IN THE COURT OF COURT OF APPEAL (CRIMINAL DIVISION)
ON APPEAL FROM THE CROWN COURT AT NORTHAMPTON
Royal Courts of Justice
Strand, London, WC2A 2LL
Before :
LORD JUSTICE RICHARDS
MR JUSTICE IRWIN
and
HIS HONOUR JUDGE BEVAN QC
Between :
The Queen | Respondent |
- and - | |
Peter Pomfrett | Appellant |
Nicholas Purnell QC and Joseph Templeton (instructed by Bankside Law) for the Appellant
Sir Derek Spencer, Mark Bryant-Heron and Rebecca Chalkley (instructed by Solicitor to HM Revenue & Customs) for the Respondent
Hearing dates : 28-30 July 2009
Judgment
Lord Justice Richards :
Peter Pomfrett appeals against his conviction on 23 January 2008, after a 3½ month trial at Northampton Crown Court before His Honour Judge Ian Alexander QC and a jury, on a count of conspiracy to cheat the public revenue. The alleged conspiracy concerned a chain or carousel fraud, referred to as a missing trader intra-Community fraud (or “MTIC” fraud), whereby Her Majesty’s Revenue and Customs (“HMRC”) was cheated out of some £25 million of VAT. It was one of a number of frauds and related money laundering offences investigated by HMRC in the period 2002 to 2006 under code-names such as “Operation Vitric” and “Operation Devout”. The prosecution in this case arose out of Operation Devout II. At the forefront of the appeal is an issue of non-disclosure. The prosecution admit that they failed to disclose a large body of material concerning Operation Vitric which ought to have been disclosed for the purposes of the appellant’s trial. The question is whether that material, now available to this court, casts doubt on the safety of the appellant’s conviction. That issue has to be considered together with a number of grounds of appeal relating to the fairness of the judge’s summing up and of the trial process. Leave to appeal in respect of the grounds other than non-disclosure was granted by the single judge. The issue of non-disclosure emerged in its present form at a later stage and is the subject of supplemental grounds for which leave was granted by the full court.
In the event that the appeal against conviction fails, there is a renewed application for leave to appeal against sentence.
The alleged conspiracy and the trial: overview
A chain or carousel fraud of the kind that occurred in this case involves a series of transactions (or nominal transactions) starting with the import of goods into the United Kingdom from a supplier within the European Union, followed by onward supplies through a number of “buffer” traders within the United Kingdom and ending with the re-export of the goods from the United Kingdom. The import and export transactions are zero-rated for VAT, whereas the onward supplies within the United Kingdom are subject to VAT. The importer, however, is a missing or “hijacked” trader who does not account to HMRC for the VAT payable on the supply to the next trader in the chain. The exporter pays VAT on the goods supplied to it by the previous trader in the chain; but since the export is zero-rated, the exporter is able to claim back from HMRC the VAT it has paid. The overall effect is that HMRC incurs a loss, whereas the fiscal position would have been neutral if the trading had been genuine and the VAT had been accounted for properly.
In this case the goods were computer processing units (“CPUs”). The nominal importer was a firm by the name of PJ Spencer, a company which had its identity hijacked for the purposes of the fraud.
The first line buffer was Aurum Jewellery Wholesale Limited (“Aurum”). A man by the name of Pravin Jogia was a director of Aurum. Between 25 June and 19 July 2002 Aurum purportedly made 56 purchases of CPUs from PJ Spencer, at a total price of £169.2 million and a VAT charge of over £25.2 million. The invoices were a sham and the VAT was not accounted for to HMRC.
The second line buffers were Lightcare Limited (“Lightcare”) and Vendon (UK) Limited (“Vendon”). Jonah Adali-Mortty was a director of Lightcare, and Arash Masoumzadeh was a director of Vendon. Aurum’s records showed 51 sales to Lightcare and 3 sales to Vendon, at a total price of £169.5 million.
The third line buffers were Globalactive Technologies Limited (“Globalactive”) and Beronvine Limited (“Beronvine”). The appellant was a director and company secretary of Globalactive; Mustafa Mehmet was another director of the company; and Timur Mehmet, the nephew of Mustafa, was an employee of the company. Globalactive is considered in greater detail below. Babak Cherazi was a director of Beronvine; Khalid Hamidi took over the role of company secretary of Beronvine on about 11 July 2002. According to the invoices found, Lightcare made 46 sales to Globalactive at a total price of £135.6 million (with a VAT charge of £20.2 million) and 4 sales to Beronvine at a total price of £13 million (with a VAT charge of £1.9 million); and Vendon made 3 sales to Globalactive at a total price of £14 million (with a VAT charge of £2 million).
Of the individuals identified above, Jogia pleaded guilty at an early stage to the conspiracy alleged. MustafaMehmetchanged his plea to guilty at the beginning of the trial. Masoumzadeh was outside the jurisdiction, in Dubai, and was not tried. Adali-Mortty, the appellant, Timur Mehmet, Hamidi and Cherazi faced trial on an indictment alleging, in the particulars of the offence charged, that “between the 1st April 2002 and the 23rd July 2002 … [they] conspired together and with Pravin Jogia, Mustafa Mehmet, Arash Masoumzadeh and other persons to cheat Her Majesty the Queen and The Commissioners of public revenue, namely monies being or purporting to be Value Added Tax payable …”. Timur Mehmet did not answer to his bail at the commencement of the trial and was tried in his absence but was represented throughout by counsel. The remaining four defendants were present and represented at the trial. All those who were tried were found guilty by the jury.
On 28 January 2008 the appellant was sentenced to 10 years’ imprisonment (with 4 days in custody on remand to count towards sentence) and was disqualified under section 2 of the Company Directors Disqualification Act 1986 for 14 years. Adali-Mortty was sentenced to 6 years’ imprisonment and disqualified for 10 years. Mustafa Mehmet was sentenced to 4 years’ imprisonment and disqualified for 10 years. Timur Mehmet was sentenced in his absence to 8 years’ imprisonment and disqualified for 12 years. Hamidi and Cherazi were each sentenced to 4 years’ imprisonment and disqualified for 8 years. Sentencing of Jogia was adjourned because he had recently had a serious medical operation, and we do not have details of the sentence later imposed on him.
The case against and for the appellant at trial
Although the indictment referred to the period from 1 April to 23 July 2002, the prosecution put its case at trial on the basis of a conspiracy to carry out a dishonest scheme during 20 days of trading between 25 June and 22 July 2002. The entire trade was said to have been rigged.
The position of each of the defendants at trial, and what the prosecution said about them, was explained as follows in the prosecution’s opening note:
“3. The defendants do not dispute there was a conspiracy to cheat. The defendants Adali-Mortty, Timur Mehmet and Babak Cherazi admit they did acts which furthered the conspiracy because they carried out the deals. But they say they did not know of any VAT fraud and acted honestly. They say they are the victims of other people’s dishonesty. They were used or hoodwinked by others who were dishonest, such as Jogia and Mustafa Mehmet. In the case of any defendant where that may be so you will acquit. The prosecution say these are lying defences, each of these defendants was dishonest because he too had agreed to help what he knew was a VAT fraud. The central issue in the case, therefore, is whether any of these defendants had a dishonest state of mind. You will decide that by looking at all the evidence including what they did or did not do or said.
4. In the case of Peter Pomfrett he denies any knowledge of the fraud and participation in the deals. If that might be so you will acquit. The prosecution contend that this is also a lying defence because he agreed to and did assist what he knew to be fraud by financing it, supervising it and receiving a substantial dishonest profit. In the case of Hamidi he denies being involved in any way whatsoever. If that might be so you will acquit. The prosecution say this is a lying defence ….
49. … The prosecution contend that the only sensible inference from all the evidence is that each defendant who was involved in the day to day operation of Aurum, Lightcare, Globalactive and Beronvine must have been aware that they were participating in an MTIC carousel fraud because such a fraud requires that movements and pricing of goods, and transfers of money are carefully synchronised and that each party has a pre-ordained and agreed role to achieve the dishonest objective. The goods have to be sold to the right people at the right price at the right time ….”
Thus the appellant’s essential defence was a denial of knowing participation in the fraud. He said that he was the innocent victim of the dishonesty of others, who had used Globalactive without his knowledge for the purposes of their fraud; and in particular, he had been duped by Mustafa and Timur Mehmet.
On the evidence at trial the genesis of the fraud charged in the indictment was obscure, though the suggestion before the jury was that it originated in, or had its epicentre in, Dubai. Certain information about this had been provided to the prosecution by Jogia after he entered his plea of guilty. That information was disclosed to the other defendants but very little of it could be deployed in evidence in the absence of Jogia as a witness. Jogia had said that he used to deal in the jewellery trade with a man called Raffi, whose companies included Masheikh Commercial Brokerage (“Masheikh”) and Emirates European General Trading in Dubai. Around March/April 2002 Raffi approached him to act, in return for a small commission, as a go-between in respect of deals with PJ Spencer and Globalactive, as Raffi did not want those two companies to meet. Jogia agreed. In May 2002 Raffi faxed him various documents, including drafts of letters to be sent to PJ Spencer and to Globalactive. Raffi also asked him to book flight tickets to Dubai for a man called Chopra and others involved in the fraud: Chopra had a hotel in Singapore but all his money was in Dubai. According to Jogia, the money from the fraud went to Raffi’s companies and was then redistributed to the others involved.
