Eurasian Natural Resources Corporation Limited v Dechert LLP & Ors

This appeal concerns applications made by the claimant (“ENRC”) in November 2024 to amend its claims for loss during the quantum phase (“Phase 2”) of two long-running sets of Commercial Court proceedings. The first set was commenced in 2017 against Dechert LLP and Neil Gerrard, a partner in that firm of solicitors (together “the Dechert defendants”). The second was commenced in 2019 against the Serious Fraud Office (“the SFO”). Those proceedings were, and continue to be, managed and tried together by Waksman J (“the Judge”).

Following an 11 week trial, mainly concerned with liability issues (“Phase 1”), on 13 May 2022 the Judge had found that:

On 21 December 2023, following a further trial of certain issues of causation and loss (“Phase 1A”), the Judge had found that the SFO’s wrongdoing was an effective cause of losses claimed by ENRC in respect of unnecessary fees, unnecessary costs and wasted management time. The Judge further found that, but for its wrongdoing, the SFO would not have commenced a criminal investigation into ENRC in April 2013 (“the CI”), an investigation which lasted until August 2023 but resulted in no charges being brought.

The losses claimed by ENRC by reason of the CI were to be assessed at the Phase 2 trial, preliminary directions for which were given by the Judge in March 2024. Pursuant to those directions, on 30 August 2024 ENRC served Further Information in which it first identified the relevant changes to its case as to loss, leading to the applications to amend. The first Case Management Conference in Phase 2 took place on 15 January 2025, at which the applications to amend were considered. The CMC resumed on 14 February 2025 after the Judge had delivered judgment on the amendment applications, in the light of which he made orders for disclosure.

The disputed aspects of the applications to amend followed ENRC’s belated realisation that certain losses it had claimed from the outset of the proceedings, mainly comprising a significant proportion of the increased borrowing costs alleged to have resulted from the opening of the CI, had not been incurred by ENRC itself but by one or more of its subsidiary companies. ENRC sought permission to amend to claim that ENRC had suffered loss by reason of alleged diminution in the value of its shareholdings in the subsidiaries, such loss matching the proportion of the relevant subsidiary’s loss equivalent to ENRC’s percentage shareholding. ENRC made it plain that it would rely solely on this “dollar for dollar” measure of its own loss, adjusted only by the tax implications that lower borrowing costs would have had on net profits.

The SFO and the Dechert defendants (the latter being interested in the increased borrowing cost amendments as Part 20 defendants to a contribution claim by the SFO in the 2019 proceedings) accepted that the disputed amendments had a real prospect of success, in particular taking no point at the amendment stage that the loss claimed was irrecoverable as reflective loss, and made no objection on the ground of cogency or particularisation. The ground of opposition was that the delay in advancing the amendments, for which there was no good reason, meant that, whereas ENRC’s documents had been preserved since 2013, no “litigation hold” had been placed on the documents of some of the subsidiaries in question, to the substantial prejudice of the defendants.

By order dated 14 February 2025 the Judge dismissed the applications in respect of the disputed amendments for reasons he had given in his judgment dated 23 January 2025, largely accepting the defendants’ objections. ENRC appealed that decision with permission granted by Males LJ.

Following the hearing of the appeal, and to assist the parties in preparing for the Phase 2 trial due to start in April 2026, this Court made an order on 23 July 2025 allowing the appeal and permitting the disputed amendments, with judgments to follow. These are my reasons for joining in the decision to allow the appeal.

At the material times, prior to January 2014, ENRC was a public limited company and the parent of a diversified natural resources group, including extensive mining operations in Kazakhstan, the Democratic Republic of Congo and elsewhere. It had been listed on the London Stock Exchange since 12 December 2007, had become a member of the FTSE 100 on 12 March 2008, and had a peak market capitalisation of US$20 billion. The group had very extensive credit facilities with numerous financial institutions.

