Various Claimants v G4S Limited

This is a case management decision made at the first CMC of three sets of claims under section 90A of and Schedule 10A to FSMA 2000. The decision relates to whether there should be a split trial, whether there should be a process for sampling of claimants, and to an application made by the defendant for further information.

The claim is for compensation in respect of losses said to have been suffered in relation to shares in the defendant, G4S, acquired, disposed of or held by the claimants, being losses suffered arising either from actions taken in reasonable reliance on certain published information which it is said contained untrue or misleading statements, or which omitted information required to be included in it, or where it is said that there was a dishonest delay in publishing information.

One requirement of the legislation relates to the state of mind of one or more persons discharging management responsibilities (“PDMRs”). Essentially, dishonesty is required.

The claims arise out of alleged misconduct relating to contracts entered into between the Government and a subsidiary of G4S, G4S Care & Justice Services UK Limited, for the provision of services in respect of the electronic tagging of offenders and the management of court facilities.

The first of the three claims, G4S1, concerns alleged overbilling (“wrongful billing”), information about which at least began to emerge in 2013 and was the subject of a settlement with the Government in 2014.

The second and third claims, G4S2 and G4S3, relate to the alleged submission of false financial models to the Government relating to the calculation of the cost of services, which were models submitted for the purposes of provisions designed to ensure that cost efficiencies were shared (the “financial model fraud”). The financial model fraud was the subject of a deferred prosecution agreement with the SFO which was announced in July 2020.

In summary, the parties' positions are as follows.

The claimants seek a split trial similar to what is currently adopted in the Allianz v RSA litigation. Following an unpublished judgment of Miles J dated 28 February 2022, this comprises a first trial covering the standing of the individual claimants (“trial 1”), but otherwise common issues rather than issues specific to individual claimants. Essentially, that means all defendant side issues: whether statements were made which were false or which omitted required information, whether there was a delay in publishing, PDMR status and dishonesty. It would not cover reliance, causation, quantum or limitation.

In any event, the claimants say that there should be sampling, although on their case that would not occur until after trial 1. The claimants also say that they have already provided a lot of information about individual claimants voluntarily, and the defendant has enough information to choose a sample.

The defendant's position is that choosing a split trial now is premature, and further, it is not possible to decide whether there ought to be a split trial, or indeed how to go about a sampling exercise, until there is another CMC following full disclosure and witness statements or, as suggested this morning, witness statements from all claimant witnesses and disclosure from a sample of claimants.

The defendant points out that the claimants' estimate of a single trial allows for a vast number of witnesses in circumstances where the defendant says it is not known whether it will be necessary to call them. The defendant has also made an application under Part 18 for further information, which it says it needs to understand the case against it.

The defendant's position is that the court should list a 10-week trial now but have another CMC following disclosure and witness statements to determine the appropriate approach to take.

I have concluded that there should be a split trial, broadly as the claimants propose, but with a parallel process to establish sampling which will entail some further information from the claimants. I also envisage that that would be followed by disclosure from sample claimants and provision for a further CMC before trial 1 to take stock, and that that CMC would also be used to decide what, if any, witness statements should be taken from claimant witnesses, which might go beyond the sample claimants.

All of these issues are interrelated, but I will deal with sampling first.

In my view a sampling process must in principle be an appropriate process to undertake. There are currently 98 master claimants operating 133 different funds or accounts. I understand that these numbers may reduce a little, but it is not suggested they will reduce materially.

If sampling is done properly, then I think it should be unlikely in practice that there would be a need for multiple further trials after a split trial involving that sample, but it is important to take careful steps in determining the sample.

In particular, the claimants selected must cover all relevant periods between them. For example, even if the statements relied on are broadly the same in different annual reports, the position may well be affected by when particular individuals were PDMRs (if they are found to have been PDMRs at all), and what they knew at particular times, including whether what they knew was or had become, having not previously been, a material fact. See in particular paragraph 3(3) of Schedule 10A in relation to omissions.

The sample must also ensure that other facts that could make a material difference and differentiate claimants in a marked way are picked up and covered by the sample.

There is no real debate between the parties that the separate categories of so-called direct and indirect reliance need to be covered, the concept of indirect reliance being used to cover tracker funds in particular. But more generally I can see the need to understand any material differentiation between claimants as regards whether individuals at claimants, or entities acting on their behalf, read or relied on particular statements in (typically) an annual report, or messages conveyed by such a document as a whole (as indeed the defendant says that the claimants need to establish) or, for example, whether a particular claimant relies only on omissions.

Those points may, for example, also determine whether the judge needs to interpret particular words in particular documents.

