United Kingdom Journals

		[2000] 2 Web JCLI

Caught in the Tangled Web

Alistair Speirs*

Newcastle Law School
University of Newcastle

<a.c.speirs@ncl.ac.uk>

* I am grateful to Professor John Alder, Joanna Gray, Dr Alison Dunn and the anonymous referee for their comments on earlier drafts of this piece.

Copyright © 2000 Alistair Speirs.
First Published in Web Journal of Current Legal Issues in association with Blackstone Press Ltd.

Summary

This casenote examines the Court of Appeal judgment in Twinsectra Ltd v Yardley and Others [1999] Lloyd's Rep Bank 438, in which the court considered the liability of a solicitor to account to a third party from whom his client had borrowed £1 million. The sum was lent subject to a stipulation that it be used for acquiring property. The court was first required to decide whether the terms of the loan created a trust. This question received an affirmative answer. The second main issue was whether the solicitor had acted in such a way as to satisfy the requirements of one or both forms of third party liability traditionally labelled 'knowing receipt' and 'knowing assistance'. The latter form of liability, following the Privy Council decision in Royal Brunei Airlines Sdn Bhd v Tan [1995] 2 AC 378, is now more properly identified as 'dishonest assistance'.

The casenote considers the court's analysis of the loan in terms of a trust as well as the test of dishonesty which the court applied. It suggests that the court was correct to conclude that a trust did arise but criticises the court's analysis. It also argues that there is some confusion evident in the application of the test of dishonesty introduced in Tan.

Contents

Introduction

Knowing receipt and dishonest assistance

Facts

(a) Twinsectra's claims

(b) The claim against Yardley

(c) The Equitable Claims

• (i) The existence of a trust
  • 1. Leach's liability
    • (a) Assistance
    • (b) Receipt
  • 2. The Companies

Comment

(a) The existence of a trust

(b) Liability of third parties associated with a breach of trust

Conclusion

Bibliography

Introduction

Increasingly, those who suffer financial loss seek to recover from peripheral participants in a transaction. This is particularly the case where the core players cannot be relied upon to satisfy judgment. Where the peripherally involved belong to a well-insured profession they become a natural target. Given the restrictions on the recovery of economic loss in negligence, an action either claiming that the defendant received funds derived from a breach of trust or that he assisted in such a breach may offer the plaintiff the only route to making good his loss. As a result equitable claims brought against recipients and accessories occupy a growth area of the law.

Knowing receipt and dishonest assistance

Prior to the decision in Royal Brunei Airlines Sdn Bhd v Tan [1995] 2 AC 378 (hereinafter Tan), knowledge of facts relating to the existence of a trust and its breach was the basis upon which a defendant was held liable to account either as a recipient of property derived from a breach of trust (knowing receipt) or as accessory to such a breach (knowing assistance)⁽¹⁾. In Tan, however, the Privy Council held that dishonesty, rather than knowledge, was the touchstone of liability to account for assistance given in a breach of trust. Accordingly, and notwithstanding the formal lack of authority of the case, this form of accessory liability should be assigned the terminology 'dishonest assistance' (see, eg Hanbury & Martin 1997, p 294). Since the issue did not arise in Tan, the basis of liability for receipt of trust funds remained undisturbed by the Privy Council.

In Twinsectra v Yardley the Court of Appeal used the concepts of dishonesty and knowledge without drawing any real distinction between these two types of liability. As a result, this case confuses rather than clarifies this area of the law.

Top | Contents | Bibliography

Facts

The appeal concerned a £1m loan made by the plaintiff (Twinsectra) for the purposes of Mr Yardley's property development business. Also named as defendants were Yardley's solicitor, Mr Leach, and 4 companies controlled by Yardley.

The loan sought by Yardley was for purchase of a site in Bradford (Apperley). Upon encountering delays by his own high street bankers Yardley was introduced to Twinsectra which was willing to advance £1m against a solicitor's undertaking. However Leach, understandably, refused to give an undertaking which would have committed himself and his partners to repay the loan and interest.

