Abstract
This article considers two areas that arise in the context of corporate criminal investigations relating to claims of legal professional privilege: the extent to which litigation privilege may attach to communications made in the context of such investigations and the difficulty of identifying the client for the purposes of legal advice privilege. These issues are of particular significance where a company is or may be the subject of an investigation by specialist prosecuting authorities, such as the Serious Fraud Office. We identify the policy considerations justifying litigation privilege and whether they continue to explain the current ambit of the privilege. With particular reference to the extent to which the privilege is capable of attaching to communications made for the purpose of working towards a potential settlement, we consider how the constraints upon its ambit operate in the context of corporate criminal investigations. In relation to legal advice privilege, we demonstrate that it is possible to give a coherent explanation of the jurisprudence in this area which, while accepting that decisions are fact-specific, should enable corporations and the courts to identify the client within the corporation with a greater degree of confidence.
Keywords
Introduction
Companies are becoming potentially criminally liable for an increasing number of offences. Examples include failure to prevent bribery 1 and more recently failure to prevent the facilitation of tax evasion. 2 Concerns about potential criminal liability can lead to internal investigations by a company, whether in response to an investigation by bodies such as the Serious Fraud Office (SFO) or due to concerns that the company has been doing something it should not in this context, which might, for example, result in a decision to self-report in order to potentially avoid prosecution. 3 Similarly, internal corporate investigations may take place in the context of preparing for and presenting evidence to an external inquiry. Internal investigations, interactions with investigating bodies and legal advice taken in relation thereto may all produce communications which may be the subject of claims of legal professional privilege, both litigation privilege and legal advice privilege.
The ambit of litigation privilege is subject to three significant constraints, namely, that litigation must be in reasonable contemplation, that it must be at least the dominant purpose of the communication and that the litigation must be adversarial. When determining whether litigation privilege attaches to communications made in the context of a corporate criminal investigation, the crucial issue is whether the 21st-century courts will apply these constraints narrowly so as to limit the ambit of a privilege which had its origins in the adversarial litigation of an earlier age. The criminal process is now less adversarial/more case managed. Further, companies and the courts are now grappling with very different regulatory and investigatory regimes which lead to increasingly blurred lines around when litigation will be in contemplation. Where litigation privilege is not available, for example, because litigation was not in reasonable contemplation at the time when an internal investigation took place, then the availability of legal advice privilege is crucial to protect communications with lawyers from disclosure. In the corporate context, legal advice privilege may be unavailable if communications with the company’s legal advisers are not made by a person or body within the company designated as the client. In addition, the type of communications taking place between the client and the lawyer impacts on the availability of legal advice privilege.
This article begins by exploring the ambit of litigation privilege in corporate criminal investigations. It identifies the policy considerations that traditionally underlie the existence of the privilege. It considers the extent to which the ambit of the privilege has been the subject of a more recent ‘policy of confinement’ by the judiciary. It also considers how the constraints upon the ambit of the privilege impact on its operation in the context of corporate criminal investigations. In assessing the potential scope of the privilege, it takes into account the fact that the role of lawyers in such scenarios goes beyond preparation for litigation and encompasses communications for purposes such as avoiding contemplated criminal litigation, perhaps by entering into a civil settlement or considering the merits of entering into a Deferred Prosecution Agreement (DPA).
The identification of the client for the purposes of claiming legal advice privilege in the corporate context is then discussed. The major issue is that while the authorities require the court to identify a client within the corporation, they provide limited guidance on how to do so. Based on consideration of attribution theory and the nature of delegated authority as it operates in the corporate context, and with reference to the membership of, and actions performed by, the client group within the corporation, we demonstrate that it is possible to give a coherent explanation of the jurisprudence in this area which, while accepting that decisions are fact-specific, should enable corporations and the courts to identify the client within the corporation with a greater degree of confidence.
The final section of the article provides advice for corporations on maximising the protection provided by litigation and legal advice privilege in the corporate context.
Litigation Privilege and Corporate Criminal Investigations
In considering how the courts should approach a claim of litigation privilege arising from a corporate criminal investigation, it is important to identify the policy considerations underlying the existence of the privilege and to consider whether they are applicable in the context of the 21st-century English and Welsh criminal process.
Unlike legal advice privilege,
4
protecting the confidentiality of communications between the legal adviser and the client does not form the sole policy justification underlying the existence of litigation privilege.
5
Rather, litigation privilege (which, arguably, can arise where proceedings are conducted by a litigant in person, with no legal adviser involved
6
) is fundamentally a creature of the adversarial trial process rather than one of the lawyer and client relationships. As Lord Rodger of Earlsferry explained, it is based on the idea that legal proceedings take the form of a contest in which each of the opposing parties assembles his own body of evidence and uses it to try to defeat the other, with the judge or jury determining the winner. In such a system each party should be free to prepare his case as fully as possible without the risk that his opponent will be able to recover the material generated by his preparations.
7
A Trend of Confinement?
