Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
Back to all news

Lucy Bone on privilege in internal investigations

This article was written for the Practical Law Employment Blog by Lucy Bone and has been reproduced with the permission of the publishers. For further information visit or call 020 7542 6664.

Internal investigations are increasingly being conducted by companies not only on regulatory grounds but also in response to employment issues such as whistleblowing and discrimination allegations. In SFO v ENRC [2018] EWCA Civ 2006, the Court of Appeal has significantly widened the scope of legal professional privilege in the context of an internal company investigation. It will now be easier for the employer to assert privilege over employees’ witness statements and other documents generated in an investigation.

Privilege: legal background

The principles of legal privilege were addressed in Three Rivers DC v Bank of England (No 6) [2004] UKHL 48. In Three Rivers v Bank of England (No 5) EWCA Civ 474, the Court of Appeal considered the scope of legal advice privilege and decided that only communications between client and lawyer were privileged. In the corporate context, only employees tasked with seeking and receiving such advice could be considered to be the “client”; communications between lawyers and other employees would not attract privilege.

This has had two practical consequences of particular importance to employment advisers:

  • First it has given rise to the “client group” model often used in investigations, whereby a small team is set up exclusively to instruct, communicate and deal with the lawyers.
  • Secondly it has meant that investigative interviews with employees outside the group would not attract privilege.

The House of Lords held in Three Rivers (No 6) that litigation privilege would only protect communications between parties or their lawyers and third parties where:

  • The litigation was proceeding or in contemplation.
  • The communications were for the sole or dominant purpose of that litigation.
  • The litigation is adversarial, not investigative or inquisitorial.

SFO v ENRC: the facts

In December 2010, ENRC received from a whistleblower allegations of corruption and fraud within a wholly owned subsidiary in Kazakhstan. By March 2011, ENRC’s general counsel formed the view that it was “on the SFO’s radar” and should upgrade its dawn raid procedures. ENRC’s audit committee instructed solicitors to investigate the allegations, and in April 2011, the solicitors warned that adversarial proceedings could occur from the internal investigation. In May 2011, ENRC engaged forensic accountants to undertake a books and records review to assess its exposure to liability under bribery legislation.

In August 2011, the Serious Fraud Office (SFO) contacted ENRC in relation to the whistleblower’s allegations which had by this time been reported in the media. The SFO referred to its self-reporting guidelines, but noted that it was not at that stage carrying out a criminal investigation. Various meetings took place, at which the SFO gave no assurance that it would not prosecute. ENRC in turn gave assurances that it was investigating, but eventually the SFO lost patience and began proceedings in 2013.

During the solicitors’ investigations, statements were compiled from the evidence given by employees and ex-employees. Documents were generated by the review of books and records, and reports prepared by the forensic accountants. ENRC sought to resist disclosure of these documents, including witness statements, on the ground of privilege.

Judgment of Andrews J at first instance

Andrews J found that none of the documents were covered by litigation privilege, as at the time they were created, litigation between ENRC and the SFO was a mere possibility rather than a likelihood. Further, the judge held that the documents had not been created for the dominant purpose of litigation, but rather to avoid it.

The Court of Appeal

In the unanimous judgment of the court (Sir Brian Leveson, Sir Geoffrey Vos and McCombe LJ), the appeal was allowed and the claim to litigation privilege was upheld.

Litigation privilege

The Court of Appeal analysed the requirement that proceedings be in reasonable contemplation by interrogating the facts in some detail, noting the dates set out above in particular. The court considered that “the whole sub-text of the relationship between ENRC and the SFO was the possibility, if not the likelihood, of prosecution if the self-reporting process did not result in a civil settlement.” (Paragraph 93.).

While noting that it was not conclusive, the court said that it was wrong of Andrews J to have disregarded the view taken by ENRC’s solicitors and their “oft-repeated advice” that there was a serious risk of prosecution and/or regulatory intervention (paragraphs 93 and 95). When the SFO specifically makes clear to the company the prospect of its criminal prosecution and legal advisers are engaged to deal with that, the court considered there is a clear ground for contending that criminal prosecution is in reasonable contemplation (paragraph 96).

The fact that a party anticipating possible prosecution will need to make further investigations before it can say that proceedings are likely does not in itself prevent proceedings being in reasonable contemplation (paragraph 98). The court was sensitive to the realities faced by corporate bodies, in particular large ones: “An individual suspected of a crime will, of course, know whether he has committed it. An international corporation will be in a different position, but the fact that there is uncertainty does not mean that, in colloquial terms, the writing may not be clearly written on the wall.”

