displayed a lack of regard for accuracy in his witness statements and in his oral evidence. His approach on these occasions appeared to be somewhat glib. These occasions included: (i) statements that S&I's insurance policy for particular risks was for up to £800K but that that limit was capable of retrospective change by the making of monthly declarations; the policy was for £500K and that policy was not capable of such change; and (ii) a statement that his diary entries included details for both offers of phones made to S&I and requests for phones; the pages of his diary which were produced to us and explained by Mr Mohammed included only the former. These instances gave us cause to consider carefully the accuracy of some of Mr Ashraf Mohammed's other responses (whilst accepting that he probably intended to be truthful in them), and also, in our view, threw light on the way he might have managed the business of S&I, and the care he may have taken with certain aspects of the conduct of that business.
  163. Lastly we note that Mr Ashraf Mohammed was aware that some of the telephones he sold to customers in Singapore and Hong Kong were smuggled into China and other jurisdictions apparently in violation of local laws. That contravention of local law is irrelevant to the matters before us, but Mr Ashraf Mohammed's acceptance of it did suggest to us that he was not as concerned to distance himself from illegality as some other businessmen may nowadays be.
  164. Mr Ashraf Mohammed is 44 years old. He has been working in the family business since he left school. That business has been conducted through S&I since about 1980. Mr Mohammed's elder brother is formally the chief executive of S&I. Mr Mohammed told us that during the relevant period his brother was in Pakistan bringing up his children, and that, in that period, Mr Ashraf Mohammed was the acting chief executive and was, broadly speaking, in charge of running its business. In 2006 he held 1% of the shares in S&I. Mr Ashraf Mohammed struck us as a capable businessman with experience in the Appellant's business sectors. We make no finding as to the extent to which Mr Mohammed's elder brother took part in the business.
  165. Mr Tariq Mohammed
  166. Mr Ashraf Mohammed told us, and we accept, that Mr Tariq Mohammed was being trained in the business in the hope that eventually he would take over from Mr Ashraf Mohammed. Mr Tariq Mohammed is 27 years of age. In our view he was not an experienced businessman and did not have that natural ability which can sometimes compensate for experience.
  167. The Business of S&I
  168. S&I started running the family electrical business in about 1980. In 2006 it had three divisions:
  169. (i) the Palson business. This had, in the month of April 2006, a turnover of about £106,000. It involved the import of small household electrical items from Spain and latterly from China;
    (ii) the Electronics division. This had a turnover of about £90,000 in the month of April 2006. It involved the importing and exporting of consumer electrical goods such as televisions, and camcorders; and
    (iii) the Communications division. In the month of April 2006 this had a turnover of some £11,500,000. This business consists predominantly of the purchase and sale of mobile phones. The phones are bought and sold in large quantities. Typically a transaction could be for 1,000 phones for a price of about £350,000.
  170. The transactions in mobile phones were principally in new phones although some transactions are undertaken in second hand phones. In the period April to June 2006 deals in second hand phones amounted to about £200,000, but deals in new phones to some £29,000,000.
  171. In 2006 and 2007 S&I occupied fairly substantial offices: there was a large open plan office about 25 feet by 100 feet, along the sides of which were other offices and meeting rooms. Below this it occupied two floors of warehouse storage. The goods of the Electronics division and the Palson division could be stored in that warehouse space, but the mobile phones generally were not.
  172. In this period S&I employed 15 or 20 employees and officers. Three of these were directly concerned with the mobile phone business: Mr Ashraf Mohammed, Mr Tariq Mohammed and Yousaf Mohammed ("Yousaf"), a cousin. Some staff spent time accounting for, recording, and administering the mobile phone business but they were not involved in arranging its transactions. The other members of staff were not involved in the mobile phone business. Yousaf spent almost all his time on the mobile phones business; Mr Ashraf Mohammed and Mr Tariq Mohammed spent some time on the other businesses. The deals in the mobile phones were negotiated and set up by Mr Ashraf Mohammed, Yousaf and Mr Tariq Mohammed only. Mr Ashraf Mohammed said, and we accept, that he would approve most of the deals.
  173. Mr Ashraf Mohammed told us that the deals were concluded in the following way: Mr Ashraf Mohammed and his colleagues would receive telephone calls from their suppliers and customers or make telephone calls to them. In these calls they would learn that a supplier had, say, 1000 Nokia 91 phones, and that a customer wished to acquire a similar quantity of such phones. As soon as they discovered such a match, they would then set about a price negotiation. That would take about an hour or so. Once prices were agreed they considered that the deal was set up. A purchase order would be faxed out to the supplier. They would then arrange for the phones to be inspected: the supplier would tell them where they could be found - that would be at the premises of a freight forwarder -and S&I would send its inspectors round to visit. It would take the inspectors about 15 minutes to get to the freight forwarder. The inspectors would conduct an initial inspection and report back.
  174. Until April 2006, when the level of cover was increased, S&I had insurance cover for consignments of goods being transported of £800,000 when the transport was by vehicle, and £500,000 when it was by air.
  175. After April 2006 the insurance did not cover goods held at specified freight forwarders, including Interken. In the period up to April 2006, only a few of S&I's sales were for goods of value £1m or more. Mr Mohammed said that in those circumstances he took a bit of a risk.
  176. S&I had not agreed with its counterparties formal written terms and conditions by which its transactions were to be governed although there may have been terms and conditions on the back of invoices. On rare occasions S&I's customers returned goods and S&I replaced them or offered a refund.
  177. S&I gave credit on occasion to its customers. There were examples in the deals in the Deal Sheets of amounts of a hundred thousand pounds or more being outstanding for several weeks. Mr Mohammed said, and we accept, that he gave credit only to customers to had known for a considerable period. There were also examples of payment being made by customers in advance of, or contemporaneously with shipment. Goods were also shipped "on hold" to some customers: in those cases S&I's financial exposure was limited to the cost of repatriation, loss of profit and cost of reselling if payment was not received. In some cases S&I made payment to its suppliers before receiving payment from its customers.
  178. Mr Mohammed kept a diary on the pages of which he made entries reflecting offers in relation to phones. The entries were short-indicating only matters such as the counterparty, the type of phone and the price. Mr Holmes had suggested that S&I maintain records of the offers (including prices) it had received in a given day for the purpose of checking whether the prices paid by S&I were in line with the market. A computer based record was kept but the extracts we saw from it did not generally record much detail of the phones being offered other than their type, and the number and price offered. The sheets appeared ill matched with the deals undertaken (see e.g. 4 and 5 May and Deal 41).
  179. The Memorandum of Understanding
  180. In 2002 S&I signed a Memorandum of Understanding with HMRC. This document was promulgated by the Commissioners as part of an exercise to limit MTIC fraud in areas such as mobile phone transactions. S&I was among the first traders to sign. The memorandum indicated steps a trader should consider before trading with a new supplier. We saw due diligence files S&I had maintained recording their adherence to those recommendations. Broadly speaking many of the recommendations were followed although there was some carelessness on occasions. Mr Mohammed said he was disappointed with the degree of co-operation exhibited by the Commissioners following the signing of the agreement.
  181. The first paragraph of the Memorandum indicated that there was "widespread acknowledgment … that VAT fraud involving mobile phones [was] widespread and growing significantly." Mr Holmes first letter to Hassan Khan indicated that HMRC's estimate of VAT loss from computer components and mobile phone fraud was £2bn per annum; Mr Holmes had communicated this to S&I at his meeting in September 2005.
  182. Notice 726
  183. At the time of the introduction of section 77A VATA, the Commissioners produced Notice 726 which addressed, in the content of those joint and several liability provisions, the steps a trader might take to protect himself from involvement in VAT fraud. Paragraph 8 provided:
  184. "8. Dealing with other businesses – How to ensure the integrity
    of your supply chain
    8.1 Checks you can undertake to help ensure the integrity of
    your supply chain
    The following are examples of checks you make wish to undertake to help establish the integrity of your supply chain.
    (1) Undertaking reasonable commercial checks to consider the legitimacy of customers or suppliers. For example:
    (2) Undertaking reasonable checks to ensure the commercial viability of the transaction. For example:
    (3) Undertaking reasonable checks to ensure the goods will be as described by your supplier. For example:
    We recommend that sufficient checks be carried out in each of the above categories to ensure that you are not caught in a fraudulent supply chain.
    8.2 Checks carried out by existing businesses
    The following are examples of specific checks carried out by existing businesses. These may also help you to decide what checks you should carry out, but this list is not exhaustive and you should decide what checks you need to carry out before dealing with a supplier or customer:
    (a) payments would be made to a third party; and
    (b) that in the case of import, the supplier and their bank shared the same country of residence.
    Paperwork in addition to invoices may be received in relation to the supplies you purchase and sell. We believe that this documentation should be kept as evidence of a transaction's legitimacy. The following are examples of additional paperwork that some businesses retain:
    Again this is not an exhaustive list, but does show some of the more common subsidiary documentation."
    (We note the emphasis in this extraxt on the suppliers and customers of the taxpayer as opposed to the more remote aspects of the supply chain.)
  185. Paragraph 4.6 of the Notice is headed "Can you tell me exactly what checks I should undertake?" The first word of the following paragraph is "No". The notice as we read it suggests checks but does not provide comprehensive guidance on identifying a link to fraud.
  186. The Notice explained that there was widespread MTIC fraud and made reference in that context to telephones.
  187. The VATman's Visits
  188. On 16 occasions in the period July 2002 to October 2004 Mr Ross visited S&I to verify its VAT returns. (Mr Ashraf Mohammed's evidence in relation to these visits was an example of his approach to accuracy and detail: he had described Mr Ross' visits in his witness statement as being "monthly" but they were not quite so. However two visits by another VAT officer also took place in that period)
  189. Mr Ross conducted inspections of S&I's records in relation to transactions in its VAT returns: he would for example check that S&I held invoices for the input tax claimed, and that it held evidence that the goods in question had been exported. The checks covered all three divisions of the business. Following the checks the VAT claims were, up until April 2006, paid in full, but there were a number of periods between late 2004 and 2005 when payments were made piecemeal or delayed.
  190. In the course of one or more of those visits Mr Ross discussed MTIC and carousel fraud with Mr Ashraf Mohammed and the provisions of Notice 726 (see above). Mr Ross' notebook entries in respect of these visits were not put in evidence before us. The paperwork given to and examined by Mr Ross, however, did not include any of the "due diligence" paperwork maintained by S&I.
  191. There was some conflict between the evidence of Mr Ashraf Mohammed and that of Mr Ross in relation to the extent to which Mr Ross had given S&I comfort that S&I was doing all that could be expected to ensure that that their mobile phone transaction were not connected with VAT fraud. In his first witness statement Mr Ashraf Mohammed noted that S&I had never been told of any deficiencies in its due diligence and that on each occasion Mr Ross had said that S&I's due diligence was more than adequate; Mr Ross denied having said that. In cross examination Mr Ashraf Mohammed reported Mr Ross's statements as indications that what S&I was doing was "more than adequate and more than he had seen in other places. Whilst Mr Ross denied making the statements ascribed to him in Mr Ashraf Mohammed's witness statement, he gave us the impression that at that time there was nothing in S&I's business which (other than the fact that it traded in mobile phones) excited suspicion, and that all he had seen was in order, although he had not seen the company's "due diligence" files.
  192. We concluded that Mr Ashraf Mohammed's first recollection of what Mr Ross had said was not accurate. We found it believable that Mr Ross would have made some favourable comment about S&I's procedures and paperwork, and even that he may have ventured a comment about the level of checking undertaken by S&I, but we so not believe that he gave any broad and firm assurance or bill of health in relation to the adequacy of its due diligence procedures. We find it likely that Mr Ross said something to the effect that S&I had done more than some other companies he had seen.
  193. The use of Mr Holmes
  194. In about September 2005 Mr Holmes' company was engaged by S&I's solicitors, Hassan Khan, to consider and report on S&I's VAT procedures. Mr Holmes provided an initial letter of advice and then monthly visit reports for each of the months from September 2005 to June 2006. Mr Holmes advice and reports were sent to Hassan Khan rather than S&I, and his fees were paid by Hassan Khan.
  195. Mr Holmes' initial letter of advice followed an initial visit to S&I on 12 September 2005. In that letter he addressed the incidence of MTIC fraud, due diligence and compliance procedures. At the visit to S&I he met Ashraf and Tariq Mohammed and concluded that they were aware that there was significant VAT fraud in the mobile phone sector and had a reasonable understanding of Notice 726.
  196. In each following month Mr Holmes made a visit to S&I. The visit lasted for about half a day. At these visits he inspected the deal packs presented to him by S&I: these included the transaction documents retained by S&I in relation to each deal. Mr Holmes checked that the invoices and documentation of export complied with the necessary formalities. This aspect of his work was concerned with compliance. Mr Holmes also met Tariq Mohammed, Ashraf Mohammed or Aziz or Yousaf on each occasion, and at some stage during his visit would have a 10 or 15 minute conversation with one or more of them. During that conversation he would discuss his audit of the deal packs and ask questions about S&I's due diligence. In our opinion he would also have volunteered some comments on the due diligence procedures, but he did not examine S&I's due diligence files.
  197. Following each visit he produced a report. The reports commented on S&I's compliance procedures but also made recommendations in a standard format in relation to its due diligence.
  198. Our impression was that Mr Holmes' approach was fairly mechanical. He paid a short visit, followed a procedure and completed a report by updating the previous month's report on his word processor. When he was given answers to questions he accepted them without seeking corroboration. He did not give, nor during his visits had he time to give, real thought to the question of whether the due diligence procedures actually undertaken by S&I were all that could be done to avoid connection to VAT fraud. His activities in our view should not have given any comfort to S&I that the transactions were not connected with fraud.
  199. In his monthly reports Mr Holmes made a number of recommendations as to due diligence. Mr Holmes had no recollection of any comment whether from S&I or Hassan Khan being made in relation to his recommendations: we find no comments were made.
  200. One such recommendation related to IMEI numbers (see para 159 below in relation to this term). In his early reports he recommended 100% IMEI number scanning; in March 2006 he noted that this had not been carried out in all cases; for May 2006 he said IMEI numbers were scanned on all goods removed from the UK. His information for these comments came from the documents presented to him. He was not told in 2005 or in early 2006 that IMEI scans were carried out in most cases.
  201. We note that Mr Holmes understood that dealers in phones who sold to UK purchasers did not conduct IMEI checks because the margin on such deals (about 50p per phone) was insufficient to cover the cost. The position was different, he said, for export sales where the margin was greater. Mr Ashraf Mohammed also indicated that margins would be tighter on sales to UK customers.
  202. Another recommendation related to stock inspection. Early on Mr Holmes recommended that S&I should ensure that 100% of the stock delivered was inspected to check it was as agreed with the supplier. That recommendation continued in later reports becoming emboldened by August 2006. At that stage he understood that 100% checks were not being carried out.
  203. He also recommended that a price log be maintained indicating offers and requests for phones and proposed the prices. This was suggested to provide evidence that deals were not being done at uncommercial prices. This recommendation was still being made in April 2006. He also made recommendations that a record be maintained of "how a deal progressed from initial enquiries through negotiations etc". No such record was produced to us. We conclude none was made.
  204. Lastly we note his recommendation that supplier and customer declarations be obtained. These would be declaration from suppliers and customers as to their status and the due diligence they had conducted. By January 2006 this recommendation was appearing in bold type. Later on it appears supplier declaration had become less of a concern but customer declarations were not being properly completed if at all.
  205. The Fraudulent Chain letters from HMRC of March and April 2006 were not discussed with Mr Holmes.
  206. Generally Mr Holmes' formal recommendations as to due diligence appear either not to have been received by S&I or not discussed by anyone with Mr Holmes, or alternatively were not acted upon.
  207. Mr Holmes regarded S&I's due diligence however as better than that of other companies with which he had been involved.
  208. IMEI Numbers and Inspections
  209. Mobile telephones have unique "IMEI" numbers. Checking IMEI numbers against a database of phones previously dealt in can identify phones which have come round in a circle. Mr Ashraf Mohammed told us that in 2000 S&I started by scanning a 10% sample of each consignment of phones to record the IMEI numbers of the sample. In about 2005 he said this was increased to a full 100% scan unless they know that the phones had come directly from the manufacturer or (rarely) in the case where a customer wished to purchase a consignment urgently. He said that the results of the IMEI scan were not kept with the deal documentation but in electronic form only because of their bulk.
  210. Mr Mohammed said that not many people in the industry kept IMEI numbers. S&I was one of the few which did. That was because scanning the IMEI numbers was time consuming and costly. It meant, he said, that S&I could do fewer deals. Mr Mohammed said he offered the results of his IMEI scans to S&I's suppliers but did not require them to keep full records of IMEI numbers or discuss with them building a database which would enable them to identify phones they had previously dealt in. What his suppliers did with the numbers was their business.
  211. Mr Tariq Mohammed explained that Aberdale Inspection Ltd (formerly Aberdale Logistics Ltd) carried out stock inspections and IMEI scanning for S&I (except in rare cases when S&I or 4G UK would inspect instead). In the relevant period S&I were charged 11p per phone for inspection and 10p per phone for the IMEI scan. The inspection took the form of an initial sample inspection - which would determine whether there were problems (such as duplicates or wrong phones) which would cause the consignment to be rejected. The results of this inspection would be reported to S&I - following which, if the initial inspection was satisfactory, a full scan and inspection would take place. Scanned IMEI numbers were checked by computer against the database of IMEI numbers of phones previously traded.
  212. Tariq and Asraf Mohammed told us that S&I rejected a consignment if phones which appeared on its data base as having been traded by it before were found within it. We accept that.
  213. We have mentioned that Mr Holmes' reports in late 2005 and early 2006 recommend full inspection and scanning of IMEI numbers, suggesting that full inspection was not being carried out. The evidence before us of Aberdale's charges persuaded us that IMEI checking was being carried out and the database had not been seen by Mr Holmes. We find that it is more likely that full inspection and IMEI scans were carried on substantially all S&I's purchases of phones in the relevant period.
  214. Visits to suppliers
  215. Mr Ashraf Mohammed said that prior to the end of 2005 almost every supplier had been visited.
  216. Mr Tariq Mohammed gave evidence of two site visits he had conducted on 14 September 2005 and of one in May 2006. Following these visits he completed a form for the due diligence files. We concluded that these visits were more about making personal contact with the people at the suppliers than about checking that the suppliers had in place systems which might alert them to fraud in their chains of supply. There was no serious testing or investigation by Mr Tariq Mohammed of those suppliers' systems.
  217. From Mr Tariq Mohammed, and from Mr Ashraf Mohammed's evidence of his contact with suppliers, we concluded that their principal interest in visiting suppliers was to get to know their suppliers as individuals and that neither of them seriously questioned or evaluated the systems their suppliers had in place to avoid connection with fraud.
  218. Credit Checks etc.
  219. S&I conducted credit checks (using Graydon and sometimes riskdisc) on its counterparties, and from June 2006 used Veracis to conduct inspections and due diligence on them.
  220. Mr Ashraf Mohammed said that credit checks helped satisfy you that your counterparties had been trading for a number of years and were in business. We felt that S&I conducted those checks because they were the kind of thing recommended by Notice 726 rather than because they thought they would provide any protection from connection with fraud in their supply chain.
  221. Redhill checks
  222. HMRC offered a facility through which a trader might check whether his counterparty was VAT registered, and his VAT registration number (VRN). In the period relevant to this appeal a trader could fax the Redhill office of HMRC with a request in relation to a counterparty. HMRC would reply indicating whether the counterparty was registered under a particular number. Its replies indicated that no other warranty was given. In the relevant period the Redhill office often had difficulty in replying to verification requests speedily. Notice 726 recommended Redhill checks.
  223. S&I sought Redhill confirmations regularly but in general did not require that a Redhill confirmation be received before a transaction could go ahead. In March 2006 Mr Holmes' report indicated that the deal packs he had examined did not all contain evidence of prior Redhill checks due to staff shortage at Redhill or the delay in the response to a request. In some cases Redhill confirmation was sought or received after the deal had been done. Where S&I dealt often with a particular supplier it might rely upon a verification received days or months before in relation to an earlier deal.
  224. An analysis prepared by HMRC's advisers, which we accept as broadly correct, indicated in relation to the deals at issue in this appeal that:-
  225. (1) Redhill confirmation had been received in respect of all the 43 purchases from NewWay Associates (one of S&I's major suppliers),
    (2) apart from NewWay, and out of 47 other deals:-
    (i) completed confirmations were received in advance in 13 cases,
    (ii) checks were not carried out in relation to 18 deals
    (iii) confirmation was sought, but there is no evidence of a reply in 3 cases;
    (iv) in 3 cases Redhill was unable to give confirmation;
    (v) in 15 deals confirmation was received after despatch.
  226. There was no suggestion by HMRC that, had Redhill confirmation been sought before each of the relevant deals have been completed, S&I would have discovered the fraud they allege. Indeed there was no suggestion that the suppliers concerned had been directly involved in fraud or gone missing. The relevance of the conduct of these checks in our view was only in relation the Appellant's attitude to due diligence.
  227. We can quite understand that a reasonable businessman might consider that getting a Redhill confirmation before completing a deal with a supplier with whom it dealt regularly and knew well, and for which it had received a Redhill confirmation in the last month or so, would not particularly enhance its comfort that its supplier was not supplying goods connected with fraud (although if its supplier was itself being directly fraudulent it might be a different matter). A policy of seeking, say, monthly reconfirmation might in those circumstances be quite reasonable.
  228. But the Appellant did not adopt such a policy. Instead, it seemed to us, it appeared to treat Redhill confirmations as something it needed to get for its paperwork. It appears to have adopted a moderately careless attitude to this receipt. That to our minds is indicative of an approach to checking on the integrity of its supply chain which was somewhat mechanical ("box ticking") and careless.
  229. Supplier declarations
  230. In 2005 and 2006 S&I sent supplier declaration forms to its suppliers for completion. Mr Holmes was instrumental in the drafting of the declaration form. These forms invited responses to questions about the due diligence being undertaken by the supplier. Included among the questions were ones which asked:
  231. (1) whether similar forms were sent to the supplier's supplier;
    (2) whether IMEI or inspection checks were conducted;
    (3) whether Redhill verification of VAT numbers was done;
    (4) whether credit checks, and Companies House checks were conducted.
  232. The information possessed by S&I and in particular the information possessed after it made further enquiries after the receipt of the Fraudulent Chain Letters gave reason to doubt the accuracy of the replies given by suppliers on these forms. In response to Mr Davis-White's question "You must have realised that in those suppliers' declarations, there was a serious question as to whether they were accurate?", Mr Ashraf Mohammed replied "Possibly, yes", and in response to a later question "in some cases one must have had a pretty good suspicion that the declarations were, in fact, in some respects false?" Mr Mohammed replied "Possibly".
  233. There was no indication that, if it was apparent from a form that a supplier was not doing a particular check, any action was taken by S&I or any consideration was given to halting trade with that supplier.
  234. Supplier declarations were not obtained from all suppliers. The task of obtaining them was left to S&I's accounts department. Mr Mohammed said that whilst they could pressurise the supplier, they could only do as much as they felt possible. The receipt of an acceptable declaration – or even a declaration – was not a condition of trade.
  235. We conclude (a) that S&I did not use these forms or the information in them as part of a serious evaluation of whether there was any danger that their supplies might be connected with fraud, (b) that the declarations were sought and kept merely in order to provide written evidence that S&I was doing due diligence, and (c) that a reasonable businessman would have concluded or strongly suspected that the declarations were in some respects false.
  236. Due diligence – inferences and conclusions
  237. Mr Ashraf Mohammed said that his due diligence was "ongoing from Day 1 when I met the customer and supplier, to get to know their businesses. You're talking to them, you're understanding them, you know how they're doing their businesses, what businesses they're doing, and you get a feel for that. [Supplier and customer declarations are] just a record that we keep for the files. The due diligence was done mainly by myself." Once you knew your supplier you would have a certain faith in him and that he would be "looking after his business as well and not doing silly things."
  238. S&I were aware of the contents of HMRC's notice 726. Paragraph 8 of that notice, set out above, emphasises the need for a trader to satisfy himself as to the integrity of the supply chain and makes suggestion as to what could be done (although in the case of what can be done its emphasis is –almost misleadingly- on the immediate supplier and customer). Material gathered in relation to those suggestions formed part of the due diligence material collected by S&I. However, it seemed to us that S&I's approach to the collection and evaluation of this material was careless. There was a flavour in the evidence of both Tariq and Ashraf Mohammed that these exercises were form filling and what really mattered was the personal evaluation of their counterparty. There was a lack of interest in the chain of supply to their supplier: we gained the impression that because S&I would not (for good commercial reasons) be told the name of its supplier's supplier, S&I felt absolved from any need to consider what happened in the supply chain before that supplier – a chain the likely existence of which S&I knew. The following pointed to that conclusion:-
  239. (i) the comments made by Ashraf Mohammed set out above;
    (ii) Tariq Mohammed's approach to the completion of the visit reports and generally the approach to suppliers;;
    (iii) the lack of scrutiny and thought over supplier declarations received – and the delegation of their procurement to the accounting department; and
    (iv) the failure to engage with Mr Holmes' recommendations either directly with Mr Holmes or through Hassan Khan
    (v) the approach to credit checks and Redhill checks described above.
  240. It seems to us that this approach did not engage with the problem highlighted in Notice 726. You may well know and trust your supplier, but trust and knowledge that he won't do anything silly is different from being able to conclude that he has taken the steps necessary to give adequate comfort on his supply chain. Someone you know and trust as a business counterparty may well misunderstand or be careless about what is required in relation to his supply chain: it is necessary to ask serious questions steadfastly, to get good answers, and to test those answers. That S&I did not do either because they were careless or because they did not understand the nature of what they needed to do. In our view what they did gave them the comfort a reasonable businessman might seek that their suppliers existed, would honour their contracts, had VAT numbers, and, probably that their suppliers would not evade VAT, but did not provide the comfort which a reasonable person would seek if he wished to conclude on balance that there was not fraud in the supply chain. That of course on its own does not mean that they knew or should have known of fraud, but it does mean that it cannot be said that the investigation and consideration they carried out was sufficient to say that they should not have known of any fraud.
  241. The Fraudulent Chain Letters : Joint and Several Liability Warnings
  242. In two letters dated 28 February 2006 and one letter dated 12 April 2006 HMRC informed S&I that, respectively:-
  243. (i) they had selected 8 transactions from those conducted by S&I in August 2005 (and which were reflected in the 08/05 VAT claim) and of those 8, 6 had been found to be connected with VAT evasion by a missing trader;
    (ii) 9 transactions had been selected from those represented in the 09/05 VAT claim; 4 of those had been traced back to a defaulting trader; and
    (iii) of 12 transactions selected from those represented in the 12/05 VAT return, one had been traced back to a defaulting trader.
  244. The letters warned that HMRC could invoke the joint and several liability provisions under section 77A VATA. Those deals were as shown in the table:
  245. Period Deals Defaulting
    Selected Deals
    08/05 8 6
    09/05 9 4
    12/05 12 1
  246. The related suppliers in respect of the Defaulting Deals were NewWay; 5 deals, Sprint: 2 deals; TEC : 1 deal; DVB : 1 deal; Future Communications : 2 deals.
  247. In the period 04/06 (a period falling after the first letter and in part after the second letter) NewWay was the Appellant's single biggest supplier supplying phones on 13 occasions to the value of some £3.1m.
  248. HMRC did not write to those suppliers warning them also that their transactions had been traced back to defaulters.
  249. On receipt of these letters S&I did three things. The directors had a long meeting with their lawyers who wrote on S&I's behalf to HMRC; they telephoned and visited the suppliers concerned; and they wrote formally to them. The only relevant aspect of the letters from S&I's lawyers is that they ask HMRC what else S&I might do. No helpful reply was received to that question. We discuss the suppliers' responses below.
  250. , Mr Ashraf Mohammed was taken aback when he discovered that HMRC had not also written to the suppliers. He said that when he contacted the suppliers he was quite heated but they calmed him down. He was told that they had had nothing from HMRC, that they had been dealing with the same people as in the past, that they had had HMRC inspections and that they had been carrying out due diligence.
  251. In relation to NewWay, Mr Ashraf Mohammed accepted what NewWay told him (orally and in reply to S&I's letter), namely that they had been trading with their suppliers for some time (and had since visited them), that they had done Redhill and credit checks and had received some form of supplier declaration (but did not indicate the form of that declaration – and in particular what comfort it gave in relation to their supplier's supplier). On the back of that he thought, well that's fine, there is nothing more we can do: we'll carry on trading with NewWay.
  252. TEC provided a little more information on the checks it conducted in relation to its suppliers. It indicated that it conducted checks on their suppliers' directors, and with their bank, and other tests directed to the existence and bona fides of its suppliers. But there was no indication that it had sought to question itsuppliers' due diligence or take any new steps.
  253. Sprint indicated that they had established due diligence procedures (including the use of an external body to enhance checks) and that they had had a recent visit from HMRC who were happy with their procedures. No steps were taken by S&I to test this assertion with HMRC. It was not clear whether this visit happened before or after the fraudulent chain came to light, and S&I did not take steps to find that out. Mr Mohammed had known the people at Sprint for 5 years or more and said he had believed that they would have taken extra steps going forward. We find that Mr Mohammed did not know the nature or extent of any extra steps Sprint would take, and that he took no measures to find out.