The exercise required to understand any material differentiation in this respect needs to extend to a level of granularity that establishes whether it is said that particular named individuals reviewed the relevant published information and relied on it or whether, for example, the only evidence a particular claimant can provide in that respect is of a general practice, or perhaps invariable practice, of reviewing that sort of information before making an investment decision.

There may also be another category of factual points related to limitation that are specific to a particular claimant. The example given to me was whether it will be alleged that there is something particular to a specific claimant that means that they could not with reasonable diligence have discovered something, as compared to another claimant. It is a point about the characteristics of claimants.

Mr Colton, for the defendant, also fairly made a point about document retention policies. If there are marked differences between different claimants in terms of document retention policies, that may affect the appropriate choice of the sample, and indeed it may be that there should be examples of claimants in different categories.

A further relevant point was raised in relation to meetings and other communications with or between individuals at the defendant and specific investors or persons acting for them. The existence and content of those communications or meetings may be relevant to the extent it is claimed that they affect reliance. Obviously that may be specific to certain investors, although the point needs to be caveated by the fact that the liability under section 90A and Schedule 10A is by reference to "published information", and not as such comments in meetings or direct communications.

I also have noted the point being suggested that those meetings or communications may be relevant to the dishonesty or alleged dishonesty of individual PDMRs. That goes to trial 1 on the claimants' split trial proposal, but it emphasises the need essentially to flush out any of those points in advance.

Another point is that in making choices about samples, it is of course going to be sensible to focus on claimant groups with the most value at stake.

In summary, on sampling, I consider that the case requires a clearly timetabled process to determine the criteria for sampling and the information needed to ensure that sampling can occur. It is very important that the parties engage collaboratively and proportionately, accepting that not every fine detail or variation can be captured, whilst aiming to identify any material differences.

I believe that I have given sufficient indication as to the core points that should be covered. The process will of course be subject to the supervision of the court, but the court hopes that the parties will be able to reach agreement without the need for further judicial intervention.

I now turn to the defendant's application for further information, and indeed disclosure. Some of what I have said already is obviously relevant to this.

G4S's case, based on parts of the decision in ACL Netherlands v Lynch [2022] EWHC 1178 (“Autonomy”), is broadly that the claimants can only run a case on untrue or misleading statements by pleading specific details of statements as having been read and relied on by particular individuals, or possibly those particular individuals having drawn an impression from the document as a whole.

The claimants disagree. They say that Autonomy does not answer the point, and specifically that the reference to reliance in paragraph 3(4) of Schedule 10A refers back to published information generally. They do not, they say, need to show reliance on particular individual statements.

The parties accept that it would be quite wrong for me to even begin to comment on or determine this point at this CMC, but it is relevant to the extent of emphasising that the claimants' pleaded case encompasses each of reliance on particular statements, the published information read as a whole, and omissions.

The defendant's position is that it says it wants to understand the parameters of the case. I understand that.

The test for requiring further information set out in paragraph 14.1(c) of the Commercial Court Guide is whether the information is strictly necessary for the case to be understood. The test as expressed in paragraph 1.2 of PD 18 is a bit more broad. That refers to the information being strictly confined to matters which are reasonably necessary and proportionate to enable a party to prepare their case or to understand the case they have to meet.

I have concluded, as effectively already indicated, that some further information is justified, namely information that will extract important potential points of difference between claimants' cases, particularly insofar as it is said that particular individuals relied on particular statements or their impression from reviewing a document as a whole, and in respect of particular meetings or direct communications with representatives of the defendant so far as claimants wish to rely on that.

This is relevant to sampling but it is also relevant to the proper particularisation of the reliance case, so that the defendant can understand the parameters of the factual and legal issues to be addressed.

Alternatively, if the response from claimants or individual claimants is, for example, that the only evidence available would be evidence of the claimant's normal procedures, which involved looking at published information such as annual reports, then I think it is important that the defendant understands that that is the way in which the case is put. Similarly, the defendant should understand if a particular claimant is basing their case only on omissions.

To the extent that this may go beyond the guidance in the Commercial Court Guide, I have firmly in mind the need to promote the potential for settlement.

I also have in mind that obtaining information about direct communications and meetings relied on is relevant to the disclosure exercise that the defendant will in any event have to undertake, particularly insofar as any such meetings or communications might be relied on in demonstrating the alleged dishonesty or knowledge of PDMRs.

So the claimants should clarify their individual cases about what individuals relied on, when, and on what statements, and also disclose or provide details of any specific meetings or communications on which they rely.

