CL-2022-000558 - [2025] EWHC 1995 (Comm)

Mr Craig Morrison KC, who appears on behalf of the First and Second Defendants, submits that there should be a split trial in this action; the first dealing with liability, the second dealing with both quantum and remedy. He says that that will save time and expense, firstly because the liability trial will be able to come on earlier than a combined trial. A combined trial could take up to 14 weeks, whereas it is estimated a liability trial would take 11 weeks. He suggests as well that costs will be saved because if one did not reach the stage of addressing quantum, then certain evidence, including certain expert evidence, would not be necessary which will amount to a costs saving.

He recognises, however, that that of course depends on the outcome of the liability trial and the possibility that, with certain liability findings, there would then have to be a second trial. He candidly accepts that if there had to be a second trial in that scenario, then costs would be likely to increase. There would also be, inevitably, a delay before that second trial could take place.

Nevertheless, his overall submission is that time and expense will be saved if liability and quantum are split. He chose to illustrate his points by reference to the draft order, and those issues which it was proposed on the part of the First and Second Defendant would take place at the second trial. He illustrates that by reference to one of the groups of issues that I have already identified, namely industrial accidents and the position of Mr Potanin. He refers me to para.73 of the list of issues, which provides as follows:

“If Mr Potanin did fail to take the steps set out above, was this a breach his duty of care and skill under clauses 3.14 and/or 4.2 of the FA? If so, would Mr Potanin having taken the steps set out above have had the effect of preventing the diesel spill and/or the beneficiation plant collapse in whole or in part? If so, what (if any) loss did Rusal suffer?”

(emphasis added)

He submits that the cost which would be incurred in ascertaining what, if any, loss did Rusal suffer would be saved if there was not a liability finding in favour of the liability of Mr Potanin. In the abstract, and without regard to the other issues in the case, it might be thought that there was a superficial attraction to such submission. However, even looking at the industrial accident issue in the abstract, if Mr Potanin was found liable, then inevitably there would then need to be a second trial as to what loss Rusal suffered, which would require expert evidence.

Mr Morrison KC, ultimately, also recognised that there is a possibility of appeals from any initial liability trial, although he suggested that this was not a case which gave rise to a large number of issues of law, and he submitted that this reduced the possibility of an appeal. However, there are undoubtedly substantial issues of construction of the contract governed by English law, and it is not difficult to foresee that an unsuccessful party, whoever that might be, might well at least seek permission to appeal and potentially could get permission to appeal in relation to what are difficult issues of construction of complex contractual provisions.

Mr Pillow KC, for Rusal, submits that far from saving time and costs, a split trial would both increase costs and would cause delay. He identified that there is a real possibility of appeals on particular issues and also that to bifurcate liability and quantum is a more complicated exercise than was suggested by Mr Morrison.

He also makes a number of points about the fact that if one removes oneself from the sphere of industrial accidents and look at the other claims like the AF claim, the Altan claim and the charities claims, then those are all rather different because they involve, for example, in relation to AF, whether or not there was a sale at an undervalue, and that will involve expert evidence in relation to that at the liability stage. He says it will be immediately seen that there is an overlap such that expert would be required at the liability stage, and there is no good reason, therefore, why there should be a splitting off. He elaborated those points in relation to a number of the claims, including the Altan and charities claim as well, where similar points could be made.

He pointed out as well that because the Defendants were proposing that remedies, as well as quantum, be split off, that it is possible that there could in fact be no less than three trials on the approach adopted by the Defendants. Indeed, he points out that in their Skeleton Argument the First and Second Defendants candidly admit as much, because footnote 13 provides:

“It necessarily follows that the six month period for remediation under the FA can only begin once the High Court reaches а decision on the appropriate remedy to award, which will – if а split trial of liability and remedy is ordered – only take place at the conclusion of the trial on remedy.”

