England and Wales High Court (Chancery Division) Decisions

IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
BUSINESS LIST (ChD)

		Royal Courts of Justice, Rolls Building Fetter Lane, London, EC4A 1NL
		26/1/2022

Master Clark:

1. This judgment deals with the consequential matters arising from my judgment dated 5 January 2022:

(1) the amount of costs to be ordered as security for the defendants' costs of the jurisdiction applications;

(2) the liability for and amount of the costs of the security applications.

The parties have agreed that, if possible, I should determine these matters on written submissions, without a hearing.

Amount of security

Principles

2. The relevant principles as to quantifying security are set out in the 2021 White Book at 25.12.7, and summarised in Pisante v Logothetis [2020] EWHC 3332 (Comm), [2020] Costs LR 1815:

"(i) The appropriate quantum is a matter for the court's discretion, the overall question being what is just in all the circumstances of the case. In approaching the exercise, the court will not attempt to conduct an exercise similar to a detailed assessment, but will instead approach the evidence as to the amount of costs which will be incurred on a robust basis and applying a broad brush (see also Excalibur Ventures v Texas Keystone [2012] EWHC 975 (QB) § 15).

(ii) In some cases, the court may apply an overall percentage discount to a schedule of costs having regard to (a) the uncertainties of litigation, including the possibility of early settlement and (b) the fact that the costs estimate prepared for the application may well include some detailed items which the claimant could later successfully challenge on a detailed assessment between litigants. There is no hard and fast rule as to the percentage discount to apply. Each case has to be decided upon its own circumstances and it is not always appropriate to make any discount.

(iii) In deciding the amount of security to award, the court may take into account the "balance of prejudice" as it is sometimes called: a comparison between the harm the applicant would suffer if too little security is given and the harm the claimant would suffer if the amount secured is too high. The balance usually favours the applicant: an under-secured applicant will be unable to recover the balance of the costs which is unsecured whereas, if the applicant is not subsequently awarded costs, or if too much security is given, the claimant may suffer only the cost of having to put up security, or the excess amount of security, as the case may be (see also Excalibur § 18).

…

(v) In determining the amount of security, the court must take into account the amount that the respondent is likely to be able to raise. The court should not normally make continuation of their claim dependent upon a condition which it is impossible for them to fulfil."

Application of the principles

D15/16's costs

3. D15/16 have prepared a costs budget for their costs of the jurisdiction applications. The amount shown in that budget (excluding the costs of the security application) is £155,419.24, and the full amount is sought. The claimant has offered security said to be valued at £75,000 i.e. just under 50% of those costs.

D2-12's costs

4. The sum sought by D2-12 is £175,000, representing about 88% of their total estimated costs for their jurisdiction application of £199,372.96, set out in a signed statement of costs dated 11 January 2022. The claimant has offered security said to be valued £100,000, which is about 50% of D2-12's total estimated costs.

Issues as to amount

5. The issues of principle which arise between the parties are:

(1) whether the starting point should be that no reduction will be made on assessment;

(2) whether the court should consider and take into account the likelihood that costs would be awarded on the indemnity basis;

(3) the extent to which the court should take into account the Guideline Hourly Rates applicable to summary assessment.

Whether starting point should be that there will be no reduction on assessment

6. In support of their submission that the starting point should be that their costs will be assessed in the amount claimed, D15/16 rely on a passage (shown in italics) in Danilina v Chernukhin [2018] EWCA Civ 1802, [2019] 1 WLR 758 at [64]:

"once it has been established that there are "substantial obstacles" sufficient to create a real risk of non-enforcement, the starting point is that the defendant should have security for the entirety of the costs and there is no room for discounting the security figure by grading the risk using a sliding scale approach."

