Mobilx Ltd v HM Revenue & Customs

This is an appeal from a decision of the VAT and Duties Tribunal in the persons of Mr Colin Bishopp as Chairman and Mr Praful Davda FCA. The appeal is against the Tribunal’s order made on 20^th May 2008 whereby it dismissed the appeals of Mobilx against HMRC’s refusal to repay input tax claimed by Mobilx in its returns for April, May and June 2006.

The case is concerned with Missing Trader Intra-Community (“MTIC”) fraud. For present purposes, this type of fraud involves the importation of goods by trader A from another member state of the EU. Trader A sells the imported goods to trader B. Trader A, the importer, fails to account for the VAT due on the sale to trader B. He does so either by disappearing or by “hijacking” another innocent person’s VAT registration. A number of sales within the importer’s state then take place: for example from trader B to trader C, C to D, D to E and E to F. These intervening traders, B, C, D and E are called “buffers”. Trader F, the so-called “broker”, then exports the goods, usually to another member state. Trader F is entitled to zero-rate this sale. In the normal course Trader F would be entitled to claim back the VAT it has paid to trader E. In net terms, because the buffer transactions are approximately neutral in VAT terms, the result of trader A’s default is that HMRC are repaying VAT that they have never received.

Often fraud of this kind is perpetrated by rings, in which those orchestrating the importation and export are connected. Once the goods are placed into circulation by the fraudulent or disappearing importer, however, it is possible for them to come into the hands of innocent traders. Those who deal in goods and reclaim VAT without any knowledge, actual or constructive, that earlier in the chain there has been a default in the payment of VAT are entitled to repayment. But those who deal in goods when they knew or should have known of VAT fraud at an earlier stage are not entitled to repayment.

The leading case in this field is the decision of the ECJ in Joined Cases C-439/04 and C-440/04 Axel Kittel v Belgium; Belgium v Recolta Recycling [2006] ECR 1-6161; [2008] STC 1537 (hereafter "Kittel"). At paragraph 51, the Court states that

At paragraph 61, the Court states

Of course, an otherwise innocent trader can only do so much to ascertain whether its supply line is “clean” or “dirty” (to use the expressions used in MTIC fraud cases). It can make enquiries of its immediate supplier, including enquiries as to the diligence with which its immediate supplier checks, in turn, on its supplier. Beyond that, the immediate supplier cannot as a matter of commercial reality be expected to reveal the identity of its own suppliers without risking being cut out of the business.

In the light of the difficulties of making enquiries beyond the immediate supplier, there is a danger in reading paragraph 51 of Kittel in a narrow sense and as suggesting that provided proper checks are carried out by the trader on a supplier, then the trader’s claims to repayment of VAT are not capable of challenge. That is not, in my judgment, a correct view. Suspicious indications obtained by a trader from carrying out due diligence checks on its supplier are one, but not the only basis from which it may properly be inferred that a trader knew or should have known of its implication in VAT fraud. The test to be applied is that set out in paragraph 61 of the Judgment, and indeed in the Court’s final determination at the end of the judgment. Paragraph 51 needs to be understood in the sense that “all reasonable precautions” may, in some cases, involve ceasing to trade in specified goods in a particular market, at least in the particular manner in which the trader undertakes that trade. Such a situation may conceivably arise where, from other indications available to the trader, the trader knew or should have known that it is more likely than not that, despite all due diligence checking, any further goods traded in the same way will be implicated in VAT fraud.

Since the hearing in the present case, Lewison J has handed down judgment in two appeals concerned with MTIC fraud: Commissioners for HMRC v Livewire and Olympia [2009] EWHC 15 (Ch). I allowed both sides to supply written submissions on any points arising from that judgment. Both sides drew attention to paragraphs 123 where Lewison J is dealing with the Tribunal’s reliance by way of analogy on Section 214 of the Insolvency Act 1986 (the section which refers to the knowledge that a diligent person having both the knowledge which the director may reasonably be expected to have and the knowledge that he actually has). He said this:

HMRC (in fact their predecessors HMCE) published a detailed guidance note, Notice 726, in August 2003 concerning the checks which it recommended traders to carry out. It is concerned principally with the notion of joint and several liability (Footnote: ¹), but both parties agree that the guidance is equally applicable to the avoiding of challenges to repayment of VAT as in the present case. Mobilx and its advisors were generally aware of the guidance in the Notice.

