The parties

1          The Appellants are Mobile Export 365 Ltd (“ME365”) and Shelford IT Limited (“SITL”) . They were at all relevant times two subsidiaries of Axxia Holdings Ltd (“Axxia”). Axxia and its five subsidiary companies are referred to in this decision as the Axxia Group. The Appellants, together with two other subsidiaries of Axxia, were engaged in trading activity (or activity including trading activity) that the Respondents, Her majesty’s Revenue and Customs (“HMRC”, which abbreviation includes the predecessors in title to HMRC) regard as involving or including activity linked to activities that are usually labelled carousel or MTIC (missing trader intra-Community) fraud. The central contention of HMRC is that the transactions for which the Appellants claim refunds of input value added tax (“VAT”) in these appeals are transactions connected to intra-Community fraud involving other traders.

2          One of the other subsidiary companies of Axxia, Live Telecoms Ltd, was originally also an appellant in these appeals. It went into administration after the initial appeals. As a result, separate legal representatives were appointed and its appeal has since been considered separately from these appeals.

3          Another subsidiary company of Axxia, Shelford Trading Company Ltd (“STC”), is the company that sold all the relevant goods to ME365 and SITL. It is therefore involved directly in every transaction of concern to the tribunal in these appeals.

4          The Appellants are both wholly owned by Axxia. The companies were not registered as a VAT group, but are all registered separately, so the Appellants are before the tribunal as separate appellants.

The decisions under appeal

5          All the decisions under appeal to the tribunal are refusals by HMRC to repay input tax claimed by the Appellants in respect of VAT on goods sold from the United Kingdom to customers in other countries. There are four groups of appeals before the tribunal, separately in respect of the monthly VAT periods 03/06 and 04/06 and separately in respect of ME365 and SITL. Within some of the groups, the denials of VAT to the company were made by single decisions for HMRC, and in others by more than one decision.

6          The sums claimed by the Appellants (in round figures) as payable input tax are in total:

            ME365                        March 2006                 £5,877,927

                                                April   2006                £9,149,525

            SITL                            March 2006                 £1,219,210

                                                April  2006                 £3,524,325.

7          In respect of each month, a letter was issued for HMRC to each of the Appellants denying it the input tax reclaimed on the grounds that it knew or had the means of knowing that the transactions were connected to fraud. (Similar decisions were issued to Live Telecoms Ltd for the same periods.) The basis for these decisions was a tracing operation conducted by HMRC into the supply chains for each of the deals or transactions for which input tax was reclaimed. HMRC maintained that these showed that all the transactions were in chains of one of two kinds, and that there was fraud by others linked to every one of them. The first kind were the “straight” chains of transactions: those where a supplier “up” the chain (that is, in the series of supplier to supplier the last of whom supplied STC) was a defaulting trader or the chain otherwise directly involved fraud. The other were “contra” chains: those where the “straight” chain showed no fraud in the chain itself (a “clean chain”) but where a supplier “up” the chain from STC was engaged in fraud related to other supplies that took place in the same VAT period as the supplies in the chain leading to STC. The result was that both the “clean chains” and the “dirty chains” were reported to HMRC in the same quarterly VAT returns by the “contra trader”. Typically, the reporting company would report output tax from sales in the “clean chains” of the same amounts as the input tax claimed from the “dirty chains”. The VAT return would record either a small net amount of output tax payable or a small net amount of input tax claimable, so concealing the fraudulent nature of the claims for the “dirty chains” behind the VAT returned for transactions in the “clean chains”.

Annex A

8          Rather than repeat the details of the specific decisions and amounts of VAT, the tribunal has summarised in Annex A all the relevant transactions which it was asked to examine. In setting the details out in this way, the tribunal has adopted from HMRC’s schedules of evidence both the date attributed to particular deals, and the deal numbers assigned to those deals. These were the deal numbers used during the hearings of the appeals and it is convenient to maintain them. The tribunal records that the attribution of separate deal numbers to each individual movement of each individual kind of mobile phone, and the specific dates given, may have served to obscure as well as to expose what was happening. This is so where, as may be the case with regard to some of the deals undertaken in these appeals, the actual bargain struck between the trader and a supplier or customer is a deal involving multiple sales at the same time. However, to avoid confusion, the tribunal uses the deal numbers from the deal schedules on which both parties presented their cases to the tribunal.

9          Annex A also identifies whether the deal was said by HMRC to be a straight deal chain involving fraud in that chain or whether it was a contra chain involving fraud by another trader involved in that chain but actually occurring in linked chains.

The Court of Appeal decision in Mobilx

10        The initial arguments of the parties were, for the reasons set out below, put to the tribunal many months before the final submissions, backed by full legal argument from the Appellants about the basis on which HMRC sought to make its contentions. Similar challenges had been put to the courts and VAT Tribunal in other appeals. At the time the final submissions were presented to this tribunal,  not only had a number of decisions been taken by the Chancery Division of the High Court about issues of direct relevance to the tribunal, but also all concerned were aware that several of those cases were under active appeal to the Court of Appeal. It was agreed by the tribunal with the parties that further hearing of the appeals would not be held over again until the decisions of the Court of Appeal were published, but that the hearing of these appeals would be completed.

11        Having given initial consideration to the decisions to be taken after the conclusion of the final days of hearing, and of the nature of the legal issues at large pending the decisions of the Court of Appeal, this tribunal took the view after the close of the hearings that it should if necessary await the decision of the Court of Appeal before finalising its own decisions.

13        Following procedure agreed by the parties, HMRC put its case first. Its case was that tax loss caused by fraud was shown to have taken place in chains of transactions leading to all the transactions relevant to the appeals. These are as listed in Annex A. Specifically:

(a)   23 of the 28 transactions by ME365 in March 2006 were in chains where the acquiring trader (that is, the trader acquiring the goods into the United Kingdom) had failed to account for the VAT payable. Those traders were:

Bargain Trading Ltd

CHP Distribution Ltd

F Options Ltd

Oracle (UK) Ltd

Roble Communications Ltd

Stella Communications UK

In the remaining 5, the acquiring trader was a contra trader (that is, a trader also engaged in contra deals involving fraud). That trader was in all cases @tomic Ltd.

(b)  The seven deals by SITL in the same month all were in chains involving a fraudulent acquiring trader. The traders, all traders also involved in deals with ME365 were: Bargain Trading Ltd, Oracle (UK) Ltd, and Roble Communications Ltd.

(c)   15 of the 27 deals by ME365 in April 2006 were in chains where the acquiring trader had failed to account for the VAT properly payable. Those traders were:

Bullfinch Systems Ltd

C&B Trading Ltd

The other 12 were chains involving contra traders – in 9 cases that was Unibrand Ltd, and in the other 3 @tomic Ltd.

(d)  The 11 deals in April 2006 undertaken by SITL and challenged by HMRC all involved contra trading. In 8 of the deals the trader concerned was Unibrand, in a further two it was Primeline (Europe) Ltd, and in the other deal @tomic Ltd.

14        HMRC offered documentary evidence, supported by oral evidence from investigating officers and others, to establish their contentions that there were tax losses attributable to fraud in the way summarised in each relevant transaction.

HMRC then contended that the Appellants knew or ought to have known that these transactions were linked to fraud.

15        The Appellants, in their response to the HMRC case and in submissions to the tribunal, challenged the HMRC case for every transaction not only with regard to the conduct of the Appellants in their own dealings but also with regard to all other evidence produced by HMRC. As part of this, the Appellants challenged the witness statements offered by officers for HMRC, and accordingly the officers were required to attend for cross-examination.

16        At the same time, it was the Appellants’ case that they had no knowledge whatsoever of any of the relevant details of these supply chains beyond their immediate contracting partners, and that they could not reasonably be expected to have such knowledge.  The central contention was that the Appellants had no knowledge of any earlier or later deals in any of the supply chains and had no reason to have such knowledge. They were therefore in no position themselves to offer evidence to assist the tribunal’s consideration of these transactions but confined themselves to challenging the quality of the evidence produced by HMRC.

17        As part of this, the Appellants put forward no evidence countering the evidence for HMRC that the above defaulting traders were shown to be defaulting traders. They tested, as they were entitled to test, HMRC’s evidence but they did not mount any serious evidence-based challenge to that aspect of the evidence for HMRC.  

18        The Appellants put forward evidence that they had, throughout, both followed HMRC guidance and conducted thorough due diligence on their suppliers and customers.

The patterns of evidence

19        The tribunal found, having considered the positions of the parties, that there is no sufficient common identity of fact across the appeals that enabled the parties, or allows the tribunal, to deal with the appeals only by way of examination of sample or typical transactions. At the conclusion of the hearings, and at the invitation of the tribunal to the parties, HMRC identified four specific transactions for detailed examination and presented the tribunal with bundles of documents about those transactions. It invited the parties to suggest sample transactions in this way for illustrative purposes so that it could consider all aspects of the evidence with regard to sample deals as it considered that it would be disproportionate for it to examine in full detail every one of the transactions in evidence in the appeals.

20        The tribunal’s examination of these sample transactions is set out in Annex B. Having examined the sample deals, the tribunal then also examined the other evidence for patterns of trade and trading terms common to those transactions and other transactions in issue in the appeals. It notes its findings on those comparisons in Annex B.

21        While the tribunal must take its decision in respect of each transaction, all these appeals were heard together by the tribunal. The tribunal finds as fact that, for the reasons below, the Axxia Group is to be treated for the purposes of these appeals as one business and financial entity despite separate VAT registrations. This is despite the way in which each of the transactions involved two separate companies within the Axxia Group. Accordingly, the tribunal approaches the appeals on the basis that the evidence before it about any aspect of any of the appeals is potentially relevant to any other aspect of the appeals.  

22        The tribunal must consider the evidence with regard to three issues, or groups of issues, not only in respect of each appeal but also in respect of each transaction within each appeal, there being no relevant common ground between the parties. For each transaction (or deal), the tribunal must consider:

            - if the transaction is connected with a tax loss, and if so      

- if it is connected with fraud, and if so

- if the appellant knew or ought to have known of the fraud.

23        The tribunal takes from the decision in Mobilx the guidance that in considering those questions it may look at the circumstances and evidence as a whole. It is entitled to consider each of the transactions in the context of all the other transactions. That approach is reflected in the analysis in Annex B.

The burden of proof

24        At the hearing, it was common ground that it was for HMRC to establish the first two of the issues. There were different views about the extent to which it was for HMRC or the Appellants to establish the position with regard to the third of those issues. It is now clear from Mobilx (at [81]) that the burden of proof falls on HMRC to show that the Appellants’ state of knowledge was such that they were not entitled to the right to deduct input tax in these appeals. The tribunal has applied that approach to all the evidence despite the conflicting approaches put to it in earlier stages of the appeals. It follows that unless HMRC can make good their contentions that in respect of each of the deals for which input tax has been refused there is both a tax loss within the chain of transactions of which that deal is part, that there was a link to fraud somewhere in that chain, and that the Appellants had the requisite knowledge or should have had that knowledge, then the Appellants are entitled to the input tax for that deal. The standard of proof is the civil standard throughout.

Case management

25        As noted above, the tribunal heard extended argument about the precise terms in which the issues should be tested and about the approach that it should take in assessing the evidence about those issues.

26        The tribunal and, on appeal, both Lindsay J and Park J gave several decisions by way of case management directions about evidence for these appeals. The parties had previously been before the Administrative Court with respect to a judicial review of the actions of HMRC affecting the Axxia Group in the year previous to those in question in these appeals with regard to repayments of input tax to the Appellants. The evidence put to that court in those hearings was available to this tribunal. The following paragraphs summarise the decisions taken on case management of direct relevance to this decision.

