Varley & Ors, R. v

(A) Overview: This is the judgment of the Court to which each of us has contributed substantially.

This is a prosecution appeal by the Appellant (“the SFO”) under s.58 and following of the Criminal Justice Act 2003 (“the CJA 2003”) against the Ruling of Jay J dated 3 April 2019 (“the Ruling”, together with other related rulings), in which he upheld the various submissions of the Respondents that there was no case to answer in respect of the count/s of the indictment which concerned each of them.

In part at least, the financial market crash of 2008 casts a shadow over these proceedings as it did for those previously pursued by the SFO against Barclays Plc and Barclays Bank Plc (collectively, “Barclays”).

Describing the impact in the US, Andrew Ross Sorkin, in Too Big to Fail (2009), spoke (at p.3) of a “near collapse of the financial system, forcing a government rescue effort with no precedent in modern history”. For her part, Gillian Tett, in Fool’s Gold (2009), remarked on this market crash as standing out because of its sheer size

Unlike some other UK banks, Barclays did not take up the October 2008 UK Government (“HMG”) package of support for British banks – preferring instead to raise capital from other sources, including various Qatari entities. The question which arose in the Barclays proceedings and which arises here is whether, in the course of doing so and in the course of a prior capital raising exercise in June 2008, criminal offences were committed.

The Respondents all held senior positions at Barclays. Mr John Varley (“JV”) was

It may be noted that Mr Christopher Lucas (“CL”) the Barclays Group Finance Director was named as a co-conspirator; however, proceedings against him were not pursued because of the state of his health. CL and JV were the only directors among the alleged conspirators.

The indictment in these proceedings (“the Indictment”) contains two counts. Each of the Respondents was charged with Count 1, whereas only JV and RJ were charged with Count 2. Both Counts allege a statutory conspiracy to commit fraud by false representation, contrary to section 1(1) of the Criminal Law Act 1977 (“the CLA 1977”). Count 1 relates to the capital raising undertaken by Barclays in June 2008 (“CR1”). Count 2 relates to the subsequent capital raising by Barclays in October/November 2008 (“CR 2”).

The Particulars of Offence under Count 1 provide as follows:

“JOHN VARLEY, ROGER JENKINS, THOMAS KALARIS AND RICHARD BOATH, between 1 May 2008 and 31

August 2008, conspired together with Christopher Lucas , dishonestly to make representations within documents relating to Barclays’ capital raising of June 2008, with the intention of making gain for themselves or another, or causing loss to another, or exposing another to a risk of loss, which they knew were untrue or misleading, in breach of section 2 of the Fraud Act 2006, namely:

i.

(In the Prospectus dated 25 June 2008) that Qatar Holding was to be paid commission of 1.5% for its subscription in shares;

ii.

(In the Prospectus dated 25 June 2008) that the aggregate costs and expenses payable by Barclays plc in connection with the Firm Placing and the Placing and Open Offer was estimated to amount to approximately £107 million; and

iii.

(In Subscription Agreements dated 25 June 2008) that Barclays had not agreed to, nor intended to pay, any additional fees, commissions, costs, reimbursements or other amounts to Qatar Holding.”

The Particulars of Offence under Count 2 provide as follows:

“JOHN VARLEY and ROGER JENKINS, between 1

September 2008 and 30 November 2008, conspired together and with Christopher Lucas, dishonestly to make representations within documents relating to Barclays’ capital raising of October 2008, with the intention of making gain for themselves or another, or causing loss to another or exposing another to a risk of loss, which they knew were untrue or misleading, in breach of section 2 of the Fraud Act 2006, namely:

i.

(In the RCI Prospectus dated 25 November 2008) that Qatar Holding was to be paid commission of 2% for its subscription in RCIs and that the net proceeds of the issue of the RCIs was expected to amount to approximately £2,905,000,000 after deduction of commissions and concessions and the expenses incurred in connection with the issue of the RCIs;

ii.

(In the MCN Prospectus dated 25 November 2008) that Qatar Holding was to be paid commission of 4% for its subscription in the Notes and that the net proceeds of the issue of the Notes was expected to amount to approximately

£3,875,000,000 after deduction of commissions and

concessions and the expenses incurred in connection with the issue of the Notes.

iii.

(In Subscription Agreements dated 31 October 2008) that there were no further agreements or arrangements entered into between Qatar Holding and Barclays; and

iv.

(In Subscription Agreements dated 31 October 2008) that Barclays had not agreed to, nor intended to pay, any additional fees, commissions, costs reimbursements or other amounts to Qatar Holding.”

As will be seen (below), it is necessary to consider and determine an application by the SFO to amend the Particulars under Count 2 of the Indictment.

It is next convenient to set out the provisions of s.2 of the Fraud Act 2006 (“the Fraud Act”):

“Fraud by false representation

(1)

A person is in breach of this section if he – (a) dishonestly makes a false representation, and

(b)

intends, by making the representation – (i) to make a gain for himself or another, or

(ii)

to cause loss to another or to expose another to a risk of loss.

(2)

A representation is false if – (a) it is untrue or misleading, and

(b)

the person making it knows that it is, or might be, untrue or misleading.

(3)

‘Representation’ means any representation as to fact or law, including a representation as to the state of mind of – the person making the representation, or any other person.

…..

(5)

For the purposes of this section a representation may be regarded as made if it (or anything implying it) is submitted in any form to any system or device designed to receive, convey or respond to communications (with or without human intervention).”

In summary, in the Ruling, Jay J held (at [626] – [627]) that:

The Ruling followed an earlier Ruling of Jay J, dated 21 May 2018 (“the May 2018 Ruling”), in which he dismissed all charges on the then indictment as against Barclays. In a nutshell, that application turned on a consideration of whether the (alleged) criminal dishonesty of senior officers within Barclays – the present Respondents – could be attributed to the corporation so as to render the Bank itself criminally liable.

The May 2018 Ruling of Jay J was, in effect, upheld by Davis LJ, sitting as a Judge of the High Court in his judgment dated 12 November, 2018 (“the Davis LJ judgment”) SFO v Barclays PLC and another [2018] EWHC 3055 (QB), in which he dismissed the SFO’s application to prefer a voluntary bill of indictment – an application pursued by the SFO in consequence of the May 2018 Ruling.

The upshot was that, despite there being a prima facie case that false representations had been made to the market, the Judge ruled both that Barclays could not be liable because the (alleged) dishonesty of senior executives could not be attributed to the Bank and that the executives could not be liable because only Barclays could make the representations in question. The SFO submits that this is a most unfortunate outcome, which ought not and cannot stand and calls for the reversal of the Judge’s Ruling. The Respondents contend that, however curious the outcome, it is a consequence of the SFO’s framing of the indictment and that there is no warrant for this Court intervening.

For completeness, Jay J made further rulings, first refusing an application on the part of the four Respondents to dismiss the charges against them (“the July 2018 ruling”) and, secondly, refusing an application on the part of RB to dismiss the charges against him (“the December 2018 ruling”). The SFO has complained of inconsistency between the Judge’s approach to those applications and his decision in the Ruling. In our judgment, nothing turns on that complaint. If the Ruling was otherwise wellfounded, this complaint would not assist the SFO; if, per contra, there is merit in the SFO’s other criticisms of the Ruling, then it does not need this complaint of inconsistency. We therefore say no more of the July and December 2018 rulings.

(B) Preparatory hearing: Further and again for completeness, we note the SFO’s complaint that its several requests for a preparatory hearing pursuant to s.7 of the Criminal Justice Act 1987 had been rejected by the Judge. By now this is water under the bridge, so we do not propose to take time over it – other than to observe that the argument for a preparatory hearing appears to us to have had considerable attraction.

(C) The role of this Court and the relevant test: It is important to underline at once the nature of our task and the applicable test. We are not sitting at first instance, hearing the matter de novo. Our role is to consider an appeal from the Judge’s Ruling. As provided by s.67 of the CJA 2003, this Court “may not reverse” a ruling on such an appeal “unless it is satisfied”:

“ (a) that the ruling was wrong in law,

(b)

that the ruling involved an error of law or principle, or

(c)

that the ruling was a ruling that it was not reasonable for the judge to have made.”

In considering whether it was not reasonable for the judge to make a ruling, the fact that a different conclusion may have been reached does not “begin to provide a basis for a successful appeal”. As Sir Igor Judge P explained in R v B [EWCA] Crim 1144 at [19]:

Where the ruling in question is a terminating ruling made on the trial judge’s evaluation of the evidence, it is recognised that the trial judge will generally be much better placed to make such an assessment, and that an appellate court will not interfere unless the “high hurdle” of showing that his decision is “outwith the range of reasonable conclusion” has been met. As Moses LJ stated in R v M and T [2009] EWCA Crim 2848 at [25]:

(D) Complexity: Before proceeding further, we note the Judge’s observations at [557] of the Ruling. When dealing with the case against RJ, TK and RB and concluding that, subject to his decision on the points of law (Issues I – III), there would have been sufficient evidence for the case on Count 1 to have proceeded against these three Respondents, the Judge said this:

“….Although the challenge for that jury in this case would be immense, and arguably at the very outer limit of the capacity of any jury, and my duty in fairly summing-up this case would be monumental, that is what the law requires….”

With respect, we should not be taken as agreeing with this observation. Like any “heavy” case, the burden on the Judge in summing-up would be considerable but we do not see this case as in any way exceptional. Equally, we see no reason why this case should be at the “very outer limit of the capacity of any jury”. Complex it may be but there is no good reason why it should be unduly complex. Its essence is relatively straightforward. In any event, it is the task of the prosecution (primarily) to ensure that a case is presented in a clear and manageable form – the very helpful Bundle of “Graphics” made available to the jury (and to us) provides a very good example of how this can be done - and it is incumbent on the Judge to case manage it accordingly. So too, as a matter of professional good practice and their duty to the Court, such case management efforts are not to be frustrated by the Defence. It is of the first importance that the prosecution of high-end cases of (alleged) fraud should not be deterred by undue complexity or cost; a failure to prosecute such cases, where it would otherwise be appropriate to do so, threatens the integrity of markets and is socially corrosive. Under our system such prosecutions entail trial by jury and we see nothing in this case raising any doubts as to the suitability of this method of trial.

