First-tier Tribunal (Tax)

You are here: BAILII >> Databases >> First-tier Tribunal (Tax) >> 3G Mobile Phones Ltd v Revenue & Customs (VALUE ADDED TAX - input tax - MTIC - fraudulent evasion of VAT) [2013] UKFTT 719 (TC) (29 November 2013)
URL: http://www.bailii.org/uk/cases/UKFTT/TC/2013/TC03099.html
Cite as: [2013] UKFTT 719 (TC)

[New search] [Contents list] [Printable PDF version] [Help]

[2013] UKFTT 719 (TC)

TC03099

Appeal number: TC/2010/06112

VALUE ADDED TAX - input tax - MTIC - fraudulent evasion of VAT - contra-trading - whether the appellant knew or should have known that its transactions were connected to fraudulent evasion of VAT– yes - appeal dismissed

FIRST-TIER TRIBUNAL

TAX CHAMBER

	3G MOBILE PHONES LIMITED	Appellant

	- and -

	THE COMMISSIONERS FOR HER MAJESTY’S	Respondents
	REVENUE & CUSTOMS

TRIBUNAL:	JUDGE GUY BRANNAN
	TYM MARSH

Sitting in public at Bedford Square, London WC1 on 16, 18, 19, 23, 24 and 26 April 2013

Liban Ahmed, Controlled Tax Management Limited, for the Appellant

Gareth Patterson, Counsel, instructed by the General Counsel and Solicitor to HM Revenue and Customs, for the Respondents

[Decision amended and re-made under Rules 37 and 38]

DECISION

Introduction

1. This is an appeal in respect of the denial by the Respondents ("HMRC", which expression is used for convenience to include HMRC's predecessor, HM Customs & Excise) of the right to deduct input tax in the Appellant's VAT returns of the months ended 30 June and 31 July 2006 (referred to as 06/06 and 07/06). HMRC's decision was notified to the Appellant by a letter dated 7 April 2010.

2. The appeals allegedly involved what is known as MTIC trading.

3. A total of five transactions took place in 06/06 and 07/06 and the total amount of input tax denied was £1,410,937.50. The transactions were documented on two invoices issues to the Appellant by its supplier. All the transactions concerned the purchase of the mobile phones.

4. The reason for HMRC's decision to deny input tax was that they considered that the transactions formed part of an overall scheme to defraud the Revenue and that the Appellant knew or should have known that its transactions were connected with the fraudulent evasion of VAT.

5. HMRC alleged that the scheme to defraud the Revenue involved contra-trading. Essentially, contra-trading is a scheme of fraudulent trading intended to disguise the VAT fraud by ensuring that a VAT refund is claimed in a different chain of transactions from that in which the fraudulent tax loss occurred. There were two contra-traders: A-Z Mobile Accessories Limited ("A-Z") and Jag-Tec Limited ("Jag-Tec").

6. The Appellant filed a Notice of Appeal dated 22 July 2010 against HMRC's decision.

7. As we shall see, the main issue in this appeal is whether the Appellant knew or should have known that its five transactions were connected with fraudulent evasion of VAT.

8. The only parties to this appeal were HMRC and the Appellant. The only witnesses to give evidence were HMRC officers and Mr Hill (the director of the Appellant). No other person was a party to these proceedings and no other party has been invited to provide evidence, to make representations and no other party was represented before this Tribunal. Accordingly, it would not be fair to treat any part of this Decision as findings made against any person who is either not a party to these proceedings or who did not give evidence.

The evidence

9. We were provided with over 40 ring binders of documents containing witness statements and exhibits.

10. So far as material to the matters in dispute, the following HMRC officers gave witness statement evidence:

(1) Stephen Hall - in relation to the Appellant;

(2) George Edwards - in relation to the Appellant's supplier, Stardex (UK) Limited ("Stardex");

(3) Roderick Stone - in relation to the background of MTIC fraud;

(4) Peter Morehead - in relation to the freight forwarder 1st Freight Limited;

(5) Katrina Wheatcroft - in relation to the contra-trader A-Z;

(6) Matthew Elms - in relation to the contra-trader Jag-Tec;

(7) Ian Simmons - also in relation to the contra-trader Jag-Tec;

(8) Nigel Humphries - in relation to the scheme of contra-trading;

(9) Daniel Payne - in relation to the Appellant's transactions through its account with First Curaçao International Bank ("FCIB"); and

(10) Peter Birchfield - in relation to payments made through FCIB.

11. Officers Hall, Edwards and Morehead gave witnesses statement and oral evidence and were cross-examined.

12. The witness for the Appellant was its director Mr Stuart Hill. Mr Hill’s evidence comprised three witness statements, oral evidence and evidence under cross-examination.

MTIC trading - legal principles

13. There have now been many appeals heard by this Tribunal in respect of alleged MTIC transactions and it is unnecessary to give an explanation of how MTIC fraud is carried out. A convenient explanation is given by Christopher Clarke J in Red 12 Trading Ltd v HMRC [2009] EWCH 2563 (Ch) [at 2].

14. There was no dispute as to the applicable legal principles, which are as follows.

15. The legal right to a deduction for input tax is enshrined in Articles 167 and 168 of Council Directive 2006/112/EC of 28 November 2006 and in sections 24, 25 and 26 of the Value Added Tax Act 1994.

16. There is no legal right to a deduction for input tax, however, where fraud is involved. There is now extensive case-law on this subject both before the European Court of Justice and our domestic courts. The position was summarised by Lewison J in the decision of the Upper Tribunal in Brayfal Ltd v HMRC [2011] UKUT 99 (TCC) as follows:

"While Brayfal’s appeal has been making its way through the system, the law has been considered by the courts on a number of occasions. It finds its latest authoritative pronouncement in the decision of the Court of Appeal in Mobilx Ltd v HMRC [2010] EWCA Civ 517. This decision was handed down on 12 May 2010, a couple of months after the revised decision of the FTT. That case examined the ramifications of the decision of the ECJ in Axel Kittel v Belgium; Belgium v Recolta Recycling Joined Cases C-439/04 and C-440/04 [2006] ECR 1-6161 (“Kittel”). What the Court of Appeal decided was:

A taxable person who knows or should have known that the transaction which he is undertaking is connected with fraudulent evasion of VAT is to be regarded as a participant and fails to meet the objective criteria which determine the scope of the right to deduct. (§ 43)

If a taxpayer has the means at his disposal of knowing that by his purchase he is participating in a transaction connected with fraudulent evasion of VAT he loses his right to deduct, not as a penalty for negligence, but because the objective criteria for the scope of that right are not met. (§ 52)

The principle does not extend to circumstances in which a taxable person should have known that by his purchase it was more likely than not that his transaction was connected with fraudulent evasion. But a trader may be regarded as a participant where he should have known that the only reasonable explanation for the circumstances in which his purchase took place was that it was a transaction connected with such fraudulent evasion. (§ 60)

The test is simple and should not be over-refined. It embraces not only those who know of the connection but those who "should have known". Thus it includes those who should have known from the circumstances which surround their transactions that they were connected to fraudulent evasion. If a trader should have known that the only reasonable explanation for the transaction in which he was involved was that it was connected with fraud and if it turns out that the transaction was connected with fraudulent evasion of VAT then he should have known of that fact. (§ 59)

If HMRC wishes to assert that a trader's state of knowledge was such that his purchase is outwith the scope of the right to deduct it must prove that assertion. (§ 81)

In answering the factual question, Tribunals should not unduly focus on the question whether a trader has acted with due diligence. Even if a trader has asked appropriate questions, he is not entitled to ignore the circumstances in which his transactions take place if the only reasonable explanation for them is that his transactions have been or will be connected to fraud. The danger in focusing on the question of due diligence is that it may deflect a Tribunal from asking the essential question posed in Kittel, namely, whether the trader should have known that by his purchase he was taking part in a transaction connected with fraudulent evasion of VAT. The circumstances may well establish that he was. (§ 82)

I should also record that it was common ground that these principles should be applied in the light of the circumstances prevailing at the date of the taxable person’s own transactions: C-354/03 Optigen Ltd v Customs and Excise Commissioners [2006] ECR I-483. "

17. We respectfully adopt Lewison J's summary of the law as a correct statement of the current position.

18. We should also add that, in relation to the issue whether a trader's transactions were connected to the fraudulent evasion of VAT, Roth J held in Powa (Jersey) Ltd v HMRC [2012] UKUT 50 (TCC) that it was not necessary that the trader was in privity of contract with a fraudulent trader. Instead, if a trader knows or should have known that the transactions which it entered into were part of a chain in which one or more of the earlier transactions were fraudulent, even if its immediate supplier was not fraudulent, the Kittel test is satisfied.

19. We also note the comments of Moses LJ in Mobilx in relation to questions of evidence, where he said (at page 1459):

“The questions posed in BSG …by the tribunal were important questions which may often need to be asked in relation to the issue of the trader's state of knowledge. I can do no better than repeat the words of Christopher Clarke J in Red 12 Trading Ltd v Revenue and Customs Comrs [2009] EWHC 2563 (Ch) at [109]–[111], [2010] STC 589 at [109]–[111]:

'[109] Examining individual transactions on their merits does not, however, require them to be regarded in isolation without regard to their attendant circumstances and context. Nor does it require the tribunal to ignore compelling similarities between one transaction and another or preclude the drawing of inferences, where appropriate, from a pattern of transactions of which the individual transaction in question forms part, as to its true nature e.g. that it is part of a fraudulent scheme. The character of an individual transaction may be discerned from material other than the bare facts of the transaction itself, including circumstantial and “similar fact” evidence. That is not to alter its character by reference to earlier or later transactions but to discern it.

[110] To look only at the purchase in respect of which input tax was sought to be deducted would be wholly artificial. A sale of 1,000 mobile phones may be entirely regular, or entirely regular so far as the taxpayer is (or ought to be) aware. If so, the fact that there is fraud somewhere else in the chain cannot disentitle the taxpayer to a return of input tax. The same transaction may be viewed differently if it is the fourth in line of a chain of transactions all of which have identical percentage mark ups, made by a trader who has practically no capital as part of a huge and unexplained turnover with no left over stock, and mirrored by over 40 other similar chains in all of which the taxpayer has participated and in each of which there has been a defaulting trader. A tribunal could legitimately think it unlikely that the fact that all 46 of the transactions in issue can be traced to tax losses to HMRC is a result of innocent coincidence. Similarly, three suspicious involvements may pale into insignificance if the trader has been obviously honest in thousands.

[111] Further in determining what it was that the taxpayer knew or ought to have known the tribunal is entitled to look at the totality of the deals effected by the taxpayer (and their characteristics), and at what the taxpayer did or omitted to do, and what it could have done, together with the surrounding circumstances in respect of all of them.'"

20. As we indicated above, contra-trading is essentially a variation on basic MTIC trading. The contra-trader attempts to disguise its export transactions by engaging in a separate series of transactions where its role involves the making of standard rated supplies. A helpful description is contained in the decision of the VAT and Duties Tribunal decision in Livewire Telecom Ltd v HMRC [2008] V&DR 131 (Dr John F Avery Jones CBE (Chairman) and Sheila Wong Chong FRICS) as follows:

“In order to demonstrate where the loss of tax arises from MTIC fraud we start with a simple example of an import of goods by X who sells them to Y who exports them. The tax on acquisition (import) by X is cancelled by input tax of the same amount, and the output tax charged on sale by X will be cancelled by input tax repaid to Y on the export, so that the United Kingdom exchequer receives no net tax. If both X and Y are fraudsters Y will have to finance the output tax charged by X, which is recovered by X not paying the output tax to Customs. The only gain by the fraud is if Customs pay the input tax to Y when the exchequer is left with a loss of the amount of the input tax; the non-payment of output tax by X is merely the recovery of what Y put in. If the exporter is innocent of that fraud he is entitled to repayment of the input tax that he has actually paid to X even though this represents tax never paid by A [the missing trader] and the exchequer is left with the same loss of the amount of the input tax.

... [T]his appeal is concerned with contra-trading. In contra-trading there are, in its simplest theoretical form, two chains of transactions. First, the “dirty chain,” in which there is a missing trader, defaulting trader, or trader using a hijacked VAT number (“missing trader” for short), comprising A (the missing trader) who is the importer of goods into the UK, who sells them to B, who sells them to C who exports the goods, and is thus in a VAT reclaim position. (For simplicity we shall use the expressions import and export for intra-Community trade, acknowledging that these are not the proper labels.) Secondly, the “clean chain,” in which there are no missing traders, comprising C, who is this time the importer, who sells to D, who sells to E, the exporter (the Appellant in this appeal is in the position of E). The effect of the clean chain is that the net input tax position of C in the dirty chain is cancelled by output VAT in the clean chain. There is no benefit to C in this as C has paid the input tax to B, and therefore C could be a trader who happens to carry out both import and export transactions unconnected with any fraud, or C could be a trader who is controlled by a “puppet master” to enter into the cancelling transactions to disguise A’s involvement in a fraud. The effect of the contra-trades is that C does not excite Customs’ attention as it is not applying for a repayment; the non-payment of tax by A is less noticeable since without a return Customs do not know how much tax A owes. The input tax reclaim that C had in the dirty chain has moved to E who is at the end of a clean chain. The only way for Customs to refuse repayment of E’s input tax is to show that E knew or ought to have known of A’s fraud in a completely different chain, and possibly of C’s involvement. Since ... the only gain from A’s fraud is the recovery of input tax by E this must imply that E is a participant in the fraud and, unless he is the puppet-master, is presumably sharing the tax recovered with someone else. As Mr Scorey pointed out it is difficult to see how a case of E having means of knowledge, rather than actual knowledge, can arise.

The nature of contra-trading is easy to state in the above way but the problem in real life is that there is no logical connection between the clean and dirty chains. First, the VAT accounting periods for C and E will not coincide; E may be on a monthly accounting period as it is a habitual exporter, but C may be on a three-monthly period, and C need only arrange that the net tax is nil during that three-monthly period by entering into transactions after E’s transactions. Secondly, the goods dealt in may be different in the two chains. Thirdly, for a particular C there may be many different equivalents to A and E, and for a particular E there may be many equivalents of C, each with more than one equivalents to A. Fourthly, C may not have deliberately entered into imports in the clean chain in order to cancel the input in the dirty chain; C may merely be both an importer and an exporter whose outputs in relation to the former happen roughly to cancel its inputs in relation to the latter. Fifthly, there may be many Bs and Ds in between the importer and exporters.”

21. In Megtian Limited v HMRC [2010] EWHC 18 (Ch), Briggs J discussed the test in Kittel and the comments of Moses LJ in Mobilx that the test should not be over-refined. In relation to "clean" and "dirty" chains involved in contra-trading he held that for the Kittel test to be satisfied it was not necessary that the Appellant should have actual or imputed knowledge of the details of the fraudulent conduct:

"31. The issue addressed by Lewison J in Livewire concerned the nature of the fraud which it was necessary to demonstrate that the broker at the foot of a clean chain knew or ought to have known was connected with his transaction. In a contra-trading case there are, at least in theory, two potentially distinct frauds. The first is that of the missing or defaulting trader at the head of the dirty chain, who intends to abscond without accounting to HMRC for the tax paid to him by his immediate buyer. The second is that of the contra-trader who seeks to use the clean chain involving the broker as a means of dishonest concealment of the first fraud. As Lewison J put it, at paragraph 102:

"In my judgment in a case of alleged contra-trading, where the taxable person claiming repayment of input tax is not himself the dishonest co-conspirator, there are two potential frauds:

(i) the dishonest failure to account for VAT by a defaulter or missing trader in the dirty chain; and

(ii) the dishonest cover-up of that fraud by the contra-trader."

The issue for Lewison J was whether a disallowance of repayment of input tax claimed by the broker at the foot of the clean chain required it to be shown that he knew or ought to have known of both of those frauds, or merely one or the other of them. He concluded that the second of those alternatives was sufficient, at least in a case where dishonesty had been established as against the contra-trader.

32. Lewison J's conclusion is set out at paragraph 103 of the judgment as follows:

"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.”

33. ….

34. … I do not read Lewison J's analysis of the issue as to what must be shown that the broker knew or ought to have known in a contra-trading case as amounting to a rigid prescription that, as a matter of law, such an analysis must be performed in every contra-trading case, such that it will be defective unless it identifies one or other of the alternative frauds as being that which the broker knew or ought to have known.

36. … Lewison J acknowledged that in many if not most cases of contra-trading, the clean chain and the dirty chain were likely to be part of a single overall scheme to defraud the Revenue. As he put it, at paragraph 109:

"Indeed it seems to me that the whole concept of contra-trading (which is HMRC's own coinage) necessarily assumes that to be so."

37. In my judgment, there are likely to be many cases in which a participant in a sophisticated fraud is shown to have actual or blind-eye knowledge that the transaction in which he is participating is connected with that fraud, without knowing, for example, whether his chain is a clean or dirty chain, whether contra-trading is necessarily involved at all, or whether the fraud has at its heart merely a dishonest intention to abscond without paying tax, or that intention plus one or more multifarious means of achieving a cover-up while the absconding takes place.

22. In POWA (Jersey) Limited v HMRC [2012] UKUT 50 (TCC) Roth J agreed with the above approach of Briggs J in Megtian, stating:

"52. However, I do not see that there is any requirement that PJL should reasonably have known the identity of the contra-trader. HMRC must establish that fraudulent evasion of VAT took place, and if the form of fraud involved was contra-trading then that is what they have to prove. But it is a misconception to consider that they must also establish that the party seeking to deduct input tax (ie, here, PJL) should reasonably have known that its own transaction was connected to (or involved in) this particular form of missing trader fraud as opposed to another form. I do not regard the Chancellor's judgment in Blue Sphere Global as authority to the contrary. Moreover, I respectfully agree with the approach expressed by Briggs J in his subsequent judgment in Megtian Ltd (in admin) v Revenue and Customs Comrs [2010] EWHC 18 (Ch) at [37]–[38], [2010] STC 840 at [37]–[38]….

[53] In any event, it is clear from the Court of Appeal judgment in Mobilx , where one of the three cases under appeal was Blue Sphere Global, that no special approach is required in a case involving contra-trading. The correct test as regards knowledge is always the same. It is the test derived from Kittel as set out at [59] of Moses LJ's judgment: see [39], above. Hence, in the section of his judgment that addressed the specific appeal in Blue Sphere Global, Moses LJ found that although the case on the facts came close to satisfying the test, the tribunal had focused unduly on whether Mr Peters, the company's sole shareholder and director, had exercised sufficient care and diligence, and what he might have found out if he had made further inquiries, and thus had failed to make a finding applying the correct test. Moses LJ concluded at [75]:

'The ultimate question is not whether the trader exercised due diligence but rather whether he should have known that the only reasonable explanation for the circumstances in which his transaction took place was that it was connected to fraudulent evasion of VAT. The tribunal might have concluded that Mr Peters should have known that the transactions into which he entered were connected with fraud, by reference to the unconventional nature of those circumstances (a finding it came close to making at para 228). But it was not the only decision within the bounds of reasonable conclusion.'