The document contains chilling warnings about the prevalence of MTIC fraud in the mobile telephone and computer equipment (including CPU) markets. In several places the document makes it clear that the obligation on the trader is to ensure the integrity of his “supply chain”. At paragraph 4.5 the Notice recognises the difficulty of checking on the supplier’s supplier, but says nevertheless that

Paragraph 4.6 of the Notice makes it clear that HMRC are not specifying exactly what checks should be undertaken. Wisely, it points out that

Paragraph 4.9 includes the following reassurance that it is not intended to interrupt lawful trade:

Section 11 (1) of the Tribunals and Inquiries Act 1992 provides that an appeal lies to the High Court if a party “... is dissatisfied in point of law" with a decision of the VAT and Duties Tribunal.

In Georgiou v. Customs and Excise Commissioners [1996] STC 463 CA at 476, Evans LJ refers to excerpts from the speeches of Viscount Simonds and Lord Radcliffe in Edwards v. Bairstow [1956] AC 14, 14-15) and observes (at 476 f-g) that

At page 476H Evans LJ set out a four stage process for examining challenges to findings of fact:

Complete absence of evidence, or the evidence being to the contrary effect, are two of the grounds on which it may be said that a tribunal was not entitled to reach a conclusion of fact. It is also well settled that a tribunal is not entitled to find serious allegations established against a party who calls relevant witnesses unless those allegations are clearly formulated and put in cross examination. As Briggs J said in HMRC v Dempster [2008] EWHC 63 (Ch) (unreported)

For a more extensive analysis of the obligation to put such allegations see Jacob LJ’s summary in Zipher v Markem [2005]EWCA Civ 267 at [57] – [61]. The principle was not seriously in dispute here.

Subject to these very tight limitations, it is not open to the High Court to conduct a review of the evidence to see whether it would have reached the same conclusion. An appellate court is poorly placed to assess the value of oral evidence given before the Tribunal. Moreover, if the analysis of the evidence is such that reasonable judicial minds might differ on the outcome, there is no basis for saying that the decision of the tribunal of first instance is wrong.

The facts are set out extensively in the Tribunal’s decision. For the purposes of this appeal, the following is all that is necessary.

Mobilx was incorporated on 16 September 2003. Its directors at all material times were Stuart Bell and his son Steven Bell who initially also each held half of the 1000 issued shares. Stanley Bell, Stuart’s brother, was at no time a director or shareholder of Mobilx, but the firm in which he was a partner became its auditors. In 2005 Stanley Bell retired from his former firm and began to work full time for Mobilx.

Stuart Bell’s experience was in accountancy and financial services. Steven Bell had a background in the mobile phone business: he had managed a subsidiary of Carphone Warehouse Limited (“CPW”); later, he worked part-time, as a consultant, for a company called Sound Solutions Limited (“Sound Solutions”). With Steven Bell’s help, Sound Solutions had begun to trade in second-hand phones purchased from CPW.

When Mobilx was formed, it was intended that Mobilx should replace Sound Solutions as the purchaser of CPW’s used phones, and develop that business.

Mr Thompson was employed by Sound Solutions at the time when Steven Bell was engaged by it as a consultant. Mr Thompson left Sound Solutions to become an employee of Mobilx in February 2004. Each of Stuart and Steven Bell transferred a quarter of their shares to him, so that they were left with 37.5 per cent holdings, and Mr Thompson had a 25 per cent shareholding. From that time on, Steven Bell dealt with sales, Mr Thompson with purchases and Stuart Bell with administration.

Mark Hetherington, an accountant with Price Waterhouse Coopers acted for Mobilx in its relations with HMRC in VAT matters. On 15 October 2003 Mr Hetherington submitted an application for registration for VAT on Mobilx’s behalf. The application indicated that Mobilx intended to engage in the “wholesale distribution and export of mobile phones”. In his covering letter accompanying the application Mr Hetherington made it clear that the company was aware that HMCE was monitoring the trade in mobile phones because of the incidence of fraud but explained why it had decided to embark on the proposed activity, and emphasised that Steven Bell (described as the “principal director”) had considerable experience of the trade. The letter included a request that Mobilx be permitted to make monthly returns.

