[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	United Kingdom House of Lords Decisions
You are here: BAILII >> Databases >> United Kingdom House of Lords Decisions >> Three Rivers District Council and Others v. Governor and Company of The Bank of England [2000] UKHL 33; [2000] 3 All ER 1; [2000] 2 WLR 1220 (18th May, 2000) URL: http://www.bailii.org/uk/cases/UKHL/2000/33.html Cite as: [1999] 4 All ER 800, [2000] 2 WLR 1220, [2000] 3 All ER 1, [2000] 2 WLR 15, [2003] 2 AC 1, [2000] UKHL 33

[New search] [Buy ICLR report: [2000] 2 WLR 1220] [Buy ICLR report: [2003] 2 AC 1] [Help]

Three Rivers District Council and Others v. Governor and Company of The Bank of England [2000] UKHL 33; [2000] 3 All ER 1; [2000] 2 WLR 1220 (18th May, 2000)

HOUSE OF LORDS

Lord Steyn Lord Hope of Craighead Lord Hutton Lord Hobhouse of Woodborough Lord Millett

OPINIONS OF THE LORDS OF APPEAL FOR JUDGMENT

IN THE CAUSE

ON 18 MAY 2000

My Lords

    Before 1979, with limited exceptions, a deposit-taking institution in the United Kingdom required no licence or other authorisation before it commenced business. There was no statutory regulation of its subsequent performance. But the Bank of England operated an informal system of supervision. The Banking Act of 1979, enacted to give effect in domestic law to the First Council Banking Co-ordination Directive of 12 December 1977 (77/780/E.E.C.), introduced a statutorily based licensing system. Subsequently, the Banking Act 1987 replaced that system. For the purposes of the First Council Banking Co-ordination Directive of 12 December 1977 (77/780/E.E.C.), the Banking Act of 1979 and the Banking Act 1987 the Bank of England was the supervisory authority in the United Kingdom. On 1 June 1998, pursuant to the Bank of England Act 1998, the Financial Services Authority assumed the Bank's powers and responsibilities under the Banking Act 1987, for the supervision of deposit-taking institutions.

    The Bank of Credit and Commerce International S.A. ("B.C.C.I."), a Luxembourg corporation, had carried on business in the United Kingdom as a deposit taking institution before the Act of 1979 came into force. When the Act of 1979 came into force B.C.C.I. came under the aegis of the new system. In June 1980 the Bank of England granted a licence to carry on business as a deposit-taking institution to B.C.C.I. Until 5 July 1991 B.C.C.I. carried on business at its principal place of business in the City of London, and at many branches elsewhere in the United Kingdom. On this date, the Bank petitioned the High Court to appoint joint provisional liquidators to B.C.C.I. The order was duly made. This resulted in the closure of B.C.C.I. in the United Kingdom, and led to the collapse of associated companies of B.C.C.I. in many jurisdictions. Thousands of depositors in the United Kingdom and elsewhere suffered substantial losses. The principal cause of the collapse of B.C.C.I. was fraud on a vast scale perpetrated at a senior level in B.C.C.I.

    The plaintiffs are more than 6,000 persons who claim to have been depositors with United Kingdom branches of B.C.C.I. The action was started in May 1993. It is unnecessary to trace the earlier procedural history of this litigation. By August 1995 the claim was formulated in a re-amended statement of claim. This is a detailed and complicated pleading. It runs to 133 pages. In outline there are two alleged causes of action. The first is based on the tort of misfeasance in public office. The plaintiffs allege that named senior officials of the Banking Supervision Department of the Bank, but not two successive Governors of the Bank, acted in bad faith (a) in licensing B.C.C.I. in 1979, when they knew that it was unlawful to do so; (b) in shutting their eyes to what was happening at B.C.C.I. after the licence was granted; and (c) in failing to take steps to close B.C.C.I. when the known facts cried out for action at least by the mid 80s. The second cause of action is based on alleged breaches of Community law, and in particular breaches of the requirements of the Directive of 1977. The alleged breaches cover the initial licensing of B.C.C.I., failure to supervise B.C.C.I. and failure to revoke the licence of B.C.C.I. The total damages claimed are apparently of the order of £550m., plus interest. In a defence the Bank comprehensively denied the material allegations under both heads of claim.

    On an application by the Bank, which was opposed by the plaintiffs, Clarke J. (now Clarke L.J.) ordered preliminary questions to be tried. This order was made on 19 July 1995 at a stage when discovery had not yet taken place. The judge directed that the questions should be tried on the assumption that the facts pleaded in the re-amended statement of claim were true. The preliminary issues were designed to test whether, if the pleaded facts are true, the causes of action based on the tort of misfeasance in public office and on breaches of community law are sustainable in law. The principal legal issues for decision were the precise ingredients of the tort of misfeasance in public office and whether the Directive of 1977 conferred rights of compensation on depositors.

    The judge tried the preliminary issues as subsequently reformulated in stages. He delivered judgments on 1 April 1996, 10 May 1996 and 30 July 1997. The first two judgments are reported at [1996] 3 All E.R. 558 and 634. These impressive and careful judgments dealt with the preliminary issues. The judge ruled that both causes of action were unsustainable. The third is an unreported judgment which considered further proposed amendments to the plaintiff's statement of claim. The judge concluded (on the assumption that his earlier rulings were correct) that the plaintiffs' claim was bound to fail and that it should be struck out. On 2 October 1997 Clarke J. struck out the re-amended statement of claim and dismissed the action. He gave leave to appeal.

    By a majority (Hirst and Robert Walker L.JJ.) the Court of Appeal dismissed an appeal and for broadly similar reasons affirmed the decisions of Clarke J. Auld L.J. dissented. These judgments are lengthy and carefully reasoned. The judgments are reported at [2000] 2 W.L.R. 15.

    The present appeal to the House, described as the plaintiffs' legal appeal, is brought by the plaintiffs with the leave of the Court of Appeal given on 21 January 1999. The order of the Court of Appeal contemplated that the House would determine "the legal issues as to the correct test for misfeasance in public office . . . before any consideration of whether the facts alleged or capable of being alleged are capable of meeting that test". At the same time the legal appeal requires the House to consider whether properly construed the Directive of 1977 confers rights on depositors. Being the court of last resort in the United Kingdom the House may only determine the Community law issue if the matter is truly acte clair.

    In a disappointingly uninformative joint statement of facts of issues the issues arising on the appeal are formulated as follows:

Whether on the assumption that the facts pleaded in the re-amended statement of claim are true the Bank is capable of being liable to the plaintiffs for the tort of misfeasance in public office.

Whether on the assumption that the facts pleaded in the re-amended statement of claim are true, the Bank is capable of being liable to the plaintiffs in damages for violation of the requirements of the First Council Banking Co-ordination Directive of 12 December 1977 (77/780/E.E.C.).

Whether on the assumption that the facts pleaded in the re-amended statement of claim are true, the plaintiffs' losses are capable of having been caused in law by the acts or omissions of the Bank.

Whether on the assumption that the facts pleaded in the re-amended statement of claim are true, the Bank is capable of being liable for the tort of misfeasance in public office to plaintiffs who were potential depositors at the time of any relevant act or omission of misfeasance by the Bank.

    A strategy which differentiates between the issues affecting the tort of misfeasance in public office and the Community law issues is necessary. It is certainly possible to state, so far as is relevant, the ingredients of the tort of misfeasance in public office. What will not be possible at this stage is to embark on the exercise contemplated by the agreed issues viz. to test at this stage the sustainability of the case pleaded in the re-amended statement of claim against the requirements of the tort as stated by the House. In granting leave to appeal the Court of Appeal realistically foreshadowed that it would be necessary to postpone the question "whether the facts alleged or capable of being alleged are capable of meeting that test" i.e. the tort enunciated by the House. That exercise will indeed require exploration at a further hearing. On the other hand, the Community law issue raises the question of interpretation whether the Directive of 1977 conferred rights of reparation on depositors. If the matter is acte clair, the House can rule dispositively on this part of the case.

The early history

    The history of the development of the tort has been described by Clarke J. and in the judgments in the Court of Appeal: see also Arrowsmith, Civil liability and Public Authorities, (1992), pp. 226-234. It is traceable to the 17th century: Turner v. Sterling (1671) 2 Vent. 24. But the first solid basis for this new head of tort liability, based on an action on the case, is to be found in Ashby v. White (1703), best reported in 1 Smith's Leading Cases (13th ed.) 253. The view ultimately prevailed that an action would lie by an elector who was wilfully denied a right to vote by a returning officer. Despite the recognition of the tort in a number of cases in the 18th and 19th centuries, the Court of Appeal in 1907 denied the existence of the tort in Davis v. Bromley Corporation [1908] 1 KB 170. But by 1981 the Privy Council described the tort as "well established:" Dunlop v. Woollahra Municipal Council [1982] AC 158, at 172F. An examination of the ingredients of the tort was still required. The first step towards that goal was the judgments in the Court of Appeal in Bourgoin S.A. v. Ministry of Agriculture, Fisheries and Food [1986] Q.B. 716. The present case is the first occasion on which the House has been called on to review the requirements of the tort in a comprehensive manner. Your Lordships are however not asked to prepare an essay on the tort of misfeasance in public office but to state the ingredients of the tort so far as it may be material to the concrete disposal of the issues arising on the pleadings in this case.

    The coherent development of the law requires the House to consider the place of the tort of misfeasance in public office against the general scheme of the law of tort. It is well established that individuals in the position of the depositors cannot maintain an action for compensation for losses they suffered as a result of the Bank's breach of statutory duties: Yuen Kun-Yeu v. Attorney-General of Hong Kong [1988] A.C. 175(P.C.); Davis v. Radcliffe [1990] 1 W.L.R. 821(P.C.) Judicial review is regarded as an adequate remedy. Similarly, persons in the position of the depositors cannot sue the Bank for losses resulting from the negligent licensing, supervision or failure to withdraw a licence: Yuen Kun-Yeu v. Attorney-General of Hong Kong; Davis v. Radcliffe. The availability of the tort of misfeasance in public office has been said to be one of the reasons justifying the non-actionability of a claim in negligence where there is an act of maladministration: Calveley v. Chief Constable of the Merseyside Police [1989] A.C. 1228, at 1238F. It is also established that an ultra vires act will not per se give rise to liability in tort: X (Minors) v. Bedfordshire County Council [1995] 2 AC 633. And there is no overarching principle in English law of liability in tort for "unlawful, intentional and positive acts": see Lonrho Ltd. v. Shell Peroleum Co. Ltd. (No. 2) [1982] A.C. 173, 187G in which the House refused to follow Beaudesert Shire Council v. Smith (1966) 120 C.L.R. 145, which was subsequently overruled by the Australian High Court in Northern Territory v. Mengel (1995) 69 A.J.L.R. 527. The tort of misfeasance in public office is an exception to "the general rule that, if conduct is presumptively unlawful, a good motive will not exonerate the defendant, and that, if conduct is lawful apart from motive, a bad motive will not make him liable": Winfield and Jolowicz on Tort, 15th ed., (1998), p. 55; Bradford Corporation v. Pickles [1895] AC 587; Allen v. Flood [1898] A.C. 1. The rationale of the tort is that in a legal system based on the rule of law executive or administrative power "may be exercised only for the public good" and not for ulterior and improper purposes: Jones v. Swansea City Council [1990] 1 W.L.R. 54, 85F, per Nourse L.J.; a decision reversed on the facts but not on the law by the House of Lords: [1990] 1 W.L.R. 1453, at 1458. The tort bears some resemblance to the crime of misconduct in public office: Reg. v. Bowden [1996] 1 W.L.R. 98.

