United Kingdom House of Lords Decisions

You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> Sempra Metals Ltd v. Revenue & Anor [2007] UKHL 34 (18 July 2007)
URL: http://www.bailii.org/uk/cases/UKHL/2007/34.html
Cite as: [2007] UKHL 34, [2007] BTC 509, [2008] 1 AC 561, [2008] Bus LR 49, [2007] 4 All ER 657, [2008] Eu LR 1, [2007] STC 1559, [2008] AC 561, [2007] 3 WLR 354

[New search] [Buy ICLR report: [2007] 3 WLR 354] [Buy ICLR report: [2008] Bus LR 49] [Buy ICLR report: [2008] 1 AC 561] [Help]

Judgments - Sempra Metals Limited (formerly Metallgesellschaft Limited) (Respondents) v. Her Majesty's Commissioners of Inland Revenue and another (Appellants)

HOUSE OF LORDS

SESSION 2006-07

[2007] UKHL 34

on appeal from: [2005] EWCA Civ 389

OPINIONS

OF THE LORDS OF APPEAL

FOR JUDGMENT IN THE CAUSE

Sempra Metals Limited (formerly Metallgesellschaft Limited) (Respondents)

v.
Her Majesty's Commissioners of Inland Revenue and another (Appellants)

Appellate Committee

Lord Hope of Craighead

Lord Nicholls of Birkenhead

Lord Scott of Foscote

Lord Walker of Gestingthorpe

Lord Mance

Counsel

Appellants:
Ian Glick QC
Rupert Baldry
Gerry Facenna
(Instructed by Solicitor's Office, HM Revenue and Customs)

Respondents:
Laurence Rabinowitz QC
Francis Fitzpatrick
Steven Elliott
(Instructed by Slaughter & May)

Hearing dates:

1 and 2 November 2006

Further heard:

16 May 2007

WEDNESDAY 18 JULY 2007

HOUSE OF LORDS

OPINIONS OF THE LORDS OF APPEAL FOR JUDGMENT

IN THE CAUSE

Sempra Metals Limited (formerly Metallgesellschaft Limited) (Respondents) v. Her Majesty's Commissioners of Inland Revenue and another (Appellants)

[2007] UKHL 34

LORD HOPE OF CRAIGHEAD

My Lords,

This is a case about the award of interest. Questions about interest usually arise where the claim is presented as ancillary to a claim for a principal sum for which the court is asked to give judgment for the recovery of a debt or as damages. Less usually they can arise where interest is sought on a principal sum which has been paid before judgment. But in this case interest is the measure of the principal sum itself.

The question is how that sum should be measured. It is agreed that the calculation of interest should be the method of measurement for the sum that is to be awarded. But the parties are at issue as to how the interest should be calculated. The choice is between simple interest and compound interest. If simple interest is used, it is agreed that it should be at the rate that is appropriate for the calculation of an award of interest under the statute. If compound interest is used, various methods of calculation are available and there is a dispute as to how it is to be calculated in this case. That issue, however, is peripheral to the important question of principle which arises on this appeal: is the claimant who seeks a remedy on the ground of unjust enrichment entitled to an award for restitution of the value of money that is measured by compound interest?

Interest: an introduction

The question of principle is much easier to state than it is to answer. But it may be helpful, before turning to the facts, to set the scene by looking briefly at the question of interest generally and then seeing how the issue in this case fits in to that wider context.

The jurisdictional routes in English law to an award of interest are to be found in statute, equity and the common law. Simple interest is available under the statute on a sum for which judgment is given for the recovery of a debt or damages or where a sum of that kind is paid before judgment: see section 35A(1) of the Supreme Court Act 1981, inserted by the Administration of Justice Act 1982, section 15(1) and Schedule 1, Part I. Interest is available in equity in cases that lie within equity's exclusive jurisdiction, especially in cases of fraud or against a trustee or other person in a fiduciary position in respect of profits improperly made. It is also available in the exercise of equity's jurisdiction in aid of rights that are enforceable at common law. In cases that lie within equity's exclusive jurisdiction compound as well as simple interest is available. As Steven Elliott, "Rethinking Interest on Withheld and Misapplied Trust Money" [2001] 65 Conv 313 puts it, when applying the inherent jurisdiction the courts have been able to craft interest awards that meet economic realities. But where equity is invoked in aid of the common law the reverse is true. In Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669 the House held, by a majority, that it would be usurping the function of Parliament if it were in equity to award compound interest in aid of the bank's common law claim for repayment of the principal sum, as the court was not authorised to award compound interest in the exercise of its common law jurisdiction under the statute.

The common law jurisdictional route is more complicated. The general rule of English common law is that the court has no power, in the absence of any agreement, to award interest as compensation for the late payment of a debt or damages: London, Chatham and Dover Railway Co v South Eastern Railway Co [1893] AC 429. The decision in that case does not fit happily with Lord Westbury's statement in Carmichael v Caledonian Railway Co (1870) 8 M (HL) 119, 131, which identified the principle that still applies in Scots law, that interest can be demanded only in virtue of a contract express or implied "or by virtue of the principal sum of money having been wrongfully withheld, and not paid on the day when it ought to have been paid." But the House felt bound to apply the law of England that was laid down by Lord Tenterden CJ in Page v Newman (1829) 9 B & C 378, 381 and endorsed by Lord Tenterden's Civil Procedure Act 1833. In 1952, however, it was recognised that loss due to late payment might be recoverable if it constituted special damage within the contemplation of the parties under the second limb of Hadley v Baxendale (1859) 9 Ex 341: Trans Trust SPRL v Danubian Trading Co Ltd [1952] 2 QB 297. This modification of the common law rule was approved in President of India v La Pintada Compania Navigacion SA [1985] AC 104.

To allow a claimant to recover special, but not general, damages for loss of the use of money is widely seen as illogical. In Hungerfords v Walker (1989) 171 CLR 125, 142 Mason CJ said that it subverted the second limb in Hadley v Baxendale from its intended purpose, which was to allow loss arising from special circumstances of which the defendant had actual knowledge in cases where the loss did not fall within the first limb because it did not arise from the ordinary course of things. The decision in London, Chatham and Dover Railway Co v South Eastern Railway Co seemed to have been based on the view that interest by way of damages was too remote: see also Trans Trust SPRL v Danubian Trading Co Ltd [1952] 2 QB 297, 306, per Denning LJ. Why then, Mason CJ asked, is the claimant not entitled to recover damages for the loss of the use of money when the loss or damage was reasonably foreseeable as liable to result from the relevant breach or tort?

The claim that is made in this case, however, is for restitution. It is presented as a claim for the time value of money by which the defendant was enriched unjustly. The claimant submits that the common law requires that it be paid a sum which represents the value of the money over the period of that enrichment, and that this sum falls to be calculated by compounding interest over that period. It has been held that in an action for money had and received the net sum only can be recovered: Moses v Macferlan (1760) 2 Burr 1005; Fruhling v Schroeder (1835) 2 Bing (NC) 78 and Johnson v The King [1904] AC 817, applying London, Chatham and Dover Railway Co v South Eastern Railway Co [1893] AC 429. But interest has been awarded at common law where restitution follows the reversal on appeal of a previously satisfied judgment: Rodger v Comptior d'Escompte de Paris (1871) LR 3 PC 465. Various other exceptions have been recognised: see Heydon v NRMA Ltd No 2 (2001) 53 NSWLR 600, 603-606, per Mason P. Furthermore the claim in this case is not for more than what was had and received by the defendant. What was had and received was the enrichment. It is the enrichment itself that is to be valued, not anything more than that.

In NEC Semi-Conductors Ltd v IRC [2006] STC 606, para 173, Mummery LJ said that the question how restitutionary relief of the kind that is sought in this case should be assessed was not settled by La Pintada, as the claim is not for an entitlement to interest, as creditors, on a debt or on damages by way of compensation for loss of the use of the money that was unjustly demanded and retained by the defendant. I respectfully agree with him, and I would approach the issue in this case from the same starting point. I would hold that it is open to your Lordships to examine this issue on the basis that the answer to it is to be found in the law of unjust enrichment. It is not foreclosed by the decisions of this House in Westdeutsche [1996] AC 669 and La Pintada [1985] AC 104, neither of which addressed the issues that arise in this case.

The ECJ's judgment

As Park J explained in the introduction to his judgment [2004] STC 1178, para 3, this case is about companies which, because they had to pay part of their mainstream corporation tax prematurely, suffered a timing disadvantage which conferred a corresponding timing advantage on the Revenue. We now know that the absence of the power to make group income elections in the case of these companies which resulted in the premature payment of corporation tax was contrary to article 52 of the EC Treaty (now article 43) which guarantees freedom of establishment: Metallgesellschaft Ltd v Inland Revenue Commissioners [2001] Ch 620. Community law requires that the companies must be provided with a remedy in domestic law which will enable them to recover a sum equal to the interest which would have been generated by the advance payments from the date of the payment of the ACT until the date on which the MCT became chargeable: para 88.

The European Court of Justice explained in para 87 of its judgment in Metallgesellschaft that the breach of Community law arose not from the payment of the tax itself but from its being levied prematurely. The purpose of the award of interest covering loss of the use of the sums paid by way of ACT is to restore the equal treatment in the levying of the tax which was guaranteed by article 52 of the Treaty. The expression "loss of use" suggests that what was primarily in contemplation was a remedy in damages. But the Court made it clear in para 81 of its judgment that it was not for it to assign a legal classification to the actions brought by the claimants in the national court to obtain this remedy:

In essence, the claim is for the time value of the money that was paid over prematurely. How that value is to be measured depends on the nature of the remedy.

In the concluding sentence of para 96 the Court recalled that, in the absence of Community rules, it is for the domestic legal system of the member state concerned to lay down the detailed procedural rules governing such actions, "including ancillary questions such as the payment of interest". That sentence must be read in the light of what the Court said in para 87 of the judgment. The claim for payment of interest covering the loss of use over time of the sums paid by way of ACT is the very objective sought in the main proceedings. It is the principal sum claimed. We are not concerned in this case with the ancillary claim under the statute for simple interest. This is not a claim for discretionary interest on a sum for which judgment is given for the recovery of a debt or damages or which is paid before judgment. Sempra accepted that the ancillary award would be made under section 35A(1) of the 1981 Act once the principal sum has been identified.

The issues

The question then is whether the calculation of the award that is required by Community law in these circumstances should be effected on the basis of compound interest as the appellants contend, or of simple interest as is contended for by the Revenue. It should be appreciated that this is the only point of substance that requires to be decided in this case. Given the decision in Metallgesellschaft, the Revenue do not now dispute liability to pay interest on the amounts paid prematurely, appropriately calculated. Moreover, for reasons which I shall explain later, there is no challenge in principle to Sempra's decision to seek to obtain this award in restitution rather than as damages.

As for the method of calculation, it was common ground before the judge that if compound interest was to be awarded it should be calculated on a conventional basis - the rate being derived from the rates of interest generally prevailing on ordinary commercial borrowings during the relevant period. But, for reasons that I shall mention later, it appears that the Revenue's statement of its position was based on a misunderstanding. Mr Glick QC accepted that, if compound interest was to be used, the rate of interest on ordinary commercial borrowings would be appropriate for an award of damages. But he said that a different approach was needed if Sempra was to be allowed to recover compound interest as a restitutionary award on the ground of unjust enrichment measured by the time value of the money that was paid prematurely.

The Court of Justice was not asked to provide an answer to the question how the principal award was to be calculated. It is highly likely however that, if it had been asked to do so in Metallgesellschaft, it would have said that its assessment was a matter for the national court, and that it would have given the same answer to the question whether it was open to the claimants to choose between the two alternative remedies. This is already implicit in its comment in para 81 that it is for the claimants, subject to the supervision of the national court, to specify the nature and basis of their actions. I agree with my noble and learned friends Lord Nicholls of Birkenhead and Lord Walker of Gestingthorpe that it would serve no useful purpose to make a further reference to the Court in these circumstances. But it became increasingly plain in the course of the argument before your Lordships that the question as to how the interest is to be calculated cannot be answered without a clear understanding of the causes of action in domestic law which are being relied upon to produce the award.

The causes of action

These causes of action were identified only in the most general terms in the second question on which the Court of Justice was asked to give a preliminary ruling in that case. They were referred to as a restitutionary right to claim a sum of money by way of interest on the ACT or, in the alternative, as a sum claimed by way of an action for damages pursuant to Community law principles. The Court gave its answer on two assumptions: first, that the actions were to be treated as claims for restitution of a charge levied in breach of Community law: para 82; and secondly, that they were to be treated as claims for compensation for damage caused by breach of Community law: para 90. On the first assumption it said that in an action for restitution the principal sum due was none other than the amount of interest which would have been generated by the sum, use of which was lost as a result of the premature levy of the tax: para 88. On the second assumption it said that an award of interest was essential if the damage caused by the breach of article 52 of the Treaty was to be repaired: para 95.

A more precise analysis of these causes of action is now needed in view of the problems that the companies face in pursuing their claim under domestic limitation of actions rules. There is little that I would wish to add to what Lord Nicholls has said about the approach that should now be taken to claims at common law for damages for interest losses suffered as a result of the late payment of money. In my opinion a decision on this point is not essential to the resolution of the question which is at issue in this case, as the cause of action with which we are concerned here is different. But I agree with him that the House should take the opportunity of departing from Lord Brandon of Oakbrook's analysis in President of India v La Pintada Compania Navigation SA [1985] 1 AC 104 and that it should hold that at common law, subject to the ordinary rules of remoteness which apply to all claims of damages, the loss suffered as a result of the late payment of money is recoverable. This is already the law where the claim is for a debt incurred by a building contractor to raise the necessary capital which has interest charges as one of its constituents: see F G Minter v Welsh Health Technical Services Organisation (1980) 13 Build LR 1, CA, 23, per Ackner LJ; Rees and Kirby Ltd v Swansea City Council (1985) 30 Build LR 1, CA; see also Margrie Holdings Ltd v City of Edinburgh District Council, 1994 SC 1, 10-11. The reality is that every creditor who is deprived of funds to which he is entitled and which he needs to run his business will have to incur an interest-bearing loan or employ other funds which could themselves have earned interest. It is a short step to say that interest losses will arise "in the ordinary course of things" in such circumstances.

I also agree with Lord Nicholls that the loss on the late payment of a debt may include an element of compound interest. But the claimant must claim and prove his actual interest losses if he wishes to recover compound interest, as is the case where the claim is for a sum which includes interest charges. The claimant would have to show, if his claim is for ancillary interest, that his actual losses were more than he would recover by way of interest under the statute. In practice, especially where the period over which interest is sought is short or where the claimant does not have to borrow money to replace the debt, simple interest under section 35A of the Supreme Court Act 1981 is likely to be the more convenient remedy.

I wish to concentrate on the approach that should be taken to the restitutionary cause of action on which Sempra prefers and is entitled to rely, which is its claim that the money was paid under a mistake. The conclusion that the court has jurisdiction to award compound interest as damages at common law is, however, a valuable one. It provides us with a building block which was missing when the House rejected the use of compound interest as a possible solution in equity in Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669. Ancillary interest was sought on a sum for which the court was to give judgment in satisfaction of the council's restitutionary claim against the bank. It was common ground that there was no jurisdiction to award compound interest in such a case at common law or by statute: per Lord Goff of Chieveley, p 690H.

The restitutionary claim

Four sample payments of ACT have been agreed upon for the purposes of this test case all of which were, sooner or later, set off against MCT. The earliest ACT payment was made on 12 October 1981, and the latest was made on 18 July 1994. The longest interval was almost ten years, and the shortest was just under one year. Tax paid in response to an unlawful demand is recoverable under the Woolwich principle: Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70. But the limitation period of six years which applies to unlawful demands runs from the date of payment. Sempra wish to take advantage of the extended limitation period that is available under section 32(1)(c) of the Limitation Act 1980. It provides that, where the action is for relief from the consequences of a mistake, the period of limitation shall not begin to run until the claimant has discovered the mistake or could with reasonable diligence have discovered it. As Park J observed in para 11 of his judgment, one of the bases on which Sempra's claim is pleaded is for restitution by reason of the ACT having been paid under a mistake of law. The effect of your Lordships' decision in Deutsche Morgan Grenfell Group plc v IRC [2007] 1 AC 558 is that it is open to Sempra to base its claim on this ground, as the longer limitation period is in its best interests: see para 51. If this is done, the claim for interest on none of the sample payments will be statute-barred.

It appears to have been assumed until the proceedings reached this House that the choice of claim is immaterial to the way in which the principal sum is to be calculated. But the observation by the Court of Justice in para 88 of its judgment that the principal sum due is none other than the amount of interest which would have been generated by the sum the use of which was lost as a result of the premature levy of the tax invites the question whether an award on this basis is available in domestic law as a restitutionary remedy. Park J referred to this passage in para 16(ii) of his judgment, after noting in para 16(i) that Sempra had formulated its claims in both ways in the alternative. He then added this comment in para 16(iii), after referring to the Court's observation in para 89 of its judgment that the sum which may be claimed by way of restitution was the interest accrued on the ACT between its payment and the date on which the MCT became payable:

The Court of Justice seems to have assumed that the basis of the award would be the same irrespective of the choice of remedy. This appears at that stage to have been common ground. But the arguments that were developed before your Lordships have shown that this assumption is no longer sustainable.

There is no doubt that a compensatory remedy for breach of Community law would look to what the taxpaying company had lost by reason of having to pay the tax early. But that, from Sempra's point of view, is not the preferred remedy. If it is to escape from the six year limitation period it must instead pursue the alternative argument that the payments were made under a mistake. This is a restitutionary remedy. So it is necessary to look more closely at the nature of this remedy, and at the basis on which a claim under it falls to be calculated. It is only when this question has been addressed and answered that it will be possible to answer with confidence the question how, if Sempra is to be provided with the restitutionary remedy to which it is entitled for its mistake as to its rights under Community law, the amount of the principal sum due must be calculated.

In Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349, 372G-373B Lord Goff of Chieveley referred to the development of a coherent law of restitution, a doctrine first recognised by this House in Lipkin Gorman v Karpnale Ltd [1991] 2 AC 548, 577-578. He said there was a general right of recovery of money paid under a mistake and that it was founded upon the principle of unjust enrichment. At p 373C he said that a blanket rule of non-recovery on the ground of a mistake of law could not survive in a rubric of the law based on that principle. This led him to conclude that there was "a general right" to recover money paid under a mistake, whether of fact or law, subject to the defences available in the law of restitution: p 375H. At p 377C he said that the common law should now recognise that restitution may be granted in respect of money paid under a mistake of law. At p 379H he said that, subject to any applicable defences, the payer was "entitled" to recover the money paid under a mistake. Throughout his speech he was addressing a common law remedy, not one that was available in equity. I think that it can now be taken as settled that, under the Kleinwort Benson principle, a cause of action at common law is available for money paid under a mistake of law: Deutsche Morgan Grenfell Group plc v IRC [2007] 1 AC 558, para 62. I also think that the time has come to recognise that the court has jurisdiction at common law to award compound interest where the claimant seeks a restitutionary remedy for the time value of money paid under a mistake.

Recognition that restitution is a common law remedy raises questions about the limits that must be set to it which would not arise if it was available only in equity. The enrichment must, of course, have been "unjust". Andrew Burrows, The Law of Restitution, 2nd ed (2002) pp 48-50 has argued that the claimant must identify positive reasons for restitution if he is to be entitled to this remedy: see also my own observations in Kleinwort Benson, pp 408C-409D. It has been suggested, from the civilian perspective, that the underlying principle is the absence of a legal ground to justify retention of the benefit: Shilliday v Smith, 1998 SC 725, 727per Lord President Rodger; Jacques Du Plessis, "Towards a Rational Structure of Liability for Unjustified Enrichment: Thoughts from two Mixed Jurisdictions", 122 South African Law Journal 142, pp154, 180-181. In Kleinwort Benson [1999] 2 AC 349 Lord Goff also accepted that the common law, having recognised the right to recover money paid under a mistake of law, must identify particular sets of circumstances in which, as a matter of principle or policy and to protect the stability of closed transaction, recovery should not be allowed: pp 382G-H, 385C-D.

