Skandinaviska Enskilda Banken AB v Conway and another

This is the judgment of the Board, to which all its members have contributed. The Board is unanimous in its advice to Her Majesty with regard to the outcome of the appeal. It is also unanimous in its reasons for arriving at its conclusions, except to the extent that Sir Donnell Deeny (with whom Lord Wilson agrees) has reached his conclusion on the first issue in the appeal, which we shall describe as the fraud point, for different reasons from the majority of the Board. Those reasons are set out in Sir Donnell’s concurring judgment, which also expresses reservations about the majority’s reasoning concerning an aspect of another set of issues, which we shall describe as the repayment issues, relating to a potential defence of change of position.

This appeal from the Court of Appeal of the Cayman Islands concerns certain share redemption payments made by Weavering Macro Fixed Income Fund Ltd (“the Company”) to the appellant, Skandinaviska Enskilda Banken AB (Publ) (“SEB”), between December 2008 and February 2009. On 19 March 2009 the Company went into liquidation. The central issue in this appeal is whether these payments constituted unlawful preferences over the other creditors of the Company within the meaning of section 145(1) of the Cayman Islands’ Companies Law (2013 Revision). The respondents (“the liquidators”) have brought these proceedings in their capacity as joint official liquidators of the Company.

The Company was incorporated in April 2003 as an open-ended investment company pursuant to the law of the Cayman Islands. It traded mainly in interest rate derivatives. Its two directors were Stefan Peterson and Hans Ekstrom, but its investment manager was an English company known as Weavering Capital (UK) Ltd (“WCUK”). The director of WCUK, and also its Chief Executive Officer and Principal Investment Manager, was Magnus Peterson, the brother of Stefan Peterson and the stepson of Hans Ekstrom. As the Company’s investment manager, WCUK undertook the Company’s trading activities. Clifford J, who heard this case at first instance, found that: “Magnus Peterson directly, and through his company WCUK, managed and controlled the Company for all purposes relevant to these proceedings. He controlled the investments and he made the material decisions about redemptions”. There was no appeal against that finding.

Those who wished to invest in the Company acquired redeemable participating shares in it. SEB, a Swedish financial institution, was such an investor. Between 2006 and 2008 SEB subscribed for shares on behalf of, amongst others, two Swedish mutual funds, HQ Solid and Catella Stiftelsefond (“Catella”). Between April 2006 and November 2007 SEB subscribed for US$8.5m of participating shares on behalf of HQ Solid. In March 2008 it subscribed for US$1m of participating shares on behalf of Catella. In each case, the Company issued such participating shares to “SEB Merchant Banking as nominee for [HQ Solid/Catella]”. SEB was registered as the holder of the shares in the Company’s register of members.

In September 2008 Lehman Brothers collapsed, and this prompted a wave of redemption requests from some of the Company’s participating shareholders, including SEB. Unfortunately, however, it ultimately transpired that the Company had been the subject of a significant fraud perpetrated by Magnus Peterson. The judge found that he had been fraudulently inflating the net asset value (“NAV”) of the Company by entering into interest rate swaps which he knew to be worthless. These swaps were concluded between the Company and a BVI company called Weavering Capital Fund Ltd (“WCF”) which he knew was not in a position to honour its obligations under the swaps. The swaps were simply used to give the impression of sustained growth when in reality the Company was suffering large losses from on-exchange trading in futures and options. The fraud was not discovered by the directors of the Company until March 2009, by which time redemption requests of over US$220m had been received. The Company was unable to pay these in full.

The mechanism by which participating shares could be redeemed was set out in articles 48-58 of the Company’s Articles of Association and was summarised in the Company’s Offering Memorandum (OM) (the latest version of which was dated 24 September 2008) in the following terms:

On 9 October 2008 SEB gave notice of redemption in accordance with these provisions for all the shares it held as nominee for Catella. On 28 October 2008 it gave the same notice in relation to the shares it held as nominee for HQ Solid. These redemption requests were processed on the 1 December 2008 Redemption Day, being the next Redemption Day following the requisite 30-day notice period, and were calculated in accordance with the Company’s published NAV per share. This NAV was markedly inflated as a result of Magnus Peterson’s fraud. Redemption payments were expected, in accordance with the OM, “generally” to be made within 30 calendar days of the 1 December Redemption Day.

The Company received other redemption requests from other participating shareholders in October 2008, totalling US$138,361,002.62 at the published NAV, which fell to be processed on the December 2008 Redemption Day (the “December Redeemers”). Additional redemption requests were also received by the Company in November and December 2008 which fell to be processed on the January and February 2009 Redemption Days (the “January Redeemers” and the “February Redeemers”). The redemption obligations incurred by the Company on these latter Redemption Days were US$54.7m in January and US$30m in February. On 17 December 2008 Magnus Peterson sent an email to the Company’s Administrator, a company called PNC Global Investment Servicing (Europe) Ltd (“PNC”), based in Dublin, in the following terms:

The email attached a spreadsheet which identified “SEB Merchant Banking as nominee for Catella Stiftelsefond” as one of the Swedish redeemers due to be paid. Accordingly, on or about 19 December 2008, SEB received payment of US$1,096,903.58 in respect of Catella’s shareholding which it duly credited to Catella’s cash account with itself. Five other Swedish redeemers received full payment on that date but no other December Redeemers received their redemption payments from the Company in December. As the end of December approached, recognising that the Company did not have enough cash to pay the other December Redeemers, WCUK took legal advice on its cash-flow position. On the same day, 29 December 2008, Magnus Peterson proposed that a letter be sent to the unpaid December Redeemers stating that a decision had been made to pay them 25% of their redemption payment, and envisaging that the remaining moneys would be paid by the end of January. The judge found that Magnus Peterson must have known by this time that the Company would never be in a position to pay all the December Redeemers, let alone pay the January 2009 redemption debt which was to fall due on 2 January. The proper course would therefore have been to suspend the redemption payments or liquidate the Company.

In the end the proposed letter was not sent until 7 January, by which time SEB had already received, on 2 January, another redemption payment of US$1,780,214.49. This represented 25% of the redemption proceeds due to HQ Solid. The Company made further partial payments to some other December Redeemers (not including SEB) across January, and by the end of the month most but not all of the December Redeemers had been paid the outstanding sums due to them. On 11 February 2009 (by which time the February 2009 redemption debt was also due) SEB received a further payment of US$5,340,643.47 constituting the remaining 75% of the redemption proceeds due to it as nominee for HQ Solid. By this time, however, the Company had in excess of US$134m in outstanding redemption obligations, being the balance of the December redemption debt, around US$50m being owed to the three largest of the December Redeemers, and the entirety of the January and February redemption debt.

In March 2009 the directors eventually learned of the fictitious nature of the swaps and the impact this had on the solvency of the Company. They resolved to suspend the determination of the NAV and the issue and redemption of shares with immediate effect. Shareholders were informed of this by letter on 11 March 2009 and on 19 March 2009 the Company was duly placed into liquidation. Subsequently, in January 2015, Magnus Peterson was sentenced to a total of 13 years’ imprisonment at Southwark Crown Court for offences including making a false instrument, furnishing false information relating to accounts and carrying on business with intent to defraud creditors.

In August 2014 the liquidators issued proceedings against SEB seeking a declaration that the three redemption payments described above were invalid as preferences under section 145(1) of the Companies Law (2013 Revision), and an order that the US$8,217,761.54 should be repaid plus interest. Section 145(1) reads as follows:

At the time of the trial before Clifford J in October 2015 there were six main issues before the judge. The first issue was to find the controlling mind of the Company. Having considered all the evidence before him over a five-day hearing he was satisfied that the two directors had delegated their powers to Magnus Peterson so that he acted as de facto director. At paras 61 and 62 of his judgment the judge concluded as follows:

On 5 January 2016 Clifford J declared that US$8,217,761.54 in payments to SEB were invalid preferences which SEB was ordered to repay with interest to the liquidators with costs. SEB then appealed that decision to the Court of Appeal.

The other issues before the judge were maintained with some alteration in the Court of Appeal. We propose to identify those issues which were pursued by the parties at the hearing before the Board.

The issues before us were referred to as the “Fraud Point”, the “30-Day Point”, the “Future Debts Point”, the “Intention to Prefer Point”, and the “Repayment Issues”. There was also a further issue, the “Amendment Point”, not advanced before Clifford J. In summary the conclusions of the courts below on these issues were as follows:

The Fraud Point: In order to establish that the Company was not insolvent at the time when the payments in issue were made, and that the payments therefore fell outside the scope of section 145(1) of the Companies Law (set out at para 12 above), SEB argued that the NAVs published by the Company were not binding as they had been inflated as a result of Magnus Peterson’s fraud, with the result that no redemptions had ever taken place in accordance with the Company’s Articles of Association, and therefore that no redeemers had ever become creditors of the Company within the meaning of section 145. In addition, SEB argued that the Company was only liable to the extent of the real (lower) NAVs. These arguments were rejected by the judge and the Court of Appeal, who held that the published NAVs were binding.

