R. v Kallakis & Anor

On 12 April 2013 there were listed before the court an application under section 36 Criminal Justice Act 1988 made on behalf of the Attorney General for leave to refer, as unduly lenient, sentences imposed for offences of conspiracy to defraud, and an appeal against sentence imposed for statutory offences of fraud and related offences. The cases raised common issues which counsel agreed made it convenient for the court to consider submissions sequentially and to prepare a joint reserved judgment. The Attorney General has requested that the court should consider afresh the approach to sentencing for the most serious offences of conspiracy to defraud. In the appeal against sentence it was contended that the starting point was manifestly too high, consecutive sentences failed adequately to reflect the principle of totality, and the appellant was given too little credit for his pleas of guilty and personal mitigation. The cases raise issues upon which the court has taken time to reflect:

The offender Achilleas Michalis Kallakis (formerly Stefanos Kollakis) was born on 3 September 1968. The offender Alexander Martin Williams (formerly Lewis) was born on 3 June 1968. They are both now aged 44 years. They were charged in an indictment containing 23 counts. In count 1 they were charged with conspiracy together and with Michael Becker, between 1 August 2003 and 13 November 2008, to defraud Allied Irish Bank PLC (“AIB”) by using various dishonest means to cause AIB to advance money for the purpose of funding the purchase of property by companies owned or controlled by Kallakis, and to provide ‘reverse premiums’ ostensibly payable on purchase to a third party. Count 1 identified 9 particulars of the nature of dishonest conduct by which the fraud was completed. Counts 2 – 20 charged 19 alternative substantive offences committed, it was alleged, in the performance of the count 1 conspiracy. They included using false instruments, obtaining money transfers by deception, fraud and money laundering. The offenders were charged in count 21 with a second offence of conspiracy, together and with Michael Becker, between 1 January 2007 and 2 May 2008, to defraud Bank of Scotland PLC by dishonestly causing the Bank to advance money for the purpose of converting a passenger ferry to a luxury yacht by producing false documents and making false representations. Count 21 identified 9 particulars of the dishonest conduct by which it was alleged the fraud was completed. In count 22 the offenders were charged with a substantive offence of fraud as an alternative to the conspiracy charged in count 21. In count 23 the offenders were charged with using a false instrument.

The first trial was terminated during the course of Mr Kallakis’s evidence. The re-trial commenced on 10 September 2012. On 16 January 2013 the jury returned verdicts of guilty upon counts 1 and 21 and was discharged from reaching further verdicts. On 17 January 2013 the trial judge, HHJ Goymer, sentenced Kallakis to concurrent terms of 7 years imprisonment and Williams to concurrent terms of 5 years imprisonment. Each offender was disqualified from acting as a director of a limited company in the United Kingdom for a period of 6 years.

The objective of the count 1 fraud was to persuade AIB to advance funds sufficient to provide the purchase price of valuable properties in London together with an additional sum purportedly to be used to cover, at least in part, what was called a ‘reverse premium’ but was in fact to be a payment made for the benefit of the fraudsters in order that the fraud could be perpetuated. The offenders worked from offices in Mayfair. Kallakis represented himself as chief executive officer of the Pacific Group of Companies. There was no such entity, at least of substance. The properties acquired were to be managed by Atlas Management Corporation Limited, incorporated in 2002, with the offender, Williams, as its managing director. While Kallakis was not named in any of the Atlas documentation, he was in ultimate control. In 2002 AIB established a property team within its corporate banking department. Its purpose was to seek and obtain business for the provision of secured loans. During the summer of 2003 Kallakis was introduced, falsely, to members of AIB’s team as the scion of a wealthy Greek family which had made its fortune in shipping. He represented that the family wished to diversify into property. Kallakis said that Sun Hung Kai Properties Limited (“SHKP”), a very substantial Hong Kong based company with whom he already did business, wished to invest in UK property but, for political reasons, wished to do so discreetly. High value property would be purchased by a Kallakis company which would become the landlord. A subsidiary of SHKP would take from the landlord a long head lease at a rent which exceeded the aggregate of rents currently being paid by existing tenants. The effect would be immediately to increase the capital value of the property while creating a shortfall between the rent paid under the head lease and the rent paid by sub-lessees. In time the rent due to the sub landlord (the SHKP subsidiary) would increase to exceed the rent due under the head lease. To cover the temporary shortfall the Kallakis purchaser company would pay to the SHKP subsidiary a reverse premium. On any re-sale, there would be a profit share between SHKP and the landlord. This was the underlying scheme which purportedly justified the several loans and purchases which followed.

For present purposes it is enough to summarise the fraud which culminated in the first of the purchases, and to identify only the scale and general nature of further frauds. Kallakis formed Andromeda Alliance Inc to be purchaser and head landlord. Michael Becker, in Switzerland, was the director. The head lessee, purportedly an SHKP subsidiary, was Causeway Capital Corporation. The proposal was for the purchase of Fitzroy House, 355 Euston Road, then occupied by Network Rail. The purchase price of the freehold was £16.5 million with costs of purchase of about £1 million. Kallakis sought a loan of £18.8 million. Comfort in the loan to value proportion was provided to AIB by a valuation obtained from commercial estate agents. If the SHKP subsidiary, Causeway Capital, entered into a head lease committing itself to an annual rent which exceeded the current rent paid by the tenant, and SHKP guaranteed the rent due under the head lease, the freehold value of Fitzroy House would increase to £23.6 million. In return Causeway would receive from Andromeda a reverse premium of £4 million. On 27 October 2003 Michael Becker wrote to AIB describing himself as the Kallakis family lawyer. The £4 million reverse premium would be provided by companies controlled by the Kallakis family. The loan application was approved by AIB on 30 October 2003. On the following day £18.8 million was paid into the client account of Mayer Brown, solicitors acting for AIB. From that sum £16.45 million was utilised by Andromeda to complete the purchase of Fitzroy House. The balance of £2.16 million was transferred to solicitors acting for Kallakis. That sum was dispersed to meet a liability of £69,000 in respect of another Kallakis matter and £2,653.37 was remitted to a Swiss bank account held by Michael Becker.

The whole edifice on which this ingenious financial proposal was based was a sham. Significant features of the deception, but by no means all, comprised:

Further and similar transactions took place. India Buildings in Liverpool was purchased for £43 million in January 2004. The loan from AIB was for £47.5 million. The sum sent to Michael Becker in Switzerland on the direction of Williams was £4,260,439.

In late 2006, 32 St James’ Square was purchased with a loan of £12 million from AIB. This time the head lessee was to be Oregon Finance Corporation, declared to be a Kallakis company, rather than a subsidiary of SHKP. Oregon was to provide a £3 million capital guarantee. Oregon was in fact a company with no net assets. A loan of £11.9 million was used to fund the purchase by Meadow Ridge Acquisition SA of which Michael Becker was the sole director. The deeds which provided guarantees for the rent payable under the head lease and, therefore, repayments of the loan, were worthless. In March 2007 the penny dropped with AIB that they should have some personal contact with SHKP. Kallakis told them that a director of SHKP’s treasury department would be passing through London on his return to Hong Kong and a meeting was arranged for 28 March. A person calling himself Jonathan Lee arrived for the meeting with the offender Williams. Kallakis, who was supposed to attend, cried off at the last minute. Lee had nothing to do with SHKP. He was an unknown individual who had been recruited to act the part of SHKP’s representative, which he did with some skill, but his efforts on behalf of the offenders were entirely fraudulent. He had been coached both as to the questions likely to be asked and the information to be provided.

In May 2007 a similar arrangement was made by which AIB advanced loans of £9 million and £1.2 million to purchase 8 Carlos Place, the offenders’ Mayfair office, and a property at Ennismore Gardens. The guarantor was again Oregon Finance.

In late 2007 AIB advanced £224 million for the purchase of 111 Buckingham Palace Road. This time guarantees from both SHKP and Oregon Finance were to be provided. False information was provided of Oregon’s asset position.

In October 2007 AIB was approached for a loan to purchase properties at 7 and 8 St James’ Square, Duke of York Street, and 7 Apple Tree Yard. The total sum lent was £152 million. After the purchase there was a surplus of £29.5 million. It was understood that this sum would be used to pay stamp duty and a reverse premium. In fact it was dispersed on the directions of the offender Williams for the benefit of the fraudsters.

In total AIB was persuaded to lend £743,345,000 to companies registered in the British Virgin Islands and controlled by the offender Kallakis. The sum remaining after payment of the necessary costs and expenses of purchase was approximately £77 million. From that sum money was dispersed to various destinations. £14 million went to Atlas Management. It was used as a fund from which to meet the shortfall between genuine rent received and the mortgage payments due to AIB. £22 million went to the Kallakis client account of Michael Becker in Lugano, Switzerland. £6.4 million went to the purchase of further property in Hungary and Mykonos, for the purchase of a ferry the subject of count 21, for the costs of purchase of a helicopter and private jet and for the purchase of a motor-car. A further £40 million went to companies associated with Kallakis. £5.6 million was dispersed to a number of different beneficiaries.