It is therefore possible, on the Defendants’ approach, that not only might there be a split trial between liability and quantum, but questions of remediation might not be possible until a third stage. Mr Pillow KC also points out, in context of fact that some of the issues involve claims, for example, for specific performance, that some of the very things which a Court would be keen to consider sooner rather than later would, on this approach, be put off to a later stage.

In this regard, he makes a number of points about the various issues which he suggested would be put off to a later trial. These include Issue 11, remediability, which provides, “Whether, insofar as breaches are established, such breaches are now irremediable and if not, whether they have been remedied.”

He submits it is inherently unattractive that that would be split off to a later hearing. He also refers to Issue 49, which is said to be appropriate for being split off, “Whether there is any continuing breach of Mr Potanin and/or Whiteleave’s aforementioned duties under the FA.”

He points out that that, at least at first blush, is not even a quantum issue and would appear to go to liability. He also identifies Issue 69, “Whether there is risk of continuing breaches of clause 3.14 if Mr Potanin is not restrained by the Court.”

That is another issue which it is proposed would be hived off and, yet, that goes to the question of injunctive relief and is not about quantum at all. It is all to do with liability and the remedies that may be available. The further illustration he gave was Issue 89, which is, “Whether Rusal is entitled to specific performance.”

Again, it is suggested it would not be appropriate to hive that off to a subsequent hearing.

In the case of Daimler AG v Walleniusrederierna & ors [2020] EWHC 525 (Comm), [25] to [32], I addressed the applicable principles that apply in relation to whether or not there should be a split trial. I identified that the Court’s power to order a split trial is part of its general powers of case management, which are set out in CPR r.3.1. CPR r.3.1(2) specifically provides that the court may:

“(f)

direct that part of any proceedings (such as a counterclaim) be dealt with as separate proceedings; …

(j)

direct a separate trial of any issue;

(k)

decide the order in which issues are to be tried.”

In the competition context, which was also the context in that case, the leading cases are The Leaflet Company Limited v Royal Mail Group [2008] EWHC 3514 (Ch), and Electrical Waste Recycling [2012] EWHC 38 (Ch), which was applied outside the context of competition law by Norris J in Hawk v Sumner, 9 November 2016, Unreported, [49] and following.

In Electrical Waste Recycling, which set out principles which are equally applicable in cases other than competition cases, Hilliard J provided guidance in the form of a non-exhaustive list of relevant factors to take into consideration whether to split the trial at [6]. The bracketed comments which follow are my own:

“Where the issue of case management that arises is whether to split trials the approach called for is an essentially pragmatic one, and there are various (some competing) variations. These considerations seem to me to include:

[factor 1] whether the prospective advantage of saving the costs of an investigation of quantum if liability is not established outweighs the likelihood of increased aggregate costs if liability is established and a further trial is necessary;

[factor 2] what are likely to be the advantages and disadvantages in terms of trial preparation and management;

[factor 3] whether a split trial will impose unnecessary inconvenience and strain on witnesses who may be required in both trials;

[factor 4] whether a single trial to deal with both liability and quantum will lead to excessive complexity and diffusion of issues, or place an undue burden on the Judge hearing the case;

[factor 5] whether a split may cause particular prejudice to one or other of the parties (for example by delaying any ultimate award of compensation or damages);

[factor 6] whether there are difficulties of defining an appropriate split or whether a clean split is possible;

[factor 7] what weight is to be given to the risk of duplication, delay and the disadvantage of bifurcated appellate process;

[factor 8] generally, what is perceived to offer the best course to ensure that the whole matter is adjudicated as fairly, quickly and efficiently as possible.

Other factors to be derived from the guidance given by CPR Rule 1.4 which reflect a common-sense and pragmatic approach may include

[factor 9] whether a split trial would assist or discourage mediation and/or settlement, and

[factor 10] whether an order for a split late in the day at the expenditure of time and cost might actually increase the costs.”