The 30 Day Point: In support of its contention that the Company was not insolvent at the time of the first redemption payment, SEB argued that on a true construction of the Articles of Association the Company had a grace period of 30 days from the 1 December redemption day to make redemption payments. Accordingly, section 145(1) was not satisfied in relation to the first redemption payment because at that date (19 December 2008) the Company was not yet liable to make payments to all the December Redeemers and was therefore solvent on a cash-flow basis. This was rejected by the judge and the Court of Appeal, who held that the Company became liable to pay the December Redeemers on the 1 December Redemption Day.

The Future Debts Point: This argument, also directed towards the existence of insolvency at the time of the first redemption payment, was advanced by the liquidators on the hypothesis that the judge had accepted SEB’s argument in relation to the 30 Day Point ie that the December redemption debt only became due and payable 30 days after the 1 December Redemption Day. If the judge were to have reached this conclusion, then the liquidators submitted that the future debts falling due at that time would in any event still be relevant to the cash-flow test of insolvency at the time of the first redemption payment. The court only needed to be satisfied that, as at 19 December 2008, the Company would not have been able to pay all of the 1 December Redemption Day payments on 31 December 2008. The judge and the Court of Appeal accepted this argument, finding that the Cayman Islands’ cash-flow test of insolvency permitted the consideration of debts which would become payable in the reasonably near future. It was possible to infer that no finance was available to the Company to fund the December redemption payments.

The Intention to Prefer Point: In relation to the requirement under section 145(1) of an intention to prefer, the liquidators submitted that Magnus Peterson’s email of 17 December relating to the Swedish redeemers was evidence of a specific intention to prefer SEB ahead of the other redeemers in relation to the first redemption payment. In addition, in relation to the second and third redemption payments, the liquidators argued that there was a specific intention to prefer SEB over those other December Redeemers and the January and February Redeemers who were not paid. Alternatively, the liquidators argued that it was possible to infer the necessary intention to prefer from the fact that, at the time of the payments, Magnus Peterson knew the Company was unable to pay its debts in full. In relation to this issue, both the judge and the Court of Appeal considered that there was a specific intention to prefer SEB (albeit for slightly different reasons) and therefore thought it was unnecessary to decide whether the requisite intention could be inferred only from the fact of payment in knowledge of insolvency (Martin JA, para 57). The judge considered that there was a specific intention, in relation to all of the redemption payments, to prefer SEB as a member of a particular class of creditors ie the Swedish redeemers, who Peterson claimed were intending to reinvest. Before the Court of Appeal SEB did not seek to challenge the judge’s conclusion in relation to the first redemption payment, but it disputed that such an intention could be in any way continuing in relation to the second and third payments. The Court of Appeal agreed, but the court nevertheless held - by virtue of the Company’s policy of paying the December Redeemers 25% of the redemption payment initially and then the balance at a later date, and the fact that this policy had been adopted at a time when Magnus Peterson knew the Company had no prospect of paying the January or February Redeemers - that these payments were nevertheless made with the specific intention to prefer SEB within the meaning of section 145(1).

Repayment Issues : In light of the conclusions set out above the judge (and, in due course, the Court of Appeal) held that the three redemption payments to SEB were preferences within the meaning of section 145(1). SEB had argued, however, that in this eventuality it was entitled to rely on the common law defences to a claim for restitution on the basis of unjust enrichment, and that it was specifically entitled to rely on the absence of any actual enrichment on its part, and on the fact that it had changed position by remitting the payments to HQ Solid and Catella. SEB also argued that the liquidators’ claim was founded on illegality and contrary to public policy in that it sought to benefit not the Company, but the unpaid redeeming shareholders. These submissions were rejected by both the judge and the Court of Appeal, who held that the common law defences were not available to a statutory claim under section 145 and that, in any event, they were not made out on the facts. The claim was also not barred as a result of illegality or public policy.

Accordingly, the first three points all concern the question whether the payments were made at a time when the Company was insolvent, and were therefore capable of falling within the ambit of section 145(1) of the Companies Law, set out at para 12 above. The first issue concerns all the payments made to SEB. The second issue, and the third issue contingent upon it, only relate to the first payment, which SEB received as the shareholder holding for Catella. The second and third payments were for HQ Solid. The fourth and fifth issues concern the question whether the payments were made with the intention to prefer SEB, and therefore met another of the requirements of section 145(1). The sixth issue concerns the consequences of the application of section 145(1). It is appropriate to deal with these issues seriatim, although there is a measure of overlap between them.

The first issue was summarised at para 16(a) above. On behalf of SEB Mr David Chivers QC submits that Clifford J and the Court of Appeal were wrong to reject SEB’s submission that the published NAVs, which had been inflated as a result of Magnus Peterson’s fraud, were not valuations at all within the meaning of the articles, alternatively were only valuations to the extent of what would have been the true valuations.

The Court of Appeal rejected SEB’s submission, largely on the basis of the opinion of this Board in Fairfield Sentry Ltd v Migani [2014] UKPC 9; [2014] 1 CLC 611. The Court of Appeal considered that it is not permissible to reopen a NAV retrospectively on the ground of fraud, whether or not the company was complicit in it. The practical reasons underlying the decision in Fairfield Sentry were essential to the operation of the company and applied equally in a situation in which the NAV was affected by fraud. Martin JA, with whom the other members of the Court of Appeal agreed, considered that the redemption of shares, once acquired, is an incident of an existing contract and brings it to an end according to its terms. It is not possible to regard it as giving rise to a new contract capable of being vitiated. There was, moreover, no difference between an internal and an external fraud in terms of the binding nature of a NAV. The whole scheme of the Company’s articles required its business to be conducted on the basis that the NAV is binding, whether it is accurate or not (at paras 29-30).

Mr Chivers submits that the fraudulent valuation of assets which occurred in this case was internal to the Company and that Fairfield Sentry is therefore distinguishable. He relies on the maxim that fraud unravels all. He submits that, as a result, the redeeming shareholders never became creditors of the Company in the sums paid to them. He accepts that, if this is correct, SEB was not entitled to receive any of the redemption payments but submits that while those payments might have been reclaimed as money had and received such a claim would now be time-barred. Moreover, he submits that SEB could never be the subject of a preference claim, nor could SEB be preferred over other persons who were not entitled to the money.

Under the contract between the Company and its members contained in the articles the directors were obliged to determine the NAV in accordance with articles 30-37. In particular, in calculating the NAV they were obliged to apply “such generally accepted accounting principles as they may determine” (article 32) and the Company’s assets were to “be valued in accordance with such policies as the Directors may determine” (article 34). Any valuations made pursuant to the articles are stated to be “binding on all persons” (article 34). However, Mr Chivers submits, first, that as a matter of public policy a fraudulent determination would not bind redeeming shareholders. Here he relies on HIH Casualty and General Insurance Ltd v Chase Manhattan Bank [2003] UKHL 6, [2003] 1 All ER (Comm) 349 per Lord Bingham of Cornhill at para 15; per Lord Hoffmann at para 68; Jones v Sherwood Computer Services Plc [1992] 1 WLR 277 per Dillon LJ at p 284. Secondly, he submits that there is nothing in the articles which purports to permit the directors to carry out a fraudulent determination of the NAV. On the contrary, he says, articles 32 and 34 make clear that any such assessment must be based on an honest assessment of the value of the assets.

In response Mr David Lord QC for the liquidators submits that this is essentially the same argument which was rejected by this Board in Fairfield Sentry as “an impossible construction” (at paras 22-23). He submits that the whole thrust of the fund’s argument in that case was that the NAV had not been correctly determined and that the reasoning of the Privy Council applies whether or not it had been or whether or not the fund itself knew it had been.

In Fairfield Sentry, the fund invested money with Bernard L Madoff Investment Securities LLC (“BLMIS”) which was, in fact, operating a Ponzi scheme and reporting fictitious returns to investors. Investors participated indirectly in BLMIS by subscribing for shares in the fund and redeeming them in accordance with the fund’s articles at a price dependent on the fund’s NAV. The articles provided that any certificate as to NAV “given in good faith by or on behalf of the directors shall be binding on all parties”. In December 2008 the directors of the fund suspended determination of the fund’s NAV, thereby terminating the redemption of shares. The liquidators of the fund sought to recover payments made to members who had redeemed shares prior to December 2008, on the ground that the payments had been made on the mistaken basis that the assets were as stated by BLMIS. They argued that the NAV would not be “determined by the directors” unless and until it was correctly determined using the true information as to the assets of the fund ascertained in the light of information which subsequently became available relating to Madoff’s fraud. This Board held that under the contract contained in the fund’s articles, the NAV was intended to be definitively ascertained at the time of the transactions and could not be varied subsequently with retrospective effect. The contrary view was untenable because on that approach the price would not be definitively ascertained for an indefinite period after the transaction had ostensibly been completed. As a result, payments made, albeit under a mistake, had discharged a contractual debt and could not be recovered on grounds of unjust enrichment, it not being suggested that the mistake was such as to avoid the contract.