27        The tribunal refused on 26 June 2007 both to allow these appeals to be dismissed under rule 19(4) of the Value Added Tax Tribunals Rules 1986 and to exclude evidence of the criminal convictions of Mr Andrew Nicolas as irrelevant to these appeals. On appeal to the Chancery Division, Lindsay J endorsed the decisions of the tribunal. His judgment was issued as Mobile Export 365 Ltd and another v Revenue and Customs Commissioners, [207] EWHC 1737 (Ch). In his judgment, which the tribunal adopts for the purposes of this decision and does not repeat, Lindsay J outlined the conduct of the appeals to that date, including decisions of the tribunal about the admission of evidence. He confirmed the refusal of the tribunal to deal with the matter as “plainly correct” and himself gave case management directions for the conduct of the appeal.

28        The tribunal notes in particular one of those directions as specifically relevant to the further conduct of the case after the directions:

            “[20] (4) Reference has been made to the relevance and admissibility of             convictions in 2000 and 2004 of Mr Nicolas for MTIC fraud. Mr Patchett-   Joyce told me that the Appellants contend that the evidence of the convictions   is irrelevant and accordingly should not be admitted because Mr Nicolas had no part in any of the relevant transactions. Nonetheless he told me that Mr Nicolas would not be giving evidence at the trial. I need only say that the    convictions appear to me to be of substantial potential significance. Much may           turn on the role of Mr Nicolas in the company and in relation to the       transactions in question. His failure to give evidence on these and other issues             may of itself be highly significant and invite adverse inferences. Accordingly I             would not exclude at the trial as irrelevant the evidence of the convictions, as          Mr Patchett-Joyce argues.”

Judgment was given by Lightman J on 20 July 2007. Following this, Mr Andrew Nicolas did give evidence at the main hearing and the evidence of his convictions was put to him by Mr Benson QC in cross-examination. The tribunal had before it records of the proceedings relevant to those convictions and the associated enforcement proceedings.

29        The next major case management decision of the tribunal was made on 1 April 2008 when, among other issues, the tribunal accepted late evidence from Mr Gary Taylor given for HMRC, refusing an application for the Appellants asking the tribunal not to admit the evidence, and declining to categorise the evidence as expert. That direction was made without a hearing. On receipt of the direction the Respondents asked for a hearing. This was held on 1 May 2008. As a result of the hearing, the tribunal revoked one part of its direction on 7 April 2008, namely the admission of evidence about an issue of French law that the Appellants contended was relevant to the questions of law in these appeals. It declined to receive this evidence.

30        The tribunal then proceeded to the main hearing of the appeal, starting on 8 May 2008. By agreement with both parties, the tribunal adjourned after hearing the evidence to an agreed later date to hear final submissions. During that hearing the Appellants, without further representation at the hearing, issued a notice of appeal against the first of the above case management directions. During the period before the final submissions a second notice of appeal was issued against the other direction. As a result, the final submissions in the appeals were stood over pending resolution of those case management appeals.

33        The tribunal was provided with considerable amounts of documentation about the appellants and the Axxia Group, about the nature of the market (or markets) in mobile telephones and about the specific deals in question.

34        HMRC presented its evidence first. The tribunal heard evidence under oath or affirmation from the following witnesses. All had provided witness statements accompanied by relevant exhibits. For convenience, the tribunal summarises in each case the evidence given to it by the witness and, save for the central witnesses to the hearing, indicates its findings about the relevance and quality of that evidence, and the extent to which the tribunal considers that the evidence assists its fact-finding.

Witnesses for HMRC

Joseph Baines

35        Mr Baines, an officer of HMRC, gave evidence about the activities of @tomic Ltd and its trading relations in April 2006. His evidence was that there were major tax losses involved in the various chains and that @tomic Ltd was fully aware of the nature of the fraudulent trade in mobile phones and that there were actual or probable defaults in all the transaction chains leading to @tomic Ltd’s exports in April 2006.

Mr Baines accepted under cross-examination that he had not pursued enquiries about third parties about whom he was asked and that various documents that might have been expected to be produced by him or other HMRC officers had not been produced. But the tribunal find that these issues did not raise material concerns about the core of his evidence about @tomic, which the tribunal accepts. This is that he identified actual or probable tax losses in all the 22 transactions undertaken by @tomic in the April 2006 VAT period, that the evidence suggested that the deal chains were contrived and they formed part of an overall scheme to defraud the revenue. His evidence is discussed further in the fourth deal examined in Annex B. The tribunal accepts his evidence that @tomic was during the relevant period conducting its business as part of an overall scheme for defrauding the revenue. That evidence is incorporated into the findings in Annex A. 

Andrew Barnett

36        Officer Barnett, as the responsible officer, made decisions on 14 11 2006 refusing to pay both Appellants input tax in respect of 03/06. Officer Barnett was instructed to complete the examination of deals within the 03/06 returns and to examine those within the 04/06 returns. He issued decisions denying the repayment of input tax to the Appellants on 29 03 2007. He put forward in evidence both the general considerations on which the extended verification of the Appellant companies claims had been pursued and the documentary evidence for HMRC to support the specific chains of which it was contended that the appellant companies’ deals formed part. He also gave evidence about the organisation of the Axxia Group and about dealings between HMRC and the Group. 

37        Mr Barnett was subject to extended cross-examination by Mr Patchett-Joyce on the evidence he produced to support his decision letters, and on the factors he took into account in contending that the Appellants knew or should have known of VAT losses caused by frauds in the various chains of transactions in which they were involved. The tribunal records that this established a number of specific errors and weaknesses in the evidence about specific deals in the chains evidenced in Mr Barnett’s general evidence in support of the decisions made by him for HMRC. For example, it was pointed out to Mr Barnett that certain information about previous transactions involving circular sales of goods that were exported to Switzerland in 2005 was only given to the Appellants after the period in question in these appeals.  This he accepted. He also accepted that at the time he took some of the operative decisions HMRC did not have available to it full evidence about some of the chains and that others revealed problems such as invalid invoices. Nonetheless, the tribunal found that he approached the analysis of the deals conscientiously and competently within the instructions and scope of operations given to him and that his evidence to the tribunal was given honestly, openly and consistently. It further finds that he willingly accepted the limitations to his evidence and, where they were shown to exist, gaps in the evidence. It did not find that the challenges to his evidence established that there was any material error in his decisions to refuse input tax, when viewed with all the evidence available to the tribunal at the hearings. It therefore accepted his evidence.  His evidence is incorporated into the findings in Annex A.

Matthew Bycroft

38        Officer Bycroft was an officer in the MTIC fraud team established by HMRC. He was the officer responsible for the assessments on ME365 on 29 03 2007 that related to deals that he decided were linked with alleged contra trading by another company. As part of this, he analysed the deal chains behind the specific transactions for which the Appellant companies claimed input tax and the due diligence undertaken by the companies. He was subject to extended cross-examination.

Again this established some gaps in the evidence and individual aspects of the analysis that were open to question. For example, Annex B includes one issue on which evidence available to Mr Bycroft was not put before the tribunal (see SITL deal 5 of April 2006). However, after consideration of all the available evidence, the tribunal finds in Annex B that it accepts Mr Bycroft’s evidence. It does so both on this point and more generally, as it finds he was conscientious and honest, and his evidence was given without attempts to misrepresent or hide weaknesses in the evidence. His evidence is incorporated into the findings in Annex A.

Guy Craddock

39        Mr Craddock was an officer of Revenue and Customs at the time of his witness statement but not when he gave evidence, having transferred to other public sector work. He gave evidence about Primeline (Europe) Ltd and its trading partners at the time relevant to these appeals. His evidence was of importance in establishing whether there was fraud in the dealings undertaken by that company, but not of more general importance to the tribunal. The tribunal notes that under cross-examination the witness was unable to explain and support all the contentions put forward in his statement. That is of importance to the transactions in April 2006 by SITL that were linked with Primeline.         

Judith Elmer

40        Mrs Elmer was instructed to review the deal chains in connection with ME365 transactions in April 2006. She gave evidence about each of the deals in that period, including both straight chains and the alleged contra trading deals. Her evidence was also subject to cross-examination. Its main value to the tribunal lay in the documents she produced to the tribunal and in the confirmation she gave of the process of checking the decisions made by HMRC.  That evidence is accepted and is incorporated into the findings in Annex A.

Peter Goulding

41        Officer Goulding gave evidence of  visits to the offices of the Appellants between 15 10 2003 and 17 05 2004 during which time he was the officer responsible for ME365. The tribunal found the evidence of officer Goulding to be reliable evidence from the officer in so far as he was involved in dealing with the Appellants, and that his evidence and notebooks fairly recorded the relationships between him as the responsible officer and officers of the Appellants group of companies at the relevant times. The tribunal finds from this evidence that the officer was satisfied at the times for which he was responsible with the relationships with the Appellant companies and that they had provided information to HMRC during that period that gave no cause for further investigation.

Fu Lam

42        Officer Lam gave evidence as the officer responsible for Unibrand at the relevant time, and in particular about the deals involving Unibrand that were contended by HMRC to be contra-deals undertaken by the company relevant to deals with, but not directly involving, the Axxia group of companies. Cross-examination of the witness did not concentrate so much on Unibrand itself as on problems with the chains involving Unibrand. He confirmed that these issues did not directly involve chains of which ME365 of SITL were parties. He also confirmed that his enquiries were focussed on the company for which he was responsible. He gave evidence showing the balance between transactions in the relevant period generating an output tax liability for Unibrand and transactions generating an input tax entitlement and that, having regard to other evidence available to him, Unibrand was engaged in fraud. The tribunal accepts that evidence and it is incorporated into the findings in Annex A.  

Jonathan Payne

43        Officer Payne gave evidence that during 2006 he was charged with examining the VAT situation of a company called F Options Limited. Officer Payne’s evidence was less satisfactory. He failed to produce documentary evidence that, in the view of the tribunal, he could have produced and should have produced to support the contentions he made in his witness statement and in oral evidence. And his oral evidence lacked the clarity and focus displayed by other officers.  However, further investigation showed that the supply to F Options was made in the name of another company, M Allen General Traders, but that the supply had been made using a hijacked VAT number. The tribunal accepts the evidence of the hijacking. It is accepted as replacing the decision detailed at [13](a) above for ME365 deals in March 2006, and that evidence is incorporated into the findings in Annex A. 

Roderick Stone

44        Officer Stone confirmed that he had no direct involvement with either of the Appellants or the disputed input tax claims. His evidence was at a more general level, concerned for instance with HMRC policy with regard to extended verification of input tax claims. And he found it necessary to revise some of his initial evidence because of errors in the figures given. He gave evidence of HMRC policy to prevent repayments to companies making input tax reclaims in cases where the Treasury had not received VAT at an earlier stage in the claim. He gave evidence about HMRC policy to stop multiple recovery (recovery of a VAT loss from a trader from more than one other trader so that the VAT recovered could exceed the VAT lost.). Linked with this, he explained how HMRC attribute a VAT loss incurred by default by one trader to another specific trader in a situation where there was a choice of the other trader to which the loss could be attributed. His evidence was that it was done partly by date and partly by value.