Count 21 represented a proposal of a somewhat different nature. The offenders were seeking a loan to finance the re-fitting of a ferry called M.V. Mercator II so as to transform it into a luxury super yacht worth €87 million. Their target was the Bank of Scotland PLC. The borrower proposed was the Mercator Shipping Corporation, a British Virgin Islands company owned by Oregon Finance. The success of the proposal depended upon satisfactory evidence of Oregon’s balance sheet position. It was represented that Oregon had a net worth of over $2 billion. The deed of loan contained a covenant that Oregon’s minimum net worth was £500 million. Evidence was produced in the form of profit and loss accounts for Oregon, a letter of confirmation from Michael Becker, and letters purportedly emanating from the Pacific Group and Atlas Management detailing companies and mortgaged assets owned ultimately by Oregon Finance and the Kallakis family trust which controlled them. These documents were utterly misleading and two of them were forged. The Bank of Scotland wanted to speak to the purported author of one of these letters. An elaborate arrangement was made by Williams to set up a virtual office in Athens and to instruct a person answering a telephone number at that office to say that the individual wanted was in a meeting and would ring back.

When in June 2007 the Bank sought further information as to the financial status of Oregon, Kallakis successfully bluffed his way to securing the loan and the agreement was executed on 17 April 2008. Oregon Finance was the guarantor. Mercator defaulted on its interest payments in September and October 2008. Work on the vessel had stopped. The guarantee was worthless. The total loss to the Bank of Scotland was €5.8 euros. At a meeting with Kallakis and Williams on 5 March 2009 Kallakis said the debt was not his responsibility; both Mercator and Oregon were “under water”. In fact Oregon was placed in liquidation on the petition of AIB in March 2009. Oregon subsidiaries had been used by Kallakis in 2006 to purchase a private jet for $44 million and a helicopter in 2008 for $8.4 million. These purchases had been made with loans from GE Capital guaranteed by Oregon. The jet and helicopter had been re-possessed and Oregon’s debts exceeded £61 million.

AIB discovered that Kallakis was a man of straw after receiving a tip-off from another bank to which he had made an approach. It was learned that he had a conviction for fraud under the name Stefanos Kollakis. AIB made enquiries of SHKB and realised in September 2008 that it had been conned.

In October 2008 AIB reported to the Financial Services Authority and the Serious Organised Crime Agency. Notice of default was served on the 14 Kallakis companies which had borrowed the money. On 21 November 2008 AIB formally called in the cross guarantees taken in respect properties which were worthless. All were sold on the same day to a property group at a paper loss of £56m. We shall explain that term when summarising the respondents’ submissions.

Kallakis was arrested at Bishopsgate Police Station on 23 April 2009. He read out a prepared statement in which he denied fraud and claimed to have been dealing with SHKB at arms length through a Hong Kong broker called Richard Lee. He acted with respect to the landlord purchasing companies only on the instructions of Michael Becker. He, Mr Kallakis, had no reason to doubt the authenticity of documents submitted to him. He made no further response to questions.

The offender Williams was arrested on 1 May 2009. He too read out a prepared statement. He claimed to have received threats from persons unknown in South East Asia but held himself ready to answer any questions which the SFO submitted in advance to his solicitor. On 13 October 2009 he read a further statement in which he blamed the Bank of Scotland for the collapse of the Mercator project. On 11 January 2010 he read his third prepared statement in which he said that AIB and some of its directors were carrying on a malicious campaign against both him and Atlas Management Corporation Limited.

On 19 May 1995 the offenders Kallakis and Williams were convicted in the names respectively of Kollakis and Lewis of conspiracy to commit forgery for which they were ordered to perform 160 hours unpaid work as part of the community service order. The fraud involved the sale of bogus heraldic crests. They changed their names before embarking upon the current frauds.

In his sentencing remarks the trial judge observed that both AIB and the Bank of Scotland had acted carelessly in their own interests. The Bank of Scotland in particular had made a loan on the mere assertion of a Swiss lawyer that Oregon Finance was good for the money. It had been advised by its own in-house lawyers as to the unsatisfactory nature of that assurance. Nonetheless, the judge observed that the offenders must themselves bear personal responsibility for their offending and that the carelessness of the Banks offered no mitigation. The central features of the fraud were the provision of forged and false information relating to the purported involvement of SHKB and the financial worth of Oregon. There were, the judge said, forged and false documents at every turn. Both offenders, he found, took an active part in the production and use of those documents. Elaborate subterfuge was used to ensure that the bankers got nowhere near the truth. The judge accepted that the offenders were taking advantage of a rising property market in the hope that they would never be called upon to make good a loss. The market would rise so as to obliterate any short fall and the fraud would not be discovered. The judge declined to place firm figures on any loss to AIB or personal gain to the offenders. The judge turned to the Sentencing Guideline Council’s (“SGC”) guideline for substantive offences of fraud. He observed:

“I am required by statute to follow those guidelines unless there is good reason not to do so in the interests of justice, and I must be prepared to justify any departure from them with sound and clear reasons.”

The judge identified the starting point for a banking or insurance fraud, fraudulent from the outset, in which a sum of £750,000 is obtained by means of a professionally planned fraud carried out over a significant period of time, as 5 years, with a range of 4 -7 years. There was a need, the judge observed, to leave head room for sentencing in respect of the most serious offences outside the range and up to the statutory maximum of 10 years.

The judge identified two questions: whether he should impose sentences outside the guideline range and whether he should impose consecutive sentences. The judge concluded that the count 1 fraud was not the worst offence of its type. In reaching this conclusion the judge took account of the fact that repayments to AIB were kept up until the summer of 2008 (by which time of course the alarm bells were ringing in the property market). The judge was influenced by the sentence of 10 years imprisonment recently imposed by Holroyde J upon Asil Nadir for fraud involving much larger sums and a very substantial breach of trust. As to the second question, the judge concluded that count 21 represented only the extension of the same conduct upon which the prosecution had relied to prove count 1. He did not consider that it was open to him to pass consecutive sentences merely because he might think that the statutory maximum was too low. In the result the judge took, in Mr Kallakis’s case, a starting point of 8 years imprisonment. He regarded him as the prime mover. The judge took account of the fact that Kallakis had not before served a sentence of imprisonment. There had been some honest business transactions. Neither of the offenders was in robust health and Mr Williams had lost his family. The judge imposed concurrent sentences of 7 years imprisonment and 5 years imprisonment respectively.

It is submitted on behalf of the Attorney General that the judge’s sentencing remarks reveal an error of approach. First, at page 3, paragraph 2 of the Guideline the SGC made clear that it did not apply to offences of conspiracy to defraud. Second, while it is accepted that consecutive sentences should not be a means by which disapproval of a statutory maximum sentence is expressed, the Sentencing Council’s (“SC”) guideline upon totality and offences taken into consideration applies to all sentences passed after 11 June 2012. At page 7 the SC specifies occasions on which consecutive sentences would ordinarily be appropriate. It is recognised that consecutive sentences may be imposed for offences of the same or a similar character where the overall criminality of the offender will not be reflected adequately by concurrent sentences. Specific examples include successive assaults on different victims on the same or different occasions. Third, Mr Edis QC submitted that the correct approach to the assessment of seriousness was explained in the appeal of Bright [2008] EWCA Crim 462, [2008] 2 Cr App R (S) 102. The then President of the Queen’s Bench Division, Sir Igor Judge, considered the circumstances in which a sentence at or close to the statutory maximum would be appropriate. It is not the court’s job to imagine the most serious of offences in the relevant category but to assess whether the offence is, within the relevant category, one of the utmost gravity. The court proceeded to observe that assertions of previous good character were unlikely significantly to reduce sentences imposed for offences of such gravity.

Mr Edis submits that the sentences do not adequately reflect the need for deterrence, one of the purposes of sentencing identified in section 142(1) Criminal Justice Act 2003. The fraud against AIB was on a truly massive scale. In respect of count 1 the starting point should have been, in Mr Kallakis’ case, at or near the maximum sentence of 10 years. A separate assessment of the seriousness of count 21 was required because, in the submission of the Attorney General, it charged a separate fraud on a different financial institution, and the overall scale of the offending could not justly and proportionately be accommodated by a concurrent sentence. The learned judge fell into error by failing to make an assessment of the overall criminality involved before reaching his decision as to whether a consecutive sentence was appropriate. As to the application of the purpose of deterrence in sentencing, the Attorney General accepts that it could not be suggested that these offenders were in some way responsible wholly or in part for the collapse of the banking system in the UK. However, it is submitted that sentence was being assessed against an economic backdrop in which bad lending against over-valued properties (represented by many different financial products and properties), with inadequate guarantees, had been the cause of a major banking crisis. In this case the whole scheme was a fraud. Fraud against banks should not be treated as victimless. Fraud on this scale caused or contributed to major loss or risk of loss to shareholders and, ultimately, the taxpayer.