  254. DVB replied by indicating that they had checked with Redhill, had a supplier declaration, their supplier was trading and had no current disputes with HMRC. There was no suggestion in their reply that they had conducted any further investigation with their suppliers. Mr Mohammed felt sure he would have asked them that and that they would have conducted further investigations but he was not told what checks their supplier was undertaking. We find that S&I did not know, and did not take steps to find out the nature and extent of DVB's suppliers' due diligence at this time.
  255. No answer appears to have been received to the letters sent by S&I to Future Communications. We accept that Mr Mohammed spoke to them and that he received some general comfort that they were undertaking some checks on their suppliers, but we find that S&I did not know and did not take steps to find out the nature and extent of those checks.
  256. We find from Mr Mohammed's oral evidence that:-
  257. (i) S&I required none of these suppliers significantly to enhance their investigations into their suppliers by for example seeking and testing the detail of the approach those suppliers took to their suppliers;
    (ii) S&I did not indicate to any of their suppliers that unless they could provide additional comfort they would not trade with them;
    (iii) S&I did not require those suppliers not to trade with those of their suppliers who had been party to the identified transactions until they had investigated the relevant chain and identified how their due diligence had been at fault.
  258. We find that generally S&I's approach was that it was not concerned to "Vouch for anybody else's due diligence" i.e. to test the robustness of their checking of their supply chain.
  259. We note Mr Mohammed's surprise that his suppliers had not also been sent letters about the fraudulent chains, but we cannot see how that could allay the concerns which those letters must have raised. These letters must have given rise to a serious concern that the steps S&I and its suppliers had thitherto taken were insufficient to prevent connection to fraud.
  260. VII Conclusions – knew or should have known
    The Appellant's awareness of VAT fraud in the mobile phone market
  261. It is clear to us that S&I were by 2006 well aware that there was VAT MTIC fraud in the mobile phone market. It gained that awareness inter alia from: (i) its signing up to the Memorandum of Understanding in 2002; (ii) discussions with Mr Ross in the course of his visits; (iii) the receipt of information about the 2003 Budget changes which introduced s.77A VATA, and Notice 726, and (iv) the discussions with Mr Holmes (Mr Holmes' engagement indicated that S&I was aware of the Commissioners' concerns about MTIC fraud).
  262. We also find that S&I was, prior to the receipt of the Fraudulent Chain Letters, aware that its business was not immune from the risk that its transactions might be connected to fraud (or at the least a reasonable businessman in possession of the information held by S&I would have been so aware). That is because (i) it had received letters from HMRC's Redhill office in 2003, 2005 and 2006 indicating that the VAT numbers of traders about whom it had enquired had had their numbers hijacked or cancelled: S&I was thus aware that some of those with whom they might have considered dealing may have been linked to fraud; (ii)in 1999 an attempt had been made to invlve S&I in MTIC fraud as a buffer. Mt Ashraf Mohammed had provided evidence for the prosection; (iii) Mr Ashraf Mohammed knew a chain of suppliers was involved in the phones he purchased and that he did not check on his supplier's supplier or earlier suppliers in the chain: he must therefore have been aware of a theoretical risk of fraud earlier in the chain.
  263. We also find that S&I would in 2006 have been aware that the degree of fraud in the market was significant and serious. The fuss made by HMRC – in Notice 726, in verification visits, in providing for Redhill checks – indicated that such fraud was a real concern.
  264. The Respondents' contentions
  265. The Respondents set out a number of factors which they say cumulatively indicate that S&I knew or should have known of the connection of their deals to fraud. We consider each of those factors in turn and the separate pointers in the opposite direction raised by Mr Beal. We then set out our conclusions on the relevant issues with our reasons.
  266. Factor 1: The deal chains revealed in the deal sheets are themselves suspicious especially where there is evidence of carousel. The profits taken at each step are suspicious: S&I, the exporter takes significantly more than the others who take a modest profit. The fact that a number of deals are carried out on one day is also suspicious.
    There were only a few Deal Chain (Nos 9, 37, and 87) where we found carousel (as to all or part) to have been proved (although there were grounds for suspicion of a carousel in a few more). We found the formal dance wherein early buffers made about 50p per phone, the penultimate buffer about £1, and S&I generally about £5 suspicious. Mr Mohammed in his oral evidence accepted that the chains of transactions generally all happened on one day and no-one held stock. He did not find the chains suspicious. Mr Mohammed in answering the Tribunal's questions accepted that profits on UK-UK transactions were lower.
    Overall we found the nature of the chains disquieting and believe that a reasonable person knowing that its transaction generally took place on one day and that UK-UK profits were lower than the profits made by exporters, would have been disturbed by that knowledge.
    Factor 2 Almost all of S&I's export deals in April, May and June 2006 were linked to fraud. HMRC say S&I must have gathered some knowledge of fraud.
    Mr Ashraf Mohammed was experienced in the mobile phone market. He told the Tribunal that to a fair degree he knew everyone of consequence in the UK phone market. S&I were conducting 40 or more transactions each month between April and June. We think it likely that some whiff of malpractice would have reached S&I in relation to some of its deals, but we do not find that S&I would, from its dealings and contacts on their own, have been able to conclude that its deals were linked to fraud (although that would not in our view have been impossible).
    Factor 3 HMRC say S&I knew the market. The export market was a small one. S&I dealt with a fairly limited number of suppliers. The relationship between S&I and its suppliers and customers was harmonious. S&I's suppliers [and their suppliers] gave references to S&I for one another. Given the small number of suppliers it was difficult to see why S&I did not cut out the middleman. That suggested that S&I was privy to fraud.
    Mr Beal says that Mr Ashraf Mohammed rightly said that if he dealt with only a small number of suppliers he could be surer that he could trust them. Although Mr Mohammed knew his suppliers well he said he would still compete with them.
    We find it odd that, for example, in some chains Diginett sold to NewWay who sold to S&I, and in others Diginett sold directly to S&I. In a market in which we are told participants spoke easily to one another and deals were fluid until the last minute, this looks like contrivance rather than chance. If it was contrivance some indication of it was likely to reach S&I.
    Factor 4 HMRC say that aspects of S&I's dealing are uncommercial.
    First, they say that there are cases where a counterparty on one occasion acted as supplier to S&I, and on another occasion as purchaser from S&I. HMRC say this makes no sense because rather than purchase from S&I that counterparty could have purchased from its own suppliers and earned for itself the profit made by S&I.
    We are not wholly persuaded by this. If S&I sourced the phones it sold to that counterparty from suppliers unknown to the counterparty, then it made its profit by its connections. But if S&I sourced phones from suppliers known to the counterparty then the possibility remains that in the bilateral conversations to set up the deal S&I got to the supplier first.
    Second, they say that most of the deals involve half a dozen or so back to back transactions concluded on the same day. They say the mobile phones typically remained in the same warehouse throughout. That is all correct. The virtually instantaneous nature of the deals points, they say, to contrivance.
    We are perturbed by the same day, back to back, feature; Mr Ashraf Mohammed accepted it as a feature of the market because "no one wanted to take stock". What, to us, is odd is that in the chains we examined it appeared in the majority of cases that the import and export and the intervening UK steps took place on the same day. Someone, somewhere would surely have held the stock for more than minutes, but why was that person almost never part of these UK chains? If the phones were moving from one longer term (more than a day or so) holder through the chain to another longer term holder, why was that first holder almost always outside the UK? Some intimation of this must, in our view, also have reached S&I.
    Third, they say it is striking that for the buffers in the supply chains leading to S&I each buffer makes a small margin, but S&I makes a much larger one. How have the buffers consistently missed out on the great opportunity seized by the Appellant?
    As noted earlier, we too were struck by this feature. The buffers in the 90 chains we examined made generally between 20p and 80p per phone traded, although very often the buffer next to S&I made a little more - £1 to £3. There were exceptions but this was the general rule. S&I generally made between £5 and £10. It was like a courtly dance.
    Mr Ashraf Mohammed said that in setting his margin he had to take account of the costs of running a big business and his finance costs. His finance cost included the financing of the VAT input tax while awaiting payment from HMRC. He said that having the money in hand enabled them to deal again and that was a cost of being out of the money. Our rough calculations (which ignored carriage and insurance costs) suggested that the margin S&I made on a phone represented a return of several hundred per cent per annum on the capital tied up in the input VAT. S&I did not appear to apply the same rates when extending trading credit to his customers. We were unconvinced by Mr Ashraf Mohammed's approach. It seems to us more likely that either there was a general acceptance in the market that exporters took a higher margin because they ran the risk of having their input tax claims denied because there was fraud in the chain, or that the margins taken at each stage were contrived by a controlling or suggesting hand. Either way the margins would provoke suspicions of connection to fraud in the mind of a reasonable man.
    Fourth the Respondents point to the Appellant's export margin remaining fairly constant over the period. We are not sure that this is a proper description of the varying margins per phone and the percentage margins per phone made by the Appellant. But our conclusions on the issue are dealt with in the preceding paragraph.
    Fifth, the Respondents say that the low level of insurance, the lack of formal terms and condition of business, and the extending of credit to customers in those circumstances is indicative of participation in fraud.
    We fail to see why this is the case. Lack of insurance, lack of clear terms and conditions, and the granting of credit are all risky practices. They point to a business taking risks (including a risk that transactions could be connected to fraud), but we can see no way in which such behaviour necessarily implies participation in fraud.
    It seems to us that a significant part of Mr Ashraf Mohammed's approach to due diligence was concerned with whether he could trust his customers to pay him and his suppliers to provide the goods. That was how he managed his risk in this respect. But this concern probably eclipsed concerns about avoiding connection with fraud.
    Sixth, the Respondents say that the phones dealt in were mostly of central European specification with plugs not intended for the UK market. That was consistent, they say, with phones being imported and exported for the purpose of MTIC fraud.
    The Appellant convinced us that it was possible to change the plugs and specifications of phones relatively cheaply so that phones intended for one market could be adopted for use in another. But we note that this would preclude quick back to back transactions.
    However, it seems to us that a reasonable businessman, being aware of the nature of MTIC fraud and finding himself exporting considerable numbers of phones which must have been imported into the UK (because they were configured for other markets) would have become concerned that he was involved in a fraudulent chain. The Appellant may have been blind to this - being used to a market in which those phones were common - but the Appellant's blindness would not prevent a sighted businessman's suspicion.
    Sixthly they note that the invoices or purchase orders sent out received for the mobile phone deals contain very limited description of the phones. Generally there was little mention of matters such as regional specification, software, languages, manuals, warranties and plugs. That indicated that neither S&I nor its customer were really interested in the phones other than as a tool for fraud.
    Mr Ashraf Mohammed said that if the detail was not on the documentation it would have been agreed orally. He also said that if, on inspection, the phones S&I was about to purchase turned out not to fit the customer's requirements a telephone call to the customer would be made to see if the differently specified phones were acceptable.
    We were not wholly convinced that in every case all the relevant details would have been agreed orally if they were not in written form. Mr Mohammed's diary entries for offers and requests for phones did not generally contain much detail, neither did the daily offers and requests lists. We take Mr Mohammed to have meant that if details were mentioned by customers they would be agreed with customer and supplier orally.
    But we do not find the lack of interest in the details indicative only of the use of the phones for fraud. It is equally consistent with the use of the phones as the commodity in an arbitrage market. It would however be something which might raise the suspicions of a reasonable businessman that there might be a connection to MTIC fraud.
    Taking these issues together we find that they do not point unequivocally to involvement in transaction connection to fraud, but they would raise in the mind of a reasonable businessman serious concerns abut such a connection.
    Factor 5 HMRC say that S&I's reaction to the Fraudulent Chain Letters was not indicative of an honest trader: the position was revealed by those letters to be of high risk and S&I's reaction was inadequate
    We found S&I's reaction in relation to their approach to the suppliers to be slipshod and mechanical. They did not engage properly with the risks. But, although this approach could have been consistent with that of a dishonest trader trying to tick boxes, we did not find it indicative of dishonesty.
    Factor 6 HMRC point to the knowledge imparted to S&I by the Fraudulent Chain Letters. We deal with this issue below.
  267. Mr Beal on the other hand points to the steps taken by S&I in its due diligence. He notes:-
  268. (i) the age and size of its business
    (ii) its good statutory accounting compliance
    (iii) that it now trades as a reverse charge trader (one of only 179 mobile phone traders who do)
    (iv) that it is an approved supplier for well known businesses
    (v) that it inspects goods and keeps IMEI numbers
    (vi) that it employed Mr Holmes
    (vii) that it conducted Redhill checks, obtained documentation from suppliers and credit agencies, and visited suppliers
    (viii) that it asked HMRC (via its lawyers) after the early Fraudulent Chain Letters, what else it could do
    (ix) Mr Ross' comments in relation to its due diligence
    The first four of these matters indicate that S&I is a serious business but have little bearing on its attitude to investigating its supply chain; the next four did not, as we have explained, indicate that it had a serious interest in the supply chain; the last might provide some measure of excuse for a failure to grapple with the issue of the supply chain, but does not affect what it did or did not do in practice and whether or not what it did or should have done would have revealed fraud. In particular we do not find Mr Ross' view of what S&I had done conclusive of, or particularly relevant to, the question for us – namely whether had S&I taken all reasonable precautions it would have discovered the frauds.
    Discussion
    Did S&I know of the connection to fraud?
  269. Mr Ashraf Mohammed steadfastly denied knowing that S&I's transactions were connected to fraud. We found him, as we have said, sometimes careless and sometimes evasive but we considered his evasiveness an indication of respect for truth. The factors advanced by the Respondents do not in our view provide compelling evidence that Mr Mohammed knew of a connection to fraud, rather, overall they suggest to us that his principle concerns were related to his business relationships with his counterparties and with satisfying the documentary requirements for the repayment of S&I's VAT claim. He did not in our view direct his mind seriously to the question of whether or not it was likely that his purchases were connected with fraud.
  270. Neither do we find that the factors advanced indicated clearly that Mr Tariq Mohammed or any other member of S&I's staff knew that it was more likely than not that S&I's deals were connected to fraud.
  271. We conclude that S&I did not know that its deals in the relevant period were connected to fraud.
  272. Did S&I take every precaution reasonably required?
  273. We do not believe that S&I took every precaution reasonably required in relation to the deals at issue. We so conclude because:-
  274. (i) S&I was aware that there was serious fraud in the grey mobile phone market;
    (ii) the Fraudulent Chain Letters made it clear that S&I's due diligence procedures had not protected it from acquiring mobile phones connected to fraud;
    (iii) neither in relation to the suppliers identified by the deals described in the Fraudulent Chain Letters (the 'identified suppliers') nor in relation to any other suppliers did S&I take the actions which would have been taken by a reasonable businessman (or which were reasonable and proportionate) to ensure that its future transactions would not be connected with fraud in these circumstances.
  275. In our view the steps which would have been so taken would have included:-
  276. (i) in relation to all its suppliers requiring them to certify that their suppliers had not only provided evidence of their existence and VAT registration and intention and ability to pay their VAT, but that they were making similar requirements of their suppliers in turn (and so on);
    (ii) ceasing to trade with any supplier until adequate believable certification had been received;
    (iii) in relation to the identified suppliers requiring that they and their suppliers (and their suppliers' suppliers and so on) submit to an independent audit of their certification before and in relation to any further transaction be undertaken with them (such an auditor being mandated not to reveal the identity of the suppliers).
    No such steps were taken. There was no serious attempt to improve or deepen the investigation of suppliers other than the identified suppliers and no steps were taken to limit trade with the identified suppliers or significantly to enhance the comfort obtained from them.
    Should S&I have known that its transactions were connected with fraud?
    (a) the identified suppliers (those indentified by the Fraudulaent Chain Letters)
  277. It seems to us that a reasonable businessman possessed of the knowledge of S&I (i.e. the knowledge of its directors and those of its employees engaged in its mobile phone business) would have come to the conclusion that it was more likely than not that purchases from the identified suppliers would have been connected with the fraud of a person earlier in the chain of supply to it. We do not here find that S&I should have known of the fraud of a particular person in the supply chain, but that it should have known, in a case where there was fraud in the chain, that it was more likely than not that there was a fraud by some earlier member of the chain.
  278. he following factors weighed in favour of this conclusion (in addition to those set out in 213 below in relation to other suppliers):-
  279. (i) S&I's knowledge that there was seriousfraud in this market;
    (ii) its knowledge that there was some possibility of fraud affecting its supply;
    (iii) S&I's knowledge that transactions involving those suppliers had been traced back to the fraud of a person in the chain of supply to S&I;
    (iv) that S&I should have known that some of these suppliers were taking a somewhat cavalier attitude towards their supplier's declarations, and should have been suspicious of the level of due diligence carried on by them and thus should have had little confidence that they were taking steps to ensure the transactions were not connected with fraud, and accordingly that S&I should have known that the likelihood of the connection to fraud in relation to deals done with those suppliers was the same after its correspondance with them resulting from the Fraudulent Chian Letters as it was beforehand.
    210. Set against these pointers to the connection of deals with these suppliers to fraud were:-
    (i) the fact that letters had not been sent by HMRC to the suppliers. We cannot see how this could have reduced any concern;
    (ii) the replies obtained orally and in writing by S&I from those suppliers. These replies provided in our view very little extra comfort: no intimation of rigorous pursuit was given, no suggestion was made that procedures were being seriously tightened or that suppliers previously involved would not be, no real extra comfort was obtained other than a proposal – accepted by Ashraf Mr Mohammed – that S&I should continue to trust them;
    (iii) that Mr Ross had (as we have found) indicated that S&I had done more than other companies he had seen, and that previous periods' VAT claims had (on occasion after some delay) been paid. But, whatever accolade Mr Ross had conferred, the due diligence undertaken by S&I had not unearthed the frauds, and the repayments had been made in respect of periods in which S&I now knew there had been fraud. These factors thus did not make the chance of connection to fraud any smaller and in fact should have caused S&I to doubt any comfort it had previously taken from its procedures or from Mr Ross.
    (iv) HMRC's letters did not indicate whether the deals in each sample which had not been traced back to a defaulter had been traced back to an importer who had paid VAT. But no comfort was available from the letters in relation to the other deals in each sample. We do not believe that a reasonable man would have concluded that it was only the identified deals which were connected to fraud.
    211. If we are wrong, and a reasonable businessman would not have concluded that it was more likely than not, merely on the factors we have listed, that supplies from the identified suppliers would be connected with the fraud of an earlier supplier in the chain, then it seems to us that, possessed of this information in relation to these suppliers such a businessman would have taken such stringent steps in relation to future deals with these suppliers that the fraud in relation to the transactions at issue with these suppliers would have been discovered or a conclusion reached that it was likely that there was fraud in the chain. We indicate the kind of steps we consider reasonable in "the other deals" para 215below.
    212. It seems to us no defence to say "well, if you were that awkward the deal would not have been done", because if you are considering the national world of what should have happened you must bring into it as invariables the sales which did happen and ask what a reasonable man would have found out. Mr Ashraf Mohammed said his discussions with the identified suppliers were initially heated; a reasonable response was in our view to be and remain heated: we believe a person reasonably (and therefore seriously) concerned about his supplier's supply chain would have found out enough to conclude that it was likely that there was a fraud somewhere in the chain.
    (b) the other deals
    213 The following factors weighed in favour of a conclusion that a reasonable person with S&I's knowledge should have known that each of the other transactions were connected to the fraud of someone earlier in the supply chain – i.e. that he would, on the knowledge he had or which he would have obtained by making all reasonable further enquiries, have come to the conclusion that, on the balance of probabilities, those transactions were connected to fraud;-
    (i) S&I's knowledge by the end of 2005 that there was a serious degree of MTIC VAT fraud in the mobile phone market and that S&I's supply chains were not immune from the risk of such fraud;
    (ii) S&I's knowledge that some of its purchases had been traced back to fraud. Only 11 of the 29 deals selected had been traced back, but HMRC's letters gave S&I no comfort that the others might not also chase back to fraud;
    (iii) S&I's knowledge that therefore its former due diligence procedures did not protect it from connection to fraud;
    (iv) a reasonable businessman's appreciation that the additional steps S&I took after the three warning letters gave it no greater comfort in relation to any of its suppliers;
    (v) the conclusion, which in our judgment a reasonable businessman would have reached on the basis of (i) to (iv), that it was likely that a significant proportion of S&I's post March 2006 transactions could, and that some would, be connected with fraud;
    (vi) the serious suspicions which would have arisen before 2006 in the mind of a reasonable businessman as a result of the knowledge (a) that "if you look at the whole industry, nobody actually does hold stock. Everybody purchases stock on a daily basis" and thus that chains of supply were involved, (b) that S&I's margins were higher than UK-UK sales and that some knowledge of the courtly dance of margins, back to back simultaneous transactions and the factors described at the end of section III above must have reached S&I, (c) that MTIC fraud usually involved an import and a chain to an export, S&I were exporting and when phones had a non-UK specification it was likely they had been imported, and as a result it was possible that many of S&I's exports were part of a chain starting with a fraudulent import, (d) of the possibility that because there was some lack of interest in the details of the phones they were being used as a vehicle for fraud as well as arbitrage, (e) of the possibility that in the same way that the information later obtained in relation to the suppliers identified from the Fraudulent Chain Letters indicated that some of these supplier's declarations were likely to be erroneous, the supplier declarations made to S&I by the suppliers could have been carelessly, recklessly or dishonestly made and that it was likely that some had been; (f) that some whiff of contrivance of the chains of supply to it or of malpractice must have reached S&I; and (g) of the other factors which we note in para 201 above which would raise serious concerns about a connection to fraud.
    214. These factors would in our judgment have given rise, in the mind of a reasonable businessman, to very serious concern, and possibly a conclusion that it was more likely than not, that each of S&I's April, May and June transactions would have been connected with fraud.
    215. Mr Beal says that S&I undertook further due diligence which would ameliorate these concerns. For reasons set out elsewhere we do not think that the additional due diligence actually undertaken should have reduced its concerns to any significant extent. Then Mr Beal says: what else could S&I have done other than stopping trading? In our opinion a reasonable businessman in these circumstances would have insisted upon the source of each new transaction being chased down the chain by an independent third party (who undertook not to divulge details of the suppliers' identities) as a condition for the completion of the transaction (or perhaps S&I could have insisted upon being given adequately secured personal indemnities from the individuals in charge of its suppliers against the loss of the input tax at stake as a result of any (or such insistence on indemnity) of fraud in the chain). In each case where there was a fraudulent trader this would in our view have revealed the fraud.
    216. Thus in these circumstances we conclude that S&I should have known of the connection to fraud. (We have dealt with Deal 8 separately).
    (c) general points
    217. In the preceding paragraphs we have concluded that there were precautions which S&I could reasonably have taken in the circumstances which would have disclosed the connection of its transactions to the frauds of the defaulting traders or the likelihood of such connection. We should make three comments on that conclusion.
    218. First, we believe that S&I thought that all that it had to do was to be satisfied about its suppliers. In our view it was wrong.
    219. Second, we believe that the steps which we think it could reasonably have taken were such that it could fairly be said that S&I had the means "of monitoring or securing compliance" by the suppliers, and thus that even if the decision in Bulves sets a higher test for the State, in this case that test is met. (Thus even if we are wrong in our second conclusion in relation to Bulves (see I(d) Bulves above) it makes no difference in this case.)
    220. Third, the steps we believe reasonable were not suggested to S&I at the time, or put to Mr Ashraf or Tariq Mohammed at the hearing. Neither were they suggested by the Respondents. We cannot see that that makes any difference.
    VIII Overall conclusions
    221. A summary of our conclusion in relation to each of the transactions impugned is set out in Appendix III.
    222. In that Appendix we indicate our conclusion in relation to each deal chain as to whether connection and fraud are proved; and where fraud is proved we indicate whether it is proved in relation to all the VAT in respect of the fraudster's sale or the VAT on its margin only.
    223. In respect of any deal in which 'no' appears in the final column our conclusion is that S&I's related input VAT is not disallowed under the Kittel principle and is thus creditable.
    223. In respect of any deal in which 'all' appears in the final column our conclusion is that all the VAT evaded by the fraudster is not deductible or creditable by S&I but that the VAT paid by buffers in the intermediate stages between the fraudster and the S&I is creditable.
    224. In respect of any deal in which 'margin only' appears in the final column our conclusion is that the VAT evaded by the fraudster – the amount equal to the VAT on its output less the creditable input tax – is not creditable to S&I but that the remainder of its input tax is creditable.
    225. In respect of any deal in which a number of those words or phrases appear in that column by reference to different alleged fraudsters or number of phones, our conclusions in relation to the input tax creditable for S&I is that it is the amount which results from the application of the conclusions above to each element of the relevant deal.
    226. We have not calculated the total allowable and non-allowable tax arising as a result of these conclusions. We adjourn the appeal for the parties to agree that figure or in default of agreement return to the tribunal.
    227. Given the absence of argument before us in relation to the Bulve decision we also give leave for either party to apply to advance oral argument for a reference to the European Court, but any such application must be made within 14 days of the release of this decision.
    Costs
  280. In the majority of deals we have found that a substantial proportion of S&I's input tax is to be denied as claimed by the Commissioners. We therefore award 80% of the Commissioners' costs to be paid by the Appellant, such costs to be assessed if not agreed by a taxing Master of the Supreme Court or a district judge of the High Court of Justice by way of a detailed assessment, and to be paid within 14 days after such assessment. However in the determination of those costs (i) costs awarded to the Appellant in interlocutory matters should be deducted, and (ii) any costs associated with the redaction of documents put before the tribunal and with the later supply of unredacted documents shall be excluded and so be borne solely by the Respondents.
  281. Our conclusions were unanimous.
  282. APPENDICES
    APPENDIX 1
    The Deal Chains
    We set out below our conclusions in relation to the alleged connection of S&I's acquisition of phones to a sale by an alleged fraudulent trader in each Deal Sheet. We have not given a full description of the chains and have in some cases abbreviated the names of relevant parties in a manner which we hope is clear to those in possession of the Deal Sheets.
    Deal 1 We found the alleged connection to have been established.
    Deal 2 The evidence for two steps in this chain derived from Spreadsheets, one of which, Mr Cameron Watson told us, had been compiled by another unnamed officer from Oracle's records, and the other, Mr Ross told us, had been compiled by another officer from Electron's records. No corroboration of the spreadsheets was available. We did not find the connection proved.
    Deal 3 The evidence for one step in this chain was confined to a spreadsheet prepared by New Way Associates. Other documentation corroborated other elements of that spreadsheet. We find the alleged connection to have been established.
    Deal 4 The deal sheet traced this transaction back two steps from S&I to Mana Enterprises. There was no evidence that Mana Enterprises was a defaulter. HMRC argued that having regard to the other members of the chain it is likely that this deal would be traced back to a fraudulent defaulter. We were not convinced that this was more likely than not. We do not find a link to an alleged fraudster has been proved.
    Deal 5 We find the alleged connection to have been established. As with Deal 3 part of the evidence for this chain relied upon a New Way spreadsheet corroborated by other documentation.
    Deal 6 We find this alleged connection to have been established.
    Deal 7 The transactions alleged in this case consisted of a chain of 500 Nokia N70s and a chain of 1000 Nokia N70s coming together to the Appellant and its onward sale of 1500 of those phones. One of the steps was supported only by a spreadsheet but other documentary evidence supported that spreadsheet. There was no direct evidence that the 500 plus 1000 phones acquired by the Appellant were the same as the 1500 it sold; but the dates and prices for the transactions convinced us that on balance they were the same phones. We find the connection established.
    Deal 8 We find the connection (in this case to the alleged contra trader) established. We comment in section V about the contra chain alleged to be associated with this chain.
    Deal 9 This was effectively two chains: one of 1200 Nokia 6280s and one of 3000 Nokia N70s. We found that the documentary evidence before us supported the connection in relation to the 1200 Nokia 6280s. But there was no documentary evidence before us in relation to an alleged sale by I Connect U Ltd to Aflecks Phones of 3000 Nokia N70s (although there was evidence of the transaction between these two parties in relation to the 1200 Nokia 6280s) other than a spreadsheet showing a purchase and sale by I Connect U of these phones; the spreadsheet however did not indicate I Connect U sold the phones to Afflecks. However (i) the only customer identified on the spreadsheet was Afflecks, and (ii) the 1200 Nokia 6280s were documented on the spreadsheet (again without mentioning Afflecks as the customer but providing corroboration for the spreadsheet). Taking the spread together with Mr Ross' evidence that the tracing back had been made, on balance we find that the Nokia N70 chain has been proved.