To address another point that has been raised by G4S, I think it would also be helpful for the claimants to clarify whether, as I would imagine to be the case, reliance was in all cases by reference to the latest published information, at least to the extent it was not inconsistent with previous information. Or, alternatively, and to the extent it is said that there was something different in older information, the extent to which they do seek to rely on that older and more historic information, or indeed later information.

All of these points will hopefully assist effective sampling and allow the defendant to understand the reliance case better. They should also assist the defendant in having a proper understanding of appropriate date ranges for its disclosure.

As I hope has become reasonably clear from what I have said, none of this should be delayed until after trial 1. This is not only because it may affect the shape of that trial to some extent, but also because some progression of the claimants' case is needed, in my view, to ensure that settlement discussions can best be facilitated. There is also a point about ensuring proper engagement by the claimants. I do appreciate that the claimants have provided a lot of information, but there is something in the defendant's complaint that much of it is generic rather than focused on facts specific to this case.

By progressing that work now I also hope to reduce, so far as possible, the gap between trials, ensure, as already indicated, that matters relevant to defendant disclosure are brought out, and try to ensure an appropriate balance and fairness in the burden between the parties.

I will not, however, go so far as to say that full disclosure or witness statements are required from all claimants before trial 1. What the approach I am describing requires is sufficient information to allow a proper sampling decision. That is likely to be achieved by questionnaires covering the points referred to and any further points the parties agree to cover, but not in as extensive a form as the defendant has so far proposed.

Once responses have been provided, an agreed process to determine the sample of claimants to proceed with would be implemented. I then propose that there is disclosure by the sample claimants and, subject to discussion at a further CMC, the provision of claimant witness statements, at least to the extent that the case is put in terms of reliance by particular individuals on particular statements or particular documents read, and in respect of any meetings or other direct communications between the parties. This may go further than witness statements from sample claimants. All this would occur in tandem with planning for trial 1.

Separately the claimants, I believe, accepted during the hearing that there was a need for a fuller explanation of the distinction between two forms of indirect reliance that they refer to, namely market and price reliance.

As I understand it, the essential distinction is between indirect reliance that depends on the market’s assessment of published information (“market reliance”), and indirect reliance in the form of a human actor's reliance on the market price (“price reliance”). It is obvious that the cases of the tracker fund claimants depend on indirect reliance, and on the definition just provided that would appear to me to mean market reliance. But I understand that that definition is a summary, and probably not the full story. I may not have fully understood it, and I am not sure the defendant has either.

I now move on to the question of split trial. I have not been persuaded that I should defer the decision until the next CMC or that I should, for example, book 10 weeks in the court diary now as the defendant suggests and look to reduce it later. Doing that raises its own issues for the court.

Overall I have determined that a split along the lines of that now adopted in the RSA case currently appears to be the best pragmatic solution but, as already indicated, in tandem with a process to allow effective sampling, in a way that allows the defendant to understand the reliance case, disclosure as regards the sample, and possibly witness statements.

There should of course be a further CMC. That will cover the issue of witness statements. It would also potentially cover whether trial 1 should include any specific points of law that might sensibly be disposed of then if it would further increase the potential for settlement. However, the court will be very cautious about being asked to decide questions of law without making factual findings. So, for example, in relation to reliance, specific points of law are going to be quite difficult for the court to decide without the full factual context.

My power to order a split trial must of course be exercised in accordance with the overriding objective. It is a pragmatic decision. I am having to take a view, which may prove wrong, about how the case will evolve. I was referred to the useful guidance in Electrical Waste Recycling v Philips Electronics [2012] EWHC 38 (Ch) at [5]-[10]. Following broadly the points referred to in Electrical Waste Recycling, the key elements seem to me as follows.

First, I am not persuaded that a single trial is either realistic or necessarily possible. The reality is that questions of quantum seem likely to be put off anyway, subject to some potential for addressing certain points of principle at an earlier stage.

The reliance case is really not straightforward. As Mr Justice Miles commented in the RSA case, it is elaborate and complex. Those comments apply here as well. It covers forms of indirect reliance. There are issues raised as to whether there needs to be conscious awareness and reliance on particular statements, whether it is enough to rely on the published information generally, and the extent to which any presumption of inducement may assist. It is complex. Questions of the counterfactual are raised, and there is a potential overlap with issues of causation.

I accept that we do not know with clarity what the number of claimant witnesses is likely to be, but I would expect it to be considerable in relation to reliance, not least because of the number of claimants who are said to be relying on direct reliance, which appears to me to amount to a substantial proportion of claimants overall.

In contrast, taking broadly the approach the claimants have put forward, issues at trial 1 would be contained. There would be a relatively small number of witnesses and there would be some relatively clear cut factual issues which are not affected by the individual claimants' position, most obviously PDMR status and knowledge and dishonesty.