The Board considers that Fairfield Sentry is distinguishable. It is correct, as Mr Lord points out, that in Fairfield Sentry there was no consideration of the operation of the fraud. In that case, however, the redemption liabilities were determined by the directors in good faith, as the articles required. The fraud which operated on the assessment of the NAV was external to the fund. Fairfield Sentry does not address the question whether a NAV would be considered to have been determined in accordance with the articles if the directors themselves had fraudulently inflated the value of the assets. By contrast, in the present case the fraud was that of Magnus Peterson and cannot be considered external to the Company. At first instance, Clifford J. found that the two de jure directors of the Company had delegated authority to Magnus Peterson:

The liquidators also point out, quite correctly, that it was PNC, and not Magnus Peterson, which was the relevant agent for the purpose of calculating the NAV and that there is no suggestion that PNC was complicit in or aware of Magnus Peterson’s fraud. PNC acted pursuant to the Administration and Accountancy Services Agreement dated 30 July 2003. The services provided by PNC included arranging for the computation of the NAV, controlling and authorising all disbursements, maintaining the register of shareholders, preparing and forwarding documents to shareholders and notifying the Adviser, the Custodian and the accounting agent of all share activity. However, PNC’s role was found by Clifford J (at paras 54-62) to be administrative only. PNC had been appointed to administer the day to day operations and business of the Company. WCUK, on the other hand, as the Investment Adviser, was appointed “to manage the affairs of the Fund” (Investment Advisory Agreement, clause 2). PNC had no part in the decision making process in relation to the payment of redemptions. Clifford J found (at paras 61-62) that the overwhelming weight of the evidence was to the effect that Magnus Peterson directly, and through his company WCUK, managed and controlled the Company for all purposes relevant to these proceedings. He controlled the investments and he made the material decisions about redemptions. He was the controlling mind in the payment of the relevant redemptions.

At this point it is necessary to say something about a submission on behalf of the liquidators, on the basis of In re Hampshire Land Co [1896] 2 Ch 743, that Magnus Peterson’s knowledge of the fraud would not be imputed to the company that he was defrauding. It is also submitted on the basis of Bilta (UK) Ltd v Nazir (No 2) [2015] UKSC 23; [2016] AC 1 that Magnus Peterson’s knowledge can be attributed to the Company for the purpose of making the SEB redemption payments but not for other purposes. It seems to the Board, however, that it is not concerned here with attributing knowledge. What matters in this case is that valuations were prepared on a fraudulent basis by the person to whom the directors had given actual authority to carry out the valuations.

The Board agrees with SEB that the relevant question in this case is not whether Magnus Peterson’s fraud tainted the contractual determination of the published NAV. As Mr Lord accepts, it clearly did. The relevant question is, rather, whether the NAV was binding. In answering that question, it is necessary to have regard to the fact that the fraud originated from the person to whom the directors had delegated their powers and who was the controlling mind of the Company. The Company was therefore in breach of its duty under the contract to act in good faith. While the Board would accept that, in general, it is vitally important that valuations are definitively ascertained at the time of the transactions and are not liable to be varied subsequently with retrospective effect, even this must yield where, as here, the fraud is internal to the company which is seeking to rely on the contract. Fraud is something apart. The Board accepts Mr Chivers’s submission that nothing in the contract constituted by the articles purported to permit the directors of the Company to carry out a fraudulent determination of the NAV and that, even if it had, such a fraudulent determination could not bind redeeming shareholders. The Board considers that the dishonest valuation of assets was not made “pursuant to these articles” and therefore was not “binding on all persons” under article 34.

Order 12 rule 2 of the Companies Winding Up Rules 2008 is relied upon by SEB as express statutory recognition that a mis-stated NAV is not binding by reason of fraud. It applies only in a solvent winding up and permits rectification of the register where a company has redeemed shares at prices based on a mis-stated net asset value which is not binding upon the company and its members by reason of fraud or default. While this acknowledges that there are circumstances in which rectification may take place on such a ground, this does not assist SEB as the whole question here is whether the NAV is not binding by reason of fraud.

Sir Donnell draws attention to the fact that the extent of any fraud lying behind the statement of a NAV may vary enormously. He raises the question of the consequences where a NAV has been influenced by a small or minimal fraud on the part of the employee who had misstated some part of the company’s assets. It is, in the Board’s view, unnecessary to address this possibility which does not arise in the present case where, as he accepts, there was undoubtedly a massive fraud.

That, however, is not the end of the matter. The Board agrees with Sir Donnell that it would be necessary for a party who wished to have the NAV avoided to bring proceedings to do so. Those proceedings would be brought against the liquidators of the Company, but notice would also have to be given to those who would be affected by a decision that the NAV was voidable. However, such proceedings would not avail SEB. As Sir Donnell explains in his judgment, SEB has not been defrauded but has, in fact, received payments on behalf of its two clients substantially in excess of the entitlement had the NAV been accurate and honest. It would necessarily be a condition of any order setting aside the NAV on the application of SEB that SEB repay the sums it received under that NAV.

For this reason alone, the fraud point fails.

The second issue has been summarised at para 16(b) above. SEB appeals against the decision of the Court of Appeal to uphold the judge’s finding that the payments to the December 2008 redeemers fell due on the 1 December 2008 redemption day rather than 30 days later.

Mr Chivers relied on the terms of the articles of association as set out at para 6 above in this sentence:

The first observation to make is that the use of the word “generally” with regard to redemption payments is indicative of the description of the practice of the Company rather than of the definition of legal rights.

The frailty of the argument is further increased by article 36 of the Company’s articles which provides as follows:

It is common case that the redemption day for the first payment to SEB was 1 December 2008. So the articles clearly say that it was a liability of the Company from then and therefore SEB was a creditor to that extent from 1 December.

One also notes article 55 of the Company’s articles, which provides that the Company should remit redemption proceeds “within such period as the Directors shall determine”.

SEB nevertheless argues that the December redemption payments were not due by December 19 and therefore the Company was not insolvent at that time.

In support of that SEB relies on the judgment of Sir William James VC in In re European Life Assurance Society (1869) LR 9 Eq 122. In considering a petition to wind up that company the Vice-Chancellor at p 127 said the following:

This decision has been a leading authority for nearly 150 years, submits counsel. It has not escaped criticism. In BNY Corporate Trustee Services Ltd v Eurosail-UK 2007 -3BL plc [2013] UKSC 28; [2013] 1 WLR 1408 at para 28 Lord Walker of Gestingthorpe noted that it was an extempore judgment “with very little reasoning”. “He did not refer to any of the authorities that had been cited. It may be unfortunate that his judgment has come to be regarded as a leading case.”

Counsel sought to argue that, whatever the position in England, where in any event there had been statutory intervention twice over the years, it was still good law in the Cayman Islands. There are two clear answers to that. First of all the Court of Appeal in the Cayman Islands was entitled not to follow this judgment at first instance by the Vice-Chancellor. They set out their reasons clearly at paras 31-35 of the judgment of Martin JA in this case for not doing so.

Secondly, for our part we consider that reliance may be placed on the decision of the Board in Culross Global SPC Ltd v Strategic Turnaround Master Partnership Ltd. In his judgment on behalf of the Board Lord Mance identified the issue:

At para 9 of his judgment one finds the relevant definition of the redemption date.

We agree with the view taken by the Board in that case. Applying that approach to the circumstances of the present case, the redemption price became a liability of the Company on 1 December 2008. As the Company was then unable to pay all of those shareholders who had served notices of redemption taking effect on that date, it was unable to pay its debts for insolvency purposes at the time the selective payments were made to SEB and certain other Swedish institutions, as the judge at first instance found. The reference to payments made generally within 30 days is part of a “supplementary procedure”.

Further support for the liquidators’ case is to be found in a further judgment of Lord Mance in Pearson v Primeo Fund [2017] UKPC 19; [2017] BCC 552. This is a case dealing with redemption funds following the exposure of Bernard Madoff and his giant Ponzi scheme. At para 13 Lord Mance said the following:

In this case we conclude that the moneys due by the Company to the redeeming shareholders from 1 December were indeed debts which had fallen due. Proceedings could have been issued for that debt even if summary judgment might not immediately have been granted. We therefore find against SEB on that point.

This point is summarised at para 16(c) above. It was expressly contingent on SEB succeeding on the 30 day point. As it has not succeeded on that point it is not necessary to deal with it further.

This point is summarised at para 16(d) above. SEB contends that the payments which it received as nominee for HQ Solid were not made with a view to preferring that company. The first of those payments was made on 2 January 2009 in the sum of $1,780,214.49. The second was made on 11 February 2009 in the sum of $5,340,643.47.

SEB acknowledges that there was a clear intention to make the full payment to SEB as nominee for Catella on 19 December 2008 with a view to preferring that redeemer.