(E) Reporting restrictions: Given the possibility of future criminal proceedings, reporting restrictions covered the proceedings before Jay J and the Ruling, as we understood it pursuant to s.4(2) of the Contempt of Court Act 1981.

On the first day of the hearing of the appeal, an oral application was made by the BBC (on its own behalf and on behalf of the media more generally), for copies to be made available of the Ruling and various other documents. There was also before the Court an application by the solicitors, Quinn Emanuel, on behalf of a client involved in civil proceedings, for copies of the Ruling and, in addition, the parties’ submissions on the issue of no case to answer before Jay J. In the event, both applications failed, save in respect of the BBC’s application for hard (but not electronic) copies of the documents it had sought. The Court’s ruling on this point is set out in the relevant Transcript (at pp. 14 and following) and does not require elaboration here. Suffice to say that the Court was acutely conscious of the long-established principles of open justice (see, Archbold, 2019, at paras. 4-3a and following) and will only depart from them when it is necessary to do so in the interests of the administration of justice. Here, reluctantly, but given the possibility of future criminal proceedings, it was necessary to do so.

This judgment is itself subject to reporting restrictions as provided by s.71 CJA 2003. The question whether an anonymised or “sanitised” version can be published in advance of those restrictions coming to an end will be ventilated with counsel when the draft of this judgment is circulated; but the prohibition on publication contained in s.71 remains in force unless or until it is varied by an order of this Court.

Much of the factual introduction which follows is gratefully adopted from the Davis LJ judgment.

In June 2008, CR1 secured £4.4 billion in additional capital for Barclays. Subsequently, CR2 in November 2008 secured an additional total of £6.8 billion. As Davis LJ expressed it (at [23]):

“The prospective investors identified – at a time when investors were difficult to attract because of the state of the banking sector – included (among others) the state of Qatar, in effect through its Sovereign Wealth Fund. In addition, the then Prime Minister of Qatar, Sheikh Hamad, was proposed as a potential investor through a BVI investment company called Challenger Universal Limited. I will, for convenience, call the various entities ‘the Qatari entities’, although I stress that they are legally distinct. RJ had a particularly close business connection with such entities.”

Davis LJ went on (at [24]) to summarise the “settled practice at the time” and “the common expectation of investors”. Subscribing investors, such as the Qatari entities, known as “Conditional Placees”, would as between themselves receive an equal commission, in addition to the “like discounted price of the shares agreed for a rights issue”.

In the case of CR1 (ibid):

“…such commission was publicly set out in the Prospectus ultimately issued to shareholders and the wider market as 1.5%; the formal Subscription Agreements dated 25 June 2008 also expressly stated that no other commissions were being paid to any of the investors. Further, in the Prospectus it was stated that the aggregate costs and expenses payable by Barclays in respect of the Placing was £107 million. (In the Prospectus, it may be added, it was stated that the Board and Barclays took responsibility for the accuracy of the information contained in it.) That figure was consistent with commission being paid to subscribers of 1.5% of their maximum investment commitment. It was not consistent with any further sum (by way of commission or otherwise) being paid to any such investors.”

In the event, as the banking crisis worsened, it became evident (at [25]) that CR1 had not solved the Barclays balance sheet problem. Willing investors were by now even harder to locate. Barclays focused again on Qatar and, also, on certain Abu Dhabi investors; this time the same Qatari investors participated in an amount of £2.05 billion. As Davis LJ explained (ibid):

“The structure of CR2 was particularly complex, involving the use of Reserve Capital Instruments, Mandatorily Convertible Notes and Warrants. Suffice it to say, Subscription Agreements and various Prospectuses were, with other documents, issued on 31 October and 25 November 2008. These contained in the relevant respects broadly the like statements and warranties as contained in the CR1 documentation. The public announcement of Barclays, and as restated in the relevant documentation, was that the Qatari entities would variously receive 2% commission on the Reserve Capital Instruments for which they subscribed and 4% commission on the Mandatorily Convertible Notes for which they subscribed (totalling £62 million) and in addition an Arrangement Fee of £66 million. The stated net proceeds for Barclays were likewise calculated on such a basis.”

In October 2008, the Qatari entities sought a loan from Barclays, originally (8 October) for US$2 billion and subsequently (29 October) for US$3 billion. The loan was primarily negotiated by JV and RJ; it was ultimately approved by the Group Credit Committee (“GCC”) on behalf of Barclays. As Davis LJ underlined (at [26]), that approval was subject to an express restriction on the use of these funds:

“…it being expressly stipulated that the loan could not be permitted for use to fund the CR2 subscription (because of an appreciation of the unlawful financial assistance provisions of s.151 of the Companies Act 1985).”

As against Barclays, it was always the SFO’s case that the loan money was used – and was always designed to be used – to fund the subscription payments.

As to CR1 and CR2, it was and is the SFO’s case that the true position was (at [28]) “very different from that being publicly stated and warranted in the respective Prospectuses and Subscription Agreements: which documents, it is said, dishonestly misrepresented the position”. The truth was that much greater sums had been paid to the Qatari entities in return for their agreement to invest in CR1 and CR2. These sums (ibid):

“…in effect had to be paid as the Qatari entities (doubtless appreciating their strong bargaining position) were insistent….”

The SFO alleged that the mechanism for achieving these additional payments was through the use of two “Advisory Service Agreements” (“ASAs”). ASA1 was dated 25 June 2008 and was made between Qatar Holding LLC and Barclays; it was signed by JV on behalf of Barclays. ASA2 was dated 31 October 2008 and was made between Qatar Holding LLC and Barclays; it was signed by RJ on behalf of Barclays.

Davis LJ succinctly summarised the ASAs as follows (at [29]):

“Both agreements were in letter form. The stated term of ASA1 was 3 years. That of ASA2 was 5 years. ASA1 comprises one page. The stated sum to be paid (of £42 million) for services to be provided is written in manuscript. That sum was to be paid in four instalments. The agreement does not specify the services to be provided in return for the £42 million; it states that Qatar Holding has agreed to provide ‘various services, as an intermediary, in connection with the development of our business in the Middle East’; and that the ‘type and scale of the services …will need to be refined by mutual agreement as our relationship develops further’. ASA2 is hardly less short. The fee is stated at £280 million. It refers to the ‘great success of the agreement to date’. It then lists, in very broad language under six heads, the nature of some of the services stated to be provided: with again a statement that ‘these will need to be refined by mutual agreement’ during the period of the agreement. The sums payable by Barclays under ASA1 thus were £42 million. The sums payable under ASA2 were £280 million. It is one feature of ASA2 that its contractual period overlaps, for all but four months with the same contractual period stated in ASA1, albeit ASA2 was to last for an additional 28 months after ASA1 terminated. It is another feature of ASA1 that the four instalments payable under it were to be paid by 1 April 2009: that is, before the end of the contractual period; and all such instalments in fact were invoiced by the Qatari entities on 13 August 2008. The instalments payable under ASA2 were 20 equal instalments of £14 million.”

The SFO’s case was that each of ASA1 and ASA2 was a sham or dishonest device (at

Overall, it was also the SFO’s case that the Barclays Board and the relevant committees – the Board Finance Committee (“BFC”) and the GCC – were kept in the dark about the true intent behind ASA1, ASA2 and the loan.

At the time when the Barclays entities remained defendants, the indictment was in broadly similar form to the Indictment as it now is (set out above), save that there were additional Counts (3 and 4), dealing with the provision of unlawful financial assistance contrary to s.151 of the Companies Act 1985 (Count 3) and the role of corporate officers (JV and RJ) in any such provision (Count 4). In the event, following the dismissal of the charges against Barclays, it was agreed that Count 4 would have to be dismissed against JV and RJ. No more therefore need be said as to the s.151 matters.

As to the Directing Mind and Will (“DMW”) of Barclays, Davis LJ expressed the matter this way (at [58]):

“…the question is, on the assumed facts, whether the alleged dishonest acts taken in conjunction with the alleged dishonest state of mind of the relevant individuals – in particular for present purposes JV, CL and RJ – can be attributed to Barclays so as to make it criminally liable. Put another way, are their (assumed) dishonest acts and intentions, for the purposes of these particular transactions, to be treated as the dishonest acts and intentions of Barclays itself?”

Jay J’s overall conclusion in the May 2018 Ruling, as summarised by Davis LJ (at [91]), was that JV, RJ and CL (or any combination of them) could not, on the evidence, be regarded as Barclays’ DMW for the purpose of CR1, CR2 and the making of the loan to the Qatar entities. The relevant responsibility and authority in this respect remained with the Barclays Board, or the BFC or GCC, regardless of the autonomy conferred on the individual Respondents in the antecedent negotiations.

In the event (at [117] et seq), Davis LJ accepted “the correctness of the essential reasoning” of Jay J in the May 2018 Ruling. As Davis LJ expressed it (at [119]), “by reference to the pleaded particulars on the indictment”, JV, CL and RJ could not be regarded as Barclays’ DMW “for the purpose of performing the functions in question”. That was “the long and short of it”. This was not (at [122]) “a matter of form over substance”. Instead, here, the form of corporate governance “is the substance”. Various other considerations (in the paragraphs which followed and to which it is unnecessary to refer) confirmed Davis LJ in these views.

Accordingly, Davis LJ dismissed the SFO’s application to prefer a voluntary bill of indictment.

We turn to the principal Issues.