As I pointed out at [28] in Daimler, the judge must undertake a “pragmatic balancing exercise” which requires assessing “how a case is likely to unfold according to whether or not there is a split (Electrical Waste Recycling at [7]).” As I noted at [29] in Daimler, if a split trial is ordered it is important that there should be a careful demarcation of the boundary between the two in terms of the issues to be dealt with at each stage (see Electrical Waste Recycling at [9]). With regard to factor 4, one example of the “excessive complexity” that a single trial can lead to is where a large number of possible permutations of loss and damage may arise, depending on the judge’s conclusion as to liability (see Leaflet Company at [7]).

As I noted at [31] in Daimler, the Court’s power under CPR r.3.12(i) to direct a split trial must be exercised in accordance with the overriding objective in each case. Relative consideration under CPR r.1.1 includes “ensuring that the parties are on equal footing”, “saving expense”, “ensuring that [the case] is dealt with expeditiously and fairly”, and “allotting to it an appropriate share of the court’s resources, whilst taking into account the need to allot resources to other cases.”

As I also noted in Daimler at [32], relevant considerations under CPR 1.4, in addition to those set out previously by Waste Recycling in para.6, include giving directions to ensure the trial of a case proceeds quickly and efficiently and the importance of the Court dealing with as many aspects of the case as it can on the same occasion.

In the Daimler case, and for the reasons set out in that case, I reached a conclusion that a split trial was not appropriate in that case. Among the factors that I identified was a consideration of whether or not there was an alleged saving in Court time between a liability trial and a combined trial. In that case I was far from convinced that such savings existed. I said as follows, at [55]:

“(1)

It is a surprising submission, even in the abstract, that two trials, one on liability and one on quantum, would take less court time than one combined trial. For the reasons set out above, the quantum trial would not necessarily advance faster as a result of certain issues being termed at a liability stage…

(2)

I am not convinced based on the information available to me that there will be a substantial saving of time overall in relation to having a split trial. In contrast, the great advantage of a combined trial is that it produces one judgment that can be appealed and the appellant court has all the relevant factual findings on all issues and can so determine matters once and for all, and have the relevant factual and expert evidence before it and associated findings of the judge. That would not be the position if there was a split trial of liability and quantum.”

At [56], I also said this:

“I consider that the reality is obvious, that with two trials and likely appeals between the two, and separate disclosure exercises followed by separate expert evidence, the costs are likely to increase and it is not appropriate for me at this stage to attempt to assess the merits of the liability defences. Even if those liability defences are potential shortcuts, they could be treacherous shortcuts in terms of delay and expense once appeals and the separate disclosure exercise and separate trials are taken into account.”

In that case, at [57], I concluded that the likelihood was that a split trial “would in fact lead to increased costs as well as potentially causing delay, which is contrary to the overriding objective.”

I also made further points which militated in favour of a combined trial. I identified at [59] that, “Considerations of trial preparation and trial management weigh strongly in favour of a combined trial.” The reasons for that that I gave included that:

“The Commercial Court is well used to dealing with large and complex commercial disputes without adopting split trials for quantum and liability.”

I also said this at [61]:

“Secondly, a split trial also impacts on judicial resources, especially once possible appeals are taken into account. In this regard, CPR rule 1.1(2)(d) is to be taken into account as part of the overriding objective, mainly the need to allot an appropriate share of court resources to a particular matter. Each of the trials which are contemplated would be a significant trial. Each would take some time apart. There must be likely to be appeals in the meantime…It might not be possible for the same judge to try both liability and quantum, particularly if there was a significant gap, as a result of appeals and the need for disclosure not given at the liability stage. Judges might have been elevated or have retired. It would be an additional burden on court resources for there to be two trials and potentially more than one set of appeals and the potential for a second trial with different judge and additional reading and preparation time. There is also the fact that splitting the trial and allocating two trials – one for liability, one for quantum – would also tie up limited available judicial resources in the Commercial Court, even over an extended period of time.”

At [62(2)], I said as follows:

“I consider that such delay as would result in separate trials is further compounded by the strong likelihood that the ruling on liability would result in appeal, however decided.”