Although centring upon a reasonably straightforward business loan, the facts of the case are complicated by what started out as an entirely separate transaction entered into by Yardley. This transaction (the Nigerian scheme) required payments to be made to Nigerian officials who would then release the sum of $49.2m⁽²⁾. This was to be distributed to the 'investors' resulting in large returns for their relatively small outlay. The scheme was promoted by a solicitor, Mr Sims. At the time at which the loan was being negotiated Sims had entered into a personal undertaking guaranteeing Yardley a return of $1.5m on his initial outlay of $75,000 within 3 months. A second undertaking (this time conditional upon the success of the scheme) was to be given against a further $50,000 'investment' by Yardley.

Yardley required a solicitor to give an undertaking to Twinsectra. He turned to Sims who, recent dealings had demonstrated, was clearly untroubled by the giving of personal undertakings.

The undertaking proposed by Twinsectra would commit Sims and his only partner, a Miss Roper, to repay £1m with interest within 4 months. Crucially, it also stipulated that the money should be retained by Sims' firm until 'applied in the acquisition of property'.

Prior to entering into the undertaking Sims became aware that Yardley had received the necessary finance for the purchase of the Apperley site from his own bank. Sims therefore asked Leach whether Yardley intended to use the loan in the acquisition of other property. Sims also asked whether he had authority to divert £34,000 of the loan to the Nigerian scheme, this being the sterling equivalent of the $50,000 proposed further investment by Yardley.

Leach was aware of the stipulation contained in the proposed undertaking. He declined to get involved, saying that this was a matter for Sims. Nevertheless, Leach took instructions from Yardley who stated that the money was to be used in connection with other property deals and gave approval for the diversion of the £34,000.

In due course Sims and Roper signed the undertaking sought by Twinsectra (hereinafter 'the undertaking'). Shortly thereafter the £1m loan was paid by Twinsectra to Sims who forwarded it (less the £34,000) to Leach to hold on Yardley's account. This was clearly a breach of the stipulation that the money was to be retained by Sims' firm until applied in the acquisition of property.

The remainder of the £1m was paid out by Leach, on Yardley's instructions, in 12 separate tranches. Of these, 3 were received by Leach's firm in payment of fees. Most of the rest, representing the bulk of the £1m, were paid to the 4 companies controlled by Yardley (hereinafter 'the companies') and then employed in various business ventures. Only 3 of the 12 payments were attributable to the acquisition of property.

Shortly after these transactions Sims was, effectively, removed from the scene when he was declared bankrupt.

Top | Contents | Bibliography

(a) Twinsectra's claims

As a first step Twinsectra obtained judgment against Sims' partner, Miss Roper, who was jointly liable on the undertaking. She was unable to satisfy judgment. A claim by her against the Solicitors' Indemnity Fund, financed by Twinsectra, failed. Twinsectra then brought the claims that were the subject of this appeal. These fell into 3 categories:

(i) against Yardley, alleging breach of contract, alternatively deceit;

(ii) against the companies, on the ground that they received money paid out in breach of trust;

(iii) against Leach, both as recipient of part of the loan (his fees) and for assisting in a breach of trust.

(b) The claim against Yardley

At first instance Carnwath J had held that Sims had acted as Yardley's agent in respect of the loan agreement and, therefore, that Yardley was liable for the failure to repay on normal contractual principles.

However, despite holding that Yardley was dishonest in not disclosing that he no longer needed the loan to finance the purchase of the Apperley site, Carnwath J did not find him liable in deceit. This decision was apparently on the basis that the dishonest representation did not undermine the agreement because Yardley was bound by it. As Potter LJ, who gave the only judgment in the Court of Appeal, pointed out, this is to misunderstand the principle. The issue is not whether the dishonesty prevented a binding agreement being formed but whether Twinsectra had relied upon the representation that the loan was needed for the purchase of the property. The evidence showed that Twinsectra had relied upon the understanding that Sims was acting for Yardley in a property venture. In reality, Sims' role extended only to giving the undertaking, receiving the £1m and passing it on. Yardley was therefore liable in the tort of deceit. This allowed for a more generous measure of damages to include the cost of the action against the Indemnity Fund.

