Header graphic for print
Global Insurance Blog International Insurance and Reinsurance News, Trends, and Cases
Posted in Case reports, UK

Disclosure Pilot Round-Up: “The Three Cs – Cooperation, Confidentiality, and Control”

After a dearth of decisions in the early months of the Disclosure Pilot, judges have begun lining up to add to the growing body of commentary on best practice for litigants.

Our overview of PD 51U’s (no-longer-new) rules can be found here.

While various judgments have covered a whole host of disclosure issues (e.g. the importance of the new Disclosure Guidance Hearing regime[1], the reiteration that Model E disclosure (the-artist-formerly-known-as Peruvian Guano) is truly exceptional[2]), the focus of this article is on three recent High Court decisions which have provided helpful clarification in the following areas:

  1. Cooperation (McParland v Whitehead [2020] EWHC 298 (Ch) (“McParland“)): parties who adopt an uncooperative approach and/or abuse the Disclosure Pilot process for tactical gain have been warned that they may face immediate adverse costs consequences.
  2. Confidentiality (SL Claimants v Tesco plc [2019] EWHC 3315 (Ch) (“Tesco“)): given that the foundation of the Disclosure Pilot is the need to adopt a cooperative approach, the courts will – now more than ever – strive to find a delicate balance between ensuring proper disclosure while preserving the confidentiality of commercially sensitive information.
  3. Control, adverse inferences, and Model C issues; Pipia v BGEO Group [2020] EWHC 402 (Comm) (“Pipia“)): a) re-affirmation of the principle that “control” for the purposes of CPR 31.8/PD 51U includes an “arrangement or understanding” (less than a legally enforceable right) which grants access to the relevant documents, b) an overview of circumstances in which adverse inferences will be drawn from failures to comply with disclosure obligations, and c) helpful commentary on the format and scope of Model C searches.

1. ‘Cooperate or else’ (McParland)

In McParland, the High Court (Chancellor Vos) fired a shot across the bows, warning litigants to cooperate, or face the (costs) consequences. After emphasising “the need for a high level of cooperation between the parties and their representatives“, Vos C made it clear that abusing the Disclosure Pilot process for tactical gain will be met with zero-tolerance (para 54):

It is clear that some parties to litigation…have sought to use the Disclosure Pilot as a stick with which to beat their opponents. Such conduct is entirely unacceptable, and parties can expect to be met with immediately payable adverse costs orders…No advantage can be gained by being difficult about the agreement of Issues for Disclosure or of a DRD, and I would expect judges at all levels to be astute to call out any parties that fail properly to cooperate as the Disclosure Pilot requires.

In broad outline, McParland concerned alleged breaches of non-compete and confidentiality clauses in a service agreement which the defendant (Mr Whitehead) had entered into with the claimant (McParland & Partners, a financial planning/advisory company).

Prior to the hearing, the parties had agreed 16 Issues for Disclosure. However, having failed to agree on disclosure models for 7 of the 16 issues, the parties sought pre-CMC guidance via the new Disclosure Guidance Hearing mechanism (see para.11 of PD 51U).

Despite first caveating that “the solicitors have tried on each side to comply with the pilot” and that his comments were “not…by way of criticism“, Vos C nevertheless proceeded to level some thinly veiled criticism at the parties’ conduct, observing that “[t]here are a number of areas in which the parties in this case have misunderstood how the Disclosure Pilot is intended to work” (para 43).

Vos C gave a number of examples of the way in which the parties’ approach had fallen short of the collaborative approach now expected by the courts:

  • In settling on a list of 16 Disclosure Issues, the parties had included a number of superfluous issues which – while appropriate for “the list of issues for trial” – were not “issue[s] that needed to appear on the list of Issues for Disclosure at all” (para 45). In stark contrast to the 16-issue length of the parties’ agreed list, there were “really only 3 Issues for Disclosure” (para 48).[3]
  • Rather than engage in a protracted debate about whether non-relevant documents generated by a wide-ranging Disclosure Issue should be redacted (“an unhelpful [approach]“), the parties perhaps should have recognised that “this issue was a classic one for model C disclosure” (para 50).

HL commentary

Perhaps because we remain in the (relatively) early days of the Disclosure Pilot, the McParland parties were afforded significant leeway (“I have…no intention of criticising the parties in this case“: para 53).

However, given that this is the latest in a series of examples of judgments extolling the cooperative virtues of the new regime[4], time is running out for recalcitrant litigants to get with the programme (para 58):

Cooperation between legal advisers is imperative. The Disclosure Pilot must not be used as an opportunity for litigation advantage. If that is attempted, the parties responsible will face serious adverse costs consequences.

2.  Confidentiality and proper disclosure: the balance to be struck (Tesco)

Part of[5] the main guidance from Tesco is that a delicate balance needs to be struck between: a) achieving proper disclosure which can be satisfactorily scrutinised and policed, and b) the preservation of confidential information.