Section 145(1) of the Companies Law, set out in full at para 12 above, provides inter alia that:

The matter was dealt with very fully by the Court of Appeal from para 41 on. At para 44 the court records the following passage from Clifford J at first instance at para 179 of his judgment.

We observe that Magnus Peterson is a convicted fraudster so the email to PNC of 17 December may just have been intended to give a plausible but entirely untruthful reason for PNC paying those creditors ahead of other creditors. Indeed that is the likely inference, partly for the reason mentioned by the Court of Appeal ie there is no evidence that SEB had any such interest in reinvesting. This reference to Swedish investors reinvesting was also to be found in an email from Peterson to PNC of 20 January 2009.

The Court of Appeal concluded in the light of two decisions of the English Court of Appeal, In re Cutts, Ex p Bognor Mutual Building Society v Trustee of T W Cutts [1956] 1 WLR 728 and In re M Kushler Ltd [1943] Ch 248, 252 that the necessary intention could be inferred by the court in accordance with general principles of inference from the available evidence.

SEB sought to rely on the following passage from the judgment of Lord Greene MR at p 252 of Kushler above:

In its written argument SEB says that commercial impropriety is at least required.

At para 55 the Court of Appeal concluded that there was insufficient material to enable the judge to infer a dominant intention throughout to prefer SEB on the grounds that it intended to invest in the Swedish fund. Martin JA, however, proceeded as follows:

It appears that, as well as the evidence from the emails and the board minutes and the witness on behalf of the liquidators, Peterson was himself a former employee of SEB, a further possible reason for him to prefer SEB. The Board concludes that the Court of Appeal was entitled to reach the conclusion that it did and that SEB fails on this point also.

This point was summarised in para 16(e) above. It only arose if the Board was with SEB on the previous issue, as to the intention behind the second and third payments. As that is not the case it is neither necessary nor appropriate to address this.

The repayment issues were summarised in para 16(f) above. To recap, SEB argued that, in the event that the payments were held to be voidable under section 145(1) of the Companies Law, the liquidators’ claim to restitution arose under the common law and was based on unjust enrichment. SEB was, it argued, therefore entitled to defend the claim on the basis that (1) it had not been enriched, since it had received the payments as the nominee of Catella and HQ Solid, and (2) it had changed its position on the faith of the payments, by remitting them to those funds, and would suffer an unjust detriment if it were now required to repay them. The judge and the Court of Appeal rejected those contentions, holding that the liquidators’ right to repayment arose under statute and not at common law, and that in any event the defence of change of position was not made out on the facts. SEB also argued that the liquidators’ claim was in any event founded on illegality and was contrary to public policy. Those contentions also were rejected by the judge and the Court of Appeal.

It may be helpful to begin by repeating the terms of section 145(1):

So understood, section 145 is consistent with the treatment of voidable preferences (or fraudulent preferences, as they have also been described) under the common law and more recently statute law. It was established by the time of Lord Mansfield that transactions by which a bankrupt discharges his debts to one or more creditors with the intention of preferring those creditors over other creditors were subject to avoidance in appropriate circumstances under the common law. That is because such transactions defeat the intention of Parliament, established in relation to bankrupt individuals since the Statute of Bankrupts Act 1542 (34 & 35 Hen 8, c 4), and subsequently applied also in relation to companies, that there should be a collective procedure for the collection and realisation of an insolvent debtor’s estate, so as to ensure its distribution in accordance with a statutory scheme providing for an equal or “pari passu” distribution among the ordinary unsecured creditors. As Lord Mansfield said in relation to a fraudulent preference in Alderson v Temple at p 2240, “it is defeating the equality that is introduced by the Statutes of Bankruptcy”.

Two other features of voidable preferences are worth noting at this stage. First, since the law governing preferences necessarily applies retrospectively, a transaction which is held to have been a voidable preference will have been valid and effective at the time when it took place. Secondly, the preference need not involve any moral blame on the part of the recipient (In re Patrick and Lyon Ltd [1933] Ch 786, 790).

Section 145 invalidates any conveyance or payment which falls within its scope, but is silent as to the consequences of that invalidation. It can be contrasted with a provision such as section 239 of the United Kingdom Insolvency Act 1986, which also deals with preferences, but requires the court, on an application by the relevant office-holder, to “make such order as it thinks fit for restoring the position to what it would have been if the company had not given that preference”. Section 241 then lists seven types of order which may be made, without prejudice to the generality of section 239. By contrast, since section 145(1) of the Cayman Companies Law is silent as to the consequences of the invalidation, those consequences must therefore be regulated by the general law which applies in the situation resulting from the avoidance of the transfer or payment in question: that is to say, by any other statutory provisions which may be applicable, or in their absence by the common law. Depending on the circumstances, there may be a variety of remedies available, personal or proprietary, at common law or in equity. They will usually include a right to proprietary or personal restitution of the property or money transferred, subject to any defences which may be available.

That conclusion is supported by a substantial body of authority. As we have explained, before preferences were rendered voidable under statutory provisions such as section 145, they were voidable at common law, as being incompatible with the scheme of distribution to which creditors were entitled under statute. An action then lay to recover the property or money which had been transferred. The point is illustrated by the case of Alderson v Temple, where the preference took the form of the delivery of promissory notes to the defendant. On the avoidance of the transaction, the notes were held to be recoverable in an action of trover. In Marks v Feldman, where the goods which were the subject of the preference had been converted into money, the appropriate remedy was held to be an action for money had and received. Kelly CB explained, at p 279, that the basis of the claim for recovery was that the preferential payment offended against the spirit, although not the letter, of the statutory bankruptcy code.

That approach continued to be followed when statutory provision was first made for the avoidance of the transaction, but no provision was made for the granting of a remedy. The point is illustrated by the case of Rousou’s Trustee v Rousou [1955] 3 All ER 486, where Danckwerts J had to decide whether the right to recover money which had been paid under a settlement which was either void under section 42(1) of the Bankruptcy Act 1914, or voidable as a fraudulent conveyance under section 172 of the Law of Property Act 1925, was a statutory right arising by implication: an analogous question to that raised in the present case. Section 42(1) of the 1914 Act provided:

Danckwerts J held that, on the avoidance of the settlement under either of these provisions, a right to restitution arose under the common law, based on the principle established in Moses v Macferlan (1760) 2 Burr 1005. He said at p 491:

Another aspect of Danckwerts J’s decision in that case, concerned with a question in private international law as to the law governing the recovery of the payment, was questioned by Sir Nicolas Browne-Wilkinson VC in In re Jogia [1988] 1 WLR 484, 495, but no doubt was expressed as to the correctness of the decision on the point now in issue.

The same conclusion as in Rousou’s Trustee was reached by Oliver J in In re J Leslie Engineers Co Ltd [1976] 1 WLR 292. That case concerned section 227 of the Companies Act 1948, which provided:

That decision was followed by the Court of Appeal in Hollicourt (Contracts) Ltd v Bank of Ireland [2000] EWCA Civ 263; [2001] Ch 555, which concerned section 127 of the Insolvency Act 1986. That section is in very similar terms to section 227 of the 1948 Act, and provides:

In In re Ahmed (A Debtor), Ahmed v Ingram [2018] EWCA Civ 519; [2018] BPIR 535, the Court of Appeal considered an argument that a statutory right of action was created by section 284(1) of the Insolvency Act 1986, which provides:

It is unfortunate that none of the authorities we have mentioned appears to have been cited to the Grand Court or the Court of Appeal in the present case. Accepting the liquidators’ submission that the cause of action arose under section 145 itself, and their further submission that a defence of change of position was therefore not available to SEB, the Grand Court referred to Professor Sir Roy Goode’s article, “The Avoidance of Transactions in Insolvency Proceedings and Restitutionary Defences”, in Mapping the Law: Essays in Memory of Peter Birks (ed Burrows and Rodger) (2006), p 299. The Board has also been assisted by Sir Roy’s Principles of Corporate Insolvency Law, now in its 5th ed (2018), edited by Kristin Van Zwieten. These contain valuable discussions of issues arising in connection with voidable preferences, but we cannot discern in them any support for the proposition that a provision in the form of section 145 creates a statutory right of recovery. On the contrary, both in his article and in his book Sir Roy carefully distinguishes between statutory provisions which make express provision for orders reversing the effect of a disposition, such as section 239 of the Insolvency Act 1986 (quoted at para 63 above), and those which contain no such provision, such as section 127 (quoted at para 69). In the article, he states at p 304 that in cases falling within provisions of the latter kind, “the consequences of invalidity are not spelled out and are left to rules of the common law”. That point is repeated at p 310 (“section 127 … says nothing about the consequences of invalidity, leaving these to be determined by the common law”), at p 311 (“These [viz remedies consequent upon avoidance under section 127] are left to the common law and typically take the form … where money was paid over and has been spent, [of] an order for repayment in a claim for money had and received”), and again at p 318 (“section 127 says nothing about the consequences of the invalidity of a disposition … These are left to the common law.”).