(c) The Equitable Claims

(i) The existence of a trust

Carnwath J had held that the claims against the companies and against Leach failed. Each depended upon the existence of a trust. The plaintiff argued that a trust had arisen because the loan was made subject to the restriction contained in the undertaking. Under the principle in Barclays Bank v Quistclose Investments Ltd. [1970] AC 567, this created an express trust and a resulting trust which attached to any portion of the loan not so used. However, Carnwath J distinguished Quistclose on the ground that the purpose (the acquisition of property) was insufficiently certain to allow for the creation of a trust of the £1m. This finding made it unnecessary to consider the equitable claims. In the absence of a trust, of course, tracing and personal claims based upon receipt or assistance did not get off the ground.

Potter LJ revisited the question of whether a Quistclose-type trust had arisen. The fact that the transaction was a loan did not exclude a trust, indeed this was the case in Quistclose itself (see Lord Wilberforce at 581D). However, more was required than merely a specific purpose for which the loan is to be used. Typically this extra feature is the requirement that the money lent should be kept separate from other assets. This was the case here. Further, Potter LJ held, it was not necessary for a Quistclose trust to arise that the parties specify the purpose with the same precision which is required to satisfy the otherwise strict requirement that a trust have 'certainty of object'. The phrase 'acquisition of property' was not so vague as to be unenforceable by a court. Accordingly, he held that the money in Sims' hands was impressed with an express trust based upon the intention of the parties as set out in the undertaking.

In fact this analysis was made unnecessary. Potter LJ relied upon a recent observation by Millett LJ in Barclays Bank v Weeks Legg and Dean [1999] QB 309 at 324 to the effect that a fiduciary relationship would normally arise whenever a solicitor holding money on a client's behalf expressly undertook to the third party who advanced the funds to use them in a particular way⁽³⁾. This fiduciary duty is owed to the third party and requires the solicitor to apply the money in accordance with the terms of his undertaking. Sims had breached this duty by paying out the money, most obviously in respect of the £34,000 to be used to bribe a Nigerian official, but also by paying over the remainder to Leach without imposing the restriction that it was to be used only in acquiring property.

1. Leach's liability

This fell to be considered in 3 parts:

(i) assistance-based liability

(a) for the portion of the £1m paid out for purposes other than the acquisition of property;

(b) for the portion which was paid out for the acquisition of property.

(ii) receipt-based liability for the small portion accepted by Leach in discharge of his own fees.

(a) Assistance

Following Tan, the four elements necessary to establish liability under this head are contained within the phrase "dishonestly assisting in a breach of trust". As set out above, the existence of a trust and its breach had been established by Potter LJ. Leach had clearly assisted in the breach; it was he who paid out the money at Yardley's direction.

The remaining element, dishonesty, posed more of a problem. Carnwath J had held that, although Leach had deliberately shut his eyes to the implications of the undertaking, he had not been dishonest. On behalf of the plaintiff, Mr Tager QC asserted that Carnwath J was only referring to conscious dishonesty; Leach was guilty of 'Nelsonian dishonesty'. He had known of Sims' involvement in the Nigerian scheme ('a scheme of doubtful legality') and the link between that scheme and the loan. He had also known of the restriction imposed by the undertaking and the diversion of the £34,000.

Counsel for Leach argued, inter alia, that he had had no reason to consider that the £1m was impressed with a trust in the hands of Sims. In Tan it had been clear that the funds were trust property. Here it was far from clear, in fact at first instance Carnwath J had held that there was no trust.

Potter LJ dismissed these arguments. As a solicitor Leach was required to maintain a high standard of honesty. He had breached that standard by deliberately closing his eyes to Sims' failure to respect the plaintiff's rights created by the undertaking. Whether these rights were regarded by Leach as equitable or merely contractual was not material.