45        The tribunal records that it accepts his evidence as a statement of HMRC policy at that time. It is not concerned to evaluate that policy in these appeals, and does not do so, but limits its consideration to ensuring that the decisions in the appeals before it are determined properly and that in that context it was satisfied that HMRC did not attempt to retain from the Appellants VAT that had already been collected or retained elsewhere. Its findings on those issues are in Annex A.

Gary Taylor

46        As recorded above, Mr Taylor’s evidence was admitted late under protest. Taking this into account, the tribunal refrained from placing any timing or similar limit on the cross-examination of the witness by Mr Patchett-Joyce. Mr Taylor was cross-examined at considerable length. The tribunal indicated that it accepted his evidence not as expert evidence in the formal sense but as evidence about the kinds of transactions into which the appellant companies had entered, the markets within which those transactions took place, the nature of the goods with which they were dealing, and the profits that could be expected to be made from transactions involving those goods in those markets at the times relevant to the appeals.

47        The tribunal does not see a need to set out Mr Taylor’s evidence in detail as the transcript records the full cross-examination. It makes the general findings that, while there were limits to Mr Taylor’s expertise with regard to certain aspects of the trade in the grey market in mobile phones, he gave his evidence conscientiously, giving his sources for specific information. While the tribunal noted Mr Patchett-Joyce’s submissions that some of those sources were not readily or economically available to the public, it did not consider that point detracted from Mr Taylor’s evidence. The Appellants were engaged in a multi-million pound business and could reasonably be expected themselves to spend some part of the proceeds on ensuring that they were well informed. The tribunal records that it was assisted in its understanding of the general market by the exchanges between Mr Patchett-Joyce and Mr Taylor, but that gave rise to no particular findings of fact of relevance.

48        Mr Taylor’s evidence was of direct relevance in two ways. First, he gave evidence about the scale of the market in certain kinds of phone at the time relevant to the appeals. Second, he gave evidence about the state of the markets at that time, and the extent to which there were profits to be made from the market. His evidence that the market prices for phones tended to be lower than the United Kingdom in states to which the Appellants sent their goods – specifically France and Dubai – carried weight as did his evidence that there was evidence of overtrading (that is, selling more than there was to sell) of stock of Nokia 3230s. The large scale sales said to have occurred in those phones during this period can be noted in Annex C.

49        The tribunal found Mr Taylor’s analysis of the ways in which a profit could be made from the grey market in phones at the relevant time to be a useful framework for considering the specific activities and business approach of the Axxia Group. This narrowed the scope of the activities of the Axxia Group to what he termed arbitrage opportunities, as against those of box-breaking, forecast failure and dumping. The tribunal accepted that there was no evidence of dumping or reliance on forecast failure by the Appellants. It comments on the significance of box-breaking below. 

 50       Witness statements were also offered by HMRC from several other officers dealing with specific aspects of transactions or particular companies. In part for reasons of time, they were not called to give oral evidence and their evidence stood unchallenged. It has been incorporated in the findings of fact in Annex A.  

Witnesses for the Appellants

51        For the Appellants, the tribunal heard evidence under oath or affirmation from the following witnesses, again supported by witness statements. As above, save for the main witnesses, the tribunal summarises the evidence given and its views of the quality of that evidence below.

52        Mr Benson QC made an application to the tribunal at the start of evidence for the Appellants on 27 May in chambers for the exclusion of some of the Appellants’ witnesses from the court while others were giving evidence. This was discussed by the tribunal with the parties, and the tribunal indicated that it would give a ruling in open court the following morning. The application was made for HMRC on the ground that it did not wish evidence of any of the key witnesses to be tainted by having heard the evidence of other key witnesses. It was opposed by Mr Patchett-Joyce on the ground that there would be no officer from the Appellants able to give him instructions during any examination. This, the tribunal observes, reflects the absence of the Appellants’ directors from the proceedings.

53        The tribunal did not accept the application, and made no order for any witness to be excluded. The tribunal noted that there was no specific rule or practice of the tribunal or of the courts with direct relevance to the application. The nearest authority was in the comments of Sir John Arnold in giving guidance to magistrates in Tomlinson v Tomlinson [1980] 1 WLR 322. His clear guidance was that the parties themselves should never be excluded from a court. The tribunal took the view that it was open to question whether it had the power to exclude any officer of a company from a hearing involving the company. But it did not rest its ruling on that consideration alone. More generally it considered that the same approach should be taken to both parties as this was not a criminal matter. The tribunal’s task was to weigh all the evidence in a complex situation where there had been considerable interaction between the witnesses on both sides over some years. So it was, as a stating point, appropriate that all witnesses should be able to give informed evidence. The tribunal also indicated to the parties that they could raise with it, if appropriate, both the argument that one witness appeared to be giving evidence “led” by an earlier witness, and the argument that a witness was not present when earlier evidence was given. The tribunal noted that the latter point was made on behalf of witnesses during the ensuring hearings.  

Iain Bramwell

54        Mr Bramwell was a director of a firm of insurance brokers specialising in providing tailored insurance solutions for businesses. He gave evidence of the firm’s involvement in insurance for the mobile phone industry. He confirmed that the firm underwrote insurance for the global transfer of goods involving multiple buyers and sellers. His evidence was that for a claim to be made the legally recognisable insurable interest had to exist only at the date of the loss. It was immaterial for this purpose that the legal title did not pass until payment was made, as long as it did exist when the claim was made. So the insurable interest would normally arise when payment was made, which was when the risk transferred. No evidence was given by him or in these appeals of any claim being made for a loss by the Appellants. The tribunal observes that there was evidence in a number of the deals of delay between the goods being traded and payment being made. On this evidence, the tribunal finds that the Appellants were not insured during that period and would not be insured unless and until the customer paid for the goods notwithstanding that the Appellants had incurred what they considered to be an insured risk before payment occurred. The tribunal finds that this evidence assisted it little in evaluating the commerciality of the deals.

Tony Deakin

55        Mr Deakin gave evidence that he had been a member of staff for a major distributor of mobile phones in Britain. He gave evidence for the Appellants in general terms of how the grey market operates and its commercial purpose. In particular, he gave evidence of price volatility and of circularity of stock. That issue is discussed in Annex C. He also gave specific evidence about the market in the United Kingdom for 2 pin phones. He offered an explanation, to counter the evidence of Mr Taylor, why it was legitimate to trade in 2 pin phones in the United Kingdom, notwithstanding that they could not be sold for use by consumers in the United Kingdom and must have been imported into the United Kingdom before the Appellants dealt in them. It emerged from cross-examination that there were trading links between the witness and individuals involved in the current appeals. Mr Deakin had worked for a company that had been supplied with counterfeit razor blades by a company called Simply Mobile Phones of which Mr Kyri Nicolas was a director. It had also been supplied with management services by a company called Europe IT Services. This was a company that had the same directors as ME365 and STC, and who were also directors of Axxia Holdings, Mr Michaelides and Mr Papaloizou (to which company the Appellants had loaned a large sum of money). This suggests a strong commercial trading link between the witness and the Appellants. It does not suggest that the witness’s evidence can therefore be treated as neutral expert evidence. With that in mind, the tribunal took his evidence into account in taking its view about the commerciality of the market in mobile phones as a whole.  

Anthony Elliott-Square

56        Mr Elliott-Square gave evidence in support of the Appellants in his role as a director of the Federation of Technological Industries, of which he was chairman at the time relevant to these appeals. He gave overview evidence about the Federation and its Best Practice Guide, about the grey market in mobile phones, about web sites used by mobile phone traders and about relations with the First Curaçao International Bank. His evidence about FCIB predated the closure of that bank by the Dutch national authorities, and is in the tribunal’s view of little weight. His evidence about the websites used by traders in mobile phones explained how the market could operate, but added little to the knowledge of the tribunal. He gave no evidence of specific assistance to the Appellants in connection with any of the deals in dispute. 

John Fraser Holmes  

57        Mr Holmes had served as an officer of HM Customs and Excise until 2001. He gave evidence that he had first-hand knowledge of investigations into VAT frauds at that time, but accepted under cross-examination that his first-hand experience was limited. He gave evidence of his advice to the Appellants, starting with his retainer as a VAT consultant to the Axxia group in 2002 at a time when it was a UK to UK phone trading company. His advice included steps that he advised should be taken to ensure due diligence in connection with specific suppliers, customers and deals. This included, for example, check lists. This was given mainly to Ms Leak, although Mr Andrew Nicolas had attended many of the meetings. His witness statement detailed advice about due diligence on each transaction. This included due diligence on suppliers, inspections of stock, and IMEI number scanning.

58        Under cross-examination Mr Holmes accepted that he had not known until the week before the main hearings in the case that Mr Andrew Nicolas, whom he had advised personally from 2002, had been convicted of VAT frauds. And the witness heard first that Mr Nicolas had been convicted of conspiracy to pervert the course of justice in connection with information obtained from a Customs officer only during the cross-examination. Cross-examination also established that this witness did not have a clear idea about the identity of the directors or of the controlling shareholders of the companies within the Axxia Group that he was advising. Nor was he aware of the previous links to the fraudulent companies of which evidence was given in the proceedings in 2004 and 2005 connected to the judicial review of HMRC decisions at that time.

59        The tribunal formed the view that Mr Holmes’ evidence was of limited value to the Appellants in establishing that they had conducted proper due diligence. The tribunal discusses in Annex B (deal 3) a specific point on which Mr Andrew Nicolas sought to rely on the evidence of Mr Holmes to support his own evidence about due diligence but where Mr Holmes gave evidence that did not support the evidence of Mr Nicolas. The tribunal prefers the evidence of Mr Holmes on that point, having also seen some of the reports he produced. It was, by contrast, significant in establishing a lack of internal due diligence within the Axxia Group despite the evidence of and for Mr Andrew Nicolas that he had turned over a  new leaf. It was clear that Mr Holmes himself also did not know, and had not been properly briefed by the Appellants, of the previous involvement of the Axxia Group in the mobile phone market as evidenced in the previous proceedings. Nor was there specific evidence that his knowledge or cautions were used by the Appellants, as was claimed, to enquire either about or beyond their immediate counterparties despite his and their knowledge of fraud in the market.

60        The tribunal finds that this undermined the credibility of the status attributed to him within the Axxia Group as someone there to ensure due diligence. It puts weight on the fact that the individual said to have been retained to advise the Axxia Group on due diligence not only had not been given responsibility in any way with ensuring that the Axxia Group stayed clear of the dangers of fraud in which its main decision maker had been directly involved, but also had not even been aware of the specific background and risks to the Group. 

Frederick Howarth

61        Mr Howarth gave specific evidence about a software package he installed at the Appellant Companies’ offices to enable them to record IMEI numbers and unit prices of stock purchased. He did this in 2003 and 2004. The records were kept in encrypted form. The data had been corrupted and he was not able to access it.

The tribunal deals with IMEI numbers in the decision below.

Claire Leak

62        Ms Leak was the company secretary to each of the companies in the Axxia Group throughout the periods relevant to these appeals. It is clear to the tribunal that her role was not merely a passive recorder of events but was more proactive and was a key part of the way the Axxia group worked in practice. Her evidence is of central importance to this appeal and is discussed more fully below. 

Andrew Nicolas

63        As anticipated by Lindsay J in the case management proceedings, the tribunal finds that the activities and evidence of Mr Andrew Nicolas are central to these appeals. It specifically records its finding that it sees no basis in fact for the submission made by Mr Patchett-Joyce to the Chancery Division, no doubt on instructions, that Mr Andrew Nicolas played no significant part in the transactions relevant to these appeals. His evidence is discussed fully below.

