Andrew Flowers (Respondent) v Enterprise Insurance Company Plc (In Liquidation) (Appellant) No 2 (Gibraltar)
Case summary
Case ID
JCPC/2025/0021
Parties
Appellant(s)
Enterprise Insurance Company PLC (In Liquidation)
Respondent(s)
Andrew Flowers
Judgment details
Judgment date
17 November 2026
Neutral citation
[2026] UKPC 40
Hearing dates
Start date
29 April 2026
End date
29 April 2026
Justices
Judgment details
Michaelmas Term
[2026] UKPC 40
LORD REED:
1. Introduction
1. The appeal in this case raises questions about the assessment of damages and equitable compensation. They concern the circumstances in which the calculation of the claimant’s loss should take account of a benefit received as a consequence of the defendant’s breach of his obligations. There is also a cross-appeal which raises questions concerning the construction of an exclusion clause in an insurance policy, and whether the policy is unenforceable on grounds of public policy.
2. The background to the claim
2. The claimant, Enterprise Insurance Company plc (“EIC”), is an insurance company which was incorporated in Gibraltar in 2003 and carried on business until 2016. It was regulated by the Gibraltar Financial Services Commission (“the Commission”). It went into provisional liquidation in July 2016, and an order for its compulsory liquidation was later made on the ground that it was insolvent. At the trial in these proceedings, the deficit in the insolvency was estimated by the liquidator at around £200 million.
3. EIC, acting by its liquidator, brought proceedings against fourteen defendants. Eleven of them were directors of EIC during the relevant period. A twelfth defendant was a former shareholder of Enterprise Holdings Ltd (“EHL”), the holding company of EIC. The remaining defendants were EHL and another subsidiary of that company, EIG Services Ltd (“EIG”). EHL and EIG, like EIC, are in liquidation. EIC has settled the proceedings against all the defendants except one, Mr Andrew Flowers. He will be referred to as “the defendant”.
4. The defendant became a director of EIC in 2003, and was its managing director from 2003 to 2014. He was also a director of EIG and EHL, and owned a proportion of EHL’s shares through a family trust. He was described in the courts below as the major figure behind the group. The claim against him is based on allegations that he acted in breach of the fiduciary duty, and the common law duty of reasonable care and skill, which he owed to EIC as a director of that company.
5. The facts which are material to the issues in the appeal, as found by Restano J, sitting in the Supreme Court of Gibraltar (2023/GSC/008) (“the judge”), and affirmed by the Court of Appeal for Gibraltar (Sir Maurice Kay P, Sir Patrick Elias and Sir Colin Rimer JJA) (2024/GCA/005), are not in dispute before us. They concern the manner in which EIC was managed, and in particular two aspects of its management which contributed to the inadequacy of its reserves to meet its liabilities. The first relates to the wrongful diversion of EIC’s funds to EIG. The second relates to the failure of EIC’s directors to establish any reserves to cover potential liabilities arising under certain insurance policies issued by EIC, referred to in these proceedings as the Icebreaker policies. Those policies are also the subject of the cross-appeal.
(1) The diversion of EIC’s funds
6. Insurance companies with a head office in Gibraltar are subject to legislation which imposes restrictions on the way in which they conduct their business. Those include a prohibition on paying dividends without having notified the Commission and obtained its consent. Such consent is not granted unless the Commission is satisfied that the company is adequately capitalised. A further restriction arises from a requirement that the company maintain a prescribed level of solvency, designed to ensure that it has sufficient capital to meet its liabilities. The basic solvency requirement, known as the “required minimum margin” or RMM, depends on a number of factors, including the type of risks being underwritten by the insurance company. The Commission has a range of powers which can be exercised if the RMM is not met, including a power to limit premium income, thereby restricting the amount of new business which can be transacted.
7. It is not possible for an insurance company to meet its RMM by borrowing, as the assets included in the calculation of the RMM must be free of incumbrances. However, if a third party, such as an unregulated parent company, obtains the loan and puts capital into the company (for example as a gift, or by subscribing for an allotment of shares), the capital counts as an asset for RMM purposes, as the liability to repay the loan is that of the third party rather than the insurer.
8. It is in part because of these considerations that some insurance companies have adopted what has been described in these proceedings as a triangular model, as a method of realising unencumbered capital. The model operates as follows. A holding company is established to own both the insurance company and a sister service company. The latter provides the insurance company with the services which it requires, and the insurance company pays it for those services. The payments generate profits in the hands of the service company, which it can distribute to the holding company in the form of dividends. The holding company can then use the dividends to inject capital into the insurance company. It may do so, for example, by subscribing for an issue of additional shares in the insurance company, paying for the shares either out of its own resources or out of sums which it has borrowed from third parties.
9. EIC was part of a corporate structure which ostensibly adopted this triangular model. It was owned by a holding company, EHL, and it had a sister company, EIG, whose name, EIG Services Ltd, suggested that it was a service company. EIC paid EIG 8% of its gross premiums – which the judge found to be about 40% of its net premiums after the payment of commission and profits to intermediaries – for marketing services purportedly rendered under an agreement entered into between them. However, in reality no such services were provided. Other payments were made by EIC to EIG, ostensibly for the use of a software system. Although this arrangement was initially genuine, from about 2009 it became another means of diverting EIC’s funds. Another payment made by EIC to EIG, ostensibly as a fee for arranging reinsurance, was also found by the judge to have no proper justification.
10. By means of these practices, large sums of money were extracted from EIC over the 12 years between 2005 and 2016 and paid to EIG, an unregulated company which could deal with the money as it saw fit. The judge found that the defendant was dishonestly involved in the making of these payments, which he knew were gratuitous.
11. EIG used part of the money wrongfully transferred to it to pay £12.3 million to the shareholders of its parent company, EHL, ostensibly as remuneration for services which they had rendered to it. The defendant, as a 50% shareholder in EHL, received half of that amount. In reality, no such services had been rendered, and the relevant invoices were fictitious. EIG paid another £16.6 million to EHL in the form of dividends.
12. The effect of the gratuitous payments made by EIC to EIG was to leave EIC under-capitalised, with a solvency margin which fell far short of the RMM. Instead of its premium income being retained and invested to establish reserves, so that it would be able to meet liabilities under the insurance policies which it had sold, a substantial proportion of its funds was dishonestly diverted to an unregulated sister company, which then paid large sums to the defendant and others, and distributed other substantial sums as dividends to an unregulated parent company, EHL.
13. EHL’s resources were derived partly from the dividends it received from EIG, and partly from borrowings of around £44 million, which it obtained by issuing bonds. EHL used some of its resources to inject capital into EIC in 2006 and in every year from 2010 to 2015, by subscribing for allotments of shares. Since EIC was already wholly owned by EHL, the issue of the additional shares did not materially increase the value of EHL’s shareholding. The purpose of the share subscriptions was to provide EIC with additional capital.
14. As explained below, the provision of the additional capital enabled EIC to allay the Commission’s concerns about its under-capitalisation, and to underwrite additional insurance business, thereby acquiring additional premium income but also incurring additional liabilities. Part of the premium income which EIC obtained from the new business was then diverted to EIG by means of the dishonest practices described above. Apart from the money which it invested in EIC, EHL also used its resources for other purposes, including the payment of consultancy and legal fees, the payment of interest and capital repayments on its borrowings, and the purchase of an interest in a business in the UK.
15. Altogether, EHL subscribed £32.3 million for additional shares in EIC. Of that total amount, £8.3 million was derived from the dividend payments which EHL received from EIG. The remaining £24 million was borrowed by EHL by issuing bonds on the Frankfurt Stock Exchange. Under the bonds, EHL undertook to make annual payments of interest on the nominal value of each bond, and to repay the nominal value of the bond on maturity, which was to occur in 2017 or 2020, depending on the bond. The bonds were issued on the basis, set out in the prospectus, that EHL would make the necessary payments using the dividends it received from EIG. In the event, EIC’s insolvency resulted in EHL’s being unable to comply with its obligations under the bonds. EHL went into liquidation before the bonds matured.
16. EHL had several other directors besides the defendant. Two of them gave evidence at the trial. Both of them understood the circulation of money around the group to be a bona fide example of the operation of the triangular model described in para 8 above. The judge accepted that neither of them knew that the payments made by EIC to EIG were other than in consideration of services genuinely provided. In an important finding of fact, the judge found that there was no evidence that EHL was aware that the payments made by EIC to EIG were made otherwise than for proper value.
17. Despite the injections of share capital, EIC continued to be under-capitalised. That is unsurprising. In the first place, EIC was covertly paying for the capital injected into it, part of which was ultimately derived from the premiums EIC received for policies it had written. The balance of the capital was borrowed by EHL on the basis that the loans would be serviced out of funds ultimately derived from EIC. Secondly, although the additional capital enabled EIC to underwrite new business, receiving additional premiums and incurring additional liabilities, a large proportion of the premiums was then dishonestly diverted to EIG, leaving EIC with the liabilities.
18. The under-capitalisation of EIC did not escape the attention of the Commission. Concerns were expressed and steps were taken, but the problem was not resolved. There were occasions when the Commission considered imposing restrictions on the business which EIC could transact, but did not proceed to do so because EIC produced ostensibly new capital. On other occasions, restrictions were imposed but disregarded.
19. The injections of capital by EHL enabled EIC to continue trading and to expand its business. However, the judge found that EIC only turned a modest profit in four years out of twelve, with very significant losses suffered from 2013 onwards. He concluded at para 675:
“The circular nature of payments was clearly not therefore in EIC’s best interests. The true position is that the recirculation of money allowed EIC to stave off a direction from the [Commission] that it cease writing business, allowed it to stagger on as a business (albeit largely unprofitably) and enabled large payments to be made to [the defendant] and other ultimate beneficial owners of the group.”
He added at para 696:
“The payments which were paid back into EIC were a means of propping EIC up which allowed it to keep writing further premium, for greater losses to be incurred, whilst allowing for ongoing substantial payments to be made out to [the defendant] and others along the way.”
(2) The Icebreaker policies
20. As well as other types of insurance policy, EIC underwrote policies for investors in a tax avoidance scheme known as “Icebreaker”, which insured the policy holders against losses which they might incur as a result of investment in the scheme. The policies were closely connected to the scheme, and the policy was one of the documents made available to prospective investors, although it was not obligatory for an investor to take out a policy. The judge found that the defendant was the dominant negotiator and underwriter of the policies, and that the scheme was fully disclosed to EIC, through the defendant, before the insurance policies were created and sold.
21. Participants in the scheme invested in a limited liability partnership (LLP), of which they became members. Typically, 20% of their investment came from their own resources. The remaining 80% was borrowed from a bank, and was returned to the members following a complex series of transactions. It was not invested by the LLP in its trading activities. The LLP invested in the intellectual property rights in a work of art such as a film, a book or a musical composition, typically by purchasing an exclusive licence. The investment was anticipated to result in a loss in its first year, when the expenditure was incurred and little income was likely to be received: indeed, clause 3 of the insurance policy referred to the “anticipated first year trading losses of the LLP”. The LLP then submitted tax returns for the year, showing the amount invested by members, including the amount borrowed from the bank, as an expense of its trade. The members submitted tax returns in which they claimed tax relief for their share of the LLP’s loss, known as “sideways loss relief”, against their taxable income. Sideways loss relief enables the members of an LLP to set their individual share of the LLP’s trading losses against their other income for income tax purposes, provided certain conditions are met, including that the trade of the LLP is being carried on as a commercial venture.