Upholding the conclusion of the Grand Court, the Court of Appeal also cited a passage in the judgment of the Board, delivered by Lord Browne-Wilkinson, in Lewis v Hyde [1998] 1 WLR 94. That case was concerned with section 309 of the New Zealand Companies Act 1955, which provided:

Read in isolation, that dictum appears to support the Court of Appeal’s conclusion. However, we do not believe that it can bear the weight which the Court of Appeal placed upon it. The case was not concerned with the question whether the statutory provision there in issue created a statutory cause of action. Lord Browne-Wilkinson was not referred to any of the authorities which we have cited. His observation as to the “effect” of the section appears in a paragraph concerned with an entirely different question. Read in its context, it should not in the Board’s opinion be treated as the expression of a considered view that the section created a statutory cause of action. When Lord Browne-Wilkinson referred to the “effect” of the section, he may have been referring to the ultimate consequence of the section’s application, that is to say, the common law obligation of the preferred creditor, following the avoidance of the payment, to repay the money.

The Board concludes, therefore, that section 145 does not create a statutory right to recover the property or payment which were the subject of the preference, but merely renders the relevant transfer or payment voidable. In the present case, the effect of the order made under section 145 is therefore to avoid the Company’s payment to SEB of the amount due to it on the redemption of its shares.

Proceeding then on the basis that the consequences of the avoidance of a fraudulent preference under section 145 depend on the general law, it is apparent that they will vary according to the circumstances. A conveyance or payment which is voidable has full effect in law and equity until it is avoided. Dealings by the recipient with the property or money in question during the intervening period are legally effective, and can therefore limit the consequences of avoidance. The effect of statutory provisions (eg as to registration of title) may also need to be considered. In principle, however, where property has been transferred and remains in the hands of the transferee, the consequence of the avoidance is to deprive the transferee of his title. The liquidator therefore has a claim to recover the property on the basis of the company’s title. If the property has passed into the hands of third parties, the liquidator may be able to trace into their hands at common law or in equity. As was mentioned earlier, that proprietorial approach was followed, in the context of personal bankruptcy, in Alderson v Temple.

In the present case, however, no claim is advanced on a proprietary basis. It is not suggested that the avoidance of the payment has affected the title to any property in SEB’s possession, and no attempt has been made to trace the money into the hands of the funds to which SEB transferred it, or their fundholders. Instead, the liquidators have based their claim to repayment on a statutory entitlement impliedly created by section 145. For the reasons we have explained, we do not accept that a claim lies on that basis. It is however accepted on behalf of SEB that, subject to its argument that it was not enriched by the payment, and its defence of change of position, the liquidators are in principle entitled to restitution of a payment which is avoided under section 145 at common law on the ground of unjust enrichment.

As counsel submitted, the law has long recognised that a claim to restitution can be brought in such circumstances by those responsible for the administration of the insolvent’s estate. Historically, the relevant form of action was indebitatus assumpsit, on a count for money had and received. In Moses v Macferlan Lord Mansfield CJ explained the basis of such claims in terms derived from equity and Roman law, stating at p 1009:

Consistently with that approach, in Marks v Feldman, as mentioned earlier, the appropriate remedy was held to be an action for money had and received. Kelly CB stated at p 281 that “if originally the fraudulent preference had consisted in the payment of a sum of money into the hands of the defendant, it can hardly be disputed, unless indeed we at once abrogate the law as to fraudulent preferences, that an action for money had and received could have been maintained against the favoured creditor”.

Following the abolition of the forms of action, the basis of such a claim was usually described as quasi-contract. Thus in Rousou’s Trustee v Rousou, in the passage cited at para 66 above, Danckwerts J explained that the right to re-transfer or repayment arose from the fact that, once the transaction had been set aside, the property or money was wrongfully in the hands of the person who had it and therefore became re-transferable or repayable to the successful party. As authority for this “quasi-contractual right”, as he described it, he cited Moses v Macferlan. The claim in In re Leslie Engineers Co Ltd was also for money had and received: see at p 299. In more recent times, following the lead given by Lord Wright in Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd [1932] AC 32, the basis of this type of claim to restitution has been described as unjust enrichment; but the legal substance of the claim remains as it was.

It may not be immediately obvious how the claim for restitution in a case such as the present fits the academic model of unjust enrichment which was adopted by Lord Steyn in Banque Financière de la Cité v Parc (Battersea) Ltd [1999] 1 AC 221, 227: a model which may not, however, readily accommodate all the situations where personal claims lie for restitution, and should not become a Procrustean bed. The reason why the law requires the repayment of the money to the liquidators (subject to possible defences) is the injustice of SEB’s benefiting from the avoided payment at the expense of the Company’s creditors. In Lord Mansfield’s words, it is money which ought not in justice to be kept. The injustice lies not between the Company and SEB, since the Company received in full the quid pro quo for which the payment was made, namely the redemption of SEB’s shares, and that is not undone by the avoidance of the payment. Nor would restitution benefit the Company, since in the winding up it has no beneficial interest in its assets, which the office holder holds on trust for the creditors in accordance with the statutory insolvency scheme. But restitution of the money to the assets available for distribution by the liquidators among the Company’s creditors is necessary to complete the process of undoing the unfair advantage which SEB obtained at their expense. The common law therefore imposes an obligation on SEB to repay the money, with a corresponding right to rank as a creditor for the amount owed to it, but only upon repayment: see Cherry v Boultbee (1839) 4 My & Cr 442.

Counsel for SEB resisted this conclusion on the ground that SEB had not been enriched by the payment. Alternatively, they argued that SEB had a defence to the claim to restitution on the ground of change of position. It is necessary to consider each of these contentions in turn.

On a count for money had and received, it was necessary to establish that the defendant had received the money in question. But there can be situations in which the receipt of the money is not in itself sufficient to establish that the law requires the defendant to repay it in the event that a mistake or a failure of consideration (or some other relevant factor) becomes apparent. The defendant may, for example, have received the money merely as an agent, and remitted it to his principal in accordance with his instructions. The term “enrichment” has been employed in recent times in order to refine the analysis.

SEB’s argument that it was not enriched is based upon the fact that, as the judge found, it held the redeemable shares as a bare trustee for two Swedish mutual funds, namely Catella and HQ Solid. The shares were issued to it expressly as the nominee of those funds, but it was (and was entered in the Company’s register of members as) the legal owner of the shares. As a trustee, SEB did not have any beneficial interest in the shares. Accordingly, it is argued, SEB received the proceeds of the redemption of the shares as a bare trustee, and had no beneficial interest in them. In these circumstances, it is argued, SEB was not enriched by the receipt of the money.

The principal authority cited in support of this contention is a dictum of Millett LJ in Portman Building Society v Hamlyn Taylor Neck (a Firm) [1998] 4 All ER 202. The defendants in that case were a firm of solicitors who had acted on behalf of the plaintiff building society. In that capacity, they received a mortgage advance from the society and paid it initially into their client account, and then, in accordance with their instructions, to the vendor of the property being purchased by the building society’s customer. The building society subsequently brought a claim against the solicitors for restitution of the mortgage advance, on the ground that it had been paid to the solicitors under a mistake of fact. The claim was struck out, and an appeal against that decision was dismissed. As Millett LJ explained at p 208, the solicitors were the agents of the building society. They received the payment from their principal, held it to the order of their principal, applied it in accordance with their principal’s instructions, and thereby obtained a good discharge.

SEB rely on Millett LJ’s statement at p 206:

There are however some fundamental differences between that case and the present. As has been explained, that was a case in which the defendant received the payment as an agent, and dealt with it in accordance with its principal’s instructions, which created a trust of the kind recognised in Quistclose Investments Ltd v Rolls Razor Ltd [1970] AC 567. Even if the defendant had been the agent of a third party to which it had remitted the money, rather than the agent of the plaintiff, any claim in restitution would have lain against its principal. In that regard, Millett LJ stated at p 207:

That point is also illustrated by the case of Challinor v Juliet Bellis & Co [2015] EWCA Civ 59, where the Court of Appeal considered that no claim in restitution lay against solicitors who had received a payment into their client trust account (again, a Quistclose type of trust) as an immediate loan to their client, and had then disbursed the money to or for the account of their client. They had received the payment as agents, not principals, and had not been enriched by its receipt. Agents may or may not act as trustees of moneys held for their principals, but they are not in either event enriched by payments made to them for the account of their principals.

SEB, on the other hand, was not the agent of Catella or HQ Solid, but was itself a registered shareholder in the Company. As the registered holder of redeemable shares, SEB, not Catella or HQ Solid, was the person entitled to be paid the proceeds of the redemption of the shares, and it was the person to whom, in law, the payment was made. It dealt with the Company as principal. The fact that the Company may be taken to have known that SEB was a nominee for the funds is neither here nor there. The Company was entitled and obliged, under the articles of association which formed its contract with SEB, to deal with SEB as the legal owner of its shares, and had no dealings with the underlying funds (which in any event lacked legal personality, as explained below) or with the investors in them.