    In this case we did look at the evidence of transactions after the Appellant's sales. This established to our satisfaction that the 6280s had been originally imported from Mingele for £194.30 each and after eight sales (most of which were on the same day) had been purchased by Mingele at £215 each.
    Deal 10 This involved two alleged chains, one of 390 phones and the other of 800 phones. The chains came together in New Way which acquired a total of 1190 phones and sold 1100 to the Appellant The deal schedule showed New Way as acquiring 300 + 800 phones in circumstances where Diginett (its predecessor in the alleged chain) had acquired 390 and sold only 300 to New Way. The documentation before us, however, indicated instead that 390 phones had been bought and sold by Diginett so that New Way acquired 390 + 800 = 1190 phones and sold 1100 to S&I. The Appellant's export to which the Deal schedule linked these acquisitions was an export of 1100 phones.
    The documentary evidence for the chains relied upon the New Way schedule already referred to and a schedule for Diginett compiled by an officer of HMRC which was corroborated by one piece of documentary evidence.

    In relation to the tracing of 390 phones to the Appellant we find the relevant connection established, but in relation to the 800 phone chain there was no documentary evidence before us in relation to the purchase or sale by Aflecks Phones (one of the earlier parties in the alleged chain).

    There was no explanation in the evidence before us which explained how the 390 + 800 = 1190 phones acquired by New Way had become the 1100 phones exported by it - i.e. no evidence as to what had happened to the other 90 phones other than the NewWay Schedule established in relation to Deal 16 which indicated to whom the balance of 90 had been sold by NewWay. However, whilst we accept that the connection is established in relation to 390 phones the absence of the documentary evidence in relation to the Aflecks sale of 800 phones leads us to conclude that to conclude that the connection is not proved in relation to those 800 phones.
    Deal 11 There was one transaction in this chain for which there was no documentary evidence before us: the sale of 1500 phones from Fonestop to London Mobile. In his oral evidence Mr Ross said that he had seen evidence that these goods were obtained by London Mobile from Fonestop.

    The number of phones said to have been purchased by London Mobile was 1500 but the number of phones the documentation (and the deal sheet) shows London Mobile as selling is 1400. That suggests a possibility that there were two deal chains, one for 1400 and one for 1500, which have somehow become mixed, or that the Fonestop sale of 1500 phones was not to London Mobile as alleged. Our impression of Mr Ross was that he was diligent and careful. On balance we accept that there was a sale of 1500 phones to London Mobile as alleged, and that London Mobile sold 1400 of them to S&I.

    The documentary evidence in relation to one step in the chain was confined to a (corroborated) spreadsheet. We accept that evidence.Mr Beal notes that the invoices for the two earliest transactions in the chain are dated 13 April 2006, whereas the later 4 transactions (ending with the Appellant's export) are dated 11 April 2006. In our view it was quite possible that an invoice may have been sent (and dated) after the sale was made, and therefore we concluded it was not impossible that the sales were connected.

    We find that the connection is proved.
    Deal 12 The documentary evidence supported the links in the alleged chain save for the sale by ETel to Lexus. However there was evidence of sale to ETel of 1840 phones on 31 March and the sale of 1550 of the same phones by Lexus to S&I on or before 4 April. Whilst there was no copy of ETel's invoice to Lexus in our bundles we were on this occasion satisfied with Mr Ross' evidence that the connection had been traced through the invoice he lists from ETel to Lexus. We conclude that the connection is proved.
    Deal 13 The Deal Sheet for this chain indicated a sale by East Midlands Engineering Supplies to MG Components and an onward sale by MG Components of 440 "Sony W800i". The remainder of the chain related to Sony W800i. However the documentary evidence for those first two sales referred to "Sony W900i", whereas the later invoices referred to "Sony W800i". Mr Ross indicated that it was possible that Sony W900i's could have been sold. Mr Beale contended therefore that the alleged chain had not been established. We noted however that the purchase order to MG Components from its purchasers had been for 440 Sony W800i and had the same date, number and price as the MG Components invoice to that purchaser. We concluded that it was likely that East Midlands and MG Components had made typographical errors on their invoices, one prompted by the other's invoice.

    The documentary evidence for one step in the chain was limited to a (corroborated) New Way spreadsheet.

    We conclude on balance that the connection is established.
    Deal 14 This involved two alleged chains, one of 625 phones and the other of 450, culminating in sales to the Appellant, who exported 1075 phones.

    In relation to the 450 phone chain, the deal sheet showed it commencing with RK Brothers. Whilst we accept the chain back to RK Brothers, there was no assertion that RK Brothers had fraudulently defaulted on the payment of VAT.

    In relation to the 625 phone chain, we saw documentary evidence (including a spreadsheet relating to Topbrandz) which in our opinion indicated a chain back to Fima Consulting Ltd, but no evidence of a purchase (as alleged) by Fima from Bodytech. There was however acceptable evidence of purchases by Bodytech from Bond Corporation, and by Bond Corporation from the alleged defaulter, Data Solutions Northern. Thus, save in relation to the Fima purchase we were satisfied that the 625 phone chain was established. In response to questions from the tribunal (arising from its examination of the evidence before the end of the hearing but detailed later), HMRC pointed to evidence of this sale in a spreadsheet which, Mr Ross had told us related to Bodytech but without comment on its provenance. This and the evidence of Mr Ross that the chain had been traced back (including the details of the invoice for the Fima purchase on the spreadsheet), the pricing and dates of the invoices, and the number of phones, convinced us, on balance, that the 625 phone chain took place as alleged, and that the relevant connection was proved.
    Deal 15 The documentary evidence for this chain did not include anything in relation to two sequential steps : the sales from Seymour Selection to J D Group, and from J D Group to the Export Company. Whilst we had Mr Ross' general evidence of timing and relevant details on the sheet, overall we did not conclude that this chain had been proved. Whilst, where documentary evidence of one step is missing, surrounding documentation may bolster the evidence of Mr Ross, in a case where there was no documentary evidence of two steps we found the surrounding evidence less compelling. We did not find the relevant connection established.
    Deal 16 The documentary evidence (which included the New Way schedule, a corroborated fonedealers schedule, and an Emmen schedule corroborated by the fonedealers schedule) for this chain provided evidence of all the steps. Mr Ross' evidence of tracing, and the pricing, dates and number of phones on the other invoices and schedules were enough to convince us, on balance, that the chain took place as alleged. We find the relevant connection is established.
    Deal 17 There was documentary evidence before us for each transaction in this alleged chain. The evidence for one step, the sale by Cobra Communications to Letting Solutions, was confined to a spreadsheet in relation to Cobra's purchases and sales. Whilst the other documentary evidence related to 1000 phones, the entry on this spreadsheet was in relation to the purchase and sale of 2000 phones. However, we saw a copy of a purchase order from Cobra for 1000 (only) of the relevant phones. We concluded that the entry of 2000 on the spreadsheet was a typographical error, and find that the connection was proved.
    Deal 18 The documentary evidence (including a corroborated spreadsheet) supported the alleged chain. We find the relevant connection to have been proved.
    Deal 19 The documentary evidence (including two (corroborated) spreadsheets) supported the alleged chain. We find the relevant connection to have been proved.
    Deal 20 The documentary evidence (including a corroborated spreadsheet) supported each link in the chain apart from a sale from Signal Telecom to DVB Ltd. Mr Ross' express evidence was that he must have seen the documentation for this step somewhere : he didn't just make it up. We find the connection to have been proved.
    Deal 21 This involved two alleged chains each involving the sale of 1000 phones. One chain is shown on the Commissioners' Deal Sheet as starting with Princeways, the other with UA Distribution, an alleged defaulter. The two chains join in New Way which thus received 2000 phones. Of those 2000, it sold 1000 to the Appellant and 1000 to other persons. The documentary evidence (including a corroborated spreadsheet) supported the transactions on the Commissioners' Deal Sheet.

    The NewWay schedule indicated that 500 of the phones it acquired from Princeways and 500 of the phones it acquired in the chain commencing with UA Distribution Ltd were sold to S&I.

    We find that a connection as to 500 phones is established back to UA Distribution, and of 500 phones to Princeways.

    Princeways was not alleged to be a defaulter. Mr Ross said that in other deals at this period in which Princeways was concerned, its purchases led back to UA Distribution. We do not doubt his recollection but find it insufficient for us to conclude that the 500 Princeways phones trace back to UA Distribution. Thus so far as those phones are concerned we do not find the connection to an alleged defaulter to have been established.
    Deal 22&23 This deal involved three alleged chains of sales of Nokia 6230i's.

    The first chain was for 2000 phones. It commenced with the sale by the alleged defaulter, UA Distribution, to USM IT. The second chain was for 500 phones starting with a sale by Aflecks Phones, and the third chain for 2953 + 300 phones starting with USM IT, an alleged defaulter. It was not suggested that Aflecks was a defaulter. (We say "+300" because the Deal Sheet did not reflect 300 phones which appeared in the copy documentation.)

    The deal chains came together with sales to New Way which thereby acquired 5453 +300 phones of which 5300 were sold to S&I who exported them, 153 were sold to Phone Shop and 300 sold to another party. New Way's schedule shows that the 5300 phones sold to S&I were made up of:

         2,953     from the third chain
         347     from the 500 phone chain
         2,000     from the first chain
         5,300

    The evidence for the first 2000 phone chain included a schedule of Fonedealers' purchases and sales corroborated by other documentary evidence, but there was no evidence of an alleged sale in chain by Electron global to Mana Enterprises. Evidence for a sale by Mana Enterprises was in the form of New Ways' spreadsheet. Mr Ross gave no evidence directly relating to the tracing of this part of the chain. We do not find the chain of 2000 phones to have been proved.

    The evidence for the chain of 500 phones included two corroborated spreadsheets but there was no evidence as regards an alleged sale by Emmen to Mana Enterprises and Mr Ross gave no direct evidence in this chain. The chain as alleged traced back to Aflecks which was not alleged to have been a defaulter. We do not find the chain of 500 phones to have been proved to trace back to an alleged defaulter.

    The evidence for the chain of 2953 +300 phones included two corroborated spreadsheets and we accept that a chain of sales starting with USM IT was established. However the spreadsheet showing the sale by USM IT to Fonedealers and its onward sale carried the legend against those transaction "please note this is a cancelled deal." This was a spreadsheet prepared by one of the Respondents' officers. On balance we find that the evidence of the later sales is sufficient to conclude that this inscription was wrong and we find the 2953 +300 chain proved, and the relevant connection established.
    Deal 24 The evidence for this chain included two corroborated spreadsheets. We find the connection proved.
    Deal 25 This deal involved two alleged chains, one starting with a sale by B

    Logistics UK of 500 Nokia 6111s, and the second with a sale by UA



    Distribution of 700 Nokia N90s. The chains came together in sales to S&I who exported the phones it had purchased to Sunico A/S. In the Nokia 6111 chain, Ilford Cellular purchased 500 phones but sold only 493 to S&I; the phones were invoiced to Ilford on 20 March (the same date as prior invoices in the chain) but invoiced by Ilford on 26 April. .

    The change in the invoicing date in relation to the Nokia 6111s and the change from 500 to 493 phones could indicate that Ilford Cellular supplied the 493 phones from a purchase different from that alleged. Ilford Cellular bought the 500 phones for £155.00 each; it sold the 493 for £138.00 each (S&I sold them for £153.00 each). That might also suggest that the phones sold to S&I by Ilford Cellular were differently sourced.

    However although a reduction in the number rather than an increase was possibly indicative of some spillage, and the date and price may well have indicated some haggling, we conclude the connection is not proved in the case of the 493 phones.

    We find that the chains are proved and the connection established in relation to the 700 phones, but not in relation to the 500 (or 493) phones.
    Deal 26 We find the alleged connection to have been proved.
    Deal 27 We find the alleged connection to have been proved.
    Deal 28 We find the alleged connection to have been proved.
    Deal 29 We find the alleged connection to have been proved.
    Deal 30 We find the alleged connection to a sale by Park Supplies have been proved. Park Supplies were not alleged by the Respondents to be a defaulter. Mr Ross said that the chances were that Park Supplies purchased from SS Enterprises but proferred no other evidence of that connection. We do not find a connection to SS Enterprise to have been proved.
    Deal 31 We find the alleged connection to have been proved.
    Deal 32 Two chains of sales were alleged - one of 3000 Nokia 7610s and the other of 1295 Nokia 7610s. The chains came together in New Way who sold 2,100 of the 4295 phones it had acquired to S&I.

    We find the alleged connection to have been proved.

    There were three features of these chains which struck us as odd and suggested an element of contrivance. First, each involved exactly the same parties. Second, that Diginett, having acquired 3000 phones from Emmen then invoiced them in two tranches, 1000 and 2000 to New Way on the same day. Third, that Diginett invoiced the 1000 and 2000 phones at £115 each, but the 1295 phones from the other chain at £117 each, all on the same day.
    Deal 33 We find the alleged connection to have been proved.
    Deal 34 We find the alleged connection to have been proved.
    Deal 35 We find the alleged connection to have been proved.
    Deal 36 This comprised two chains. The first chain related to 1000 Samsung D800s. We find that chain proved and the connection established. The second chain related to 1000 Nokia N70s. There was documentary evidence supporting each link in that chain bar a sale from Ocean Connection to DVB.

    Mr Ross explained that DVB had schedules and the schedules showed purchases of both 1000 Samsung D800 and 1050 Nokia N70s from Ocean Connection, but when he had asked DVB for the invoices only the invoice for the D800s was produced.

    In the papers before us there was a purchase order from DVB to Ocean Connection for 1000 Nokia N70s dated two days before the invoice date alleged in the deal schedule. The order specified a price of £198.50 per phone, the deal schedule showed an invoice price of £189.00.

    We concluded that it was likely that the agreed price had changed before the deal was completed and that there had been a sale as alleged. We therefore find the Nokia N70 connection also to have been proved.
    Deal 37 We find the alleged connection to have been proved.
    Deal 38 We find the alleged connection to have been proved.
    Deal 39 We find the alleged connection to have been proved.
    Deal 40 We find the alleged connection to have been proved.
    Deal 41 The deal sheet alleged a chain relating to 500 Nokia 91s starting with Bullfinch Systems going through four intermediary dealers and then to S&I.

    Purchase order and invoice documentation supported the alleged chain.

    There were also copy release notes before us, in which instructions were given by one party to a freight forwarder to release the goods to another person. These release notes in the case of the first three intermediaries in the chain were addressed to Turners Forwarding but the final intermediaries' release note was addressed to Hawks. Mr Beal suggested that two sets of phones had been confused and that the goods at Turners were destined for a person other than S&I so that despite the chain of invoices and purchase orders, the 500 phones eventually sold by S&I had not derived from Bullfinch but from a different source. Mr Ross suggested that the goods may have been moved from Turners to Hawks by Why Systems (S&I's supplier) or by Transglobal Trading (Why's supplier). Transglobal, on the invoice values in the deal schedule, made a profit of 25p per phone or £125 in total on its deal, and Why made £1 per phone or £500 in total; but the cost of moving the goods to a different freight forwarder we believe could well be less than £500.

    We concluded that the connection was proved.
    Deal 42 We find the alleged connection to have been proved. We also find it likely that there was a carousel in relation to 2000 of the phones. (Mr Beal pointed to a discrepancy in the number in this regard which found explicable).
    Deal 43 We find the alleged connection to have been proved.
    Deal 44 We find the alleged connection to have been proved.
    Deal 45 The Respondents alleged two chains of sales and purchases: one of 2000 Samsung D600s starting with USM IT, the other of 1000 of the same phones starting with LTH Ltd. The chains came together with a sale by New Way to S&I of 500 of the 2000 phones in the first chain, and a sale by Future Communications to S&I of the whole of the 1000 phones in the second chain. S&I then exported all 1500 phones it had received.

    In the first chain the sales shown on the deal sheet we re:

         Seller     No. of phones sold
         USM IT     2000
         Horas Art Foundation     2000
         Fonedealers     2000
         Electron Global     2000
         Diginett     500
         New Way     500 (sold to S&I)


    However the documentation before us showed a sale by Diginett of 3,500 of these phones to New Way. Thus Diginett had acquired an additional 1,500 phones. There was nothing to indicate whether the 500 phones sold by New Way came from the 2000 or the 1500. Accordingly we could not find that it was more likely than not that they came from the 2000.

    Thus as regards the 2000 chain we do not find it proved. We were also unable to find that there was a scheme pursuant to which USM IT purchased, sold and defaulted in respect of 2000 phones and S&I purchased 500 phones.

    The documentation before us supported the second chain of 1000 phones. There was one discrepancy in relation to the freight forwarder specified: for all the sales other than one between Cobra and Letting Solutions the freight forwarder (at whose premises the phones were said to be held) was Hawks; for that one it was Pauls Freight. The possible explanations are (i) that another chain has been muddled with this one (ii) there was a movement of the goods, or (iii) that there was a clerical error in completing a standard form. Given the other evidence of and aspects of the chain we think it more likely that it was a clerical error. We find the chain of 1000 and the connection proved.
    Deal 46 S&I sold 2000 phones, 200 of which we find it bought from Princeways (which was not alleged to be a defaulter), and 1800 of which we find derived from a chain of purchases and sales going back to the alleged defaulter SS Enterprises GB.
    Deal 47 We accept Mr Ross' evidence that this connection has been established.
    Deal 48 We accept Mr Ross' evidence that this connection has been established.
    Deal 49 We accept Mr Ross' evidence that this connection has been established. We note that the members of the chain are the same and in the same order as Deal 47.