In Kleinwort Benson, at p 382G-H, Lord Goff gave three examples of the circumstances that he had in mind: defence of change of position, the defence of compromise and the defence of settlement of an honest claim, adding that the scope of the last was a matter of debate. This was not, and was not intended to be, an exhaustive list. He said that it was possible that other defences might be developed from judicial decisions in the future. A valuable review of the defences that may be available in Scots law is to be found in Robin Evans-Jones, Unjustified Enrichment, Vol 1, Chapter Ten. As he explains in para 10.01, the relationship between a claim and a defence in the law of unjustified enrichment is to be found in Lord Kyllachy's observation in Credit Lyonnais v George Stevenson & Co Ltd (1901) 9 SLT 93, 95:

The use of the word "equitable" in this context must, of course, be understood in the light of the fact that in Scotland equitable principles are part of the common law. But it shows that, in principle, the right of recovery must be accompanied by appropriate defences to prevent unfairness. Protecting the stability of closed transactions is the paradigm case for such a defence.

There is no need to pursue these arguments any further in this case. The question whether there is an unjust factor has already been settled. As the European Court of Justice has explained, there was no legal ground for the retention of the enrichment. The unjust enrichment principle supports the free-standing cause of action to recover interest, which is the measure of the enrichment. It has not been suggested that a restitutionary award by way of interest would give rise to injustice, so long as it was appropriately calculated.

The claim in Kleinwort Benson [1999] 2 AC 349 was for recovery of the money that had been paid to the local authorities outside the six-year limitation period. It was not necessary to explore how the restitutionary remedy was to be calculated, as there was no issue as to the amounts that were recoverable. The principal sums due were the same, whether the measure of the remedy was the amount by which the local authorities had been enriched or the amount of the Bank's loss. But Lord Goff went much further in his explanation of what he referred to as a coherent law of restitution at common law than he felt able to do in Westdeutsche Landesbank (see [1996] AC 669, 686A-B). His speech in Kleinwort Benson provides us with another vital building block. Recognition that the court has jurisdiction to award compound interest at common law is a short, but logical, step in the further development of the restitutionary remedy. It follows from the fact that the right to recover money paid under a mistake is available at common law To treat the choice of remedy in unjust enrichment as discretionary would, in my opinion, be inconsistent with the common law right that gives rise to it.

The basis of the award

I turn then to the basis on which the restitutionary award should be calculated. In Shilliday v Smith 1998 SC 725, 727 Lord President Rodger said that anyone who wants to glimpse something of the underlying realities in the law of unjust enrichment must start from the work of Professor Peter Birks. In the essay which he contributed to Restitution, Past, Present and Future, Essays in Honour of Gareth Jones (1998), Misnomer, p 1, Professor Birks said that the whole thrust of the law of restitution is towards defining and analysing the event which most commonly brings it about, which is unjust enrichment. Restitution is the response to unjust enrichment, and unjust enrichment is the event which triggers the response. The name of the event ought to predominate over the response. So, he argued, the subject ought to be called unjust enrichment. That is the starting point and, because the concept is one of enrichment not of damages, it determines the nature of the response.

In his introduction to the book which he called Unjust Enrichment (2nd edition, 2005) pp 3-4, he drew attention to another terminological difficulty. He explained that the law of restitution is the law of gain-based recovery, just as the law of compensation is the law of loss-based recovery:

So the remedy of restitution differs from that of damages. It is the gain that needs to be measured, not the loss to the claimant. The gain needs to be reversed if the claimant is to make good his remedy.

Professor Birks then subjected the law of unjust enrichment to a five question analysis - an exercise that he had been conducting ever since he began his discussion of the subject in "Six Questions in Search of a Subject: Unjust Enrichment in a Crisis of Identity" 1985 JR 227. As his fourth question he asked what kind of right it is that the claimant acquires: p 163. He said that the choice always lies between a right in personam (a personal right) and a right in rem (a property right). As to the content of the right in personam, which is the relevant right in the context of Sempra's claim against the Revenue, he repeated that the right in personam which arises from unjust enrichment is a right to restitution: pp 167-168. This, as he explained, is a right that the enrichee give up his gain. Elaborating on this point, he said that in the context of unjust enrichment the everyday meaning of "restitution" is stretched so that it reaches all givings up, with no hint of a restriction on giving back. His reference to "givings up" requires further analysis. But there is no suggestion here that, once it is established that the enrichment was unjust, calculation of the remedy that this gives rise to is discretionary.

The question then is whether the claimant in unjust enrichment must nevertheless have suffered a loss corresponding to the defendant's enrichment. In Unjust Enrichment 2nd ed, pp 167-168, Professor Birks said that there was no need for this to be the measure of the enrichment:

I would apply the reasoning in these passages to the claim for interest in this case. A remedy in unjust enrichment is not claim of damages. Nor is it a contractual remedy, so there is no need to search for an express or an implied term as the basis for recovery. The old rules which inhibited awards of interest to ancillary interest on sums due on contractual debts or on claims for money had and received do not apply. The essence of the claim is that the Revenue was unjustly enriched because Sempra paid the tax when it did in the mistaken belief that it was obliged to do so when in fact it was being levied prematurely. So the Revenue must give back to Sempra the whole of the benefit of the enrichment which it obtained. The process is one of subtraction, not compensation.

But Professor Birk's use of the word "disgorgement" is controversial, and it is misleading when applied to the facts of this case. Steven Elliott makes this point when he says, in the context of the equitable jurisdiction, that disgorgement interest is calculated to strip profits the fiduciary has made through the use of trust property: "Rethinking Interest on Withheld and Misapplied Trust Money" [2001] 65 Conv 313, 316. Mr Rabinowitz QC said that Sempra was seeking restitution, not disgorgement. Disgorgement would be appropriate if the claimant was seeking to recover the actual profits that the defendant had made as a result of the enrichment. But Sempra does not ask for an account of profits, nor does it invite the court to look at the use, if any, to which the money was actually put by the Revenue. Furthermore it is accepted that the claim is personal, not a proprietary one. But, as in cases of property other than money where the claim includes restitution for the value of the use of the asset that was transferred, subtraction of the enrichment from the defendant includes more than the return of the money that was transferred at its nominal or face value. That value, in this case, has already been accounted for. The subject matter of Sempra's claim is the time value of the enrichment. This is the amount that has to be assessed.

In this case the enrichment consists, not of the payment of a sum of money as such, but of its payment prematurely. As Professor Birks pointed out, the availability of money to use is not unequivocally enriching in the same degree as the receipt of money: Unjust Enrichment, 2nd ed, p 53. But money has a value, and in my opinion the measure of the right to subtraction of the enrichment that resulted from its receipt does not depend on proof by Sempra of what the Revenue actually did with it. It was the opportunity to turn the money to account during the period of the enrichment that passed from Sempra to the Revenue. This is the benefit which the defendant is presumed to have derived from money in its hands, as Lord Walker puts it in para 180. The Revenue accepts that the money it received prematurely had a value, but it says that the restitutionary award should take the form of simple interest. I do not think that such an award would be consistent with principle. Simple interest is an artificial construct which has no relation to the way money is obtained or turned to account in the real world. It is an imperfect way of measuring the time value of what was received prematurely. Restitution requires that the entirety of the time value of the money that was paid prematurely be transferred back to Sempra by the Revenue.

All this points to the conclusion, subject to what I say later about onus (see paras 47, 48) that, for restitution to be given for the time value of the money which was paid prematurely, the principal sum to be awarded in this case should be calculated on the basis of compound interest.

I recognise, of course, that in Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669 this House held that in a claim at common law for money had and received the claimant was entitled only to simple interest under section 35A of the Supreme Court Act 1981 and, by a majority, that it would not be appropriate for equity to award compound interest on the principal sum in aid of the bank's common law claim. As my noble and learned friend Lord Mance points out in his analysis of that case, the argument throughout was that there was no power at common law to award compound interest. But I agree with Lord Nicholls and with my noble and learned friend Lord Scott of Foscote that Sempra's restitutionary claim is available to it at common law. Once it is accepted that losses caused by late payment are recoverable under the restitutionary remedy at common law irrespective of the position in equity, the problem that was addressed in Westdeutche disappears.

Furthermore the interest in question in the present case is, as the Court of Justice [2001] Ch 620 stressed in para 88 of its judgment, the principal sum itself. In my opinion the decision in Westdeutsche does not address this point. We were not asked to overrule that decision, because it is distinguishable on this ground. Furthermore, the basis of Sempra's claim, as the common law has now recognised, is unjust enrichment. I do not think that it is open to the common law, when it is providing a remedy in unjust enrichment, to decline to apply the principle on which that remedy is founded when the principal sum to be awarded is being calculated. As Lord Nicholls points out (see para 99), there is now ample authority to the effect that interest losses which are recoverable as damages should be calculated on a compound basis where the evidence shows that this is appropriate. The same rule should be applied to the restitutionary remedy at common law.

Compound interest in the EU

It appears that the practice in a majority of states in the EU is to award simple interest ancillary to a principal sum that is to be paid by way of damages - the main exceptions being Poland and the Netherlands, and Germany where compound interest may be claimed as part of the damages: Ashurst, "Study on the conditions of claims for damages in case on infringement of EC Competition rules" (2004), p 87. In Corus UK Ltd v Commission of the European Communities (Case T-171/99) [2002] 1 WLR 970, para 60, the Court of First Instance said that, according to the principle generally accepted in the domestic law of the Member States, the amount recoverable for loss consisting of the loss of use of money over a period of time is generally calculated by reference to the statutory or judicial rate of interest, without compounding. But there is now ample evidence that it is the practice of the Commission to redress the advantage which the recipient of money has obtained contrary to EU rules by requiring that interest to be paid on that money is compounded at commercial rates.

Article 14 of Council Regulation (EC) No 659/1999 of 22 March 1999 (OJ 1999, L83, p1), laying down detailed rules for the application of article 93 of the EC Treaty (now article 88 EC), provides that, when negative decisions are taken in cases of unlawful aid, the aid to be recovered is to include interest at an appropriate rate fixed by the Commission. In its communication on the interest rates to be applied when aid granted unlawfully is being recovered (2003/C 110/08; OJ 203 C110, p 21) the Commission referred to its letter to Member States of 22 February 1995 in which it said that, for the purpose of restoring the status quo, compound interest at commercial rates provided a better measure of the advantage improperly conferred on the recipient of unlawful aid, and that for many years it had been its standard practice to include in its recovery decisions a clause requiring interest to be calculated on this basis.

In para 14 of the preamble to Commission Regulation (EC) No 794/2004 of 21 April 2004 implementing Council Regulation (EC) No 659/1999 laying down detailed rules for the application of Article 93 of the EC Treaty (OJ 2004 L140, p 2) the Commission stated:

In an action brought on 2 August 2004 by the Commission against the French Republic (Case C-337/04; OJ 2004 C239, p 9) in which it claimed that the French Republic had failed to fulfil its obligations in relation to state aid, the Commission said that it did not agree that the simple interest method used by the French authorities to calculate interest to be paid on the amount to be recovered was the right method:

The choice of method to be used in this case is, of course, a matter for the national court applying domestic law. But it is reassuring, although hardly surprising, to find that the compound interest method is the Commission's method of choice in the field of Community law in not dissimilar circumstances. The claim for interest in this case may be regarded as a fortiori of these examples. It is not ancillary to a principal sum for which recovery is being ordered. It is the principal sum that is recoverable for the time value of the sums that were paid prematurely.

Compound interest in domestic law

The fundamental point, however, is this. Compound interest is a necessary, and very familiar, fact of commercial life. As the Law Commission said in its Consultation Paper on "Compound Interest" (2002, No 167), para 4.1, the obvious reason for awarding compound interest is that it reflects economic reality. In its "Discussion Paper on Interest on Debt and Damages" (No 127, 2005), para 8.18 the Scottish Law Commission said that it endorsed the view of the Law Society of England and Wales in their response to the Law Commission's Consultation Paper that "simple interest never provides a full indemnity for the loss to the litigant." In para 8.38 the Scottish Law Commission said, having examined the arguments either way, that it was inclined to the view that the case against the compounding of interest was essentially a case against interest itself. Computation of the time value of the enrichment on the basis of simple interest will inevitably fall short of its true value. Such a result would conflict with the principle that applies in unjust enrichment cases, that the enrichee must give up to the claimant the enrichment with, as Professor Birks put it in Unjust Enrichment (2nd ed), p 167, no hint of a restriction to giving back. In my opinion the compounding of interest is the basis on which the restitutionary award in this case should be calculated.

Measurement

The virtue of simple interest is its simplicity. That cannot be said of compound interest, which can be calculated in different ways leading to different results. This creates the potential for dispute, which is one of the more important objections to its use generally. Hence the Law Commission's recommendation that, if a power to award compound interest were to be introduced, the Civil Procedure Rule Committee should have power to provide the courts with guidance on when to award compound interest: "Pre-Judgment Interest on Debts and Damages" (Law Com No 287), para 5.38. The evidence for the Revenue in this case was that the nature of the financial relationship between the Government and the Bank of England was such that it was impossible to measure the amount of the interest earned or saved by the Revenue, or by the government generally, on the ACT payments that were made by Sempra. The effect of its evidence was that it was wrong to assume that the Government invested the payments that it received on the basis that it would receive a compound return for it in the commercial market. But the judge did not, in the event, have to resolve this issue. It was common ground before him that if compound interest was to be awarded it should be calculated on a conventional basis - the rate being derived from the rates of interest generally prevailing on borrowings during the relevant period.

Mr Glick accepted during the first hearing of this appeal that there was no challenge to the judge's adoption of the idea that a conventional approach should be adopted to the calculation of compound interest. He explained that this was on the assumption that it was not being suggested that the basis of the claim was what the Crown achieved by way of a return on the premature payments. He accepted that a conventional rate should be adopted because Sempra's position was understood to be that the amount of the award would be the same whether it was for restitution or for damages. This too was the position as it was understood to be by Park J: [2004] STC 1178, para 19. His order of 16 June 2004 states, in para 3, that the interest rate or rates to be used for the calculation of interest during the premature tax period should be a conventional rate or rates. Your Lordships sought further clarification of the Revenue's position on this point, among others that arose out of the way the case was argued at the first hearing. In the light of the written submissions that were received, the case was put out for further oral argument.

At the further hearing Mr Rabinowitz submitted that the Revenue should be held to the concession that had been made on this issue on its behalf in the courts below and in this House. Mr Glick said that a different approach was needed if Sempra were to be allowed to recover compound interest as a restitutionary award for the time value of the money while it was in the hands of the Revenue. He submitted that, if a conventional rate of interest was to be used because of the difficulty of establishing the amount of the Government's benefit, it should be arrived at by reference to the rate at which the Government could have borrowed money during the relevant period - by the issue of Treasury Bills, for example.

The formula that was agreed for the award of compound interest was intended to be the same for all claimants in the GLOs. This was that the interest should be calculated without reference to the particular circumstances of the parties. It was to be fixed conventionally at the generally prevailing rates of interest which lenders were able to obtain commercially. As Mr Rabinowitz puts it, the measure is the reasonable price that would have to be paid to borrow the money. There is an obvious advantage to be gained by adopting a formula that will apply to all these cases and avoid the leading and analysis of complex evidence. But the concession that the cost of borrowing should be adopted was made before Sempra' argument was refined in this House in the course of argument. It is no longer obvious that an approach which looks only to the rates of interest which lenders are able to obtain commercially fits the unusual facts of this case. I would hold that justice requires that Mr Glick should be allowed to withdraw his concession. His suggestion that in this case the conventional rate should be arrived at by reference to the rate of Government borrowing is attractive. But questions of this kind are normally approached objectively by reference to what a reasonable person would pay for the benefit that is in question: Attorney General v Blake [2001] 1 AC 268, 278 F-G. Bearing this in mind, can Mr Glick's subjective approach be justified in principle?

As Lord Goff said in Lipkin Gorman v Karpnale [1991] 2 AC 548, 578C-E, and stressed again in Kleinwort Benson [1999] 2 AC 349, 385A-F, the recovery of money in restitution is not, as a general rule, a matter for the discretion of the court. A claim to recover money at common law is a matter of right. But the enrichment has to be measured, even in cases such as this where it is impracticable to identify the amount of the benefit obtained by the enrichee. The basis of the restitutionary right is the unjust enrichment principle. This suggests that, if a conventional rate of interest is to be adopted, it should be one which is appropriate to the enrichee's circumstances. In Scots law, the claim to recover an unjustified enrichment is regarded by the common law as an equitable one: Morgan Guaranty Trust Co of New York v Lothian Regional Council, 1995 SC 151, 155E-G, 166B. In Scotland, from the earliest times, the law has been permeated by equity: John W Cairns, "Historical Introduction" in A History of the Private Law of Scotland, eds Reid and Zimmermann (2000), vol 1, pp 98, 138. So the court, applying equitable principles, may take into account a whole range of circumstances: see Robin Evans-Jones, Unjustified Enrichment, Vol 1, para 9.54. I do not see why, although the claim in English law is a matter of right at common law, the enrichment principle on which it is based should not lead to the same result when an English court is measuring what needs to be reversed by it. This would require no more than a small step forward to assimilate the common law with equity and create, at least to this extent, a unified remedy: see Andrew Burrows, "We Do This At Common Law But That In Equity" (2002) 22 Oxford Journal of Legal Studies, 1, 15.

A further question as to the measure of loss has been raised by my noble and learned friend Lord Mance (para 233). Why, he asks, should there be an onus on the recipient to displace a conventional or objective measure? The basic test of recovery should be, he suggests, to look to actual benefit. Otherwise the effect would be incorrectly to reverse the onus. Your Lordships did not hear argument on the question of onus, and it has received little attention in the authorities. But in Morgan Guaranty Trust Company of New York v Lothian Regional Council, 1995 SC 151, 165 I said that, once the pursuer has averred the necessary ingredients to show that prima facie he is entitled to the remedy, it is for the defender to raise the issues which may lead to a decision that the remedy should be refused on grounds of equity. This approach was based on the principle that a party ought not be required to produce proof of matters that are unlikely to be within his own knowledge.

That observation was, of course, made in the context of a legal system whose common law principles are informed by equity. But I think that it is capable of being applied here too. Once the claimant has shown that prima facie he is entitled to a restitutionary remedy, direct knowledge of the extent of the benefit, if any, that has been received can be assumed to lie with the recipient. It is open to the recipient to demonstrate that there was no actual enrichment when the money fell into his hands notwithstanding the opportunity to turn it to account. But the case for the Revenue was not that it did not use the money at all. On the contrary, its evidence was that, because of the nature of the financial relationship between the Government and the Bank of England, it was impossible to measure the amount of interest earned or saved by it, or by the Government generally, on the sample ACT payments paid by Sempra. It was not that there was no actual benefit, but that the benefit was extremely difficult to quantify. It seems to me that, on this evidence, the assumption that the Revenue derived some benefit from the receipt of the money prematurely has not been displaced, and that this justifies resort to a conventional rate of interest as the measure of that benefit.

The proposition that a conventional rate should be used leaves open for further discussion questions of detail such as how that rate is to be arrived at and what rests should be adopted. The enrichment principle indicates that these questions should be resolved by looking at the circumstances of the enrichee. The use of ordinary commercial rates of interest, at ordinary rests, would be appropriate if those rates were relevant to the enrichee's circumstances. But I would hold that it is open to the enrichee to show that it would have been able to borrow money at rates or on terms more favourable to it than those available in the ordinary commercial market. If it can do that, then ordinary rates and other terms must give way to those that are relevant to the circumstances of the enrichee. The unusual position of the Revenue has been sufficiently demonstrated. It seems to me that Mr Glick's suggestion is in accordance with principle, and I would adopt it.

Conclusion

For these reasons I agree with Lord Nicholls and Lord Walker that Sempra's claim for restitution ought to be measured by an award of compound interest at conventional rates calculated by reference to the rates of interest and other terms applicable to borrowing by the Government in the market during the relevant period. I would vary para 3 of the judge's order to that effect. I would delete para 2 of the Court of Appeal's order because the assumption on which it was based, that ordinary commercial rates of interest would be used, is being departed from. Otherwise I would dismiss the appeal.

LORD NICHOLLS OF BIRKENHEAD

My Lords,

Legal rules which are not soundly based resemble proverbial bad pennies: they turn up again and again. The unsound rule returning once more for consideration by your Lordships' House concerns the negative attitude of English law to awards of compound interest on claims for debts paid late.

We live in a world where interest payments for the use of money are calculated on a compound basis. Money is not available commercially on simple interest terms. This is the daily experience of everyone, whether borrowing money on overdrafts or credit cards or mortgages or shopping around for the best rates when depositing savings with banks or building societies. If the law is to achieve a fair and just outcome when assessing financial loss it must recognise and give effect to this reality.

Unhappily this is still not altogether so. To a significant extent the law remains out-of-step with everyday life in the 21st century. In the first half of the 19th century the common law adopted a restrictive rule: unpaid debts do not carry interest, either compound or simple. This was an exception to the ordinary common law principles applicable to recovery of damages for breach of contract.

