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)

		Rolls Building 7 Rolls Building Fetter Lane London EC4A 1NL
		15/01/2021

Mr Justice Fancourt :

This judgment comprises the following parts and sections:

I. Introduction	1-21
II. The Relevant Factual Allegations	22-37
III. The Law of Knowing Receipt Issue	38-117
(a) The nature of liability for knowing receipt	38-46
(b) The Macmillan v Bishopsgate case	47-55
(c) More recent authority	56-74
(d) The claimants' arguments	75-88
(e) The argument based on the Land Registration Act 2002	89-106
(f) Conclusions	107-117
IV. The Saudi Arabian Law Issue	118-206
(a) Preliminary matters	118-123
(b) The Disputed Securities in Saudi Arabian law	124-136
(c) The parties' cases	137-140
(d) The expert witnesses on Saudi Arabian law	141-150
(e) The attacks on the credit of the expert witnesses	151-168
(f) The sub-issues for decision	169-173
(g) Characterisation	174-181
(h) Remedy against third party	182-185
(i) Effect of registration	186-199
(j) The National Commercial Bank shares	200-202
(k) A claim for compensation only?	203-206
V. The Valuation Issue	207-263
(a) Introduction to the block discount question	207-212
(b) The expert witnesses	213-216
(c) Is a block discount excluded in principle?	217-227
(d) The right basis of value to apply	228-250
(e) Evidence of market value	251-260
(f) The level of applicable discounts	261-263
VI. Conclusion	264-265

I. Introduction

1. This claim is brought by joint official liquidators of a Cayman Islands registered company, Saad Investments Company Limited ("SICL"), which is the beneficiary of Cayman Islands trusts. The claim is to recover from a Saudi Arabian bank the value of shares in five Saudi Arabian companies ("the Disputed Securities") that were transferred to the defendant bank on or about 16 September 2009 in breach of trust ("the September Transfer").
2. The transferor was the trustee, Maan Al-Sanea, and the purpose of the September Transfer was to discharge part of the debt he owed the defendant bank, Samba Financial Group ("Samba"). Samba credited Mr Al-Sanea's account with the market value of the Disputed Securities on the day of transfer, some 801 million Saudi riyals.
3. Samba still holds the Disputed Securities but the claimants do not claim them in specie. They claim instead an account of their value on either the date of the September Transfer or the date of judgment, together with income derived from the Disputed Securities since the date of the September Transfer and interest, by way of equitable compensation.
4. For reasons that I will shortly explain, only three substantive issues were tried:

i) Whether the effect of Saudi Arabian law, as the governing law of the September Transfer, was to extinguish SICL's rights in the Disputed Securities even if Samba had knowledge of SICL's interest ("the Saudi Arabian Law Issue");

ii) Whether the claim, pleaded by the claimants as governed by Cayman Islands or English law, must fail if SICL's interest was so extinguished ("the Law of Knowing Receipt Issue");

iii) The value of the Disputed Securities at the date of the September Transfer and at the date of judgment – in reality, this was only a dispute about whether a "block discount" should be applied to the quoted prices of the Disputed Securities on the Saudi Arabian stock exchange on those days ("the Valuation Issue").

5. The Law of Knowing Receipt Issue raises an important question of whether a transferee, who upon receipt obtains title to the property free from a beneficiary's equitable proprietary interest, can nonetheless be personally liable in equity for knowing receipt because he received the property with sufficient knowledge that the transfer was a breach of trust. That is not a question that arises in relation to English property under the general law of England and Wales: a transferee who has sufficient knowledge of a breach of trust cannot be a bona fide purchaser of a legal estate without notice of the breach of trust. It may however arise under a statutory scheme for registration of title, e.g. under the Land Registration Act 2002, where a disponee's title cannot be questioned on the basis of any unregistered limitation affecting the owner's powers of disposition, even if the transfer is unlawful (s.26(3) of the 2002 Act) and notice or knowledge of an unprotected equitable interest is irrelevant to priorities (ss. 28, 29 of the 2002 Act). It may also arise, as in this case, where under the applicable foreign law of the property the transferee's title may have priority to the interest of the beneficial owner and knowledge of that interest is irrelevant. In such cases, it raises difficult questions – not for the first time – about the true juridical nature of liability in knowing receipt.
6. The claimants' claim is advanced on the basis that it is irrelevant whether under Saudi Arabian law Samba's title extinguished or overrode SICL's proprietary interest – because under Cayman Islands or English law, whichever is the law of this claim, Samba received the Disputed Securities for its own purposes with sufficient knowledge that they had been transferred in breach of trust. Alternatively, the claimants argue that under Saudi Arabian law the effect of registration of Samba as owner of the Disputed Securities is not to deprive SICL of the ability to assert against Samba its equitable interest.
7. The claim has a long and complex procedural history. In brief outline, it is as follows.
8. The first claim started on 14 August 2013, when the joint official liquidators of SICL claimed a declaration under s. 127 of the Insolvency Act 1986 that the September Transfer was a void disposition of property belonging to SICL. Samba applied to stay those proceedings on forum non conveniens grounds and a stay was granted on 28 February 2014 by the then Chancellor of the High Court, Sir Terence Etherton.
9. On appeal against that order, the argument in the Court of Appeal was whether the claim was bound to fail on the basis that the lex situs of the shares, Saudi Arabian law, did not recognise common law trusts and so the claimants could not establish that SICL had beneficial ownership of the shares. The Court of Appeal considered that both the identity of the governing law of the trusts and the content of relevant Saudi Arabian law were arguable, depending on the exact facts found at trial, and so it allowed the appeal on 4 December 2014.
10. The parties argued a further appeal to the Supreme Court on, principally, a narrow issue of whether there could in Cayman Islands or English law (the two being materially identical) be a valid trust of shares in companies in a jurisdiction where equitable interests in property were not recognised. After hearing that argument, the Supreme Court invited written argument on broader questions, including whether there was a "disposition" within the meaning of s. 127 of the Insolvency Act 1986 by the September Transfer if SICL did have equitable proprietary interests in the Disputed Securities.
11. The Supreme Court decided on 1 February 2017 that a common law trust of shares of a company registered in a jurisdiction where no division of legal and equitable interests was recognised was valid, so that SICL had a beneficial interest; however, it also decided that there was no "disposition" of SICL's beneficial interest in the shares by the September Transfer. Mr Al-Sanea transferred the legal title to the shares and the equitable interest of SICL was extinguished, on the assumption (which was made at that time, for the purpose of the argument, because no allegation of knowledge of breach of trust had yet been made) that Samba was a bona fide purchaser of the shares without notice. (If Samba was not a bona fide purchaser without notice then, under English law, it took the shares subject to SICL's equitable interests, which ex hypothesi were not disposed of.) The Supreme Court did not need to and could not decide whether Samba acquired good title to the Disputed Securities under Saudi Arabian law, but it recognised that possibility if, contrary to the joint official liquidators' case, Saudi Arabian law applied to decide that question.
12. Recognising that there were other ways in which the joint official liquidators might seek to plead a case against Samba, the Supreme Court allowed time for them to make that application and did not immediately strike out the claim.
13. The joint official liquidators then applied to amend the particulars of claim to plead a claim in knowing receipt. In case permission to amend was refused on limitation grounds, they also issued a new claim, which is the claim that in the event was before me at trial. Samba issued an application to strike out the new claim and that application and the application of the liquidators to amend the 2013 claim were heard together by Birss J. On 6 December 2017, he granted permission to amend and dismissed the strike out application. Samba obtained permission to appeal only on the question of whether permission to amend the 2013 claim should have been refused on limitation grounds. In due course, that appeal succeeded and the 2013 claim was dismissed, leaving the new claim to proceed to trial.
14. In the meantime, the parties had filed statements of case. Samba denied any relevant knowledge of the Cayman Islands trusts or of a breach of trust by Mr Al-Sanea. It also contended that it obtained good title to the shares which extinguished any equitable interest that SICL had, such that the claim in knowing receipt must fail. A first Case Management Conference took place before HH Judge Klein on 20 November 2018. He gave detailed directions including a direction for standard disclosure, leading to a trial in October 2020.
15. Samba did not comply with the order for disclosure and, in my judgment handed down on 8 April 2020 (Byers v Samba Financial Group [2020] EWHC 853 (Ch)), I held that Samba must be debarred from defending the claim save for certain limited issues. These were the only issues where it was sufficiently clear at that stage that a fair trial of them could not be prejudiced by the absence of any disclosure from Samba and that it would be unjust to deny Samba a trial of them. They were:

i) The governing law of the September Transfer;

ii) If Saudi Arabian law is the governing law of the September Transfer, whether its effect is to extinguish SICL's rights in the Disputed Securities even if Samba had knowledge of SICL's interest;

iii) If Saudi Arabian law is the governing law of the claimants' claim, whether that claim as pleaded must fail (as a matter of that law);

iv) If Cayman Islands or English law is the governing law of the claimants' claim, whether that claim must fail (as a matter of that law);

v) The value of the Disputed Securities at the date of the September Transfer and at the date of trial.

It has at all times been common ground that it is unnecessary to decide which of English law and Cayman Islands law is the governing law of the claim because there is no material difference between them.

16. Among the issues that Samba sought but was not permitted to defend were, first, the governing law of the transactions between Mr Al-Sanea and SICL between 2002 and 2008 that gave rise to the allegation by the claimants of valid Cayman Islands trusts ("the Six Transactions") and the law governing the effect of the Six Transactions; and second, the identity of the governing law of the substantive claim. The claimants contended that the applicable law in each case was Cayman Islands or English law; Samba contended for Saudi Arabian or Bahraini law.
17. At a further hearing on 24 April 2020, it was agreed that since the claimants' primary case was that Cayman Islands or English law was the governing law of the claim, and its claim in Saudi Arabian law was only an alternative basis of claim in case it was wrong on its primary case, issue (iii) above did not need to be tried unless and until the claimants amended their claim to rely on Saudi Arabian law as the governing law of their claim. They have not done so.
18. By the start of the trial, the claimants had effectively conceded issue (i) on the basis that Saudi Arabian law must be the governing law of the effect of the September Transfer of the Disputed Securities. In the event, the parties' valuers agreed all valuation issues except for the issue of whether a block discount should be applied and the values on the judgment date. The values on the judgment date will have to be assessed (on an agreed basis, subject to the decision on the block discount) after this judgment is handed down, if the claimants succeed.
19. The effect of my decision was that all factual questions other than the content of Saudi Arabian law and the valuation issues were deemed to have been resolved in accordance with the claimants' pleaded case. In a further ruling on 24 April 2020 (Byers v Samba Financial Group [2020] EWHC 1006 (Ch)), I held that judgment would be entered for the claimants at trial, without the need for them to prove their case by calling evidence on those matters, subject only to the outcome of the permitted trial issues. Permission to appeal that ruling was refused by the Court of Appeal by order dated 10 July 2020. The position at the start of trial was accordingly that only three discrete issues (the Saudi Arabian Law Issue, the Law of Knowing Receipt Issue and the Valuation Issue) fell to be tried.
20. It is now common ground that the claimants only need to succeed on either the Saudi Arabian Law Issue or the Law of Knowing Receipt Issue to succeed on their claim. That is because Samba's argument as to why this claim must fail, as a matter of law, is that – absent a pleaded allegation of dishonest assistance in a breach of trust – the claimants must prove what Samba called an "undestroyed proprietary base" after Samba's receipt of the Disputed Securities, i.e. that SICL's beneficial proprietary interest in the property was not extinguished at the moment of receipt by Samba. Samba contends that an assertion of a constructive trust based on knowing receipt of property transferred in breach of trust cannot succeed if, at the moment of receipt, the recipient acquires good title to the property and the beneficiary's proprietary interest is extinguished or overridden.
21. If the claimants succeed on that issue, because receipt of trust property with knowledge of a breach of trust is sufficient under English or Cayman Islands law, it matters not whether in Saudi Arabian law Samba's registration as owner of the Disputed Securities gave it unencumbered title. Equally, if the claimants succeed in proving that SICL's interest was not overridden under Saudi Arabian law, it matters not to them – though it may be of considerable importance to equity lawyers – whether they are right on the Law of Knowing Receipt Issue.

II. The Relevant Factual Allegations

22. Before turning to those two issues, I must set out the factual basis of the claimants' claim that Samba is liable as a knowing recipient of the Disputed Securities and Samba's defence to it, so far as relevant now.
23. The Six Transactions, under which the Cayman Islands trusts of shares that included the Disputed Securities were created, were of two types. The first type is a share sale agreement made between Mr Al-Sanea and SICL, under which SICL agreed to buy and Mr Al-Sanea agreed to sell the beneficial ownership in shares and Mr Al-Sanea agreed to remain (or become) the registered nominee owner, with all risk and all income being for SICL, and income held for it as trustee by Mr Al-Sanea. In one case, the vendor of the shares is Ahmad Hamad Algosaibi & Brothers Company ("AHAB"), and Mr Al-Sanea agreed to be the nominee owner as trustee for SICL. Each of these sale agreements has an express choice of law clause, with Bahraini or Saudi Arabian law specified as the governing law. In each agreement, Mr Al-Sanea agreed to deal with the shares in accordance with SICL's directions.
24. The second type of transaction comprises a bill of sale of shares for payment within 60 days and a simple form declaration of trust of those shares, both of which documents were created by Mr Al-Sanea and SICL on the same day. The declaration of trust states that the shares, though standing in the name of Mr Al-Sanea, "in fact belong to the buyer" (SICL) and that Mr Al-Sanea will hold them in trust for SICL, deal with them as directed by SICL, and complete all documentation necessary for that purpose.
25. It is clear therefore that, under both types of transaction, SICL was solely beneficially entitled to the shares and had the right to direct Mr Al-Sanea to transfer the shares to another trustee, if it so wished.
26. Given that, by default, the claimants' case about the lawfulness and effect of the Six Transactions under Cayman Islands law is proved, nothing further turns on the content of the Six Transactions. By virtue of the decision in Akers v Samba Financial Group [2017] UKSC 6; [2017] AC 424, the trusts were valid even though the lex situs of the trust property does not (as is common ground in this trial but was assumed by the Supreme Court) give direct effect to the common law trust or recognise any division between legal and equitable ownership.
27. The claimants' re-amended particulars of claim make detailed allegations about the state of knowledge of Samba at the date of the September Transfer. By default, the facts alleged are taken to be proved against Samba. However, the precise factual allegations are of some importance to the resolution of the Law of Knowing Receipt Issue. It is therefore appropriate to set them out here.
28. In support of its allegation that Samba knew that the Disputed Securities were being held on trust for SICL by Mr Al-Sanea, the claimants rely on Mr Al-Sanea and Mr Algosaibi being directors of Samba from 2003 to 2006 (in Mr Al-Sanea's case) and from 2003 to 2009 (in Mr Algosaibi's case). They also plead that Samba's knowledge is to be inferred from various matters, including SICL's own relationship with Samba as a borrower in its London branch and Samba's knowledge of SICL's assets, including those (such as shares in Saudi Arabian companies) that must have been held by a nominee on its behalf.
29. Further or alternatively, the claimants plead that:

"41.1 In the context of what Samba actually knew, a reasonable bank in its position would have appreciated that (alternatively would or ought to have made inquiries or sought advice which would have revealed the probability that):

41.1.1 The Relevant Securities (and therefore the Disputed Securities) were held by Mr Al-Sanea on trust for SICL; and

41.1.2 The September Transfer was a breach of trust; and/or

41.1.3 In the light of what Samba actually knew, Samba recklessly failed to make such inquiries about the September Transfer, the Relevant Securities and the Disputed Securities as an honest and reasonable bank would make: had Samba made such inquiries, it would have learned that Mr Al-Sanea held the Relevant Securities (and therefore the Disputed Securities) on trust for SICL."

30. In support of that allegation, the claimants rely on certain documents relating to SICL's financial position that Samba received, so that Samba knew of SICL's shareholdings in five Saudi Arabian banks, being the same five banks in which the Disputed Securities were held, including Samba Financial Group itself. On that basis, it is alleged that Samba would have appreciated or ought to have appreciated that SICL owned the Disputed Securities at the date of the September Transfer and that such ownership was through a nominee, which was very likely to have been Mr Al-Sanea, given his ownership and control of SICL and his ownership of large quantities of shares in each of the five banks. Further, Samba knew of the liquidation of SICL and the Cayman Islands worldwide freezing order against SICL and Mr Al-Sanea made on 24 July 2009, such that Samba should have been profoundly suspicious about the right of Mr Al-Sanea to transfer the Disputed Securities to Samba and should have made further enquiries before entering into the September Transfer.
31. On that basis, the claimants plead that:

"… upon its receipt of the Disputed Securities:

43.1 an equity arose and remains between SICL and Samba in relation to the Disputed Securities; and

43.2 Samba became accountable to SICL as a constructive trustee of the Disputed Securities, and remains so."

(emphasis added)

32. Samba at trial did not dispute that these allegations of knowledge of various types (taken to be proved) were sufficient in principle to establish the overarching requirement for knowing receipt liability, namely that "the recipient's state of knowledge should be such as to make it unconscionable for him to retain the benefit of the receipt": per Nourse LJ in Bank of Credit and Commerce International (Overseas) Ltd v Akindele [2001] Ch 437 at 455. It is therefore unnecessary to analyse further the extent to which particular allegations are or are not sufficient in themselves to give rise in principle to liability.
33. Two matters in particular should be noted, however. First, neither in the re-amended particulars of claim nor in the amended reply to Samba's amended defence have the claimants alleged that Samba acted dishonestly. An allegation that a person has failed recklessly to make the enquiries that an honest person would have made is not an allegation of dishonesty but of a type of knowledge (sometimes called constructive knowledge) that suffices to make it unconscionable for the recipient to keep the property received. The claimants in argument – though at times they came close to alleging that Samba was an accessory to theft of the Disputed Securities – did not (and could not) argue that their pleaded case could be taken as an allegation of dishonesty, such as would be required to establish liability as a constructive trustee for dishonest assistance in a breach of trust.
34. Second, and connected with the first point, the claimants expressly plead that the equity on which they rely arose upon receipt by Samba of the Disputed Securities. No allegation is made of any prior equity in favour of SICL arising from dealings between SICL, Mr Al-Sanea and Samba. The claimants' claim is therefore solely a claim based on the receipt by Samba of SICL's property with sufficient knowledge that it was transferred to SICL in breach of trust.
35. Samba's Amended Defence pleads that the effect of the September Transfer on property rights in the Disputed Securities is determined by Saudi Arabian law, as the lex situs of the Disputed Securities. Saudi Arabian law therefore determines whether SICL's equitable interest was extinguished or overridden by the September Transfer. Under Saudi Arabian law, title to property is indivisible; Mr Al-Sanea was the registered legal owner of the Disputed Securities and the September Transfer passed good title in them to Samba, which itself obtained unqualified registered title. Any equitable interest of SICL was therefore extinguished by the September Transfer.
36. Para 61 of the Amended Defence pleads:

"Even if (contrary to Samba's case) the claim against Samba is governed by English (or Cayman Islands) law:

61.1 The cause of action in unconscionable receipt requires the claimant to have retained a proprietary interest in the assets received by the defendant.

61.2 As set out in paragraphs 53 to 56 above, under Saudi Arabian law, which as the lex situs governed the proprietary effects of the September Transfer, any proprietary interest that SICL may have held in the Disputed Securities was extinguished.

61.3 This provides a complete defence to the claim. "

37. This plea is the origin of issues that Samba at one stage suggested should be tried as preliminary issues and the issues that I held in my judgment of 8 April 2020 that Samba should be permitted to defend (see at [177] of that judgment). They are whether, as a matter of English and Cayman Islands law, the claimants' claim requires SICL to have retained a proprietary interest in the Disputed Securities immediately following the September Transfer, and if so whether, applying Saudi Arabian law, registration of Samba as shareholder of the Disputed Securities ended SICL's equitable proprietary interest.

III. The Law of Knowing Receipt Issue

(a) The nature of liability for knowing receipt

38. I will address first the issue of whether, for a claim in knowing receipt to succeed, it is necessary for there to be an "undestroyed proprietary base" to the claim, that is to say that the equitable interest of the claimant in the property is not destroyed or overridden at the moment of receipt, but the recipient thereafter - at least for a time, however short - holds property that belongs in equity to the claimant. It matters not if the recipient thereafter disposes of the property: personal liability is established if knowledge precedes the disposal.
39. For reasons previously given, such overriding in relation to English property could only arise in English law under a statutory registration code; it cannot arise under the general law. Samba contends that it arises here in relation to Saudi Arabian securities. That is because of its pleaded defence that under Saudi Arabian law, applied by English or Cayman Islands conflicts of law rules to determine priorities in relation to the Disputed Securities, registration of Samba as proprietor of the Disputed Securities extinguishes any rights of SICL in the Disputed Securities and gives Samba good title irrespective of its knowledge of SICL's interest. Since, as is common ground, the transfer to Samba occurred at the moment of Samba's registration, Samba contends that it only received shares to which it had good title, not property belonging to SICL.
40. As previously explained, the claimants dispute that Saudi Arabian law has any application in an in personam as opposed to a proprietary claim; but if it does, they submit that the relevant question under Saudi Arabian law is not whether SICL has a valid proprietary claim but whether it has a valid personal claim for compensation from Samba.
41. Liability for knowing receipt of property transferred in breach of trust is often described as liability as a constructive trustee, though that term has been criticised for its potential to mislead. In a frequently cited paragraph from his judgment in Paragon Finance plc v DB Thakerar & Co [1999] 1 All ER 400, Millett LJ made observations about two categories of case at p.409:

"A constructive trust arises by operation of law whenever the circumstances are such that it would be unconscionable for the owner of property (usually but not necessarily the legal estate) to assert his own beneficial interest in the property and deny the beneficial interest of another. In the first class of case, however, the constructive trustee really is a trustee. He does not receive the trust property in his own right but by a transaction by which both parties intend to create a trust from the outset and which is not impugned by the plaintiff. His possession of the property is coloured from the first by the trust and confidence by means of which he obtained it, and his subsequent appropriation of the property to his own uses a breach of that trust… In these cases the plaintiff does not impugn the transaction by which the defendant obtained control of the property. He alleges that the circumstances in which the defendant obtained control make it unconscionable for him thereafter to assert a beneficial interest in the property.

The second class of case is different. It arises when the defendant is implicated in a fraud. Equity has always given relief against fraud by making any person sufficiently implicated in the fraud accountable in equity. In such a case he is traditionally though I think unfortunately described as a constructive trustee and said to be 'liable to account as constructive trustee'. Such a person is not in fact a trustee at all, even though he may be liable to account as if he were. He never assumes the position of a trustee, and if he receives the trust property at all it is adversely to the plaintiff by an unlawful transaction which is impugned by the plaintiff. In such a case the expressions 'constructive trust' and 'constructive trustee' are misleading, for there is no trust and usually no possibility of a proprietary remedy; they are 'nothing more than a formula for equitable relief': Selangor United Rubber Estates Ltd v Cradock (No.3) [1968] 2 All ER 1073 at 1097, [1968] 1 WLR 1555 at 1582 per Ungoed Thomas J."