The documents sent by Raffi to Jogia were in evidence before the jury, and the fax details showed that they had been sent from Masheikh in Dubai. This invited an inference that someone else, in Dubai, was behind the fraud, but what Jogia said about it, including the involvement of Raffi, Chopra and others, was not in evidence.
Also in evidence was a letter of introduction written by Jogia to Globalactive on 7 May 2002, in terms corresponding to the draft sent to Jogia. Entries in Globalactive’s cashbook provided evidence of two payments of £667.20 to Emirates on 7 May for travel: and the appellant said in evidence that Timur Mehmet had visited Dubai that day.
On receipt of Jogia’s letter of introduction of 7 May, the appellant wrote a reply on the same day, thanking Jogia for the letter, indicating a willingness to trade with Aurum, and asking for a trade reference and for clarification of Aurum’s trade classification. Globalactive subsequently sent completed purchase orders to Aurum, but there was no evidence that the trading between them went ahead. During the period of trading relied on by the prosecution at trial, sales by Aurum were made to Lightcare and Vendon as second line buffers, and then on to Globalactive and Beronvine as third line buffers, rather than direct from Aurum to Globalactive.
There was evidence from a Customs & Excise officer, Mr Davies, that he had attended the offices of Globalactive on 18 April 2002 and was informed that the company had set up a practice of contacting the Redhill Business Centre to validate VAT registration numbers of potential new suppliers. The appellant confirmed to him that the company would continue to do so. The appellant’s evidence was that Aurum’s classification as a jewellery business prompted him to query, in his reply to Jogia’s letter of 7 May 2002, whether Aurum was allowed to trade in computer hardware. A note was subsequently sent to Globalactive by a VAT specialist (marked for the attention of Mustafa Mehmet but seen by the appellant) confirming Aurum’s VAT registration and that it could deal in any goods. But the appellant had also sent a request to the Redhill Business Centre for confirmation that Aurum was a bona fide company registered for VAT purposes; and he followed this up by sending them a copy of the VAT specialist’s note. He did not receive a satisfactory response: the Redhill Business Centre in fact referred Aurum to a local customs officer to check why a jeweller was trading in CPUs. The appellant said that because his concerns about Aurum were not satisfied he gave instructions that no-one at Globalactive was to trade with Aurum without his consent. The defence relied on these matters as evidence of the appellant’s innocence. The prosecution’s case was that as a knowing participant in the fraud the appellant was checking whether it was safe for Globalactive to deal with Aurum.
Another element in the prosecution case was evidence from a representative of Barclays Bank that sudden activity, with turnover bigger than expected in the Globalactive bank account, gave rise to a concern that the company was being used for a VAT fraud, and the bank decided for commercial reasons to give notice of closure on 16 May for 15 June. The appellant wrote to the bank to complain about the notice of closure, and following negotiations an extension was granted to 23 July. There was evidence that the appellant took steps to obtain an alternative banking facility and managed to secure the opening of a bank account at Bank of Scotland with effect from 9 August 2002. Globalactive, however, ceased to trade after 27 July.
As a major plank of the prosecution case, the jury were taken through a detailed schedule of the various chains of transactions entered into by the companies concerned, including the payments made in respect of them (many of which were made to third parties rather than to the suppliers of the goods), to show how the fraud operated in practice.
When it came to the defence case, the appellant gave evidence and called a number of witnesses. He said that he played very little part in Globalactive and did not know that the Mehmets were using the company for VAT fraud. He explained how he became involved with the company. He had regretted not taking up a previous opportunity to become involved in a computer business. He knew Timur and family because his sons and Timur went to the same school. Timur had originally sought funding for a company to refurbish phones, which was not successful. Timur had returned in 2001 with a fresh proposal for Globalactive, which had been set up by Mustafa Mehmet. They needed capital to fund the VAT element in export sales before the VAT could be reclaimed from HMRC. At the time the appellant’s funds were tied up, but when he sold his country club business his position changed. He visited two companies (IT Wholesale and DVD Trading) and was impressed, so he agreed to become involved in Globalactive. He became a director and company secretary of the company, brought in his own accountants and communicated with the bank, informing them that he would be dealing with the company and that he was proposing to invest a considerable sum to enable trading to commence. The business was to be the buying and selling of computer hardware both at home and overseas. On 19 March 2002 he put £500,000 into the company and obtained 51% of the shares: the £500,000 was only intended to be a short-term loan.
According to the appellant, Globalactive had been looking for offices and it made sense for it to use the offices, and the secretarial and accountancy services, of the golf club business. He became sole bank signatory in order to safeguard finances. Timur requested online banking facilities in order to make payment instantly in a fluctuating business. The accountancy and bookkeeping were done by the appellant’s staff, whom he trusted. Checking procedures were carried out as requested by Customs & Excise. He tracked what was happening financially by asking for preparation of a deal sheet for a given period. He looked at cash flow occasionally, knowing it was important. He did not get involved in the trading side. He understood that Timur had the goods sold before they were purchased because of a wish not to be left with unsold stock. The appellant himself was out of the United Kingdom regularly twice a month and was planning retirement. Although the business was dealing in very large sums of money, he had experience and trusted the bookkeepers. He did not involve himself in the day to day business but was busy increasing the value of his remaining assets. He normally attended the office at about 8 a.m., and Mustafa Mehmet arrived at about 10 a.m., around which time the appellant tended to leave. He could see that trading was hectic and non-stop, and he was unable to conduct his own business at the same time because there were too many distractions. If he returned to the office it was not until after the banks closed. He looked at the cash flow at the end of the trading day: Timur and Mustafa showed him what the company owed and what was owing to it.
The appellant said that from an early stage Customs & Excise had asked him to check and verify the people the company was purchasing from. Unless he checked and it was verified he would not allow them to trade. He was aware of, and was concerned about, missing traders. He gave instructions for every enquiry to go through the system.
The prosecution case had included evidence of the payment to Globalactive of two sums of about £1 million each, on 17 April and 10 May respectively, from the client account of the company’s solicitors, Tolhurst Fisher. In his evidence in chief the appellant said that the money was from a contact of Timur’s but he gave no details. He was cross-examined extensively about this. The relevant files had been in the possession of Tolhurst Fisher and the defence had had difficulty in obtaining them, so that they became available to the appellant only shortly before he gave evidence. Asked at the outset of the cross-examination who the contact of Timur’s was, he said that he had now had an opportunity to look at the files and it was an investment company called Atlantic (it was in fact Atlantic Custodian SA; we will refer to it as “Atlantic Custodian”). He said he did not know who controlled Atlantic Custodian and he could not recall any other details about it, such as whether it was a limited company or where it was registered. At a meeting also attended by Timur and, he believed, arranged by Timur, he met two men who were either agents of or worked for Atlantic Custodian. He did not recall their names or where precisely they met, though it was in a hotel in the City or West End. He thought that one of the men was possibly American and the other Asian but he could not give further details about them. The meeting resulted in an oral agreement that Atlantic Custodian would advance an initial sum of £1 million to Globalactive, with the possibility of calling down further funds if Globalactive required further investment. It was left to the parties’ solicitors to draw up a written agreement.
Later, after prosecuting counsel had had a chance to look at the Tolhurst Fisher file, there was further extensive cross-examination about the documents. They showed the process of finalising a Co-operation Agreement between Atlantic Custodian and Globalactive. Instructions were being given to Tolhurst Fisher by the appellant, with no evidence of any involvement by Timur Mehmet. A lawyer by the name of David Post in Virginia, USA, was acting for Atlantic Custodian. At one point the appellant requested that a copy of the draft Agreement showing the amendments proposed by Tolhurst Fisher be faxed to a Mr Davey at a Dubai number. The Agreement in its final form provided for the parties to undertake “the Project” (namely the trading of electrical and electronic goods and components and computer parts and peripherals) for one year, with Atlantic Custodian to pay Globalactive the initial funding of £1 million on signature, and Globalactive to pay Atlantic Custodian 25% of the gross profit on the trading. The Agreement was signed on behalf of Globalactive by the appellant and Mustafa Mehmet. The first payment of £1 million was transferred to Tolhurst Fisher soon afterwards. A second payment of about £1 million followed later. Repayments were made just before the closure of Globalactive’s bank account at Barclays Bank. It was put to the appellant that he was using Atlantic Custodian as a means of investing his own monies to finance Globalactive’s part in the fraud.
Another area of cross-examination concerned large payments that had been made by Globalactive to a company called HIFX Limited in Canada. It was put to the appellant that they were a means of extracting some of the fruits of the fraud into a foreign jurisdiction.
More generally, there was lengthy cross-examination about the nature and extent of the appellant’s involvement in, and knowledge of, Globalactive’s day to day operations.