Consequently, Leach was liable to account for the sums paid out for purposes other than the acquisition of property. With little real analysis, however, Potter LJ held that Leach was not liable in respect of the part of the £1m which had been employed in the acquisition of property. He stated that, although outwith the letter of the undertaking, these payments were within its spirit.

(b) Receipt

Having reversed Carnwath J on his finding as to the existence of a trust it followed that Leach was liable for the (small) part of the loan which he received beneficially, given the view taken at first instance on Leach's state of knowledge. Acting like Nelson qualifies as category (ii) type knowledge under the 5-pronged approach adopted by Peter Gibson J in Baden, Delvaux and Lecuit v Societe Generale pour Favoriser e Developpement du Commerce et de l'Industrie en France SA [1993] 1 WLR 509 at 575. Accordingly, Leach was liable to account for the sums he received in discharge of his firm's fees.

2. The Companies

Twinsectra sought a tracing order in respect of the parts of the loan received by the companies. A proprietary remedy was preferable because the companies were by now in administration. Potter LJ held that this claim succeeded. The funds had left Sims' hands in breach of trust. This was true not only of the money expended in clear breach of the undertaking but also those parts of the loan which had been employed in the acquisition of property. It followed that a proprietary tracing order was available in respect of all of the original £1m which had been received by the companies .

Top | Contents | Bibliography

Comment

(a) The existence of a trust

It is trite law that trusts, in general, must be for the benefit of ascertainable beneficiaries (Morice v Bishop of Durham (1805) 10 Ves 522). The exceptions are charitable trusts and a small number of recognised purpose trusts. The latter are 'troublesome, anomalous and aberrant' and ought not to be increased in number (per Harman LJ Re Endacott [1960] Ch 232 at 251). The Quistclose trust poses a problem in that it is not immediately apparent that there is a beneficiary. Where the loan is to be used to pay an identifiable individual or class of individuals it may be thought that the individual or class satisfies the requirement. Most of the cases cited by Potter LJ had such person or persons as the target of the purpose. In Gibert v Gonard (1884) 54 L J Ch 439 it was a vendor, in Quistclose shareholders and in Re Rogers (1891) 8 Morr 243, Toovey v Milne (1819) 2 B & Ald 683 and Carreras Rothmans Limited v Freeman Matthews Treasure Ltd [1985] Ch 207 creditors.

Some cases appear harder to analyse in terms of a beneficiary. In Re Northern Development (Holdings) Ltd (unreported, 6 October 1978) Megarry VC employed the principle of Re Denley's Trust Deed [1969] 1 Ch 373⁽⁴⁾ that, although the trust was expressed to be for a purpose, it was actually for the benefit of identifiable individuals. The Vice-Chancellor's approach has been criticised by Millett (see Millett 1985 at 280). It certainly is not available where the purpose is the acquisition of property. Where, then, was the beneficiary in Twinsectra?

The solution reached by Potter LJ was to label the Quistclose trust a 'quasi-trust' which does not need 'certainty of objects'. The use of the latter phrase, in the context, seems to suggest some confusion between the requirement of 'certainty of objects' (the need for identifiable beneficiaries) and the separate requirement that the terms of the trust must be sufficiently clear to enable the court to enforce the trust. A non-charitable purpose trust may fail both because there is no identifiable beneficiary and because it is too vague to be controlled by the court (see Re Astor's Settlement Trusts [1952] Ch 534 per Roxburgh J at 547).

It is submitted that it is not necessary to depart from conventional trust analysis to explain the Quistclose trust even where the purpose cannot be construed as benefiting identifiable persons. The solution is contained in Millett's article (Millett 1985). Millett recognised that the sum loaned remained the beneficial property of the lender until employed in the permitted purpose. In Sims' hands the £1m is held on trust for Twinsectra, subject to a power to pay out for the purpose of acquisition of property. The trust satisfies the beneficiary principle; it does have 'certainty of objects'. 'Certainty of intention' is demonstrated both by the restriction imposed and the requirement that the loan be held separate from other funds. The remaining question is whether the stated purpose is clear enough to be enforced by the court. In Re Challoner Club, The Times, November 4, 1997, it was the intention of the directors of the club to create a trust of money donated to secure the club's future. However, the terms on which the money was to be used were not sufficiently clear, leading Lloyd J to hold that there was no trust. By contrast, in Twinsectra Potter LJ rightly concluded that 'the acquisition of property' was a phrase that the court could construe and enforce.

