JSC COMMERCIAL BANK PRIVATBANK v IGOR VALERYEVICH KOLOMOISKY & Ors

This judgment is concerned with the claimant’s application for further orders in relation to the first defendant’s disclosure of 26 chains of WhatsApp messages. The WhatsApp chats were collected from images of the first defendant’s mobile telephones. They have been disclosed in heavily redacted form. The claimant submitted that the redactions were unjustified and sought an order that 17 of them be disclosed to its solicitors unredacted.

Mr Tim Akkouh QC, who argued this application for the claimant, submitted that if I was not satisfied that all the redactions must have been unjustified, I could either inspect the unredacted documents myself or I could make what he called a Hollander order, named after a suggested form of confidentiality club order, described in Hollander on Documentary Evidence (14^th edn, 2021 at para 10-15). This would require disclosure of the unredacted messages to named members of the claimant’s team at Hogan Lovells and counsel. There could then be either an attempt to agree the way forward based on the claimant’s lawyers having had sight of the relevant material or a return to court for further argument on a more informed basis.

The original disclosure orders with which it was said that the first defendant has not complied were made by Mann J on 26 June 2020 in accordance with CPR PD 51U. He ordered extended disclosure and set out in the attached disclosure review document the Issues for Disclosure and the models to be applied.

The Issues for Disclosure included issues relating to (a) the control exercised by the first defendant and the second defendant over the claimant; (b) the first and second defendants’ ownership and control of other entities and assets relevant to the claimant’s claim: the borrowers, the intermediary borrowers, the new borrowers, the third to eighth defendants and those alleged to be principals of the third to eighth defendants; and (c) the first and second defendants’ knowledge and involvement in the transactions relating to the claimant’s claim. One of the sub-issues is whether the employees of the claimant remained loyal to the first and second defendants in relation to the conduct of the claimant’s affairs after the time it was nationalised in 2016.

The disclosure review document appended to Mann J’s order obliged the parties to give model E disclosure in relation to issues concerning ownership and control of the other entities I have described in category (b) above. Model D disclosure was ordered in relation to other Issues for Disclosure and sub-issues falling within the three categories I have identified above. In accordance with the description of model D disclosure in para 8.3 of CPR PD 51U, Mann J ordered that any party giving model D disclosure was required to search for and disclose narrative documents, which means that documents must be disclosed if they are relevant only to the background or context of material facts or events and not directly relevant to the Issues for Disclosure themselves (App 1 to PD 51U).

The court’s jurisdiction to grant the relief sought by the claimant arises under para 17 of CPR PD 51U. This provides that, where there has been or may have been a failure adequately to comply with an order for extended disclosure, the court may make such further orders as may be appropriate, including an order requiring a party to undertake further steps, such as further or more extended searches to ensure compliance with an order for extended disclosure, and orders to produce documents or make a witness statement explaining any matter relating to disclosure. The applicant must satisfy the court that making an order is reasonable and proportionate having regard to the factors referred to in para 6.4 of CPR PD 51U.

Redaction of disclosed documents is dealt with by para 16 of CPR PD 51U as follows:

It was common ground at the hearing that while, for the purposes of carrying out the model D and model E search-based extended disclosure in accordance with para 8 of CPR PD 51U, the exercise is to be done by reference to the Issues for Disclosure, a para 16.1(1) redaction is only permissible if the redacted data is irrelevant to any issue in the proceedings. Once a document has been identified for disclosure by application of model D or model E, the question of whether parts of it can be excluded from inspection raises different issues, not least because the need for proportionality in the conduct of the search will not continue to be a relevant factor, anyway to the same extent. Why therefore should the positive step of redacting data relating to an issue in the proceedings be appropriate merely because that data did not also relate to an identified Issue for Disclosure?

As Snowden J explained in WH Holding Ltd v E20 Stadium LLP [2018] EWHC 2578 (Ch), at [37], it is well known that the court will normally be satisfied by a statement from a solicitor with responsibility for the disclosure process that the redaction in question has been properly made. However, where there has been heavy redaction of many documents, the court is justified in adopting greater vigilance to ensure that the right to redact is not being abused or too liberally interpreted, recognising all the while that the burden is on the applicant to make out a case for inspection.

Where the court has doubts as to whether the redactions have been properly made, there are several solutions that have been adopted. One is to order that the redactions be reconsidered in light of the court’s ruling with or without a more specific statement as to why a redaction has been made. A second is that the documents be provided confidentially to the inspecting parties’ lawyers – i.e., the Hollander order described above – a solution adopted by Cockerill J in Recovery Partners GP Limited v Rukhadze [2021] EWHC 1621 (Comm), [65]-[67]. A third is the approach adopted by Snowden J in WH Holding, where he decided (at [36]) to inspect the documents himself. The court will often be reluctant to take that course, although it is sometimes the only practical way forward: see Snowden J’s discussion of Bank Austria AG v Price Waterhouse [1997] CLY 464 and Atos Consulting v Avis plc [2007] EWHC 323 (TCC) at paras [33] to [35] and [40] of his judgment in WH Holding.