The respondents submitted that the judge, who had presided at both trials, was in the best position to evaluate the seriousness of the offending by reference to the culpability of the offenders and the harm, if any, which they either caused or foresaw. It was submitted that the judge made no error of approach in assessing count 1 against the statutory maximum sentence of 10 years imprisonment and then posing to himself the question whether the sentence for count 21 should be concurrent or consecutive. It was conceded that the guideline for statutory fraud offences was not strictly applicable. However, both the Attorney General and the respondents acknowledged that the principles which inform that guideline were undoubtedly relevant. The judge did not apply the fraud guideline table since he adopted a starting point of 8 years for Mr Kallakis. The respondents argued that the judge was right to conclude that the count 1 offence was not of such gravity that the statutory maximum was deserved principally because it was unlikely that it would result in financial loss to AIB. He was also right to conclude that the additional offending revealed by count 21 did not require a consecutive sentence. There was a second victim but the deception exercised was in many respects similar to that used to commit the count 1 fraud.

The respondents submitted that no particular prominence is given in section 142(1) to the place of deterrence in sentencing. It is given equal prominence with (a) punishment, (c) reform and rehabilitation, (d) the protection of the public and (e) the making of reparation. No particular necessity for deterrent sentencing arose by reason only of the economic climate in which the offenders were convicted. The collapse of the banking system was caused by banks placing profit before prudence. They too were taking risks in the market; they too failed to anticipate the collapse of the property market in which they were taking those risks. The judge explicitly did not treat the negligence of the lenders as mitigation of the offences. The judge was right to conclude that the offenders had not targeted a vulnerable victim and he accepted mitigation on their behalf that they believed an inexorable rise in property values would prevent loss and conceal guilt. The sentences imposed were, it was submitted, appropriate by comparison with the facts in the appeal of Bright. In that case 1,000 employees had lost their employment. The total loss occasioned on collapse of IIG Plc was in the region of £1 billion. The court reasoned that a sentence of 7 years imprisonment was not inappropriate in the circumstances of that case. It was submitted that it must be relevant to the sentencing exercise that it was improbable that any net loss would be caused to AIB.

The terms of the eventual sale by AIB of the properties to the property company on discovery of the fraud contained a clause which entitled AIB to take 35% of any profit made on re-sale. The evidence before the judge was that one of the properties had been valued at £500m. That valuation was not conceded by the prosecution but it was reputable. If a sum of that magnitude was received on re-sale, all of AIB’s £49m loss, as it was subsequently calculated, would be wiped out and the bank would make a net profit of £100m. This is the explanation for the judge’s decision to treat the loss to AIB as purely a “paper” loss.

Nicholas David Andrew Levene appeals with the leave of the single judge against a total sentence of 13 years imprisonment imposed on 5^th November 2012 at the Crown Court at Southwark by HHJ Beddoe. Some weeks earlier, on 24^th September 2012, he had changed his plea to guilty on re-arraignment to 14 of the 19 counts on the indictment which he faced. The remaining 5 counts were ordered to lie on the file on the usual terms. The counts to which he pleaded guilty, and the sentences imposed in respect of each of them, are as follows: On count 1, an offence of obtaining a money transfer by deception contrary to section 15A of the Theft Act 1968, 5 years imprisonment; on count 2, false accounting contrary to section 17(1)(a) of the Theft Act 1968, 3 years imprisonment; on counts 3 and 4, allegations of fraud contrary to section 1 of the Fraud Act 2006, 5 years imprisonment. All these terms were ordered to run concurrently. On counts 7, 8, 10, 11, 13, 14 and 15, all allegations of fraud, terms of 8 years imprisonment were imposed and ordered to run concurrently with each other but consecutively to the 5 years on counts 1 to 4. Counts 6, 16 and 17, also allegations of fraud, each attracted terms of 6 years imprisonment ordered to run concurrently with the earlier terms. The total sentence was therefore 13 years imprisonment.

The maximum sentence on each of the offences to which the defendant pleaded guilty was 10 years imprisonment save for count 2 in respect of which the maximum term was 7 years imprisonment.

The case involved what has become known as a “Ponzi” fraud. The appellant had a background as a City trader, having worked in the financial sector since 1984. From about 2000 he traded honestly on his own account but during the period of his offending, 2005 to 2009, he sought and received substantial deposits from clients to make investments which he did not make. Instead he used the money as if it were his own. Some of the investors did receive their money back with what appeared to be substantial ‘gains’ but those gains were funded by money from new investors. As the level of new investment required to keep the fraud going grew so did the magnitude of the crash when inevitably it occurred.

The total sum obtained by the appellant by fraudulent misrepresentation, as reflected in the specimen counts to which he pleaded guilty, was £32 million. He spent over £18 million on lifestyle expenditure which included yachts, property, travel, vehicles, credit cards, schooling and family. However, between January 2005 and October 2009 he received approximately £252 million from investors of which only £183 million was returned. The overall loss which he caused is estimated as being in the region of £100 million. For a considerable period during the offending, worldwide freezing orders were in place when the appellant was being sued by investors for the return of monies. The appellant ignored those court orders. They were in place for two months in 2008 and three months in 2009.

Counts 1 and 4, although separated in time are connected. Tadco Ltd was a Jersey incorporated company owned by a Mr Rad. In April 2005 Tadco wanted to invest through the appellant in 140,000 shares in Imperial Energy. The shares were to be purchased and held in trust. The cost was £571,261.03. However, the appellant did not buy the shares; instead the money went into his bank account and then was moved to another account three days later. Mr Rad, believing the investment had been made, had no further contact with the appellant until May 2008 when he contacted him again intending to purchase further shares in Imperial Energy. At a meeting the appellant agreed to do so and two further tranches of money were paid. That money went into the appellant’s account and then to a company involved with betting and a firm of solicitors. Mr Rad subsequently chased up the transfer of those shares but received no response from the appellant. Tadco intended that the shares would be sold as part of a takeover, realising £3,875,000. At a meeting the appellant suggested to Mr Rad that rather than take the profits he should roll them over into a new rights issue being offered by another company but Mr Rad declined that offer. The appellant gave Tadco a cheque for £4,115,000 but it was dishonoured. Civil proceedings followed which included a freezing order. Some repayments were made to Tadco but not the total invested and certainly not the profits envisaged.

Count 2 was an offence of false accounting which occurred between 9^th and 30^th September 2005. Investigations revealed that a fraudulent system was up and running at that time as there was an email from the appellant to a man called Raphael Haas ordering forged bank statements. Those forged statements showed that money paid to the appellant by a Mr Caring was paid in favour of Spreadex Ltd as Mr Caring had intended. However the genuine statements showed that the money went to various other destinations and not to Spreadex.

Count 3 relates to funds placed in the hands of the appellant by Blenheim Properties Group which was Yigal Ahouvi’s holding company. Mr Ahouvi agreed for the appellant to place an order for £14.1 million worth of shares in Delek Global Real Estate, which would be held in trust for Blenheim Properties Group. In April 2007 Blenheim Properties Group transferred £14.9 million to an account in the name of Niblick Investments, a vehicle used by the appellant, and the appellant was instructed to buy the shares. Later that month the appellant falsely stated that he had purchased the shares. Instead the money was moved around different bank accounts in different countries over a significant period of time. Payments of substantial sums were made to various people including to a Mr Coleman. In May 2007 he was paid £1,313,969 and that payment was an example of the Ponzi method: the money from one victim used to pay off another’s ‘investment’. The appellant also transferred £5,320,000 in relation to a Contract for Difference (CFD) for 7,150,000 shares. City Index brokers, seeing the size of that transaction, also entered a CFD with City Index. The money invested by Blenheim was not used to buy the shares agreed. Three months after the deal was supposed to be completed the appellant was told to sell the shares and transfer the money. As the shares had never been purchased and there was no money to transfer the appellant tried to stall Blenheim. Blenheim pursued litigation against the appellant and a worldwide freezing order was obtained, in force between June and August 2008. Some repayments were made but they fell far short of the initial outlay.

Count 6 reflects investments which Julan Shah the director of DVS Property Ltd tried to make through the appellant. He sought to acquire shares in Lloyds TSB, Xstrata PLC, Natural Gas, ENEL, and Rio Tinto PLC. Of the first transfer of money the appellant used £100,000 to pay the Blenheim solicitors to keep them at bay. The rest of the transfers the appellant used for his own purposes and none of the money was used to buy the shares Mr Shah believed he was purchasing. Mr Shah did receive a payment of £311,430.12 in February 2009 and was told that it represented profits from the Xstrata and Lloyds shares. That was untrue as the money actually came from the victims in counts 7 to 11. However Mr Shah believed what he was told and so made further significant investments of £197,000 and £980,000. When Mr Shah attempted to obtain his returns the appellant stalled him. A deadline of August 2009 was missed and Mr Shah began litigation, claiming a loss due to his dealings with the appellant of £2 million.