Monthly VAT returns are a concession allowed to assist a trader to obtain quicker repayment and hence increased cashflow. HMRC are understandably reluctant to grant this concession when they suspect that the trader is involved in MTIC fraud or in dealing in goods on which VAT has not been paid. Rapid repayment supplies liquidity into the supply chain, and increases the rate of losses of revenue.

HMCE did not immediately grant the application for registration, and a meeting was arranged to take place on 17 November 2003. It was attended by Stuart and Steven Bell, Mr Hetherington and one of his colleagues, and by two HMCE officers, one of whom was Bob Martin. Mr Martin was told of the directors’ intention that Mobilx would buy used handsets from CPW and other well-known retailers and sell them initially to customers in China and South Africa, though sales elsewhere, including within the EU, were a future possibility. Although HMCE evidently nurtured some doubts, Mobilx was registered for VAT on 27 November 2003, with effect from 1 October.

The Commissioners accept that the due diligence which Mobilx carried out into its suppliers and customers was reasonable, and better than that carried out by most other traders in mobile phones and computer components. Before the Tribunal the Commissioners did not identify, or rely on, any inadequacies in it. The due diligence Mobilx carried out, however, went no further than its immediate suppliers and customers.

Following its registration, Mobilx was not granted its requested monthly returns. Mobilx submitted its first return on 7^th January 2004, covering transactions in December 2003. There were four transactions in this period, two relating to second-hand mobile phones purchased from CPW and two to computer chips or CPUs. In the subsequent three month period the transactions related to a mixture of second hand phones and CPUs. At a meeting to discuss HMCE’s reluctance to allow monthly returns on 12^th February 2004, HMCE was clearly aware that Mobilx was trading in CPUs. It stated that these CPU transactions were under investigation.

HMCE wrote immediately after the meeting of 12^th February 2004 saying that there were “concerns” about traders in Mobilx’s chains of supply. The monthly return concession was still not forthcoming. It was explained that “tax at risk” was a vital component of the consideration of that issue. After further requests, HMCE wrote again on 28^th April refusing the concession. The letter stated that, on investigation of Mobilx’s first return for December 2003, an instance of a VAT loss in the supply chain had been identified.

Mobilx requested monthly returns again in July 2004. This time HMRC’s response was to say that HMCE were prepared to reconsider their earlier decision to refuse monthly returns.

On 14 September Mr Johnson and Mr Martin met Mr Hetherington and Stuart and Stanley Bell. Mr Martin told the Bells that HMRC had investigated the chains of transactions in the CPUs which Mobilx had bought and sold in May and June 2004. There were 24 such chains; in 17 of them a defaulter had been found and, Mr Martin said, there were concerns about four of the remaining chains. The other three were still under investigation. In those circumstances permission to make monthly returns would not be granted.

On 12 October 2004 HMCE told Stuart Bell, in a telephone conversation, that of Mobilx’s eight transactions in July, four had been fully traced, all to defaulters, and of its 17 August transactions, seven had been fully traced, all to defaulters.

On 17^th January 2005 HMCE wrote stressing the widespread abuse of the VAT system and fraud in the mobile telephone and computer chip market. The letter disavows any allegation of knowing involvement on the part of Mobilx. Nevertheless it explained that every one of Mobilx’s chains of supply that had been fully investigated for the months of May, June, July and August 2004 had led back to a defaulting trader. The letter included a table like this:

Mr Hetherington replied for Mobilx on 8 February, saying:

After more pressure from Mobilx, its request to be placed on monthly returns was granted on 3^rd June 2005. The letter granting the concession stated:

On 7 July 2005 Mr Martin of HMCE told Mobilx that HMRC (as it now was) had discovered that some of the CPUs in which Mobilx had dealt in previous months had entered the UK more than once, an indication that they had been used for fraudulent purposes. The affected CPUs were identifiable because the boxes in which they were packed bore Customs stamps. Steven Bell accepted that Mobilx had dealt in such chips, as they were offered at a reduced price and were acceptable to their customers in the United States, who re-packaged them before onward sale. He did, however, agree that Mobilx would not deal in such chips again.