22. The scheme was the subject of legal proceedings in the UK. In proceedings between some of the LLPs and the Commissioners of Revenue and Customs (“the Revenue”), it was accepted that the LLPs were carrying on the trade of exploiting intellectual property rights, and that the trade was carried on with a view to profit. However, it was held that the 80% of members’ investments which the LLPs received pursuant to the bank loans was not expended for the purposes of that trade and was accordingly not a deductible expense: Acornwood LLP v Revenue and Customs Comrs [2016] UKUT 361 (TCC); [2016] STC 2317. Subsequent proceedings established that sideways loss relief could not be claimed by members of the LLPs even with respect to such losses as were accepted to have been incurred by the LLPs, as the trade of the LLPs was not being carried on as a commercial venture: Seven Individuals v Revenue and Customs Comrs [2017] UKUT 132 (TCC); [2017] STC 874.
23. Under the insurance policies, EIC undertook to indemnify the members against the loss of the 20% contribution that they paid to the LLP at the end of five years, adjusted for any profits received from the LLP. EIC undertook that risk for a premium of 2.75% of the amount at risk. EIC was off risk if either (a) at the end of five years, the member’s share of the LLP’s income equalled or exceeded their 20% contribution (ie the member achieved a 100% return on the amount actually invested by the LLP), or (b) the Revenue agreed, or a court or tax tribunal decided, that the LLP’s first year trading losses were allowable for tax purposes and that the LLP’s tax return for the relevant year was correct.
24. Clause 11.4 of the policy provided the insurer with a defence to liability in the following terms:
“The Insurer shall have no liability under this Policy where the Insurer reasonably believes, and has reasonable grounds for such belief, that the LLP and the Insured have colluded in an attempt to engineer the failure of the LLP leading to insufficient income being generated causing a claim to arise on this Policy.”
25. The judge found that, as the defendant knew, the way in which the scheme was intended to work was fixed before a participant became a member of the scheme, and the participant had no ability to alter it before or after joining. There was no evidence that any policy holder took any decision open to them which was designed to engineer the failure of the LLP they joined.
26. EIC did not establish any reserves to cover potential liabilities under the Icebreaker policies; nor was there any reinsurance of the risk. The gross premiums received were less than £1 million. The liabilities were estimated by the liquidator at £31.4 million.
3. The claim and the defence
27. Against that background, EIC’s claim against the defendant is based on a number of causes of action, and seeks a variety of remedies. The causes of action include breach of the fiduciary duty which the defendant owed EIC as its director, and breach of his common law duty to exercise reasonable skill and care in relation to its management. The breaches specified include the diversion of EIC’s funds, the failure to ensure that EIC retained sufficient funds to meet its solvency requirements and to fulfil its obligations to policy holders, the underwriting of the Icebreaker policies, and the failure to create appropriate reserves after the policies were written, or, if adequate reserves could not be created, to recognise then that the company was insolvent and thus reduce the company’s losses.
28. The remedies sought include damages and equitable compensation. In relation to those, EIC pleads that it has suffered loss and damage as a result of the defendant’s breach of fiduciary duty and negligence in respect of the payments wrongfully made to EIG, and that the defendant is liable to pay damages or equitable compensation in an amount which equates to those payments. Further or in the alternative, in relation to the defendant’s failure to ensure that EIC held sufficient reserves to meet its liabilities to policy holders, EIC states that it has suffered loss and damage in the amount by which its losses would have been less had the defendant not acted in breach of duty. The measure of loss claimed on that basis is the difference between the position EIC was in as at the date in 2016 when the directors concluded that it was insolvent, and the position it would have been in at that date if (a) the Icebreaker policies had not been underwritten, or appropriate reserves had been established for them, or if such reserves could not have been established, a decision had been taken that EIC was insolvent, or (b) the payments to EIG had been recalled, or economies made at any stage before 30 June 2010 (the date as at which it is alleged that EIC was insolvent), or a decision taken in any year from 30 June 2010 onwards to wind up the company.
29. In reply, the defendant denies that EIC has suffered any loss, and relies, amongst other matters, (a) on the amount of capital introduced into EIC by EHL in return for share allocations, ie the £32.3 million, and (b) on a contention that the Icebreaker policies did not give rise to any risk, and therefore did not require the creation of any reserves.
30. In relation to the first of these points, the defendant contended at trial that when equitable compensation or damages were assessed, credit should be given for the amounts subscribed by EHL for new shares in EIC. His case was that the capital injections formed part of the same overall course of dealings as his alleged breaches of duty, they increased EIC’s assets, and hence they reduced the loss suffered.
31. In relation to the second point, the defendant relied at trial on clause 11.4 of the Icebreaker policy. He argued that it resulted in EIC coming off risk if, as was the case, the whole amount of the member’s contribution to the scheme, including the amount which was borrowed from a bank, was not put at risk, or if the member was not a genuine investor.
32. The defendant pointed out that one of the policy holders whose claim, said to arise once the tax avoidance scheme was found to be ineffective, had been rejected by EIC had made an application to the UK’s Financial Ombudsman Service, challenging EIC’s decision not to meet his claim. The initial assessment of the Ombudsman Service was that it was reasonable for EIC to rely on clause 11.4 of the policy as a basis for declining the claim.
33. The defendant also emphasised that applications for compensation under the UK’s Financial Services Compensation Scheme (“FSCS”), made by a group of persons insured under the Icebreaker policies, had been rejected on a number of grounds. One of the grounds, as summarised by Lang J in subsequent proceedings for judicial review, was that the Icebreaker partnerships, of which the insurance policies formed an integral part, were failed tax avoidance schemes and were accordingly void as being contrary to public policy.
34. An application for permission to apply for judicial review of that decision, on the ground, inter alia, that in determining that the policies were void as being contrary to public policy the FSCS acted irrationally or unreasonably, was refused on the papers by Sir Ross Cranston, sitting as a High Court judge. The renewed application was then refused at an oral hearing by Lang J: R (Kennedy) v Financial Services Compensation Scheme Ltd [2021] EWHC 3039 (Admin). She considered that the challenge to the FSCS’s decision was unarguable, stating at para 10 of her judgment:
“The defendant was entitled to take the view that contracts which seek to avoid tax legislation can engage the illegality doctrine …”
She added at para 11 that it was open to the FSCS to conclude that the policies were constituent components of an investment scheme deliberately designed to facilitate tax avoidance.
35. An application for permission to appeal to the Court of Appeal was also unsuccessful. Coulson LJ noted that the Icebreaker scheme had been found to be a tax avoidance scheme, and concluded:
“Accordingly … if – as the [FSCS] found – those policies were a critical part of the Icebreaker Schemes, then those policies too would be contrary to public policy.”
4. The judgments below
(1) The judgment at first instance
36. After a lengthy trial, the judge found substantially in favour of EIC. In an impressive judgment which dealt carefully with a plethora of issues, he found that the defendant had acted in breach of his fiduciary duty to act honestly and in good faith in the best interests of EIC, and had breached his common law duty to exercise reasonable care and skill, when he allowed the payments under the marketing services agreement, the overpayments under the software agreement, and the payment of the arrangement fee to be made to EIG. Declarations to that effect were made.
37. The judge also found that the defendant had breached his duty to exercise reasonable care and skill when he wrote the Icebreaker business on behalf of EIC and failed to establish any reserves. The business should either never have been written, or appropriate reserves should have been established, which would not have been possible. The defendant should have made adequate inquiries about the businesses that the LLPs were investing in; he should have obtained independent legal and tax advice about the Icebreaker scheme; he should have provided clear instructions on the exclusion clause to be included in the policies; he should have ensured that policy holders signed an appropriately drafted proposal form; and he should have carefully considered reserves and reinsurance protection. A declaration to that effect was made.
38. The judge rejected the defendant’s argument based on clause 11.4. He noted that liability was excluded if the insurer reasonably believed, and had reasonable grounds for such belief, that the insured “colluded with the LLP to engineer the failure of the LLP”, which, he said, connoted active steps being taken to make the investment fail. There was no evidence that any such steps had been taken. The judge did not say in terms whether he also rejected the submission that the policies were unenforceable on grounds of public policy, but his conclusions are only consistent with his having done so. He recorded EIC’s submission that there was almost a century of jurisprudence at the highest appellate level establishing the lawfulness of tax avoidance schemes. He observed that the case of Patel v Mirza [2016] UKSC 42; [2017] AC 467, to which Coulson LJ had referred when considering the application for permission to apply for judicial review of the FSCS decision, arose from a case of established illegality, which was not the position in relation to the Icebreaker scheme.
39. The judge also found that the defendant breached his duty to have regard to the interests of creditors and to keep EIC’s solvency under review from 30 June 2010, on the basis that EIC was insolvent from that date, having failed to make proper reserves for its insurance business.
40. The judge entered judgment against the defendant for equitable compensation and damages, to be assessed at a later hearing. In that regard, it was argued on behalf of EIC that no credit should be given for the capital injected into the company by EHL. When the money went out of EIC, it was unconditionally owned by EIC; but when EIC later received sums from EHL, that was in exchange for shares in EIC, and formed part of a separate transaction. Reference was made to Villeneuve v Gaillard [2011] UKPC 1 (“Villeneuve”), where Lord Walker said at para 93:
“There is no general principle that in assessing either common law damages or equitable compensation for breach of fiduciary duty, losses should be reduced by adventitious gains in separate transactions between the same parties”.
On the other hand, the defendant argued, under reference to Bartlett v Barclays Bank Trust Co Ltd (Nos 1 and 2) [1980] Ch 515, Target Holdings Ltd v Redfern [1996] AC 421 (“Target Holdings”), and AIB Group (UK) plc v Mark Redler & Co Solicitors [2014] UKSC 58; [2015] AC 1503 (“AIB v Redler”), that account should be taken of the capital injections made by EHL into EIC, since they had reduced EIC’s loss.
41. The judge rejected the defendant’s argument. He emphasised that there was no suggestion that EHL’s purchase of shares in EIC was other than a genuine transaction. It followed that EIC had given value in exchange for the money it received, namely the shares, and no credit was therefore appropriate. Although he did not rely on this point, it is also relevant to note that the judge found that once the improper payments had been made by EIC to EIG, there was nothing that constrained the directors of EIG or EHL to re-invest any part of that money in EIC.
(2) The judgment on appeal
42. Virtually all the judge’s conclusions were challenged on appeal. The finding of negligence summarised in para 39 above was reversed, although the finding of fact that EIC was insolvent from June 2010 was upheld. The other findings of negligence, dishonesty and breach of fiduciary duty were upheld. For present purposes, it is necessary to focus on the issues relating to (a) the assessment of damages and equitable compensation, and (b) the Icebreaker policies.