7. However, as this passage (and [57], also relied upon by D15/16) shows, the Court of Appeal in Danilina was not concerned with the likely amount recovered on assessment. In that case, security had been granted under the non-residence condition (CPR 24.13(2)(a)) and the Judge had applied a discount to the amount awarded to reflect a "sliding scale" of risk of non-enforcement. The Court of Appeal held that this was wrong. By contrast, security in this claim was awarded on the impecuniosity condition. The claimant does not (and could not) contend that D15/16's costs should be discounted to reflect a sliding scale of risk of enforcement.
8. Accordingly, in my judgment, Danilina is not authority for the proposition of the breadth put forward by D15/16, and is not relevant to this application.
9. The appropriate approach in this application is, as stated by Gross J (as he was) in Texuna International Ltd v Cairn Energy plc [2004] EWHC 1102 (Comm), at [23(vi)]:

"as in this case a claimant is impecunious, this provides 'an objective justification for the court exercising its discretion to make an order for payment of the full amount of the costs likely to be ordered against a claimant if unsuccessful in the litigation'"

10. This requires an evaluation of the amount that the successful applicants are likely to recover, on a detailed assessment, taking into account the various uncertainties identified in paragraph 2 above.

Relevance of prospect of indemnity costs

11. D15/D16 submitted that if they succeeded, then they would be entitled to costs on the indemnity basis on two grounds, derived from the summary by Tomlinson J in his costs judgment in Three Rivers District Council v Bank of England [2006] EWHC 816 (Comm) at [25]:

(1) the claim is speculative, weak, opportunistic or thin;

(2) where the unsuccessful allegations are the subject of extensive publicity, especially where it has been courted by the unsuccessful claimant.

12. Similarly, D2-12 submitted that security on the indemnity basis should be awarded where there was a "real prospect" of costs being awarded on the indemnity basis: relying on Phones 4U Ltd (In Administration) v EE Ltd [2020 ] EWHC 1943 (Ch) [2020] Costs LR 1065, Roth J at [31] to [35]. They submitted that the novel (and they say unmeritorious and misconceived) basis of the claim, meant that there was such a real prospect.
13. There are two first instance decisions in which security was granted by reference to the indemnity basis: Danilina and Re Ingenious Litigation [2020] EWHC 235 (Ch). As Roth J observed in Phones4U, it is necessary to consider the factual context of the claims:

"In Danilina, the claims were heavily dependent on the claimant's evidence, and Teare J held that if the claims failed there was a real possibility, if not a probability, that this was because the court found that she was being dishonest. Ingenious was very different in that in those proceedings the core allegation against the defendants was that they made fraudulent misrepresentations about the tax schemes they were promoting. It was for that reason that Nugee J felt that, should the court reject those allegations at trial, that could lead to an award of indemnity costs."

14. Contrary to D2-D12's submissions, Roth J expressly rejected (at [23]) the proposition that Danilina and Ingenious express a general test for a higher level security as being simply that there is "a real possibility" or "reasonable possibility" of an ultimate award of indemnity costs. He also noted that the Court of Appeal decision of Stokors SA v IG Markets Ltd [2012] EWCA Civ 1706 was not cited in either Danilina or Ingenious.
15. In Stokors, the defendants sought an order for security at a high percentage of their costs on the basis that the claimants were making serious allegations of dishonest conduct against professional men, which could have the outcome of ruining their careers. Upholding the judge's refusal to order security at that higher level, Tomlinson LJ (as noted, the first instance judge in Three Rivers), in a judgment with which Munby and Lewison LJJ agreed, observed (at [42]):

"The only basis for seeking 80 per cent of the estimate to which the judge referred in his judgment was the suggestion that the defendant might recover indemnity costs. For what it is worth, I have never heard of security for costs being awarded on a more generous basis for that reason,"

16. He continued

"but in any event the judge here decisively rejected that suggestion at paragraph 31 of his judgment. He concluded that there was no material before him which would justify such an approach. Rightly, as I think, he declined to go into the merits of the case when considering what was the appropriate quantum of an order for security for costs ..."

17. In this case, in accordance with established principle^[1], I have not reached a conclusion as to the merits of the claim, nor as to whether it is brought in good faith.
18. However, the approach urged by the defendants would require me to consider the merits of the claim, without having heard any detailed argument on them, and where the judge hearing the Jurisdiction Applications will be required to carry out that task. It would in my judgment be wrong in principle for me to do so. Whether, if the defendants succeed, they will be entitled to indemnity costs will depend upon the basis of that success, and will be a matter for the Judge.
19. I am not willing therefore to determine the amount of security to be ordered in the defendants' favour on the basis that it would be assessed on the indemnity basis.
20. In addition, I am fortified in this conclusion by the statement by Etherton C, with whose judgment Richards and Patten LJJ agreed, when reversing the decision of a judge to award indemnity costs in Arcadia Group Brands Ltd v Visa Inc [2015] EWCA Civ 883 , at [83]:

"The weakness of a legal argument is not, without more, justification for an indemnity basis of costs, which is in its nature penal. The position might be different if proceedings or steps taken within them are not only based on a plainly hopeless case but are motivated by some ulterior commercial or personal purpose or otherwise for purely tactical reasons unconnected with any real belief in their merit."