In The Director of the Serious Fraud Office v Eurasian Natural Resources Corporation Ltd, Andrews J identified a trend of confining the scope of litigation privilege. 11 The essence of such confinement is the existence of three constraints that the courts have imposed upon its operation, namely, that (a) litigation must be in progress or in contemplation; (b) the communications must have been made for the sole or dominant purpose of conducting that litigation; (c) the litigation must be adversarial, not investigative or inquisitorial. 12
The trend of confinement had commenced in the 19th-century jurisprudence which restricted the ambit of the privilege to material ‘obtained for the purpose of litigation’ 13 and indicated that it only arose where ‘litigation [was] existing or contemplated between the parties’. 14 In justifying the imposition of the dominant purpose test, Lord Edmund-Davies in Waugh v British Railways Board believed that ‘the public interest is, on balance, best served by rigidly confining within narrow limits the cases where material relevant to litigation may be lawfully withheld [because] [j]ustice is better served by candour than by suppression’. 15 When determining that the operation of the privilege is confined to adversarial proceedings, Lord Jauncey of Tullichettle in In Re L (A Minor) (Police Investigation: Privilege) indicated that, ‘in…proceedings…which are primarily non-adversarial and investigative as opposed to adversarial, the notion of a fair trial between opposing parties assumes far less importance’. 16
Regarding the trend of confinement, in Eurasian, Andrews J referred to the speech of Lord Scott of Foscote in Three Rivers District Council and Others v Governor and Company of the Bank of England (No. 6), in which Lord Scott had opined that ‘[c]ivil litigation conducted pursuant to the current Civil Procedure Rules (CPR) is in many respects no longer adversarial’ and that ‘[t]he decision in In re L warrants…a new look at the justification for litigation privilege’.
17
It is suggested that contrary to Lord Scott’s view, while In Re L, like Waugh, is clearly a decision that confirms the limits of litigation privilege, it, again like Waugh, is not a decision that is inconsistent with the underlying rationale of litigation privilege as a creature of adversarial proceedings. Indeed, neither the decision in In Re L nor the fact that civil proceedings conducted under the CPR are less adversarial than was the case in the 19th century warrant a new look at the justification for litigation privilege. What is warranted is an examination of the nature of specific types of proceedings for the purposes of which communications were made in order to determine whether the adversarial justification for litigation privilege is applicable in their specific context (which it was held not to be both in the context of the care proceedings that In Re L concerned and in that of the private, non-statutory, inquiry which Three Rivers No. 6
18
concerned). For example, where disclosure of communications with expert witnesses is sought in the context of care proceedings but the communications were made for the purposes of criminal proceedings, the communications are privileged because they were made for the purposes of adversarial litigation.
19
Moreover, in order for litigation to qualify as adversarial for the purposes of the privilege, it is not necessary for proceedings to possess all of the adversarial features of the traditional 19th-century criminal or civil trial. Rather, as the Upper Tribunal (Administrative Appeals Chamber) indicated when deciding that the privilege attached to communications made for the purposes of proceedings before the First-tier Tribunal,
20
[l]itigation privilege was available for the old prerogative order proceedings and applies to modern judicial review proceedings, although many of the features of a fully contested adversarial contest would be absent in such cases. We see no basis for not regarding proceedings before the First-tier Tribunal as litigation for the purposes of legal professional privilege in both of its aspects
Criminal Proceedings, Criminal Investigations and Attempts to Settle
So far as criminal proceedings are concerned, the Criminal Procedure Rules (CrimPR) impose a variety of case management duties and confer a variety of case management powers upon the court. The Criminal Procedure and Investigations Act 1996 imposes disclosure obligations upon both the prosecution and, to a lesser but still significant extent, the defence. Even so, there is no doubt that the modern English and Welsh criminal trial remains, in essence, an adversarial contest in which the defence is entitled to rely upon the traditional adversarial safeguard of litigation privilege. In his report, which catalysed the creation of the CrimPR,
21
Auld LJ took ‘as given…a continuation of a trial procedure that is in the main adversarial…’. Specifically, when considering whether the defence should be required to disclose unused experts’ reports as a means of combatting ‘expert shopping’, he opined that, [s]o long as our system remains adversarial, I can see no proper basis upon which the defence should be required to disclose material of this or any sort that is unfavourable to their case. There is undoubtedly a lack of parity between the prosecution and the defence in this respect, but that is a necessary consequence of where the burden of proof lies.
22
A related issue not considered in Eurasian is whether the process of agreeing and approving a DPA amounts to adversarial litigation. In Serious Fraud Office v Rolls-Royce PLC, 24 Levison P, considering an application for approval of a DPA, did not appear to question the view of Rolls-Royce that legal professional privilege could arise in relation to the interview transcripts produced during an internal investigation. Given that privilege had been waived by Rolls-Royce, there was no analysis by Levison P of the merits of the claim. Since the process of obtaining a DPA, under Schedule 17 to the Crime and Courts Act 2013, is, essentially, one of the agreements between the prosecutor and the person whose prosecution is under consideration followed by an application to the court by the prosecutor for approval of the DPA, it is submitted that the process via which a DPA is obtained does not amount to adversarial litigation. It is suggested, however, that communications made for the purpose of considering whether it is desirable to attempt to follow a DPA route rather than defend a prosecution, or for the purpose of avoiding prosecution by enhancing the likelihood that a DPA might be obtained, should properly be classified as falling within the ambit of the privilege.