(a) The assessment of damages and equitable compensation
43. In considering whether credit should be given for the sums paid into EIC for new shares, Elias JA, with whose judgment the other members of the court agreed, cited the judgment of Lord Sumption in Tiuta International Ltd v De Villiers Surveyors Ltd [2017] UKSC 77; [2017] 1 WLR 4627 (“Tiuta”), para 12:
“The general rule is that where the claimant has received some benefit attributable to the events which caused his loss, it must be taken into account in assessing damages, unless it is collateral. In Swynson Ltd v Lowick Rose LLP [2017] UKSC 32; [2018] AC 313, para 11, it was held that as a general rule ‘collateral benefits are those whose receipt arose independently of the circumstances giving rise to the loss’.”
44. In Elias JA’s view, the judge had erred in treating EHL’s purchases of the shares as separate transactions from those causing the loss. In his opinion, there was plainly a close link between the payments made to EHL via EIG and the sums paid back to EIC. That was always a feature of the triangular model described at para 8 above, albeit that there was no legal obligation binding EHL to pay anything to EIC. Indeed, in the context of considering an allegation that EIC had unlawfully provided financial assistance for the purchase of its own shares – an allegation which the judge rejected – the judge had held (para 863) that “the prospect of capitalising EIC from the [wrongful] payments was on Mr Flowers’ radar from the outset”. These were not, therefore, unrelated transactions; they were, to use Lord Sumption’s words, payments which were “attributable to the events which caused his loss.”
45. Elias JA went on to accept that although the transactions were related, with the consequence that any benefit received by EIC could not be regarded as collateral, the payments should not be taken into account when assessing the loss suffered by EIC if EIC had provided good consideration to EHL in return. However, it was unreal to say that when EHL received consideration in the form of the shares issued by EIC, it received something of value. EHL was always the sole shareholder of EIC, and the addition of further shares gave it nothing of substance. In short, the payments from EHL were not collateral benefits but were directly attributable to the events causing the loss, and their effect was to provide a direct benefit to EIC. They should therefore be offset against the losses suffered. The same conclusion was reached in relation to equitable compensation. Applying the approach set out by Lord Browne-Wilkinson in Target Holdings, equitable compensation was designed to make good the loss suffered. It would be inequitable to fail to give credit for those payments.
46. Accordingly, the Court of Appeal ordered that “credit should be given for £32.3 million (comprising £8.3 million paid by EHL into EIC for new shares and £24 million paid for the subscription of new shares paid out of a loan which EHL had taken on the Frankfurt bond market where EIC was the original source of the funds)”.
(b) The Icebreaker policies
47. In relation to the defendant’s negligence in underwriting the Icebreaker policies and failing to establish any reserves against potential liabilities, the Court of Appeal held that neither clause 11.4 nor public policy operated to exclude EIC’s potential liability under the policies to investors in the LLPs.
48. In relation to clause 11.4, Elias JA noted that since it was argued that the clause provided a defence to a claim by any policy holder, the facts which excluded liability must be inherent in how the scheme worked. However, as the defendant accepted, an investor could not affect the way the scheme operated. If an investor chose to join the scheme, it was on the terms proposed by the promoters of the scheme: there was no room for individual negotiation over the structure of the scheme itself, nor over how the funds were invested. An investor could not be said to be “colluding” with the LLP merely by investing in the scheme on the terms offered. Nor could the members be described as attempting to “engineer” the failure of the LLP: they neither selected the project for a particular scheme, nor were they involved in its exploitation. Elias JA recognised that it was not necessary under clause 11.4 for there to be an act of collusion for the forbidden purpose: it was enough that EIC had reasonable grounds to believe that to be the case. However, there were no reasonable grounds for such a belief. It was not a reasonable inference from the facts that there had been any relevant collusion or any attempt to engineer the failure of the LLP.
49. A further problem was that, if clause 11.4 had the effect for which the defendant contended, the policy holder was never being protected against any risks whatsoever: the exclusion clause would apply immediately the policy came into effect and EIC agreed to underwrite the risk. Yet all the elements of the scheme were known to EIC. It would have designed a bespoke policy, framed to deal with these specific investments, which on its terms could never come into effect. The court would be very slow to interpret the policy in a way which had that effect.
50. It was submitted on behalf of the defendant that the investors and the LLPs had colluded to misrepresent the true nature of the scheme to the Revenue and EIC, by representing that the entire investment, including the 80% which was borrowed, was an expense of the trade, when in fact the borrowed 80% was never used to drive the trade forward.
51. Elias JA noted that the proceedings had not been defended on the basis that any misrepresentations were made to EIC, and a defence based on misrepresentation would have raised different issues, such as the avoidance of the policy, and tax evasion rather than tax avoidance, with the consequence that the scheme would be unlawful. Far from there being any misrepresentation, there was never any attempt to conceal the fact that the loan was not going to be put at risk: that was a key feature of the scheme, and the judge was satisfied that the defendant always appreciated this.
52. The Court of Appeal also rejected the argument based on public policy. Elias JA observed that, when dealing with a tax avoidance scheme, the court does not start with an abstract view that such a scheme is immoral or unacceptable because it employs artificial devices or stratagems designed to bring itself within the terms of the statute conferring the benefit. The only relevant question is whether it succeeds in its objective, which it will do if it falls within the scope of the relevant provision. It was true that the adoption of a purposive approach to the interpretation of revenue statutes had made it more difficult to adopt successful tax avoidance schemes, but that was not because the fundamental principle had changed: the question continued to be whether the scheme fell within the relevant statutory provision.
5. The grounds of appeal and cross-appeal
53. EIC appeals against the decision of the Court of Appeal on three grounds. First, it is argued that the court erred in law in holding that credit ought to be given for the payments made by EHL to EIC, thereby reducing the sums payable to EIC by the defendant. It is argued that the court was mistaken in considering that the payments were “attributable to the events which caused [EIC’s] loss” within the general rule explained by Lord Sumption in Tiuta, para 12. It ought to have found that they were collateral benefits which arose independently of the circumstances giving rise to the loss, as explained by Lord Sumption in Swynson Ltd v Lowick Rose LLP [2017] UKSC 32; [2018] AC 313 (“Swynson”), para 11, or were “adventitious gains in separate transactions”, as explained by Lord Walker in Villeneuve at para 93.
54. Secondly, it is argued that the court erred in holding that it would be inequitable to fail to give credit for the EHL payments, and paid insufficient regard to the inequitable result of allowing credit. Three points are made in that regard:
(1) Giving credit for the payments would permit the defendant to set off against his personal liability for dishonest breaches (a) in the case of the payments made by EHL out of its own resources, monies received by reason of a third party’s decision to choose to use some of the dishonestly extracted money to purchase shares in EIC; and (b), in the case of the payments made by EHL out of amounts which it had borrowed, monies received by reason of a third party’s decision to borrow monies from third party investors, namely the bondholders, and to invest some of those monies in EIC.
(2) EIC provided shares in return for the EHL payments. Since the defendant at all material times had an ownership interest in EHL, he would have a double benefit from the monies dishonestly extracted – the new shares, and credit against the dishonestly extracted monies.
(3) The EHL payments permitted the defendant’s dishonest scheme, which he used to extract £12.3 million for his and his fellow shareholder’s benefit, to continue so that he could dishonestly extract further money from EIC.
55. Thirdly, it is argued that the EHL payments were not true benefits to EIC, because they were introduced to keep EIC afloat and directly produced additional claims liabilities.
56. The defendant cross-appeals on two grounds. First, it is argued that on a proper construction of the Icebreaker insurance policy, clause 11.4 meant that EIC carried no insurance risk in respect of the Icebreaker scheme, so that underwriting such policies could not be negligent. The purpose of clause 11.4 was to ensure that EIC would not be at risk if the LLP’s loss was contrived rather than commercial, and that was precisely the situation which was found to exist. Only a small fraction of the investors’ money was truly invested, yet the full amount was treated as a trading expense so as to create artificial losses. The Icebreaker arrangements were of the very nature described in clause 11.4: a collusive attempt to engineer failure.
57. Secondly, it is argued that the Court of Appeal erred in holding that public policy did not render the Icebreaker policies void or unenforceable. The principles in Patel v Mirza required the court to deny enforcement of transactions which formed part of, or furthered, a scheme found to be contrary to law and the public interest. Enforcing such policies would be contrary to the public interest in maintaining the integrity of the tax system, and in preventing the court’s processes from being used to give effect to a scheme found to be unlawful or abusive.
6. Should a deduction be made from the damages or equitable compensation payable to EIC in respect of the payments made to it by EHL for the issue of shares?
(1) Damages
58. In answering the question whether the payments made by EHL to EIC should be taken into account in the assessment of damages, it is helpful to begin by distinguishing two aspects of the matter. The first is the question whether EIC benefited from the payments. The second is the question whether, if it did, the benefit should be taken into account. The Board will consider the principles relevant to each of those questions in turn.
(a) Was there a benefit?
59. In deciding whether the receipt of a payment constituted a benefit which ought to be taken into account in the assessment of damages, the starting point is the purpose of an award of damages, namely to compensate the claimant for the loss caused by the defendant’s breach of his obligations. In calculating the claimant’s loss, it is necessary to compare the claimant’s actual position with the position in which it would have been if the defendant’s obligations had been performed, so that damages can make up the difference: see, for example, Nykredit Mortgage Bank plc v Edward Erdman Group Ltd (No 2) [1997] 1 WLR 1627, 1631.
60. Accordingly, the correct approach to assessing whether the claimant has received a benefit which should be deducted in the assessment of damages is not, ordinarily at least, to keep a running account of amounts paid out by the claimant as a result of the wrongful conduct and amounts received, and to assess damages on the basis of the net balance. Instead, it is necessary to examine the claimant’s overall financial position prior to trial, and to compare that with its hypothetical position if the wrongdoing had not occurred. The claimant’s actual position may have been improved by the receipt of a payment or other advantage which has resulted from the defendant’s wrongdoing. If so, applying the general principle that damages are designed to compensate the claimant for the loss caused by the defendant’s breach of his obligations, that benefit should be taken into account in the assessment of damages. However, the receipt of a payment will not necessarily have improved the claimant’s actual position. For example, if the claimant gave valuable consideration for the payment, then there will not have been any net benefit: the claimant merely received what it was entitled to. Equally, if the result of the payment was that the claimant made greater losses than it would otherwise have done (for example, because it carried on trading unprofitably for longer than it would have done without the payment), then the payment is unlikely to have constituted a benefit.
61. The question whether payments constituted a benefit has been considered in a number of cases, including cases concerned with dishonest schemes in which some of the moneys wrongfully abstracted from the claimant were returned to it in order to keep the scheme running, so that further sums could be abstracted from it. The relevant cases are discussed below.