Counts 7, 8 10 and 11 arise out of the appellant’s dealings with Brian Souter and his sister Ann Gloag who were well known business people in Scotland. In early 2009 they were indirectly approached by the appellant and persuaded to invest through him. After discussions with the appellant they agreed to invest £5 million in a forthcoming rights issue in Xstrata. The money was paid on 3 March 2009 to a firm of solicitors called Needleman Treon. Before the payments were made there had been substantial negotiations during which the appellant said that he had access to the rights issue, despite not being an existing shareholder, as he had connections from his days as a trader. There had been an over-allotment but he did not have enough money to take advantage. Four documents were drawn up: a letter of undertaking, a declaration of trust, a personal guarantee from the appellant, and a stock transfer form. The appellant moved the investment money of £5m to various accounts. An overdraft was cleared at a London bank account. On 16 March 2009 (the day a freezing order was obtained by Tadco) £1,922,000 went to a Jersey bank account. From the Jersey account money went to R3 Investments, another victim of the Ponzi scheme. There were also significant payments to the appellant’s bank account in Tel Aviv and from there to various places, including a payment to Tadco. Around £1 million was used to buy Xstrata rights and that showed a profit of £211,665. Mr Souter and Ms Gloag were persuaded to part with a further £5 million for the purchase of shares in HSBC. Once again the appellant failed to purchase the required shares and used the money to make various payments including to DVS Property and City Index. In August and September 2009 meetings took place. The appellant admitted he had only bought a fraction of the shares and had used the balance for spread betting. He confessed at that time to having a personal debt of £30 million and said he had gambling and personal problems.

Mr Bowman and Mr Clink were two further victims of the appellant’s fraud and their dealings with him are reflected in Count 13. They invested £499,800. Both had previously been employed by Orsus Property Consultants. Orsus invested through Ice Mountain Investments and Mr Bowman and Mr Clink were beneficiaries of Orsus. Once again the appellant proposed investments in Xstrata, HSBC, ENEL, and Rio Tinto. Mr Bowman and Mr Clink conducted some due diligence checks and believed the investment would be made through IAM and that the appellant was a director of that company. A written agreement was signed on 4^th March 2009 and the money for the investments went to the appellant’s Jersey bank account where he rapidly treated it as his own. He made payments to Freedex Ltd, DB Jewellery, Eagle Portfolio Corporation and a further payment to Blenheim’s solicitors. Ice Mountain believed that their original investment was now worth £2.5 million and wanted it back. No payment was forthcoming as no investments had been made. In September 2009 Mr Bowman issued a deadline and from the correspondence he received from the appellant it appeared that by that stage he had begun to lose track of what he had done with his investors’ money. Ice Mountain obtained a freezing order in respect of the appellant’s Jersey assets.

In March 2009 three American businessmen, Messrs Mermelstein, Klein, and Sofer, learned of an opportunity to invest in HSBC share rights through the appellant. The fraud the appellant perpetrated on them is the subject of Count 14. They instructed a solicitor, Simon Holden, to act on their behalf. The appellant told Mr Holden that he would be able to acquire the rights on behalf of his clients as he had contacts with large financial institutions. Documents were drawn up to clarify that the money the appellant was to receive was to be invested as directed. This was done at the time of the Tadco freezing order. The appellant received £702,739 from Mr Mermelstein, £344,043 from Mr Klein, and £688,078 from Mr Sofer. Again the money was not invested. In May 2009 Mr Holden chased the appellant for the profit from the investments and the appellant used delaying tactics. On 19 May the appellant emailed Mr Holden saying that £2,557,200 would be transferred within 48 hours. It was not. On 3 June 2009 Mr Holden’s firm received £1 million on behalf of the appellant and a further £1.6 million on 5 June.

However, £1 million came from the victim in count 15 who was Glenn Coleman. He made an investment with the appellant in 2006 and believed he had received a profit on that investment in 2007 when he received a payment of £1.3 million. In fact the source was the Blenheim investment. In 2009 the appellant told Mr Coleman that he had acquired options to purchase shares in ENEL. He said they were very high yield and Mr Coleman could double his money in 24 hours. Mr Coleman pledged £1 million. Once again the appellant failed to purchase the shares and used the money for his own purposes. When Mr Coleman requested his money the appellant tried to delay him. Mr Coleman confronted the appellant in late August 2009 and was told that he had not bought the shares but had invested the money elsewhere. Mr Coleman demanded his money back but did not receive it.

The final two counts to which the appellant pleaded guilty were counts 16 and 17 which relate to the appellant’s dealings with Russell Bartlett, a director of the R3 Investment Group. In October 2008 he transferred £1 million to the appellant for payment in respect of the rights issues of RBS and Barclays PLC. That was eventually repaid to Mr Bartlett with a return. In May 2009 Mr Bartlett, having previously received an attractive profit, was induced to make a further investment. He transferred £500,000 to the appellant to purchase shares in ENEL. The appellant used most of that money to pay existing investors to help keep the fraud going. The appellant also approached Mr Bartlett about a rights issue in Rio Tinto. However Mr Bartlett first wanted to know when he would get his reward from the ENEL investment he had made. The appellant said he would get that money on 5 July 2009. Mr Bartlett agreed to invest £700,000 in Rio Tinto. The money was transferred to the appellant’s bank account in Israel and was rapidly dispersed. It was not spent on any Rio Tinto shares. Mr Bartlett chased the appellant for a return on his investments and read in newspapers of the situation with Mr Souter and Ms Gloag. The appellant told him there had been an accounting error and that money had been repaid to Mr Souter and Ms Gloag. It had not. Eventually on 28^th September 2009 the appellant wrote to Mr Bartlett and acknowledged that he owed R3 Investments £1,774,000. It was never paid.

The fraud eventually collapsed under the weight of litigation and on 7^th October 2009 the appellant was declared bankrupt following a series of undefended judgments against him.

On 15^th October 2009 the appellant and his solicitor attended court in relation to civil proceedings arising from his offending. He was, by then, aware that at some stage there would be an inquiry into his activities. From court the appellant went with his solicitor to the Serious Fraud Office, self- reported and was interviewed under caution. This marked the commencement of the serious fraud investigation. The appellant was further interviewed on 1^st February 2011. He made admissions as to the facts of his activities but maintained that, with the agreement of his investors, he had a free hand as to the investments he made. On 28^th February 2011 the appellant made no comment when further interviewed and on being charged on 4^th March 2011 he said that he had done nothing wrong.

In due course, the case was transferred to the Crown Court and a plea and case management hearing was listed for 12 July 2011. That hearing was adjourned by Judge Rivlin QC until 28 September 2011 and in doing so the judge stated that maximum credit would be given for pleas of guilty entered on that date. On 28 September 2011, at the adjourned PCMH, the appellant, consistent with his case as put in interview and when charged, pleaded not guilty to all counts in the indictment and the trial was fixed for a 12 week slot starting on 22 October 2012. In September 2011, while Mr Edis QC and Mr Burke QC were engaged together in another case, Mr Burke informally told Mr Edis that in his judgment the case against the appellant was unlikely to be a fully contested trial.

On 3 October 2011, there was a bail hearing before HHJ McCreath at which Mr Burke said that the appellant’s case was that he had had a free hand to invest money as he saw fit; the appellant’s clients perceived him as a man with the Midas touch and “they were content for him to do what was necessary”. There was a further directions hearing on 17 February 2012 before HHJ Leonard QC who reminded the defence that any credit for a plea would rapidly diminish after 6 March 2012.

During 2012, the defence were seeking disclosure and the time consuming and expensive exercise of reviewing documentation, a substantial proportion of which the appellant was familiar with as the documents had been obtained from him, was carried out. Although it is right to note that prosecuting counsel, on the basis of Mr Burke’s informal indication, did not carry out all the pre-trial preparation they would otherwise have undertaken, such as the preparation of admissions and a jury bundle, the prosecution, nevertheless, had to make some preparations for trial and had to warn its witnesses, many of whom lived abroad, to be available for trial. At the end of July 2012, the appellant’s solicitors informed the prosecution that it would be inappropriate for them to file a defence statement. Mr Burke contacted prosecuting counsel about the case in the first half of August 2012, and matters were left on the basis that a consultation would be held with the appellant as a result of which it was expected that an indication would be given of proposed pleas. Discussions as to the acceptability of pleas followed and that resulted in the listing of 24 September 2012.