In Eurasian, Andrews J
25
treated the avoidance of adversarial litigation as a purpose that fell outside the ambit of the privilege, though she did accept that [i]n theory, it is conceivable that documents could be generated for the purpose of assisting a company to persuade the SFO not to prosecute but also, if that failed, to help it mount a defence to criminal proceedings; but the evidence in this case does not establish such a dual purpose, let alone that the latter purpose was the dominant one. ‘if litigation is reasonably in prospect, documents brought into being for the purpose of enabling the solicitors to advise whether a claim shall be made or resisted are protected by privilege, subject only to the caveat that that is the dominant purpose for their having been brought into being. in the first instance the hope [may well have been] that litigation would be unnecessary [it was not] possible to distinguish between the purpose of taking legal advice concerning one’s rights or obligations, where that necessitates acquiring information from third parties, and the additional purpose of using that information in aid of litigation, should litigation be necessary, as long as litigation is reasonably in prospect. That is not…an example of a dual purpose which prevents the purpose of using information in aid of litigation from being dominant.
29
Andrews J in Eurasian held 32 that documents created for the purpose of showing them to the other party are not privileged, and that this is so whether they are created to persuade the other party to settle or to persuade the other party not to commence proceedings. Based upon the arguments immediately above, it is suggested that litigation privilege would attach to documents created for the dominant purpose of enabling the legal advisers to advise whether to try to persuade the SFO to follow a civil settlement route (or enter into a DPA), even though it was also intended that if this route were followed, privilege would be waived and they would be disclosed to the SFO. Indeed, Hanna suggests that where documents are created for the purpose of showing them to the other party, they may still be created for the purposes of litigation and thus litigation privilege may still attach to them, with privilege being waived if and when they are disclosed to the other party. 33 It is certainly arguable that such documents should be regarded as privileged if they were created at a stage when the legal advisers and client were still deciding whether to go down a civil settlement or DPA route with the intention that they would be taken into consideration when making that decision and only disclosed (with consequent waiver of privilege) if the ultimate decision was to follow a civil settlement or DPA route.
An issue that has the potential to give rise to significant practical problems when determining whether litigation privilege attaches to communications made prior to the commencement of criminal litigation is at what stage prior to such commencement criminal litigation may properly be said to be in contemplation. To adopt a commonly applied formulation of this test, the question is when is criminal litigation ‘reasonably in prospect’? 34 The Court of Appeal in United States of America v Philip Morris Inc and others, recognising that ‘[s]ome concepts are difficult to express in words’, considered that the judge in the instant case had correctly concluded that litigation privilege does not attach where there is a ‘“mere possibility” of litigation’, ‘a distinct possibility that sooner or later someone might make a claim’ or ‘a general apprehension of future litigation’, but also accepted that the judge had not suggested that in order for litigation to be reasonably in prospect ‘there must [be] a greater than 50% chance of litigation’. 35 In confirming its view that litigation had not been reasonably in prospect at the relevant time, the Court of Appeal had recourse to ‘the traditional justification for litigation privilege’, concluding that the retainer that the case concerned ‘had nothing to do with the preparation of the brief for a trial’. 36
In Eurasian, Andrews J rejected the submission that once an SFO criminal investigation is in reasonable contemplation, then a criminal prosecution will also be in reasonable contemplation on the basis that the two do not necessarily equate. 37 She accepted that there may be circumstances in which the two do equate (where there is an awareness of circumstances that will make a prosecution likely once they are discovered) but treated the question as one which is fact specific and which must be considered ‘on a case by case basis’. She suggested that knowledge of accusations only raises a real prospect of prosecution once it is discovered that there is some truth in the accusations or at the very least that there is some material to support the allegations of corrupt practices. 38 She distinguished criminal from civil proceedings on the basis that the only inhibition on the commencement of civil proceedings that lacks a foundation is the imposition of a retrospective sanction, 39 whereas the commencement of criminal proceedings is subject to the satisfaction of a test that has an evidential limb and a public interest limb. 40
The Significance of Eurasian
Carter regards the ‘alarm’ caused by Andrews J’s decision in Eurasian as ‘substantially overstated’ and its encroachment upon the protection provided by the privilege as not serious, because her decision was one that was dependent upon the specific facts of the case before her. 41 He suggests that this fact-specific treatment of Eurasian is one that might commend itself to the Court of Appeal. That Eurasian is a case that can readily be distinguished on its facts was effectively confirmed by the Chancellor of the High Court, in a case in which litigation privilege was claimed and Eurasian was cited but was not regarded as ‘determinative’, the Chancellor, accepting that the application of the dominant purpose test required him to make a fact-specific determination and also the fact-specific nature of the question whether litigation had been in reasonable contemplation, recognised that he had to focus on the facts of the case before him and not those of Eurasian. 42 In a more recent decision of the criminal division of the Court of Appeal, R v Jukes, 43 the court agreed with Andrews J’s view from Eurasian concerning the nature of those circumstances in which criminal prosecution is in reasonable contemplation. Their Lordships held that litigation privilege did not attach to communications in circumstances in which there was no evidence that, at the time when the statement was made, the company ‘had enough knowledge of what the investigation would unearth’ such that ‘it could be said that they appreciated that it was realistic to expect the Health and Safety Executive to be satisfied that it had enough material to stand a good chance of securing convictions’. 44 They indicated that the mere fact that the Health and Safety Executive normally prosecutes, ‘where there is a death and on the face of it a breach of duty’, did not compensate for ‘the critical absence of evidence’. Moreover, the fact that the statement had taken the form of a s. 9 statement 45 was, in itself, not sufficient to establish reasonable contemplation of prosecution.