The ingredients of the tort

    It is now possible to consider the ingredients of the tort. That can conveniently be done by stating the requirements of the tort in a logical sequence of numbered paragraphs.

(1) The defendant must be a public officer:

It is the office in a relatively wide sense on which everything depends. Thus a local authority exercising private-law functions as a landlord is potentially capable of being sued: Jones v. Swansea City Council. In the present case it is common ground that the Bank satisfies this requirement.

(2) The second requirement is the exercise of power as a public officer:

    This ingredient is also not in issue. The conduct of the named senior officials of the Banking Supervision Department of the Bank was in the exercise of public functions. Moreover, it is not disputed that the principles of vicarious liability apply as much to misfeasance in public office as to other torts involving malice, knowledge or intention: Racz v. Home Office [1994] 2 A.C. 45.

(3) The third requirement concerns the state of mind of the defendant.

    The case law reveals two different forms of liability for misfeasance in public office. First there is the case of targeted malice by a public officer i.e. conduct specifically intended to injure a person or persons. This type of case involves bad faith in the sense of the exercise of public power for an improper or ulterior motive. The second form is where a public officer acts knowing that he has no power to do the act complained of and that the act will probably injure the plaintiff. It involves bad faith inasmuch as the public officer does not have an honest belief that his act is lawful.

    The distinction, and the availability of an action of the second type, was inherent in the early development of tort. A group of cases which began with Ashby v. White (1703), reported in 1 Smith's Leading Cases (13th ed.) 253, concerned the discretionary refusal of voting rights: see also Drewe v. Coulton (1787) 1 East 563n; 102 E.R. 217; Tozer v. Child (1857) 7 El. & Bl. 377: 119 E.R. 1286; Cullen v. Morris (1819) 2 Stark 577; 171 E.R. 741. In the second group of cases the defendants were judges of inferior courts, and the cases concerned liability of the judges for malicious acts within their jurisdiction: Ackerley v. Parkinson (1815) 3 M. & S. 411; 105 E.R. 665; Harman v. Tappenden (1801) 1 East 555; 102 E.R. 214; Taylor v. Nesfield (1854) 3 El. & Bl. 724; 118 E.R. 1312. These decisions laid the foundation of the modern tort; they established the two different forms of liability; and revealed the unifying element of conduct amounting to an abuse of power accompanied by subjective bad faith. In the most important modern case in England the existence of the two forms of the tort was analysed and affirmed: Bourgoin S.A. v. Ministry of Agriculture, Fisheries and Food [1986] Q.B. 716. Clarke J. followed this traditional twofold classification. He expressly held that the two forms are alternative ways in which the tort can be committed. The majority in the Court of Appeal commented on "a rather rigid distinction between the two supposed limbs of the tort" and observed that there was "the need to establish deliberate and dishonest abuse of power in every case:" [2000] 2 W.L.R. 15 at 67C-D. As a matter of classification it is certainly right to say that there are not two separate torts. On the other hand, the ingredients of the two forms of the tort cannot be exactly the same because if that were so there would be no sense in the twofold classification. Undoubtedly there are unifying features, namely the special nature of the tort, as directed against the conduct of public officers only, and the element of an abuse of public power in bad faith. But there are differences between the alternative forms of the tort and it is conducive to clarity to recognise this.

The present case is not one of targeted malice. If the action in tort is maintainable it must be in the second form of the tort. It is therefore necessary to consider the distinctive features of this form of the tort. The remainder of my judgment will be directed to this form of the tort.

    The basis for the action lies in the defendant taking a decision in the knowledge that it is an excess of the powers granted to him and that it is likely to cause damage to an individual or individuals. It is, not every act beyond the powers vesting in a public officer which will ground the tort. The alternative form of liability requires an element of bad faith. This leads to what was a disputed issue. Counsel for the Bank pointed out that there was no precedent in England before the present case which held recklessness to be a sufficient state of mind to ground the tort. Counsel argued that recklessness was insufficient. The Australian High Court and the Court of Appeal of New Zealand have ruled that recklessness is sufficient: Northern Territory v. Mengel (1995) 69 A.J.L.R. 527; Garrett v. Attorney-General [1997] 2 N.Z.L.R. 332; Rawlinson v. Rice [1997] 2 N.Z.L.R. 651. Clarke J. lucidly explained the reason for the inclusion of recklessness [1996] 3 All E.R. 558, 581:

"The reason why recklessness was regarded as sufficient by all members of the High Court in Mengel is perhaps most clearly seen in the judgment of Brennan J. It is that misfeasance consists in the purported exercise of a power otherwise than in an honest attempt to perform the relevant duty. It is that lack of honesty which makes the act an abuse of power."

The Court of Appeal accepted the correctness of this statement of principle: [2000] 2 W.L.R. 15, 61G-62A. This is an organic development, which fits into the structure of our law governing intentional torts. The policy underlying it is sound: reckless indifference to consequences is as blameworthy as deliberately seeking such consequences. It can therefore now be regarded as settled law that an act performed in reckless indifference as to the outcome is sufficient to ground the tort in its second form.

    Initially, counsel for the plaintiffs argued that in this context recklessness is used in an objective sense. Counsel said that the distinction was between subjective or advertent recklessness in the sense used in Reg. v. Cunningham [1957] 2 Q.B. 396 and objective recklessness as explained in Reg. v. Caldwell [1982] AC 341 and Reg. v. Lawrence [1982] A.C. 510. The latter ingredient is present where in a case of an obvious risk the defendant failed to give any thought to the possibility of its existence: see Smith and Hogan, Criminal Law, 9th ed., (1999) pp. 60-69. Smith and Hogan trenchantly observed, at p. 67:

"The Caldwell test fails to make a distinction which should be made between the person who knowingly takes a risk and the person who gives no thought to whether there is a risk or not. And, on the other hand, it makes a distinction which has no moral basis. The person who, with gross negligence, fails to consider whether there is a risk is liable; but the person who considers whether there is a risk and, with gross negligence, decides there is none, is not liable. The right solution, it is submitted, is to go back to the Cunningham test which appears to have been entirely trouble-free in practice."

Counsel argued for the adoption of the Caldwell test in the context of the tort of misfeasance in public office. The difficulty with this argument was that it could not be squared with a meaningful requirement of bad faith in the exercise of public powers which is the raison d'être of the tort. But, understandably, the argument became more refined during the oral hearing and counsel for the plaintiffs accepted that only reckless indifference in a subjective sense will be sufficient. This concession was rightly made. The plaintiff must prove that the public officer acted with a state of mind of reckless indifference to the illegality of his act: Rawlinson v. Rice [1997] 2 N.Z.L.R. 651. Later in this judgment I will discuss the requirement of reckless indifference in relation to the consequences of the act.

(4) Duty to the plaintiff

    The question is who can sue in respect of an abuse of power by a public officer. Counsel for the Bank argued that in order to be able to claim in respect of the second form of misfeasance, there must be established "an antecedent legal right or interest" and an element of "proximity". Clarke J. did not enunciate a requirement of proximity. He observed [1996] 3 All E.R. 558, 584B):

"If an officer deliberately does an act which he knows is unlawful and will cause economic loss to the plaintiff, I can see no reason in principle why the plaintiff should identify a legal right which is being infringed or a particular duty owed to him, beyond the right not to be damaged or injured by a deliberate abuse of power by a public officer"

The majority in the Court of Appeal held that "the notion of proximity should have a significant part to play in the tort of misfeasance, as it undoubtedly has in the tort of negligence:" [2000] 2 W.L.R. 15, 66A. Counsel for the Bank argued that both requirements are essential in order to prevent the tort from becoming an uncontrollable one. It would be unwise to make general statements on a subject which may involve many diverse situations. What can be said is that, of course, any plaintiff must have a sufficient interest to found a legal standing to sue. Subject to this qualification, principle does not require the introduction of proximity as a controlling mechanism in this corner of the law. The state of mind required to establish the tort, as already explained, as well as the special rule of remoteness hereafter discussed, keeps the tort within reasonable bounds. There is no reason why such an action cannot be brought by a particular class of persons, such as depositors at a bank, even if their precise identities were not known to the bank. The observations of Clarke J. are correct.

    In agreed issue 4 the question is raised whether the Bank is capable of being liable for the tort of misfeasance in public office to plaintiffs who were potentially depositors at the time of any relevant act or omission of misfeasance by the Bank. The majority in the Court of Appeal and Auld L.J. held that this issue is unsuitable for summary determination. In my view this ruling was correct.

(5) Causation

    Causation is an essential element of the plaintiffs' cause of action. It is a question of fact. The majority in the Court of Appeal and Auld L.J. held that it is unsuitable for summary determination. That is plainly correct. This conclusion disposes of agreed issue 3 so far as it relates to the tort of misfeasance.

(6) Damage and Remoteness

    The claims by the plaintiffs are in respect of financial losses they suffered. These are, of course, claims for recovery of consequential economic losses. The question is when such losses are recoverable. It would have been possible, as a matter of classification, to discuss this question under paragraph 3 in which the required state of mind for this tort was examined. It is, however, convenient to consider it under the traditional heading of remoteness.

    On the assumption that the other requirements can be established, counsel for the plaintiffs argued that the plaintiffs should be able to recover all reasonably foreseeable losses suffered by them. In support of this argument he had the advantage of a powerfully reasoned dissenting judgment by Auld L.J. Counsel for the Bank argued that the rule is more restrictive. He supported the conclusion of the majority in the Court of Appeal. The judge had held that the plaintiffs must prove that the Bank actually foresaw the losses to the plaintiff as a probable consequence. This part of the judgment at first instance provided the reason for the judge refusing to allow the proposed amendments and striking out the claims. The majority observed [2000] 2 W.L.R. 15, 102A:

"[The] formulation, however, may have been too favourable to the plaintiffs. In view of the stringent requirements of the tort of misfeasance in public office, the more appropriate question may be: Is it reasonably arguable that the Bank at any stage made an unlawful and dishonest decision knowing at the time that it would cause loss to the plaintiffs? To that question, in the light of our analysis of the evidence, the answer is plainly 'No.'"

Counsel adopted this formulation as his primary submission. In the alternative he submitted that the test stated by Clarke J. should be adopted.