The question of how the para 16.2 explanation of the entitlement to redact is to be given was considered by Butcher J in Eurasian Natural Resources Corpn Ltd v Dechert LLP [2020] EWHC 1002 (Comm). He said in a passage at para [91] with which I agree: “I consider that what is ordinarily required under para 16.2 is a list of documents which have been redacted which identifies for each the reason for the redaction, namely whether it is irrelevance and confidentiality or privilege”. He then made suggestions as to how that could conveniently be done and went on to stress at para [92] that depending on the case it may also be desirable for an additional clear explanation of the claim of entitlement to redact to be given, particularly where the basis for redaction is not apparent.

This approach picked up what Sir Geoffrey Vos C had said in UTB LLC v Sheffield United [2019] EWHC 914 at para [83] in the context of a para 14 claim to privilege. It also seems to me to be consistent with what Chief Master Marsh said in Astra Asset Management UK Limited v MUSST Investments LLP [2020] EWHC 1871 (Ch) at para [20] when explaining that a generic description which simply says ‘irrelevant and confidential’ may prove to be insufficient:

A similar approach had been adopted in WH Holding, which was decided before CPR PD 51U had come into force. A schedule explaining the reasons for the redactions in more detail had been directed, even before Snowden J carried out the inspection that he did (see paras [19] to [25] of his judgment). It will sometimes be the case that the exercise of explaining why each separate redaction has been made will help the solicitor for the disclosing party to focus more clearly on the possible breadth of the issues in the proceedings, whether or not they are also Issues for Disclosure.

The claimant submitted, and I accept, that I should approach the present application against the background that only a small proportion of the documents disclosed by the first defendant come from what it calls his own sources. The reason for this is that he does not use a personal email account whether for personal or business purposes, he does not use a desktop computer to store or create electronic documents and the only forms of social media account he has are WhatsApp and Viber. This means that his WhatsApp messages have the potential to be a particularly important and valuable source of relevant information.

When the WhatsApp chats were first disclosed in June 2021, they comprised 350 pages of chats with a total of c.6,209 messages, all but 272 of which had been redacted or partially redacted. In a number of instances, it was not possible to identify the counterparty to the chat. Fieldfisher confirmed that the redactions had been applied on the ground that the redacted data is irrelevant to any issue in the proceedings and confidential. Each redaction was therefore said to fall within para 16.1(1) of CPR PD 51U. There was no specific explanation of why particular chains of WhatsApp messages had been redacted, but from the description in their covering letter the explanation which appeared to apply most naturally to the material redacted from all of the messages was “information about unrelated commercial transactions and other commercial information unrelated to the issues in these proceedings.”

Over the course of the succeeding months, there was correspondence between Fieldfisher and Hogan Lovells in which Hogan Lovells pressed Fieldfisher for a fuller explanation of the reasons behind the redactions and sought to challenge the extent of the redactions that had been made.

This correspondence eventually culminated in Fieldfisher’s letter of 8 March 2022, under cover of which they disclosed further information, both in relation to the identity of the counterparties to the WhatsApp messages and by increasing the total number of unredacted messages from 272 to 422. In their covering letter, Fieldfisher described these as a small number of messages in respect of which they accepted that it was “at least arguable that they may be relevant to the issues for disclosure”. However, they said that they considered them unlikely to be of any particular significance to the issues in dispute in the proceedings. Fieldfisher also said in their letter of 8 March 2022 that they remained “satisfied that the majority of messages that were redacted are both irrelevant and confidential, and (insofar as applicable) do not provide relevant background or context to the communications which have been disclosed”.

Fieldfisher’s statements (a) that they were only arguably relevant to the issues for disclosure and (b) that they were unlikely to be of any particular significance to the issues in dispute in the proceedings appears to me to illustrate that they had adopted an approach to relevance which was too narrow. In my view there are two reasons for this.

The first is that the way in which Fieldfisher expressed themselves in their letter of 8 March indicates that they may well have misunderstood the distinction between identified Issues for Disclosure and “any issue in the proceedings”. When explaining the re-review exercise, they appear to have said that they did so by reference to the Issues for Disclosure. If, as seems likely from what they said, that was the approach they adopted, I think that they were wrong to do so. As I have already explained the Issues for Disclosure relate to the need to disclose a document in the first place, not to the redaction of part of it.