64        The tribunal agrees with Mr Benson QC that it should take into account Mr Nicolas’ record of criminal convictions when considering his evidence and the role he played in the Axxia Group. It is clear from the decision of Lindsay J (set out at [20] above) that it is entitled to do so. The convictions are plainly relevant to any consideration of the role Mr Andrew Nicolas played in the events that gave rise to these appeals. The tribunal was presented with the relevant formal evidence by officer Bycroft.

65        The tribunal accepts that Mr Andrew Nicolas was convicted on indictment at the Crown Court at Woolwich on 11 September 2000 of conspiring with others to cheat the public revenue and, on a guilty plea, was sentenced to four years imprisonment. He was charged together with nine others in connection with a multi-million pound evasion of VAT by MTIC fraud.

66        Mr Andrew Nicolas was further convicted on 21 December 2000 of conspiracy to pervert the course of justice and of conspiring to handle stolen goods. He was sentenced, on pleading guilty, to 2 years imprisonment consecutive on the previous sentence. He was charged with seven others. The offence involved corruptly obtaining information from a Customs officer working on customs frauds and money laundering.

67        Following these convictions Mr Andrew Nicolas and others were subject to confiscation proceedings on which a ruling was handed down by HH Judge Collins in the Woolwich Crown Court on 11 April 2003. This concluded that Mr Nicolas had no pre-fraud assets at the relevant time, and had limited assets available save for an interest in a house. In a further ruling by Judge Collins in the Central London County Court given after a hearing on 12 December 2003 and 6 April 2004, the judge identified further assets.

68        Further, on 25 January 2005, on the authority of Mr Justice Newman in the Administrative Court (CJA/61/1998), Mr Nicolas was committed into custody for a year for contempt of court in breaching a restraint order placed on him by Mr Justice Hidden in 1998 with regard to his assets and those proceedings.    

Kyri Nicolas

69        Mr Kyri Nicolas, the son of Mr Andrew Nicolas, joined the Axxia Group in 2002 having worked with another phone company. In 2007 he was the “trading director” for ME365 and STC Ltd, although not formally a director of either company. Indeed, he held no formal position in any of the Axxia Group at the times relevant to these appeals. He gave evidence in his witness statement that the Appellants outsourced their procurement and due diligence functions to STC. He also gave evidence about due diligence and how business was conducted. His evidence is discussed more fully below.

70        The Appellants put lengthy submissions to the tribunal about the law relevant to the entitlement of HMRC to refuse to repay input tax to the Appellants. The tribunal now has the advantage of the full argument before, and decisions of, the Court of Appeal on several of the issues raised before the tribunal. The decisions of the Court of Appeal in Mobilx also resolve some of the inconsistencies in approach in earlier decisions of both the Chancery Division and of the First-tier Tribunal and VAT Tribunal to which this tribunal was taken.

71        The tribunal is bound to follow, and readily follows, the guidance of the Court of Appeal in Mobilx. As already noted, that decision also resolves, so far as this tribunal is concerned, the questions raised at an earlier stage in this appeal of a reference of questions to the European Court of Justice. The tribunal can therefore deal without full discussion or further argument with several of the issues put before it, and it does so as follows.

72        The tribunal does not accept the Appellants’ argument that HMRC has no right to refuse input tax in circumstances such as those in these appeals without specific domestic United Kingdom legislation. HMRC is entitled to rely on European Union law alone: see Mobilx [49].

73        The tribunal does not accept the Appellants’ argument that the Appellants can be expected to be aware of the conduct or fraudulent actions or intent only of those parties with which the Appellants have a contractual relationship. In other words, the actions of the Appellants are not to be judged by reference to the Kittel test only in relation to their dealings with their immediate counterparties. The knowledge of the Appellants is to be judged with regard both to earlier and subsequent transactions: see  Mobilx [43], [62].

74        The tribunal does not accept the Appellants’ argument that the refusal of input tax to the Appellants by HMRC amounts to, or should be viewed as, a penalty, or in some other way as a breach of the Appellants’ protected rights under the European Convention on Human Rights: see Mobilx  [63] to [65].

75        The tribunal does not accept the Appellants’ argument that the actions of HMRC in refusing input tax to the Appellants constituted, without further specific allegations,  a breach of any fundamental or general principle of European law such as legal certainty or proportionality on the basis only that it was applying the decision in Kittel: see Mobilx [61], [62], [66].

76        It follows from the above that the tribunal does not consider that it is required or entitled, as part of these appeals, to consider whether HMRC’s decisions to withhold input tax from the Appellants in respect of a tax loss where they might also have withheld that same input tax from some other party to a chain of transactions involving  that tax law are proportionate or otherwise open to challenge. It is sufficient that the tribunal is given, and accepts, evidence that there has been no duplication of recovery. If the tax loss has been recovered from some other taxable person, then it must follow that that tax loss is no longer a loss in the relevant chain.

This is consistent with the emphasis placed by the Court of Appeal in Mobilx on the decision of the European Court to extend the scope of those who can be identified with a fraud. It must follow that the result will normally be that more than one trader is identified with a single fraud.

77        Likewise, the tribunal does not accept that the Appellants are assisted by the principle of fiscal neutrality. This must clearly be examined in the context of the test laid down in Kittel as now applied in the United Kingdom in Mobilx and in now way limits it.

83        The evidence and submissions about the identity of the directors and shareholders of the Axxia Group companies was neither clear nor consistent. The tribunal considers it necessary to examine the issue and make its own findings about points put, together with a selection of relevant documents, to the relevant company officers in cross-examination.

84        The tribunal finds that Axxia Holdings Ltd was established in July 2003. The idea was that of Andrew Nicolas. It was set up to be used, in his words, “as a vehicle to reverse the shares of the trading vehicles into it”. Those trading vehicles included both the Appellants. The tribunal accepts that the timing of the establishment was during the period when court confiscation proceedings against Mr Nicolas, following him pleading guilty to MTIC fraud charges, stood adjourned. The proceedings, before Judge Collins, had taken place in March and April 2003, but were adjourned to be completed in December 2003 and February 2004.

85        The tribunal finds that Mr Nicolas’ involvement in Axxia and its trading subsidiaries were not disclosed in the proceedings or to the judge. It also finds that, following from the findings of the judge in the confiscation proceedings, the only personal assets available to Mr Nicolas to introduce into the Axxia Group at that time were either assets obtained either after or conterminous with those proceedings and after the frauds for which he was convicted or were assets not disclosed in those proceedings.  

Axxia Holdings

86        Turning to the structure of the “trading vehicles”, the tribunal starts with the company returns to Companies House of Axxia Holdings Ltd for the year ending 31 July 2004. These recorded that the board of directors of the company as at the end of the year consisted of C Papliozou, L Michaelides, and Andrew Nicolas. The company secretary was Claire Leak, and she made the returns. The directors’ report shows that these directors were appointed on 14 July 2004, when a previous director retired. It also shows that at that date the only director who was a shareholder was Andrew Nicolas, who held 800 £1 ordinary shares at that date. The total number of these shares issued is stated to be 1,000. The importance of this is confirmed in note 6 to the financial statements:

            “The company was under the control of Mr A Nicolas throughout the current

            Period. Mr A Nicolas is a director and the major shareholder.”

That is confirmed by details in the annual return made for the company by Claire Leak. This shows the 800 shares held by Mr Nicolas, with three other shareholders. All three are companies, two being British companies and the third based in the British Virgin Islands. It is noted elsewhere that Mr Nicolas was allotted 798 shares. The tribunal finds that the other two shares were shares that Mr Nicolas had previously acquired from the original shareholders from the formation agents that established the company. It therefore finds that Mr Andrew Nicolas was the controlling shareholder of Axxia Holdings Ltd, and therefore of the companies it controlled, continuously from the time he acquired the two shares from the original shareholders.   

STC

87        The Axxia Holdings Ltd return confirms that in 2004 it was the sole shareholder in two controlled subsidiaries: STC and ME365. It became controlling shareholder of STC on the day of its formation in July 2003 as STC gave a shareholding of 99 ordinary shares to Axxia Holding Ltd. It had a 51 per cent shareholding in Live Telecoms Ltd. Contemporary documents confirm that STC was a wholly owned subsidiary of Axxia Holdings Ltd. The person signing as the company secretary was Claire Leak.   The two directors of STC are shown as Lakis Michaelides and C Paploizou. These are the two directors of Axxia Holdings Ltd other than Mr Andrew Nicolas.

ME365

88        The internal interlinking of the Axxia Group at all relevant times is confirmed by the detail of the ME365 annual report for the year to July 31 2006, signed in November 2007. These confirm that the ultimate controlling party of ME365 in both that year and the previous accounting year was Mr Andrew Nicolas as 80 per cent shareholder in Axxia Holdings Ltd. Further, ME365 paid Axxia Holdings Ltd £147,000 in the year to July 2006 for management services. The tribunal received no direct information about those management services, but notes that the company secretary of ME365 – the person who might be expected to provide or organise those services - is Claire Leak, the company secretary of Axxia Holdings Ltd.

89        The directors of ME365 throughout the year were C Papaloizuo and L Michaelides, the same individuals as were the directors of STC and were directors also of Axxia Holdings.  They were paid no emoluments in 2005 or 2006. No dividends were paid. Ms Leak gave evidence that the group decided that it was easier to pay those involved from the STC payroll rather than from each of the companies. Those accounts show that Mr Andrew Nicolas was paid £410,000 for the years 2005 and 2006 together. They also show sums paid in consultancy fees to a company called Classic Hotels Ltd. Ms Leak’s evidence was that the directors of that company were Mr Papaloizou and Mr Michaelides.  Monthly invoices from that company to the Axxia group totalled £283,000 in 2006. No substantive evidence was offered of any services provided by the company to the Axxia group. Further, £500,000 was lent in December 2005, and a further £2.5 million in March 2006 on unsecured loans to these directors. Ms Leak’s evidence was that this was linked to property development.

90        The ME365 accounts show a number of changes in the structure and functioning of ME365 in that year compared with the previous year. The number of administrative staff dropped from 5 to 1, with 1 member of management staff in both years. At the same time the service charge paid to Axxia Holdings increased from £28,000 for the year to £147,000. Total turnover from sales dropped from £246 million to £118 million. Turnover from sales to countries outside the European Union dropped from £168 million to £10 million, while sales within the European Union increased. During the year, £112 million worth of goods were purchased from STC.

SITL

91        The relevant annual accounts of SITL were for the same period, to 31 July 2006, and were signed on the same day. They confirm that the ultimate control lay with Mr Andrew Nicolas. The sole director throughout the period was S Solomonides. He was paid emoluments of £70,000 for the year to July 2006. The accounts record the repayment to Mr Solomonides of a loan of £147,000 and interest in that year. The company secretary was, again, Claire Leak. The company paid the same amount by way of management charges to Axxia Holdings as did ME365 in both 2005 and 2006. But, unlike ME365, its staff remained constant at 1 manager and 1 administrator. The turnover from sales dropped from £45 million for 2005 to £36 million, all of which is recorded in 2006 as “overseas”. During the year it purchased £34 million worth of goods from STC.

92        The tribunal understands that during the year in question other Axxia Group companies had significant staff based in the same buildings as the Appellants but those companies took no part in these proceedings.