In addition to the appellant’s own evidence, he adduced supporting evidence from his architect (who gave evidence about their projects together and the appellant’s pattern of work) and from two long-serving employees, namely Theresa Calvey and Sandra Dee (who both gave evidence about the system in the Globalactive office and the nature of the work carried out by the appellant on the one hand and by Mustafa and Timur Mehmet on the other), in addition to character evidence. The appellant had said that a third employee, Christine Killigrew, would also be called as a witness, but in the event this did not happen.
By the end of the trial the prosecution was able to summarise the appellant’s defence as follows (and the fairness of the summary was not disputed by Mr Purnell). At Timur Mehmet’s invitation the appellant made a short-term loan of £500,000 to Globalactive, and became the sole bank signatory and majority shareholder to protect his investment. The loan was soon repaid. Timur arranged further finance with Atlantic Custodian. It was really Timur’s company; he controlled it; he was the driving force behind it. The appellant had other business interests. He set up procedures including verification with Redhill. This included Aurum. He was not satisfied with the results of the verification of Aurum and told Timur and Mustafa Mehmet not to trade with it. He supplied premises and three staff who authorised online payments. He spent only 10 minutes or half an hour a day on Globalactive’s business. The trading room was manic so he would leave. He was really only concerned with the cash flow. On a daily basis he saw the amounts owed and owing at the end of the day, and he saw the deal sheets monthly. His profit was about £120,000. He believed it was honest trading. He trusted the Mehmets and when he heard Mustafa Mehmet had pleaded guilty he was disappointed and shocked.
The issue for the jury was whether the appellant’s evidence, supported by the witnesses called on his behalf, was or might be truthful.
In his closing speech, prosecuting counsel made various general points, including the need for the jury to consider such matters as the machinery of the conspiracy, how events dovetailed and locked together, what each defendant did or said, who benefited from the fraud (“Following the money is a sound principle in these cases because they are committed for money”), whether a defendant had lied on oath, whether two or more defendants had cooperated in the way they conducted their defences in court, and, in the case of the appellant, whether he had got witnesses to come to lie on his behalf.
It was submitted that the appellant and his witnesses had lied and that the only sensible inference from a number of matters taken together was that the appellant was a knowing party to the cheat, together with Mustafa and Timur Mehmet. The matters relied on were as follows. First, what the appellant did: (i) his 51% controlling shareholding; (ii) he was sole bank signatory; (iii) he had arranged for his companies’ accountant to be Globalactive’s accountant; (iv) he had arranged for his companies’ solicitors, Tolhurst Fisher, to act for Globalactive in connection with the Atlantic Custodian agreement; (v) he wrote to the VAT registration unit to effect speedy VAT repayments; (vi) he signed most of the VAT returns; (vii) he wrote five letters to HMRC about the VAT affairs of Globalactive in 2003; (viii) he dealt with Barclays, especially to extend the life of the account; (ix) he tried to obtain banking elsewhere and succeeded in securing further banking with Bank of Scotland; (x) he gave instructions to Tolhurst Fisher in connection with the written agreement with Atlantic Custodian; (xi) he gave instructions for the written agreement with Atlantic Custodian to be sent to Mr Davey in Dubai; (xii) he signed the Atlantic Custodian agreement; (xiii) he took the lead at the meeting with the Customs & Excise officer Mr Davies on 18 April 2002; (xiv) he supplied the accommodation for Globalactive; (xv) he supplied his own long-standing and loyal staff to assist in the trading and administration; (xvi) he authorised loans to IT Wholesale totalling £1.4 million with a view to them being repaid when Globalactive got another bank account; (xvii) he answered the letter from Jogia on 7 May 2002 written to implement the fraud blueprint received from Dubai; (xviii) he initiated the verification of the Aurum VAT number; (xix) he wrote the memo that there was to be no trading with Aurum; and (xx) he wrote letters to customers about damaged goods.
The prosecution relied further on the way he gave evidence, referring to his three refrains of “I can’t remember”, “it was Timur Mehmet” and “I was not there”. It was submitted inter alia that he had been unable to give a convincing reason why he loaned £500,000 with no written agreement to a company that had never traded, to engage in trade he knew nothing about; that he and Theresa Calvey had lied in evidence about the role of Timur Mehmet; that he was a successful property developer and financial sophisticate who liked to have hands-on dealings with all his companies; that he lied about the circumstances of the £2 million loan from Atlantic Custodian, which was said to be the perfect arrangement for a fraudulent venture (though it was not possible to tell whether the appellant had laundered the proceeds of sale of the country club through Atlantic Custodian or whether the money had been provided by other fraudsters); that he lied about his lack of any connection with Dubai, as shown by his request for the Atlantic Custodian agreement to be faxed to Davey in Dubai; that he lied about the profit made by Globalactive; that he lied about the atmosphere in the trading room; that he and his witnesses lied about the level of financial supervision he gave, about the role they played in the bank payment process, and in saying that they did not pay any attention to the payment instructions from Lightcare and Vendon.
The issue for the jury was put very clearly by the judge in his summing up: the appellant’s criticisms of the summing up, which we consider later, are not directed to this point. The jury’s verdict shows that they were sure that the appellant lied in evidence and that he was a knowing participant in the fraud.
The nature of the undisclosed material
As we have said, the prosecution accepts that it ought to have disclosed but failed to disclose to the defendants a body of material relating to Operation Vitric. The Operation Vitric fraud was a multi-handed MTIC fraud causing a VAT loss of £100 million.. The proceedings relating to it were stayed on 18 May 2005 as an abuse of process on grounds of non-disclosure of documents. The indictment charged a man by the name of Eric Rachar with being a party to the main conspiracy to cheat, alternatively with money laundering. This was based on the fact that seven of his companies received third party payments (i.e. payments to persons other than the suppliers of the goods) throughout the life of the fraud, amounting in total to over £40 million.
There were three stages to the Operation Vitric fraud:
Stage 1 ran from 30 November 2001 to 19 April 2002. A company called Tradex, acting as a first line and then a second line buffer, bought directly or indirectly from nine missing traders and sold to fourteen companies, including sales of £9.4 million to Globalactive in the period 8-19 April 2002 (this was virtually the first trading carried out by Globalactive). Tradex made third party payments to Rachar companies totalling £29 million.
Stage 2 ran from 22 April to 17 May 2002. A company called Cytech bought from a single missing trader and sold to seven companies, including sales of £56 million to Globalactive. Cytech made third party payments to Rachar companies totalling £2.7 million.
Stage 3 ran from 21 May to 27 June 2002. A company called Rage bought indirectly from two missing traders and sold to nine companies, but none of those sales was to Globalactive. Rage made third party payments to Rachar companies totalling £10.5 million.
Much of the money paid by Tradex, Cytech and Rage to Rachar companies was withdrawn by Rachar in cash and dispersed through a web of companies. The case against Rachar was based on repeated money laundering.
The Operation Vitric fraud was followed by the Operation Devout frauds, which themselves fell into two phases. Operation Devout I was a fraud executed between 12 April and 30 April 2002, causing a VAT loss of £12.5 million. The machinery of the fraud involved the extraction of part of the VAT by third party payments to eleven money laundering accounts of persons who were not involved in CPU trading. There was no material showing any involvement of Rachar in this fraud. Jogia was alleged to have been involved in it, and was arrested on 22 July 2002 in the Devout I operation because of his laundering of £2 million through the account of Aurum. As a result of Jogia’s plea of guilty to participation in the Devout II fraud and his provision of information to the prosecution, the case against him in relation to Devout I was not proceeded with.
Operation Devout II related, as we have indicated, to the fraud charged in the trial of the appellant and his co-defendants. The fraud ended on 22 July 2002 because of the arrest of Jogia on that day in the Devout I operation. There was some evidence of involvement by Rachar in Devout II, including the making of payments to a company, Camross, owned by him, but he was not proceeded against as a defendant in the Devout II trial because, on the information available at the time to prosecuting counsel, he was (wrongly) regarded as a minnow in the context of Devout II.
The parties agreed, for the purposes of this appeal, a statement of facts arising from the Operation Vitric documents which should have been, but were not, disclosed to the appellant for the purposes of his trial. We do not need to set out in detail the 131 facts so agreed. In summary:
The first section of the statement relates to Rachar and his companies and includes the fact that he was a target in the Operation Vitric investigation and was under covert surveillance from May to June 2002, and that he was the director of companies called LaSale Ltd (the registered address of which was his home in London W2), Camross Ltd, Danakil Ltd, Midthorpe Ltd, Tantivy Ltd, Infrasonique Ltd, City & London Ltd, and Managing Agents Ltd.
The next section concerns Rachar’s partner or wife, Jane Campbell, who lived with him in London, also had an address in Brisbane, Australia, and was the company secretary of LaSale, Camross, Danakil, Midthorpe, Tantivy and Infrasonique (which changed its name to City & London International Ltd).
There follow two sections concerning the relationship between Rachar, David Post, Sun Trust (a bank in Miami, Florida, which held an account in the name of David Post), Atlantic Custodian and IMS. They show in effect that, inter alia, Atlantic Custodian was controlled by Rachar through IMS (which was the corporate director of Atlantic Custodian and was itself owned by Rachar, with Jane Campbell as company secretary), and that the two payments of £1 million or thereabouts made by Atlantic Custodian to the Tolhurst Fisher client account and thereby to Globalactive were made on Rachar’s instructions and funded by payments from Rachar’s other companies: the mechanics involved payments into David Post’s Sun Trust account and the giving of instructions by David Post to Sun Trust for payments to Tolhurst Fisher. There are also references to the involvement of a solicitor by the name of Bernard Halpin in the affairs of Atlantic Custodian. The payments made by Atlantic Custodian to Tolhurst Fisher are touched on further in some later sections.