The SFO announced the opening of the CI on its website on 25 April 2013, stating that “[t]he focus of the investigation will be allegations of fraud, bribery and corruption relating to the activities of the company or its subsidiaries in Kazakhstan and Africa”. The same day Moody’s downgraded ENRC’s corporate family rating from Ba3 to B1 and placed that rating on review for further downgrade. Shortly afterwards Moody’s explained that the SFO’s criminal investigation and boardroom changes at ENRC were “credit negative” and that the downgrade was “in response” to those developments. In May 2014 Standard and Poor’s stated that it “viewed ENRC’s access to funding as constrained by the risks related to an ongoing [SFO] investigation”.

ENRC was de-listed from the London Stock Exchange on 25 November 2013 and ceased to be a public limited company on 16 January 2014. It is currently in the ultimate beneficial ownership of ERG SARL.

In paragraph 59 of its Particulars of Claim in the 2019 proceedings, ENRC set out its claim for “consequential loss”, alleging that the ongoing CI had damaged ENRC’s reputation and “resulted in… losses in the form of its loss of a chance to borrow funds from, or on more advantageous terms from, lenders who have declined to deal with ENRC and/or demanded increased rates of interest for the same reasons”. ENRC contended that such losses should be assessed at an inquiry as to damages after liability and other issues had been determined.

ENRC disputed its obligation to provide further particulars of its consequential loss at that stage, but on 10 January 2020 its solicitors volunteered some further information, including that, as at 31 December 2012, ENRC had borrowings of US$5.833 billion. The average interest rate that ENRC paid in relation to its borrowings increased from 5.9% in 2012 (before the CI was opened) to 7.4% in 2014 (the year after the CI opened). The increase equated to US$90 million per year. No suggestion was made that the borrowing was by entities other than ENRC.

The SFO’s solicitors responded on 13 January 2020, making the point that if specific transactions with specific lenders were not identified until a quantum trial, there was a real risk that relevant counterparties would cease to maintain their records or employees would move on, hindering the SFO’s defence.

On 13 May 2020 the Judge directed split trials, with the consequential losses being dealt with in Phase 2, including disclosure in relation to those issues. The Judge did state, however, that if there were to be further amendments to the pleadings, they should be made then and not left until later. No further amendments to the consequential loss claim in the 2019 proceedings were proposed until the applications now under consideration.

By way of the proposed amendments, ENRC seeks to claim increased borrowing costs for the period 2013-2016, totalling US$91.3m (rising to US$290m including interest). However, it now recognises that only US$12.6m (14%) of that increased borrowing cost related to loans in ENRC’s own name, the remaining US$78.7m relating to debt held at subsidiary level. In particular (i) US$35.8m (39%) related to the indebtedness of ENRC Finance Limited and ENRC NV, themselves 100%-owned holding companies below ENRC in the corporate structure; (ii) US$30.5m (33%) related to the debts of Kazakhstan subsidiaries named SSGPO JSC and TSC Kazchrome JSC, 100% and 99.56% owned by ENRC NV respectively; and (iii) the remaining US$12m (14%) was in relation to the borrowings of five other ENRC subsidiaries. ENRC was a guarantor of much of the debt held by the subsidiaries, so that it was an obligor for over 80% of the total group debt.

The amendments also sought to address the fact that certain costs and fees claimed as a result of the wrongdoings of the SFO and the Dechert defendants were in fact incurred by subsidiaries and were not recharged to ENRC. But the focus of the applications and of this appeal was on the increased borrowing costs, by far the larger claim.

As referred to above, the documents of ENRC were preserved as from 2013 as were those of the two holding companies, ENRC Finance Limited and ENRC NV. Those of the new parent company, ERG SARL, were also preserved from its incorporation in December 2014. The documents of the other borrowers in the group, however, were not subject to a litigation hold.

After setting out the procedural history, the legal principles (as to which there was no dispute below or on this appeal, but to which I will return later in this judgment) and the parties’ positions, the Judge addressed the lateness of the application as follows:

“31.