62. Finally, in relation to the general principles applicable to this topic, the evidential burden of proving that the claimant received a benefit which should be deducted rests on the defendant: see, for example, Mead v Clarke Chapman & Co Ltd [1956] 1 WLR 76, where the burden of proving that a child’s loss of dependency, following her father’s death, had been reduced as a result of her mother’s remarriage was held to rest on the defendant; Cheltenham Borough Council v Laird [2009] EWHC 1253 (QB); [2009] IRLR 621, para 56, where the loss resulting from a misrepresentation was alleged to have been reduced by the receipt of a benefit; and Sainsbury’s Supermarkets Ltd v Visa Europe Services LLC [2020] UKSC 24; [2020] Bus LR 1196 (“Sainsbury’s”), para 211, where the same approach was applied in the analogous context of alleged mitigation of loss.
(b) The deductibility of financial benefits
63. The law governing the deductibility of benefits is complex, and for the purposes of the present case it is necessary to consider three aspects. They are, however, closely related; and they share the objective of ensuring that the award of damages will compensate the claimant for the net loss for which the defendant can fairly and reasonably be held responsible.
(i) Benefits resulting from the wrongful conduct of the defendant
64. The wrongful conduct of the defendant may not only cause harm to the claimant, but may also result in a compensating benefit or advantage. A familiar example is the breach of a contract for the supply of goods. The defendant may supply the claimant with goods which do not meet the requirements of the contract but which nevertheless have a market value. In such a case, the value transferred is normally taken into account in the computation of the claimant’s loss. That follows from the compensatory aim of an award of damages. The point is illustrated by Smith New Court Securities Ltd v Scrimgeour Vickers (Asset Management) Ltd [1997] AC 254 (“Smith New Court”), where the claimant had been induced, as a result of the defendant’s fraudulent misrepresentation, to pay a price for shares which was above their market value. Lord Browne-Wilkinson summarised the relevant principles as including the following (p 267):
“(3) … the plaintiff is entitled to recover by way of damages the full price paid by him, but he must give credit for any benefits which he has received as a result of the transaction; (4) as a general rule, the benefits received by him include the market value of the property acquired …”
65. What is meant by a benefit being received “as a result of the transaction”? This involves more than factual or “but for” causation: the defendant is not entitled to take credit for every benefit of which his conduct was a “but for” cause, any more than he is liable for every loss of which his conduct was such a cause. In relation to both losses and benefits, the law takes account only of those which are sufficiently closely connected to the defendant’s conduct.
66. Some of the clearest illustrations of the point are cases concerned with benefits resulting from action taken by the claimant in response to the defendant’s wrongful conduct. Those cases are discussed below. But there are also illustrations where there was no intervening conduct of the claimant. For example, in Needler Financial Services Ltd v Taber [2002] 3 All ER 501 (“Needler”), the claimant sought damages from the defendant for negligently advising him to transfer the benefits to which he was entitled under an occupational pension scheme to a personal pension plan with a mutual assurance company. The mutual fund subsequently demutualised, and the claimant, as a member, received an allocation of shares. He later retired, and found that his pension was less than he would have received if he had remained in the occupational pension scheme. The question was whether the value of the shares should be taken into account in the assessment of damages. Sir Andrew Morritt V-C, applying the approach adopted in the mitigation case of Hussey v Eels [1990] 2 QB 227, discussed below, considered that “the relevant question is whether the negligence which caused the loss also caused the profit in the sense that the latter was part of a continuous transaction of which the former was the inception” (para 24). He held that the benefit in question was not deductible, because it “did not flow, as part of a continuous transaction, from the negligence” (para 26).
67. A number of cases, of particular relevance in the present context, have involved the claimant’s being induced by wrongful means to make payments, and part of the payments subsequently being used by the defendant, or entities associated with the defendant, in ways which were argued to be of benefit to the claimant. Such cases have raised both the question whether the claimant received a benefit, and also the question whether the benefit was to be regarded in law as a result of the defendant’s wrongful conduct.
68. The most influential of these cases has been Komercni Banka AS v Stone and Rolls Ltd [2002] EWHC 2263 (Comm); [2003] 1 Lloyd’s Rep 383 (“Komercni”). The claimant bank was induced by deceit to issue a series of letters of credit in favour of the defendants. Their accomplice was a customer of the bank, which had requested the issue of the letters of credit. Sums of money derived by the defendants from the letters of credit were paid into accounts held at the bank by their accomplice and associated companies. The bank applied the payments in discharge of debit balances on the accounts. The defendants argued that the discharge of those liabilities constituted a benefit to the bank which should be set against the amount of which it had been defrauded.
69. Toulson J took as his starting point Lord Browne-Wilkinson’s dictum in Smith New Court that the claimant must give credit for any benefits which he has received as a result of the transaction. He considered that the question whether an alleged benefit had been received “as a result of the transaction” could not be determined by the application of a “but for” test. Where the wrongful conduct consisted of causing the victim to enter into a venture or transaction which he would not otherwise have entered into, and the wrongdoer alleged that the victim had received a subsequent benefit which he would not have received but for entering into the venture or transaction, Toulson J considered that “the question to be asked is whether the receipt of the benefit was not merely a result of the venture or transaction, in a historical sense, but was part of the complex of obligations and benefits intrinsic, ie belonging naturally, to the venture or transaction” (para 167).
70. Toulson J analysed the case before him as involving a series of transactions, since each time the bank accepted fraudulently presented documents, it accepted liability under a separate letter of credit, with a separate right to reimbursement by its customer. If the liability which the customer incurred as a result of the issue of one of the letters of credit was discharged using money derived from the dishonest scheme, then the consequence was that no loss was suffered in respect of that letter of credit. However, if money derived from the fraud was used to discharge an unrelated debt owed to the bank, then no benefit was derived by the bank, even if the debt would not otherwise have been repaid. As Toulson J stated (para 171), “it would be wrong that a customer who could not repay his overdraft should be able to pay off the overdraft with funds obtained from the bank by deceit, and then defeat the bank’s claim for deceit by saying that there had been no loss from his deceit (since the money obtained from it had been used to repay a debt that he could not have otherwise have paid)”. The repayment of the overdraft would discharge the debt owed by the customer to the bank, but it would not discharge the fraudster’s liability in damages arising from the deceit. The position was the same if the fraudster used the proceeds of his fraud to discharge a liability owed to the bank by a third party.
71. Toulson J’s approach to causation was followed by Hamblen J in Cheltenham Borough Council v Laird [2009] EWHC 1253 (QB); [2009] IRLR 621. The case concerned a claim by an employer against an employee who had obtained her position by means of a misrepresentation. The employee maintained that she had provided a wide range of benefits to the employer, which ought to be taken into account in diminution of damages. Hamblen J considered that the employee’s performance of the duties of her employment had been paid for by the employer, by way of the employee’s salary, and did not constitute a deductible benefit.
72. Toulson J’s approach to the question whether the claimant had benefited from the use of misappropriated funds was again followed in National Bank Trust v Yurov [2020] EWHC 100 (Comm). The case concerned a breach of duty by the shareholders controlling a bank, who procured that it entered into loan agreements with shell companies associated with them, without adequate security. The principal amount of the loans was not repaid, and the bank suffered a colossal loss. Bryan J rejected a contention that the bank had received a benefit, which should be set against its claim for damages, from the fact that some of the money advanced to borrowers had been used to service the debts of other borrowers, by meeting the interest payments due. The interest was something to which the bank was entitled under its agreement with the borrower in question, and for which it had given consideration. The fact that it received the interest to which it was entitled did not discharge to any extent the shareholders’ liability for their breach of duty. The only person who benefited from that application of the misappropriated money was the borrower whose liability to pay the interest was discharged.
73. Toulson J’s approach to causation was applied by the Court of Appeal in Tuke v Hood [2022] EWCA Civ 23; [2022] QB 659, when rejecting a submission that the defrauded seller of classic cars had to give credit for the “time value” of the (insufficient) money he had received from the fraudster. Andrews LJ stated (para 47):
“what Mr Tuke did, or may have done, with the cash he received after he received it, is irrelevant. If he had used the money to gamble, and had won a further £1million, those winnings would not be brought into account in the computation of damages, any more than a loss of £1million would increase the recoverable damages. The gain or loss would not be a result of the transaction but of his own independent acts and decisions.”
Andrews LJ also observed that “a dishonest wrongdoer cannot expect the court to make ‘tender presumptions’ or to exercise discretions in his favour. He must strictly establish an entitlement to the credit which he claims” (para 35).
74. The Board has set out the approaches adopted in these cases as the different ways in which the causal test was formulated can be helpful. However, the question is one of fact and degree which must be answered by considering all the relevant circumstances and forming an overall judgment as to the sufficiency of the causal connection. The objective is always to arrive at a result which is just and reasonable.
(ii) Benefits resulting from the conduct of the claimant
75. Losses which result from the claimant’s failure to take reasonable steps in response to the defendant’s wrongdoing are treated in law as having been caused by his own conduct rather than that of the defendant (see, for example, Credit Suisse Life (Bermuda) Ltd v Ivanishvili [2025] UKPC 53; [2026] AC 533, paras 101-105). By parity of reasoning, a benefit which the claimant receives by reason of taking reasonable steps in response to the wrongdoing must normally be brought into account in the computation of his loss. As the principles relating to the mitigation of loss rest, like the principles discussed earlier in relation to losses and benefits flowing from the defendant’s wrongful conduct, on the directness of the connection between the wrongdoing and the benefit or loss, it should not make any difference which analysis is adopted, if the case is one where either might arguably be apposite.
76. The mitigation principle was authoritatively explained in the case of British Westinghouse Electric and Manufacturing Co Ltd v Underground Electric Railways Co of London Ltd [1912] AC 673 (“British Westinghouse”). The claimant responded to the defendant’s breach of a contract for the sale of machinery, which proved to be defective, by purchasing other machinery from a third party, which was more efficient than that supplied by the defendant even if it had not been defective. The House of Lords held that the gains made from the greater efficiency of the replacement machinery were to be taken into account in assessing damages. Viscount Haldane LC described the relevant principle at p 689:
“when in the course of his business [the claimant] has taken action arising out of the transaction, which action has diminished his loss, the effect in actual diminution of the loss he has suffered may be taken into account …”
The Lord Chancellor added at p 690:
“The subsequent transaction, if to be taken into account, must be one arising out of the consequences of the breach and in the ordinary course of business.”
Those dicta apply equally to damages in tort, mutatis mutandis: see, for example, The Liverpool (No 2) [1963] P 64, 77–78.
77. An illustration of the mitigation principle which is helpful in the present context is the decision of the Court of Appeal in Hussey v Eels [1990] 2 QB 227. The claimants were induced to buy a bungalow by a negligent misrepresentation that it had not been subject to subsidence. In consequence, they paid more for the property than it was worth. Rather than effecting repairs, or selling the house at its market value, they sought planning permission to demolish the bungalow and erect two other buildings on the land. After repeated applications, they succeeded in obtaining permission. They then sold the land to developers, with the bungalow and the planning permission, at a profit reflecting the development value of the land. The issue was whether that profit had to be taken into account in the assessment of damages. Mustill LJ posed the question, “Did the negligence which caused the damage also cause the profit – if profit there was?” (p 241), and answered that it did not:
“It seems to me that when the plaintiffs unlocked the development value of their land they did so for their own benefit, and not as part of a continuous transaction of which the purchase of land and bungalow was the inception.”