A section concerning the invoices and payments between companies involved in the Operation Vitric fraud shows numerous third party payments, along the same lines as subsequently occurred in the Operation Devout II fraud. A further section details third party payments made offshore from United Kingdom companies controlled by Rachar to a British Virgin Islands company called Managing Agents (Europe) Ltd, of which Rachar was the director. Another section refers to significant amounts of money being transferred by Rachar through the intermediary of a Singapore company called Cedgar, and includes a statement of account referring to very substantial payments made by Cedar to Mr Harbhajan S Chopra and Mrs Surjit K Chopra. It also details third party payments made by Globalactive to various Rachar companies both in the context of the Operation Vitric fraud between 2 May and 21 May 2002 and, as regards a single payment of £41,000 to Camross on 5 July 2002, in the context of the Operation Devout II fraud.
There are sections relating to Rachar’s trading in CPUs through his companies in the Far East and his apparent intention to start new carousels through Singapore. It is also recorded that Rachar was connected with a company called Diva Pte Ltd in Singapore; and on 3 May 2002 there was trade between Globalactive and Diva, and an invoice was sent by Globalactive for the attention of “Mr Kumar” (who, it was suggested by Mr Purnell, was Chopra).
Finally, there are two sections relating primarily to Masheikh in Dubai. They detail various dealings between Rashar and Masheikh. They also refer to a transaction on 9 April 2002 in which Timur Mehmet (i.e. Globalactive) purchased CPUs from Tradex and sold them on to Masheikh; and transactions on 16/17 April 2002 in which Globalactive imported CPUs from PCC Europe and sold them on to Masheikh, but received payment on 29/30 April 2002 not from Masheikh but from Rachar’s company Midthorpe.
The effect of non-disclosure on the safety of the appellant’s conviction
The appellant’s supplemental grounds and the written submissions of Mr Purnell QC on the issue of non-disclosure dwell at length on the litigation history, detailing the pre-trial requests for disclosure and further matters which, it is said, ought to have acted as triggers causing a revision of the decision not to disclose the Operation Vitric material. In our view no useful purpose is served by going over that history. As was common ground, the issue for the court, applying R v Pendleton [2001] UKHL 66, [2002] 1 WLR 72, is whether in the light of the additional material now disclosed the conviction is safe. We should therefore concentrate on the consequences of non-disclosure for the safety of the appellant’s conviction. So far as the reasons for non-disclosure are concerned, it suffices to record that at the hearing before us Sir Derek Spencer QC, who was leading counsel for the prosecution at the trial as on the appeal, accepted full responsibility for the decisions made and acknowledged that he should have co-ordinated more effectively the disclosure process and should have insisted that the Operation Devout disclosure team went deeper than they did into the Operation Vitric material. We should also make clear that there was no suggestion of bad faith in relation to the non-disclosure.
For the appellant, Mr Purnell submitted that the additional material fell into three specific categories: (i) evidence establishing that Rachar was the organising mind of both Operation Vitric and Operation Devout II, acting as the international banker and extracting the money through his corporate vehicles and thence to investment vehicles such as Atlantic Custodian to distribute the proceeds and invest in further carousels; (ii) evidence that the fraud in Operation Devout II was put into effect in order to divert fraudulent funds from Operation Vitric to the next cycle of fraudulent activity and in anticipation of a third cycle which was intended to operate through Singapore entities; (iii) evidence that Chopra and Raffi were personal associates of Rachar and were used by Rachar to give effect to his fraudulent schemes (or, as it was also put in the course of oral submissions, those three and the solicitor used by them, Halpin, formed a group constituting the directing minds of the fraudulent schemes). The way in which they planned Globalactive’s role in the fraud, and the funding to be provided through Atlantic Custodian, was highlighted by rough diagrams drawn by Rachar. The evidence demonstrating those various matters would also have provided the missing explanation for the “genesis” of the fraud charged, namely the aborted early transactions which were intended by Rachar, Chopra and Raffi to be undertaken directly between Aurum and Globalactive. It would also have corroborated the appellant’s case that his correspondence with the Redhill Business Centre of Customs & Excise (the intelligence-gathering centre that was specially designated to track and identify missing trader activity) was a genuine enquiry: Redhill refused to clear Aurum as a trader in CPUs and the appellant subsequently directed that Globalactive was not to trade with Aurum. A further point to emerge from the additional material was that Davey, who was mentioned in the context of the Atlantic Custodian documents available at trial, could now be seen to have links with Chopra and to have been involved in the fraud.
Mr Purnell submitted that the case against the appellant had proceeded on an artificial basis as if it was a self-contained conspiracy, owing to the tunnel vision of prosecuting counsel in keeping back the Operation Vitric material. If the additional information had been available at trial, it would have changed the landscape of the entire case and would have enabled the appellant’s defence to be advanced in a very different context. Rachar would have had to be joined as a defendant or, if not joined, named in the indictment as a co-conspirator, and the prosecution would have brought out the roles of Rachar and his associates, thereby fundamentally changing the nature of the prosecution case. If the prosecution had not brought out those matters, the defence would have done so. This was not possible on the material available to the defence at the trial, owing to the limitations in the information provided by Jogia and the impossibility of cross-examining on it since it came from a co-defendant who was not giving evidence. The appellant’s defence that he had been the innocent victim of the dishonesty of others could have been advanced much more effectively with the benefit of the additional material. It would have enabled a picture to be built up of a group of people who knew one other and were masterminding the overall fraud but with whom the appellant had no association. The picture of Timur Mehmet conspiring with Raffi, Chopra and Jogia to have in place a conspiracy with a respectable front would have supported the defence case that the appellant was duped by sophisticated and experienced fraudsmen. The question whether a defendant has been duped can only be answered fairly if one can identify the persons by whom he is said to have been duped. The additional material would also have helped to negative the prosecution contention that the appellant was financing and supervising the fraud. All the matters now included within the 131 agreed facts could and would have been deployed by the defence.
One specific and important area in relation to which Mr Purnell submitted that the case would have been fundamentally different was the cross-examination of the appellant on the documents relating to Atlantic Custodian. The fact that Atlantic Custodian was controlled by Rachar and was being used by him as a means of channelling monies from a previous fraud into this further fraud would have been brought out in the course of the prosecution case. Prosecuting counsel would not have called for the Atlantic Custodian file during cross-examination of the appellant. Counsel would not have cross-examined on the basis that the appellant was using Atlantic Custodian as a means of investing his own money for the purposes of the fraud or as a means of obtaining payments as the principal beneficiary of the fraud. The entire story would have been available, instead of just half of it; and whilst the appellant’s evidence that he knew nothing about the people behind Atlantic Custodian would have been the same, he would have been able to show who was in fact behind it and how it was being used, through the instrumentality of David Post and Bernard Halpin, to channel monies from Rachar and others into the fraud whilst creating a barrier between them and the appellant. Prosecuting counsel had told the jury to “follow the money”, but the additional material presented an altogether different picture of where the money came from and how it was processed. All this, in relation to a matter that occupied so much time at the outset of the appellant’s cross-examination, was capable of having a material effect on the jury’s assessment of his credibility; and there would have been an evidential basis for inviting the jury to draw an altogether different inference from that put forward by the prosecution as to the appellant’s knowledge of the conspiracy.
For the prosecution, Sir Derek submitted that the undisclosed material would not have fundamentally changed the prosecution case at trial or the appellant’s explanations or evidence. If the material had been drawn to his attention, it would not have caused him unilaterally to add anything from Operation Vitric into the trial, since the Operation Vitric trial had been stayed for non-disclosure and he regarded it as tainted and “a can of worms”. He also considered the case in respect of Operation Devout II to be near the limits of manageability. Thus, he would not unilaterally have added Rachar into the indictment as a defendant or as a named conspirator, or have alleged a conspiracy going wider than that in fact put forward at the trial. He accepted, however, that he might have acted differently in response to a defence statement raising these matters. He would also have been prepared, if asked, to make admissions about Rachar’s role and about the involvement of Raffi, Chopra and others as directing minds of Masheikh. He pointed out that the prosecution case throughout was that there were other persons, including the people behind Masheikh, acting as ringmasters of the fraud. In any event he could not have stopped the defence from bringing in the substance of all the material now disclosed. He submitted, however, that the court should not concern itself with how the material might have got in front of the jury or what the modalities of the trial might have been. The question was whether, if it had got in front of the jury, it might reasonably have led the jury to a different verdict.
As to that, Sir Derek submitted that the material would not have changed the central issue, namely the appellant’s defence that he had been an innocent dupe and had not been involved in the day to day operations of Globalactive, or caused the jury to view in a different light the evidence of guilt that they heard. On the contrary, it would have been strongly supportive of the prosecution case and would have demolished the appellant’s defence.