(b) Liability of third parties associated with a breach of trust

It appears that the distinction between knowing receipt and dishonest assistance stands. This has recently been confirmed by the Court of Appeal in Houghton v Fayers, The Times, February 14, 2000. Lord Nicholls in Tan saw a clear distinction between knowledge and dishonesty. Knowledge was an 'inapt criterion' for accessory liability whereas the question of dishonesty was a meaningful one which was capable of being given a meaningful answer (see Tan, p 107).

The distinction between the two forms of third party liability is justified by the different bases of liability. Liability for assistance is fault-based (Tan, p 104). Liability for receipt derives from restitutionary principles (Tan, p 103)⁽⁵⁾.

In Twinsectra, muddled use was made of the terms. Potter LJ referred variously to 'dishonest recipients' (para. 111), 'dishonest assistance', 'knowing assistance' (para. 114) and "knowing (i.e. dishonest) assistance" (para. 116). The term 'Nelsonian dishonesty' is particularly unhelpful and was described as 'unfortunate' by Aikens J in Bank of America v Arnell [1999] Lloyd's Rep Bank 399. It can only serve to cloud the issue of the basis of liability and should not be permitted to enter the legal lexicon. The phrase 'Nelsonian knowledge' has a pedigree (see Baden, p 576) and is understandable. Nelson knew what he would see if he used his good eye. It is far less clear that, when he clapped the telescope to his bad eye, he was 'not acting as an honest [captain] would in the circumstances' (Cf. Tan, p 105j).

The dangers of confusing the terms is illustrated by the decision in Bank of America v Arnell. In this case Aiken J appears to have been misled by reference to Twinsectra to the extent that he held that the defendant had not been dishonest and therefore was not guilty of knowing receipt. He had observed the defendant in the witness box and had formed the view that she was not dishonest. This was to ask and to answer the wrong question in two respects.

Turning to Leach's liability, the case raises three important points on accessory liability. First, is the question of whether dishonesty in the abstract is sufficient to impose liability for dishonestly assisting in a breach of trust or whether there is a requirement that the defendant know of the existence of the trust. In Tan Lord Nicholls held that dishonesty was not only a necessary but also a sufficient ingredient (at p 109a). Fox LJ in Agip (Africa) Ltd v Jackson and others [1991] Ch 547, took the view that a defendant who acted dishonestly could not escape liability by claiming that he thought his conduct was only directed at evasion of exchange controls (at p 569). However, in Brinks Ltd v Abu-Saleh, The Times, October 23, 1995, although it was not necessary to his decision, Rimer J indicated that he was of the view that knowledge of the existence of a trust was necessary (see Ferris 1996). Leach, it was argued, did not know that there was a trust to be breached. However, the distinction between Leach and Mrs Elcombe (the defendant wife in Brinks) was that Leach knew all of the facts relating to the existence of the trust and the breach. If, not unreasonably, he did not know that the sum which Twinsectra had loaned was the subject of a trust it was because he made a mistake of law.

Second, was the question of whether Leach was dishonest. Carnwath J thought not but Potter LJ held that this was because he was considering whether Leach had been subjectively dishonest. Tan makes clear that the concept of honesty imposes an objective standard leavened by a subjective element based upon the characteristics and knowledge of the particular defendant (p.105-6). This is not, however, to be equated with the subjective element found within the criminal law test of dishonesty⁽⁶⁾. Recent cases have struggled with this point⁽⁷⁾. One of the factors which is to be had regard to is the ordinary course of business (see Tan, p 107e). However, as Ferris points out, the Privy Council was not concerned to ask whether Tan was acting in accordance with accepted practice in the industry (see Ferris 1996). Potter LJ was clearly satisfied that he could properly decide that an honest solicitor would not have acted as a conduit for funds which came to him as the result of a breach of an undertaking.