For these reasons, the court found that Andrews J was wrong to suggest a general principle that litigation privilege cannot attach until either a defendant knows the full details of what is likely to be unearthed or a decision to prosecute has been taken. The fact that a formal investigation has not commenced will be one part of the factual matrix but will not necessarily be determinative (paragraph 100).

The court accordingly found that at the time the documents were created, proceedings were reasonably in contemplation in April 2011 and certainly by August 2011. The court then turned to consider whether the documents were created for the dominant purpose of litigation.

The court reminded itself that the exercise of determining dominant purpose is one of fact and the court must take a realistic and commercial view of the facts. Moreover, there may be a duality of purpose. The key question in the ENRC appeal was whether it would be reasonable to regard ENRC’s dominant purpose as being to investigate the facts to see what had happened and then deal with compliance and governance, or to defend the proceedings in contemplation (paragraph 104).

Noting that regulatory and compliance regimes provide a “stick” to enforce legal standards, the court thought that an investigation for “compliance” and “governance” purposes could, where there is a clear threat of criminal investigation, be “brought into the zone where the dominant purpose may be to prevent or deal with litigation.” (Paragraph 109.). The need to investigate allegations of fraud and corruption was a subset of the dominant purpose of defending contemplated proceedings.

The court said that it was in the public interest that companies should be prepared to investigate allegations from whistleblowers or journalists, prior to going to a prosecutor such as the SFO, without losing the benefit of privilege for the work product and the consequences of their investigation (paragraph 116). Otherwise the temptation may be to not investigate at all.

The fact that the solicitors prepare a document with the ultimate intention of showing it to the other side does not deprive the preparatory legal work of litigation privilege. Thus iterations of a response to a claim or discussions of a letter would nevertheless attract privilege (paragraph 102). Even if litigation was not the dominant purpose of the investigation at its very inception, the court accepted as sufficient that the evidence showed that it swiftly became the dominant purpose (paragraph 111).

Not only were proceedings in reasonable contemplation, but the first instance judge ought to have held that the documents were brought into existence for the dominant purpose of resisting or avoiding those or some other proceedings (emphasis added) (paragraph 113).

Legal advice privilege

The court held that legal advice privilege was not the central issue in the case but still addressed the “forceful” arguments made, including by the Law Society as intervener. It began its discussion of this issue by identifying what Three Rivers (No 5) had actually decided: that communications between an employee of a company and its lawyers could not attract legal advice privilege unless that employee was tasked with seeking and receiving such advice on behalf of the client (paragraph 123).

It noted that the 19th century authorities were decided when the concept of privilege was in its infancy, and further that most of these authorities concerned small businesses or individual litigants. In a modern multinational corporation, the information on which legal advice is sought is unlikely to be in the hands of the main board or the client group. If the multinational corporation cannot claim privilege over communications between its lawyers and employees outside the client group, it will be disadvantaged as compared to smaller companies or individuals. The court considered that this anomaly could not be justified and that basis of privilege should be applicable to all clients, whatever their size or reach (paragraph 127). Moreover, English law was out of step with other common law jurisdictions, which was undesirable given the international context in which business operates (paragraph 129).

The court was clear that, it if had been open to it to depart from the House of Lords decision in Three Rivers (No 6), it would have done. Although it remains good law until overturned by the Supreme Court, a clear signal has been given of the judicial receptiveness to challenges to the narrow definition of “client” for the purposes of legal advice privilege.

The SFO has now confirmed that it will not be appealing to the Supreme Court. Accordingly Three Rivers (No 6) remains the authoritative statement of the law on legal advice privilege.


While the test of litigation privilege has been significantly widened by the Court of Appeal’s decision, any employer hoping to assert privilege will need to consider their position carefully:

  • Each case turns on its own facts. The court will judge claims to privilege in the light of the facts as a whole. Companies must therefore take bespoke advice and regularly update the advice as matters progress.
  • The views formed by the company’s legal advisers should be taken into account by the court, though they will not be conclusive.
  • It is optimal to consider privilege at the outset of an investigation, and then maintain a clear documentary trail to assist in supporting assertions of privilege.
  • The fact that the company needs to conduct further enquiries before it can say with certainty that litigation will follow does not prevent proceedings being in reasonable contemplation.
  • It may be arguable that privilege applies to documents even though the litigation was not in contemplation until a short time after their creation.
  • In relation to the dominant purpose test, privilege will extend to documents for the purpose of avoiding proceedings as well as actively litigating those proceedings.
  • Privilege may exist over documents that are intended to be shown to the other side including for the purpose of settlement.
Related Members
Lucy Bone
Shortlist Updated