This is a case where… the matters which are the subject of the amendment could and should have been pleaded very much earlier. ENRC has of course always known that a large proportion of the lending relied upon was made not to it but to its subsidiaries. The claim for increased borrowing costs due to the CI was made in the original 2019 Particulars of Claim. Further, although the CI did not end until 2023, the claim in respect of increased borrowings is in fact limited to the years 2013 to 2016. In reality, ENRC and/or its legal team took their eye off the ball in terms of accurately pleading the increased borrowing costs claim once the parties’ attention had turned to the management of the claim generally and in particular the hiving off of quantum to Phase 2. So it is a late amendment, though not a very late one in the sense of one which jeopardises an already fixed trial date.”

The Judge then addressed the relevance of the subsidiaries’ documents, considering separately “whose documents are relevant to the negotiation of the loans themselves” and “whose documents are relevant to the claim for diminution in value”. It is clear from the context that the Judge used the term “relevant” as shorthand for relevant and disclosable by ENRC.

As for the negotiation of the loans, the Judge considered ENRC’s contention that it was not even clear that disclosable documents of the subsidiaries were no longer available, and that it might be found that they still exist if the amendments are allowed and the position is then investigated. The Judge rejected that argument in the following terms:

“50.

However, all of that is far too speculative. What would have been useful is if ENRC had already checked with the subsidiaries and sought documents relating to the loans, for example by way of keywords, and if documents did emerge, then perhaps ENRC could have argued that the absence of a litigation hold did not matter because there were no responsive documents. However, it has not undertaken that exercise and it is somewhat surprising that it has not.

51.

I should add here that in this context ENRC does not argue that it could not have imposed a litigation hold years ago, or could not now seek documents from the subsidiaries, because such documents were not within its control. While ENRC, in the past, has said that it did not have legal control of such documents, Mr Pillow accepted that the position was or may be different in terms of practical control. At any rate, the control argument was not made in the context of the current application.”

The Judge also considered ENRC’s argument that no documents of the subsidiaries would be disclosable in any event because (i) it was the treasury and finance team at ENRC (and then ERG) which conducted negotiations with lenders on behalf of the entities in the wider group; (ii) it was therefore implausible that, even if representatives of subsidiaries in Kazakhstan were sometimes involved in discussions with lenders, anything significant to a material borrowing arrangement would not have been discussed at a treasury level; (iii) substantial amounts of documents from the treasury team were available; (iv) only documents passing between ENRC/ERG and the lender would be relevant: documents internal to subsidiaries were unlikely to be of any probative value. He then recorded the SFO’s contrary contention that there were many kinds of documents relating to the subsidiaries that would be relevant to the borrowing costs, to which experts considering the cause of increases in those costs would wish to have regard. The Judge recognised the force of ENRC’s response, but ultimately rejected its position in the following terms:

“64.

As to that point, ENRC's response is essentially to say that this cannot be right. It may be appropriate if one was to start if the ground up, as it were, so as to work out what a hypothetical lender to any given subsidiary, absent the CI, would have charged it in terms of interest. However, the exercise here is different, because it is to ask what difference the CI made to the actual lenders in terms of their deliberations. I follow that, but without more detail I cannot say for sure that at least some of the documents sought might not be relevant and it is difficult to be sure where we have not yet had a hearing on the disputed items in the DRD [Disclosure Review Document]. So what I am being asked to do now by ENRC is effectively to decide that dispute in advance for the purpose of the amendment application, which does seem rather back to front to me.”

After making the further points that the subsidiaries’ documents might be relevant in considering interest rates charged to comparator companies and in relation to the question of whether borrowing cost increases might have been mitigated, the Judge concluded as follows:

“67.