42. Liability for knowing receipt self-evidently falls into the second class of these cases of so-called constructive trust. There is no prior fiduciary relationship and the recipient first receives the trust assets by an unlawful transaction.
43. In the 1990s, the development of the law of knowing receipt became entangled with the parallel development and much broader reach of restitutionary principles. At that time, the language of restitution became the lingua franca of most if not all property-based and financial remedies. In El Ajou v Dollar Land Holdings plc [1993] BCC 698, Millett J was concerned with a claim to recover money received by the defendant, allegedly with knowledge that it represented the proceeds of fraud. Under a sub-heading "Knowing Receipt" at p. 716F, his Lordship said:

"The plaintiff seeks a personal remedy based on "knowing receipt". As I have previously pointed out, this is the counterpart in equity of the common law claim for money had and received. The latter, at least, is a receipt-based claim to restitution, and the cause of action is complete when the money is received: see Lipkin Gorman v Karpnale Ltd [1991] 2 AC 548 at p. 572. So, in my judgment, is the former, unless arbitrary and anomalous distinctions between the common law and equitable claims are to be insisted upon."

44. In the Court of Appeal's decision in the same case, [1994] 2 All ER 685 at 700g, Hoffmann LJ identified the essential elements of a claim to enforce a constructive trust on the basis of knowing receipt:

"For this purpose the plaintiff must show, first, a disposal of his assets in breach of fiduciary duty; secondly, the beneficial receipt by the defendant of assets which are traceable as representing the assets of the plaintiff; and thirdly, knowledge on the part of the defendant that the assets he received are traceable to a breach of fiduciary duty."

The emphasis on traceability of the assets is explained by the particular facts of that case, where the main issue was not whether the defendant could establish a better title than the plaintiff but whether the stolen money could be traced into funds in the hands of the defendant.

45. It is clear from these cases that liability in knowing receipt arises where a stranger to the trust receives property to which he is not entitled, knowing that the transfer to him was a breach of trust or fiduciary duty. The equity arises upon receipt of property that belongs to another. In Williams v Central Bank of Nigeria [2014] UKSC 10; [2014] AC 1189, Lord Sumption JSC said:

"[31] The essence of a liability to account on the footing of knowing receipt is that the defendant has accepted trust assets knowing that they were transferred to him in breach of trust and that he had no right to receive them. His possession is therefore at all times wrongful and adverse to the rights of both the true trustees and the beneficiaries. No trust has been reposed in him. He does not have the powers or duties of a trustee, for example with regard to investment or management. His sole obligation of any practical significance is to restore the assets immediately. It is true that he may be accountable for any profit that would have been made or any loss that would have been avoided if the assets had remained in the hands of the true trustees and been dealt with according to the trust. There may also, in some circumstances, be a proprietary claim. But all this is simply the measure of the remedy. It does not make him a trustee or bring him within the provisions of the Limitation Act 1980 relating to trustees."

46. The emphasis in this paragraph on the absence of a right to receive the assets and possession that is at all times "wrongful" suggests that liability in knowing receipt should not arise where the recipient is, as a consequence of the transfer or otherwise, entitled to the assets. If the primary duty of the knowing recipient is to restore the trust property, with a secondary obligation to account for any profits received and losses suffered by the trust, it is not obvious how such duties can exist if the recipient is entitled to retain the assets for himself.

(b) The Macmillan v Bishopsgate case

47. The question of whether liability as a knowing recipient can exist where the recipient obtains good title to the property under local law arose in an acute form in Macmillan Inc v Bishopsgate Investment Trust plc (No.3) [1995] 1 WLR 978. The key issue in that case was whether English law or New York law should apply to determine whether transferees of shares in a New York corporation were liable to return the proceeds of sale of the shares to the plaintiff. The claim was both a proprietary claim and a personal claim against each of the recipients of shares, with a claim for compensation and/or damages for breach of constructive trust. The plaintiff sought to characterise it as essentially a restitutionary claim and the defendants sought to characterise it as essentially a proprietary claim. The decision of Millett J at first instance, and the decision of the Court of Appeal, were central to the arguments before me of the claimants and Samba, and both parties spent much time and ink seeking to tease out the true ratio of the decision and apply it to the facts of this case.
48. At this stage, I will set out the relatively long passage from Millett J's judgment on which both parties relied. It is at pp.988C – 990E of the report:

"There is at the outset a fundamental disagreement between the parties as to the proper characterisation of the dispute for the purposes of English conflict of laws. The defendants insist that the question at issue is concerned with the priority of competing interests in a chose in action. Macmillan insists that its claim lies in restitution and, being brought by an equitable owner, must be decided in accordance with equitable principles. Even its claim to the return of the Berlitz shares or their proceeds in specie, it submits, which can loosely be described as proprietary, was in truth no more than an invocation of the power of a court of equity, acting in personam as a court of conscience, to require the defendants to take whatever steps were necessary to restore the shares to Macmillan's ownership. Accordingly, Macmillan submits, even if the question of priorities falls to be decided under New York law, this would at best result in a conflict between the English choice of law rules governing priorities and those governing restitutionary claims which would still need to be resolved. In order to resolve that conflict, Macmillan submits, it is first necessary to identify the purpose for which it is sought to characterise its claim. In the present case that purpose is to determine whether the defendants or any of them is under any obligation to Macmillan to restore the relevant Berlitz shares or their proceeds. Such a claim is properly regarded as a restitutionary claim, and the liability of the defendants to satisfy the claim as a restitutionary obligation.

In my judgment that argument is circular. It is manifestly correct to characterise Macmillan's claim as lying in restitution, but that is only the first step in the analysis. In order to ascertain the applicable law under English conflict of laws, it is not sufficient to characterise the nature of the claim: it is necessary to identify the question at issue. The English law of restitution makes a fundamental distinction between the unjust enrichment of the defendant which is occasioned by depriving the plaintiff of his property and enrichment which results from a wrong done to the plaintiff by the defendant. In the first category of case the plaintiff's restitutionary claim is said to have a proprietary base. The enrichment of the defendant is at the direct expense of the plaintiff and is matched by a corresponding diminution of his assets. The plaintiff brings the claim in order to recover his own property and must succeed, if at all, by virtue of his own title. In the latter class of case his claim arises from a breach of fiduciary or other obligation on the part of the defendant. The distinction is that drawn by equity between the claim of an equitable owner to recover his property, or compensation for the failure to restore it, from a person into whose hands it has come and a claim by a plaintiff in respect of a breach of fiduciary obligation owed to him. In the former case he relies upon his continuing equitable interest in the property under an express or resulting trust; in the latter upon an equity between the parties which may in appropriate circumstances give rise to a constructive trust. The distinction, which is crucial, may have been lost sight of in the language of some of the more recent decisions on knowing receipt.

Macmillan's claim is of the former kind. In respect of the Berlitz shares there was no relationship of any kind between Macmillan and any of the defendants. There is no equity between them. In the absence of such an equity, any liability of the defendants to restore the shares or their proceeds to Macmillan or to pay compensation for their failure to do so must be based upon Macmillan's continuing equitable ownership of the shares. In the language of restitution, Macmillan's claim must rest upon "an undestroyed proprietary base". Such a claim cannot succeed against a party who has under the applicable law acquired a title to the shares which is superior to that of Macmillan.

It is no answer to assert that a claim which invokes the intervention of equity is a claim in personam and part of the law of remedies, and - a highly dubious proposition - as such governed by the lex fori. The principles which I have endeavoured to state not principles of English conflict of laws, but of equity and our domestic law of restitution."

49. Millett J then proceeded to illustrate the difference between the two types of restitutionary claim by reference to Norris v Chambres (1861) 29 Beav 246 (affirmed 3 De G.F.&J. 583). He drew a distinction between a claim specifically to enforce an existing equitable right and a claim against a third party with notice of the right, and explained:

"The difference between the two cases is that in the second there is no equity or privity between the parties which the court can enforce except such equity, if any, as may arise from the transferee's notice; while the sufficiency of such notice to affect the transferee's title is a matter for the lex situs. If, by that law, the transfer to the defendant extinguished the plaintiff's interest notwithstanding the defendant's notice, the plaintiff no longer has any proprietary interest upon which he can base his suit in England. The common law position is the same: see Cammell v Sewell (1860) 5 H.& N. 728."

50. In considering the decision and reasoning of Millett J, it is important to bear in mind, first, that in 1993 the language of restitution was all-pervasive; and second, that the central issue in the case was a conflict of laws issue, which itself depended on the correct characterisation of the issue that lay between the parties. Nevertheless, if one looks through the language of restitution to the equitable and common law basis of the claims in that case, there was undoubtedly a proprietary claim, since the proceeds of sale of the Berlitz shares had been retained by the defendants to await the outcome of the trial, but there was also an in personam claim based on knowing receipt: the plaintiff claimed equitable compensation from the alleged knowing recipients for breach of constructive trust, in circumstances in which there was no prior relationship between them.
51. The decision cannot therefore be taken to have been narrowly decided only on the basis that the proprietary claim raised a priorities issue that had to be determined in accordance with New York law and that the defendants were bona fide purchasers for value without "New York notice" of the breach of duty. Millett J also clearly decided that the claim in knowing receipt raised the same issue because that type of knowing receipt claim depends on a proprietary base. It was not a claim in dishonest assistance, where receipt is legally irrelevant, but a claim in knowing receipt, which is similarly vulnerable to a defence that the recipient acquired good title to the property. This is clear from the passages of the judgment that I have set out above. If it were the case that priority of title under New York law was irrelevant to liability in knowing receipt, because the equity arises from what the recipient knew at the moment of the transfer, the Judge would, I consider, have identified this difference and explained that the validity of Macmillan's personal claim therefore depended on whether the defendant banks received the shares with sufficient knowledge of the breach. That might not have changed the result, since banks that did not have New York notice of Macmillan's interest would not have had sufficient knowledge of a breach of duty, but it would have required the Judge to decide the governing law of the equitable or restitutionary claim.
52. The claimants in this case argue that, on its true interpretation, the claim against Samba, which is not brought as a proprietary claim but as a personal claim for compensation, is a claim in Millett J's second category, viz a claim by a plaintiff in respect of a breach of a fiduciary obligation owed to him. In my judgment it is clear that this is not so. That category, in Millett J's taxonomy, is one where there is a pre-existing fiduciary obligation, or where an equity arises otherwise than upon receipt of trust property, e.g. where a defendant dishonestly assists in a breach of trust, or where the equity arose independently of the receipt of property. The position is, perhaps, somewhat confused by the reversal of the two categories in the Judge's explanation of the decision of the court in Norris v Chambres. It is nevertheless sufficiently clear that Millett J considered that a claim based on equitable ownership of property, where there is no equity against the defendant before his receipt of the property, cannot be maintained if, upon receipt, the defendant acquires good title to the property under the governing law. As explained in that case, it is not the form of the claim that is determinative but the issue that is raised.
53. On appeal from Millett J, the Court of Appeal was concerned with the conflict of laws issue. It considered that the real issue was whether the bank defendants acquired title to the shares in good faith and for value, without notice of any beneficial interest of Macmillan. In addressing Macmillan's argument that the governing law should be the law of the restitutionary obligation, Staughton LJ noted that Millett J identified two different categories of restitutionary claim, in one of which the real issue was one of priority, and said:

"It is clear that Macmillan's claims in the present case are to some extent proprietary. Mr Oliver asserts that they are receipt based. But he needs to do more than show that the defendants received the shares; he must also plead, in effect, that they are Macmillan's shares; and the statement of claim does indeed say that. Millett J. described this requirement as "an undestroyed proprietary base"….

I am prepared to accept that Macmillan's claim is restitutionary in nature; and I would accept without deciding that rule 201 of Dicey & Morris, 12^th ed. determines what system of law governs such a claim. But the issue is not, or not any longer, whether Macmillan have a cause of action for restitution; it is whether the defendants have a defence on the ground that they were purchasers for value in good faith without notice of Macmillan's claim."

His Lordship agreed with Millett J that the law of New York applied to determine that issue.

54. Auld LJ agreed with Millett J's categorisation of the issue:

"In my view, the Judge correctly identified the transaction for this purpose via his identification of the issue. The authorities relied on by Mr Oliver were all cases where there was privity of contract or some fiduciary relationship between the parties stemming from more than mere receipt of property with notice of another's claim to an interest in it. That is not so here. The negotiations and agreements in England proceeding the transfer were not with Macmillan; there was no privity of contract between the parties and, apart from the claimed equity which Macmillan relies upon to support its "receipt-based restitutionary claim", no equitable or other fiduciary relationship between them."

55. Aldous LJ said:

"Macmillan submitted that their claim was in essence a claim for the performance of an obligation by the defendants to restore their property or the proceeds or the value of the property. That, it was said, was a claim in equity for restitution. That is true, but to succeed it involves establishing a number of facts, including that they owned the shares and that they were transferred to the defendants in breach of trust. The reply of the defendants is that the shares are registered in their names and they were bona fide purchasers for value without notice.

The issue between the parties concerns the title to the shares and, in particular, whether Macmillan or the defendants have the better title. The issue is one of priority. I agree with the judge…"

(c) More recent authority

56. The next case in which a similar issue was raised is Lightning v Lightning Electrical Contractors Ltd (1998) 23 TLI 35. The defendant company bought property in Scotland with funds allegedly provided by Mr Lightning. In Scottish proceedings, the Sheriff Court held that Mr Lightning's licensee was not entitled to resist the company's claim to possession. In English proceedings, Mr Lightning claimed a declaration that the property was held by the company on resulting trust for him. The Court of Appeal held that there was no issue estoppel nor any jurisdictional impediment to hearing the claim. That was because there was an existing equity between the parties, both of whom were amenable to the jurisdiction. It further held that the law governing the relationship of alleged resulting trust was English law, notwithstanding the Scottish situs.
57. Having referred to earlier authorities, Peter Gibson LJ said:

"As was pointed out by Millett LJ when sitting at first instance in Macmillan Inc v Bishopsgate Trust (No 3) [1995] 1 WLR 978 at page 989 (commenting on Norris v Chambres (1891) 29 Beavan, 246, affirmed 3 De Gex Fisher and Jones 583), where a plaintiff invokes the in personam jurisdiction of the English court against a defendant amenable to the jurisdiction and there is an equity between the parties which the court can enforce, the English court will accept jurisdiction and apply English law as the applicable law, even though the suit relates to foreign land. In contrast if the equity which is asserted does not exist between the parties to the English litigation, for example where there has been a transfer of the property to a third party with notice of an equity but by the lex situs governing the transfer, the transfer extinguished the plaintiff's equity, the English court could not then give relief against the third party even though he is within the jurisdiction."

This explanation of the difference between enforcing an existing equity that exists inter partes and enforcing an equity against a third party who acquired property under foreign law is revealing. It establishes that, even when the English court has jurisdiction, if the third party obtains good title by the lex situs, his notice of the claimant's equitable interest cannot found a claim against him. Since Lightning was a case in the former category, that explanation is not the ratio of the decision, but nevertheless it is stated as part of the explanation for the decision in that case by a highly experienced and respected equity judge. Millett LJ, another such equity judge, was a member of the court and agreed with Peter Gibson LJ.

58. The issue of whether a recipient of property with knowledge of a breach of trust is liable if under local law he obtains good title arose in the Privy Council decision in Arthur v The Attorney-General of the Turks & Caicos Islands [2012] UKPC 30. The issue arose under the law of the Turks & Caicos Islands, which includes a Torrens style system of land registration. It was that legislation that arguably gave a registered proprietor unimpeachable title regardless of his knowledge of a breach of trust involved in the registrable transfer of the land.
59. The Attorney-General advanced a proprietary and personal claim against Mr Arthur. Mr Arthur argued that his title was good and indefeasible, on the true construction of the Turks & Caicos Registered Land Ordinance, and that accordingly there could be no claim against him in knowing receipt regardless of any knowledge that the land was transferred to him in breach of fiduciary duty. The appeal turned on the true construction and effect of the Registered Land Ordinance. Giving the decision of the Board, Sir Terence Etherton said at [38]:

"Although the claims to personal and proprietary relief are separate, the appellant in the present case seeks to defeat them both by the same argument resting on the provisions of the RLO which, he asserts, have the effect that the appellant never received any trust property since any trust was eliminated at the moment of registration. The respondent has not sought to argue that, even if the proprietary claim is barred by the provisions of the RLO, the personal claim can nevertheless be advanced. Both sides appear to have proceeded on the assumption that knowing receipt claims, even though for personal relief, are properly viewed as a vindication of pre-existing property rights and are parasitic on those property rights and so are inappropriate against a purchaser who takes free from the prior trust interests by virtue of the Torrens system in question: see "Knowing Receipt and Registered Land" by Matthew Conaglen and Amy Goymour in Constructive and Resulting Trusts (op cit)."

60. It follows that the Privy Council did not need to decide whether that assumption was right, because it decided that on the true construction of the statute any registered title was subject to "any duty or obligation to which [the proprietor] is subject as a trustee", which included the obligations of a constructive trustee arising at the time of receipt. The decision was explained in the following way:

"The expression 'any duty or obligation to which he is subject as trustee' is quite general and refers to the personal obligations of the registered proprietor as trustee whether or not the transferee is registered as trustee and even though beneficial interests under a trust are excluded from the register. The constructive trust for knowing receipt is imposed as a matter of law at the moment of receipt. On the appellant's case, the Property was received at the moment of his registration as proprietor. At that moment, all the essential requirements summarised by Hoffmann LJ in El Ajou were satisfied: there was a disposal of property in breach of fiduciary duty; the property was traceable as representing the property of the Crown; and the appellant had the requisite knowledge."

In context, it is clear that Sir Terence was not deciding that, apart from the particular terms of the Ordinance, Mr Arthur would be liable as a constructive trustee notwithstanding his registered title. His Lordship explained that, given that liability for knowing receipt arises in principle at the moment of receipt, such a liability would naturally fall within the words "any duty or obligation to which he is subject as a trustee". The Ordinance could not both confer perfect title and make the registered proprietor subject to such a liability.

61. The article by Conaglen and Goymour to which Sir Terence referred explains why, contrary to observations of the Law Commission when the Land Registration Act 2002 was prepared and enacted and some other judicial dicta, a registered proprietor of land in England and Wales, who had knowledge of an irregularity in the exercise of the transferor's powers of disposition, should not be liable as a constructive trustee of a beneficiary's interest in the land if his registration gives him priority over that interest. It will be necessary to return to that issue when addressing the claimants' argument that a personal claim in knowing receipt is unaffected by the perfection of the transferee's title to the property.
62. The next authority to which I need to refer is the decision of the Supreme Court in Akers v Samba Financial Group. As previously explained, the Supreme Court had to decide, first, whether there could in English law be a valid trust of property the lex situs of which did not recognise beneficial or equitable ownership, and second, whether there was in any event a disposition of SICL's beneficial interest for the purposes of s. 127 of the Insolvency Act 1986. In doing so, some of their Lordships made assumptions about the effect of Saudi Arabian law on the transfer to Samba of the Disputed Securities; others considered the matter assuming that English law applied. The appeal was argued on the assumption that Saudi Arabian law does not recognise the institution of the common law trust or a division between legal and equitable ownership.
63. The leading judgment was given by Lord Mance JSC, with whose reasoning (or at least conclusion) the other Justices agreed in varying degrees. Lord Neuberger of Abbotsbury PSC agreed with Lord Mance that there was no "disposition" of an equitable interest within s. 127, and further agreed with other issues discussed in Lord Mance's, Lord Sumption's and Lord Collins' judgments. Lord Sumption agreed with the result but not expressly with the reasoning of Lord Mance. Lord Collins agreed with Lord Mance's conclusion on the relevance of the Hague Convention and on the order proposed. Both parties seek to rely on various dicta in the judgments of Lord Mance, Lord Neuberger and Lord Sumption.
64. Lord Mance, addressing the nature of the equitable rights of a beneficiary, said at [18]:

"The initial inquiry is therefore whether an equity subsists, which it will prima facie do at common law, so long as the relevant property (original or substitute) does not pass into the hands of a transferee for value of the legal interest without notice of the equity. But a further issue may arise under the law of the situs of the relevant property."

His Lordship then explained that the situs of the shares of a company is the place of its incorporation or their registration, and continued at [20]:

"It is established by Court of Appeal authority (and was not challenged on this appeal) that, where under the lex situs of the relevant trust property the effect of a transfer of the property by the trustees to a third party is to override any equitable interest which would otherwise subsist, that effect should be recognised as giving the transferee a defence to any claim by the beneficiary, whether proprietary or simply restitutionary: Macmillan Inc v Bishopsgate Investment Trust plc (No 3) [1996] 1 WLR 387."

65. Samba draws attention to the words "whether proprietary or simply restitutionary" as indicating that the overriding effect of the lex situs will defeat not only a proprietary claim but also a personal restitutionary claim, such as the claim brought by the claimants. It seems to me that Lord Mance was summarising and accepting in principle the effect of the decision of the Court of Appeal in Macmillan v Bishopsgate, not establishing any wider proposition. What he meant by "restitutionary" is therefore coloured by the content of that decision. The Supreme Court did not however have before it for decision the question of whether a personal claim in knowing receipt could nonetheless succeed.
66. At [42], Lord Mance said:

"The issue before the court in the light of the expanded submissions which it has received is whether SICL has any basis for alleging that there was a disposition of property within the meaning of section 127. Viewing the matter in the light of the common law principles set out in paras 21-34 above, I would regard the present trusts not only as intended to create, but also as creating equitable proprietary interests in the Saudi Arabian shares, enforceable at common law at least as between SICL and Mr Al-Sanea and anyone else other than a transferee from Mr Al-Sanea in circumstances giving the transferee a good title under Saudi Arabian law. But, in the context of the present issues under section 127, there is to my mind a considerable case to be made for saying that it cannot matter. The definition of "property" in section 436 is wide enough to embrace both equitable proprietary and purely personal interests."

67. The conclusion is that valid trusts creating property interests were created, which were capable of being enforced by SICL against anyone other than someone who acquires better title to the Disputed Securities – under Saudi Arabian law, since this is the lex situs of the shares. However, whether the trusts created proprietary or personal rights did not matter for the purposes of section 127.
68. The claimants contend that this observation must be read in the context of a claim that only raised the invalidity of the transfer of the equitable interest in the Disputed Securities, and that Lord Mance therefore cannot have been considering whether a fault-based (or receipt-based) personal claim against Samba could nevertheless succeed. However, at [51], Lord Mance, in explaining the distinctiveness of legal and beneficial interests under a trust, says:

"Where an asset is held on trust, the legal title remains capable of transfer to a third party, although this undoubted disposition may be in breach of trust. But the trust rights, including the right to have the legal title held and applied in accordance with the terms of the trust, remain. They are not disposed of. They continue to be capable of enforcement unless and until the disposition of the legal title has the effect under the lex situs of the trust asset of overriding the protected trust rights. If the trust rights are overridden, it is not because they have been disposed of by virtue of the transfer of the legal title. It is because they were protected rights that were always limited and in certain circumstances capable of being overridden by virtue of a rule of law governing equitable rights, protecting in particular (under common law) bona fide third party purchasers for value…"

69. Lord Sumption in his judgment identifies these trust rights as two-fold: a true proprietary right and a personal right to have the trust administered according to its terms. He emphasises that the personal rights will survive the destruction of the proprietary rights since they are independent of each other. The claim against a third party recipient may be proprietary or personal. Although Lord Sumption notes that (because it does not recognise equitable ownership) the law of Saudi Arabia will treat Mr Al-Sanea as the owner of the entire interest in the Disputed Securities, he does not (unlike Lord Mance) mention the effect on the priority of Samba under that law. The reason was doubtless that, as he noted, at that time there was no claim to make Samba accountable as a constructive trustee. It is for that reason that, in [88] and [89], Lord Sumption considers the position of Samba in principle under English law. His concern is the issue of whether SICL's rights under the trust were disposed of.
70. Those paragraphs read as follows:

"SICL's problem is not that it lacked a beneficial interest in the shares but that Mr Al-Sanea did not dispose of that interest by transferring the shares to Samba. Mr Al-Sanea purported to transfer the legal interest to Samba. That was the only interest that he had. He did not purport to dispose of SICL's interest. Only SICL could do that, and it did not do so. The disposition of the legal interest did not itself extinguish any equitable interest of SICL in the shares. It only meant that that interest fell to be asserted against Samba, subject to the usual equitable defences. Samba's position in law was that it took the shares on a bare trust to restore them to the beneficial owner, unless it was a bona fide purchaser for value without notice. Since Samba gave value in the form of the discharge of Mr Al-Sanea's debt, its liability to restore the shares must depend on whether they are accountable on the basis of notice. Section 127 is irrelevant to the disposition of the only interest which matters for present purposes, namely SICL's equitable interest in the shares.