The judge said that the offending for which he had to pass sentence was, as far as he could tell, unprecedented for a fraud of this type. The appellant’s offending was on a “massive scale with the whole panoply of aggravating features”. Those features included the following: This was a well planned and professionally executed fraud; it involved huge sums and huge profits; there were multiple victims whose trust the appellant had betrayed; the fraud had extended over a long period of time; the appellant had taken steps to keep the fraud concealed and moved money out of the reach of his creditors when the fraud began to unravel; losses caused by the appellant’s conduct amounted to approximately £100 million. The judge also considered the mitigating features of the case. Those he identified as: the appellant’s self reporting to the Serious Fraud Office; the appellant had shown some signs of remorse although the judge said it was difficult not to see those signs as in reality self pity; the appellant’s previous good character and, of course, his guilty pleas.

Mr Burke QC, on behalf of the appellant, concedes that because of the scale of the appellant’s fraudulent and dishonest conduct, the sentencing judge was justified in taking a starting point in excess of the maximum term available for any individual count and in order to reach that starting point to consider imposing consecutive sentences. However, he pursues two grounds of appeal. Firstly, he submits that the starting point for sentence of 17 years taken by the judge was out of line with earlier decisions of this court and is, therefore, manifestly excessive. He relied in particular on a comparison between the facts of the present case and those considered in the two judge court (Dobbs and Globe JJ) which heard the appeal of Shephard [2012] EWCA Crim 1523. Secondly, Mr Burke argued that the discount given to the appellant, for his guilty pleas, was inadequate in the circumstances of this case.

Mr Burke submits that the judge wrongly rejected his submission that the fact the appellant’s business activities were not fraudulent from the outset mitigated the seriousness of his offending.

Mr Burke also drew our attention to what he said is the absence of some aggravating features present in other frauds. The victims of these frauds were not “vulnerable” victims in the sense referred to in the Definitive Guideline issued by the Sentencing Guidelines Council “Sentencing for Fraud – Statutory Offences” (page 19, paragraphs 4 and 5); the appellant was not an organiser or a planner or prime mover in a fraudulent enterprise carried out by a number of individuals acting together (see Definitive Guideline, page 8, paragraph 24) although it is right to say that the appellant engaged a professional fraudster, Raphael Haas, to produce fraudulent documentation for him; the losses suffered in this case by individual investors had no wider consequence to the banking and financial system as had been the case, for example, in Gokal (unreported 11 March 1999).

At page 3, paragraph 2 the SGC’s guideline on sentencing for the statutory offences of fraud appears the following:

“2.

This guideline applies to sentencing for statutory offences of fraud. This guideline does not cover the common law offence of cheating the public revenue or conspiracy to defraud. The common law offence of cheating the public revenue is generally reserved for the most serious and unusual offences (Mavji [1987] 84 Cr App R 34) and where a sentence ‘in excess of the statutory maximum for other offences would be...proper’ (Ward [2005] EWCA Crim 1926). As such cases are unusual, no proposals are made for sentencing offenders convicted of this offence. It would be open to a court to have regard to the principles expressed in the guideline for fraud against HMRC when sentencing an offender convicted of cheating the public revenue, but it should be used only as a point of reference as higher starting points are likely to be necessary. Sentencers should continue to refer to existing guidance from the Court of Appeal (Criminal Division) when sentencing these offences.”

Sentence for cheating the public revenue is at large while the maximum sentence for conspiracy to defraud is 10 years imprisonment (section 12(3) Criminal Justice Act 1988). In Mavji this Court observed that the common law offence of cheating the public revenue was properly charged in a case of serious VAT fraud notwithstanding that the statutory offence of fraudulently evading payment of VAT was available for more conventional offences and, in an appropriate case, the Court should not be inhibited from imposing sentences in excess of the maximum for the statutory offence.

At page 4, paragraph 7 the guideline states that the primary consideration when sentencing fraud offences is the seriousness of the offending behaviour. The court must have regard to the five purposes of sentencing set out in section 142(1) Criminal Justice Act 2003. In Wilkinson [2010] 1 Cr App R (S) 628 the Court of Appeal confirmed that it would be for the sentencer to determine how those five purposes should be applied on the facts of the case. In that case, which involved gun crime, the protection of the public was paramount.

At page 5, paragraph 10 the guideline reminds the sentencer of the provisions of section 143(1) of the Act:

“(1)

In considering the seriousness of any offence, the court must consider the offender’s culpability in committing the offence and any harm which the offence caused, was intended to cause or might foreseeably have caused.” [emphasis added]

At pages 5-7 the guideline explains the application of the guideline Overarching Principles: Seriousness to offences of fraud. Fraud offences generally involve the highest level of culpability because the offender intended to bring about a gain to the offender or another or to cause loss or the risk of loss to the victim. Planning and ‘professional’ activity will aggravate the offence. Fraud may cause considerable social and economic harm beyond its immediate financial impact. In assessing harm the primary consideration is the loss to the victim or to the community at large. Where the harm which results is greater or lesser than that intended the harm caused should be judged in the light of the offender’s culpability. In general the greater the loss the more serious will be the offence but financial loss may not reflect the full extent of the harm caused by the offence. The court should also take into account:

As to (e) at paragraph 52 above, the guideline advises (paragraph 20) that in cases where the loss resulting does not correspond to the loss intended the sentencer should select a starting point based upon the amount obtained and adjust the assessment of seriousness to reflect the degree of loss actually caused by the offence. Where there was no intention to cause loss (for example, because the offender intended to obtain credit through fraud and to repay the sum advanced) “a court should use the starting point corresponding to no financial loss and, where a loss occurs, adjust the assessment of seriousness to reflect the degree of loss”. When the fraud is unsuccessful and results in no loss “the court should use the starting point corresponding to the amount which the offender intended to obtain and adjust the assessment of seriousness to reflect the fact that no loss has resulted”.

At pages 7-11 the guideline identifies aggravating and mitigating factors of seriousness. Higher levels of culpability are indicated by planning, an intention to cause more ‘harm’ than resulted, operating in groups, ‘professional’ offending, a high level of profit, an attempt to conceal or dispose of evidence, deliberate targeting of vulnerable victims and abuse of a position of trust. A more than usually serious degree of harm is indicated by multiple victims, vulnerable victims and property of high value to the victim or substantial consequential loss. Of particular relevance to the assessment of seriousness are the number of offenders and their role in the offence, fraud carried out over a significant period of time, the use of another person’s identity and the lasting effect on the victim. Mitigating factors include peripheral involvement and that a legitimate entitlement became fraudulent by non-disclosure. Personal mitigation may comprise a voluntary cessation of offending, complete and unprompted disclosure, voluntary restitution and, perhaps, exceptional financial pressure.

The guideline makes provision for sentencing in banking and insurance fraud and obtaining credit through fraud at pages 23 and 24. Starting points and ranges are selected which are based upon the “amount obtained or intended to be obtained” for all three categories of fraud. It follows that the starting point will be identified by reference to the sum obtained or intended to be obtained but will be adjusted downwards to reflect the fact (paragraph 20 of the guideline), if it is the case, that no loss or reduced loss resulted.

At section D, pages 13-15, the guideline draws attention the range of additional orders available to the sentencing court including, for example, confiscation, compensation, deprivation, financial reporting and serious crime prevention orders. We emphasise that the impact of such orders has formed no part of the submissions we have received in the present cases.

The general principle identified at page 5 of the guideline is:

“1.

All courts, when sentencing for more than a single offence, should pass a total sentence which reflects all the offending behaviour before it and is just and proportionate. This is so whether the sentences are structured as concurrent or consecutive. Therefore, concurrent sentences will ordinarily be longer than a single sentence for a single offence.

2.

It is usually impossible to arrive at a just and proportionate sentence for multiple offending simply by adding together notional single sentences. It is necessary to address the offending behaviour, together with the factors personal to the offender as a whole.”

As to the question whether sentences should be consecutive or concurrent the guideline states:

“There is no inflexible rule governing whether sentences should be structured as concurrent or consecutive components. The overriding principle is that the overall sentence must be just and proportionate.”

At page 6 the guideline states that the general approach should be to (1) consider the sentence for each individual offence, referring to the relevant sentencing guidelines, and (2) determine whether the case calls for concurrent or consecutive offences. At pages 6 and 7 the guideline gives examples of circumstances in which concurrent and consecutive sentences may be appropriate. Where a series of smaller thefts or frauds is committed the sentencer will (page 6) generally assess the seriousness of the offending by reference to its overall criminality and impose concurrent sentences each representing that assessment. However (page 7), offences of the same or a similar character may attract consecutive sentences if the overall criminality revealed would not otherwise be sufficiently reflected, particularly where there are separate victims or offences committed on separate occasions against the same individual. Examples of specific situations are as follows:

“When sentencing for similar offence types or offences of a similar level of severity the court can consider:

▸ whether all of the offences can be proportionately reduced (with particular reference to the category ranges within sentencing guidelines) and passed consecutively;

▸ whether, despite their similarity, a most serious principal offence can be identified and the other sentences can all be proportionately reduced (with particular reference to the category ranges within sentencing guidelines) and passed consecutively in order that the sentence for the lead offence can be clearly identified.