(A) Introduction: The SFO’s primary case in relation to CR1 and CR2 is that JV (subject to Issue IV) and CL (the two directors among the alleged conspirators), made false representations within the meaning of s.2 of the Fraud Act, relating, as will be recollected, to the “disguised commissions” payable to the Qatari entities, for which each was personally liable. The other Respondents, RJ, TK and RB in respect of Count 1/CR1 and RJ in respect of Count 2/CR2, conspired with them to do so. The point hinges on JV and/or CL incurring personal liability in respect of the representations in question – the “direct route” to liability.

In the Ruling, the Judge upheld the Respondents’ submission of no case to answer in this regard; neither JV nor CL had made a false representation within s.2 of the Fraud Act in the company prospectuses or the SAs. Any statements were and could only be made by Barclays not by individual directors: Ruling, esp. at [182], [185], [186], [189] and [197]. Moreover, the Judge (at [178] and [187]) refused the SFO’s application to amend the Indictment to include Particulars as to the Warrant Prospectus under Count 2.

The SFO submits that what it describes as the Judge’s “binary approach” was in error. Common sense, policy and authority all pointed to the directors being personally liable for such false statements. In her oral submissions, Ms Darlow QC said this:

“….it is axiomatic that directors can be regarded as making statements within company prospectuses, and there are very strong public policy considerations to support that conclusion in order to give efficacy to the efforts of the common law and civil and criminal statutes to protect the wider market and safeguard the integrity of rights issues from false statements contained in public facing company documents….”

The Judge had confused or conflated the question of making a statement with the question of identifying Barclays’ DMW and, similarly, with the separate question of identifying the issuer (or publisher) of company documents.

The Respondents retort that the Judge was right, essentially for the reasons he gave. Far from there being a lacuna in the law, any problems were of the SFO’s ownmaking in the framing of the Indictment and, as Mr Purnell QC put it, its “obdurate” pursuit of market-facing representations – the more so, given the absence now of any corporate defendant. There was a difference between “making” a statement and “causing” a statement to be made; so too, a distinction was to be drawn between the

“responsibility regime” for directors and personal liability. Statements in company documents, especially the prospectuses, were made by Barclays alone.

In addressing this Issue, we deal separately with the equity prospectus (CR1), the SAs (CR1 and CR2), the MCN and RCI prospectuses (CR2) and the Warrant Prospectus (CR2).

(B) CR1: The equity prospectus: A key ingredient of CR1 was the equity prospectus, dated 25 June 2008 (“the CR1 Prospectus”). The production of such a prospectus was required by ss.73A and 85 of the Financial Services and Markets Act 2000 (“FSMA”) and the Prospectus Rules (as in force on 1 June 2008, “the Prospectus Rules”). R.5(2)(b)(i) of the Prospectus Rules (giving effect to s.84(1)(d) FSMA) provided that where the issuer of the transferable securities is a body corporate, in addition to the issuer “each person who is a director of that body corporate when the prospectus is published” is responsible for the prospectus.

All the directors of Barclays, including JV and CL signed DRLs (i.e., Directors’ Responsibility Letters). These were addressed to Barclays – not to the market; they were thus not market-facing. Nonetheless, they are of very considerable importance. Insofar as material, the DRLs provided as follows:

“1.

I have read a proof of the Prospectus dated 22 May 2008 (the final version of which is expected to be published on or around 18 June 2008), and I understand that the Prospectus will constitute a prospectus prepared in accordance with the prospectus rules made under Part VI of ….[FSMA]…, as amended (‘FSMA’)(the ‘Prospectus Rules’).

2.

I understand that the Prospectus is required by section 87A(2) FSMA to contain the information necessary to enable investors to make an informed assessment of the assets and liabilities, financial position, profits and losses and prospects of the Company and of the rights attaching to the New Barclays Shares…

3.

I understand that pursuant to Prospectus Rule 5.5.3R(2) I will be required to take responsibility for the Prospectus (and any supplementary prospectus) and that the Prospectus will contain a declaration in the following terms….

‘The Barclays Directors, whose names appear at paragraph 2 below, and Barclays accept responsibility for the information contained in this document. To the best of the knowledge of the Barclays Directors and Barclays (who have taken all reasonable care to ensure that such is the case), such information is in accordance with the facts and does not omit anything likely to affect the import of such information.’

4.

To the best of my knowledge and belief (having taken all reasonable care to ensure that such is the case), all statements of fact in the prospectus relating to the Company and its subsidiary undertakings and all statements relating to myself are true and accurate in all material respects and are not misleading…..

6.

There are no material facts or considerations omitted from the Prospectus which to my knowledge would make any statement in the Prospectus misleading and there is no other information known to me or which could on reasonable enquiry be known to me whose omission makes any statements or opinions in the Prospectus misleading.

7.

I accept responsibility for the information contained in the Prospectus and confirm that to the best of my knowledge, having taken all reasonable care to ensure that such is the case, the information contained in it is in accordance with the facts and does not omit anything likely to affect the import of such information.

8.

I hereby authorise the naming of myself as a director of the Company in the Prospectus and I authorise the issue and publication of the Prospectus….in the form in which it is approved for issue by resolution of the board of directors of the Company or a duly authorised committee of the board ….and in particular authorise the inclusion of a statement as to my responsibility as a director and I undertake to accept responsibility in the terms set out in the Approved Document. I understand that a responsibility statement is regarded as including expressions of opinion. This authority and undertaking is notwithstanding the fact that I may not attend the meeting of the board …., or the meeting of the committee of the board, which approves the final form of the relevant document or see the final form of any document approved in this manner prior to its publication or posting.

11.

I have reviewed the memorandum prepared by Clifford Chance LLP entitled ‘Memorandum On Directors’ Responsibilities And Liability For Public Documents’ distributed at the board meeting on 28 May 2008….

12.

I will inform you immediately if I become aware, at any time before dealings begin in the New Barclays Shares on the London Stock Exchange:

(a)

that any statement of fact or expression of opinion contained in the Prospectus becomes or has become untrue or inaccurate; and

(b)

of any other fact, the omission of which renders any such statement or expression misleading.”

The CR1 Prospectus, at Part X, para. 2, named, amongst others, JV and CL as directors of Barclays. Part X, para. 1 was in these terms:

“1.

Responsible Persons

The Directors, whose names appear at paragraph 2 below, and Barclays accept responsibility for the information contained in this document. To the best of the knowledge of the Directors and Barclays (who have taken all reasonable care to ensure that such is the case), such information is in accordance with the facts and does not omit anything likely to affect the import of such information.”

With regard to the CR1 Prospectus, we are, with respect, unable to share the difficulty felt by the Judge or to agree with his conclusion. We find this aspect of Issue I straightforward.

While the DRLs do not themselves constitute market-facing statements on the part of the directors – addressed, as they were, solely to Barclays – the DRLs fed in to the statement contained at Part X, para. 1 of the CR1 Prospectus (“the Directors’ Responsibility Statement” or “DRS”). That statement was indisputably marketfacing. It says in terms that both the directors and Barclays accept responsibility for the matters there set out. There is no good reason why that wording should not mean what it says and every good reason (in terms of market integrity) why it should. On the basis that it does, the irresistible conclusion is that the directors (and Barclays) “made” that statement and there is no reason whatever to conclude otherwise.

“Make” or “made” is an ordinary English word, to be given its ordinary and natural meaning; it should not be overlaid with complexity. The question of whether JV (subject to Issue IV) and CL “made” the DRS is not to be conflated with any issue as to the DMW of Barclays, an irrelevancy for these purposes. Likewise, the obvious fact that Barclays – not JV or CL – “issued” the CR1 Prospectus is neither here nor there. Further, as the focus is on whether JV and/or CL “made” the DRS, such differences (if any) as there might be between “making” a representation and

“causing” a representation to be made are beside the point. Still further, the

“responsibility regime” for directors in no way tells against their personal liability; it would be surprising if it did and if questions of “responsibility” and “liability” were unconnected.

Accordingly, a reasonable jury, properly directed, would be capable of convicting a director in respect of the DRS, provided that it constituted or contained a false representation within the meaning of s.2 of the Fraud Act. Whether in fact the DRS constituted or contained a false representation of this nature would be a question of fact for the jury to resolve. On this simple ground, there was no proper basis for allowing the submission of no case to answer in respect of the CR1 Prospectus. Furthermore, even if the Judge’s conclusion in respect of JV is upheld on Issue IV, the fact that CL had a case to answer suffices for the proceedings to continue against RJ, TK and RB.

Elaboration is almost unnecessary with regard to the CR1 Prospectus, but we go on to outline, briefly, why this conclusion, as to directors’ personal liability for misstatements in prospectuses or other company documents, fits comfortably within the longstanding framework of law in this area.

In Derry v Peek (1889) 14 App Cas 337, a claim for deceit was brought against the directors of a company, founded upon a false statement in the prospectus. The action

Nor does Derry v Peek stand alone. The earlier case of Henderson v Lacon (1867) 5 Eq 249, contains observations in the judgment of Sir W Page Wood VC to like effect. Thus, at p.262, the Vice-Chancellor spoke of the directors incurring personal liability for false statements in a prospectus, provided they were “fixed” with the guilty knowledge or “scienter”, necessary for an action of deceit. That scienter was clearly fixed “…from the moment you find a representation concerning their own acts which is incorrect, and which they must be taken to have known to be incorrect, and to have knowingly stated, and thereby to have misled the party complaining of the misrepresentation”. Again, there does not appear to have been any resistance to the notion of directors incurring personal liability for statements in a prospectus.

Possfund v Diamond [1996] 1 WLR 1351 was concerned with shares bought in the market after flotation in reliance upon a prospectus containing a misrepresentation as to the value of the company. The issue in these (admittedly) civil proceedings was whether the company’s directors and advisers owed a duty of care to purchasers in the “aftermarket”. Lightman J dismissed the application to strike out the claim, holding that it was arguable. The judgment contains (at pp. 1358 et seq) a valuable summary of the common law and statutory schemes providing protection to investors in respect of prospectuses. For immediate purposes, however, the importance of the case lies in the absence of any suggestion that a director could not incur personal liability for a misrepresentation in a prospectus.