    It will be necessary to give a brief account of the decisions in which this issue was considered. It was first touched on in Bourgoin S.A. v. Ministry of Agriculture [1986] Q.B. 716. At first instance Mann J. had spoken of foreseeable losses. Oliver L.J. quoted and endorsed the relevant passage. In Northern Territory v. Mengel 69 A.L.J.R. 527, at 540 the majority in the Australian High Court adopted a test of "a foreseeable risk of harm" for which it relied on Bourgoin. In the present case Clarke J. concluded that in using the word "foreseeable" in Bourgoin Mann J. must have meant "foreseen" and that the same applies to the adoption of the relevant passage by Oliver L.J. Before the judgments in the Court of Appeal in the present case the Court of Appeal of New Zealand adopted the conclusions of Clarke J. as well as his explanation of Bourgoin: Garrett v. Attorney-General [1997] 2 N.Z.L.R. 332; Rawlinson v. Rice [1997] 2 N.Z.L.R. 651. In England the Court of Appeal and Divisional Court have on a number of occasions approved the reasoning of Clarke J. These decisions include the following: Lam v. Brennan [1997] 3 P.L.R. 22 (C.A.); Reg. v. Chief Constable of the North Wales Police, Ex parte A.B. [1999] QB 396 (D.C.); Barnard v. Restormel Borough Council [1998] 3 P.L.R. 27 (C.A.); W. v. Essex County Council [1999] Fam. 90 (C.A.) While it is unnecessary to discuss these decisions it is relevant to point out that in the North Wales Police case the Lord Chief Justice expressed agreement with the view that the tort is only established if the officer had knowledge that he had no power to do the act complained of and that the act would probably injure the plaintiff. He paid tribute to the "extended consideration and most helpful summary" by Clarke J. at [1999] QB 396, 413B.

    The issues have been canvassed in great depth in written and oral argument. Taking into account all the matters advanced the choice before the House can be narrowed down. So far as the majority was minded to adopt a stricter test than Clarke J., encapsulated in the words "knowing at the time that [the decision] would cause damage to the plaintiffs," they went too far. A test of knowledge or foresight that a decision would cause damage does not readily fit into the standard of proof generally required in the law of tort, and specifically in the case of intentional torts. Moreover, this test unnecessarily emasculates the effectiveness of the tort. The real choice is therefore between the test of knowledge that the decision would probably damage the plaintiff (as enunciated by Clarke J.) and the test of reasonable foreseeability (as contended for by counsel for the plaintiffs).

    It is now necessary to return to the Bourgoin case. While all judges are prone to error and imprecise language from time to time, it is difficult to say that Mann J. and Oliver L.J. used the word "foreseeable" when they meant "foreseen." It is sufficient to point out, as the majority of the Court of Appeal did, at p. 484D, that there was no focus in the Bourgoin case on the choice which is now before the House. In these circumstances the observations in Bourgoin on this particular issue do not greatly assist.

    It is true that Clarke J. made new law. He relied on the special nature of the tort. He reasoned from legal principle. It is true that the earlier decision of the majority in the Mengel case runs counter to the conclusion of Clarke J. But apart from the Mengel case there has however been no judicial support for a foreseeability test. And there has been no academic criticism of the view of Clarke J. that a test of foreseeability is not enough in this tort. Given that his ground-breaking first instance judgment has been pored over by many judicial and academic eyes, this is a factor of some significance. Nevertheless, it is necessary to consider the merits of the competing solutions from the point of view of principle and legal policy.

    Enough has been said to demonstrate the special nature of the tort, and the strict requirements governing it. This is a legally sound justification for adopting as a starting point that in both forms of the tort the intent required must be directed at the harm complained of, or at least to harm of the type suffered by the plaintiffs. This results in the rule that a plaintiff must establish not only that the defendant acted in the knowledge that the act was beyond his powers but also in the knowledge that his act would probably injure the plaintiff or person of a class of which the plaintiff was a member. In presenting a sustained argument for a rule allowing recovery of all foreseeable losses counsel for the plaintiffs argued that such a more liberal rule is necessary in a democracy as a constraint upon abuse of executive and administrative power. The force of this argument is, however, substantially reduced by the recognition that subjective recklessness on the part of a public officer in acting in excess of his powers is sufficient. Recklessness about the consequences of his act, in the sense of not caring whether the consequences happen or not, is therefore sufficient in law. This justifies the conclusion that the test adopted by Clarke J. represents a satisfactory balance between the two competing policy considerations, namely enlisting tort law to combat executive and administrative abuse of power and not allowing public officers, who must always act for the public good, to be assailed by unmeritorious actions.

    It is undoubtedly right, as counsel for the plaintiffs pointed out, that the mental element required for the tort of misfeasance in public office means that it is not an effective remedy to deal with state liability for breaches of Community law: Brasserie du Pêcheur S.A. v. Federal Republic of Germany; Reg. v. Secretary of State for Transport, Ex parte Factortame (No. 4) Joined Cases C-46/93 and C-48/93 [1996] QB 404. This consideration cannot, however, affect the decision of the House on the tort. If there is a gap it must be for Community law to fill it. And our courts will loyally apply Community law.

Conclusion on misfeasance in public office

    For the reasons given the requirements of the tort are as set out.

My Lords, I have had the advantage of reading in draft the speech of Lord Hope of Craighead. He has demonstrated with compelling logic that the Directive of 1977 was not intended to confer rights on individual depositors. I am persuaded that the matter is truly acte clair.

It will be necessary to take account of the following matters:

The question whether the existing re-amended statement of claim reveals a sustainable cause of action based on the tort of misfeasance in public office will have to be considered at a further hearing of the Appellate Committee.

The next hearing will include the questions whether the Court of Appeal was right to affirm the decisions of Clarke J. who refused to allow proposed amendments and struck out the action.

The shape of the case has been altered. The requirements of the tort, so far as relevant to the present case, have today been authoritatively stated. The allegations of breaches of Community law can no longer found a cause of action.

In these circumstances I take the view that at a further hearing there should be available a new draft pleading by the plaintiffs reflecting the altered position. And I draw attention to the fact that the Court of Appeal referred to "the facts alleged or capable of being alleged", that being a reference to the circumstances in which it is proper to strike out an action.

Given this untidy procedural position, it may be necessary when the parties are ready for an Appeal Committee to consider the future progress of the matter.

    For the reasons given by Lord Hope of Craighead, I would dismiss the appeal on Community law issues. In the light of my statement of the requirements of the tort of misfeasance in public office I would adjourn this part of the appeal for further argument.

LORD HOPE OF CRAIGHEAD

My Lords,

    I have had the advantage of reading in draft the speeches prepared by my noble and learned friends Lord Steyn and Lord Hutton. As regards the tort of misfeasance in public office, I am in full agreement with what they have said as to the essential elements of the tort and the requirements which must be satisfied. The question with which I wish to deal is whether the appellants have a basis for an action of damages against the Bank in Community law.

Community Law

    The appellants' claim that they are entitled to damages for losses caused by breaches of Community law is based upon the following allegations. First, it is alleged that in June 1980 the Bank granted to B.C.C.I. S.A. a full licence to carry on business as a deposit-taker deliberately contrary to the scheme laid down by the First Council Banking Co-ordination Directive (77/780/E.E.C.) of 12 December 1977 on the co-ordination of laws, regulations and administrative provisions relating to the taking up and pursuit of the business of credit institutions ("the Directive of 1977") and the Banking Act 1979 when it knew that the relevant criteria in Schedule 2 to the Act of 1979 were not fulfilled. Then it is alleged that, at all times after the grant of the licence in June 1980 until the eventual closure of B.C.C.I. S.A., the Bank continued to act contrary to the scheme laid down by the Directive of 1977, the Act of 1979 and the Banking Act 1987 in the respects described in paragraphs 40 - 44 of the re-amended statement of claim. The essence of this further allegation is that, contrary to the scheme laid down by the Directive of 1977, the Act of 1979 and the Act of 1987, the Bank concluded that it had no discretion or power to revoke the licence to carry on business as a deposit-taker when it knew that B.C.C.I. S.A. had conducted and was conducting its affairs in a way which threatened the interests of its depositors. Furthermore, it permitted B.C.C.I. Overseas to carry on a deposit-taking business when it knew that the business of that company was carried on in a manner which might affect the soundness of B.C.C.I. S.A. and place its depositors at risk. It is also alleged that throughout this period the Bank failed to supervise both B.C.C.I. S.A. and B.C.C.I. Overseas to the detriment of the depositors.

    Shortly put, and based upon these allegations, the appellants' Community law argument proceeds along these lines. Under the E.E.C. Treaty the Directive of 1977 has direct effect in the United Kingdom. The United Kingdom discharged its obligations under the Directive by enacting the Banking Acts of 1979 and 1987. The Bank was at all material times the supervisory authority in the United Kingdom for the purposes of both the Directive of 1977 and the Act of 1979 and the Act of 1987. The Bank as an emanation of the state is liable to the depositors for failing in its functions as supervisory authority to give full effect to the Directive of 1977. National courts are required by Community law to protect the rights which individuals derive from Community law, including those which are derived from a Directive. Accordingly the appellants are entitled, as parties who were intended to benefit from the Directive of 1977, to rely upon its terms against the Bank in order to obtain damages.

    The appellants maintain that the Directive of 1977 was intended and designed to protect the savings of depositors. They say that, in order to achieve this purpose, it imposed certain well-defined Community law obligations on the competent authorities of member states in relation to the authorisation and supervision of banks, and that it conferred on depositors and potential depositors corresponding Community law rights against the competent authorities to have these obligations fulfilled. That being so, their Community law rights under the Directive of 1977 must prevail over the requirement in national law to prove bad faith or dishonesty as a prerequisite of the tort of misfeasance in public office under the common law. And they must prevail over the Bank's right to seek exemption from liability under section 1(4) of the Banking Act 1987, which provides:

"Neither the Bank nor any person who is a member of its Court of Directors or who is, or is acting as, an officer or servant of the Bank shall be liable in damages for anything done or omitted in the discharge or purported discharge of the functions of the Bank under this Act unless it is shown that the act or omission was in bad faith."

    The main issue of Community law which arises in this appeal from these allegations is whether the Bank is capable of being liable to the appellants in damages for violations of the Directive of 1977, on the assumption that the facts pleaded in the re-amended statement of claim are true.

    There appeared to be no real dispute between the parties on this point in the course of the argument which was presented to your Lordships, and both the majority judgment in the Court of Appeal [2000] 2 W.L.R. 15, 70E - 80D and the judgment of Auld L.J., pp. 106G - 110G have dealt with the whole matter in great detail and with admirable clarity. Nevertheless I think that it is necessary for me to explain at the outset of this chapter the criteria which must be fulfilled before a Directive can be relied upon as a source of rights before a national court.

    Community law, as it has been developed by the European Court of Justice, is capable of conferring upon individuals the right to claim damages from a national authority by one or other or both of two distinct routes. The purpose of the right to claim damages is to ensure that provisions of Community law prevail over national provisions. This is because the full effectiveness of Community law would be impaired if individuals were unable to obtain redress in the national courts of the relevant member state when their rights were infringed by a breach of Community law: Brasserie du Pêcheur S.A. v. Federal Republic of Germany; Reg. v. Secretary of State for Transport, Ex parte Factortame Ltd. (No. 4) (Joined Cases C-46/93 and C-48/93) [1996] QB 404, 495, para. 20. The first route by which the right to claim damages against the state or an emanation of the state for the non-implementation or misimplementation of a Directive may be asserted is based upon the principle of direct effect. This is the principle which was established in Community law by N.V. Algemene Transport- en Expeditie Onderneming van Gend & Loos v. Nederlandse administratie der belastingen (Case 26/62) [1963] ECR 1. The second route is based upon the principle of state liability.