Relevance of summary assessment guidelines

21. D15/D16 submitted that the guideline rates were inapplicable where, as here, the hearing is sufficiently lengthy that detailed assessment will be ordered. However, the recently updated Summary Assessment Guide (2021 White Book, 3^rd supplement, p117) recognises that they "may also be a helpful starting guide on detailed assessment".
22. In a case such as this, the following factors mean that, in my judgment, the guidelines rates are of limited assistance: the value of the claim is well over $4 billion; it involves considerable factual, technical and legal complexity; it has an international element; and has important implications for the Bitcoin system as a whole. This favours a higher level of rate than the guidelines.

Amount of security to be ordered: discussion and conclusions

23. The claimant makes a number of criticisms of the defendants' costs. Applying Pisante, I approach the evidence as to the amount of the defendants' costs on a robust basis and applying a broad brush.
24. As to D2-12, the primary criticism is excessive use of grade A time, including the use of 3 grade A fee earners. As D2-12 point out, one of these only contributes 0.9 hours. I do not consider use of 2 grade A fee earners in a claim of this value and complexity to be plainly unreasonable or disproportionate, and the extent of that use is a matter for detailed assessment.
25. As to D15/16, the claimant's criticisms can be summarised as follows

(1) excessive hourly rates;

(2) tendency to significantly overestimate costs, since their estimated costs for their security application were £66,000, and only £53,870 is now being claimed;

(3) excessive estimated costs where incurred costs are already substantial;

(4) incurred costs on Issue/Statements of Case (£25,313) are excessive where the particulars of claim were only required to be considered for the purpose of the jurisdiction application and no work on the defence should have been undertaken.

26. As to (2), this is of no relevance to the question of the likely amount recoverable on a detailed assessment of the costs of the jurisdiction application. (3) and (4) are matters for the detailed assessment itself, and consideration of them is not in my judgment required by the robust, broad brush approach with which this task should be conducted.
27. As to (1), the criticism of D15/16's solicitors' hourly rates, and whether they justify any reductions of themselves, I set out those hourly rates, together with guideline rates (and, for comparison, D2-12's solicitors' rates).

Fee earner grade	Guideline rate	D15/16	D2-12
A	512	790 ($1,100)	590.76
	512	288 ($400)	599.25
	512		442
B	348
C	270
D	186	288 ($400)	216.75
	186		174.25
	186		178.50

28. Although one of D15/16's grade A fee earners has a rate well above the guideline rate, the other grade A earner is well below, and their average combined rate is just below the guideline rate. Furthermore, because the estimated time to be spent of the higher rate grade A earner is half that of the lower rate earner, the blended rate for their estimated costs is £393 (and the position is approximately the same for the incurred costs).
29. I do not therefore consider that any reduction in the amount of security to be ordered is justified by the hourly rates of D15/16's solicitors.
30. Having concluded that this is not a case in which security should be awarded on the basis that indemnity costs will be awarded, and there being no approved costs budget by reference to which the award can be made, I turn the question of the appropriate discount.
31. In Danilina at first instance ([2018] EWHC 2503 (Comm) at [17]), it was said that the usual order was for 60-70% of the incurred and expected costs. In Ingenious (at [101]), Nugee J (as he was) expressed the view that "65% or 2/3 is more typical", though he in fact awarded 70%. Having rejected the claimant's criticisms of D15/16's solicitors' hourly rates, I consider that the appropriate discount is 30% of the defendants' costs, so that security for 70% of the defendants' incurred and estimated costs for the jurisdiction applications should be ordered:

D2-D12: £199,372.96 x 70% = £139,561.07

D15/16: £155,419.24 x 70% = £108,793.46

Manner of security

The parties' proposals

32. D2-12 seek security by way of payment into court or money held by the claimant's solicitors to the order of the Court.
33. D15/16 seek security by way of payment into court, or bank guarantee given by a

reputable first-class bank in London on reasonably acceptable terms.