Since then successive statutes have made general provision for courts to award interest in many instances. This provision is limited to simple interest. The statutes make no provision for compound interest.

In 1984 your Lordships' House curtailed the scope of the restrictive common law rule. Despite this, by common accord the current position is not yet coherent or satisfactory. So your Lordships are being called upon to consider the implications of this restrictive rule once more. Your Lordships have to consider how far the common law should still abide in a world where present-day economic reality is not allowed to intrude.

Advance corporation tax and the EC Treaty

The context in which the question of compound interest arises on this appeal is a claim by a taxpayer company Sempra Metals Ltd ('Sempra') for compensation in respect of the United Kingdom's breach of article 52, now article 43, of the EC Treaty. In 2001 the Court of Justice of the European Communities ('the ECJ') decided that one aspect of this country's advance corporation tax regime, as it was in force until 1999, contravened article 52. The offending statutory provisions concerned group income elections. These were contained latterly in section 247 of the Income and Corporation Taxes Act 1988 ('ICTA') as it stood until April 1999.

In short, from 1973 until 1999 a company was liable to pay advance corporation tax ('ACT') on the amount of its dividends. ACT could be set off against the company's liability to corporation tax on its profits, colloquially known as its mainstream corporation tax liability, when it fell due: see sections 14(1) and 239 of ICTA. In practice there was always an interval of at least eight months between any payment of ACT and a subsequent set off against mainstream corporation tax.

Section 247 enabled parent companies and subsidiary companies jointly to make an election having the effect of excluding dividends paid by a subsidiary to its parent from the obligation to pay ACT. This option, however, was available only when the parent company and its subsidiary were resident in the United Kingdom. A group income election, as it was known, was not available if the parent company was resident outside the United Kingdom. The ECJ held this provision contravened parent companies' freedom of establishment, contrary to article 52 of the Treaty: see Metallgesellschaft Ltd v Inland Revenue Commissioners and Hoechst AG v Inland Revenue Commissioners (Joined cases C-397 and 410/98) [2001] Ch 620. Metallgesellschaft was the previous name of Sempra, the respondent to the present appeal.

In its decision the ECJ considered the remedial consequences of this contravention of the Treaty. The court observed that in the ordinary course it is for national law to settle ancillary questions relating to the reimbursement of charges improperly levied, such as the payment of interest, including the rate of interest (paragraph 86). But in the present cases the claim for payment of interest covering the loss of use of the money paid by way of ACT was not ancillary. It was the very objective sought by the claimants. Where the breach of Community law arises, not from payment of the tax itself, but from its being levied prematurely, the award of interest represents the reimbursement of what was improperly paid. An award of interest is 'essential' in restoring the equal treatment guaranteed by article 52 (paragraph 87).

The court declined to discuss the form of remedial proceedings in a national court. That was a matter for claimants, subject to the supervision of the national court. But the court did make brief observations on the two forms of remedial proceedings ventilated before it. In an action for restitution, the principal sum due is none other than the amount of interest which would have been generated by the sum whose use was lost as a result of the premature levy of the tax (paragraph 88). In a claim for compensation for damage caused by the breach of Community law, the interest available to the claimants had it not been for the inequality of treatment 'constitutes the essential component' of their Community right (paragraph 93). The award of interest 'would therefore seem to be essential' if the damage caused by the breach of article 52 was to be repaired (paragraph 95).

Accordingly, in answer to a second question, the ECJ gave guidance to the following effect. Where subsidiary companies have been obliged to pay ACT in respect of dividends paid to their non-resident parents when liability to pay ACT would have been avoided had the parent companies been resident in the United Kingdom, article 52 requires that the resident subsidiary companies and the non-resident parent companies should have 'an effective legal remedy in order to obtain reimbursement or reparation of the financial loss which they have sustained and from which the authorities of the member state concerned have benefited as a result of the advance payment of tax by the subsidiaries' (paragraph 96). The court continued:

The present case

The present case is a test claim under a group litigation order made to manage the numerous claims brought against the Inland Revenue as a result of the Hoechst decision. Sempra traded, as a principal and as a broker for clients, in metals listed on the London Metal Exchange. Sempra is resident in the United Kingdom. Its parent company is resident in Germany. Sempra's claims are founded on the amounts of ACT paid in respect of four sample dividends Sempra paid to its parent company. The dividend payments comprised £2.5million paid in July and September 1981; £2million in January 1985; £1.8million in April 1989; and £21million in May 1994.

In very round figures, for the exact amounts are not material on this appeal, Sempra paid ACT in respect of these dividends as follows: £1.07million in October 1981; £809,000 in April 1985; £369,000 in July 1989; and £3.23million in July 1994. All these ACT payments were subsequently set off against Sempra's mainstream corporation tax. The intervals between the payment of ACT and the time of set off varied considerably. The shortest period was just under one year, the longest almost ten years. Of the ACT totalling £1.07million paid in October 1981, £812,000 was not utilised until July 1991. With the exception of £1.56million utilised in July 1996, in each case the set off occurred before the issue of the writ in these proceedings.

For completeness I should add that this appeal proceeds on the basis that Sempra and its parent company would have made a group income election had this not been precluded by the UK residence requirements of section 247 ICTA.

At first instance Park J decided that the compensation due to Sempra should be calculated on a compound basis: [2004] STC 1178. The Court of Appeal, comprising Chadwick, Laws and Jonathan Parker LJJ, dismissed an appeal by the Inland Revenue: [2005] STC 687. The Court of Appeal made clear that interest should be computed by compounding at the same periodic rests as those by reference to which the applicable rate of interest is fixed. The Inland Revenue has now appealed to your Lordships' House.

Compound interest and Community law

The detriment suffered by a taxpayer by the premature payment of tax is loss of use of the money for the period of prematurity. So if a taxpayer had to borrow the money, and his claim is for damages, his loss comprises the cost of borrowing the money for the period of prematurity. Alternatively, if the taxpayer's reparation claim is framed in restitution, the Inland Revenue's unjust enrichment comprises the benefit of having use of the money for the period of prematurity. Either way the essence of the taxpayer's claim is for an amount of money by way of interest in respect of the tax paid prematurely.

The Hoechst decision makes plain that domestic law cannot exclude payment of interest in such cases. But, if not explicit, it is plainly implicit in the ECJ's answer to the second question read in the context of the judgment as a whole that assessment of the amount of interest is a matter for the member state concerned, provided always the member state's rules satisfy the principles of equivalence and effectiveness. The member state's rules must be not less favourable than those governing the same or analogous domestic proceedings, and they must not render practically impossible or excessively difficult the exercise of the rights conferred by Community law.

A primary factor governing the amount of interest payable on a debt is the rate of interest. This, par excellence, is a matter for national law: see paragraph 86 of the Hoechst judgment, already noted. Another factor is whether the interest is simple or compound and, if compound, the intervals between the compounding dates. The rate of interest being a matter for national law, it would be very odd if the choice between simple and compound interest were not also a matter for national law. To my mind it is clear from the Hoechst decision that, even in the context of the present type of case, this choice is a matter for national law: provided always that national law satisfies the requirements of the twin principles of equivalence and effectiveness.

In the present case there is no difficulty or dispute about the principle of equivalence. The rights conferred on citizens in the United Kingdom by article 43, previously article 52, of the EC Treaty are rights falling within section 2(1) of the European Communities Act 1972. They are rights to which effect must be given in this country without further enactment. Accordingly a breach of the article 43 prohibition is characterised in English law as breach of a statutory duty: see the well-known analysis of Lord Diplock in Garden Cottage Foods Ltd v Milk Marketing Board [1984] 1 AC 130, 141.

The Inland Revenue accepts that, in addition to this primary claim in tort, Sempra has an alternative claim for restitutionary relief on two grounds: tax paid pursuant to an unlawful demand, and payments made under a mistake of law. The first of these two grounds is based on Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70, the second on Deutsche Morgan Grenfell Group Plc v Inland Revenue Commissioners [2007] 1 AC 558.

Sempra accepts that these three causes of action satisfy the Community law requirement of equivalence. Issue has been joined on the effectiveness of these domestic remedies.

As already foreshadowed, the crux of the dispute on effectiveness concerns the availability of compound interest in respect of the wrongful levying of ACT. The Inland Revenue recognises that interest is payable in respect of the tax paid prematurely in the form of ACT. But it contends that under English law the courts do not have power to award compound interest save in cases of fraud and misapplication by a fiduciary. The Revenue contends, further, that an award of simple interest would be an effective legal remedy for Sempra in the present case.

So I turn to consider the provision English law makes for the payment of interest in respect of the three causes of action relied upon by Sempra. I shall then consider whether this remedial provision satisfies the Community law principle of effectiveness.

Interest losses and damages

I start with the broad proposition of English law that as a general rule a claimant can recover damages for losses caused by a breach of contract or a tort which satisfy the usual remoteness tests. This broad common law principle is subject to an anomalous, that is, unprincipled, exception regarding one particular type of loss arising in respect of one particular type of claim. The exception comprises claims for interest losses by way of damages for breach of a contract to pay a debt. The general common law principle does not apply to such claims. Damages are not recoverable in cases falling within this exception.

The origin of this exception to the general principle is not impressive. Nor is its subsequent history. The late Dr F A Mann once said this exception showed the common law of England 'at its worst': "On Interest, Compound Interest and Damages" (1985) 101 L Q R 30, page 47.

I can start in 1829. This is when matters took an unfortunate turn. In that year Lord Tenterden CJ delivered the judgment of the Court of King's Bench in Page v Newman (1829) 9 B & C 378. Captain Page had loaned various sums of money to Mr Newman while they were prisoners of war together at Verdun in 1814. In 1819 Mr Page claimed repayment of £135 plus interest. The court held that, in the absence of agreement, money lent does not carry interest. The reason was one of practical convenience. The contrary rule would be 'productive of great inconvenience', because 'it might frequently be made a question at nisi prius whether proper means had been used to obtain payment of the debt, and such as the party ought to have used' (page 381). In other words, difficulties might arise in determining whether a claimant had taken reasonable steps to mitigate his loss.

Shortly thereafter Lord Tenterden promoted the Civil Procedure Act 1833 (3 & 4 Wm 4 c 42), known colloquially as Lord Tenterden's Act. Section 28 empowered juries to award interest on 'debts or sums certain' in defined circumstances. These circumstances were strictly interpreted by the courts. Section 28 proved to be of limited value in meeting the perceived shortcoming in the common law.

In 1893 the problem came before your Lordships' House in London, Chatham and Dover Railway Co v South Eastern Railway Co [1893] AC 429. The appellant company claimed money due on the taking of an account together with interest. The official referee who took the account allowed interest under Lord Tenterden's Act. On appeal the company contended that, even if it was not within the statute, it could recover interest by way of damages for wrongful detention of its debt. The House rejected the submission. The House decided that at common law a court had no power to award interest by way of damages for the late payment of a debt.

The House reached its conclusion with reluctance. Lord Herschell LC said that a person wrongfully withholding money ought not in justice to benefit by enjoying the use of that money (page 437). He considered Lord Tenterden's 'inconvenience' reasoning in Page v Newman was not a satisfactory basis for excluding a claim to interest by way of damages in cases where justice requires it should be awarded. He also considered the statutory provision was 'too narrow for the purposes of justice' (pages 440-441). But the combination of three factors made it impossible to reopen the question. The factors were the decision in Page v Newman, the intervention of the legislature, and the absence over the preceding 60 years of any case giving practical effect to the more liberal views of Lord Mansfield CJ and other judges.

Some 40 years later Parliament enacted section 3 of the Law Reform (Miscellaneous Provisions) Act 1934. This replaced section 28 of Lord Tenterden's Act. Section 3 empowered courts, in proceedings to recover any debt or damages, to award simple interest on the amount for which judgment was given for the period from when the cause of action arose to the date of judgment. This provision applied more widely than the much criticised section 28 of Lord Tenterden's Act but there were still huge gaps. There was no provision for the payment of interest where a debt was paid after proceedings for its recovery had been commenced but before judgment. Nor did the section apply where a debt was paid late but before the inception of proceedings.

In 1952 Denning LJ, not for the first time, showed the way forward. Trans Trust SPRL v Danubian Trading Co Ltd [1952] 2 QB 297 concerned a claim for damages for breach of contract in failing to provide a confirmed credit. As an aside, and with the support of Romer LJ, Denning LJ said the only principled ground on which damages can be refused for non-payment of money is remoteness. He suggested that when a special loss is foreseeable at the time of the contract as the consequence of non-payment, such loss might well be recoverable: page 306. In other words, the common law rule confirmed in the London, Chatham and Dover Railway case is limited in its scope. A particular loss arising from the late payment of money may well be recoverable where it meets the ordinary remoteness test applied to losses arising from breaches of contract.

This suggestion was taken up by the Court of Appeal in Wadsworth v Lydall [1981] 1 WLR 598. The court held that interest charges incurred by Mr Wadsworth as a result of Mr Lydall's late payment of money were recoverable as damages for breach of contract. Brightman LJ noted, at page 603, that in the London, Chatham and Dover Railway case your Lordships' House was concerned only with a claim for interest by way of general damages. In a much-quoted passage he continued:

Hadley v Baxendale

(1854) 9 Exch 341

Clearly, Brightman LJ's generalised reference to 'the principle of Hadley v Baxendale' was intended to be no more than a reference to that case as the locus classicus on the remoteness test applicable in breach of contract cases.

In the following year section 3 of the Law Reform (Miscellaneous Provisions) Act 1934 was superseded by a new statutory provision, inserted into the Supreme Court Act 1981 as section 35A: see the Administration of Justice Act 1982, section 15. This section is still in force. For present purposes it is sufficient to note that the effect of section 35A is much the same as section 3 of the 1934 Act save that, unlike the 1934 Act, the new provision covers also the case where a defendant paid his debt after the inception of proceedings but before judgment: section 35A(3). There is still no provision in the Act for debts paid late but before the inception of proceedings. Nor is there provision for compound interest.

The Pintada decision

I come next to the important decision of this House in the Pintada case: President of India v La Pintada Compania Navigacion SA [1985] 1 AC 104. The case concerned a claim by ship owners for interest on freight and demurrage paid late. Lord Brandon of Oakbrook delivered the leading speech, with which all their Lordships agreed. Lord Brandon approved the decision in Wadsworth v Lydall. He held that, contrary to common belief, the London, Chatham and Dover Railway case applied only to claims for interest by way of general damages. It did not extend to claims for special damages. He said this interpretation of the London, Chatham and Dover Railway case would reduce considerably the scope of that case as generally understood: page 127.

Unfortunately Lord Brandon's analysis has given rise to its own difficulty. In ordinary legal usage general damages comprise losses which must be pleaded and proved but which are quantified in money terms by the court. Special damages comprise losses which must be pleaded and proved in money terms. To take a simple example: damages for the loss of a limb are an instance of general damages, damages for the cost of medical treatment are special damages. With both general and special damages questions of remoteness may arise.

The difficulty with Lord Brandon's approach is that he adopted a different criterion when distinguishing general and special damages. He said that in this context the difference between general and special damages corresponds to the difference between damages recoverable under the first part of the rule in Hadley v Baxendale (1854) 9 Exch 341 (general damages) and damages recoverable under the second part of that rule (special damages).

Use of this criterion produces a widely criticised result. Certainly, on its face, the result is extraordinary. The first limb of Hadley v Baxendale applies to losses arising according to the ordinary course of things, the second limb applies to losses which may reasonably be supposed to have been in the parties' contemplation when they made the contract as the probable result of a breach. The application of this distinction in the present context means that interest losses arising from the late payment of money in the ordinary course of things, however serious these may be, are irrecoverable, but other losses are recoverable. The distinction thus drawn was castigated by Staughton J in President of India v Lips Maritime Corporation [1985] 2 Lloyd's Rep 180, 185:

In the High Court of Australia Mason CJ and Wilson J observed that this subverts the second limb of Hadley v Baxendale from its intended purpose: Hungerfords v Walker (1989) 171 CLR 125, 142.

Lord Brandon did not explain why this distinction should be made. The rationale may perhaps be gleaned from what he said when your Lordships' House revisited the subject three years later in President of India v Lips Maritime Corporation [1988] 1 AC 395. That case concerned a claim to recover exchange currency losses as damages for late payment of demurrage. The House held that, not being a claim for interest losses as damages for late payment of a debt, this claim lay outside the common law exception to the general rule. Lord Brandon emphasised that the only reason the House did not depart from the London, Chatham and Dover Railway case when deciding the Pintada case was the subsequent intervention of the legislature in 1934 and again in 1982. Departure would produce an undesirable conflict between the right to recover interest at common law and the statutory entitlement to recover interest on a discretionary basis, page 424.

A possible inference from this may be that in Lord Brandon's view entitlement to recover interest losses arising in the ordinary course of things would conflict with the statute but entitlement to recover other interest losses would not. If this is the rationale I must respectfully disagree with Lord Brandon. Clearly, conflict with section 35A of the Supreme Court Act 1981 is to be avoided. But this important consideration does not necessitate or justify the use made in the Pintada case of the distinction between the first and second limbs of the rule in Hadley v Baxendale. Section 35A is of general application. It applies generally to proceedings for recovery of a debt or damages. The court is empowered to award simple interest on a discretionary basis. If, as the House held in Pintada, interest losses falling within the second limb of Hadley v Baxendale are recoverable at law despite the statute, the same must surely be true of interest losses falling within the first limb. The policy considerations underlying this general statutory provision cannot apply differently to interest losses according to which limb of a remoteness rule they fall within.

Sempra's claim for damages

I have almost reached journey's end on this issue. I pause to note one point which is clear beyond a peradventure. In the present case Sempra is not claiming damages for interest losses caused by late payment of a debt. Sempra is claiming interest losses as damages for breach of a statutory duty. As such Sempra's claim does not fall within the exception to the general common law rules. Accordingly Sempra's damages claim is subject to the same rules as apply generally to damages claims in tort. Subject to satisfying those rules Sempra is entitled to recover damages in respect of the losses of interest, whether simple or compound, it sustained by reason of the wrongful levying of ACT.

This being so, no difficulty arises in English law from the fact that the ACT exacted unlawfully was mostly utilised by being set off against Sempra's mainstream corporation tax liability before Sempra commenced proceedings. Sempra's cause of action arose when it suffered loss by having to provide the ACT. The subsequent utilisation of the ACT by way of set off did not extinguish Sempra's existing claim for damages.

The restrictive common law exception today

This conclusion suffices to dispose of the damages issue in the present case. I go further, in view of the wide-ranging arguments presented to your Lordships. The common law should sanction injustice no longer. The House should recognise the remnant of the restrictive common law exception for what it is: the unprincipled remnant of an unprincipled rule. The House should erase the remains of this blot on English common law jurisprudence. The House should do so by taking to its logical conclusion the step initiated by the House in 1984 in the Pintada case. This would accord with an observation of Lord Woolf in Westdeutsche Landesbank Girozentrale v Islington Borough Council [1996] AC 669, 733. Lord Woolf noted that, despite the lapse of time since the London, Chatham and Dover Railway case was decided in 1893, 'the courts will be prepared to limit the application of that decision where this can be done in accordance with principle and it is appropriate to do so.'

In the Pintada case the House made clear that, contrary to the general understanding of the effect of the London, Chatham and Dover Railway case, claims for damages for interest losses suffered as a result of the late payment of money are not taboo. That is plainly right. Those who default on a contractual obligation to pay money are not possessed of some special immunity in respect of losses caused thereby. To be recoverable the losses suffered by a claimant must satisfy the usual remoteness tests. The losses must have been reasonably foreseeable at the time of the contract as liable to result from the breach. But, subject to satisfying the usual damages criteria, in principle these losses are recoverable as damages for breach of contract. This is so even if the losses consist of a liability to pay borrowing costs incurred as a result of the late payment, as happened in Wadsworth v Lydall [1981] 1 WLR 598. And this is so irrespective of whether the borrowing costs comprise simple interest or compound interest.

To this end, if your Lordships agree, the House should now hold that, in principle, it is always open to a claimant to plead and prove his actual interest losses caused by late payment of a debt. These losses will be recoverable, subject to the principles governing all claims for damages for breach of contract, such as remoteness, failure to mitigate and so forth.

In the nature of things the proof required to establish a claimed interest loss will depend upon the nature of the loss and the circumstances of the case. The loss may be the cost of borrowing money. That cost may include an element of compound interest. Or the loss may be loss of an opportunity to invest the promised money. Here again, where the circumstances require, the investment loss may need to include a compound element if it is to be a fair measure of what the plaintiff lost by the late payment. Or the loss flowing from the late payment may take some other form. Whatever form the loss takes the court will, here as elsewhere, draw from the proved or admitted facts such inferences as are appropriate. That is a matter for the trial judge. There are no special rules for the proof of facts in this area of the law.