It is arguable, as Lord Neuberger of Abbotsbury PSC observes, that the transfer of the legal interest in movables may constitute a "disposition" of an equitable interest if its effect is that the equitable interest is extinguished. But the difficulty about the argument, and the reason why I would reject it, is that equitable interests arise from equity's recognition that in some circumstances the conscience of the holder of the legal interest may be affected. When the asset is transferred to a third party, the question becomes whether the conscience of the transferee is affected. On the facts pleaded in the present case, the equitable interest of SICL was defeated not by the act of the transferor (Mr Al-Sanea) but by absence of anything affecting the conscience of the transferee (Samba). The rules of equity which protect transferees acquiring in good faith and without notice are among the fundamental conditions on which equitable interests can exist without injustice."

71. The claimants interpret that passage as indicating that, were it established that Samba had notice of the breach of trust, there would be a valid claim to make Samba accountable as a constructive trustee. In my judgment, that interpretation assumes too much. Lord Sumption was concerned principally with explaining the nature of an equitable interest in property and in what circumstances it was liable to be defeated, which was of importance on the question of whether SICL's interest had been disposed of by the September Transfer. Although (as the claimants further submit), by allowing time to amend the particulars of claim following judgment, the Supreme Court recognised at least a real possibility of a personal claim in constructive trust, Lord Sumption did not consider whether such a claim could succeed against someone who was not equity's darling but who did acquire good title to the Disputed Securities under Saudi Arabian law.
72. I therefore do not consider that Akers v Samba Financial Group provides the answer to the Law of Knowing Receipt Issue.
73. In Courtwood Holdings S.A. v Woodley Properties Ltd [2018] EWHC 2163 (Ch), the claimant alleged (as assignee of a previous owner of a property) that numerous defendants received either the property or proceeds of its sale and other profits knowing of breaches of fiduciary duty by the agent of the previous owner. The claim failed on the basis that the sale of the property to the first defendant was not a direct result of the alleged breach of fiduciary duty, as Court of Appeal authority required (Brown v Bennett [1999] 1 BCLC 649). The Judge, Nugee J, gave an unreserved judgment at the start of the trial dismissing the claim on that basis, but he was persuaded the next day to withdraw it and hear the evidence. In his later reserved judgment, the Judge reached the same conclusion. He stated that the claim had been brought only on the basis of knowing receipt following a breach of fiduciary duty and quoted from his unreserved judgment at [201]:

"59. The foundation of the claim in knowing receipt seems to me quite different. It is that a person has got their hands on property which belongs to somebody else, in this case [SFPL]. If that is the analysis -- and I read a short passage from Lord Sumption's judgment in Williams [ie Williams v Central Bank of Nigeria [2014] UKSC 10] at [31] "the essence of a liability to account on the footing of knowing receipt is that the defendant has accepted trust assets knowing that they were transferred to him in breach of trust and that he had no right to receive them… His sole obligation of any practical significance is to restore the assets immediately." which suggests, when dealing with a limitation point, but that that was the analysis that he adopted, because he said that the obligation of the recipient was to restore the assets immediately -- the foundation of that is that the assets do not belong in equity to the recipient; and the foundation of the fact that the assets do not belong to the recipient in equity is that the transfer by which the assets were transferred is a flawed transfer. It may be a voidable transfer, it may indeed, for example if a company's assets are disposed of in a way that is ultra vires, be an entirely void transfer. But what gives the equity to the claimants is not the knowledge of the defendants by itself, or antecedent breaches of duty, but the fact that the transaction which is impugned is not one which transfers a good title to the recipient. It is in those circumstances that the recipient, unless a bona fide purchaser for value without notice, is liable, if he still has the property, to give it back, and can be made liable to account as constructive trustee, whether he still has the property or not, if he received it in circumstances that make his receipt unconscionable."

The judge then referred to Lewin on Trusts (19^th ed), at para 42-044, which stated:

"A breach of trust is usually essential because it is the basis on which the beneficial title is retained by the beneficiaries and does not pass to the recipient. If the beneficial title does pass to the recipient there is no occasion for the imposition of liability in equity on the recipient under the knowing receipt head of constructive trusteeship."

The same passage appears in the current 20^th edition at para 42-044.

74. The reasoning of Nugee J in Courtwood Holdings seems to me to support the argument of Samba that a claim in knowing receipt cannot succeed unless the claimant has a continuing proprietary interest following the impugned transfer. Thus, if the equitable interest of the claimant has passed (e.g. because the transfer is not itself a breach of trust) or if (as per the decision in Akers v Samba Financial Group) the equitable interest has been overridden or extinguished as a consequence of the transfer, the claimant no longer has the basis for a claim in knowing receipt, though he may have a claim for breach of trust against the transferor or a claim for dishonest assistance against the recipient. There is a consistent line of case law, starting with Macmillan v Bishopsgate and ending with Courtwood Holdings, where that has either been decided or assumed to be correct when specifically adverted to.

(d) The claimants' arguments

75. Against that review of the authorities, I come to the arguments advanced by the claimants as to why, as a matter of principle, the extinguishing or overriding of SICL's equitable interest should not preclude a claim in knowing receipt.
76. Their starting point is that proprietary and personal claims against a constructive trustee co-exist but are independent. The loss of a proprietary claim does not defeat a personal remedy; indeed, that is usually where the latter is most needed. That is clearly right, as far as it goes, in that the further transfer of the trust property and dissipation of the proceeds would defeat any proprietary claim against the recipient, though there would be no answer to a personal claim to make good the value of the trust assets.
77. Next, the claimants submit that it is important, when considering the authorities, to distinguish between what is in substance a proprietary claim and what is a personal claim because different rules and principles apply. They submit that this is not a case about priority of title but about an equitable wrong: taking property when one knows that it belongs to another, whose consent has not been given. They ask, rhetorically: how can an international bank with a branch in London accept shares from one customer, which it knows belong to another customer and are being transferred in breach of trust, just because the shares are in Saudi Arabian companies? If the shares were in English companies, Samba would not be a bona fide purchaser without notice of the breach of trust and SICL's claim would succeed.
78. The claimants submit that inability to pursue a property-based claim – because the property has been dissipated, or because someone has acquired better title to it – does not prevent them from pursuing a personal remedy to make good the trust, by payment of compensation. Thus, even if under Saudi Arabian law Samba acquired unimpeachable title to the Disputed Securities, that would not prevent the court from granting a personal remedy against Samba, requiring it to make compensation for SICL's loss. A claimant can avoid the issue of priority of title, as the claimants here have sought to do, by bringing their claim as a personal claim. In that case, they submit, the only relevant questions are whether Samba received the Disputed Securities with sufficient knowledge of the breach of trust.
79. The claimants submit that Macmillan v Bishopsgate was a truly proprietary claim, and that the El Ajou case demonstrates how and why an in personam claim is different. Foreign law is irrelevant to an in personam claim where the defendant is within the jurisdiction: the relevant law is then the lex causae, which here, by default, is Cayman Islands or English law. It is inappropriate, where such a personal claim is brought and the lex causae is determined to be English or Cayman Islands law, to have regard to foreign law at all. A personal claim is only defeated if a valid defence under Cayman Islands or English law exists, i.e. if the defendant is a bona fide purchaser for value without notice of the equity. The cause of action is complete at the moment of receipt of the property.
80. The claimants submit that, according to Millett J's taxonomy in Macmillan v Bishopsgate, their claim would be regarded as a category 2 claim, unlike the claim of Macmillan. Macmillan's claim was a category 1 claim because there was a proprietary claim that would have succeeded, subject to the defence of bona fide purchaser without notice. The claimants' claim is not a proprietary claim but is only a claim for equitable compensation. Alternatively, if on a correct analysis this claim is a category 1 claim, Millett J's taxonomy is wrong in principle.
81. I am not persuaded by this attempt to distinguish Macmillan v Bishopsgate. Macmillan's claims were both proprietary and personal, and for the reasons I have already given, the judgment must be taken to be addressing and disposing of both aspects of the claim. The approach of the judge was approved by the Court of Appeal. Although the claimants here have brought a personal claim only, Samba in fact retains the property in which -- if the claimants are right in their analysis -- their equitable interest remains. There is therefore no material factual distinction between the two cases. Further, there was no pre-existing equity between SICL and Samba, as was the case in Macmillan v Bishopsgate: the equity only arises if Samba holds SICL's property with knowledge that it was transferred in breach of trust.
82. This is a case in which, in Millett J's words, the claimants claim that Samba has been enriched by depriving SICL of its property, which enrichment results from a wrong done to SICL. The claimants bring this claim in order to recover the financial value of SICL's property that was taken from it: it is a claim on behalf of an equitable owner to recover its property or compensation for the failure to restore it, and therefore depends on SICL's continuing equitable interest. It is not a claim for breach of a fiduciary obligation owed by Samba to SICL. As Millett J said, it is no answer to assert that the claim is in personam and as such is governed by the lex fori or lex causae: the principled distinction between two different kinds of constructive trust claim and its consequences is part of the law of England ("... not principles of English conflict of laws, but of equity and our domestic law of restitution.").
83. The claimants' reliance on El Ajou as epitomising the difference between a personal and a proprietary claim for breach of trust is also unpersuasive. Foreign law had no application in that case, first, because it had not been pleaded and proved (p.715C), and second because the defendant was resident and the funds were received in England (p.716D-E). The live issues at trial were whether there was evidence that the funds represented the claimant's monies; whether the defendant had sufficient knowledge of the fraud; and whether the ability to trace in equity was lost because the funds had passed through civil law countries. The defendant did not suggest that it or previous recipients had better title than the plaintiff – it could not claim to be a bona fide purchaser without notice. Millett J considered that the issue of whether a charge could have been obtained against intermediate recipients of the funds was irrelevant. As Millett J himself commented in Macmillan v Bishopsgate:

"That case was concerned with tracing, not title. It was not disputed that the defendant received the money from the fraudsters - it had remained in their ownership or under their control throughout - and that they could not extinguish their victim's beneficial interest in the money of which they had defrauded him. The question was whether the money which the defendant received could be identified with the proceeds of the fraud, whether it represented other property of the fraudsters not derived from the fraud. The present is the converse case. It is concerned with title, not tracing. There is no doubt that the Berlitz shares in which the defendants claim security interests are the selfsame shares in which Macmillan's beneficial interest formerly subsisted. That is not in dispute. The question is whether any of the defendants has acquired an interest in the shares which is superior to that of Macmillan."

84. The claim of Mr El Ajou failed at first instance on the ground that the defendant, through its directors, had insufficient knowledge that the funds were the proceeds of fraud. That decision was reversed on appeal by the Court of Appeal, which nevertheless agreed with the Judge's conclusions on the tracing aspects of his decision.
85. The claimants also relied on the decision of North J in Mercantile Investment and General Trust Co v River Plate Trust, Loan and Agency Co [1892] 2 Ch 303, which they say was approved by the Court of Appeal in Macmillan v Bishopsgate. In the Mercantile Investment case, an American domiciled company owned land in Mexico and created an equitable charge on it for a debt, also secured by a debenture. The land was then transferred to an English company with notice of the charge and expressly subject to the payment of the debt. The debenture holders sued, claiming entitlement to a first legal charge and realisation of the land to pay the debt. The English company argued that the only remedy was against the land, by action in Mexico, there being no privity between it and the debenture holders. In other words, the English company was denying an in personam remedy in England.
86. North J, having referred to previous authority in 1796 (Lord Cranstown v Johnston 3 Ves 170) on the jurisdiction of the court to determine issues relating to foreign land, said:

"Applying that to the present case, I say it would be most unconscionable to allow the defendants here, who have registered their assignment in Mexico subject to the obligations created in favour of the Plaintiffs, who have obtained the land at a consideration measured to some extent by the existence of those obligations and the taking by the English Company upon themselves of the burden of satisfying those obligations; in my opinion it would be as unconscionable as anything could be to say that now, because they had registered their transfer before the hypothecation to the Plaintiffs had been registered, they are at liberty to set the Plaintiffs at defiance altogether. It was said that the case I have cited went upon fraud. Such a fraud as there was in that case would equally exist in the present case if the English Company were attempting to do what their counsel claims for them a right to do."

87. The court exercised its jurisdiction in personam to prevent the English company from acting unconscionably by taking the land while denying its obligation to discharge the debt. That jurisdiction could be exercised by granting relief to prevent the English company from dealing with foreign land inconsistently with its obligations. This was a straightforward case of relief being granted to prevent equitable fraud: the English company had only taken the land on the basis that it was liable for the debt. It was not alleged that it had received the land with notice of a breach of trust by another. It is notable that in the passage cited by North J from the judgment in Lord Cranstown v Johnston, the Master of the Rolls had said: "It is said, what if the sale [of the foreign land] had been to a third person? I am glad I have not to determine that." There was therefore no decision in either of those cases on the approach to the applicable law (or jurisdiction, as it would have been seen at that time) in the event that a third person's title to the land were involved.
88. The Court of Appeal in Macmillan v Bishopsgate referred to the Mercantile Investment case, but as an example of a case where, although the English court would not decide questions of title to foreign land, they would nevertheless grant an equitable remedy against a defendant within the jurisdiction (pp. 398G, 408B). The question in this case is different: does a claimant seeking a personal remedy against a knowing recipient of trust property need to have a continuing equitable interest in the property?

(e) The argument based on the Land Registration Act 2002

89. By the stage of closing submissions, the claimants placed considerable reliance on the effect of registration of title to land in England and Wales in support of their argument that no continuing equitable interest was necessary. They submit that although a registered proprietor's title is unaffected by any impropriety in the transfer to him or by any unregistered interest, so that his title is not subject to the beneficiary's proprietary interest, nevertheless the beneficiary can maintain a claim against the proprietor in knowing receipt if he had knowledge that the transfer was a breach of trust. Accordingly, they say, the law of England and Wales recognises that a claim for a personal constructive trust remedy may succeed even though, by registration, the recipient obtains priority over the beneficial interest in the property. This is a cogent and important argument that requires careful consideration of the material that is said to support it. If it is right, it would support the claimants' case that no continuing proprietary interest is needed under English or Cayman Islands law for a knowing receipt claim.
90. The background to this argument is a decision of the Privy Council, on appeal from the New Zealand Court of Appeal, Frazer v Walker [1967] AC 569. A wife forged her husband's signature on a re-mortgage deed and then defaulted. The mortgagee sold the property and the purchaser, who had acted in good faith without notice of the forgery, became registered as proprietor. He claimed possession from the husband. The judge and the Court of Appeal granted the claim. The 'indefeasibility of title' provided by the Land Transfer Act had only limited exceptions, including the case of actual fraud by the proprietor. The Privy Council dismissed the husband's appeal.
91. Lord Wilberforce, who gave the judgment of the Board, described the system of 'indefeasibility of title' as not protecting the proprietor against such cancellation or correction of title as the Act permits, or against claims in personam (p.580G) (though in that case the husband claimed no personal remedy against any respondent). His Lordship held that previous authority established that title registered pursuant to a void instrument of transfer is valid and immune to adverse claims, and then made a further observation in these terms:

"First, in following and approving in this respect the two decisions in Assets Co. Ltd. v Mere Roihi and Boyd v Mayor, Etc., of Wellington, their Lordships have accepted the general principle that registration under the Land Transfer Act, 1952, confers upon a registered proprietor a title to the interest in respect of which he is registered which is (under sections 62 and 63) immune from adverse claims, other than those specifically excepted. In doing so they wish to make clear that this principle in no way denies the right of a plaintiff to bring against a registered proprietor a claim in personam, founded in law or in equity, for such relief as a court acting in personam may grant. That this is so has frequently, and rightly, been recognised in the courts of New Zealand and of Australia: see, for example, Boyd v Mayor, Etc., of Wellington and Tataurangi Tairuakena v Mua Carr.

Their Lordships refer to these cases by way of illustration only without intending to limit or define the various situations in which actions of a personal character against registered proprietors may be admitted. The principle must always remain paramount that those actions which fall within the prohibition of section 62 and 63 may not be maintained."

92. It seems to me that, here, Lord Wilberforce was saying only that the decision reached by the Board was determined on the basis of the particular relief being sought and under the terms of the Land Transfer Act. His Lordship said that the Act did not determine what other in personam claims might succeed; only that the title conferred by the Act was immune from challenge, and so claims that challenged the proprietor's title could not succeed. The cases referred to demonstrate that title cannot be set aside (absent fraud) by a beneficiary on the basis of his own proprietary rights but that personal obligations affecting the transferee (such as under the contract of sale, or a pre-existing fiduciary duty, or arising from his own acts upon transfer) may be enforced against him. The distinction is shown by a passage in the judgment of the Privy Council given by Lord Lindley in Assets Company, Ltd v Mere Roihi [1905] AC 176 at 204-205, another case on appeal from New Zealand under the Land Transfer Acts:

"Then it is contended that a registered owner may hold as trustee and be compelled to execute the trusts subject to which he holds. This is true; for although trusts are kept off the register, a registered owner may not be beneficially entitled to the lands registered in his name. But if the alleged cestui que trust is a rival claimant, who can prove no trust apart from his own alleged ownership, it is plain that to treat him as a cestui que trust is to destroy all benefit from registration. Here the plaintiffs set up an adverse title and nothing else; and to hold in their favour that there is any resulting or other trust entitling them to the property is, in their Lordships' opinion, to do the very thing which registration is designed to prevent."

93. In my judgment, as regards personal claims Frazer v Walker does not establish more than that registration of indefeasible title does not rule out the possibility of such claims against the proprietor, though it does determine that a claim will not lie if it is based on a rival proprietary interest. The various types of possible personal claim were not explored. The decision does not establish that a knowing receipt claim can succeed notwithstanding registration.
94. Part 3 of the Land Registration Act 2002 deals with dispositions of registered land, including an owner's powers of disposition and the effect on priorities of a registered disposition. Section 26 provides:

"(1) Subject to subsection (2), a person's right to exercise owner's powers in relation to a registered estate or charge is to be taken to be free from any limitation affecting the validity of a disposition.

(2) Subsection (1) does not apply to a limitation –

(a) reflected by an entry in the register, or

(b) imposed by, or under, this Act.

(3) This section has effect only for the purpose of preventing the title of a disponee being questioned (and so does not affect the lawfulness of a disposition)."

Thus, an owner's registered title cannot be challenged on the basis that a transfer was void or voidable because of a limit on the power of the proprietor to sell the land. Sections 28 and 29 of the 2002 Act establish the rules on priorities, including that a purchaser for valuable consideration will upon registration have priority over earlier unprotected interests. Priority is a separate question from whether a transfer was in excess of owner's powers: see, on the distinction between title and priorities, Ali v Dinc [2020] EWHC 3055 (Ch) at [299] – [315].

95. The 2002 Act was preceded by a consultation paper and then a draft bill and commentary prepared by the Law Commission. The consultation paper, Land Registration for the Twenty-first Century (1998) Law Com No. 254 and the draft bill and commentary entitled "Land Registration for the Twenty-first Century: a Conveyancing Revolution" (2001) Law Com No. 271, address the question of whether the doctrine of notice should be relevant to priorities arising from land registration. The Law Commission considered that the law provided a wide range of personal remedies against those that in some way behave improperly, which it demonstrated by a number of examples, including the following:

"(1) If A transfers trust property to B in breach of trust and B knows or (perhaps) has notice of this, B is liable as constructive trustee for "knowing receipt" of trust property. Liability is personal and not proprietary and the obligation is to make restitution for the loss suffered by the trust. It has been assumed that this form of liability may apply where the trust property transferred is registered land and the rights of the beneficiaries have not been protected, so that as a matter of property law, the transferee takes the land free of the trust. "

A footnote refers to Eagle Trust plc v S.B.C. Securities Ltd [1993] 1 WLR 484 at 503-504, a decision of Vinelott J in a case about knowing receipt of money. The case concerned the degree of knowledge or notice that was required to be alleged and proved for a claim in knowing receipt. The passage identified by the Law Commission discusses how a purchaser of unregistered land is bound by proprietary interests of which he has constructive notice and the greater requirement of some degree of knowledge to support a claim in knowing receipt. The judge comments that if title is registered, questions of notice are irrelevant, but that it would be surprising if the beneficial owner had no remedy and anomalous if his rights were different depending on whether the land was registered or unregistered. Vinelott J then says that he expresses no opinion on such questions.

96. Paragraphs 4.8 to 4.11 of the commentary with the draft bill explain how, if the person exercising owner's powers did not have power to sell, the disponee's title could not be challenged but the disposition would not be rendered lawful. Paragraph 4.11 states:

"… where the disposition is in fact unlawful, the consequences of that unlawfulness can be pursued so long as these do not call into question the validity of the disponee's title. The example may be given of trustees of land, A and B, who had limited powers of disposition, but who failed to enter a restriction on the register to reflect this fact. If they transfer the land to a buyer, C, in circumstances that were prohibited by the trust, they would commit a breach of trust. Furthermore, although C's title could not be impeached, the protection given by Clause 26 does not extend to any independent forms of liability to which she might be subject. Thus if C knew of the trustees' breach of trust when the transfer was made, she might be personally accountable in equity for the knowing receipt of trust property transferred in breach of trust."

The only authority cited in support of this proposition is BCCI v Akindele, as to the degree of knowledge required for knowing receipt. That case says nothing about whether a claim for knowing receipt might lie against a registered proprietor or someone who acquires good title under foreign law.

97. The Explanatory Notes for the 2002 Act include exactly the same commentary on section 26.
98. In para 6-051 of Megarry & Wade: The Law of Real Property (9^th ed.), the following more cautious commentary is given:

"… the disponee's title cannot be questioned if it transpires that [the disponor] did not have power to make the particular disposition. This protection is limited to that purpose and does not affect the lawfulness of the disposition. This means that if, for example, trustees sold land without obtaining the consent of a beneficiary, which was required under the trust, where there was no restriction in the register to indicate the need for consent, the purchaser's title could not be challenged. The trustees would still have committed a breach of trust and would be liable for that breach. It is possible that, although the disponee's title could not be challenged, the disponee might be personally implicated in the breach of trust and be accountable in equity for the knowing receipt of property transferred in breach of trust."

99. The editors of Ruoff & Roper on The Law of Registered Conveyancing state, at para 13.003, that s. 26 of the 2002 Act only prevents the title of the disponee from being questioned. In a footnote, they paraphrase and refer to para 4.11 of the Law Commission's commentary, cited in [96] above.
100. The claimants also rely on an observation of Henderson J in Haque v Raja [2016] EWHC 1950 (Ch). That decision was made on an application by a respondent to discharge a proprietary freezing order on the ground that the applicant had no arguable case. The applicant argued that the previous registered owner of a property held it on resulting trust for him, and that that owner and the respondent dishonestly arranged the transfer of the property to the respondent in order to defeat his interest. The Judge held that there was a serious issue to be tried about the applicant's alleged interest, but that any such interest was overridden by the registration of the respondent as proprietor. The Judge also rejected as unarguable the alternative way in which the applicant put his case, namely that the respondent was a constructive trustee on the basis of dishonest knowing receipt of the property in breach of trust. Addressing that argument, Henderson J said:

"If the requisite degree of knowledge on the part of Mr Khan is established, his liability as a constructive trustee arises as a matter of law and attaches to the Property while it remains in his ownership. It is a liability which affects his conscience directly, and is not dependent upon the survival of the claimant's original beneficial interest as one which binds the Property in his hands. This way of putting the claim is therefore unaffected by the technicalities of overreaching and land registration, as Ms Williamson rightly accepted."