The placing of Mobilx onto monthly returns did not mean that HMRC were not still investigating VAT fraud in its supply chains. On 24 August 2005 HMRC wrote to Mobilx stating that it was investigating its July 2005 return, notwithstanding the fact that “based on the evidence currently available, your claim has been repaid”. A similar letter was written in respect of the August 2005 return, with similar statements about investigations. Mobilx were never informed of any dirty chains in respect of those two months. Thus by the end of January 2006 Mobilx had not been informed of any dirty chain later than September 2004 (of which they were notified in January 2005).

At a meeting on 1^st February 2006 Mr Martin of HMRC told Mobilx that he had traced one of Mobilx’s December 2005 chains to a defaulter and that he would write again concerning his findings. Mr Martin said that the incidence of defaulting traders “may require review of measures adopted”.

On 7 February 2006 a “hijack incident” occurred. Mobilx bought a consignment of CPUs from Sound Solutions and sold them to a US customer called Smith. HMRC’s Redhill office learnt during the course of the day that the goods had previously been dealt in by someone who had, it was suspected, hijacked a genuine trader’s registration, and informed Mobilx, by telephone, of that discovery. Mobilx immediately arranged to have the goods removed from the aircraft on which they had already been loaded, and they were thereafter returned to Sound Solutions, which agreed to rescind the transaction and repay the price of the goods.

Further meetings took place between HMRC representatives and Mobilx on 15^th and 29^th March 2006. At the first meeting Mr Martin explained that he was currently undertaking checking of transaction lines to establish the presence and extent of defaulting traders within the chains in which Mobilx were involved. At the second meeting Mr Martin explained that all chains traced so far had gone back to defaulting traders. This led Stuart Bell to enquire whether they were sourced from the same supplier. Mr Martin explained the process in which he was engaged would enable Mobilx to identify suppliers where there were defaulting traders further up the chain. His note records that in providing this knowledge of defaulting chains, HMRC would expect the traders to review the relevant supplier’s due diligence and ask questions if they repeatedly got stock appearing in defaulting chains. As the Tribunal records, Mr Martin’s note also includes the following passage:

The promised details of the traced transaction chains, enabling identification of the relevant suppliers, and relevant to the January 2006 return were supplied with a letter of 7 April 2006. The information enabled Mobilx to discover that one dirty chain came through its supplier Leisure Communications Limited, one through Rapid Global Limited and the remaining chains through 21^st Century Trading Limited. The letter stated that:

HMRC claimed to have sent two further letters to Mobilx, relating to its returns for December 2005 and January 2006. But the Tribunal found that these letters were never received by Mobilx, and there is no challenge to that finding in this appeal by HMRC.

In fact, it is now accepted by Mobilx, and was accepted before the Tribunal, that all 85 transactions undertaken by Mobilx in the relevant period led back to fraudulent traders. It is not suggested that Mobilx were in any way knowingly implicated in VAT fraud. The case against them is that they should have known.

Each of Mobilx’s directors and Mr Thompson benefited to the tune of over £1 million in the year to end December 2005 from the activities described.

The Tribunal approached the issue it had to decide by asking itself the question first posed by another Tribunal in Dragon Futures (decision dated 25^th October 2006) at [75]:

The Tribunal approached its conclusions in the following way:

The Tribunal expanded on the second of these factors as follows:

As to the third factor, which the Tribunal described as “decisive”, it concluded as follows:

Mr Philip Jones QC, who argued the appeal on behalf of the Administrators of Mobilx, advanced the following two principal grounds of appeal:

That the Tribunal applied an incorrect test in law;

That the Tribunal was not entitled to reach the conclusion that Mobilx knew or should have known that if it continued to deal in CPUs its transactions were more likely than not to be tainted with fraud.

Mr Jones drew attention to the fact that the Tribunal said at paragraph [9] of its decision that the test set out in paragraph 51 of Kittel was qualified by that in paragraph 61. This, he said, led the Tribunal into error when dealing with his submissions about the steps taken by Mobilx at paragraph 98:

Mr Jones submitted that the test set out in the judgment in Kittel is a single test: and that paragraphs 51 and 61 are simply opposite sides of the same coin. Thus if the Tribunal was satisfied that Mobilx’s conduct satisfied paragraph 51, and that all reasonable precautions had been taken, it was not open to the Tribunal to come to an adverse conclusion in relation to paragraph 61.