The third point relates to the puzzling decision on the funds which were applied in the acquisition of property. Potter LJ held that no equitable remedy arose against Leach. He put this on the basis that if there was a breach of the undertaking it was only of the letter rather than its spirit. However, this held true only for Leach and not the companies (para. 112).

Equity holds trustees to a high standard and does not usually excuse because the breach was only technical⁽⁸⁾. The transfer by Sims to Leach without imposition of a restriction was, as the Court found, a breach of trust. This must be true whatever the eventual destination of the money. Leach assisted in that breach of trust. Why was he not liable to account for the money that was used in acquisition of property? A better answer than that advanced by Potter LJ is that he did not act dishonestly in respect of these sums. The test is whether an honest solicitor would receive funds from another solicitor, knowing that solicitor was technically in breach of an undertaking, but also knowing that the funds were to be applied according to the lender's stipulation. It is submitted that he would. It is further submitted that there is no bar to holding that an individual acted honestly as to part of a transaction but dishonestly as to the rest.

Top | Contents | Bibliography

Conclusion

Twinsectra raises a number of issues. It goes further than the authorities which are relied upon in holding that a trust arises by virtue of a wide stipulation in an undertaking. It is submitted, however, that the phrase 'quasi-trust' should be avoided and that the trust which arises by reason of such a stipulation is susceptible to explanation on usual trust principles. It also illustrates that Tan has not dispensed with the need to examine closely the defendant's state of knowledge and has introduced some confusion between the equitable law and criminal law meanings of dishonesty.

In Tan, Lord Nicholls was of the view that 'in most situations there is little difficulty in identifying how an honest person would behave' (at p 106c). Twinsectra suggests that the issue may not be so easily resolved in more cases than anticipated by Lord Nicholls. In particular, should a defendant argue that he was acting in accordance with ordinary commercial practice this may raise difficult issues for the court to resolve. Is a defendant to be permitted to introduce expert evidence? As in the medical negligence decision in Bolam v Friern Hospital Management Committee [1957] 1 WLR 582, the question arises as to whether a sector may set its own standard. The answer must surely be that it may not.

The decision in Twinsectra offers a cautionary tale to solicitors, in particular. Leach took some care to distance himself from his client's dishonest dealings. In reality, his approach had more similarity with that of Pontius Pilate than Nelson. The Court was correct to hold that the distance he attempted to put between himself and the transaction did not justify relieving him of liability.

Top | Contents

Bibliography

Ferris G (1996) 'The advice of the Privy Council in Tan' 30 Law Teacher 111.

Hanbury & Martin (1997) Modern Equity, 15th ed., (London: Sweet & Maxwell).

Millett P J, QC (1985) 'The Quistclose Trust: Who Can Enforce It?' 101 Law Quarterly Review 269.

Lord Nicholls (1998) 'Knowing Receipt: The Need for a New Landmark' in Cornish, W (ed) Restitution, Past, Present and Future: Essays in honour of Gareth Jones (Oxford: Hart) Chapter 15.

(1) These terms derived from the much-cited dictum of Lord Selborne LC in Barnes v Addy (1874) LR 9 Ch App 244 at 251-252.

(2) The facts of the scheme align with what is now recognised as a familiar scam.

(3) This case concerned the liability of solicitors (acting for property purchasers) on standard form undertakings given by them to banks providing purchase monies to the said clients.

(4) A trust for the maintenance of a sports ground was for the benefit of employees who were to enjoy it.

(5) Consequently some commentators argue that liability should be strict, see e.g. Lord Nicholls (1998).

(6) Cf. R v Ghosh [1982] 1 QB 1053.

(7) See e.g. Mortgage Express v Newman [1999] EGCS 123 and Abbey National v Solicitors' Indemnity Fund [1997] PNLR 307.

(8) See e.g. Boardman v Phipps [1967] 2 AC 46.