The question whether the recipient of a voidable preference is enriched by the receipt does not therefore depend upon whether the recipient receives the money as a trustee for someone else, or upon the terms of the trust. Although a bare trustee has a functional resemblance to an agent, there are nevertheless material legal differences between a trustee and an agent, which apply to bare trustees just as to other trustees. A trustee who is not also acting as an agent acts as a principal in transactions with third parties, including the Company in the present case. It was, for example, SEB which was owed the redemption moneys. A trustee who is not acting as an agent is enriched at common law by the payments which he receives, since the common law ignores the equitable interest of the beneficiaries. He can accordingly be an appropriate defendant in an action for the restitution of money paid to him as trustee.

These points are illustrated by the case of In re Morant [1924] 1 Ch 79, in which a trustee in bankruptcy sought to recover a preference avoided under section 44(1) of the Bankruptcy Act 1914, which provided:

There are practical reasons, as well as the reasons of legal principle which we have explained, why a trustee who acts as principal should be regarded as being enriched by a payment which he receives in that capacity. The alternative is to join the beneficiaries as defendants to the claim to repayment; but that may often be inconvenient and expensive, if the beneficiaries are numerous or difficult to trace. It may even be impossible, if for example the beneficiaries cannot be ascertained (as, for example, where the trust confers wide discretionary powers, or where the beneficiaries are ascertainable only on the occurrence of some future event).

The conclusion that a trustee is enriched by the receipt of money in that capacity is consistent with the decision of the Divisional Court in King v Stewart (1892) 66 LT 339, where money had been paid to the defendant trustee in the mistaken belief that it was due under a contract between him and the plaintiff. The trustee was held to be the appropriate defendant, although the money had been remitted by him to the cestui que trust. A more recent illustration is In re Thirty-Eight Building Ltd [1999] BCC 260, a case brought under section 239 of the Insolvency Act 1986, where the trustees of a pension fund, rather than the beneficiaries of the scheme, were held to be the appropriate defendants.

There is also an analogy with the decision of the High Court of New Zealand in Springfield Acres Ltd (In Liquidation) v Abacus (Hong Kong) Ltd [1994] 3 NZLR 502, where it was argued that a trustee could not be liable for the knowing receipt of funds transferred in breach of a fiduciary duty, since it did not receive the funds for its own benefit. The contention was rejected. Henry J observed at p 511 that the trustee “was not merely an agent … and the fact that it had fiduciary obligations itself to others in accounting for its use of the funds is beside the point”. Another illustration is the case of Port of Brisbane Corpn v ANZ Securities Ltd (No 2) [2001] QSC 466; [2002] QCA 158; [2003] 2 Qd R 661, in which the defendant received money as a bare trustee before remitting it in accordance with its instructions. An argument that it had not been enriched by the receipt of the money was advanced at first instance but rejected, and authorities concerned with the position of agents were distinguished. The argument was not renewed on appeal, where the focus was on the defence of change of position.

There remains the question whether a defence of change of position is available. The expression “change of position” has been employed in recent times to describe one type of situation in which it may be unjust, and therefore inappropriate in an action founded on essentially equitable considerations, to allow a claim to restitution on the basis of unjust enrichment. This idea can be traced back to Moses v Macferlan, where after explaining that the action was equitable in nature, Lord Mansfield CJ stated at p 1011:

Despite that broad statement, however, the defence of change of position has received only partial recognition in English law, as Lord Goff of Chieveley acknowledged in Lipkin Gorman v Karpnale Ltd [1991] 2 AC 548, 578. Lord Goff explained the general nature of the defence at p 580:

The counterpart of finding that it would be inequitable to require the defendant to repay the money which he received is that it would be inequitable for the plaintiff to pursue a claim for its recovery. That way of considering the matter has been adopted in a number of the authorities: see, for example, the judgment of the Board in Dextra Bank & Trust Co Ltd v Bank of Jamaica [2002] 1 All ER (Comm) 193, para 38.

The potential importance of a defence of change of position is particularly evident in relation to claims to recover voidable payments long after they were received. In such a situation, the recipient may in good faith have parted irretrievably with the money, without notice of the risk of avoidance. In such a case, it is possible that an order for restitution against the recipient would not in reality restore the position to what it was before the voidable payment was made, but would effectively leave the recipient worse off than if he had never received the money in the first place. Consistently with Lord Goff’s explanation of the defence of change of position, one would therefore expect that a trustee, like any other innocent recipient of money paid on a basis which has failed, might be excused from the obligation to repay it in such circumstances.

That general proposition gains support from the case of Challinor v Juliet Bellis & Co, where the Court of Appeal accepted that, if a claim in restitution had otherwise lain against an agent who had received moneys in trust and disbursed them in good faith to the beneficiary, a defence of change of position would have been available. The availability of the defence was also accepted in the Australian case of Port of Brisbane Corpn v ANZ Securities Ltd, where the trustee recipient was found to have acted in good faith and to its detriment (if it was required to make restitution) in paying out the money in accordance with its instructions.

The question which arises, therefore, is whether the defence is available to the recipient of a payment which constituted a voidable preference. In that regard, the primary argument of the liquidators is that section 145 creates a statutory entitlement to repayment which is unqualified, and that the common law relating to restitution is simply irrelevant. We have rejected that argument for the reasons explained earlier.

Their alternative argument is that a defence of change of position is inconsistent with the statutory aim of section 145, since it would subvert the fundamental principle of pari passu distribution of an insolvent company’s assets. In support of this argument, they cite the article by Professor Sir Roy Goode which was mentioned earlier (para 71 above). In that article, Sir Roy accepted at p 303 that claims under sections 238 (transactions at an undervalue), 239 (preferences) and 423 (transactions defrauding creditors) of the Insolvency Act 1986 are based on unjust enrichment, but argued (as he does also in his Principles of Corporate Insolvency Law (op cit), para 13-144), that to allow a defence of change of position would be inconsistent with the policy underlying those provisions, namely the restoration of value to the company for the benefit of its creditors.

As we have explained, preferences were struck down by the courts long before legislation was enacted to deal with them. The courts avoided preferences, exercising common law powers, on the basis that the preferential payment of one or more creditors would otherwise defeat the intention of Parliament in enacting a statutory scheme for the distribution of a bankrupt’s assets, under which his ordinary creditors received equal treatment. In doing so, the courts gave effect to a principle of public policy which also underlies what has become known as the anti-deprivation rule (that a person cannot effectively provide for his property to cease to form part of his estate upon his bankruptcy), and what has become known as the pari passu rule (that a party cannot effectively contract out of the distribution of his estate in accordance with insolvency legislation). The relevant principle was described by Lord Walker in Belmont Park Investments Pty Ltd v BNY Corporate Trustee Services Ltd (Revenue and Customs Comrs intervening) [2011] UKSC 38; [2012] 1 AC 383, para 121, as “a general principle of public policy which (in the traditional phrase) prevents a fraud on the insolvency statutes”. In other words, the common law supports insolvency legislation by refusing to give effect to arrangements which would defeat its operation.

Insolvency legislation itself now contains anti-avoidance provisions designed to prevent the circumvention of the statutory scheme of distribution of an insolvent individual’s or company’s estate among its creditors. Sections 238, 239 and 423 of the Insolvency Act 1986 are examples, which contain their own claw-back provisions, subject to provisions for the protection of third parties in particular situations which Parliament has specified. Section 145 of the Companies Law is another example of an anti-avoidance provision. As has been explained, however, it does not contain a claw-back provision, and therefore requires the liquidator to rely on a common law right to recover the property or money transferred to the preferred creditor, following the avoidance of the transfer or payment.

The question which therefore arises in this case is whether, in the context of a claim by a liquidator for the restitution of money paid to a preferred creditor following the avoidance of the payment under section 145, the common law gives priority to the operation of the statutory scheme of distribution over the detrimental impact which recovery may have upon the creditor against whom the claim is made.

The Board answers that question in the affirmative. As a matter of principle, the well-established principle of public policy which is applied by the courts so as to deny effect to arrangements in fraudem legis does not, of its nature, pick and choose according to the detriment which may be suffered by the defendant as a consequence of its operation. As a matter of practice, the court has not been referred to any judicial decision in which a change of position has been held to be available as a defence to a claim brought at common law for the recovery of a preference (our own researches have disclosed one dictum, in In re Morant at p 87, which does not appear to have been cited in any subsequent English case). It would be remarkable if such a defence were in principle available, without its having an established basis in centuries’ worth of case law on insolvency.