[54] By contrast with the tribunal in Blue Sphere Global, the FTT here emphasised that the test was 'not whether PJL took adequate precautions, but whether it knew or had the means of knowing that its transactions were connected with fraud': [2009] UKFTT 360 (TC) at [127]. Based on a thorough consideration of all the surrounding circumstances, it found that PJL knew or must have known that it was engaged in an artificial, contrived market, and that finding applies to the three transactions that were part of a contra-trading chain as much as to all the others. This ground of appeal is accordingly misconceived."

23. Leave to appeal from the judgment of Roth J was refused by Moses LJ in the Court of Appeal (POWA (Jersey) Limited v HMRC [2013] EWCA Civ 225).

24. Finally, it was common ground that the standard of proof is the normal civil standard of proof (i.e. the balance of probabilities) and, as explained by Lewison J in Brayfal (above), that the burden of proof lies upon HMRC.

Terminology

25. Appeals involving MTIC disputes have evolved a set of expressions all of their own. For example, the exporter (such as the Appellant) who seeks a repayment of VAT is usually called the "broker". Companies higher up the chain of transactions, between the broker and the importer, are usually referred to as "buffers". Contra-traders are involved in "dirty chains", which trace back to a fraudulent tax loss, and "clean chains" where no tax losses are involved and which are used to disguise export transactions which would otherwise have been carried out by the contra-trader. In this decision, we employ these terms for convenience, without in any way pre-judging the issues before us. Also for convenience we have tended to use “import” and “export”, rather than the technical VAT expressions “acquisition” and “despatch”.

Issues in dispute

26. Prior to the hearing, the Appellant accepted that:

(1) HMRC had identified the correct supply chains in both the Appellant's deal chains and those of the two contra-traders (A-Z and Jag-Tec);

(2) a fraudulent tax loss had been proved in respect of each of the "dirty" supply chains;

(3) HMRC's evidence regarding the actions and trading patterns of the contra-traders was accurate and that, therefore, none of HMRC's witnesses in respect of the contra-traders was required to give oral evidence;

(4) the two contra-traders (A-Z and Jag-Tec) had acted fraudulently by using the Appellant's transaction chains (ie “clean chains”) to mask the fraud in their "dirty" supply chains;

(5) there had been an orchestrated fraud involving both the Appellant's supplier and customer during the periods 06/06 and 07/06; and

(6) the Appellant had a general knowledge of fraud in its business sector.

27. The Appellant therefore accepted that there had been fraudulent evasion of VAT and that its transactions were connected to such fraudulent evasion.

28. The issue in this appeal was, therefore, whether the Appellant knew or should have known that its transactions were connected to the fraudulent evasion of VAT.

The Appellant - background

29. The Appellant was incorporated on 25 February 2002. Mr Stuart Hill was the director of the Appellant from that date and for all periods material to this appeal. Mr Hill's father was the company secretary.

30. The Appellant was registered the VAT with effect from 1 November 2002. The application form for registration was signed by Mr Hill and described the intended business activities as "telecommunications." This form (VAT 1) estimated the Appellant's annual turnover as £1 million and indicated that all sales would be to other EC Member States but the purchases would not be made from other EC Member States. Form VAT 1 also indicated that the Appellant expected to receive regular repayments of VAT.

31. The principal place of business of the Appellant according to Form VAT 1 was that of the Appellant's accountant. Mr Hill's home address was in Stanmore, Middlesex and, in practice, he ran the business from his home address.

32. HMRC granted the Appellant's application for registration but imposed the condition that the Appellant was required to submit monthly VAT returns.

33. On a post-registration visit by an HMRC officer in 2002, Mr Hill confirmed that the principal place of business of the Appellant was indeed his home address. He also confirmed that the main business activity was the wholesale of mobile phones. He stated that he had previously worked for Stardex. This point is significant because Stardex features as the immediate supplier to the Appellant in all five deals under appeal. The officer noted in his report:

"Stuart Hill obviously knows the mobile phone trade inside out. Can't really refuse application this time."

34. In fact, Mr Hill had worked in the telecommunications sector since 1999. He had previously worked selling mobile phones in telesales, working for a company called Euro Cellular. Thereafter, he was employed by Stardex buying and selling mobile phones.

35. After the Appellant’s business had been set up in 2002 and Mr Hill became its director, Mr Hill continued to sell mobile phones for Stardex on a commission basis.

36. From the date of the Appellant's registration for VAT in 2002 until the periods in 2006 which are under appeal, the Appellant submitted VAT returns which in most periods of significant activity claimed VAT repayments. Mr Hall, the HMRC officer who gave evidence in respect of the Appellant, summarised the Appellant's VAT returns (most of which was signed by Mr Hill), the repayment claims, made. For convenience these summaries are set out in the Appendix to this decision (omitting EC acquisitions, which were nil in every period).

37. The Appellant made an application on 24 October 2005 to open an e-banking account with FCIB. FCIB was a bank based in the Netherlands Antilles.. In 2006 FCIB was investigated in relation to alleged money-laundering. Its banking licence was subsequently revoked.

38. The Appellant's application to open an FCIB account gave the signatory as Mr Hill, his passport number and the telephone number for his Stanmore home. The e-mail address given was that of Mr Hill's business e-mail account. The application also gave the Finchley address of the Appellant's accountant.

The deals under appeal

39. The VAT return for 06/06 was received by HMRC on 5 July 2006. It related to the purchase of the mobile phones on an invoice dated 30 June 2006. The return for 07/06 was received by HMRC on 3 August 2006. It related to the purchase of mobile phones on an invoice dated 28 July 2006.

40. In the return for 06/06 the Appellant claimed an input tax deduction of £900,952.45. Of this, £898,537.50 related to two transactions involving the purchase of 18,000 mobile phones (11,000 and 7,000 units respectively). The two transactions were contained on the same invoice issued by Stardex to the Appellant.

41. In the return for 07/06 the Appellant claimed an input tax deduction of £529,840.19. Of this, £512,400 related to three transactions involving the purchase of 10,000 mobile phones (2,000, 3000 and 5,000 units respectively). All three transactions were contained on the same invoice issued by Stardex to the Appellant.

42. Thus, there are five deals under appeal which relate to two invoices. The two deals on the first invoice in 06/06 and three deals on the second invoice in 07/06. In each deal Stardex, Mr Hill's former employer, was the Appellant's immediate supplier.

43. In all five deals the Appellant exported the mobile phones to a company based in Marbella, Spain called Complementos De Exportacion Multifunctionales SA ("CEMSA").

44. The five deals involve different makes of mobile phones but all the consignments comprised of European specification mobile phones with a 2-pin plug. The 2-pin plug meant that the phones could not be used in the UK without replacing the 2-pin plug with a standard UK 3pin plug or using a Traveller to UK plug adapter. It was pointed out by Mr Hill in his evidence, and it was not disputed, that all mobile phones used by consumers in this country were European specification mobile phones and they were all imported into the UK: no mobile phones mobile telephone were manufactured in this country.

45. As already indicated, the Appellant accepts that both Stardex and CEMSA were involved, in relation to these transactions, in an organised VAT fraud. We shall go into further detail about Stardex and CEMSA later in this decision.

46. We now turn to consider the five deals in greater detail, looking at the deal chains which HMRC have pieced together and which the Appellant accepts to be correct. The evidence in respect of these deal chains was given by Mr Hall and was not challenged.

06/06 Deal 1

47. KOM Team SARL ("KOM Team"), a company based in France, sold 11,000 Nokia N80 mobile phones to Jag-Tec, a trader based in the UK (and one of the two fraudulent contra-traders involved in these appeals) for £3,047,000 on an invoice dated 30 June 2006. Jag-Tec's purchase order was dated 29 June 2006.

48. Jag-Tec sold the 11,000 Nokia N80 mobile phones to Stardex for £3,058,000 (plus £535,150 VAT) on an invoice dated 29 June 2006.

49. Stardex sold the 11,000 Nokia N80 mobile phones to the Appellant for £3,080,000 (plus £539,000 VAT) on an invoice dated 30 June 2006.

50. The Appellant sold the 11,000 Nokia N80 mobile phones to CEMSA for £3,267,000 on an invoice dated 30 June 2006.

51. CEMSA sold the 11,000 Nokia N80 mobile phones to Vundera SA ("Vundera"), a company based in Latvia, for £3,273,600 on an invoice dated 26 July 2006.

52. The invoices issued by Stardex, the Appellant and CEMSA also related to 06/06 Deal 2.

06/06 Deal 2

53. KOM Team sold 7,000 Sony Ericsson W900i mobile phones to A-Z for £2,030,000 on an invoice dated 30 June 2006. A-Z's purchase order was dated the same date.

54. A-Z sold the 7,000 Sony Ericsson W900i mobile phones to Stardex for £2,040,500 (plus £357,087.50 VAT) on an invoice dated 30 June 2006.

55. Stardex sold the 7,000 Sony Ericsson W900i mobile phones to the Appellant for £2,054,500 (plus £359,537.50 VAT) on an invoice dated 30 June 2006 (which, as noted, also related to 06/06 Deal 1).

56. The Appellant sold the 7,000 Sony Ericsson W900i handsets to CEMSA for £2,177,000 on an invoice dated 30 June 2006.

57. CEMSA sold the 7,000 Sony Ericsson W900i mobile phones to Vundera for £2,181,340 on an invoice dated 26 July 2006.

07/06 Deal 1

58. KOM Team sold 2,000 Nokia 8800 mobile phones to A-Z for £576,000 on an invoice dated 21 July 2006. A-Z's purchase order was dated the same date.

59. A-Z sold the 2,000 Nokia 8800 mobile phones to Stardex for £578,000 (plus £101,150 VAT) on an invoice dated 27 July 2006.

60. Stardex sold the 2,000 Nokia 8800 mobile phones to the Appellant for £588,000 (plus £102,900 VAT) on an invoice dated 28 July 2006.

61. The Appellant sold the 2,000 Nokia 8800 mobile phones to CEMSA for £623,300 on an invoice dated 28 July 2006.

62. CEMSA sold the 2,000 Nokia 8800 mobile phones to Vundera for £624,500 on an invoice dated 4 August 2006.

63. The invoices issued by Stardex, the Appellant and CEMSA also related to 07/06 Deals 2 and 3.

07/06 Deal 2

64. KOM Team sold 3,000 Nokia N80 mobile phones to Jag-Tec for £831,000 on an invoice dated 24 July. Jag-Tec's purchase order was dated 21 July 2006.

65. Jag-Tec sold the 3,000 Nokia N80 mobile phones to Stardex for £834,000 (plus £145,950 VAT) on an invoice dated 28 July 2006.

66. Stardex sold the 3,000 Nokia N80 mobile phones to the Appellant for £840,000 (plus £147,000 VAT) on an invoice dated 28 July 2006.

67. The Appellant sold the 3,000 Nokia N80 mobile phones to CEMSA for £891,000 on an invoice dated 28 July 2006.

68. CEMSA sold the 3,000 Nokia N80 mobile phones to Vundera for £892,800 by invoice dated 4 August 2006.

07/06 Deal 3

69. KOM Team sold 5,000 Nokia N91 mobile phones to A-Z £1,445,000 on an invoice dated 21 July 2006

70. A-Z sold the 5,000 Nokia N91 mobile phones to Stardex for £1,450,000 (plus £253,750 VAT) on an invoice dated 27 July 2006.

71. Stardex sold the 5,000 Nokia N91 mobile phones to the Appellant for £262,500 (plus £262,500 VAT) on an invoice dated 28 July 2006.

72. The Appellant sold the 5,000 Nokia N91 mobile phones to CEMSA for £1,590,000 on an invoice dated 28 July 2006.

73. CEMSA sold the 5,000 Nokia N91 mobile phones to Vundera £1,593,250 on an invoice dated 4 August 2006.

Input tax claimed

74. As already explained, Stardex issued two invoices in respect of the five transactions under appeal (two transactions in 06/06 and three transactions in 07/06) and the Appellant issued two invoices in respect of the five sales of mobile phones to CEMSA (again, two transactions in 06/06 and three transactions in 07/06).

75. Consequently, the Appellant claimed input tax in respect of the five transactions as follows:

(1) 06/06 Deal 1 - £539,000

(2) 06/06 Deal 2 - £359,537.50

(3) 07/06 Deal 1 - £102,900

(4) 07/06 Deal 2 - £147,000

(5) 07/06 Deal 3 - £262,500

The contra-traders: Jag-Tec and A-Z

76. The Appellant accepted that Jag-Tec and A-Z were fraudulent contra-traders.

77. In each of the five deals under appeal the supply chain began with the same EU supplier, KOM Team.

78. Jag-Tec was a participant in two deals: 06/06 Deal 1 and 07/06 Deal 2. A-Z was a participant in the three other deals: 06/06 Deal 2 and 07/06 Deals 1 and 3. The mobile phones were sold by Jag-Tec to Stardex, and were then sold (at the standard rate of VAT) to the Appellant. The Appellant sold the goods to the Spanish trader, CEMSA, and thereby created a VAT repayment claim. CEMSA then sold the goods to Vundera in Latvia.

79. HMRC's evidence in respect of Jag-Tec and A-Z was not disputed. Nonetheless, we record our findings in respect of the two contra-traders in Appendix 2 to this decision.

80. It was clear from the evidence of HMRC's witnesses, none of which was challenged by the Appellant, that Jag-Tec and A-Z were part of a circle of contra-traders carrying out a logistically complex, highly organised and sophisticated fraud on HMRC.

Stardex - the Appellant’s supplier

81. Stardex was registered for VAT on 25 February 1999.

82. Maria Prouost was the main shareholder. She used different dates of birth. Ms Prouost also spelt her name "Proust" and "de Prouost", although for simplicity we refer to her in this decision as Maria Prouost.

83. A previous director of Stardex (from January 1999 to March 1999), Jass Raykanda, was imprisoned for 4 1/2 years for MTIC VAT fraud in 2006.

84. Mr Hill commenced employment with Stardex in or around November or December 2000, having previously been employed for 6 months by a company called Euro Cellular. Mr Hill worked from home and did not work at Stardex’s premises. Mr Hill remained in Stardex's employment until late 2001 when he left Stardex to set up the Appellant.

85. During his time at Stardex, Mr Hill remained based in Stanmore, trading from home. Maria Prouost lived in Stockport and carried on the business of Stardex in Manchester, but Mr Hill frequently met her in London. As noted above, even after Mr Hill left Stardex he continued to trade on behalf of Stardex and was remunerated on a commission basis. There was no evidence to indicate the period of time for which this commission basis of trading lasted although Mr Edwards' evidence suggested that the commission basis may have continued up to the break in trading by the Appellant in April 2003. However, it is clear that the Appellant repeatedly traded, as principal, with Stardex from late 2003 until 2006.

86. Mr Edwards accepted that during the period in which Mr Hill was employed by Stardex HMRC had not written a letter to Stardex notifying it of any problems in relation to its supply chain.

87. In the year ended 31 March 2006 Stardex's turnover was £232,457,000 and £171,281,000 in the year ended 31 March 2007. Stardex employed five people who were known to HMRC, although Stardex's accounts for the year ended 31 March 2006 indicated that Stardex employed three people and four people in the following twelve month period and nine people in the year ended 30 September 2008.

88. Stardex and the Appellant entered into their first transaction at some time in the period December 2002 to January 2003. Mr Humphries, in his evidence, accepted that his assertion in his witness statement that the first transaction between Stardex and the Appellant occurred in April 2005 was incorrect.

89. Stardex did not carry out due diligence on the Appellant until February 2006 i.e. a few months before the periods to which these appeals relate. In that month Stardex received a company profile report dated 23 February 2006 which it had commissioned from the firm of solicitors, Halliwells, (see below) in respect of the Appellant. The report concluded that the Appellant was a "low risk."

90. Mr Hill was aware that Stardex was not an authorised distributor of any mobile phone manufacturer.

91. In his evidence, Mr Hill said that he had an excellent relationship with Maria Prouost. As well as having a close business relationship, Ms Prouost would also discuss personal problems with Mr Hill and they knew a number of personal details about each other. On the other hand, Mr Hill described Ms Prouost as being "very, very hard.” He also described her as "aggressive" and "she's just beyond control." Mr Hill said that he could not understand why Ms Prouost would put him in the position that he currently found himself (i.e. entangled in an alleged MTIC transaction). Mr Hill plainly regarded Ms Prouost is a hard-nosed businesswoman but with whom he had a close business relationship founded on mutual trust.

92. As described below, the Appellant took a break from trading from around April 2003 until September 2004 following the introduction of "joint and several liability" provisions in the Budget 2003. The VAT returns made by the Appellant for the period 04/03 - 09/04 exhibit a general decrease in trade (with the exception of 03/04 where there were outputs of £1,174,225, 05/04 where there were outputs of £838,176 and period 08/04 outputs of £878,750: see Appendix). A comparison with the returns made by Stardex for the same period shows that Stardex's level of business also declined in a similar fashion. Stardex's average outputs of approximately £46,800,000 per month in the periods 05/02 - 03/03 fell to average monthly outputs of £581,600.

93. Mr Humphries’s evidence was that Stardex had acted as a buffer trader in six fraudulent contra-trading schemes.

94. Stardex was the Appellant’s supplier in all five of the deals currently under appeal.

95. HMRC denied Stardex input tax of £176,244.50 in respect of the period 06/06 and £59,933.72 in respect of the period 07/06. Stardex did not appeal these decisions.

96. Stardex notified HMRC of its intention to appeal against the decision to deny input tax in respect of 08/06 but withdrew its appeal in September 2011.

97. Stardex regularly acted as a broker and had long-established contacts in the EU and Dubai. Mr Edwards, in his evidence, questioned why Stardex would need to involve the Appellant when it was perfectly capable of exporting goods itself.

98. In all the deals in which Stardex supplied the Appellant, the Appellant exported the goods. None of the chains could be traced downwards to a retailer or a wholesaler known to supply retailers or backwards to a manufacturer or an authorised distributor.

99. On the basis of the evidence provided by HMRC in this appeal, including the FCIB evidence in relation to IP addresses, the Appellant now accepts that Stardex was a party to an orchestrated VAT fraud.