The Judge however rejected the inadequately pleaded allegations of dishonesty as having no real prospect of success and noted that the claim in knowing receipt had not been based on any other grounds.

101. It is evident that Henderson J considered that a claim in knowing receipt might be maintained against a registered proprietor whose title had priority under the 2002 Act over the beneficial interest. However, the claim in knowing receipt was based on allegations of dishonesty, which if proved would have established liability for dishonest assistance in a breach of trust. The matter was not argued on the basis of an equity arising from knowing receipt alone, and the point that an independent claim in constructive trust could in principle be maintained was conceded by counsel for the defendant, who had better arguments to pursue on an interim hearing.
102. The claimants however submit that, under the Land Registration Act 2002, a constructive trustee may acquire unchallengeable title to the property in priority to a beneficiary's interest in the property (assuming that the beneficiary has no registered interest or overriding interest under Schedule 3 to the 2002 Act) yet be amenable to a personal claim for knowing receipt in breach of trust. Such a claim may be advanced, they submit, not just in a claim for dishonest assistance in a breach of trust but also in a case where the equity only arises on receipt of the property with knowledge of a breach of trust.
103. This question has been carefully analysed in the context of the 2002 Act by Matthew Conaglen and Amy Goymour in an article entitled Knowing Receipt and Registered Land (Chap 5 in C. Mitchell: Constructive and Resulting Trusts (2010)) to which I have already referred. They consider whether the comments of the Law Commission can be right and whether a purchaser of land, whose title upon registration has priority over the equitable interests of the beneficiary notwithstanding the unlawfulness of the transfer, can be liable in knowing receipt. Their conclusion is that knowing receipt claims vindicate the beneficiary's property rights and that accordingly a registered proprietor with priority should not be liable in knowing receipt. The personal knowing receipt claim is necessarily parasitic upon the property right, which is postponed to the proprietor's interest. Further, the structure of the law of land registration intends the registered proprietor to have priority:

"… knowing receipt should be inapplicable against a registered purchaser who can claim the benefit of section 29. In general terms, outside the context of registered land, the knowing receipt claim is parasitic on the proprietary nature of the beneficiaries' equitable interests in the trust property: it is a claim to vindicate those property rights once they are no longer able to be vindicated in specie. The purpose served by such claims runs directly counter to the purpose of section 29, which is to protect purchasers from the effect of pre-existing interests irrespective of whether the purchaser has notice of those interests."

104. That conclusion is partly a function of the law of knowing receipt and partly a function of the policy of the 2002 Act. Where unregistered land is concerned, a purchaser with sufficient knowledge of a breach of trust could not be a bona fide purchaser without notice and so would not have priority over the beneficiary's equitable interest in the land. That means that a knowing receipt claim would not be defeated. But that is because the applicable law does not give priority to the recipient over the beneficiary's interest in those circumstances and the purchaser is subject to the continuing proprietary interest of the beneficiary. Contrary to the obiter observations of Vinelott J in the Eagle Star case (at [95] above), which were made in the context of the 1925 property legislation, it is not surprising that the policy underlying land registration produces a different result. The differences between the quality and effect of unregistered and registered title are greater as a result of the changes introduced by the Land Registration Act 2002. There is no longer any assumption that registered conveyancing is designed to or should have the same effect as unregistered conveyancing.
105. Conaglen and Goymour do, however, expressly state that registration of title should be no bar to a claim for dishonest assistance in a breach of trust. In such a claim, the basis of liability is not the receipt of trust property with knowledge but participation in a fraud. That could include a case where the receipt of property by a particular person in a particular country is an integral part of the fraud: the receipt is then incidental, being merely the particular form taken by that person's participation (see Dubai Aluminium Co Ltd v Salaam [2003] 2 AC 366 at [87], per Lord Millett, at [108] below).
106. I consider that the comments made by the Law Commission (referred to in [95] and [96] above) would therefore be accurate in such a case, but inaccurate in a case of 'pure' knowing receipt, where no dishonesty is alleged against the transferee whose property rights are given priority by the 2002 Act. The more cautious opinion expressed by the editors of Megarry & Wade is accurate if by "personally implicated in the breach of trust" is meant someone who gave dishonest assistance and not a 'pure' recipient with knowledge of a breach of trust. But, absent dishonesty or a pre-existing equity, a registered transferee of land, whose title has priority under the 2002 Act over the equitable interest of the beneficiary, is not liable in knowing receipt, even if he has knowledge that the transfer was made in breach of trust. Notice and knowledge are irrelevant to the scheme of priorities in registered land under the 2002 Act except to the extent that the Act otherwise provides (as it does e.g. in Sched 3, paras 2(c), 3(1)). The equitable interest of the beneficiaries is overridden by the registration of the proprietor and therefore there is no basis for an argument that the proprietor must deal with trust property as if he were a trustee, respecting the proprietary interest of the beneficiary.

(f) Conclusions

107. Underlying the claimants' arguments that there should in principle be a personal remedy is the contention that liability in knowing receipt arises from the wrongful receipt of property by the transferee; in other words that it is fault-based liability. The transferee knows (sufficiently) that the property belongs to another and that it has been wrongly transferred to him. He therefore is not entitled, as against the beneficiary, to receive and retain the property for himself. But in my judgment that argument fails to draw a sufficiently clear distinction between liability for dishonest assistance and liability for knowing receipt. As previously observed, the claimants were inclined to treat Samba at times as if it had assisted Mr Al-Sanea to steal the Disputed Securities from SICL, but that is not their pleaded case.
108. The distinction between dishonest assistance and knowing receipt is often blurred as a matter of fact because a defendant can be liable for both dishonest assistance and knowing receipt; but as a matter of law the distinction is clear: see BCCI v Akindele at pp. 448F-450F and the following observation of Lord Millett in Dubai Aluminium Co Ltd v Salaam [2003] 2 AC 366 at [87]:

"Dishonest receipt gives rise to concurrent liability, since the claim can be based on the defendant's dishonesty, treating the receipt itself as incidental, being merely the particular form taken by the defendant's participation in the breach of fiduciary duty; but it can also be based simply on the receipt, treating it as a restitutionary claim independent of any wrongdoing: see John v Dodwell & Co Ltd [1918] AC 563."

Developments in the law of constructive trusts in this area since 1995, when the Privy Council decided (in Royal Brunei Airlines Sdn Bhd v Tan [1995] 2 AC 378) that dishonesty was the essential requirement for ancillary liability for assisting a breach of trust, have emphasised the difference between claims in dishonest assistance in a breach of trust and claims of knowing receipt of trust property with sufficient knowledge of a breach of trust: see e.g. BCCI v Akindele at p.455E-F.

109. Dishonest assistance is truly fault-based – the equity arises from the dishonesty of the defendant in assisting a trustee to commit a breach of trust – but knowing receipt unconnected with dishonesty is different, at least at the moment of receipt. The recipient is not liable in such a claim for wrongly agreeing to receive the property. Although Lord Sumption referred to the receipt as being "wrongful" from the outset, that is in the sense that the transfer is a breach of trust and the recipient has no authority under the trust to receive the property. The principal duty of a knowing recipient is to deal with the property once received as if he were a trustee of it and to restore it to the trust, as Lord Sumption explained. It is unconscionable for him to do otherwise.
110. The knowing recipient's liability depends on his knowledge that the property he receives is trust property and is to be dealt with in that way. His receipt is not wrongful in the sense that he has acted dishonestly or culpably (unless he has also dishonestly assisted in the breach of trust), but his liability to deal with the property as if he were a trustee arises at the moment of receipt because of his knowledge that the property is trust property. If the transferee then deals with the property otherwise than as a trustee should (whether by failing to restore it to the trust or by dealing with it as his own) he is at fault and will be liable for the consequences. In those circumstances, a personal claim against the transferee can properly be said to be fault-based, but the reason for liability is that the transferee has knowingly dealt with (or retained) property that belongs to the trust inconsistently with his duty. If the property is not trust property, there cannot be liability of that kind.
111. In my judgment, the issue of principle – whether a knowing receipt claim that does not allege dishonesty requires the claimant to have a continuing proprietary base for it – is decided (in this court at least) by Millett J in the Macmillan v Bishopsgate case, which decision was approved by the Court of Appeal in that case. The claimant must be able to assert that the defendant received his property and was obliged to deal with it as if he were a trustee of it. As Millett J said in terms, that is a matter of the English law of knowing receipt, not a matter of English conflict of laws. If the recipient was from the outset entitled to deal with the property as his own, the claim cannot succeed.
112. No other case has decided this question as part of its ratio, but the dicta of Peter Gibson LJ in the Lightning case and of Nugee J in the Courtwood Holdings case encapsulate the principle: the transfer cannot be one that gives the transferee good title in priority to the beneficiary's interest. The Arthur case turns on the specific provisions of the Land Registration Ordinance, which prevented registration from doing that; and the Privy Council did not decide what the position would have been if that Ordinance did not expressly make a transferee's title subject to a constructive trust claim. The observations of their Lordships in Akers v Samba Financial Group cannot, in my judgment, be taken to be directed to the particular issue as it has crystallised in the amended claim in this case. Nor are the dicta of Henderson J in Haque v Raja directed to that issue.
113. The claimants nevertheless submit that the applicable law on the question of knowing receipt is the lex causae, which by default is here English or Cayman Islands law; and that accordingly the question of whether under Saudi Arabian law good title to the Disputed Securities passed to Samba is irrelevant. They submit that they have brought a personal claim, not a proprietary claim, and so Samba's title to the shares is irrelevant. But whether the lex situs is applied by English or Cayman Islands law depends on the issue that has to be determined. If, as a matter of English or Cayman Islands law, the decisive issue is whether SICL retained a proprietary interest in the Disputed Securities following the September Transfer, English and Cayman Islands law will, under their conflict of laws principles, apply the lex situs of the Disputed Securities to determine what if any rights of SICL in the Disputed Securities survived the registration of Samba. Accordingly, the argument that the claim is a personal claim under Cayman Islands or English law does not take the claimants any further.
114. There is scope for a difference of view (as demonstrated by the cogent arguments in this case) about the principle or principles that should apply on a 'pure' knowing receipt claim, in particular whether the beneficiary's ownership immediately prior to receipt together with the recipient's sufficient knowledge of a transfer in breach of trust should establish liability. However, as the law stands, authority and the greater weight of judicial observation appears to support the conclusion for which Samba contends, namely that a claim in knowing receipt, where dishonest assistance is not alleged, will fail if, at the moment of receipt, the beneficiary's equitable proprietary interest is destroyed or overridden so that the recipient holds the property as beneficial owner of it.
115. It is true that liability as a knowing recipient will then depend in such cases on the law applicable to the transfer of the trust property. Where it is English law, the beneficiary's proprietary interest will be undisturbed unless the recipient is equity's darling or the statutory equivalent. Where foreign law applies, the result may be different. However, it is not surprising that the rights of an equitable owner are vulnerable in different ways, according to different laws, if the trust property is foreign property: that is merely to give respect and deference in a principled way to the law by which third party rights in foreign property are governed. The fact that a Cayman Islands trust owning shares in foreign companies is a valid trust does not mean that the lex situs is irrelevant to a dispute with a third party about ownership of the shares. The claimants suggest that it would be bizarre for a knowing receipt claim to produce different results if a misappropriated basket of shares contained English and Saudi Arabian stocks. But that is because they regard liability as following from knowledge and receipt alone, without any requirement of a continuing proprietary base. If English and Cayman Islands law has that requirement, it means that the lex situs may make a difference to the result, but that makes foreign law no more a charter for thieves, as the claimants at one stage suggested, than it makes land registration a vehicle for fraud.
116. It also seems to me to be wrong in principle to treat a 'pure' knowing receipt claim (where no allegation of dishonesty is pleaded) as being unconnected with the proprietary rights of the beneficiary. It is of the essence of such a claim that the beneficiary asserts that the recipient has, or had, the beneficiary's property. That is so whether the claim is formulated as a proprietary or personal cause of action. In my judgment, a knowing recipient must have held trust property, not property to which from the moment of receipt he had good title. I consider it was this principle that underlay Millett J's analysis in the Macmillan v Bishopsgate case, distinguishing between those constructive trust claims that assert an equity based on a proprietary interest and those claims that assert an equity based on a prior relationship or on dishonest involvement in a fraud. While the Supreme Court may revisit the whole analysis of claims in knowing receipt, the Court of Appeal endorsed Millett J's approach and I consider that I should follow it.
117. I therefore decide the Law of Knowing Receipt Issue in favour of Samba and determine that, absent a continuing proprietary interest in the Disputed Securities at the time of Samba's registration, the claim in knowing receipt as pleaded will fail.

IV. The Saudi Arabian Law Issue

(a) Preliminary matters

118. My conclusion on the Law of Knowing Receipt Issue means that it is necessary to decide the effect under Saudi Arabian law of the September Transfer on the rights of SICL in the Disputed Securities. The claimants must prove that they still retained a proprietary interest in the Disputed Securities following the September Transfer, such that Samba held SICL's property and – by virtue of its knowledge of a breach of trust – became liable to deal with the Disputed Securities as if it were a trustee of them.
119. The Saudi Arabian Law Issue, as I have named it, conceals two separate questions, each of which needs to be decided:

i) First, under Saudi Arabian law, what was the effect of the September Transfer on the interest or rights that SICL had in the Disputed Securities? By default in this action, it was determined that prior to the September Transfer SICL was an equitable owner under a valid Cayman Islands trust governed by Cayman Islands law.

ii) Second, was any interest or right that SICL had, following the September Transfer, a proprietary interest in the Disputed Securities, sufficient under the English or Cayman Islands law of knowing receipt for the claimants to maintain such a claim against Samba?

120. It has been common ground throughout this litigation that Saudi Arabian law does not recognise (in the sense of give legal effect to) the common law trust as such. First, the courts of the Kingdom of Saudi Arabia do not apply foreign law and so, even though the Six Transactions created valid trusts, those courts would not apply Cayman Islands law to determine their effect. They would seek to give some effect to the Six Transactions in accordance with Saudi Arabian law (the parties disagree about how far they would go in doing so). Second, there is no distinction in Saudi Arabian law between legal and beneficial ownership in the way that the common law recognises it (but the parties disagree to what extent Saudi Arabian law recognises 'segmented' ownership at all).
121. It was on the basis of non-recognition of the common law trust that the validity of the trusts created by the Six Transactions was argued in Akers v Samba Financial Group and the Supreme Court held them to be valid. The starting point is therefore that SICL had an equitable proprietary interest in the Disputed Securities before the September Transfer, but that it would not have been recognised as such by the Saudi Arabian courts.
122. The claimants' first argument is that, since Saudi Arabian law does not recognise SICL's beneficial interest as such and does not apply foreign law, it cannot adversely affect the interest of SICL because under Saudi Arabian law there was no such interest. The claimants rely on a statement by Samba's expert witness that:

"As a matter of Saudi Arabian law the September Transfer had no effect on SICL's purported rights in the Relevant Securities, because SICL never had any rights in the Relevant Securities".

That is not merely a matter of semantics. If the effect of non-recognition is that proprietary rights under foreign law are left unaffected, SICL's proprietary interest continues. However, the expert witnesses were agreed that Saudi Arabian law would seek to give effect to the Six Transactions in terms of Saudi Arabian law, not fail to recognise them entirely. It is agreed that Saudi Arabian law recognises the transactions that gave rise to trusts under Cayman Islands law and gives them the effect that they would have had under Saudi Arabian law. It is not right, therefore, to say that Saudi Arabian law leaves untouched SICL's interest in the Disputed Securities.

123. That leaves the claimants with the argument that the trusts validly created under Cayman Islands law cannot be affected by Saudi Arabian law. That is true in one sense, in that the trusts remain governed by Cayman Islands law; but the conflict of laws rules of Cayman Islands law and English law, whichever is the governing law of the claim (and which it is makes no difference to the result), refers the effect of a transfer of shares in a foreign company to the lex situs. Whether Samba, as registered transferee of the Disputed Securities, has better title to them than SICL is determined by reference to Saudi Arabian law. That question depends on the priority that Saudi Arabian law gives a registered proprietor of shares over the rights of SICL, such as they are under Saudi Arabian law, or anyone else. The claimants therefore cannot succeed simply on the basis that Saudi Arabian law has nothing to do with the trusts.

(b) The Disputed Securities in Saudi Arabian law

124. It is necessary at this point to say something about the nature of the Disputed Securities, how such securities are held in Saudi Arabia and the law that applies to them.
125. The Disputed Securities comprise large holdings in five different Saudi Arabian joint stock companies ("JSCs"). All of them are large banks, including Samba Financial Group itself. The holdings on the date of the September Transfer were as follows:

Arab National Bank:	889,797 shares
Banque Saudi Fransi:	4,248,146 shares
National Commercial Bank:	2,129,250 shares
Samba Financial Group:	7,130,044 shares
Saudi British Bank:	3,319,346 shares

126. Of these JSCs, all but National Commercial Bank ("NCB") were listed and publicly quoted on the Tadawul (the Saudi Arabian stock exchange) at the date of the September Transfer. All these JSCs are now publicly quoted on the Tadawul.
127. The relevant legislation in Saudi Arabia governing the ownership of shares in JSCs was then the Companies Regulation 1965, now the Companies Regulation 2015. Article 99 of the 1965 Regulation allowed shares to be registered in the name of a shareholder or made out to a bearer. Bearer shares were uncommon and were abolished by the Companies Regulation 2015. Under both Regulations, shares in a closed JSC must be registered in the company's register of shareholders, to include the name, nationality, residence and profession of the shareholder and the amount paid up.
128. Article 102 of the 1965 Regulation, which applied at the date of the September Transfer (now superseded by article 109 of the 2015 Regulation), is central to the dispute between the expert witnesses on Saudi Arabian law. In the translation of a standard reference work, it provides:

"Nominal shares are transferred by registration in the shareholders register, prepared by the company, and containing the names, nationalities, domiciles and occupations of the shareholders, the serial numbers of shares and the part paid on account of the share. Said registration shall be endorsed on the share. The assignment of a registered share shall, in regard to the company and third parties, be reckoned the date of being entered in the said register. Bearer shares shall be negotiated by mere delivery."

129. Article 109(2) of the Companies Regulation 2015 states that negotiation of shares of companies listed on the Tadawul shall be in accordance with the provisions of the Capital Market Regulation 2003 ("the CMR 2003"). This supersedes article 102 of the 1965 Regulation so far as listed shares are concerned.
130. The official translation of Article 21 of the CMR 2003 provides:

"Securities listed on the Exchange shall be traded through transactions among brokers, each on behalf of its client, and shall be evidenced by entries in the Exchange records, in accordance with the provisions of Chapter Four of this Law, unless such transactions are excluded from trading pursuant to the rules and instructions issued by the Authority."

Chapter Four of the CMR 2003 provides for the establishment of the Securities Depository Centre ("the SDC"). Subject only to authorised exceptions, therefore, listed securities can only be traded in the SDC.

131. Article 25 of the CMR 2003 requires the Capital Market Authority ("CMA") to establish a committee known as the Committee for the Resolution of Securities Disputes ("the CRSD"), which has jurisdiction over disputes falling under the CMR 2003 and the other regulations, rules and instructions issued by the CMA and the Tadawul with respect to public and private actions.
132. Article 26 of the CMR 2003 establishes the SDC as the only entity in the Kingdom of Saudi Arabia authorised to operate deposits, transfers, settlement, clearing and registration of ownership of securities traded on the Tadawul.
133. Article 27 is central to the dispute between the experts and states, in official translation:

"a. The registration of ownership of Securities traded on the Exchange and the settlement and clearance of Securities shall be made by entries in the Depository Centre's records. Ownership of securities traded on the Exchange must be registered with the Depository Centre in order to be protected against third-party claims. The Depository Centre's records will also report pledges or other claims related to the Securities traded on the Exchange.

b. The Depository Centre shall be the sole entity to register all property rights in Securities traded on the Exchange. The final mentions reported in the records of the Depository Centre shall serve as conclusive evidence and proof of ownership of the Securities indicated therein together with the encumbrances and rights associated therewith, subject to the provisions of paragraph (d) of this Article.

c. Registration of ownership of Securities shall be effective from the time of final verification by the Depository Centre of the authenticity of the ownership documents. The Depository Centre shall promptly register all transactions effected upon being reported to and received by the Depository Centre with no delay. If the Depository Centre has reason to doubt actual or legal facts or consequences related to the registration of ownership or if the Depository Centre receives any notice that registration will cause damages to third parties, the Depository Centre may make a preliminary registration and, if it does so, it shall immediately commence an appropriate process to decide how the final registration for such Security shall be effected.

d. A person who believes that there is an error in the information entered into the registry so that the registry needs to be corrected or otherwise amended should make a written request to the manager of the Depository Centre or the person appointed by the manager to receive such requests. The Depository Centre shall correct or amend the registry after confirming the validity of the comments and information that are requested to be corrected or amended in the registry. Such correction or amendment can only be effected after notice and opportunity to comment by the personal persons the registry identifies as owning the Security, and giving them a reasonable opportunity to comment on the required correction or amendment.

e. The Depository Centre shall issue a certificate of registration upon request by the investor….

f. Complaints about decisions with respect to the registration of Securities listed on the Exchange shall be brought before the Committee."

The "Committee" is the CRSD.

134. There is no dispute between the parties that the above regulations apply, in the context of Islamic law, to determine the effect of the September Transfer.
135. The Listing Rules 2004, as amended, issued pursuant to the CMR 2003, require listed securities to be registered and settled centrally through the SDC (rule 12(c)). Samba asserted and the claimants did not dispute that such registration and settlement can only be carried out by authorised persons. An authorised person must not hold client assets unless it is authorised to provide custody services. In such circumstances, the title to the account must make clear to what client the assets belong. Where the securities are held with the SDC, they must be held in an account with the relevant client's name.
136. The Securities Depository Centre Rules 2012 are similar in effect to the CMR 2003, though they did not exist in that iteration at the time of the September Transfer. Article 21(1) provides that the electronic register "shall be considered as conclusive evidence of ownership of the relevant Security and any associated rights" and that "no other documents shall be recognised for this purpose". Article 34 provides that all restrictions on ownership and on "conducting any action on deposited Securities", and any pledge, must be registered, and that the securities are not considered subject to any such restriction or pledge until it is registered. It was not suggested that the previous rules applying at the date of the September Transfer were different in any of these respects.