That view derives support, in the first place, from the decision of the Supreme Court in Wilson v First County Trust Ltd (No 2) [2003] UKHL 40; [2004] 1 AC 816. The case concerned a consumer credit agreement which had not been executed in accordance with the relevant legislation. The consequence, under the legislation, was that the creditor could not recover the money advanced. An argument that the creditor was entitled to recover the money at common law, on the ground of unjust enrichment, was rejected, since such a result would defeat the intention of Parliament. Lord Nicholls of Birkenhead stated at para 49:

As Lord Nicholls noted at para 50, that result was consistent with the approach adopted by the House of Lords in the earlier cases of Orakpo v Manson Investments Ltd [1978] AC 95 and Dimond v Lovell [2002] 1 AC 384, which arose in similar circumstances, and where claims based on unjust enrichment were likewise rejected. In the latter case, Lord Hoffmann observed at pp 397-398 that Parliament contemplated that a debtor might be enriched consequential upon non-enforcement of an agreement pursuant to the statutory provisions. It was not open to the court to say this consequence was unjust and should be reversed by a remedy at common law. In the present case also, given the absence of any trace of a defence of change of position in centuries’ worth of case law on the recovery of preferences, and the absence from section 145 itself of any provision for the protection of preferred creditors, we do not consider that it is open to the Board to conclude that the legislature did not contemplate that the effect of section 145 might be to enable preferential payments to be recovered from their recipients, notwithstanding detrimental consequences. Once that point is reached, the position is clear: the Board cannot override the intention of the legislature by providing a common law defence.

Further support for that conclusion can be derived from a number of Canadian authorities. They all display, in broad harmony with the views of the courts below, a deep-seated disinclination to allow a change of position defence to undermine recovery of voidable preferences for the benefit of the general creditors.

In Principal Group Ltd v Anderson (1994) 29 CBR (3d) 216; 164 AR 81; 52 ACWS (3d) 486, Cairns J rejected change of position as a defence to a voidable preference claim under section 95 of the Bankruptcy and Insolvency Act 1985. That provided that payments made within three months prior to bankruptcy with a view to preferring a creditor were to be “deemed fraudulent and void as against the trustee in the bankruptcy”. The judge held first that it would be contrary to public policy to permit that defence to succeed in relation to a payment motivated by fraud, even though the recipients were innocent. Secondly he held that:

On an appeal from that decision specifically concerning the availability of change of position as a defence, sub nom Ernst and Young Inc v Anderson (1997) 147 DLR (4th) 229, the Alberta Court of Appeal held that a change of position defence could not be maintained where a statute placed a public duty upon an office holder to make the relevant recoveries. In so holding, the court relied on the decision of the Board in Maritime Electric Co Ltd v General Dairies Ltd [1937] AC 610, that an equitable estoppel could not be pleaded as a bar to the performance of a statutory duty, and the decision of the Supreme Court of Canada in Kenora (Town) Hydro Electric Commission v Vacationland Dairy Cooperative Ltd [1994] 1 SCR 80, 110 DLR (4th) 449, holding that the same approach should be applied to a defence of change of position where it was incompatible with the performance of a statutory duty. The court said, at p 233:

In the Ernst and Young case, the Alberta Court of Appeal also identified at p 234 an additional reason why a defence of change of position is inapt:

In these cases, a defence of change of position was denied partly on the basis that the moneys received under a fraudulent preference were the proceeds of fraud. We do not consider that that reasoning is in accordance with modern English law about the nature of a voidable preference, or the law of the Cayman Islands. But in both cases concern was also expressed that to allow a change of position defence to a claim for the recovery of a preferential payment would run counter to the statutory scheme for the division of the insolvent company’s property among its creditors, since in the cases where it applied the anti-avoidance provision, under which preferential payments are voidable, would be deprived of practical effect. The reasoning also acknowledged that a claim designed to restore to a company in liquidation the assets which ought to be available for distribution among its creditors is different in nature from a conventional claim to restitution of a benefit transferred from one individual to another by a defective transaction. In the latter situation, it is the private interests of the two parties which are mainly in issue, and the equitable considerations requiring the benefit to be returned to the plaintiff can be cancelled out by equitable considerations arising from a change of position on the part of the defendant. In the former situation, on the other hand, the liquidator’s claim is brought in order to give effect to a statutory scheme which has been enacted by the legislature mainly in the public interest for the protection of creditors as a class.

In answer to those points, counsel for SEB have referred to some recent decisions of English courts in which the considerations relevant to a change of position defence have been treated as relevant also to the application of provisions of the Insolvency Act 1986.

The case of 4Eng Ltd v Harper [2009] EWHC 2633 (Ch); [2010] BCC 746 concerned a claim under section 423 of the Insolvency Act 1986, which deals with transactions at an undervalue and gives the court a discretionary power to make such order as it thinks fit for restoring the position to what it would have been if the transaction had not been entered into. A change of position was treated as being relevant to the exercise of that discretion. No change of position was however established on the facts, as it was not proved that the recipient of the money had used it for any purpose for which she would not otherwise have used her own funds. In the later case of BAT Industries plc v Sequana SA [2016] EWHC 1686 (Ch); [2017] Bus LR 82, para 523, Rose J stated that the matters referred to in 4Eng were “relevant to the question of what is the appropriate relief to be granted and do not provide a complete defence to the claim under section 423”.

The case of Rose v AIB Group (UK) plc was concerned with section 127 of the Insolvency Act 1986 (quoted in para 69 above), which gives the court a discretionary power to validate dispositions which would otherwise be avoided. Where a disposition is avoided, the section is silent as to recovery of the property or payment in question, as was explained earlier, and leaves it to the general law. In that context, it was accepted in Rose that a personal claim to restitution could be met by a defence of change of position. Mr Nicholas Warren QC, sitting as a Deputy Judge of the High Court, stated at para 41 that he did not know why the defence should not be available in cases where, for instance, a creditor did not know and could not have known of the existence of a winding up petition. He observed that “in other cases where payments can be treated as void or ultra vires, it is commonplace that restitution is available subject to restitutionary defences”. He contrasted section 127 with sections 238 and 239 (dealing respectively with transactions at an undervalue and preferences), noting at para 42 that, in contrast to the latter sections, no remedy was provided for recovery under section 127 and the matter was left to the general law. However, the defence failed on the facts, as the creditor was aware at the time when it received the payment of the risk of its being invalidated under section 127.

That approach has been followed in some other cases concerned with restitution following avoidance under section 127: see, for example, Clark v Meerson [2018] EWHC 142 (Ch); [2018] BPIR 661, para 47, and Officeserve Technologies, para 41. In the latter case Judge Paul Matthews, sitting as a High Court judge, thought that there would have to be “some good policy reason” why a change of position defence should not apply. In every case, the defence has failed on the facts. A different approach was adopted, in relation to section 284, in In re D’Eye (A Bankrupt) [2016] BPIR 883, where the registrar commented at para 49 that “we are not in Goff and Jones territory because we are not dealing with unjust enrichment generally but a particular statutory regime”.

The judgments in 4Eng Ltd v Harper and Rose v AIB Group (UK) Ltd have been criticised, for example by Professor Sir Roy Goode in his Principles of Corporate Insolvency Law (op cit), para 13-144. In his article “4Eng v Harper: Restitution and Insolvency” (2011) 24 Insolvency Intelligence 91, Simon Davenport QC argues that the wholesale recognition of a change of position defence to recovery claims arising from statute and made in support of the pari passu principle would be wrong in principle. He concludes that the fact that rejection of such a defence “may lead sometime to ‘harsh results’ in the language of Lord Goff is in the nature of the regime: the interests of the general body of creditors are put above those of the individual recipient”: p 95.

This is not the occasion on which to decide whether or not the reasoning in those cases was correct. They were not concerned with voidable preferences, and the statutory provisions in question were materially different from section 145 of the Companies Law. There is however nothing in them which would lead the Board to doubt the correctness of the conclusion which it has reached as a matter of principle in the light of authorities such as Wilson v First County Trust, and which is supported by the Canadian authorities it has cited.

Given that conclusion, it is not strictly necessary for the Board to consider whether a defence of change of position might otherwise have been made out on the facts of the present case. It is however appropriate to make some brief observations about the reasoning of the courts below on this point, particularly since concerns about that reasoning prompted the Board to consider in such detail whether the defence was in principle available.

It is not suggested that SEB had the rights of indemnity against beneficiaries which are available to a bare trustee under the common law. SEB nevertheless had the supposed benefit of contractual indemnities from the two Swedish funds, which were on their face designed to protect it from claims by third parties, including the present claim by the liquidators. Having heard expert evidence of Swedish law, the judge held that those indemnities had from the outset been worthless, mainly because the two funds lacked legal personality. A main reason why the judge held (obiter) that a change of position defence, if it had been available, would have failed, and the sole reason why the Court of Appeal reached the same conclusion, was that the effective cause of SEB being out of pocket if it had to repay the liquidators was the ineffectiveness of its indemnities, rather than the payment of the redemption proceeds to the two funds. Martin JA, with whom Morrison and Field JJA agreed, stated at para 65:

The judge also gave as reasons for his conclusion that a change of position defence would have failed the fact that SEB was, as trustee, obliged to account to its beneficiaries in respect of redemption proceeds, and that, as registered owner of the shares, it rather than its beneficiaries was the person liable to repay a voidable preference. Again the majority of the Board doubt whether these points truly impact upon a change of position defence. The first wrongly assumes that a change of position must be voluntary, while the second goes to the question whether SEB was enriched, as the Board has already explained.