When sentencing for two or more offences of differing levels of seriousness the court can consider:

▸ whether some offences are of such low seriousness in the context of the most serious offence(s) that they can be recorded as ‘no separate penalty’ (for example technical breaches or minor driving offences not involving mandatory disqualification);

▸ whether some of the offences are of lesser seriousness and are unrelated to the most serious offence(s), that they can be ordered to run concurrently so that the sentence for the most serious offence(s) can be clearly identified.”

Our attention has been drawn to previous decisions of the Court. In Attorney General’s References 48-51 of 2002 (Paulssen and Others) [2002] EWCA Crim 3165, [2003] 2 Cr App R (S) 36 the offenders were sentenced for conspiracy to commit a ‘Ponzi’ fraud. An investment firm which was dishonest from the start carried on business in the office of a solicitor. It attracted funds of $9.8m which were not invested as represented but distributed to the offenders. The fraud created victims of many small investors. The total loss was $7.7m. Three of the four offenders played key roles in the fraud. The Attorney General invited the Court to hold that the starting point should have been at or near the 10 year maximum. There was, by reason of the breach of trust with investors, a comparison to be drawn with the guidance case of Clark [1998] 2 Cr App R (S) 95 upon sentencing in cases of theft from employers. For thefts of £1m or more sentences of 10 years or more may be merited. The Court derived ‘considerable assistance’ from Clark. The Court concluded that the fraud conducted was not the most serious of its kind because the sum obtained could have been higher and the fraud was not in operation ‘over long’. There was no deliberate targeting of small investors and there had been a prolonged delay between discovery and trial. In these circumstances the Court accepted a submission that the starting point should be 7 years.

Bright [2008] EWCA Crim 462, [2008] 2 Cr App R (S) 102 was another case of conspiracy to defraud but its form was very different. The offender controlled a group of insurance companies, the Independent Insurance Group. The business was commenced in 1988, developing into the ninth largest insurance company in the UK. In 2001 the group was placed in liquidation. Ultimate losses were likely to be £1 billion and 1,000 employees had lost their jobs. £400m had been paid to policy holders by the Financial Services Compensation Scheme funded by the insurance industry. It was not suggested that the company had been set up as a vehicle for fraud. The dishonesty was a response to trading difficulties. The appellant and others orchestrated the manipulation of the company’s accounts and reserves to give the false impression that it was trading profitably while concealing the existence of claims worth many millions of pounds. Secondly, the existence of reinsurance contracts which were unlikely to be worth their face value was concealed having the effect that the balance sheet was inflated. The company declared a profit of £22 million in 2000. It had in fact suffered a loss of £30 million. The sentencing judge found that the position of the company had been misstated by more than £220 million. The appellant’s activities had put beyond reach the possibility of a conventional rescue package. Many in the industry had been deceived: employees, directors, accountants, auditors, shareholders, creditors and policy holders. The appellant had blamed innocent employees for the dishonest practices of the company.

The trial judge accepted submissions that the nature of the deception was indistinguishable from a conspiracy to commit the statutory offence of fraudulent trading, the maximum sentence for which was 7 years and not 10 years imprisonment, and that although there had been convictions for two offences of conspiracy to defraud there was one overall conspiracy which had been split for case management reasons to reflect two different forms of deceit. This Court held that the judge had not been bound to treat the maximum notional sentence as 7 years imprisonment. Conspiracy to defraud had been, by reason of the variety of components of the fraud, the correct charge and the maximum sentence was 10 years imprisonment.

The question arose as to the circumstances in which it might be appropriate to impose the maximum sentence. Sir Igor Judge, then President of the Queen’s Bench Division, said at paragraphs 29 and 30:

“29.

The maximum sentence permitted by statute is, of course, very rarely imposed, and nowadays when there has been a guilty plea, effectively never. Such sentences should be reserved for those cases which, at the end of the trial and within the statutory context, can fairly be regarded as crimes of the utmost gravity. It is sometimes loosely said that the maximum sentence should be reserved for the worst case of its kind, and from this, imaginative counsel for the defendant will urge examples of cases of greater criminality than the offence established against his client. The argument however is founded on the misapprehension that if a realistically more serious case can be imagined, the imposition of the maximum sentence is precluded. That is why we repeat, the maximum sentence permitted by statute is reserved not for the worst possible case which can realistically be conceived, but for cases which in the statutory context are truly identified as cases of the utmost gravity.

30.

The principle is long established. In R v Amber and Hargreaves, unreported, November 24, 1975, but referred to in Current Sentencing Practice at A1-4CO1 a case involving offences of corruption in the context of bribery of prison officers, Lawton LJ observed:

“It is of course a principle of sentencing that maximum sentences should only be passed for the worst kind of offence. But it is to be borne in mind that when judges are asking themselves whether they should pass the maximum sentence, they should not use their imagination to conjure up unlikely worst possible kinds of case. What they should consider is the worst type of offence which comes before the court and ask themselves whether the particular case they are dealing with comes within the broad band of that type. Where the maximum sentence is low, the band may be wide.”

It is rare for sentencing decisions from the mid 1970s to continue to provide assistance over 30 years later. However the principle was recently endorsed in R v Butt [2006] 2 CAR (S) 364, where it was said that the “enunciation of principle bears repetition”. We repeat and endorse the principle.”

The appellant argued that his personal mitigation had not been reflected sufficiently in sentence. At paragraph 35 the Court said:

“35.

Our conclusion is that in this particular case not much, if any, allowance would be appropriate. It is not unknown for major white collar fraudsters - and that is what the appellant is – to be individuals of apparent impeccable rectitude and good character. Indeed this very reputation helps them to establish this standing in business which, for a variety of different reasons, later becomes the vehicle for fraud. At the time when he was committing these offences, he was a President of the Institute of Insurers, apparently a pillar of respectability. Suspicion would not fall on him, and if it did, could be brushed aside as unworthy. It is always sad to see a successful individual fall from heights which he has achieved through his own hard work. But the appellant’s criminal behaviour means that all his wounds were and are self inflicted.”

The Court considered a submission that delay between the commencement of the investigation and trial was a feature of personal mitigation. At paragraph 37 the President said:

“47.

It was suggested that some allowance should have been made for the delay between the first investigations with the appellant and the date of sentence. He was first interviewed in January 2002 by the liquidators of the company, and by the Financial Services Authority in November 2004. The prosecution, applying that term broadly, was not responsible for any delay. The investigation involved study of literally a million documents. During his interviews the appellant was concerned to conceal the truth rather than reveal it. The case therefore had to be prepared and it eventually proceeded as a trial. The practice of allowing a discount against sentence to a defendant who pleads guilty is pragmatic. In this case an honest response to the interview process, and an indication of a guilty plea, would have avoided a huge investigation and a mammoth trial. It would be absurd for the defendant whose criminal activity requires massive investigation and whose lies contribute to the delay in bringing him to justice, then to seek to take advantage of the delay as a feature in mitigation.”

The Court concluded that even if the correct approach had been to treat 7 years as the maximum sentence of imprisonment (which it was not) it was unlikely that the Court would have reduced the trial judge’s sentence of 7 years imprisonment. No reduction was appropriate.

In Hibberd and Allen [2009] EWCA Crim 652 the appellants were charged in two counts with conspiracy to defraud, each representing deception practised on two different losers, Royal Bank of Scotland PLC (“RBS”) (count 1) and Lombard North Central PLC (“Lombard”) (count 2). The sentencing judge concluded that the appellants had funded their failed business by obtaining large sums of money by way of loans dishonestly from the losers. RBS lost over £1.586 million and Lombard £450,000. Much of the proceeds had been expended on high living. The judge imposed consecutive sentences on each count. Hooper LJ giving the judgment of the Court said at paragraph 41:

“41.

Mr Cox and Mr Corre complain that the sentences, relating as they did essentially to the same fraud should not have been consecutive. This court is usually concerned not with whether sentences are concurrent or consecutive but whether the total sentence is wrong in principle or manifestly excessive. On the other hand, the victims of the two counts, RBS and Lombard, were part of the same conglomerate and Clark and Patterson shared the same office. For that reason we think that in this case the maximum sentence which could have been passed for the two counts is 10 years’ imprisonment, being the maximum sentence for conspiracy to defraud, see Criminal Justice Act 1987, section 12(3). But it is an aggravating feature of the case against the appellants that they were involved in fraud which had two distinct manifestations: see Bright [2008] EWCA Crim 462 at paragraph 13...”