CEMSA - EU customer

100. CEMSA was a Spanish company based in Malaga. Its director was a Mr Stephen Russell, an Irish national. Mr Hall's evidence was that the Spanish tax authorities suspected CEMSA of being engaged in illegal activities.

101. CEMSA was the Appellant's customer in all five deals under appeal.

102. On the basis of the evidence provided by HMRC in this appeal, including the FCIB evidence in relation to IP addresses, the Appellant now accepts that CEMSA was a party to an orchestrated VAT fraud. CEMSA has featured many times in MTIC fraud appeals before this Tribunal. We have no doubt, on the evidence in this case, that the Appellant was correct now to accept that CEMSA was a fraudster and we so find.

103. The Appellant's first deal with CEMSA was dated 15 November 2005. Mr Hill’s evidence was that on this occasion CEMSA was so desperate to obtain the mobiles phones, which the Appellant was able to supply, that when asked by the Appellant for a deposit CEMSA offered to pay in full for the goods before they were shipped.

104. We deal with due diligence in respect of CEMSA later in this decision.

Appellant's knowledge of MTIC fraud

105. In his evidence, Mr Hill accepted that he had a general awareness of fraud in the industry but said that he had no idea that fraud was so rife and that HMRC officers did not inform him of this.

106. We have to say that we found Mr Hill's evidence on this point (i.e. that he had no idea that MTIC fraud was rife in his industry) unconvincing.

107. As described above, Mr Hill took a break from trading in mobile phones from April 2003 until the autumn of 2004 (as he appears to have informed Halliwells when preparing their report that the break lasted from around April 2003 to September 2004 - see below) In fact, as described above, the Appellant undertook significant levels of business 03/04, 05/04 and 08/04. This followed the introduction in the Budget 2003 of the new VAT provisions aimed at combating MTIC fraud - the so-called "joint and several liability" provisions. Mr Hill said that he had been sufficiently concerned about the impact of these provisions that he withdrew from trading in order to protect himself and said that, in the meantime, he intended take advice on the likely impact of the new legislation.

108. It seemed to us unlikely that Mr Hill would take the drastic step of withdrawing from mobile phone trading on account of the joint and several liability provisions without being aware that MTIC fraud was widespread in the mobile phone industry.

109. Mr Hill's own evidence was that while he was working at Stardex he became aware of MTIC fraud in February 2001 where the dangers of making third-party payments (i.e. making payments to parties other than the Appellant's immediate supplier) had been drawn to his attention by HMRC.

110. Moreover, the scale of MTIC fraud was brought to Mr Hill's attention in letters from HMRC on 28 July 2003 and 9 March 2004 which noted that the fraud involved mobile phones and that "the current estimate of the VAT loss from this type of fraud in the UK alone is between £1.7 and £2.6 billion per annum." Mr Hill accepted that he had received these letters and that they made clear that MTIC fraud was a huge problem for HMRC.

111. In addition, Mr Hill and his accountant, Mr Price, had attended a meeting on 24 April 2003 at HMRC's offices in London to clarify the position as regards the Appellant's supply chain including Stardex, about which HMRC evidently had doubts. We should note, in this context, that Mr Hill was advised throughout by Mr Price and we find it highly unlikely that any competent accountant would have failed to advise a client active in mobile phone trading of the dangers and prevalence of MTIC fraud.

112. Furthermore, the Appellant received letters from Mr Stone of HMRC (who had also written the letters of 28 July 2003 and 9 March 2004 mentioned above) of 11 January 2005 and 13 January 2005 informing the Appellant that IBO s.a.r.l. ("IBO") and Ashcor Associates Ltd (“Ashcor”), both companies with which the Appellant had recently dealt (in December 2004 and October 2004 respectively), had been deregistered for VAT purposes. It is true that neither letter specifically referred to MTIC fraud, but we consider that in context it was obvious that the de-registration was related to MTIC fraud and we found Mr Hill's protestations to the contrary lacking in credibility. We also note that at no time did Mr Hill enquire of IBO and Ashcor why they had been deregistered even though, in the case of Ashcor, Mr Hill had traded with it since his days at Cellcom. His lack of curiosity seemed to us strange if he genuinely did not know why these trading partners had been deregistered.

113. For these reasons, we concluded that Mr Hill (and consequently the Appellant) not only had a general awareness of MTIC fraud but was well aware that it was a major problem and widespread in the mobile phone industry.

114. We also find that once the Appellant resumed trading, after the break referred to above, it followed the same pattern of trading which it had previously pursued. We could see no significant difference and none was suggested by the Appellant.

FCIB and Banking Evidence

115. The banking evidence in relation to FCIB was given by Mr Payne and by Mr Birchfield. Mr Payne's evidence included an analysis of the documentation relating to the opening of FCIB accounts, IP addresses used, statements of account and an analysis of the five deals relevant to the current appeal. Mr Birchfield's evidence considered the evidence of Mr Humphries and Mr Payne. Mr Birchfield analysed transactions in relation to a number of brokers who feature in relation to the six contra-traders referred to by Mr Humphries (see paragraph 114 above).

116. The evidence of Mr Payne and Mr Birchfield was not challenged.

Payments through FCIB accounts

117. Mr Payne analysed the computerised records of FCIB.

118. In relation to the five deals under appeal, all the parties banked with FCIB. Mr Payne identified the sums paid by each company in the deal chain and the time of each payment.

119. In all five deals payments were made by Vundera to CEMSA, from CEMSA to the Appellant, from the Appellant to Stardex, from Stardex to one of the contra-traders (A - Z or Jag-Tec), from the contra-trader to Kom Team (an EU trader). In all five deals Kom Team then paid a company called RCCI High Tech Limited ("RCCI"), a company based in Cyprus. In four out of five deals RCCI paid a company called SNV Worldwide Limited ("SNV"). SNV then paid the money (in one or more instalments) to Vundera. In other words, the money went round in a circle. In the fifth deal RCCI made a part payment to a company called Satt Telecom which on-paid this amount to Vundera. RCCI made a payment of the balance to a company called Pridewell which on-paid the amount to a company called Zonna GmbH which it then on-paid as part of a larger payment to the Vundera. In other words, the fifth deal also involved circular payments.

120. Mr Payne analysed the timing of the payments. He concluded that the money used by the Appellant to pay Stardex in some transactions had originated with Stardex before travelling in a circle to reach the Appellant.

121. Thus, in 06/06 Deal 1, Stardex's payment to Jag-Tec was made in two tranches £393,150 at 12:27 PM on 27/07/06 and £3,200,000 at 12:54 PM on 26/07/06. Jag -Tec then paid Kom Team in two tranches £247,000 at 2:30 PM on 27/07/06 and £2,800,000 at 3 PM on 26/07/06. Kom Team paid RCCI in two tranches £244,250 at 2:54 PM on 27/07/06 and £2,800,000 at 3:09 PM on 26/07/06. RCCI paid SNV in two tranches £241,500 at 15:00 PM on 27/07/06 and £2,800,000 at 3:15 PM on 26/07/06. SNV paid Vundera in two tranches £263,350 at 6:54 PM on 26/07/06 and £3,200,000 at 7:42 PM on 26/0706. Vundera paid CEMSA in two tranches £3,200,000 at 7:42 PM and £73,600 at 7:51 PM both on 26/07/06. CEMSA paid the Appellant £3,267,000 at 8:03 PM on 26/07/06. The Appellant paid Stardex £3,266,000 at 8:21 PM on 26/07/06. In other words, from start to finish, the main part of the payment circle took seven hours, the Appellant paying on the funds it received from CEMSA to Stardex within 18 minutes.

122. In 06/06 Deal 2, Stardex paid A-Z at 11:18 AM on 25/07/06. A-Z then paid Kom Team at 7:33 PM on the same day. The payment moved round the circle (from A-Z to Kom Team to RCCI to SNV to Vundera to CEMSA), with CEMSA paying the Appellant £2,177,000 at 8:09 PM on 25/07/06 and the Appellant paying Stardex £2,175,000 at 8:39 PM on the same day. In other words, the payments moved round in a circle so that Stardex received its payment from the Appellant only 1 hour 6 minutes after Stardex's supplier, A-Z, paid KOM Team and the Appellant paid Stardex 30 minutes after receiving payment from CEMSA.

123. In 07/06 Deal 1, the payments started with SNV. SNV made its payment to Vundera at 6:30 PM on 04/08/06, Vundera paid CEMSA at 6:33 PM on the same date, CEMSA paid the Appellant £623,300 on 04/08/06 at 6:39 PM and the Appellant paid Stardex £1,515,000 (this amount also included payment for 07/06 Deal 2) on the same date at 7:03 PM. In other words, payments to and from four FCIB accounts (from SNV to Vundera, Vundera to CEMSA, CEMSA to the Appellant and the Appellant to Stardex) took only thirty-three minutes. The Appellant therefore paid Stardex within 24 minutes of receiving the money from CEMSA. Stardex paid the contra-trader A-Z at 12:09 PM on 07/08/2006, A-Z paid Worldwide Wholesale at 12:45 PM on the same date, Worldwide Wholesale paid Grange Solutions at 12:48 PM on the same date and, finally, Grange Solutions paid SNV at 1:24 PM on the same date.

124. As regards 07/06 Deal 2 the payment flows took place on 04/08/06 and commenced with Stardex. From the time when the first payment was made by Stardex to Jag-Tec it took six hours twelve minutes for payments to flow round the circle back to Stardex. The payment chain was: Stardex to Jag-Tec to KOM Team to RCCI to SNV to Vundera to CEMSA to the Appellant to Stardex. Payments involving four FCIB accounts, starting from Jag -Tec through to SNV took only fifteen minutes (from Jag-Tec the Worldwide Wholesale, Worldwide Wholesale to Grange Solutions and Grange Solutions to SNV). CEMSA paid the Appellant £891,000 on 4 August 2006 at 5:48 PM and the Appellant paid Stardex on the same date £1,515,000 (which included an amount in respect of 07/06 Deal 1) at 7:03 PM. In other words, the Appellant paid Stardex 1 hour and 15 minutes after receiving payment from CEMSA.

125. Finally, in relation to 07/06 Deal 3, as regards the different payment flows, payments made on 10/08/06 started with the Appellant and took 17 hours and 51 minutes to return to the Appellant. From A-Z to the Appellant it took one hour and three minutes for the money to leave A-Z and to be received by the Appellant (going through five different FCIB accounts). As regards payments made on 11/08/06 the money started from A-Z and ended with Stardex. The money took three hours forty-five minutes to pass through seven FCIB accounts. CEMSA paid the Appellant £969,000 on 10 August 2006 at 9:30 PM and £621,000 on 11 August 2006 at 4 PM. The Appellant paid Stardex £970,000 on 10 August 2006 at 1:39 AM and £620,400 on 11 August 2006 at 5:24 PM. Thus the Appellant effectively paid on to Stardex the sums which it had received from CEMSA within 4 hours and 9 minutes and 1 hour and 24 minutes respectively.

126. Mr Payne also observed that the amounts paid by the Appellant to Stardex were always less than the amount invoiced by Stardex to the Appellant and that the Appellant only paid on what it had received from CEMSA, sometimes weeks after the invoices had been raised. We shall return at this point about a shortfall in payments later. As regards the timing differences between payments and invoices, the position was as follows:

(1) 06/06 Deal 1 - the invoices were dated 30 June 2006 and the payments were made on 26 and 27 July 2006.

(2) 06/06 Deal 2 - the invoices were dated 30 June 2006 and the payments were made on 25 July 2006.

(3) 07/06 Deal 1 - Stardex and the Appellant's invoices were dated 28 July 2006 and the payments were made on 4 and 7 August 2006.

(4) 07/06 Deal - Stardex and the Appellant's invoices were dated 28 July 2006 and the payments were made on 4 August 2006.

(5) 07/06 Deal 3 - Stardex and the Appellant's invoices were dated 28 July 2006 and the payments were made on 10 and 11 August 2006.

127. Mr Payne also noted that 14 of the transactions in the five deals under appeal used the same IP addresses. Taken together with the witness statement of Mr Letherby, it was plain that Jag-Tec, Kom Team, CEMSA, Vundera, A-Z, RCCI and Pridewell by using the same IP addresses as other parties in the payment circle participated in an orchestrated circular set of transactions. It should be noted, however, that the Appellant did not use the same IP address as any other participant in the payment circle.

128. At one point we noted that the payments recorded in the FCIB evidence in respect of each transaction, when aggregating the profits and losses made by each participant in the chain of payments, equalled zero. Mr Ahmed explained this curious but innocent mathematical phenomenon entirely to our satisfaction and we draw no adverse inference from it.

129. Mr Birchfield's evidence included a spread-sheet which analysed the flow of funds between the parties involved in payment circles which related to the six contra-traders referred to above. His evidence was that these transactions exhibited similar circularity of payments as can be seen in the five transactions under appeal.

FCIB: account opening

130. As already noted, the Appellant applied to open an account with FCIB on 24 October 2005. Maria Prouost gave a reference to FCIB supporting the Appellant's application. The other referee was a Mr Dipam Patel, the managing director of Secure Trader Network Ltd.

131. Mr Hill said that he had opened an account with FCIB because UK banks were withdrawing from the mobile phone market. The Appellant had previously banked with the Bank of Scotland but Mr Hill said that that account had been closed and the Appellant had opened up a new account with Abbey National. When the Abbey National account was closed the Appellant went back to Bank of Scotland and opened a new account. Mr Hill was asked to produce documents to support this sequence of opening and closing accounts and, although agreeing to look for documentation, no documents were produced (notwithstanding a reminder from the Tribunal).

132. In cross-examination, Mr Hill was asked what Maria Prouost had said about FCIB. Mr Hill reported Ms Prouost as saying: "it was a very good bank, it was secure." It was suggested to Mr Hill that his use of the word "secure" indicated that he was using FCIB because his account would not then be transparent as regards HMRC. Mr Hill denied this and said that he intended to use the word "secure" to indicate that his money would be safe and that he had to use a non-UK bank because those banks were withdrawing from the mobile phone market. On balance, we do not draw any adverse inference from Mr Hill's use of the word "secure" - it is quite possible that he used the word merely to indicate that he is money would be safe with FCIB.

133. In any event, it was clear from the questionnaire dated 24 October 2005 that Mr Hill completed in order to open the FCIB account that the Appellant had an account with Bank of Scotland (and had had the account for 18 months) and that the Appellant still had an account with Bank of Scotland in October 2006. We know this latter point because Mr Hill produced certain bank statements in relation to a Bank of Scotland account which covered a period from July to October 2006. Mr Hill said that this was a different account from the Bank of Scotland account which he had in October 2005. We accept his evidence on this point (ie that the two accounts were different) because the address for the Bank of Scotland branch contained on the questionnaire (a Glasgow address) was different from that contained on the bank statements (an Edinburgh address) leading us to conclude that there were two different Bank of Scotland accounts, although exactly when those accounts were opened and/or closed is unclear.

134. CEMSA had applied to open an account with FCIB in November 2004 and Stardex had done so in September 2004.

135. As already noted, the Appellant's first deal with CEMSA was dated 15 November 2005. Mr Hill's evidence was that he had been speaking with Mr Russell of CEMSA for about two weeks before that date. One of the exhibits to Mr Hill's evidence was a fax from CEMSA dated 14 October 2005. The fax was addressed to "Stuart". The opening paragraph of the fax read as follows:

"Further to our telephone conversation I take this opportunity to introduce my company and I look forward to commencing a commercial relationship with your company which I have no doubt will be fruitful to both our companies."

136. The fax later gave details of CEMSA's two bank accounts - the first in the Isle of Man and the second with FCIB.

137. In cross-examination, Mr Hill said that the date on the fax was incorrect and that CEMSA had contacted him in November.

138. At the top of the fax there were fax markings dated 29 December 2005 and 15 November 2005. It seemed to us neither of these markings could correctly record the date on which the fax was prepared (even if may have been sent later). Since the fax was effectively CEMSA's introduction to the Appellant it made no sense for this to have been prepared on the date on which CEMSA and the Appellant entered into their first deal nor for it to have been sent five or six weeks later. In any event, Mr Hill's evidence was that he had been in communication with Mr Russell of CEMSA for some time ("a couple of weeks") prior to the first deal taking place. We therefore conclude that the date on the fax was correct and that contact between the Appellant CEMSA had first been established around the middle of October 2005 at approximately the same time or shortly before the Appellant applied to open its FCIB account.

139. The internal FCIB document recording the details of the Appellant's application to open an FCIB account recorded a Mr Abid Ahmed Mirza as being the "Beneficial Owner" and "Director". HMRC's evidence linked Mr Mirza with a business associate called Mr Yassen Patel, who had connections to MTIC fraud. Mr Hill denied any knowledge of Mr Mirza. Mr Ahmed took us to other similar FCIB account forms relating to other companies where additional names were shown in the same part of those forms as Mr Mirza's name in the form relating to the Appellant. However, on each occasion the account application completed by the prospective customer contained these additional names. The Appellant's account application form contained no reference to Mr Mirza. On balance, therefore, we concluded, as Mr Ahmed submitted, that the account application form reflected an internal error by FCIB.

140. We were not persuaded by Mr Hill's evidence, which was challenged in cross-examination, that he had opened the Appellant's FCIB account because he had no option as a result of UK banks withdrawing from the mobile phone market. At the time the Appellant opened its FCIB account it still had a Bank of Scotland account and the Appellant failed to produce any documentary evidence indicating that in October 2005 the Bank of Scotland was threatening to close the Appellant's account. Secondly, the Appellant opened its FCIB account at around the same time that it started dealing with CEMSA and that Maria Prouost of Stardex had both recommended FCIB and had acted as the Appellant's referee. We infer that the reason why the Appellant opened its FCIB account in October 2005 was linked to the fact that it was about to trade with CEMSA.

Did the Appellant pay Stardex's invoices in full?

141. HMRC argued that the Appellant had failed to pay the full amount invoiced to it by Stardex in respect of all five deals under appeal. The Appellant, on the other hand, submitted that it had made full payment of all amounts invoiced to it by Stardex in respect of these deals but that the payments were made partly through its FCIB account and partly through its account with the Bank of Scotland (Edinburgh branch).

142. In respect of 06/06 Deal 1 Stardex invoiced the Appellant on 30 June 2006 for £3,080,000 plus £539,000 VAT, amounting to a total of £3,619,000. The Appellant paid Stardex on 26 July 2006 from its FCIB account £3,266,000 i.e. less than the gross amount owed.

143. In respect of 06/06 Deal 2 Stardex invoiced the Appellant on 30 June 2006 for £2,054,500 plus £359,537.50 VAT, amounting to a total of £2,414,037.5. The Appellant paid Stardex on 25 July 2006 from its FCIB account £2,175,000 i.e. less than the gross amount owed.