The SFO placed some emphasis on the relative similarity between the wording contained in s.397(1)(a) and (c) FSMA and that found in s.2 of the Fraud Act (set out above). As expressed in the SFO’s skeleton argument:

“Liability under FSMA is dependent upon the concept that the actus reus of making a statement in a document published by a company (such as a trading statement or prospectus) could be committed by a director of that company.”

It would be curious, the SFO submits, if a knowingly misleading, false or deceptive statement made in a company document and considered in the context of protecting market integrity, was capable of giving rise to personal criminal liability under s.397(2) FSMA but a false representation within s.2 of the Fraud Act was not so capable. We accept this SFO submission as to the analogous FSMA regime and, for this reason too, would not be at all attracted to ruling out the direct route in the circumstances of this appeal unless driven to do so – which we are not at all persuaded we are. For completeness, in this context, the SFO drew our attention to a number of FSMA sentencing authorities where no question had been raised as to the appropriateness of convictions of officers of companies for false or misleading statements published by the companies in question. It is unnecessary to refer to those authorities, given the absence of discussion as to the convictions but, if to a distinctly limited extent, they are supportive of the SFO’s submission.

Furthermore, we cannot think that the risk of personal criminal liability would occasion surprise in the commercial world generally. It certainly could not do so, specifically, in the case of the Barclays directors who had signed DRLs. It will be recollected that para. 11 of the DRLs made reference to a Clifford Chance memorandum, which the directors in question confirmed they had reviewed. Though obviously in itself of no legal force, it is instructive to note that para. 4 of that memorandum is headed “THE PROSPECTUS – CRIMINAL LIABILITY” and the very first sentence thereunder reads as follows:

“Criminal liability for the Prospectus may arise under FSMA, the Fraud Act 2006, the Theft Act 1968 or the common law offence of conspiracy to defraud.”

We conclude, with no real hesitation, that the direct route is available to the SFO, via the DRS in the CR1 Prospectus:

It follows that we part company with the Judge insofar as he held that there was no case to answer in respect of either JV or CL making a false representation within s.2 of the Fraud Act arising out of the CR1 Prospectus. The Judge’s conclusion involved, with respect, an error of law, so entitling this Court to intervene pursuant to s.67(a) and/or (b) of the CJA 2003. Under this heading, therefore, the case can proceed

We add only this. First, we have not approached the DRS as a statement made by JV and/or CL as agents of Barclays. We prefer the view that, in accordance with the ordinary meaning of language, both the directors (JV and CL) and Barclays made and accepted responsibility for that statement as principals. If, however, it could be said that the DRS had been made by JV and/or CL as agents of Barclays, that would be of no assistance to the Respondents. In such circumstances, Lord Hoffmann’s trenchant observation in Standard Chartered Bank v Pakistan Shipping Corpn [2002] UKHL 43; [2003] 1 AC 959, at [22], would be squarely in point:

“No one can escape liability for his fraud by saying: ‘I wish to make it clear that I am committing this fraud on behalf of someone else and I am not to be personally liable’.”

Secondly, if for any reason the DRS was not (arguably) a false representation made by JV and/or CL and was instead to be treated as a false representation made by Barclays only, then we would accept Ms Darlow’s alternative (or “adjunct”) submission – namely, that the wording of the DRS entailed the contemporaneous adoption by JV and/or CL of Barclays’ false representation. As explained in Ormerod & Laird, Smith, Hogan and Ormerod’s Criminal Law (15^th ed., 2018) (“Smith, Hogan & Ormerod”), at p.939, the offence under s.2 of the Fraud Act can be committed by any person who makes a false representation. Thus (ibid), D will be liable for his personal representations and:

“….might also be liable for representations of third parties which he can be said unambiguously to have adopted…..”

(C) The SAs: CR1 and CR2: The SAs formed part of both CR1 and CR2 and featured in both Counts on the Indictment. With regard to CR1, Barclays agreed SAs with five investors; four of those SAs were signed by CL, one by JV. With regard to CR2, the SAs entailed a commitment by investors to take MCNs, RCIs and Warrants from Barclays. Six SAs were made between Barclays and the investors in question; all were signed by CL. In all cases, the SAs were signed by the individual director “for and on behalf of” Barclays.

The CR1 SAs, dated 25 June 2008, each contained a warranty from Barclays, as follows:

“8.

WARRANTIES AND BARCLAYS UNDERTAKINGS

8.1

Barclays hereby warrants to the Investor that each of the Warranties in Part A of Schedule 1 (Warranties) is true, accurate and not misleading as at the date hereof.”

In turn, Schedule 1, Part A included the following provision:

“(B)

Other than as disclosed in the Subscription Agreements and the Draft Prospectus, and in respect of the subscription of Barclays Ordinary Shares under the Subscription Agreements: (i) there are no further agreements or arrangements entered into between the Investors and Barclays; and,

(ii)

Barclays has not agreed to, nor intends to pay any fees, commissions, costs, reimbursements or other amounts to the Investors.”

Mutatis mutandis, the CR2 SAs, dated 31 October 2008, contained warranties on the part of Barclays to like effect.

The false representation is again said to flow from the (alleged) disguised commission payments to the Qatari entities.

The SFO “direct route” case pursuant to the SAs turns on their signature by CL (and, in one instance, JV). It is not and cannot be suggested that the warranties were given by either JV or CL.

As set out in the Ruling (at [167], [171] - [172] and [174] – [175]), it was not in dispute that the signing of the SAs was “mechanical” in the sense that the anterior decision had been made by the Board and whoever signed the SAs would be merely executing them. Moreover, the signing of the SAs took place pursuant to a delegation by the Board (it would seem through the BFC) to “any Director, the Company Secretary or the Group General Counsel” to do so.

Two questions arise. The first goes to the signature as such. In our judgment, the fact of signature, even though a matter of happenstance as to which individual signed, cannot be dismissed in the manner contended by the Respondents, as a mere “formal execution”. In determining whether the signature of the SAs was capable of amounting to a false representation within s.2 of the Fraud Act, the factual nexus between the director signatory and the SAs would be relevant. If, for example, the SAs were signed for and on behalf of Barclays by an individual director who knew that the warranties were false or misleading, having been involved in a conspiracy beforehand to make false representations, then the question of the applicability of Lord Hoffmann’s observations in Standard Chartered would call for careful consideration. That said, given the terms of the SA, and the absence of any DRS, there would remain the difficulty of showing how that would involve a false representation made by a conspirator rather than by the company alone.

Secondly, however, the signature point does not stand alone. The Respondents are charged with statutory conspiracy under s.1(1) of the CLA 1977. That section provides as follows:

“(1)

….if a person agrees with any other person or persons that a course of conduct shall be pursued which, if the agreement is carried out in accordance with their intentions…

(a)

will necessarily amount to or involve the commission of any offence or offences by one or more of the parties to the agreement… …..

he is guilty of conspiracy to commit the offence or offences in question.”

The category of authorised signatories was far wider than the conspirators themselves. Thus, the terms of the delegation were such that any director could have signed the SAs; so too, the Company Secretary or Group General Counsel. This meant that it was a matter of chance who signed them and there was no evidence of an agreement between the conspirators to a course whereby JV or CL would necessarily do so. In short, with regard to the SAs, the (alleged) agreement upon which the SFO’s case rests would not necessarily have amounted to or involved the commission of an offence “by one or more of the parties to the agreement”.

Accordingly, and for this reason (whatever view might be taken of the signature point), the SFO’s case with regard to the SAs does not satisfy the requirement of s.1(1) of the CLA 1977. We dismiss the appeal insofar and to the extent that it rests upon the SAs.

(D) The RCI and MCN Prospectuses: CR2: We take this point summarily. As explained in the Davis LJ judgment (at [25]), the structure of CR2 included (inter alia) Reserve Capital Instruments (“RCIs”) and Mandatorily Convertible Notes (“MCNs”). Not being equity prospectuses, the RCI and MCN prospectuses were not preceded by DRLs; nor did either contain a DRS. These prospectuses were not signed by an individual director, so they were not at all comparable with the SAs in that respect. When asked by the Court, Ms Darlow very fairly accepted that, by themselves, the RCI and MCN Prospectuses (and thus Particulars i and ii under Count 2 of the indictment) could not be relied upon in support of the direct route or, hence, any appeal. Whether the RCI and MCN Prospectuses have any relevance to these proceedings depends on the fate of the SFO appeal against the Judge’s refusal of its application to amend the Particulars under Count 2 of the Indictment in respect of the Warrant Prospectus, the matter to which we next turn.

(E) The Warrant Prospectus: CR2: As already seen, the Particulars under Count 2 of the Indictment contain no reference to the Warrant Prospectus dated 25 November 2008 (“the Warrant Prospectus”), which formed part of CR2.

Before the Judge, the SFO renewed an earlier unsuccessful application to amend the Indictment, so as to add the following Particulars iii, iv and v under Count 2 – with the existing Particulars iii and iv (set out above) becoming Particulars vi and vii.

“iii.

(In the Warrant Prospectus dated 25 November 2008) that the estimated issue costs of the RCI and Warrants were £95 million and that the estimated issue costs of the MCNS were £175 million.

iv.

(In the Warrant Prospectus dated 25 November 2008) that, to the best of the knowledge of John Varley, who had taken all reasonable care to ensure that such was the case, the information contained in the Prospectus was in accordance with the facts and did not omit anything likely to affect the import of such information.

v.

(In the Warrant Prospectus dated 25 November 2008) that, to the best of the knowledge of Christopher Lucas, who had taken all reasonable care to ensure that such was the case, the information contained in the Prospectus was in accordance with the facts and did not omit anything likely to affect the import of such information.”

Thus, the nub of the allegation that the SFO was seeking to introduce pursuant to the proposed amendment was that the RCI and MCN issue costs had been knowingly under-stated by the £280 million which Barclays had agreed to pay the Qatari entities under ASA2.