Mr Mark Cunningham QC who appeared for HMRC submitted that the Tribunal was clear that there was only one test, as revealed by its reliance on the question posed in Dragon Futures. The test the Tribunal applied was, he submitted, the correct one.

I do not think there is anything in this ground of appeal. The Tribunal rightly focussed ultimately on the question in paragraph 61 of Kittel. For present purposes that is all that matters. In my judgment, it is clear from the Tribunal’s decision that it may have viewed paragraph 51 of Kittel as dealing only with precautions in the form of due diligence checks on the supplier, as opposed to the more drastic precaution of ceasing the type of trade in question altogether. That may be what led it to make the criticism of Mr Jones’ submission that I have set out above. Whether or not that is so, the manner in which the Tribunal approached the ultimate question is not, in my judgment, seriously open to criticism.

Mr Jones approached the second ground of appeal by attacking systematically each of the three “important factors” on which the Tribunal based its conclusion. As Mr Cunningham politely reminded me, it is advisable to subject attacks such as these to the four stage analysis in Georgiou to avoid exceeding this Court’s limited powers of redress in relation to findings of fact.

Mr Jones characterised the Tribunal’s first factor as a finding that Mobilx deliberately misled its advisors and HMRC about the nature of its intended business when it applied for VAT registration. The detailed finding is set out in paragraph 76 of the decision:

There is a difference between a lack of candour and deliberate deceit. They are at opposite ends of the spectrum of standards of good faith in business dealing. The Tribunal nevertheless regarded this as a serious allegation. Furthermore it considered that Steven Bell had given “unconvincing evidence” in his account of Mobilx’s change of trade, which made clear that the decision to trade in CPUs was not made until after the application for registration.

There was no allegation in HMRC’s Statement of Case that Mobilx knew at the date of its application that it would trade in CPUs: the allegation does refer to the decision to “involve itself in the sales of” CPUs, but that is as far as it went. There is certainly no allegation of lack of candour.

The Tribunal plainly regarded this finding as of significance. At [99] it refers back to a lengthy passage in the decision which deals with it. It is clear that it attached importance to the finding that the trade in CPUs was contrary to Mobilx’s declared intention and always had been.

Mr Steven Bell was not cross examined by HMRC’s counsel on that basis. He was asked some questions by the Tribunal Chairman at the end of his evidence, but it was not suggested to him then that he had a pre-formed intention to trade in CPUs at the date of the VAT registration, or that his company’s dealings with HMRC had lacked candour in any respect.

Mr Stuart Bell gave clear evidence in his witness statement that it was Mobilx’s intention when it applied for registration to trade in mobile phones. In paragraph 45 he explains how the trade in CPUs began as a result of a suggestion by Adam Thompson in December 2003. Stuart Bell was not cross examined at all, let alone on this evidence.

Beyond this, counsel for HMRC expressly disavowed any claim that Mobilx had misled HMRC. The only reliance placed by HMRC in this area was that Mobilx left an area of relatively “safe” trade with CPW and dived into a more risky area: the shark-infested waters of trade in CPUs. Such an allegation does not really add anything to other evidence in the case, that it was known by way of background that trade in these markets was subject to considerable risk. Indeed the Tribunal had summarised this part of the evidence by saying at [90] that:

The Tribunal plainly attached much more significance to this factor than mere factual background. Moreover Mr Cunningham relied in this appeal on these findings of lack of candour and unconvincing evidence as part of his attack on the evidence of Mr Bell, despite the fact that these had not been put to him.

The Tribunal was not in my judgment entitled to make this finding against Mobilx. It is a fundamental aspect of civil litigation that parties do not learn for the first time in a judgment or decision of serious adverse allegations against them. They must be given a proper opportunity of dealing with them before they can form a building block of any substance in the case against them.