    In the van Gend & Loos case it was held that article 12 of the E.E.C. Treaty (now article 25 E.C.) prohibiting customs duties on imports and exports had to be interpreted as producing direct effects and creating individual rights which national courts must protect. The decision in that case has been applied by the European Court in a large number of cases to other articles of the Treaty which the Court has construed as having direct effect in member states. Later decisions of the European Court have made it clear, in the light of the provisions of the third paragraph of article 189 of the E.E.C. Treaty (now article 249 E.C.) which provides that a Directive shall be binding as to the result to be achieved upon each member state, that Directives as well as articles of the Treaty are capable of conferring directly effective rights upon individuals provided certain conditions are satisfied. In Becker v. Finanzamt Munster-Innenstadt (Case 8/81) [1982] ECR 53, 70-71 the court made the following observations as to the conditions for the application of the direct effect principle to a Directive:

"22. It would be incompatible with the binding effect which article 189 ascribes to Directives to exclude in principle the possibility of the obligations imposed by them being relied on by persons concerned.

"23. Particularly in cases in which the Community authorities have, by means of a Directive, placed member states under a duty to adopt a certain course of action, the effectiveness of such a measure would be diminished if persons were prevented from relying upon it in proceedings before a court and national courts were prevented from taking it into consideration as an element of Community law.

"24. Consequently, a member state which has not adopted the implementing measures required by the Directive within the prescribed period, may not plead as against individuals, its own failure to perform the obligations which the Directive entails.

"25. Thus, wherever the provisions of a Directive appear, as far as their subject matter is concerned, to be unconditional and sufficiently precise, those provisions may, in the absence of implementing measures adopted within the prescribed period, be relied upon as against any national provision which is incompatible with the Directive or in so far as the provisions define rights which individuals are able to assert against the state."

    Two things should be noted about the observations which the European Court made in paragraph 25 of its judgment. The first is that in order for there to be liability under this principle, which the Court of Appeal in the judgments in this case has described as Becker-type liability, the rights said to have been conferred by the Directive must be "unconditional and sufficiently precise". The second is that a distinction is made between relying upon a Directive to nullify some provision in national law which is incompatible with the Directive in order to give effect to rights under Community law, and relying upon Community law itself to give a right to claim damages in the national courts for breach of an obligation of Community law. The appellants seek to rely on each of these two branches in this case.

    In Francovich v. Italian Republic (Joined Cases C-6/90 and C-9/90) [1995] ICR 722 the European Court established the conditions for state liability, which is described in the judgments in the Court of Appeal as Francovich-type liability. The European Court had to deal with two issues in that case. The first concerned the direct effect of the provisions of Council Directive 80/987/E.E.C. which determined the rights of employees in the event of their employers' insolvency and which Italy had failed to implement: this was Becker-type liability. The second concerned the existence and scope of the liability of the state for damage resulting from Italy's breach of its obligations under Community law: this was Francovich-type liability. In regard to the direct effect route the Court said at p. 768, para. 12 that there were three points to be considered, in order to see whether the provisions of the Directive which determined whether the rights of employees were, in Becker-type liability terms, "unconditional and sufficiently precise" to enable them to recover under this route. These were:

"the identity of the persons entitled to the guarantee provided [by the Directive,] the content of that guarantee and the identity of the person liable to provide the guarantee."

    In regard to the conditions for state liability for failure to implement a Directive, the European Court made these observations, at p. 772:

"39. Where, as in this case, a member state fails to fulfil its obligation under the third paragraph of article 189 of the Treaty to take all the measures necessary to achieve the result prescribed by a Directive, the full effectiveness of that rule of Community law requires that that there should be a right to reparation provided that three conditions are fulfilled.

"40. The first of these conditions is that the result prescribed by the Directive should entail the grant of rights to individuals. The second condition is that it should be possible to identify the content of those rights on the basis of the provisions of the Directive. Finally, the third condition is the existence of a causal link between the breach of the state's obligation and the loss and damage suffered by the injured parties.

"41. Those conditions are sufficient to give rise to a right on the part of individuals to obtain reparation, a right founded directly on Community law."

As Lord Slynn of Hadley said in Reg. v. Secretary of State for Transport, Ex parte Factortame Ltd. (No. 5) [1999] 3 WLR 1062, 1072E, state liability is conditional on there being a grant of rights to individuals by the Directive, that the contents of these rights is clear and that the loss suffered is shown to be caused by the state's breach.

    In Dillenkoffer v. Federal Republic of Germany (Joined Cases C-178/94, C-179/94, C-188/94, C-189/94 and C-190/94) [1997] QB 259 the Court restated the conditions for state liability in the light of a number of cases with which it had dealt subsequently to the Francovich case. It did so in a manner which, in paragraph 22 of the judgment, applied the tests which Francovich had laid down for the direct effect, or Becker-type, route to the state liability, or Francovich-type, route. The relevant paragraphs are set out in the judgment at pp. I-4878-9:

"20. The court has held that the principle of state liability for loss and damage caused to individuals as a result of breaches of Community law for which the state can be held responsible is inherent in the system of the Treaty: Francovich v. Italian Republic (Joined Cases C-6/90 and C-9/90) [1995] ICR 722, 772, para. 35; Brasserie du Pêcheur S.A. v. Federal Republic of Germany; Reg. v. Secretary of State for Transport, Ex parte Factortame Ltd. (No. 4) (Joined Cases C-46/93 and C-48/93) [1996] QB 404, 496, para. 31; Reg. v. H.M. Treasury, Ex parte British Telecommunications Plc. (Case C-392/93) [1996] QB 615, 654, para. 38 and Reg. v. Ministry of Agriculter, Fisheries and Food, Ex parte Hedley Lomas (Ireland) Ltd. (Case C-5/94 [1997] QB 139, 160, para.24. Furthermore, the court has held that the conditions under which state liability gives rise to a right to reparation depend on the nature of the breach of Community law giving rise to the loss and damage: Francovich [1995] ICR 722, 772, para. 38; Brasserie du Pêcheur [1996] QB 404, 497, para. 38, and Hedley Lomas [1997] QB 139, 160, para. 24.

"21. In Brasserie du Pêcheur, at p. 499, paras. 50 and 51; British Telecommunications [1996] QB 615, 655, paras. 39 and 40, and Hedley Lomas, at p. 160, paras. 25 and 26, the court, having regard to the circumstances of the case, held that individuals who have suffered damage have a right to reparation where three conditions are met: the rule of law infringed must have been intended to confer rights on individuals, the breach must have been sufficiently serious, and there must be a direct causal link between the breach of the obligation resting on the state and the damage sustained by the injured parties.

"22. Moreover, it is clear from the Francovich case which, like the present cases, concerned non-transposition of a Directive within the prescribed period, that the full effectiveness of the third paragraph of article 189 of the Treaty requires that there should be a right to reparation where the result prescribed by the Directive entails the grant of rights to individuals, the content of those rights is identifiable on the basis of the provisions of the Directive and a causal link exists between the breach of the state's obligation and the loss and damage suffered by the injured parties.

"23. In substance, the conditions laid down in that group of judgments are the same, since the condition that there should be a sufficiently serious breach, although not expressly mentioned in Francovich, was nevertheless evident from the circumstances of that case."

    In the Court of Appeal it was observed in the majority judgment [2000] 2 W.L.R. 15, 77C that the main difference between the parties, as to the basic principles to be applied, was as to whether a Directive, once it has been transposed into national law, ceases to be the immediate source of rights enforceable by an individual claimant in his national court. For the Bank it was submitted that there is a clear and well-established principle that the Directive is supplanted at that stage by rights under the national law. For the appellants it was submitted that, as the obligation on member states is to ensure that the Directive is applied in practice, a Directive can be the immediate source of enforceable rights under the Becker-type principle even if it has been transposed, and correctly transposed, into national law. This is in order to ensure its effectiveness as to the result to be achieved in conformity with the third paragraph of article 249 (ex 189) E.C..

    The majority in the Court of Appeal [2000] 2 W.L.R. 15, 80A-B found some support in Norbrook Laboratories Ltd. v. Ministry of Agriculture, Fisheries and Food (Case C-127/95) [1998] ECR I-1531 for the view that there may be a category of Directives in relation to which a member state's obligation of proper implementation is not restricted to a once-for-all legislative process, but also requires a continuing administrative process. Auld L.J., who was of the opinion that the appellants were entitled to rely on Francovich-type liability, regarded the debate on this point as academic: p. 139E. But, in the course of his development of this point at pp. 139F-142B, he gave two further reasons for rejecting the Bank's argument that implementation of the Directive deprived the appellants of recourse to the Directive under the Becker-type principle of liability to which, in my view, great weight should be attached. Although the debate on this issue did not receive the same prominence in the arguments which were presented to your Lordships as Mr. Lasok Q.C. for the Bank addressed the main part of his submissions to the terms of the Directive, I think that Auld L.J.'s observations are worth recording here in order to set the scene for an examination of the Directive.

    The first point which Auld L.J. made was that it could be said that the precondition of liability for damages of bad faith on the part of the Bank or its officers in a common law action for misfeasance in public office and as introduced in section 1(4) of the Banking Act 1987, to the extent that they derogate from the Directive: p. 141B, "misimplement" the Directive: p. 139F. His second point was that recent decisions of the European Court, including the Norbrook case, indicate that in the main the Court is indifferent to the precise route by which it gives effect to a Directive. As he pointed out at p. 141E-G, neither the Act of 1979 nor the Act of 1987 transposed the Directive of 1977 word for word, and the rights of redress which he found in the Directive were wider than those dependent on proof of bad faith as required by section 1(4) of the Act of 1987 and the common law action of misfeasance in public office. Having noted the differences between the two approaches - one that the United Kingdom legislation properly construed effectively implemented the Directive, the other that it did not fully implement the Directive with the consequence that the United Kingdom courts must have direct recourse to it - he concluded at p. 141H-142C:

"As Josephine Steiner observed in 'Coming to Terms with E.E.C. Directives,' [(1990)] 106 L.Q.R. 144, 146 the question whether a Directive has been correctly implemented can only be assessed by reference to the Directive itself, with the result that it can rarely be disregarded. Clearly, as a result of the court's ready application of the Francovich principle, the two approaches can shade into one another. Whichever route is taken, the answer on matters of Community law should be the same, with the result that it should prevail over United Kingdom law, including the common law as to misfeasance in public office and section 1(4) of the Act of 1987, where the latter frustrates it: see Amministrazione delle Finanze dello Stato v. Simmenthal S.p.A. (Case 106/77) [1978] ECR 629, 643-645, paras. 16-26; the Brasserie du Pêcheur case [1996] QB 404, 502, paras. 72-73."

