A reminder of the sheer size, and major dangers, of group litigation can be seen in the judgment today in Sharp & Ors v Blank & Ors [2020] EWHC 1870 (Ch).  The judgment relates to the costs of the action where the claimants were liable to pay the defendants’ costs in excess of £30 million.  The judge considered issues relating to the liability to pay, the position of the individual claimants and a litigation funder, interest on costs and an interim payment on costs.


“… there is the theoretical possibility that an aggregation of the assessed reasonable and proportionate incurred costs and the reasonable and proportionate budgeted costs results in a total that is itself beyond what is reasonable and proportionate (to a degree that provides a “good reason” within CPR3.18) But in truth these possibilities are remote and would be adequately covered by a discount of 10% on the approved costs. In MacInnes (supra) Coulson J said (at [28]) that he thought that such a discount was the maximum deduction appropriate in a case where there is was an approved costs budget. Although the matter is always one for the exercise of individual discretion in the particular circumstances of the case, I find that guidance helpful.”


In an earlier judgment Sir Alastair Norris had dismissed the claimants’ action for damages against individual directors of Lloyds TSB Group. This judgment is concerned with the costs of that action.


The claimants argued that, because they had “succeeded” on some issues they should only pay 65% of the costs. This argument was rejected by the judge.

    1. By my judgment (the reference to which is [2019] EWHC 3096 (Ch)) I dismissed the claims of the Claimants listed in the GLO Register (“the Claimants”). The Defendants submit that this is a case for the straightforward application of the “general rule” in CPR 44.2(2)(a) and seek an order that Claimants pay their costs of and incidental to the action, to be assessed (in default of agreement) on the standard basis. The Claimants submit that this is a case for making “a different order” under CPR 44.2(2)(b) and argue for an order that they pay only 65% of the Defendants’ costs.
    2. The foundation of that argument is that although the Claimants failed overall they did succeed on two issues in the case; and that CPR 44.2(4)(b) directs the Court to take into account in deciding what order to make about costs whether (amongst other things) a party has succeeded in part of his or her case even though not wholly successful.
    3. I do not accept the argument of the Claimants;

(a) It is a commonplace that a successful party will not succeed on every aspect of its case. But notwithstanding that very frequent occurrence in litigation, the general rule still applies. Costs are determined by reference to overall success.

(b) Although no authority is needed to support that observation, the point was pithily summarised by Gloster J. in HPL Kidson’s v Lloyds Underwriters [2008] 3 Costs LR 427 at [11].

(c) A degree of caution is needed against a too-ready departure from the general rule for the reasons explained by Jackson LJ in Fox v Foundation Piling [2011] EWCA 790 at [62].

(d) There is no reason in principle why a party who succeeds in establishing one element of his cause of action but fails to establish the others should be regarded as partially successful. The Claimants established that the Defendant directors were in breach of a duty of care in respect of statements and in breach of an equitable duty of disclosure (because the directors did not cause to be noted in the Circular the existence of a specific Bank of England support facility for HBOS or the existence of the facility covered by the Lloyds Repo). But the Claimants failed to establish that such a state of disclosure was causative of any loss, nor did they establish the amount of any loss in fact suffered.

(e) The Claimants submitted that although they only established breach of duty in relation to two items of disclosure, those allegations lay at the heart of their case. That may happen to be so from their perspective, but it certainly was not so from my perspective. At trial the Claimants attacked the Transaction across an extremely broad front (even after they had abandoned significant parts of their pleaded case pre-trial). They did not abandon their “recommendation” case, so that had to be fought out and adjudicated (and required the bulk of the evidential review and fact-finding): and that outcome had an effect upon the “disclosure” case. They made a multi-faceted case about disclosure; including cases of deliberate concealment, of duties of care in relation to market announcements and briefings, of misstatements about the value of HBOS and about the amount of capital a stand-alone Lloyds would have to raise, and of the obligation of directors to disclose every piece of information they themselves had considered. None of this succeeded. Given the breadth of the attack the degree of success was small.