144. In relation to 07/06 Deal 1 Stardex invoiced the Appellant on 28 July 2006 for £588,000 plus £102,900 VAT, amounting to a total of £690,900. The Appellant paid Stardex on 4 August 2006 from its FCIB account £1,515,000 (this incorporated the amount due in respect of 07/06 Deal 2).

145. In respect of 07/06 Deal 2 Stardex invoiced the Appellant on 28 July 2006 for £840,000 plus £147,000 VAT, amounting to a total of £987,000. The Appellant paid Stardex on 4 August 2006 from its FCIB account £1,515,000 (this also incorporated a payment in respect of over 7/06 Deal 1).

146. Finally, in relation to 07/06 Deal 3 Stardex invoiced the Appellant on 28 July 2006 for £1,500,000 plus £262,500 VAT, amounting to a total of £1,762,500. The Appellant paid Stardex on 4 August 2006 from its FCIB account £970,000 and £620,400 on 10 August 2006 and 11 August 2006 respectively i.e. less than the gross amount owed.

147. Thus, HMRC is submitted that the Appellant had failed to pay Stardex full amount due on its invoices argued that this was evidence of the lack of commerciality of the deals between Stardex and the Appellant.

148. The Appellant produced evidence in the form of its bank statements in respect of its account with the Bank of Scotland (Edinburgh branch). These bank statements had, we understand, previously been produced to and examined by HMRC.

149. The account showed two debits in respect of cheques drawn on the account which the Appellant said related to 06/06 Deals 1 and 2 (which were both contained on the same Stardex invoice). The details of the payments were as follows: 1) £300,000 on 01/08/06 - cheque number 14 and 2) £292,037.50 on 09/08/06 - cheque number 18. Adding those amounts to the amounts paid in respect of those deals through the Appellant's FCIB account referred to above, produced a total of £6,033,037.50 i.e. the total amount invoiced in respect of 06/06 Deals 1 and 2.

150. Similarly, in respect of 07/06 Deals 1, 2 and 3, the bank statements showed a payment by the Appellant of £335,000 on 09/08/06 - cheque number 20. Adding that amount to the amounts paid through the Appellant's FCIB account referred to above, produced a total amount of £3,440,400 which was the total amount invoiced by Stardex to the Appellant in respect of those three deals.

151. Mr Hill's evidence was that all these cheques had been paid to Stardex in respect of the invoices relating to the five deals under appeal.

152. HMRC did not accept that the cheques drawn on the Bank of Scotland account were used to "top up" payments to Stardex as the Appellant contended. HMRC contended that there was no commercial logic in using a combination of electronic transfers from the FCIB account and cheques from Bank of Scotland account. HMRC referred to the evidence of Mr Hill that one of the reasons he had opened the FCIB account was to speed up payments into and out of his account. Mr Hill had said in examination in chief:

"… I was being paid - prior to me opening my FCIB account, I was being paid from FCIB accounts and it was taking too long for the money to reach my supplier. So if the payment came from my FCIB account into my Abbey National, it would take two days that the money to actually physically come to me. Then if I was paying another FCIB account, it would take another two days for the money to go the other way. So I thought if I have a FCIB account as well, it will eliminate that process and the transactions can be completed within one or two days."

153. In cross-examination, Mr Hill accepted that there was no documentary evidence to show that the payments went to Stardex because, he said, the cheques had not been returned. He said that the reason that two accounts had been used was because there was not enough money in one account to cover the total invoice and that is why he had paid Stardex from two accounts.

154. Mr Patterson put it to Mr Hill that in the Far Consulting report (see below under heading "Due Diligence") it stated that the Appellant had only one bank account through which deals were traded i.e. the FCIB account. Mr Hill explained that because at the time of the report the Appellant only had the FCIB account and that the Bank of Scotland account (Edinburgh branch) had only been opened at the beginning of July. Mr Hill further stated that that when this Bank of Scotland account was opened he was unable to make telegraphic transfers from the account.

155. Mr Hill was cross examined on the source of the payments into the Bank of Scotland account. Mr Hill explained that one credit to the account of £605,927.07 was a repayment from HMRC. However, there were also credits of £355,997.84 on 14 July 2006 and £599,715.29 on 26 July 2006. There was no evidence as to the source of these payments.

156. Finally, we note that in the FCIB records in relation to the Appellant's FCIB account, the payment of £3,266,000 from the Appellant to Stardex on 26 July 2006 was stated to be a "final payment." In addition, we note that none of the three cheques drawn on the Appellant's Bank of Scotland account was credited to Stardex's FCIB account, although there was no evidence to the effect that Stardex's FCIB account was its only a bank account.

157. Our conclusion on the above evidence, is that, on the balance of probabilities, HMRC have failed to satisfy us that the Appellant did not pay Stardex's invoices in respect of the five deals under appeal in full. The three cheques drawn on Stardex's Bank of Scotland account exactly matched the unpaid balance due to Stardex and, therefore, we infer that it is more likely than not that these payments were made to Stardex.

Freight Forwarder: 1^st Freight

158. The Appellant, as is common with mobile phone traders, did not store stock itself but rather used the services of freight-forwarders. In relation to the five deals under appeal, the Appellant used a freight-forwarder called 1st Freight Ltd ("1st Freight") which was based in Chadwell Heath, Essex.

159. 1st Freight was also used by Jag Tec, A-Z, Kom Team and Stardex. Mr Hill agreed to use 1st Freight on the basis of a recommendation from Maria Prouost. Mr Hill said that she had told him that she had heard rumours of "stock swapping" being practised by some freight-forwarders. On the basis of this recommendation, Mr Hill opened an account with 1st Freight, although there was no written contract evidencing the terms of the Appellant's relationship with this freight-forwarder.

160. Mr Hill did not visit 1st Freight or carry out any due diligence on this company, save in respect of verifying its VRN number. On 3 May 2006 Mr Hill asked HMRC to verify 1st Freight's VRN number. HMRC replied that the number supplied by Mr Hill did not match their records. Mr Hill said that the situation had subsequently been rectified, but there was no documentary evidence to this effect. Mr Hill was not sure when he first dealt with 1st Freight:

"I think the first time I dealt with 1st Freight, off the top of my head, would have been February, March or April, I don't know. Without looking, I don't know."

161. We consider that it is more likely that Mr Hill started to deal with 1st Freight at the beginning of May 2006 when he verified 1st Freight's VRN number.

162. Mr Morehead's evidence was that an analysis of the outbound international consignment notes (the Convention Merchandises Routiers or “CMRs” for short) provided by 1st Freight indicated that more pallets were being shipped out of 1st Freight's warehouse than it had capacity to handle. Mr Morehead had analysed approximately 500 CMRs issued by 1st Freight in the period over 9/02/2006 to 29/09/2006 (in respect of which that the consignee was CEMSA, with deliveries being made on its behalf to GR Distribution in St Folquin, outside Calais).

163. Mr Morehead referred to a meeting that HMRC officers had with a Mr Brandon Burrha of 1^st Freight on 29 November 2006 when Mr Burrha estimated the storage capacity of 1st Freight's warehouse to be 100 pallets (if double stacked). The CMRs dated 30 June 2006 amounted to 410 pallets. These included pallets in respect of 06/06 Deals 1 and 2. The same point was true for the number of pallets included on CMRs for other days. Thus, for example, on 27 April 2006 the CMRs record 314 pallets being shipped, 401 pallets on 28 April 2006, 388 pallets on 25 May 2006, 458 pallets on 26 May 2006, 652 pallets on 31 May 2006 and 321 pallets on 29 June 2006.

164. Moreover, the Appellant's goods were transported by a haulier called K. Elsey. In relation to the CMRs dated 30 June 2006 K. Elsey's vehicle (registration Y54FVX) is shown as having transported 62 pallets. Mr Morehead's evidence was that industry standards required a trailer to carry a maximum of 33 pallets (European size) or 26 pallets (UK size). The pallets were delivered to G R Distribution in St Folquin in France (just outside Calais) on behalf of CEMSA. (We note that G R Distribution also received goods on behalf of Vundera). Indeed, the vast majority of goods in the period 09/02/2006 - 29/09/2006 were delivered to G R Distribution on behalf of CEMSA. Information obtained by HMRC from the French authorities stated that GR Distribution was a buffer company which never undertook actual business activities.

165. Mr Ahmed argued that it was possible that a vehicle may have made two trips within the day. Although possible, we consider this to be unlikely. The fastest journey time from Chadwell Heath to St Folquin is approximately 2 1/2 hours via the Channel Tunnel so that the journey time for two trips would be a minimum of 10 hours, leaving out of account traffic delays, queues at the Channel Tunnel, Customs checks, meal breaks and loading and unloading times.

166. In addition, during a visit in February 2007 HMRC had attempted to obtain from K. Elsey the relevant vehicle tachographs that corresponded with the CMRs but the haulier claimed that its tachographs had been taken by the Vehicle and Operators Service Agency (“VOSA”) - a claim that had previously been made by other hauliers used by 1st Freight.

167. When cross-examined by Mr Ahmed, Mr Morehead's attention was drawn to the exhibits attached to Mr Hall's witness statement which included various CMRs, Eurotunnel tickets and Certificates of Shipment completed by 1st Freight. Mr Morehead acknowledged that the CMR in respect of 06/06 Deal 1 contained a date of 30 June 2006 but bore a receipt stamp of GR Distribution dated 10 July 2006 in respect of 5500 Nokia N 80 mobile phones and another receipt stamp dated 13 July 2006 in respect of another consignment of 5500 Nokia N 80 mobile phones (obviously the total number of 11,000 mobile phones had been split into two consignments). These were accompanied by Certificates of Shipment in respect of each consignment both showing a date of shipment as 30 June 2006 but issued by 1st Freight at Chadwell Heath on 18 July 2006. In addition, there were Eurotunnel tickets dated 10 July and 12 July 2006 in respect of the respective vehicle registration numbers shown on the Certificates of Shipment. It will be noted that the 12 July 2006 date on the Eurotunnel ticket appears to be a day earlier than the GR Distribution receipt stamp of 13 July 2006 on the second CMR.

168. In respect of 06/06 Deal 2 the CMR was dated 30 June 2006 but bore a GR Distribution receipt of 12 July. There was a Eurotunnel ticket dated 12 July 2006 and a Certificate of Shipment showing a shipment date of 30 June 2006 issued by 1st Freight on 18 July 2006.

169. Mr Ahmed's basic point was that the date on which the goods were actually transported to France was not the same as that shown on the CMR. Thus, Mr Ahmed submitted that Mr Morehead's calculation of the number of pallets which were carried on a single vehicle was of doubtful accuracy. Whilst we accept the difference in dates of the CMRs and the date of shipment, it does not, in our view, detract from the substance of the point made by Mr Morehead, viz that the CMRs consistently recorded shipments in excess of the capacity of 1st Freight's warehouse when the position was viewed over an extended period of time.

170. Moreover, the discrepancies between the date of receipt by GR Distribution, the Eurotunnel tickets, the date of the CMRs and the dates on the Certificates of Shipment indicated to us that 1st Freight consistently falsified its CMRs and Certificates of Shipment.

171. This point is reinforced by the evidence of Mr Hall in relation to a CMR which, according to Mr Hall’s evidence, was produced by the Appellant at a meeting between the Appellant and HMRC (attended by Mr Hill and his accountant, Mr Price) held in September 2006. In relation to 06/06 Deal 1, the Appellant provided HMRC with an inspection report from 1st Freight for 11,000 Nokia N80s dated 30 June 2006 indicating that a full inspection took place. In addition, the CMR in question indicated that these mobile phones were transported by an Irish haulier called Daytona Transport Ltd, based in Donegal, from Donegal on 22 June 2006 for delivery on behalf of Jag-Tec to 1st Freight in Chadwell Heath, Essex.

172. This CMR bore a number of stamps and had a manuscript annotation "42 A". A box stamp with a space for insertion of an "invoice number" had been hand completed with the number "1396." Mr Hall could not trace any invoice numbered "1396."

173. In addition, a further copy of a CMR was held by HMRC. This document was blank save for the stamps and handwritten annotations noted on the CMR for the 11,000 Nokia N 80s. This second CMR did not show the sender, consignee, place of delivery, a description of goods and bear any signatures. It did, however, bear the manuscript annotations "42 A" and "1396." The stamps (which included 1st Freight's stamp in the "Goods Received" box) and manuscript annotations were identical to those on the completed CMR and they were in an identical position.

174. Mr Hall's conclusion, which we share, was that the Daytona Transport and 1st Freight stamps had been applied to the form prior to the entry details of the goods being moved being completed and that this cast doubt on the genuineness of the “completed” CMR.

175. Mr Hall also noted that the invoice pursuant to which Kom Team sold the 11,000 Nokia N 80s to Jag Tec was dated 30 June 2006. It is hard to reconcile this with the CMR dated 22 June 2006 which appears to show delivery of the mobile phones to 1st Freight on behalf of Jag-Tec eight days earlier.

176. For these reasons, we consider that the evidence shows that 1st Freight was engaged in fraudulent activity i.e. the falsification of CMRs and that its CMRs consistently showed more goods being dealt in through its warehouse than the warehouse accommodate. In his closing submissions Mr Ahmed accepted there was sufficient evidence to establish that 1st Freight was a party to the fraud.

177. We would also observe that Mr Hall’s oral evidence, which was unchallenged, was that it was his understanding that the 22 June 2006 CMR was handed to HMRC by the Appellant at a meeting on 5 September 2006. Jag Tec was Stardex's supplier. Mr Hill consistently maintained throughout his evidence that he did not know the identity of Stardex's suppliers in relation to the transactions under appeal. In cross-examination, Mr Hill denied that the Appellant had handed over that CMR to HMRC.

178. Mr Hall's evidence that the CMR was handed over by the Appellant at the 5 September 2006 meeting (which Mr Hall did not attend) was delivered orally in examination in chief. By contrast, in his witness statement (paragraph 145) Mr Hall stated that:

"Under Deal 1 period 06/06, 3G has provided an inspection report from 1st Freight for 11,000 Nokia N 80s dated 13 June 2006 indicating that a full inspection took place. A CMR is held which indicates that these phones were collected by Daytona transport from Donegal Town on 22 June 2006 for delivery on the half of Jag Tec to 1st Freight in Chadwell Heath, Essex." (Emphasis added)

179. It seems to us that Mr Hall in his witness statement might be drawing a distinction between the document provided by the Appellant (an inspection report) and a document which HMRC already held obtained from other sources (e.g. from officers dealing with either Jag Tec or 1st Freight). Accordingly, we find that it is more likely than not that the disputed CMR did not come from the Appellant.

Inspection of the goods

1st Freight

180. It was accepted by both parties that Mr Hill did not personally inspect the goods to which the five deals under appeal related. Indeed, at the meeting on 5 September 2006 Mr Hill told HMRC that he had occasionally inspected goods but had not done so for some time.

181. In respect of the 06/06 Deals 1 and 2, 1st Freight's inspection reports dated 30 June 2006 stated that 100% of the phones (i.e. a total of 18,000 mobile phones) had been the subject of a "full inspection" and that 100% of the IMEI numbers had been "notated". The report stated that 1st Freight had carried out the inspections on the instructions of the Appellant. By two invoices dated 3 July 2006 1st Freight invoiced the Appellant a total of £1,250 (excluding VAT) for these inspections at a rate of 10 pence per unit. In other words, 1st Freight said that they had inspected 10,000 mobile phones in respect of the three deals.

182. The 1st Freight’s report in respect of 07/06 Deals 1, 2 and 3 also stated that 1st Freight had carried out "100% full inspection of the… goods and notated 100% of the IMEI numbers."

183. Mr Hall (and Mr Morehead) queried whether it was realistic for 1st Freight to inspect that number of mobile phones in a 24-hour period. Mr Hall calculated that in order for 18,000 units to have been inspected by one person in 24 hours each inspection would have lasted less than five seconds.

184. The persuasiveness of Mr Hall and Mr Morehead's evidence in this regard was diminished by the fact that neither officer knew exactly how IMEI numbers were scanned. For example, Mr Morehead accepted that he did not know whether there were barcodes that could be scanned on the outside of a pallet of mobile phones or whether it was necessary to break down the large cardboard boxes within the pallet. Mr Hall also accepted, in answer to questions from the Tribunal, that he could not explain in detail how the scanning process worked.

185. It is true, however, that it would have been unlikely that each mobile phone was individually examined. It is not clear what was meant by a "100% full inspection" but it seems to suggest that the inspection did not involve sampling. It seems to us, therefore, unlikely that a full inspection of all the goods was carried out in the time available but it was not impossible for the IMEI numbers to be scanned. For example, they could have been (and we were not told) printed on the outside of the packaging so that they were accessible with a hand-held bar-code scanner without breaking down the pallets.

186. Mr Hall and Mr Morehead also noted in their evidence that the schedule of IMEI numbers recorded by 1st Freight (and provided to HMRC by the Appellant) indicated that they were in sequentially numbered batches. For example, in relation to 07/06 Deal 3 the last IMEI numbers scanned preceded the first IMEI number scanned in a deal for another trader which also took place on 31 July 2006. Thus the last IMEI number in the Appellant's deal was 35835300018800 and the first IMEI number in the other party's deal was 35835300018801. Mr Morehead's evidence was that 16,550 IMEI numbers in respect of Nokia N 91s were scanned and 10,000 of these numbers were in sequential order.

187. Similarly, in relation to 07/06 Deal 2 the last number scanned preceded the first IMEI number scanned in respect of another trader's deal.

188. Mr Morehead calculated the number of mobile phones traded via 1st Freight according to the CMRs received and the total amounted to 1,627,530 in the period 9 February to 26 September 2006. Duplication of numbers occurred on seven separate occasions involving 10 separate clients of 1st Freight and occasioned the duplication of 1699 IMEI numbers. A review of the scanning showed that there were 21 blocks of IMEI numbers are different phones for 18 different clients where the IMEI numbers followed on in sequence from one block to another. This equated to some 24,602 IMEI numbers (or 30.9% of the numbers scanned) running in numerical order. Mr Morehead considered it was unlikely that this sequential phenomenon could have occurred by chance. This was particularly so when the mobile phones being traded through a chain of traders (KOM Team, to either A-Z or Jag Tec, from one of them to Stardex and then to the Appellant). Mr Morehead concluded that the scanning had not actually taken place. We accept that conclusion.

189. Mr Hall's evidence was that 131 of the mobile phones sold to the Appellant by Stardex had previously been scanned by HMRC officers visiting freight forwarders and entered into HMRC's national database, suggesting that they had previously been traded within the UK.

