I often hear the buzz words ‘opt out proceedings’ mentioned in the context of group litigation, and given that most English litigators won’t have experienced them, it struck me that it would be useful to explain such proceedings to readers. As a starter, ‘opt out’ is where representative proceedings are brought by a single claimant on behalf of a class but without the consent of the class (or at least of all of the class), and where there is scope for members of the class to say ‘not in my name’ and to opt out of the representative proceedings. There are only two routes for opt out proceedings in English civil ligation; either collective proceedings brought in the CAT or proceedings brought in the County or High Courts under CPR 19.8. This article will focus on the Court rather than (competition only) tribunal route.
Jargon
Something to get out of the way is terminology. CPR 19.8 is the representative action. It is not the opt out process. The default position is that CPR 19.8 has a universal and comprehensive application to the whole class with no opt out option at all. It is only if the court permits an opt out that this option arises under 19.8(4)(a). Equally whilst we can talk about ‘opt in’ processes the reality is that all routes outside CPR 19.8 are by definition opt in; claimants have to come forward and seek vindication of their claim whether by a GLO, a multi-claimant claim form or some other form of group action.
It is also possible for a representative action to be structured on an opt in basis where the same interest is shared only by those members of the class who opt in (i.e. who come forward to bring the claim). This avenue has not been traditionally used for representative actions and immediately throws up questions as to why other opt-in procedures are not suitable. This very point was relied upon by Michael Green J in Wirral Council v Indivior Plc [2023] EWHC 3114 (Comm) in ordering that Wirral could not act as the representative pursuant to CPR 19.8(2). The case is on appeal and should be heard later this year.
The representative action has undergone something of a resurgence in recent years. It was originally fashioned as a rule designed to achieve practical justice where wrongs were committed against a large class of members. It then went through a phase of narrow construction of the “same interest” requirement for such an action, under which case law if a defence could be raised against some but not all members the class did not have the same interest (Emerald Supplies v BA [2010] EWCA Civ 1284 [64]). That is no longer the case following the Supreme Court decision in Lloyd v Google [2021] UKSC 50. The recent approach is to refocus on the purpose of the representative action procedure (to permit efficient proceedings which bind a multitude of parties because they have the same interest) which requires the representative action to be a flexible tool where divergent interests (such as where a particular defence might apply to one set of claimants but not to another set) are no bar to having the same interest (Lloyd v Google [72]).
Where there are conflicting interests it may be still possible to proceed by way of a representative action by splitting the class along the lines of the conflict and having separate representatives. Thus one could have class A which is not exposed to a limitation defence and class B which is exposed to a limitation defence. In practice it will be much more complex, as (to stay with the example of limitation) there will often be uncertainty as to when an extended limitation period under s.32 of the Limitation Act 1980 terminates by reason of the claim being discoverable by reasonable claimant diligence. This will give rise to problems in identifying which claimants fall within which class. It also remains to be seen whether multi-class representative actions are economically viable. This is because of the impact on the availability and terms of funding and ATE once multiple classes come into play. For instance if there is just one class which in aggregate has claims worth £50 million that will be likely to obtain funding on better terms than if there are two classes of claims each worth say £25 million, because funding terms generally improve the higher the value of the claim and the higher the funding required. It therefore seems likely that in practice unless the multiple classes are each very large and high value a conflict is likely to make the terms of the representative action less attractive to claimants and at worst could render it non-viable. That said, the Court of Appeal in UK Trucks Claim Ltd v Stellantis NV [2023] EWCA Civ 875 [88] was not presented with such practical problems when it decided that a class conflict could be resolved by splitting the class into two classes with different legal teams, experts and funders.
The opt out procedure involves the burden of some kind of notification/advertising of the representative action in order for members to know of their right to opt out, as happens in CAT collective proceedings after certification. Once a member is notified of such a right why would they want to opt out? It might be thought that a representative action is a free option where the members of the class stand only to gain; if they win they get a favourable judgment (which they cannot enforce without permission of the court) and if the action fails they ordinarily will not be on the hook for costs which is also the position in the CAT. The answer from the experience in other jurisdictions and the CAT is that there remain reasons to opt out because the representative action may (i) seek more limited damages than an individual might be able to claim (perhaps for the purposes of efficiency of the representative action or so as to emphasise the same interest), (ii) lead to the loss or diminution in value of alternative forms of redress (such as a compensation scheme) and (iii) damage commercial relationships with the defendant (a factor which really applies to institutional or large commercial claimants rather than retail claimants).