The directors

93        The tribunal records that it received no evidence from, and little evidence about, the directors of the appellant companies. It notes in particular that Mr Michaelides and Mr Papaloizou were directors of  ME365, STC and Axxia Holdings throughout the relevant period. Neither offered to give any evidence on behalf of the Group nor took any other part in these proceedings. It also records that it was not supplied with any relevant records of meetings of the boards of directors of any of the companies. It was unable to establish from any of the usual sources what role, if any, the directors played in the companies.  Ms Leak gave evidence that Mr Michaelides and Mr Papaloizou were both present “several times a month” and that they took part in the overall direction of the companies. The tribunal saw and heard no other evidence to confirm this. Nor was any evidence offered by Mr Solomonides, although his remuneration and other evidence suggested a significant involvement in the activities of the Axxia Group.

The management of the Group

94        In the absence of that evidence, the tribunal forms its conclusions about the direction and management of these companies on the basis of that absence of evidence, the documentary evidence that was available, and the evidence given by the company secretary of all the companies, Claire Leak, the ultimate controlling shareholder of all the companies, Mr Andrew Nicolas, and his son, Mr Kyri Nicolas.

95        The tribunal finds that it is not possible to view the separate companies in the Axxia Group in isolation in terms of finance, management or direction at the times relevant to these appeals. Mr Kyri Nicolas stated in evidence that “we would all muck in and help each other”. This followed evidence about how he would help those working in other companies in the group. But in reply to the following question in cross-examination from Mr Benson QC he stated that at the same time “each company had its role and each company had its own directors and its own shareholders.” The tribunal finds that there was no significant difference in the shareholding  between ME365, SITL and Axxia Holdings. Control of all of them lay with Mr Andrew Nicolas. The overlap in membership of the boards of directors and the company secretary has been noted.  The tribunal finds that Mr Nicolas’s description that they all worked closely together is an accurate summary of the situation and that his evidence that the companies were nonetheless separate was not supported by the other evidence. The tribunal finds that the reality was that there were no divisions of any significance in the management and running of the companies or in their finances. It follows, and the tribunal finds, that what was known, or should have been known, by any of the companies or those managing them was known or should have been known by all of them.

96        Of the roles of the individuals involved, the tribunal was given evidence as follows. It was originally, at the start of the appeal process, told that the role of Mr Andrew Nicolas was not relevant. He gave evidence only after the unsuccessful appeal to the Chancery Division by the appellants on that point.  In evidence, Mr Andrew Nicolas confirmed that in terms of the Axxia Group and staffing he was “actually doing quite a lot”. He confirmed that he was in charge of strategy, grey markets and overall direction. Ms Leak’s evidence was that he was present “two or three days out of every five”. The tribunal finds that he was, during the relevant period, the person with central management and control of the Group and in particular of the Appellants. 

97        The tribunal further finds that he was prepared to exercise that control without regard to either the formal structure of the Axxia Group or the roles supposedly being played by others. The tribunal noted above the appointment of Mr Holmes as a consultant, but the complete failure to inform Mr Holmes of Mr Nicolas’s history despite the obvious importance of that information to any advice to be given by Mr Holmes. Further, Mr Nicolas was prepared to accept in evidence that he had on several occasions lied or otherwise misled Claire Leak with regard to matters relevant to her role as company secretary. That is consistent with the view originally taken by the Appellants that his evidence was not of relevance to this tribunal in these appeals.

It is also consistent with the absence of any evidence from the other individuals named as directors of these companies or of any decisions attributable to those directors.

98        The evidence of Mr Andrew Nicolas has previously been the subject of adverse comment by the judiciary. It is clear to this tribunal that following his conviction for carousel fraud offences, Mr Nicolas senior decided to go back into the same area of commercial activity, and that he did so using both the knowhow and financial resources gained from his previous activities. It is also clear to the tribunal that the Appellants did not wish to disclose, and attempted to conceal, the extent to which he was and is involved in both the ownership and the central management and control of the Axxia group of companies. He gave evidence to the tribunal only after his legal advisers failed in attempts to keep his evidence off the record. He admitted candidly that he had failed to “turn over a new leaf” (to use Ms Leak’s phrase), that he had been involved in the VAT frauds and in a failure to disclose assets, and that he had misled not only the court but also his own company secretary about this.

99        At the same time Mr Andrew Nicolas appeared to be trying to convey the approach that he had been involved in a very simple fraud and that he did not really understand what was going on as MTIC frauds developed. In evidence he put it this way:

            “It is very easy. The difficulty I have in this tribunal is to say that I decided to          do the best I could. Even the stuff that I learned from the missing trader days, I tried to implement for the good of the company. I know my track record is not    particularly good, but I tried my best throughout these companies’ trading        periods, from 2002 to the present day. “

The tribunal rejects that evidence save only in the limited sense that Mr Nicolas wished the Axxia Group to make money and from early 2006 (and before) that involved any activities that would support the input tax claims. The tribunal does not accept that Mr Nicolas was not aware of the current extent of fraud in the grey market at the time relevant to these appeals. It finds that he personally had all the contacts, or could readily make them, that he needed to be fully informed of the state of the market, including the various kinds of fraud being operated within the market, and that he had an active knowledge of such fraud and of some of those who perpetrated it. It does not accept that his knowledge about the trading activity, or alleged trading activity, of the Axxia Group, which was at all times his group, was limited only to the direct sellers to the Group and the direct purchasers from the Group. 

100      It was clear to the tribunal, and it finds, that Mr Nicolas was well informed about the market in which the Axxia Group was operating and about fraud in that market, and that anything that Mr Andrew Nicolas knew or reasonably ought to have known was knowledge or information available to the Axxia Group as a whole and therefore the Appellants. By contrast, his evidence in this appeal is neither reliable nor credible. 

Mr Kyri Nicolas

101      Mr Kyri Nicolas had no formal role in any of the companies, either as office holder or shareholder.  He acknowledged that he was a director and shareholder of another company with his uncles, but that was not directly associated with the activities of the Axxia Group. He described his role in the group in the relevant periods as looking after the export side of the business. He ran a team of four people working in STC, SITL and ME365.  His activities included relations with customers and in particular due diligence. He agreed that Mr Philippou was the sales director of STC who bought in the stock. Mr Solomonides helped with the paperwork. Ms Leak stated that Mr Solomonides worked alongside Mr Kyri Nicolas and Mr Kyri Nicolas took the responsibility for the trading. She stated that Mr Philippou was not the sales director of STC and that he was under the direction of Mr Kyri Nicolas, who was in charge of everyone for the purposes of day-to-day trading. Further, her evidence was that he was involved in all the deals relevant to these appeals in that he and his team organised all the deals, and that he would have approved each purchase as well. 

102      Mr Kyri Nicolas, in evidence, took the repeated view that in conducting the trades for the Group his responsibility was to know about his immediate counterparties. He was clearly aware of the importance of due diligence but did not accept that this involved him or the Axxia Group in any enquiries beyond the immediate suppliers to the Group or customers of the Group.  

103      There was no evidence before the tribunal of any “Chinese wall” or other limitation on the relations between Mr Kyri Nicolas and his father in these deals. The general tone of the evidence suggested, and the tribunal finds, that relations between the two were strong and that they would have been free to discuss, and would have discussed together, any issue of concern to Mr Kyri Nicolas in handling deals or counterparties. The tribunal finds that he would follow his father’s guidance, and others would follow his instructions. It finds that any decision by Kyri Nicolas not to investigate or consider risks beyond the immediate counterparties to his deals was therefore either an approach adopted for the purposes of the evidence he chose to give the tribunal or an approach adopted more generally as a way of avoiding knowledge that he might otherwise reasonably have acquired. It does not accept that he was entirely open with the tribunal about the extent of his knowledge beyond the counterparties.

Claire Leak

104      Ms Leak confirmed that she was responsible within the group for administration, banking and finances but not trading. She has worked with the Axxia Group from 2002. Her initial evidence was that she was responsible for day to day running of the businesses along with Mr Kyri Nicolas and Mr Solomonides. In particular, she stated that she was responsible for ensuring that all relevant due diligence requirements had been completed before releasing funds. Consistently with that she gave evidence in her statements about the due diligence procedures followed. 

105      Ms Leak was the last witness to give oral evidence, so had the advantage of listening to most of the proceedings at the main hearing, including the evidence of most other witnesses. In that context, the tribunal particularly noted the extent to which, in giving evidence, she either formally declined to give evidence or commented that matters were not her concern. She specifically refused, after a warning by the tribunal about self-incrimination, to answer a question about assets she jointly owned with Mr Andrew Nicolas and the failure of either of them to declare the assets in his net assets statement made to the court in previous proceedings. She declined to answer questions about various financial transactions of companies in the Axxia Group because they reflected on the personal position of Mr Andrew Nicolas or others. Mr Benson QC also put to her a conflict in her sworn evidence to this tribunal and her sworn evidence in previous proceedings about a significant loan by the Axxia Group to a third party company. The witness’s general response was that she could not recall details. This approach she applied under cross-examination on one day to the evidence that she had given to the tribunal the day before when she was questioned on the matter without reference to her previous sworn evidence. She was also questioned about the requests for IMEI numbers, a matter on which both Mr Goulding for HMRC and Mr Howarth for the appellants had given evidence. Again her answers were that she had no clear recollection. The same pattern appeared in her evidence about due diligence. Her response to a number of questions was that this would have been for Mr Kyri Nicolas or Mr Solomonides to check, and it was they, rather than her, that had assembled the disclosed information about due diligence. Finally Mr Benson QC put to her evidence she had given to the court in the confiscation proceedings, pointing out to her that as part of that evidence she had there been warned by the judge about her evidence. As part of that he put to her that she knew that part of the proceeds invested in STC had derived from fraud. She accepted that she had found out about this only during the process of attending to give evidence in the court proceedings.

106      The tribunal records that it finds the evidence of Ms Leak of little value both in terms of her claimed oversight of banking and finance and her claimed central role in coordinating due diligence. She was clearly not privy to the full reasons for some major financial transactions of the group of companies of which she was not only the company secretary but which she represented to the tribunal fell within her specific responsibilities in connection with banking and finance. At the same time, she initially claimed overall responsibility for due diligence in discussion in particular with Mr Frazer Holmes. But when specific deficiencies were put to her, her response was that the responsibility was that of others. The tribunal was left with the impression that she had not in fact had any real input into due diligence procedures and it finds that her evidence does not assist the argument of the Appellants that they both had a system for conducting proper due diligence and used that system. 

107      More generally, Ms Leak’s evidence was clearly lacking in credibility given her failure on oath to explain her failure on oath to disclose to the court assets in which she had an interest in the confiscation proceedings. The tribunal finds, overall, that her evidence was unreliable and evasive. It formed the clear impression that she knew more about the activities of the Axxia Group and the markets within which it was operating than she was prepared to disclose even in answer under oath to direct questions.  

108      Various other individuals were stated to be the directors, either formally or in name, of the companies in the Axxia Group. The tribunal finds that none were in reality the decision makers. The decisions were those of the senior and junior Mr Nicolas. Their knowledge and sources of knowledge were central to the knowledge of, and available to, the Group. The tribunal does not accept that any due diligence system put in place in the Axxia Group was of sufficient strength either itself to lead to a conclusion contrary to their views or to empower Ms Leak or others to change a decision taken by those two individuals. 

109      The tribunal therefore concludes that the question whether the Appellants knew or ought to have known of fraud in connection with their transactions can be answered by reference in particular to the knowledge of Mr Nicolas senior and Mr Nicolas junior and what they ought reasonably to have appreciated. It follows from these findings that those who took the decisions in the Axxia Group had clear knowledge of the existence of active fraud in the market in which they chose to continue trading and a full awareness of the need to ensure that their own transactions were kept separate from that fraud. They knew of this because of the involvement of the Axxia decision makers in the previous proceedings before the courts concerning exactly these kinds of issues and because of their employment of HMRC officers (including the officer whose employment gave rise to criminal charges) to assist them.  