Since “[p]arties should not be required as the price of vindicating their rights [i.e. the relevant civil claim] to prejudice their proprietary rights“, the courts will be astute – in appropriate circumstances – to help the parties to find “tailored [solutions] affording greater flexibility” (para 81-82, per Hildyard J).

Tesco is one of the cases arising out of Tesco plc’s 2014 accounting irregularities scandal[6]. In long-running, complex litigation, a number of groups of claimants (here, the so-called “SL Claimants“) alleged various breaches of FSMA’s misstatement provisions (s.90A, Schedule 10A).

In order to succeed at trial, the SL Claimants would need to demonstrate reliance on Tesco’s allegedly misleading statements (see e.g. FSMA, Schedule 10A(3)(4)). Since the SL Claimants had invested in Tesco’s shares via an investment fund (the “Frankfurt Trust”), Tesco sought disclosure of the underlying formulae/model which formed the basis of investment decisions made (on behalf of the SL Claimants) by the Frankfurt Trust (para 76).

Essentially, Tesco’s objective in seeking such disclosure would be to show that – on the basis of a counterfactual scenario in which the relevant accounting statements had been true – the SL Claimants (via the Frankfurt Trust’s investment model) would have made the same investment decision anyway, such that it cannot be said that there was sufficient reliance on those (allegedly misleading) accounting statements (para 80).

The SL Claimants (naturally) took a contrary view. Citing the commercial sensitivity of the underlying formulae (described by the SL Claimants as “the life’s work, essentially, of the gentleman behind this“), the SL Claimants proposed a compromise. Instead of Tesco being given the “full, unredacted and native version of the model“, the SL Claimants suggested that Tesco – “instead of running its own tests” – could inform the SL Claimants of the tests it wanted to run and could then observe the Frankfurt Trust running those experiments via video-link (para 79).

Hildyard J recognised “the particular confidentiality and sensitivity of the formulae” in question. However, after citing House of Lords authority that “the burden is on the party resisting disclosure to persuade the court that [preserving confidentiality] can be done without eroding the rights of the other party“, Hildyard J nevertheless considered that “some better balance [than the SL Claimants’ proposals] must, if at all possible be achieved…” (para 80-81).

To this end, the court invited the parties to go away and come up with a better solution, suggesting that an improved compromise might be found via “the instruction of a highly trusted third party nominated by Frankfurt Trust and approved by the court to supervise the process…” In the absence of agreement, Hildyard J warned that “I shall have to adopt the rather blunt and binary approach left to the court” (para 82).

HL commentary

The interests of one party in preserving its confidential information will often be in direct conflict with the right of another party to receive proper disclosure[7]. Tesco demonstrates that the courts will – as far as possible – encourage and assist the parties to find eminently sensible solutions which achieve the requisite delicate balance between confidentiality and disclosure.

3.  Meaning of “control”; scope of Model C disclosure (Pipia)


In Pipia, the High Court (Andrew Baker J) re-affirmed the principles that:

  1. For the purposes of CPR 31.8/PD 51U, “control” includes an “arrangement or understanding” (less than a legally enforceable right) which grants access to the relevant documents.
  2. Such an arrangement/understanding does not need to provide free and unfettered access to the relevant materials (an argument which had been run by the party resisting disclosure). Rather, the access can be of a more limited quality/scope.

The first of those items was common ground between the parties (consistent with a long line of appellate authority: see e.g. Lonrho Ltd v Shell Petroleum Co Ltd (No 1) [1980] 1 WLR 627). However, BG Group (the defendant) argued that “as a matter of principle an arrangement must grant an unfettered right of access in order for it to qualify as a control arrangement” (para 41).

The court disagreed, holding that there was no such general rule that – in order for there to be “control” – the access must be free or unfettered. In so finding, Andrew Baker J gave the following example (emphasis added; para 52):

…suppose a standing consent is provided by a subsidiary to its parent to provide to the parent…any written resolution of the subsidiary’s Board relating to some defined subject matter. That would give the parent control over such Board resolutions…The unconditionality of the consent would give it the quality required to amount to control. But it would amount to control only over Board resolutions relating to the defined subject matter

The underlying case concerned the validity of the sale of certain assets (in order to enforce security upon a defaulted loan) held by a Georgian company (Rustavi) which was owned by the claimant, Mr Pipia. The loan which Rustavi had defaulted on had been provided by a Georgian bank (BG Georgia); the assets had been bought by EUI Investments Ltd (EUI). The alleged controversy arose because Mr Pipia’s case was that EUI’s purchase of the assets had been funded by BG Georgia (which, if true, generated a conflict of interest).