    We also note that in this chain and Deal 47, S&I buys from Diginett. In deal 45 S&I buys from New Way who in turn buys from Diginett. S&I therefore knew and dealt with at least one of New Way's suppliers.
    Deal 50 We accept that the connection has been established. Mr Beal pointed to inconsistent information in the documentation as to the location of the phones: they appear at the start of the chain to be at Hawks and then later, appear to be found at Point of Logistics. On examining the documents we concluded that it was likely that the goods did not go to Point of Logistics (whereas a purchase order says that the goods should be shipped to Point of Logistics the corresponding invoice says ship via Hawks). We therefore accept this chain and find the connection proved.
    Deal 51 Mr Beal says that there is no evidence of the sale by the alleged defaulter, Cyberweb, at the start of this chain save a schedule prepared by an unidentified HMRC officer in relation to the alleged purchaser from Cyberweb, which was Horus Art. We found that the same schedule recorded a purchase order a copy of which was in evidence in relation to Deal 49: that gave us confidence in the schedule. That the schedule refers to "De Souza" but we accept that this is a mistaken shortening of Eduardo D'Souza, who ran Horus Art Foundation. We find the connection established.
    Deal 52 Mr Ross produced a schedule prepared by another officer as the only documentary evidence of one step in the alleged chain. There was no corroboration of the schedule. However, we accept Mr Ross' evidence that S&I's sale can be traced back through the chain to the alleged defaulter, and find the connection proved.
    Deal 53 Again Mr Ross produced an uncorroborated schedule prepared by another officer as evidence of one step in this alleged chain. The schedule however declared the source of its primary data. We accept Mr Ross' evidence that S&I's sale can be traced back to the alleged defaulter. We find the connection proved.
    Deal 54 We accept Mr Ross' evidence that S&I's acquisition can be traced back to the alleged defaulter. We find the connection proved.
    Deal 55&56 The alleged chain started with the sale of 2000 phones by Cyberweb. Four transactions later, 2000 phones were acquired by New Way, and New Way sold 1300 to S&I, who sold them to two different purchasers. In other chains where New Way was involved and in which it split or amalgamated acquisitions, we were able to trace the phones through a schedule prepared by New Way. No such schedule was exhibited for this split. That exposed the possibility that New Way had acquired these 1300 from a different source. However, the dates and prices of the transaction convinced us that on balance the 1300 came from the 2000. We therefore find those chain established, and the connections proved.
    Deal 57 We accept Mr Ross' evidence that S&I's acquisition can be traced back to the alleged defaulter. We find the connection proved.
    Deal 58 We do not find that it is proved that S&I's acquisition traces back to an identified defaulting trader.

    There was evidence that S&I's acquisition could be traced back to Shelford Trading, but it was not alleged that it was a defaulter. Mr Ross relied on some documentary evidence to link the phones back to an import of phones into the UK from a Cypriot company but could not identify vendors and purchasers later in the chain. The mere fact of an original import does not to our minds provide evidence that there was a defaulter in the chain: the VAT may have been paid. Neither could we find a satisfactory connection between the documented import and Shelford's sale.
    Deal 59 We accept Mr Ross' evidence that S&I's acquisition can be traced back to the alleged defaulter. We find the connection proved.
    Deal 60 Mr Beal noted that the documentation before us contained a purchase order which did not match the exhibited invoice. It appeared to be for a different sale. Its presence in the documentation did not prevent us relying upon the sale invoice for the particular step.

    Mr Beal also highlighted that the documentation indicated that the goods were shipped from the UK on 26 May but arrived on 30 May. This concern related to issues after S&I's sale and did not affect our consideration of the UK chain.

    We find the alleged connection to have been proved.
    Deal 61 The documentation before us was evidence that there were two chains, one of 3000 and one of 2500 phones, each of which in the UK started with Cyberweb. The chains culminate in acquisitions by New Way of 3000 phones from the first chain and 1500 from the second, New Way in turn sold 200 phones to S&I. The following diagram illustrates the alleged later steps in the chain:-

    Image 1

    The sales by Diginett and New Way were invoiced on 16 May 2006; the other sales between 11 and 13 May. New Way's sale price to S&I is uncharacteristically below its acquisition prices from Princeways and Diginett.

    Mr Ross told us that he relied upon a New Way spreadsheet (of the type we had already seen) which was not in our bundle to link the 200 phones sold to the Princeways and Diginett purchases. There was no evidence before us in relation to the split by Diginett, or as to whether or not Diginett may also have acquired a separate consignment of 1500 phones otherwise than from Electron Global. The difference between the dates of Diginett's disposal and acquisition dod not point to a connection.

    Whilst we accept Mr Ross' evidence of the existence of a New Way spreadsheet, we heard and saw nothing to indicate which of the 200 phones sold by S&I came from Princeways and which from Diginett and we were not convinced that Diginett had split the purchase as alleged.

    On balance we did not find the connection proved.
    Deal 62 We accept Mr Ross' evidence, including his reliance upon an undisclosed New Way spreadsheet, that S&I's acquisition can be traced back to the alleged defaulter.
    Deal 63 We accept that S&I's sale can be traced back to the alleged defaulter. We were not dissuaded form this conclusion by some discrepancies in the dating of some of the documents. The trail was in our view sufficiently clear.
    Deal 64 . The documentary evidence established to our satisfaction the transactions of purchase and sale on the Respondents' deal sheet. These showed the invoiced sales by the 7 parties in the chain, of 1000 Nokia N91s all on 25 May 2006.

    One aspect of the chain troubled us. Regal Portfolio bought the phones at £387.50 and sold phones at £338.00 making a loss of £49.75 per phone or £49,750 in total. We find it extraordinary that in transactions concluded back to back (as we assume they must have been) on the same day that such a loss would be made. That suggest that Regal may have been involved in another chain of 1000 Nokia N91s and that there is some mix up between the chains such that S&I's phones do not trace back to the alleged defaulter but to some other person.On balance however we find that S&I's acquisition does trace back to the alleged defaulter.
    Deal 65 (The Deal Sheet contained a typo: Diginett's sale was in voiced on 25 May not 2 May) We accept Mr Ross' evidence that S&I's acquisition can be traced back to the alleged defaulter. Mr Beal's; specific concerns related to transactions after S&I's sale.
    Deal 66 The Respondents' deal sheet alleged a chain of 7 sales starting with Crossview Consortium, the alleged defaulter, and finishing with S&I (after which S&I exported the phones). The first 6 sales were of 2232 phones after which point New Way Associates, having acquired 2232, sold 1000 to S&I. The assumption in the Deal Sheet is that the 1000 phones sold to S&I were sourced from those 2232.

    In other deals involving New Way, a schedule prepared by New Way showing the source and split of various transactions was part of the documentary evidence before us. There was none exhibited in this case. There was no oral evidence (in chief, cross examination or re-examination) from Mr Ross to indicate whether or not there was such a schedule. An attested copy schedule is one thing; a witness' recollection of the existence of one is another; simply assuming it might exist is yet different still.

    However, the dates of the invoices and the prices of the phones suggest that the 1000 phone sale was part of the 2232 phone chain.

    On balance we are satisfied that S&I's acquisition can be traced back to the sale by the alleged defaulting trader.
    Deal 67 The alleged chain started with the sale of 1900 Samsung D500s on 1 June 2006. All the other sales in the chain, down to the sale to S&I were invoiced on the same day.

    We found that the documentary evidence enabled us to conclude that each of the transactions in the alleged chain took place. Mr Beal noted that a schedule (of purchases and sales by 4A Developments) was relied upon which failed adequately to identify the model of phone being sold. The spreadsheet entry was however corroborated by a 4A Developments sales invoice which did specify the model. We were satisfied with that aspect of the evidence.

    However there were two aspects of the chain which gave us concern. First, whereas the early sales in the chain had been of 1900 phones the last two sales (by PLC Communications to New Way, and to S&I by New Way) were of 1950 phones. There was no indication of the source of the additional 50 phones, and nothing, other than dates and prices, to dispel a suspicion that the 1950 phones might have been differently sourced by PLC, with its 1900 being sold elsewhere. Also PLC acquired the 1900 for £97 each and (on the same day, back to back it is effectively alleged) sold them at a loss for £95. In the absence of PLC's loss we would have been inclined to find that the dates of the transactions and general level of prices indicated the requisite connection, but the loss weighed heavily against that. In addition Mr Beal pointed to an apparent change in the freight forwarder between the beginning and end of the chain which for a chain with small margins invoiced on the same day seemed odd.

    We do not find that the connection was proved. We were unable to otherwise conclude that there was a scheme pursuant to which DBP Trading purchased, sold and defaulted in respect of 1900 phones and S&I purchased 1950 phones.
    Deal 68 We accept Mr Ross' evidence that S&I's acquisition can be traced back to the alleged defaulting trader.
    Deal 69 We find S&I's acquisition can be traced back to the alleged defaulter. (&72)
    Deal 70 The documentary evidence before us supported the transactions in HMRC's deal sheet. We had two concerns:-

    (i) the documentary evidence for the first sale in the chain was a spreadsheet which Mr Ross said was evidence of the primary documentation. We were not told who had produced it; and(ii) Mana Enterprises bought 1750 phones but sold only 900 to New Way. There was no evidence other than the (same) date of those two transactions and their similar price to point to the 900 being sourced from the 1750

    New Way bought 900 phones and sold 500. There was a New Way schedule to link the purchase and sale.

    Overall we concluded that this chain was proved.
    Deal 71 This chain started with sales of 1788 phones which eventually were bought by Lexus. Lexus then sold 700 to S&I. The dates and prices convinced us that the 700 were sourced from the 1788. We find that S&I's acquisition of phones can be traced back to the alleged defaulting trader.
    Deal 72 We find that S&I's acquisition can be traced back to the alleged (see 69) defaulter.
    Deal 73 The documentary evidence satisfied us that the sales specified on the deal sheet took place. The first five sales in the chain were each of 3000 phones and were all invoiced on 1 June 2006, the last such sale being to Lexus Telecom. Lexus Telecom sold 1408 phones to S&I on 22 June. It bought the phones at £82 each and sold them at £79 each. The difference in dates and numbers and the price differential indicated a possibility that Lexus had sourced the 1408 otherwise than from the 3000, but the documentation indicated that the goods, rather than being held at a freight forwarder, had been delivered to Lexus and then later collected from Lexus for S&I.

    We find that the phones acquired by S&I can be traced back to the alleged defaulter.
    Deal 74 We find that the phones acquired by S&I can be traced back to the alleged defaulter. In so finding we have accepted a spreadsheet which Mr Ross indicated was prepared by an HMRC officer.
    Deal 75 We find that the phones acquired by S&I can be traced back to the alleged defaulter.
    Deal 76 The chain alleged by HMRC involved 2000 phones but there was documentation in the bundle which related to the transfer of 3000 of the same phones between the same parties on dates close to the dates of the alleged transactions. We concluded that the documents relating to the 2000 phones were satisfactory evidence of the alleged chain and that the documents relating to 3000 phones related to a separate chain. We find that S&I's acquisition can be traced back to the alleged defaulter.
    Deal 77 We found that S&I's acquisition can be traced back to the alleged defaulter. In so doing we relied upon copies of invoices for all steps in the chain other than the first. There was no invoice or purchase order relating to the first transaction but its happening was confirmed to our satisfaction by documents from a freight forwarder which were corroborated by the copy invoices we saw.
    Deal 78 The Respondents allege a chain in which 2000 phones were sold in three transactions, the last of which was a sale by Jos (UK) to New Order Trading which then sold 1000 phones to New Way. It in turn sold 700 phones to S&I.

    There was no documentation before us for the sale by Jos (UK) to New Order. There was however a document from New Order to Hawk, a freight forwarder, indicating that 2000 phones had been allocated to New Order and that they should be made available for inspection by New Way. It was possible that either (i) New Way purchased all or none of 2000 of these phones but that there was a second separate sale of 1000 phones to New Way or (iii) that in the end New Way purchased only 1000 of these 2000 phones. If the former is the case the chain to S&I is not established.

    This uncertainty caused us to conclude that the chain from S&I to the alleged defaulter had not been proved. Neither were we able to conclude that there was a scheme pursuant to which Heathrow Business Solutions purchased, sold and defaulted in respect of 2000 phones and S&I purchased 700 phones.
    Deal 79 We find thag sold to S&I.
    Deal 80 We find that S&I's acquisition can be traced back to the alleged defaulter.
    Deal 81 We find that S&I's acquisition can be traced back to the alleged defaulter.

    The Deal Sheet indicated that this deal involved a carousel: the phones starting with Sunico on 6 July and finding their way to S&I and then being exported back to Sunico on 10 July. Mr Beal noted that the dates of the release notes offered as evidence of the beginning of this chain were faxed on anomalous dates. We did not find the carousel proved.
    Deal 82 We find that S&I's acquisition can be traced back to the alleged defaulter.

    We were not dissuaded from this conclusion by some relatively small discrepancies in some of the documentation.

    As with Deal 81 the Deal Sheets indicated a carousel starting and ending with Sunico. The documentation did not evidence the alleged initial sale. We heard no direct evidence from Mr Ross on the issue. We do not find the carousel proved.
    Deal 83 The Respondents alleged two chains of supply of 1000 and 1030 phones culminating in sales to Lexus Telecom of 2030 in total, of which Lexus & 84 Telecom sold 1830 to S&I which exported them in two sales.

    We accept that the chain from the alleged defaulter to Lexus was established.

    The sale of the 1830 phones by Lexus took place 14 days after this purchase and at lower price per phone than Lexus had purchased the phones. It was therefore possible that Lexus had sourced these phones elsewhere.

    On balance we concluded that it was not proved that S&I's purchase traced back to the alleged defaulter or was otherwise connected to the alleged defaulter's default.
    Deal 85 We find that S&I's acquisition traces back to the alleged defaulter.
    Deal 86 We find that S&I's acquisition traces back to the alleged defaulter.
    Deal 87 We find that S&I's acquisition traces back to the alleged defaulter.