The approach in Jukes and Eurasian has, most recently, been endorsed by the Divisional Court, though in circumstances in which its views upon the issue of privilege clearly did not form part of the ratio of the court’s decision. 46 Conversely, in giving leave to appeal in Eurasian itself, the Civil Division of the Court of Appeal had previously indicated that the appeal in Eurasian had a ‘real prospect of success’. 47 It is worth noting that in Jukes, the Criminal Division of the Court of Appeal made no reference to previous jurisprudence of the Court of Appeal or the House of Lords/Supreme Court concerning the ambit of litigation privilege, merely relying on the first instance decision in Eurasian. Nor did they refer to the subsequent decision of the High Court in which Eurasian was not regarded as determinative. 48 It also held that the party attempting to rely on privilege in that case (not the company), who had never been the client of the legal adviser, had not been entitled to claim privilege on his own behalf (the company had not attempted to claim it). Thus, even if the statement in Jukes was privileged, the claim of privilege in that case would have failed. Fundamentally, post Jukes it still appears to be open to the Court of Appeal in Eurasian to treat the decisions in Eurasian and Jukes as not being ‘determinative’ but at most as being decisions on the specific facts of the two cases, with the possibility of arguing that Jukes was decided per incuriam and/or could be distinguished on its facts.
The area of Andrews J’s judgment that Carter regarded as being ‘open to possible question’ (while recognising that the fact-specific nature of that judgment limited the extent to which it was necessary to question it) concerned the relationship between criminal investigations and reasonable contemplation of criminal proceedings. He asserted that, a criminal investigation, or the reasonable expectation of a criminal investigation, can give rise to a reasonable expectation of litigation. That litigation might be by way of challenge to the lawfulness of a search warrant, or of the conduct of a search, or of a notice requiring production of material. A belief that a thorough investigation will result in a decision not to prosecute does not preclude a reasonable expectation that litigation of some sort will arise. One example is the fact that sometimes prosecutions are instituted but then abandoned. 49
Moreover, Hanna questions Andrews J’s implicit assumption that a party who lacked an arguable case would proceed with civil proceedings on the basis that ‘it is a questionable empirical assumption [which] seems to ignore the opportunity costs’. 50 He also points out that where the basis upon which a criminal investigation is in reasonable contemplation is that there is ‘verifiable or substantiated evidence…of wrongdoing’, it is arguable that criminal proceedings will also be in reasonable contemplation. 51
As was seen above, the Criminal Division of the Court of Appeal indicated that the mere fact that the Health and Safety Executive normally prosecutes in particular circumstances is not in itself sufficient to justify the conclusion that relevant communications were made in reasonable contemplation of litigation. 52 It is suggested, however, that there will be circumstances in which it is proper to treat a suspect who knows or believes herself to be innocent as having a reasonable contemplation of criminal prosecution once, for example, an arrest has been made or a search warrant has been issued. The fact that the suspect knows or believes herself to be innocent does not inherently make such contemplation unreasonable since it must be common knowledge that persons who are subsequently found not guilty of criminal offences have still been prosecuted in the first place. Why should this be any different for a corporation/its directors or employees who are faced with an SFO investigation, even if they believe that they are not guilty of any offences? The SFO’s Annual Report and Accounts 2016–17 indicates during 2013–17 average conviction rates of 70.3% when calculated by defendant and 82.6% when calculated by case. 53 In 2016–17, the SFO had ‘around 70’ active criminal investigations, with 25 defendants being charged, 16 prosecutions concluded or in progress, 3 investigations concluded without charge, 33 defendants awaiting trial and 2 DPAs secured. Thus, given the number of prosecutions that arise out of a relatively limited number of SFO investigations and the fact that the conviction rate is not 100%, once such an investigation commences there must be potential for suspects who know or believe that they are innocent to still be in reasonable apprehension of prosecution which might result in a not guilty verdict.