The Court reduced Hibberd’s sentence upon count 2 to 12 months imprisonment consecutive to the judge’s sentence of 4 years upon count 1 in order to achieve an overall sentence of 5 years imprisonment in his case. Allen’s sentences of 2 ½ years and 12 months imprisonment consecutive were upheld.

Another ‘Ponzi’ scheme was the subject of the appeal against sentence in Pollett and Others [2013] EWCA Crim 359. The target of the fraud was largely the British expatriate community in Mallorca. Many of the investors were retired and unsophisticated in financial matters. Between 2001 and 2009 some £10 million was invested. Just over £4.6 million was returned to investors and over £5 million was lost. The lion’s share of the benefit went to Hirst who had pleaded guilty to an earlier, less sophisticated, scheme in 1992. He was sentenced to 9 years imprisonment for the count 1 conspiracy to defraud and concurrent sentences for money laundering offences charged in counts 9 and 11. Rafferty LJ said at paragraphs 56 and 57:

“56.

This case is said to be by no means the worst of its type. There are more sophisticated frauds - cases where the main players understand how authentic investment works and create a complex camouflage which deceives seasoned professionals and distorts markets. There are cases where the fraudsters steal every penny invested and divert it beyond the reach of the victims and the authorities. The starting point in this case should have been of the level identified in Paulssen and Hibberd. Hirst should have received tangible credit for his plea. Although very late, it was not preceded by the creation of an elaborate factual defence which the Crown had to refute. Hirst suffers a very serious illness. Only modern medication keeps alive. The court should in its discretion take this into account for a man of his age.

57.

Refusing Hirst leave, the single judge said:

“Although emphasis is placed by the applicant on AG Ref 48-51 of 2002 [2003] 2 Cr. App. Rep.(S) 36, none of the three considerations identified in paragraph 28 of the judgment of Kennedy LJ so as to drive the court to a sentence below the statutory maximum applies here (with regard to the first consideration, in view of the time which the scheme was operated). There were additional aggravating considerations in the applicant’s case, not least his previous comparable offending and his involvement of others, including his wife and his innocent son. The Judge’s sentencing observations were wide-ranging but, as I read them, he weighted the relevant considerations and was not influenced by irrelevant matters. Although the sentence that he passed was severe, it was not so long that the Court of Appeal should reduce it.”

58.

We agree...”

In Shephard (paragraph 46 above) the appellant, who had previous convictions for fraud, pleaded guilty to charges of conspiracy to use false instruments and conspiracy to commit fraud by false representations. He obtained loans of £90,000 and €20m by providing false and forged documents as evidence of personal wealth. Some of the money was recovered. The sentencing judge adopted a starting point of 11 years and discounted it to 7 years imprisonment for personal mitigation and the guilty pleas. On being informed that the maximum sentence on each count was 10 years imprisonment the judge reduced the total sentence to 6 years 3 months. Globe J, referring to Leaf [2007] EWCA Crim 802, noted that consecutive sentences could properly have resulted in a sentence which exceeded the maximum for each count. The court rejected the argument that 10 years was the ceiling for sentence. In Gokal (11 March 1999, unreported) the prosecution arose from the BCCI fraud causing losses of hundreds of millions of dollars. A sentence of 14 years after a trial was upheld. In H and P [2006] EWCA Crim 2385 a starting point of 12 years was upheld where loss to the Revenue was £41.5m but the appellant was not the organiser. In Shephard the court reduced the starting point to 8 years. The court did not consider that the offences were the most serious of their kind and there was little or no evidence of personal gain.

On the subject of consecutive sentences the court requested and has received from the parties four previous judgments of the Court of Appeal in cases of conspiracy to cheat the revenue. In Attorney General’s Reference No 136 of 2006 (Craig Johnson) [2007] EWCA Crim 2837, the Court considered sentences of 10 ½ years imprisonment for an MTIC fraud and 2 years imprisonment consecutive for money laundering the proceeds of a second fraud of the same character, charged in a separate indictment. The Attorney General argued that since the sentencing judge had indicated that a sentence of 8 years imprisonment for the second offence would have been appropriate had the judge been sentencing for that offence alone, the judge had made too great a discount for the principle of totality. Johnson argued that the sentence of 12 ½ years in total was excessive. Hughes LJ, giving the judgement of the Court, said at paragraphs 19 and 20:

“19....The way in which courts deal with the problem of total sentence is ordinarily to assess the whole of the criminality before them and to arrive at an appropriate total sentence.

20.

Accordingly, we approach this problem by asking ourselves if the judge had been dealing with both the first indictment and the money laundering offence on the second indictment, with a trial on the first and a plea of guilty on the second, all at once and had passed a sentence of 12 and a half years, would that have been unduly lenient. We accept that it could have been longer but we are satisfied that it could not be said to be unduly lenient. It is a very substantial sentence. In those circumstances we do not propose to entertain the application to revise the sentence upwards and for the same reasons we refuse the application of Johnson to revise the sentence downwards.”

In Namer [2007] EWCA Crim 2749, [2008] 2 Cr App R (S) 24 (134), the Court approved the use of consecutive sentences where the appellant had returned to the same type of fraud on the revenue after an interval of 18 months.

In Randhawa and Others [2012] EWCA Crim 1, [2012] 2 Cr App R (S) 53 (298) the Court upheld a sentence of 15 years imprisonment imposed after trial for an MTIC fraud in which £19m had been fraudulently reclaimed from the Revenue and £7.8m actually paid. The appellant was an organiser of the fraud. In Ravjani [2012] EWCA Crim 2519 the Court held not arguably excessive a sentence imposed of 17 years imprisonment after the trial of a principal offender whose fraud resulted in loss to the Revenue of £100m.

In our judgment, there is no ambiguity in the principles to be applied in sentencing for the most serious offences of fraud, nor in the assessment whether consecutive sentences should be imposed to reflect the overall seriousness of similar offending. No review of the appropriate level of sentencing for serious fraud is required on the facts of the present cases. We acknowledge that in individual cases, of which this is one, the judgment may not be a simple one.

While the SGC’s guideline on statutory offences of fraud expressly did not apply to the present offences of conspiracy to defraud (because the offence of conspiracy may and usually does embrace a wider range and duration of deceptive behaviour) the underlying principles of sentencing for fraud apply both to the substantive offence and to conspiracy to defraud. The fact that the maximum sentence for the statutory offence and for conspiracy to defraud is the same makes the comparison of some, if ultimately limited, value.

The count 1 conspiracy in which the victim was AIB was clearly more serious than the count 21 conspiracy in which the Bank of Scotland was the victim. Seriousness comprised both culpability and harm. The culpability of the offenders was at the highest level because they set out and persisted over a significant period with planning, determination and audacious dishonesty to commit a commercial fraud with international proportions and by that means to obtain sums of over £730 million from the loser. By section 143(1) Criminal Justice Act 2003 ‘harm’ is to be treated as harm which was intended by the offender, or was caused by or may foreseeably have been caused by the offence. A downward adjustment in property prices could have resulted in enormous losses to the bank. That is the risk which any investor in the property market takes and is the risk taken by these offenders with the bank’s money. The security which the offenders knew to be the essential prerequisite of the loans was an outrageous but apparently convincing sham. Good fortune may in the result prevent a net loss to AIB. However, the consequence of the fraud was not merely the substantial risk of financial loss to the bank but included the disruption of its normal business and the need to take exceptional measures to investigate and protect the bank’s position. Nevertheless, as the guideline on the statutory offence recognises, a fraud which results in substantial financial loss, particularly if the loss is suffered by vulnerable victims deliberately targeted, is more serious than a fraud which does not.

In our view, notwithstanding the judge’s mistaken belief that he was bound by the statutory offence guideline, his approach to identification of the starting point accorded with the principles stated in the guideline and the decisions of the Court of Appeal. The judge rejected the submission that lack of care by the bank constituted mitigation of the offence. The offenders were the protagonists; they were not led into temptation by lax standards of security. There remained a risk of actual loss of some millions of pounds. The judge acknowledged that the maximum sentence was available for offences of the most serious kind and concluded that a starting point outside the range offered by the guideline was required. However, he concluded that the present fraud was not in the most serious category. The judge reached that view because the offenders had maintained mortgage repayments until the summer of 2008. In this conclusion we consider that the judge was generous to the offenders. The reason why it was essential that the fraudsters continued to make mortgage repayments was that if they did not the sham would be discovered and the tap which provided funding for the continuing fraud would be switched off. In our judgment, had substantial financial loss to the bank been established the starting point for the count 1 offence would properly have been the maximum for the offence. We conclude, however, that a starting point of 8 years for Mr Kallakis would properly reflect the probability that, unlike the most serious such offence, no financial loss would be realised. In the result, we do not consider that the judge’s selection of a starting point of 8 years can be regarded as unduly lenient.