    We also find that in this Deal there was a carousel.
    Deal 88 We find that S&I's acquisition traces back to the alleged defaulter.
    Deal 89 We accept Mr Ross' evidence (including his reliance on a spreadsheet) that S&I's acquisition traces back to the alleged defaulter.
    Deal 90 We find that S&I's acquisition traces back to the alleged defaulter. We noted that Sunico A/S sold 500 phones to Silus BV who sold to Phone City. On 24 July 2006 S&I sold 499 phones to Sunico A/S. There was in this something akin to a carousel clouded by the addition of 1000 phones by the first UK trader in the chain
    APPENDIX 2
    The 'Defaulters'
    The Alleged Defaulters
    In relation to each alleged defaulter we set out in the headings of each section below the Deals (the 'specified deals') in which that company (they are all companies) appears as the defaulter in the Deal Sheets. We use the term 'import' to include the technical term 'acquisition'. We use the term 'Regulation 25 notice' to refer to a notice given to a trader under regulation 25 of the VAT General Regulations requiring the completion of a VAT return to a date earlier than the normal date.
    The Commissioners' case was, broadly, that the first entity shown in the chain on those sheets, or if different, the first UK entity between which and S&I there was in the chain no non-UK entity was a defaulter. Furthermore they say that either there was a default in relation to all the VAT on that entity's sale (because there was no relevant input tax to set against it since its purchase was by import) or that there was an earlier unidentified defaulter in the chain before the first disclosed entity. We set out our conclusions in relation to these contentions as well.
  283. East Midlands Engineering Services ("EMES") (Deals 1, 3, 6, 13)
  284. Mrs Mee told us that in April 2006 she received from another branch of HMRC information that EMES had opened an account with a freight forwarder and had sold mobile phones. As a result she and another officer made a visit to EMES. They were told by Mr Goy, a director, that the company's business was distributing plumbing equipment and that it did not trade in mobile phones.
    In May 2006, HMRC obtained from European Wireless, invoices which appeared to have been issued to it by EMES for mobile phones. We accept that such phones had been supplied to European Wirelesss. European Wireless also provided HMRC with copies of EMES' VAT certificate and utility bills, and a copy of Mr Goy's passport. The director of European Wireless also said that he had met Mr Goy in a pub in Nottingham. Mrs Mee visited EMES again on receipt of this information and interviewed Mr Goy and the two other persons involved in the management and administration of the business. Mr Goy and his colleagues denied any involvement in mobile phone sales and expressed concern that someone else had a photocopy of Mr Goy's passport.
    Mrs Mee said that Mr Goy still maintains that EMES has not done any mobile phone deals. EMES has not been assessed with the VAT attributable to the sales made in its name. Mr Ross told us that none of the phone invoices had been entered on EMES' VAT return. We accept that.
    We accept Mrs Mee's evidence. We conclude that either another person "hijacked EMES' VAT registration" (i.e. used its name and VAT number on invoices dishonestly raised in its name) with the intention of evading the payment of VAT on them, or that EMES did in fact make these sales and deliver the invoices but did so intending dishonestly to evade payment of the VAT attributable to them.
    EMES was the 'fraudulent trader' identified by the deal sheets in relation to deals 1, 3, 6 and 13. In each case an invoice in its name was made out to MG Components Ltd. We find that the VAT attributable to those invoices was fraudulently evaded. (We find it highly unlikely that the seller of the phones (whether EMES or the hijacker) had related input VAT which would cancel out all or some of the VAT on the sales).
  285. Oracle (UK) Limited (Deal 2)
  286. Mr Cameron Watson gave evidence of the investigation into Oracle's affairs in which he participated. We accept his evidence that:
    (i) in the first half of 2006 Oracle was importing mobile phones in
    bulk and selling them to UK businesses;
    (ii) some of those transactions involved payments by those businesses of amounts which constituted almost all the sales price being made to directly third parties, leaving a payment to Oracle which was substantially less than the VAT it had to account for on its sale, and thus leaving it without any apparent means of recovering and financing that VAT;
    (iii) Oracle's trading records for the period February to April 2006 had been taken by HMRC and had been used to produce total assessments for that period of some £23m. The assessment, made in July 2006, had not been appealed by Oracle. Oracle had not produced a VAT return relating to all or any part of that period;
    (iv) Oracle continued to correspond with HMRC on other matters up to at least November 2006;
    (v) the July assessment has not been paid, and Oracle was the subject of a winding up order made in March 2007. Since then there has been no communication from Oracle; and
    (vi) his investigations into the period 01/06 (the period ending 31 January 2006) revealed that Oracle had claimed VAT input credit to which it was not entitled, and had claimed VAT credit for cancelled sales which had in fact taken place but in relation to which Oracle claimed that its purchaser had not paid.
    The evidence as a whole indicated to us that Oracle did make supplies in February to April 2006 on which it did not intend to pay the relevant VAT and did not pay it. Those supplies included that relevant to Deal 2. We conclude it is likely that the Deal 2 phones were imported by Oracle. We conclude that Oracle fraudulently evaded the payment of all the VAT attributable to its supply in Deal 2.
  287. Realtech Limited (Deal 5)
  288. Mr Cole told us:-
    (i) Realtech was registered in September 2005;
    (ii) following a visit he made to Realtech in January 2006 at which he had raised concerns about whether it intended to trade, one of the directors had provided evidence of an intention to trade. At a later visit that director indicated it would not trade in mobile phones but four days later wrote to say that it would do so.
    (iii) following unsuccessful attempts to contact the company in March 2006, Mr Cole made successful visits in April and May in which he collected trading records;
    (iv) notices were served on Realtech requiring the speedy production of VAT returns for the 04/06 period, and for a period shortened to end on 24 May. The returns were received on 31 May.
    (v) the returns omitted significant sales and purchases.
    (vi) assessments made in November 2006, and January 2007 were returned by the Post Office. Letters sent to a director at his home address in February, July and August 2007 were not replied to.
    (vii) no appeal has been made against any of the assessments and no communication has been received from Realtech since mid 2006.
    (viii) in May 2006 Mr Cole recognised that Realtech had been making payments to parties other than its suppliers ("third party payments") and had been involved in deals which traced back to defaulting traders. (The evidence before us did not indicate whether third party payments had been made to UK or non-UK entities.) There was concern that it was acting as a "buffer". In the autumn of 2006 he was told by other officers that chains of phone transactions had been chased back to Realtech but could not be chased back further.
    (ix) Mr Cole assumed, in the absence of records of, or co-operation from, Realtech, that Realtech had acquired the goods it sold by importing them, and therefore that no input tax credit was attributable to its sales. On that basis assessments were raised for all the relevant output tax.
    Mr Beal says, and we agree, that a mere failure to pay an assessment is not evidence of fraud. Here, however, there was more than that. Here was an odd period of no activity and equivocation about mobile phone sales, there was then a period of little contact at the end of which a threat to deregister produced details of phone dealings, there was the preparation of VAT returns which omitted substantial transactions, there was the making of third party payments, and then a period of no communication and no payment. This was not the behaviour of a trader who got into difficulties and put his head in the sand; it was the behaviour of a trader who was up to no good. It seems to us that this trader intended to evade the VAT which was due for its period of phone trading.
    We are less clear about the amount of VAT evaded. There appears to have been an initial assumption by HMRC that Realtech was acting as a buffer in which case the VAT on its sales may have been offset by input tax ("may" because if it knew or should have known its purchase traced back to a defaulting trader the input tax would not have been recoverable), but then there was a change of view, and a conclusion that at least some of its purchases were imports, and the assumption was therefore made that no input tax was due.
    There was no evidence before us which related directly to the question of whether Realtech had imported the phones it onsold in Deal 5. There were purchase records indicating that very many purchases had been treated as subject to input VAT and indicating therefore UK suppliers. As a result it is hard for us to conclude that VAT on the whole of its sale price (rather than VAT on its margin – assuming, as is more likely, a profit) was sought to be evaded.
    The Respondents say that the facts set out above (including the underdeclarations of sales and third party payments) make it likely that Realtech was part of an MTIC deal chain in which either Realtech was an importer or a purchaser in a deal chain starting with a VAT evading importer.
    There was nothing in the evidence before us which enabled us to conclude that Realtech was the importer of the phones in Deal 5. Whilst Realtech's unusual pattern of trader and its evasion of VAT clearly indicate that it was involved in VAT fraud, there was not sufficient evidence for us to be able to conclude that the phones in Deal 5 were likely to have been imported by a person who evaded all the VAT in their onward sale.
    We conclude that it is more likely than not that Realtech sought fraudulently to evade and did evade the VAT on its margin, but it has not been proved that it fraudulently evaded any larger sum or that the phones in Deal 5 were acquired in a deal chain starting with a VAT evading importer.
  289. Attic Attack Limited (Deal 7)
  290. Mr Selwood told us:
    (i) that Attic Attack started trading in Loft conversions in October 2005 and was VAT registered from 1 April 2006;
    (ii) after some unsuccessful attempts when he had left letters messages, Mr Selwood visited the company on 11 April after being told by another branch of HMRC that it had imported CPUs. He met Mr Cavenagh, a director on 8 May. Mr Cavenagh said that in April he had given a business card to a taxi driver who had suggested that he should start dealing in mobile phones. After that he had received a lot of documentation. He said he then panicked (he may at this stage also have received letters from HMRC) and sent letters saying he would not pursue the deals, but when he was shown copies of release notes addressed to freight forwarders purporting to come from Attic Attack and signed by him he admitted he may have signed them. He then provided a box of documentation which included details of sales of mobile phones;
    (iii) Mr Selwood received from colleagues in HMRC details of other Attic Attack invoices dated April 2006 for mobile phones;
    (iv) An assessment had been made on Attic Attack for some £3m which included VAT in relation to its sales in Deal 7;
    (v) the assessment has been appealed and the appeal was waiting to be heard. Mr Selwood understand the basis of the appeal to be that Attic Attack did not make the supplies;
    (vi) Attic Attack were still trading.
    We accept this evidence.
    The documentary evidence in relation to Deal 7 included a copy bank statement of Siverline, the ostensible purchaser from Attic Attack. There were payments recorded which could have been for the phones in Deal 7 but no evidence as to whom the payments had been made.
    HMRC submit that if the appeal is successful then the phones must have been supplied by another person who intended fraudulently to evade the VAT, and if the appeal is unsuccessful that Attic Attack will be unable to pay the VAT and that it never intended to pay it.
    We agree with the first part of this submission. If Attic Attack did not supply and invoice the phones then it seems more likely than not that someone else made the supply with the intention of dishonestly evading VAT on it.
    Whilst we cannot determine Attic Attack's appeal, it does on the limited evidence available to us seem more likely than not that it supplied the phones and as a result likely that it has a VAT liability. The issue for us is therefore whether the Respondents have provided cogent evidence that Attic Attack did not intend to pay the VAT. The evidence before us in relation to this issue was:-
    (i) Mr Selwood's evidence that Mr Cavenagh had told him that in his conversation with the taxi driver he had said that Attic Attack was in need of additional funding in April;
    (ii) Mr Selwood's evidence of Mr Cavenagh at first denying dealings and then admitting to them.
    If Attic Attack was the importer and onward supplier in a contrived chain in which third party payments were made, it may not have received adequate payment for the goods it sold so and would not have had the funds to pay its VAT liabilities, and so probably would not have intended to pay it. But we have no evidence that it did not receive payment for Deal 7 or that it does not have a valuable right yet to receive payment and thus to fund its VAT liabilities; and evidence of Mr Cavenagh's apparent panic and change of mind, although it may create our suspicion, could also be explained in other ways. We did not hear Mr Cavenagh.
    Thus although we believe it is quite possible that Attic Attack fraudulently did not intend to pay the VAT we did not have before us sufficient cogent evidence to conclude that it was more likely than not that that was the case in the face of Attic Attack's continued trading.
  291. [blank]
  292. I Connect U Limited (Deals 9, 10, and 16)
  293. Mr Ross told us that on 11 April 2006 he visited I Connect U after hearing from another part of HMRC that it was dealing in large quantities of mobile phones. He delivered a Regulation 25 notice requiring it to complete a VAT return for the period to 10 April. He was told on his visit that I Connect U had not done any deals. On 12 April he called again, this time he was told by the same person that there had been some trade. He picked up the return and the company's records from its accountants on 13 April. Those records showed that a number of sales had not been declared although the related imports had been. He also concluded that input tax had been fraudulently claimed to offset a VAT liability declared. He made an assessment for some £3.8m which was subsequently increased to £3.9m. The assessment has not been appealed or paid and I Connect U has not since contacted HMRC. It is in liquidation. The amounts assessed include the VAT in respect of I Connect U's sales in Deals 9, 10 and 16. We believe these sales were of phones imported by I Connect U.
    We find that I Connect U fraudulently intended to evade payment of and did not pay the VAT in respect of its sales in Deals 9, 10, and 16. The company's actions in relation to its VAT return and its failure to respond in any way to the assessments convinced us that it was more likely than not that it intended never to pay the relevant VAT.
    I Connect U's supplier in each deal was Mingele, a non-UK entity. We conclude that it was likely that I Connect U was entitled to no input tax credit and accordingly that the whole of the VAT on each sale was fraudulently evaded.
  294. Skywide Limited (Deals 11 and 17)
  295. From Miss Badminton's evidence we find as follows:-
    Skywide was registered in 1999. It corresponded with HMRC on its VAT matters until September 2005. Between September 2005 and 12 April 2006 no communication was received in the name of Skywide and letters from HMRC were returned. Then on 12 April 2006 a change of address letter was received in Skywide's name, but the director's signature on the letter did not appear to match his signature on earlier letters. There is no record of any communication to HMRC from Skywide since 12 April 2006. Letters, assessments and notification sent since then to Skywide's initial address, the addresses of its directors and its new address have gone unanswered with one exception. That exception was in relation to a letter sent to the Company Secretary at his registered home address: Miss Badminton received a call from the occupier who said that the addressee was no longer there.
    Information from Skywide's customers indicated that Skywide (or someone purporting to be Skywide) required its customers to make payments to non-UK entities.
    From the records of other companies details were obtained of goods sold, or said to have been sold, by Skywide. No VAT return was made by Skywide in respect of these sales but HMRC officers have sought to demand payment of the VAT on the sales by assessments or letters asserting a debt due to the Crown. (The former being relevant to sales made by a VAT registered entity, the latter to sales by a non registered entity. Skywide was at various times deregistered).
    The VAT on sales attributed to Skywide in Deals 11 and 17 has been notified to Skywide under these letters or assessments.
    We conclude from Miss Badminton's evidence that it is likely that another person made the supplies attributed to Skywide in Deal Sheets 11 and 17, and that such person dishonestly intended to evade payment of, and did not pay the VAT thereon, and that it is likely that no input tax credit was available to it.
  296. FX Drona Limited (Deal 12)
  297. From Mr Mendes' evidence we find as follows.
    FX Drona was VAT registered on 2 February 2006 following an application which stated that its business would be making supplies of graphic animations and multimedia development services. Its directors changed on 15 March 2006. On 30 March HMRC concluded from a freight forwarder's records that FX Drona was importing mobile phones. On 31 March FX Drona was served with a Regulation 25 notice requiring the completion in short order of a VAT return up to that date. The notice was delivered at the time of a meeting with the company's officers at which HMRC were told that FX Drona's supplier was "Electron in Bath". Electron Global appears as a 'buffer' in other deal chains.
    No return has been received. A letter sent to the company on 11 July 2006 was returned "moved away". HMRC made assessments on FX Drona in respect of transactions undertaken by it of which it had become aware. The assessments include that related to Deal 12 and total some £33m. No appeal or communication has been received by HMRC in respect of the assessments. The Company is now in liquidation.
    We conclude that FX Drona intended dishonestly to evade the payment of VAT attributable to transactions including that relevant to Deal 12: we can think of no other likely explanation for the change of directors and business, the sudden increase in sales and the failure to communicate with HMRC.
    We reach the conclusion independently of whether or not FX Drona was acting as a buffer or as an importer. However, despite the possibility, raised by the mention of Electron, that FX Drona was acting as a buffer, we concluded that it was more likely it had been an importer in relation to Deal 12 because the evidence in relation to the chain showed that payment instructions had been given by FX Drona to the company to which it sold to make payment to a non-UK entity.
  298. Data Solutions Northern Ltd (Deal 14)
  299. From Mr Marescaux' evidence we find the following facts:-
    (i) Data Solutions was registered for VAT on 31 March 2005. It declared its business as dealing in data. At the relevant times thereafter it had one director, Keith Thelwell;
    (ii) in July 2005 HMRC obtained evidence that it was dealing in items common in MTIC fraud. At visits in August 2005 and May 2006 by HMRC Mr Thelwell accepted he was trading in CPUs and mobile phones and give details of his suppliers (who were UK entities). There were discussions with Mr Thelwell about the due diligence he conducted on his suppliers and customers;
    (iii) Data Solutions wrote to HMRC in June 2006 providing deal information. Following abortive visits, HMRC wrote on 11 September asking Data Solutions to contact them or be deregistered. They were later deregistered (we conclude no reply was received);
    (iv) on 27 September 2006 Data Solutions was put into voluntary liquidation. In the report to creditors Mr Thelwell blames the liquidation on the pressure of having to comply with VAT due diligence requirements and the adverse effect that was having on his health;
    (v) Data Solutions submitted VAT returns for the periods up to 02/06 but did not submit them for 05/06, 08/06 and the period to deregistration. The turnover for 02/06 was stated as £47m;
    (vi) on the basis of records taken from Data Solutions, and information from its customers and suppliers HMRC officers computed its VAT liabilities for the missing periods and Mr Marescaux made assessments totalling some £21m. These assessments included an amount in respect of Deal 14;
    (vii) initial assessments were made in lesser amounts on the basis that Data Northern was a buffer and was entitled to input tax credit, but later information from a freight forwarder indicated that it had been an importer of CPUs in April 2006. On this basis assessments were revised to allow input tax credit only where HMRC had evidence of acquisition from a VATable UK counterparty;
    (viii) in relation to Deal 14 HMRC had no evidence as to whom Data Solutions acquired from. The assessment therefore allowed no input tax credit;
    (ix) no payment has been received in respect of these assessments.
    Mr Beal noted that in deals 41, 48, 50, 54, 76 and 89 Data Solutions appears as a buffer sandwiched between UK entities. Data Solutions invoiced Deal 14 in April, Deals 41, 48, 50 and 54 in May, and Deals 76 and 89 in July.
    Counsel for HMRC say that although Data Solutions was a buffer in other deals 'that is no reason to conclude on the evidence that it was only a buffer' in Deal 14,
    Data Solutions failure to render VAT returns and to account for VAT in the period May to September was followed by voluntary liquidation in September. This suggests at the least some recklessness as to whether VAT was properly accounted for. We note that on 11 May 2006 Mr Thelwell was recorded as saying to two of HMRC's officers that he had never purchased from EC countries but that the information obtained from the freight forwarders, Humber Freight and ASR Logistics in relation to CPU transactions in taking place in April 2006 and from ASR Logistics in relation to Nokia 8800 phones in May 2006 indicated purchases from Megatek Sarl. (Deal 14 was invoiced in April 2006). We accept the evidence of that conversation and those purchases and conclude that Mr Thelwell was not being honest.
    Taking all this together we conclude, on balance that it is more likely than not that Data Solutions intended dishonestly to evade its VAT liabilities in respect of transactions in the period April to September 2008.
    We are not convinced that the evidence shows that Data Northern was an importer in relation to Deal 14 rather than a buffer. It seems to us that the general pattern of the later deals in which it was a buffer suggest that it was more likely that it was a buffer (noting what appears to be an exception in the preceding paragraph but one) in relation to this deal and as a result that the VAT sought to be evaded was limited to the VAT in its margin.
    The Respondents say that the evidence makes it likely that Data Solutions was set up as a key participant in an MTIC fraud and that in Deal 14, if Data Solutions was not the importer, then it was the purchaser in a deal chain starting with an importer who fraudulently did not account for the VAT on its sale.
    However, in our assessment the evidence as to Data Solutions' activities does not enable us to conclude that it is more likely than not that its purchase of the phones in Deal 14 was connected to an import in which the importer fraudulently evaded VAT on its onward sale.
  300. Apollo Communications Centre Limited (Deal 15)
  301. From Miss Badminton's evidence we find:-
    (i) Apollo was VAT registered in February 2004. Its sole director at relevant times was Mr Rahman;
    (ii) in April 2006 HMRC officers visited Apollo's premises, removed records and delivered (on 24 April) a notice requiring a speedily produced VAT return for a shortened period. Following the second visit Mr Rahman telephoned to say that he would not be able to pay VAT of £24m and asked to be deregistered. He indicated that his customers were supposed to pay the VAT element to him and the rest to his suppliers. His customers now said they were waiting for payment from their customers before making payment. So far he had not received anything. He was going to issue credit notes;
    (iii) the required VAT return was delivered on 25 April. It showed a VAT liability of £4,035;
    (iv) from the records obtained by HMRC the VAT due for the relevant period was £23.9m odd;
    (v) the difference between HMRC's calculation and that of Mr Rahman resulted from Mr Rahman's inclusion of a number of credit notes for deals done in April 2006. It appears that Mr Rahman indicated to HMRC that it was his customers' refusal to pay which prompted the issue of the credit notes;
    (vi) HMRC issued various additional assessments between April and November 2006. The VAT eventually assessed was in the region of £45m;
    (vii) no further communication was received by HMRC from Mr Rahman. No appeal has been made;
    (viii) in April 2006 Menzies Corporate Restructuring wrote to HMRC indicating that they were assisting Apollo with its financial affairs. At some time between then and August 2007 Apollo was put into liquidation;
    (ix) a disqualification order was made against Mr Rahman. He signed a statement in which he said:
    "I caused Apollo to effect millions of pounds of purchases and sales from 7 April to 24 April 2006 … acting against written advice provided by HMRC in instruction [sic] that millions of pounds of Apollo's sale revenue be paid to an overseas company … and causing Apollo to continue to sell to customers that failed and continued to fail to pay Apollo the millions of pounds that they owed it for the goods received, thereby causing Apollo to trade at a significant risk of a VAT fraud being perpetrated against HMRC …";
    (x) the credit notes which reduced the VAT liability in Apollo's VAT return included one in respect of Deal 15, and the assessed VAT includes VAT calculated without any credit in respect of that credit note;
    (xi) a schedule prepared by Howes Percival indicates that in relation to many deals in April 2006 Apollo acted as importer.
    Mr Beal noted that Mr Rahman does not admit involvement in fraud: he merely admits that he exposed Apollo to the risk of fraud.
    We find it likely, from Mr Rahman's statement in April 2006, that Apollo's customers did not pay Apollo because they had been instructed by Apollo to pay, and had paid, the monies due to third parties; and that the credit notes were issued in response to a realisation that Apollo would not receive payment. It seems likely that those third parties were overseas and that Apollo's suppliers were therefore overseas. It is unclear whether Mr Rahman knew or understood, at the time such instructions were given, that the making of such payments would deprive Apollo if the ability to pay the VAT it would owe, but it is clear to us that someone must have known, and that Mr Rahman or another party dishonestly procured the payments with the intention that Apollo would not be able to pay its VAT liability including its VAT liability in relation to Deal 14.
    Mr Beal noted that there was evidence that the liquidator of Apollo had corresponded with Gara Technologies (which was Apollo's customer in Deal 14), and that Gara was offering £5,000 in settlement to Apollo. It did not seem to us that this affected our conclusion: the output VAT payable by Apollo on its Deal 14 sale was about £150k; £5,000 even if in settlement of a claim ten times that size (and there was no evidence before us in relation to that) would not have enabled Apollo to pay its VAT liability. It seems likely that Apollo was importing and had no input tax credit to set against its output liability.
  302. Goodluck Employment Services Ltd (Deal 18)
  303. From Mr Patterson's evidence we find:-
    (i) Until the later part of 2005 Goodluck supplied the services of employees. It told HMRC that it made no supplies between 1 September 2005 and 3 November 2005, and in December informed HMRC of a change of address.
    (ii) In March 2006 HMRC concluded that Goodluck was importing from the EU. A Regulation 25 notice was served requiring a return for the period to 24 April 2006.
    (iii) At a visit to Goodluck on 25 April HMRC's officer Andy Monk said that the director "blamed all MTIC trading on an administrator who was off ill" (to quote Mr Patterson's witness statement).
    (iv) Between May 2006 and September 2007, 14 assessments or amendments thereto or notification thereof were sent to Goodluck.
    (v) Goodluck went into liquidation in January 2007.
    (vi) A letter sent to Goodluck on 29 January 2007 was returned "gone away". Letters to the Director and the company sent on 6 August 2007 were returned.
    (vii) HMRC record outstanding liabilities of some £17m.
    (viii) The VAT attributable to Goodluck's sale in Deal 18 forms part of the VAT assessed.
    Mr Ross told us (and we accept) that no payments had been received in respect of these assessments.
    We accept that it is likely that Goodluck undertook the transactions in respect of which HMRC made assessments. We accept that it is likely that Goodluck was importing the phones it sold and was entitled to no input tax credit. We find the failure to communicate with HMRC after May 05 indicative of an intention not to pay the VAT in respect of those transactions. It seems to us more likely than not that that intention was formed at the time of Goodluck's sale in Deal 18.
  304. GPA International Ltd (Deal 20)
  305. From Mr Bennett's evidence we find:-
    (i) GPA was VAT registered in July 2004 as a drinks retailer;
    (ii) it provided VAT returns for the 09/04, 12/04 and 03/05 periods declaring outputs of about £15k per month on average. It made nil returns for 06/05 and 09/05 and a modest return for 12/05.
    (iii) On 1 June 2005 the original officers of the company resigned and Amanulla Raja (it may be that his name was Raja Amanulla) was appointed as director;
    (iv) No further VAT returns have been made;
    (v) Following receipt of information from another branch of HMRC that GPA was dealing in significant quantities of mobile phones (some of which had been imported from a Spanish vendor in April 2006) Mr Bennett visited GPA's declared place of business on Friday 28 April 2006 (Deal 20 was invoiced on 21 April 2006). The premises were shuttered and closed. The new officers of the company had given it as their private address. Mr Bennett did not believe the premises could provide residential accommodation. Mr Bennett left a Regulation 25 notice requiring a VAT return for the period to 28 April 2006.
    (vi) On his return Mr Bennett e-mailed GPA saying that he was arranging the cancellation of its VAT number. On Tuesday 2 May he received a phone call purporting to be on behalf of Amanulla Raja to say that he would visit his accountant on Thursday 4 May.
    (v) By a letter dated 3 May 2006 Amanulla Raja said that he would be back in his office on 8 May and would arrange for his accountant for the production of the return which had been required by the Regulation 25 notice.
    (vi) On 4 May 2006 Amanulla Raja e-mailed Mr Bennett saying that GPA was still trading, his mother was sick, and he was in London on business. He said he would be back on 18 May and be in contact then. He did not want his VAT number suspended. He gave no address or telephone contact details.
    (vii) No further communication has been received from GPA's officers.
    (viii) GPA was put into liquidation on 21 June 2006.
    (ix) Assessments were made on GPA on 20 June 2006 for £4m odd, 1 August 2006 for £4m odd, and on 10 further occasions. The assessments related to sales of mobile telephones in March and April 2006. The VAT in respect of GPA's sale in Deal 20 is included in those assessments. The assessments have been neither paid nor appealed.
    In point (v) above we note the evidence that GPA was importing phones. The Spanish Vendor concerned was Jacub Impex. Mr Beal notes that documents 13331-4 indicate that Jacub Impex operated in the UK and were themselves importers. We do not draw that conclusion from those documents: Jacub Impex' operational address is in Spain; the only evidence relating to the UK is a request on its purchase order that the goods be delivered to the UK; that is consistent also with a sale by Jacub Impex to a UK importer and the direct delivery of the phones by Jacub Impex' supplier to its UK customer.