(f) The Claimants submitted that the Defendants should have conceded the case on the disclosure duty (by which I think they mean “the part of the disclosure case that ultimately succeeded”) and that by so doing costs would have been saved. But even the measure of success achieved by the Claimants was so achieved on a fine balance: as paragraph [855] of my judgment and its surrounding paragraphs seek to make clear.

(g) It is, of course, not the law that a successful party can only be deprived of the costs of an issue if he has unreasonably resisted that issue: any more than it is the law that he should be deprived of the costs of the issue simply because he lost it. In singling out an issue for separate treatment by way of costs I think the court must look for some objective ground (other than failure itself) which alongside failure distinguishes it from other issues and causes the general rule to be disapplied. In F & C Alternative Investments Holdings v Barthelemy Davis LJ described it as “a reason based on justice” ([2013] 1 WLR 548 at [47]). Without seeking in any way to be prescriptive, I think the distinctive ground may relate (amongst other things) to the comparative weakness of an argument (even if not unreasonably maintained) or to the necessity for particular evidence relevant only to that issue or to extensive and intensive legal argument directed to that issue which gives it an especial significance in the costs context. But such a factor is missing here. The law both on “misstatement” and “sufficient information” was all but agreed. The Claimants failed to establish the deliberate concealment they alleged. They failed to establish the disclosure duties for which they contended. They succeeded in establishing a breach of the conceded duty by demonstrating that the Defendants had not correctly assessed the materiality of two matters. The key issue of “materiality” was determined by reference to evidence (about the nature of ELA and the Lloyds Repo and what resort to ELA and to the Lloyds Repo told one about HBOS as part of the Enlarged Group) adduced in relation to the “recommendation case”.

  1. This is a case in which the general rule should apply, and costs should follow the event. But that then raises the question of what this means for individual Claimants. It may well be that many of the 5800 Claimants never foresaw this as a real question because they thought that they were litigating risk-free. But most unfortunately that is not the case.


There were 5,800 claimants litigating under a Group Litigation Order.  The order stated that the claimants were severally liable.  They may, however, have been under-insured.

  1. The Claimants (or at least those who were Claimants as at 14 January 2015) have primary ATE cover of £6.5 million in respect of any legal obligation to pay costs to the Defendants under an order of the Court. The costs claim of the Defendants exceeds £30 million. For the excess over the £6.5 million primary cover the Claimants are reliant on the provisions of a Deed of Indemnity granted in their favour by Therium. The Claimants’ openly stated position has always been that there are no concerns over Therium’s ability to satisfy that indemnity (or indeed any direct liability in costs that might be owed to the Defendants by Therium). In part that derives from the fact that Therium has itself effected excess layers insurance in respect of its liability under the Deed of Indemnity to any Claimant against whom an adverse costs order is made. This excess layers insurance totals a further £14.95 million (bringing the aggregate insurance cover available to the Claimants and to Therium to £21.45 million: still short of the Defendants costs claim). Even this level of cover is affected by the insolvency of some of the insurers. In these circumstances there is a (most regrettable) risk that individual Claimants may face a several liability for costs to the extent that it overtops their direct ATE cover.
  2. The Defendants seek to smooth the way for this. They seek to define the class of liable Claimants simply as those on the GLO register at the conclusion of the hearing (though they would be equally content with those on the register at the commencement of the trial). Ms Davies QC submits that this would avoid undertaking the “unreal exercise” (as she described it) contemplated by paragraph 10(4) of the GLO (apportioning the costs incurred by the Defendants each quarter on common issues amongst the Claimants on the GLO register at the commencement of that quarter by reference to the size of each such Claimant’s claim as a proportion of the aggregate claims at that date).
  3. I do not accept this submission. I do not think it is possible as part of the process of settling the terms of the order after trial simply to apportion the costs liability amongst a class of Claimants as at a single date. I agree with Ms Davies QC that the task of identifying the costs of the Defendants quarter by quarter, and then of calculating the several liability of each Claimant for that quarter’s costs according to the ratio which his claim bore to the aggregate value of claims in that quarter, is very burdensome; and if the nature and extent of the Claimants’ record-keeping makes it impossible to complete, then the terms of the GLO may have to be revisited. But until that point Mr Hill QC is right in his submission that the terms of the costs order made by me cannot change the liability of the Claimants amongst themselves and must reflect the terms of paragraph 10(4) of the GLO.
  4. The Claimants will therefore between them severally bear the costs of the Defendants (the Defendants’ costs to be the subject of assessment on the standard basis in default of agreement), each several liability for a share of those costs being ascertained in accordance with the terms of paragraph 10(4) of the Order dated 6 August 2014.
  5. It is, I think, premature to order an inquiry into what is the several liability for a share of the costs borne by each Claimant. But the order should contain a permission for either party to apply for such an inquiry.