190. In addition, 154 mobile phones bought by the Appellant were unusable on the manufacturer's network as the numbers had been blocked. Mr Morehead explained that if a mobile phone operator disables the mobile phone then the IMEI number for the disabled phone would be registered on the mobile phone industry's shared database (Central Equipment Identity Register ("CEIR")). In addition, if a mobile phone is lost or stolen its IMEI number will be registered on the Stolen Equipment National Database ("SEND"). Neither Mr Hall nor Mr Morehead knew whether those databases were available to the public (and thus the Appellant) in 2006. It would have been impossible, in Mr Hall's view, for the batches of IMEI numbers to be sequential when 131 of the mobile phones had been blocked.

A1 Inspection

191. The Appellant's evidence was that the IMEI numbers recorded by 1st Freight would be forwarded by the Appellant to a company called A1 Inspection Ltd ("A1 Inspection"). A1 Inspection maintained a database for the Appellant and compared each batch of IMEI numbers in respect of mobile phones purchased by the Appellant against numbers already entered on the Appellant's database. A1 Inspection would inform the Appellant if any duplication was found. The Appellant's evidence contained invoices from A1 Inspection for the services supplied to the Appellant.

192. Mr Hill accepted that that there was no evidence that A1 Inspection were asked to or in fact did check the Appellant's mobile phones against any wider database other than mobile phones that the Appellant had dealt in.

193. On 30 January 2007 HMRC had written to the Appellant and its accountant asking to see the original A1 Inspection file. However, the Appellant failed to provide these documents. On the fourth day of the hearing the Appellant supplied us with copies of a number of e-mails from the Appellant to A1 Inspection asking A1 Inspection to add IMEI numbers (which were attached to the original e-mails but which were not provided to us) to the Appellant's database. HMRC did not accept that these e-mails asked A1 Inspection to check the IMEI numbers against the Appellant's database. We note, however, that a number of the e-mails asked A1 Inspection to contact Mr Hill if there were any problems. Furthermore, an e-mail dated 7 September 2006 from A1 Inspection to Mr Hill noted that: "I have added 3600 Nokia N 91 to your database and no duplicates were found."

194. One of the e-mails from the Appellant dated 16 October 2006 requested IMEI numbers from A1 Inspection in respect of the deals under appeal. This enclosed a reply (the date of which was not recorded) from A1 Inspection to Mr Hill stating:

"Please find attached Excel Spreadsheets containing your Imeis for the months of June, July and August. If u[sic] can confirm these are fine that would be most appreciated as I don't have a listing of all the Imeis you have sent across to be checked on your database."

195. Although it was not clear to us exactly what was meant by this final paragraph, it seemed clear enough to us that A1 Inspection regarded it as their job to check IMEI numbers supplied by the Appellant against the database held by A1 Inspection in respect of the Appellant's mobile phones. As regards why these documents had not been supplied to HMRC, Mr Hill said that A1 Inspection had had their computers seized by HMRC. He could not, however, point to any reply to HMRC's letter of 30 January 2007 explaining this to HMRC.

196. Moreover, the e-mail from the Appellant to A1 Inspection dated 31 July 2006 which asked A1 Inspection to add the IMEI numbers relating to the three 07/06 deals was dated 31 July 2006 i.e. three days after the date of the invoices by which the Appellant bought and sold the mobile phones which were subject of those three deals. Mr Hill, in cross examination, did not agree that any inspection request was therefore made too late and was merely "window dressing". He said that he could have cancelled the deal and recovered the goods. In our view, asking A1 Inspection to check the IMEI numbers after the Appellant had bought and sold the relevant mobile phones was inconsistent with the Appellant taking reasonable care to ensure that it was not buying goods in which it had previously dealt and was more consistent with "window dressing" then with genuine due diligence.

197. The managing director of A1 Inspection was a Mr Hersh Patel. Mr Hersh Patel's brother, Mr Dipam Patel, had written a reference for the Appellant in relation to the opening of its FCIB bank account. Mr Dipam Patel was a director of Secure Trader Network Ltd ("Secure") (the business address of which was the same as A1 Inspection). The Appellant had purchased mobile phones from Secure in the latter part of 2004 and the first quarter of 2005. A1 Inspection was used by the Appellant to inspect the goods that it was buying from Secure. Mr Hill was unconcerned by Secure's apparent lack of independence, saying that he regularly used A1 Inspection to record IMEI numbers and for inspections of goods and that A1 Inspection was a separate company from Secure. This did not seem to us to be a convincing answer in relation to independence, although the issue was not in our view central to the matters we have to decide.

Insurance

198. Mr Hill's evidence was that the Appellant had insurance for all its goods up to June 2006 "to protect against anything that ever happened to [the Appellant's] stock." Mr Hill said (in his oral evidence but not in his witness statement) that on 15 June 2006 the Appellant's insurance lapsed through an oversight but that he only discovered this in July 2006 i.e. after the deals in 06/06. He then said (and this was contained in his witness statement) that he took the commercial decision not to renew the insurance having had no problems for 3 1/2 years. Mr Hill estimated the annual cost of insurance to be £50,000 (although in the Halliwell's report on the Appellant - see below - the cost was estimated to be approximately £22,000 per annum). As a result, none of the goods (with a value of approximately £8 million) comprised in the five deals under appeal were insured when they were transported to CEMSA. Mr Hill said that he had decided to look at the insurance issue again in September 2006.

199. The invoices and stock offers for the 06/06 and 07/06 deals bore the legend "CIF", the well-known acronym for "cost, insurance and freight", meaning that the Appellant had responsibility for paying for insurance and transport. In relation to the 06/06 deals, Mr Hill said that, as already indicated, he had not realised that his insurance had lapsed. However, in relation to the 07/06 deals Mr Hill explained the inclusion of "CIF" as simply being part of the invoice template and that he did not want to advertise the fact that his goods were not insured. Mr Hill accepted that this was misleading as regards CEMSA and that he was willing to lie in this regard.

200. At a meeting (which lasted for one hour twenty minutes, although Mr Hill claimed to have only attended the meeting for 20 minutes) with HMRC on 5 September 2006 attended by Mr Hill and his accountant (Mr Price) Mr Price is recorded as having told the HMRC officers that the Appellant's goods were covered by an insurance policy. It was put to Mr Hill in cross examination that there was no reference to the fact that for all five deals under appeal there was no insurance. Mr Hill replied that he did not know at what stage of the meeting this question was asked. It may have been said before he arrived. Alternatively, Mr Hill said that he may have gone to the toilet or gone outside for a cigarette when Mr Price discussed insurance with the officers.

201. On 17 November 2006, HMRC wrote to Mr Price asking for a copy of the current insurance policy as the previous policy had expired in June 2006. On 22 November 2006, Mr Price wrote to HMRC explaining that the Appellant's policy had indeed expired in June 2006 and that the Appellant had not renewed the policy because it had become clear that the future conduct the business was:

"extremely uncertain as a direct result of the withholding of our client’s funds. In the light of these circumstances, our client has deliberately chosen not to renew the insurance policy on the grounds of expense."

202. The note of the meeting of 5 September 2006 and HMRC's letter of 17 November 2006 indicated, however, that the policy of insurance which expired in June 2006 had been provided amongst other documents at that meeting.

203. We did not, however, find Mr Hill's evidence on this point credible. The note of meeting refers to Mr Hill at various points. It therefore seems more likely to us that Mr Hill was present when the insurance issue was discussed. At no point in this meeting (or in his witness statements) did Mr Hill explain that the 06/06 deals were not insured by virtue of an oversight or (although this was mentioned in his first witness statement) that a deliberate decision had been taken not to insure the 07/06 deals. The various explanations given about failure by Mr Hill to mention the insurance issue at the September 2006 meeting suggests strongly to us that the Appellant's evidence on this point was unreliable. We accept that Mr Price included the expired insurance policy in the papers which HMRC took away from the meeting but the failure by Mr Hill to mention its expiration was nonetheless misleading. In any event, the explanation tendered by Mr Price in his 22 November 2006 letter could not, in terms of timing, explain the deliberate decision taken not to insure the goods in the 07/06 deals.

Passing title to the goods

204. The invoices issued by Stardex to the Appellant in respect of the deals under appeal all stated:

"Titles [sic] in the goods remain with Stardex (UK) Limited until full and final payment is received."

205. Mr Hill accepted that in all five deals under appeal the Appellant had shipped the goods (worth in excess of £8 million) “on hold” to GR Distribution on behalf of CEMSA in France before either receiving payment from CEMSA or paying Stardex. The Appellant would, in each case, then instruct 1st Freight by fax to release the goods to CEMSA once CEMSA had made payment.

206. Mr Hill's evidence was that he had a verbal agreement with Maria Prouost whereby she permitted him to export the goods prior to receiving payment. In Mr Hill's view, once Stardex had released the goods to the Appellant that was the equivalent of obtaining title. No documents were produced to us to evidence any such agreement.

207. By way of background, in March 2003 the Appellant bought 2,000 Nokia mobile phones from Stardex and sold them to a company called Britwap. Britwap did not pay the Appellant and the Appellant was forced to sell the mobile phones elsewhere at a loss of approximately £11,000. On this occasion, Mr Hill negotiated with Britwap that they should pay his transportation costs. It seems to us that the lesson that Mr Hill should have derived from his experience with Britwap was that it was risky to allow his goods to be exported before receiving full or part-payment.

208. Moreover, there was no indication that the Appellant transacted business on standard conditions of sale which are often incorporated by reference on invoices etc. in commercial transactions. We find this strange because conditions of sale will usually regulate the rights and remedies of the parties to a contract for the sale of goods. In this case, however, notwithstanding the high value of the goods traded the Appellant operated without any obvious contractual terms covering matters such as title, risk, warranties, default, force majeure, claims procedure/ defective products, terms of payment, responsibility for duties and taxes (if any), governing law and arbitration: in other words all the usual issues which one would expect to see addressed in a set of standard conditions of sale in respect of the international sale of goods. The only legend on the Appellant’s invoices was “CIF”. This legend would indicate that the Appellant was responsible for the cost of transporting the goods and for their insurance to the point of delivery to the purchaser or its agent, but did not deal with other issues.

209. It seems to us highly unlikely that in an arms-length commercial transaction a trader such as Stardex, particularly in light of the reservation of title legend on its invoices, would allow over £8 million worth of goods to be released into the possession of its customer (the Appellant) in a way which then allowed the Appellant to export the goods to a third party (CEMSA). Moreover, Mr Hill described Ms Prouost as an aggressive and hard-nosed businesswoman - a picture at odds with the relaxed attitude of Stardex in allowing goods to pass into the hands of the Appellant and to be transported out of the country before receiving payment. It seems equally unlikely that in an arms-length transaction an exporter would deliver the goods to a foreign warehouse (GR Distribution) on behalf of its customer (CEMSA), with whom it had been trading for less than a year, without receiving some payment or at least a deposit.

210. In reaching this conclusion we are aware that, in its first transaction with CEMSA in November 2005, CEMSA paid the Appellant in full before the goods were shipped, although Mr Hill explained the reason for this up-front payment as occurring because CEMSA were "desperate" to obtain the goods. Furthermore, we are also aware that in an earlier deal (Mr Hill, in fact, described it as the first deal that the Appellant transacted on its own behalf) in December 2002 where the Appellant bought from Starmill and exported to a Portuguese company on terms, as recorded on the Appellant's invoice to the Portuguese customer that a 10% deposit was paid and the 90% balance was required on inspection in Portugal. However, there was no indication in relation to the five deals under appeal that the Appellant required CEMSA to pay any deposit and or other payment upfront, leaving the Appellant open to the risk that, if CEMSA rejected the goods, and having to pay transportation costs of bringing the goods back to the UK and possibly having to re-sell the goods at a loss.

Manner of trading and mark-ups

211. As already noted, all the transactions under appeal the goods concerned European specification mobile phones which had 2-pin plugs. Thus, unless the plugs were changed for standard 3-pin UK plugs, the mobile phones could not be used in the UK without a travel adapter. Since no mobile phones are manufactured in the UK it would have been clear to the Appellant that in relation to the five deals under appeal the mobile phones had been imported into the UK, had passed from Stardex to the Appellant and from the Appellant to CEMSA.

212. Mr Hill was aware that CEMSA intended to on-sell the goods that it had bought from the Appellant. Mr Hill accepted that he knew that there must be at least five wholesale traders in the chain.

213. Mr Hill also knew that Stardex exported mobile phones. He knew that there were other wholesalers of mobile phones in France and Spain from which CEMSA could have bought mobiles phones if it had known their identities. Mr Edwards’ evidence was that Stardex had good contacts in Europe and Dubai and regularly exported goods.

214. The Appellant did nothing to the goods - it did not alter them in any way or substitute a three pin for the two pin plugs.

215. The Appellant's mark-ups, and those of other traders in the chain, on the five transactions under appeal are summarised in Appendix 3.

216. From Appendix 3, it will be seen that the contra-traders Jag-Tec and A-Z received a profit of between 0.35% - 0.52%. Stardex achieved a profit margin between 0.69% - 3.49%. The Appellant achieved a profit margin of approximately 6% in every deal. CEMSA always achieved a mark-up of 0.20%.

Due Diligence

Introduction

217. It was not disputed that the Appellant carried out what might be described as "standard" due diligence in relation to its trading partners i.e. verification of the VAT registration of immediate trading partners, obtaining corporate documents such as certificates of incorporation, obtaining letters of introduction, trade references, directors’ ID etc. In our view, such due diligence material would provide a trader with very limited reassurance that its trading partners were bona fide. In practice the organisers of a contra-trading fraud would be bound to ensure that the broker’s immediate trading partners had these standard documents in order.

218. What is unusual in this case is that the Appellant commissioned due diligence reports from the solicitors Halliwells on CEMSA and also obtained a letter (rather than report) from Halliwells in relation to Stardex. In addition, the Appellant commissioned a Halliwells report on itself and on Swift Communications Ltd. The Appellant paid Halliwells £25,000 for these reports. Halliwells were a large and reputable law firm with offices in, for example, Manchester, Sheffield and London.

219. There is no suggestion of any conscious wrongdoing by Halliwells or that the role played by Halliwells was anything other than an innocent one. Nonetheless, as will be seen, Halliwells' letter in respect of Stardex and their report in respect of CEMSA appeared to have significant shortcomings from the Appellant’s point of view and it is now clear that Halliwells’ reassuring findings were very wide of the mark: it is now accepted that Stardex and CEMSA were parties to the MTIC fraud with which these appeals are concerned.

220. Mr Hill's evidence was that Halliwells had first approached him to promote their services to him. They offered to do a number of reports on the Appellant's main suppliers and customers. Mr Hill was aware that Halliwells acted for Stardex and it seems that they came to Mr Hill on Stardex’s recommendation: "They came to me from Stardex - on behalf of Stardex…." The head of indirect tax at Halliwells, Mr Chris Chipperton, had previously acted for Maria Prouost and Stardex when he worked at one of the major accounting firms.

221. We now turn to the Halliwells due diligence and other due diligence in relation to the main parties concerned in this appeal.

Halliwells letter in relation to Stardex

222. Halliwells wrote the Appellant a short letter dated 23 May 2006 in relation to Stardex. We quote below the relevant paragraphs:

"We can confirm that Stardex (UK) Limited is a client of Halliwells LLP and has fully completed our client acceptance procedures for anti-money-laundering purposes. The main shareholder of Stardex, Maria Prouost, is a client of Halliwells and formally was a client of mine [the head of indirect tax at Halliwells] at Ernst & Young LLP.

Halliwells LLP are engaged to conduct on-going reviews of the Stardex (UK) Limited supply chain for both suppliers and customers. In our view this client has conducted the most extensive due diligence of any trader in this sector. It is legal opinion of Halliwells LLP that Stardex (UK) Limited is fully compliant as regards to the joint and several liability provisions in s 77A VAT Act 1994. Every Stardex transaction is individually verified and all suppliers have been independently visited by Halliwells LLP for the purposes of reviewing the supplier's processes and bona fides. Halliwells LLP provides Stardex (UK) Limited with written reports of their reviews to provide a risk rating on each trading company of either low, medium or high. Trading partners are rejected if they do not get the appropriate approval rating.

Stardex (UK) Limited has a good relationship with HM Revenue & Customs and suffers no undue delay in its monthly VAT claim. The level of business can be evidenced by company reporting in annual accounts filed with Companies House, a copy of the latest filed accounts for year ended 31 May 2005 are attached. Stardex (UK) Limited is [sic] well-established business dealing in both mobile telephones and computer processing units. Stardex (UK) Limited trades both within the United Kingdom and internationally. The company has been in business for over 10 years.

We enclose a certified copy of the passport of Mr J McGeechan the company secretary and a registered solicitor in Scotland with SJ Hamilton & Co. For security purposes our client does not allow us to divulge either personal photographs, sample signatures of directors or shareholders, or visits to trading premises."

223. The letter concluded with further information about Stardex including its Company Number, its address and enclosed a copy of its certificate of incorporation. Finally, the letter provided confirmation that Stardex banked with the Royal Bank of Scotland and FCIB.

224. We note that in the footer to the letter the file is described as: "correspondence/letter to Mobile Express". Mobile Express was the EU customer of Stardex in deals between April and June 2006 in which Stardex bought from the contra-trader Red House. It is therefore indicates that Halliwells used this letter on more than one occasion and suggests that this letter was written in their capacity as solicitors to Stardex rather than as an adviser to the Appellant. This conclusion is reinforced by references in the letter to Stardex being Halliwells' "client."

225. There are a number of points to be noted about this letter. First, Stardex was a pre-existing client of Halliwells and Maria Prouost had been a contact of Mr Chipperton when he had worked at Ernst & Young. Although Halliwells had been commissioned by the Appellant to write this letter, it is plain that there was a conflict of interest - Stardex were a client of Halliwells. HMRC criticised the letter as not being an independent report, that Mr Hill should have been aware of this fact and placed limited weight upon its contents. In our view this criticism was justified. In cross-examination, Mr Hill professed not to accept (or, it seemed, to understand) this point, stating that he had instructed Halliwells and therefore they worked for the Appellant rather than Stardex, although acknowledging that Stardex were a prior and existing client of Halliwells. Mr Hill was plainly, in our view, an intelligent businessman and we found his apparent failure to grasp this basic point in relation to an obvious conflict of interest unconvincing.

226. Secondly, the Halliwells letter in relation to Stardex covered one and a half pages (with 13 pages of appendices). This is in stark contrast to the reports that Halliwells produced in relation to CEMSA which ran to over 40 pages (including appendices), in relation to the Appellant itself which ran to over 100 pages (including appendices) and in relation to Swift Communications which (with appendices) ran to over a 50 pages.