    Mr Bennett also referred to two police investigations into the affairs of GPA. We do not know the results of the investigations and decline to conclude that an investigation is a pointer to guilt.
    Mr Bennett also noted that it had been reported to him that GPA had underdeclared its sales in the 12/05 quarter (after the change of officers) and had been assessed for £112,601 in respect of the underdeclaration.
    It seems to us that when the new officers took over GPA it made significant supplies with no intention of reporting them to HMRC. The company's response to HMRC's contact was, it seems to us, limited to a response in respect of the cancellation of its VAT registration – a registration which enabled it to participate in further sales in which the purchaser would have been able to seek to claim input VAT. No significant response was received in relation to the proper reporting and accounting for VAT.
    We find that GPA dishonestly intended to evade the payment of VAT in respect of its transaction in mobile phones in March and April 2006 and did not pay that VAT. That VAT included the VAT attributable to Deal 20.
    HMRC say that the evidence makes it likely that GPA was being used as a key participant in MTIC fraud and that the transaction in Deal 20 was part of an MTIC deal chain in which either GPA was the importer, or was the purchaser in a chain starting with an importer who fraudulently evaded VAT. Given the evidence that GPA was importing phones and its sudden blooming and death, we think it more likely than not that it imported the phones in Deal 20 and accordingly that the VAT evaded by it was the whole of the VAT on its sale with no input tax credit.
  306. UA Distribution Ltd (Deals 19, 21 as to 2000 phones, 22, 24, 25)
  307. From Mr Smith's evidence we find:-
    (i) UA Distribution was VAT registered in September 2005. Its sole director at all material times was Mr Azam.
    (ii) UA Distribution made VAT returns for the 12/05 and 03/06 periods, and, in accordance with two Regulation 25 notices made returns for the shortened periods 1 April-25 April 2006, and 26 and 27 April 2006. The returns were in good time. The company was deregistered on 27 April 2006.
    (iii) Mr Smith took away from UA Distribution its deal packs relating to the purchase and sale of goods. He used them to audit those VAT returns.
    (iv) The two April returns included a claim for the credit of £1,959,358 of input tax. The company's documents indicated that this was input tax on purchases from Worldwide Enterprises and that the goods purchased had been exported. Mr Smith concluded that these transactions were fictitious. He made an assessment on UA Distribution on that basis and notified UA Distribution on 29 June 2006.
    (v) The company's documents in relation to the 03/06 return also indicated an input tax credit claim in respect of exports to an Italian customer. Mr Smith concluded that these were fictitious.
    (vi) The VAT returns included VAT in respect of UA Distribution's sales in Deals 19, 21, 22, 24 and 25.
    (vii) After his collection of the April VAT returns from Mr Azam, Mr Smith explained that he intended to disallow the input tax on the Worldwide Enterprises purchases. Mr Azam said he would gather evidence to support these transactions.
    (viii) Since then Mr Smith has heard nothing from Mr Azam, but:
    (a) UA Distribution's solicitors requested a
    reconsideration of the deregistration in July or
    August 2006;
    (b) UA Distribution was dissolved on 28 August
    2007. It had filed no accounts;
    (c) a tax adviser representing Mr Azam or UA Distribution contacted HMRC on 10 July 2008
    suggesting that bad debt relief should be available to cancel out the £1.9m assessment,
    and evincing a wish that the company appeal.
    HMRC replied that it would deal only with the
    Liquidator of the company.
    (ix) No appeal has been made against the assessment. No payment has been received.
    This was not a case where sales had not been declared or VAT returns not returned. Mr Azam appeared to have been co-operative. Persons representing the company had contacted HMRC. The company's officers and representatives had not arranged its transactions and then disappeared. However the assessment had not been appealed and the tax had not been paid. We address two preliminary issues: first whether Mr Smith's £1.9m assessment was justified, and second whether VAT bad debt relief would be available to eliminate it. If the assessment is not justified or if bad debt relief were available to eliminate, there would be no loss of tax and no fraud - whatever the motives of these involved.
    Mr Smith told us, and we accept that:-
    (i) the input tax related to three Worldwide Enterprises invoices to UA Distribution;
    (ii) the documents taken from UA Distribution indicated that the purchaser was an Italian company;
    (iii) there was no evidence of the removal of the goods purported to have been sold from the UK;
    (iv) the documents suggested the goods had been held at the freight forwarder Point of Logistics but another officer told him that Point of Logistics denied all knowledge of the transactions;
    (v) there were documents asking the Italian customer to pay the full invoice amount to a specified account with First Curacao International Bank, but no identification of the account holder.
    We also note that no further information on the deals was provided by Mr Azam. We conclude that it is likely that the deals were fictitious and that their inclusion in the VAT return was a dishonest attempt to evade the VAT attributable to the company's sales.
    Among the documents relating to UK sales Mr Smith also found requests from UA Distribution to its customers to make payments to third parties. These requests were made in relation to each of Deals 19, 21, 22, 24 and 25.
    Mr Azam told Mr Smith that UA Distribution had received no payment for its UK sales. The payment instructions in relation to the specified Deals provided for payment to UA Distribution of some £2.3k in total with the remainder of the purchase price being paid to third parties. So UA Distribution had £2.3k with which to finance some £187k of output VAT due on those deals. It seems likely to us that Mr Azam was telling Mr Smith that he had not received the sums corresponding to the £2.3k.
    Regulations 165-172B VAT Regulations 1995 provides for the giving of bad debt relief. It is clear to us, if a trader requires payment to be made to a third party, that bad debt relief is available in respect of that payment only if it is not made and that failure to pay is reflected in the accounts of the supplier. There was no evidence which suggested that the third party payments had not been made. We conclude that bad debt relief was unlikely to have been available in respect of them.
    On the evidence before us the greatest possible bad debt relief claim available to UA Distribution in respect of the April returns would have been that proportion of its output tax which the amount it did not receive, £2.3k, represented of the sales price of those deals. That amount is about 0.2% of its assessed liability.
    We conclude that HMRC was not paid almost all the VAT due in respect of UA Distribution sales on the specified Deals and that UA Distribution dishonestly intended to evade paying it.
    In each of the specified deals UA Distribution we find imported the phones. The VAT evaded was thus unreduced by any input tax credit.
  308. B Logistics Ltd (Deal 25)
  309. Miss Glover gave evidence in relation to B Logistics. She had not taken any part in its relationship with HMRC but exhibited documents drawn from HMRC's records prepared by other HMRC officers or received by HMRC relating to it. We did not hear evidence from the relevant officers directly.
    We have no hesitation about finding the following facts from that evidence:
    (i) B Logistics was VAT registered in November 2005. At all material times it had a single director Mr Rafiq;
    (ii) B Logistics made only one VAT return: for the period from registration to 31 January 2006;
    (iii) The VAT shown as due on that return was £146.25 which was paid. HMRC have received no payment since then;
    (iv) On 3 April 2006 HMRC wrote to the company requiring the production of business records and indicating that if they were not forthcoming it would be deregistered;
    (v) On 6 April the company's accountants wrote to ask for more time;
    (vi) On 11 April HMRC wrote to the company to say that it had been deregistered with effect from 3 April 2006.
    (vii) Assessments for £1.3m, £602km £13k, and £72k were issued to the company on19 October 2000, 15 December 2006, 15 May 2007 and 7 March 2008;
    (viii) the total debt recorded as outstanding from the company in HMRC's books is £1,989k. No payment of the assessment has been made.
    (ix) the assessments include output VAT in relation to Deal 25;
    (x) B Logistics was put into liquidation on 29 November 2006.
    Miss Glover's witness statement does not indicate whether any communication had been received from the company or its representatives after 6 April 2006. Miss Glover's statement however appears to list all the items she could find in HMRC's records and we infer that no such communication was received.
    The report of a visit on 30 March 2006 by another officer to B Logistics exhibited by Miss Glover indicates that Mr Rafiq said that the company was dealing in bulk sales of mobile phones, that it had done 50 or 60 deals since 14 February 2006, and that it had made payments ("third party payments") to persons other than its suppliers in respect of its purchases at the request of those suppliers and that many of those payments had been made to foreign bank accounts. We accept that Mr Rafiq said words to this effect. HMRC requested B Logistics' records at that visit. After correspondence with its accountant banking records were provided on 28 April. Other records appear to have been provided to make the assessment dated 19 October 2006 (see 14/4908).
    The company's failure to render VAT returns and to engage with HMRC's assessments persuades us that it intended dishonestly to evade payment of its VAT liabilities in respect of transactions undertaken in February, March and April 2006 including the VAT in respect of its sale in Deal 25.
    Mr Beal suggested that B Logistics had acted as a buffer in relation to Deal 25. If it had then (arguably) its VAT liability on its sale would be reduced by the input tax on its purchase from a UK supplier, and thus restricted to the VAT on its margin (if any).
    The assessment for the VAT on Deal 25 was made on 2 May 2007. The letter accompanying the notice of assessment states that it is made on the basis of information in the records of Xcel Solutions, and asks for further details. There is no record of such details being provided.
    The schedule (14/4908) to the assessment for £1.3m made on 19 October 2006 indicates that B Logistics was purchasing 90% of its phones from UK suppliers. There is no indication in the accompanying letter as to whether or not B Logistics purchased these phones from a UK supplier or imported them. Nor is there any indication as to whether it made any profit on the transaction. Despite the evidence of some third party payments to foreign bank accounts, we do not find it proved that the phones were imported. On balance we find it likely that some margin was made and thus that there would have been some VAT liability.
    Our conclusion is thus that we are satisfied that B Logistics evaded and dishonestly intended to evade payment of the VAT on whatever margin it made in respect of Deal 25.
    The Respondents say that the evidence indicates that Deal 25 was part of a chain in which either B Logistics was the importer or which started with an importer who fraudulently evaded VAT on its onwards sale.
    There was no evidence to suggest that B Logistics was the importer and no direct evidence as to whom the importer would otherwise have been , or whether or not it was fraudulent. We are asked to conclude that the combination of a deal chain tracing back to B Logistics, and B Logistics' failure to account and response to assessments means that it is likely that either B Logistics imported the phones which were the subject of the specified deal or that someone else did and fraudulently evaded the VAT. We are unable to do so on the available evidence.
  310. USM IT Limited. (Deals 22 & 23 (as to 2953 phones), 26, 31, 32, 34, 37-40,
  311. 42-45)
    Miss Davis and Mr Ross gave evidence in relation to USM. Miss Davis had taken part in the assessment of USM but a significant part of her evidence consisted of reports and other documents prepared by other officers and stored in HMRC's files (paper and electronic). We have accepted the evidence of matters such as the registration and deregistration of the company and letters sent to it which Miss Davis took from HMRC's records. We find:-
    (i) USM was VAT registered from 1 September 2004 to 10 May 2006;
    (ii) in the five quarterly VAT periods to 31 January 2006 the company's net sales were about £53,000 per quarter. USM's turnover for the period between 7 February and 14 March 2006 was £44m;
    (iii) 14 assessments were made on the company between 5 June 2006 and 15 February 2008 for a total of about £17m. The assessments cover supplies made between February and May 2006;
    (iv) no VAT returns were made after the 01/06 period. No response appears to have been made to a Regulation 25 notice requiring the company to submit a return for the period to 9 May 2006;
    (v) no communication has been received by HMRC appealing or contesting any of the 14 assessments;
    (vi) USM went into liquidation in January 2007. Although Miss Davis does not say so explicitly we infer from the tenor of her witness statement that no payment has been made in respect of any of the assessments, (and Mr Ross in a statement dated November 2007 indicated that at that time the aggregate of the assessments stood at £11m and no payment had been made);
    (vii) records relating to deals in February 2006 and up to 10 March 2006 were collected from USM on 14 March. Further records were seized in August 2006;
    (viii) the assessments noted in (iii) above have been compiled on the basis of the records referred to in (vii) and information from other officers responsible for traders who have offered USM invoices in support of input tax credit;
    (ix) the records indicate that in a number of cases (Miss Davis' witness statement suggests all for February deals and for all deals up to 10 March 2006, although USM IT's invoice of 10 February 2006 to Global Logistics (Doc 4680) gives USM IT's account as the only account for payment) USM instructed its customer to make payment of the bulk of its purchase price to a third party leaving USM with inadequate funds to meet its VAT liability;
    (x) the reports by other officers indicate inconsistencies, and some lack of availability or helpfulness, of USM's officers. It is difficult to be sufficiently certain that these behaviours were the result of dishonest intention;
    (xi) mail sent to USM's declared place of business has been returned "gone away" since October 2006.
    The sudden change in the turnover of USM, the failure to return the VAT due, the failure to contact HMRC, and the payment instructions which deprived USM of the funds to pay its VAT all point in our view to a scheme to evade payment of the VAT properly due on the transactions undertaken by USM in February to May 2006. We conclude that USM dishonestly intended to evade payment of the VAT due in that period.
    Miss Davis explains the assessment of USM in relation to the Deals specified in the heading of this section. (Each of the 14 deals are reflected in one or other of the assessments referred to above.) The company's records show that the phones for Deals 26 and 38 were imported. Thus the assessment of VAT for those deals gave no input tax credit. But she says that the company's records do not deal with the remaining 12 deals. She says at para 47 of her statement "as the records that are available … confirm that USM always purchased the goods from Europe … no allowance has been made for the recovery of the input tax." However at paragraphs 28 and 29 she indicates that USM may have been supplied by persons (presumably in the UK) using hijacked VAT numbers, and, at paragraph 48, that certain goods sold to Fonedealers were purchased from a UK company. We also note that at document 4687 TEC, as USM's supplier gives payment instructions : TEC was a UK company.
    We conclude that USM intended to evade the VAT attributable to its sales in the specified deals. In relation to the following deals there was evidence which enabled us to conclude that the VAT intended to be evaded was the whole of the VAT on its sale:-
    Deal 22 & 23: In Deals 22 and 23 the chains trace back to two sales by
    USM. In one (for 2000 phones) it buys from UA Distribution
    who we find bought from Premiston outside the UK. In the
    other for 2953 phones there is evidence of a purchase earlier in
    the chain from Premiston in the form of payment instructions.
    We conclude that either USM imported in this second deal or
    that an earlier party importing from Premiston fraudulently
    evaded the VAT on its sale.
    Deal 26 : Miss Davis' evidence that the company's records
    showed the phones were imported.
    Deal 37 : Evidence of purchase by USM from Premiston in Florida.
    Deal 38 : Miss Davis' evidence that the company's records showed the
    phones were imported, and evidence of payment
    instructions given by USM to its customer to make payment "to my supplier Marubo International SL". It seemed likely that
    this supplier was a non-UK party. Mr Beal notes that I Partner gave its customer payment instructions using the same words:
    he says therefore that USM's instructions do not necessarily
    indicate that it was the importer from Marubo. We
    acknowledge that it is not absolutely clear but on balance we
    think that the instructions point to that conclusion.
    Deal 39 : As deal 37.
    Deal 40 : As deal 37
    Deal 42 : As deal 37
    Deal 44 : As deal 37 together with release notes.
    Mr Beal points to documents which suggest that Premiston was operating within the UK and was itself the importer. Premiston's invoices give an Estonian address but do not otherwise indicate whence it is operating. Premiston gives instructions to Interken in Deal 37 to release to USM phones "allocated to myself by Intertech SARL" – thus the phones were allocated to Premiston while they were in the UK and then allocated and released by Premiston in the UK to USM. In deal 42 the documentary evidence includes a sale by Sunico A/s to Premiston for delivery to Hawks in the UK.
    It seems to us that the allocations instructions given by Premiston in Deal 37 indicate that the supply of those phones by Premiston to USM did not involve their removal from the UK, and accordingly, as a result of s.7(2) VATA, their supply took place in the UK. Their supply was clearly in the course of a business by Premiston and it seems likely to us that Premiston was a person who would be liable to be registered for VAT under Schedule 1 VATA. As a result the supply of the phones to USM would be a taxable supply within section 4 VATA on which Premiston should have charged VAT. Premiston's invoices however bear no VAT number and therefore would not satisfy the requirement in Regulation 29(2)(a) of the VAT Regulations 1995 that USM should hold a VAT invoice. It also seems unlikely that the tailpiece of Regulation 24(2) would apply. As a result we conclude that even if Premiston was acting from the UK, ie even if USM's purchase from Premiston was not an import no import tax credit was available to USM and the VAT evaded by USM was the whole of the VAT on its sale.
    That leaves Deals 31, 32, 34, 43 and 45. In the light of the evidence of USM's purchases in the other deal chains and the nature of the deal chains, we conclude that it is likely that USM imported the phones in those deals or otherwise was not entitled to any input tax credit. We therefore conclude that in relation to each of the specified deals USM IT intended to evade VAT equal to the whole of the VAT on its sale.
  312. Colston Association Limited (Deals 27 and 28)
  313. From Ms Humphrey's (and Mr Ross' evidence in relation to point (iii) below) evidence we find:-
    (i) Colston was VAT registered on 29 December 2005. Its VAT registration application declared its business as management consultancy and its expected turnover £180k;
    (ii) the VAT return form for the first VAT period (to 28 February 2006) was returned "gone away";
    (iii) no VAT returns were submitted by the company;
    (iv) mail sent by HMRC to the company's recorded place of business was returned marked "gone away" on 4 May and 22 May 2006;
    (v) on 26 April 2006 a report by an officer of HMRC describes a telephone call with a Mr Marsh who is said to have said that he had acquired Colston and that it was involved in the wholesale of electrical stock such as mobile phones, but became reticent when asked about details of the business;
    (vi) the officer who spoke to Mr Marsh indicated that HMRC had acquired evidence of imports by Colston in the paperwork of freight forwarders. We accept such evidence was found and that it indicated imports by Colston even though the documents were not exhibited;.
    (vii) assessments totalling some £16m and representing sales of over £60m were made on Colston on the basis of evidence of its sales collected by other officers. Assessments were made for a Crown Debt under paragraph 5 schedule 11 on the basis of invoices issued since Colston was not registered. No communication has been received by HMRC in relation to these assessments;
    (viii) the company is now in liquidation;
    (ix) VAT in relation to Colston's sale in Deal 28 is included in the assessments but there was no evidence that the VAT on Deal 27 had been assessed.
    The failure to provide VAT returns, the failure to communicate with HMRC, the difference between the activities declared in the registration form and the actual activities, and the volume of transactions undertaken and not accounted for so soon after registration convince us that the company intended dishonestly to evade VAT on its sales after its registration. Those sales included those in Deals 27 and 28.
    In both Deals 27 and 28 Colston passed to its customers instructions to pay a Cypriot company. From these, and taking into account the evidence of imports of phones in the freight forwarders documents, (and acknowledging that it was possible that the Cypriot company was selling from within the UK), we conclude that it is likely that Colston imported the phones in Deals 27 and 28.
    We find that Colston intended to evade VAT on the relevant deals. Further we find that since it was likely that the phones in these Deals were imported by it, it intended to evade, and evaded, all VAT on the VAT on its sale.
  314. Computec Solutions Ltd (Deal 29)
  315. From Mr Reardon's evidence we find:-
    (i) Computec was VAT registered on 1 November 2004. It made nil VAT returns for the six periods up to 28 February 2006;
    (ii) a Regulation 25 notice was served requiring a return for the period 1 March to 5 May 2006;
    (iii) 17 VAT assessments or amendments thereto were issued to the company in the period July 2006 to April 2008;
    (iv) no appeal or other communication has been received by HMRC from the company since 5 May 2006 other than in connection with its liquidation;
    (v) a winding up order was made against Computec on 24 January 2007;
    (vi) the assessments made on Computec related to sales invoices issued by the company and obtained by HMRC officers visiting Computec's customers. The assessments include VAT in respect of Computec's sale in Deal 29. They have not been appealed;
    (vii) the assessments total some £105m and relate to invoices to the value of over £600m issued in the period 3 April to 9 May 2006;
    (viii) Computec purchased the goods it sold in this period from suppliers in the EU.
    We find that Computec intended dishonestly to evade its VAT liability in relation to the sales made between 3 April and 9 May (thus including that on Deal 24). It lay dormant for almost two years and then completed £600m of sales in a month and failed to account for, or return, the VAT on these, or otherwise to communicate with HMRC. The VAT evaded does not fall to be reduced by input tax. These were not the actions of a honest trader.
  316. Park Supplies/SS Enterprise Ltd (Deal 30).
  317. As we note in our findings on Deal 30, we found S&I's sale traced back to Park Supplies. Mr Ross said the chances were that it traced back further to "SS Enterprises" but we were unable to find that this was proved. At section 23 below we deal with whether or not "SS Enterprise GB" was a defaulter. We conclude that it was. There was no evidence however that Park Supplies was a defaulter.
    We do not find it proved that Deal 30 traced back to a defaulter.
  318. Bullfinch Systems Ltd (Deals 33, 35, 36, 41, 48, 50, 54)
  319. From Mr Laing's evidence we find:-
    (i) Bullfinch was VAT registered on 1 August 2005. Its declared business was software and security information with an expected turnover of £1m;
    (ii) Bullfinch agreed to provide security for VAT of £43,000 in October 2005;
    (iii) Bullfinch submitted VAT returns for 10/05 and 01/06 declaring sales of about £7m and £1.5m respectively. No return was made for 04/06. A Regulation 25 notice requiring a return for the period to 11 May 2006 did not result in the rendering of a return;
    (iv) 29 assessments have been issued to Bullfinch in respect of the period 1 February to 1 May 2006. They amount to some £54m. The assessments were made between May 2006 and May 2008;
    (v) no appeal or other communication has been received by HMRC in respect of the assessments other than the letter of 12 May referred to below. They have not been paid;
    (vi) in connection with the service of the Regulation 25 notice HMRC indicated that they considered some £2m of VAT was due and were deregistering Bullfinch. The director of the company wrote to HMRC on 12 May 2006 expressing surprise at HMRC's conclusions and saying that when the returns were completed the VAT would be paid. He takes offence at HMRC's suggestion that he would not pay VAT due. He indicated a desire to appeal and that he was considering taking legal advice. On 23 May a telephone call from a person representing himself as Bullfinch's bookkeeper indicated that the papers would be ready by 30 May.
    (vii) no further communication was however received from Bullfinch. A letter addressed to its declared place of business was returned in May 2007. A letter written to the director's home address in September 2007 resulted in a letter from, and then a call to, his sister who said that she and his family had been trying to contact him without success for a couple of years. In May 2008 officers of HMRC were told that Bullfinch no longer occupied offices at its declared place of business and had left in 2007 or before without leaving a forwarding address.
    (viii) Bullfinch's records were seized by HMRC but Mr Laing had been unable to locate them.
    (ix) the assessments were mainly based on information from other HMRC officers who indicated that Bullfinch had made supplies to traders they were responsible for, and on documents from freight forwarders indicating the import of the goods sold. An early assessment was based on information linking sales to imports of the goods by Bullfinch.
    The initial periods in which VAT returns were made, and the initial response of Bullfinch's period to HMRC requests and requirements did not obviously point to people intending to evade VAT liability. But in the light of the later failure to communicate, of its removal from its premises, and of the untraceability of its personnel, its actions look to us more like the actions of people trying to put off investigation while they arrange to disappear without paying the VAT due.
    Mr Beal suggested that HMRC's seizure of Bullfinch's records explained its failure to complete VAT returns. That may be the case (even if, as Mr Laing suggested, the seizure occurred four or five months after the return was due), but we would have expected an honest trade in that position to make representations to HMRC that that was the case, and to make enquiries about how to calculate and pay VAT due.
    We conclude that Bullfinch dishonestly intended to evade the VAT payable in respect of its activities in February to May 2006.
    Mr Laing indicated that none of the assessments made against Bullfinch covered any of the Deals specified above. We deal with this issue in the section of our decision I(h): "Tax Loss".
    The release notes obtained by HMRC from freight forwarders indicated that Bullfinch was importing in April 2006 from EU suppliers on a large scale. The evidence before us of these release notes did not include any relating to the specified deals, but we conclude that it is likely in its circumstances that the phones which were the subject of the specified deals were also imported by it and accordingly that the VAT it evaded and intended to evade was the whole of the VAT on its sales.
  320. MidWest Communications Ltd (Deal 36)
  321. From Mr Bycroft's evidence we find:
    (i) that MidWest's acquisition and sale in Deal 36 was one of a number of importations by it of mobile phones by it and their later sale to UK customers (in the case of Deal 36 this was corroborated by documents supporting the deal sheets and indicating a purchase from a non-UK company);
    (ii) the Deal 36 transaction and other such transactions were reflected in MidWest's VAT return for the period to 2 May 2006;
    (iii) also reflected in that VAT return were nine transactions which were documented as the purchase of phone cards from a UK company and their export to an Italian company;
    (iv) the input tax credit on the phone card deals of some £58m broadly cancelled the VAT due on the mobile phone transactions leaving VAT due of £8k odd;
    (v) MidWest had given instructions to its customers in the mobile transactions to make payment to a third party. These payments were about 17% above the purchase price of the mobiles. It is assumed that they more than satisfied MidWest's obligations on its purchases, but they left MidWest without the funds to pay the VAT due on the mobile sales;
    (vi) the transactions in the phone cards were fictitious: they were said to be purchased from a company whose details were wrong, said to be sold to a saddlery company, and were cards of a type which did not appear to be sold by the card manufacturer;
    (vii) HMRC assessed MidWest for £58m odd – disallowing the input tax credit on the claimed phone card transactions;
    (viii) MidWest has not paid the assessment nor appealed against it. The company was put into liquidation on 11 August.
    We conclude that MidWest dishonestly intended to evade the payment of all the VAT attributable to its mobile phone sales including that for Deal 36. It carried out a plan so to do by claiming fictitious input tax. Its arrangements for payment for the mobile phone sales left it without funds to pay the VAT on them absent fictitious input tax credit It may have had a right to repayment from those it overpaid but its failure to pursue any such right suggest clearly to us that its intention always was to evade payment of the tax.
  322. LTH Limited (Deal 45)
  323. From Miss Bushby's evidence we find:-
    (i) LTH was VAT registered on 9 August 2004. It declared its business as retailing children's wear. On 27 July 2005 its accountant wrote to say LTH had ceased trading on 4 July 2005 and asked for its registration to be cancelled. Then on 17 November 2005 a director of LTH wrote requesting that the registration should not be cancelled and indicating a change of address.
    (ii) that address was visited 6 months later on 17 May 2006. There was no trace of LTH : the premises had been taken over by someone else in mid-April.
    (iii) on 22 May a Mr Tilstone phoned Miss Bushby and said he had purchased the company and its business now included trade in mobile phones buying from inside and outside the UK. Mr Tilstone has not been in contact since. Mr Tilstone gave various addresses in that conversation. HMRC's letters to those addresses have been returned "not known".
    (iv) details of sales made to other traders by LTH in the period 1 April to 19 May and from freight forwarders indicated that LTH had been importing mobile phones and selling them in the UK. No documentary evidence of such imports was put before us. Assessments on LTH were issued to LTH in respect of these transactions. The assessed VAT was some £27m (representing some £154m + VAT of sales. The assessments include VAT in respect of LTH's sale in Deal 45.
    (v) no appeal against, correspondence in relation to, or payment of those assessments has been received by HMRC.
    (vi) a winding up order was made for LTH on 29 November 2006.
    We conclude that between 1 April and 19 May LTH conducted transactions in mobile phones with the intention of dishonestly evading the VAT therein: the sudden increase in its activity is not on its own proof of evasion, although it is to our minds somewhat suspicious, but its failure to return or pay the VAT and the disappearance of those involved coupled with this sudden trade, convinces us of its fraudulent intent to make hay while it was VAT registered but not to pay the VAT.
    The only evidence of import was Miss Bushby's evidence that information from freight forwarders indicated some imports, and the reported comment of Mr Tilstone that the business included buying from outside and inside the UK. We do not find ourselves able to conclude that LTH imported in Deal 45 or that there was an earlier fraudulent importer in the chain.