110      It also follows that the tribunal is unable in this decision to rely on the evidence of Mr Andrew Nicolas or Ms Leak, and is able to rely on the evidence of Mr Kyri Nicolas only to the extent that it is confirmed elsewhere by the evidence. 

111      The tribunal now turns to specific aspects of the evidence put before it.

112      The first questions for the tribunal to answer are whether it was satisfied on the evidence that HMRC had established on the balance of probabilities that there was a tax loss in each of the supply chains that led to the transactions for which the Appellants were reclaiming input tax and whether that tax loss was connected to fraud.

113      Each of the transactions is detailed in Annex A. As noted in Annex A, each of the facts recorded in the Annex for each deal constitutes a finding of fact by the tribunal. 

114      The evidence accepted from individual officers of Revenue and Customs in evidence to the tribunal is recorded above to the extent that that evidence was given orally to the tribunal. The findings in Annex A also take into account the findings in Annex B about both the specific deals analysed in Annex B and the other deals identified as having common features to those deals.

115      Save for the issues raised with regard to the witness evidence about Primeline and F Options as defaulting traders named in the original decisions, the tribunal accepts the decisions as correct with regard to tax losses. In doing so, it draws on findings in Annex B. With those exceptions, it finds that there were tax losses as contended by HMRC in all the chains identified in Annex A.

116      With regard to F Options Ltd, the tribunal finds on the evidence in [43] above, that the transaction involved the use of a hijacked VAT number in the name of M Allen and accordingly there was a tax loss in the chain and that the use of the hijacked number is clear evidence of fraud. Annex A records this finding, noting the defaulter as “M Allen”.

117      With regard to Primeline, the tribunal noted above that Mr Craddock, the officer with responsibility for this company, was unable under cross-examination fully to establish the evidence given in his statement. The tribunal revisited his statement and the supporting documentation in the light of the answers he gave under cross-examination. His written evidence was supported by the production of approaching 1,000 pages of documentation relating to 46 separate deals identified for April 2006. The detailed deal schedules supporting the analysis of these deals showed recurring patterns of suppliers and customers undertaking rapid sales evidenced by standard simple documentation with payments directed through First Curaçao International Bank (“FCIB”). Mr Craddock gave evidence, which the tribunal accepts, that there has been a tax loss from Primeline with no subsequent recovery of VAT not paid. While he could not establish that for all the transactions on which he relied, the tribunal finds on his evidence, taken as a whole, meets the standard required to establish the tax loss. The tribunal also accepts the evidence of fraud given in his statement.

118      Accordingly the tribunal formally finds, as recorded in Annex A, that there was a tax loss in each of the chains of suppliers, and the contra-trades, leading to every one of the deals in the table (that is, all the deals under appeal).

119      A second general finding from Annex A is that the specific deals to and from the Axxia Group with which HMRC took exception all took place, at least in broad terms, as alleged. In each case phones described by the make and model set out were supplied to the Axxia Group from a supplier based in the United Kingdom (and therefore making the supply subject to VAT) and supplied by it by way of an intra-Community acquisition or export (and therefore subject to an entitlement in principle to a refund of the input tax paid by the Appellant).

120      A third general finding from Annex A is that there are patterns and aspects of the series of transactions that suggest that the deals were not, as contended, simply the product of “best price” sales to willing customers from the best supplier available on the day.  These patterns and aspects are considered by the tribunal in Annex B and Annex C.

121      Annex B sets out the detailed findings of fact of the tribunal concerning the four transactions selected by HMRC at the invitation of the tribunal for its detailed investigation. Both parties were invited to nominate deals. The Appellants did not respond to the invitation. The tribunal did not select these deals itself. In respect of each of the deals, the tribunal has noted all available documentary evidence (starting from the bundle of documents assembled by HMRC) and all relevant references from the transcripts of evidence. It has concentrated on the available details of the chains of transactions as put forward by HMRC, and whether these deals evidenced fraud and  tax losses and then on the evidence of due diligence and enquiry by the Appellants about the deal.

122      The tribunal finds that each of the four transactions took place broadly as indicated in so far as there was a series of supplies of phones through a chain of suppliers to the Axxia Group and then from the Group to a customer. This confirms the finding of the tribunal that the supplies of goods took place as contended by the Appellants and as recorded in Annex A.

123      The tribunal found several indicators that the deals were not genuine market deals at all stages, and that there was probably fraud in the deals. Further, the evidence in one of the deals suggested that the claimed intra-Community acquisition of goods probably did not take place although the relevant counterparties purported to represent that it did.

124      The first of the deals examined was SITL deal 2 of March 2006. The tribunal finds it to have sufficient common features to be regarded as part of the same overall transaction as deal 1 of SITL and deal 3 of ME365 to the same customer on the same day. This is, in the tribunal’s view, an illustration both that individual deals on a day may properly be regarded commercially as parts of a single transaction and that there was no practical distinction between the deals carried out using ME365 and those carried out using SITL. It finds clear evidence of fraud in the chain of transactions leading to the supply to the Axxia Group. It accepts the evidence of HMRC that one of the companies in the chain, Oracle, is a defaulting trader. Oracle is the company said to have received the goods into the United Kingdom on an intra-Community acquisition, although the available evidence is that the goods were probably in the United Kingdom when the supply was alleged to have been made. In any event, the tribunal accepts that there was a tax loss in the chain.  

125      The tribunal examined in detail all the available documentary and oral evidence about the conduct of the companies that supplied, and were supplied by, the Axxia Group in SITL deal 2 of March 2006. Its detailed conclusions are in Annex B. The tribunal finds that the Axxia Group’s investigations of both the supplier and the customer were less than satisfactory. There were features of both relationships that lead the tribunal to find that the deals were not arms length deals conducted in an open competitive market. The tribunal further finds that this deal, in common with most of the deals undertaken in the disputed period by the Axxia Group, evidences an unexplained link between the company supplying the Axxia Group, the Group, the company being supplied by the Axxia Group and, behind the supplying company, recurring patterns of chains of brokers. This is clear from Annex A and the full scheduled evidence of deals produced by HMRC which, save for a few small points irrelevant to this general comment, the tribunal accepts as proved on the balance of probabilities.

126      The tribunal records similar findings with respect to the other three deals examined in detail, and adopts into the decision and Annex A all findings of fact in Annex B.

127      The tribunal concludes from this that there are significant failings in the evidence produced by the Appellants that the deals were, as was argued for the Appellants, deals taking place in the open market after full due diligence by the Appellants and at arm’s length.

128      In each of the four deals the tribunal accepts HMRC evidence of a defaulting trader and of fraud in the chain, and finds accordingly.

129      As recorded in Annex B, the tribunal finds common elements between these deals and others and accepts the HMRC evidence as indicating a tax loss and fraud in those deals also. This includes findings about @tomic Ltd, a counterparty involved in a series of contra deals involving the Appellants. 

130      HMRC put considerable weight in its submissions about the failure of the Appellants to show that they had kept proper checks, as requested by HMRC, on the IMEI numbers on the phones with which they were trading. The tribunal’s view, reinforced by the guidance of the Court of Appeal in Mobilx, is that the IMEI evidence is of importance and is to be given weight in the tribunal’s analysis of what the Appellants knew or ought to have known of fraud connected with their deals.

An IMEI (international mobile equipment identification) number is unique to every mobile phone produced in Europe. The number is a string of digits recorded in the phone and in bar-coded form on the packaging of the phone. Each IMEI number is a sting of digits identifying the reporting body (for these purposes the manufacturer), the model number, and a unique unit serial number, followed by a check digit. It is therefore a number that not only identifies the make, model and serial number of the phone but also contains a check digit to identify if a number has become corrupted. Logging IMEI numbers is one of the ways in which a trader in the mobile phone market can look “up” the chain beyond its immediate suppliers notwithstanding commercial confidentiality. It can also check on sales from its customers. This is because a consistent record of IMEI numbers will show if the trader handles the same phone twice. In addition, HMRC keeps a national record of IMEI numbers scanned at ports during supplies overseas (a system referred to as NEMESIS), so can inform a trader whether a particular phone has entered or left the United Kingdom on any previous occasions (and if so, when). A trader cooperating with HMRC can therefore monitor the extent to which its phones are involved in other deals involving fraud.

131      The tribunal accepted without argument the technical aspects of IMEI numbers summarised in the previous paragraph. It accepted the evidence from HMRC of NEMESIS checks (checks through internal HMRC systems of numbers collated from its own scans of goods coming into or leaving the jurisdiction). It does not consider that it need repeat those technical aspects of that evidence. It accepts that the Appellants were not aware of, and had no access to, any data from NEMESIS save in so far as this was made available to them by HMRC or as part of these proceedings. But it also finds that the Appellants were fully aware of the importance of both IMEI checks and the NEMESIS system.

132      HMRC submitted that there were two aspects to the IMEI evidence on which it was entitled to rely. The first was the evidence from its own analyses of scans of goods being transferred out of the jurisdiction by ME365 or SITL. The second was the failure by the Axxia Group to produce clear evidence to show that it had undertaken, as it contended it had undertaken, its own checks on IMEI numbers.

133      Evidence of the results of IMEI checks through NEMESIS were given by Officer Barnett. Only some of this evidence related directly to the period in question and could be identified specifically with deals before the tribunal.

134      One report that was clearly identified as linked, the tribunal finds, is NEMESIS batch report 1022. This relates to scans of Nokia N6230i phones sent to Midcom and logged as ME365 deal 20 of March 2006. The HMRC evidence, presented by officer Barnett, was that of the 656 phones scanned from this batch, 224 of the phones had been scanned in NEMESIS batch ME365 937 and 205 in ME365 batch 938, with 262 being identified in SITL batch 262. This shows a significant number of the phones in this deal as having previously been presented by the Axxia Group to HMRC – with a duplication rate of over 30 per cent on each report. The evidence is that batches 937 and 938 were presented as goods being consigned to Farouk & Suhail on a deal dated 28 02 2006. Separately from this, the tribunal notes that the IMEI numbers in the batch report are not entirely consistent, some starting 3570660 and others 3583680. That separately suggests that the phones were not all of the make and model declared.

135      While officers accepted under cross-examination that some of this information had not been fed back to the Appellants at the times relevant by HMRC, the tribunal does not regard that as removing the relevance of the evidence. It may be that HMRC has not shown that it directly warned the Appellants about this information but that does not detract from the clear evidence of a high level of circularity involving an established customer of the Appellants. If the Appellants maintained proper records, as they said they did, they knew this.

136      NEMESIS ME365 batch report 1323 also illustrates this. This links to ME365 deal 3 of 13 03 2006, a consignment of N 6230i phones to Farouk & Suhail. 779 phones were scanned. The evidence shows a duplication rate with phones in batches 939 and 940 presented by ME365 and 949 and 960 presented by SITL. The duplication varied from 2 per cent in batch 960 to 16 per cent in batch 949. Batches 939 and 940 were consignment presented to HMRC as going to Farouk & Suhail (on 28 02 2006) as was batch 949 (on 13 03 2006). This latter batch appear to be part of deal 2 by SITL in March 2006, a supply of 5,000 N 6230i invoiced on that date, so directly relevant to these appeals. This raises other questions, as it appears on its face to suggest that the phones were presented for consignment twice at the same time. That may suggest that this aspect of that report may not be reliable, or that there was deliberate double counting. 