The Warrant Prospectus was required to be and was preceded by DRLs, addressed to Barclays, dated 20 November 2008 (“the November DRLs”), signed by JV and CL. Para. 1 of the November DRLs recorded that the signatory had read a “proof of the Prospectus dated 18 November 2008 (the final version of which is expected to be published on or around 25 November 2008)”. Para. 12 of the November DRLs was in the same terms as para. 12 of the DRLs set out above and dealing with future material changes.

The proposed amended Particulars were underpinned by the contents of the Warrant Prospectus, which contained the following DRS (“the November DRS”):

“The directors of the Warrant Issuer (the ‘Directors’), whose names appear on pages 55 and 56 of this document, and the Warrant Issuer accept responsibility for the information contained in this Prospectus. To the best of the knowledge of the Directors and the Warrant Issuer (having taken all reasonable care to ensure that such is the case), the information contained in this Prospectus is in accordance with the facts and does not omit anything likely to affect the import of such information.”

The names of JV and CL appeared at p.55 of the Warrant Prospectus. Footnote 5, at p.75 is in these terms:

“For accounting purposes

-

….the net proceeds of RCIs and Warrants of £2,905m (representing the £3,000m issuance, net of estimated issue costs of £95m)….

-

….the net proceeds of the MCNs of £3,875m (representing the £4,050m issuance, net of estimated issue costs of

£175m)…”

Mr Purnell QC, for the Respondents, very realistically and properly made plain that there was no suggestion of prejudice flowing from the amendment (if permission to amend was otherwise justified). It is therefore unnecessary to take time considering the history of the proposed amendment. It suffices instead to record the Judge’s reasons for refusing permission to amend (at [187] of the Ruling):

“There is a further difficulty with the Warrants Prospectus in connection with Count 2. This is not so much procedural as substantive. The SFO’s reliance on the representations in the MCN and RCI Prospectuses creates no difficulty on the facts, and also recognises that it is the co-existence of the £280M with the statements made in them that engenders the falsity. The argument becomes strained and artificial in relation to statements made in the Warrants Prospectus because these were true even if the £280M is taken into account. The SFO relies on recondite statements in unaudited financial information referred to in the Warrants Prospectus which were introduced very late in the day. This was not new information. There is really no evidence that JV and CL applied their minds to this, either generally or in the context of the DRLs signed on 20^th November. Although I have no difficulty with the submission that for the purposes of s.5 of the Fraud Act 2006 specific intent can be proved by demonstrating that the directors must have appreciated that the extra £280M would falsify the MCN and RCI Prospectuses, I accept the defence submission that this becomes entirely artificial, indeed untenable, in relation to the Warrants Prospectus. This is a synthetic argument which aspires to transcend a legal difficulty. My January 2019 ruling was correct, although not given for all the right reasons, and I will not reverse it.”

Ms Darlow QC’s submission is straightforward. Whether or not the statements were “recondite” and contained in the footnotes did not matter; the three prospectuses were all part of CR2 and were not independent documents; JV and/or CL must have known that the public documents contained false representations by reason of the £280 million “hole”. The amendment should be allowed; the case based on the false representation in the Warrant Prospectus as to the financial information in the RCI and MCN prospectuses was fit to be left to the jury to decide.

Mr Purnell QC focuses on timing in particular, together with the evidential basis for the proposed amendment; by this route he sought to uphold the Judge’s reasoning and conclusion. The November DRLs were dated 20 November. As those DRLs stated in terms, the draft Warrant Prospectus to which the signatories had specific regard was that of 18 November. The 18 November draft Warrant Prospectus contained nothing comparable to footnote 5 in the (25 November) Warrant Prospectus. There was nothing to show that any of the alleged conspirators “could have ever conceived that the warrant prospectus would contain this recondite information which was not present in the prospectus when the directors had signed their responsibility letters”.

Notwithstanding Mr Purnell’s excellent submissions, we cannot agree with the Judge’s reasoning or conclusion. We think there is a case fit to go to the jury that JV (subject to Issue IV) and/or CL must have known of the false representations in the public-facing financial documents by reason of the agreement in ASA2 to pay the £280 million. Ultimately, whether the jury conclude that there was such knowledge is a matter for them. It is at that stage that the “recondite” nature of the material in question can fall for consideration. As to the timing point, we accept Ms Darlow’s submission that late changes to the Warrant Prospectus were capable of giving rise to personal criminal liability.

For the reasons given, we are persuaded that the Judge erred in law, so entitling this Court to intervene, here too, pursuant to s.67(a) and/or (b) of the CJA 2003. Accordingly, we allow the appeal from the Judge’s Ruling on this point and permit the amendment of the Indictment to include the new Particulars (iii. – v., under Count 2) in the Warrant Prospectus.

(F) Ramifications: It follows, therefore, that the SFO’s case by way of the direct route, is fit to go to the jury in respect both of Count 1 (the CR1 Prospectus) and Count 2/CR2 (the Warrant Prospectus and its cross-reference to the RCI and MCN Prospectuses). This outcome affects all the Respondents, save that in the case of JV alone, it is subject to our conclusion on Issue IV.

(A) Introduction: The SFO’s second proposed route to legal liability was “Innocent Agency”. A useful definition of “innocent agency” (cited with approval in R v

Two general considerations can conveniently be mentioned at this stage. First, the liability of the perpetrator (acting through the innocent agent) is that of a principal, not an accessory. This is an important consideration for the SFO, given the difficulties facing an allegation of statutory conspiracy based on accessory liability (see below). With regard to liability as a principal, as set out in Smith, Hogan & Ormerod, at p.183:

“If D1 and D2 agree to employ an innocent agent, E, both D1 and D2 are liable as principal offenders for E’s acts when they are committed….. The innocent agent’s acts are considered the acts of both conspirators.”

Secondly and as the Judge held (Ruling, at [197(1)]), the doctrine of innocent agency is capable of applying to statutory conspiracy under s.1(1) of the CLA 1977. Innocent agency has long been available in the context of a common law conspiracy: Bull and Schmidt (supra). There is nothing in the CLA 1977 to suggest that Parliament intended to disapply the common law rule in the case of statutory conspiracies. This is obviously sensible; consider, for example, two individuals who agree to send a letter bomb through the post, employing an innocent agent to put the bomb into a post box. There is a strong policy interest in discouraging conspirators from seeking to commit offences through unsuspecting third parties.

(B) The Ruling: In short summary, while accepting (as already noted) that innocent agency was capable of applying to a statutory conspiracy, the Judge ruled that the innocent agency route to liability was precluded to the SFO even if there was otherwise a case to answer on the merits against all the Respondents.

First (“point 1”), the Judge held that the SFO’s case on innocent agency failed for the same reason as its case on the direct route; only the company could ever have committed the offence under s.2 of the Fraud Act; no individuals could have

“committed an offence in the public-facing documents” and, therefore, they could not be regarded as principals: see, esp., at [225], [228] and [229].

Secondly (“point 2”), the Judge held (at [235] – [244]) that, on its true construction, s.2(2) of the Fraud Act could never apply to a case of innocent agency. At [236], he said this:

“….s.2(2) has two elements: the representation as made must be untrue; the person making it must know that. The representation is not ‘false’ unless these two elements are in place.”

On this footing, if the maker of the representation lacks knowledge, there is no false representation for the purposes of the provision. Interposing, it may of course be added that if the agent had such knowledge, he would not be an “innocent” agent. It followed that the doctrine of innocent agency was inapplicable to all cases concerning s.2(2). This outcome was “uncomfortable” and “close to being absurd…and may well have been unintended”. However, the language of s.2(2) was clear; this was a criminal statute and the “principle of legality” applied. Many of these difficulties arose because the SFO had chosen to frame the indictment in a particular way.

(C) The rival cases: Ms Darlow QC submits that the doctrine of innocent agency was a feature of the common law and plugged the lacuna that would otherwise exist in the law if conspirators could agree to commit crimes through unsuspecting third parties. The Judge had been wrong to conclude that there was no room for innocent agency here. Point 1 was a world away from other examples where innocent agency was inapplicable. As to point 2, the construction of s.2(2) adopted by the Judge was absurd - and absurd constructions were to be avoided in penal as well as non-penal statutes. S.2(2) should instead be construed so that “the actus reus is the making of a false representation, the mens rea is dishonesty, intention and knowing that it is or might be untrue or misleading”.

Mr Purnell QC submits that the Judge’s decision was correct and should be upheld. There was no “closed category” of cases where innocent agency was inapplicable. It was so here, having regard to the specifics of the case. Moreover, innocent agency was precluded by the clear definition of false representation contained in s.2(2) of the statute. The Fraud Act had not been enacted in haste (anything but) and the construction favoured by the Judge was not absurd.

As became clear following exchanges between the Court and Mr Purnell QC, points 1 and 2 and only those two points were taken in opposition to the applicability of innocent agency here. To repeat, point 1 was that only Barclays could make the representations. Point 2 went to the construction of s.2(2) of the Fraud Act. If wrong on both those points, Mr Purnell expressly disclaimed any “point 3”. While he submits that it was “impossible to conceive” of the Board of Barclays “with its governance processes and….collective decision-making” being the agent through which individuals could make the indicted representations, that was a component of points 1 and 2, not a separate point 3. It follows that matters relating to the factual context and the panoply of Barclays corporate governance such as those alluded to in the Davis LJ judgment (at [46] and, especially, [126]) are for any trial in this matter and (other than as part of points 1 and 2) do not comprise an objection at this stage to the applicability of innocent agency. We understand the realism of Mr Purnell’s stance, especially having regard to the unavoidable factual content of any point 3 (along the lines we have outlined) – very much including the likely debate as to the improbability of Barclays being an innocent agent not exercising independent judgment (cf. Stringer, supra, at p.17) on the one hand, set against the impact of the (alleged) fraud on Barclays’ decision-making process, on the other.