    In the result, although the appellants' case under Community law is put in different ways and is based upon both types of liability, the conditions which the appellants must satisfy in order to establish a right to damages against the Bank under each route are so closely analogous that they can be taken to be, at this stage of case, the same. The critical questions in this appeal, following the language of paragraph 22 of the judgment in the Dillenkofer case, are whether the Directive of 1977 entails the grant of rights to individual depositors and potential depositors and whether the content of those rights is identifiable on the basis of the provisions of the Directive.

    Clarke J. [1996] 3 All E.R. 558, 602A held, contrary to the submissions which Mr. Stadlen Q.C. for the Bank had made to him, that an important underlying purpose of the Banking Acts 1979 and 1987 was to protect savers, including both existing savers and future savers, and that the same was true of the Directive of 1977. But he went on to say at p. 602B that this was not enough to impose any obligation on the Bank which gave rise to a right in the savers to claim damages for a breach of it. Having examined the terms of the Directive, he held at p. 614J that it was not intended to confer rights upon savers, even although the underlying purpose of supervision of credit institutions was to be for their benefit. In the Court of Appeal [2000] 2 W.L.R. 15, 82F the majority understood the Bank's position to be that it did not dispute that one of the Directive's purposes was the protection of depositors. Auld L.J., at p. 110H, put the matter in this way:

"It is plain that one of the purposes of the Directive was the protection of depositors. The plaintiffs say it was the main purpose. The Bank, to the extent that it recognises it as a purpose of the Directive at all, says it was subsidiary to that of beginning the process of harmonisation. As I have mentioned, the judge regarded it as an important underlying purpose."

    It was suggested by the appellants in the course of the hearing before your Lordships that the Bank had changed its position on this point having realised, as Lord Neill put it, that once the concession was made that one of the purposes of the Directive was the protection of depositors it was on a slippery slope from which it now wished to extricate itself. In my view however the position which the Bank has adopted both in its written case and in the oral argument advanced on its behalf by Mr. Lasok is based upon a more substantial argument than that which might be thought to have been suggested by that criticism. I am not convinced that there has been, in substance, any change of position on the part of the Bank from its position as Auld L.J. understood it to be. There is however a more important point. The appellants still rely, and take as their starting point on this whole issue, on the proposition that one of the purposes of the Directive was to protect depositors. This is a significant step in the argument which they then advance that the Directive also imposed obligations on the Bank which conferred corresponding rights upon which they are entitled to base their claim of damages. The question which the Bank has raised is not only as to the accuracy of the appellants' description of the purpose of the Directive but also as to its relevance as a starting point to an examination of the articles of the Directive in order to discover what rights, if any, they conferred on depositors.

    The appellants' submission is that it is evident from the legislative background to the Directive, its terms and cases decided by the European Court that a principal purpose of the Directive of 1977 was the protection of depositors. They rely on article 57(2) of the E.E.C. Treaty with reference to which the Directive of 1977 was enacted, on observations in the opinion of the Economic and Social Committee ("E.C.O.S.O.C.") mentioned in the preamble to the Directive, on recitals and articles set out in the Directive and on several decisions of the European Court of which the most important is Société Civile Immobilière Parodi v. Banque H. Albert de Bary et Cie (Case C-222/95) [1997] E.C.R. I-3899.

    I do not think that the appellants derive any assistance from article 57(2) of the E.C.C. Treaty (now, in a revised form, article 47 (2) E.C.). Its relevance is not in doubt. Article 253 E.C. (formerly article 190) requires Community instruments such as Directives to state the reasons on which they are based and to refer to any proposals or opinions which were required to be obtained pursuant to the Treaty. The duty to give reasons will normally require specification of the Treaty article on which the measure was based: Craig and de Búrca, E.U. Law 2nd ed. (1998), p. 120. In Commission of the European Communities v. Council of the European Communities (Case 45/86) [1987] E.C.R. 1493(the Tariff Preferences case) a Council measure was annulled in part by the European Court because the legal basis of the measure had not been specified. In this case the only article of the Treaty which is referred to by the Directive of 1977 is article 57. This is one of a group of articles which appear in Title III (Free Movement of Persons, Services and Capital), Chapter 2 (Right of Establishment) of the Treaty. Paragraphs (1) and (2) of article 57, in the terms which were in force in 1977, provided:

"1. In order to make it easier for persons to take up and pursue activities as self-employed persons, the Council shall, on a proposal from the Commission and after consulting the Assembly, acting unanimously during the first stage and by a qualified majority thereafter, issue Directives for the mutual recognition of diplomas, certificates and other evidence of formal qualifications.

"2. For the same purpose, the Council shall, before the end of the transitional period, acting on a proposal from the Commission and after consulting the Assembly, issue Directives for the co-ordination of the provisions laid down by law, regulation or administrative action in member states concerning the taking up and pursuit of activities as self-employed persons. Unanimity shall be required on matters which are the subject of legislation in at least one member state and measures concerned with the protection of savings, in particular the granting of credit and the exercise of the banking profession, and with the conditions governing the exercise of the medical and allied, and pharmaceutical professions in the various member states. In other cases, the Council shall act unanimously during the first stage and by a qualified majority thereafter."

    The purpose to which reference is made in the first line of article 57(2) is that of the mutual recognition of qualifications which is the subject of article 57(1). The appellants base their argument that a purpose of the Directive was to protect "savers" on the reference to the protection of "savings" in the second sentence of article 57(2). But this reference appears in provisions which laid down those matters in regard to which proposals were to be dealt with unanimously and those which could be dealt with by qualified majority. The granting of credit and the exercise of the banking profession are taken as two examples of "measures concerned with savings." This seems to me to be no more than a recognition that an ability to protect savings is one of the qualifications which member states will normally require those who wish to grant credit or exercise the banking profession to satisfy. Recognition that this was so, that such measures would need to be co-ordinated throughout member states and the making of provision for the voting formula to be adopted in regard to such matters is one thing. A purpose to direct that provision must be made for the protection of savers and depositors under Community law over and above the protections available under the national law of each member state is quite another. I do not find anything in the wording of the article as a whole to suggest that the protection of individual depositors and potential depositors against loss could be regarded as a purpose for which Directives were to be issued under it.

    In Federal Republic of Germany v. European Parliament and Council of the European Union (Case C-233/94) [1997] ECR I-2405 the European Court made certain observations about article 57(2) of the Treaty. This was in the context of a challenge to the Deposit Guarantee Scheme Directive 94/19/E.E.C. by Germany on the ground that, contrary to its preamble, article 57 could not constitute the sole legal basis for the Directive as it did not merely regulate banking operations but was aimed at increasing protection for consumers. The fact that the court rejected this challenge might seem at first sight to provide support for the view that the protection of consumers was a purpose for which Directives could be issued under article 57. But the background to the Directive is important to a proper understanding of the reason why the challenge to its legal base did not succeed. The court had held in earlier cases that member states were entitled in certain circumstances to adopt or maintain measures which were justified on public interest grounds, such as the protection of consumers, which constituted an obstacle to free movement within the Community: p. I-2450, paras. 16 and 17. At p. I-2456, para. 41 the court said that article 57(2) of the Treaty authorised the Parliament and the Council to issue Directives with a view to abolishing obstacles of this kind. It was apparent that such an obstacle was to be found in the fundamental differences between the deposit-guarantee systems existing in the various member states, so the laws on those systems were to be harmonised in order to facilitate the activity of credit institutions at Community level. At p. I-2459, para. 48 the court held that there had to be a high level of consumer protection concomitantly with the right of establishment and the freedom to provide services which the Directive aimed to promote. It referred to "the general result" which the Directive sought to achieve, which was a considerable improvement in the protection of depositors within the Community. In that particular context it was legitimate for the Directive to adopt measures which would render the domestic measures for the protection of consumers otiose. The special circumstances which led to that decision are absent in this case.

    Consultation with E.C.O.S.O.C. was required by the second paragraph of article 100 of the E.E.C. Treaty (now article 94 E.C.) prior to the issuing of the Directive. But the observations in its opinion on which the appellants rely do not seem to me to advance their argument. In paragraph 1.1.3 the point was made that "the lack of harmonisation of member states' legislation, whose main purpose in each country is to provide security for depositors and to protect savings, is liable to create serious disparities with regard to that objective, indeed even certain dangers." In paragraph 1.4.1 it was stated that the ultimate aim was to harmonise the authorisation requirements for financial institutions in all the member states. The appellants say, under reference to these and other passages in the opinion that the committee recognised that the main purpose of legislation concerning banking regulation was to provide security for depositors and to protect savings. I am willing to accept that this is so. No doubt the committee recognised that the protection of savings is a necessary part of every system at national level for the regulation of credit institutions whose business it is to receive from the public deposits and other forms of repayable funds. But the point to which it was drawing attention in its opinion was the need for the harmonisation of authorisation requirements, without which there would be likely to be serious disparities between the member states. The Community law purpose which was indicated by its observations was that of the harmonisation of regulatory measures affecting the right of establishment with a view to eliminating these disparities. I do not find any indication here that the committee saw the purpose of the Directive as being to confer Community law rights on individual depositors.

    In the Parodi case [1997] E.C.R. I-3899, p. I-3923, paras. 24-25 the European Court said that the Directive of 1977 was no more than a first step towards the mutual recognition by member states of authorisations issued by each of them to credit institutions, and that it confined itself to imposing a number of minimum conditions on member states. Member states were to be obliged to require authorisation of all credit institutions wishing to commence banking activity within their territory of origin, but this was to be subject to minimum requirements and without prejudice to other conditions of general application laid down by national laws. The question which was raised in that case was whether national legislation requiring authorisation in order to supply banking services was precluded by the Treaty where the bank concerned was already established and authorised in another member state. The court said at p. I-3922 in paragraphs 20-22 of the judgment that, in view of the special nature of certain provisions of services, specific requirements imposed on the provider that were attributable to the application of rules governing that type of activity could not be regarded as incompatible with the Treaty, and that the banking sector was a particularly sensitive area from the point of view of consumer protection. At p. I-3924, para. 26 the following observation was made:

"It must therefore be accepted that, as Community law stood at the time of the facts in the main proceedings, there were within the banking sector imperative reasons relating to the public interest capable of justifying the imposition by the member state of destination of conditions regarding access to the activity of credit institutions and their supervision which could go beyond the minimum conditions required by the first banking Directive and already implemented in the member state of origin."

    The conclusion in the Parodi case was that member states were entitled to apply their own consumer protection measures in the banking sector, pending the entry into force of the measures in the Second Council Directive 89/646/E.E.C. on the co-ordination of laws, regulations and administrative provisions relating to the taking up and pursuit of the business of credit institutions and amending the Directive of 1977 which rendered the national measures otiose. The appellants rely on the observations in the judgment about the need to protect consumers in the banking sector in support of their argument as to the purpose of the Directive of 1977. But, as I read these observations, they were made, not with reference to the purpose of the Directive of 1977, but in order to justify the application of national measures by a member state during the period prior to the entry into force of the Directive of 1989. The purpose of the Directive of 1977 was to begin the process of harmonisation of national laws so as to remove barriers to the provision of banking services throughout the single market, but without weakening or impairing the protection of depositors. The protection of depositors was seen therefore not as a purpose of the Directive but as a constraint on the provision of banking services to the public which had to be recognised.