The Tribunal’s second important factor is characterised by Mr Jones as a finding that:

I have set out the Tribunal’s reasoning in paragraphs [100] to [102] of its decision above. It seems to me that two rather different considerations were in play in this reasoning. The first is a suggestion that easy profits spell fraud, and the directors should have realised this. The second and quite different question is that whilst good money is being earned, a Nelsonian attitude to warning signals is likely to be taken. The first consideration is based on a link between the profits and illegality. The second is a suggestion that the directors simply shut their eyes to the objective evidence, such as warnings from HMRC. In my judgment the Tribunal was relying on both of these considerations.

That the Tribunal was relying on the first of these suggestions is clear from the question it suggested the directors should have asked and answered, namely “how it was possible to make so much money so easily”?

It was not part of HMRC’s pleaded case that easy profits were a badge of fraud. Indeed Mobilx’s accounts were only admitted into the case at a late stage, and for an entirely different purpose.

It was never suggested to any witness before the Tribunal that easy profits were a badge of fraud. The cross examination in relation to benefits and profits was limited to a suggestion that Steven Bell was motivated by greed rather than the long term building of a company. Furthermore it was not expressly suggested to him that the amounts of money being earned had led him to ignore the warning signs, or to fail to investigate his sources of supply.

There was again no cross examination of Stuart Bell, the director with responsibility for financial matters: he was not cross examined at all. The Tribunal recognised this difficulty at paragraph [101]. In those circumstances it is unfortunate that in paragraph [102] the Tribunal attributes to Stuart Bell a comment of Stanley Bell, from which it seems to draw support for its conclusions – including a conclusion, or at least a suspicion, that Stuart Bell was swept along by Steven and Adam Thompson. This was simply an error on the part of the Tribunal, but it is indicative of the importance of putting matters of this nature in cross examination to the relevant witness. If that had been done, the Tribunal would have been less likely to have made this mistake.

I do not consider that the Tribunal was entitled to reach the conclusion which it did on this second important factor. Again this was a serious accusation which the company had no proper opportunity to answer. One of the underlying propositions, namely that profitable business is necessarily fraudulent, was simply not established.

The alternative suggestion that high profits had led the company to take its eye off the ball was also not put to any witness. Moreover, as Livewire and Olympia makes clear, if the information available to a company is such that a competent director would have made further enquiries or stopped trading altogether, I am not sure how it helps to establish that the actual directors shut their eyes. On the other hand if the available information was not sufficient to prompt the competent director in that way, the fact that the directors shut their eyes will not put the company on notice.

I turn therefore to the third important factor. This is characterised in the Grounds of Appeal as a finding that Mobilx ought to have known that every transaction in which it engaged led back to a defaulter.

The Tribunal in fact characterises the third factor as “the director’s response to the fact that every traced transaction led back to a defaulter,” but I agree with Mr Jones that the import of this finding is as he describes it. It is clear from the Tribunal’s analysis which follows that it considered that the reasonable and proportionate response to the information Mobilx was receiving concerning the dirty chains was to cease trading, or at least cease trading in that way. This was plainly a highly material finding. The Tribunal describes it as decisive.

Mr Jones does not challenge the Tribunal’s finding at paragraph [106] that every one of the chains of transactions in CPUs which HMRC had been able to trace by the end of March 2006 led back to a defaulter: indeed this is a finding that was recorded to have been made on the Administrators’ concession. Nevertheless he says that the Tribunal could not on the evidence properly come to the conclusion that a reasonable and proportionate response would have been to cease to trade.

Mr Jones’ principal point is that the suggestion that Mobilx should have ceased trade, or radically modified its trading method, is glaringly inconsistent with the attitude of HMRC at the time, whose advice, even at the meeting of 29^th March 2006, two days before the relevant period, was nothing like as draconian. He draws attention to the long periods when Mobilx were not notified of any dirty chains, and HMRC’s grant of the concession of monthly returns, a decision to which the extent of tax risk is highly material. In support of these submissions he relies on the fact that if the case of constructive knowledge is to be supported, it must be supported on the basis of information given by HMRC to Mobilx, as no reliance on other factors apparent to Mobilx (as is sometimes available) was pursued by HMRC. However he submits that the Tribunal failed to engage with the submissions and evidence relied on by the Administrators in this respect.

Mr Jones backs this up these submissions by saying that the person who in fact attended the meetings with HMRC was Stuart and not Steven Bell. Stuart Bell’s evidence about the meetings was therefore unchallenged.