(D): Point 1: Point 1 can be swiftly disposed of. It falls away in the light of our conclusion on the direct route, discussed above. We held there that individual Respondents were capable of committing offences, as principals, under s.2 of the Fraud Act in the public-facing documents. That conclusion suffices to eliminate the Judge’s objection to the applicability of the doctrine of innocent agency under point 1.

(E) Point 2: Point 2 turns on the true construction of s.2 of the Fraud Act. With respect and for the reasons which follow, we are unable to accept the Respondents’ and the Judge’s construction of the section.

First, there is no intrinsic reason why innocent agency should be inapplicable to s.2 of the Fraud Act.

We accept, of course, that there are limitations on the doctrine of innocent agency. As expressed in Glanville Williams (op cit, at p.316), the doctrine can be used “only when it is plausible to say that the defendant did the forbidden act in the forbidden circumstances”. The Law Commission Working Party is again of assistance:

“A person is not guilty of committing an offence through an innocent agent when the law provides or implies that the offence can be committed only by one who complies with a particular description which does not apply to that person, or specifies the offence in terms implying personal conduct on the part of the offender. ”

Thus, a woman could not be convicted of rape - as a principal - through the innocent agency of a man (though there is no bar to a woman incurring accessory liability for rape where the facts so warrant it). Put another way, for innocent agency to be relied upon, the charge must not entail contradicting the statute defining the offence (Glanville Williams, op cit, at p.320) and the principal must not lack “some characteristic essential for liability as a principal” (Blackstone’s Criminal Practice 2019, at A4.18).

So too, we are content to proceed in agreement with Mr Purnell’s submission that there is no “closed list” of offences where innocent agency cannot be invoked.

However, put shortly, the present matter is far-removed from the difficulties encountered in such offences as rape or bigamy. There is, here, no or certainly no relevant, lack of any “characteristic essential for liability as a principal”.

Secondly, we accept the SFO’s submission that prior to the Fraud Act, innocent agency has long formed the basis of convictions that today would most likely be charged as fraud by false representation. Butt (1884) 15 Cox CC 564, is a telling example. Butt was charged under the Falsification of Accounts Act 1875, with making or concurring in the making of a false entry. Butt collected a sum of money from Sheppard for his employer; he did not, however, account for the full sum but only for a lesser amount. The entry of the lesser amount in his employer’s cash book was made by a clerk, Elford, who was an innocent agent in the matter. Butt’s conviction was upheld, Lord Coleridge CJ saying this (at p.567):

“ It is contended…that the statute is not broken, because the person who made the entry did not know it was false, and the person who did know it was false did not make the entry. There is high authority that, where a man who knew of the falsity of the representation he was making made such representation by means of an agent who was ignorant of its falsity, there is no fraud…but that was in a civil action, and is, I believe a decision not universally approved of. This is clearly a false entry as far as Sheppard is concerned. It purports to represent receipts from the persons who have been entered as making payment of such receipts, and it seems to me clear that the prisoner either made it with the innocent hands of Elford, or concurred in the innocent hands of Elford making it. I am of opinion that this conviction was perfectly right, and must be upheld.”

As it seems to us, if the Judge’s construction of s.2 is correct, then Butt could not be convicted under that section today. At the very least, that would be a startling outcome, the more especially in the light of the purpose of the Fraud Act and the mischief at which it was directed (see below). (Only in passing, we underline that we express no view on what the position would now be in a civil claim on those facts and would not wish to be taken as concurring with the observations in that regard in Butt.) 103. Thirdly, the mischief at which the enactment of the Fraud Act was aimed included the technical nature of the existing law: Smith, Hogan and Ormerod, at p.928, fn. 20. The objective (inter alia), as expressed by the Law Commission, in its paper, Fraud (Law Com No. 276, July 2002), at para. 1.4, was:

“…to ensure that the scope of the criminal law of fraud is wide enough to enable fraudsters to be successfully prosecuted and appropriately sentenced, without being so wide as to impose unacceptable restrictions on personal freedom, or so vague as to infringe the principle of the rule of law.”

Fourthly, against the background already outlined, it would be curious if Parliament, when enacting the Fraud Act, had impliedly intended to disapply the longstanding application of the common law doctrine of innocent agency in this area – so narrowing rather than widening the scope of the criminal law, thus at odds with the scheme of the legislation. That scheme is emphasised by the fact that the new law, unlike the old, is conduct rather than results based. Under s.2, there is no need to prove a result of any kind. The principal aim is “to make the offence easier to prove…”: Smith, Hogan and Ormerod, at p.937. Moreover, a further and likewise curious consequence of the Judge’s construction of s.2(2) is that while innocent agency would be excluded in respect of s.2(2), it would plainly remain available under s.2(5) where a representation is inputted to a system or device by way of “human intervention”.

Fifthly, although the Judge was correct to treat the Fraud Act as a penal statute and therefore to construe it strictly, an absurd construction, producing objectionable and undesirable consequences, is to be avoided in this sphere, as elsewhere: R v McCool [2018] UKSC 23, at [23] – [26], per Lord Kerr. A strict construction is one thing; an absurd construction is another.

Sixthly, powerful support for a construction different from that adopted by the Judge is furnished by two leading textbooks. Blackstone’s Criminal Practice (2019) says this (at para. B5.13):

“Section 2 defines the elements that make up a false representation. By s.2(2) a statement that is literally true may be treated as false if it is misleading, although this is complicated by the introduction in s.2(2)(b) of what is in effect a mens rea requirement masquerading as part of the actus reus…..”

So too, Smith, Hogan & Ormerod, while remarking (at p.953) that the Act “rather oddly” provides that “a statement is only false if D knows it is or knows it might be false” analyses the matter as follows (at p.937):

“The actus reus requires proof that D made a representation, which is untrue or misleading, and the mens rea requires proof that D knew the representation was, or knew that it might be, false and he acted dishonestly in making the representation, and with intent to gain or cause loss or expose to a risk of loss.”

Seventhly, in the light of all these overwhelming considerations, we cannot agree with the Judge’s construction of s.2(2), urged upon him by the Respondents. With respect, that construction leads to absurdity, having regard to the context and scheme of the legislation. We decline to accept that the Legislature intended the consequences which would follow from this construction and there is no good reason for the legislature knowingly to have opted for it. We decline to subscribe to the counsel of despair that the legislation might, unintentionally, have produced a construction productive of absurdity; though the wording of s.2(2) is, admittedly, not free of difficulty, we are not persuaded that we are driven to reach so unpalatable a conclusion.

For our own part, in agreement with the SFO’s case and supported by the views expressed (in particular) in Smith, Hogan & Ormerod, we are instead persuaded that the actus reus, the conduct element of the offence, is the making of an objectively untrue or misleading representation. The mens rea, or mental element is made up of the requisite knowledge, dishonesty and intention. The actus reus is therefore contained in s.2(2)(a), with the mens rea found in s.2(2)(b), s.2(1)(a) and s.2(1)(b). Such a construction occasions no difficulty when no question of agency is involved and permits the legislation to give effect to Parliament’s intention of simplifying and widening the scope of the law in this area. It remains a strict construction and does no violence to the language of s.2, provided only that the language of the section is read in context and the section is construed as a whole. To reiterate, the contrary construction, by necessarily precluding the doctrine of innocent agency, narrows the ambit of the criminal law when contrasted with its scope prior to the passing of the Fraud Act. For that, there is simply no warrant.

(F) Conclusion on Issue II: For the reasons given, we are persuaded that the Judge erred in law in this regard, thus permitting our intervention under s.67 (a) and/or (b) of the CJA 2003. We think there is a case to answer in respect of the innocent agency route and allow the appeal on this Issue from the Judge’s decision as set out in the Ruling. Innocent agency is applicable to both CR1 and CR2 and, hence to both Counts 1 and 2. It applies to all the Respondents, save that in the case of JV, it is subject to the outcome on Issue IV. Its fate at trial on the factual issues for the jury

Under this heading, on the premise that Barclays was not an “innocent agent”, the SFO’s case was that the Respondents committed the statutory offence of conspiracy by “participation” in it – incurring secondary liability by procuring the issue of false prospectuses by Barclays. That, by reason of the May 2018 Ruling and the Davis LJ judgment, Barclays had not committed an offence was neither here nor there. This argument was rejected by the Judge in the Ruling, at [249] and elsewhere.

As Ms Darlow QC accepted in oral argument, this further Issue only arises if the “innocent agency” route to liability is not available to the SFO. In the light of our conclusion on Issue II, this Issue therefore does not need to be determined. Given its complexity, we do little more than note it and draw back from expressing any concluded view.

The complexity arises in this way:

In the light of Kenning, Ms Darlow QC’s central submission is that the present case was one of a secondary party (allegedly) procuring the offence; procuring was different from aiding and abetting because there was far greater certainty that the result would be achieved. The authority of Kenning did not extend to bind us in respect of procuring, as distinct from aiding and abetting. In any event, there was a difficulty with the judgment of Lord Phillips in that liability in conspiracy crystallised upon the making of the agreement; it did not depend on the inevitability or certainty that the substantive offence would be committed.

Mr Purnell QC counters by submitting that there was no proper basis for distinguishing Kenning (or Hollinshead), given that (as already noted) Hollinshead was itself a case of procuring. In any event, no “special route” could be fashioned out of “procuring”, because s.1(1) of the CLA 1977 required that the offence committed, would be committed by one or more of the conspirators. That requirement could not be satisfied on the facts of the present case.

As already foreshadowed, we express no concluded view on this Issue. The question of whether procuring is different from aiding and abetting, so that Kenning is not binding, is not necessarily straightforward. It best awaits a case where the outcome turns on it. In any event and assuming that the SFO could have overcome that legal hurdle in its path, Mr Purnell QC’s factual submission struck us as formidable though, again, we have not reached a decision on it. Beyond this, we do not go.