But an unparticularised and unproved claim simply for 'damages' will not suffice. General damages are not recoverable. The common law does not assume that delay in payment of a debt will of itself cause damage. Loss must be proved. To that extent the decision in the London, Chatham and Dover Railway case remains extant. The decision in that case survives but is confined narrowly to claims of a similar nature to the simple claim for interest advanced in that case. Thus, that decision is to be understood as applying only to claims at common law for unparticularised and unproven interest losses as damages for breach of a contract to pay a debt and, which today comes to the same, claims for payment of a debt with interest. In the absence of agreement the restrictive exception to the general common law rules prevails in those cases.

The common law's unwillingness to presume interest losses where payment is delayed is, I readily accept, unrealistic. This is especially so at times when inflation abounds and prevailing rates of interest are high. To require proof of loss in each case may seem unduly formalistic. The common law can bear this reproach. If a party chooses not to prove his interest losses the remedy provided by the law is to be found in the statutory provisions.

I must emphasise that limiting the scope of the restrictive common law exception in this way does not lead to a result which conflicts with the legislation, or with the underlying legislative policy, for two reasons. First, section 35A of the 1981 Act is not an exhaustive code. It is not intended to be an exhaustive code. Section 35A does not displace any jurisdiction the courts themselves have to award interest. This is made plain by subsection (4). Courts of equity, for instance, have long exercised a jurisdiction to award interest, including compound interest, in certain circumstances. Likewise with the Late Payment of Commercial Debts (Interest) Act 1998, now amended by the Late Payment of Commercial Debts Regulations 2002 (SI 2002 No 1674). This Act implies into certain types of contracts a term to the effect that a qualifying debt carries interest in accordance with the provisions of the Act. But a debt does not carry interest under a term implied by the Act if, or to the extent that, a right to demand interest on it, which exists by virtue of any rule of law, is exercised: section 3(3).

Secondly, section 35A is concerned with interest on debts and damages. The section says nothing about the principles to be applied by a court when assessing the amount of damages for which it gives judgment. The section does not preclude a court from taking interest losses into account when awarding damages for breach of contract. This has long been the general understanding. This is shown by the string of reported cases where interest losses have been recovered as damages in claims for breach of contract or in respect of torts, or would have been so recovered if the losses had been proved. These interest losses have included losses calculated on a compound basis where appropriate. Among the cases are Brandeis Goldschmidt & Co Ltd v Western Transport Ltd [1981] 1 QB 864, Swingcastle Ltd v Gibson [1991] 2 AC 223, Nigerian National Shipping Lines Ltd v Mutual Ltd [1998] 2 Lloyd's Rep 664, Hartle v Laceys [1999] 1 PN 315, and Mortgage Corporation v Halifax (SW) Ltd [1999] 1 PN 159.

For these reasons I consider the court has a common law jurisdiction to award interest, simple and compound, as damages on claims for non-payment of debts as well as on other claims for breach of contract and in tort.

Interest benefits and restitution

Against this background I turn to the two restitutionary causes of action asserted by Sempra. Sempra's claim is that it paid ACT in response to an unlawful demand and under a mistake of law. On both these bases the Inland Revenue's receipt of ACT comprised unjust enrichment at the expense of Sempra. Of these two formulations Sempra much prefers the second because of the extended limitation period applicable under the Limitation Act 1980, section 32(1)(c). Sempra's claim is that under both causes of action restitution requires the Inland Revenue to pay Sempra the value of the benefit the Inland Revenue obtained by having use of the money Sempra paid as ACT.

In principle this claim is unanswerable. The benefits transferred by Sempra to the Inland Revenue comprised, in short, (1) the amounts of tax paid to the Inland Revenue and, consequentially, (2) the opportunity for the Inland Revenue, or the Government of which the Inland Revenue is a department, to use this money for the period of prematurity. The Inland Revenue was enriched by the latter head in addition to the former. The payment of ACT was the equivalent of a massive interest free loan. Restitution, if it is to be complete, must encompass both heads. Restitution by the Revenue requires (1) repayment of the amounts of tax paid prematurely (this claim became spent once set off occurred) and (2) payment for having the use of the money for the period of prematurity.

In the ordinary course the value of having the use of money, sometimes called the 'use value' or 'time value' of money, is best measured in this restitutionary context by the reasonable cost the defendant would have incurred in borrowing the amount in question for the relevant period. That is the market value of the benefit the defendant acquired by having the use of the money. This means the relevant measure in the present case is the cost the United Kingdom government would have incurred in borrowing the ACT for the period of prematurity. Like all borrowings in the money market, interest charges calculated in this way would inevitably be calculated on a compound basis.

The present position in English law

The present state of English law does not accord with this analysis. At present the court is considered to have no jurisdiction, that is, no power to make an award of compound interest on a personal claim for restitution of a sum of money paid by mistake or following an unlawful demand. The court has power to make an award of simple interest under section 35A of the Supreme Court Act 1981. But, as the authorities now stand, English law does not recognise that a restitutionary award at common law should include restoration to the claimant of the time value of the money he transferred to the defendant and which the defendant enjoyed by having the money in his possession.

How did this divergence from reality come about? How is it that, to paraphrase Lord Goff's words in Westdeutsche Landesbank Girozentrale v Islington London B C [1996] 669, 691, English law is revealed as incapable of doing full justice? Once more the answer lies in the historical origin of this area of the law. The restitutionary cause of action relied upon by Sempra derives from a contrived extension of the indebitatus assumpsit form of action. A plea of 'money had and received to the plaintiff's use' was regarded as apt where a defendant was under an obligation 'from the ties of natural justice' to refund money to the plaintiff. This plea appears to have been in common use from the early 17th century. The law implied a debt, and gave a cause of action, 'as it were upon a contract', in the well known words of Lord Mansfield CJ in Moses v Macferlan 2 Burr 1005, 1008. Lord Mansfield likened this to a claim 'quasi ex contractu, as the Roman law expresses it'.

That was in 1760. Heavily influenced by the fictitious nexus with a claim for breach of contract, the courts set themselves against awarding interest on a claim for money had and received: see Walker v Constable (1798) 1 Bos & Pul 306, De Havilland v Bowerbank (1807) 1 Camp 50, per Lord Ellenborough, De Bernales v Fuller (1810) 2 Camp 426, Depcke v Munn (1828) 3 Car & P 112, per Lord Tenterden CJ, and Fruhling v Schroeder (1835) 2 Bing (N C) 78. This approach culminated in the Privy Council decision in Johnson v The King [1904] AC 817. On a claim for repayment of money paid by mistake the Board considered an order for payment of interest would be inconsistent with the law as settled in London, Chatham and Dover Railway Co v South Eastern Railway Co [1893] AC 429. In other words, in the London, Chatham and Dover Railway case your Lordships' House had decided that at common law a court had no jurisdiction to award interest on the late payment of a debt. The position was the same regarding an award of interest on a claim for repayment of money paid by mistake because a claim for repayment of money paid by mistake is founded on an implied contract.

This fiction of an implied contract lingered long in the law. It continued to govern the ambit of the remedy for money had and received. In 1914, for instance, in Sinclair v Brougham [1914] AC 398, 417, Viscount Haldane LC said the remedy was given only 'where the law could consistently impute to the defendant at least the fiction of a promise'. This fetter on the principled development of the law of restitution was not finally removed until the celebrated decision of your Lordships' House in Lipkin Gorman v Karpnale Ltd [1991] 2 AC 548. It is now accepted law that a claim for restitutionary relief is not founded on a fictitious implied contract or 'quasi-contract'. This is a false and misleading characterisation of the nature of claims for restitution as a remedy for unjust enrichment.

The Lipkin Gorman decision was in 1991. Meanwhile the law on the courts' inability to award interest at common law on restitutionary claims remained as settled by Lord Mansfield CJ in 1760. The common law regarding claims for interest as damages for non-payment of debt had developed to some extent. The unsatisfactory rule established authoritatively in the London, Chatham and Dover Railway case was partly ameliorated by the decision in President of India v La Pintada Compania Navigacion SA [1985] 1 AC 104. No such ameliorating development took place in the law of restitution.

So it is perhaps not surprising that as recently as ten years ago the law concerning interest in restitution cases was still generally understood to be as enunciated by Lord Mansfield, Lord Ellenborough and Lord Tenterden. Hence it came about that in 1996, in Westdeutsche Landesbank Girozentrale v Islington London Borough Council [1996] AC 669, counsel opened an appeal in your Lordships' House on a claim for compound interest on money paid under interest rate swaps held to be void with these words:

The court has jurisdiction to award simple interest under section 35A of the Supreme Court Act 1981, because 'debt or damages' in section 35A includes any sum of money recoverable by one party from another: see BP Exploration Co (Libya) Ltd v Hunt (No 2) [1983] 2 AC 352. But no interest, whether compound or simple, is recoverable at common law. Sometimes interest, compound as well as simple, is recoverable in equity.

The appeal in the Westdeutsche case proceeded on that footing. Lord Goff of Chieveley recorded this, noting that the central issue was whether there is jurisdiction in equity to award compound interest in a case such as the instant case: pages 690-691. The majority of the House, comprising Lord Browne-Wilkinson, Lord Slynn of Hadley and Lord Lloyd of Berwick, held there is no such jurisdiction. Lord Goff and Lord Woolf dissented. They held an award of compound interest in equity is necessary to grant full restitution. In exercise of its equitable jurisdiction the court has power to make such an award in personal claims in restitution.

In these unusual circumstances I consider it is open to your Lordships' House on this appeal to re-examine the basic point of law conceded and not argued on the Westdeutsche appeal, namely, whether interest may be awarded by the courts in exercise of their common law jurisdiction to grant personal restitutionary relief. Further, I consider your Lordships should undertake this task. Having only recently been released from the shackles of implied contract and, hence, the restraints of the London, Chatham and Dover Railway case, the law of restitution should now have the opportunity to develop as a coherent body of principled law. The decision of the House in a case where this point was conceded and assumed cannot properly stand in the way.

If the House takes this opportunity I venture to repeat there can only be one answer on this important question of law. Nobody has suggested a good reason why, in a case like the present, an award of compound interest should be denied to a claimant. An award of compound interest is necessary to achieve full restitution and, hence, a just result. I would hold that, in the exercise of its common law restitutionary jurisdiction, the court has power to make such an award. I agree with the thrust of Mummery LJ's observations on this point in NEC Semi-Conductors Ltd v Inland Revenue Commissioners [2006] STC 606, 642-643, paras 172-175. To that extent I would depart from the decision on the Westdeutsche appeal.

If this approach is adopted the unfortunate decision in the London, Chatham and Dover Railway case will be effectually buried in relation to the payment of interest for non-payment of a debt and in relation to the payment of interest for having the use of money in personal restitution cases. The law will achieve a principled measure of consistency between contractual obligations and restitutionary obligations. The common law in Australia has developed in this way. The common law in England should do likewise.

I add that, as with awards of compound interest as damages for non-payment of a debt, so also with awards of compound interest as restitutionary relief in respect of a defendant's unjust enrichment: such awards do not conflict with section 35A of the Supreme Court Act 1981. As already noted, section 35A is concerned with interest on 'a debt or damages'. An amount of money recoverable as restitutionary relief falls within this phrase. Section 35A bites on that amount. But section 35A says nothing about the principles to be applied by the courts at the anterior stage when assessing the amount of money required to achieve full restitution.

Further, as with the damages claim in the present proceedings, so also with the two restitutionary claims, no difficulty arises from the fact that Sempra's ACT payments were mostly used before the inception of proceedings. The Inland Revenue had the benefit of the use of each payment of ACT for at least eight months. Setting off a payment of ACT against Sempra's mainstream corporation tax liability did not extinguish the Inland Revenue's restitutionary liability in respect of the interest benefits it had by then obtained from the ACT payments.

Measuring the value of the use of money

I mentioned above that in cases of personal restitution the value of the use of money is prima facie the reasonable cost of borrowing the money in question. I should elaborate a little on this, noting first that a comparable objective measure is well established in the analogous case of valuing the benefit derived by a defendant from unauthorised use of the claimant's land or goods. In the modern terminology these are instances of restitution for wrongdoing as distinct from restitution for unjust enrichment. The Earl of Halsbury's chair and Lord Shaw's horse are famous hypothetical examples of the application of this 'user' principle: see The Mediana [1900] AC 113, 117, and Watson, Laidlaw & Co Ltd v Pott, Cassels and Williamson (1914) 31 RPC 104, 119. If the unauthorised use causes injury, damages will be recoverable. If the unauthorised use does not cause damage the defendant must still recompense the plaintiff for the benefit he unjustly received. This distinction was drawn explicitly in cases such as Whitwham v Westminster Brymbo Coal and Coke Co [1896] 2 Ch 538 and Penarth Dock Engineering Co Ltd v Pounds [1963] 1 Lloyd's Rep 359. In Attorney General v Blake [2001] 1 AC 268, 278, I summarised the ordinary measure of the benefit in this type of case as the price a reasonable person would pay for the right of user.

The time value of money, measured objectively in this way, is to be distinguished from the value of the benefits a defendant actually derived from the use of the money. The latter value is not in point in the present case. Sempra retained no proprietary interest in the money it paid to the Inland Revenue, and it has no interest in the 'fruits' of that money. Sempra's claim is a personal claim against the Inland Revenue in respect of the benefits it transferred to the Revenue. The value of those benefits should be measured as described above.

In the present case there can be nothing unjust in requiring the Inland Revenue to pay compound interest, by way of restitution, on the huge interest free loan constituted by Sempra's payment of ACT. But this will not always be so. For instance, a recipient of a payment made by a mistake shared by both parties might make no actual use of the money. He might pay the money into a current account at a bank yielding little or no interest. When the mistake comes to light he repays the money. In such a case, depending on the circumstances, it might well be most unfair that he should be out of pocket by having to make an additional payment, whether as compound interest or even simple interest, in respect of the 'time value' of the money he received.

Here, as elsewhere, the law of restitution is sufficiently flexible to achieve a just result. To avoid what would otherwise be an unjust outcome the court can, in an appropriate case, depart from the market value approach when assessing the time value of money or, indeed, when assessing the value of any other benefit gained by a defendant. What is ultimately important in restitution is whether, and to what extent, the particular defendant has been benefited: see Professor Burrows, The Law of Restitution, 2nd ed, (2002), page 18. A benefit is not always worth its market value to a particular defendant. When it is not it may be unjust to treat the defendant as having received a benefit possessing the value it has to others. In Professor Birks' language, a benefit received by a defendant may sometimes be subject to 'subjective devaluation': An Introduction to the Law of Restitution (1985), page 413. An application of this approach is to be found in the Court of Appeal decision in Ministry of Defence v Ashman [1993] 2 EGLR 102. Whether this is to be characterised as part of the 'change of position' defence available in restitution cases is not a matter I need pursue.

Other submissions

The Inland Revenue pointed out that the general position under the United Kingdom tax regime, both for direct and indirect taxes, is that taxpayers who pay tax late are required to pay simple interest, and taxpayers who overpay tax are entitled to repayment together with an amount based on simple interest. Mr Glick QC submitted there is no compelling reason why the taxpayers in the present actions should receive more favourable interest treatment than taxpayers with other claims based on domestic law.

The point is not without force. But this is now water under the bridge. In Woolwich Equitable Building Society v Inland Revenue Commissioners [1993] AC 70 a similar submission was made regarding the availability of interest under section 35A, as opposed to the statutory repayment supplements. The point was upheld by Lord Keith of Kinkel and Lord Jauncey of Tullichettle. The majority of the House, however, took a different view. A restitutionary cause of action arose, with the usual consequences regarding interest, when the building society made the tax payments required by the ultra vires regulations. This was so, even though in the result the building society received more favourable interest treatment than other overpaying taxpayers

The Woolwich case concerned a claim to simple interest under section 35A. But on this point no sensible distinction can be drawn between an overpaying taxpayer's right to seek interest under section 35A and, as in the present case, an overpaying taxpayer's entitlement to an award of interest, simple or compound, as damages or as an element of substantive restitutionary relief.

The Inland Revenue also submitted that Sempra's restitutionary claim based on mistake stands apart from Sempra's other two causes of action. Sempra's claims for damages for breach of statutory duty and restitution in respect of tax paid pursuant to an unlawful demand are directly founded on the United Kingdom's breach of the Treaty. This is not so with the claim based on mistake. The claim based on mistake is founded on Sempra's own mistake. The fact that Sempra's mistake arose because of this country's breach of the Treaty is not part of Sempra's cause of action. This distinction, it was submitted, provides a principled justification for treating Sempra's mistake-based claim differently from its other claims so far as compound interest is concerned.

Here again this point has already been decided adversely to the Inland Revenue. The effect of the decisions of this House in Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 and Deutsche Morgan Grenfell Group Plc v Inland Revenue Commissioners [2006] 3 WLR 781 is that money paid by mistake can be recovered, whether the mistake is of fact or law. Money paid by way of tax does not stand on a different footing. In principle the restitutionary consequences are the same for tax payments made by mistake as they are for other payments made by mistake.

The seriously untoward consequences this may have for the Inland Revenue flow from the open-ended character of the extended limitation period prescribed by section 32(1)(c) of the Limitation Act 1980. Parliament has now recognised this extended period should not apply to payments of tax made by mistake: see section 320 of the Finance Act 2004.

The judge's order

The effect of the judge's order as clarified in the Court of Appeal was that the Inland Revenue should pay interest for the period of prematurity on a compound basis at a 'conventional' rate. No distinction was drawn between the rate applicable to the claim for damages and the rate applicable to the restitutionary claims.

This order should stand so far as it relates to the claim for damages. The reference to a 'conventional' rate was intended to dispense with the need for protracted investigation of the financial affairs of the parties and of other claimants. On the claims for damages a conventional rate should be taken to refer to the rate at which a substantial commercial company could borrow the amounts in question in the market at the relevant time.

The United Kingdom government can of course borrow more cheaply than commercial companies. With the restitutionary claims therefore the reference to a conventional rate should be taken to refer to the rate at which the government could borrow the relevant amounts in the market at the relevant times. The need for this distinction came to light as a result of the much fuller submissions on the law of restitution presented to the House.

I add one further note on the form of the judge's order. The order provided for payment of simple interest pursuant to section 35A for the period from the date of set off until judgment. Sempra did not challenge this provision. So this provision in the order will stand. But I am not to be taken as accepting that compound interest was unavailable for this period. Sempra's financial losses caused by payment of ACT did not wholly cease at the date of set off. Sempra remained out of pocket for the unpaid interest, and its financial losses in this regard continued to accrue up to judgment. Similarly, as to the restitutionary claims: after the date of set off the Inland Revenue continued to derive interest benefits from the benefits it had already obtained from having use of the ACT payments.

An effective remedy

I can now state my conclusion on whether English law provides an effective remedy for the United Kingdom's breach of article 43 of the Treaty. In my view it plainly does. For the reasons given above, compound interest is available under English law when quantifying the extent of Sempra's losses and when quantifying the extent of the Inland Revenue's unjust enrichment.

I would not refer a question to the ECJ for preliminary ruling. A reference to the ECJ is not necessary. It would serve no useful purpose. English law provides for compound interest to be awarded or taken into account when quantifying the financial remedies available to Sempra under the three asserted causes of action. There would be no point in seeking a ruling on whether Community law requires payment of compound interest. Subject to the variation in the judge's order mentioned above, I would dismiss the appeal.