137      The evidence includes several other NEMESIS reports showing significant levels of duplication. The tribunal does not consider it necessary or proportionate to follow through each of them. It finds that while there may be scanning errors in some of the information, there is sufficient evidence showing significant duplication produced by HMRC to make the question of duplication and therefore of IMEI checks directly relevant to the determination of these appeals. That duplication is direct and specific evidence that goods previously provided by the Appellants to its customers must have been circulated by those customers and subsequent suppliers back into the United Kingdom and to the suppliers to STC. It shows that the Appellants were using previously circulated goods and not brand new goods. Further, the evidence showed the timing of this circulation. It was rapid: the classic evidence of carousel fraud. Those duplications, which on their own evidence should have been known to the Appellants, put them on clear notice that they were dealing both with circulating goods and with chains involving companies that were circulating the goods. The clear inference is that they knew or should have known that they were handling goods that had been connected with carousel fraud and that their suppliers and customers may have been connected with such fraud. Separately, some of the IMEI records suggest that the supplies were of mixed products, including some phones of a model number other than that recorded for the transaction. That also should have put the Appellants on warning to ensure the integrity of their supplies separately from any issues of VAT fraud.  

138      Those conclusions put directly in issue the extent to which the Axxia Group took the steps it contended it had taken to identify and deal with IMEI numbers and duplication. 

139      Instead of producing this evidence, the Appellants produced evidence to explain their failure to produce this evidence. This included the evidence of Frederick Howarth. His evidence was that he had supplied the Axxia Group in 2003 with bespoke software, enhanced in 2004. The software was designed to perform two functions. The first was recording prices discussed while a trader was on the phone discussing prices with a supplier. The other was to hold scanned barcodes displayed on mobile phone stock held by a freight forwarder and emailed to a trader as an Excel spreadsheet. The software was designed to keep the data confidential and so encrypted the information held by jumbling the IMEI numbers.

140      The evidence was that the database became corrupted, and so the Appellants were unable to download data from it. Mr Holmes was unable to establish why or how the system became corrupted, surmising that it might have been an uploaded file that corrupted it or it might have been overloaded. He could not explain how a corruption could be uploaded on a properly protected system. As to the resulting loss of all records, he confirmed in oral evidence to the tribunal that he anticipated that the Axxia Group would have back-up systems to hold their data, but he was unable to offer evidence about any such back-up. Nor was he able to assist further either about the problems with the database or of the refusal to allow access to HMRC officers to see for themselves how it had failed.

141      Ms Leak was asked about IMEI numbers and their provision in cross-examination. It was put to her that there was unchallenged evidence that officer Goulding had asked for IMEI numbers from Axxia in 2004 but they had not been supplied. She was unable to recall this. Her evidence was that he had been offered access in the previous year but had declined it. She also accepted that there had been problems with the software in 2004. The tribunal notes from this evidence and that of Mr Howarth that neither of them could recall the request for data in 2004, and further that a request for data made later to the Appellants’ solicitors was, according to Ms Leak, not passed to her so was not answered by her. The tribunal finds this lack of recall unconvincing. It accepts the evidence that requests were made, and repeated, by HMRC from 2004.

142      It is clear from this evidence that the Appellants accepted that IMEI scans were relevant to due diligence when handling mobile phones, and to their own due diligence in particular. It is also clear that they knew of the value of the evidence available from IMEI scans. The tribunal finds that this is why the Axxia Group both required IMEI numbers on inspections and commissioned its own software to analyse the IMEI numbers reported.

143      Submissions were made by both parties about the significance of the absence of evidence from the Appellants about IMEI checks actually undertaken and the occasions on which HMRC had asked for details. The tribunal finds as follows. The Appellants were aware of the importance of IMEI information both to the Group and to HMRC. HMRC had more than once put them on notice of its interests in their records of these numbers. There is clear evidence of repeated duplications of supply in the phones being handled by the Appellants in the period relevant to these appeals. This involved direct circularity of supply through customers with whom the Group continued to trade. The Appellants were clearly aware of that problem in general terms, and of its significance. They were clearly aware, at least from 2004, that they could take steps to avoid such involvement. That being so, the tribunal would expect traders such as the Axxia Group to have taken reasonable steps with regard to these aspects of their deals. The tribunal must therefore take a view on the evidence of failure in the software system being an answer to the failure of the Appellants to have this evidence.

144      The evidence for the Appellants was inconsistent. What Ms Leak referred to as a “glitch” first occurred in 2004. From that time the Appellants were aware both that HMRC wanted IMEI information from them and that they had internal problems with their own handling of the information. The evidence offered to the tribunal was that the Appellants had a system that had worked but had then become corrupt in such a way that evidence could not be made available later to HMRC. But it did not give evidence of what it did to ensure either a backup or alternative forms of record-keeping. Nor did the Appellants produce any evidence that the records, before they became corrupted, either had or had not indicated any duplication of supplies. If there was evidence of duplication, then the tribunal would have expected to see other evidence, perhaps checked with Mr Holmes, about the assessment of the due diligence implications of any duplication. The tribunal does not accept the evidence that was given is a sufficient explanation of the failure to produce any evidence of IMEI checks during 2006 or of the failure to assist HMRC in seeing if HMRC officers could access the corrupted data. If the internal system was working at an earlier time, then the NEMESIS checks show that the Appellants had direct knowledge that they were handling circulating goods that they had previously supplied to continuing customers. The Appellants had received that information separately from their internal systems. It came in the form of emails of data collated on inspection of goods in the warehouses. It could become corrupted only after receipt and inclusion in the faulty database. Yet the Appellants knew of the risk of corruption in 2004. There was no evidence that the Appellants attempted either to introduce a back-up to the corrupted system or to keep alternative forms of record such as the emails sent to them. 

145      There are two ways this might be explained. One is that the Appellants chose to blame, or create, a systems failure to explain the non-production of evidence that it did not wish to produce. The other is that they introduced the checks without taking them seriously, so did not bother to take action as a result of the checks and consequently could produce no evidence that they had taken any action resulting from the checks. Either explanation leads the tribunal to conclude that the Appellants chose not to use IMEI evidence to assist them in ensuring that they stayed away from handling circulating goods.

146      IMEI numbers indicate whether the specific phones scanned are the kinds of phones they are supposed to be. Evidence from HMRC shows, and the tribunal finds, that the IMEI scans revealed evidence of mixed phones in supplies. Linked with this, IMEI numbers also show whether a batch of phones is a single batch coming from the manufacturer as part of a batch of phones manufactured at the same time and with the same specification or whether the batch had been assembled from different batches manufactured at different times and possibly with different specifications. This information is clearly of value to traders concerned with the specific identity and condition of phones. Mr Kyri Nicolas identified such concerns in evidence. The tribunal was offered no evidence that the Appellants had taken specific action resulting from any evidence of wrong phones or mixed batches being supplied as a result of IMEI records. Such a check did not need a database, least of all a bespoke database. The records could be checked by a simple visual scan of the numbers in the lists emailed to the Group as part of the inspection reports it commissioned. There is no record that this was done or that any action was taken indicating that this information was used. The tribunal does not accept that the corruption of the database, if that occurred, explains the total absence of any alternative evidence about such checks.

147      In summary, the tribunal accepts the evidence produced by HMRC showing duplication in supplies made by the Axxia Group and of mixed supplies and mixed batches of phones in the supplies. This is not rebutted by any evidence produced by the Appellants. The tribunal does not accept the explanation for the absence of any rebutting evidence from the Appellants. In particular, it does not accept the evidence that any failure in the bespoke software both explains why the Appellants could not produce any other evidence about IMEI records and why they failed to accept the offer from HMRC to assist them with regard to this evidence. The tribunal finds that either the Axxia Group chose to withhold evidence of IMEI checks or that the Group wilfully failed in 2006 to carry out or act on IMEI checks despite knowing of the importance to the Group of such checks and despite the Group’s representations to HMRC and the tribunal that checks had taken place – or a combination of both. It does not accept that there is any other probable explanation for the absence of positive specific evidence on IMEI numbers for the periods in question in these appeals. The tribunal considers that either interpretation entitles it to draw adverse conclusions against the Appellants about their failure to produce specific evidence relating to monitoring of IMEI numbers by the Axxia Group. Specifically, the Axxia Group knew or should have known that some of its customers, as well as some of its suppliers, were linked to circulating goods, but the Group made no enquiries about this.

148      With regard to the pricing software, Mr Holmes confirmed that this was not interactive software, or a way of monitoring the web, but was purely a way in which Axxia staff could record for themselves prices discussed with other parties. In other words, it was merely a notebook recording internal information. The tribunal saw a little of this in the evidence of Mr Kyri Nicolas. The tribunal finds that this was, as stated in evidence, merely a machine-based record of information collated on particular days, and no more than that. It is of little significance.

149      The evidence for the Appellants was that all its trades were fast moving transactions taking place in a competitive market in a situation where the Appellants, and the Axxia Group as a whole, could have no knowledge of the supply chains behind their immediate supplier and the sales chains beyond their immediate customer. That was, in particular, the approach taken by Kyri Nicolas who, as the tribunal finds above, was the individual directly responsible for the decisions both to buy phones and to sell them.

150      The tribunal put that view to the test by examining the findings it had made in Annex A by reference to the evidence of the pricing of goods by the Appellants and the margins realised on the deals. Annex C presents the identical information to that in Annex A, but in another order. While Annex A presents the information chronologically by reference to each Appellant, Annex C presents the same information in the alphabetical order of the phone (or other item) make and model supplied by the Appellants during the two months in question, and within supplies of the same model by descending order of price. The tribunal produced this simply by using the table sort function available as part of the standard Word software on the table in Annex A.

151      The tribunal records its findings from that presentation in Annex C, and adopts those findings into its decision. Viewed in this way, the tribunal finds patterns in the trading details that are not random and are not explained by the evidence given by the Appellants to the tribunal.

152        Witnesses for the Appellants were asked about markup and pricing. The evidence given by them suggested that the individual Appellants were working to mark up levels for the individual companies, with a separate arrangement in place under which STC was paid a consistent flat rate sum for each phone on each deal. The tribunal rejects that evidence. And it rejects the evidence that the only explanation for the pricing arrangements for individual deals was that these were what the market produced at the time. The evidence in Annexes A and C show clearly that there were consistent levels of markup for the Axxia Group as a whole that suggested that the two-stage arrangement for buying in and onselling the phones was merely the way the Axxia Group chose to present the deals. The structure of the Group and the way it worked, as found above, show no internal competition within the Group. Rather, the way in which the selling companies paid the buying company within the Group was part of the internal transfer pricing of the Group, and of no commercial significance. The tribunal finds that the two-stage element of the deals is to be ignored. The proper view to be taken of the return or markup on supplies is by reference to the Axxia Group as a whole.

153      The tribunal finds a consistent return, or margin, on supplies made through the Group, of 6 per cent on most of the straight deals. It does not accept that this was mere coincidence. The same mark up, as found in Annex C, applies regardless of the details of individual transactions and despite considerable variation in the price of what appear on the face of the paperwork to be identical products. In other words, the return to the Axxia Group was not dependent on any identifiable individual criterion of a particular deal. That finding must bear considerable weight given that the transactions in question were not the first ventures of the Appellants into the market place. They had been engaged in handling purchases and sales of mobile phones for some years and must have gained experience and market knowledge from this. Yet there was no evidence produced to the tribunal of any specialism in approach, for example, to sales of a particular model. Further, the records kept by the Appellants and the documents used in the transactions (as described in part in Annex B) do not bear out the evidence that pricing was sensitive to specific details about the way phones were boxed or other minor aspects of the models. If that was so, then the tribunal would expect the paperwork to record those details. It did not. Alternatively, the tribunal would expect the Appellants to keep internal records of such details. None were produced. 