Legal Advice Privilege and Corporations—Identifying the Client Group and Preserving Privilege
It is well demonstrated that there are difficulties surrounding the identification of the client for legal advice privilege purposes in the corporate context. 54 These difficulties are clearly significant where legal advice is sought in the context of a corporate criminal investigation/an internal investigation, particularly where litigation privilege does not attach to the communications. Difficulties primarily arise because a corporate body must inevitably act through agents. Whether legal advice privilege attaches to communications may depend upon the particular authority of these agents to instruct/communicate with the company’s lawyers and the types of activities undertaken by the agents. Unless employees of the company possess the requisite authority and therefore constitute ‘the client’, legal advice privilege will not attach to communications between them and the corporation’s legal advisers. If litigation privilege does not apply, the communications will therefore not be protected by legal professional privilege at all. 55 The approaches taken by the courts to determine which agents of the company constitute the client can seem lacking in clarity and certainty. Using attribution theory and considering the nature of delegated authority, it is possible to rationalise these approaches and draw conclusions regarding the delegation of authority within a company to a client group, and the actions of that group and the company’s legal advisers, to ensure that communications between them benefit from legal advice privilege.
Identifying the Client
The starting point in answering the question ‘who is the client?’ in the corporate context will normally be the board of directors, as the body within the company having authority to seek legal advice on behalf of the company. 56 Hildyard J refers to the ‘attribution’ argument, 57 referring to the judgment of Aikens J in Winterthur Swiss Insurance Company and Others v AG (Manchester) Limited (in liquidation) and others. 58 As expressed by Aikens J, the argument goes that, in the corporate context, legal advice privilege cannot attach to communications made by employees to the legal adviser, where those employees are not themselves part of the directing mind and will of the corporate client. The status of communications made by such employees as far as legal advice privilege is concerned is analogous to that of communications by or to other independent third parties who are not the client, which would equally not be covered by legal advice privilege. 59 Hildyard suggests that the restrictions of the attribution argument will often reflect reality, as a corporation is likely to restrict the authority to seek or receive legal advice on its behalf to ‘an individual or body’, that is its directing mind and will. 60 However, the board can usually delegate its authority 61 so to what extent can the authority to instruct legal advisers (i.e. the authority to form part of the directing mind and will of the corporate client for the limited purpose of preparing and communicating instructions) be delegated effectively within the company without losing privilege.
The decision in Three Rivers District Council & Others v The Governor & Company of the Bank of England (No. 5) 62 (Three Rivers) indicates that a designated person or body can be the client for legal advice privilege purposes. It was conceded in Three Rivers that the body established by the Bank of England (the Bingham Inquiry Unit (BIU)) to deal with communications between the Bank and the Bingham Inquiry, and in that context to seek and receive advice from the lawyers, was the client. 63 There was therefore no further analysis of the position that the BIU held in the Bank and the attribution theory was not explored. However, it could be argued that the BIU represented the directing mind and will of the Bank for the purpose of instructing/communicating with the Bank’s lawyers. The position was that three officials of the Bank of England were appointed by the Governor to deal with all communications between the Bank and those conducting the Bingham Inquiry. In connection thereto, the BIU sought and received legal advice from the Bank’s solicitors. It is not apparent that the authority delegated to them by the Governor meant that they and they alone made decisions regarding presentation of evidence to the Bingham Inquiry. In essence, ‘The preparation and communication of information and instructions to the bank’s legal advisers to enable them to advise pursuant to their retainer was the central role performed by the BIU’. 64 So, the BIU appears to have been given authority to create information and communicate with the lawyers, which included giving them instructions.
It could be argued that the restrictive approach adopted in Three Rivers of requiring the court to identify a specific client within the company was a consequence of the court applying public policy considerations in order to impose a constraint on the ambit of the privilege. Higgins suggests that in Three Rivers, there were public policy issues at play in the court’s decision-making, to prevent large amounts of information being withheld from disclosure. 65 It is certainly true that reference was made to a national institution gathering evidence for a Government inquiry, although this was in the context of assessing the dominant purpose for the preparation of documentary material, 66 rather than in relation to restricting the client to the BIU. However, it is not such a stretch to suggest that the public nature of the Bank 67 may have been some incentive to restrict privilege to communications from and to the BIU rather than encompassing a wider group. 68 It is possible that such an approach could be taken as a way to keep the ambit of the privilege within reasonable bounds where a company had chosen to delegate authority to a wide group, designated as the client, in order that as many communications as possible benefit from privilege. Such an approach might well be attractive to a court, faced with privilege issues where a large number of employees are within a group designated to communicate with the company’s lawyers, and the court suspects that this is a contrived device intended to attach privilege to as wide a variety of communications as possible. 69 The public interest concerns around courts having access to as much relevant material as possible upon which to base their judgments, and the effect of privilege, apply quite generally. 70 Whether the nature of the organisation asserting privilege brings some secondary public interest issue with it is yet to be seen. For example, this secondary issue is unlikely to apply to most companies or limited liability partnerships but may apply to quasi-public bodies such as Universities. The question is, does the nature of these organisations mean that privilege is more likely to be restricted through a narrow reading of who within the organisation represents the client? Three Rivers perhaps hints that this might be the case.