My Lords,

Having had the great advantage, before writing this opinion, of reading the opinions of all my noble and learned friends I gratefully adopt their analysis of the previous case law regarding awards of interest on tortious, contractual and restitutionary claims and, since I cannot improve upon or usefully add to that analysis, I shall content myself with expressing my concurrence with the conclusion which appears to me to have been reached by all my noble and learned friends, that interest losses caused by a breach of contract or by a tortious wrong should be held to be in principle recoverable, but subject to proof of loss, remoteness of damage rules, obligations to mitigate damage and any other relevant rules relating to the recovery of alleged losses. That conclusion, however, does not dispose of this appeal. It deals with, and accepts, Sempra's interest claims to the extent that they are claims for compensation for loss. It does not deal with Sempra's claims to the extent that they are claims in restitution. As to those claims, I regret that I am unable to accept the conclusion of the majority of your Lordships that the restitutionary remedy allowed by the law of this country to a person who has made a payment under a mistake can entitle the claimant to recover not only the payment but also interest thereon that the person under the obligation to make repayment has never received or to recover the value of an assumed benefit derived from the mistaken payment that that person has never in fact enjoyed. On this issue I am in complete agreement with what my noble and learned friend Lord Mance has said. The conclusion of the majority, in my respectful opinion, confuses the remedy for a payment made under a mistake with the remedy for loss caused by a wrongful act. The wrongful act may be tortious, or a breach of contract, or it may be, as here, a breach of some statutory obligation. But, whichever it is, the claimant's cause of action is for loss caused by the wrongful act. Where, on the other hand, the cause of action is for recovery of money paid, or property transferred, by mistake, the remedy, if the cause of action can be made good, is one of recovery, or "disgorgement" as some of your Lordships have put it. There is no need to plead or prove a wrongful act on the part of the recipient against whom the remedy is claimed. If the money has actually earned interest, the claimant ought, in my opinion, subject to change of position defences, to be able to recover not only the money but also the interest. This is not because the interest belongs in law to the claimant. The claim is an in personam claim, not a proprietory one. It is a claim based on the unjust enrichment of the recipient who has received money he had no right to receive or, having received it, to retain. The remedy, therefore, subject to change of position defences, should restore to the claimant the extent of the unjust enrichment but no more. If interest had been earned on the money it ought, in my opinion, to be held to be prima facie recoverable as part of the restitutionary remedy for the unjust enrichment. Whether, if some other tangible benefit had been obtained by the recipient, that too, if quantifiable in money terms, should be accounted for to the claimant does not arise on this appeal and should be left to be decided when it does arise. But if interest had not been earned then, whether or not interest could or should have been earned, the restitutionary remedy should not, in my opinion, allow the recovery of anything other than the money itself. In such a case the remedy for the recovery of interest would not be a restitutionary one but a claim for compensation to the claimant for having been kept out of his money. It may be that in some cases a claimant would have both a restitutionary remedy for the return of money paid by mistake and a claim for compensation for loss caused by a wrongful act that had induced the mistaken payment. But if a claim is to be made for interest that cannot be, or has not been, shown to have been received by the recipient of the mistaken payment, the claim must, in my opinion, be prosecuted as a claim for compensation and, as such will be subject among other things to the rules applicable to such claims. These will include limitation of action rules. If the distinction between claims for compensation for loss on the one hand and claims for restitution on unjust enrichment grounds on the other hand is not recognised, incoherence of legal principle will, in my opinion, be the result.

A confusion between compensatory claims and restitutionary claims seems to me apparent in this case and starts with the judgment of the European Court of Justice (the ECJ) in the conjoined Metallgesellschaft and Hoechst cases [2001] Ch 620. This confusion appears to me to have been not the fault of the ECJ but a consequence of the way in which the questions put to the ECJ by the High Court had been formulated and to have been compounded by a misreading of the thrust of the ECJ's answers to those questions. The second of these questions asked whether, in the event that the ACT regime was not consistent with Community law -

The Advocate General re-formulated this question in paragraph 41 of his advice

and, in paragraph 45, expressed the principle underlying the remedy to be afforded to Sempra:-

unlawful charge

This passage seems to me a very important one. A member state must not profit from the levying of an unlawful charge and the person required to pay the unlawful charge must not suffer from having done so. This is language describing the approach to a remedy for a wrongful act. It is not language apt for describing a remedy for money paid by mistake. The Advocate General then went on to cite a highly relevant passage from the Dilexport Srl v. Amministrazione delle Finanze dello Stato (Case C-343/96) [1999] ECR I-579, para 25:

This citation was repeated by the ECJ at para.85 of its judgment.

The clear thrust of this is that provided under domestic law a remedy is available to subsidiaries such as Sempra that prevents the infringing member state from profiting from its wrongful act and does not prevent the subsidiary from recovering the loss caused to it by the wrongful act, and provided the remedy satisfies the principles of equivalence and effectiveness, Community law will be satisfied.

However the UK Government had submitted to the ECJ that under English law "no action for interest in respect of the loss of the use of monies which were ultimately set off against the paying company's corporation tax liability would lie" and that it was for national law not Community law to determine whether interest should be paid in respect of the prematurely paid ACT (see para.46 of the Advocate General's opinion and para.79 of the ECJ's judgment). The ECJ made clear that this submission could not be accepted. It stressed that "it is not for the Court of Justice to assign a legal classification to the actions brought by the claimants before the national court" (para.81) but then went on to consider the scope of the remedy required by Community law to be made available "first on the assumption that the actions … are to be treated as claims for restitution of a charge levied in breach of Community law" (para.82) and "secondly, assuming that the claimant's claims are to be treated as claims for compensation for damage caused by breach of Community law" (para.90).

This is the point, in my opinion, at which confusion creeps in. The "restitution" of which the ECJ was speaking was restitution required in consequence of the levying of a tax charge in breach of Community law. But it was not restitution of the tax itself that was in question. The obligation of the Revenue to repay the tax was not in doubt. Most of that tax had been "repaid", before the commencement of recovery proceedings, by means of set-off against the mainstream corporation tax that had become due. In paragraph 82 the ECJ said that the question was whether the breach by the member state of Community law

"Re-imbursement"? The use of this word suggests that the ECJ had in mind a loss suffered by the taxpayer. In paragraph 86 the ECJ referred to "the re-imbursement of charges improperly levied". Here, by contrast, the word "re-imbursement" was clearly referring to the repayment of the improperly levied charge. But in paragraph 87 the ECJ said this:

covering

the cost of loss of the use

This passage recognises the compensatory nature of the claim and the use of inverted commas in relation to "re-imbursement" underlines the point that an award of interest to the subsidiaries who had prematurely paid ACT would not be re-imbursement in the ordinary sense of the word. The award of interest to cover the cost of the loss of use of the sums paid by way of ACT would not constitute the restoration to the taxpayer of something that the taxpayer had previously had. It indicates an intention that the taxpayer should be put in the position in which the taxpayer would have been if the ACT had not been paid. This is compensation for "loss of use" of the money, not restitution.

The confusion between the two concepts is apparent again in paragraphs 88 and 89

would have been generated by the sum,

use of which was lost as a result of the premature levy of the tax.

and that sum may be claimed by way of restitution

The emphasised words in paragraph 88 show again that the ECJ had in mind compensation to the subsidiary for loss. Paragraph 89 says that that sum "may be claimed by way of restitution". But these paragraphs proceed on the assumption referred to in paragraph 82, namely, that the claims for recovery of interest were to be treated as claims for restitution. The ECJ did not say, and it would have been quite contrary to its own well established jurisprudence for it to have said, that the actions for payment of interest had to be treated as claims for restitution (see paragraph 196 and 197 of Lord Mance's opinion and the cases there cited). It is for domestic law to classify the actions.

The confusion becomes the more stark when the passages from the ECJ judgment to which I have referred are examined in the context of the type of restitutionary action on which Sempra is relying. In the Deutsche Morgan Grenfell Group case [2006] UKHL 49; [2007] 1 AC 558 your Lordships held that an action for restitution could lie where money had been paid under a mistake of law. This was a clarification, or perhaps a development, of previous case-law which had restricted restitutionary actions for recovery of money paid by mistake to cases where the operative mistake had been a mistake of fact. Sempra, it is said, paid the ACT under a mistake of law, that is to say, in ignorance that the ACT statutory regime was inconsistent with Community law and accordingly unlawful. It is common ground that this circumstance entitles Sempra to compensation for the loss caused to it by the unlawful levying of ACT. The payment of compensation is not a restitutionary remedy; it is a remedy for tort or for breach of statute (it does not matter which description is to be preferred). The ECJ cannot, in my opinion, be taken to have intended to reform our domestic law on restitution. Suppose this House in the Deutsche Morgan Grenfell case had ruled that a remedy for money paid under mistake of law was not available under domestic law. Would it be said that that ruling would have been contrary to Community law? That could only be said if the alternative domestic law remedy, a remedy allowing compensation for loss caused by unlawful conduct, were in some respect or other inadequate to comply fully with the Community law requirement that domestic law provide a full remedy "covering the cost of loss of the use of the sums paid by way of ACT" (para.87), essentially a requirement for a compensatory remedy. And on what coherent basis of domestic law can a restitutionary remedy for money paid under mistake include an award of the interest "which would have been generated by the sum, use of which was lost as a result" (para.88 of the ECJ judgment) of the payment? In many, perhaps most, cases the recipient of money paid under a mistake will have committed no wrongful act at all. So why should the recipient compensate the payer for the payer's loss? On what basis should a recipient who has received no interest be made liable to "disgorge" interest? As a matter of principle, in my opinion, an innocent recipient cannot be required to disgorge something that the recipient has never had.

In summary, the ECJ was concerned in the Metallgesellschaft/Hoechst cases to ensure that domestic law provided a full remedy that took account of the consequences both to the Revenue and to the taxpaying subsidiary of the premature payment of ACT. It was not concerned to reform the law of restitution so as to ensure that a sufficient remedy in restitution as well as a sufficient remedy for compensation was available to the subsidiary.

The reality of the case being put forward by Sempra, and the other subsidiaries, is that they were constrained to pay ACT in reliance on a statutory scheme that was not compliant with Community law and was therefore unlawful, and that they should be compensated accordingly. This is a tort/breach of statute claim. Once the claim for the return of the ACT has been satisfied, the residue of the claim, i.e. the claim for interest, is not a claim for restitution other than in a highly contrived sense - hence the use by the ECJ of inverted commas around the word "re-imbursement" in paragraph 87. The re-modelling of the law relating to restitution is not necessary unless the compensation remedy is, in some respect or other, inadequate to meet Community law requirements.

The law relating to monetary awards in actions where the claim for the award is based on a wrong for which the defendant is responsible has itself become, in my opinion, somewhat incoherent. The textbooks and the cases speak of compensatory damages, aggravated damages, exemplary damages, punitive damages, restitutionary damages and even, in the United States, curative damages. These adjectives sometimes mask the legitimate purposes for which damages may be awarded. But where A has committed a wrong as a result of which B has suffered a loss or has been deprived of something of value, then, provided the wrong gives B a cause of action, B can claim compensatory damages. The purpose of the compensation is, subject to remoteness of damage rules, to place B in the position in which he would have been if the wrong had not been committed. Where a wrong committed by A has caused B to be wrongfully deprived for a period of time of a sum of money, the extent of the loss should be measured by the value to B of that sum for that period. There may be actual evidence of what B would have done with the money. The actual evidence may show that B has lost nothing. On the other hand it may show that he has lost a great deal, but of course some of that loss may be too remote to enable the claim for its recovery to succeed. At the least it can usually be said that by being deprived of the money B has lost the opportunity of leaving it on deposit at the bank for the period in question. On that sensible footing B can claim compensation measured by the interest the sum would have produced if simply left with the bank for that period. Interest would have accrued at the bank's usual rate of interest on deposits and with the usual rests allowed by the bank's terms. An award of interest compounded in the manner referred to would result, in my opinion, from an ordinary application of legal principles applicable to the assessment of compensatory damages for tort or breach of statute. The application of these principles would satisfy the ECJ requirement that domestic law provide a full remedy covering the loss of use of the sums paid by way of ACT.

The Advocate General in Metallgesellschaft/Hoechst referred also to the need for the domestic remedy to prevent the member state in question from making a profit out of its breach of Community law (para.45). Compensation assessed on the basis I have referred to, would satisfy the need referred to by the Advocate General.

It is certainly the case that where alternative remedies are available to a claimant, the claimant can choose which of the remedies to pursue. And it is the case that, since your Lordships have held (in the Deutsche Morgan Grenfell case) that an action for restitution in respect of money paid under a mistake of law is available to Sempra, that Sempra can choose whether to pursue its claim for interest via an action for compensation for loss or an action for restitution. There is, however, as it seems to me, a problem standing in its way if a restitutionary action is its preferred vehicle. The problem is the nature of its pleaded case.

Sempra's writ, as amended, claimed

This, therefore, was a combination of a claim for restitution of money paid pursuant to a demand (unlawful as the ECJ had held) by the Revenue or paid under a mistake of law, and a claim for compensation for loss of use of the money so paid. But the claim for restitution was supplemented by the following paragraph:

loss of use of sums so paid

It is apparent, therefore, that the claim, whether expressed as a claim for restitution or as a claim for compensation, was based on Sempra's "loss of use" of the money it had paid to the Revenue as ACT. It was not based on the "unjust enrichment" of the Revenue in having received that money prematurely. I would accept that a claim to recover interest on the money prematurely paid that the Revenue had actually obtained, or, perhaps, to recover the value of a saving actually achieved by the Revenue in not having to borrow equivalent sums, would prima facie be acceptable in a restitutionary remedy. This would be a recovery by Sempra of an actual benefit the Revenue had obtained from the premature payments. It would not represent Sempra's "loss of use of sums so paid". But there was no evidence that the Revenue derived any such actual benefit from the premature payments. The cited paragraph taken as a whole describes a claim for compensation dressed up, for limitation of action reasons, as a claim in restitution. The claim as described could, in my opinion, only succeed if it were based on a wrongful act. It constitutes a tort claim not a restitutionary claim. Sempra's Re-Re-Amended Statement of Claim pleads not only a case of compensation for tort/breach of statute but also that in making the ACT payments Sempra was mistaken as to the validity of the ACT statutory regime and would not otherwise have paid the ACT. Paragraph 11C then pleads as follows:

The paragraph 11C allegations about investing and making savings were either denied by the Revenue or were not admitted. The Revenue said they were required by statute to place their revenue receipts in a Bank of England account over which the Treasury had control (see para.7D of the Re-Amended Defence). Park J, in his judgment, examined the interest issue solely from the point of view of Sempra (see paras.31 to 43). He asked himself, in effect, what Sempra had lost by its loss of use of the ACT. He took the view that the ECJ judgment required interest to be awarded in respect of that loss whether the claim was for compensation or was a restitutionary claim. This, in my respectful opinion, was a misreading of the Metallgesellschaft/Hoescht judgment. Because of this the judge did not ask himself whether a benefit had been obtained by the Revenue from the mistaken ACT payments, whether that benefit was quantifiable in money terms, and, if so, how it should be quantified. The same features are apparent in Chadwick LJ's judgment in the Court of Appeal. In paragraph 49 he said, by way of summary, this -

the loss suffered by the taxpayer

full compensation for the loss of the use

compensation

This may be unexceptional and correct in relation to an action for compensation. It is not, in my opinion, in the least apt in an action for restitution. In an action of restitution it is the position of the recipient of the mistaken payment, and the benefit, if any, that the recipient has obtained from the money, on which attention should be concentrated.

The majority of your Lordships have taken the view that the Revenue's unjust enrichment consists of the benefit of having had Sempra's money available for use for the period between payment of the ACT and set-off (see Lord Nicholls of Birkenhead para 66) and that the domestic law on restitution should be developed so as to allow a restitutionary claim for interest, including compound interest, in order to place an objective value on that benefit. My Lords I find myself unable to accept the premise on which this proposed development is based, namely that the mere possession of mistakenly paid money - and accordingly the ability to use it if minded to do so - is sufficient to justify not simply a restitutionary remedy for recovery of the money, but a remedy also for recovery of the wholly conceptual benefit of an ability to use the money. Why should a restitutionary remedy be concerned with a benefit that is no more than conceptual? The mistake that had been made and on which the restitutionary claim is based was the mistake of the payer, Sempra in the instant case. The restitutionary claim is not based on any wrongdoing on the part of the recipient, the Revenue in the instant case. If it were so based it would be a compensation claim attracting a different limitation period. So it is not so based. I therefore repeat the question, differently phrased - why should an innocent recipient of a payment that the claimant has made by mistake be required to restore to the claimant anything additional to the amount of the payment and any actual benefit that the recipient has obtained from the payment?

My noble and learned friend Lord Nicholls, in paragraphs 116 to 118 of his opinion, has treated the "time value of money, measured objectively." (para 117) as the measure of a benefit of which restitution should be made and, accordingly, as the basis on which the Revenue should be required to make restitution to Sempra of the benefit of receiving the "huge interest free loan constituted by Sempra's payment of ACT" (para 118). But what is the value of a loan if the recipient has done nothing with it, has derived no actual benefit from it and has paid it back to the lender? An additional payment required to be made by the payee in those circumstances is not restitution; it is compensation to the payer for being out of his money. The cases referred to by Lord Nicholls in paragraph 116, and referred to also by Lord Mance in paragraph 230 of his opinion, all concerned compensation to the victim for a wrongful act that had caused no actual loss to the victim but had provided a benefit to the wrongdoer. They were cases where the compensation held to be payable to the victim was measured by the value of the benefit taken by the wrongdoer. These cases established a valuable and coherent principle of compensation for cases in which the victim of a wrong had suffered no actual loss but, in my respectful opinion, have nothing to do with cases where there is no wrongdoer, no case for compensation, but simply the need for restitution of the extent of an unjust enrichment that has been received. If there has been no enrichment, the mistaken payment having been restored, the question of what is just or unjust does not arise.

In my opinion, the inquiry that should have been made in the present case was whether the Revenue had derived any actual benefit from the premature receipt of Sempra's money. The rate at which the Government can borrow in the market at the relevant times would be, in my opinion, irrelevant unless there were evidence to justify the conclusion that the Government's borrowing in the market was less because of that premature receipt than it would otherwise have been. There was no such evidence.

Accordingly, in my opinion, this is not a case in which, on the evidence before Park J, any interest, whether simple or compound, should be ordered to be paid by the Revenue as part of a restitutionary remedy to Sempra for its mistaken payment of ACT. I agree that, in relation to Sempra's cause of action in tort, Sempra is entitled to recover its losses brought about by the Revenue's unlawful demand for ACT and that those losses could well include compound interest on the sums paid. But, as is common ground, Sempra's tort remedy is, for Limitation Act reasons, abbreviated.

Summary

In my opinion the remedy which enables Sempra, and other subsidiaries in the same position, to be compensated for the loss caused by premature payment of ACT is a remedy in tort or for breach of statute. That remedy complies fully with the requirements of domestic law demanded by the ECJ in the Metallgesellschaft/Hoechst case. A remedy for recovery of money paid under a mistake of law cannot lead to compensatory damages unless it is also based on a wrongful act. To the extent that it is so based it will be an action in tort or for breach of statute. The claim to recover interest based upon the interest that would have been earned if the money used to pay the ACT had instead been placed on deposit with a bank, or upon the interest that would have been saved if the money had been used to repay borrowings, is a claim for compensatory damages. To dress that claim up as a restitutionary remedy for money paid under a mistake of law is unacceptable. To alter the claim to one requiring disgorgement by the Revenue of the benefit it has derived from the premature payments is, in principle, acceptable but fails for the want of any evidence that the Revenue did derive any benefit from the payments. A benefit cannot be simply assumed. Mere possession of money for a period is not a benefit to be valued and disgorged. I would, therefore, dismiss the appeal so far as it relates to Sempra's claim in tort but allow the appeal so far as it relates to Sempra's claim in restitution.

There is a final point to which I should refer, namely, whether, if the law does allow Sempra a claim to interest as a restitutionary remedy, the remedy should be regarded as a remedy as of right or as an equitable, and therefore discretionary, remedy.

The discussion about whether interest on money paid by mistake can be recovered as part of a restitutionary remedy has led your Lordships to consider claims to interest on debts, on contractual damages, and on tortious damages as well as on money paid by mistake. I concur with your Lordships in concluding that interest, whether simple or compound, can represent an item of contractual damages or tortious damages, subject to the normal rules applicable to such claims. I take the view that a claim to interest as part of a restitutionary remedy for money paid by mistake can and, subject to change of position defences should be, accepted if the interest has actually been earned, but not otherwise.

As to interest on debts, Parliament has, in section 35A of the Supreme Court Act 1981, provided a discretionary power for the court to award simple interest where proceedings are brought for the recovery of the debt, or for the recovery of contractual or tortious damages. This power enables interest to be awarded notwithstanding that the interest could not, for remoteness of damage or other reasons, qualify for award as an item of damages. As Lord Mance has pointed out (para. 221 of his opinion), the Arbitration Act 1996 conferred a discretionary power on arbitration tribunals to award, unless the parties had otherwise agreed, simple or compound interest on sums awarded in the arbitration and the Late Payment of Commercial Debts (Interest) Act 1998 provided, in certain circumstances, a right to simple interest on certain commercial debts. Moreover, where statutory debts are concerned - fiscal debts owing to the Revenue and council tax owing to local authorities are obvious examples but there are many others - payment of interest on debts that are overdue for payment is either provided for by statute or cannot be recovered. In these circumstances it seems to me that it would be inappropriate for your Lordships, acting judicially, to seek to improve on what Parliament has enacted regarding the payment of interest on unpaid debts.