That dictum was the source of the test applied in Needler (para 66 above). One might add that the development value was inherent in the land when the claimants purchased it; and they alone were responsible for the efforts required for it to be realised: something which they could have done for their own benefit whether or not there had been negligence on the part of the defendant. For all those reasons, it would have been unreasonable and unjust for the defendants to receive the benefit of the development value in the assessment of damages.
78. Another helpful authority on this topic is Fulton Shipping Inc of Panama v Globalia Business Travel SAU (formerly Travelplan SAU) of Spain [2017] UKSC 43; [2017] 1 WLR 2581 (“Fulton Shipping”), where the owners of a ship responded to the premature termination of a charterparty by selling the vessel, thereby avoiding a capital loss occasioned by a fall in the market. The capital value of the vessel was not brought into account in the assessment of damages. It was pointed out that the sale of the vessel was the exercise of a proprietary right which the owners could have exercised at any time irrespective of the charterparty or its termination. The capital value of the vessel was an asset which the owners had obtained for their own benefit and at their own risk when they purchased it, and nothing to do with the defendants.
(iii) Benefits received as a consequence of the claimant’s loss
79. The claimant may also receive a benefit as a result of the loss which he has suffered. Examples would include the receipt of unemployment benefits by a claimant who has been wrongfully dismissed, or the receipt of the proceeds of an accident insurance policy by a claimant who has been injured in an accident. Prima facie, the fact that the benefit has offset or alleviated the loss which the claimant has sustained means that it should be taken into account in the assessment of damages. To disregard it would prima facie result in the claimant being over-compensated, since he would be found entitled to damages in respect of loss which had already been offset by that amount. Accordingly, there is a general principle, subject to exceptions, that benefits received by the claimant in consequence of the loss which he has suffered are deducted in the assessment of damages.
80. In relation to the exceptions to the general rule, an appropriate starting point is the decision in Bradburn v Great Western Railway Co (1874) LR 10 Ex 1, where the Court of Exchequer Chamber rejected a contention that damages for injuries caused by the defendant’s negligence should be reduced by the amount which the claimant had received under an accident insurance policy which he had taken out prior to the accident. The case is notable as an early example, the correctness of which has never been doubted, of the court’s refusing to reduce damages by reason of a benefit which was received in consequence of the loss which the claimant had suffered, and which offset that loss. Another relatively early example is the decision in Redpath v Belfast and County Down Railway [1947] NI 167, where it was held that payments made from a charitable fund to the victims of a railway accident should be left out of account in the assessment of damages.
81. Those cases were considered in Parry v Cleaver [1970] AC 1, where the question was whether a disability pension, received by the claimant from a police pension fund, should be taken into account in the assessment of damages for personal injuries. The House of Lords held by a majority that it should not. Lord Reid observed that, in deciding whether sums received as a result of the accident should be deducted from the claimant’s losses, there was no universal rule: the common law had “treated this matter as depending on justice, reasonableness and public policy” (p 13).
82. Lord Reid went on to consider the two main categories of case which had been the subject of earlier decisions: voluntary payments made for reasons of benevolence, and insurance payments. He rejected a causal explanation of decisions concerned with voluntary payments made in order to alleviate the claimant’s loss, such as the payments in Redpath v Belfast and County Down Railway. He considered the real explanation to be that “[i]t would be revolting to the ordinary man's sense of justice, and therefore contrary to public policy, that the sufferer should have his damages reduced so that he would gain nothing from the benevolence of his friends or relations or of the public at large, and that the only gainer would be the wrongdoer” (p 14). Similarly, Lord Reid described a causal explanation for treating insurance payments as non-deductible, as in Bradburn v Great Western Railway Co, as “artificial and unreal” (ibid). The real reason for disregarding such payments, where the policy was taken out by the claimant, was that “it would be unjust and unreasonable to hold that the money which he prudently spent on premiums and the benefit from it should enure to the benefit of the tortfeasor” (ibid).
83. Lord Reid applied the same reasoning to the disability pension in the case before the House, and rejected an argument that the pension should be taken into account because it came from the same source as the wages of which the claimant had been deprived (p 15):
“This seems to be founded on an idea of remoteness which is, I think, misconceived. Remoteness from the defendant’s point of view is a familiar conception in connection with damages. He pays damages for loss of a kind which he might have foreseen but not for loss of a kind which was not foreseeable by him. But here we are not dealing with that kind of remoteness. No one has ever suggested that the defendant gets the benefit of receipts by the plaintiff after his accident if they are of a kind which he could have foreseen, but not if they are of a kind which he could not have foreseen, or vice versa … Surely the distinction between receipts which must be brought into account and those which must not must depend not on their source but on their intrinsic nature.”
Lord Reid went on to reiterate that “public policy must enter largely into our decision” (p 19), and based his conclusion on the view that “a decision that pensions should not be brought into account in assessing damages at common law is consistent with general principles, with the preponderating weight of authority, and with public policy as enacted by Parliament…” (p 20).
84. The speeches of Lord Pearce and Lord Wilberforce were to similar effect. Lord Pearce rejected a causal analysis at pp 33-34, and like Lord Reid referred to “fairness and justice and public policy” (p 37). He emphasised what he described as the character of the pension rights in question: “they are intended by payer and payee to benefit the workman and not to be a subvention for wrongdoers who will cause him damage” (ibid). Lord Wilberforce emphasised at p 40 the difference between the computation of the claimant’s loss and “the totally different question whether, after the loss has been so computed, any gain or benefit is to be deducted against it”. He also observed at pp 41-42 that it was “impossible to devise a principle so general as to be capable of covering the great variety of benefits from one source or another which may come to an injured man after, or because, he has met with an accident”.
85. It is apparent from Parry v Cleaver that considerations of justice, fairness and public policy lie at the heart of this area of the law. The point is also illustrated by Palatine Graphic Arts Co v Liverpool City Council [1986] QB 335, which concerned statutory compensation for disturbance, assessed on the same basis as common law damages. The claimant firm had been required to relocate because of the compulsory purchase of its former premises. It received a government grant as an inducement to relocate to premises in Liverpool. Applying Parry v Cleaver, the Court of Appeal held that it would be contrary to public policy to take the grant into account, since that would be inconsistent with the policy of the legislation under which it had been paid.
86. This area of the law was reviewed in Hodgson v Trapp [1989] AC 807. The question in the case was whether attendance and mobility allowances paid to the claimant under social security legislation should be taken into account in the assessment of damages for personal injuries, where the allowances were paid in consequence of the injuries which the claimant had sustained. The House of Lords held that they should be taken into account, applying a general rule which Lord Bridge of Harwich explained at p 819, in a speech with which the other members of the Appellate Committee agreed:
“If, in consequence of the injuries sustained, the plaintiff has enjoyed receipts to which he would not otherwise have been entitled, prima facie, those receipts are to be set against the aggregate of the plaintiff’s losses and expenses in arriving at the measure of his damages.”
87. Lord Bridge acknowledged that there were exceptions to that general rule. Classic examples were, first, moneys accruing to the claimant under policies of insurance for which he had paid the premiums (Bradburn v Great Western Railway Co), and secondly, moneys received by the claimant from the bounty or benevolence of third parties motivated by sympathy for his misfortune (Redpath v Belfast and County Down Railway). Lord Bridge observed that the reasoning relied on by courts in support of other exceptions “has, I think, invariably been based on the application to a greater or lesser degree by analogy of the same reasons as are thought to justify the primary exceptions” (p 820). He added (ibid):
“The difficulty, which has been widely recognised, is to articulate a single precise jurisprudential principle by which to distinguish the deductible from the non-deductible receipt. As Lord Reid said in Parry v Cleaver [1970] AC 1, 13: ‘The common law has treated this matter as one depending on justice, reasonableness and public policy.’”
88. Lord Bridge concluded that, in the end, “the issue in these cases is not so much one of statutory construction as of public policy” (p 823). Approaching the question on that basis, Lord Bridge noted that the statutory benefits and the damages claimed were designed to meet the same costs, and observed that to allow double recovery in such a case at the expense of both taxpayers and insurers seemed incapable of justification on any rational ground (p 823).
89. The foregoing cases were all concerned with damages for personal injuries, but a similar approach has been adopted in other contexts. The point is illustrated by the case of Hamilton-Jones v David & Snape [2003] EWHC 3147 (Ch); [2004] 1 WLR 924, where the claimant sued her solicitor for negligence in allowing her children to be removed by their father to Tunisia. The question arose whether her claim for damages in respect of the costs incurred in visiting the children in Tunisia should be reduced by amounts which her late mother had contributed. Neuberger J held that no such deduction should be made. Referring to Lord Reid’s explanation in Parry v Cleaver of the decision in Redpath v Belfast and County Down Railway, cited in para 82 above, Neuberger J said at para 74:
“I can see no reason, in logic or policy, as to why the principle should be limited to personal injury claims. On the contrary … There is no reason why payments made to the claimant by third parties, particularly when those payments are made out of natural love and affection for the claimant, should be credited to the benefit of the person whose negligence has harmed the claimant, and to the disadvantage of the claimant. Indeed, it would seem almost absurd if that were the law.”
90. In some cases where the claimant receives damages in respect of loss which has been offset by a benefit received from a third party, the damages may be held on trust for the third party, thus avoiding double recovery: see, for example, Hunt v Severs [1994] 2 AC 350. However, that consideration does not affect the damages for which the defendant is liable.
(iv) The authorities on deductibility relied on in the present case
91. Turning, in the light of this discussion, to the authorities cited by the courts below, the principal authority relied on was the judgment of Lord Sumption in Tiuta, citing his own earlier judgment in Swynson. In Swynson the claimant, acting on negligent advice from the defendant, made a number of loans to a borrower, which the borrower was unable to repay. The claimant then sought damages from the defendant for the loss resulting from the non-payment of the loans. By the time damages came to be assessed, some of the loans had been repaid. The Supreme Court held that the claimant’s loss was thereby reduced. That was an entirely unsurprising result. The claimant was seeking compensation for the loss arising from the borrower’s failure to repay the loans. The repayment of some of the loans meant that there was no loss in respect of those loans.
92. However, it was argued in Swynson that a loan made by a third party to the borrower, out of which it repaid the loans made by the claimant, was res inter alios acta, with the result that the repayment of the loans did not reduce the damages payable by the defendant. This was confused thinking. The case did not concern a benefit received by the claimant which might or might not be taken into account in the assessment of damages, depending on considerations of the kind discussed above. The repayment of the loans simply meant that there was no loss caused by the defendant’s wrongdoing, so far as those loans were concerned. That was explained by Lord Sumption at para 13:
“the payments made by [the third party] to [the borrower] and by [the borrower] to Swynson to pay off the 2006 and 2007 loans could not possibly be regarded as collateral. In the first place, the transaction discharged the very liability whose existence represented Swynson’s loss.”