It seems clear that different persons within the company may meet the requirements of being the directing mind and will of a company for different purposes. 71 It may seem surprising that, in Three Rivers, the court stated that the Governor of the Bank of England was not the client. 72 However, the decision was fact specific ‘for the purposes of this application’, and the point arose in the context of a hypothetical question concerning the position had the Governor noted his recollection of relevant events and passed them to the BIU for transmission to the legal advisers. 73 The Court of Appeal had previously recognised that the Governor had appointed the BIU to deal with the legal advisers 74 ; presumably the Governor could, had he wished, have decided to perform that function himself. Indeed, where a board delegates authority to instruct a legal adviser, it seems that both the Board and the agent will be clients for the purposes of legal professional privilege. 75 Thus, it seems that the client may have multiple personalities, represented not only by the board but also by each person or group which the corporation authorised to instruct the lawyers for particular purposes. 76 In Menon and others v Herefordshire Council, 77 Lewis LJ held that communications between employees of a Local Authority and the Authority’s in-house legal team were covered by legal advice privilege because he accepted that all employees of the Local Authority were authorised to obtain legal advice from the in-house team in relation to the work they carried out as and when necessary. In AB v Ministry of Justice, 78 Baker J held that, in the absence of evidence of specific delegation to another entity, a head of department had implicit authority to seek legal advice from the department’s in-house lawyers, which was covered by legal advice privilege. In these cases, the legal advice sought did not concern potential criminal liability; however, the position would clearly be the same if it had.
Ho suggests that there are two tenable interpretations of the rationale underlying the decision in Three Rivers. Ho regards the Bank of England as the client and the BIU as either communicating with the lawyers ‘as the client (being employees who personified the BOE for purposes of privilege) or on behalf of the client (as employees authorised by the BOE to conduct privileged communications on its behalf with Freshfields’). 79 Given that the BIU was not merely acting as a conduit, communicating instructions from the Bank of England to its legal advisers, but was itself preparing those instructions, it is the former rather than the latter interpretation that appears to reflect the reality of the situation in Three Rivers and is consistent with the attribution principle. Loughrey suggests that the crucial point in determining whether communications between an employee of a company and a lawyer are privileged is the kind of authority the employee is exercising on behalf of the company. 80 This is important because, as Loughrey points out, ‘a company can be said to act or communicate through all of its employed agents for one purpose or another’. 81 What is the appropriate extent of delegated authority that is required in order to confer ‘client’ status? The crucial distinction is between an intermediate agent authorised to communicate instructions prepared by the client to the client’s legal advisers (who in our view would not form part of the directing mind and will of the corporate client for the purpose of preparing and communicating instructions) and an agent authorised by the client to prepare instructions to the lawyers on the client’s behalf (who would, via delegation of authority, form part of the directing mind and will of the corporate client for that purpose). 82 Although legal advice privilege can attach to communications between the client and the lawyer where they are made through ‘an intermediate agent of either’, 83 this does not apply where such an agent goes beyond being a channel of communication, for example, by bringing material into existence. 84 In contrast, where an agent is authorised to prepare and communicate instructions, it seems that communication of those instructions to the client’s legal advisers is privileged even if the legal advice is to be communicated back to the client itself and not to the agent. 85
An additional distinction is between authority of the two types mentioned immediately above and authority merely to provide information to the client or the client’s legal advisers or merely to conduct preparatory work in compiling such information. Authorising an employee to provide factual information/evidence to the company’s legal advisers or to the designated client within the corporation does not make that employee the client for the purposes of legal professional privilege. 86 Similarly, where the purpose of a committee or group set up by the company is merely to produce such material to be used by the client in instructing legal advisers, communications from it to the corporation or its legal advisers will not be covered by legal advice privilege. 87 In addition, documents of a factual nature, where the dominant purpose for preparation is not deemed to be that of obtaining legal advice, do not fall within the ambit of legal advice privilege. 88
A possibility is that in establishing a designated client group, a corporation might include employees whose activities in reality amounted to evidence gathering or evidence provision. This could potentially result in communications by the group not being covered by legal advice privilege. This could be the case whether or not such inclusion was a device to artificially extend the ambit of the privilege. The device of enlarging this group to absorb as many employees as possible may therefore partially fall at this hurdle, before a court has to consider any artifice in the designation of the client. An analysis of these sorts of activities in Three Rivers was not undertaken in relation to the BIU; however, it would seem that such an analysis would be relevant within the group designated by the corporation as the client. The wide nature of the group might make it more likely that a court will examine this point in more detail. Property Alliance Group Ltd v Royal Bank of Scotland plc 89 provides an example of an order for inspection being made by the court following uncertainty around the precise nature and role of an internal committee (the Executive Steering Group or ESG) and therefore whether a claim to legal advice privilege in relation to ESG documents was correctly made out. The ESG was set up within the Royal Bank of Scotland (RBS) in response to regulatory investigations into the rigging of LIBOR currencies. Following examination of the relevant documents, the court concluded that they formed ‘part of “a continuum of communication and meetings” between Clifford Chance and RBS, the object of which was the giving of legal advice as and when appropriate’. 90