227. Thirdly, Mr Hill accepted in cross-examination that the Halliwells letter of 22 May 2006 was the only substantive due diligence that the Appellant had performed on Stardex (although we understand that routine VAT registration checks were made) since it began trading with Stardex more than 4 1/2 years earlier. Whilst we would accept that in the period immediately after his employment with Stardex Mr Hill could reasonably have felt comfortable that he knew enough about Stardex to be confident in trading with them, that level of comfort could not reasonably be sustained for a period of over four years.

228. Fourthly, in the light of the evidence put forward by HMRC to the effect that Stardex was deeply and repeatedly involved with a web of fraudulent contra-traders, Halliwells’ statement that every Stardex transaction was individually verified and that all suppliers were independently visited by Halliwells the purposes of reviewing the supplier's processes and bona fides is remarkable. We call to mind Mr Humphreys' evidence that Stardex bought directly from all six of the ring of contra-traders. It does lead to questions about the level of competence and professional attention that was engaged in this task. Nonetheless, although that statement (and others e.g. "[Stardex] has conducted the most extensive due diligence of any trader in the sector.") now, with the benefit of hindsight, makes uncomfortable reading, a bona fide recipient of that letter would have been justified in drawing comfort from it.

229. Finally, HMRC made much of the fact that Stardex would not allow them to divulge personal photographs or signatures of directors and shareholders. We do not place so much weight on this as HMRC and it could have seemed plausible to Mr Hill that Stardex might refuse to do this for security reasons. What did catch our attention, however, is the statement that Stardex did not allow visits to trading premises. This seems very odd in relation to a purported due diligence exercise. If it is correct that Stardex commissioned Halliwells to send out this sort of letter to its trading partners then it is strange that Stardex did not permit Halliwells to visit their offices. In our experience, it would be highly unusual for a law firm to carry out due on a company without visiting its premises, interviewing staff and inspecting its files.

230. Mr Hill accepted that he too had never visited Stardex’s premises.

Halliwells’ report in relation to CEMSA

231. Beyond the usual verification of trade application, incorporation details and VAT registration check, it was accepted that the Appellant had not performed due diligence on CEMSA when it first traded with CEMSA in November 2005 and Mr Hill had never visited their offices or met Mr Russell. We discuss the issue of Dun & Bradstreet reports below.

232. Halliwells' report in relation to CEMSA was dated 29 June 2006, although it was sent to the Appellant by registered post on 5 July 2006. This was almost one week after the deals in 06/06 were concluded. The contents of the report had, Mr Hill said, been discussed by Halliwells with him in the course of June i.e. before the written report was sent to him and before the 06/06 deals were concluded.

233. Apart from the appendices, the report ran to 6 pages. It concluded as follows:

"Although some checks could not be undertaken during our visit (because Mr Russell had forgotten that it was a local public holiday when we agreed the date of our visit), CEMSA seems a well-founded and established business which is happy to have a good local profile. It has been in business for a number of years, Mr Russell presents himself as an experienced businessman with a long track record in this business sector.

We would rate this business as low risk."

234. As noted above, CEMSA was run by Stephen Russell, an Irish national. The report states that Mr Russell declined to offer any proof of personal identity, although he proffered a business card and was happy for an identification photograph to be taken. Mr Hill accepted that this failure to provide personal ID had not been mentioned to him by Halliwells over the telephone. Mr Hill never met Mr Russell and never visited CEMSA's premises personally.

235. The report noted that CEMSA had two bank accounts, one in the Isle of Man and the other with FCIB. No query seems to have been raised over the fact that an established business with a substantial turnover (see below) based in Marbella in Spain had, apparently, no Spanish bank account.

236. The report noted that creditworthiness details about CEMSA were held with a company called e-informa, a Spanish associate of Dun & Bradstreet. Halliwells obtained a copy of the e-informa report. They did not provide a formal translation, but included the following information concerning CEMSA’s turnover for 2002 to 2004 as follows:

2002 Euros 85.3million

2003 Euros 109.3 million

2004 Euros 9.8 million

Remarkably, the report related that Mr Russell had estimated the sales revenue for 2005 as Euros 500 million (approximately £300 million at the then exchange rates) and that that figure would be substantially exceeded for the current (2006) financial year. The report noted:

“No explanation was given for the substantial fluctuation in sales turnover in recent years.”

237. This very substantial increase in gross income arose in the context of other comments in the report to the effect that CEMSA did not have a functional web-site and operated “largely by word of mouth and by personal recommendation.”

238. When asked in cross-examination whether the increase in turnover was remarkable, Mr Hill replied:

"A little, but nothing that would have flagged me up, you know. Obviously with every report, even the report on my company and even on this company… there are things that obviously will draw my attention to. But I'm looking at the whole package, okay, and what I got from this report was that this company was low risk and I was okay to trade with them."

239. It seemed to us extraordinary that this increase in turnover did not prompt further enquiries either by Halliwells or Mr Hill. A large increase in turnover by a company may have an entirely innocent explanation - although in the case of CEMSA, we now know this to be improbable - but at the very least it calls for enquiry. The purpose of “due diligence” is not simply to ask questions and obtain information as an end in itself but to analyse and evaluate the information obtained.

240. The sense of unease about the thoroughness of this report was heightened by the fact the report outlined CEMSA’s own due diligence:

"Mr Russell says that he uses the VIES websites to validate the VAT numbers of the traders involved in each deal CEMSA is involved in. He has met most of his customers and suppliers face-to-face. He ensures that he obtains a Certificate of Incorporation, VAT certificate and a trade application from each trading partner (see CEMSA's documents at Appendix K). He pointed out that there is no obligation under Spanish law to carry out formal due diligence on suppliers or customers. He says he is comfortable with his own commercial risk analysis, particularly as he does not offer credit and retains title to goods until he receives payment."

241. In some respects, this is one of the most critical paragraphs of the report. A trader in the position of the Appellant would be concerned to know what steps CEMSA took to ensure that its customers and suppliers were bona fide traders. The Appellant, however, could have obtained very little comfort from this paragraph. Obtaining certificates of incorporation and VAT certificates etc., although necessary, would tell a trader very little about the bona fides of a trading partner. Also, simply meeting "most" customers and suppliers face-to-face would reveal very little useful information unless accompanied by probing enquiries. When asked whether CEMSA's statement that it had no obligation to perform due diligence on its customers and suppliers rang any "alarm bells", Mr Hill replied: "A little bit."

242. Furthermore, it is clear that much of the information contained in the report, other than that obtained from publicly available documents, was provided by Mr Russell and that there was little or no independent verification of the statements that he made.

243. In the light of the above, although it is important not to be wise with the benefit of hindsight, it is hard to understand how Halliwells reached the conclusion that the fraudster CEMSA was "low risk." It is difficult to avoid the conclusion that Halliwells' due diligence in respect of CEMSA was superficial and unsatisfactory.

244. Furthermore, it emerged in cross-examination of Mr Hill that his main concern in relation to Mr Russell of CEMSA was to satisfy himself that Mr Russell was, as Mr Hill put it, "loaded" i.e. wealthy. He repeatedly used this word to describe Mr Russell. At one point Mr Hill said:

"The most important thing I needed to know was: was this guy loaded, okay, was he telling me what was [sic] - was it correct and was he a risk."

245. It was hard to avoid the impression that by focusing on whether Mr Russell was "loaded", Mr Hill was not applying his mind to the appropriate question, viz whether CEMSA was a bona fide trader.

Halliwells report in relation to the Appellant

246. Halliwells also prepared a report on the Appellant which Mr Hill described as being for his own purposes and not for the purposes of being shown to third parties. This report was dated June 2006.

247. It is not necessary for us to dwell at length on this report since it does not bear directly on the Appellant's state of knowledge in relation to its trading partners. However, at paragraph 10 of the report ("Conclusion and recommendations") Halliwells stated:

"10.3 Halliwells LLP are satisfied with the attention to detail displayed within the company around the issued due diligence and also with the generally efficient and clearly profitable way that the business is run.

10.4 This report contains the caveat that Halliwells LLP is in the process of completing the supply chain review, so comment cannot be made on the risks contained within the actual chain, although it is clear that the company has done much to its processes to minimise any VAT risks.

10.5 Halliwells make the following recommendations:

· Complete a full supply chain review to ascertain any risks inherent in the supply chain;

· Complete due diligence visits to complement the existing internal processes;

· Maintain a fully documented rejected business file; and

· Verify that [the Appellant] have their own dedicated file on the A1 Inspection file.

10.6 Subject to the above recommendations, Halliwells LLP are happy to rate [the Appellant] as being in the low risk category in this industry sector."

248. No evidence was produced to indicate that Halliwells ever completed their supply chain review. The recommended rejected business files were likewise not produced in evidence and it is not clear whether they were ever brought into existence.

249. The report contained an appendix headed "Standard Trading Terms". In fact, the appendix contains nothing apart from the address and company registration and VAT number, contact information and bank details.

250. In addition, Halliwells commissioned Far Consulting Ltd ("Far Consulting") to produce a financial due diligence review of the Appellant. A draft of this report was produced in evidence.

251. Finally, the June 2006 report on the Appellant was not the first report written by Halliwells on the Appellant. As already noted, a Halliwells report dated 23 February 2006 on the Appellant was commissioned by Stardex. That report rated the Appellant as "low risk." In addition, the report indicated that the Appellant sought "pre-payment on exports." We note that there was no comparable statement in the June 2006 Halliwells report.

Due diligence - general

252. As a general matter, in the course of his evidence Mr Hill said that he placed emphasis and importance upon knowing his trading partners rather than formal due diligence enquiries, including supply chain checks.

253. There was also some debate about whether the Appellant had obtained Dun & Bradstreet reports in respect of its trading partners. No such reports were produced in evidence because Mr Hill said that he had been unable to find them. Mr Hill said that he made Dun & Bradstreet checks on trading partners which he did not already know. Mr Patterson, for HMRC, queried whether these reports had existed because they had not been produced and it would have been clear to the Appellant from an early stage that these records would have been important in relation to the issue of due diligence.

254. In paragraph 30 of the Far Consulting report there is a reference to the Appellant making "Dun & Bradstreet checks which were maintained on the permanent diligence file." This suggests that the Appellant did make some Dun & Bradstreet checks in respect of trading partners. In addition, in the purchase ledgers of the Appellant attached to one of Mr Hall's witness statements there was a debit for £3,787.50 on 6 March 2006 in favour of Dun & Bradstreet which Mr Hill explained represented the cost to the Appellant of Dun & Bradstreet reports on its trading partners. Mr Hill said in re-examination that he had obtained a report in respect of CEMSA and the information it contained gave him no cause for concern. We conclude that the Appellant did obtain some Dun & Bradstreet reports but there was insufficient evidence to establish in respect of which companies reports were obtained. It seemed odd that Mr Hill had mislaid such clearly material documents and we found his explanation why he had not asked Dun & Bradstreet to send him duplicates unconvincing. Mr Hill stated:

"… I didn't feel it was necessary because, as far as I was concerned, all my due diligence in relation to everything was then done by Halliwells. You're talking about stuff going back into 2005. I wasn't aware that I was going to have to be having to give evidence on stuff in 2005. I thought we were here to discuss the fact that my repayment was denied the 2006."

255. It must have been clear to Mr Hill from an early stage after the denial of his claim for repayment of input tax that his due diligence in relation to his trading partners was an important issue and we found his reaction to the question of duplicate Dun & Bradstreet reports less than credible.

256. In any event, as far as we are aware, there was no evidence of any Dun & Bradstreet report having been obtained in respect of CEMSA or any other trading partner prior to March 2006.

Reliability of Mr Hill's evidence

257. The reliability of Mr Hill's evidence was repeatedly challenged in cross-examination. We have considered Mr Hill's evidence in considerable detail and we have concluded that he was an unreliable witness. On a number of occasions his replies seemed to us to be evasive or lacking credibility.

258. We set out below examples where we considered Mr Hill's evidence was unreliable.

259. Mr Hill claimed that his accountant, Mr Price, had not informed him of concerns raised by an HMRC officer at a meeting at the accountant's offices in August 2004 in which the officer expressed concern about evidence of "circularity" in the March 2004 return. This related to a period shortly after the Appellant had resumed trading after the period of inactivity following the Budget 2003. It seems to us highly unlikely that any responsible professional adviser would fail to mention such a matter particularly in circumstances where his client had taken a break in trading as a result of his concerns about fraud in his industry.

260. In December 2004 the Appellant had sold a consignment of mobile phones worth more than £300,000 to IBO. On 11 January 2005 HMRC wrote to the Appellant notifying it that IBO Sarl had been deregistered for VAT purposes with effect from that date. Mr Hill was asked about his reaction to the deregistration of a company that he had dealt with so recently. Mr Hill said that his reaction was to ask himself whether this was a mistake. Also, he noted that this was a general letter that was not specifically related to the transactions he had just undertaken with IBO Sarl. He said that HMRC's letter did not tell him what had happened. When asked whether he had contacted any of the HMRC officers with whom he regularly had contact for clarification, Mr Hill replied that if there had been a problem he assumed HMRC would contact him. We found Mr Hill's responses unconvincing and evasive.

261. On 13 January 2005 HMRC sent to the Appellant another letter, this time about Ashcor Associates, informing the Appellant that Ashcor Associates had been deregistered. The Appellant had bought mobile phones worth over £200,000 from Ashcor Associates in May 2004 and almost £300,000 worth of mobile phones in October 2004. Mr Hill said that he didn't know what deregistration letters, such as the one in respect of Ashcor Associates, meant:

"Like I said, when I used to get these letters, I don't know what they mean, apart from the fact - what they're telling me, as far as I'm concerned, customers [Customs] are telling me, "You want to deal with Ashcor Associates in the future, make sure they've got a valid VAT number or they may be an issue in regard to your reclaim."

262. Mr Hill indicated that he did not understand these letters in relation to IBO Sarl and Ashcor Associates (and other similar letters in relation to other deregistered traders) to relate to MTIC fraud.

263. It is true that these letters did not specifically refer to MTIC fraud, but in the circumstances of an industry in which fraud so concerned to the Appellant that it ceased trading for a period of time in the wake of the Budget 2003 and Mr Hill's obvious extensive knowledge of the mobile phone trading industry we found Mr Hill's replies wholly unconvincing.

264. In relation to Dun & Bradstreet reports and obtaining duplicate reports, we have already indicated that we did not find Mr Hill's evidence reliable.

265. We have already referred to the meeting on 5 September 2006 between HMRC officers, Mr Hill and Mr Price. Mr Hill claimed that he may have gone outside to the toilet or for a cigarette when insurance was being discussed. In addition, the note of that meeting records that Mr Hill, when asked why the Appellant's UK bank accounts had been changed for the fourth time, gave "very little explanation." In cross-examination Mr Hill was asked why he gave such little explanation. Mr Hill replied:

"Maybe he asked us in a way that I didn't understand, but he did not specifically turn around and say to me, "Mr Hill, you have had four bank accounts in this amount of time, what is your explanation", and then I just didn't give him an answer. That was not the question. I don't remember him asking that. Maybe he asked Robert [ Mr Price], you know, while I was outside having a cigarette or something like that, I don't know."

266. For a meeting which Mr Hill claimed to have attended only for 20 minutes (as opposed to the one hour 20 minutes duration recorded on the manuscript version of the note), a surprising number of significant issues seem to have been discussed, according to Mr Hill, in his absence or which he appears not to have remembered. In relation to the question about bank accounts, the note recorded the question as being asked of Mr Hill, not Mr Price. We considered Mr Hill's replies in cross-examination both as regards insurance and bank accounts and to be evasive and untruthful.

267. We note also that Mr Hill admitted that he had been prepared to lie to CEMSA in respect of the existence of insurance cover.

268. We heard evidence that the Appellant had made a series of short-term (usually just a matter of a few days) loans to one of its suppliers, Unique Distribution Ltd ("Unique Distribution") in 2003 and 2004 during the Appellant's break from trading following the Budget 2003. Unique Distribution was an authorised distributor. The total amount the loans was approximately £1.7 million, although the total outstanding at any one time did not seem to exceed £250,000. Loans were made in every month from July 2003 to February 2004 (two loans were made in September 2003 and a final loan was made in April 2004). Mr Hill was asked about these loans:

"A. There were loans made, I am not disputing that fact at all.

Q. Spell it out.

A. I lent Unique Distribution money.

Q. Right. Go on.

A. What else is there to say, I let them money?

Q. Why were you lending money to Unique?

A. Because they asked me to lend the money.

…

Q. So why did Unique want to borrow these sums?

A. Because they didn't have enough money at the time to conduct their business. So they were looking to source money for short periods of time, like short-term loans."

269. In the event, nothing seemed to turn on this curious arrangement with Unique Distribution. However, Mr Hill's reluctance fully to explain the arrangement seemed to us to indicate a degree of evasiveness.

270. We concluded that Mr Hill's evidence had to be treated with considerable caution.

Discussion

General

271. As already discussed, the question in this appeal is whether the Appellant knew or should have known that its five deals under appeal were connected to the fraudulent evasion of VAT.

272. It is clear from the FCIB evidence that the fraudulent contra-trading scheme of which these deals were pre-arranged by a mastermind. The conspiracy required that deals in the “clean” and “dirty” chains were coordinated so that the supplies made by the fraudulent contra-trader in the "clean" chains roughly approximated, and thereby disguised, the exports made in the "dirty" chains. None of this necessarily indicates that the Appellant knew of this, but as a fact it cannot be in dispute.

273. We should make it clear that we have applied the principles of law set out in an earlier part of this decision, particularly the judgments of Lewison J in Brayfal Ltd v HMRC [2011] UKUT 99(TCC), Moses LJ in Mobilx Ltd v HMRC [2010] EWCA Civ 517 (including, especially, Moses LJ's approval of the comments of Christopher Clarke J in Red 12 Trading Ltd v Revenue and Customs Comrs [2009] EWHC 2563 (Ch) at [109]–[111]) and the decision of Briggs J in Megtian Limited v HMRC [2010] EWHC 18 (Ch).

274. Our decision also proceeds on the basis that, as the director and sole shareholder of the Appellant, the actions and knowledge of Mr Hill should be attributed to the Appellant. No argument to the contrary was put to us on behalf of the Appellant.

275. We have reached the conclusion after carefully considering all the evidence cumulatively that the Appellant knew that its five deals were connected to the fraudulent evasion of VAT.