34. The claimant has proposed in its written submissions, and in a proposed draft order, that it provide security by way of digital assets, namely either Bitcoin Satoshi Vision or (if considered more acceptable) Bitcoin Core, by

(1) transferring to its solicitors Bitcoin to the value of the security ordered plus a 10% "buffer" ("the Bitcoin");

(2) instructing its solicitors to provide to the defendants' solicitors

(i) written confirmation that it holds the Bitcoin on an undertaking that it be used on behalf of the claimant in satisfying any adverse costs order against it in the jurisdiction applications;

(ii) the public addresses of the Bitcoin.

35. The 10% "buffer" is directed towards addressing the volatility in the value of the Bitcoin. In addition, the draft order includes a mechanism for topping up the value of the Bitcoin to the value of the security ordered plus the 10% buffer.
36. Importantly, the draft order provides for the enforcement of any costs order in favour of the defendants by transfer of the Bitcoin from the claimant's solicitors to the defendants.

Legal principles

37. The principles applicable to the court's discretion as the manner of security are set out by Popplewell J (as he was) in Monde Petroleum SA v Westernzagros Ltd [2015] EWHC 67 (Comm); [2015] 1 Lloyd's Rep. 330 at [61]:

"It is conventional to order security to be given either by payment into Court or by the provision of a guarantee from a first class London bank. That practice recognises that the security should be in a form which enables the defendant to recover a costs award made in its favour at the trial from funds which are readily available, such that there is little risk of delay or default in enforcement. Although security may be ordered in an alternative form, that form should be such as to fulfil the same function, so as to allow simple and swift enforcement of a costs order from a creditworthy source. In practice any such alternative form of security must be such as can properly be regarded in these respects as at least equal to, if not better than, security by payment into Court or provision of a first class London bank guarantee. See Belco Trading Co. v Condo [2008] EWCA Civ 205 at paragraphs [6] to [9] and Versloot Dredging BV v HDI Gerling Industrie Versicherung AG [2013] EWHC 658 (Comm) at paragraph [10]."

38. More recently, in Infinity Distribution Ltd (in administration) v Khan Partnership LLP [2021] EWCA Civ 565, the Court of Appeal set out the principles to be applied in determining the form of security to order when the claimant proposes an alternative form of security that is not the usual payment into court:

(1) In exercising its discretion to make an order for security under CPR 25.12 and 25.13(1)(a), a court should have regard to all the relevant circumstances: [32].

(2) When exercising any power given to it under the rules, including under CPR 25.12 and 25.13(1), a court is obliged by CPR 1.2(a) to seek to give effect to the overriding objective, which, by CPR 1.1(2), includes, so far as practicable, ensuring that the parties are on an equal footing and ensuring that the matter is dealt with fairly: [33].

(3) The task of the court is to "weigh up the respective pros and cons and strike a fair balance between the interests of the parties", and this balancing of pros and cons "is likely to be the primary consideration": [34] – see also [35].

(4) If, on an application for security, two different forms of security would provide equal protection to the defendant, the court should, all else being equal, order the form which is least onerous to the claimant (emphasis added): [45].

Discussion and conclusion

39. In addition to its written submissions, the claimant filed, on 18 January 2022, a further witness statement dated 18 January 2022 of Dr Wright, which I have read de bene esse. In it, Dr Wright proposes and sets out factual matters and arguments in support of the court ordering security to be provided in the form of digital assets, namely either Bitcoin Satoshi Vision or so-called Bitcoin Core. The defendants object to the admission of the statement.
40. This evidence not having been filed or served before judgment on the security applications was handed down, the claimant requires the court's permission to rely upon it. No application for permission has been made. It is evidence that could have been put before the court at the hearing, and no explanation is put forward to explain why it was not. Insofar as it advances arguments, I have considered those arguments (which effectively duplicate those found in its counsel's written submissions). Insofar as it raises factual matters to which the defendants have not had an opportunity to respond, it would be unfair in my judgment to permit the claimant to rely upon those matters, and I disregard them.
41. I note however that the statement includes evidence that:

(1) since (as previously evidenced) the claimant does not have a bank account, it is "impractical" for it to obtain a guarantee from a reputable first-class English bank;

(2) in order to provide security, the claimant would have to exchange digital assets for pounds sterling; and this would give rise to a CGT liability.