The second is that Fieldfisher’s characterisation of some of the chats as being unlikely to be of any particular significance to the issues in dispute in the proceedings seems to me to betray an approach to relevance that may have been wrong. This can be illustrated by consideration of some of the examples on which Mr Akkouh made submissions. In relation to these chats, Mr Akkouh made clear that the claimant was not alleging that Fieldfisher were in any way untrustworthy. The claimant’s criticism was simply that Fieldfisher had obviously taken an unduly narrow view of the scope of what was capable of being relevant to any issue in the proceedings for the purposes of para 16 of CPR PD 51U. Although I will not go through the ten example chats one by one, it is informative to focus on some of what has now been unredacted.

The first example is a chain of messages between the first defendant and the second defendant. Most of this chain has been redacted because, like the other redacted material, it was said to be about unrelated commercial transactions and other commercial information and it was said that that information was therefore unrelated to the issues in these proceedings.

The claimant disputed this and submitted that how the first and second defendants dealt with each other in relation to their jointly controlled assets, the character and closeness of the business relationship between them and the way in which (and extent to which) they operated together is relevant to an issue in the proceedings. In particular, such matters were relevant to the two control issues I described at the beginning of this judgment each of which is plainly an Issue for Disclosure.

Those matters were also said to be relevant to the claimant’s more specific control allegations, not just that the first and second defendants were close business associates (an issue which is not in dispute in the proceedings) but also that, where control was exercised by one of them with the express, implied or standing approval of the other, that control was exercised on behalf of both of them. The question of standing approval and its impact as between the first and second defendant is an important part of the claimant’s case and is in dispute.

Mr Akkouh submitted that the nature of the continuing business relationship between the first defendant and the second defendant, even some time after the date of nationalisation, may well cast light on the nature and extent of their pre-nationalisation acts of control over the claimant. The closeness of the relationship between the first defendant and the second defendant is to that extent and in that context relevant to an issue in the proceedings.

As the description as to why the redactions have been carried out refers only to commercial transactions and commercial information, I agree that these chats must have been electronic discussions between the first and second defendants on commercial issues. I also agree that there is a strong probability that they will therefore have a direct bearing on the ability of the claimant and the court to obtain a proper understanding of the true nature of the business relationship between them. The mere fact that the commercial transactions to which the exchange of messages related were not transactions with which these proceedings are directly concerned, and took place sometime later, does not in my judgment mean that the messages will necessarily be irrelevant to any issue in the proceedings within the meaning of para 16.1(1).

The second example is the chat between the first defendant and Mr Dubilet, who was the claimant’s CEO prior to nationalisation and seems from unredacted material to have been able to obtain information for the first defendant’s benefit from the claimant’s employees post-nationalisation. It seems improbable that this would have happened if Mr Dubilet was not accustomed to do what the first defendant wanted pre-nationalisation. Acting in accordance with the first defendant’s post-nationalisation instructions or wishes, whatever the context is or may be, is the kind of response which might be said to be at least consistent with the first defendant’s control of the claimant pre-nationalisation.

The claimant drew attention to a number of instances from the unredacted messages which indicate a close continuing relationship post-nationalisation. It was submitted that this itself was capable of reflecting a pre-nationalisation relationship of control by the first defendant over an entity of which Mr Dubilet was the CEO (i.e., the claimant) and submitted that it was most improbable that there were not more examples in the WhatsApp messages.

I think that the claimant is right to contend that documents reflecting or constituting communications between the first defendant and the claimant’s pre-nationalisation CEO are likely to be relevant to the question of his control of the claimant pre-nationalisation even where they took place after nationalisation had occurred. The nature and very fact of their communications post-nationalisation will tend to inform an assessment of the relationship between Mr Dubilet and the first defendant pre-nationalisation, and the extent to which the first defendant did in fact exercise control over the claimant through his relationship with Mr Dubilet as its CEO.

There are several other examples of WhatsApp chats between the first defendant and other individuals who played important roles in relation to the claimant’s affairs prior to nationalisation. The claimant contended that the nature of their respective relationships with the first defendant, the way he and they continue to correspond with each other post-nationalisation, and the subject matter of that correspondence will be relevant at least as narrative background, and possibly more directly than that, to the extent of the first defendant’s control of the claimant.

The first of these is the post-nationalisation WhatsApp chats between the first defendant and Mr Oleksandr Granovsky, a businessman who seems to have been corresponding post-nationalisation with the first defendant about the affairs of the corporate defendants, and in particular by disclosing what appears to have been confidential information relating to legal proceedings brought against them by the claimant. Some, but not all of those chats have now been disclosed in unredacted form, but in the light of the nature of the confidential information which has been disclosed, I agree with the claimant that the relationship more generally may well relate to the nature and extent of the first defendant’s control of the claimant. It seems to me that these messages may not have been reviewed with that possibility in mind.