93. The case was not, therefore, one which required a detailed analysis of the distinction between deductible benefits and collateral benefits. However, Lord Sumption considered the matter at para 11 of his judgment. He began by stating a general rule, to which there was said to be one exception:
“The general rule is that loss which has been avoided is not recoverable as damages, although expense reasonably incurred in avoiding it may be recoverable as costs of mitigation. To this there is an exception for collateral payments (res inter alios acta), which the law treats as not making good the claimant’s loss.”
94. Lord Sumption went on to acknowledge (ibid) that “[i]t is difficult to identify a single principle underlying every case” in which the law treats a benefit as collateral, and therefore to be disregarded in the assessment of damages. He continued (ibid):
“Broadly speaking, collateral benefits are those whose receipt arose independently of the circumstances giving rise to the loss. Thus a gift received by the claimant, even if occasioned by his loss, is regarded as independent of the loss because its gratuitous character means that there is no causal relationship between them. The same is true of a benefit received by right from a third party in respect of the loss, but for which the claimant has given a consideration independent of the legal relationship with the defendant from which the loss arose. Classic cases include loss payments under an indemnity insurance: Bradburn v Great Western Railway Co (1874) LR 10 Ex 1. Or disability pensions under a contributory scheme: Parry v Cleaver [1970] AC 1. In cases such as these, as between the claimant and the wrongdoer, the law treats the receipt of the benefit as tantamount to the claimant making good the loss from his own resources, because they are attributable to his premiums, his contributions or his work.”
Lord Sumption added that the position “may be different if the benefits are not collateral because they are derived from a contract (say, an insurance policy) made for the benefit of the wrongdoer … [o]r … from steps taken by the claimant in consequence of the breach, which mitigated his loss” (ibid).
95. These obiter remarks were expressly prefaced by the words “broadly speaking”, and by an acknowledgment that it is difficult to identify a single principle underlying every case. They do not, therefore, purport to be a comprehensive account of the relevant legal principles. They have, however, sometimes been treated as if they contain all that is necessary to decide difficult cases. It may be helpful to explain why they should not be treated in that way: why, indeed, they risk over-simplifying the position.
96. The description of collateral benefits as “those whose receipt arose independently of the circumstances giving rise to the loss” is broadly apt to distinguish non-deductible from deductible benefits, so far as concerns benefits resulting from the wrongful conduct of the defendant. As has been explained, the question in relation to benefits in that category, put shortly, is whether the receipt of the benefit was not merely a result of the defendant’s conduct in a factual or “but for” sense, but was sufficiently closely connected to it to be treated in law as being the result of the defendant’s wrongdoing. A benefit whose receipt arose independently of the circumstances giving rise to the loss would be the converse of such a case. In relation to benefits resulting from the claimant’s conduct, the relevant test, put shortly, is whether the claimant acted in the way he could reasonably be expected to act in the circumstances in which he was placed by the breach. Again, a benefit whose receipt arose independently of those circumstances is the converse of such a case.
97. The situation is different in relation to benefits received by the claimant in consequence of his loss. The relevant test in relation to the deductibility of such benefits is not one of causation. Far from Parry v Cleaver and Bradburn v Great Western Railway Co being classic cases where there was no causal relationship between the benefit and the loss, it was made clear in Parry v Cleaver that those decisions are more convincingly justified on other grounds. As has been explained, the focus since Parry v Cleaver has been on the character of the benefit in question, as Lord Sumption recognised earlier in para 11 of his judgment: “the critical factor is not the source of the benefit in a third party but its character”. In cases concerned with insurance or voluntary payments, such as Bradburn v Great Western Railway Co and Redpath v Belfast and County Down Railway, a particularly important factor has been the intention with which the benefit was conferred (see, for a fuller discussion, Gatwick Investment Ltd v Liberty Mutual Insurance Europe SE [2026] UKSC 14; [2026] 1 WLR 2006).
98. The subsequent case of Tiuta concerned a claim against a valuer for negligence in the valuation of security for a loan advanced by the claimant to a borrower. In accordance with the terms on which the loan was advanced, part of it was used to repay an earlier loan which the claimant had advanced to the borrower on the same security. Since the repayment of the first loan was a condition of the advance, it followed that the amount being advanced in reliance on the negligent valuation was in reality the amount by which the second loan exceeded the first. That was therefore the amount of the loss suffered by the claimant as a consequence of the valuer’s negligence.
99. Nevertheless, it was argued that the borrower’s use of the second loan to repay the first loan should be disregarded as a collateral benefit to the claimant. As in Swynson, this was confused thinking. As Lord Sumption said at para 13, the discharge of the first loan out of the money advanced under the second loan “was plainly not a collateral benefit” in the sense in which that expression is used in the law. Lord Sumption pointed out the similarity of the case to Swynson, and repeated at para 12 the explanation of collateral benefits which he had given in the earlier case:
“The general rule is that where the claimant has received some benefit attributable to the events which caused his loss, it must be taken into account when assessing damages, unless it is collateral. In Swynson Ltd v Lowick Rose LLP [2017] UKSC 32; [2018] AC 313, para 11, it was held that as a general rule ‘collateral benefits are those whose receipt arose independently of the circumstances giving rise to the loss’. Leaving aside purely benevolent benefits, the paradigm cases are benefits under distinct agreements for which the claimant has given consideration independent of the relevant legal relationship with the defendant, for example insurance receipts or disability benefits under contributory pension schemes.”
100. Similar comments apply as in relation to Lord Sumption’s dictum in Swynson. The “general rule” that a benefit received by the claimant is deductible if it is “attributable to the events which caused his loss” is a fair summary of the position in relation to benefits resulting from the defendant’s wrongdoing. It also reflects the position in relation to benefits resulting from the claimant’s conduct in response to the wrongdoing, at a high level of generality. But the later part of the passage is more difficult to accept, and, since the case before the court in Tiuta did not concern situations of that kind, was obiter dictum. Far from insurance receipts and disability benefits under contributory pension schemes being paradigm cases of benefits arising independently of the circumstances giving rise to the loss, it has been accepted since Parry v Cleaver that their non-deductibility cannot be satisfactorily explained on that basis. Nor can it be said, as a general rule, that benefits are not deductible if they are “benefits under distinct agreements for which the claimant has given consideration independent of the relevant legal relationship with the defendant”, since that is often true of benefits resulting from the claimant’s conduct which are properly taken into account, such as the wages earned by a wrongfully dismissed employee who finds alternative employment.
(c) Damages in the present case
(i) Was there a benefit?
101. The first question which arises is whether EIC received a benefit from the payments made to it by EHL when it subscribed for shares. As explained earlier, EHL made repeated payments to EIC over the course of several years in return for the issue of additional shares, in order to improve EIC’s margin of solvency, which had been diminished by the wrongful diversion of its moneys to EIG. The EHL payments enabled EIC to continue trading, thereby earning additional premium income but also incurring additional liabilities. Some of the additional premium income was diverted to EIG by means of further sham transactions, and there was in addition a failure by the defendant to maintain adequate reserves to meet EIC’s liabilities under the policies it had written. It has been established that the defendant is in principle liable to the claimant for the wrongful diversion of EIC’s moneys to EIG, and for the failure to make adequate provision for liabilities.
102. Against that background, two points have been made in relation to the question whether EIC benefited from the payments. First, EIC gave consideration for the payments it received, by providing shares in return. Secondly, since the EHL payments resulted in EIC incurring additional liabilities on policies for which inadequate provision had been made, and enabled the defendant to arrange further wrongful payments to EIG, their result was to increase EIC’s losses.
103. The Board is not persuaded that the fact that consideration was given for the payments in itself refutes the contention that a benefit was received. Certainly, the additional shares which EIC issued to EHL in return for the payments were items of property, and their issue resulted in the creation of rights and obligations, including EIC’s obligation to maintain its share capital. The shares undoubtedly constituted consideration for the amounts paid for the subscription. However, the relevant question in the present context is not whether EIC gave consideration for the payments it received from EHL, but whether EIC received a benefit. In the circumstances of this case, the new shares did not add significant value to EHL’s assets, since EHL already owned the whole of EIC’s share capital. The value of EHL’s shareholding would not have been materially different if the money which it paid to EIC had been gifted, so that the increase in EIC’s capital was reflected in the value of EHL’s existing shares, rather than being diluted by the issue of additional shares. As Lord Reid remarked in Parry v Cleaver at p 13, “in this chapter of the law we must have regard to realities rather than technicalities”. Viewed from that perspective, the allotment of the additional shares was not inconsistent with the payments constituting a benefit to EIC.
104. The contention that the EHL payments did not constitute a benefit to EIC because they resulted in EIC incurring additional claims liabilities without adequate provision, and in further sums being wrongfully diverted from EIC to EIG, is supported by the judge’s findings. As explained at para 19 above, he concluded that the circulation of moneys via EHL “was clearly not therefore in EIC’s best interests”, but “allowed it to stagger on as a business (albeit largely unprofitably) and enabled large payments to be made to [the defendant] and other ultimate beneficial owners of the group”; and that the amounts which were paid back into EIC “allowed it to keep writing further premium, for greater losses to be incurred, whilst allowing for ongoing substantial payments to be made out to [the defendant] and others along the way.”
105. Those conclusions are important because, as has been explained, the question whether EIC received a benefit is not to be decided simply on the basis that it received some payments. The damages awarded have to be such as will place EIC in the position in which it would have been if the defendant had performed his duties. The relevance of the payments to the computation of EIC’s loss therefore depends on whether, overall, EIC was better off prior to trial by reason of receiving them. The judge’s conclusions are inconsistent with that possibility. The fresh money was a disbenefit for EIC, because it resulted in more potential claims not covered by adequate provision, and it also resulted in more moneys being wrongfully diverted to EIC. At the very least, it cannot be said to have been established that EIC benefited from receiving the payments; and, as has been explained, that was a matter in relation to which the evidential burden rested on the defendant.
106. The Board’s conclusion is, therefore, that the defendant has not satisfied the evidential burden of proving that EIC received a benefit from EHL when EHL introduced further capital into EIC by purchasing additional shares in EIC. Given the existing relationship between the two companies, the fact that EIC gave EHL consideration in the form of new shares may not, of itself, have precluded that money from being regarded as a benefit received by it from EHL. But the judge’s findings were clear that EIC was not better off by having received that money and it was not therefore a benefit to EIC in the relevant sense.
(ii) Was the benefit deductible
107. Given that conclusion, the question whether any benefit was deductible does not arise. However, the Board will address it on the hypothesis that a benefit was received, as the differing conclusions of the courts below indicate that some discussion of the issues may be helpful.
108. Both parties, and the courts below, have focused on the question whether the benefits said to have been received by EIC, in the form of the payments made by EHL for share subscriptions, should be regarded as the result of the defendant’s wrongful conduct. As has been explained, the defendant caused EIC to enter into a series of transactions with EIG, under which EIC paid EIG for non-existent services. Each time EIC was induced to enter into such a transaction, it transferred moneys belonging to it to EIG for no consideration, and the defendant thereupon incurred a liability to EIC in damages. EHL responded to EIC’s consequent under-capitalisation by providing it with financial support over a period of years, during which the defendant’s misconduct continued.