(c) The parties' cases

137. Before the September Transfer, the shares in NCB are taken to have been registered in Mr Al-Sanea's name in NCB's register of shareholders. The shares in the four listed banks are taken to have been registered at the SDC in Mr Al-Sanea's name or on his behalf, with no restriction or interest of SICL registered. There is no pleaded case that Mr Al-Sanea was wrongly registered as the owner of those shares at that time. Although, under the law of the Cayman Islands, SICL could have directed Mr Al-Sanea to transfer them to another nominee, it had not in fact done so by September 2009. Mr Al-Sanea held the shares on trust for SICL, but there was no error at that stage in the share registers.
138. It is necessary to consider what rights in relation to the registered shares SICL had in Saudi Arabian law. It is at this point that the opinions of the two experts and the parties' cases diverge.
139. The claimants submit that:

i) the CRSD or a Saudi Arabian judge of the Commercial Court or any other court seized of the matter would well understand the character of a foreign law trust, since beneficial and nominal ownership and fronting of ownership on behalf of others are (and were in 2009) embedded in Saudi Arabian financial and commercial law;

ii) the CRSD or a judge would seek to give approximate effect to a trust, by analogy with other concepts and principles of Saudi Arabian law under which 'segmented' ownership of assets is recognised;

iii) the CRSD or a judge would therefore recognise that the true owner of the Disputed Securities was SICL, by virtue of the Six Transactions, and would give effect to SICL's rights not just against Mr Al-Sanea, who had direct obligations to SICL, but also against Samba, as a person who had acquired SICL's property with knowledge that it belonged to SICL;

iv) on the basis of SICL's rights against Samba, the CRSD or court would order rectification of the registers of ownership to give effect to SICL's claim, in the case of the listed shares pursuant to article 27(d) on the basis that Samba's registration was an error, and so SICL's proprietary rights prevail.

140. Samba submits that:

i) a Saudi Arabian judge would recognise the nature of the relationship between SICL and Mr Al-Sanea arising from the Six Transactions as being one under which Mr Al-Sanea had contractual obligations to SICL, which as a matter of Islamic law should be enforced;

ii) however, neither the CRSD nor a judge would go behind the registered ownership of Samba in order to give effect to dealings between SICL and Mr Al-Sanea, where those dealings were neither registered as giving SICL an interest in the Disputed Securities nor (in the case of the listed securities) carried out as a transaction at the SDC by an authorised person, in accordance with the CMR 2003 and the Listing Rules;

iii) the registration of Samba as proprietor of the shares in NCB and the listed securities was conclusive as against any third party, subject only to a power to correct the register for a clerical error or a mistake made within the SDC (in the case of the listed securities) in recording the settlement of the sale to Samba;

iv) the modern system of registration of ownership of listed securities, which was designed to compete with the stock exchanges of the west, would be undermined if registration could be upset on the basis of a claim that a registered transferor did not have power to transfer his registered shares because of an off-market transaction under foreign law between the transferor and a third party (SICL);

v) any knowledge of the transferee about the relationship between the transferor and the third party is irrelevant to the conclusive character of registered title: if SICL is the owner in Saudi Arabian law it does not need to rely on knowledge: it can rely on its ownership; if it is not the owner in Saudi Arabian law, it has no claim except against Mr Al-Sanea;

vi) so far as the NCB shares are concerned, the position is the same even though article 27 of the CMR 2003 does not apply.

(d) The expert witnesses on Saudi Arabian law

141. The expert witness called by the claimants was Professor Chibli Mallat. He is the principal of Mallat Law Offices in Beirut and is a Lebanese national, a member of the Lebanese Bar and an avocat at the Cour d'Appel in Beirut. He is now Presidential Professor of Law Emeritus at the University of Utah, but obtained his first doctorate at SOAS in London and was there subsequently director of the Centre of Islamic and Middle Eastern Law. He then concentrated on practice in Beirut and elsewhere in the Middle East for many years. He has also been a visiting Professor at Princeton and has taught Islamic and Middle Eastern law at both Yale and Harvard as well as many other distinguished law schools, and is widely published on those subjects. Between 2001 and 2017 he was the EU Jean Monnet Professor in European Law and Director of the Centre for European Studies in Beirut.
142. Professor Mallat is a native Arabic speaker but, although he has practised widely and been a consultant in many cases in different countries, he has not practised in the Kingdom of Saudi Arabia or been concerned in court proceedings in that country. It follows that he has not previously advised on or given evidence about provisions of the Companies and Capital Market Regulations in Saudi Arabia that are directly in issue in this case (he accepted that he had never seen the register of shares at the SDC), though he has expertise in Islamic finance law more generally. He has previously given expert evidence on Saudi Arabian law once, in the Cayman Islands proceedings between AHAB and SICL and others.
143. My impression of Professor Mallat, from the evidence that he gave, was that he was immensely knowledgeable, particularly about Islamic law but also about the fundamental tenets of English, French and Roman law. A better qualified student of comparative Middle Eastern and Western law it would probably be hard to find. There is no question in my mind that he was more than amply qualified to give evidence about Islamic law and did so with some authority. Professor Mallat was scholarly in approach, but able to provide a broad overview of legal principle. He was able to read and understand the relevant legal texts and Regulations in Arabic.
144. Professor Mallat is experienced in litigation in Middle Eastern countries, which contributed to his ability to explain important principles of Islamic law and their application across various countries' legal systems, but he did not have relevant knowledge of litigation in Saudi Arabia. He said that to some extent he relied on a professional colleague, Dr Adli Hammad, a Saudi national and practitioner at Hammad & Al-Mehdar law firm in Jeddah, for knowledge about Saudi Arabian practice and procedure; indeed, earlier expert reports on interlocutory issues were made by Professor Mallat and Dr Hammad jointly. The Court did not hear from Dr Hammad and cannot assess his contribution or the basis of his expertise.
145. I found that Professor Mallat was (subject to the issue addressed in section (e) of this Part below) persuasive on matters of Islamic law principle and how such principle infuses statutory law and impacts on the interpretation of statutory law; indeed, statutory law in Saudi Arabia would be regarded and interpreted as part of the Islamic law of that country. He was emollient in his approach to his evidence, treating other views with respect and seeking out areas of common ground, rather than sticking obstinately to his own opinions or written evidence. He was somewhat diffident in manner but quietly persuasive. I formed the view, however, that he was not well-placed to speak with real authority on the application and effect of the Companies Regulation and the Capital Market Regulation and related legislation, or on practice in the CRSD or the Commercial Court or relevant committees in Saudi Arabia, and further that the Saudi Arabian case law on which he relied was not always to the point or fully analysed.
146. Samba called Mr Andreas Haberbeck as an expert witness. Mr Haberbeck could not have been more different from Professor Mallat. In the first place, he is not an Arab and does not speak or read Arabic fluently. He is a long-standing practitioner in Saudi Arabia (over 32 years), whose practice largely involves commercial cases on behalf of international clients, and he is a specialist and has published a number of articles on maritime and shipping law and marine insurance in Saudi Arabia. Like Professor Mallat, he learned his Islamic law at SOAS and was then called to the Bar of England and Wales. After 8 years practising in City of London law firms, he moved to a firm in Jeddah and has remained as a practitioner in the Kingdom of Saudi Arabia since then.
147. He is not qualified as a lawyer in Saudi Arabia (or any Middle Eastern country) and only studied modules in Islamic law as part of LLB and LLM degrees in the University of London. He holds the status in Saudi Arabia of legal advisor (mustashar qanuni), by virtue of which he is authorised to advise and represent clients but not to appear as an advocate before Saudi courts. He therefore conducts litigation in Saudi courts and may attend courts, boards and committees that resolve disputes, but he accepted that his Arabic was in any event not good enough to act as advocate, or to understand everything that was said in hearings, though he was able to follow them. It was therefore curious that his expert reports included substantial appendices that were in Arabic only, which he accepted he could not translate.
148. It was however clear that Mr Haberbeck had considerable practical experience and knowledge of Saudi Arabian law, as practised in the courts of that country, including the Board of Grievances, the Commercial Court, the Committee for Resolution of Banking Disputes and the CRSD. He had himself drafted pleadings in cases before these courts except the CRSD and was currently involved in 25 cases in the CRSD. He has experience of the attitude and approach of judges in those courts. He therefore has experience of cases in Saudi Arabia raising issues about ownership of property and registered shareholdings. I was also persuaded from his oral evidence that he had a detailed and complete understanding of the cases that were relied upon by both expert witnesses in support of their opinions.
149. I found Mr Haberbeck to be direct, firm and self-confident in the way that he expressed his opinions, but that he was on occasions rather dogged in the defence of his views and rarely allowed himself to be taken away from them, to consider different points of view. He tended to see the cases that were cited and relied upon as compartmentalised, rather than as expressions in different contexts of broader principles, as was Professor Mallat's approach. So, e.g., he regarded the Jeddah Syrian market case on which Professor Mallat relied (case no. 302/2/J of 1409 Hijri) as being specifically a case about the abuse of powers by a government official, even though it established liability for interference in a contract with a third party, and as being irrelevant in other contexts (indeed, he said, such liability could not exist in other contexts).
150. It follows from the above summary that I heard evidence of Saudi Arabian law from two witnesses who were very different in their expertise, experience and approach. Nevertheless, they both assisted me in different ways to understand the approach that would be likely to be taken under Saudi Arabian law to the rights that SICL claimed adversely to the interest of Samba. I feel less confident about and place less reliance on two aspects of Mr Haberbeck's evidence. First, his evidence about underlying principles of Islamic law. Though the essentials are readily grasped, real expertise seems to me to depend on a broader study of the texts and an in-depth knowledge and understanding, such as Professor Mallat undoubtedly has. Second, his evidence about the meaning of statutory provisions, which would depend to some extent on an ability to read Arabic as a first language. Correspondingly, I feel uneasy about relying on Professor Mallat's opinions on the approach that would be likely to be taken by the CSRD in particular, and Saudi Arabian courts more generally, to claims adverse to registered shareholders and to the operation of the SDC, the CMR 2003 and associated statutes, because he had no relevant experience of those matters.

(e) The attacks on the credit of the expert witnesses

151. There was a sustained attack on the credit of both expert witnesses on the basis that the opinions that they expressed in their reports prepared for trial were notably different from the opinions that they appeared to express in reports prepared for the resolution of interlocutory issues in the original claim in 2013. In 2013, the parties were engaged in a procedural struggle about whether the claim should be struck out, as being bound to fail, or stayed on the basis that Saudi Arabia was the forum conveniens for the then claim that the September Transfer did not vest SICL's equitable interests in Samba. The joint official liquidators contended that the governing law was Cayman Islands (or English) law and that they could not have a fair trial in Saudi Arabia because its law did not recognise proprietary beneficial interests or give effect to the concept of trusts. Samba contended that the governing law was Saudi Arabian law, which does "recognise" such interests and so SICL could bring a claim there to assert its proprietary rights.
152. Professor Mallat's 2013 report says that "[i]nsofar as the common law trust concept is concerned, it is correct to say that the law of Saudi Arabia does not recognise the concept of trusts"; that it is somewhat similar to the concepts of "waqf" and "amaana" but that the Six Transactions were not attempting to nor were they capable of creating such a structure – they were "incomprehensible in a Saudi Arabian legal context as waqf"; and none of the meanings of amaana were capable of encompassing arrangements such as the Six Transactions because amaana is not a split between a legal owner and a beneficial owner. The common law trust principles pleaded in the Particulars of Claim "have no application in Saudi Arabian law". The report did not in terms say (other than in relation to the claim to interest) that the claim as brought could not be brought before a Saudi Arabian court, but in written and oral submissions to the Chancellor in January 2014 it was contended by the joint official liquidators on the strength of Professor Mallat's evidence that SICL could not get relief in Saudi Arabian courts because it had no recognised proprietary interest in the Disputed Securities under Saudi Arabian law.
153. In his expert report for the trial, Professor Mallat says, by way of summary, that a Saudi Arabian court would not ignore SICL's rights arising under Cayman Islands trusts, but would characterise them by reference to concepts analogous to the trust concept, such as mahassa, amaana, wadi'a and waqf; and would have no difficulty applying such concepts by analogy, to give effect to SICL's benefit (manfa'a); and that where shares are transferred without a beneficiary's authority to a third party, the transfer is rescindable by the beneficiary. In more detail, Professor Mallat explains that a judge would use the concepts of waqf and amaana to enable him to understand common law trusts and give effect to them. A judge would be more concerned with the overriding concepts of harm and fault and seek to fashion a remedy:

" 'Characterising' the Six Transactions into categories familiar to him in Saudi Arabian law, a Saudi Arabian judge therefore has several routes open to him. In the presence of these instruments, the judge will not accept that the Six Transactions had no effect whatsoever and rule as if they simply did not exist. Nor will he tolerate a harm done to a party without providing a remedy. The judge would consider that SICL had sufficient interests concerning the Disputed Securities, such that he would permit it to bring a claim in respect of them."

154. Professor Mallat then explains in his report that, having brought such a claim, if SICL could establish that Samba had knowledge of its beneficial interest and so of Mr Al-Sanea's lack of authority, the court would order rescission of the September Transfer and rectification of the register of shareholders to vest the shares in SICL's approved nominee.
155. Taxed in cross-examination with apparent inconsistency between the 2013 report, which said that SICL did not have an interest that a Saudi Arabian court would recognise, and the trial report which said that SICL would succeed in a claim in Saudi Arabia based on its ownership of the Disputed Securities, Professor Mallat said that beneficial ownership as understood in English law is alien to Saudi Arabian law but that all the evidence that he gave in 2013 was correct, but nevertheless a Saudi Arabian judge would give effect to SICL's "manfa'a", on the basis that it is an interest or use of another's property. The only other point that he made to explain a change between 2013 and 2020 was that since 2013 very many case reports have been published in Saudi Arabia for the first time, which has given him more of an insight into the way that judges decide cases.
156. I find that there is inconsistency between Professor Mallat's opinions in his 2013 report and his trial report. The 2013 report is at least materially incomplete, if the trial report contains Professor Mallat's complete and honest opinions, and the trial report must be doubtful if the 2013 report contained his complete and honest opinions and was correct as expressed, as Professor Mallat insisted that it was. I accept Samba's argument that, had Professor Mallat's opinions as given in 2020 been expressed in his 2013 report, the joint official liquidators could not have argued that they could not have a fair trial of their proprietary claims in Saudi Arabia. I cannot accept that the further opinions of Professor Mallat in his trial report are attributable to the publication of reports of cases, since Professor Mallat expresses them largely as matters of first principle of Islamic law about theft, honouring agreements and violation of authority. In any event, Professor Mallat did not consider that there was any material difference in his opinions.
157. This raises awkward questions about how and why the two different expert reports are so different. There are various possible explanations but, whatever is the real reason, the reliability of Professor Mallat as a witness supposedly giving his unvarnished, independent opinion is adversely affected. I must therefore exercise some caution about what parts of his evidence I am able to accept.
158. The claimants correspondingly attacked the credit of Mr Haberbeck for expressing different opinions in his 2013 report from those in his trial report.
159. In his 2013 report, Mr Haberbeck said at para 34:

"I disagree with the following statement set out in Paragraph 24.1 of the Particulars of Claim: 'The law of Saudi Arabia does not recognise the concept of trusts'"

He further said in that report that the institution of waqf has many elements of a trust and that the concept of trust is often used in connection with the status of amaana, but that it was not a trust as understood in English law, and that the terminology in the Six Transactions would be recognised by a Saudi Arabian judge as an attempt to create an amaana, and that use of terminology such as trustee and nominee was consistent with that. His conclusion was that SICL could not lawfully (under Saudi Arabian law) have acquired an interest in the Disputed Securities and that neither a waqf nor an amaana had been validly instituted; but that the claimants "would be entitled to bring proceedings against SAMBA in Saudi Arabia in relation to SICL's alleged ownership of the Relevant Securities, and SICL's alleged entitlement to compensation against SAMBA…".

160. Mr Haberbeck was asked in cross-examination to explain how that evidence squared with paragraph 60 of his trial report:

"Saudi Arabian law does not recognize the concept of a trust within the meaning of the term under Article 2 of the Convention on the Law Applicable to Trusts and on their Recognition dated 1^st July 1985 …"

He affirmed that Saudi Arabian law does not recognise the concept of a trust or beneficial ownership, but that his two reports are not inconsistent because his 2013 report was written in the context of whether a Saudi Arabian judge would understand what the parties (to the Six Transactions) were attempting to do. However, the 2013 report was addressing various, separately identified issues, including:

"(d) Whether Saudi Arabian law recognizes the concept of trust as understood in Common Law jurisdictions;

(e) Whether the terminology used in the Share Sale Agreements and the Declarations of Trust has application in Saudi Arabia and whether the purported sales of the Relevant Securities were valid under Saudi Arabian law",

and paragraph 34 of the 2013 report was under the heading "The Concept of Trust under Saudi Arabian Law".

161. Mr Haberbeck then explained that common law trusts are "recognised in the sense that Saudi Arabian law recognises their existence. They are not enforceable under Saudi law." He further explained that the assertion in his trial report that SICL had no rights that are recognised by Saudi Arabian law meant that they were not recognised because they would not be given effect by a Saudi Arabian court: rather like a gambling debt, there would be no enforceable right - "if you wanted to go against a third party, like Samba here, you would have no enforceable right. It would in that sense not be recognised". The trust would only be 'recognised' (not in the sense of enforced) as a fronting arrangement (day 3 transcript, pp. 387-391).
162. Mr Onslow QC submitted that the charge of giving inconsistent evidence was unfounded, on the basis that in his 2013 report Mr Haberbeck was merely saying that there was no jurisdictional bar to SICL bringing a claim in Saudi Arabia and was not addressing the merits of the claim; and that the point about recognition of trusts was about recognition of their existence, not giving effect to them. There may be partial force in that argument because in his first report Mr Haberbeck did say that Mr Al-Sanea's holding shares for SICL was unlawful under Saudi Arabian law and that no property in the shares passed to SICL. However, it is clear that if Mr Haberbeck's current evidence is right he was not giving the full picture in his 2013 report, and that using the word "recognise" without fuller explanation is apt to mislead. Mr Haberbeck did not say in 2013 that Saudi Arabian law would not enforce the interest of a person under a trust against someone other than the trustee, and that there would therefore be no valid claim, which is now his evidence. Had he given that evidence in 2013, as he should have done if his current evidence is true, Samba would have been in real difficulty in arguing that the claimants could have an effective trial of their proprietary claim in Saudi Arabia.
163. There is also doubt in my mind about the completeness of Mr Haberbeck's evidence at trial. He now takes the position that Saudi Arabian law does not "recognise" the common law trust in the sense of giving any effect to the interests of beneficiaries. The position is more nuanced than that, as shown by the detailed evidence of Saudi Arabia's recognition of the existence of trusts, beneficiaries and nominees in the context of financial, banking and companies legislation, which require persons to have regard to such matters. It is fair to say that Mr Haberbeck exhibited much of that legislation, but he did not explain its significance.
164. It is evident from that material that the statutory law of Saudi Arabia requires effect to be given in various ways to interests of persons who are not the legal owners of property. For example, the Rules Governing Anti-Money Laundering & Combating Terrorist Financing and the Rules Governing the Opening of Bank Accounts impose strict requirements on banks and money exchangers to monitor the use of offshore structures, including trusts, to obscure the identities of those controlling funds and to comply with 'know your customer' principles to identify the beneficial owners of all accounts and transactions, with specific requirements on trustee and nominee accounts and definitions of "beneficial owner", "nominee" and "trustee". Even more pertinently, the CMA's own Anti-Money Laundering and Counter-Terrorist Financing Rules (2008) require authorised persons to identify and verify beneficial ownership and control in relation to all clients, and the CMA Listing Rules require a person to be treated as a beneficial owner of shares if he has the ultimate beneficial ownership or control of the shares, whether through a chain of companies or otherwise (art. 49). The CMA's Authorised Persons Regulations (2005) impose on authorised persons fiduciary duties that are immediately recognisable to any equity lawyer, and the Regulations provide for authorised persons to hold shares in their names on behalf of identified clients. So there is no doubt that the CRSD or a Saudi judge in the Commercial Court or Committee for the Resolution of Banking Disputes would understand that a fiduciary may hold shares on behalf of the real owner, albeit that Saudi Arabian law provides for the real owner to be identified in the register. In his report, Mr Haberbeck suggested that all this was concerned only with a registered owner acting as a front for another person, but it clearly goes wider than this.
165. I am therefore left with real doubt about the completeness of Mr Haberbeck's evidence. I have the same concerns as in the case of Professor Mallat about how contact between expert witness, client and lawyers could have led to incompleteness and inconsistency of the expert evidence. I also have doubt as to the depth of Mr Haberbeck's expertise on general principles of Islamic law, as distinct from commercial and securities law and practice in Saudi Arabia. For example, he expressed a strong view that Professor Fadel, the author of Fiduciary Principles in Classical Islamic Law Systems (Chap. 28 in The Oxford Handbook of Fiduciary Law) and a respected scholar of Islamic law, was wrong on certain matters of basic principle, and based the contrary view in part on something that he was taught by a professor at SOAS over 40 years previously.
166. I therefore exercise caution too in accepting Mr Haberbeck's evidence.
167. That potentially leaves me with difficulty in deciding whose evidence to accept as probably right. However, despite some headline differences of opinion, there was a significant degree of common ground between the experts, and, in the case of both expert witnesses, some answers (including concessions) given in the course of cross-examination on which I feel that I can safely place reliance. I heard both experts being cross-examined at length, and the way in which they answered questions and explained some of the cases referred to gave me significant help in assessing the credibility of their opinions on certain issues and sub-issues that arise in this case. There is also, as regards the CMR 2003 and associated legislation, an inherent likelihood of certain evidence about it being correct, because the purpose of the legislation was not in dispute: it is designed to support an efficient, transparent, secure and reliable stock exchange on a modern western model, and to facilitate competition with other such markets, to which objectives uncertainty or insecurity would be inimical. Although I warn myself against seeking to interpret words in the official translations of the legislation as if they were original English legislation, the purpose of the statutory provisions is generally self-evident and forms a useful basis on which to consider the likelihood of the evidence that I was given by one or other expert being probably right.
168. I also conclude, having heard the detailed evidence of both witnesses, that I can and should give more weight to the evidence of the expert who is likely to have greater experience and expertise in relation to a particular matter or question. That is, of course, subject to the caution that I have previously voiced about placing reliance on each of them, and so my assessment depends not just on what they said but on matters of inherent plausibility of their opinions, their persuasiveness and consistency in explaining their opinions orally and the extent to which they are supported by exhibited material. In general, on the basic principles of Islamic law and the way in which they would be understood and applied by a judge in a particular context, Professor Mallat as a scholar and Arabic speaker with wider experience of Islamic law in practice across the Middle East is in my judgment likely to be more reliable in his assessment; whereas Mr Haberbeck is in my view much better placed to provide an informed and reliable opinion on how in practice the CRSD, the Commercial Court and various judicial committees in Saudi Arabia will apply the capital markets and companies legislation.

(f) The sub-issues for decision

169. The experts agree in their joint statement that Saudi Arabian law comprises Islamic law and Saudi Arabian legislation, as interpreted by competent Saudi Arabian courts, and that under Saudi Arabian law the Six Transactions are agreements between SICL and Mr Al-Sanea. They also agree that Saudi Arabian courts do not apply foreign law, but that a court would seek to give effect to the Six Transactions according to a Saudi Arabian law understanding of their nature.
170. It is not in dispute that the Companies Regulation 1965 governed registration and ownership of the NCB shares in 2009 and that under that Regulation Samba had at least prima facie good title to the shares from the date of its registration. The same is agreed in relation to the registration of the listed securities, and it is agreed that the ability of SICL to challenge Samba's ownership of those shares depends entirely on article 27(d) of the CMR 2003. There was no disagreement that listed shares are traded through transactions by authorised brokers and concluded by registration, save for expressly exempted cases.
171. The main issues between the parties' experts are whether registration in NCB's share register is conclusive of ownership or only prima facie evidence and susceptible to correction or rectification, and if so in what circumstances; and as regards the listed securities, what kind of error may form the basis of a successful challenge to the registered title of the owner. One important underlying issue is whether the CRSD or a Saudi Arabian court would give the Six Transactions only contractual effect as between SICL and Mr Al-Sanea or whether, by analogy with one or more constructs of Islamic law, it would characterise SICL's interest as being one of ownership and capable of being asserted as such against a third party such as Samba. Clearly, if Mr Haberbeck is right in saying that the Six Transactions would only be recognised as having contractual effect, SICL would have no claim against Samba.
172. In their closing written submissions, the claimants suggested that there were only two remaining questions of expert evidence with which the court had to engage:

i) whether Saudi Arabian law would protect the claimants' undoubted interests under the Cayman Islands trusts; and

ii) whether registration of the Disputed Securities gives a completely indefeasible title in those shares to Samba free of SICL's interest.