Lord Mance has made clear in his opinion that recovery in a claim in restitution for money had and received of interest on the sum of money in question should be "contingent upon proof of actual benefit acquired by the recipient". I respectfully agree, but am unable to agree that that recovery should be regarded as discretionary. The recovery is, under established common law rules, subject to change of position defences. What, in a particular case, would constitute a sufficient change of position to enable the recipient to resist an order for recovery of interest actually earned, or of the value of some other benefit acquired by the recipient by use of the principal money, would be a subject of judgmental decision in each case. The decision might, in the ordinary use of language, depend on where the judge thought the equities lay, what the judge thought was fair (see the discussion about change of position in Goff & Jones The Law of Restitution 7th ed (2007) Chap. 40). But this would not turn the remedy into a discretionary one. In my opinion, the restitutionary remedy, a common law remedy, has all the ingredients for a fair and just result to be reached. I can see no need for an ancillary equitable jurisdiction and your Lordships should not, in my opinion, construct one.

LORD WALKER OF GESTINGTHORPE

My Lords,

Introduction

I have had the privilege of reading in draft the opinions of my noble and learned friends Lord Nicholls of Birkenhead and Lord Hope of Craighead. I am essentially in agreement with them, and I too would dismiss this appeal, largely for the reasons which they give. But because of the importance of the issues before the House I add some observations of my own.

The issue in this appeal is the principles to be applied in quantifying the award to be made to Sempra Metals Ltd ("Sempra"), by way of damages or restitution, for the detriment that Sempra suffered as a result of the United Kingdom's corporation tax regime being contrary to EU law. That the regime did contravene EU law was established by the Court of Justice ("the ECJ") in a judgment delivered on 8 March 2001 in Metallgesellschaft Limited v IRC and Hoechst AG v IRC (joined cases C-397/98 and C-410/98) ("Hoechst") reported at [2001] Ch 620. Sempra is the same company, under a new name, as Metallgesellschaft Limited.

The precise issue originally before the Court is set out in para 9(ii) of a Group Litigation Order ("the GLO") made by the Chief Chancery Master on 26 November 2001. The issue is subdivided into two elements:

It can be said at once that the answer to the second element of the issue is not in dispute. Park J decided that the interest should be simple interest under section 35A of the Supreme Court Act 1981 (as inserted by the Administration of Justice Act 1982), leaving the rate or rates to be determined later if the parties could not agree. There is no appeal from that (though I agree with Lord Nicholls' observation, in para 129 of his opinion, that Sempra might have challenged it). It is the first element of the issue that has given rise to acute controversy. Park J decided ([2004] STC 1178) that the calculation should be effected on the basis of compound interest (with the rate or rates and appropriate rests to be determined later if the parties could not agree) and his decision was upheld (subject to a small drafting amendment) by the Court of Appeal ([2005] STC 687). That is the issue which is now before your Lordships' House.

The quantification issue is only one of several important issues, with far-reaching financial implications, that have been or are being litigated under the GLO. There are also other group litigation orders raising similar points in connection with Article 43 or Article 56 of the Treaty. The background of the statutory material as to various group reliefs (and in particular group income elections postponing liability for payment of ACT) is by now very familiar to practitioners in this field. It has been covered in many judgments, the most recent being the speeches in this House in Deutsche Morgan Grenfell Group plc v IRC [2007] 1 AC 558 ("DMG"). It is unnecessary to repeat the statutory material in any detail. For present purposes the essential point is that a United Kingdom subsidiary which paid a dividend to a holding company established in another member state was deprived of the opportunity of making a group income election and so deferring payment of ACT until mainstream corporation tax ("MCT") became payable on the subsidiary's profits. This period would normally be between eight and a half and seventeen and a half months in duration (Peter Gibson LJ gave a full explanation of this in Pirelli Cable Holding NV v IRC [2004] STC 130, para 8) but the deferment might be longer if (because of double taxation relief or for some other reason) the subsidiary had insufficient taxable profits to attract liability to MCT at least as great as the ACT already paid. When sufficient MCT did sooner or later become payable (and was paid before the issue of any writ) what the subsidiary lost was the time value of what it had paid in ACT for a period which had a minimum duration of eight and a half months, but could in exceptional circumstances continue for several years. In this case the shortest period was just under one year and the longest was almost ten years (the details are set out in paras 62 and 63 of Lord Nicholls' opinion).

The hearing before Park J occupied three days. This may seem a surprisingly short period since (as your Lordships were told) there was a very large volume of evidence in the form of witness statements from expert and non-expert witnesses and copious documentary evidence of a financial nature. In the event, however, the expert witnesses were not cross-examined, and oral evidence from the non-expert witnesses was deferred. That was because the judge, at the urging of Mr Glick QC for the Revenue, and without strong opposition from Mr Rabinowitz QC for Sempra, decided that the amount of any damages or restitutionary award should be computed on what Mr Glick called a "conventional" basis—that is, on general principles rather than by reference to the particular financial circumstances of the claimant (or the financial circumstances, so far as they sensibly admit of enquiry, of the Revenue): see [2004] STC 1178, paras 13 and 18 to 22. It seems that the judge may have been given an over-estimate of the number of other cases which raise this issue: between 50 and 70 was, your Lordships were told, much too high a figure. The judge's decision that the award should be quantified on a "conventional" basis was not a ground of appeal before the Court of Appeal. But this point has to some extent been reopened before your Lordships.

Should there be another reference under Article 234 EC?

The first matter that your Lordships have to consider is whether to make a reference to the ECJ under Article 234 EC. At the beginning of the appeal hearing the House decided not to make an immediate reference but to proceed to hear argument on the substantive issues raised in the appeal. Having heard the argument I feel sure that was the right decision, and that your Lordships should resolve this appeal without a further reference.

As I have already noted, Sempra is the same company, under a new name, as Metallgesellschaft Ltd, the plaintiff in the proceedings (Ch 1995 M 7327) commenced by writ on 23 November 1995. On 2 October 1998 Neuberger J referred five questions to the ECJ under Article 234 EC. The ECJ's judgment of 8 March 2001 ([2001] Ch 620) provided detailed answers to the first, second and fifth questions. The answer to the second question (in paras 77 to 96 of the judgment) has already given the ECJ's considered view on the central issue in this appeal. This part of the judgment is not without its difficulties, as the argument before your Lordships has amply demonstrated. It will be necessary to make detailed reference to the ECJ's judgment at a later stage. But I venture to suggest that its general thrust is that causes of action and remedies for any breach of EU law are matters for national law and the national courts, subject always to the very important EU principles of equivalence and effectiveness. The ECJ spelled out some of the implications, in relation to remedies, of the principle of effectiveness. In my opinion it would serve no useful purpose to make a further reference seeking fuller elucidation of the principles stated in paras 77 to 96 of the judgment, or practical guidance as to how those principles should be implemented. I think it is predictable that on a further reference the answer would amount to quod scripsi, scripsi.

I appreciate that the questions put to the ECJ in Hoechst did not refer in terms to compound interest. But had they done so the answer would surely have been that any choice between simple and compound interest was for the national court, subject always to the principles of equivalence and effectiveness. There is the complication that the Court of Appeal has on 6 February 2006 made a reference to the ECJ in a value added tax case, British Telecommunications plc v Commissioners for HM Revenue and Customs C-185/06, raising three questions in relation to tax paid by mistake but voluntarily (and not as a result of any unlawful demand or an unlawful tax regime) in circumstances where the tax authority did not contribute to the mistake and had no reason to be aware of it (the mistake was in BT's accounting software). The third of the questions relates specifically to simple or compound interest. However your Lordships were told (in the course of the further oral argument on 16 May 2007) that the reference is likely to be withdrawn. In any case it is speculative whether the ECJ's answer to the third question (if it arises, which is far from certain) will assist in the resolution of the appeal now before your Lordships, because both the facts and the legal issues are so different. In my opinion we must make what we can of the guidance already given by the ECJ in this very case.

Interest under national law

It is common ground, since this House's decision in DMG [2006] 3 WLR 781, that domestic law offers Sempra three possible causes of action in respect of its premature payments of ACT under a tax regime which contravened EU law. These are

(1) an action for damages (in the nature of an action for breach of statutory duty: see R v Secretary of State for Transport ex parte Factortame Ltd (No. 7) [2001] 1 WLR 942, paras 143-158);

(2) a restitutionary claim for repayment of tax unlawfully exacted;

(3) a restitutionary claim for money paid under a mistake of law.

All these claims are put forward in Sempra's much-amended statement of claim. The ECJ has made clear in Hoechst that the appropriate remedy is to be determined by the national court (and therefore, so far as national law permits a claimant to choose between different causes of action available to him, by the claimant). The Revenue's position is that whichever of the three roads Sempra chooses to follow, it cannot lead to an award of more than simple interest. Nothing in the ECJ judgment, the Revenue argues, requires compound interest. The Revenue argues with particular vigour that compound interest cannot be awarded if Sempra follows the mistake of law route.

In Hoechst the ECJ deliberately avoided taking up any position as to the content of national law (the Advocate-General had in his opinion, paras 12 and 46, referred to President of India v La Pintada Compania Navigacion SA [1985] AC 104 —"La Pintada"—but the ECJ did not go into any of that detail). But it is impossible to apply the principles of equivalence and effectiveness without first considering the present state of English law as to the award of interest (either as interest on a principal sum, or as damages, or as a restitutionary remedy). So in my view your Lordships have no alternative but to embark on that burdensome task. The burden is to some extent lightened since your Lordships have the benefit of counsel's careful and wide-ranging researches, reflected in the judgments below in this case, and also in the judgments given in the Court of Appeal in NEC Semi-Conductors Ltd v Inland Revenue Commissioners [2006] STC 606 (especially Mummery LJ at paras 163-175). It is nevertheless an unwelcome task. The Law Commission has recently expressed the view (Pre-Judgment Interest on Debts and Damages Law Com. No. 287, para 1.15) that:

There are also some pertinent observations by Lord Goff of Chieveley in Westdeutsche Landesbank Girozentrale v Islington LBC [1996] AC 669, 684 ("Westdeutsche"):

In his opinion Lord Nicholls has traced the history of the rule in Page v Newman (1829) 9 B&C 378 since it was first stated, in wide and blunt terms, by Lord Tenterden CJ (at 380):

My noble and learned friend has described how this questionable rule survived reviews by this House in London, Chatham and Dover Railway Co v South Eastern Railway Co. [1893] AC 429, La Pintada [1985] 1 AC 104 and Westdeutsche [1996] AC 669, but has been progressively eroded by statutory intervention and by judicial recognition that the rule does not exclude the award of interest as special damages. The point was aptly expressed by Ward LJ in Hartle v Laceys [1999] Lloyds Reports PN 315, 327:

The thrust of Lord Nicholls' observations on interest benefits and restitution (paras 101 et seq of his opinion) might be encapsulated similarly as being 'not about interest on restitution but about interest as restitution.'

I respectfully agree with Lord Nicholls that the distinction which Lord Brandon drew in La Pintada between damages under the two limbs in Hadley v Baxendale (1854) 9 Exch 341 is unsound, and should no longer be used (see the observations of Mason CJ and Wilson J in the High Court of Australia in Hungerfords v Walker (1989) 171 CLR 125, 141-142). I cannot usefully add anything about interest as damages. But I wish to add something about interest on (or as) a restitutionary award.

Interest in Unjust Enrichment

It is a topic which has until recently attracted relatively little attention from English restitution scholars. The most extensive treatment may be the chapter entitled "Interest" by Professor Francis Rose in Birks & Rose, eds. Lessons of the Swaps Litigation, (2000) pp291-328. There is also a useful survey by Professor Andrew Burrows in his The Law of Restitution (2nd ed, 2002) pp53-56. English and Australian authorities on the topic are considered in chapter 28 of Mason & Carter, Restitution Law in Australia (1995) pp945-967, and in part 1 of Edelman and Cassidy, Interest Awards in Australia (2003) pp92-115. Like Lord Hope I have been assisted by an article by Steven Elliott, "Rethinking Interest on Withheld and Misapplied Trust Money" (2001) 65 Conv.313. I gratefully acknowledge my debt to all these materials.

If Moses v Macferlan (1760) 2 Bur 1005 is recognisably the first leading case in English restitution law, then the first monograph on English restitution law can probably be identified as the work by Sir William Evans entitled "An Essay on the Action for Money Had and Received". It was published in 1802 and dedicated to Sir Edward Law (later Lord Ellenborough). It is reprinted in [1998] RLR 3. In its opening paragraphs, Evans identified the subject-matter of his study as

Evans quoted as a "proper introduction" to the subject the famous passage from the judgment of Lord Mansfield CJ in Moses v Macferlan (at p 1012):

ex aequo et bono

The last chapter (IX) of the essay is concerned with damages (which must be understood as meaning restitutionary awards). It provides a convenient starting-point for considering interest on restitutionary awards. After referring to Dutch v Warren (1721) (best reported within Moses v Macferlan (1760) 2 Burr 1005, 1010) Evans stated:

Walker v Constable

Walker v Constable (1798) 1 Bos & Pul 306 was a claim for the return (with interest) of a deposit paid under an abandoned contract. The relevant part of the report of the decision of the Court of Common Pleas is as follows (at p 307):

Moses v Macferlan

In Moses v Macferlan, Lord Mansfield did not make any reference to interest as such. But the report shows that he was presented with arguments that the count for money had and received (which was, for procedural reasons, a popular form of action) should not be extended too far. In that context he explained (at p 1010) that the count for money had and received had advantages for the litigants on both sides:

ex aequo et bono

Mr Rabinowitz submitted that Lord Mansfield's reference to the defendant being liable "no further than the money he has received" did no more than exclude any consequential loss. The fact that the judgment went on to refer to Dutch v Warren supports this. In any case the practice of not awarding interest on the count for money had and received seems to have been partly procedural. It was also strongly influenced by the rule in Page v Newman, under which the Court looked for an express or implied term as to interest. This appears from de Havilland v Bowerbank (1807) 1 Camp 50 where the report shows that the point was reconsidered by Lord Ellenborough at the instance of the Attorney-General (at p 52):

The same rule was reaffirmed by Tindal CJ and the other judges of the Court of Common Pleas in Fruhling v Schroeder (1835) 2 Bing (NC) 78. Again the rule in Page v Newman (although not mentioned as such) seems to have been decisive (at p.79):

A similar approach was taken in an early 20th century appeal to the Privy Council from Sierra Leone, Johnson v The King [1904] AC 817. Johnson was a contractor who had grossly overcharged the government for his services in moving granite to a naval installation in Freetown. He presented vouchers for moving over 4,000 tons of granite when he had actually moved less than 450 tons. He was sued for the amount overpaid, a sum of just over £8,000, together with interest. But although he had by then been prosecuted, convicted and imprisoned, and though fraud was pleaded in the civil claim, at trial the case was not rested on fraud. The Privy Council held that that was fatal to the claim for interest (Johnson had paid into court the principal sum claimed). Lord Macnaghten stated (at p.821):

London, Chatham and Dover Ry Co v South Eastern Ry Co

Again, a principle applicable to contractual debts was being applied to a restitutionary claim, because it was treated as quasi-contractual in nature (this point is well made in Edelman and Cassidy Interest Awards in Australia, part 1 p 93).

My Lords, the constant search, in these old cases, for an express or implied term as to interest reflects the fact that unjust enrichment was for a very long time regarded as a sort of appendage to the law of contract. That is how Lord Mansfield put it in Moses v Macferlan (1760) 2 Bur 1005, 1008,

quasi ex contractu

It was therefore natural that the rule as to interest on contractual debts should be applied to restitutionary awards. The term "quasi-contract" should not have survived Lord Wright's great speech in Fibrosa Spolka Akcyjna v Fairbairn Lawson Combe Barbour Ltd [1943] AC 32, 61, though in fact it lingered on for another generation.

It may have received its final quietus with the judgment of Robert Goff J in BP Exploration Co (Libya) Ltd v Hunt (No 2) [1979] 1 WLR 783 (upheld by this House [1983] 2 AC 352). In his judgment Robert Goff J, as well as restating the principles underlying unjust enrichment, decided that the wording of section 3(1) of the Law Reform (Miscellaneous Provisions) Act 1934 was wide enough to include interest on restitutionary awards, so that there is a statutory power (now found in section 35A of the Supreme Court Act 1981) to award simple interest, at the Court's discretion, on a restitutionary award. That was plainly a step forward, but it may have distracted attention from considering where (apart from statute) restitutionary interest stood once unjust enrichment had been set free from the fettering fiction of quasi-contract.

The Court of Chancery was more generous in giving interest on restitutionary awards, as appears from the judgment of Malins VC in In re Maria Anna and Steinbank Coal and Coke Company (McKewan's case) (1877) 6 Ch D 447, 455-456 (upheld by the Court of Appeal, without a reasoned judgment on this point, at p 462). Interest was ordered as of course against a fiduciary agent who failed to account (Harsant v Blaine, Macdonald & Co (1887) 56 LJQB 511, followed by the High Court of Australia in Bayne v Stephens (1908) 8 CLR 1). But at equity too there were anomalies: in particular a personal restitutionary claim to recover a mistakenly paid legacy seems not to have carried interest, although a proprietary restitutionary claim did so: see Re Diplock; Diplock v Wintle [1948] Ch 465, 506-507 (following, on the personal claim, Gittins v Steele (1818) 1 Swanst 199, a decision of Lord Eldon LC) and at 558 (on the proprietary claims). Gittins v Steele was discussed by Lord Goff and Lord Woolf in Westdeutsche [1996] AC 669, 694, 730.

In short the state of English law as regards interest in respect of restitutionary claims (considered as a genus) has been at least as confused as that in respect of compensatory claims. It is regrettable that (for reasons adverted to by my noble and learned friend Lord Mance, whose opinion I have had the advantage of reading in draft) the whole subject was not more fully investigated in Westdeutsche. In the event, this House's decision to depart from Sinclair v Brougham [1914] AC 398 (a possibility not even hinted at, so far as I can see, in the Court of Appeal [1994] 1 WLR 938) threw the argument on interest into disarray, and the three members of the House who decided not to develop the equitable jurisdiction to award interest relied strongly on the absence of full argument (see Lord Browne-Wilkinson at p 718, Lord Slynn of Hadley at pp 718-719 and Lord Lloyd of Berwick at pp 738-739).

The crucial insight in the speeches of Lord Nicholls and Lord Hope is, if I may respectfully say so, the recognition that what Lord Nicholls calls income benefits are more accurately characterised as an integral part of the overall benefit obtained by a defendant who is unjustly enriched. Full restitution requires the whole benefit to be recouped by the enriched party: otherwise "the unravelling would be partial only" (Lord Nicholls in Nykredit Mortgage Bank plc v Edward Erdman Group Ltd [1997] 1 WLR 1627, 1637).

That was a case where money paid in damages had to be refunded in consequence of an appellate judgment. The same principle has been applied by differently constituted divisions of the Court of Appeal of New South Wales and by the Full Court of the Supreme Court of South Australia (Heydon v NRMA Ltd (No 2) (2001) 53 NSWLR 600; Roads and Traffic Authority v Ryan (No 2) [2002] NSWCA 128 (16 May 2002); Cornwall v Rowan (No 2) [2005] SASC 122 (1 April 2005)). In the first of these cases Mason P (at pp 604-606) cited from his judgment in National Australia Bank Ltd v Budget Stationery Supplies Pty Ltd (23 April 1997, unreported). Having set out a long catalogue of cases in which the London, Chatham rule had been bypassed, Mason P continued:

London, Chatham

He also noted the doubts as to a "free-standing" right to interest expressed in the High Court in Commonwealth of Australia v SCI Operations Pty Ltd (1998) 192 CLR 285, 316-7.

I wish to add a word about terminology. In the course of argument the expression 'disgorgement interest' was frequently used, sometimes (if I understood counsel correctly) as a synonym for restitutionary interest, sometimes as a species of restitutionary interest, and sometimes in contrast to restitutionary interest. It is hard to make progress through that sort of confusion. There is a clear need for a vocabulary, generally understood and accepted, to distinguish between (1) proprietary claims which may involve tracing in equity (as in Attorney General for Hong Kong v Reid [1994] 1 AC 324); (2) personal claims for an account of profits (that is, for a sum equal to the profits actually made by the defendant); and (3) personal claims for interest which represents (in a more or less conventional way) the benefit which the defendant is presumed to have derived from money in his hands.

