The State of Qatar v Banque Havilland

David Mumford QC and Thomas Munby (with Hugo Leith of Brick Court) act for the State of Qatar in proceedings in the Commercial Court against Banque Havilland SA (the “Bank”). In an important decision on 30 July 2021, David Edwards QC (sitting as a Judge of the High Court) made a number of orders for further disclosure against the Bank, including, significantly: (a) an order that the Bank give disclosure of an investigation report and associated documents prepared for it by PwC (the “PwC Report”); (b) unusually, an order that the Bank’s solicitors should re-review on a “Model E” basis certain categories of disclosure which had previously been given; and (c) an order for further steps to be taken to identify “known adverse documents”.

The background

Qatar’s claims arise out of the “Blockade” by a number of Arab nations which commenced in June 2017. Qatar alleges that the Bank – which is owned by the family of the well-known financier David Rowland – participated in a conspiracy with one of its employees and certain other banks to further the aims of the Blockade and harm its economic interests, by (inter alia) manipulating the market in Qatari-issued currency and bonds. In 2017 the Bank’s employee prepared a presentation (the “Presentation”), which described elements of a plan for such manipulative trading, details of which were then revealed by the press after an apparent leak in November 2017.

The litigation privilege claim

The leak of the Presentation prompted the Bank to notify is regulators in Luxembourg and London, and to instruct PwC to conduct a forensic investigation into how the Presentation had come to be created, whether it had been disseminated and how it come into the hands of the press. The resulting PwC Report was issued in June 2018 and provided to the Bank’s Luxembourg regulators.

The Bank maintained that the PwC Report was subject to litigation privilege, it having been prepared at a time when adversarial proceedings by emanations of Qatar and/or the Bank’s regulators, arising out of what had been publicly revealed about the Presentation, were reasonably in contemplation; and it having been prepared (it was said) for the sole purpose of collecting evidence and enabling legal advice to be given in connection with those proceedings. The Bank further maintained that such privilege was not lost upon the provision of the PwC Report to its regulator, the regulator being subject to Luxembourg professional secrecy restrictions and the report having been provided on the basis of a “limited waiver” only.

One important feature of the factual background was that the Bank received a letter from lawyers acting for Qatar in December 2017 requesting that the Bank put in place a “litigation hold” on its documents. This was after the Bank had commissioned the PwC Report (in November 2017) but before it had been issued (in June 2018). The Bank argued that this helped demonstrate not just that litigation was reasonably in contemplation, but also that the report was prepared for the purpose of that litigation, since such matters had to be judged at the time when the supposedly privileged document was created, not when the instruction to create it was given. The Judge accepted Qatar’s submission in response that – unless there was evidence that the Bank’s purpose had changed over time (which there was not) – the most important point in time with reference to which to judge the Bank’s purpose was when the instruction was given.

Contemplation of adversarial proceedings

Reviewing the contemporaneous evidence and a witness statement from the responsible executive at the Bank, the Judge concluded that – although the Bank clearly believed that the Presentation “could have significant legal, regulatory and legal consequences” – this was too general an apprehension to support a claim for litigation privilege: something “more concrete” was needed. So far as concerned possible adversarial proceedings by the Luxembourg regulator, whilst an investigation could shade into an adversarial proceeding, on the facts the regulator’s involvement had not moved beyond the investigative stage prior to the PwC Report being issued. Matters with the Bank’s London regulators had not progressed beyond initial information gathering; and, despite the litigation hold letter from Qatar’s lawyers in December 2017, there was no evidence that the Bank in fact anticipated a claim by it at that stage.

Litigation purpose

In any event, the Judge was not persuaded that the dominant (let alone, as claimed, sole) purpose of the PwC Report was anticipated litigation: it was also prepared (a) to find out the facts, including how the Presentation had been leaked; and (b) to satisfy the Luxembourg regulator that the Bank was acting appropriately, and enable it to answer its questions. Importantly, even though interactions with the regulator might have become adversarial at some point, satisfying the regulator was not (in the circumstances) a litigation purpose. The Judge also gave limited weight to the fact that PwC had been retained not by the Bank directly, but by its local lawyers: the evidence showed that the Bank had decided to engage PwC before lawyers were retained, and channelling the instruction through the lawyers appeared to have been “with the aim of improving the prospect of a successful claim for privilege”. Finally, the receipt of the litigation hold letter from Qatar’s lawyers did not bring about a change in the dominant purpose of the PwC investigation.

The PwC Report (and associated documents) was therefore not subject to litigation privilege. It was therefore unnecessary for the Judge to resolve the question of whether the provision of the report to the Luxembourg regulator involved a waiver of privilege; or (as Qatar had submitted) whether a document which had always been intended to be provided to a potential adversary and in fact provided to it could not, by its nature, be litigation privileged.

The order for a “Model E” re-review of previous disclosure

Disclosure under “Model E” (i.e. of documents that may lead to a “chain of enquiry”) is reserved for exceptional cases. The Judge found that this was such a case, so far as concerned the transcripts of telephone conversations between certain custodians (including David Rowland and his son Edmund) within a particular date range. He directed a re-review of those transcripts on a Model E basis (disclosure having originally been ordered on a Model D basis), for four reasons: (a) a transcript of one conversation which the Judge accepted was plainly relevant on a Model D basis had been missed during the original review; (b) the transcripts that had been disclosed contained references to communications about relevant matters having been taken “off line”; (c) it was a case of covert conspiracy and the phone recordings, containing unguarded comments by key alleged protagonists, might lead to the identification of other documents concerning the genesis and dissemination of the Presentation; and (d) other potential sources of information were not available (notebooks having been lost, a phone wiped and certain emails deleted). The Judge was therefore satisfied that a fuller review might well lead to a train of enquiry which would result in the disclosure of other relevant documents.

Known adverse documents

“Known adverse documents” are those which a party is actually aware (without undertaking any further search than already undertaken) both (a) are or were previously within its control and (b) are adverse (paragraph 2.8 of the Disclosure Pilot). In the case of a company or organisation, paragraph 2.9 of the Disclosure Pilot stipulates that awareness for these purposes is that of any person with accountability or responsibility for the events or the circumstances which are the subject of the case, or for the conduct of the proceedings. It is generally understood that these provisions require reasonable and proportionate checks to be undertaken with persons with such accountability or responsibility to ascertain whether they are aware of adverse documents.

In the present case the Bank only made such checks of certain employees and former employees; for other custodians who had accountability or responsibility for the relevant events, it obtained documents under the Bank’s control and contacted them to ask if they used personal email accounts or devices for work purposes. The Bank argued that this was sufficient to check for known adverse documents; but the Judge disagreed: checking whether an individual had access to particular sources of potentially relevant documents is not the same as checking whether they are aware of adverse documents (to state the obvious, the answer to the question whether an individual used a personal email account for work purposes could be “no”, even if the individual was aware of adverse documents). Further, the Judge refused to accept that there was any difficulty in sufficiently educating custodians about the issues in the case for them to be able to answer questions as to their awareness of adverse documents. The Bank was therefore directed to make further enquiries.

The take-away from this is that the requirement to check with relevant individuals for known adverse documents is an important safety net which operates additionally to, and is not satisfied by, the usual steps to collect and search potentially relevant sources of documents.

To view the full judgment, please click here.