He submitted that the appellant was proactive in negotiating the terms of the co-operation agreement with Atlantic Custodian and that the additional material would have shown by inference that he was negotiating directly with Rachar, as well as that the £2 million was money from Rachar. It would also have supported the complicity in the fraud of Davey in Dubai, to whom the appellant instructed his solicitors to send a copy of the Co-operation Agreement. It would not have altered the thrust of the cross-examination of the appellant with regard to Atlantic Custodian, which was to demolish the appellant’s evidence that the deal was all down to Timur Mehmet. The correspondence in the Tolhurst Fisher file already gave the lie to that: there is no mention of Timur, as opposed to the appellant, in any of the documents, and it was evidently the appellant who was acting in practice as the client of Tolhurst Fisher. The additional material contains nothing to help the appellant’s position in relation to such cross-examination. It would merely have raised lots of further questions concerning his relationship with Rachar. The only point in the cross-examination at trial that would not have been made if the additional material had been available was the suggestion that the £2 million had been put up by the appellant himself; but that was not of major significance in the context of the cross-examination as a whole and was downplayed in the prosecution closing speech. It cannot have been determinative of the jury’s verdict.
The additional material would also have shown Globalactive’s involvement in fraud prior to the period of the conspiracy alleged in the actual trial (25 June to 22 July). In particular, Globalactive had transactions with Tradex and Cytech which formed part of the Operation Vitric conspiracy (and evidence produced by the defence at the trial showed that the appellant personally asked for and obtained verification of Tradex from Redhill on 3 April 2002, and also of Rage on 4 July 2002, though no actual trading took place with Rage). The appellant’s loan of £500,000 could be seen as having funded this part of the fraud, thereby destroying a point made by the defence that the loan had been repaid weeks before the fraudulent transactions comprised in the conspiracy with which the appellant was charged. The involvement of Globalactive over this longer period would also have reduced the possibility that the appellant and his staff were ignorant of the fraud. Further, since all the trading could be seen to have been rigged from first to last, without any period of legitimate trading prior to the period of the conspiracy alleged at trial, the appellant’s story about keeping away from the trading room because of the noise would have been unsustainable.
More generally, the additional material would have strengthened the prosecution case that Mustafa and Timur Mehmet and the other fraudsters (now shown to include Rachar and his associates) had no need of an innocent target, as the appellant claimed to have been, since they had the knowledge and the finance to engage in the fraud without an innocent “front” such as the appellant. Moreover Rachar, an experienced and very successful international fraudster, would not have countenanced using Globalactive without the appellant’s complicity, in circumstances where the appellant was the major shareholder and controlled the bank account. Timur Mehmet was an employee, not director, and was unable to make payments. Without the appellant’s complicity the fraud might have been discovered at any time or have been impeded because payments could not be made.
A further respect in which the additional material would have strengthened the prosecution case was by showing that the sole purpose of the payments to HIFX was to launder the profits of fraud and that the entries in the ledger pages of the cash books were bogus. The payments went to a Canadian company, and Rachar is Canadian, although there is nothing to show that he controlled the company or what happened to the money.
In assessing those competing submissions, we have directed ourselves in accordance with the decision of the House of Lords in R v Pendleton (cited above). Pendleton was a murder case in which the court received fresh evidence, including a statement made by the appellant at the time of the murder and a report and oral evidence by a forensic psychologist relating to the reliability of the appellant’s admissions to the police in 1985. The question was whether, in the light of the fresh evidence, the conviction was unsafe. The decision of the House of Lords lays down how the Court of Appeal should approach such a question. It was common ground between counsel before us that the same principles apply in the present case to our assessment of the safety of the appellant’s conviction in the light of the additional material that should have been disclosed to the defendants and is now available to the court.
The leading speech in Pendleton was that of Lord Bingham of Cornhill. At paragraph [15] he quoted from the speech of Viscount Dilhorne in Stafford v Director of Public Prosecutions [1974] AC 878. In that speech Viscount Dilhorne rejected an argument that in a fresh evidence case the Court of Appeal should only apply the proviso to section 2(1) of the Criminal Appeal Act 1968 in its then existing form if it was of the opinion that, if the jury had been properly directed, they would inevitably have come to the same conclusion; and he observed that, while the Court of Appeal might find it a convenient approach to consider what a jury might have done if they had heard the fresh evidence, the ultimate responsibility for deciding the question rested with the court and the court alone. Lord Bingham continued:
“16. In hearing any appeal against conviction the Court of Appeal will ordinarily have a considerable body of material before it: grounds of appeal; transcripts of the judge's summing up to the jury and any relevant passages in the evidence and of any material rulings given before or in the course of the trial; plans, photographs and so on. And although the court does not have the jury's reasons, it does have the jury's verdict. From this, some inferences may always be drawn. If the issue is consent, the jury must, to convict, have been sure that the victim did not consent. If the issue is pure identification, the jury must, to convict, have been sure that the evidence identifying the defendant was accurate and reliable. If a proper judicial direction has been given, it will ordinarily be safe for the Court of Appeal to infer that the factual ingredients essential to prove guilt have been established against the defendant to the satisfaction of the jury. But the Court of Appeal can rarely know, save perhaps from questions asked by the jury after retirement, at what points the jury have felt difficulty. The jury's process of reasoning will not be revealed and, if a number of witnesses give evidence bearing on a single question, the Court of Appeal will never know which of those witnesses the jury accepted and which, if any, they doubted or rejected.
17. My Lords, Mr Mansfield is right to emphasise the central role of the jury in a trial on indictment. This is an important and greatly-prized feature of our constitution. Trial by jury does not mean trial by jury in the first instance and trial by judges of the Court of Appeal in the second. The Court of Appeal is entrusted with a power of review to guard against the possibility of injustice but it is a power to be exercised with caution, mindful that the Court of Appeal is not privy to the jury's deliberations and must not intrude into territory which properly belongs to the jury.
...
19. It is undesirable that exercise of the important judgment entrusted to the Court of Appeal by section 2(1) of the 1968 Act should be constrained by words not to be found in the statute and that adherence to a particular thought process should be required by judicial decision. Thus the House in Stafford v Director of Public Prosecutions [1974] AC 878 were right to reject the submission of counsel that the Court of Appeal had asked the wrong question by taking as the test the effect of the fresh evidence on their minds and not the effect that that evidence would have had on the mind of the jury. It would, as the House pointed out, be anomalous for the court to say that the evidence raised no doubt whatever in their minds but might have raised a reasonable doubt in the minds of the jury. I am not persuaded that the House laid down any incorrect principle in Stafford, so long as the Court of Appeal bears very clearly in mind that the question for its consideration is whether the conviction is safe and not whether the accused is guilty. But the test advocated by counsel in Stafford and by Mr Mansfield in this appeal does have a dual virtue to which the speeches I have quoted perhaps gave somewhat inadequate recognition. First, it reminds the Court of Appeal that it is not and should never become the primary decision-maker. Secondly, it reminds the Court of Appeal that it has an imperfect and incomplete understanding of the full processes which led the jury to convict. The Court of Appeal can make its assessment of the fresh evidence it has heard, but save in a clear case it is at a disadvantage in seeking to relate that evidence to the rest of the evidence which the jury heard. For these reasons it will usually be wise for the Court of Appeal, in a case of any difficulty, to test their own provisional view by asking whether the evidence, if given at the trial, might reasonably have affected the decision of the trial jury to convict. If it might, the conviction must be thought to be unsafe.”
In applying that approach to the particular facts of Pendleton’s case, Lord Bingham observed at paragraph [26] that “given the primacy of the jury, it must always be a ground for concern if the jury have never considered a potentially important aspect of the case”; and at paragraph [27] he said that no-one could now be sure what would have happened if the psychologist’s evidence had been available at the time of the trial, but the defence might in at least three respects have been conducted differently (the first being the obviously important consideration that the appellant might have been called to give evidence on his own behalf). At paragraph [28] he concluded that in the light of the uncertainties and the fresh psychological evidence, it was impossible to be sure that the conviction was safe, and that was so whether the members of the House asked whether they themselves had reason to doubt the safety of the conviction or whether they asked whether the jury might have reached a different conclusion.
Applying those principles, we have asked ourselves in the present case whether, in the light of the additional material, we have reason to doubt the safety of the conviction; and having reached a provisional view on that matter, we have tested it by asking whether the additional material, if available at trial, might reasonably have affected the decision of the trial jury to convict. We accept Sir Derek’s submission, illustrated by the decision in R v Kenedy (Hamidi) [2008] EWCA Crim 2817, paragraph [23], that the mere fact that the material satisfied the test for disclosure does not decide the question (not least because any potential assistance that the material gives the defence may be greatly outweighed by the help it gives the prosecution case). The question has to be decided by consideration of the overall impact of the additional material on the case.
In reaching our decision we have given very careful consideration to the written as well as oral submissions of both counsel. Long though this judgment is, it provides only a summary of the matters covered by counsel and does not detail every matter relied on even in oral submissions. We are also acutely conscious of the fact that we have seen only a fraction of the evidence in this very long trial and, of course, that we do not know the detailed process of reasoning by which the jury reached their verdict.