At sub-para.(5):

“As I have already foreshadowed, the Court of Appeal would only have issues of liability before them, not issues of quantum. The result is that the quantum trial would remain inevitable in the context of the follow-on damages claim and, depending on the findings of the Court of Appeal, issues of liability might remain to be resolved. By contrast, if there was one trial then the Court of Appeal would have all the appealed issues of liability in quantum before it and this would bring certainty and finality.”

I also noted at sub-para.(6) that:

“There would be a delay in that case of up to three years in determining issues of quantum and I do not regard that as either satisfactory or appropriate, having regard to the overriding objective.”

On the facts of that case, at [63] I concluded that a split trial would not take place long before a full trial on liability and quantum and went on to conclude that it was not clear that there would in fact be a significant time or cost saving from having one trial as to liability and a further trial on quantum as opposed to a combined trial.

I also addressed at [68] the question of settlement. I noted that:

“The Defendants also submitted that trial and liability taking place in advance of a trial on quantum would encourage the parties to settle by defining the scope of the relevant commerce and defining more clearly the total value of the claim.”

However, I concluded at [69] that the settlement was more likely to be hindered than helped by a split trial. I expressed the view in that case that the possibility of settlement is maximised if all the issues were prepared and tried together.

I also considered at [70] that it was improbable that settlement would be possible until the true quantum of the claim – or the various permutations on the quantum – was apparent in the context of the fact that the sums at stake were very large. Overall in that case, at [77] I concluded that the application to split the trial was inappropriate and that the matter should proceed to trial on liability and quantum.

I consider that all the matters there identified apply equally on the facts of the present case, as shall be seen in the discussion section that follows.

In this regard, and as was identified by Peter MacDonald Eggers KC, sitting as a Deputy Judge of the High Court, in the case of Jinxin Inc. v. Aser Media PTE Limited and others [2022] EWHC 2431 (Comm) at [23]:

“The fact remains that the decision to split what would otherwise be a single trial into more than one trial each dealing with defined issues is a step out of the norm, where in most cases there will be a single trial determining all of the issues arising in an action. Accordingly, there must be a real and substantial advantage if a split trial were ordered to take place.”

See also at [26]:

“Unless a split trial can be justified as a means of resolving the disputed issues in action in accordance with the overriding objective with clear benefits over and above those of a single trial, the peril exists that a split trial will add considerably to the parties’ costs burden, will delay the conclusion of the action (with an unappealing drain on the Court’s resources) and/or will lead to unanticipated difficulties.”

I agree with the sentiments expressed in those paragraphs, which are consistent with the views I expressed in the Daimler case and also with what was said by Hildyard J in the Electrical Waste Recycling case.

I am satisfied that far from there being a saving in Court time and costs if liability and quantum were split in this case, it will inevitably lead to additional costs. The superficial attraction of identifying certain of the issues in the liability stage, such as that in relation to the Industrial Accidents Claim and Issue 73, does not bear examination.

As was identified by Mr Pillow KC, in relation to other important aspects of the claim, AF, Altan and the Charities Claims, each by their very nature involves a consideration and valuation, for example as to whether there was a sale at an undervalue or the like, which will inevitably involve expert evidence at that time. So, such expert evidence will be needed at the breach stage and, if quantum was to be bifurcated off, then there would need to be further expert evidence and the experts, who may well be the same experts, would then have to come back and give evidence twice. That is inherently unattractive and is likely to increase both costs and lead to delay.

Secondly, and as I foreshadowed, and as Mr Pillow KC identified, there is the very difficulty of the First and Second Defendants’ proposal to actually identify which issues should be split off. It is clear, as I have already identified, that many of the issues that are identified as being split off in fact either go to breach or go to remedy and logically should be determined at the first trial. As I have already noted, and as the First and Second Defendants are quite candid about, the approach that they suggest could result in the question of remediation only being dealt with at a much later stage with the possibility only of three trials.