The ECJ judgment in Hoechst

In his opinion Lord Hope has fully analysed the effect of paragraphs 77 to 96 of the judgment of the ECJ in Hoechst [2007] Ch 620. I respectfully agree with his analysis. While leaving causes of action and remedies to the national court, the ECJ has made clear that the national court must award an effective remedy. The ECJ emphasised (para 88) that

Without referring in terms to La Pintada [1985] AC 104, it made clear that the rule in that case must not be applied.

Similarly in the paragraphs of its judgment dealing with a compensatory award the ECJ distinguished (para 93) R v Secretary of State for Social Security, Ex p Sutton (Case C-66/95) [1997] ICR 961 as a case where entitlement to interest was not an essential component of the claimant's right, stating that:

Instead it compared the cases with Marshall v Southampton and South West Hampshire Area Health Authority (Teaching) (No 2) (Case C-271/91) [1994] QB 126, in which the requisite full compensation had to include an allowance for the time value of money. This case is even stronger because the time value element is not ancillary but absolutely central to the claim.

Conclusions

The judgment of the ECJ is in my opinion a powerful encouragement for this House to reconsider the basis on which a monetary award reversing unjust enrichment can and should take account of the time value of money. In modern economic conditions simple interest does not provide full compensation in a case where unjust enrichment has lasted for a significant period (a fact which is now reflected, as Lord Hope points out, in the practice of the European Commission). As Hobhouse J said in Westdeutsche at first instance, [1994] 4 All ER 890, 955,

Wallersteiner v Moir No 2

National Bank of Greece SA v Pinios Shipping Co, The Maira

Hobhouse J was of course delivering his judgment before this House decided not to follow Sinclair v Brougham [1914] AC 398, and so at a time when there appeared to be an indisputable equitable jurisdiction to award compound interest in a case of this sort. The House's decision in Westdeutsche could be distinguished (except in relation to those cases, mentioned in para 33 of Park J's judgment, where amounts of ACT were actually repayable as principal sums) on the basis that in Westdeutsche the House was not concerned (or at any rate did not perceive itself as concerned) with a case where part of the principal sum had been repaid before issue of the writ (in fact the position was more complicated, as the Court of Appeal pointed out in IM Properties plc v Cape & Dalgleish [1999] QB 297 at pp 305 (Waller LJ) and 308 (Hobhouse LJ)). But such a distinction would be anomalous and might be thought to leave the law in an even less satisfactory state.

Lord Nicholls and Lord Hope propose to cut through the thicket of problems by recognizing a restitutionary remedy available as of right at common law, subject to the Court's power to resort to "subjective devaluation" in order to avoid injustice in hard cases. This would be following a course which, in Westdeutsche, was not so much rejected as assumed not to be open. I must confess that my own inclination would be to take the course which this House came very close to taking, but ultimately drew back from taking, in Westdeutsche: that is to extend the court's equitable jurisdiction to award compound interest. Before your Lordships the law has been much more fully investigated, and in my opinion there are compelling reasons for departing from Westdeutsche, and recognising the force of Lord Goff's and Lord Woolf's powerful dissenting speeches in that case.

Both Lord Goff [1996] AC 669, 695-697 and Lord Woolf (at 721-723) saw their preferred solution as an extension of equity's auxiliary jurisdiction in order to make good the inadequacy of a common law remedy. That would in my opinion be a principled development in the still-evolving relationship between equity and the common law (see Professor Andrew Burrows, "We Do This at Common Law but That in Equity" (2002) 22 OJLS 1). It would be a further and permissible step in the progress which this House has made towards developing what Lord Goff (in Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349, 372G) referred to as "a coherent law of restitution."

In Westdeutsche Lord Goff and Lord Woolf both considered (at 691 and 723 respectively) that on the facts of that case compound interest was required in order to achieve complete restitution and reverse unjust enrichment. Lord Woolf recognised (at 722) that the exercise of the auxiliary jurisdiction would (as with all equitable remedies) be discretionary, and (at 724) that compound interest should not be awarded if the facts were such that the defendant would not have earned compound interest. Awards of simple interest under section 35A of the Supreme Court Act 1981 are also discretionary, but the Court's exercise of its discretion causes few difficulties in practice. In my opinion this is clearly a case in which compound interest should be awarded, since (i) it is a case where Community law requires full restitution; (ii) the defendant is economically powerful and sophisticated and must be supposed (as the agreed "conventional basis" seems to recognise) to have taken full advantage of its premature receipts of ACT; and (iii) it is not suggested that the claimant has been at fault or has been dilatory in making or pursuing its claim.

I feel some apprehension about the suggested conclusion that compound interest should be available as of right, subject only to an exception for "subjective devaluation," a concept normally applicable to benefits in kind (see Birks, An Introduction to the Law of Restitution, revised 1989 edition p.109, quoting Robert Goff J in BP Exploration Co (Libya) Ltd v Hunt (No 2) [1979] 1 WLR 783, 799). It is true that the time value of money (as opposed to money itself) may be regarded as a "non-money benefit", as Birks does in Unjust Enrichment, 2nd ed. P.53. But it is a benefit which can readily be quantified in money terms; that has been, for many centuries, the function of interest. The discretionary nature of an equitable award of interest provides the necessary flexibility, though I would expect the principles for the exercise of the discretion to develop along familiar and predictable lines.

In this case either the common law route or the equitable route leads to the same conclusion. The appropriate exercise of discretion is to order the Revenue to pay compound interest at a conventional rate calculated by reference to the average cost of government borrowing during the relevant period. I would therefore dismiss the appeal and make the order proposed by Lord Hope.

My Lords,

Introduction

This is a further appeal in the series relating to advance corporation tax ("ACT"). It has involved two hearings, one of two days in November 2006 and a further day's hearing at the House's request in May 2007. ACT was until 5th April 1999 levied under s.14 of the Income and Corporation Taxes Act 1988. But it was held by the European Court of Justice in Joined Cases C-397/98 and C-410/98, Metallgesellschaft Limited and Others v. Commissioners of Inland Revenue and HM Attorney General [2001] Ch 620, to discriminate unlawfully against companies with parents resident in other Member States. Lord Hoffmann, in Deutsche Morgan Grenfell Group Plc v. H. M. Inland Revenue Commissioners [2006] UKHL 49, [2007] 1 AC 558 described the conceptual basis of ACT:

Metallgesellschaft Hoechst

The Pintada

President of India v La Pintada Cia Navigacion SA

The issue now is whether a company unlawfully required to pay ACT is entitled, under European or domestic principles, to interest for the loss of use of money on a compound or only a simple interest basis. Under a Group Litigation Order made 26th November 2001, Sempra Metals Limited ("Sempra"), as Metallgesellschaft Limited is now known, is a test claimant for the determination of this issue, and four sample dividends have been identified for consideration. Their relevant details are as follows:

Dividend Payment Date	Dividend Amount (£)	ACT Payment Date	ACT Amount (£)	Set off Date	Set off Amount (£)
23/7& 21/9/81	2.5m	12/10/81	1,071,428.57	1/7/90 1/7/91	a) 259,206 b) 812,223
4/1/85	2.0m	10/4/85	808,625.00	1/7/94	808,625.00
4/4/89	1.8m	10/7/89	369,143.12	1/7/94	369,143.12
25/5/94	21.0m	18/7/94	3,230,002.49	1/7/95 1/7/96	a) 1,665,358 b)1,563,644.19

As the table shows, often very substantial periods of years passed before Sempra was able to set off the ACT payments which the 1988 Act called upon it to make. Sempra started the present proceedings by writ issued 6th March 1996, by when Sempra had not been able to set off the whole of its fourth payment of ACT, although it was able to do so shortly afterwards. But the only claim made in the present proceedings relates to the periods prior to set off. The claim as pleaded is for damages to compensate for the loss of use of the ACT or, by later amendment, for restitution in respect of the loss of use of the ACT during such periods. The course of argument before the House led Sempra to reformulate the latter claim as a claim for the benefit, actual or notional, obtained by the Revenue by having the use of the money.

Both Park J ([2004] EWHC 2387 Ch); [2004] STC 1178 and the Court of Appeal, consisting of Chadwick, Laws and Jonathan Parker LJJ, ([2005] EWCA Civ 389; [2005] STC 687) held that Community law, and in particular the principles laid down by the European Court in the Metallgesellshaft case, require English domestic courts to give a full remedy or full compensation in order to restore equal treatment and that only an award of compound interest would fully achieve this. Park J accepted a submission by the Inland Revenue that such interest should be calculated on a "conventional basis", that is by fixing a single market-based rate for all claimants, irrespective of whether they were net borrowers or depositors and of any actual rates at which they borrowed or lent (paragraphs 18 to 22). On this last point his decision was not appealed, but Chadwick LJ giving the only full judgment in the Court of Appeal commented that "if the matter is to be approached on the basis of 'one rate suits all', then (as it seems to me) there is really no alternative to adopting a borrower's rate rather than a lender's rate" (paras. 50-52). The Court of Appeal added the rider that interest should be computed by compounding at the same periodic rests as those to reference to which the applicable rate was fixed (para. 54).

Against the award of compound interest the Revenue now appeals. The Revenue submits that the remedy to be provided in respect of ACT payments made under the discriminatory legislative scheme identified by the European Court in the Metallgesellschaft case depends on English domestic law. It submits that, whatever form the remedy takes under English domestic law, only simple interest is recoverable, but that, if this is wrong, distinctions should be drawn between the various remedies. Unfortunately, until the submissions before the House, no such distinction appears to have been suggested (cf paragraph 11 of Park J's judgment), but the matter is one of pure law. Three possible domestic remedies exist: (a) a claim for damages for or analogous to breach of statutory duty or (b) a restitutionary claim based on or by analogy with the principle in Woolwich Equitable Building Society v. Inland Revenue Commissioners [1993] AC 70 and (c) a restitutionary claim based on or by analogy with the principle of mistake of law recognised in the House's decisions in Kleinwort Benson Ltd. v. Lincoln C.C. [1999] 2 AC 349 and Deutsche Morgan Grenfell Group Plc v. Inland Revenue Commissioners [2007] 1 AC 558. Sempra, I understand, accepts that it would, before entry of any final judgment, have to elect between, on the one hand, the first and, on the other hand, the second and third remedies. I leave aside whether the second and third remedies may combine at some higher level of abstraction, such as absence of any basis for the payment or retention of the ACT. For domestic law purposes there is on any view a significant difference between them: the third remedy would, under section 32(1)(c) of the Limitation Act 1980, potentially postpone the commencement of the limitation period until the time when Sempra discovered or could with reasonable diligence have discovered that the ACT was not due. Sempra and other claimants intend to identify that time with the European Court of Justice's decision on 8th March 2001 in the Metallgesellschaft case. Without such a postponement, interest losses prior to 6th March 1990 would on the face of it be time-barred, a matter of great relevance to ACT paid by Sempra on the first three sample dividends shown in the table above. If the third remedy proves available, Sempra will be likely therefore to elect for it, rather than for the first. Finally, the Revenue submits that, even if the nature of the interest awarded is a matter of Community law as the courts below considered, there is nothing in Community law to require or to make appropriate an award of compound interest.

The Metallgesellschaft case

I start with the European Court's reasoning and decision in the Metallgesellschaft case [2001] Ch 620. The second question put by the High Court was

Brasserie du Pêcheur SA

Federal Republic of Germany

Secretary of State for Transport

Ex p

Factortame Ltd (No 4)

[1996] QB 404

Secretary of State for Social Security

Ex p Sutton

[1997] ICR 961

The European Court's answer in paragraph 96 and in its ruling was as follows:

This answer establishes the principles by reference to which English courts must approach Sempra's claim. Domestic law is required to provide "an effective legal remedy in order to obtain reimbursement or reparation of the financial loss which they [in casu, Sempra] have sustained and from which the authorities of the member state concerned have benefited as a result of the advance payment of tax by [Sempra]". The answer involves no requirement that the remedy should take any particular form, an impression confirmed by the following paragraph in the European Court's reasoning:

Nevertheless, Mr Rabinowitz QC for Sempra submits that the European Court determined, as a matter of Community law, that a remedy must be available in domestic law in both restitution and damages, and that in each case it must consist in allowing recovery of compound interest. He bases this submission on ensuing paragraphs of the European Court's reasoning, which discuss the position firstly "on the assumption that the actions brought by the claimants ….. are to be treated as claims for restitution of a charge levied in breach of Community law" (paragraphs 82-89) and secondly "assuming that the claimants' claims are to be treated as claims for compensation for damage caused by breach of Community law" (paragraphs 90-95). Again, on the face of it, the Court's introduction to each of these sets of paragraphs indicates no more than an intention to discuss the two alternative possibilities envisaged in the bracketed phrase in paragraph 81. However, in paragraphs 82 to 89 the European Court rested its reasoning regarding restitution firmly on "well-established [European] case-law", in terms which, it may be suggested, require a restitutionary remedy to be available. In my opinion, that would be a mistaken understanding of the Court's actual or likely intentions. In paragraph 84, the Court was emphasising that, in circumstances like those under discussion, Community law requires there to be a (not any particular) right to a refund:

Amministrazione delle Finanze dello Stato v SpA San Giorgio

[1983] ECR 3595

Barra v Belgian State

BP Supergas Anonimos Etairia Geniki Emporiki-Viomichaniki kai Antiprossopeion v Greek State

[1995] ECR I-1883

Dilexport Srl v Amministrazione delle Finanze dello Stato

[1999] ECR I-579

Kapniki Mikhailidis AE v Idrima Kinonikon Asphaliseon

[2000] ECR I-7145

Société Comateb v Directeur Général des Douanes et Droits Indirects

[1997] ECR I-165

Dilexport

Mikhailidis

A reading of the cases cited in paragraph 84 shows the European Court consistently stating that "repayment may be sought only within the framework of the conditions as to both substance and form, laid down by the various national laws applicable thereto" (cf the San Giorgio case, para. 12) and in the Dilexport case, paras. 24-25 it stated that

Aprile v Amministrazione delle Finanze dello Stato

[1998] ECR I-7141

The Court was also repeating its previous jurisprudence when it said in paragraph 85 of its present judgment that it is for the domestic legal system

In paragraph 86, the Court further stated that it was for national law

In paragraphs 87 to 89 it reiterated why, in the circumstances of the present case, an award of interest is not an ancillary matter for the national court, but is required by Community law:

Despite the language of "restitution", the Court appears clearly to have been focusing throughout on Sempra's loss of the money, rather than on any benefit that the Revenue may or may not have had from use of the money. Before the House, as I have said, Sempra's attention shifted to the latter.

I do not read the concluding words of paragraph 89 as meaning that interest must be made available by way of restitution, but as meaning that, if the right to a refund of charges levied in a Member State in breach of rules of Community law is under national law provided by way of a restitutionary remedy, then that remedy must as a matter of Community law include a right to interest. In paragraphs 90 to 95, the European Court dealt with the alternative way in which a remedy might be provided, that is by an action for damages for breach of Community law. It pointed out that it had already held, in paragraph 87 of its judgment in Brasserie du Pêcheur SA v. Federal Republic of Germany; R v. Secretary of State for Transport, Ex p Factortame Ltd (No 4) (Joined Cases C-46 and 48/93); [1996] QB 404, 503, that

So it drew the conclusion that the remedy, if provided by way of an action for damages for breach of Community law, must also include a right to interest. I see no reason why the Court should have insisted that the domestic law remedy or right to interest required by Community law must be provided in both ways - that is both by way of restitution and by way of a claim for damages. It is necessary, but also sufficient, to provide it by one of the alternative routes that the Court was discussing.

The consideration that a restitutionary claim based on mistake of law might, if available, yield a much longer limitation period does not, in my opinion, bear on the effectiveness of the relief required by the European Court. The contrary was not suggested in the submissions put before the House. Time-limits for particular claims are par excellence a matter for domestic legislation and courts. Moreover, the appropriateness of an extended time limit in this context is questionable. Indeed, Lord Hoffmann in Kleinwort Benson [1999] 2 AC 349 recognised that "allowing recovery for mistake of law without qualification, even taking into account the defence of change of position, may be thought to tilt the balance too far against the public interest in the security of transactions", adding that "The most obvious problem is the Limitation Act, which as presently drafted is inadequate to deal with the problem of retrospective changes in law by judicial decision" (p.401D-E). With effect from 8th September 2003 (and so not applicable to the subject-matter of this appeal), s.320 of the Finance Act 2004 in fact means that the provisions of s.32(1)(c) of the Limitation Act 1980 no longer apply to mistakes of law relating to a taxation matter under the care and management of the Revenue. So the existence of a six year limitation period in respect of any particular claim for damages or restitution does not mean that such claim would not be an effective remedy.

The requirement of an effective remedy in national law

I turn to the content of the requirement that national law should provide, in one form or another, an effective legal remedy. Both Park J and the Court of Appeal concluded that, whatever domestic law might otherwise provide, the reasoning in the Metallgesellschaft case required domestic law to make an award of compound interest. That was, in Park J's opinion (paragraph 27), because "only compound interest will fully restore equal treatment" and, in Chadwick LJ's words (paragraph 53), because English domestic rules of interest "must yield to the overriding requirement that the domestic court gives full compensation". The European Court in the Metallgesellschaft case used the phrase "full compensation" in paragraph 94 in the context of consideration of its previous decision in Marshall v. Southampton and South West Hampshire Area H.A. (Teaching) (No. 2) (Case C-271/91) [1994] QB 126. It there held that a maximum limit on awards of £6,250 and the absence of a power to award interest in employment tribunal proceedings were inconsistent with the Equal Treatment Directive (76/207/EEC). As to interest, the Court said that, where financial compensation was the method adopted by national law to achieve the objective of equal treatment of men and women as regards employment, then

and

The question whether interest should be compound or simple was not raised in Marshall, although the interest which the Industrial Tribunal had purported to award by analogy with the s.35A of the Supreme Court Act 1981 - and which the Employment Appeal Tribunal and Court of Appeal held that it had no power to award - must have been simple interest.

While an award of interest is an essential component of the "effective" legal remedy which national courts are bound to afford, I doubt whether this means, as a matter of Community law, that it is essential that there should be compounding. In an article On Interest, Compound Interest and Damages (1985) 101 LQR 30, 42-46, Dr F. A. Mann QC drew attention to the powerfully rooted objection to compound interest in many European jurisdictions, and in the USA. A Study on the conditions of claims for damages in case of infringement of EC competition rules dated 31st August 2004 prepared for the Commission by Messrs Ashursts noted that in the great majority of member states, interest awarded in such claims was not compounded. In Société Roquette Frères v. Commission of the European Communities (Case 26/74) [1976] ECR 677, cited by the European Court in paragraph 86 of its present judgment, the Court held that it was a matter for the discretion of the national court whether to make any award of interest ancillary to an order for the recovery of charges levied by the French State on the basis of a Commission regulation which was inconsistent with regulations made by the Council of Ministers. In its present judgment, the Court held that an award of interest was mandatory, because the claim for interest was the very objective sought by the proceedings, rather than ancillary. In Directive 2000/35/EC of 29 June 2000 on combating late payment in commercial transactions, the European Parliament and Council adopted a simple interest approach, but provided for a rate of 7% over the rate applied by the European Central Bank to its most recent main refinancing operation. In Commission Regulation (EC) No 794/2004 of 21 April 2004 implementing Council Regulation (EC) No 659/1999 and laying down detailed rules for the application of article 93 of the EC Treaty (which was aimed in cases of unlawful state aid at restoring the situation existing before the aid was unlawfully granted), the approach adopted was for the Commission to fix a recovery interest rate and for that rate to be compounded annually. Against this background, I doubt whether it can be said (or that the European Court would say) that national law would necessarily be failing to give an effective remedy, if, under national principles governing analogous domestic claims, simple rather than compound interest was the rule.

In the present case, Park J concluded that a single "conventional" market-based rate should, as a matter of Community law or alternatively under national law, be adopted (para. 22), and the Court of Appeal added the rider that it should be the borrower's rate, compounded at the same periodic intervals as those by reference to which the rate was set. But this was on the basis that any recovery, whether by way of damages or on a "restitutionary" basis, should be measured by reference to Sempra's loss, which is no longer Sempra's position with regard to restitution.

In the light of the above it is necessary to consider English domestic legal principles governing interest. But this must be on the basis that it is not permissible for English law to decline to give any interest at all or to treat an award of interest as a purely discretionary matter (cf paragraphs 87-89 of the European Court's judgment cited in paragraph 198 above and paragraph 5 of Lord Hoffmann's judgment in the Deutsche Morgan case cited in paragraph 189 above). Neither approach would be consistent with the European Court's conclusion that, because reimbursement or reparation in respect of the loss of use which Sempra has sustained and from which the Revenue has as a result benefited is the "very objective" or "sole object" sought by Sempra, English law must provide an (though not any particular) effective legal remedy in respect thereof.