276. We set out below our reasons for reaching these conclusions.

The Appellant's knowledge of the high level of MTIC fraud

277. The Appellant accepted that Mr Hill was aware that there was fraud in the mobile phone industry as a whole - his knowledge was described as a general awareness. For the reasons given earlier in this decision, we consider that Mr Hill had considerable knowledge of MTIC fraud and that this fraud was rife in the mobile phone industry.

Relationship with Stardex and Maria Prouost

278. The Appellant accepted that Stardex and Maria Prouost were part of the conspiracy to defraud HMRC.

279. Mr Hill had a close business relationship with Maria Prouost for over five years. At her invitation, he originally worked for Stardex (although there was no evidence that at that time Mr Hill was aware of any fraudulent trading by Stardex). In addition, when he subsequently set up his own business, he continued to trade on behalf Stardex on a commission basis even though he was competing with Stardex on a principal basis. Futhermore, Maria Prouost confided matters of a personal nature to Mr Hill. Mr Hill opened up an FCIB account on the recommendation of Maria Prouost and she assisted the process by providing a reference for the Appellant. The account was opened at around the time when the Appellant first traded with the fraudster CEMSA. Maria Prouost also recommended the freight-forwarder 1st Freight, which the Appellant now accepts was part of the fraud and in respect of which he carried out no significant checks. It also appears that Halliwells came to the Appellant via Maria Prouost. Mr Hill repeatedly traded with Stardex and his evidence was that he was in frequent, almost daily, telephone contact with Maria Prouost and often met her in London.

280. Both the business of Appellant and Stardex's businesses showed a significant reduction in the level of activity in the period 04/03 to 09/04 i.e. in the period after the introduction of the joint and several liability counteraction provisions in the Budget 2003. We infer from this that the business of the Appellant and that of Stardex were closely linked.

281. Moreover, Mr Hill's evidence in relation to Maria Prouost seemed to be inconsistent. On the one hand he presented her as an aggressive and hard-nosed businesswoman but on the other hand she was, according to Mr Hill, so relaxed that she was content for the Appellant to ship outside the UK valuable consignments of mobile telephones, for which the Appellant had paid nothing. This did not ring true and we considered that this undermined Mr Hill's claims that he had no knowledge of Maria Prouost's fraudulent activities.

282. Furthermore, Mr Hill's evidence in relation to his knowledge of Stardex's suppliers was inconsistent. At one point he claimed that he did not know the identity of Stardex's suppliers (save those with whom he had dealt while employed by Stardex), but subsequently (when seeking to explain why he did not by-pass Stardex and deal directly with their suppliers) he appeared to know who they were. This was another aspect where we regarded Mr Hill's evidence is unsatisfactory.

283. Apart from routine verifications of VAT registration and corporate documents, the Appellant carried out no due diligence on its trading partner Stardex until May 2006. Even then the Appellant's due diligence consisted of obtaining a letter from Halliwells which was by no means independent since the firm already acted for Stardex. As we have already said, in the period immediately after Mr Hill left Stardex's employment we could understand that he may have considered that he knew Stardex well enough not to undertake extensive due diligence. That excuse, however, could not be an indefinite one. We consider that the reason why Mr Hill did not undertake more extensive due diligence on Stardex at an earlier stage than May 2006 (e.g. 2004) was because his relationship with Maria Prouost and Stardex was much closer than he has admitted.

Relationship with CEMSA

284. The Appellant did not undertake substantive (i.e. apart from routine verification of VAT registration and corporate documents) due diligence in relation to CEMSA until June 2006, having undertaken its first transaction with CEMSA in November 2005. It is, however, possible that a Dun & Bradstreet report was obtained in or after March 2006, although we treated Mr Hill’s evidence on this point with some caution.

285. We consider that undertaking due diligence with CEMSA at the same time as or immediately before entering into transactions which were connected with fraud (i.e. the transactions under appeal) was a remarkable coincidence.

286. A similar coincidence, which we have already noted, was that the Appellant opened a bank account with FCIB at or around the time of its first transaction with CEMSA in November 2005. As the Appellant's application form to FCIB indicated, the Appellant already had an existing bank account with Bank of Scotland. We did not consider Mr Hill's explanation that he opened an account with FCIB because UK banks were closing the accounts of mobile phone traders to be convincing. There was no evidence that Bank of Scotland were threatening to close the Appellant's account and, although invited to produce the relevant correspondence and agreeing to search for the relevant correspondence, Mr Hill failed to produce any further documentation.

287. In the context of the evidence as a whole, we considered that the two coincidences (of the Appellant's suddenly undertaking substantive due diligence on CEMSA by means of a Halliwells report at the same time as or immediately prior to entering into the deals under appeal and the coincidence of the Appellant opening in FCIB account at or around the time when it first began trading with CEMSA) can most readily be explained by the Appellant being aware that CEMSA was involved in fraudulent activity.

288. Mr Ahmed submitted that on the first three occasions that the Appellant dealt with CEMSA the Appellant paid a deposit and that it was inconceivable that those controlling the fraud would have wanted the Appellant to be paid in advance. Mr Ahmed submitted that if Mr Hill was not knowingly involved in the earlier Stardex/CEMSA transactions, it was unlikely that he knew that the later deals under appeal were also connected with fraud. It seems to us that even if (which we doubt) earlier deals may not have been entered into by the Appellant knowing that they were connected with fraud, it does not follow that later deals were innocent. Those later deals must be looked at in the light of all the circumstances surrounding them and those circumstances indicate to us that the Appellant did have knowledge that the five deals under appeal were connected with fraud.

FCIB evidence

289. The FCIB evidence, particularly the circularity of payments and the common IP addresses used by some participants (not including the Appellant), indicated that the five appealed deals were part of a complex and carefully organised fraud which must have had an organiser or mastermind. All the parties banked with FCIB - a fact which, on itself, is a strange coincidence. Surely, if UK banks were closing down the accounts of mobile phone traders those traders would open accounts in a variety of different jurisdictions. Instead, they all opened accounts with FCIB in the Caribbean.

290. In the five transactions under appeal, the Appellant received payments from CEMSA and paid Stardex through their FCIB accounts. The on-payments were made by the Appellant within 18 minutes (06/06 Deal 1), 30 minutes (06/06 Deal 2), 24 minutes (07/06 Deal 1), 1 hour and 15 minutes (07/06 Deal 2) and 4 hours 9minutes and 1 hour 24 minutes (07/06 Deal 3). In 07/06 Deal 3 the Appellant's on-payment to Stardex was made at 1:39 AM. In summary, the rapidity of the on-payments by the Appellant in the context of rapid payments in the circular chain suggests to us that the Appellant was well aware of the need to play its role in a pre-arranged series of payments.

291. Moreover, in the light of this evidence, it seems to us more likely than not that the organiser of the fraud would have needed to be certain that the chain of transactions and payments would proceed as planned. It seems to us highly unlikely that the mastermind behind the fraud would have allowed an innocent party to take part in the transaction chains. It would have been necessary to ensure that the goods were bought from and sold to and the payments made by and to the correct (pre-planned) parties. The five appealed deals were part of an elaborate larger contra-trading fraud of considerable sophistication where the contra-trader was seeking to disguise its transactions in the corresponding "dirty" chain. HMRC’s evidence to this effect was not challenged. Organising and balancing the "clean" and "dirty" chains, with the consequent need to organise paperwork, transport, inspections and payments was complicated enough in itself without introducing into the equation unwitting parties whose behaviour, because of their lack of knowledge, could not be easily predicted. Furthermore, introducing an innocent party to act as broker would always run the risk that the innocent trader might "smell a rat" and take its concerns to HMRC.

292. Mr Ahmed argued that introducing an "innocent dupe" to act as broker in a transaction may have benefits to the organising fraudster in that the fraudster's money was not at risk in relation to the VAT repayment claim. Whilst this is true, and it is possible that in some cases this attraction may conceivably have led to innocent parties becoming involved in MTIC transactions, we very much doubt whether this advantage would outweigh the disadvantages outlined in the preceding paragraph.

293. Furthermore, there seemed no good reason why the Appellant opened an FCIB account in November 2005. We have already remarked upon their coincidence in the timing of the opening of this account with the Appellant's first deal with CEMSA. The Appellant argued that it opened an FCIB account largely because UK banks were withdrawing from the mobile phone trading sector. Nonetheless, the Appellant's FCIB application made clear that the Appellant had an account with the Bank of Scotland and, accordingly, there seemed no pressing need to open another bank account. We consider that it is more likely than not that the Appellant opened the FCIB account in order to participate in transactions which it knew would in some manner be connected with fraudulent evasion of VAT. As Briggs J pointed out in Megtian it is not necessary for HMRC to show that a trader was aware of the particular type of fraud or whether the fraud occurred in its supply chain or another chain.

Insurance

294. The evidence was that until the five deals under appeal the Appellant always insured its goods. In these five deals five consignments of valuable mobile phones were shipped from the warehouse of 1st Freight to the warehouse of GR Distribution near Calais which received the goods on behalf of CEMSA.

295. Mr Hill's evidence was that he had overlooked the fact that the contract of insurance had expired shortly before the deals in 06/06 and that when he entered into the deals and authorised the shipment of goods he was unaware that he was not covered by insurance. As regards the deals in 07/06, Mr Hill said that he had taken a commercial decision not to insure the goods. His evidence was that the cost of insurance in the period June 2005 to June 2006 had been approximately £25,000. Because the level of business and increased throughout the year the renewal premium was likely to be greater. Mr Hill said that he decided not to insure the 07/06 goods and also decided to wait until September to decide whether to renew the policy.

296. We did not find Mr Hill's evidence credible. We note that at no stage in the investigation of the transactions by HMRC or in any of Mr Hill's three witness statements was it claimed that the failure to insure the goods comprised in the 06/06 deals was the result of an oversight. This claim was only made when Mr Hill gave oral evidence. We did not find this "oversight" claim to be credible - it seemed to us to be a last-minute invention. Moreover, whilst taking a commercial decision not to insure the goods in the 07/06 deals saved Mr Hill from paying the premium, we could not understand why he would take this risk on this deal when in all previous deals he had made sure that his goods were insured. Indeed, the use of the "CIF" legend by the Appellant on its invoices etc. indicated that it was standard practice for the Appellant to insure its goods. Furthermore, the fact that the premium for the insurance policy was likely to increase simply reflected the Appellant's increased turnover which, in turn, would have increased its profits. We consider it more likely that the explanation for the lack of insurance in respect of all five appealed deals was not a question of cost but that the Appellant knew that there was no risk because the deals were not genuine arm's-length transactions.

297. In addition, as already indicated, we did not find Mr Hill's claim that, when HMRC officers were incorrectly told at the meeting in September 2006 that the goods were insured, he had not been present to be credible. In our view, although the subsequent provision of documents by Mr Price to HMRC alerted HMRC to the fact that the goods have been transported without insurance, this was an attempt to mislead HMRC.

298. We attach considerable significance to Mr Hill's failure to insure the goods in the deals under appeal. It seems to us a clear indication that he knew that these were not genuine commercial deals as regards which he was at risk.

Passage of title and contractual terms

299. We already discussed the fact that the Appellant appeared to trade significant volumes of high-value goods in the five transactions under appeal without having any standard conditions of sale. The Appellant accepted that there was no written contract in place with Stardex or CEMSA. Consequently, if any dispute had arisen the Appellant's legal rights and remedies would have been unclear. We consider that this is not consistent with genuine arm's-length trading.

300. We have seen, in relation to the Britwap transaction, that if the Appellant's overseas customer defaulted on the transaction the Appellant would be exposed to the cost of transporting the goods back to the UK or to another destination. In the Britwap transaction the Appellant ensured that Britwap bore these costs, but it would surely be standard practice to enshrine such an obligation in a written agreement or standard conditions of sale. Moreover, in the Britwap transaction the Appellant suffered a loss on resale of the mobile phones of approximately £11,000. Obtaining a deposit from a customer in respect of goods shipped abroad would have offered the Appellant some degree of protection against both the risk of additional transportation costs and resale losses. It is true that the Appellant sometimes required deposits to be paid, but did not do so in any of the transactions under appeal.

301. Furthermore, as we have seen, Stardex included a reservation of title clause on its invoices indicating that title to the goods did not pass until it had received full payment. Notwithstanding that provision, the Appellant agreed to sell the goods to an overseas customer prior to any payment being made to Stardex. There was no evidence to support the Appellant's claim that there was a verbal agreement with Stardex which would enable the Appellant to export without payment and, in any event, such an agreement would seem inconsistent with the picture that Mr Hill painted of Maria Prouost as a hard-nosed businesswoman. We did not find Mr Hill's evidence on this point credible.

302. In our view, the absence of contractual terms and the illogicality of exporting goods which the Appellant did not own strongly indicated to us that it was more likely than not that the five appealed deals were not genuine commercial transactions but were contrived transactions which the Appellant knew were connected with the fraudulent evasion of VAT.

Manner of trading and mark-ups

303. We have seen that the Appellant repeatedly made a mark-up of approximately 6% in each of the five transactions under appeal. Although he knew Stardex was an exporter as well as a UK to UK trader it never seemed to strike Mr Hill as odd that Stardex was selling to him rather than making the larger profits that accrued to an exporter. Moreover, Mr Hill was aware that there were at least five wholesale traders in the chain. Nonetheless, Mr Hill did not find it curious that these mobile phones (which could not be used in the UK without the plugs being changed or an adapter being used) passed from hand to hand, at each stage presumably having a mark-up applied to them, before leaving the UK and passing through the hands of at least two other traders.

304. It seems to us that the basis of the five transactions under appeal lacked commercial logic. We think it is unlikely that Stardex, a company in essentially the same business sector as the Appellant, would have given up the chance of making the more substantial profits to be made on exports rather than repeatedly contenting itself with the role and the profits of a buffer trader. Moreover, there is something unreal about an alleged market where goods (which in their current form cannot be used in the UK) enter the UK pass through various hands at ever increasing prices (but no payment being made until after the end of the process) before being exported again to another wholesaler.

305. The consistency of the mark-ups made by the Appellant also suggests to us that the deals under appeal did not constitute genuine arm's-length trading but rather contrived transactions. The Appellant makes a consistent mark-up of almost exactly 6% in every deal regardless of the quantity, the per unit value of the mobile phones or the type of mobile phone. The Appellant painted a picture of the mobile phone wholesaling market as one in which prices fluctuated considerably (e.g. the Britwap transaction). It is, therefore, hard to see how such consistent profits could be made unless there was contrivance.

Due diligence

306. As already noted above, Moses LJ in Mobilx, supra at paragraph [82] said:

"Tribunals should not unduly focus on the question whether a trader has acted with due diligence. Even if a trader has asked appropriate questions, he is not entitled to ignore the circumstances in which his transactions take place if the only reasonable explanation for them is that his transactions have been or will be connected to fraud. The danger in focussing on the question of due diligence is that it may deflect a Tribunal from asking the essential question posed in Kittel, namely, whether the trader should have known that by his purchase he was taking part in a transaction connected with fraudulent evasion of VAT. The circumstances may well establish that he was."

307. It is clear from the context that Moses LJ was considering due diligence in the context of the claim that a trader should have known that its deals were connected with fraud.

308. In this case, we have to consider whether the due diligence carried out by the Appellant had any bearing on the initial question whether the Appellant knew that its transactions was so connected.

309. It will be apparent that what we have referred to as "standard" due diligence such as checking VAT registration numbers, company details, references, directors’ ID etc. - all of which the Appellant carried out - is unlikely in most cases to be of much assistance to a trader in determining whether fraud has occurred (either the contra- trader's fraudulent concealment or the missing or defaulting trader in the "dirty" chain) higher up the chain i.e. beyond its immediate supplier. We quote the words of this Tribunal in Mayfair Executive Ltd v HMRC [2011] UKFTT 148 (TC) (Judge Nowlan and Ms Bridge):

"Even without the guidance of the Court of Appeal, to the effect that we as a Tribunal should look far more to the overall circumstances, rather than just dwell on the due diligence, it was fairly obvious that much of the due diligence would inevitably fail to reveal the chain to fraud. The reality of the planning behind these transactions that is now very evident, to which we referred … above, meant that the broker’s immediate trading partners would virtually always be duly incorporated, duly registered, and accounting for VAT on their slim margins. And it was equally obvious that if and when due diligence questions were put to the fraudster, the fraudster would obviously lie.

…[T]he Appellant was aware that due diligence was in two respects a fairly hopeless way of ascertaining whether transactions were connected to fraud. Firstly, the great likelihood was that the organisers would ensure that the immediate parties, either side of the exporter, would pass all the standard-form tests. They would, in other words, be duly incorporated companies, with valid VAT registrations and inevitably they would pass the Redhill and Europa checks. Secondly, at the point in the chains where those requesting answers in due diligence questionnaires of the fraudster, or those aware that they were participating in fraudulent chains, it was always slightly unlikely that the answers to the questions would reveal this. "

310. In this case, the Appellant appears to have obtained some Dun & Bradstreet reports from March 2006 onwards although we considered that there was insufficient evidence to establish as regards which companies reports are being obtained. Furthermore the Appellant retained Halliwells to report on CEMSA and also received a letter in respect of Stardex from Halliwells, although we do not accept that this letter was written to the Appellant in its capacity as a client of Halliwells. What we have to consider is whether the Appellant was honest in engaging in due diligence or was simply using due diligence as a smokescreen or a shield in case HMRC investigated its transactions. In our view, the latter is more likely to be the case for the following reasons.

311. The crucial due diligence reports related to CEMSA and Stardex - the Appellant's immediate trading partners (although we recognise that the Appellant also commissioned Halliwells to produce reports on other trading partners). The Appellant had traded with CEMSA since November 2005. Beyond the "standard" due diligence there is nothing to indicate that the Appellant thought it worthwhile to carry out due diligence until immediately before the five deals currently under appeal. Certainly, Mr Hill did not visit CEMSA. Furthermore, it is clear from the report that much of the information contained in it simply reported what Mr Russell had said, with very little third-party verification. There was no explanation why CEMSA, a company which did not have a working website, was able hugely to increase its turnover nor did Halliwells think to question this.

312. Likewise, the report on Stardex - in fact a short letter - was only received in May 2006. We have already indicated that whilst Mr Hill could probably have felt confident in dealing with Stardex in the period immediately after he left their employment, that confidence could no longer reasonably exist after, in our view, 2004. We would have expected a more comprehensive report to have been obtained. In addition, it was perfectly plain that the Halliwells letter in relation to Stardex was not independent. It appeared, from the footer, to be a standard form letter which had been sent to other recipients.