Stuart Bell’s unchallenged evidence dealt with the correspondence with HMRC as well. He said, for example, in relation to the letter granting Mobilx monthly VAT returns in June 2005:

The Tribunal was faced with weighing factors which pointed in opposite directions. The bedrock of HMRC’s case was constituted by the communications it sent to Mobilx. These were

the communications in February 2004, September 2004, October 2004, January 2005 concerning the fact that all chains investigated led back to defaulters;

the further dirty chain notified at the meeting in February 2006;

the notification at the meeting on 29^th March 2006 that all chains so far investigated led back to defaulting suppliers.

Against this, the Tribunal was aware that HMRC had continued to pay Mobilx’s claims, and in June 2005 placed Mobilx on monthly returns. The Tribunal accepted that Mobilx’s directors took comfort from these matters: see paragraph [98] of the decision. In paragraph [108] the Tribunal concluded that these matters did not excuse Mobilx from compliance with the Kittel test.

I cannot in the end see any error in the manner in which the Tribunal approached this question. It is true that Stuart Bell was not cross examined, and that he was the party dealing with HMRC. But the question in the end for the Tribunal was what Mobilx should have known on the basis of the information supplied to them. I think the Tribunal was entitled to find from the evidence before it that Mobilx should have known that on the balance of probabilities all its transactions were leading back to defaulting traders.

Mr Jones argued that the three important factors relied on by the Tribunal were cumulative, and that once any of the factors fall away then the conclusion must go as well. I accept that submission to the extent that I believe it is right that this Court is entitled to come to its own conclusion on the evidence which was before the Tribunal.

Notwithstanding my decision about the first two factors, I have come to the conclusion that the evidence before the Tribunal did establish on the balance of probabilities that Mobilx should have realised based on objective factors that its transactions were connected with fraud.

Firstly, I have already indicated that I consider that the Tribunal was entitled to come to the conclusion that Mobilx should have realised that all of its chains were likely to lead back to defaulting traders, unless it ceased trading or significantly changed its manner of doing trade. It is clear that the Tribunal regarded this finding as taking HMRC most of, if not all, the way home. I agree.

Secondly, there was further evidence before the Tribunal. Although I have rejected the unjustified (and in my view unnecessary) finding that Mobilx were less than candid with HMRC about going into the CPU business, the evidence clearly established that both the mobile phone and the CPU businesses were ones in which MTIC fraud was rife. It seems to me that the relevant point here was not so much what Mobilx told HMRC, but what Mobilx were actually doing. There is ample evidence on which to conclude that Mobilx was well aware that the business it was in was one where it is easy to become involved in MTIC fraud. Against this background, the fact that all its transactions were leading back to defaulters should have alerted a competent company to the fact that its trade was the result of fraud.

Thirdly, the information available to Mobilx, and of which it was established that it was aware, included information that traders are expected to verify the integrity of their supply chains. Due diligence on one’s immediate buffer supplier may be all that one can do. But if, despite due diligence on the immediate supplier, chains are being identified as dirty, more drastic action is required. A reasonable and proportionate response by April 2006 at the latest was either radically to alter the method of trading or get out of it altogether. Mobilx did not do this.

Fourthly, there was ample evidence to show that the way in which Mobilx was carrying on its trade was not protecting it from becoming implicated in dirty chains. Checks on suppliers were not doing the trick. So the fraud was not just present in the business as a whole: it was affecting the very supply chains Mobilx was actually using.

It is true, as the Tribunal accepted, that the directors took comfort from the actions of HMRC. But the company has to exercise independent judgment, not delegate its judgment to HMRC. I agree entirely with the Tribunal when it said that “there must come a time when a trader, told that every one of his purchases followed a tainted chain, is compelled to recognise that without a significant change in his trading methods every one of his future purchases is more likely than not also to follow a tainted chain”. The trader is not entitled, when that point has been reached, to wait for HMRC to tell him to cease to trade. Moreover, as the Notice explained, HMRC’s advice is not intended to create a shield for fraud.

Putting all these matters together, I have reached the same conclusion as the Tribunal, although I have done so by a more direct and different route. Mobilx should have known that all its transactions were more likely than not to be implicated in MTIC fraud. The appeal must be dismissed.