In my judgment, and at this stage of the proceedings, and without a detailed debate on DRD issues, I do not think it right to conclude that none of the documents which would have been caught by a litigation hold on subsidiaries could be relevant, nor can I say that, even if they were relevant, they would only be marginally so, or that it would be disproportionate, especially given the amounts claimed here, to have regard to them…”

The Judge then considered the potentially different position in relation to the borrowing of ENRC Finance Limited and ENRC NV, accounting for US$35.8 million, given that all of their documents were preserved. The Judge nonetheless concluded that the amendments in relation to the borrowing of those companies still faced the same problems, stating:

“69… [I]f, as ENRC says, they are both mere holding companies, the documents which would be relevant are those of their indirect subsidiaries, which of course takes us back to Kazchrome, SSGPO and the rest.”

As regards the diminution in value claim, the Judge noted that ENRC’s “dollar for dollar” claim was a stark one, and that ENRC does not put its claim in any other way. The Judge nevertheless rejected ENRC’s contention that that “all or nothing” approach removed the need for any disclosure from the subsidiaries as to the diminution in the value of ENRC’s shareholding in those companies, stating:

“73.

Indeed…part of the expert input…concerned what would be required for a proper assessment of the diminution in value claim. In that regard, even though ENRC has no alternative diminution in value or other case for loss, if the court is to be persuaded that the dollar-for-dollar method is not the correct one, I can see that, in order to challenge it, the SFO and the Dechert Defendants’ experts will need to consider the particular activities, standing and attributes of each subsidiary in order to explain why the dollar-for-dollar approach does not work. For that reason, the lack of documentation now at subsidiary level constitutes also a real prejudice to the Dechert Defendants and the SFO, which could have been avoided if the claim in respect of the subsidiaries’ borrowing had been made at the outset. This particular prejudice is not one which exists for the claim in respect of ENRC’s own borrowing, since by definition there is no diminution of value claim there.

74.

I should add that ENRC somewhat downplays the significance of its amendments on the basis that all the court is doing is making a rough and ready assessment as to whether the CI caused increased borrowing costs and that ENRC’s case here is simple. That, I think, underestimates what in fact would be involved in dealing with these very large claims.”

The Judge therefore concluded that there would be real prejudice to the Dechert defendants and the SFO if the amendments to the borrowing costs claim were allowed. As for the countervailing prejudice to ENRC, the Judge stated:

“76.

On the other hand, and by the same token, ENRC will suffer considerable prejudice if the amendment is not allowed because it will be deprived of most of the very substantial increased borrowing costs claimed. However, here it really is the author of its own misfortune, since it could and should have pleaded this claim properly at the outset, namely, around five years ago. That point is given added weight when one recalls that the SFO actually pressed ENRC for further information on the borrowing costs claim in 2019 and 2020, thereby emphasising the need for proper particularisation. There is yet further weight given to the point by the fact that I said on 13 May 2020 that any amendment should be made then and not later. Here, all of the material facts must have been known by ENRC in 2019, just as now.”

In the light of the above, and given that there was no good reason for the amendments being sought as late as they were, the Judge held that it was clear that those relating to increased borrowing costs must be refused. For the same reasons, the Judge also refused those relating to fees and costs incurred by the subsidiaries and compound interest paid by the subsidiaries.

ENRC challenged the Judge’s decision on five grounds:

As indicated above, the principles applicable to amendments were not contentious and, in so far as relevant to the issues on this appeal, can be found in the following summaries, starting with that undertaken by O’Farrell J in Municipio de Mariana v BHP Group (UK) Limited and others [2024] EWHC 23, adopted by the Judge in his judgment:

“16…On an application by a party to amend its pleading, where there are potential issues of lateness or adverse impact on the trial date, the following principles are applicable, as set out in CIP Properties (AIPT) Ltd v Galliford Try Infrastructure Ltd [2015] EWHC 1345 (TCC) per Coulson J (as he then was) at [19] and Quah Su-Ling v Goldman Sachs International [2015] EWHC 759 per Carr J (as she then was) at [36]-[38]:

(i)

In exercising the court’s discretion whether to allow an amendment, the overriding objective is of the greatest importance. Although the court will have regard to the desirability of determining the real dispute between the parties, it must also deal with the case justly and at proportionate cost, which includes (amongst other things) saving expense, ensuring that the case is dealt with expeditiously and fairly, and allocating to it no more than a fair share of the court’s limited resources.