They contend that the answers to these questions are 'Yes' and 'No' respectively.

173. I consider that, in view of the evidence of Saudi Arabian law given by the expert witnesses and the common ground that exists, the relevant questions to be addressed now are the following:

i) Would Saudi Arabian law characterise SICL's rights as being those of an owner of the shares, or only those of a person to whom Mr Al-Sanea owed personal obligations under an agreement (in other words: how would the Six Transactions be understood and characterised in a Saudi Arabian court) ("characterisation")?

ii) If the answer is "the interest of an owner", does Islamic law (and therefore Saudi Arabian law) in principle provide a remedy for SICL against a third party purchaser of SICL's property, if the third party knew of the lack of authority of Mr Al-Sanea to sell ("remedy against third party")?

iii) If the answer to (ii) above is that such a remedy would in principle exist, whether the effect of the Saudi Arabian legislation governing share registration is to preclude that remedy against a registered proprietor in a case such as this, where SICL's interest derives from an unregistered, off-market transaction ("effect of registration").

The legislation governing share ownership is part of the law of Saudi Arabia and is to be interpreted consistently with Islamic law, so it may be artificial to separate out sub-issues (ii) and (iii) in this way, but nevertheless it is a convenient way of analysing the differences between the experts.

(g) Characterisation

174. On the first issue, Mr Haberbeck contends adamantly that a Saudi Arabian judge would only give effect to the Six Transactions under Saudi Arabian law as a contract. However, his principal reasons for that, as originally formulated, were that the holding of shares by Mr Al-Sanea for SICL was unlawful and that a beneficial interest under a common law trust would not be recognised in Saudi Arabia. Given that in this action the trusts are to be treated as valid and that it is clear in the context of domestic capital markets and share ownership that Saudi Arabian law recognises the concepts of beneficial ownership, trustees, nominees and the fronting of transactions for the true owner, I am satisfied that Professor Mallat's opinion is probably right, namely that a Saudi Arabian judge would regard the Six Transactions as more than just a contract between SICL and Mr Al-Sanea and instead as making SICL the underlying owner of the trust property. It is true that Professor Mallat played down the distinction between property rights and personal rights, as not being the dichotomy that it is under English law, but his point was that the main focus would be on SICL's rights and the harm that had been done to them, not on whether what SICL had was property:

"… he [the judge] will look at the contracts [the Six Transactions], he will look at the intent of the parties and he will say there is an interest there. He will recognise that this sort of interest can be a mahassa, a bailment or a waqf without ascribing – and it may be justified of my opacity in this [sic] because the judge will not worry about this. He will look at the contracts and will want to give them effects to see whether SICL has a right somewhere hidden there and if it is the right, what nature is this right. And if this right has been taken away from it unlawfully then it has to be remedied."

175. It is evident from the modern legislation associated with banks and securities that property rights are well understood as a concept, and the protection of property and money is a fundamental tenet of Islamic law. This is established by foundational texts from the Qur'an and the hadiths – "believers to honour contracts" and "not to eat each other's monies unlawfully" are the best known Qur'anic verses, and are supplemented by important hadiths including "the interdiction to sell what one doesn't own" and "the need for the hand to return what it has unlawfully taken".
176. I do not find the argument of Mr Haberbeck that Saudi Arabian law would "recognise" a common law trust but give it only contractual effect convincing. It is clear that Saudi Arabian statutory law has – to the extent necessary in modern business life – embraced the concepts of true underlying ownership and registration in the name of a front or nominee, and takes account of the fact that these same questions arise in the context of trusts and similar 'offshore' legal structures. The detailed references to trusts, beneficial owners, nominees and the like in the Saudi Arabian capital markets and banking legislation, cited above, demonstrates that these concepts have significance within the Kingdom of Saudi Arabia as regards ownership of and dealings in Saudi Arabian shares. Commercial Courts in Saudi Arabia clearly need to understand and apply concepts such as beneficial ownership in order to give effect to Saudi Arabian domestic legislation. It would be incoherent of the law to have regard to such matters in connection with finance and share transactions and yet to have no regard to such concepts of underlying ownership when seeking to characterise in Saudi Arabian law terms a lawful 'offshore' transaction in shares of Saudi Arabian JSCs. Both expert witnesses persuaded me that Saudi Arabian law is far from incoherent in this or any other relevant respect.
177. I also consider that Professor Mallat is probably right that a Saudi Arabian judge would – as a result of familiarity with concepts such as waqf, amaana and muhassa, which to varying degrees recognise and give effect to different rights of parties in the same property – be able to understand that SICL would have ownership rights in the trust property, even though Mr Al-Sanea was the apparent owner of it, without needing to find a precise analogue in Islamic law for the interest that SICL had. A judge would be able to use a variety of imprecise analogues to understand the intended effect of the Six Transactions, namely that SICL would be the real owner and Mr Al-Sanea the front with fiduciary obligations. In terms of SICL's ability to control, benefit from and direct Mr Al-Sanea how to deal with the shares, that much is expressly stated in the share sale and trust documents and would readily be understood as written.
178. My conclusion is more likely to be correct in the case of the CRSD or a judge of the Commercial Court, or indeed the Committee for Banking Disputes, where, according to Mr Haberbeck (whose evidence on this point I accept), a claim by SICL would be heard. The judges in those courts or committees would be likely to be familiar with the detailed provisions in the money-laundering, 'know your client', banking and capital markets legislation addressing – and requiring banks and other commercial parties to recognise and take account of – underlying ownership.
179. Samba argues that the court should not accept Professor Mallat's opinion because, ultimately, he placed his reasons firmly on the basis that a Saudi Arabian judge would recognise that what SICL had was a "manfa'a", i.e. the Disputed Securities were being held for its benefit, whereas, as he ultimately accepted, manfa'a is (in Roman Law terms cited in Professor Schacht's Introduction to Islamic Law) usufructus, not ius in re aliena, so that it does not give rise to any ownership interest in SICL, only the benefit of use of the assets. The debate on this issue underlines that there is no precise analogue in Saudi Arabian law for the common law trust. I do not find that Professor Mallat was contending that the case was only one of manfa'a, which the judge would understand and apply directly; but rather that manfa'a is a concept of Islamic law that would help a judge to understand the nature and effect of the Six Transactions. In any event, the concept of fronting and concealed but underlying ownership in domestic Saudi Arabian law provides a sufficient and good analogy.
180. For these reasons, I do not accept that a Saudi Arabian judge would be likely to treat SICL's interest as being only a matter of contract between it and Mr Al-Sanea. I consider that it is probable that a judge would recognise that SICL had rights of an owner in the subject-matter of the Six Transactions. Although I accept Mr Haberbeck's evidence that the Commercial Court restaurants case on which Professor Mallat relied (case no. 13509 of 1438 Hijri) was about the ownership of licences rather than land itself, the licences were nevertheless registered in the name of the Saudi front, whereas the underlying and true owner was treated as the foreign investor. The Commercial Court gave effect to that underlying ownership. The latter's rights prevailed because the Commercial Court recognised that he was the true owner and was entitled to a remedy against the front. That case was not of course one that involved a third party, but it depended on the court's recognition of formal rights (business licences) being held by one person on behalf of another.
181. I therefore decide the first issue, characterisation, in favour of the claimants: a Saudi Arabian judge would characterise SICL's interest under (assumed) lawful Cayman Islands trusts as an ownership interest, not only a contractual right.

(h) Remedy against third party

182. The second issue I must decide is whether – apart from the statutory provisions for registration of shares – SICL's underlying ownership would enable it to claim its property from a third party, assuming that the third party acquired it from Mr Al-Sanea with knowledge of the fact that he did not have authority to deal with it. That is a broad question of Islamic law about the extent to which an owner's rights can be vindicated where he has suffered harm - not against the person who directly caused it but against someone who received the property with knowledge of the lack of authority.
183. Basic norms of Islamic law to which I have referred suggest that there should be a valid claim. Professor Mallat said that the famous mainstay of Islamic law, that one cannot "eat the monies of another unduly", establishes that vindication of ownership is a very strong element of the legal system and that in general ownership cannot be violated without a remedy being available. Mr Haberbeck said that knowledge was irrelevant, since if a claimant is the owner he simply relies on his ownership (a claim called ghasb). He denied the availability of a general tort claim, which he says is limited to personal injury or damage to property.
184. Professor Mallat had two parallel theories. One was that there is an overriding principle of Islamic tort law that where harm has been suffered the Saudi court will provide a remedy against someone who has caused it. The second is that an owner whose property is sold by an agent without authority can rescind the sale and recover the property from anyone who bought with knowledge of the lack of authority. The two principles seemed to connect, in that a buyer who knows that his seller did not have authority to sell will not be able to say that he did not cause or was not a cause of the true owner's harm. For that purpose, therefore, knowledge may be an important ingredient for the liability of a third party. As to this, Mr Haberbeck stoutly maintained that there is no theory of vicarious liability in Saudi Arabian law ("no one is responsible for the acts of another") and for that reason Samba could not be liable for any wrongs done by Mr Al-Sanea. However, Professor Mallat's theory means that Samba is at fault and causes harm itself, by reason of its knowledge of the want of authority.
185. Once it is recognised that SICL has the interest of an owner and that harm has been caused to it by fault on the part of Samba, which sufficiently knew that Mr Al-Sanea held the Disputed Securities for SICL and that he had no authority to transfer them to it, I consider that a Saudi court would probably provide a remedy against the party with knowledge to recover the property. This is supported by text cited and translated by Professer Mallat from article 94 of the Hanbali Majalla, a respected and authoritative statement of Islamic law:

"[In case of] receipt of money belonging to a third party from the party in possession of that money without authorization from the owner, if he [the party in possession] was allowed to pass it on, the money shall be deemed to be in trust [amaana] with the latter [the third party] if the former [the possessor] is a trustee [amin]; otherwise the money shall not be deemed so [in trust]. If he [the possessor] cannot pass on the money, liability shall fall on both [the possessor and the third party]. Another possibility in law is that liability falls on the former alone."

Professor Mallat explained that the other possibility referred to arises where the recipient does not have knowledge of the want of authority. This is supported by a passage from Al Mughni (Chap 7, 397-8) the most famous Hanbali treatise on classical law, written by ibn Qudama:

"If the seizor deposits the misappropriated (object), or mandated a person in selling it, and delivers it to him and the object perishes, the owner can ask for compensation from either one he chooses; the seizor because he interfered between the owner and his property, and he has proven the misappropriating hand on the object seized, and the depositee or the agent proven in possession of the misappropriated object without right (unlawfully, bi-ghayr haqq). If the owner requests compensation from the seizor, and they (the depositee and/or the agent) did not know of the misappropriation, compensation is on the seizor, and he [the owner] cannot turn for compensation to anyone else. And if the owner makes them (the depositee and/or the agent) pay for compensation, they can turn to the seizor for the value or rent which they had to compensate, because they entered (into the transaction) with the understanding that they are not liable in any of this; but if they knew that the object was misappropriated, compensation is on them [alone]"

In the light of these passages and the evidence of the principles of Islamic law generally, I accept Professor Mallat's opinion about the ability to claim against a third party recipient with knowledge. The second issue, remedy against a third party, is therefore decided on the balance of probability in the claimants' favour.

(i) Effect of registration

186. Against that background, the third issue will determine whether SICL had a claim against Samba for the Disputed Securities or for compensation.
187. Professor Mallat's opinion is that a judge would apply the same principles that I have summarised under the second sub-issue above to reach the conclusion that an "error" (in terms of art. 27(d) of the CMR 2003) had been made and that SICL's nominee was entitled to be registered in place of Samba. The claim, he said, could be made on the basis that SICL had rescinded the September Transfer and then applied to have its nominee registered in place of Samba, if Samba did not voluntarily re-transfer the Disputed Securities.
188. I agree with the claimants that Mr Haberbeck was wrong in his written report to suggest that only a "clerical error" would be corrected under article 27(d). Under sustained cross-examination and with apparent reluctance, Mr Haberbeck eventually conceded that if a fraud had been perpetrated within the SDC, so that a true owner was de-registered or a purchaser's shares were deliberately registered in the name of another, article 27(d) could apply, even though no clerical error had been made. However, he insisted, it could only be used to correct something that had gone wrong within the SDC system itself; it could never be used to give effect to a transaction that had taken place outside the system, since the whole purpose of the SDC register was to ensure that only authorised and registered transactions take place affecting title to shares, in order to maintain the reliability and transparency of the register. To admit claims based on extraneous dealings would undermine the whole basis of the authorised registration system.
189. The claimants criticised Mr Haberbeck for only developing this theory in the witness box, when forced to give up his contention that only "clerical errors" fell within article 27(d). However, that is not the case. In para 21(a) of his trial expert report, Mr Haberbeck says:

"Based on several reported decisions of the CRSD, it is clear that the CRSD will not accept disputes over shares unless the disputed transactions were conducted in accordance with the applicable rules of the Capital Market Regulation and delegated legislation issued thereunder."

190. In paragraphs 43 and 44 of his expert report in response to Professor Mallat's trial expert report, Mr Haberbeck says:

"I do not agree with Professor Mallat's statements that SICL is or would have been entitled to call for the Disputed Securities to be registered in the name of its nominee in place of Mr Al-Sanea, or later SAMBA. Article 27(b) of the Capital Market Regulation … States, among other things: "The final mentions reported in the records of the Depository Centre shall serve as conclusive evidence and proof of ownership of the Securities indicated therein together with the encumbrances and rights associated therewith, subject to the provisions of paragraph (d) of this Article." Article 27(d) of the Capital Market Regulation provides for corrections of errors in the registry within the confines of the system established by Capital Market Regulation, and does not extend to transactions which were concluded outside that system, such as the Six Transactions.

The above is confirmed by CRSD Case 23-36, to which I referred in paragraph 44 of my tenth report, which was an action by the purported purchaser of listed shares against Tadawul to have the ownership records concerning the shares changed. Tadawul refused because the purchase was not conducted in accordance with the procedures under the Capital Market Regulation, and made the following statement to the court:

"Tadawul is not concerned with the sale contracts executed outside the Exchange or proving their validity. Instead, it is concerned with proving the ownership of transactions executed in the Exchange, settled and have their prices cleared under the controls is registered in the records of Tadawul Center of Securities Deposition in consistence is with Article (27) of the Capital Market Law."

The CRSD dismissed the claim because the purported share purchase had not been conducted in accordance with Article 21 of the Capital Market Regulation. In its reasoning, the CRSD stated:

"Whereas CRSD found that the claimant had not fulfilled the procedures required for transferring the ownership of the shares under consideration according to the said Article, CRSD can not oblige Tadawul Trading Company to transfer the ownership of shares prior to the fulfilment of such procedures. Hence, CRSD decided claimant's ineligibility for what he claims the respondent for." [sic]

In view of this decision, and the other cases referred to in paragraph 46, below, there is no possibility that the Six Transactions could have been used as the basis of an application to force Tadawul to change the ownership records of the Relevant Securities."

191. Pursuant to article 27(d), any challenge in relation to the register would first be made to the SDC and then, by complaint about its decision, to the CRSD. Professor Mallat accepts that article 27(d) is the only means by which the registration of Samba could be rectified. Unless and until rectified, Samba's ownership is conclusive.
192. Mr Haberbeck refers in his expert reports to 7 other cases in which the CRSD dismissed claims for lack of jurisdiction, where the claim depended on transactions conducted otherwise than in accordance with the CMR 2003. One of these was upheld on appeal by a decision of the Appeal Committee for the Resolution of Securities Disputes. Professor Mallat accepted that the CRSD will only entertain cases relating to transactions through the SDC:

"Q. … the CRSD would only exercise jurisdiction over transactions which had been conducted within the confines of the Capital Markets Law; in other words, transactions that had been –

A. Yes, you are right.

Q. .. on exchange.

A. Sorry, I misunderstood you. I thought that it was had an exclusive right to adjudicate such disputes. No, it is true, it can only look at shares that are listed.

Q. It is not just that it can only look at shares that are listed, if it is being asked to look into transactions in the shares, it has to be satisfied that the transactions have taken place on the exchange through authorised persons. That is right, is it not?

A. Yes.

Q. We probably do not need to go through them, but there are a number of cases, quite a large number of cases, that Mr Haberbeck refers to that support precisely that proposition. That is right, is it not?

A. If I am following you correctly, yes it is right."

193. I accept Mr Haberbeck's evidence and opinion about the jurisdiction of CRSD to entertain a claim by SICL under article 27(d). The conclusion is inherently credible and supported by cases referred to, including the case referred to in Mr Haberbeck's responsive expert report (para [190] above) and another decision of the CRSD, case no. 28/29 (decision 572/L/D1/2009G of 1430 H), in which a claim under a will of a testator was made against a registered proprietor of shares. The proprietor was a front for a foreign investor, and so the beneficiaries' claim depended on proving an agreement between the proprietor and the testator. The CRSD held that it had no jurisdiction because the transaction giving rise to the claimants' claim had occurred off the Exchange, not in compliance with regulations and through an authorised broker. The CRSD noted that the jurisdiction point "relates to the public order" and so should be taken by the Committee, even if not taken by one of the parties.
194. What is less clear is whether SICL might have fared better in the Commercial Court or before the Committee for Banking Disputes, or even in the General Courts. Mr Haberbeck gave evidence that, although proceedings could be issued in the Committee for Banking Disputes, the Commercial Court or the General Courts, it is likely that those courts would decline jurisdiction over a dispute relating to bank shares. He accepted, however, that a claimant could apply to the Committee for the Determination of Jurisdictional Disputes at the Supreme Council of the Judiciary, which would nominate a court or tribunal to make a binding determination, to ensure that a claimant at least gets through the doors of one court. That may overcome the jurisdictional bar, but it does not mean that the court will see fit to grant a remedy.
195. In his responsive evidence, Professor Mallat disagreed that any of these courts would decline jurisdiction, though he has no experience of practice in Saudi Arabia to support such an assertion, nor was he able to point to a decision in which jurisdiction had been accepted in a dispute relating to a claim of this kind. In his oral evidence he accepted that the CRSD would not entertain disputes arising from off-market dealings but felt that "it may be that the Commercial Court will look at it". He had referred in his 2013 report to two cases in which the Board of Grievances (which preceded the Commercial Court) and one case in which the General Court in Riyadh appeared not to consider absence of registration to be determinative of the claim. These cases were therefore important in supporting Professor Mallat's opinion that SICL could obtain redress in a court other than the CRSD.

i) The first case – case no. 592/1/Q/1427 Hejra: a shareholder was refused entry to a general meeting of a company. He had failed to bring with him, as was required, evidence of share ownership in the form of the share certificate (the shares were not listed). He sought to invalidate the company's resolutions and claimed compensation. This was refused on the basis that the company was right and not at fault in refusing entry. Professor Mallat said that the Board appeared to be open to other proof of ownership, such as the share certificate, not just evidence of registration. Mr Haberbeck said that this was a case about failure to prove ownership. Making allowances for the limits of understanding given by reading a translation of a judgment on foreign law, it appears to me that Mr Haberbeck is right: the decision was that the company was in the right in stipulating in its advertisement of the general meeting for production of a share certificate to gain entry to the meeting, which the claimant did not bring. The fault was his. This case says nothing about whether registration is conclusive, whether the title of a registered proprietor will be rectified based on a claim by a true owner, or whether the Board had jurisdiction in such a case.

ii) The second case – folio no. 88/16 of 5/4/1427 Hejra: two people bought equal numbers of shares in an unlisted company in 1992 and the share certificate for all the shares was in the name of one of them. Further instalments of capital were subscribed by the registered owner. The claim was for delivery of half the shares and compensation. The judge decided that the claimant had made an admission that the shares had been properly sold and that therefore the only issue was an account between them, relating to the purchase price and the sale proceeds. The claim for delivery of the shares and compensation was dismissed. Professor Mallat said in his 2013 report that there was no mention of registration as a condition of transfer and the judge used various types of evidence to ascertain the existence of the transfer. In his oral evidence he conceded that the case was not totally apposite, but that it was clear that the judge accepted that it was not simply a matter of registration: the judge could have relied solely on registration but did not do so. Mr Haberbeck said that the case turned on the admission that the shares had been sold at the claimant's request and that the court did not have to decide any question of ownership. I prefer Mr Haberbeck's explanation, which accords with the language of the translation of the case. It is clear that the defendant had sold the shares, and so would not have been the registered owner of them. The company was not a party and presumably its share register was not available in any event. The only issues were whether the claimant had authorised the defendant to sell the shares and the state of account between them. This case does not support Professor Mallat's opinion because there was no issue about underlying ownership to be decided.

iii) The third case – 1601/1/Q of 1427 Hejra: the dispute was whether the claimant had paid capital to become a 50% partner in an establishment (mu'assasa) that then became a limited liability company. The claimant alleged an oral agreement and relied on cheque stubs and oral evidence of witnesses. The dispute was a factual one as to what had been agreed and paid, and for what purpose. The owner of the company swore an oath that he did not receive money from the claimant for a partnership share. The court noted that the claimant had not adduced any proof of his allegations and therefore accepted the defendant's oath ("the rule under the sharia is that the oath belongs to him who denies"). Professor Mallat commented that the Board did not determine the matter on the basis of non-registration as a shareholder but considered the evidence about the original arrangement. In his oral evidence he suggested that the case was about evidence and accepted that it was not about registration, but it undermined the suggestion that registration was absolute. Mr Haberbeck said in his trial report that the case was about a sleeping partner of a partnership, not about shareholdings in a company and had nothing to say about registration of share ownership. In my judgment, Mr Haberbeck is right: there was no claim to company shares; just a claim to be a partner in a mu'assasa. The case was about proof of allegations that did not depend on registration and pre-dated the formation of the company. It says nothing about what proof is required to establish that a registered proprietor of shares is not the owner of them, or whether the Board would have jurisdiction to decide such a claim.

196. Following a discussion of these cases, Professor Mallat was challenged about the basis for his opinion that any court would recognise that SICL had a sufficient interest to disturb Samba's registration as owner. Professor Mallat then seemed to retreat from his suggestion that the register would be ordered to be rectified so as to confer title on SICL's nominee:

"Q. I think where we are is this, that a judge would not understand, you say, anything about the word "proprietary" and "proprietary interests" and so on. He would regard Samba's legal ownership of the shares as unaffected by any suggested proprietary claim because Saudi law does not recognise any proprietary claim or interest other than sole legal title.

A. I think I dispute the process in which the judge will approach this. You are really in this format putting in his language a matter that he will not be comfortable with, he will not be able to relate to because he does not have the same, as I understand it, sharp division between proprietary rights under…

Q. He will not recognise any interest other than legal title. I think I can put it even simpler than this.

A. Yes.

Q. Your position is that the Saudi court would award compensation to SICL against Samba because SICL has been deprived of its interest in the shares?