109. The Court of Appeal treated the defendant’s wrongful conduct as constituting, in law, the cause of EHL’s payments to EIC in return for the allocation of additional shares. As explained above, Elias JA considered that the judge had erred in treating EHL’s purchases of the shares as separate transactions from the transactions between EIC and EIG which caused the loss. Given the triangular model, there was in Elias JA’s view a close link between the payments made by EIC to EIG, those made by EIG to EHL, and those made by EHL to EIC. In that connection, he referred to the judge’s statement (para 863) that “the prospect of capitalising EIC from the [wrongful] payments was on Mr Flowers’ radar from the outset”. These were not, therefore, unrelated transactions; in his view, they were, to use Lord Sumption’s words in Tiuta, payments which were “attributable to the events which caused his loss.” The Court of Appeal therefore concluded that the capital payments from EHL to EIC were deductible from the damages payable by the defendant to EIC.
110. On the other hand, the claimant argues that the payments made by EHL were not sufficiently closely linked to the defendant’s wrongful diversion of EIC’s funds to EIG, particularly since EHL took independent commercial decisions which were not part of the defendant’s dishonest scheme, and the larger part of the amount which EHL paid to EIC was borrowed from third parties.
111. It is clear that a chain of transactions intervened between the defendant’s wrongdoing in causing EIC to make unjustified payments to EIG, and EIC’s receipt of payments from EHL in return for the issue of additional shares. Although there was little evidence concerning EIG’s decision-making, and it is not as clear as it might have been whether EIG’s payment of the dividends to EHL followed naturally from its receipt of EIC’s funds, that appears to be a reasonable inference from what occurred. On that basis, EHL’s receipt of the dividends could be regarded as flowing from the defendant’s wrongdoing.
112. EHL then had to decide how to use the dividends it received. The judge found that EHL was not party to the defendant’s scheme, but was acting in good faith. Its adherence to the triangular model meant that it could be expected to use its resources to support EIC, at least up to a point: there could presumably come a point at which it would be commercially unwise to continue to do so, given EIC’s persistent financial problems. EHL therefore had to take a series of independent decisions: as to whether to continue to support EIC; if it decided to do so, how much to subscribe for additional shares; and whether to fund such subscriptions out of its own resources or out of borrowings on the financial markets. In the event, most of the support which it gave to EIC was funded by borrowings from third parties, who purchased the bonds which EHL issued, and who are now EHL’s creditors.
113. The different views taken by the judge (albeit without detailed consideration of the point) and by the Court of Appeal as to whether the defendant’s wrongdoing was, in law, the cause of the payments funded out of the dividends which EHL received – in effect, the recirculation of part of the moneys wrongly taken from EIC – illustrate how finely the issue is balanced on the particular facts of this case. On a second appeal, the Board would be slow to interfere with the conclusion of a lower court on a matter of this kind, turning as it does on an evaluation of the facts. On the other hand, the position is less equivocal in relation to the payments made by EHL which were funded out of its borrowings on the financial markets. Although the Court of Appeal drew no distinction between those payments and the payments made out of EHL’s own resources, it appears to the Board to be much more difficult to regard the former as being directly attributable to the events causing EIC’s loss. EHL’s issue of the bonds, and its investment of the proceeds of the bond issues in EIC, were in the Board’s judgment clearly distinct from the defendant’s wrongdoing: an independent response to the consequences of that wrongdoing, rather than something for which responsibility can fairly and reasonably be attributed to the defendant. It is true that EHL relied on the dividends it received from EIG to pay the interest due on the bonds; but in so far as that might have constituted a benefit caused by the defendant’s wrongdoing, it was a benefit to EHL, discharging its liability to the bondholders, rather than discharging the defendant’s liability to EIC.
114. There is, however, a more fundamental reason why the payments made by EHL, on the hypothesis that they constituted a benefit, should not be regarded as diminishing the defendant’s liability to EIC in damages. The argument in support of a close causal connection between his wrongful conduct and the EHL payments, and therefore in support of treating those payments as a deductible benefit, is essentially that the defendant reasonably anticipated that EHL would inject further capital into EIC, with the consequence that the circulation of misappropriated funds would enable his wrongful conduct to continue. However, when EHL decided to subscribe for additional shares in EIC, its intention was to increase EIC’s working capital so as to enable it to continue trading. It had no intention to discharge a dishonest director of his existing liabilities to EIC, or to shield him from liability for future wrongdoing.
115. In those circumstances, it would be unjust and unreasonable if the payments made by EIC’s parent company in order to support it in the face of its financial difficulties were to be treated as diminishing the defendant’s liability for the wrongdoing which had brought about those very problems. Although the present case is different in some respects from the cases concerned with voluntary payments or other forms of gratuitous assistance by third parties which were discussed earlier, such as Redpath v Belfast and County Down Railway and Hamilton-Jones v David & Snape, there is an analogy. EHL’s payments also were effectively a gratuitous payment to the victim of loss; and it is just as important in the present case as in those earlier cases that the law should avoid a result which would be unjust and unreasonable, and therefore contrary to public policy.
116. The Board concludes, therefore, that the payments made by EHL to EIC should not be taken into account in the assessment of the damages for which the defendant is liable. It has not been established that EIC benefited from the payments, and any such benefit would not properly have been deductible in any event.
(2) Equitable compensation
117. Since the defendant was in dishonest breach of his fiduciary duty to EIC in relation to the payments made to EIG, the claimant has remedies in equity as well as at common law. The remedy sought is equitable compensation, and the arguments on this aspect of the case have been presented on the basis that the relevant inquiry is as to the extent of the loss sustained by EIC as a result of the defendant’s breach of duty, assessed in accordance with the decisions in Target Holdings and AIB v Redler.
118. The Board has heard no argument as to the relationship between those decisions and other authorities supporting the view that a director in breach of fiduciary duty is liable to make restitution of the moneys wrongfully paid out. For example, in Revenue and Customs Comrs v Holland [2010] UKSC 51; [2010] 1 WLR 2793, Lord Hope said at paras 48-49:
“Then there is the question whether the correct remedy for any breach of the duties of a director not to make unlawful payments of dividends is damages or equitable compensation for the net loss sustained by the company, or restitution or restoration of the amount of the unlawful dividends without any inquiry into the loss sustained … I agree with the Court of Appeal that the obligation is to restore the moneys wrongfully paid out. This … is the established remedy. Where dividends have been paid unlawfully, the directors’ obligation is to account to the company for the full amount of those dividends…”
See also paras 124 (Lord Walker) and 146 (Lord Clarke).
119. More recently, in Mitchell v Al Jaber [2025] UKSC 43; [2026] AC 758, Lord Hodge, Lord Briggs and Lord Sales, giving the judgment of the Board, said at para 93:
“Where a trustee misappropriates trust property or (as here) a fiduciary misappropriates property under his management and control, then there is little doubt as to the general objective of a court of equity in awarding compensation to the beneficiary (or the principal: here, a company) if the misappropriated property cannot be returned in specie … It is to restore to the trust fund at the expense of the defaulting trustee or fiduciary (or to the company where its property is misappropriated by a director) the value of the property misappropriated …”
The decision in Target Holdings was explained at paras 98–99 on the basis that it was not a case in which trust property had been misappropriated for the trustee’s benefit, but one where the commercial object of the trust had been fulfilled notwithstanding the default of the defendant solicitor. The decision in AIB v Redler was explained at para 100 on the basis that it was another case of negligent failure by solicitors acting for a lender to do their job properly, where the client’s loss had been assessed on a just and equitable basis.
120. In the light of those authorities, there is a question as to the relevance of Target Holdings and AIB v Redler to the present case. That is not, however, a question on which the Board has been addressed. It should only be noted that the argument not advanced, that the defendant is liable to reimburse the money wrongfully misappropriated, would plainly result in the share subscriptions by EHL being disregarded.
121. Reliance is placed on Target Holdings because of Lord Browne-Wilkinson’s observations as to certain common features of equitable compensation and damages at common law. As Lord Browne-Wilkinson acknowledged at p 434, there are also some differences between the assessment of compensation for the breach of equitable obligations and the assessment of damages for breaches of common law duties: a point also illustrated by more recent authorities such as Recovery Partners GP Ltd v Rukhadze [2025] UKSC 10; [2026] AC 209 and Mitchell v Al Jaber, para 97.
122. Against that background, the question in the present case is whether, in quantifying the amount of equitable compensation for the defendant’s dishonest breach of his fiduciary duty to EIC, credit should be given for the payments made to EIC by EHL for the share subscriptions. It has been explained that no such credit would be given when assessing common law damages. No reason has been suggested why the relevant principles of equity should be any more favourable to a delinquent fiduciary than those adopted by the common law. The Board therefore concludes that in equity, as well as at common law, the payments made by EHL to EIC should not be taken into account in the assessment of compensation.
7. Should clause 11.4 of the Icebreaker policy be interpreted as meaning that EIC carried no insurance risk in respect of the Icebreaker scheme?
123. Clause 11.4 of the Icebreaker policy provides:
“The Insurer shall have no liability under this Policy if the Insurer reasonably believes, and has reasonable grounds for such belief, that the LLP and the Insured have colluded in an attempt to engineer the failure of the LLP leading to insufficient income being generated causing a claim to arise on this Policy.”
124. As the Court of Appeal noted, an investor in the Icebreaker scheme could not affect the way in which the scheme operated: there was no room for individual negotiation over the structure of the scheme itself, nor over how the funds were invested by the LLP. An investor cannot readily be said to have been “colluding” with the LLP merely by investing in the scheme on the terms offered. Nor can investment in the scheme be described as “an attempt to engineer the failure of the LLP”. Whether the LLP succeeded or failed depended on whether the intellectual property in which it invested could be profitably exploited; and the investor had no control over the LLP’s selection of the intellectual property in which it chose to invest. Accordingly, even if the LLP turned out to be financially unsuccessful, that was not something engineered by the investor. Nor was there any evidence that any policy holder took any decision open to them which was designed to engineer the failure of the LLP they joined.
125. The principal argument before this court, as before the Court of Appeal, was that the scheme inherently involved “collusion” in “an attempt to engineer the failure of the LLP”, because of a misrepresentation that the 80% of the investor’s contribution which was borrowed was used for the purposes of the trade, when in reality it was not. However, it has been found as a fact that there was no such misrepresentation, either to the Revenue or to EIC. The true nature of the scheme was transparent and was well understood. Furthermore, even if such a misrepresentation had been made, it would not amount to collusion between the investor and the LLP in an attempt to engineer the failure of the LLP, although it might raise other issues which have not been pleaded, such as tax evasion and the effect of misrepresentation on the contracts entered into by EIC.
126. In these circumstances, there were no grounds on which EIC could reasonably believe that there had been any relevant collusion or any attempt to engineer the failure of the LLP.