  324. SS Enterprise GB Ltd (Deal 46)
  325. From Mr Saul's evidence we find:-
    (i) SS Enterprise was VAT registered on 3 April 2006 and deregistered on 11 May 2006.
    (ii) a regulation 25 notice was served requiring a VAT return for
    this shorter period. The return was completed and timeously given to HMRC.
    (iii) Mr Saul's examination of paperwork taken from SS Enterprises on 11 May indicated that the return was not accurate. He made an assessment for £4.4m odd. This related to the import and sale to UK customers of mobile phones. Later information on the company's sales caused a reassessment to £14.2m odd. That assessment includes the VAT in respect of Deal 46.
    (iv) When Mr Saul visited the company's premises in September 2006 they appeared empty and uninhabited.
    (v) No appeal has been made against the assessments. No amount has been paid.
    We note that Mr Saul's evidence refers to "SS Enterprise GB Ltd" whereas the deal documentation referred to "SS Enterprise Ltd". Both however appeared to have the same address and we concluded that it was likely that the deals were done by SS Enterprise GB Ltd.
    It seemed clear to us that the officers of SS Enterprises used the period of its registration to make VATable supplies in respect of which they had no intention of paying VAT. Once contacted by HMRC it shut up shop and effectively disappeared. We find that it dishonestly intended to evade payment of VAT, and did evade such payment, in respect of Deal 46.
    We find that SS Enterprise GB purchased the phones in Deal 46 from a Cypriot company, MacDelta. The documentation before us did not contain any copy of an invoice from MacDelta. Mr Beal suggest that without a copy of that invoice we cannot tell whether the goods were imported from Cyprus (or any other country) or whether MacDelta provided SS Enterprise GB with a valid UK VAT invoice which would have enabled it to reclaim input VAT. The only evidence before us as to the place of MacDelta's supply was that MacDelta had a Cypriot VAT registration and a spreadsheet indicated it was a Cypriot company. Whilst this suggests that MacDelta was operating in and from Cyprus it leaves open the possibility that, like Premiston in relation to USM (see 16 above), MacDelta was selling to USM goods which were already in the UK. On that evidence alone, we do not find it possible to conclude that it was more likely than not that SS Enterprise GB imported the goods or that MacDelta did not supply a valid VAT invoice. As a result we can only conclude that the VAT it evaded and intended to evade was the VAT on its margin.
  326. Cyberweb Limited (Deals 47, 49, 51, 52, 55, 56 and 61)
  327. From Mr Dean's evidence we find the following facts which in our view point towards a fraudulent intent by Cyberweb or a person acting in its name to evade the VAT attributable to its sales in the specified deals:-
    (i) Cyberweb was VAT registered on 1 April 2004. Its initial business appears to have been providing technical support and training for computers. It submitted nil VAT returns for the periods 08/04, 11/04, 02/05 and 05/05. No further returns were submitted;
    (ii) Between 12 and 16 May 2005 Cyberweb, or a person acting in its name and using its VAT number made supplies of mobile phones. These supplies included those relevant to the specified Deals. The supplier of the phones related to the deals had imported them;
    (iii) in the period 25 May 2006 to 18 October 2007 assessments were issued in respect of the supplies under the specified Deals and other supplies. Some 8 assessments were sent to Cyberweb. Later seven were cancelled and re-addressed to the "person purporting to be Cyberweb". These seven were for a total of some £3.6m, the remaining one for £1.078m (which was the first assessment and made on 25 May 2006). Save as noted below no communication was received from Cyberweb in relation to these assessments. None have been paid;
    (iv) the documentation on which the assessments were based was either invoices bearing Cyberweb's name held by those to whom the supplies had been made or release notes collected from freight forwarders bearing Cyberweb's name and instructions;
    (v) in November 2006 (at a time when three of these assessments had been issued) HMRC instigated proceedings to wind up Cyberweb;
    (vi) a director of Cyberweb, Mr Bates, told HMRC officers on 11 May 2006 that he had no intention of entering the mobile phone market; on 16 May he told the officers that when he had spoken on 11 May the deals had not been finalised and identified the customer in the deals which gave rise to a £1,078k assessment. He is also recorded as saying, in relation to those deals (where the phones had been imported from Premiston and sold to Horus) that Horus had paid Premiston and the "commission" due to Cyberweb had not been received;
    (vii) save as noted below no response from Cyberweb or its directors has been received by HMRC to letters written after 18 April 2007.
    Mr Dean told us, and we accept, that there were differences between the documentation which supported the first assessment (25 May 2006 for £1,078k) and that for the others. Those differences, and differences between the freight forwarders used in the first assessment deals and the later deals pointed, in our view, to another person's involvement – and possibly a person not authorised to act on behalf of Cyberweb – in those deals. Of the Specified Deals we find all but Deal 52 are included in the first assessment (Mr Dean could not find the schedule showing this in the bundles before us but we accept his evidence that they were). Deal 52 is included in the assessments eventually issued to 'the person purporting to be Cyberweb'.
    We find that there were discussions between HMRC and various of Cyberweb's officers on 10 May, 11 May, 12 May and 16 May 2006. Thereafter there was nothing until April 2007 when, on 13 April a director, Mr Williets, called to speak about an assessment he had received for Cyberweb, and on 16 April another director, Mr Bates, returned HMRC's call (made to a number given by Mr Williets). Mr Bates claimed the mobile phone deals were nothing to do with him and agreed to suggest a date and time for a meeting. That he did on 18 May 2007 suggesting a meeting on 20 May 2007. Mr Dean did not respond to Mr Bates' suggestion. Mr Dean did write five months later, in October 2007, to Mr Bates requesting a meeting but no response has been received.
    Mr Bates willingness to meet HMRC on 20 May 2007 suggests that he was not attempting to disappear and that he wished to contest HMRC's assessments and its allegations that it had made supplies of mobile phones. But his earlier concession in relation to the Premiston-Horus deals in mobile phones is inconsistent with a denial of any involvement. Also we find that when an indication of his involvement on behalf of Cyberweb was put to him in May 2006 he gave at best an equivocal reply. We think that it is more likely than not that Mr Bates in his capacity as director of Cyberweb, and as its agent, consented to the deals which are reflected in the £1,078k first assessment. We think it likely that Mr Bates either knew that the arrangements for those deals were such that the VAT relating to them would go unpaid, or that another person so knew and persuaded him to do the deals thinking that Cyberweb would be liable for VAT on the "commission" only. His later denial of involvement and his rather shifty dealings with HMRC in May 2006 suggest to us that the first of those alternatives is the more likely. We therefore find that Cyberweb intended dishonestly to evade, and did evade, the VAT attributable to all the specified deals other than Deal 52.
    For deals 49 and 51 there were release notes from Premiston, an Estonian company to Hawkes directing the release to Cyberweb of phones allocated at Hawkes to Cyberweb. For the reason set out in relation to USM and Premiston at section 16 of this Appendix we conclude that Cyberweb had no right to input tax credit in respect of these purchases.
    It seemed to us likely that Cyberweb obtained the phones in Deals 47, 51, 55 and 56 in similar circumstances and that accordingly no input tax credit is available to it.
    As regards Deal 52 we find that either other persons acting for Cyberweb authorised it, or that persons purporting to act for Cyberweb (with or without Mr Bates consent) authorised it and that they did so with the intention of evading the VAT attributable to the sale (in relation to which we find there was no attributable input tax).
  328. Tresle Limited (Deal 53)
  329. From Mrs Okola's evidence we find:-
    (i) Tresle was VAT registered on 9 August 2005 and declared an expected trade in power tools and accessories;
    (ii) at a visit on 24 January 2006 Mr Aziz, ,the director, indicated that the expected business had not taken off but he had carried out some deals for the sale of new cars. He was warned not to participate in third party payments, and said he would not become involved in them;
    (iii) at a visit on 16 May 2006 Mr Aziz told HMRC's officers that he had started dealing in mobile phones the previous week and had done 62 deals so far. He said the goods he traded in were held at freight forwarders and that payment instruction from his suppliers were sent on to his customers. He expected to be paid by his supplier (rather than his customer – which was odd) but he had received no payment to date;
    (iv) invoices and other paperwork collected from Mr Aziz on 16 May indicated that Tresle had acquired the phones from a Cypriot supplier and invoiced UK customers whom it had instructed to make payment as requested by its supplier. The payment instructions left Tresle with insufficient funds to pay the VAT on its sales;
    (v) an assessment was made on Tresle for some £9.9m on 5 June 2008 for deals including those evidenced by Tresle's records. This includes Deal 53. Further assessments were made between 28 March 2007 and 20 March 2008. None of the assessments have been appealed. The outstanding amount is £13.1m. No payment has been made.
    The records of HMRC's officers produced by Mrs Okola indicate that at one time Tresle was considered to have been a buffer, buying from UK suppliers, but the evidence of the documents taken from Tresle indicate that it was importing from non-UK suppliers. The documents supporting the deal sheets indicate that in Deal 53 Tressle purchased from MacDelta, a Cypriot company. We conclude therefore there was no input tax to be credited against its output liability.
    We find either that Tresle engaged in the sales it made in May 2006, including Deal 53, knowing that it would not have the funds to pay the VAT on its sales and intending to evade that VAT or that someone else dishonestly engineered that result.
  330. AAA Multilink Ltd (Deals 57, 60, 62 and 65)
  331. From Mrs Parsons' evidence we find:-
    (i) AAA was VAT registered on 1 October 2005 declaring an intended business of online retail furniture and wine. Its director was Mr Stumner and the company secretary was Charlotte Schofield;
    (ii) AAA delivered a VAT return for its first period (12/05) but has delivered no other VAT returns;
    (iii) a colleague in a different part of HMRC told Mrs Parsons that documents from a freight forwarder suggested that AAA appeared to be dealing in mobile phones. On 26 May 2006 she made a visit to AAA at which a Regulation 25 notice requiring a return to that date was served. AAA was deregistered on the same day;
    (iv) a few days later, on 30 May Mr Stumner phoned Mrs Parsons and arranged a meeting with HMRC for 31 May. In that telephone call he told Mrs Parsons that he had begun to trade in mobile phones and his suppliers and customers were in the UK. The meeting was postponed and took place at AAA's accountant's offices on 7 June;
    (v) At the 7 June meeting Mr Stumner said he had intended to begin trading in mobile phones but had not done so;
    (vi) On the basis of invoices obtained by HMRC from Fonedealers which were addressed by AAA to Fonedealers HMRC assessed AAA for £7.3m VAT on 12 September
    (vii) on 15 September 2006 Mr Stumner phoned Mrs Parsons querying the assessment and three days later sent a fax in which he said he had not undertaken any of the assessed transactions;
    (viii) Mrs Parsons was unable to obtain any other evidence that the transactions represented in the assessment had been undertaken by AAA. She withdrew the assessment and raised an assessment for the same amount addressed to "the taxable person purporting to be AAA";
    (ix) each of the specified deals is reflected in the assessment made.
    The Fonedealers' documents indicate fairly clearly that either AAA were supplying the mobile phones or that another person was supplying them using AAA's details. If AAA were supplying them then Mr Stumner's representations were clearly intended to evade the payment of the related VAT by AAA; if the supplies were by another person then it seems to us likely that that person was fraudulently intending to evade VAT on the sales.
    Mr Beal notes that Mrs Parsons said in evidence that no steps had been taken to contact the company secretary Charlotte Schofield who had signed the VAT registration form. We do not believe that HMRC's failure so to do affects our conclusion in the previous paragraph. It seems unlikely to us that Charlotte Schofield intended to include any of the sales on a VAT return or to pay the related VAT.
    Mr Beal also notes that whilst the invoices obtained from Fonedealers were exhibited there was no exhibit of a genuine AAA invoice. Thus it could not be determined whether or not the invoices obtained were genuine. However in our opinion this does not affect our conclusion above: if the invoices were genuine, AAA was fraudulent; if they were false another person was.
    The only evidence before us as to the possible source of the phones supplied was:
    (i) Mr Stumner's comment recorded on 31 May 2006 that his suppliers and customers were in the UK (see (iv) above). Mr Stumner's statements to HMRC appear to have been somewhat contradictory so we are not inclined to place too much weight on this; and
    (ii) the documents obtained from the freight forwarders which indicate that AAA (on the person purporting to represent AAA) was importing the phones from Premiston.
    We conclude it is likely either that the phones were imported prior to this sale by AAA (or the person acting in its name) and as a result no input tax credit is available to reduce the tax assessed or that the sales to AAA did not carry an input tax credit for reasons similar to those relating to Premiston discussed at para 16 of this section above.
  332. Deal 58: no specified defaulter. See our conclusions in relation to the deal chain on Deal 58.
  333. 3D Animations Limited (Deals 59, 63 and 64)
  334. (i) 3D was VAT registered on 3 May 2006 with a declared intended business of design multimedia and animation graphics;.
    (ii) four weeks later on 1 June 2006 an HMRC officer visited its declared place of business because other officers had reported that substantial amount of mobile phones had been imported by 3D from EC companies. There was no-one present. Letters were left giving a Regulation 25 notice requiring completion of a VAT return to 1 June, warning of deregistration, and requesting business records;
    (iii) no reply to any of those letters and no other communication from 3D has been received by HMRC since 1 June 2006. Letters addressed to its director have also not been replied to;
    (iv) on the basis of information obtained by other officers from other traders; assessments were made on 3D for a total of about £130m representing £876m of sales in the period 3 May 2006 to 1 June 2006. No response or appeal in relation to these assessments has been made;
    (v) the assessments include the VAT in respect of the sales said to be made by 3D in the specified deals;
    (vi) 3D animations went into compulsory liquidation on 20 September 2006.
    The documents from the freight forwarders and the evidence of other traders' purchases convinces us that it is likely that 3D made the supplies assessed. We think, in view of the almost immediate disappearance of its officers, that the supplies were made by 3D rather than by another person in its name.
    The evidence of the freight forwarders' documentation indicating the release of goods by a Cypriot trader to 3D convinces us that it is likely that the phones sold by 3D were imported by it and that the VAT due on its sales was not reduced by any input tax credit.
    We conclude that 3D dishonestly intended to evade, and did evade, the VAT attributable to its sales in the specified Deals.
  335. Crossview Consortium Limited (Deals 66 and 73)
  336. From Ms Hirani's evidence we find:-
    (i) Crossview was VAT registered on 17 October 2005. Its declared intended business was software consultancy and development;
    (ii) at a visit to Crossview on 27 February 2006 officers of HMRC were told that it intended wholesale trade of electrical products. The record of the conversation indicated that trade in mobile phones was not excluded;
    (iii) Crossview submitted a VAT return for the period to 31 March 2006;
    (iv) on 2 June 2006 Ms Hirani visited the declared address of Crossview. She spoke to a receptionist who told her that she had not seen anyone from Crossview for three weeks. Ms Hirani then spoke to the premises manager who told her that a Mr Arekapadi, (a director of the company who had some three months previously given notice to Companies House of his resignation) was outside the UK for 3 weeks. He gave Ms Hirani office and mobile contact numbers for Mr Arekapadi. Ms Hirani rung the office number and was told by a Mr Khan that Mr Arekapadi was outside London for the day but would be back. Ms Hirani asked Mr Arekapadi to be told that she had delivered a Regulation 25 notice and that a VAT return had to be completed by 5 June;
    (v) on 5 June, when she went to Crossview's place of business, no one was available;
    (vi) between August 2006 and April 2008 a number of assessments and amendments to assessments were issued to Crossview and letters sent to Crossview;
    (vii) letters sent by HMRC's place of business were returned "moved away July 2006" or "not at this address";
    (viii) Crossview's former accountant, when contacted by HMRC in January 2007, said she had not heard from Crossview since submitting its 2004 accounts;
    (ix) attempts to get alternative addresses from Crossview from its bank and one of its customers have been unsuccessful;
    (x) no contact, and no appeal against the assessments has been received from Crossview since 2 June 2006;
    (xi) the assessments made include amounts in respect of Crossview's sales in Deals 66 and 73. The assessments were based on information from Crossview's customers. There was no direct evidence before us that Crossview imported the phones;
    (xii) no assessment has been paid and HMRC record an outstanding debt of some £5m.
    In June 2006, the Companies House records indicate that the company secretary and its sole director was Winko Palan.
    There was no evidence that HMRC had attempted to contact Winko Palan in June 2006. Although letters were sent to him in and mid-May 2008, there was no specific correspondence with Winko Palan asking about the company's activities in June 2006.
    The failure of any officer or employee of Crossview to get in touch with HMRC or to attempt to challenge the assessments or to provide a VAT return, and the evidence that the company was selling goods convinces us that the company had no intention of paying the VAT attributable to these sales. Whilst an attempt by HMRC to question or contact Winko Palan in June 2006 could have been useful to HMRC and the response might have been relevant to our conclusion, ample opportunities remained thereafter for the company to contact HMRC, and those opportunities were not taken.
    There was no evidence directly relating to the source from which Crossview acquired the phones. But we saw copies of payment instructions from Crossview's customers in Deals 66 and 73 given to their customers requiring that payment in respect of their transactions in these deal chains to be made respectively to Danish and Swedish entities. Those instructions in our view make it likely that the phones were either imported by Crossview or by a predecessor in the chain of which Crossview was the first identified member.
    HMRC ask us to conclude that, if Crossview did not itself import the phones, the importer of the phones fraudulently evaded the payment of VAT on its sale. In this case we believe that that is the more likely conclusion: the third party payment instructions and Crossview's evanescent appearance lead us to the conclusion that either Crossview imported the goods and intended to evade all the VAT on its sale, or that there was an earlier importer who evaded and intended to evade the VAT on its sale.
  337. DBP Trading Limited (Deal 67)
  338. From Mr Strachan's evidence we find that DBP, having previously traded in the building sector, started to trade in electrical commodities wholesale. Sales rose from some £52k per quarter (05/05) to £100 million per quarter (02/06). The returns and records for the periods up to 02/06 indicate that DBP was buying from UK suppliers and selling on to UK customers (acting, HMRC would say, as a buffer). In this period it made third party payments at the behest of its suppliers, and dealt with persons who had hijacked other traders' VAT numbers.
    On 27 June 2006 a Regulation 25 notice was served, and there followed communications with DBP's bookkeeper who appears to have been cooperative with HMRC, supplying business records and analysis. The bookkeeper indicated however that a large part of the records were missing as they had been stolen. (We make no finding that they were stolen.) The records and analysis identified many back to back sales and purchases, but purchase information was missing in respect of 100 deals in the period February to May 2006.
    In his discussion of the assessments made by HMRC on DBP (it appears that no returns were made after the 02/06 return), Mr Strachan noted two deals in which CPUs were bought by DBP from a non-UK supplier, but gave the impression that at least in the first months of 2006 DBP generally seemed to be purchasing from UK companies rather than importing the goods.
    Deal 67 was invoiced on 1 June 2006. The documents and analysis obtained from the bookkeeper provide no information as to whether goods were imported or not. An assessment was made on DBP which included tax in respect of the sale in Deal 67 but so far as we could tell without any credit for corresponding input tax.
    Correspondence (of sorts) continued between HMRC and one of the company's directors until March 2007, and in June 2007 a winding-up order was made against the company. The assessments made and outstanding (including that in respect of Deal 67) totalled some £61m in September 2007.
    It seems to us quite likely that up to February 2006 DBP bought and sold within the UK making a small margin on its transactions and intending to pay the VAT on that margin. Mr Stachan's evidence suggested that DBP was not particularly diligent or discerning in its appraisal of its business partners and their practices and may have turned a blind eye to their imperfections, but our impression is that it expected to be liable for and intended to pay VAT on its margin.
    After February 2006 it is clear that its transactions greatly increased, and that it entered into transactions in which it purchased from non-UK suppliers. But it was by no means clear to us that even a majority of its purchases so emanated. The £61m debt includes assessments made in respect of sales for which no input tax credit was given as the result of an absence of records rather than as the result of records indicating importation.
    Were it proved that DBP had imported the phones in Deal 67 then on the evidence before us as to the behaviour of one of its directors we would have concluded on balance that it intended to evade the whole of the VAT on that sale. But if, as seems equally likely, the sale of phones in Deal 67 was of phones which it had not imported, DBP's liability would have been on that margin then the maximum liability it could be said to have evaded was that on that margin (unless it were shown, which it was not, that it knew or should have known of fraud earlier in the chain such that it would not be entitled to its input tax).
    On Mr Strachan's evidence, a director of DBP wrote to HMRC in January 2007 and March 2007 indicating respectively a change of address for DBP and that it had ceased to trade. Thereafter no further communication was received from him. The same director is reported as providing the bookkeeper with a bag of records late in 2006. But some of those records were missing as the result of what was said to be a theft; we find it suspicious that business records were said to be stolen. It seems to us possible that following the increased sales and HMRC's attention to DBP in June 2006 and thereafter, the director took fright felt able only to provide details of changes in address and cessation of trading, and relied on the bookkeeper to deal with the accounting for VAT; but did not feel able to contest the large assessments made by HMRC. But on the whole it seems to us more likely that after March 2006 DBP did start importing phones and knew that its VAT liabilities would significantly increase (because of the absence of input tax credit) but had become involved in crooked practices from which it could not extricate itself and as a result of which it would not be able to pay the VAT.
    We find it proved that DBP intended to evade its VAT liability on the margin it made on Deal 67 but on the margin only
    The Respondents submit that if DBP was not an importer then the evidence makes it likely that it was the purchaser in a deal chain starting with an import in which the importer fraudulently evaded the VAT on its sale. We do not find that the evidence makes this conclusion in relation to Deal 67 more likely than not.
  339. Subumma Limited (Deal 68)
  340. From Mr Patterson's evidence we find:-
    (i) Subumma was VAT registered on 1 November 2002 declaring its business to be "International Simple Voice Resale (telecom)". It submitted VAT returns for the five periods up to that ending on 31 December 2003 (but see (x) below);
    (ii) Subumma had a single director who was in contact with HMRC in 2003;
    (iii) in 2005 and early 2006, letters sent by HMRC to Subumma's address were returned "addressee unknown";
    (iv) papers collected from Freight Forwarders by HMRC indicated that in May and June 2006 phones were being released to and by Subumma; these records indicated that Subumma was buying from UK suppliers and selling to UK customers;
    (v) a letter sent to Subumma on 9 January 2007 indicating that an assessment in respect of sales to Fonedealers was to be issued was returned "Not here";
    (vi) further assessments were made in 2007, the later ones being sent to the director's home address;
    (vii) the director wrote to HMRC on 11 February 2008 following a telephone call. The director indicated no knowledge of sales to Fonedealers;
    (viii) following attempts by HMRC in mid May 2008 to contact the director, he visited HMRC's office on 20 May 2008. He was shown an invoice purporting to come from Subumma to Fonedealers and said it was a forgery. A visit to the director's home was arranged to view the records held;
    (ix) following the visit Mr Patterson concluded that Subumma's VAT number had been 'hijacked'. This conclusion was reinforced by changes made in the Companies House records for Subumma in which, under a suspect hand, details of a home address had been given for the directors at which it was established that they did not live;
    (x) HMRC deregistered Subumma on 8 June 2006 and received no VAT returns for the periods 03/04 to 03/06. Assessments which had been made on Subumma on the basis of Freight Forwarder's records and other information, were cancelled. There is no recorded outstanding debt due to HMRC from Subumma;
    (xi) the VAT in respect of the sales purportedly made by Subumma, including the VAT in respect of the sale in Deal 68, is some £1.2m. This amount would be reduced by any input tax relevant to purchasers by the supplier of these goods. The information held by HMRC suggests that the purchases relevant to these sales were from UK suppliers so that input tax credit could be available.
    In our judgment the behaviour of the director of Subumma did not indicate that Subumma had fraudulently intended to evade the VAT attributable to the sale attributed to it in Deal 68. We concur with Mr Patterson's conclusion that it is likely that another person hijacked Subumma's VAT number and acted in its name but without authority.
    Our conclusion that it was likely that someone hijacked Subumma's VAT number leads us to the conclusion that that person was acting dishonestly and with the intent to evade VAT. We therefore find that there was a fraudulent intention to evade VAT in relation to the sale in Deal 68 to Fonedealers purportedly by Subumma.
    Mr Davis-White and Mr Robertson submitted that although there was no direct evidence that the mobile phones the subject of Deal 68 were imported from the EC, all the surrounding circumstances including the chain of deals in the deal sheet entitle us to infer that they were. He says that the transaction has all the hallmarks of a classic MTIC fraud and a relevant tax loss was caused by it.
    The evidence before us (see (iv) above) suggested, however, in relation to that sale that an input tax credit would have been available (unless the supplier knew or ought to have known of fraud earlier in the chain – as to which there was no evidence) and thus that the VAT intended to be evaded was prima facie that on the margin only.
    The Respondents effectively invite us to conclude that because there was some fraud by the hijacker, and because there was a chain, it is likely that there was fraud earlier in the chain in relation to a sale by an importer. Whilst absent other evidence, the dishonesty involved in hijacking a number is likely in our view to extend to a dishonest scheme to avoid all the VAT so that it is likely that VAT was evaded by the importer and that evasion was known about by the hijacker (why else hijack a VAT number?), the evidence here was different – see (iv) above.
    We conclude that fraud is established only in relation to the margin made on the relevant sale.
    Mr Beal suggests that no tax loss is established because no assessment has been made by HMRC. For reasons set out earlier, we do not agree.
  341. Rukford Limited (Deals 69 and 72)
  342. The Respondents contend that Rukford's VAT number was hikacked and that the supplies attributed to Rukford in the Deal sheets for the specified deals were made by a person purporting to be Rukford who intended to evade the VAT attributable to those supplies.
    From Mr Dean's evidence we find:-
    (i) Rukford was VAT registered on 27 October 2003 declaring an intention to trade in building and construction;
    (ii) the company's sole director was a Mr Forden. Mr Forden gave the company's address as being the same as his own, 5 Ullswater Road, SE17 0AL, but he moved at sometime before July 2005, to Lordship Lane, SE22 which he left at some time before January 2008. HMRC's attempts to trace him since 2005 have been unsuccessful;
    (iii) the company was deregistered from 12 April 2008;
    (iv) the director of UK Communications, the person to whom Rukford purportedly supplied phones in Deals 69 and 72 provided HMRC's officers with Rukford's VAT number which he said had been given to him orally. UK Communications held invoices said to come from Rukford but:-
    (a) "Rukford" was spelled "Ruckford"
    (b) the invoices did not carry a VAT number.
    The director also indicated that he had visited Rukford's premises in West Norwood and dealt with a Mr Abbass. The transactions in question took place in Spring 2006 about a year after Mr Forden, Rukford's director, appears to have moved to SE22. There is no Companies House record of Mr Abbass being associated with Rukford.
    (v) release notes addressed to the Freight Forwarder, First Forward, from UK Communications and from Rukford were in similar format;
    (vi) an invoice to UK Communications from "Rukford" and release notes addressed to First Forward from "Rukford" contain 5 Ullswater Road (Rukford's first declared address) as the address of the company but show its postcode as SE27 6AL rather than SE27 0AL;
    (vii) release notes obtained from First Forward indicated that Rukford had been importing goods from a Cypriot company and on-selling them;
    (viii) Mr Dean concluded that Rukford's VAT number had been hijacked. He assessed "the person purportedly to be Rukford" with VAT of some £4m which derived from transactions including the sale in Deal 69 and 72. UK Communications was eventually treated as eligible for input tax credit in respect of the supplies purportedly received from Rukford (presumably on the assumption that it did not have the requisite knowledge of any fraud);
    (ix) the assessments were sent to UK Communication and first Forward for transmission to the assessed person. No reply or appeal has been received.
    We conclude that in relation to the phones which were the subject of Deals 69 and 72 someone was intent upon dishonestly evading VAT in relation to their first supply in the UK. That person may have been Rukford acting through the disappeared Mr Forden or another person authorised to act on its behalf, or may have been UK Communications purporting to acquire the goods from Rukford rather than directly from the supplier, or may have been another person representing himself as authorised to act for Rukford. Whoever it was it is clear to us that VAT was intended to be evaded.