The second is the chats between the first defendant and Ms Svetlana Melnikova. She was an employee of the claimant and the two chains of messages with her have now been fully unredacted. The claimant contends that the fact that Fieldfisher described them in their 8 March letter as messages which were unlikely to be of any particular significance to the issues in dispute reflects an unduly narrow approach to relevance. I agree. They do at the very least bear on the ownership or control of PrimeCap which is an issue in the proceedings and is one which, as Fieldfisher will have known, is said by the claimant to be an important one.

The third is the chats between the first defendant and Mr Timur Novikov, of which only one out of 259 messages is unredacted. It is said that, because Mr Novikov had several important roles as head of the claimant’s investment department and the “boss” of PrimeCap, it is inherently improbable that more of the chats are not disclosable, taking the proper approach to relevance which I agree should be taken. I too find it surprising that, given the known relationship between the first defendant and Mr Novikov, so many of these chats are said not to relate to any issue in the proceedings.

In relation to a number of other examples, the claimant’s submissions focussed on the improbability of WhatsApp chats not being relevant to questions of control of the entities I identified at the beginning of this judgment, once it is appreciated that relevance must be approached in a rather broader manner than the manner that appears to have been adopted to date. Further, I think that the issue of continuing employee loyalty (identified as a sub-issue) is a good illustration of why it is that post-nationalisation exchanges are likely to throw light on and be relevant to pre-nationalisation control. The way in which the first defendant’s relationship with those employed by or associated with the relevant entities post-nationalisation may have reflected their loyalty is likely to be manifested in their messages and chats. That manifestation may be relatively subtle, but it will be no less relevant for that. I agree that as time goes by, the relevance of this sort of communication may diminish, but as a matter of principle, it seems to me that the manner of communication and the topics on which they chose to communicate will relate to an issue in the proceedings. I am not satisfied that the extent to which this is the case has been fully appreciated by Fieldfisher.

In light of these matters, I think that there has been or may have been a failure adequately to comply with an order for extended disclosure because of the approach that has been adopted by the first defendant and Fieldfisher to questions of relevance in the redaction of the WhatsApp messages. The jurisdiction to grant relief under para 17 of CPR PD 51U is therefore engaged. The question which then arises is the nature of the relief that it is appropriate to grant having regard to the overriding objective and the factors such as the nature and complexity of the proceedings and the other matters listed in para 6.4 of CPR PD 51U.

I am not persuaded that the first defendant should simply be ordered to disclose the entirety of the WhatsApp chains in unredacted form, which is the principal head of relief sought by the claimant. While I think that there will be some more, and probably many more, messages which will be disclosable if the correct approach to relevance is adopted, I think it is also quite possible that this will not be the case in respect of every chat. I do not think that the stage has been reached at which blanket disclosure is warranted.

Nor do I think it is appropriate (anyway at this stage) for the messages to be disclosed into the type of confidentiality club contemplated by a Hollander order. Although the point was not explored in detail during the hearing, Mr Haydon said that Hogan Lovells are involved in proceedings against the first defendant in other jurisdictions, a factor which is capable of giving rise to difficulties if they as a firm come under disclosure obligations elsewhere. It may prove to be the case that these concerns are not well-founded, but I am not in a position to say that that is certainly the case.

I have also considered whether the court should inspect the material itself, but have decided that, even though I am docketed to this case and have the knowledge of the issues which flows from dealing with a number of heavy interlocutory applications, inspection by me is not the most satisfactory way forward.

In my view, the right relief is to direct a further review of the redactions of all of the WhatsApp messages, having regard both to the need to assess them against all of the issues in the proceedings and not just the Issues for Disclosure and taking account of the views that I have expressed in this judgment about the breadth of what is capable of being relevant. I shall also direct that the first defendant instructs Fieldfisher to prepare a schedule which identifies in relation to each WhatsApp message in respect of which redaction is sought to be maintained, the names of the recipient, the date and time of the message and a generic description of the subject matter of the exchange without disclosing any of what Fieldfisher are satisfied amounts to irrelevant and confidential information. If but only if each of these pieces of information is identical in relation to more than one redaction within a single chain, the messages in respect of which that information is given may be grouped and dealt with together.

I appreciate that this task may prove to be time-consuming, and in some cases would be regarded as disproportionate. However, as required by para 17.2 of CPR PD 51U I have had regard to the factors identified in para 6.4 and am satisfied that in the particular circumstances of these proceedings (including most especially the very limited disclosure of documentation from what the claimant called the first defendant’s own sources and the nature and complexity of the case) the carrying out of such an exercise is one that is reasonable and proportionate having regard to the overriding objective. I hope that the parties will be able to agree on an appropriate timescale for it to be done. The schedule is to be verified by a witness statement.

If, as a result of that exercise, there remain any issues in dispute between the claimant and the first defendant, it remains open to the claimant to seek further relief under para 17 of CPR PD 51U, in respect of which I will if appropriate rule without a further hearing.