We accept that with the additional material the landscape of the case, as Mr Purnell put it, would have been significantly different and the appellant’s defence would have been advanced in a different context. The conspiracy alleged would have fallen to be considered within a much wider context of fraudulent activity than appeared at the trial itself. It may be, as Mr Purnell submitted, that this would have enabled aspects of the defence to be advanced more effectively and would have prevented the prosecution from making certain points against the appellant. It would not, however, have altered the essential nature of the prosecution case against him, or his defence, or the evidence he gave in support of that defence. The central issue would still have been his defence that he was an innocent dupe who was not involved in the day to day operations of Globalactive and was unaware that Globalactive was participating in fraudulent activity. That defence was rejected by the jury, who must have found that the appellant had lied about his lack of knowledge of the fraud. In our judgment, the overall effect of the additional material would have been seriously damaging to the defence rather than helpful to it and would not have warranted a different conclusion as to the appellant’s credibility or his knowledge of the fraud.
We do not accept that the involvement of sophisticated fraudsmen such as Rachar and his associates in the organisation and financing of the fraud made it more likely that the appellant was duped. As Sir Derek submitted, it would have strengthened the prosecution case that those responsible for the fraud had no need to involve the appellant as an innocent dupe, since they had the knowledge and finance to engage in the fraud without an innocent front of that kind; and that they would not have countenanced using Globalactive without his complicity, in circumstances where he was the major shareholder and controlled the bank account.
The cross-examination of the appellant in relation to Atlantic Custodian would, to a certain extent, have had a different shape to it. It would already have come out that Rachar was behind Atlantic Custodian and was using the company as a vehicle for financing the fraud. Thus it could not have been suggested that the appellant was channelling his own money through Atlantic Custodian to finance Globalactive’s participation in the fraud (but that suggestion, whilst made in cross-examination, was downplayed in the prosecution closing speech, where the alternative that the money had been provided by other fraudsters was advanced). Moreover, even if there had still been late disclosure of the Tolhurst Fisher file, it is possible that no point would have been made of it and it is unlikely that it could of itself have given rise to any adverse inference against the appellant. But the appellant’s evidence concerning his knowledge of, and dealings, with Atlantic Custodian would have been the same, as would the main thrust of the cross-examination in relation to those matters. The prosecution’s point that the documents evidenced the appellant’s detailed involvement in the negotiation of the Co-operation Agreement, and no involvement by Timur Mehmet, would have lost none of its force. That the appellant’s negotiations were with the major fraudster Rachar would have been an additional point, and a very damaging one. Although there was no direct evidence of association by the appellant with Rachar, or that the appellant knew Rachar’s identity, his instruction to fax the draft of the Agreement to Davey in Dubai would have had an even greater significance because of the links shown to exist between Davey and the fraudsters in Dubai; and, in any event, the cross-examination directed to the implausibility of the appellant having no real idea of the identity of the people with whom he was negotiating would have had an even greater resonance to it.
The evidence that Globalactive was involved in the latter stages of the Operation Vitric fraud, so that it could not be said that the company had been engaged in legitimate business prior to the period relied on by the prosecution at the appellant’s trial in respect of the Operation Devout II fraud, would also have been damaging to the appellant’s defence, for the reasons advanced by Sir Derek.
Having considered those and the other matters raised by counsel, and having looked at them in relation to the case as a whole, we have formed the view that the additional material provides no reason to doubt the safety of the appellant’s conviction.
In order to test that view, we have gone on to ask ourselves whether the additional material might reasonably have affected the jury’s decision to convict. In that connection we have considered the various matters that might have been particularly influential in the jury’s assessment of the appellant’s credibility, including Mr Purnell’s submission that the beginning of the appellant’s cross-examination, concerning Atlantic Custodian, must have been particularly significant yet would have proceeded in a materially different way if the additional material had been available at trial. We have also borne in mind that if, as Sir Derek asserted and we are inclined to accept, the additional material would not have caused the prosecution to add Rachar as a defendant in the trial, the very fact that Rachar was not being pursued for his role in the fraud could have been deployed forcefully as a forensic point in the appellant’s favour. In general, however, the matters that we have taken into account in reaching our own provisional view on safety would have been of equal importance for the jury’s decision. Taking everything together, we cannot see a sustainable basis for concluding that the additional material might reasonably have affected the jury’s decision.
Our conclusion, therefore, is that the non-disclosure of the additional material, taken by itself, did not affect the safety of the appellant’s conviction. Before reaching a final conclusion on safety, however, we need to consider the other grounds of appeal.
The summing up: alleged fragmentation of the defence case
The judge’s summing up extended over some two and half days. He adopted an essentially chronological approach, interleaving the prosecution and defence cases into a narrative history or dealing with them in the context of various topic headings. It is submitted that this had the effect of fragmenting the appellant’s defence and that, whilst the essence of his case would have been clear to the jury, the evidence in support of it was not pulled together or summarised in any coherent way so as to enable the jury to understand it as of a piece. The force of the defence was thereby diluted and diminished, and the appellant was deprived of a fair trial.
In developing this submission Mr Purnell pointed to the various different contexts in which the judge dealt with items of evidence supporting the appellant’s case; but we think it unnecessary to recite the detail. He also contrasted the position of the appellant with that of the defendants who did not give evidence (including the absent Timur Mehmet), in relation to each of whom the judge gave the jury, by way of reminder, a summary of the submissions that had been made by their counsel. He submitted that it was incumbent on the judge to include a similar section in relation to the appellant; the more so since the case for Timur Mehmet, which the judge did summarise, was inimical to the appellant.
In our judgment it would have been better if the judge, in addition to dealing with matters in the way he did, had given the jury a short summary of the defence of each of the defendants, especially as this was done for the defendants who did not give evidence. We are satisfied, however, that his omission to follow that course did not render the summing up unfair or the conviction unsafe. That the jury were aware of the essential nature of the appellant’s case is not in dispute. It seems to us that they must also have been well aware of the main points of evidence relied upon in support of that case. It was accepted by Mr Purnell that no vital point was omitted from the judge’s summing up, albeit they were not pulled together as he contended should have been done. The jury had also heard a closing speech from Mr Purnell which, we are told, took about 5 hours; and whilst we accept that the judge’s summing up, especially after a long trial, has a different focus and one cannot rely on counsel’s closing speech alone, the combined effect of the closing speech and the summing up must have been to leave the jury in no doubt as to the appellant’s defence.
The summing up: the judge’s new points
A further complaint about the summing up is that in the course of it the judge introduced a number of points which, it is said, had not been canvassed in evidence or submissions and some of which were simply wrong. The specific matters raised are as follows.
First, having referred to Jogia’s introductory letter of 7 May 2002 to Globalactive and to the appellant’s prompt reply to it, the judge said “I do not have any note of Mr Pomfrett giving an explanation of why he should have chosen to answer that letter and not Mr Mehmet who was after all running the trading”. The fact was that the appellant was not asked for an explanation during his evidence. The submission is that, if the judge wished to raise the point, he should have asked a question in the course of evidence rather than making a comment of this nature in his summing up. Against that, Sir Derek Spencer told us that the point had already been made in the prosecution closing speech.
Second, the judge referred to a visit by the Customs & Excise officer, Mr Davies, to the premises of Globalactive on 9 April 2003 while the appellant was in Spain. Mr Davies spoke mainly to one of Globalactive’s employees, Ms Killigrew, but he also spoke to the appellant by phone. The judge attributed to the appellant information recorded in Mr Davies’ notebook about the places and companies with which Globalactive was trading (including that there had been trading with Dubai). There was in fact an issue, explored in cross-examination of Mr Davies, as to whether that information had come from Ms Killigrew rather than the appellant. This matter was raised promptly by Mr Purnell with the judge, who thereupon reminded the jury of the relevant cross-examination. In our view that dealt with the matter satisfactorily.
Third, in relation to a particular transaction examined by Mr Davies, concerning export sales by Globalactive to Masheikh in April 2002, the judge commented “you will not find anywhere in those Global accounts any payments at all coming in from [Masheikh] for the goods that apparently they were buying from Global”, and that it was not known why that was so. In fact Globalactive did receive a payment, but from Midthorpe rather than from Masheikh. Mr Purnell submitted that the judge’s comment was adverse to the appellant because the gravamen of the prosecution case was that since he was in charge of money he must have known about the various payments. We accept that submission, though we do not regard the comment as one of great significance in the context of the case as a whole. We think that Sir Derek went too far in submitting that this was not a point against the appellant at all.
Fourth, the judge carried out a detailed mathematical exercise, which had not been done in evidence or submissions, to show that there were considerable variations in the number of boxes of CPUs sold in the various transactions: he told the jury that they might like to ask themselves whether this had been done deliberately by the organisers so that the sales would not stand out to any assurance officer looking at the documents. Although Mr Purnell described this point as being generally anti-defendant, it can have had no material impact on the appellant since there was no suggestion that he had anything to do with the number of boxes comprised within individual transactions.