For the reasons we have explained, however, we consider that a defence of change of position is in any event unavailable.

For the foregoing reasons, the Board has concluded that a claim to restitution of the amount paid by the Company to SEB cannot be met by a defence of change of position. As a consequence, the liquidators are in its opinion entitled to recover from SEB the amount which they received and remitted to the funds on whose behalf they were acting.

The Board should not depart from this case without acknowledging that the rejection of a defence of change of position may be capable of leading to harsh results. That possibility has been recognised in recent times in a number of jurisdictions. In England and Wales, for example, the provisions of the Insolvency Act 1986 dealing with transactions at an undervalue, preferences and transactions defrauding creditors provide a defence of bona fide purchase in order to protect third parties dealing with the person who obtained property from the company: see sections 241(2) and 425(2). In New Zealand, section 296 of the Companies Act 1993 also provides protection to the bona fide purchaser from the company’s transferee, and in addition provides the transferee himself with what amounts to a statutory defence of change of position. Whether it would be desirable to provide a measure of protection under the law of the Cayman Islands, either to creditors of the company or to third parties dealing with such creditors, may be a question worthy of consideration, but is not a matter for the Board.

Although pleaded Mr Chivers did not press the Board with regard to the contentions that his client should succeed because of illegality and public policy. The fact that the directing mind of the Company was guilty of a fraud does not render illegal the liquidators’ attempts to recover payments made unlawfully in the lead up to the insolvency. On the contrary they are discharging a legal duty to recover moneys owed properly to the Company and distribute these according to law among the creditors. The cause of action here of the liquidators is not founded on an illegal act but rather to redress the unlawful acts of paying out to some creditors in preference to others. Similarly public policy dictates that the courts should assist the liquidators in doing this. There is no public policy reason to the contrary in favour of SEB.

For all of the above reasons the Board will humbly advise Her Majesty that this appeal should be dismissed.

The Board is obliged to counsel for their learned oral and written arguments and to the solicitors for the careful and professional preparation of the papers relating to the case.

I have expressed my own views in the judgment of the Board on the issues on which we are agreed. I acknowledge that the repayment issues required considerable analysis and elucidation. We are all agreed that the appellant (“SEB”) is not entitled in law to avail of a defence of change of position to the claim by the liquidators, who, in any event, I do not consider to be unjustly enriching themselves as they are seeking to recover funds unlawfully dispersed and redistribute them to the creditors according to law. I observe that SEB never sought to argue for delay on the part of the liquidator as a defence. I do not agree, with respect, with the “serious reservations” of the majority as to the correctness of the findings of the Grand Court and Court of Appeal on this issue, as set out at paras 119-123 of the judgment of the Board.

SEB contends that in paying away the money to their clients they changed their position to their detriment. But they received the money on foot of their contractual relationship with their clients. They were registered shareholders in Weavering Macro Fixed Income Fund Ltd (“the Company”) as part of that relationship. On foot of instructions from the clients they redeemed the shares and paid the money received into the accounts of their clients. I incline to the view that the Court of Appeal in Cayman were correct in holding that there was no change of position in that seamless process.

SEB will now be at a loss not due to a change of position but for a different reason. It believed that it had effective indemnities from its clients. It called an expert in the relevant law, that of Sweden, Mr Alf-Peter Svensson, at the trial at first instance before Clifford J. But as he sets out in his judgment at paras 219 to 229 this witness gave evidence in cross-examination that, for reasons that the judge set out and accepted, these indemnities were “worthless”. The judge found at para 223 that SEB “never had the ability” to recover on the indemnities. Neither client was a mark for recovery by the time of the trial. But that was not the case in March 2009 when the Company went into liquidation, yet no steps were taken by SEB to protect its position eg by freezing the moneys in the accounts of their client received from the Company. This would be relevant if a defence existed as to the exercise of the court’s discretion in relation to any such defence. The failure of SEB to take effective steps to protect itself would render it more equitable to let it bear the loss, rather than the shareholders in the Company, of which, in any event it was one.

The judge, at para 229, found that even if a defence of change of position was available to SEB it had not been established on the facts. The Court of Appeal agreed. I consider they were entitled to do so.

We are also agreed on the outcome of the issue regarding fraud but I arrive there by a slightly different route which I set out in this judgment. My caveat reflects the approach to the issue of fraud tainting the net asset valuation (“NAV”) of the Company on the basis of which SEB received share redemption payments between December 2008 and February 2009 as set out in our main judgment.

SEB contends that the liquidators cannot recover against them because the payments were made to them on foot of a fraudulent assessment of the net asset value of the shares in December 2008. Its counsel relied on a dictum of Lord Denning in Campbell v Edwards [1976] 1 WLR 403 at 407, a contract case. “Fraud or collusion unravels everything.”

The respondents point out that the assessment of the NAV was in fact carried out by PNC, not by Magnus Peterson. No criticism is made of PNC. Nor is there any allegation of fraud against the two directors of the company. But the matter does not end there.

PNC arrived at their NAV on the basis of information provided by Magnus Peterson. As found at first instance he was the controlling mind of the company. He had arranged an elaborate fraud with a related company to give the appearance of assets in the company which did not exist. The reality therefore is that the NAV was indeed tainted with fraud as it was based on calculations and claims which were fraudulent and provided by Magnus Peterson.

What flows from fraud of this nature? Does it render the NAV and the payments to redeeming shareholders that were based on that NAV void ab initio or voidable on application to the court? Firstly, one notes that article 34 of the Articles of Association expressly says that the NAV is binding on all parties.

Secondly, the decision of the Board in Fairfield Sentry Ltd v Migani [2014] UKPC 9; [2014] 1 CLC 611 is of considerable assistance. The Privy Council there was concerned with a similar clause. The company concerned had invested very heavily in the funds run by Bernie Madoff. It transpired that Madoff was running a huge Ponzi scheme. Mr Chivers QC points out that in that case there was no finding of fact that the company was a party to Madoff’s fraud. It is to be contrasted with the situation here where the controlling mind was a fraudster who largely, although not completely, guided the actions of the administrator, as the judge at first instance found. However the judgment of Lord Sumption in that case nevertheless has considerable relevance to the facts before us.

I agree with the Board in Fairfield Sentry that a situation where a NAV, stated in the articles to be binding, was in fact to be retrospectively fixed or altered would tend to render the operation of the company and its share dealing unworkable. Even in a situation where there was fraud on the part of the company, that consideration would point against an automatic invalidity on the part of the transaction based on the false NAV, but would not prohibit a subsequent application to void the transaction. While Fairfield is distinguishable on the facts the principles set out therein are against the appellant.

Thirdly, it must be borne in mind that the extent of any fraud lying behind the statement of a NAV might vary enormously. Here there was undoubtedly a major fraud but it is nevertheless apparent that the company had sufficient assets or credit to make payments to redeeming shareholders of some US$90m. A NAV might have been influenced by a small or minimal fraud on the part of the employee who had misstated some part of the company’s assets. It would be quite wrong that such a marginal factor should render all transactions based on that NAV void ab initio.

In my view the NAV was and is binding on all parties until and unless successful proceedings are brought before the Grand Court by a party which has suffered loss to have the NAV in a transaction based upon it avoided for fraud.

This approach would be consistent with Order 12 rule 2 of the Companies Winding Up Rules. Martin JA at para 29 of the judgment of the Court of Appeal points out that section 112 of the Law applies only in the case of a solvent winding up but it does give a remedy for rectification of the register between members. It does not assist SEB on the facts here.

It would be necessary for a party who wished to have the NAV declared void to bring proceedings to do so. Those proceedings would be against the persons responsible at that time for a company. In this case that would be the joint liquidators. But it would have to be on notice to those who would be impacted by a decision of the court to conclude that the NAV was properly voidable. All the other persons who had sold on the basis of the allegedly fraudulent NAV would be entitled to be heard as a finding may adversely affect them. Such persons are not before this court and were not before the courts below.

On the facts of this case the conclusion that the NAV is voidable for fraud does not assist SEB. They have not been defrauded. On the contrary they received far more on behalf of their two clients than they were entitled to if an honest and accurate NAV had been stated at that time. If SEB brought proceedings and the NAV was set aside it would follow that the payments made under that NAV to SEB would be ordered to be repaid.

The view that fraud unravels everything reflects in particular the need for a court not to act as an engine of fraud assisting the fraudster against his victim. The courts will strive to compensate those who have suffered loss as a result of fraud and to recover compensation from the fraudster. But that is not the situation here. SEB received an inflated amount for its shares. It then chose to pass that money on without further ado to its clients. It was not defrauded.

On the other hand the liquidators are seeking to recover the moneys from SEB and, we were told, some 24 other redeeming shareholders in other proceedings, in order to distribute it in a lawful fashion amongst all the shareholders who sought to redeem prior to 19 March 2009. Presumably the shareholders who invested in the last months of the Company or chose not to redeem their shares will be left with nothing.

The Board is agreed that the fraud point does not assist SEB.