    In Criminal Proceedings against Romanellii (Case C-366/97) [1999] All E.R. (E.C.) 473, 479, para. 12. the Court said that it was clear from the Directive of 1977 and the Directive of 1989 that the protection of savings constituted one of the objectives of the measures taken to co-ordinate credit institutions. Here again, taken in its context, this observation seems to me to do no more than recognise the point already made in the Parodi case [1997] E.C.R. I- 3899, I-3922, para. 21 that, as a matter of fundamental principle, restrictions on the freedom to provide services under the Treaty must be justified by imperative reasons in the public interest which are objectively necessary to guarantee the protection of the recipient of services and which do not exceed what is necessary to attain these objectives. I do not find in these observations support for the argument that a purpose of the Directive of 1977 was to promote or protect the interests of individual depositors.

    In my opinion the question whether the Directive of 1977 granted rights to individual depositors and potential depositors must be answered by examining the recitals and the articles of the Directive itself without any pre-conception as to its purpose based upon these extrinsic materials.

    The Directive of 1977 contains fifteen recitals and fifteen articles. The appellants rely in particular on the third, fourth, fifth and twelfth recitals and on articles 3, 6, 7 and 8. Article 3 is relevant to their allegation that the Bank acted contrary to the Directive when it granted a full licence to B.C.C.I. S.A. to carry on business as a deposit-taker. Articles 6 and 7 are relevant to their allegation that it failed in its duty under the Directive to supervise B.C.C.I. S.A. and B.C.C.I. Overseas. Article 8 is relevant to their allegation that it had a duty under the Directive to revoke the licence which it had granted to B.C.C.I. S.A. But it is necessary to have regard to some of the other recitals and articles in order to understand the overall effect of the Directive. The question in the case of each of the allegations against the Bank is whether, in terms of the conditions for Becker-type liability which were applied to Francovich-type liability in paragraph 22 of the Dillenkoffer case [1997] QB 259, 292:

"the result prescribed by the Directive entails the grant of rights to individuals, the content of those rights is identifiable on the basis of the provisions of the Directive and a causal link exists between the breach of the state's obligation and the loss and damage suffered by the injured parties."

    As Lord Neill pointed out, the appellants do not need to show that depositors were the only persons in whose favour obligations were imposed or on whom rights were conferred by the Directive. But in order to satisfy the Dillenkoffer conditions they must be able to demonstrate that the result to be achieved by the Directive entailed the grant of rights to depositors and potential depositors as well as to the credit institutions operating in several member states whose activities were to be authorised and supervised by the competent authorities. A triangular or tripartite relationship is implied by this argument, between the competent authorities and the credit institutions on the one hand and the competent authorities and the depositors on the other. It is not too difficult to see, as the majority in the Court of Appeal observed [2000] 2 W.L.R. 15, 84G that the Directive conferred rights on the credit institutions which were affected by it. What the appellants have to do is to show that third parties to these arrangements, depositors and potential depositors, were also granted rights by the Directive on the application to its terms of the Dillenkoffer test.

    The first two recitals record the fact that the Treaty prohibited any discriminatory treatment from the end of the transitional period and that, in order to make it easier to take up and pursue the business of credit institutions, it was necessary to eliminate the most obstructive differences between the laws of the member states as to the rules to which these institutions were subject. The third, fourth and fifth recitals are in these terms:

"Whereas, however, given the extent of these differences, the conditions required for a common market for credit institutions cannot be created by means of a single Directive; whereas it is therefore necessary to proceed by successive stages; whereas the result of this process should be to provide for overall supervision of a credit institution operating in several member states by the competent authorities in the member state where it has its head office, in consultation, as appropriate, with the competent authorities of the other member states concerned;

"Whereas measures to co-ordinate credit institutions must, both in order to protect savings and to create equal conditions of competition between these institutions, apply to all of them; whereas due regard must be had, where applicable, to the objective differences in their statutes and their proper aims as laid down by national law;

"Whereas the scope of these measures should therefore be as broad as possible, covering all institutions whose business is to receive repayable funds from the public whether in the form of deposits or in other forms such as the continuing issue of bonds and other comparable securities and to grant credits for their own account; whereas exceptions must be provided for in the case of certain credit institutions to which this Directive cannot apply."

    The appellants rely upon the reference to the overall supervision of credit institutions in the third recital, upon the phrase "in order to protect savings" in the fourth recital and the reference in the fifth recital to the receipt of repayable funds from the public as indications that it was a purpose of the Directive to protect depositors. Taken in their context however these recitals seem to me to show that the Directive had a quite different purpose. This was, as the first step in a process which would have to proceed by successive stages, to co-ordinate the conditions for the supervision of all institutions of the kind mentioned in the fifth recital operating in several member states, bearing in mind the need for the co-ordinating measures to protect savings on the one hand and for them to create equal conditions of competition on the other.

    The sixth to ninth recitals declare that the eventual aim of the harmonisation process was to introduce uniform authorisation requirements throughout the Community for comparable types of credit institution and that, while at the initial stage it was necessary to specify only certain minimum requirements to be imposed by all member states, the eventual aim could be achieved only if the particularly wide discretionary powers which certain supervisory authorities had for authorising credit establishments were progressively reduced. The tenth and eleventh recitals state that the purpose of co-ordination was to achieve a system whereby credit institutions having their head office in one of the member states were exempt from any national authorisation requirement when setting up branches in other member states, but that a measure of flexibility might nonetheless be possible in the initial stage. The twelfth recital, which explains the means by which the gradual approximation of the systems for the monitoring of solvency and liquidity of credit institutions established by the member states was to be brought about, begins with this proposition on which the appellants rely:

"Whereas equivalent financial requirements for credit institutions will be necessary to ensure similar safeguards for savers and fair conditions of competition between comparable groups of credit institutions; …"

Here again however the point being made with regard to safeguards for savers and fair conditions of competition between institutions seems to me to be directed to the two requirements which the measures of co-ordination which the Directive was to lay down would have to satisfy, bearing in mind the fact that these measures would to some degree restrict the fundamental principle of freedom of establishment. Nothing turns on the wording of the remaining three recitals.

    The first two articles of the Directive are comprised in Title I which deals with definitions and the scope of the Directive. Article 1 contains a number of definitions, including those of the expressions "credit institution" and "branch". It is worth noting that there is no definition of any expression referring to individuals in whose favour rights might be said to have been intended to be created by the Directive. If the result to be achieved was to entail the granting of rights to individuals such as savers or depositors, I would have expected a definition such as that included in article 2 of the Directive (90/314/E.E.C.) concerning package travel, package holidays and package tours which defines the expression "consumer." The meaning and effect of this Directive was considered in the Dillenkoffer case [1997] QB 259, in Verein fur Konsumenteninformation v. Österreichische Kreditversicherungs A.G. (Case C-364/96) [1998] E.C.R. I-2949 and in Rechberger and others v. Republic of Austria (Case-140/97) (unreported), 15 June 1999. In each of these cases the European Court held that article 7 of the Directive gave "consumers" a right to be reimbursed or repatriated in the event of the insolvency of the tour operator. The absence of a definition of that kind from the Directive of 1977 suggests that it was not the intention when the Directive was being drafted to grant rights under the Directive in favour of individuals or any group or class of individuals.

    In a series of cases, referred to as "the German environmental cases," claims were brought by the Commission against Germany for its failure to implement Directives which laid down various requirements to be observed by member states in relation to water and air quality: e.g. Commission of the European Communities v. Federal Republic of Germany (Case C-131-88) [1991] ECR I-825; Commission of the European Communities v. Federal Republic of Germany (Case C-298/95) [1996] E.C.R. I-6747. As Auld L.J. noted [2000] 2 W.L.R. 15, 133A, each of these Directives required member states to take specific measures to ensure that water or air was of the quality prescribed by the Directive, but they said nothing about the conferment of rights on individuals. Nevertheless the European Court held that the purpose of certain of the provision of these Directives was to create rights and obligations for individuals. In Case 131/88 [1991] ECR I-825, 867 in para. 7 the Court said:

"The Directive at issue in the present case seeks to protect the Community's groundwater in an effective manner by laying down specific and detailed provisions requiring the member states to adopt a series of prohibitions, authorisation schemes and monitoring procedures in order to prevent or limit discharges of certain substances. The purpose of those provisions of the Directive is thus to create rights and obligations for individuals."

    I agree with the appellants that these cases demonstrate that the potential width of the class of persons granted rights does not militate against the conclusion that the relevant provisions of these Directives were intended to create rights. This does not in itself mean that the persons intended to be granted rights are not sufficiently identifiable. But the cases also demonstrate that the question whether provisions in a Directive create rights and obligations for individuals depends in each case on the subject matter of the Directive, on the context and on the nature and purpose of the provisions which are in issue. The environmental cases were concerned with the protection of human health. This is a matter of concern to everybody, as we all share the environment in which we live. So the absence of a definition of the individuals who were granted rights by the Directives was of no importance. As the court said in Case C-298/95 [1996] E.C.R. I-6747, 6760:

"15. As the Commission points out, one of the purposes of the Directives at issue is to protect human health through the monitoring of the quality of waters which support, or could support, fish suitable for human consumption…

"16. In those circumstances, it is particularly important that Directives should be transposed by measures which are indisputably binding. In all cases where non-implementation of the measures required by a Directive could endanger human health, the persons concerned must be in a position to rely on mandatory rules in order to assert their rights. . . "

    Article 2 of the Directive of 1977 begins with the proposition that the Directive "shall apply to the taking up and pursuit of the business of credit institutions." It then sets out a number of exceptions and qualifications, but these are not significant for present purposes.

    The articles on which the appellants mainly rely are in Title II of the Directive, which bears the headnote "Credit institutions having their head office in a member state and their branches in other member states" and comprises articles 3 to 8. But it is necessary to have regard also to article 10 which is included among the general and transitional provisions in Title IV, as the Bank's contention is that when article 3 is read together with article 10 it is clear that it did not apply to B.C.C.I. S.A. which had already taken up business as a credit institution before the coming into force of the Banking Act 1979. Title III deals with the situation where credit institutions which have their head offices outside the Community have branches in a member state. There is nothing in article 9, which is the only article in Title III, which is relevant to this case.

    The first two paragraphs of article 3 provide:

"1. Member states shall require credit institutions subject to this Directive to obtain authorisation before commencing their activities. They shall lay down the requirements for such authorisation subject to paragraphs 2, 3 and 4 and notify them to both the Commission and the Advisory Committee.

"2. Without prejudice to other conditions of general application laid down by national laws, the competent authorities shall grant authorisation only when the following conditions are complied with:

- the credit institutions must possess separate own funds

- the credit institutions must possess adequate minimum own funds

- there shall be at least two persons who effectively direct the business of the credit institution.

"Moreover, the authorities concerned shall not grant authorisation if the persons referred to in the third indent of the first sub-paragraph are not of sufficiently good repute or lack sufficient experience to perform such duties."