Following Three Rivers, it seems that a company is free to decide, in any given situation, who within the company the client actually is for the purposes of legal advice privilege. The Model Articles for Private Companies Limited by Shares 91 allows the board of directors to delegate any of the powers conferred on them under the articles to ‘such person or committee…to such extent; in relation to such matters…as they think fit’. 92 In the Model Articles, the powers conferred on the directors constitute exercise of all the powers of the company, 93 which would include the power to instruct lawyers and receive their advice. Therefore, there is no requirement that this delegation be solely to a director or committee of directors—delegation can include employees who are not members of the board, either individually (such person) or as part of a committee. The board of directors could decide that a particular group within the company should have the delegated power to instruct lawyers and therefore be designated as personifying the directing mind and will of the client for this specific purpose, or could delegate to a specific officer or employee. The effect would be to make communications to and from the lawyer to members of that group or to that person subject to legal advice privilege. Alternatively, adopting a more restrictive approach to delegation, the board could choose to prepare instructions itself and merely delegate to an intermediate agent the authority to communicate those instructions to the legal advisers. 94 The mere fact that a committee charged with preparing information to be communicated to a corporation’s legal advisers forms an internal organ of the corporation does not in itself mean that the committee is the client for the purposes of legal advice privilege. 95 Whether the committee possesses delegated authority to instruct the legal advisers on behalf of the corporation will be a question to be resolved on the facts.
Membership of the Client Group
In general, in terms of day-to-day activities, it may well be that a company is content to leave instructing a lawyer to a director or directors, who prima facie will represent the directing mind and will of the company 96 and will constitute the client. However, in the context of particular matters, including where internal investigations into potential criminal liability are being undertaken, it may be useful for practical purposes to establish a delegated group within the company, perhaps beyond the directors, to instruct the lawyers and thus possess the appropriate degree of delegated authority to ensure that communications with the lawyer are covered by legal advice privilege. It may be that if the membership of this group is established at the outset in such a way as to ensure that communications between it and the legal adviser are privileged, subsequent changes in membership, if managed properly, will not result in a loss of privilege. In Three Rivers (No. 3), it was apparent that the composition of the BIU was increased by other Bank officials from time to time during the process of responding to the Bingham Inquiry to enable the BIU to ‘carry out speedily and efficiently the various research and analysis tasks that were required to furnish information and accurate instructions to the Bank’s legal advisers’. 97 Although employees of the Bank outside the BIU, including the Governor, were deemed analogous to third parties from outside the Bank, these changes in composition did not appear to have had any effect on the privileged status of communications between the BIU and the solicitors. It appears that privilege existed not in consequence of the specific offices held by these employees but because of their temporary incorporation into the committee to which the power to instruct had been delegated, that is, the BIU. It is suggested that the key issue is why the specific employee has been seconded into the committee. If the purpose is to prepare instructions, then privilege would appear to attach. If the true purpose is to provide the committee with factual information possessed by the employee and the secondment is contrived to provide a cloak of privilege to communications not otherwise privileged, the danger is that some communications within or by the committee might not be privileged. Arguably, it may be possible to include an external third party, such as an accountant, as a member of such a committee. However, similar concerns will arise. If the accountant is included to provide factual information or legal or accountancy advice, then privilege may not attach. 98
Practical Guidelines
Do current suggestions for corporate clients grappling with issues of who is likely to be designated as the client for the purposes of privilege work? For example, Fortnam and Lobo advise establishing ‘at the outset, before giving instructions and receiving advice, the nominated body/representative who is to deal with the external lawyers’. 99 They suggest that the retainer letter can be used to document who the nominated body or individual is, 100 a point echoed by Preston-Jones and Paterson. 101 Arguably, it may not be sufficient merely to identify a person or body as having authority to communicate with the legal advisers, as this might merely authorise communication of instructions as an intermediate agent for communication or gathering of evidence. What would seem to be desirable is identification of the person or group that is specifically authorised to create and communicate instructions to the legal adviser. In addition, is the retainer letter enough to guarantee that the nominated body is regarded as the client for privilege purposes? Should this be supplemented by, for example, an appropriate board minute or partners meeting minute, to record the delegation of authority to the relevant body and, perhaps crucially, the ambit of that delegated authority? It would seem that the safest approach in terms of delegated authority is to ensure that the client group has the authority to create information and communicate with the lawyers, including giving them instructions. It does not appear to be necessary to give only this group authority to make decisions following receipt of legal advice. Legal advice received by the client group can be disseminated within the company (e.g. to the board of directors) without losing its privileged status. 102 Attention must also be paid to the actions of members of the client group. Where a group is accepted as the client, it seems less likely that the actions of members of the group will be minutely scrutinised, but it is not impossible that this may happen. Care should be taken to ensure that actions taken by members of the group, such as the preparation of factual records, are taken for the purpose of obtaining legal advice. Moreover, it may be that the retainer letter is itself created by the designated client if that person or body falls in a category of persons or bodies who have been authorised by the company to instruct legal advisers, whether in-house or external, as and when required in appropriate circumstances relating to that person’s portfolio within the organisation. In such circumstances, it will presumably be other evidence, such as a board minute, rather than the retainer letter itself that will be crucial to evidencing authority to act as the client.