  1. For the Defendants Ms Davies QC argued for a payment of £18,750,00 being (i) 50% of the Defendants’ pre-budget incurred costs of £9,705,570; (ii) 100% of the Defendants’ budgeted costs of £10,903,627; and (iii) irrecoverable VAT at 18.9% (roughly £3m). As I have indicated above, the total costs claim of the Defendants is some £30m, because of increases in budgeted costs arising from a lengthening of the trial and other factors. These factors the Defendants will pray in aid of an increase in approved costs at the assessment (under the provisions of CPR 3.18): but they accept that budgeted costs are what must guide an order for interim costs, although they ask me to stand back and look at the interim costs sought in the context of the total costs claimed.
  2. For the Claimants Mr Hill QC accepts items (i) and (iii) but contests item (ii). A payment of 100% of the approved costs is not, he submits, “a payment on account” of such costs. He argued for a payment on account of 80% of the budgeted costs. The reason advanced for the discount was that the incurred costs had still to be the subject of assessment in the traditional way, and the aggregate of the assessed incurred costs and the budgeted costs then measured for proportionality. This might lead to a re-assessment of the proportionality of the overall costs, a real issue notwithstanding that the Claimants’ damages claim was of the order of £385m.
  3. The task in hand is to assess “a reasonable sum on account of costs”: this will not exceed an estimate of the likely level of recovery allowing for an appropriate margin of estimation error. For incurred costs in the instant case that margin has been covered by a discount of 50% on the claimed costs. For budgeted costs a detailed estimation process has already been undertaken to reach a figure which the Court considers both reasonable and proportionate for the Defendants to expend. That affords a large amount of certainty as to what the likely costs recovery will be for those phases of the litigation. That is why Coulson J in MacInnes v Gross [2017] 4 WLR 49 at [25] (in a passage approved by the Court of Appeal in Harrison v University Hospitals Coventry & Warwickshire NHS Trust [2017] 1 WLR 4456 at [40]) said that almost always the approved costs budget provided the starting point for assessing a payment on account, because the costs management process had established it to be a reasonable and proportionate sum.
  4. The outcome of the costs management process does not completely eliminate the possibility of an estimation error in the “payment on account” process. First, there is always the possibility that the paying party may on assessment establish that there is a “good reason” (sufficient for the purposes of CPR 3.18) ultimately to assess some costs below the figure approved in the costs budget. Second, there is the theoretical possibility that an aggregation of the assessed reasonable and proportionate incurred costs and the reasonable and proportionate budgeted costs results in a total that is itself beyond what is reasonable and proportionate (to a degree that provides a “good reason” within CPR3.18) But in truth these possibilities are remote and would be adequately covered by a discount of 10% on the approved costs. In MacInnes (supra) Coulson J said (at [28]) that he thought that such a discount was the maximum deduction appropriate in a case where there is was an approved costs budget. Although the matter is always one for the exercise of individual discretion in the particular circumstances of the case, I find that guidance helpful.
  5. In my judgment a reasonable sum to award by way of interim costs is £17m. (being half of the claimed incurred costs, plus 90% of the budgeted costs, plus VAT thereon at 18.9%, rounded down in the Claimants’ favour). That sum will fall due for payment 42 days after the handing down of this judgment.