    We also think it more likely than not that the goods were imported by Rukford (or one of the other persons mentioned above) and that the VAT sought to be evaded in relation to the relevant sale was the whole of the VAT on the sale.
    Mr Beal notes that HMRC permitted UK Communications input tax credit in respect of its purchase. There may have been arguments available to HMRC that UK Communications were not so entitled. This does not affect our conclusion.
  343. RS Sales Agency Ltd (Deals 70, 74 and 75)
  344. From Mrs Parsons' evidence we find:-
    (i) Mr Sodawala was a sole trader registered for VAT from 14 May 2003 trading under the name RS Sales Agency;
    (ii) On 20 February and 3 March 2006 Mr Sodawala requested that his VAT number be transferred to the company, RS Sales Agency Ltd. The business activities declared for the company included mobile phones and other electronic goods; the activities declared for Mr Sodawala had been limited to clothing;
    (iii) HMRC refused to transfer the VAT number;
    (iv) Mr Sodawala was advised to make an application for VAT registration of the company afresh. No application was received;
    (v) On 17 May 2006 Mr Sodawala telephoned HMRC to ask if he could import electrical equipment using his VAT registration number.;
    (vi) Mrs Parsons visited Mr Sodawala's declared place of business on 3 July 2006. No-one was present. She left a letter requesting contact. None was made. She wrote on 5 July 2006 to say his VAT number was cancelled. On the next day an accountant phoned on behalf of RS Sales Agency. Mrs Parsons asked him to get Mr Sodawala to contact her. He has not contacted her;
    (vii) Mr Sodawala made VAT returns for the periods 12/03 to 03/06 (except for 12/05 with 03/06 being a nil return). (In para 14 of her witness statement Mrs Parsons says no declaration was made after 1 Oct. 2005, but a copy return was included in our bundle for 03/06.)
    (viii) RS Sales Agency Limited issued sales invoices which bore Mr Sodawala's VAT number ;
    (ix) HMRC initially raised assessments against Mr Sodawala in respect of sales they understood to have been made by RS Sales Agency (or RS Sales Agency Ltd) but cancelled these assessments and, having allocated RS Sales Agency Ltd a VAT number, assessed that company instead;
    (x) a winding-up order was made against the limited company on 13 December 2006. The Official Receiver's report indicated that Mr Sodawala said that a third party was in control of the company. The Official Receiver had not been able to trace that person;
    (xi) the assessments made included VAT in respect of Deals 70 and 75 but there was no evidence that an assessment had been made in respect of Deal 74.
    Mr Beal says that by registering and assessing the limited company rather than Mr Sodawala, HMRC were the authors of their own loss since by doing so – by "transferring" the debt to the limited company – they rendered the enforcement of the debt far less likely since Mr Sodawala was more likely to have assets to satisfy the debt that was a £1,000 company. This seems wrong to us : if the company was the supplier, it was liable to VAT, if Mr Sodawala was the supplier, it, and not Mr Sodawala, was liable. The question is who was the supplier? On Mrs Parsons' evidence it seems more likely to us that it was the company.
    Mr Beal notes that Deal 74 does not appear to have been assessed. For the reasons we set out in section I we believe that to be irrelevant to the question of the tax loss; we find it does not affect our conclusion on the question of fraud.
    The failure of either Mr Sodawala or someone else on behalf of the company to contact HMRC in relation to the assessments or at all, the failure to return and pay VAT, the attempt to use the wrong VAT number, and the apparent decision to make supplies through a limited company rather than personally (with the consequent effect on personal liability), persuade us that it is likely that RS Sales Agency Limited did intend to evade the VAT attributable to the sales in the specified deals.
    There was no evidence before us as to whether the phones sold by RS Sales Agency had been imported or bought from a UK supplier, other than the evidence of Mr Sodawala's telephone call asking whether he could use his VAT number for the company's import of phones. That was evidence by Mrs Parsons of a note taken by another officer – a note we saw without seeing the officer in question. With hesitation we find that it is just about more likely than not that the phones in the specified deals were imported.
  345. JD Telecoms UK Limited (Deals 71, 83 and 84)
  346. The Respondents' case in relation to these deals is that J D Telecom's VAT number was hijacked by another person who fraudulently evaded the VAT on the sales.
    Mrs Quinn's evidence was that in June 2006 she went to J D Telecom's declared place of business, and found no one there and a "to let" sign above the door. She left a letter with a regulation 25 notice, but when she went the next day to pick the return up it was clear that the letter had not been opened. We accept that. However in July J D Telecom's accountants got in touch, objected to J D Telecom's deregistration (which had happened in the meantime), explained that they had received the regulation 25 direction after its specified due date. They later provided a VAT return for the period to 31 May 2006 claiming a VAT repayment of £2m. That claim is currently unpaid and is being verified by HMRC.
    Mrs Quinn told us that that verification exercise was started by her colleague officer Neale. Officer Neale has retired from HMRC and was not called to give evidence before us. Mrs Quinn said that in the period when officer Neale was looking into J D Telecom's affairs he was contacted by other officers of HMRC asking him to raise assessments in respect of supplies made by J D Telecom to the traders those officers were concerned with in relation to supplies made between 31 May 2006 and 30 June 2006. She told us that Mr Neale had issued two assessments but that J D Telecom's accountants had written to dispute the assessments saying that J D Telecom had not made the supplies alleged, and enclosing VAT return for the shortened period to 30 June 2006. We saw the letter from the accountants, which contained supporting documentation.
    Mrs Quinn told us that Mr Neale had investigated the supplies alleged to have been made by J D Telecom which gave rise to the disputed assessments and had written a report. That report was not before us. She said that in that report Mr Neale had concluded that those supplies had not been made by J D Telecom because:-
    (i) the invoice format and sequencing where goods had been acquired from the EC was different;
    (ii) the address and bank account details were different.
    For those reasons it appears that Mr Neale accepted J D Telecom's accountants protestations that the supplies had not been made by it but had been made by another person hijacking its VAT numbers.
    The assessments on J D Telecom were cancelled and fresh assessments issued to "the person purporting to be J D Telecom" and sent for onward forwarding to the Freight Forwarder associated with the relevant sales. These assessments total some £6.8m and include amounts assessed in respect of the specified deals. No communication or appeal has been received in relation to these later assessments.
    Mr Beal says that the Respondents' position is self-contradictory: they accept the accountant's remonstrations that J D Telecoms did not make the sales reflected in the assessments but appear to question their assertion in relation to the VAT reclaim of £2m for 05/06. He points to the absence of any direct evidence before us in relation to Mr Neale's decision on the assessments other than the accountants' letter.
    If the sales which were the subject of the assessments were in fact made by J D Telecom, then the accountants' representations indicate that someone on behalf of J D Telecom intended to evade the VAT in respect of them and probably intended to evade it at the time of sale. The evidence of a possible repayment claim does not affect this conclusion.
    If, on the other hand, the sales were not made by J D Telecom, then someone was in our opinion falsely representing themselves as acting for J D Telecom with the object of evading the relevant VAT.
    Either way, therefore, someone intended fraudulently to evade the VAT on the relevant sales.
    There was no evidence before us that the phones in the specified deals had been imported rather than purchased from a UK supplier, other than the second hand evidence that Mr Neale had said that the documentary evidence showed that where invoiced goods had been acquired from the EC the invoice format was different. On balance, and only just, we think this is enough to conclude that the goods were imported and therefore that the VAT evaded was not reduced by any input tax credit.
  347. DTM Provision Limited (Deals 80, 82, 86, 87)
  348. From Mr McBrine's evidence we find:-
    (i) DTM was VAT registered on 1 March 2006. Its declared business was making catering supplies. It was deregistered on 27 June 2006.
    (ii) DTM has made no VAT returns;
    (iii) the only oral contact with officers of the company (other than that referred to at (v) below) was four telephone conversations with Mr Robertson a director in January, February and March 2006;
    (iv) the company gave notice of a change in its place of business on 22 May. In the same month the Companies House records indicates that Mr Robertson resigned and a Mr Pearce was appointed director. In the previous month the company secretary was changed to Ian Shaw;
    (v) visits to the company's declared address, and a calling card left, in June 2006 resulted in a telephone call from a Mr Pearce who denied being a director of DTM although his date of birth was that on the Companies House register;
    (vi) HMRC have taken no action to trace Mr Robertson or Mr Shaw;
    (vii) assessments have been made on DTM for £23m in total. These assessments include the VAT in respect of the sales by DTM in the specified deals. The assessments were made on the basis of reports from other officers that sales were made by DTM;
    (viii) no appeals, payment or communication has been received in respect of these assessments.
    Mr Beal notes that no action was taken to trace Mr Robertson or Mr Shaw. He says that there is insufficient evidence of fraud.
    The sales reflected in the specified deals were invoiced in July 2006, some two months after Mr Robertson's declared replacement by Mr Pearce, and three months after the change in company secretary. It seems to us that if Mr Robertson truly ceased to be connected with the company in May 2006, then he had no role in relation to the July sales; and if he continued to act on behalf of the company after May and into July then (a) his apparent replacement by the seemingly unwilling Mr Pearce is odd if not suspicious, and (b) the failure of anyone to respond to the assessments made on the company and its deregistration are of greater significance than any failure to contact Mr Robertson.
    In our judgment the cessation of contact, the lack of returns, the making of sales of mobile phones so soon after registering as the mater of catering supplies, and the failure to respond to the assessments show that it is more likely than not that J D Telecom intended fraudulently to evade the VAT on the sales in the specified deals.
    The evidence as to whether J D Telecom was an importer in relation to the specified deals was, again, somewhat limited. It as confined to Mr McBrine's evidence that the assessments had "been raised on the basis that DTM acted as an importer", that, in about 160 deal chains investigated by HMRC, nearly every one had DTM at the start of the chain, but wherever a supplier to DTM was identified it had been an EU company. Taken with the lack of any appeal or communication from DTM in respect of the assessments we find this (just) enough to conclude that in the specified deals DTM imported the phones. As a result we find that it fraudulently intended to evade the whole of the VAT on its relevant sale.
  349. Vision Soft (UK) Limited ("VSUK") (Deal 76)
  350. From Ms L'Argents' evidence, which was confined to the production of information from HMRC's records, we find:-
    (i) VSUK was VAT registered on 15 March 2005 having declared its business to be software development and indicating that it would not generally supply hardware. It produced an company profile which seemed seriously to overstate the work it had carried out given that other declarations indicated it had not started trading;
    (ii) on 1 June 2006 its director changed and a new director was appointed, a Mr Shafiq;
    (iii) VSUK rendered VAT returns for the periods up to May 2006. All were nil returns;
    (iv) reports from HMRC officers indicated that VSUK made significant sales of mobile phones in July 2006. Copies of release notes addressed to freight forwarders indicated that VSUK was importing from the EU and selling to UK companies;
    (v) a visit by HMRC officers on 6 July 2006 to VSUK's declared place of business found that it was a serviced mail box in a block of flats;
    (vi) VSUK was deregistered from 13 July, 2006;
    (vii) assessments were issued to VSUK between December 2006 and October 2008. These assessments include amounts for VAT in respect of the specified dates. They total over £10m. None have been questioned, appealed or paid;
    (viii) VSUK was compulsorily wound up on 16 January 2008.
    Ms L'Argent indicated that there was evidence that in relation to transactions other than the specified deal VSUK had made third party payments to its director, Mr Shafiq. Even though the payment made was large (some £2.4m), we are not persuaded that making a payment to a director is an indication of fraud.
    We find it very likely that VSUK intended fraudulently to evade the VAT in respect of its sale in Deal 76. We find it likely that it made the supplies in respect of which it was assessed, and its failure to contact HMRC in any way after the sudden explosion of its sales and after July 2006 indicative of an intention not to pay the relevant VAT.
    The evidence that VSUK was an importer of the goods sold in Deal 76 was a release note from Silas BV, its supplier releasing the phones to Data Solutions Northern which was VSUK's customer's customer. We conclude that the VAT sought to be dishonestly evaded on its sale was the whole of its output VAT.
  351. Pearl Cosmetics Limited (Deals 77, 79 and 85)
  352. There is one feature of Pearl's case on which we should comment. Pearl was deregistered and assessed following information obtained from freight forwarders about its sales. It did not reply to assessments and buried its head in the sand. It was put into liquidation. When the Insolvency Service managed to contact the director it turned out that it was likely that the company's number had been hijacked. Thus this provided an example of a legitimate trader burying its head in the sand and not replying to HMRC correspondence.
    From Mr Ross' evidence we find:-
    (i) Pearl was VAT registered on 22 November 2004 declaring its business to be in cosmetics;
    (ii) the letter containing the VAT return form for 02/06 was returned to HMRC as returned mail;
    (iii) the VAT registration was cancelled on 23 November 2006 after abortive attempts by HMRC officers to contact its officers;
    (iv) documentation obtained from a freight forwarder indicated that Pearl was importing goods and selling on to UK customers. This documentation included that evidencing the chains in the specified Deals. Assessments were made on Pearl;
    (v) on 11 October 2007 there was contact from representatives of Pearl and, later, a meeting between HMRC officers and its director. The director explained that Pearl had only made one purchase - of cosmetics, of which it had sold some, and that no other business had been done;
    (vi) the release notes in the documentation from the freight forwarder included notes purporting to come from Pearl. But the signatory of the notes was not a recognised officers or employee of Pearl, and the address given for Pearl on the notes appears mis-spelled or misstated;
    (vii) the documents before us in respect of Deal 77 include payment instructions from RX Tech (Pearl's customer) to its customer RK Brothers to make payment direct to Koonmarket.
    There was no evidence before us that any assessment had been made by HMRC on any person other than Pearl in respect of the VAT attributable to the specified deals.
    Mr Beal suggested that the third party payment instructions recorded at (vii) above belied Pearl's involvement in the transaction. We do not so find. They are consistent with Pearl giving such instructions to RK Brothers.
    It seems to us that either persons acting with Pearl's authority entered into the release transactions but did not intend to declare the VAT thereon, or that another person fraudulently purported to act for Pearl with the intention of evading the VAT on the sale. In the former case the reaction of Pearl's director at the meeting noted at (v) above indicates that it is likely that the relevant persons were intending to evade the VAT by not telling the director about their actions, or not giving that information to HMRC. We conclude that the VAT in relation to the specified Deals was fraudulently evaded.
    The release notes from the freight forwarder indicates that the phones which were the subject of the sales by Pearl or a person purporting to be Pearl, were imported. Thus there is no relevant input tax credit. We find that someone dishonestly intended to evade and did evade payment of VAT in respect of the sales in the specified deals. For the reasons set out earlier we do not consider that the failure to make any assessment on any person other than Pearl in respect of these sales affects our conclusion as to fraud, or indicates that there was no relevant tax loss.
  353. Heathrow Business Solutions Ltd (Deal 78)
  354. From Mr Lewis' evidence we find:-
    (i) Heathrow was VAT registered in October 2004. Its declared business was IT and recruitment;
    (ii) since registration the company has not submitted a VAT return (Mr Lewis' evidence on this was at first unclear since his statement indicated that nil returns had been rendered for two specific periods but later said that it had never submitted a VAT return. Mr Lewis explained, and we accept that by his earlier reference to "nil returns" for the two periods he meant that nothing was sent in);
    (iii) on a visit to the company's declared place of business on 30 June 2006 there was no trace of the company and he was told it had never operated from those premises;
    (iv) release notes obtained by HMRC from freight forwarders indicate the import of goods by Heathrow and their transfer to UK companies in June 2006;
    (v) on the basis of these documents HMRC made a number of assessments on the company;
    (vi) no appeal or communication has been received by HMRC from Heathrow in relation to those assessments or otherwise. No payment has been made.
    Mr Beal made three points. First he noted that Mr Lewis accepted that his witness statement had been drafted by another person for him while he was away ill. Having heard Mr Lewis in person we were not persuaded that his evidence was any less reliable or complete as a result. Second he noted that Mr Lewis accepted that a copy of a letter Mr Lewis said had been sent the last known director was not on the file. We accept Mr Lewis' evidence that the letter was sent. Third he says the allegation of fraud is based simply in the absence of contact from the trader and that is insufficient.
    It seems to us that the allegation of fraud is based on more than just lack of response from the trader. We accept that the freight forwarders' documents indicate that it was likely that Heathrow was trading; we find it suspicious that the goods in which it was trading indicated a different trade from its originally declared trade; we find its absence from its declared place of business suspicious; and we find that these findings when coupled with a failure to make returns, to contact HMRC, to appeal assessments and to reply to letters is enough to conclude that the trader did not intend to pay the VAT attributable to the sales. We accept that a trader faced with an HMRC enquiry, with large assessments and demanding letters might bury his head in the sand, but we believe that it is likely even in that case that eventually the trader would if it was honest make some response to HMRC. On balance we conclude that Heathrow intended to evade the VAT on its sales.
    The documentary evidence for Deal 78 persuaded us that Heathrow had made the relevant sale. Mr Lewis was unable to say whether or not the related VAT had been assessed on Heathrow. We do not find that any lack of such assessment affects our conclusions that VAT was due and that it was evaded, and, for the reasons set out elsewhere, do not find that the lack of assessment means that there was no loss of tax.
    There was no direct evidence before us that related to Deal 78 which indicated whether Heathrow had been an importer of phones rather than having acquired than from a UK supplier. Release notes in relation to other transactions consistently indicated that Heathrow was an importer in all its transactions. We find on balance that it was likely to have been an importer in Deal 78.
  355. Parfums (UK) Limited (Deal 81)
  356. From Mr Read's evidence we find that HMRC's records show and we accept that:-
    (i) Parfums was VAT registered on 6 October 2003 declaring its business to be Importer and Exporter of toiletry items. It had one director, a Mr I Patel;
    (ii) Mr Patel, through Parfums, conducted a modest trade in the period to 31 August 2005. He was the only person in respect of whom PAYE returns were made in the period from incorporation to 28 February 2006;
    (iii) Mr Patel said that in May 2006 he had negotiated with a Mr Khan in relation to a possible purchase by Mr Khan of the company. Mr Patel said that the deal had not gone through;
    (iv) when HMRC made assessments on Parfums in relation to transactions in which its officers believed Parfums had made sales in July 2006, Mr Patel and his accountants indicated that they believed that someone else was using Parfums VAT number;
    (v) Parfums rendered nil VAT returns from 02/04 to 02/06;
    (vi) the Companies House records for Parfums indicated that I April 2006 Mr Khan was registered as a director and that the registered office had been changed;
    (vii) HMRC was contacted in May 2006 by Mr Khan and others asking whether they had received a letter abut the change of address of Parfums;
    (viii) three assessments were made on Parfums between November 2006 and April 2007 in respect of sales which HMRC believed had been made in July 2006 by Parfums. Other assessments have been made for a "dummy" hijacker;
    (ix) no response other than that detailed above has been received by HMRC in relation to the assessments and none have been paid; the last contact was from Parfums' accountants in December 2006 saying that Mr Patel was attempting to find out who had been using the company's VAT number;
    (x) the assessments include VAT in respect of Parfum's sale in Deal 81.
    The freight forwarders documents indicated that the phones which were the subject of Deal 81 were imported from Lisbon. Whilst the release was direct from the Lisbon seller to Parfums' customer we concluded that it was likely that Parfums had acquired the goods from the Lisbon seller. Thus no input tax credit was available against the VAT on their sale.
    Mr Beal makes three points: first he says that a letter was not sent to the company about procedures to avoid and combat MTIC fraud until after its deregistration. It does not seem to us that this bears on the question of whether there was fraud in this case. Second he says that HMRC had contact details for Mr Khan but did not attempt to contact him. Mr Read's evidence however was that he did not know whether or not any attempt had been made to contact him: all he had done was to interrogate HMRC's database. We find the lack of evidence of whether Mr Khan was contacted and if so what he said disturbing. Third he says it is difficult to see how Mr Khan can have been added as a director on Companies House records without the consent or connivance of the existing officers. We note that section 288 Companies Act 1985 requires the company to send to the registrar of companies details of a change in directors. It seems to us possible that such a notification could have been properly or fraudulently sent. Fourth he noted that Mr Read said he thought that Parfums' VAT number may have been hijacked although he could not guarantee it. In our view whether or not Mr Read thought it was is irrelevant to our determination.
    In our judgment it is likely that someone or some people dishonestly intended to evade VAT, and did evade VAT in relation to the sales attributed to Parfums in the documentation for Deal 81. That person could have been Mr Khan or Mr Khan and others hijacking Parfums' number, it could have been Parfums acting through Mr Khan or Mr Khan and others as its authorised agents, or it could have been Mr Patel (though we doubt the depth of his involvement). In any event the failure to provide any return in relation to the VAT on the sales and the absence of any communication in relation to these suggests in these circumstances a fraudulent intention to evade the VAT on the sale (we find that the phones were imported and no input tax credit was available).
  357. Phone City Limited (Deals 88, 89 and 90)
  358. Mrs Wanat gave evidence of a prolonged investigation into Phone City's VAT affairs for the periods 11/05, 02/06, 05/06 and a final period to 26 July 2006. She and other officers of HMRC had contact with the officers of Phone City in the year after 26 July 2006. This was not a case of the directors disappearing, but of a failure to produce records, and of VAT returns which indicate reclaims of VAT which HMRC investigated. We find from her evidence that:-
    (i) between October 2005 and July 2006 Phone City was engaged in the wholesale of mobile phones;
    (ii) the company's VAT returns for this period showed:
    Period Sales VAT due/(repayment)
    11/05 £96m (£38k)
    02/06 £19m (£3.8k)
    05/06 £181m (£908k)
    Final (£246k)
    (iii) after some delay repayment was made by HMRC in respect of the 11/05 period;
    (iv) strenuous efforts were made by Ms Wanat and other HMRC officers to obtain the records which underpinned the other returns. These efforts were mainly unsuccessful;
    (v) strenuous efforts were also made to contact the officers and employees of the company. When contacted they generally professed little knowledge of its operations and indicated that others had more knowledge. They blamed each other.
    (vi) for the final period the VAT returns showed an input tax credit of some £27m. Despite the reports and requirements made no evidence to support this claim was produced;
    (vii) details of sales made by Phone city to other traders obtained by HMRC from those other traders indicates that Phone City had underdeclared its output tax in the final period by some £4.2m;
    (vii) for the 05/06 period the absence of evidence of export of certain sales and the underdeclaration of other sales led Ms Wanat reasonably to conclude that £2.7m was due from Phone City rather than £0.908m due to it;
    (viii) the company is now in liquidation. HMRC have not taken any civil or criminal proceedings against its officers. Assessments made against the company total some £34m. None have been appealed or paid.
    The failure to produce records to support the VAT returns despite the efforts of Ms Wanat and others, and the evidence that the VAT returns were incomplete and incorrect when taken with the way in which the officers and employees washed their hands of responsibility suggests strongly to us that the returns deliberately understated Phone City's liability and that the company intended to evade its true VAT liabilities in these periods, and fraudulently claimed repayment of VAT. We so find.
    The evidence supporting the deal sheets for deal 90 included a release note from Silus BV releasing the phones to Phone City's purchaser. This dealt in typescript with "1500 units" but someone had written in manuscript on the notes "500 pcs" ie 500 phones. Mr Beal suggests that this manuscript annotation casts doubt in the alleged link between the release note and the later sale invoice by Phone City of 1500 phones sufficient to make it impossible to conclude that the 1500 phones later sold were imported from Silus BV. We do not agree: the manuscript note is in handwriting which appears to be different from that of the signatory of the release notes. We conclude that the two documents are linked and that in that deal Phone City imported the phones (despite the possibility that there was a UK purchaser between Phone City and Silus BV) and accordingly that its VAT liability in respect of that sale was not reduced by any input tax credit.
    There was no such evidence in relation to Deals 88 and 89. We are unable to conclude that Phone City was not entitled to input tax credit on those phones.
    We conclude in relation to deal 90 that Phone City fraudulently intended to evade and did evade the VAT on its sale. In relation to deals 88 and 89 we conclude that it intended to evade VAT on its margin only.
    In relation to those other deals the Respondents submit that the evidence makes it likely that either Phone City was the importer or that it was a purchaser in a deal chain starting with an importer who fraudulently evaded the VAT on its sale. We are unable to reach that conclusion : there was insufficient evidence that another person had been fraudulent or that Phone City had imported the phones. We conclude that in relation to Deals 88 and 89 only that Phone City intended to evade VAT on its margin.
    APPENDIX III
    Deal Number Chain Proved Alleged Defaulter Default proved Comment Input tax denied
    1 yes EMES yes   all
    2 no Oracle yes   none
    3 yes EMES yes   all
    4 No Mana? No   No
    5 yes Realtech margin only   margin only
    6 yes EMES yes   all
    7 yes Attic None   none
    8 yes     Contra trade: connection not proved noone
    9 yes I ConnectU yes   all
    10 390-yes; 800: no ConnectU yes   390-all 800 - none
    11 yes Skywide yes   all
    12 yes FX Drona yes   all
    13 yes EMES yes   all
    14 450: n o connection to alleged fraudster; 625 -yes 625: Data Sols Northern 450: no 625:margin only   450 - none 625- margin
    15 no Apollo yes   none
    16 yes ConnectU yes   all
    17 yes Skywide yes   all
    18 yes Goodluck yes   all
    19 yes UA Distribution yes (99%)   all
    20 yes GPA yes   all
    21 500-yes 500-no UA Distribution yes   500- all 500-none
    22&23 2000 - no 3253-yes 2000:UA Distribution 2953: USM IT 500: Afflecks yes margin only no   2000- none 2953-margin 500- none
    24 yes UA Distribution yes   all
    25 700 - yes 493-no 700:UA Distribution 500:Biogistics UA:yes(99%)
    Boilog:margin only
      700-all 493-none
    26 yes USM IT yes   all
    27 yes Colston yes   all
    28 yes Colston yes   all
    29 yes Computec yes   all
    30 yes Park Supplies no   none
    31 yes USM IT yes   all
    32 yes USM IT yes   all
    33 yes Bullfinch yes   all
    34 yes USM IT yes   all
    35 yes Bullfinch yes   all
    36 yes-both 1000:Bullfinch 1000:Midwest yes yes   all
    37 yes USM IT yes   all
    38 yes USM IT yes   all
    39 yes USM IT yes   all
    40 yes USM IT yes   all
    41 yes Bullfinch yes   all
    42 yes USM IT yes   all
    43 yes USM IT yes   all
    44 yes USM IT yes   all
    45 2000- no 1000- yes 2000:USM IT 1000:LTH Ltd yes yes   2000-none 100-margin
    46 200 -no 1800 -yes 200:Princeways 1800:SS Enterprise GB no
    yes
      200- none 1000- all
    47 yes Cyberweb yes   all
    48 yes see p84 - Bullfinch or Data Northern? Bullfinch: yes   all
    49 yes Cyberweb yes   all
    50 yes Bullfinch yes   all
    51 yes Cyberweb yes   all
    52 yes Cyberweb yes   all
    53 yes Tressle yes   all
    54 yes Bullfinch yes   all
    55&56 yes Cyberweb yes   all
    57 yes AAA Multilink yes   all
    58 no n/a no   none
    59 yes 3D Animations yes   all
    60 yes AAA Multilink yes   all
    61 no Cyberweb yes   none
    62 yes AAA Multilink yes   all
    63 yes 3D Animations yes   all
    64 yes 3D Animations yes   all
    65 yes AAA Multilink yes   all
    66 yes Crossview yes(someone)   all
    67 no DBP Trading margin only   none
    68 yes Subumma margin only   margin only
    69&72 yes Rukford yes   all
    70 yes RS Sales yes   all
    71 yes JD Telecoms margin only   margin only
    72 yes Rukford yes   all
    73 yes Crossview yes(someone)   all
    74 yes RS Sales yes   all
    75 yes RS Sales yes   all
    76 yes Vision Soft yes   all
    77 yes Pearl Cosmetics yes   all
    78 no Heathrow Bus yes   none
    79 yes Pearl Cosmetics yes   all
    80 yes DTM Provision yes   a
    81 yes Parfums (UK) yes   all
    82 yes DTM Provision yes   all
    83&84 no JD Telecoms margin only   none
    85 yes Pearl Cosmetics yes   all
    86 yes DTM Provision yes   all
    87 yes DTM Provision yes   all
    88 yes Phone City margin only   margin only
    89 yes Phone City margin only   margin only
    90 yes Phone City yes   all
    CHARLES HELLIER
    TRIBUNAL JUDGE
    RELEASED: 18 May 2009


BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: http://www.bailii.org/uk/cases/UKFTT/TC/2009/TC00076.html