Opt out benefit v risks
An interesting question is whether an opt out, which shrinks the class bringing the claim, increases or should increase the exposure to costs of those members who do not opt out. A defendant might argue that once a member had notice and did not opt out they have by their conduct authorised the representative action on their behalf and should be liable for a share of the costs if the action fails. There are arguments going the other way (including authorities under earlier rules and the difficulty of deemed notice) but this appears to be an untested point. However, the risk of exposure to costs for members who did not opt out is a further reason in itself for members who are risk adverse to opt out.
Opting out does of course avoid the costs risk on the representative action. But does it provide a cake and eating it scenario so the opt outers avoid a risk on costs but can pursue a form of ‘follow on’ claim if the representative action is successful, taking advantage of the successful findings or precedent? Unsurprisingly, in most circumstances the answer is very likely to be ‘no’. In principle those who opt out can bring their own actions (this is also the case in the CAT). But the opt outs are not bound by and cannot benefit from any favourable judgment in (or settlement of) the representative action, albeit any second claim largely duplicating the first will be able to take advantage of any binding legal findings in the first, and in practice factual findings are unlikely to diverge either on the same points. However, those seeking to piggyback on others’ earlier successes will face the hurdle of limitation, which often will have expired by the time of a judgment/settlement. This problem does not apply to CAT collective proceedings, as the duration of the collective action does not count towards limitation for a later individual claim by a member who has opted out. The second hurdle is overcoming arguments that it is abusive under Henderson v Henderson to adopt this precise course because the opt outs could and should have stayed in the represented class or else advanced their own claim immediately (if they did want to pursue a claim outside the representative action). This will be fact sensitive, but there is likely to be an uphill battle to persuade a court that ‘I wanted to wait and not take the benefits/disbenefits of the representative action’ is a just approach. If not, that would undermine the purpose of a representative action and lead to the vice of multiple proceedings on the same or very similar facts (the worst outcome from any court’s perspective given the waste of court resources that is incurred- see the earlier article by Adam Kramer KC as to how this duplication concern has played out in the Pan NOx emissions litigation). Some may remember that one claimant group in the RBS Rights Issue litigation sought to be able to join the GLO register, have their claims stayed without any further costs liability and only revive the claims if other claimants won against RBS. This free rider proposal was given short shrift by Hildyard J in Greenwood v Goodwin [2014] EWHC 227 (Ch) [47-48]. However, experience teaches that there may be exceptional facts which might justify a follow-on claim, although these are likely to be few and far between. This is all to be contrasted with a GLO which has a cut-off date which has the practical effect of corralling all the claims to issuing and joining the register before the cut off. Those who do not do so and want to bring a claim will have to overcome the hurdle of seeking relief from sanctions. One can immediately see the increased certainty that a GLO brings to a defendant in respect of the risk of ‘follow on’ claims.
Claimant/Defendant side
The representative action is generally seen as a ‘pro-claimant’ tool and there are obvious reasons for that. However, it does have some benefits from a defendant’s perspective. By having only one action it limits the defendant’s costs to one set of disclosure, witness statements and so on and avoids responding to and managing multi-fronted litigation. It also achieves closure once and for all in respect of the relevant event and on a timeline which is likely to be much faster than waiting for limitation to expire (which can be very long once a claimant is able to claim a limitation extension). What might also appeal to a defendant is the ability to bring a representative action itself against a class of potential claimants and to limit it to legal issues in respect of which the defendant considered itself to be very strong. The danger is that the court might not so limit the representative action and be persuaded to broaden it to include issues where claimants might fare better. This is especially so when there is so little clarity as to how courts will deal with representative actions in practice.
There is no doubt that many of the complexities of opt out representative actions have yet to be worked through by the courts and the CAT. Let’s see how much more there is to say within the next five years (note to self to compare and contrast this blog in five years’ time).