  1. Two issues arose in relation to interest on costs. First, whether the Court should award pre-judgment interest on costs. Second, from what date should the Court order that interest at judgment debt rate should run?
  2. CPR44.2(6)(g) provides that the Court may order interest on costs from a date before judgment. It is common ground that this jurisdiction will generally be exercised where a party has had to put up money to pay its solicitors and has thereby either lost the beneficial use of that money or has had to borrow it. It is also common ground that the rate awarded will be determined by weighing the factors identified by Sharp LJ in Jones v SoS for Energy and Climate Change [2014] EWCA Civ 363 at [17]-[18], generally with the aim of identifying a commercial rate in the circumstances. In the instant case there is no argument about the rate. The Claimants submit (and the Defendants accept) that, if ordered for any period, the appropriate rate is base rate simpliciter. The issue is whether interest should be awarded at all.
  3. For the Claimants Mr Hill QC submits that it should not. His argument was that the Claimants obtained (in the face of opposition from the Defendants) a costs management order so that they could secure an accurate assessment of the costs risks that they faced and could compare it with the level of ATE insurance available to them. The Defendants did not signal an intention to claim pre-judgment interest on costs at any time when costs budgets were under consideration, and a possible claim for interest was not factored in to the level of ATE cover obtained. The figure at which Defendants invited the Claimants to secure insurance cover (with which request they broadly complied) did not include any element of interest at costs.
  4. For the Defendants Ms Davies QC submitted (i) that interest on costs does not form an element of recoverable costs for the purpose of costs budgeting so that it is not surprising that it was not mentioned in the costs budget debates; (ii) there had been no representation that interest would not be claimed; (iii) no-one could during the costs budgeting process know for what period or at what rate interest would run and it was for the Claimants to assess their exposure on that account and (if desired) cover it; (iv) there was no justice in the current shareholders in Lloyds being deprived of compensation for the loss of the beneficial use of their money simply to benefit a group of past shareholders.
  5. I accept that the Defendants had a commercial interest in the level of ATE cover obtained by the Claimants because of the great difficulties attending costs recovery under the GLO. But it was ultimately for the Claimants to decide against what risks to insure and what risks to bear themselves. A claim for pre-judgment interest on costs is commonplace, and it was for the Claimants to decide whether any protective measures were required, not for the Defendants to call for them. I shall exercise the discretion in the way in which it is customarily exercised and order the Claimants to pay interest on the Defendants’ costs at the applicable Bank of England base rate from the date of payment of each invoice until the earlier of (i) payment of such costs or (ii) the date from which interest at the rate prescribed by the Judgments Act 1838 become payable.
  6. This brings me to the second issue: from what date will the judgment debt rate apply? CPR 40.8 provides interest shall run from the date of judgment unless the rules make a different provision or “the Court orders otherwise”. The Defendants submit that the general rule (or what they would call “the default position”) should apply, whilst the Claimants submit that the Court should order “otherwise” and direct that interest at Judgment Act rate (viz. 8%) should run from a date 6 months after judgment.
  7. The essential question is: what, having regard to all the circumstances of the case, do the interests of justice require? That is the distillation of the thorough analysis undertaken by Leggat J (as he then was) in Involnert v Apigrange [2015] EWHC 2834 (Comm) and summarised at paragraph [20]. Normally the interests of justice will indicate that the rate of interest applicable in the event of non-payment of a judgment debt should only run from the time when (i) the amount of that debt is known (or can at least be the subject of fair estimation for the purposes of offers of compromise) and (ii) the essential mechanics of payment can be completed.
  8. Given (i) the likely size and complexity of the bill to be submitted for assessment; (ii) the complications inherent in triggering the insurance and indemnity arrangements which it was in the interests of both Claimants and Defendants should be put in place (because of the terms of the GLO) and the uncertainties that have since arisen; and (iii) the fact that, pending payment, the Defendants are receiving interest on unpaid costs at a commercial rate, in my judgment a period of about four months from the date of this judgment should elapse before interest at judgment debt rate is payable on unpaid costs. 13 November 2020 is a convenient date.