Under this Issue, we address the SFO’s Grounds of Appeal, 5 and 6, going to the evidential sufficiency of the case against JV. The relevant test, here too, is contained in s.67, CJA 2003 (already set out).

In the present case, in addressing whether there was a case which could properly be left for the jury, the Judge identified the crucial question for each conspirator as being whether there was sufficient evidence of knowledge by him that no genuine services would be provided under ASA1 (for Count 1) and ASA2 (for Count 2) – i.e. that they were sham agreements.

Although the SFO had been resistant to the Judge’s insistence that this was a necessary element of its case, ultimately this was not and is not challenged.

The SFO invited us to consider first the evidential sufficiency of the case on Count 2. JV’s evidential footprint was greater in relation to Count 2 and if, as the SFO contends, the Judge’s approach was unreasonable and in error, then it submits that that would provide relevant context for the consideration of Count 1. We did consider the Counts in that order but, for the purpose of this judgment, we shall address the Counts chronologically. In so far as the SFO made criticisms of the Judge’s approach in relation to both Counts, these are addressed under Count 2.

With regard to the knowledge of JV in relation to ASA1, the principal findings made by the Judge are as follows:

In seeking to show that the Judge’s conclusion was not reasonable, the SFO emphasises, in particular, the following:

The SFO contends that (i) for ASA1 to be a stand-alone agreement for value, the Qatari entities would have had to have abandoned the agreed “subscription” fee of 3.25% just days later, (ii) that a reasonable jury would have been entitled to conclude that JV must have known that this proposition was unrealistic, and (iii) that JV must have known that the ASA was being used as a mechanism to pay what in truth were hidden commission fees and was not a genuine agreement for value.

Particular criticism is made of the Judge’s finding that the 3.25% fee to which JV agreed was a “composite” fee. Mr Brown QC for the SFO contends that up to this point all the documents refer to fees in relation to the proposed subscription and that this can be the only basis upon which the fee was being agreed. This was in reality a subscription fee and the Judge should have so found.

As Mr Purnell QC points out, this is to assume what the SFO has to prove. It is not in dispute that JV knew that an overall fee for the Qatari entities was agreed at 3.25% and that a means of paying them the extra 1.75% had to be found. The means or mechanism of so doing was yet to be identified, but it did not have to be unlawful or dishonest.

As the Judge explained at [306]:

“….The fact in issue for the jury is always the following: was ASA1 a disguise for an additional fee for subscribing or was it the intention of the parties that genuine services would be provided? Proof of the mechanism, without more, does not prove that the parties did not intend that genuine services be provided. The parties could have reached an agreement to that very effect notwithstanding that all the various elements of the SFO’s case were satisfied. This is because RJ and Sheikh Hamad could lawfully have agreed that Qatar would provide services and be paid for them on the basis of the 1.75%....”

JV was not involved in the discussion of the means by which the extra 1.75% was to be paid or in the development and negotiation of ASA1. JV did come to know that ASA1 was the agreed means of doing so and he signed it. As the Judge made clear, however, knowledge of the “mechanism” of ASA1 does not connote knowledge that it was a dishonest mechanism. The critical question, as the Judge correctly identified, was whether ASA1 was a sham, in that it was not intended that any services were to be provided under it, and whether JV knew that.

On its face, ASA1 was an agreement for services and the Qatari entities could easily provide services of real value under it. Barclays had been seeking a strategic relationship and such relationships were of commercial importance to it. Unlike the other conspirators, JV had not been involved in the development of ASA1 or participated in the exchanges which might suggest knowledge that it did not mean what it said. As Mr Purnell QC submits, JV had no more than “book-end involvement” in the development and execution of ASA1.

The Judge found that “the inference that JV must have been aware” that the “solution” to be found for paying the extra 1.75% “was going to have to be unlawful is not supported by the evidence”. Nor was there any direct evidence relating to JV to support the inference that JV knew that the solution found, ASA1, was in fact unlawful. On its face it was a lawful agreement for value. Barclays wanted a strategic relationship with the Qatari entities. ASA1 seemingly provided such a relationship and there was the possibility of services of real value being provided.

In the light of JV’s lack of involvement in the development and negotiation of ASA1, and in any of the exchanges potentially implicating the other defendants as identified by the Judge, we consider that the Judge was entitled to treat the case against JV differently to that against the other defendants.

For reasons set out below, we reject the criticisms of the Judge’s general approach, in so far as relevant to Count 1.

In all the circumstances, we consider that the Judge was entitled to rule that there was insufficient evidence upon which a reasonable jury could convict JV on Count 1. On any view that ruling was within the range of reasonable conclusions open to him.

With regard to the knowledge of JV in respect of ASA2, the principal findings made by the Judge are as set out below. Additional details and explanatory comments derived from the documents have been added in brackets. By way of general background these events took place during the height of the financial crisis (see above) and the Judge found the threat to Barclays during the last week of October 2008 to be “existential”:

At [601]-[614], the Judge considered whether there was a case to answer against JV in the light of the general findings set out above. The Judge observed that he found “the determination of whether there is a case to answer against JV on the issue of knowledge extremely difficult to resolve”.

The Judge’s conclusions were as follows:

The SFO contends that the Judge erred in his approach in various ways and that any and all of these errors are material to his “extremely finely balanced” decision and render his conclusion unreasonable. The principal alleged errors in approach are:

Failing to recognise the relevance of the findings made in relation to Count 1 when considering Count 2.

In relation to ASA1 the Judge stated as follows [601]:

“The SFO relies on the ASA1 history against JV even if there is no case to answer against him on Count 1. I was not impressed by that: on this premise, JV’s knowledge of ASA1 is a factor which, if anything, can be deployed in his favour, because his assumption would have been that it was legitimate. The SFO has JS’s point about what further advice do we need, but there is no evidence that JV was aware of the nature of the

“services”, if any, that had apparently been provided under ASA1.”

The SFO contends that this was a wrong approach and that, even if there was no case against JV on Count 1, the relevant factual context relating to Count 1 included the following: (i) that all investors were to be paid a fee of 1.5%; (ii) that JV knew and approved the agreement with the Qatari entities on 5 June which included a composite fee of 3.25%; (iii) that JV was aware that efforts were being made to pay the Qatari entities the extra 1.75% and (iv) that JV had not been “altogether forthcoming” with the Board.

The Judge was clearly right to consider that no adverse inference could be drawn against JV in relation to Count 1 in the light of his conclusion that there was no case to answer on that Count. Background facts (i) to (iii) reflect what the Judge found to be legitimate aspects of the arrangements made. In so far as the SFO suggests that adverse inferences should nevertheless be drawn from those facts, that would be inappropriate.

The SFO seeks to find support for its approach by relying on the Judge’s comment at [500] that “There may be some force in the contention that JV should have been entirely forthcoming with the Board and the BFC, if for no other reason that these arrangements were not free from difficulty, controversy and the taint of suspicion, and the issue could always return to bite Barclays’ back”. However, as Mr Purnell QC points out, the Judge was not here finding that Mr Varley himself had been suspicious of the genuine nature of the ASA; rather his point was concerned with potential public perception in the event of later controversy.

The main relevance of ASA1 as factual context was that Barclays already had an ASA for three years at a fee of £42m. That raised questions as to the need for a further ASA for five years at a fee ultimately agreed of £280m. The Judge, however, recognised this and referred to the observation made by JS of “what further fees do we need”. But, as the Judge found, JV was not aware of what services had or had not been provided under ASA1. ASA2 also provided for more specific services, for a longer period and over a wider geographical area.

The Judge accordingly had appropriate regard to the findings made in relation to Count 1 and the factual context and there was no error in his approach.

Wrongly giving weight to JV’s positive good character.

The Judge made reference to JV’s good character when considering the case against him (e.g. [492] and [504] in relation to Count 1 and [601] in relation to Count 2). The SFO contends that this was wrong in principle. In this connection they relied on a single sentence from the judgment of the Court given by Hallett LJ in R v Bush, Scouler [2019] EWCA Crim 29. At [64] Hallett LJ stated, without comment, that the trial judge, Sir John Royce, in his terminatory ruling, having concluded that there was no case to answer, added a “few further matters” which included the statement that the respondents’ good character was “irrelevant to his decision” but that “nothing in their character supported the Prosecution case”.

In principle, we agree with the SFO that the “good character” of a defendant is ordinarily irrelevant when considering a submission of no case to answer. At this stage, the sole focus is on the strength of the prosecution case not the evidence, including good character, which may be adduced by the defendant. On any view, the credibility aspect of good character is irrelevant at the half time stage of a trial. The propensity aspect of good character is, perhaps, less straightforward. Over and above the consideration that good character evidence forms part of the defence and not the prosecution case, caution should be exercised in placing reliance on the propensity aspect of good character, given that the direction ultimately given may have to be tailored in the light of all the trial evidence. We would not, however, accept that it is impermissible in all cases to have regard to propensity when considering the inherent probability of the prosecution case. Indeed, Mr Brown QC for the SFO accepted that it was arguably relevant in that context, and that was the context in which the Judge was referring to good character (see, for example, the reference to good character at [284] in a section headed “Inherent Probability”).

In the final analysis, whether the Judge was correct to refer to good character in relation to the inherent probabilities does not, however, matter since he made it clear that if he had otherwise considered there was a case to answer on the evidence then JV’s good character “would have to yield” [504]. It was not therefore material to the decision he reached.

Inappropriately “entering into the arena” and making findings on matters for the jury.

The SFO submits that on occasion the Judge usurped the function of the jury and inappropriately entered the arena. We reject this as a general criticism. It is simply not made out. The Judge had well in mind his duty to take the prosecution case at its highest and to consider that case from the perspective of a hypothetical reasonable jury, properly directed.

The more troubling matter here and the high-water mark of the SFO’s complaint was the Judge’s comment in relation to the evidence of Mr Agius at [606]:

“606.

