High Court orders "mass disclosure": parties only to exclude unequivocally irrelevant documents

The judge's starting point was that most large disclosure reviews now involve a large body of electronic documents being reduced via electronic filtering to a more manageable number of documents, which are then manually reviewed by a qualified individual, capable of discerning which documents are relevant to the issues in dispute. The judge pointed out that such a process inevitably results in a significant volume of documents being discarded at the outset, without having been reviewed by a person.  In cases with a particularly large number of documents, the temptation will be to "make the electronic filter more aggressive", which could compound the risk of relevant material being excluded.

The judge opined that, where one could not be confident as to the efficacy of this electronic filtration process, it was "fatally flawed" and should not be adopted. He did not specify the circumstances in which the receiving party might legitimately have cause to question such a process, but he did consider those circumstances to be present in this case. Several factors appear to have influenced this conclusion. The first was that the parties had struggled over an extended period to agree an approach to disclosure, despite undertaking significant work to compile two Disclosure Review Documents. The second was that these proceedings were closely related to other proceedings jointly before the High Court and the CAT, where a conventional disclosure regime had not worked satisfactorily and had led to a series of specific disclosure applications being made at the pre-trial review. The third was a more general observation that in competition claims, the "issues for disclosure are more diffuse", leading to difficulty in crafting disclosure regimes which can robustly identify relevant documents.

At a case management conference in July 2022, the judge therefore raised with the parties a possible alternative process, which he described as involving "less the attempt to identify and produce relevant documents to the Receiving Party, and more a process of excluding unequivocally irrelevant and privileged documents from production, whilst providing the rest to the Receiving Party for the Receiving Party itself to review" (emphasis in the original).

The judge described the Claimants as having "reacted with forceful opposition" to his proposal. The essence of their concerns was articulated by the Court of Appeal in Nichia Corporation v Argos Limited [2007] EWCA Civ 741. In that case, the Court of Appeal said that "[i]t is wrong just to disclose a mass of background documents which do not really take the case one way or another" and that there was "a real vice in doing so" for several reasons. Lawyers from the receiving party would be compelled to engage in an unnecessarily extensive, and potentially disproportionate, document review, and there was an increased risk of irrelevant documents being imported into the case and included in the trial bundle, each of which would risk generating substantial and unnecessary "downstream costs". Further, there was a real risk that important documents would be overlooked; subsumed within a mass of irrelevant material.

The judge was not persuaded by the Claimants' objections. He opined that Nichia Corporation was a case of its time and that disclosure processes were now more sophisticated, such that the concerns articulated by the Court of Appeal could be mitigated, or altogether avoided. The receiving party would no longer be required to conduct an eyeball review of all documents disclosed; on the contrary, concept grouping, predictive coding and other AI-based searches could now be used to dissect the disclosure in a targeted manner determined by the receiving party. In the judge's view, the more acute risk in this particular case was that relevant documents would be mechanically excluded at the outset, without any manual review at all.

The judge held that a process of "massive overdisclosure" (his description, albeit one he said was tendentious and ungenerous), ought to be adopted where the following four conditions are satisfied:

• There is a real risk that if a standard disclosure process is adopted, using disclosure models to identify documents responsive to particular issues, relevant documents will be missed.
  
  
• There is no danger of the process being used to oppress any of the parties to the litigation, which might occur, for example, if the receiving party had limited resources.
  
  
• The risk of disclosing privileged material is contained to the levels of any standard process of disclosure.
  
  
• Confidential material, whether relevant or irrelevant, is appropriately protected.

In the instant case, the judge held that the first two conditions were satisfied, and that an appropriate regime could be devised to mitigate the risk of inadvertent disclosure of privileged material, and preserve confidentiality over confidential documents.

As regards the protection of privileged material, he suggested that there should be a separate search, targeted specifically at the identification of privileged material, and that a qualified person should review all documents withheld from production on grounds of privilege. The judge was not prescriptive as to how this should work, but ordered each party to produce a statement describing the process it had adopted.

As regards the preservation of confidentiality, he decided to buttress the CPR r.31.22 regime by imposing a de facto confidentiality ring in respect of all documents disclosed, rather than just a specified class. Specifically, he ordered that all disclosed documents would be held on a platform chosen by the receiving party, and that the receiving party would maintain a record of which individuals accessed those documents (and when). External counsel and experts would be able to access the disclosure without requiring the consent of the disclosing party and without having to sign an undertaking. However, the disclosing party would be required to consent to anyone else being permitted access, including the in-house counsel of the receiving party, and any such individuals would be required to sign a confidentiality undertaking, with a penal notice. Any dispute about whether a particular individual should be granted access to disclosed documents would be resolved by the judge on the papers. The judge suggested that this regime should avoid the need to establish an "inner" confidentiality ring.

The judge concluded that the disclosing parties in this case would be unable proportionately to filter a large body of electronic documents, using established technological methods, without creating an intolerable risk that relevant documents might be discarded unseen. Yet he also concluded that the receiving parties would be able to search and analyse the documents disclosed, using the same or similar techniques, such that the risks of relevant documents being overlooked, or of an onerous and costly eyeball review, were negated. While there may be some tension between these two positions, the judge appears to have considered that the balance lay with ensuring, insofar as possible, that relevant documents were not excluded from consideration, allowing the receiving party the opportunity to determine, in the manner it saw fit, how those documents should be searched and analysed. This is a notable inversion of the traditional approach and seems to represent something of a departure from the prevailing trend, which underpinned PD57AD, of taking a more targeted approach to disclosure.

The judge appeared to consider that this more expansive approach would be particularly suitable for competition claims, which are excluded from PD57AD. However, none of the four conditions that he suggested had to be satisfied before "mass disclosure" should be ordered, are necessarily limited to competition claims. Indeed, they would arguably apply in many disputes between well-resourced parties, where the disclosure process is likely to involve the harvesting of large quantities of electronic documents.

It will therefore be interesting to see whether this approach gains further traction, in competition cases or more generally, and whether practitioners' experience vindicate the judge's expectation that the Court of Appeal's concerns in Nichia Corporation can now be overcome.