Mr Pipia sought, and the BG Group resisted, the disclosure of documents held by certain BG Group subsidiaries. Mr Pipia argued (successfully) that “control” was found in the form of two letters which granted the BG Group access to documents pertaining to the above claim held by those BG Group subsidiaries.

Adverse inferences

The court in Pipia also gave an overview of when the court will draw adverse inferences from a litigant’s failure to obtain disclosable documents from third parties:

  1. Disclosure orders (pre-trial; para 36): failing to make “sensible requests” to a third party might lead to an inference that documents held by that third party are within the litigant’s control for the purposes of disclosure orders “(the logic being that the disinclination to make sensible requests stems from an appreciation that they would be responded to in such a way as would evidence the existence of that right).
  2. Findings on the merits (mid-/post-trial; para 38): at trial, that same failure to obtain documents from a third party can result in adverse inferences going to the merits.

In any event, Andrew Baker J went on to hold that no such inferences were to be drawn here because there was no evidential basis on which to suspect the BG Group of foul play.

Model C disclosure

Finally, Pipia contains helpful guidance on framing Model C requests. Andrew Baker J confirmed that, once the Model C request has been agreed/court-approved (and the search has been run reasonably and proportionately), the disclosing party must disclose all responsive documents, without filtering for relevance.

The key guidance for litigants is therefore that the initial Model C request itself must be framed sufficiently narrowly in terms of relevance, because a further relevance-filtering process upon disclosure/production is not permissible (para 67; emphasis in the original):

The intention and effect of the language used in CPR PD 51U is this, namely that any documents located upon a reasonable and proportionate search that fall within the scope of a Model C Request adopted as part of directions for Extended Disclosure will be disclosed. Model C Requests therefore should be defined with that end result in mind; and a request for a disclosing party to search for “any or all documents relating to” a topic is not, to my mind, a Model C Request at all, it is merely a definition of a category of documents that might or might not contain anything of any real interest…

HL commentary: adverse inferences

A growing body of (first instance) judgments[8] seems to be in favour of drawing adverse inferences in appropriate circumstances. This may have curious consequences. It appears that, in one breath, the court in Pipia professed that “absent control” there is no duty to ask a third party to provide documents (“…it is no part of the litigating party’s disclosure obligations to ask such a third party to provide documents…“; emphasis in the original). However, in the very next breath, the court revived – by the back door – the very same duty whose existence it had just denied (para 38; emphasis in the original):

It can be proper, if the conclusion then is that the related party would probably have assisted if asked…to draw an adverse inference against the litigating party going to the merits…from its failure to seek that assistance.”

Regardless of the technical distinction the court drew between whether the adverse inference is drawn at the (pre-trial) disclosure stage or (mid-/post-trial) on the merits, in reality, adverse inferences force parties – however indirectly – to do what they can to avoid that inference. This is an obligation in all but name.

It remains to be seen how enthusiastic other judges will be to draw such adverse inferences. However, it may well be that Pipia and its progeny have ushered in a new era where parties to civil litigation will need to be cautioned by their legal advisers along the following lines (which will be eerily reminiscent to any police drama fan/criminal law practitioner):

You do not have to request documents from third parties (which are not in your “control”). But, it may harm your defence if you do not make such requests. Any request you fail to make may be given in evidence against you.[9]

Watch this space.


[1]               See Vannin Capital PCC v RBOS [2019] EWHC 1617 (Ch) at para 4, per Smith J.

[2]               See Kings Security Systems Ltd v King [2019] EWHC 3620 (Ch) at para 33-37, per Master Kaye; McParland v Whitehead [2020] EWHC 298 (Ch) at para 13, per Vos C.

[3]               See also to the same effect Kings Security Systems Ltd v King [2019] EWHC 3620 at para 20.

[4]               See e.g. para.78 of another judgment of Vos C: UTB LLC v Sheffield United & Others [2019] EWHC 914 (Ch).

[5]               This case also provides helpful commentary on disclosure proportionality (even in cases where the, exceptional, option of Model E disclosure is appropriate: para 5-8)), and the extent to which confidentiality is lost over a document referred to in open court (para 32-43).

[6]               See the SFO’s January 2019 Deferred Prosecution Agreement press release here.

[7]               See also the Court of Appeal’s resolution (in favour of disclosure) of the conflict between Iranian banking secrecy laws and HM Treasury’s right to receive proper disclosure, in the context of Iranian sanctions: Bank Mellat v HM Treasury [2019] EWCA Civ 449 at para 73-79.

[8]               The court in Pipia cited the High Court case of Ardila Investments v ENRC [2015] EWHC 3761 (Comm), per Males J (as he then was).

[9]               c.f. The equivalent criminal law caution given upon arrest: “You do not have to say anything. But, it may harm your defence if you do not mention when questioned something which you later rely on in court. Anything you do say may be given in evidence” (PACE, Code C, para.10.5).