Fifth, in going through a detailed schedule of the transactions, the judge commented at one point that the supplier appeared to have been underpaid, and that “no documents have been recovered as to why anybody [i.e. nobody] is complaining about not getting quite large sums of money or sometimes being paid large sums of money”. Mr Purnell submitted that the judge’s comment might be taken to imply that the appellant must have been aware that this was bogus trading. Two considerations reduce the force of that submission. First, Mr Purnell himself, as he told us, spent a long time going through the schedule with the jury to show that every CPU supplied was paid for by somebody, even though the payments were not always direct payments between supplier and purchaser. Secondly, since the appellant’s evidence was that he was not involved in the day to day trading, the part of the judge’s summing up devoted to going through the schedules had little, if any, direct impact on his case.
The sixth and final matter raised under this ground concerns the judge’s comment that Globalactive could have continued to trade in August 2002 and did not have to repay the money to Atlantic Custodian: although the Barclays bank account had been closed on 22 July, an account with Bank of Scotland was opened on 9 August. Mr Purnell submitted that this overlooked the evidence that the company needed an account with an online facility for the purposes of trading, and the Bank of Scotland account lacked that facility. Moreover the evidence was that Mustafa and Timur Mehmet had gone to Cyprus at the end of July. The monies were repaid to Atlantic Custodian because the Barclays account was being closed and all outstanding liabilities were settled. Sir Derek’s response to all of this was that what happened in August was well after the event and not a point of substance.
In addition to the comments we have made when summarising the specific matters raised under this ground, we have a number of more general observations. First, the judge’s comments on the evidence have to be considered in the context of the direction he gave the jury at the outset of his summing up, in clear and conventional terms, that the evidence and facts were entirely for them. Secondly, in so far as it is said that the judge erred in his summary of the evidence, it was incumbent on counsel to draw the errors to the judge’s attention, so that they could be put right at the appropriate time: that was how the second matter, concerning the evidence of Mr Davies, was dealt with, and it removed any reasonable concern. Thirdly, if the judge was minded to make a new point in his summing up, it would have been prudent for him to canvass it in advance with counsel in the interests of accuracy and fairness, especially in a trial of this length and relative complexity. Thus it might have been better in the circumstances for the judge not to have made, without notice to counsel, some of the points on which this ground of appeal is based.
We are, however, satisfied that the matters identified by Mr Purnell, whether taken individually or in the aggregate, did not render the summing up as a whole unfair and did not have any material impact on the fairness of the appellant’s trial. They do not provide a ground for doubting the safety of the conviction.
Receipt of instructions by counsel for Timur Mehmet in his client’s absence
Timur Mehmet had attended pre-trial hearings when required to do so and had entered a plea of not guilty to the indictment but, as we have said, he absented himself from the trial itself. His counsel informed the judge that he was at an unknown location, was in continuing communication with his solicitors by means of email and had been advised that he should attend trial and was in breach of his bail conditions. The judge ruled, on the prosecution’s application, that he should be tried in his absence and that he should be represented by leading and junior counsel and solicitors. In so ruling the judge observed that if any matter arose on which they did not have instructions there was the possibility of email communication, so that any unfairness that might otherwise arise through, for example, new evidence coming to light or new matters becoming part of the case could be resolved by the legal team obtaining instructions. Immediately after the ruling Mr Purnell put down a marker concerning that issue but he considered it unnecessary to pursue it at the time.
As the trial proceeded, there was a developing conflict between the appellant’s case and that of Timur Mehmet. Then there came a time, in the course of the appellant’s evidence in chief, when there had to be an adjournment to enable counsel to examine a large number of additional documents just produced by a Customs & Excise officer, Mr Bright. An indication was given by Timur’s counsel that he would need to take instructions on the new material because he might need to deploy it against the appellant. This prompted an application by Mr Purnell for a ruling to the effect that Timur’s counsel should be restricted to the instructions received prior to Timur’s failure to surrender and that no fresh instructions should be taken in the course of the trial. The judge ruled against that application. One of the grounds of appeal is that the judge’s ruling was wrong and that the opportunity for Timur’s counsel to receive and deploy instructions from, and give advice to, an absent defendant led to the appellant having an unfair trial. It is said that the damaging effect on the appellant’s case was all the greater because the judge had restricted the making of any adverse comment in respect of Timur’s absence and had ruled that no inference could be drawn against him under section 35 of the Criminal Justice and Public Order Act 1994 from his failure to give evidence.
The principles governing the trial of a person in his absence are set out in the decision of the House of Lords in R v Jones (Anthony William) [2002] UKHL 5, [2003] 1 AC 1. One of the matters covered is the representation of the absent defendant. As to that, Lord Bingham of Cornhill stated at paragraph [15]:
“Secondly, it is generally desirable that a defendant be represented even if he has voluntarily absconded. The task of representing at trial a defendant who is not present, and who may well be out of touch, is of course rendered much more difficult and unsatisfactory, and there is no possible ground for criticising the legal representatives who withdrew from representing the appellant at trial in this case. But the presence throughout the trial of legal representatives, in receipt of instructions from the client at some earlier stage, and with no object other than to protect the interests of that client, does provide a valuable safeguard against the possibility of error and oversight. For this reason trial judges routinely ask counsel to continue to represent a defendant who has absconded during the trial, and counsel in practice accede to such an invitation and defend their absent client as best they properly can in the circumstances. The current legal aid regulations provide for that contingency …. It is in my opinion a practice to be encouraged when the defendant absconds before the trial begins.”
The question whether the legal representatives of an absent defendant are entitled to receive and act on fresh instructions during the course of the trial was not the subject of argument in Jones, and nothing said by their Lordshipsprovides a direct answer to it. Lord Bingham, in stating at paragraph [15] that an absent defendant “may well be out of touch” with his legal representatives, evidently also contemplated that an absent defendant might be in touch with his legal representatives, but he did not go on to consider the implications as regards the giving of further instructions. Similarly, he referred at paragraph [12] to a multi-defendant trial, noting that a defendant who absconded might gain a wholly unjustified advantage unless he were tried with the others in his absence, but he did not go on to consider the implications for the co-defendants of a defendant being tried with them in his absence, or how the trial judge’s discretion was to be exercised so as to ensure fairness to the co-defendants in those circumstances. It is, however, fair to say that there is nothing in Jones to suggest any distinction of principle between instructions given by the absent defendant before he absconded and those given by him after he absconded.
Nor is any distinction drawn in the Bar Council’s guidance on the subject. Paragraph 15.3.2 of section 3 (“Written Standards for the Conduct of Porfessional Work”) of the Bar Code of Conduct states:
“If for any reason the barrister’s professional client does not withdraw from the case, the barrister retains an absolute discretion whether to continue to act. If he does continue, he should conduct the case as if his client were still present in Court but had decided not to give evidence and on the basis of any instruction he has received. He will be free to use any material contained in his brief and may cross-examine witnesses called for the prosecution and call witnesses for the defence.”
Mr Purnell submitted that in any case where a defendant absconds and is tried in his absence, his counsel is limited to acting on instructions received before the date when the defendant absconded. He said that it would be contrary to the public interest to confer an advantage on an absconding defendant by enabling his counsel to cross-examine witnesses, including any co-defendants who give evidence, on the basis of fresh instructions received in the course of the trial.
We disagree. In our judgment, no principled distinction can be drawn between instructions received before and instructions received after the date of absconding. If counsel were not permitted to take into account fresh instructions received from the absent defendant, it would increase unnecessarily the possibility of error or oversight against which, as Lord Bingham said in Jones at paragraph [15], representation by counsel provides a valuable safeguard. It could lead to counsel having to advance a case wholly at odds with the defendant’s true case. An example suggested by the court in the course of argument was that of a pub fight in which the defendant lied in interview as to presence (because he did not want his girlfriend to know he had been there), was proved by DNA evidence to have been present, but in an email to his legal representatives after absconding admitted that he had been present but said he had been acting in defence of himself and a co-defendant: it would be highly unsatisfactory for counsel in those circumstances to be precluded from advancing the defendant’s true defence and to be required instead to advance a hopeless defence of absence from the scene. It could also prevent counsel from being able to deal effectively with new evidence or new issues, even where the absent defendant had valid points to make in relation to them: the possibility of taking instructions on any new matter that might arise in the course of the trial was one of the considerations the judge had properly in mind in ruling in the first place that Timur Mehmet could be tried in his absence. If it is generally desirable that a defendant be represented even if he has voluntarily absconded, it is equally desirable in our judgment not to impair the effectiveness of that representation by laying down an artificial restriction on the instructions upon which counsel can act. The public interest does not require a cut-off point at the date of absconding.