The words "before commencing their activities" in paragraph 1 of article 3 indicate that its function was to deal with the authorisation of credit institutions with their head offices in one member state which were seeking to carry on business in another member state after that member state had implemented the Directive. The appellants claim in their re-amended statement of claim that B.C.C.I. S.A. was incorporated in Luxembourg in 1972, that it carried on business there as a bank under an authorisation issued by the Luxembourg authorities and that it established an office in London in the same year through which it had been carrying on business as a banker and deposit-taker for several years before it was granted a licence under the Banking Act 1979 by the Bank. On these facts it is plain that, as B.C.C.I. S.A. had already commenced its activities as a credit institution in the United Kingdom prior to the implementation of the Directive in domestic legislation by the Banking Act 1979, article 3 did not apply to it. The Bank says that no authorisation procedure was required by the Directive to enable it to continue with its activities. The fact that B.C.C.I. S.A. was granted a licence under the Banking Act 1979 which required all credit institutions to submit to the licensing procedure was a matter of domestic law only, not of Community law.

    The Bank submits that this conclusion is confirmed by the first sub-paragraph of paragraph 1 and paragraphs 3 and 4 of article 10, which provide:

"1. Credit institutions subject to this Directive which took up their business in accordance with the provisions of the member states in which they have their head offices before the entry into force of the provisions implementing this Directive shall be deemed to be authorised. They shall be subject to the provisions of this Directive concerning the carrying on of the business of credit institutions and to the requirements set out in the first and third indents of the first sub-paragraph and in the second sub-paragraph of article 3(2).

…

"3. If a credit institution deemed to be authorised under paragraph 1 has not undergone any authorisation procedure prior to commencing business, a prohibition on the carrying on of its business shall take the place of withdrawal of authorisation.

Subject to the first sub-paragraph, article 8 shall apply by analogy.

"4. By way of derogation from paragraph 1, credit institutions established in a member state without having undergone an authorisation procedure in that member state prior to commencing business may be required to obtain authorisation from the competent authorities of the member state, concerned in accordance with the provisions implementing this Directive. Such institutions may be required to comply with the requirement in the second indent of article 3(2) and with such other conditions of general application as may be laid down by the member state concerned."

The appellants claim in their reply to this submission that the head office of B.C.C.I. S.A. was in London, not Luxembourg. They point out that there were no authorisation procedures for credit institutions in the United Kingdom prior to the coming into force of the Banking Act 1979. Furthermore all credit institutions were required to undergo the licensing procedure under that Act, irrespective of whether they had previously taken up business in the United Kingdom. The appellants also say that, even if in making that requirement the United Kingdom was exercising its right of derogation under article 10(4), this would not affect the Bank's obligation to grant authorisation only on the conditions permitted by the Directive.

    On the assumption that the facts stated in the re-amended statement of claim are true, there may be some force in the appellants' argument that deemed authorisation under article 10(1) does not apply to B.C.C.I. S.A. as there were no "provisions" regulating the business of credit institutions when it commenced its activities in this country. But on balance it seems to me that this is to construe the word "provisions" too narrowly. An institution which was legitimately carrying on business here under the system of law in force at the time, as B.C.C.I. S.A. was doing because there was no prohibition to the contrary, could be said to be doing so "in accordance with the provisions" of the member state within the meaning of article 10(1). That provision does not stipulate any particular requirements which those provisions had to satisfy. In any event, this still leaves article 10(4), which clearly does apply to B.C.C.I. S.A. as it had not undergone an authorisation procedure in the United Kingdom before commencing business here. The derogation provided by this paragraph, which permits a member state to require credit institutions which are legitimately carrying on business in that state to obtain authorisation on such conditions as may be laid down by that member state, is inconsistent with the view that B.C.C.I. S.A., which was already legitimately carrying on its activities as a credit institution in the United Kingdom, required to be authorised under article 3(1).

    The appellants submit as a general principle that, where a member state voluntarily accepts obligations under Community law, it cannot escape its liabilities by saying that it need not have assumed these obligations in the first place. In support of this proposition they rely on Wagner Miret v. Fondo de Garantía Salarial (Case C-334/92) [1993] ECR I-6911, which concerned Directive 80/987 relating to the protection of employees in the event of the employer's insolvency. The Directive permitted member states to exclude certain categories of employee from the scope of the protection, and a list of the excluded categories of employee was set out in an annex to the Directive. Spain requested the exclusion of one category of employee only, and the exclusion of that category was entered in the list. It did not request the exclusion of the category of employee to which Mr. Wagner Miret belonged. The European Court held that he was entitled to the protection of the Directive, and that it was no answer to say that Spain could have excluded that category from that protection if it had chosen to exercise the option to do so. It seems to me that that case provides no support for the argument that, in the reverse situation which arises under the Directive of 1977, the United Kingdom was obliged to require an institution which was already legitimately carrying on business here to be authorised. A requirement made under article 10(4) is voluntary, not obligatory. The exercise of the option to make that requirement cannot affect the scope of the obligation under article 3. I think that it is clear, as a matter of principle, that the voluntary incorporation by a member state of a provision in a Directive into national law which it is not obliged to incorporate under Community law does not give rise to a Community law obligation. The scope of that provision is a matter for determination by the national courts as a part of the domestic law of the member state. The European Court may assist the national court in construing the Directive, but it does not follow that the obligation in question is a Community law obligation: see Leur-Bloem v. Inspecteur der Belastingdienst/Ondernemingen Amsterdam 2 (Case C-28/95) [1998] QB 182, 209, paragraphs 33-34.

    For these reasons I would hold that the Bank was not obliged by article 3(1) of the Directive to require B.C.C.I. S.A. to obtain authorisation as a condition of continuing to carry on its business in the United Kingdom. But even if it was obliged to do so as a matter of Community law, I do not find a sufficient indication, in the conditions for authorisation of credit institutions which are set out in article 3(2), that the result to be achieved by the Directive entailed the granting of rights to individuals or groups of individuals affected by their activities. The purpose of article 3(1), as indicated by the first, second and eighth recitals, was to take the first step towards the introduction of uniform authorisation requirements for comparable types of credit institution having their head office in one member state and their branches in other member states. The obligations which it imposed were designed to bring to an end, in a manner which was consistent with the nature of the business carried on by credit institutions, any discriminatory treatment as between the laws of member states with regard to establishment and the provision of these services.

    Article 6(1) provides:

"1. Pending subsequent co-ordination, the competent authorities shall, for the purposes of observation and, if necessary, in addition to such coefficients as may be applied by them, establish ratios between the various assets and/or liabilities of credit institutions with a view to monitoring their solvency and liquidity and the other measures which may serve to ensure that savings are protected.

To this end, the Advisory Committee shall decide on the content of the various factors of the observation ratios referred to in the first sub-paragraph and lay down the method to be applied in calculating them.

Where appropriate, the Advisory Committee shall be guided by technical consultations between the supervisory authorities of the categories of institutions concerned."

The appellants rely on this article as the basis for their claim that the Bank was under an obligation owed to the depositors to supervise the activities of B.C.C.I. S.A. and B.C.C.I. Overseas at all times during the relevant period. Auld L.J. said [2000] 2 W.L.R. 15, 113B-C:

"Article 6 deals with supervision. In my view, and contrary to that of Clarke J., at p. 616, it imposed on regulators immediate duties of a technical banking nature to 'ensure that savings are protected,' and it did so in advance of the process and achievement of co-ordination. It is clear from the wording of the article and the context of the Directive as a whole, concerned as it is with 'the taking up and pursuit of the business of credit institutions' (my emphasis), that the intended purpose of the supervision was to ensure, as a minimum, continuing compliance with the requirements of authorisation under article 3."

    In my opinion however article 6, although concerned with supervision, had a more limited purpose in view. As the twelfth recital indicates, it imposed a duty on the supervisory authorities, pending subsequent co-ordination, to formulate structural ratios which would make it possible for the national authorities to co-operate with each other in the setting of standards, or coefficents, to ensure the sound management of credit institutions which in due course would be co-ordinated between member states. The ultimate aim was to set equivalent financial standards which would, in terms of the recital, achieve the twin requirements noted by E.C.O.S.O.C. of ensuring "similar safeguards for savers and fair conditions of competition between comparable groups of credit institutions." It did not impose a duty of supervision. The assumption on which it proceeds is that the competent authorities in each member state would be performing that function under the national law of that member state. No minimum standards of supervision or other criteria are laid down in the article. The whole emphasis is on co-operation between the supervisory authorities, with a view to harmonisation in due course of the means by which the performance of credit institutions carrying on business in more than one member state could be monitored. The protection of savings was assumed to be the purpose of the monitoring system. But it was not necessary in order to establish observation ratios and their co-ordination between member states to impose a Community law duty of supervision or to grant rights in that regard to individuals or groups of individuals.

    Article 7 is also concerned with supervision. Here again Auld L.J. was of the view that it imposed a duty to supervise: p. 113G. But I think that the duty which it imposed was one of co-operation between the supervisory authorities. Paragraph 1 of the article provides:

"1. The competent authorities of the member states concerned shall collaborate closely in order to supervise the activities of credit institutions operating, in particular by having established branches there, in one or more member states other than that in which their head offices are situated. They shall supply one another with all information concerning the management and ownership of such credit institutions that is likely to facilitate their supervision and the examination of the conditions for their authorisation and all information likely to facilitate the monitoring of their liquidity and solvency."

As in the case of article 6, article 7 assumed the existence in each member state of a competent authority or competent authorities whose function it was to supervise the activities of credit institutions in that member state.

    Prior to the implementation of the Directive of 1977 there were no supervisory authorities in either the United Kingdom or Denmark. So it was necessary, to give effect to the Directive, for authorities to be set up with the function of supervising credit institutions operating in those member states. An obligation to do so is not expressed in article 7. It is to be found in the third paragraph of article 189 of the E.E.C. Treaty (now article 249 E.C.), which provides that a Directive is to be binding as to the result to be achieved but leaves the choice of form and methods to each member state. Here again however it is necessary to distinguish between the duty under Community law for an authority to be set up under the national law of each member state whose function it was to supervise and the duty under article 7 of the supervisory authorities of each member state to co-operate. I do not think that article 7, which refers to the sharing of information "likely to facilitate" supervision, examination and monitoring, imposed a duty under Community law to supervise. The absence from the article of any prescribed system of supervision, and of any criteria or standards to be applied by the supervisory authority, is an indication to the contrary. It is noteworthy that, although the appellants claim that there was a general duty to supervise, they do not point to the breach by the Bank of any particular duties of supervision imposed by either article 6 or article 7.