(ii)

Therefore, such applications always involve the court striking a balance between injustice to the applicant if the amendment is refused, and injustice to the opposing party and other litigants in general, if the amendment is permitted.

(iii)

The starting point is that the proposed amendment must be arguable, coherent and properly particularised. An application to amend will be refused if it is clear that the proposed amendment has no real prospect of success.

(iv)

An amendment is late if it could have been advanced earlier, or involves duplication of steps in the litigation, costs and effort. Lateness is not an absolute, but a relative concept. It depends on a review of the nature of the proposed amendment, the quality of the explanation for its timing, and a fair appreciation of the consequences in terms of work wasted and consequential work to be done.

(v)

It is incumbent on a party seeking the indulgence of the court to be allowed to raise a late claim to provide a good explanation for the delay.

(vi)

A very late amendment is one made when the trial date has been fixed and where permitting the amendment would cause the trial date to be lost. Parties and the court have a legitimate expectation that trial fixtures will be kept.

(vii)

Where a very late application to amend is made the correct approach is not that the amendments ought, in general, to be allowed so that the real dispute between the parties can be adjudicated upon. Rather, a heavy burden lies on a party seeking a very late amendment to show the strength of the new case and why justice to him, his opponent and other court users requires him to be able to pursue it. The risk to a trial date may mean that the lateness of the application to amend will of itself cause the balance to be loaded heavily against the grant of permission.”

The question of “lateness” had also been considered in this Court in ABP Technology Ltd v Voyetra Turtle Beach Inc [2022] EWCA Civ 594, [2022] E.T.M.R 33 by Birss LJ (with whom Coulson and Nicola Davies LJJ agreed):

“23.

When considering whether to exercise the discretion to permit an amendment provided by Part 17.3 of the Civil Procedure Rules, there are several factors to bear in mind. One of these factors is lateness. Coulson J (as he then was) summarised the relevant authorities on ‘lateness’ in CIP Properties v Galliford Try [2015] EWHC 1345 (TCC) and, for the purposes of this case, explained the following key principles at paragraph 19:

“(a)

The lateness by which an amendment is produced is a relative concept… An amendment is late if it could have been advanced earlier, or involves the duplication of cost and effort, or if it requires the resisting party to revisit any of the significant steps in the litigation (such as disclosure or the provision of witness statements and expert's reports) which have been completed by the time of the amendment.

…

(c)

The history of the amendment, together with an explanation for its lateness, is a matter for the amending party and is an important factor in the necessary balancing exercise…. In essence, there must be a good reason for the delay.

…

(f)

Prejudice to the amending party if the amendments are not allowed will, obviously, include its inability to advance its amended case, but that is just one factor to be considered... Moreover, if that prejudice has come about by the amending party’s own conduct, then it is a much less important element of the balancing exercise.” (my emphasis)

24.

The simple point about lateness is that it calls for an explanation justifying the lateness. That is because an amendment which might otherwise be allowed, could well be refused if its lateness has caused unjustifiable prejudice to the other party. Therefore an explanation is needed in order for the court to work out whether or not it is a case in which, despite the prejudice caused by the lateness, nevertheless the balance comes down in favour of allowing the amendment.

25.

Examples of the kinds of prejudice a late amendment might cause were given by Coulson J in CIP Properties at paragraph 19(e): “at one end of the spectrum, the simple fact of being ‘mucked around’ … to the disruption of and additional pressure on their lawyers in the run-up to trial.…, and the duplication of cost and effort…at the other. If allowing the amendments would necessitate the adjournment of the trial, that may be an overwhelming reason to refuse the amendments….”