313. Why was it that the Appellant obtained the Stardex letter and the CEMSA report in May and June 2006? Mr Hill wanted us to believe that this was simply a marketing initiative by Halliwells, although it was fairly clear that Halliwells came to Mr Hill via Maria Prouost. We did not believe Mr Hill's evidence on this point. The flurry of due diligence in May and June 2006 immediately before the Appellant entered into five substantial transactions which were connected with fraud seems to us, in the context of all the evidence, too much of a coincidence.

314. Stardex, the company which it is now accepted was involved in the fraud, had commissioned a Halliwells report on the Appellant in February 2006. In context, it is hard to see how this report was intended by Stardex to be anything other than window-dressing to protect it against HMRC enquiries. Although this was not a report commissioned by the Appellant, it does serve to demonstrate that a Halliwells report was clearly used by Stardex other than for bona fide purposes.

315. Mr Ahmed submitted that the Appellant had paid Halliwells £25,000 for their various reports and that, therefore, the reports were unlikely simply to be "window-dressing." That, however, has to be seen against the background of the very significant profits which the Appellant made (or would have made if its VAT repayment claim had been met) from the five transactions under appeal.

316. Furthermore, it was clear from his evidence that Mr Hill seemed mainly interested in Halliwells' conclusion i.e. that CEMSA or Stardex was "low risk" rather than in the means by which Halliwells arrived at that conclusion. In the case of CEMSA, Mr Hill was plainly influenced by the fact that Mr Russell was "loaded" - this seemed to be one of the main factors which prompted Mr Hill to regard Mr Russell as a suitable trading partner.

317. In addition, throughout his evidence Mr Hill placed considerable emphasis on "knowing" his trading contacts, rather than any investigative due diligence. Even if his own evidence as to the innocence of his relationship with Maria Prouost and Stardex is to be believed (and, as discussed, we do not accept Mr Hill's evidence on this point) the shortcomings of this approach are now manifest: he believed he had an excellent relationship with Maria Prouost yet he now accepts that she was the fraudster.

318. For these reasons, we consider that issue surrounding the Appellant's due diligence make it more likely than not that the Appellant knew or should have known that its transactions were connected with fraud.

The Freight-Forwarder: 1st Freight

319. The Appellant accepted that, on the basis of the evidence produced by HMRC, 1^st Freight was a participant in the fraud.

320. As already discussed, Maria Prouost had recommended 1^st Freight to the Appellant and had plainly alarmed Mr Hill by referring to "stock swapping" in relation to other freight-forwarders. Nonetheless, Mr Hill carried out no substantive due diligence in relation to 1^st Freight, the company which would handle over £8 million worth of goods traded by the Appellant. Mr Hill had not dealt with 1^st Freight before, yet he did not visit their premises to satisfy himself, for example, in relation to security. He did not enter into any written agreement with1^st Freight.

321. The one check that Mr Hill did carry out was to verify 1^st Freight's VAT registration number. HMRC replied that the number Mr Hill has supplied did not match 1^st Freight's number. Mr Hill said that he had supplied further documentation and that the VAT registration number had come back verified. There was, however, no documentary evidence to this effect.

322. We considered that Mr Hill's dealings with1^st Freight were, to put it at its lowest, commercially imprudent and suggested that he knew that there was no risk involved in the handling of the goods because the deals were contrived.

Credibility of Mr Hill's evidence

323. For the reasons given earlier in this decision, we have concluded that Mr Hill's evidence was unreliable.

Decision

324. We have based our decision on the totality of the evidence. For the above reasons, we have concluded that the Appellant knew that its transactions were connected with the fraudulent evasion of VAT. Any one reason in isolation might have been insufficient to justify the conclusion that the Appellant must have had actual knowledge, but the points described above cumulatively, in our view, justify that conclusion.

325. Accordingly, this appeal is dismissed.

Costs

Rights of Appeal

327. This document contains full findings of fact and reasons for the decision. Any party dissatisfied with this decision has a right to apply for permission to appeal against it pursuant to Rule 39 of the Tribunal Procedure (First-tier Tribunal) (Tax Chamber) Rules 2009. The application must be received by this Tribunal not later than 56 days after this decision is sent to that party. The parties are referred to “Guidance to accompany a Decision from the First-tier Tribunal (Tax Chamber)” which accompanies and forms part of this decision notice.

GUY BRANNAN

TRIBUNAL JUDGE

RELEASE DATE:

APPENDIX 1

SUMMARY* OF APPELLANT’S VAT RETURNS 2002-2006 - all figures expressed in £

*The Appellant made no EC acquisitions in any period

Period	12/02	01/03	02/03	03/03
Output tax	4936.11	26,979.04	15,162.29	2851.04
Input tax	37,273.70	99,844.06	210,598.27	194,696.16
Net tax assessed	32,339.59 CR (repayment)	72,865.02 CR (repayment)	195,435.98 CR (repayment)	191,845.12 CR (repayment)
Outputs	263,767	603,166	1,302,941	1,203,792
Inputs	225,352	571,827	1,207,106	1,103,533
EC supplies	235,561	449,000	1,256,300	1,187,500

Period	04/03	05/03	06/03	07/03
Output tax	2139.99	729.17	0.00	2406.25
Input tax	53,983.46	81.99	0.00	187.15
Net tax assessed	51,843.47 CR (repayment)	647.18	0.00	2219.10
Outputs	298,228	6182	0.00	13,750
Inputs	308,647	468	0.00	1325
EC supplies	286,000	2015	0.00	0.00

Period	08/03	09/03	10/03	11/03
Output tax	3160.94	1050	175	175
Input tax	160.73	0.00	281.22	40.88
Net tax assessed	3000.21	1050	106.22 CR (repayment)	134.11
Outputs	18,063	6000	1000	1000
Inputs	918	0.00	1607	233
EC supplies	0.00	0.00	0.00	0.00

Period	12/03	01/04	02/04	03/04
Output tax	1925	480.90	1925	146,804.88
Input tax	177.03	0.00	177.03	201,304.37
Net tax assessed	1747.97	480.90	1747.97	54,499.49 CR (repayment)
Outputs	11,000	2500	11,000	1,174,225
Inputs	1152	248	1152	1,150,311
EC supplies	0.00	0.00	0.00	335,340

Period	04/04	05/04	06/04	07/04
Output tax	175	60,462.50	0.00	0.00
Input tax	20.73	141,843.56	0.00	0.00
Net tax assessed	154.27	81,381.06 CR (repayment)	0.00	0.00
Outputs	1000	838,176	0.00	0.00
Inputs	121	810,535	0.00	0.00
EC supplies	0.00	492,676	0.00	0.00

Period	08/04	09/04	10/04	11/04
Output tax	0.00	0.00	0.00	0.00
Input tax	144,338.64	0.00	51,086.15	75,672.42
Net tax assessed	144,338.64 CR (repayment)	0.00	51,086.15 CR (repayment)	75,672.42 CR (repayment)
Outputs	878,750	0.00	315,000	462,000
Inputs	825,584	0.00	291,921	432,634
EC supplies	878,750	0.00	315,000	462,000

Period	12/04	01/05	02/05	03/05
Output tax	77,125	0.00	80,150	61,670
Input tax	171,013.12	86,436.51	221,126.59	173,787.48
Net tax assessed	93,838.12 CR (repayment)	86,436.51 CR (repayment)	140,976.59 CR (repayment)	112,117.48 CR (repayment)
Outputs	1,026,500	515,347	1,321,758	1,035,144
Inputs	977,218	497,878	1,266,959	996,157
EC supplies	585,500	0.00	622,000	0.00

Period	04/05	05/05	06/05	07/05
Output tax	36,575	62,037.50	44,493.75	50,750
Input tax	213,132.52	319,318.62	245,840.77	491,558.94
Net tax assessed	176,557.53 CR (repayment)	257,281.12 CR (repayment)	201,347.02 CR (repayment)	440,808.94 CR (repayment)
Outputs	1,279,125	1,470,241	1,922,150	3,278,432
Inputs	1,226,380	1,827,363	1,418,407	3,128,590
EC supplies	0.00	0.00	0.00	0.00

Period	08/05	09/05	10/05	11/05
Output tax	50,750	51,100	89,775	0.00
Input tax	543,452.81	540,625.93	588,728.05	646,178.10
Net tax assessed	492,702.18 CR (repayment)	489,525.93 CR (repayment)	498,953.05 CR (repayment)	646,178.10 CR (repayment)
Outputs	3,278,432	3,260,425	3,541,702	3,916,600
Inputs	3,128,590	3,120,103	3,409,867	3,758,249
EC supplies	0.00	0.00	0.00	2,259,000

Period	12/05	01/06	02/06	03/06
Output tax	0.00	0.00	0.00	0.00
Input tax	0.00	577,025.38	0.00	3518.35
Net tax assessed	0.00	577,025.38 CR (repayment)	0.00	3518.35 CR (repayment)
Outputs	0.00	3,487,226	0.00	0.00
Inputs	0.00	3,302,096	0.00	26,207
EC supplies	0.00	0.00	0.00	0.00

Period	04/06	05/06	06/06	07/06
Output tax	1037.75	0.00	0.00	0.00
Input tax	365,416.23	599,715.29	900,952.45	529,842.19
Net tax assessed	364,378.48 CR (repayment)	599,715.29 CR (repayment)	900,952.45 CR (repayment)	529,842.19 CR (repayment)
Outputs	2,212,870	3,605,000	5,444,000	3,104,300
Inputs	2,088,098	3,426,967	5,148,579	3,027,893
EC supplies	2,207,000	3,605,000	5,444,000	3,105,300

APPENDIX 2

Contra-Traders Jag-Tec and A-Z and the Evidence of Mr Humphries

Jag-Tec- contra -trader

(1) Jag-Tec's sales for the period 06/06 totalled £110,228,000. However, the amount of VAT to be paid (£19,232,664.81) was largely offset by the figure for input tax being reclaimed (£19,228,644.89), leaving a small payment to be made to HMRC of £4,019.98.

(2) Jag-Tec completed 58 purchases (another two deals were cancelled) in 06/06. 37 of these 58 deals involved Jag-Tec acting as the broker i.e. purchasing goods from a UK supplier at the standard rate of VAT and then exporting goods to customers in other EU member states (i.e. "broker deals"). All 37 of these broker deals traced back to a VAT loss arising either from defaulting traders or a hijacked VAT registration.

(3) The remaining 21 deals involved Jag-Tec purchasing goods from traders in other EU member states and selling them at the standard rate of VAT to UK traders (i.e. "acquisition deals"). The Appellant's Deal 1 in 06/06 featured in one of these acquisition deal chains.

(4) Stardex often features as the buffer company and in Jag-Tec's broker deals the goods are often exported to CEMSA in Spain, as well as to other EU-based purchasers such as Evolution SARL, in France

(5) The result of the 58 deals in 06/06 was that Jag-Tec's repayment claim in respect of its broker deals was offset (and thereby disguised) by the output tax in respect of its acquisition deals. The outcome was the small amount of VAT which was paid by Jag-Tec to HMRC as noted above.

(6) In the period 09/06, Jag-Tec completed 32 purchases. In 16 of these deals Jag-Tec acted as the broker, as explained above. All 16 of these broker deals were traced back to a fraudulent VAT loss arising from defaulting traders.

(7) The remaining 16 (of the total 32 deals) in 09/06 involved Jag-Tec entering into acquisition deals, as explained above, which included the Appellant's Deal 2 in period 07/06.

(8) Again, the result of the 16 broker deals and the 16 acquisition deals was that Jag-Tec made a small repayment claim from HMRC of £344.59. Thus, the large repayment claim that would have been made by Jag-Tec was masked and, in HMRC's submission, effectively "transferred to" the brokers, such as the Appellant, in Jag-Tec's acquisition chains.

A-Z- contra-trader

(9) As already explained, A-Z sold mobile phones to Stardex, which then sold those goods to the Appellant in Deal 2 in period 06/06 and in Deals 1 and 3 in period 07/06.

(10) A-Z's sales for the period 08/06 (which covered June, July and August 2006) totalled £110,738,335. The amount of output tax which would otherwise have been payable (£11,119,595.16) was almost entirely offset by the input tax being reclaimed (£11,095,436.10), with the result that A-Z made only a small payment of VAT to HMRC of £24,159.06.

(11) A-Z undertook 122 deals in 08/06. 52 of these deals involved A-Z acting as the broker. 51 of the 52 broker deals were traced back to a VAT loss arising from defaulting traders or hijacked VAT registrations.

(12) Of those 122 deals, 14 were buffer deals which were also traced back to fraudulent tax losses.

(13) The remaining 56 deals in the period 08/06 were acquisition deals. These included three of the deals currently under appeal referred to above.

Evidence of Nigel Humphries

(14) Mr Humphries, a senior HMRC officer, gave evidence concerning the similarities between the transaction chains of Jag-Tec and A-Z and those of four other contra-traders in the period March - June 2006:

(15) Kingswood Trading Limited ("Kingswood");

(16) Opportunities Recruitment International Limited ("Opportunities");

(17) Red House International Limited ("Red House"), and

(18) Starmill International Limited ("Starmill").

(19) Mr Humphries’ evidence was not challenged.

(20) In particular, Mr Humphries noted that the transactions passing through the six contra-traders had been structured in a very similar manner and were all connected to fraudulent tax losses. His evidence considered the acquisition deals of Jag-Tec and A-Z, which included purchases from KOM Team and sales through buffers, including Stardex, which then sold to brokers (the role played by the Appellant in these appeals), which then sold the mobile phones to EU traders, (such as CEMSA in these appeals.

(21) KOM Team took part in deals involving all six contra-traders. CEMSA featured as the EU purchaser in relation to five of the six contra-traders and Stardex featured in relation to all six contra- traders. It should be noted that Stardex usually played the role of a buffer trader which bought directly from the contra-trader (a “front-line buffer”), although in some cases it bought from a front-line buffer and on-sold to a broker. In two instances, where the contra-trader was Red House, Stardex played the role of the broker selling to Navigo in Italy and to Mobile Express in the Isle of Man.

(22) In April 2006 the Appellant was a broker in relation to a deal involving the contra-trader Starmill - a transaction not under appeal. In this transaction Stardex had bought from Starmill, the Appellant bought from Stardex and then sold to CEMSA. The Appellant had previously dealt directly with Starmill in 2003 i.e. shortly after he had left Stardex. Mr Hill had also dealt with the Starmill before he had joined Stardex.

(23) From March to June 2006 the six contra-traders sold to ten buffers traders who sold to 42 brokers. The 42 brokers had six EU customers, with CEMSA featuring repeatedly.

(24) Mr Humphries noted that in these deal chains much of the profit accrued to the brokers and that this profit could have been enjoyed by the contra-traders had they sold directly to EU customers, bypassing the brokers.

(25) The EU trader in one chain was a company called Evolution SAAL (which failed to produce documentation to the French authorities, had no business establishment in France and was deregistered in 2007). Evolution SAAL, like CEMSA, made sales to Vundera, a Latvian company (which had no business establishment in Latvia) suggesting that its bona fides were in doubt.

(26) In July and August 2006 only two of the six contra-traders (A-Z and Jag-Tec) were importing from EU suppliers. The pattern of transactions in July and August was similar to that in earlier months, with most goods being sourced by the contra-traders from Kom Team and being sold by the UK brokers (one of which was the Appellant) to CEMSA and Evolution (although the Appellant did not sell to Evolution in this period).

(27) In August 2006, Jag-Tec acquired goods from MS Enterprise, on sold to Stardex which either sold them directly to brokers (or to intermediate traders which then on sold to brokers).

(28) In August 2006, A-Z acquired goods from a French supplier, City Trading, in three transactions. The goods in these transactions were sold by two brokers, the Appellant and Vortech, CEMSA. City Trading was, according to the French authorities, a missing trader which had never rendered any VAT returns.

APPENDIX 3

Mark-ups

Deal	Trader	Units	Net sale price	Mark-up
06/06 1	Kom Team	11,000	£277.00
	Jag-Tec	11,000	£278.00	0.36%
	Stardex	11,000	£280.00	0.72%
	Appellant	11,000	£297 .00	6.07%
	CEMSA	11,000	£297.60	0.2%
06/06 2	Kom Team	7,000	£290.00
	A-Z	7,000	£291.50	0.52%
	Stardex	7,000	£293.50	0.69%
	Appellant	7,000	£311.00	5.96%
	CEMSA	7,000	£311.62	0.2%
07/06 1	Kom Team	2,000	£288.00
	A-Z	2,000	£289.00	0.35%
	Stardex	2,000	£294.00	1.73%
	Appellant	2,000	£311.65	6.00%
	CEMSA	2,000	£312.25	0.19%
07/06 2	Kom Team	3,000	£277.00
	Jag-Tec	3,000	£278.00	0.36%
	Stardex	3,000	£280.00	0.72%
	Appellant	3,000	£297.00	6.07%
	CEMSA	3,000	£297.60	0.20%
07/06 3	Kom Team	5,000	£289.00
	A-Z	5,000	£290.00	0.35%
	Stardex	5,000	£300.00	3.49%
	Appellant	5,000	£318.00	6.0%
	CEMSA	5,000	£318.65	0.20 %

First-tier Tribunal (Tax)

Introduction

The evidence

MTIC trading - legal principles

Terminology

Issues in dispute

The Appellant - background

The deals under appeal

06/06 Deal 1

06/06 Deal 2

07/06 Deal 1

07/06 Deal 2

07/06 Deal 3

Input tax claimed

The contra-traders: Jag-Tec and A-Z

Stardex - the Appellant’s supplier

CEMSA - EU customer

Appellant's knowledge of MTIC fraud

FCIB and Banking Evidence

Payments through FCIB accounts

FCIB: account opening

Did the Appellant pay Stardex's invoices in full?

Freight Forwarder: 1st Freight

Inspection of the goods

1st Freight

A1 Inspection

Insurance

Passing title to the goods

Manner of trading and mark-ups

Due Diligence

Introduction

Halliwells letter in relation to Stardex

Halliwells’ report in relation to CEMSA

Halliwells report in relation to the Appellant

Due diligence - general

Reliability of Mr Hill's evidence

Discussion

General

The Appellant's knowledge of the high level of MTIC fraud

Relationship with Stardex and Maria Prouost

Relationship with CEMSA

FCIB evidence

Insurance

Passage of title and contractual terms

Manner of trading and mark-ups

Due diligence

The Freight-Forwarder: 1st Freight

Credibility of Mr Hill's evidence

Decision

Costs

Rights of Appeal

TRIBUNAL JUDGE

RELEASE DATE:

Contra-Traders Jag-Tec and A-Z and the Evidence of Mr Humphries

Jag-Tec- contra -trader

A-Z- contra-trader

Evidence of Nigel Humphries

Freight Forwarder: 1^st Freight