42. Moreover, even in this new evidence, the claimant has not given evidence as to

(1) its overall financial position;

(2) whether it could raise the necessary funds from (or obtain a bank guarantee with the assistance of) outside sources e.g. Dr Wright, or other backers or interested sources;

(3) whether it could raise the necessary funds by using the Bitcoin as security;

(4) how it is funding its own legal costs.

There also remains no suggestion that the claim will be stifled if security of the usual type is ordered.

43. The claimant accepts, and I can take judicial notice of, the high level of volatility in the value of Bitcoin. It does not, as security, meet the criteria in Monde Petroleum. Thus, even though the claimant claims that providing the usual form of security would impose a burden on it, this is not a case where all other things are equal.
44. The security offered by the claimant would not result in protection for the defendants equal to a payment into court, or first class guarantee. It would expose them to a risk to which they would not be exposed with the usual forms of security: namely of a fall in value of Bitcoin, which could result in their security being effectively valueless. The top-up provisions proposed by the claimant do not fully meet this risk, because if the claimant did not comply with the order, there would be a substantial risk that enforcement of the obligation could not be achieved before judgment in the jurisdiction applications. Furthermore, the draft order envisages any liability for costs to be satisfied by the transfer of the Bitcoin, which would be an additional occasion when the defendants would be subjected to the risk of a fall in value.
45. I decline therefore to order security in the manner proposed by the claimant.

Costs of the security applications

Liability - legal principles

46. The court's decision as to costs is a discretionary one. However, the general rule is that the unsuccessful party pays the successful party's costs: CPR 44.2(2)(a). The court retains a discretion to make a "different order": CPR 44.2(2)(a).
47. CPR 44.2(4) directs the court to have regard to all the circumstances, including (so far as relevant):

"(a) the conduct of all the parties;

(b) whether a party has succeeded on part of its case, even if that party has not been wholly successful; …"

48. CPR 44.2(5) provides that the conduct of the parties includes (so far as relevant):

" …

(b) whether it was reasonable for a party to raise, pursue or contest a particular allegation or issue;

(c) the manner in which a party has pursued or defended its case or a particular allegation or issue; …"

Liability: discussion and conclusions

49. It is clear in my judgment that the defendants are the successful parties in the security applications: they obtained an order for security where the claimant resisted it "as a matter of principle".
50. The claimant submitted that the costs claimed by the defendants should be discounted by about 76%. This, it said, reflects the fact that the defendants lost on "the majority of issues", and the reasonableness and proportionality of the costs incurred.
51. It is correct that of the 4 grounds advanced by the defendants, they succeeded only on one, the impecuniosity condition. The claimant submitted that it was unnecessary and unreasonable to have relied on the unsuccessful grounds.
52. D15/16 submitted that there was only one issue in the claim, namely whether security should be granted, and that they succeeded on that issue. Whether the impecuniosity ground was satisfied was, they said, pursued as their primary case. The fact that they did not succeed in showing that the other conditions were satisfied did not, they said, justify a costs order in which the issue of security was divided into various conditions.
53. Neither side referred me to any authority as to the principles to be applied when the court is considering whether to make an issue-based costs order. These are set out in para 44.2.10 of the 2021 White Book. They reveal the delicate balance to be struck between giving proper weight to the success of the successful party, and discouraging a "kitchen sink" approach to litigation:

"1. The rules themselves impose no requirement to the effect that an issue-based costs order should be made only "in a suitably exceptional case", and none is to be implied, although "there needs to be a reason based on justice" for departing from the general rule, and that the question of the extent to which costs of a particular issue are to be disallowed should be left to the evaluation and discretion of the judge, "by reference to the justice and circumstances of the particular case" (F&C Alternative Investments (Holdings) Ltd v Barthelemy (No.3) [2012] EWCA Civ 843; [2013] 1 WLR 548, CA, at paras 47 and 49 per Davis LJ (a case where a proportionate costs order, made in relation to two issues on which the parties who had succeeded overall had not succeeded, was upheld)).

2. The reasonableness of taking failed points can be taken into account, and the extra costs associated with them should be considered (Antonelli v Allen, The Times, 8 December 2000, unrep. (Neuberger J); Sycamore Bidco Ltd v Breslin [2013] EWHC 583 (Ch); [2013] 4 Costs LO 572 (Mann J)).