There appears to be no authority specifically on the question of whether an amendment at the early case management stage of the quantum phase of a split trial should be classified as “late”, but in my judgment the question is readily answered by application of the approach identified by Birss LJ above. Accordingly, if the amendment could have been made earlier (as the amendments in this case clearly could have been), it is properly to be considered “late”. But lateness is a relative concept. Therefore, as with any application to amend, if it is made at an early stage of the relevant phase of the proceedings, before decisions as to what disclosure should be given and before witness statements and expert reports are exchanged, the lateness will far more readily be excused by the Court and it will be harder for the party resisting the amendment to demonstrate that the prejudice it would suffer by the amendment being permitted would outweigh the prejudice occasioned by its refusal.

As regards the approach of an appellate court to appeals against a discretionary or evaluative assessment by a first instance judge, the relevant principles are again to be found in ABP Technology, where the court at [21] approved the following summary by Saini J in Azam v University Hospital Birmingham NHS Foundation Trust [2020] EWHC 3384 (QB):

“50.

An appellate court will only interfere with a discretionary evaluation where an appellant can identify one or more of the follows errors:

(i)

a misdirection in law;

(ii)

some procedural unfairness or irregularity;

(iii)

that the Judge took into account irrelevant matters;

(iv)

that the Judge failed to take account of relevant matters; or

(v)

that the Judge made a decision which was "plainly wrong".

51.

Error type (v)… means a decision which has exceeded the generous ambit within which reasonable disagreement is possible.

52.

...The appellate court’s role is to police a very wide perimeter and it will be rare that a judge who has exercised a discretion having regard to relevant considerations will have come to a conclusion outside that perimeter... It needs to be underlined that an appellate court in an appeal such as the present is exercising a CPR 52.21(1) “review” power. It is also well-established that the weight to be given to specific factors is a matter for the trial judge and absent some wholly unjustifiable attribution of weight, an appellate court must defer to the trial judge.”

It is important to recognise that at the time the Judge heard and decided the amendment applications in January 2025 he had been managing the proceedings for several years and had done so with the conspicuous skill and fairness to be expected of a Commercial Court judge of his considerable experience and expertise. He had deep familiarity with the parties and the issues and fully understood the history of the proceedings and the context of the applications, with the result that his case management orders demand even more respect than that customarily accorded to such decisions. Nevertheless, I am satisfied that, in this one instance, the Judge erred in his approach, causing him to reach the wrong decision.

The Judge’s approach was that, once it was determined that the amendments were late and that there was no good reason for the delay, it was for ENRC to demonstrate that there was no possibility that any documents relevant to the issues raised by the amendment had ceased to be available. As ENRC failed in that task, the Judge regarded the risk of non-availability he had found to exist as necessarily seriously prejudicial to the SFO and the Dechert defendants. He further regarded such prejudice as outweighing the prejudice to ENRC in losing a very substantial part of its loss claim because ENRC was the author of its own misfortune.

The result of that approach is that ENRC’s amendments were refused even though it is perfectly possible, on the Judge’s analysis, that (i) he might have determined in due course (had the amendments been allowed) that no or only minimal categories of documents were disclosable, or (ii) any disclosable documents might have been found still to exist or to have been lost prior to the date a litigation hold should have been put in place or (iii) any disclosable documents that were lost due to the delay might not have resulted in prejudice to the defendants of a serious nature or at all. Further, the above aspects could have been assessed by the Judge, had he allowed the amendments, at latter stages of the proceedings, including at trial, and the amended claims struck out if a fair trial was not possible due to the non-retention of documents.

In my judgment the approach the Judge adopted was wrong in a number of respects, but the overarching point is that it regarded an entirely uncertain and unquantified risk of injustice to the defendants as outweighing the certain and substantial injustice to ENRC, notwithstanding that it was unnecessary to reach a final view on that balancing exercise given the very early stages of the Phase 2 process in which the issue arose. That was not, in my view, consistent with the overriding objective of determining the real dispute between the parties and was not necessitated by any countervailing considerations of proportionality, efficiency or expedition.