So far as litigation privilege is concerned, where documents are prepared for the purposes of contemplated litigation, they should be marked as such. 103 While the designation will not be determinative, because the court is not required to accept the evidence of a party as to its intention at the relevant time, the test applied is an objective test and the evidence the court will consider in deciding whether the privilege arose includes evidence of the parties’ intentions at the relevant time. 104 In addition, marking documents in this way seems particularly desirable where automated software is used to scan large numbers of documents for legally privileged content. 105
Conclusion
The policy basis of litigation privilege is found in the concept of adversarial trial, that is, the right of a party not to disclose to an adversary either the party’s brief or materials that came into existence for the brief. The ambit of the privilege is subject to three major constraints, restricting its operation to adversarial litigation, to circumstances in which litigation has commenced or is in reasonable contemplation and to communications made for the purposes of litigation (dominant rather than sole purpose sufficing). While it might be argued that changes in the nature of criminal proceedings justify the adoption of a ‘policy of confinement’ in relation to the application of litigation privilege, it is suggested that criminal proceedings remain primarily adversarial and that the policy rationale underlying the existence of the privilege remains valid in their context (though it is accepted that existence of contemplation of a criminal or internal investigation does not, in itself, justify the application of the privilege). There is first instance authority for the proposition that litigation privilege does not attach to communications for the purpose of avoiding criminal litigation but it is suggested that the better view is that such communications (unlike communications for the purpose of avoiding a criminal investigation) do fall within the ambit of litigation privilege. Similarly, it is suggested that litigation privilege should potentially attach to communications for the purpose of considering the desirability of following a DPA or to communications made for the purpose of enhancing the likelihood of obtaining a DPA. The fact that a criminal or internal investigation is contemplated or in progress does not in itself mean that criminal litigation is in reasonable contemplation. A recent first instance civil authority, followed without citation of other authorities by the Criminal Division of the Court of Appeal, suggests that criminal litigation may only be in reasonable contemplation if the client is aware of the circumstances that make a prosecution likely—that is, where there is at least knowledge of material supporting the truth of the allegations. It is suggested that the better view is that the questions that the court has to determine are fact specific and that even where a client believes on the information available to it that it is innocent, this does not mean that the client cannot reasonably contemplate prosecution if the relevant prosecuting authority tends to prosecute in circumstances of the type that the client finds itself in.
The approach taken by the courts when considering claims for legal advice privilege in the corporate context involves identifying, within the company, a client, either an individual or a client group. The consequence of this approach is that only communications between this client and the company’s lawyers will be protected by legal advice privilege and then only if the communications from the client are of a particular type, that is, those communications that relate to instructing the lawyers and creating information relevant to those instructions.
When identifying a client within the company, attribution theory enables the company to create a directing mind and will for the purpose of communicating with the company’s lawyers and with delegated authority to do so. In this context, if a designated group within the company only has authority to act as a conduit for instructions from, for example, the board of directors to the lawyers, then this group is not the client—it does not have the necessary characteristics for directing mind and will to be so. If the group has the authority to prepare instructions for the lawyers on behalf of the company, then the group has the necessary characteristics to be the directing mind and will of the company and is therefore the client, for this specific purpose. It seems also to be the case that there can be more than one client within the company for legal advice privilege purposes, so, for example, both the group with delegated authority to instruct the lawyers and the board could constitute the client. Similarly, various employees within a company might all have the authority to instruct the company’s in-house or external lawyers for purposes relevant to their respective portfolios. In terms of the evidence needed to establish both the nature of delegated authority and to whom it has been delegated, it seems that an appropriately drafted letter of retainer and a board minute giving the group the authority to create information and communicate with the lawyers, including giving them instructions, are the optimal approach.
Although it may seem attractive to designate a very wide group of people within the company as the client, with authority to communicate with the company’s lawyers, there do seem to be dangers with this approach. This device may encourage the court to look much more closely into the activities of members of the group, particularly if artifice is suspected. If a member of the group is essentially simply providing factual information possessed by him or her, and inclusion in the client group is contrived to try and provide a cloak of privilege to communications not otherwise privileged, the risk is that some communications from the client group to the lawyers may not be covered by privilege. Even where artifice is not suspected, the point remains that the wider the client group, the more likely it is that some members of it are not in fact engaged in communications of the sort that would ordinarily be covered by privilege. Rather than creating a very wide client group at the outset, a better approach may be to second employees to the group as required, something that clearly took place in Three Rivers, subject to the caveat regarding artifice.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.