3. Where the circumstances of the case require an issue-based order in the form of an order expressed by reference to the costs of the issue, that is what the judge should make; however, generally, because of the practical difficulties which this causes, the judge should hesitate before doing so and, where practicable, the order should be expressed as a percentage or with reference to a distinct period of time (r.44.2(7)) (Multiplex Constructions (UK) Ltd v Cleveland Bridge UK Ltd [2008] EWHC 2280 (TCC); [2009] 1 Costs L.R. 155 (Jackson J) at para.72(iv)).

4. There is no automatic rule requiring an issue-based cost order in the form of a reduction of a successful party's costs if he loses on one or more issues (HLB Kidsons v Lloyds Underwriters [2007] EWHC 2699 (Comm); [2008] 3 Costs LR 427 (Gloster J) at para.10). The mere fact that the successful party was not successful on every last issue cannot, of itself, justify an issue-based costs order (J Murphy & Sons Ltd v Johnson Precast Ltd (No.2) [2008] EWHC 3104 (TCC); [2009] 5 Costs LR 745 (Coulson J) at para.10).

5. The courts recognise that in any litigation, especially complex commercial litigation …, any winning party is likely to fail on one or more issues in the case (possibly issues on which the losing party could have taken steps to protect himself, at least to an extent, to costs liability)."

54. In addition, it is well established that unreasonableness is not a pre-condition to making an issue-based order: see, e.g. Capita (Banstead 2011) Ltd v RFIB Group Ltd [2017] EWCA Civ 1032; [2017] C.P. Rep. 38; [2017] 4 Costs L.R. 669, CA at para. 41.
55. Applying these principles, I turn to consider whether the defendants' failure to establish the other 3 grounds it relied upon justifies departing from the general rule. I reject D15/16's submission that there was only one issue in the application, namely whether they were entitled to security. Each of the grounds would have justified the court considering whether to exercise its discretion to grant security and are, in my judgment "issues" for the purpose of the principles set out above.
56. The defendants' failures on the nominal claimant and enforcement avoidance condition do not in my judgment of themselves justify departing from the general rule. These were relatively short points, primarily of law, and took up relatively short parts of the hearing.
57. The position as to the non-residence condition is different. There was substantial evidence as to the factual position, and argument both as to the legal principles and the application of those legal principles to the facts. In my judgment, the defendants' failure on this ground justifies a reduction of 15% in the costs recoverable by them.

Amount

58. D15/16's costs total £53,870.36, including the costs of written submissions following judgment. The claimant's costs schedule filed before the hearing for D15/16's application totalled £73,206.25 – it has not filed a further schedule for its post judgment costs.
59. D2-12's costs, again including post judgment costs, total £33,328.55. The claimant's schedule for their application totalled £55,755, and again no post judgment schedule has been filed.
60. The primary criticisms made of the amount of D15/16's costs are as to rates and excessive use of grade A time on both attendances and documents. I have already rejected the criticisms made of the rates. Similarly, much of the grade A time is spent by the lower grade A fee earner, whose rate of $400 (£288) is well below the guideline rates. I therefore also reject this criticism. I agree that £732.48 in photocopying and "miscellaneous charges" should not be allowed: see SCP PD 47, para 5.22(4), (5).
61. I therefore summarily assess D15/16's costs of their security application in the sum of 85% of £53,137.88 = £45,167.20.
62. As to D2-12's costs, the claimant submitted that these should be reduced to reflect the high level of duplication of the work done by solicitors acting on behalf of the two sets of defendants. I accept that there is a degree of duplication, but this is in my judgment sufficiently reflected in the amount of D2-12's costs, which are some £20,000 less than those of D15/16.
63. I therefore summarily assess D2-12's costs of their security application in the sum of 85% of £33,328.55 = £28,329.27.
64. The parties should therefore file a draft order, agreed if possible, reflecting the above, as soon as practicable, and in any event, by 2pm on Friday, 28 January 2022. If the order cannot be agreed, the parties' respective proposed orders should be filed, together with short submissions on the points of difference.

Note 1   See the judgment of Hamblen LJ (with whom Sir Stephen Richards and Longmore LJ agreed) in Danilina v Chernukhin [2018] EWCA Civ 1802, at [69]-[70]    [Back]