127. If there were any doubt about the matter, it is a well-established principle of insurance law that an exclusion clause “must be construed in a manner which is consistent with and not repugnant to the purpose of the insurance contract”: Impact Funding Solutions Ltd v Barrington Support Services Ltd [2016] UKSC 57; [2017] AC 73, para 7, per Lord Hodge. To treat the act of entering into the scheme as an act of collusion of a kind which fell within the scope of the exclusion clause, so that anyone investing in the scheme would automatically be excluded from benefiting under the insurance policy designed specifically for the protection of such investors, would not merely be repugnant to the purpose of the policy, but would negate its purpose entirely. A court would be very slow to reach that conclusion, and there is no basis for doing so in the present case.
8. Were the Icebreaker policies void or unenforceable for reasons of public policy?
128. Tax avoidance schemes are not unlawful. Nor are insurance contracts related to such schemes. There is therefore no basis on which issues might arise of the kind which were discussed in Patel v Mirza and more recent authorities concerned with the consequences of illegality. Nor is there any trace of a common law doctrine that arrangements designed to avoid tax are void as being contrary to public policy. On the contrary, as Lord Hoffmann observed in Norglen Ltd v Reeds Rains Prudential Ltd [1999] 2 AC 1 (“Norglen”), at pp 13-14:
“If the question is whether a given transaction is such as to attract a statutory benefit, such as a grant or assistance like legal aid, or a statutory burden, such as income tax, I do not think that it promotes clarity of thought to use terms like stratagem or device. The question is simply whether upon its true construction, the statute applies to the transaction. Tax avoidance schemes are perhaps the best example. They either work (Inland Revenue Comrs v Duke of Westminster [1936] AC 1) or they do not (Furniss v Dawson [1984] AC 474). If they do not work, the reason, as my noble and learned friend, Lord Steyn, pointed out in Inland Revenue Comrs v McGuckian [1997] 1 WLR 991, 1000, is simply that upon the true construction of the statute, the transaction which was designed to avoid the charge to tax actually comes within it. It is not that the statute has a penumbral spirit which strikes down devices or stratagems designed to avoid its terms or exploit its loopholes. There is no need for such spooky jurisprudence.” (Emphasis added.)
129. There have been two developments in the law which have made it more difficult in practice for some kinds of tax avoidance scheme to succeed; but they have not involved any departure from the fundamental principle, stated by Lord Hoffmann, that the question is simply whether upon its true construction the statute applies to the transaction.
130. The first development concerns statutory construction. As the House of Lords explained in Barclays Mercantile Business Finance Ltd v Mawson [2004] UKHL 51; [2005] 1 AC 684, the modern approach to interpretation is to have regard to the purpose of a particular provision and to construe its language, so far as possible, in the way which best gives effect to that purpose. However, until WT Ramsay Ltd v Inland Revenue Comrs [1982] AC 300, the interpretation of tax legislation was based predominantly on a linguistic analysis. Equally importantly, the courts treated every element of a composite transaction which had an individual legal identity (such as a payment of money, a transfer of property, or the creation of a debt) as having its own separate tax consequences, whatever might be the terms of the statute. As Lord Steyn said in Inland Revenue Comrs v McGuckian [1997] 1 WLR 991, 999, in combination those two features – a literal interpretation of tax statutes, and an insistence on applying the legislation separately to the individual steps in composite schemes – allowed tax avoidance schemes to flourish to the detriment of the general body of taxpayers.
131. The significance of the Ramsay case was to do away with both those features. First, it extended to tax cases the purposive approach to statutory construction which was orthodox in other areas of the law. Secondly, it established that the tax analysis of the transactions in question depended on the relevant statutory provisions, purposively construed. The resultant position was summarised in the Barclays Mercantile case at para 32:
“The essence of the new approach was to give the statutory provision a purposive construction in order to determine the nature of the transaction to which it was intended to apply and then to decide whether the actual transaction (which might involve considering the overall effect of a number of elements intended to operate together) answered to the statutory description … As Lord Nicholls of Birkenhead said in MacNiven v Westmoreland Investments Ltd [2003] 1 AC 311, 320, para 8: ‘The paramount question always is one of interpretation of the particular statutory provision and its application to the facts of the case.’”
132. The “paramount question” described in the last sentence of that passage remains the same as that identified by Lord Hoffmann in Norglen. Accordingly, although the adoption of a purposive approach to statutory construction has made it more difficult to design successful tax avoidance schemes, that is not because of any departure from the fundamental principle, but because the method of analysing whether a scheme falls within the statutory provision has changed.
133. The second development has been the adoption by legislatures in some jurisdictions, including the UK, of a more vigorous approach to addressing certain forms of tax avoidance. Part 5 of the Finance Act 2013 introduced the general anti-abuse rule “for the purpose of counteracting tax advantages arising from arrangements that are abusive”: section 206(1). There are also many targeted anti-avoidance rules, including one which is directly relevant to the Icebreaker scheme. Section 74ZA of the Income Tax Act 2007, introduced by the Finance Act 2010, has the effect that sideways loss relief is not available if the loss results from relevant tax avoidance arrangements, which are defined to include arrangements “the main purpose, or one of the main purposes, of which is the obtaining of a reduction in tax liability by means of sideways relief”. In Seven Individuals this was one of the grounds for denying the tax relief to one of the appellants. It would no doubt have caught other participants in the Icebreaker scheme whose tax affairs were governed by UK law if their investments had been made after the date from which section 74ZA took effect.
134. An important feature of these legislative provisions is that they have no effect on the lawfulness, validity or enforceability of the arrangements to which they apply. Their only effect is that the arrangements do not attract the tax benefit they were designed to obtain.
135. As evidence that the common law is setting its face against tax avoidance schemes, the defendant cited some observations of Lord Walker in Pitt v Holt; Futter v Futter [2013] UKSC 26; [2013] 2 AC 108, para 135:
“In Futter this court declined to permit the appellants to raise for the first time the issue of mistake, primarily because there was no sufficient evidential basis for considering that issue for the first time on a second appeal … Had mistake been raised in Futter there would have been an issue of some importance as to whether the court should assist in extricating claimants from a tax-avoidance scheme which had gone wrong. The scheme adopted by Mr Futter was by no means at the extreme of artificiality … but it was hardly an exercise in good citizenship. In some cases of artificial tax avoidance the court might think it right to refuse relief, either on the ground that such claimants, acting on supposedly expert advice, must be taken to have accepted the risk that the scheme would prove ineffective, or on the ground that discretionary relief should be refused on grounds of public policy. Since the seminal decision of the House of Lords in WT Ramsay Ltd v IRC [1980] AC 300 there has been an increasingly strong and general recognition that artificial tax avoidance is a social evil which puts an unfair burden on the shoulders of those who do not adopt such measures. But it is unnecessary to consider that further on these appeals”.
136. It is necessary to place these remarks in context. The appeals in Pitt v Holt and Futter v Futter were not concerned with the validity of tax avoidance arrangements at common law, but with the question whether equitable relief should be granted where, as the result of a mistake, arrangements resulted in an unforeseen tax liability.
137. In Pitt v Holt an award of damages to the victim of a serious accident was settled by his receiver on a discretionary trust. The trust could have been established without any immediate inheritance tax liability if it had complied with statutory requirements intended to apply to trusts of that kind. However, those requirements were not complied with because, when executing the settlement, the receiver’s advisers failed to consider the question of liability for inheritance tax upon the transfer of the assets into a discretionary trust. As a result, large inheritance tax liabilities arose. The first question for the court was whether the receiver’s exercise of her powers should be set aside on the ground that she had acted in breach of her fiduciary duty by failing to give proper consideration to a relevant matter. The court held that since the receiver had obtained and acted on professional advice, she was not in breach of her fiduciary duty merely because the advice turned out to be wrong. The second question was whether the settlement should be rescinded on the ground of mistake. The court held that it should, on the basis that it would be unconscionable to leave the mistake uncorrected. The court based its conclusion on the fact that the receiver had made a grave mistake, and that the trust could have complied with the statutory requirements for tax relief without any artificiality or abuse.
138. The case of Futter v Futter concerned family trusts under which the father had a life interest, his wife had a reversionary life interest, and the capital was eventually to go to their children. After taking legal advice, the trustees exercised their powers under one of the trusts in such a way that the father became absolutely entitled to the fund, and they exercised their powers under the other trust so as to appoint capital to each of the children immediately. The purpose of these transactions was to transfer assets out of the trusts in such a way as to avoid incurring a charge to capital gains tax. However, the trustees’ legal advisers overlooked a statutory provision which undermined the premise on which the transactions were based. As a result, large capital gains tax liabilities arose. As in Pitt v Holt, the question for the court was whether the trustees’ exercise of their powers should be set aside because they had acted in breach of their fiduciary duty. As in Pitt v Holt, the court held that the trustees had not acted in breach of their duty, since they had acted on professional advice, even though the advice turned out to be wrong. Accordingly, their exercise of their powers could not be set aside on that ground.
139. As an alternative argument, the trustees also sought to rely on mistake as a ground for equitable rescission: the argument which succeeded in Pitt v Holt. However, as Lord Walker stated in the opening sentence of the passage quoted above, the court refused to allow the issue of mistake to be raised in the Futter case, as there was not a sufficient evidential basis for considering it for the first time on a second appeal. What followed was therefore obiter dictum. It was in that context that Lord Walker envisaged the possibility that it might be appropriate to refuse equitable relief in some cases of artificial tax avoidance. That is a far cry from suggesting that contracts forming part of tax avoidance schemes are unlawful or void as being contrary to public policy.
140. The defendant also relied on the decisions of Lang J and Coulson LJ on the application for judicial review of the determination of the FSCS, discussed at paras 34 and 35 above. Each of those decisions concerned a permission application, and was determined by a single judge without full argument. The decisions have no authority as precedents. The authorities to which the Board has referred are inconsistent with the proposition that the common law doctrine of public policy invalidates either a contract forming part of tax avoidance arrangements, or an insurance policy underwriting the risk of loss arising out of such arrangements.
141. Drawing these threads together, it can be seen that Parliament has to some extent regulated the area of tax avoidance, but it has not rendered tax avoidance arrangements unlawful. The common law has also developed, in Ramsay and subsequent cases, by applying principles of statutory construction to tax legislation which have made it more difficult for tax to be avoided by means of artificial arrangements. It is also possible that a court may refuse to exercise a discretion in equity in some cases of artificial tax avoidance, as Lord Walker suggested, applying conventional principles for the grant or withholding of equitable relief. But there is no basis for holding that even the most artificial tax avoidance schemes are unlawful, or that the relevant contracts are void as being contrary to public policy.
9. Overall conclusions
142. For the foregoing reasons, the Board concludes:
- That no deduction should be made from the damages or equitable compensation payable to EIC in respect of the payments made to it by EHL for the issue of shares.
- That clause 11.4 of the Icebreaker policy should not be interpreted as meaning that EIC carried no insurance risk in respect of the Icebreaker scheme.
- That the Icebreaker policies were not void or unenforceable for reasons of public policy.
The Board humbly advises His Majesty that the appeal should be allowed, and that the cross-appeal should be dismissed.