154      Taking that evidence with the evidence of the patterns of sellers to the Group and buyers from the Group, the tribunal concludes that HMRC has shown that the deals were not market deals undertaken at arms’ length. If they were genuine market transactions as contended by the Appellants, how did the Axxia Group repeatedly secure their 6 per cent markup, no more and no less?  The tribunal finds that this cannot be explained only by the evidence offered to the tribunal by the Appellants. That evidence was of an ever-changing fast-moving market place. The tribunal considers that the evidence suggests an alternative analysis. The Appellants were able to offer bridging finance while, they thought, the goods supplied to and through them were acquired in another European state under an intra-Community acquisition so that the input tax paid by the Appellants to STC could be  reclaimed. In exchange, in most of the deals, they were given a turn of 6 per cent. That required some agreement beyond the kind of pure market driven dealing represented in the Appellants’ evidence.

155      Were the Appellants were engaged in genuine sales of goods from and into the market, or were they receiving a “going rate” for the services being offered by the Axxia Group in financing the movement of goods as part of a scheme involving others from the United Kingdom and abroad until VAT was reclaimed and a profit realised? This would explain why the chains of supplies through the Axxia Group showed a strong association between particular suppliers to the Group and particular customers of the Group. It would explain why the Group, and in particular the Appellants, could realise a consistent return for their sales. It would explain why the Appellants were not concerned about either the specific details of the phones being purchased and sold or of ensuring that the legal and commercial details of each transaction were in place so as to minimise risk. Further, the same approach applied to other parts of the chains put in evidence to the tribunal. That would explain why the chains evidenced to the tribunal allowed each of the traders participating in a chain of transactions to make a small predetermined turn out of a supply. It would explain the links seen between the suppliers “up” the chains from the Axxia Group as evidenced in Annex B. It would also explain why the Appellants did not wish to know whether the goods they were handling had been circulated before.

156      Annex B contains findings on fraud in connection with the deals examined there, and similar deals.

157      HMRC contended that there was also fraud in all other chains of transactions relevant to the deals in these appeals, both straight and contra. The tribunal has accepted and found that there was fraud in each chain relevant to the deals in Annex A. Mr Benson QC relied on the following factors in submitting that the evidence showed that the Appellants knew or should have known that all their deals were connected with fraud.

158      He first relied on the specific criminal record of Mr Andrew Nicolas, noting that he controlled the Axxia Group, and therefore the Appellants, through the relevant period. The tribunal agrees that Mr Nicolas’ criminal convictions are directly relevant to the decisions it must make in this appeal, and that it is entitled to take those convictions into account as they were properly put in evidence by analogy with section 101(1) of the Criminal Justice Act 2003. The tribunal records above both its findings with regard to the Axxia Group, to the extent to which Mr Nicolas was involved in the control and direction of the Group companies, and with regard to the weight to be attached to his evidence.

159      He drew attention to the failure of the following officers of the Appellants to give any evidence: Costas Philippou and Solomon Solomonides, and suggested that the tribunal is entitled to take that absence of evidence into account. The tribunal agrees, noting also the failure of Mr Michaelides to give any evidence. All three were officers of companies in the Axxia Group and clearly individually benefitted from this. The tribunal does not accept that their evidence would not have assisted the tribunal.

160      The tribunal has indicated above its view of the evidence tendered by Ms Leak and Mr Kyri Nicolas. It agrees with Mr Benson QC that the Appellants’ evidence from those officers and employees of the Appellants that did give evidence fell far short of full and frank disclosure on which it could safely rely.

161      Mr Benson QC drew attention to evidence put before the tribunal resulting from the judicial review proceedings with regard to transactions in May and June 2005.  A comparison of the deal sheets for those periods with the deals listed in Annex A of this decision show that most of the suppliers who supplied the Axxia Group in 2005 were not supplying it directly in 2006. Instead, they were supplying through a new intermediate broker. The tribunal accepted the schedule of such occasions put forward in the final submissions for HMRC. Those companies appear in the column headed S2 (the supplier to the immediate supplier). In most cases the tribunal notes that the supplier in S1 (the immediate supplier to the Group) was Mitek. It regards as relevant to this its finding that all supplies from Mitek were routed to one of two associated French companies.  It agrees with Mr Benson QC that this is not a random pattern or something that can readily be explained by sales in the open market of a kind of which the Appellants gave evidence. It therefore accepts that on the balance of probabilities the Appellants were aware of the link through Mitek to its previous suppliers. The tribunal examined in detail the links between the Axxia Group and Mitek in deal 3 in Annex B. Those links started in late 2005, so fitting the pattern identified by Mr Benson QC. The tribunal records in Annex B its findings about the due diligence undertaken by the Appellants in connection with Mitek. It notes, in particular, the conflict between Mr Andrew Nicolas and his consultant about why trade took place to such an extent with Mitek while it also notes the evidence as to the solvency of Mitek. This evidence as a whole confirms the weight to be put on Mr Benson’s submission.

162      He also put weight on the evidence produced by HMRC, and the failure to produce evidence by the Appellants, with regard to IMEI numbers. The tribunal examined this issue above, and agrees with Mr Benson that this is further evidence to support the HMRC contention that the Appellants knew or ought to have known of fraud connected with their transactions, given the clear evidence that they knew or had an easy means of knowing that they had handled circulating phones that had been through their own trading books before.

163      The tribunal also accepts and puts weight on two further linked submissions for HMRC. It finds, and records in Annex A, the pattern of links between particular suppliers to the Group and particular customers of the Group. It examined in Annex C the markup or return on sales across those deals. It finds that this evidence shows a pattern that is not evidence of a random, open, arms’ length, market, but something that suggests some deal or understanding of which the tribunal was given no evidence but which indicates links with the suppliers and customers otherwise that in the open market.

164      The tribunal further accepts the submission for HMRC that the large sum (£ 2 million) apparently deposited by a customer (Farouk and Suhail) as a deposit for sales that were not in evidence to have occurred suggests that the relationship with that customer was not an open market relationship. The tribunal notes from other evidence that Farouk and Suhail were one of the companies to whom the Appellants had previously supplied goods but which goods were found on NEMESIS checks to have returned to the United Kingdom through a chain of suppliers that resulted in the appellants again supplying them out of the United Kingdom. Taken together there is strong evidence in the view of the tribunal that these dealings were not ordinary market dealings and that the Appellants knew or must have known that there were connections to fraud in these dealings.

165      Another way in which the Appellants sought to persuade the tribunal that it had carried out due diligence about its suppliers and customers was by the evidence of Mr Holmes.  In the view of the tribunal the evidence of Mr Holmes attracts more weight with regard to what he was not told by the Appellants than with regard to what he told the Appellants.

166      A further element of relevance is the information given by HMRC to the appellants. The tribunal accepts that in previous trading periods the relevant officers had been satisfied with the conduct of the Appellants transactions. But HMRC had at the same time notified the Appellants of tax losses in various supply chains. The tribunal rejects Mr Andrew Nicolas’ evidence that HMRC had never given the Appellants written notice of missing traders in its chains. 

167      The tribunal must evaluate all these factors against the background of the kind of market in which the Appellants said they were operating, and the kind of market evidenced both by Mr Taylor and by the considerable evidence offered by HMRC officers in supporting the evidence of both the straight chains and the contra chains. In particular, there is repeated evidence of chains with ever-increasing prices as one supply rapidly follows another, but yet no evidence of any final customer. The supplies were routinely conducted in ways that presented legal and commercial risks, for example the evidence that payment occurred in some cases some time after the goods had moved to another jurisdiction and that paperwork was minimal.  

168      The tribunal has recorded above its finding that all the transactions subject to these appeals were transactions that resulted in tax losses to the revenue linked to fraud by parties in those chains of transactions.

169      The remaining question is whether the tribunal is satisfied that HMRC has shown it, to the required standard of proof, that the tests in Kittel, as explained in Mobilx, are satisfied. The tribunal notes the closing comments of Moses LJ, and considers that these are entirely apposite to these appeals:

            “[84]   … circumstantial evidence … will often indicate that a trader has     chosen to ignore the obvious explanation as to why he was presented with the opportunity to reap a large and predictable award over a short space of time…

            [85] … A trader who chooses to ignore circumstances which can only          reasonably be explained by virtue of the connection between his transactions     and fraudulent evasion of VAR, participates in that fraud, and, by his own         choice, deprives himself of the right to deduct input tax.”

170      The tribunal finds in conclusion that the only reasonable explanation for the conduct of the Appellants in the periods in question is that they knew that the transactions they were undertaking could secure the profit levels they realised without the need for rigorous due diligence or proper safeguarding of supplies and finances and that the reason for this is that they knew that the transactions were connected with fraud. They undertook deals at margins suggesting levels of profit without regard to the precise nature of the mobile phones being supplied, and in some cases at a markup that the tribunal finds to be agreed for the transactions on the day rather than the deal. They had the means of knowing a great deal about general fraud in the market and indeed clearly knew of it. They were dealing with phones that had been handled in fraudulent deals and they knew this or, if they had taken reasonable steps of a kind that they suggested they had taken, should have known this. They had been put on notice of the lack of integrity of some of the chains of supplies to their suppliers and from their customers and that they had the relevant information readily available to them. It is clear that the appellants’ claim to have conducted or attempted to conduct rigorous due diligence is not borne out by the documentary evidence produced to the tribunal. The tribunal has in particular not accepted the Appellants’ evidence and submissions about its inability to produce IMEI evidence while it has accepted the evidence of HMRC, as noted above, about circulating phones.

171      Above all, it is clear that those witnesses before the tribunal who had first hand knowledge of how the Axxia Group conducted its deals were less than open and honest in their evidence to the tribunal while others with such knowledge chose to give no evidence to the tribunal. On recurring occasions the evidence given by those individuals was inconsistent with the documentary evidence of the deals actually undertaken and the facts explored in the three Annexes to this decision. Many of those inconsistencies were put to the witnesses by counsel. Where they were not, the tribunal accepts the submission of Mr Benson QC that it can rely by analogy on similar evidence and similar challenges. The tribunal has indicated above why, where the documents referred to in the Annexes tell one story and the Appellants’ witnesses another, it prefers the evidence of the documents. The tribunal is satisfied that the Appellants were given the fullest opportunity in the extended hearings to make any point they wanted to make and to answer any point they wanted to answer about such inconsistencies.

172      The tribunal concludes that the test in Kittel, as explained in Mobilx, is satisfied with regard to all the deals listed in Annex A. If the Appellants did not know that the deals that the Axxia Group of which they were completely integrated parts undertook at the relevant times were connected directly with fraud, and in some cases the tribunal is satisfied that they did know this, then they should have known this because the only reasonable explanation for such ignorance was that they chose to ignore circumstances that for which the only reasonable explanation was a connection between their transactions and fraudulent evasion of VAT.

173      The appeals therefore fail.

174      The parties are at liberty to apply to the tribunal with regard to costs.

175.     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.

DAVID WILLIAMS

TRIBUNAL JUDGE

RELEASE DATE: 6 August 2010