Interest in English law

The attitude of English law to interest has been inhibited and undistinguished. As this appeal developed, the House was invited radically to reshape it. This is a not unattractive invitation. But we need to do so with some caution in the light of the last two hundred years of the law's development. We must navigate using the reference points of precedent, Parliamentary intervention and analogy, and we should bear in mind the limitations of judicial knowledge and the assistance offered by a series of Law Commission reports.

Although old and in some respects unsatisfactory, the case-law of the late 18th and early 19th centuries remains the relevant backdrop. Neither in debt nor in a claim for money had and received could interest be recoverable (save where agreed or in the case of a negotiable instrument): cf Walker v. Constable (1798) 1 B & P 306, De Bernales v. Fuller (1810) 2 Camp. 426 and Depcke v. Munn (1828) 3 C & P 112 (all cases of claims for money had and received), and Page v. Newman (1829) 9 B & C 378 (interest not recoverable on loan not repaid when due). In Arnott v. Redfern (1826) 3 Bing 352, 359 Best CJ referred to the general rule that interest was not recoverable as one which

But he suggested that interest might be recovered in a case of "unjust detention", by which he meant wrongful refusal of payment after demand for payment. In Page v. Newman, Lord Tenterden CJ dismissed this suggestion, at p 381, on the basis that if there were such a rule "it might frequently be made a question …. whether the proper means had been used to obtain payment of the debt", which "would be productive of great inconvenience" and the court ought not to depart from the "long-established" contrary rule.

In London Chatham and Dover Railway Company v. South Eastern Railway Company [1893] AC 429, 438 the House discounted as not "very satisfactory" rules (the last, that interest might be recovered where proof was "given of the money being used") suggested by Lord Ellenborough in De Havilland v. Bowerbank (1807) 1 Camp. 50, 52 (another claim for money had and received). The Act then promoted by Lord Tenterden in 1833 (3 & 4 Wm. 4, c. 42) introduced only a very limited extension of the circumstances in which interest might be recovered. Under s. 28 interest might on the trial of any issue or on any inquisition of damages be allowed, if thought fit, but only on "debts or sums certain …. payable by virtue of some written instrument at a certain time, or if payable otherwise, then from the time when demand of payment shall have been made in writing" giving "notice to the debtor that interest will be claimed from the date of such demand". S. 29 also permitted an award of damages in the nature of interest on the value of goods in actions of trover or trespass de bonis asportatis.

Claims in debt and for money had and received were thus treated together in the early authorities, and it is I think clear that the wording of Lord Tenterden's Act covered both. This was indeed assumed by counsel in argument in Fruhling v. Schroeder (1835) 2 Bing (NC) 78, 80, where Tindal CJ held, following the early decisions, that "in money had and received, the net sum only can be recovered". The decisions of Brandon J in The Aldora [1975] QB 748 and Robert Goff J in B. P. Exploration Co. (Libya) Ltd. v. Hunt (No 2) [1979] 1 WLR 783, 835-7, affirmed in this House at [1983] 2 AC 352, 373-4, that restitutionary claims fell within the replacement 1934 Act were not revolutionary; the second case merely concerned a particular problem about bringing a claim under the Law Reform (Frustrated Contracts) Act 1943 within the statutory language.

I am unpersuaded by the academic suggestion (Edelman & Cassidy, Interest Awards in Australia (2003), Chap. 5, p. 93) that the law's attitude to interest in claims for money had and received was shaped by the fiction that such claims arose "as it were upon a contract". The full passage from Lord Mansfield's judgment in Moses v. Macferlan (1760) 2 Burr 1005, 1008 indicated where "the defendant be under an obligation, from the ties of natural justice, to refund", then "the law implies a debt, and gives this action, founded in the equity of the plaintiff's case, as it were upon a contract ('quasi ex contractu', as the Roman law expresses it)". Lord Mansfield went on carefully to distinguish the features of claims in contract and for money had and received, and showed that he were well aware of the different considerations actually underpinning these different claims, consensus on the one hand and natural justice or equity in a broad sense on the other. It seems to me that the law's attitude to interest was shaped not by conceptual confusion, but by policy-driven concerns (however debatable) regarding interest which may well have had historical roots, and which find an echo to this day in modern legislation and Law Commission reports (including their most recent report Pre-judgment Interest on Debts and Damages dated 23 February 2004, (Law Com No 287) where it shaped their recommendations: cf especially at paras. 5.15 through 5.40). They also found an echo in the speech of Lord Lloyd of Berwick in Westdeutsche Landesbank Girozentrale v. Islington London BC [1996] AC 669, 741A-F.

The legal position after Lord Tenterden's Act was considered in the London, Chatham and Dover case [1893] AC 429, where De Havilland, Arnott v. Redfern and Page v. Newman were examined. On the facts, the Act was inapplicable, and the House held, with some regret, that no claim could be made by the plaintiffs at common law "by way of damages in respect of the wrongful detention of their debt" (per Lord Herschell LC at p.437). Lord Herschell thought that the limits imposed by the Act "seem to be too narrow for the purposes of justice". But having regard to Page v. Newman "and to the statute passed subsequently with obvious reference to it by the Legislature", and the absence since that time" of any contrary case he did not "think it would be possible nowadays to reopen the question, even in this House, and to hold that interest under such circumstances could be awarded" (pp.440-1). Lord Shand regretted that English law was not like the law of Scotland where "it is the common and ordinary practice, in bringing an action for money which is due, to conclude not only for the payment of that money but for the payment of interest upon it from the date of citation or service of the summons, and interest is decreed as a matter of course on whatever balance is found to be due" (p. 443). But I note that Lord Shand spoke of interest only from citation or service, and also of an award that would, I understand, be of simple, not compound interest. In Johnson v. The King [1904] AC 817 the Privy Council held, consistently with the above, that, fraud aside, interest could not be recovered in law or equity on monies recoverable as having been paid under a mistake. But it was submitted to the Board in support of an award of interest that "Johnson, who was a trader, must have made a profit by the use of the money which was in his hands for a year", and the way in which the Board dealt with this submission is of some interest, viz: "That is very probable, but there is no proof of it" (p. 822). The case was one where the monies received were paid into court, and accepted in partial satisfaction of the claim, during the proceedings for their recovery, but the plaintiff had continued to trial to seek interest as "special damage for the loss of use of the money during the periods of detention".

In 1934 s.3(1) of the Law Reform (Miscellaneous Provisions) Act 1934 gave effect to the recommendations on interest in the Second Interim Report of a distinguished Law Revision Committee, chaired by Lord Hanworth MR (Cmd 4546). The Act expanded the circumstances in which interest might be awarded, to provide for the discretionary award of simple interest on any debt or damages for which judgment was given by a court. In June 1978 the Law Commission issued a Report on Interest (Law Com No 88; Cmnd. 7229) which recommended, first, a general rule that contract debts (but not "debts due in quasi-contract") should carry simple interest at a statutory rate (paras. 218 and 221), and, second, that the provisions of the 1934 Act should be replaced by a wider discretionary power for the court in any proceeding for debt or damages to award simple interest in respect of sums recovered or paid without a trial or before judgment (paras. 236-9).

Parliament adopted only the second recommendation when in 1982 it replaced the provisions of the 1934 Act with s. 35A of the Supreme Court Act 1981. In the meanwhile, there had been a common law development. Building on remarks of Lord Denning in Trans Trust Sprl v. Danubian Trading Co. Ltd. [1952] 2 QB 297, the Court of Appeal in Wadsworth v. Lydall [1981] 1 WLR 598 allowed recovery of interest as "special damages" for failure to pay £10,000 to complete an agreement to dissolve a partnership. On the strength of the agreement, the claiming partner had committed himself to complete a purchase of another property which he was unable to complete on time, so becoming liable to pay his seller the interest for non-completion.

In the President of India v. La Pintada Co. Nav. SA [1985] AC 104, the House re-affirmed the principle in the London Chatham case. Lord Brandon of Oakbrook (in a speech with which all other members agreed) observed in passing that the London, Chatham and Dover decision "was regarded as applying to any form of damages" (cf p.115F). However, he identified some common law exceptions, as well as the statutory interventions discussed above. One exception was Admiralty law, under which simple interest had been awarded on damages recovered in a damage action for more than a century and on sums recovered in a salvage action since 1975, and it had been assumed, he said, that interest would likewise also be recovered in an action for a contractual debt (p.115G-H). Another was equity, in which Chancery courts regularly awarded simple interest in respect of equitable remedies "such as specific performance, rescission and the taking of an account", and compound interest was awarded "when they thought that justice so demanded, that is to say in cases where money had been obtained and retained by fraud, or where it had been withheld or misapplied by a trustee or anyone else in a fiduciary position" (p.116A-B). Lord Brandon also identified and approved the "important judicial inroad into the previously accepted understanding of the scope of the decision" made in Wadsworth v. Lydall (p 125G-H).

The House revisited the topic in President of India v. Lips Maritime Corporation [1988] AC 395, where a claim for a currency exchange loss through late payment of demurrage failed on the straightforward basis that demurrage is liquidated damages, and the law knows no such thing as a claim for damages for failing to pay damages. Lord Brandon, in a speech with which three other members of the House agreed, again considered the scope of the London, Chatham and Dover case at p.424B-G and indicated that (despite the understanding about its application to "any form of damages" - cf paragraph 213 above) its scope is strictly confined to claims to recover interest as damages for late payment of a debt. Claims for other losses (such as exchange losses) by way of damages for late payment of a debt differed significantly because there was no "previously established law comparable to that laid down in the London, Chatham and Dover …. case" and "no statutory provisions …. which could conflict with any right of recovery at common law" (p. 424D). Accordingly "claims to recover currency exchange losses as damages for breach of contract, whether the breach relied on is late payment of a debt or any other breach, are subject to the same rules as apply to claims for damages for breach of contract" (p.424E).

As my noble and learned friend, Lord Nicholls of Birkenhead, points out in his opinion at paragraphs [85 to 89], it is difficult to follow or accept Lord Brandon's explanation of the term "special damages" by reference to the second rule in Hadley v. Baxendale (1854) 9 Exch. 34: see also the criticisms by Dr F. A. Mann in On Interest, Compound Interest and Damages (1985) 101 LQR 30, 34-35 and by Staughton LJ in President of India v. Lips Maritime Corporation [1985] 2 Lloyd's Rep. 180, 185. Since Lord Brandon described the first rule in Hadley v. Baxendale as dealing with "general damages", the difficulty is, if anything, greater. The two limbs of Hadley v. Baxendale are the practical expression of a single principle (inspired by the civil law) that parties should only be liable for damages which were when they contracted within their contemplation in the event of a breach. The precise line between the two limbs is itself hazy - especially in the modern legal environment where it is axiomatic and integral to contractual construction that it occurs not in a vacuum, but in the light of all surrounding circumstances within the parties' knowledge. When seeking to distinguish the two limbs for present purposes, Lord Brandon quoted Lord Denning's citation in the Trans Trust case, at p 306, of a passage from Bullen & Leake (3rd ed., at p.51) stating that the ground on which the London, Chatham and Dover case rested was "that interest is generally presumed not to be within the contemplation of the parties" (the important word "not" is omitted from the quotation at p.124H in the Appeal Cases report of the La Pintada case). But Lord Denning's statement as to the general position does not exclude the existence of transactions, the very terms and nature of which makes it patent without more that loss of interest is likely to result from breach (i.e. falling within the first rather than the second limb of Hadley v. Baxendale). Further, when the Court of Appeal in Wadsworth v. Lydall [1981] 1 WLR 598 distinguished the London, Chatham and Dover case on the ground that it concerned general rather than special damages, I think it clear that they meant by special damages any damages provable under either limb of Hadley v. Baxendale: cf p. 603F-G. Indeed, the award of interest in Wadsworth v. Lydall itself was based as much on what was self-evident from the nature of the particular transaction as on any special communication: see per Brightman LJ at p.602E-F and Ormerd LJ at p.605 D-F.

It is, nevertheless, still fair to assume that loss of interest is not within the parties' contemplation under many everyday contracts. It is in this context material that a contracting party has the opportunity to stipulate for, or reserve the right to claim, interest in the event of breach. The distinction emphasised by the House in Koufos v. C. Czarnikow Ltd. [1969] 1 AC 350 between, on the one hand, what is to be taken as within contracting parties' reasonable contemplation and, on the other hand, what may be said to be reasonably foreseeable for the purpose of a claim for purely tortious damages remains good. So, to take some examples within the first limb: a householder who pays a builder's or shop's account late or very late should not (in the absence of contractual agreement or special circumstances) be taken to have within his or her contemplation for the purposes of Hadley v. Baxendale damage in the form of interest (or a fortiori compound interest); in contrast, a loan contract made specifically to enable an individual to refinance pre-existing loans or complete a property purchase could appear of its nature very likely, if broken, to give rise to the individual incurring damage by way of interest. Again, in the business to business sphere, while the business context may make it easier to show that loss by way of interest was within the parties' reasonable contemplation, this cannot follow automatically. The Late Payment of Commercial Debts (Interest) Act 1998 (to which I refer later in this opinion) cannot be relegated to the status of a measure providing for simple interest either as a convenient and easier alternative to damages or in situations where no interest loss at all has been suffered. The present case should not therefore be seen as a charter for claims, still less for claims on a compound basis, in respect of interest losses following a breach of contract, where there is no contractual stipulation for its recovery, simply because it can be said that the situation was one where loss of interest might foreseeably, and did in fact, follow on breach. Neither general principle nor the cautious approach of Parliament and the Law Commission (to which I refer further below) would justify this. Before loss by way of interest is recoverable as damages, the higher threshold of reasonable contemplation must be crossed. Whether this has occurred depends on the particular case. But it is not appropriate to draw a line between the two limbs of Hadley v. Baxendale in the manner suggested in the La Pintada and Lips cases. Loss of interest is recoverable as damages for breach of contract, if it was within the reasonable contemplation of the parties, in the sense explained in Koufos v. C. Czarnikow Ltd., under either limb when the contract was made and is specifically pleaded and proved on that basis.

The rule in the London, Chatham and Dover case being thus confined, no reason appears why claims for loss of interest in tort, should be subject to special rules. In tort, there is generally no possibility of stipulating for the recovery of interest and the test of reasonable foresight, as opposed to contractual contemplation, operates as a significantly less strict pre-condition to recovery. The possibility of recovering such loss in tort has been recognised without reference to the London, Chatham and Dover case in Brandeis Goldschmidt & Co Ltd v. Western Transport Ltd [1981] QB 864, 873 (Brandon LJ), Swingcastle Ltd. v. Alastair Gibson [1991] 2 AC 223 (where the defendant valuer was not in contractual relations with the claimant lender (cf p.227D-F), and was referred to accordingly by Lord Lowry at p.239C-D as the "tortfeasor"), Nigerian National Shipping Lines Ltd. v. Mutual Ltd. (The Windfall) [1998] 2 Ll.R. 664, (Rix J) and The Mortgage Corporation v. Halifax (SW) Limited [1999] 1 Ll.R Prof. Neg. 159 (HHJ Laurie). La Pintada was, on the other hand, referred to in I. M. Properties Plc v. Cape & Dalgleish [1999] QB 297 and Blue Circle Industries Plc v. Ministry of Defence [1999] Ch 289, but in each case the Court of Appeal was concerned with claims to interest at conventional rates under s. 35A of the Supreme Court Act 1981 on moneys which had been recovered in mitigation of the claimants' loss prior to action brought. In the former case, the Court concluded, correctly in my view, that the similar award made in the Westdeutsche case had been made per incuriam. In both cases, the Court acknowledged the possibility that the claimants might have been able to plead and prove a claim for special damage consisting of loss of interest, but none had been.

In Westdeutsche Landesbank Girozentrale v. Islington London BC [1996] AC 669 the argument proceeded, not surprisingly, on the basis that there was no power at common law to award compound (or, indeed, apart from statute, simple) interest (cf per Lord Goff at p.684G-H, Lord Browne-Wilkinson at p.700H and 717E-H, Lord Slynn at p.718F-H (although referring to the common law "as presently formulated"), Lord Woolf at p.722C-F and 730H-731A and 731F and Lord Lloyd at p.737B-C). The common law recognises a claim for money had and received, but, in the absence of a proprietary claim (and none is advanced by Sempra in this case), the money received belongs to the recipient while he holds it. Any interest received thereon thus also becomes the recipient's. The common law does not recognise a claim against him for the use had of money had and received. Professor Birks points out that Lord Mansfield made clear in Nightingale v. Bevisme (1770) 5 Burr. 2589 (citing Moses v. Macferlan (1760) 2 Burr. 1005) that "'money received' meant what it said: the words could not be extended to other things" (cf An Introduction to the Law of Restitution, p 113). The subsequent cases cited in paragraph 206 above underline the point. The situation differs where there is a proprietary claim. The fruits or offspring of my property in your hands are mine.

Once a proprietary claim was ruled out, the issue in Westdeutsche Landesbank Girozentrale v. Islington London BC [1996] AC 669 became whether compound interest was recoverable in equity. The bank had paid £2.5 million to the council under a swap agreement, later established to be ultra vires the council. "Interest" payments totalling £1,354,474 were made by the council to the bank in 1988-89. The bank then commenced proceedings to recover the balance of the £2.5 million which it had paid, plus compound interest. The Court of Appeal awarded compound interest on the balance from time to time outstanding from 18th June 1987. Until the hearing before the House of Lords it was assumed in the light of Sinclair v. Brougham [1914] AC 398 that the council was liable as a fiduciary, and the claim for interest was made in equity on this basis (under the exception recognised by Lord Brandon in La Pintada). But, in the House of Lords, the House decided to hear argument on the correctness of Sinclair v. Brougham, and in the event held that it was wrongly decided. The width of equity's power to award compound interest came thus into focus, although only at the end of long argument on other matters and in circumstances where the council was unwilling to devote more resources to a further day's submissions (cf pp.679C, 717G-H and 738F-G).

By a majority (Lord Goff and Lord Woolf dissenting) the House refused to award compound interest in equity. The majority felt that any amendment of the law was a matter for Parliament: cf per Lord Browne-Wilkinson at pp. 717G-718C, Lord Slynn at pp.718H-719A and Lord Lloyd at pp.738D-741G, especially at p740D-741A). I have noted above the Court of Appeal's conclusion in the I. M. Properties case that the order made in the Westdeutsche case was per incuriam so far as it awarded simple interest on the balance outstanding from time to time but paid by the date when action was commenced. The Court of Appeal assumed, rightly in my view, that this order was made under s.35A of the Supreme Court Act 1981. There is no sign of any submission that equity enabled an award of simple interest on a sum recovered prior to action brought. All three of the majority in the Westdeutsche case concentrated on rejecting the minority's argument that equity could award compound interest in aid of a personal common law claim for money had and received.

In 1996 Parliament enacted the Arbitration Act giving an arbitration tribunal express power, unless otherwise agreed by the parties, to "award simple or compound interest from such dates, at such rates and with such rests as it considers meets the justice of the case" (section 49(3)) In 1998 Parliament enacted the Late Payment of Commercial Debts (Interest) Act 1998, later amended by the Late Payment of Commercial Debts Regulations 2002 (SI 2002 No. 1674), providing for a right to simple interest on the unpaid price of goods or services as from any agreed date of payment, or otherwise as from 30 days after the supplier's performance or notice to the purchaser of the amount of the debt or, where that amount is unascertained, the amount claimed as the debt.

Following the Westdeutsche decision, the Law Commission in its Seventh Programme of Law Reform (1999) Law Com 259. p. 11, recommended an examination of the courts' power to award compound interest. This led to its consultation paper (No. 167) in 2002 and its "Report on Pre-Judgment Interest on Debts and Damages" (No. 287) of 23 February 2004. The Law Commission proceeded in the light of authority on the basis that at common law no interest was recoverable on a claim for money had and received (para. 2.42 of the consultation paper). It understood (para. 5.16 of the Report) that:

To this, one might add that it is not very many decades ago that double digit inflation was known in this country, and experience in the Privy Council of appeals from the West Indies, where it is still prevalent, includes banking cases where claims have multiplied several times in size with compound interest.

To allay the fears it had identified, the Law Commission observed in para. 5.17 of the Report that compound interest was already paid in common situations and that

In para. 5.27-5.28 it addressed three reasons put to it for preferring simple to compound interest (viz to protect poor consumers, to discourage claimants from deliberately delaying claims and to avoid the cost of calculating compound interest), by saying (para. 5.28) that