The judge reduced his starting point by 12 months. He said the offenders had made some efforts to rehabilitate themselves before engaging in the present frauds. He made allowance for indifferent physical and mental health respectively and, in Mr Williams’ case, the collapse of his marriage as a result of these offences. In our judgment, there was no significant personal mitigation available to the offenders. They had committed fraudulent offences together on a previous occasion and Williams had served a custodial sentence for passport offences. We are doubtful whether any discount from the starting point was appropriate.

We turn, secondly, to the question whether a consecutive sentence should have been imposed for the count 21 fraud. We do not accept that a consecutive sentence would have indicated a view that the maximum sentence for conspiracy to defraud was inadequate. On the contrary, the effect of the concurrent sentence is, in our view, to give the impression that the offenders have entirely escaped the consequences of a serious fraud in which substantial loss has resulted. It is true that the nature of the fraud was similar and that it overlapped in time with the count 1 conspiracy but Bank of Scotland was a separate victim, separately targeted, and, unlike the count 1 offence, a substantial loss was realised. We have no doubt that a consecutive sentence for the offence was required, subject to the principle of totality. Had the offenders been sentenced for the count 2 offence standing alone it is our view that the appropriate starting point after a trial would have been 6 years imprisonment.

Standing back from the individual counts, we consider, finally, what was the just and proportionate sentence for this offending as a whole. In our judgment that sentence was, for the leading and dominant offender, Kallakis, 11 years imprisonment, and, for his lieutenant Williams, 8 years imprisonment. We shall achieve that result by quashing the concurrent sentence of 7 years imprisonment upon count 2 in Kallakis’ case and substituting a sentence of 4 years imprisonment consecutive; and by quashing the concurrent sentence of 5 years concurrent upon count 2 in Williams’ case and substituting a consecutive term of 3 years imprisonment.

We turn first to the assertion that the judge failed to accept mitigation that the fraud was not perpetrated from the outset. A careful reading of the judge’s sentencing remarks shows that the judge did not proceed on the basis that the appellant’s conduct was fraudulent from the outset. What the judge rejected was the submission that the fraud developed as the appellant began to face trading losses. In rejecting that submission, the judge said:

“In his mitigation to me, Mr Burke sought to suggest that it was not a fraud from the outset but one which developed as you began to face trading losses. I cannot accept that. If, as he put it, you had a net worth in 2004 of some £16m, and you sold in February 2005, your shares in the company known as Trio for £4.13m, I can see no excuse or explanation at all for the commission by you of count 1 on this indictment in April 2005…What is revealed by that offence and by what so similarly follows over the following four years is a determined and utterly dishonest course of offending for your own aggrandisement…If what Mr Burke told me is right about the success story which you apparently represented in 2005, then had you chosen to follow a path of honesty thereafter, you would probably have made, as you had before, a very good living that would be the envy of most. But for you that was not enough. I have, of course, read all that has been put before me on your behalf … It is probably what your father wrote which struck me as much as anything. He describes yours as a lifestyle beyond comprehension and understanding, explained only by that monster greed. In my judgment, he is right. That is the explanation, for it is that for which you are truly addicted, greed, and a lifestyle that you did not really have the skills or the imagination legitimately to acquire.

… . . . . . . You did as you did because you wanted to and because you thought that you could get away with it.”

The judge was finding that this was not a case in which an investment manager had suffered genuine investment losses which he had attempted in panic to cover with other investors’ money; it was a case in which the appellant had effectively stolen investors’ money to fund an extravagant lifestyle. To complete the picture, however, it was necessary to recognise that the business in which the appellant was engaged was not set up to commit fraud from the outset but was trading honestly before the appellant took advantage of his trusted position to remove funds for his own purposes.

Previous decisions of the Court of Appeal demonstrate that the assessment of seriousness of offences of fraud is an intensely fact sensitive exercise and it is clear that the judge gave anxious consideration to the facts of the appellant’s case. The appellant made personal gains of not less than £18 million which he expended on sustaining an extravagant lifestyle and further sums which he said had been lost in gambling. The difference between the sum of money received from clients and the sums returned to them was some £69m. The total loss occasioned was £100m. In the main the losers were individuals who trusted the appellant to make investments on their behalf. They were not otherwise vulnerable. Of its kind we consider that this was a fraud within the top category of seriousness and, as Mr Burke concedes, the sentencing judge was not limited to the maximum sentence which applied to any one offence of fraud. The judge correctly made an assessment of the seriousness of the appellant’s overall criminality and imposed consecutive and concurrent sentences to reflect that assessment. In the course of so doing the judge was applying the principle that the total resulting from consecutive sentences should be a just and proportionate reflection of the overall criminality.

However, even within the top category of offences which cause permanent loss to individuals, there are factors that can mitigate seriousness. The three principal features of the case which provide some mitigation of the offences are that the business was not fraudulent from the outset, vulnerable victims were not targeted and the appellant self-reported to the Serious Fraud Office. The degree to which these factors reduce the seriousness of the offending is modest. The decision to appropriate clients’ funds for the appellant’s own purposes was a gross breach of trust made possible by the appellant’s reputation acquired during a period of honest trading. The level of sums which the appellant was used to handling was unlikely to be available to victims who were financially vulnerable. It was the wealth profile of the appellant’s clients which enabled him to fleece them of such huge sums of money. By the time the appellant visited the Serious Fraud Office the net had closed around him. He bowed to the inevitable having earlier defied court orders restraining his assets in order to perpetuate the fraud. We agree with the judge that no significant mitigation was available for delay. The appellant prevaricated for many months before tendering his pleas of guilty. However, the appellant is entitled to some credit for the fact that, although he was denying criminal responsibility, he gave information to the SFO which amounted to an acknowledgement of the underlying case against him.

The feature of the appellant’s offending which, in our view, required a very substantial sentence was the determination with which the appellant continued fraudulently to obtain large sums of money knowing that by so doing he was increasing the damage which would ultimately result to his clients, and his defiance of legitimate attempts by the losers to recover their money by ignoring orders made in the civil courts. The scale of the eventual loss was huge. The judge was, in our view, entitled to find that the losses caused by a Ponzi fraud of this kind were unprecedented in the United Kingdom. We conclude, after balancing these factors, that the appropriate starting point was not 17 years but 15 years imprisonment.

When explaining the way in which he approached the question of credit for pleas of guilty, the judge said:

“Consistent with the sentencing guidelines in respect of pleas of guilty, I give you credit for those pleas of guilty but it cannot be as much as Mr Burke, at one stage, appeared to suggest that it should be. You did not plead guilty at a stage on either of the two occasions when judges of this court before me suggested that full credit was still available to you. I cannot replace [sic] the informality of discussions between counsel late last year which, on closer analysis, did not reveal confirmed instructions that you would plead guilty until in fact August of this year, for pleas of guilty. Since July of 2011, this matter could have been listed at any time before September of this year for you to be re-arraigned. A delay in pleading guilty is not explained by some outstanding and peripheral issues of disclosure still being attended to. As the guidelines indicate, the maximum credit is available for those who plead guilty at the first opportunity and not for those who do so only after every avenue, as might enable a guilty defendant to avoid the responsibility of which he is and has almost certainly always been aware, has been explored by those he chooses to represent him…I can only conclude that the delay since July 2011 had, for the most part, been because you hesitated until July of this year to commit yourself to those pleas of guilty. But I do give you credit for them and I recognise even as late as they were formally given, much has been saved in terms of trial preparation and there had to be no trial which would have lasted several weeks…In my judgment [the credit for your guilty pleas] should be nearer to 20% rather than 25%...”

Mr Burke correctly states that pleas of guilt in heavy fraud cases have a particular benefit in avoiding the need for costly, long and burdensome trials (see R v Buffrey 1993 Crim App Rep (S) 511). However, he proceeded to make the bold submission that, in light of the history of the case, which we have described, this appellant should have been given a full one-third credit for the pleas he entered in September 2012. We disagree. In our judgment, the judge’s approach to assessing the credit to which this appellant was entitled for his late guilty pleas cannot be faulted. The appropriate discount was 20%.

It follows that in our judgment the appropriate total sentence was one of 12 years imprisonment. We acknowledge that this results in a reduction of sentence of 12 months only. Nonetheless, this is on any view a very lengthy sentence and the appellant should receive the advantage of the court’s review of seriousness after an analysis of the guidelines and the decided cases. We shall achieve the result we intend by quashing the sentences of 8 years imprisonment upon counts 7, 8, 10, 11, 13, 14 and 15 and substituting sentences of 7 years imprisonment upon each of those counts, to be served concurrently with each other but consecutively to the sentences imposed upon counts 1 to 4. Otherwise the judge’s orders will remain undisturbed. To this extent the appeal is allowed.