A. Excellent, yes. I would subscribe to that.

Q. And what that would involve would be the Saudi court acknowledging, well Samba has the shares now, it has legal title to the shares, and there is no interest here that can upset that legal property but because they have been, you would say, taken in dishonest or other circumstances by virtue of Mr Al-Sanea's dishonesty - and Samba knew that the shares were being held on trust - the court would award compensation for the loss of that interest? That is your case, is it not? That is your argument?

A. I think it is close enough, yes. I would quibble, but for the purpose of clarity, I think we have the useful shared territory for his Lordship.

……

Q. So we are left in a position that your evidence is, and we see it in your reports, that SICL would have a claim for compensation but could not upset the transfer of the title as such to Samba?

A. I think it would yes, but it would be immaterial. The harm would be at the bottom of it. The court will try to understand how the deprivation of that interest would need to be fully remedied.

Q. But it is a deprivation by virtue of the transfer of legal title to Samba. That is what we have here.

A. Yes.

Q. By that means, whatever property interest, other interest, SICL previously had has been lost to it?

A. No, I mean, SICL has benefits arising from the contracts and there were various types of contracts."

197. In my judgment, this evidence should be taken as a frank acceptance that there is at best a claim for compensation against Samba, not a claim for the return of (or rectification of the register showing title to) the Disputed Securities. Apart from the three cases that he relied on, Professor Mallat has provided no material that supports the view that he expressed in his written reports, viz that article 27(d) would encompass a claim by SICL based on the Six Transactions. I find that Mr Haberbeck must be right that it would not be credible for a claimant in SICL's position to be able to rely on an off-market transaction with a registered proprietor in order to assert ownership of shares, which would then be capable of binding a purchaser in an authorised transaction through the SDC exchange, who has obtained registered title to the shares. Such a possibility would undermine the credibility of a system of authorised dealings in shares that is transparent, efficient and secure and be contrary to its objectives, which are (among others):

"1. Regulate and develop the Exchange, seek to develop and improve methods of systems and entities trading in Securities, and develop the procedures that would reduce the risks related to Securities transactions.

…..

5. Seek to achieve fairness, efficiency and transparency in Securities transactions.

6. Regulate and monitor the full disclosure of information regarding Securities and their issuers, the dealings of informed persons and major shareholders and investors, and define and make available information which the participants in the market should provide and disclose to shareholders and the public."

(CMR 2003, article 5)

The word "fairness" cannot be plucked from this article and used to justify a proprietary remedy for a person who has been defrauded, as the claimants sought to do. There are much wider considerations of fairness that apply in this context. It is clear from the Rules that any interests that affect a proprietor's registered title are to be noted on the register, not hidden behind the curtain that registered title designedly provides.

198. Professor Mallat did not dispute the underlying purposes of the registration scheme, or that article 27 was designed to further them:

"Q. …transparency and efficiency requires that someone who buys listed shares in Saudi Arabia knows that he is getting security of title, does he not?

A. Yes.

Q. He has to be sure that what he is buying he is getting what he is buying?

A. That is right.

Q. And unless other adverse interests are registered he is always going to be entitled to assume that he is getting a completely secure title?

A. Yes, a title in any context, whether it is ownership of a share or an ownership of a piece of land requires stability, transparency and security, yes.

Q. We will look at Article 27 in due course, but what Article 27 is designed to ensure is that all registrable interests, including claims of the incumbrances, appear on the register so that the acquirer, the transferee of the shares knows exactly what he is getting?

A. Yes, of course.

Q. And he takes free, to use that expression, of any interests that do not appear on the register. That is Saudi law. That is registration everywhere?

A. That is right. Yes, it is true."

At a later stage in his cross-examination, Professor Mallat agreed that article 27 has to be read against a background of the efficiency, transparency and all other features of a well-regulated market.

199. I find no persuasive evidence that a court or committee in Saudi Arabia would have allowed these purposes of the Tadawul to be compromised by directing Samba to transfer the Disputed Securities to SICL's nominees, or otherwise by recognising SICL's ownership. The CRSD case (no. 2247, decided on 1 May 2018) cited by Professor Mallat (the heirs case) was not a case of a beneficial owner asserting rights against a purchaser from the legal owner but of heirs of the owner of shares asserting their entitlement on his death. I was told that there is a specific exception in the CMR 2003 for such cases, where the claimants' entitlement as heirs does not depend on an authorised and registered transaction. I find that it is probable that no court or committee other than the CRSD would accept jurisdiction in a claim to assert SICL's ownership since, as was common ground, a claim to upset Samba's title had to be brought under article 27(d).

(j) The National Commercial Bank shares

200. So far as the NCB shares are concerned, Professor Mallat said that the same principles would apply to those unlisted shares, namely that registration in the share register is prima facie evidence of ownership, but that evidence can be adduced to show that the real owner is someone else, and rectification of the share register sought on that basis. Mr Haberbeck contended that registration was conclusive evidence of ownership, so an unregistered transfer of shares (i.e. the Six Transactions) is invalid against third parties. He pointed out that there was no case that had been produced in which a beneficial owner had successfully claimed rectification of the share register, but equally no case in which it was said that such a claim could not be made. He accepted that his opinion was based on his reading of article 102 of the Companies Regulation.
201. In my judgment, registration is prima facie evidence and, until displaced, is conclusive as to ownership. But there is nothing to prevent a claim for rectification of the share register, if the claimant has a sufficient basis for it. For example, if an agent had by deceit procured registration in his own name, it would be extraordinary if article 102 prevented a claim for rectification. It would provide less scope for claims to correct the register than is provided in the case of listed securities. So Mr Haberbeck must be wrong to say that registration is conclusive if by that he meant (as I understood him to mean) that title could not be disturbed. But the position may be different where the defendant is not the agent but a purchaser from the agent who is now registered as proprietor of the shares. The CMR 2003 are themselves evidence that registration principles often override adverse claims. The Companies Regulation also provides for registration and for it to affect third parties (art. 102).
202. The outcome as regards the NCB shares involves a conflict between the purpose of registration under the Companies Regulation and basic principles of Islamic law. My conclusion on sub-issues (i) and (ii) above mean that I agree with Professor Mallat that, had SICL claimed to have Mr Al-Sanea replaced as registered owner by another nominee before the September Transfer, it would probably have succeeded before a Saudi Arabian court. Professor Mallat's opinion is that the same principle applies where the 'fiduciary' transfers the shares without authority to a third party: once the transfer is rescinded, the third party cannot resist a claim to rectify the register in favour of a nominee of the 'beneficiary'. There was nothing, other than Professor Mallat's assertion that the position as regards unlisted securities is the same as under the CML 2003, to support that opinion. The position under the CML 2003 does not support that opinion, for reasons that I have given. In those circumstances, I conclude that I cannot be satisfied on a balance of probabilities that such a claim would succeed. The fact of registration of title is a distinct consideration, and such a claim is therefore different both from a claim to recover (say) a painting or vase from a purchaser who bought knowing that the seller was not authorised to sell, and from a claim against the seller. The reason is that the buyer has relied on registration of title.

(k) A claim for compensation only?

203. Professor Mallat may be right that a claim for compensation – premised on the fact that SICL had been deprived of its property – would exist in Saudi Arabian law in the Commercial Court or before the Committee for Banking Disputes. This is uncharted territory so far as a claim against a registered owner of shares is concerned: neither expert witness has been able to provide a persuasive analogy in the case law, or from learned texts, that deals with the significance of registration. Professor Mallat relied on the Jeddah market case (DM, Majmu'a 1403-1426, at 336-47) in which the head of the municipality of a new airport prevented a party to a building contract from complying with it by seizing its generators, leading to liability for unlawful acts directly harming the claimant construction company. He contends that this establishes that there can be liability generally for wrongful interference in a contract, but Mr Haberbeck disputed this. In his opinion this case turned on the fact that it was an administrative court that decided it: the claim was against a government official, brought under a particular statute, and was a ghasb claim. The case on any view does not begin to address the consequences of the Disputed Securities being registered at the time in the name of Mr Al-Sanea and now registered in the name of Samba. Neither do any of the other cases cited by Professor Mallat. If there can be liability in such circumstances, something may depend on the extent of Samba's knowledge of particular facts and the rights that Samba believed itself to be exercising vis-à-vis Mr Al-Sanea, or on the extent of its duty as a bank to inquire.
204. I am unable to decide, on the evidence before me, that it is probable that a case against Samba for compensation would succeed. Such a claim would of course mirror, in its essentials, the claim under English or Cayman Islands law that the claimants bring in this action. I suspect that such a compensation claim against a registered (bank) transferee of securities, where a proprietary claim cannot succeed, would be res integra for a Saudi Arabian court. Certainly, none of the cases to which I have been referred comes close to giving guidance on how a Saudi Arabian judge might apply established principles of Islamic law in that context. I am not satisfied that I can rely on Professor Mallat's opinion, as it appears to be based on cases that do not approximate to the circumstances of such a claim. I am not persuaded that a claim which it appears has never been brought in Saudi Arabia would be more likely than not to succeed.
205. In any event, any such claim would be premised on the loss of SICL's property, not an assertion of continuing proprietary rights. The availability of such a claim could not amount, as a matter of English or Cayman Islands law, to a continuing proprietary interest in the Disputed Securities. The question of what rights SICL had as against Samba that survived the September Transfer is a matter governed by Saudi Arabian law, but any such rights must then amount in English and Cayman Islands law to a proprietary interest (or, at least, a claim of entitlement to recover property that is likely to succeed) for there to be a continuing proprietary base to the claimants' knowing receipt claim. A claim for compensation for loss of property is not on any view, as a matter of English law, a proprietary interest or claim.
206. The claimants' claim therefore cannot succeed: SICL had no continuing proprietary interest in the Disputed Securities after the September Transfer capable of supporting a claim against Samba in knowing receipt.

V. The Valuation Issue

(a) Introduction to the block discount question

207. My conclusions on the Law of Knowing Receipt and Saudi Arabian Law Issues are sufficient to dispose of the claim, but in case my decision on either of those issues is overturned on appeal, I will give my conclusion and reasons on the Valuation Issue. This is whether, in valuing the Disputed Securities, there should be a discount ("block discount") from the quoted market price on the relevant dates for each of the substantial holdings in the five banks, and if so the size of each discount.
208. The relevant dates are: for all the holdings except NCB, the date of the September Transfer (see [125] above for a list of the holdings at the date of the September Transfer); and for all five holdings, the date on which judgment is given. The expert valuers were of course unable to value at the date of judgment and for the purpose of exposition of the block discount issue they valued at 31 January 2020, a convenient date shortly before expert reports were exchanged and filed. At that date, the number of shares in each holding and the percentage of the relevant bank's issued equity represented by the holding were as follows:

Arab National Bank:	2,053,375 (0.14%)
Banque Saudi Fransi:	7,080,242 (0.59%)
National Commercial Bank:	4,258,500 (0.14%)
Samba Financial Group:	15,844,541 (0.79%)
Saudi British Bank:	6,638,691 (0.32%)

209. The issue of block discount arises because Samba contends that the appropriate basis of value for the Disputed Securities is their "market value", a concept defined in International Valuation Standard 104, namely as:

"…the estimated amount for which an asset or liability should exchange on the valuation date between a willing buyer and a willing seller in an arm's length transaction, after proper marketing and where the parties had each acted knowledgeably, prudently and without compulsion."

Detailed notes in IVS 104 explain what is to be understood by these terms, but for present purposes it is sufficient to note that it represents a compromise between the seller's desire to obtain the highest price that they can reasonably obtain for the asset and the buyer's desire to acquire the asset at the lowest price that the buyer can reasonably achieve in the market on the valuation date.

210. Where small numbers of shares are to be sold on an active public exchange, the price per share can readily be taken to be the quoted reference price on the day in question. No further adjustment is necessary, save perhaps for costs of sale. But where a large block of shares is to be sold on a single day, the reference price is unlikely to be obtainable, at least where the size of the block far exceeds the average daily traded volume ("ADTV") for that stock, as is the case for each of the holdings of the Disputed Securities. The market has insufficient liquidity to absorb the sale of the holding without a bespoke transaction or transactions at a negotiated price. If the entirety of a large holding is to be sold on a particular day, a discount may well be required to achieve the sale. (Correspondingly, if a buyer needs urgently to acquire a large holding, a premium may be necessary to induce more sellers to sell on a particular day.)
211. If the holding is to be sold in smaller parcels of shares over a longer period, there may still be an impact on price from the cumulative volume being sold, depending on the liquidity in the market and the impact (if any) on sentiment from large volumes of sales, and the seller is further exposed to the risk (but not the certainty) of a fall in the share price from other causes before the shares are fully sold. The amount of the block (or, in the United States, 'blockage') discount is therefore likely to reflect the alternatives that are reasonably available to the seller. The seller will not willingly accept a lower price than they can realise by other means.
212. Samba's pleaded case was that a discount of 5% - 15% was appropriate for each of the holdings in listed shares.

(b) The expert witnesses

213. The expert witnesses called by the parties on the valuation issue were Mr James Worsnip FCA, on behalf of the claimants, a managing director of AlixPartners UK LLP, who has lengthy experience in the valuation of businesses and securities, and specialises in valuation assignments in the banking and capital markets sectors; and Mr Luke Steadman FCA, on behalf of Samba, a partner of Alvarez & Marsal Disputes and Investigations, LLP, who is a specialist forensic accountant and his firm's International Financial Reporting Standards technical partner and head (or joint head) of its Global Arbitration Practice and its contentious valuation team.
214. Both expert witnesses clearly had the experience and expertise needed to give a relevant and informed opinion on the question of block discount generally, though neither was able to point to specific experience in connection with selling or buying large blocks of shares on the Tadawul in Saudi Arabia or on a similar exchange, or in valuing blocks of such shares.
215. My impression of Mr Worsnip was that he was taking rather extreme positions in order to support the argument that an orderly liquidation basis of value was appropriate, and in particular that no adjustment on such a valuation was required for the risk of movement in the market over time adverse to the seller; and further, that even if market value was the right basis of value to take, there could in principle be no block discount on such a valuation either. He was rather more persuasive in his analysis of the market evidence and in his criticisms of Mr Steadman's modelling assumptions and reliance on academic studies.
216. Mr Steadman was an extremely polished and fluent witness, who clearly understood in considerable depth the issues that he was addressing. His explanation for the use of the market value basis of value and the type of put option model were very persuasive, but I felt that on occasions his own intelligence and understanding led him to go too far in justifying quantitative models that produced what, on standing back, one might regard as surprisingly large discounts for stocks of the quality with which he was concerned. Notably, most of Mr Steadman's original discount figures and some of his final figures exceeded the applicable range of discounts pleaded by Samba, but no attempt was made by Samba to amend its pleaded case. Mr Steadman's discounts reduced significantly over the course of his three reports, as he made corrections for errors pointed out by Mr Worsnip and adopted some of the adjustments that Mr Worsnip considered appropriate. Even so, I was left with the impression that not enough objectivity and scepticism had been applied to the results, probably because there was an intellectually satisfying argument for applying the models without further adjustment or allowances. Overall, although Mr Steadman was an impressive witness, I felt that rather too much machinery and not enough objective judgment went into his percentage discounts.

(c) Is a block discount excluded in principle?

217. The claimants' position is, first, that no block discount is applicable, as a matter of principle. They submit that the nature of liability for knowing receipt requires the defendant to restore the trust property and pay compensation for any losses, in the same way that a true trustee would be liable; or, if it is unable to restore property in specie to the trust fund, the trustee must pay the value of the misappropriated property at the date of judgment, together with any consequential losses. The focus in a substitutive performance claim is on what is required to restore the trust fund, and therefore on the objective value of the property that the trustees should be able to produce, using the full benefit of hindsight. As the editors of Lewin on Trusts (20^th ed.) say, at para 41-010: "the trustee must pay sufficient compensation to put the estate back to what it would have been had the breach not been committed".
218. The claimants therefore argue that any concept of discount, predicated on the amount that could be realised if a large block of shares were sold on a single day, is irrelevant to the question of what would be needed to restore to them the value of the fund. The focus should instead be on what money would be required to restore the value of the Disputed Securities to the fund, which if anything is the cost of purchasing them, not the price obtainable on selling them. The claimants have not pleaded that if the September Transfer had not occurred the Disputed Securities would have been sold by SICL at any stage; they are therefore entitled to be placed in the same position as if the Disputed Securities were still held on trust for them.
219. The claimants cite In re Bell's Indenture [1980] 1 WLR 1217, in which a defaulting trustee sought to argue that the full value of the misappropriated funds should not be paid since they would have attracted estate duty had the funds remained in the trust. That argument was rejected by Vinelott J at pp. 1235 – 1236:

"There can to my mind be no doubt that a trustee who has himself defrauded his beneficiaries by taking trust moneys for his own purposes is liable to restore the moneys he has taken without regard to any fiscal liabilities that might have fallen on the trust fund if he had not misappropriated the funds. If as a result of the decision of the Estate Duty Office not to charge estate duty on the restored fund there is a windfall, the windfall cannot be allowed to benefit the defaulting trustee. Equally, if a trustee has sold an investment in order to misappropriate the proceeds, he must restore the investment or if it is shown or admitted that the investment would have been sold at a later date (as in the case of Churchill Farm) restore its value at that later date."

220. The claimants further note that Samba in fact retains the Disputed Securities, so there has been no diminution in their value as a result of a block discount, and applying such a discount would give Samba an unmerited windfall. They further point out that, upon the September Transfer, Samba valued the Disputed Securities in their books at the market price, without any discount, and that Samba should be required to disgorge the value that it ascribes to the Disputed Securities, not merely make good any loss that SICL suffered.
221. As to this primary case that any discount is wrong in principle, Samba accepts that the claimants' claim is for substitutive performance and argues that this requires it to pay the objective value of the property, that is to say its market value or monetary value. It refers to a number of textbooks and learned articles for this proposition, which use the expressions "objective value" and "monetary value", but not "market value"; nor are they concerned with how objective value of a large holding of shares in listed companies is to be ascertained.
222. Samba argues that market value is the most appropriate measure of objective value of the Disputed Securities and that any analogy with Re Bell's Indenture is misplaced, since that was not a valuation case at all. The value at which Samba holds the Disputed Securities in its books is a holding or accounting value, not a measure of the objective value; and Samba has obtained no gain by holding the shares (other than the dividends for which it accepts it must account to the claimants, if liable). It submits that the claimants have elected not to claim the Disputed Securities themselves, despite the fact that Samba still holds them (or the shares that now represent them), and so are limited to a claim for their objective value.
223. Samba points out that neither of the expert witnesses contends in their evidence that fair value or investment value (where, they agree, no block discount would be appropriate) is an appropriate measure of the objective value of the shares. Mr Worsnip says that an orderly liquidation value basis of value is appropriate, and Mr Steadman says that market value is the obvious basis of value to use.
224. The first question must be whether, as a matter of law, when seeking to identify the objective value of an asset in connection with a trustee's liability for substitutive performance, a measure of value based on what could be realised for the asset in a transaction is inappropriate. If this were so, the proper measure of value would have to be the holding or investment value of the property, since subjective value to the beneficiaries is plainly antithetical to the accepted measure of the objective value of the property. Investment value is defined in IVS 104 as:

"… the value of an asset to a particular owner or prospective owner for individual investment or operational objectives. Investment Value is an entity specific basis of value. Although the value of an asset to the owner may be the same as the amount that could be realised from its sale to another party, this basis of value reflects the benefits received by an entity from holding the asset and, therefore, does not involve a presumed exchange. Investment Value reflects the circumstances and financial objectives of the entity for which the valuation is being produced. It is often used for measuring investment performance."

Thus, investment value would be a value specific to the trust, which may have (or have had) as its objective the holding of the asset, with no intention of selling it.

225. I am unable to conclude that the measure of liability for substitutive performance must always be the investment value of the misappropriated asset. There are likely to be many cases in which market value at the valuation date is higher than the long-term holding value of the asset. Many trust assets may be held otherwise than to achieve investment or operational objectives. It is notorious that, in economic terms, utility value of an asset may be substantially lower or higher than its exchange value. I do not agree with the claimants' arguments that an assumed transaction in the property in question is wrong in principle as a basis for assessing the objective value of trust assets. In many cases, market value will be the most appropriate basis on which to assess the objective value of the property; however, market value is not necessarily the right basis of valuation in every case. I will come back to this point shortly.
226. The second question is whether, on the facts of this case, if not in every case of substitutive performance, the appropriate measure of value in principle is investment value. The factors on which the claimants particularly rely are summarised above. The Disputed Securities are not alleged to have been held for any purpose other than as an investment, nor on a particular long-term holding strategy, and since 2009 they would have been held subject to the joint official liquidators' realisation strategy in SICL's liquidation. The Disputed Securities are freely traded on the Tadawul and have no special characteristics, other than the relatively large size of the holdings. The fact that the claimants are claiming substitutive performance rather than the Disputed Securities themselves suggests that there is no particular significance in having the Disputed Securities rather than their value.
227. I therefore reject the argument that investment value is appropriate on the facts of this case. Neither expert valuer was of the opinion that it was the most appropriate basis of value in this case.

(d) The right basis of value to apply

228. The next question is what objective basis of value is appropriate in this case. The fact that the Disputed Securities are retained by Samba does not seem to me to make any difference to the right valuation approach: Samba has not made any gain from holding them that it would not otherwise be liable to disgorge.
229. I do not accept that market value is automatically the appropriate basis of value. According to IVS 104:

"It is critical that the basis (or bases) of value be appropriate to the terms and purpose of the valuation assignment, as a basis of value may influence or dictate a valuer's selection of methods, inputs and assumptions, and the ultimate opinion of value. (10.1)

"Valuers must choose the relevant basis (or bases) of value according to the terms and purpose of the valuation assignment. The valuer's choice of a basis (or bases) of value should consider instructions and input received from the client and/or its representatives. However, regardless of instructions and input provided to the valuer, the valuer should not use a basis (or bases) of value that is inappropriate for the intended purpose of the valuation… (20.2)"

So the purpose of the valuation must be borne in mind and, subject to that and any assumptions explicitly required, there is a valuation judgment to be made about the right basis of value.

230. The purpose of the valuation in this case is to estimate (in the sense in which IVS 104 uses that word) in money terms the sum that will put the trust in the same position as if the misapplied property were still held for its benefit. The most appropriate basis of value should therefore be used to achieve that objective.
231. The reason why market value will not always be the right basis of value is that it excludes special circumstances or any element of value available only to a specific owner or purchaser, and excludes from the applicable market those with a special interest in purchasing. To use a commonplace example, 'marriage value', or 'synergistic value' is sometimes released when two assets in different ownership are brought together. This often arises with two parcels of land or different interests in a single parcel, which together enable a valuable development to be carried out; or with shareholdings that individually are minority holdings with no control but together may give a single holder control of the company in general meeting or a 75% share of the votes. The two parcels or holdings together are worth more (by the amount of the marriage value) than the sum of the individual parts. Thus, land or a shareholding might have special value to one buyer.
232. This is not subjective value; it is objective value that is only capable of being fully released by a single owner. If property had synergistic value in the hands of a trust, that could be a reason for using a basis of value different from market value, or for making a special assumption about the notional transaction. The issue of synergistic value does not arise in this case, but it demonstrates why market value, as defined, will not necessarily be the right basis of value.
233. In this case, the claimants' fall-back position is that market value is an inappropriate basis and that the most suitable basis is liquidation value, assuming an orderly transaction rather than a forced transaction. This is defined by IVS 104 as being:

"…the amount that would be realised when an asset or group of assets are sold on a piecemeal basis. Liquidation Value should take into account the costs of getting the assets into saleable condition as well as those of the disposal activity".