Second, two trials or even more than two trials are not inevitable. Obviously if the claimants fail at the trial 1, there will be a substantial saving in costs on any basis. If they succeed, the claimants' position is that there is then a high prospect of settlement, with the same result.

I am not convinced by that second point without a better level of understanding by the defendant of the claimants' case, but I hope that that will be achieved by a proper process of sampling and the provision of information, as I have sought to outline. I note that in the RSA case there was both sampling and disclosure from sample claimants.

If the defendant does understand the case, including a better understanding of how it differs between claimants, then I can see real scope for settlement discussions.

Third, costs. I accept that multiple trials, certainly two trials or potentially even more, may ultimately be more expensive. But there would be material savings if a later trial or trials could be avoided. The defendant's suggestion that there could be up to four trials is in my view unlikely to be realistic.

Fourth, the risk of so-called bifurcated appeals. That means separate appeals from separate trials. That risk is always there, but insofar as trial 1 will deal with fact-heavy points, such as PDMR knowledge and honesty, there is obviously significantly less scope for appeal. That point could be affected by bringing legal points of principle into trial 1, which will be a point to be weighed up in due course.

Fifth, trial preparation. A split trial would confine the work for trial as opposed to aiming for a single trial, which would be extremely cumbersome.

Sixth, can the split be defined? The answer to this is it can be defined fairly readily. On the whole, the common issues are largely discrete and can be considered separately from other issues such as reliance.

Seventh, multiple witnesses and duplication or overlap of evidence. I seek to avoid that with careful management. In particular, any evidence of either party about meetings or other communications directly between the parties should be dealt with in full at trial 1. That would include evidence from claimant witnesses about those meetings or communications, including evidence relevant to reliance.

As regards expert evidence, it appears that there will be limited expert evidence, if there is expert evidence at all, for trial 1. An issue has been raised about whether expert evidence is appropriate in relation to the materiality of omissions, and a potential overlap with the need to consider materiality in a second trial as regards reliance and the impact on a reasonable investor.

Again, I think that could be addressed by seeking to ensure that any materiality-related evidence that could overlap is obtained in full, or at least the relevant expert reports are, for trial 1, and if necessary findings made at trial 1 in respect of that.

Finally, eighth, what is the best course overall to ensure that the case is adjudicated as fairly, quickly and as efficiently as possible? I have talked about facilitating settlement. As regards the overall time period to the end of a final trial, that would obviously be longer than with a single trial. But we already have a long period since the events in question occurred. Whilst that will increasingly affect the quality of recollections, and would do so more to the extent those recollections are required in a later trial rather than the first, the already long gap raises an issue about quality of recollections anyway. I am also proposing that we shorten the gap between trials so far as possible by ensuring that work is done not only to achieve sampling but to have disclosure from sample claimants and potentially witness statements, at least where those go to the recollection of specific events rather than, for example, general investment practices.

I have taken account of the potential unfairness in placing a primary burden on the defendant in relation to trial 1, but I consider that to be tempered by the points I have already discussed about sampling, the provision of further information and disclosure, and potentially witness statements.

I have also considered the question of document preservation. This is potentially more a claimant rather than a defendant risk, because document destruction may weaken their case on reliance, but concerns about preservation of documents should to an extent be alleviated if disclosure is ordered, at least for the sample.

Further, as regards any claimant who relies on specific individuals reviewing particular documents, if that position is made clear as part of the sampling process then it is obvious that identification of relevant individuals should in practice assist with document preservation.

I am also taking into account that the outcome of trial 1 could narrow the focus of the second trial, for example, if it is found that there was no PDMR or no PDMR with the requisite knowledge for certain periods, or no concealment of material facts.

In addition, I have taken into account the potential for prejudice. This is a concern that if a judge decides in trial 1 that a statement has a particular meaning, then it may be more likely that a witness in a later trial would say that he of course understood it in the same way and indeed relied on it, perhaps making it harder for the defendant to say that even though the court interpreted a statement in a particular way, the claimant did not.

However, I hope that the process I have described to flush out specific reliance claims, and potentially take witness statements from relevant individuals, will alleviate that.

A final general point on this is a really key point about making a trial manageable for the trial judge, and facilitating the handing down of a judgment within a reasonable time frame. Overall that important consideration favours, in broad terms, the claimants' proposal.

Whilst I appreciate that going through the sampling process first could better inform the trial split or whether there should be a trial split, for example, by making the number of witnesses clearer, I do not think that this is sufficient to justify delaying a decision rather than implementing a clear plan now so that a case which has already been on foot for some time can proceed.