  1. By order dated 13 January 2020 Therium was joined as an additional party for the purposes of costs. Therium accepts, in principle, that as a commercial funder it is liable to pay costs awarded against the Claimants. But it submits that it should be so liable only (i) to the extent that the Claimants do not satisfy the adverse order and (ii) to the extent of the funding that Therium actually provided (i.e. subject to “the Arkin cap”: see Arkin v Borchard Lines Ltd [2005] 1 WLR 3055 at paras [38]-[43]).
  2. To take the second suggested limitation first: as at 29 January 2020 the true nature of “the Arkin cap” was due for consideration by the Court of Appeal the following month in Davey v Money. For Therium Mr Marven QC submitted that the appropriate course was to adjourn consideration of the extent of its liability to await the decision of the Court of Appeal. For the Defendants Ms Davies QC submitted that the correct course was for me to make an order without the restraints imposed by “the Arkin cap” (following the first instance decision of Snowden J in Davey v Money [2019] Costs LR 39) but to give permission to apply once the outcome of Davey v Money in the Court of Appeal was known.
  3. The Court of Appeal’s decision in Davey v Money (reported at [2020] 1 WLR 1751) was delivered on 25 February 2020. I have considered its terms. In essence it holds that “the Arkin cap” is not a binding rule, but simply guidance given to individual judges, who retain a complete discretion in relation to third party costs orders: and whilst the extent of the third party’s investment may be a factor to be weighed, so also might be the potential return of the third party funder from a successful costs investment (along with other factors). I summarise paragraph [38] of the Court of Appeal’s judgment and refer to the factors which influenced the disapplication of “the Arkin cap” recorded at paragraph [19] of that decision.
  4. I do not know enough about the detail of the funding arrangements effected by the Claimants with Therium properly to exercise the discretion in relation to the entirety of the Defendants’ costs claim (even if I were to do so on a provisional basis and grant permission to apply). But I do know enough to conduct a limited exercise.
  5. I do know that Therium funded the Claimants’ costs of some £17m together with the premiums on the excess layers insurance above £6.5m (up to £21.45m). It is therefore possible to say that even if “the Arkin cap” were to be applied the amount of the interim payment on account of costs ordered above would fall below that cap. I can therefore safely make an order on that basis, with a “permission to apply” as a “failsafe”. The extent of Therium’s liability beyond that must be adjourned for further consideration. If the parties (who, as commercial entities, I would expect to engage in an alternative dispute resolution process) cannot resolve the issue and cannot agree directions for a resolution by the Court then I (or another judge) will give directions so that the matter can be resolved.
  6. I can now revert to the first issue. It is accepted that a third-party costs order is in principle appropriate. I see no reason in principle why the liability of Therium (which has indemnified the Claimants) should be secondary and not simply joint and several in the usual way. Mr Marven QC accepted that if I made an order for an interim payment of costs in (say) 21 days which the Claimants did not satisfy, then Therium’s liability would be triggered (without the Defendants having to seek recovery of a several share from each Claimant). But such “default” on the part of the Claimants would seem to result from a failure on the part of Therium to honour its indemnity to them. I cannot see what is achieved by postponing any liability of Therium direct to the Defendants for 21 days. Therium’s liability should be joint and several.