An orderly liquidation basis is then defined as being:

"…the value of a group of assets that could be realised in a liquidation sale, given a reasonable period of time to find a purchaser (or purchasers), with the seller being compelled to sell on an as-is, where-is basis. The reasonable period of time to find a purchaser (or purchasers) may vary by asset type and market conditions".

234. There are significant differences between Market Value and Liquidation Value. Under the former, a reasonable period for preparing and exposing the asset to the market in the most appropriate manner to achieve the best price is deemed to have happened before the valuation date; whereas, under the latter basis of value, the valuation date precedes a reasonable period that is needed to find a purchaser willing to buy on an as-is, where-is basis. The value on a Liquidation Value basis therefore reflects the price that will be paid at the end of that reasonable period, which will be long enough for the seller to prepare the asset and market it properly and for the buyer to do due diligence. The consequence is that risks associated with preparation for sale and delay in agreement on a price will be reflected in the Liquidation Value, whereas Market Value reflects a price paid on the valuation date. Further, the seller on a market value basis is a willing seller; on a liquidation basis he is a compelled seller. A willing seller will only sell for the best price that can reasonably be obtained from a willing buyer in a negotiation; a compelled seller will sell for whatever he can get.
235. The claimants submit that market value is inappropriate because any trustee selling the Disputed Securities would be obliged to take care to maximise the realisable value of the property. To sell a large holding of shares on a single day would be a fraud on the trustee's power of management if it caused the reference price to fall substantially. Samba as trustee therefore could not lawfully sell the Disputed Securities in such a way as to cause the price to fall and would be obliged to take steps to avoid such loss. This is what in fact the claimants have been doing, by instructing stockbrokers to realise SICL's other assets in such quantities that unnecessary impact on market prices and loss to SICL is avoided. The claimants therefore submit that any hypothetical sale of the Disputed Securities should be treated as having been undertaken pursuant to such a strategy, without incurring any block discount for the trust assets. On that basis, they submit, market value is inappropriate if a sale of the Disputed Securities on a single day would incur a block discount.
236. In my judgment, the claimants are right in principle that, when valuing trust property, the court should adopt a basis of value that properly reflects the nature of the property as trust property. The monetary equivalent of the trust property is the money that would be realised by a trustee on a sale that was authorised by its powers of management, not in breach of trust. Only in that way will the defaulting trustee have restored the position of the trust, if not in specie then the full monetary equivalent of the trust property. If a notional sale on the transaction assumed in the standard definition of market value would be a breach of trust (e.g. because it would fail to avoid a diminution in the price that could reasonably be avoided, or because it would otherwise not obtain the full value of the trust property) that assumed transaction must be the wrong basis of value. On the other hand, if an alternative strategy could not reasonably achieve a better price, or if market value would itself reflect the price that should be obtained by the trustee, then market value is likely to be the right basis of value. Where the claimants go wrong is to assume or conclude that the true block discount resulting from the market value sale hypothesis could be avoided by pursuing a different realisation strategy. I will return to this and explain why it is so.
237. However, first, Mr Worsnip considers that there would be no block discount in any event, because the tendency of a sale of a large block of shares to depress the market price will be matched by the upwards pressure on price that a willing buyer of a large block will have, so that if a transaction is assumed on the valuation date between a willing seller and a willing buyer it will take place at the reference price, with no upwards or downwards adjustment. I do not accept that argument: it confuses the assumption that a sale is concluded on the valuation date between willing parties with the reality of the state of the market on that date. There is an assumption that a transaction takes place but there is no assumption about the state of the market. There is no requirement to assume – contrary to known facts about the average volume of a given stock that is traded on a daily basis – that on the valuation date there is a buyer who is willing to buy a large block of shares at the reference price or above. What has to be assumed is that a notional seller and a notional buyer in the market reach an agreement, but there is no assumption about the terms of the deal that is struck.
238. Second, Mr Worsnip argued that a strategy of "dribbling out" the shares in each of the holdings over a protracted period, in small quantities each day, would be likely to prevent any downwards impact on the reference price, so that no block discount is appropriate. That raises two issues: whether the reference price would be unaffected by such a strategy, and how the value should reflect the risk that the reference price may fall owing to extraneous factors over the period of time required to dribble out the holding. Mr Worsnip considered that there would be no impact on the reference price from pursuing such a strategy at volumes of 20-25% ADTV for each stock and that no adjustment was required for the risk of a falling market.
239. Mr Steadman disagrees with Mr Worsnip on both issues. He considers, principally on the basis of the work and models of Professor Kissell (The Science of Algorithmic Trading and Portfolio Management), that even trading at a relatively small percentage of the ADTV of a stock has an effect (other than one attributable to temporary lack of liquidity) on the reference price. He also argued that Mr Worsnip was wrong in failing to allow for the risk of a falling market. Both experts referred in detail to the academic studies of a number of learned authors, including Professor Finnerty's work on discounts for lack of marketability, Dr Pratt's work on share discounts and premiums, various put option models used to calculate the cost of trading shares at a future date (Finnerty and Black-Scholes) and studies on the effect of sales of blocks of shares (Keim and Madhavan, Alzahrani and others). It is unnecessary to consider their work in any detail.
240. Mr Worsnip's approach was to value the Disputed Securities on the basis of orderly liquidation value, using the Finnerty put option model to assess the cost of being able to sell the whole block of shares at a future date at the prevailing reference price. The Finnerty model was created to value discounts for illiquid stock where there is a lack of marketability. Mr Worsnip said that basis was most appropriate to the strategy of dribbling out the Disputed Securities over a longer period of time after the valuation date. He contended that no reduction in value was necessary for market impact because the dribbling out of shares was at only 20-25% of ADTV, which would have no effect on the reference price. The cost of the put option that enabled the entire holding to be sold on by the end of the reasonable period for marketing and sale was expressed as a percentage discount on the reference price of each stock on the valuation date.
241. Mr Worsnip did not accept that any adjustment was appropriate for the risk that the reference price might fall for extraneous reasons over the dribble out period. He considered that it was equally likely that the price might rise over that period (as on average the price of a good quality stock does), and that the seller would be compensated by the payment of dividends in the meantime, and that therefore the risk of a fall in price did not translate into any discount. On that basis, it was not necessary, he said, to pay for a more expensive put option (the Black-Scholes model used by Mr Steadman) that guarantees the seller that the holding will be sold at a future date at the reference price on the valuation date. All that was needed was an option to sell the required volume of shares at the quoted price, whatever it might be, by the end of the reasonable period. He said that the more expensive put option gave the seller an additional advantage: if the price rose, the option would not be exercised and the shares would be sold at the market price.
242. In my judgment, this approach does not equate to the orderly liquidation basis of value, nor is it persuasive in identifying the likely cost of dribbling out shares over a protracted period. Under the liquidation basis of value, the value is a day 1 value that reflects the cost and time taken to prepare a collection of assets for sale at an uncertain future date, on terms that have not yet been agreed, after a reasonable period for preparation and marketing has elapsed. In almost all cases, therefore, the value will be less, often significantly less, than the market value of the assets, which reflects a negotiated sale concluded for cash on the valuation date after a reasonable period for marketing has already expired. Liquidation value reflects greater cost, delay and substantial risk that there will be no strong demand for the assets and that, over the period during which the sale is delayed, the aggregate price that can be obtained may fall. I accept Mr Steadman's evidence that with listed securities there is always a risk of a fall in price and risk carries with it a cost, which impacts on value. That is why the Black-Scholes model of put option (which guarantees the valuation date price) is more expensive than the Finnerty model put option. It is no answer to say that the price equally might rise. The seller is not assumed to be willing to retain the asset to obtain the benefits of ownership: he is a seller; and the fact that the option may not be exercised if the price rises is likely to be factored into the price for the option.
243. I also consider that Mr Worsnip is wrong to assume that dribbling out shares over a protracted period at a rate of 20-25% of ADTV cannot itself impact on the price obtainable. Mr Worsnip's assumption is based on US and EU regulatory guidance on corporate stock buy-backs (US SEC rule 10b-18 and Commission Regulation (EC) No. 2273/2003), which permit buy-backs of shares at up to 25% of ADTV without thereby raising an issue about market abuse or manipulation. The fact that trading at such levels is deemed not to amount (of itself) to market abuse does not mean that if trading at those levels is carried on for a protracted time it can have no impact on the market price. Although Professor Kissell's study is not directly applicable to the circumstances of valuing large blocks of bank stocks in Saudi Arabia, it provides some evidence of the possible effect on market price of selling relatively small quantities of stock.
244. In carrying out his fall-back valuation, Mr Worsnip has therefore not given effect to the liquidation basis of value that he purports to apply. He has not done so because he has assumed that there is no allowance required for the risk of the share value falling after the valuation date, before a sale can be concluded, and because he assumes that the dribbling out of shares over a protracted period or the preparation during that period of the sale of a large block of shares would have no impact on the share reference price. It is for those reasons, principally, that despite using a liquidation basis of value rather than market value, his values are significantly higher than Mr Steadman's market values.
245. Mr Steadman's approach to determining market value is to calculate the appropriate discount from the valuation date reference price to reflect the cost of the other alternatives that are available to the seller. He accepts that the only viable alternative is to dribble out the shares, in the way that Mr Worsnip proposes, but he recognises that in doing so the seller runs two risks: that the exercise of dribbling out shares over a protracted period will be visible and will impact on price, and that the price may otherwise fall over the period. The only way to avoid these risks is to dribble out the shares at the fastest rate that will have no substantial impact on price and to obtain insurance against the risk of the share price falling over that period.
246. Mr Steadman's model is that suggested by Joseph Estabrook in Chapter 7 (Blockage Discounts) of Reilly & Schweihs Handbook of Advanced Business Valuation (2000). Mr Steadman calls it a least-cost on-exchange approach to valuing a block discount because it compares different rates of dribbling out a holding of shares and takes whichever rate provides the lowest aggregate cost of (i) a put option at the valuation date price over the dribble out period and (ii) the assessed price impact of that velocity of sale. It is described in this way by Mr Estabrook:

"Another method of estimating a blockage discount suggests that there are two separate and distinct cost components of blockage that should be measured. The first cost component is price pressure. This can be defined as the impact on stock price when a large block of stock depresses the market and lowers the price that can be obtained for the stock. The second cost component is market exposure. This is the cost associated with bearing the risk of holding a position in the marketplace without the ability to close the position, for a specified period of time. The appropriate blockage discount is that combination of price pressure and market exposure that produces the least cost to the seller of the block. Using this approach appears to satisfy both requirements of the tax regulations: (1) that the taxpayer show that the length of time is reasonable, and (2) that the sale of smaller blocks of stock would depress the sales price.

In estimating price pressure, the factors listed earlier should be taken into consideration. The valuation analyst should analyze the daily price volatility in relation to trading volume over at least a three- to six-month period (longer if deemed appropriate) in order to identify the amount of additional price movement that is created by the trading of the additional shares. From this analysis, the amount of the excess volatility or negative pricing pressure created by selling the shares is estimated.

Market exposure is estimated by calculating the cost of buying a put option on the subject company shares (1) at a strike price equal to the traded sales price at the valuation date…, (2) exercisable in the number of days determined under the different trading period scenarios, (3) based on the results of the analyst's due diligence. The cost of the put option is determined under a conventional option model (e.g., the Black-Scholes option pricing model). Such a model takes into consideration the five variables of stock price, exercise price, risk-free rate, time to expiration, and volatility (all on an annualized basis)."

A long dribble out period may result in a small discount for price pressure, but the long period will require a costly put option to insure against the risk of market falls from other causes. Correspondingly, a short dribble out period would reduce the risk of a fall in the market but the pressure on share price from the volume traded is likely to be greater. The two factors are clearly connected in the cost of the put option, and somewhere between the extremes will be a dribble out period that is long enough not to cause significant impact on price but not so long as to make the cost of the option excessive. Mr Steadman's (and Mr Estabrook's) approach is to find the optimum period that reduces the overall cost.

247. Mr Steadman was criticised in cross-examination for relying on this alternative method that was suggested only in an out-of-date textbook, but his answer was that it is an orthodox method of identifying the least cost of liquidating a holding where there is both price pressure from the size of the holding and risk from market exposure. He rejected the criticism that he was pre-selecting one period during which the holding would be liquidated and explained that his method looks at many different possible periods with a view to identifying the most cost-effective. The approach taken in Estabrook seems to me to be entirely logical if one accepts that there can be price pressure resulting from selling a stock at relatively low velocity and that the time dimension is a risk not a neutral factor.
248. In my judgment, the method used by Mr Steadman is an appropriate means of identifying the cost of the seller's alternative to selling on a single day and incurring a block discount. I accept the evidence of Mr Steadman that there may well be some price pressure caused by dribbling out the holdings in the Disputed Securities, given the relatively large numbers involved, and that the risk of a fall in the market price over the period is one that carries a cost.
249. Both hypothetical seller and hypothetical buyer in a market value negotiation are assumed to be knowledgeable and prudent, and so will be well-informed about the seller's alternatives to agreeing a price for a single transaction on the valuation date. I accept Mr Steadman's evidence that any block discount in the market is likely to reflect the parties' knowledge of the likely cost of the alternatives available to the seller, though it will not necessarily be identical to that cost, otherwise there is no incentive to the seller to sell to the buyer and the buyer would lose out on the bargain. If market value is determined in this way, as Mr Steadman proposes, it does not give rise to a discount that could reasonably have been avoided by a trustee selling the Disputed Securities: it gives rise to a discount that does not exceed the cost and exposure of any alternative course open to the trustee. That being so, Mr Steadman's use of market value as a basis of valuation is not open to the claimants' criticism that its assumptions are incompatible with the duties and powers that Samba would have as a trustee of the Disputed Securities.
250. For these reasons, I consider that the market value basis of valuation propounded by Mr Steadman is the appropriate basis of value in this case. In that context, the Black-Scholes put option model used by Mr Steadman as a component of the discount is clearly the right model for a market value valuation: it would be necessary to guarantee the receipt of the valuation date price over an extended dribble out period. The Finnerty model is appropriate only if one is seeking to guarantee sale of a large holding at whatever might be the reference price on a future date. However, the orderly liquidation value is inappropriate because it will expose the seller to a much greater discount to reflect price uncertainty, delay and the risk of a fall in the market from extraneous factors. Mr Worsnip's dribble out strategy is therefore better seen as the means of seeking to identify the approximate amount of a block discount than as a measure of liquidation value.

(e) Evidence of market value

251. The claimants also argued that if market value is the right basis of value it is unnecessary to resort to quantitative modelling because there are comparable transactions that demonstrate that a block discount is inappropriate. If there is evidence of no discount having been agreed in the market on a sale of a similar size of holding on a date close to a valuation date, that is likely to be the best evidence available.
252. The claimants rely on two sets of comparable transactions. First, Samba's own purchase of the Disputed Securities and a sale of a 2.2% holding in Samba Financial Group at about the same time as the September Transfer; second, privately negotiated deals in shares in the five JSCs over an extended period from 2011 to 2020.
253. Samba's acquisition of the Disputed Securities in part satisfaction of Mr Al-Sanea's debt (whether as principal or as guarantor of SICL's liability to Samba) is recorded in its client statements of account for its own treasury and for Mr Al-Sanea at the same price, calculated on the basis of the quoted market price for each of the four listed banks on 26 September 2009. There is no discount shown. Other documents that might have cast light on how the September Transfer came about and was recorded are not available, owing to Samba's failure to give disclosure. The value in the client accounts may reflect fair value or accounting value (mark to market accounting) in compliance with IFRS 13 rather than market value; nevertheless it appears that Samba credited Mr Al-Sanea's account with the undiscounted market value of the Disputed Securities.
254. This was a transaction in which Samba accepted the Disputed Securities in partial discharge of Mr Al-Sanea's debts. By the date of the September Transfer, worldwide freezing injunctions had been granted against Mr Al-Sanea and his companies by the Saudi Arabian Monetary Authority, the Grand Court of the Cayman Islands and the High Court in London. Samba had been notified of those injunctions. Samba was very exposed to Mr Al-Sanea's debt. The release of debt for the Disputed Securities was not an open market transaction and may not have been at arm's length. Samba as a creditor of Mr Al-Sanea had a special interest in exchanging illiquid debt of questionable value for liquid assets. I reject Mr Worsnip's assertion that Samba was not a special purchaser. Standing back, Samba's treatment of this settlement between it and Mr Al-Sanea in its own books cannot be seen as good evidence of whether in the open market a sale of entire holdings such as the Disputed Securities would be undiscounted.
255. Little is known about the sale of a 2.2% holding in Samba Financial Group save that it was acquired without a discount by the Saudi Arabian Public Pension Agency. The seller is unknown, as are the circumstances of the sale. This evidence, so far as it goes, shows that on occasions, at about the date of the September Transfer, there may be an institutional buyer for large blocks of shares in Saudi Arabian banks, who may be willing to buy without a discount. But it cannot be said that this one transaction establishes a general practice in the market, or that the same or a similar buyer might have been interested in the Disputed Securities at the time of the September Transfer. Clearly, it says nothing about the state of the market at the judgment date.
256. The second set of comparable transactions is a collection of privately negotiated deals in the shares of the five JSCs from January 2011 to 2020. So far as possible, those deals that were seller-initiated are identified (the experts agree that buyer-initiated deals do not provide good comparable evidence) and exceptional and potentially misleading transactions have been excluded. Mr Steadman also excludes from his analysis deals that took place at no discount, whereas Mr Worsnip reasons (rightly, in my judgment) that it is likely to be more accurate (and avoid prejudicing the outcome of the issue) to regard half of these has having been seller-initiated, rather than all of them as buyer-initiated. The data is not a complete set, however, and the majority of transactions are at lower volumes than the size of the holdings in the Disputed Securities; much of the evidence is from transactions in 2019. These factors lessen the reliance that can be placed on these transactions as presenting a complete and reliable picture – in the event neither valuer relied on them to any degree in expressing his final opinion – but in my judgment they are of some value nonetheless in indicating the range of discounts at which seller-initiated private deals of shares in the five JSCs take place on the Tadawul, albeit not at a time close to the September 2009 valuation date.
257. On balance, I preferred Mr Worsnip's analysis of these transactions. His arguments for excluding outliers and including half of the transactions at the quoted price were more compelling than Mr Steadman's approach to them. I was not persuaded by Mr Steadman that it is right to assume that all or most of the 2019 transactions should be disregarded. Mr Worsnip analysed these transactions as showing discounts ranging from 1.6% for Samba Financial Group to 4.3% for Saudi British Bank. This evidence is in my judgment of some value, at least as a cross-check against the size of discount produced by quantitative modelling.
258. Turning to that, on which Mr Steadman exclusively relied for his final opinion, I am not persuaded by the reliance that he ultimately placed on the Keim and Madhavan paper (The Upstairs Market for Large-Block Transactions: Analysis and Measurement of Price Effect (1996)) for the conclusion that there would be a significant permanent effect on the value of the stocks of the five JSCs on account of low-level trading. That paper was not addressing impact on stock values for companies of the size and importance of the five Saudi Arabian JSCs with which this case is concerned, where significant public information and analysis about them is generally available. Mr Steadman assumes quite a lot in contending that the Keim and Madhavan conclusions can apply to such companies in the same way that they apply to smaller US companies. I do not accept that the significant degree to which relatively low-level trading in stocks of smaller US companies has a permanent impact on share price can be applied directly to large Saudi Arabian bank stocks, though I do accept that there may be some impact that goes beyond the effect of temporary illiquidity in the market.
259. Despite those reservations, I nevertheless accept in principle Mr Steadman's basic approach to valuing the cost to a seller of the alternative sale strategy, in that it will involve identifying and determining a degree of price pressure and exposure to market volatility that can only be avoided by purchasing a Black-Scholes type of put option covering a defined dribble out period. Mr Steadman has used Professor Kissell's I-Star model (frontier markets 2012 calibration) to attempt to calculate the price pressure on various velocities of dribble out of the holdings. Mr Worsnip challenged the appropriateness of the frontier markets calibration on the basis that the Tadawul was more akin to western European markets than other Gulf or Central Asian states. However, as Mr Steadman convincingly explained, the principal characteristics of the Tadawul are that it was (and remains to some extent in 2020) a local, mainly retail market, with no derivatives market or exposure to short selling, which makes Mr Worsnip's comparison with western Europe misplaced.
260. Mr Worsnip also challenged the reliability of the I-Star model on the basis that it was not developed for the purpose of ascertaining block discounts; has not been sufficiently tested to apply it to large blocks of securities such as the Disputed Securities; and is proved to be unreliable for quantities of shares exceeding 15% ADTV. I accept that these are criticisms that can fairly be made and they do have an impact. In the final iteration of his modelling, Mr Steadman broke down the daily trades into sizes of blocks on which the model had been fully tested, which had the effect of reducing the impact. I agree with the claimants, however, that those adjustments did not address all the doubts about the applicability of the modelling.

(f) The level of applicable discounts

261. In his reply expert report, Mr Steadman concluded that the following block discounts applied to the holdings in the Disputed Securities:

Sept 2009 Jan 2020

ANB 9.3% 3.8%

BSF 26.5% 7.5%

Samba 16.9% 9.2%

SBB 26.0% 8.1%

NCB 3.4%

(No discount is required for NCB in September 2009 because the claimants elected before the start of the trial to have the NCB shares valued as at the date of judgment.) Mr Steadman's original report had contended for discounts in the region of 25% to 35% higher in all cases. In comparison, Mr Worsnip's discounts (calculated on a wholly different basis) ranged between 1.26% - 5.4% for the September 2009 valuation date and between 0.78% - 2.56% at the judgment date. Both valuers accepted that the discounts were appropriately higher in September 2009 than in January 2020.

262. The levels of discount proposed by Mr Steadman are surprisingly high, though a direct comparison with Mr Worsnip's discounts is inappropriate because he has carried out his assessment on a different basis. They are substantially higher than the only hard evidence of discounts resulting from the 2011-2020 privately negotiated trades. I am concerned that Mr Steadman has used, to the exclusion of all other evidence, a calibration of the I-Star model that creates results that exceed what is inherently likely for stocks of this quality. My doubt is not about the appropriateness of the Black-Scholes put option as such (though the price pressure impact will affect the price of the put option) but about the output of the I-Star model as a means of assessing the degree of price pressure.
263. Doing the best that I can in the light of all the evidence that I have read and heard relating to this issue, including the privately negotiated deals, and bearing in mind that the costs associated with the alternative realisation strategy only inform and do not dictate the level of block discount for a sale in the market on the valuation date, I will reduce Mr Steadman's discounts by 30% across the board at both valuation dates. The resulting discounts are, in my judgment, probably about the level of discounts that would be negotiated on the valuation dates for the respective holdings of the Disputed Securities. The September 2009 reduced discounts will enable the expert valuers to calculate and agree the value of the Disputed Securities (other than NCB) on that valuation date. Similarly, the judgment date discounts will enable them to calculate and agree the prices as at the date on which I hand down this judgment, should the claimants so elect.

VI. Conclusion

264. For the reasons that I have given in relation to the Law of Knowing Receipt Issue and the Saudi Arabian Law Issue above, I dismiss the claimants' claim. Had I reached a different conclusion on liability, I would have determined that a block discount was applicable in the case of each of the holdings of the Disputed Securities in the following amounts:

Sept 2009 Judgment date

ANB 6.5% 2.7%

BSF 18.6% 5.2%

Samba 11.8% 6.4%

SBB 18.2% 5.7%

NCB 2.4%

265. It remains for me to express my gratitude and admiration for the preparation and presentation of the parties' cases on these complex issues. The presentation and argument on both sides, both oral and written, was of the highest quality and, in the best traditions of the English Bar, I was given all the assistance that a judge could reasonably ask for.