Mr Agius’ moral outrage that JV did not discuss the £280M fee with him did not particularly impress me at the time. He could not have separated after-acquired knowledge from what he knew at the time. We now have convincing evidence that a nine-figure fee for advisory services was discussed at Board level, and Mr Agius has forgotten that. Mr

Agius’ judgment that JV was a man of utmost integrity is also relevant here. JV’s ethics appear to have differed from others in bank. Mr Agius was not aware of the brevity of the timescales in the context of the opportunity he believes he should have been afforded. It is possible that the outrage would have been better directed elsewhere.”

It is submitted that the impression made by a witness is an obvious jury matter and there is plainly force in that contention. Moreover, given the understandable concern as to JV’s failure to mention the £280M fee to Mr Agius, with whom he was in close contact, we very much doubt that we ourselves would have expressed the matter in this way. That, however, is not the question and we bear in mind that the Judge saw and heard Mr Agius give evidence. That evidence was to the effect that Mr Agius apparently had no recollection of the relevant Board meetings and had seemingly forgotten that advisory services for a fee of £115m had been raised at the 26 October 2008 meeting. Viewed in that light, we are unable to treat the Judge’s comment, essentially confined to Mr Agius’ moral outrage, as making good this SFO criticism. We deal separately (below) with the inferences to be drawn from JV’s silence with regard to the £280M.

Mr Brown QC also objects to the limitations placed by the Judge on his questioning of Mr Agius and we were taken to the relevant parts of the trial transcript. In particular, Mr Brown QC complains about not being allowed to ask Mr Agius about whether he would have expected JV to mention the £280m fee to him. In our judgment this was a matter of discretionary trial management. In any event, regardless of what Mr Agius may have said, this was a point which the SFO could and did make in submissions. We are not persuaded that this SFO contention has any real force.

Failing to approach the case of JV in the same way as he did in finding a case to answer against the remaining defendants and CL.

Mr Brown QC places particular emphasis on the reliance placed by the Judge, in relation to the case against those other than JV, on the lack of commercial negotiations in respect of the advisory services. At [555] the Judge stated as follows:

“555.

Additionally, it is a compelling feature of the evidence, or rather its absence, that there was no separate commercial negotiation between both parties about the advisory services that would be provided, putting to one side their value. This was all done through the lawyers. I have not been shown any proper instructions given to the lawyers from Barclays’ side.”

This was a comment made in relation to ASA1, but JV was not involved in its evolution or negotiation. ASA2 set out the services to be provided in more detail, and lawyers were involved in the drafting of each of its iterations.

As to the value of the services, and the increase to £280m, the Judge recognised this as an important feature in the case against JV on Count 2, as borne out, for example by [604], where the Judge recognised that:

“….If JV had no real choice, and Qatar appreciated that, the possibility of dishonesty becomes more substantial. As the fee rocketed upwards, would a reasonable commercial man begin to believe that what was driving this was not genuine services but an unswerving and intractable desire to secure as much money as possible from an ailing British bank?”

In our judgment there was no imbalance as alleged in the Judge’s approach.

Wrongly considering that the case against JV would have involved him participating in a different conspiracy and determining the application on that basis.

The Judge found that there was no case that JV joined the conspiracy on 24 October 2008, as the SFO had contended, but nevertheless went on to consider the possibility that he joined on 29 October 2008 when he and RJ struck the final deal on ASA2 for a fee of £280m. In [611] the Judge described this as being a “different conspiracy”.

The SFO contends that the Judge thereby erred. The period of the indicted conspiracy was 1 September 2008 to 30 November 2008. A conspirator may join the alleged conspiracy at any time during that period. If JV joined on 29 October 2008 he would have been joining the existing CL/RJ conspiracy in which they all agreed to carry their criminal scheme into effect. That did not involve different objectives or different intentions. In substance, in objective, in method, it was exactly the same conspiracy.

In our judgment, insofar as the Judge was treating this as a separate conspiracy in law, we would be unable to agree with him. However, we think that the Judge (if at some length) was doing little more than making the point that, if there was no case fit to go to the jury on JV’s participation in the conspiracy as at the time of the Board meeting on 26 October, then developments between 26 and 31 October did not transform the position. Having regard to the individual perspective of JV, if ASA2 was not a sham on 26 October, when the Board was apprised of a payment in the region of £115M or £125M, how, it might be said, did it become a sham when a payment of £280M was agreed on 31 October? That, in reality and as it seems to us, was the timing question with regard to JV which the Judge was addressing (and to which we return below).

Accordingly, we do not consider that the Judge erred in his approach, or, if he did, that any such error was material to his decision on Count 2, based on his evaluation of all the evidence. In so far as the alleged errors in approach also apply to Count 1, we reach the same conclusion.

In arriving at that conclusion, we have also had regard to the more subsidiary points made by the SFO in relation to Count 2, such as comparisons to the position of those not charged, the Judge’s characterisation of the prosecution case as mechanistic, the inference drawn by the Judge at [582], and the relevance of the description of ASA1 as being a “great success”. None of these matters undermine the conclusion that he reached.

Having failed to demonstrate a specific error in approach, the SFO is accordingly left with having to meet the “high hurdle” of persuading this Court that on the evidence

In this regard, the SFO emphasises the following findings made in relation to JV and Count 2: (i) JV had a number of direct contacts with the Qatari entities when negotiating CR2; (ii) JV had said that the Qatari entities would be “very demanding on the economics”; (iii) JV, when he agreed the CR2 deal with Sheikh Hamad on 24 October 2008, knew that the demand for “additional value” was going to go into another ASA, then for £125m, and later for £280m, as agreed to by JV.

Against that background the SFO contends that:

In the light of these facts and matters it is submitted that the Judge should have concluded that there was a sufficient evidential case to put before the jury and it was unreasonable for the Judge to conclude otherwise.

The Judge’s reasoning proceeded in stages. First, he considered the evidence that JV was part of the conspiracy on 24 October 2008 and he concluded that there was no evidence to support such a case. We do not consider this conclusion to be unreasonable, so far as it concerns JV. At this stage ASA2 was being considered by the lawyers, who were advising that it was lawful and irrelevant to financial assistance and commissions, provided that it was on commercial terms. The lawyers were also aware that fees of £125m were contemplated. As the Judge found, “the £125M fee in the advisory agreement, or an amount in that region, had been approved by the lawyers” and “it is inconceivable that JV was somehow negotiating with Sheikh Hamad without being aware that favourable legal advice had been given” [576].

This was then followed by the Board meeting of 26 October 2008, at which the Judge found, as borne out by the evidence he refers to, that the ASA or co-operation agreement was mentioned by JV at a fee level of £115m.

JV was accordingly going into further negotiations with Sheikh Hamad in relation to an ASA with fees of over £100m which had been put before the lawyers and the Board. He was also going into that negotiation on the basis, on the Judge’s findings, that he had no knowledge of or involvement in any conspiracy involving ASA1 or, up to that time, ASA2. Essentially, all that changed thereafter was the increased fee which the Qatari interests demanded and JV and RJ agreed to.

The Judge raised the question of whether such an increase would have caused JV to question the commerciality of what was being agreed, but what JV had to realise, for the ASA to be a sham, was that this meant that no genuine services were to be provided, not merely overvalued services. Moreover, the context was that an agreement for services valued at over £100m had been before the Board. JV would also have had to agree with RJ to go ahead with ASA2 in the light of that knowledge and to participate in a conspiracy, about which he had only just learnt, to further a sham agreement and make false representations. The Judge described this as “inherently highly unlikely” and we cannot say that that was a view not reasonably open to him.

There were, plainly in our view, factors which provided support for the SFO case, such as, in particular, the fact that there was a case to answer against RJ, JV’s involvement with RJ in the negotiations for ASA2 and the failure to mention the increased fee of £280m to Mr Agius or, according to the Board Minutes, to the Board. These give us pause for concern; the failure to mention the significantly increased figure to Mr Agius (with whom, as already underlined, JV was in close contact) was striking; so too was the failure to ensure that the £280M was mentioned to the Board and, indeed, recorded in the Minutes, as soon as possible after the 31 October agreement. Our concern is heightened by the Judge’s mistaken interpretation (at [582]) of the Board Minutes of 27 October; it seems clear that the fees in question were not the ASA2 fees but related instead to the placees. Still further, we have not, in all this, overlooked JV’s lack of attention to the detail of the services to be provided by the Qatari entities or to the basis for the £280M fee. All these are troubling matters.

Conversely, we cannot ignore the following features of the case, posing obvious difficulty for the SFO, keeping well in mind the clear demarcation between the roles of Judge and Jury:

In the upshot, we find Issue IV finely balanced. That, of itself, is a matter of no little significance, having regard to the relevant test under s.67(c). In the event, we are not persuaded that (the error at [582] apart), the Judge failed to have regard to any significant evidence or had regard to evidence which he should not have done. Ultimately and whatever our reservations, the Judge was in a far better position to reach an overall conclusion as to whether that evidence provided a sufficient case to be left to the jury than this Court. It is possible that another judge might have reached a different conclusion, but that is not the test. The SFO has to show that the conclusion to which the Judge came was not within the range of reasonable conclusions open to him. Having carefully considered the parties’ extensive written and oral submissions, and the limited evidence which we have been shown, we do not consider that the SFO can overcome this “high hurdle”.

In our judgment the SFO have not established that the Judge’s ruling in respect of JV on Count 2 involved an error of law or principle or was not a reasonable ruling to have made.

For the reasons outlined above, the SFO’s appeal on Issue IV must be dismissed.

For completeness, although the Respondents made not infrequent criticism of the framing of the indictment, nothing has turned on that. We have of course approached the matter in the light of the charges brought rather than those which might have been brought.

In the light of our conclusions on Issues I, II and IV, the proceedings are at an end so far as concerns JV. However, a trial can proceed against RJ, TK and RB on Count 1 and RJ on Count 2. In all the circumstances, that trial should proceed before a different High Court Judge, to be designated by the President of the Queen’s Bench Division.