As Mr Pillow KC submitted and illustrated (I consider rightly), if a split trial was to take place and if the First and Second Defendants were correct as to when that trial should commence, which the court has yet to determine and will determine later in the course of the case management conference, if one took a split trial from January 2028 – which is the First and Second Defendants’ proposal, with an 11-week trial – it is unlikely there would be a judgment before the end of 2028. There is a real possibility, I consider, that there could then be at least applications for permission to appeal and potentially permission to appeal being granted in relation to liability issues. That itself would be unattractive in terms of the fact that the Court of Appeal would not have all issues before them, and also there would be delay of course whilst that appellate process was dealt with.

But even leaving aside that appellate process, assuming there were findings in relation to liability, there would then had to be at quantum trial, and that quantum trial could itself take three to four weeks, and almost inevitably that would result in a further significant period before there could be such a trial. Based on current lead times, it is quite possible that that could be a year or 18 months thereafter, by which stage one would be into mid-2030 in terms of quantum. By the time one had a judgment in relation to that, one would be at the end of 2030. Then, on this hypothesis, there would then be the remediation period, whether that be six months or otherwise, which would take one into 2031.

If there was a dispute about remediation that would then necessitate another trial, which might well not start until the end of 2032 into 2033. On any view that would mean that a period of up to seven years would have passed from the present time before there was overall finality. Each of those stages of course could also carry the possibility of applications for permission to appeal to the Court of Appeal and potentially appeals to the Court of Appeal, which would introduce further delay in the overall timescale. In the meantime, and as in Daimler, the chances of settlement would, I consider, be reduced in circumstances where none of the parties would know what the final outcome might be. That would be all the more acute if quantum was split off so that the sums involved could themselves not be quantified.

Far from saving time and costs, I consider that a split trial of liability and quantum, a fortiori liability versus quantum and remedies, would lead to considerably increased costs and delay on what is a realistic scenario.

For present purposes, it would not be appropriate to consider the merits of the respective claims which are advanced, but on any view no one is suggesting that the claims advanced are not arguable. Therefore, one possible outcome is that those claims might succeed. I consider, for the reasons I have identified, and those which I set out in Daimler, which I consider apply also in the present case, not only would there not be a real and substantial advantage if a split trial was ordered, but in fact there would be real and material disadvantages of not dealing with all issues, save remediation, at the first trial.

Those relate to increased costs, the possibility of experts having to give evidence on more than one occasion, the use of Court resources in relation to potentially two separate trials, if not three, the fact that the appellate process could intervene, and the Court of Appeal would not have, at any one time, all the issues before it to be determined at the same time, and the fact that the prospect of settlement would also be reduced all militate against a split trial. For all those reasons, and having regard to the overriding objective, and the need to proceed expeditiously and justly to a final outcome in an action which already relates to events a number of years ago, I consider that a split trial would not be appropriate.

Accordingly, the trial will proceed as one trial, dealing with both liability, quantum and remedies, leaving over the possibility of remediation, which of course may or may not apply, depending on the outcome of the trial.

I also consider that such a combined trial of liability and quantum is capable of being dealt with within a period of 14 weeks, as compared to 11 weeks for liability only. That itself and the relatively modest difference between an 11-week trial and a 14-week trial is a further factor which shows that dealing with liability and quantum would be an appropriate course. On any view, if it was necessary to deal with quantum and remedies separate, very much more than 14 weeks overall would be needed, with very considerable costs introduced by the bifurcation process, including, by way of example only, the associated brief fees for a second substantial trial which, due to the passage time, would require substantial preparatory work and reading in, which would not be necessary in the event of a combined trial. Ditto the Court resources utilised for a 14-week trial and, if necessary, any appeal therefrom would be far less than bifurcated proceedings, which could result in both multiple appeals to the Court of Appeal and court resources being used on at least two, if not potentially three, trials.

I will now address case management directions to ensure the expeditious progress, and trial, of the action as a whole.