It will, of course, always be important for the judge to keep an eye on the overall fairness of the proceedings and to exercise his discretion appropriately to prevent any unfair manipulation of the process by an absent defendant to the disadvantage of co-defendants. But Mr Purnell’s submission on the issue of fresh instructions was put forward as one of principle, applicable in every case irrespective of the circumstances, not on the basis that there was a particular point in the present case where the judge ought, in the exercise of his discretion, to have intervened to stop or restrict cross-examination of the appellant by counsel for the absent Timur Mehmet. In any event, Mr Purnell’s arguments appeared to us to make too much of the damage that could be done to a co-defendant, even in a cut-throat situation, by cross-examination on behalf of an absent defendant. He conceded that in his speech to the jury he was “entirely dismissive” of what was said against the appellant on behalf of Timur Mehmet. The fact that Timur was not present at the trial to be cross-examined himself was no doubt one of the points deployed in support of that dismissive approach. It is unnecessary for us to consider the precise extent to which comment could properly be made on Timur’s absence. Although Mr Purnell submitted to us that the problem of cross-examination on behalf of Timur was exacerbated by the limitations the judge imposed on counsel’s freedom to make such comment, he accepted that he was not making any separate challenge to the judge’s rulings as to comment on absence or as to the drawing of an inference under section 35 of the 1994 Act. Since we do not accept that there was any error in the judge’s ruling about cross-examination on behalf of Timur on the basis of fresh instructions, any question of exacerbation of an error falls away.
Visits by junior prosecuting counsel to see the judge alone
This issue relates to visits by junior counsel for the prosecution to the judge’s room at the judge’s request, without attendance by other counsel in the case and, as is contended, without any assurance that what took place in the room was recorded or properly documented. When, early in the trial, Mr Purnell raised with the judge his concern about such a visit, the judge made clear in open court that it concerned an administrative matter that the judge needed to deal with in respect of other, unrelated cases. The judge’s remarks also suggest that what took place in his room was recorded. Sir Derek told us, on the basis of information provided by junior counsel and the instructing solicitor, that there had previously been ex parte PII hearings before the judge in an unrelated case. He submitted that, whilst it was not known why the judge wanted to see junior counsel on the occasion to which Mr Purnell took objection, the judge would have been entitled to see counsel in his room in connection with ex parte PII matters. He also informed us that on two subsequent occasions the judge invited junior counsel to his room to find a file, but after Mr Purnell’s intervention junior counsel asked Sir Derek to forbid him from making further such visits. Mr Purnell took issue with junior counsel’s recollection, assuring us that visits to the judge’s room took place regularly.
To the extent that there is a factual dispute, it is one that we do not need to resolve. Mr Purnell accepted that these visits were not concerned with the case against the appellant or his co-defendants and did not put them forward as rendering the appellant’s conviction unsafe, though he described them as an irregularity that added to the unfairness of the appellant’s trial. In our view, if they gave rise to any problem, it was in relation to the other trials with which they were concerned. It was nonetheless right of Mr Purnell to draw the matter to the court’s attention. Whilst there may well be a good explanation for visits of this kind for the purpose of an ex parte PII application, we would stress the need for full and careful compliance with the requirements of Part 25 of the Criminal Procedure Rules in relation to the making of such applications, the importance of ensuring that the proceedings are recorded, and the need for discretion in how the matter is handled. If the visit to the judge’s room relates to something other than PII, for example if the judge is simply seeking the assistance of counsel in finding a document, it should be dealt with on notice to, and with agreement of, all counsel in the relevant case, so as to avoid any impression of unfairness.
This ground of appeal is, however, of no assistance to the appellant in the present case.
Appeal against conviction: conclusion
For convenience, we have dealt with each of the grounds of appeal in turn and have explained in relation to each why we do not consider it to provide a basis for doubting the safety of the conviction. Having regard to the way in which Mr Purnell put his case, we have also considered whether they have any different effect when looked at together. We are satisfied, however, that they provide no greater assistance to the appellant when taken in the aggregate than when examined individually.
It follows that in our judgment the appellant’s conviction is safe and his appeal against conviction must be dismissed.
Appeal against sentence
We turn to consider the renewed application for leave to appeal against sentence. The appellant was 55 years of age at the time of conviction (he is now 56). He had no relevant previous convictions. He was a self-made man who had effectively retired from his other business interests by the time he became involved in Globalactive. He had arranged the sale of his country club and golf club businesses and was spending a substantial amount of time in Spain.
In sentencing him, the judge said that he was “no doubt a very clever and well organised person and as the jury have found a thoroughly dishonest person”. The judge was satisfied that he controlled the detailed finances of Globalactive with a very careful eye. As to the Crown’s case that the appellant was the financier of the fraud, the judge observed that “[i]t is undoubtedly the fact that you primed the pump of Global’s trading but that is not necessarily the same as saying you were the financier of the fraud”. He summarised the appellant’s involvement in Globalactive, the appellant’s investment of £500,000 and the receipt of payments totalling £2 million from Atlantic Custodian, which money “was available in the Global bank account … for what use you decided upon”. He had no doubt that the fraudulent exercise was intended to start on 7 May but that the concerns of Customs & Excise about Aurum caused the appellant to abort the process of Aurum purporting to sell to Globalactive and led to the delay in setting up Lightcare as second line buffer and Globalactive moving down to third line buffer. The judge continued:
“When the scheme got going on 25 June the very first transaction ended with Global selling to Perles France, which enabled Global to later make repayment claim for the VAT element. However Global had to outlay £982,713 payments to third party accounts. They sold to Perles for £864,000 but … there is no receipt of this sum. In all the other transactions Gobal sell on to a UK company and I accept … that thereafter money flows into the Global account from the next traders in the chain.
The evidence shows Global in effect financed this fraud and became its paymasters, paying the largest portion of the monies out of the UK. This puts you in a different category to the other defendants, you were in control of the finances. I reject your evidence that you were only concerned in a general overseeing way. There is in my judgment little difference between the instigator and organiser of these frauds and the financier of them. However I cannot on the evidence find that you were the actual top dog in this conspiracy who remains unknown.”
On that basis, and having taken into account the appellant’s personal mitigation, the judge imposed a sentence of 10 years’ imprisonment.
The authorities to which the judge referred in setting that and the other sentences included Attorney-General’s References Nos. 88 etc of 2006 (Brian John Meehan and Others) [2007] 2 Cr App R (S) 28 (at page 155). In that case the offenders had been convicted of cheating the public Revenue in a missing trader fraud. The court held that those who organise such fraudulent activity can and should expect sentences well into double figures after trial (paragraph [18]). The role of the first three offenders, however, was limited to running the buffer companies which handed over the VAT to the organiser of the fraud; they were not involved in any deeper way in the planning, organising or running of the scheme. For them, a substantial period of imprisonment was clearly required because of the money involved (the total VAT lost as a result of the involvement of the first and second offenders was just under £24 million, and in the case of the third offender it was £28 million), but there should be a substantial discount from the sort of sentence to be expected by an organiser. The court held that “the right bracket for this type of offending, with the amount of money involved in these cases, after a plea of guilty, would have been six to eight years” (paragraph [19]).
Mr Purnell submitted that even on the evidence before the judge it was unfair and inappropriate to focus on the appellant as the judge did, treating him as playing the crucial financier’s role within the fraud. In any event the additional material shows a very different situation, in which Globalactive played a satellite role within a much larger fraud, and that role was itself organised and substantially funded by Rachar and his associates. The appellant’s role was limited to his involvement in running Globalactive as a buffer company. The appellant’s £500,000 investment in Globalactive did not make him the financier of the fraud and he should not be sentenced as a financier. He was a facilitator rather than a financier or organiser.
Mr Purnell also advanced a submission to the effect that, since the appellant could have been charged with the statutory offence of conspiracy covering the same criminality and carrying a 7 year maximum sentence, it was wrong for the prosecuting authority to bring a common law charge of conspiracy to cheat or for the judge to impose, in respect of such an offence, a sentence in excess of the maximum for the statutory offence. He accepted that the decision of the Court of Appeal in R v Ward [2006] 1 Cr App R (S) 66 (at page 356) was against him, but he signalled in his written submissions that he would seek leave to argue that that case and the authorities it followed were wrongly decided; and he submitted orally that the charging of conspiracy to cheat so as to give a higher sentencing range should be limited to the most exceptional cases and that it was not appropriate for this appellant, a fortiori in the context of the landscape disclosed by the additional material.
We are satisfied that the approach in R v Ward is correct and should be followed, and we reject the contention that the appropriate level of sentencing in this case was constrained by the maximum of 7 years that would have applied to the statutory offence of conspiracy. In our view the level of sentencing indicated in Attorney-General’s References Nos. 88 etc of 2006 (cited above) for an offence of cheating the public Revenue in a missing trader case provides a useful guide for the offence of conspiracy to cheat of which the appellant was convicted.
On the information available to the sentencing judge, the appellant’s sentence of 10 years was within the range properly open to him. In particular, he was entitled on the evidence he had heard to treat the appellant as the financier of the fraud, albeit, as he said, “not the actual top dog”. But we accept Mr Purnell’s submission that the additional information now available shows the appellant to have had a lesser role in the overall scheme of things than appeared on the evidence at trial. He was neither an organiser of the conspiracy nor a principal financier of it: that role belonged to Rachar and his associates. The appellant’s role was essentially that of running a buffer company, though it was a particularly important buffer company and was used as the vehicle for the introduction, through Atlantic Custodian, of finance for the fraud.
In all the circumstances we consider that the appropriate sentence for the appellant would be one of 8 years’ imprisonment, the same as that imposed on Timur Mehmet. We will therefore grant leave to appeal and allow the appeal to the extent of quashing the sentence of 10 years’ imprisonment and substituting a sentence of 8 years (with the 4 days in custody on remand still to count towards that sentence).