In my judgment the starting point should have been to consider the nature, extent and timing of the amendments proposed. A claimant which had proved very serious wrongdoing on the part of the defendants was now, at the pre-CMC stage of the quantum phase, seeking to re-formulate its claim for very substantial losses which had always been pleaded (amounting to some US$128 million including interest) due to a late realisation as to which companies in a group had incurred them. Given that the re-formulated loss claim passed the prospects and cogency tests, to refuse such an amendment would be hugely prejudicial to the claimant.

I do not consider that it was appropriate significantly to discount the weight to be given to the prejudice to ENRC in rejection of the amendments by reason of the timing of the applications. Whilst they were technically “late” in the sense discussed above (and ground 5 of ENRC’s appeal therefore fails) and no good reason could be advanced for the delay, they were made at the outset of the quantum phase, at a time when issues as to disclosure were yet to be determined. The Judge placed great store in the fact that ENRC resisted the defendants’ request for further information of the increased borrowing costs claim in the early stages of the proceedings, but no order was made requiring the provision of such information and all quantum issues were “put off” to Phase 2. The Judge also stressed that he had stated in 2020 that any amendments to the pleading should be made at that time, repeating that when giving preliminary directions on Phase 2 in March 2024, but again such indications were informal and in no way debarred ENRC from applying to amend at the start of Phase 2. In my judgment the Judge’s understandable view that problems would have been avoided had ENRC focussed on and corrected its loss pleading when pressed by the defendants and invited by the Judge himself, overly affected his view as to the timing of the amendments when made and the lack of sympathy he afforded ENRC in that regard.

The second stage was to consider whether the evidence demonstrated that permitting the amendments would cause injustice to the defendants that outweighed the substantial injustice to ENRC if the amendments were rejected. As it was suggested such prejudice arose from the fact that documents might not be available by reason of the delay in making the applications to amend, in my judgment the following questions needed to be addressed:

It is apparent that the Judge did not reach a concluded view on the first question, but based his decision on an assessment that he could not rule out the possibility that there were disclosable documents. In relation to the second question, the Judge did not consider at all the fact that some or all unavailable documents (if any) may have been lost or destroyed in the period between 2013 and the date when the applications should have been made. The Judge did not expressly address when that date would have been, although it is clear that he considered that ENRC should have revised its pleading soon after his instruction to consider amendments in 2020. That still means that documents might have been lost in the seven prior years, regardless of the criticism levelled at ENRC for lateness.

It is also apparent that the Judge did not expressly address the third or fourth questions, save to hold that the risk of prejudice to the defendants which he perceived outweighed the injustice to ENRC due to the fact that ENRC was the author of its own misfortune.

Had the Judge approached the question of the balance of injustice in stages as set out above (whether or not in that order), it would have been clear, in my judgment, that he did not have any or any sufficient evidence at that time of prejudice to the defendants capable of outweighing that to ENRC by disallowing the amendments. The proper approach would have been to allow the amendments, but to keep under review, if necessary, whether any subsequent failure to provide disclosure of the subsidiaries’ documents rendered a fair trial of the issues impossible, notwithstanding the possibility of making allowances or drawing adverse inferences at trial: see Republic of Mozambique v Credit Suisse International [2023] EWHC 1650 (Comm).

I would add that the Judge’s approach effectively regards a failure to place a litigation hold on documents as being seriously prejudicial in itself to the opposing party. Whereas I fully accept that a party must take steps to preserve documents when litigation is in prospect (and the same is an express requirement of the rules, see PD 31B paragraph 7 and PD 57AD paragraph 3) a failure to do so does not carry any automatic debarring sanction and does not in itself cause injustice unless and until it is established that disclosable documents have not been preserved with prejudicial effects. The Judge’s approach elevates a formal requirement over the substantive question of its effect on the parties and the justice of granting or refusing an application to amend.

It follows that I regard the criticisms of the Judge’s approach in grounds 1 and 2 of ENRC’s appeal as made out and as justifying allowing the appeal. In those circumstances I do not consider that it is necessary to address ground 3 or ground 4.

I concurred in the decision to allow the appeal for the reasons that have been given by Phillips LJ.

I also concurred for those reasons.