    The appellants rely upon the observations of Advocate General Sir Gordon Slynn in Municipality of Hillegom v. Hillenius (Case 110/84) [1985] E.C.R. 3947, 3948, where he said that article 7 of the Directive of 1977 "provides that the supervisory authorities of the various member states shall collaborate closely in order to supervise credit institutions operating in more than one member state." I do not think that this comment, which simply repeats the wording of paragraph 1 of the article, is in any way inconsistent with the view which I have formed, that the duty under article 7 is a duty to collaborate in order to assist the competent authorities in the performance of their supervisory functions under national law. The observations which the European Court made in its judgment, at p. 3963, paras. 26 and 27, to the effect that the Directive was designed to facilitate the overall monitoring of credit institutions operating in more than one member state by the competent authorities of the member state in which the credit institution has its head office and about the need for the monitoring of banks through supervision within a member state and the exchange of information by the competent authorities to function properly, do not go to the length of suggesting that the court saw the Directive as entailing the grant of rights to individual depositors.

    The appellants also rely on Carbonari v. Università degli studi di Bologna (Case C-131/97) (unreported), 25 February 1999. That was a case which was concerned with the direct effect of provisions in two Directives (82/76/E.E.C. amending 75/362/E.E.C. and 75/363/E.E.C.), the first of which related to training periods and remuneration of doctors. Detailed provisions were included about the training requirements, and there was a provision in the Annex to the Directive of 1975 that the posts which specialists were to hold while in full-time training were to be "subject to appropriate remuneration." The European Court in para. 44-47 of its judgment applied the first limb of the Becker test. The right of the medical students to appropriate remuneration during their training period was in itself unconditional and sufficiently precise. But the Directives were not unconditional as to which institution was to bear the obligation to pay the remuneration. They did not include any Community definition of the remuneration which was to be regarded as appropriate or the methods by which it was to be fixed to enable the national court to determine the body liable to pay it or the level at which it was to be paid. Such definitions were to be a matter for the member states when they were implementing the Directive. In other words, as the Directive did not itself define as a matter of Community law what was appropriate remuneration or lay down a Community law method to fix that amount, the medical students had no Community right which they could enforce to obtain payment.

    The appellants say that the Carbonari case supports their claim that articles 6 and 7 imposed a duty of supervision notwithstanding that the precise methods or forms of supervision are not specified in the Directive. In my opinion however the case tends to support the Bank's argument. None of the provisions in articles 6 and 7 of the Directive of 1977 define any categories of individual on whom rights were being conferred, nor do they state in obligatory terms that the credit institutions "shall be subject to appropriate supervision" by the competent authorities. Even if such an obligation in general terms could be said to be implied, the absence of even the slightest amount of detail as to the system of supervision required by Community law which was to be adopted and enforced by the national courts would make it impossible to say that, as matter of Community law, the obligation to supervise was unconditional and sufficiently precise to satisfy the Becker-type liability test.

    The appellants also rely on article 8 which deals with the withdrawal of authorisation. Paragraphs 1 and 2 of this article provide:

"1. The competent authorities may withdraw the authorisation issued to a credit institution subject to this Directive or to a branch authorised under article 4 only where such an institution or branch:

does not make use of the authorisation within 12 months, expressly renounces the authorisation or has ceased to engage in business for more than six months, if the member state concerned has made no provision for the authorisation to lapse in such cases;

has obtained the authorisation through false statements or any other irregular means;

no longer fulfils the conditions under which authorisation was granted, with the exception of those in respect of own funds;

no longer possesses sufficient own funds or can no longer be relied upon to fulfil its obligations towards its creditors, and in particular no longer provides security for the assets entrusted to it;

falls within one of the other cases where national law provides for withdrawal of authorisation.

"2. In addition, the authorisation issued to a branch under article 4 shall be withdrawn if the competent authority of the country in which the credit institution which established the branch has its head office has withdrawn authorisation from that institution."

In my opinion the key word in paragraph 1 is the word "only" which precedes the list of the various situations in which authorisation may be withdrawn. It seems to me that this is a limiting provision, as indicated by the second recital, with a view to eliminating differences between the laws of member states. The reference in sub-paragraph (e) to cases where "national law provides for withdrawal of authorisation" ensures that the matter is not left to the administrative discretion of the competent authority in that member state.

    Auld L.J. said [2000] 2 W.L.R. 15, 114G that he recognised the fact that, while paragraph 2 was obligatory, paragraph 1 was in general terms permissive in the various circumstances specified. But his view was that the article should be read as imposing a duty on competent authorities to withdraw authorisation in the circumstances referred to in sub-paragraphs (c) and (d). He added this explanation at p. 114H:

"It is inconceivable that the Directive should be read so as to require banking regulators to insist on certain minimum requirements of authorisation and to supervise to ensure continued satisfaction of them, yet leave them with a discretion, unspecified as to criteria, to permit continuance of trading without check or condition when those requirements are no longer met."

On my reading of the Directive as a whole however, it is to be regarded as laying down a series of provisions with a view to the harmonisation of the criteria to be applied to credit institutions having their head office in a member state and their branches in other member states. That being so, there is nothing surprising in an approach to the withdrawal of authorisation in the circumstances referred to in paragraph 1 which permits withdrawal in these, and only these, circumstances but provides that withdrawal in the situation mentioned in paragraph 2 is to be obligatory. I do not think that there is anything here which entails the grant of rights to individuals to insist upon the withdrawal of authorisation in the circumstances mentioned in paragraph 1. For their part, the credit institutions could hardly object to a decision to withdraw authorisation in the circumstances mentioned in sub-paragraphs (c) and (d) of paragraph 1 which, in order to protect savings, it was obviously necessary to include in the list of circumstances in which the withdrawal of authorisation was to be permissible.

    Looking back at the Directive as a whole, the key to a proper understanding of its purpose and effect seems to me to lie in the fact that it was the first step in a process of harmonisation of provisions for the regulation of credit institutions carrying on business within the Community. It was about the removal of barriers to the right of establishment under article 52 of the E.E.C. Treaty (now article 43 E.C.). It confined itself to imposing a number of minimum conditions and prohibitions on member states as to the authorisation and supervision of credit institutions having their head offices in another member state or having their head offices outside the Community. It was based upon an appreciation of the fact that credit institutions require regulation in order to protect savings. So any measures of harmonisation had to meet the twin requirements of protecting savings on the one hand and creating conditions of equal competition between credit institutions operating in more than one member state on the other. It placed duties of co-operation on the competent authorities where a credit institution was operating in one or more member state other than that in which its head office was situated. But it stopped short of prescribing any duties of supervision to be performed by the competent authority within each member state. It is not possible to discover provisions which entail the granting of rights to individuals, as the granting of rights to individuals was not necessary to achieve the results which were intended to be achieved by the Directive.

    For these reasons I am unable, with great respect, to agree with Auld L.J.'s conclusions [2000] 2 W.L.R. 15, 102H-103A that the Directive of 1977 imposed clearly defined obligations on member states and on their regulatory bodies and that in doing so it gave rise to corresponding Community law rights in depositors to enforce those obligations by an action of damages. I prefer the views of Clarke J. [1996] 3 All E. R. 558, 616D-E where he said:

"The true position, as it seems to me, is that the Directive was not intended to require the imposition of a duty to supervise upon the supervisory authority because, whatever the underlying purpose of the system of supervision, the immediate purpose of the Directive, rather like that in R. v. International Stock Exchange of the U.K. and the Republic of Ireland Ltd., ex parte Else (1982) Ltd., R. v. International Stock Exchange of the U.K. and the Republic of Ireland, ex parte Roberts [1993] Q.B. 534, was a first step towards harmonisation of the systems in the member states, which was assumed to and no doubt did exist. Its purpose was not to lay down the duty to supervise or radically to alter the existing systems, but, even if was, it was not (as I see it) to confer rights upon either savers or other creditors."

    The majority in the Court of Appeal did not examine the provisions of the Directive in detail. But at p. 84G-H they said this:

"In our judgment it is reasonably clear that the Directive of 1977 does confer enforceable rights on credit institutions which are to be supervised under its provisions. In particular, a credit institution which was refused authorisation for reasons incompatible with article 3, or which had its authorisation withdrawn which for reasons incompatible with article 8, would (in the event of non-transposition of the Directive of 1977) most probably have had a right to challenge the Bank's decision on the basis that the Becker conditions [1982] ECR 53 were satisfied. But there is not in our judgment any sufficient clarity or certainty in the position of depositors (or other customers of credit institutions)."

Subject to the qualifications that I have substituted the Francovich test as applied in Dillenkoffer for the Becker test, on the view that it is more precise and covers both types of liability, and that I do not accept that credit institutions were to be "supervised under" the provisions of the Directive, I agree with the conclusions of the majority on this issue. I would therefore dismiss the appeal on the Community law claim.

    In the courts below neither party asked for a preliminary reference under article 234 E.C. (ex article 177 E.E.C.) on the issues relating to the appellants' claim under Community law. Clarke J. said, at p. 625E-F, that he would have referred the matter as to the meaning and effect of the Directive of 1977 had the parties not requested him to give a judgment on the Community law issue without doing so. In the Court of Appeal the majority said, at p. 85D, that they did not regard the question of Becker-type liability as acte clair, but that as a matter of discretion the court had decided not to make a reference. The parties' unanimity in asking the court not to make a reference had been a significant factor in that decision. That unanimity was departed from in your Lordships' House. The appellants, having regard to the observations in the courts below, to the fact that there was a dissenting judgment in the Court of Appeal and to what they saw as a change of position by the Bank as regards the purpose and intent of the Directive of 1977, have asked for a reference under article 234 E.C. (ex article 177 E.E.C.) on the question whether the Directive conferred rights, and if so what rights, on depositors which they can exercise against the competent authority designated by the member state for the purpose of carrying out that member state's obligations under the Directive.

    I am of the opinion that it would not be appropriate for a reference to be made to the European Court on the critical question in this case, which is whether the Directive of 1977 entailed the granting of rights to individual depositors and potential depositors. I consider that this matter, on which I understand your Lordships to be unanimous and on which we have had the benefit of very full and helpful submissions both orally and in writing from both sides, is acte clair. So I would decline the appellants' request that we should make a reference in this case.

    I would make the same order as that proposed by my noble and learned friend Lord Steyn.

LORD HUTTON

My Lords,

    The action for misfeasance in public office has long been recognised by English law. In some of the cases in the 18th and 19th centuries there are statements that the plaintiff must establish that the defendant was actuated by malice towards him and intended to injure him in the way in which he discharged his public duty; in modern cases this is referred to as "targeted malice." In Harman v. Tappenden (1801) 1 East 555, 102 E.R. 214 Lawrence J. said, at pp. 562-563, 217:

"There is no instance of an action of this sort maintained for an act arising merely from error of judgment. Perhaps the action might have been maintained, if it had been proved that the defendants contriving and intending to injure and prejudice the plaintiff, and to deprive him of the benefit of his profits from the fishery, which as a member of this body he was entitled to according to the custom, had wilfully and maliciously procured him to be disfranchised, in consequence of which he was deprived of such profits. But here there was no evidence of any wilful and malicious intention to deprive the plaintiff of his profits, or that they had disfranchised him with that intent, which is necessary to maintain the action."

But other cases suggest that an unlawful act done with an improper motive is sufficient to constitute the tort.

In Tozer v. Child (1857) 7 E.+B. 377, 119 E.R. 1286 Lord Campbell C.J. directed the jury that:

"the defendants were not necessarily liable in this action, although the plaintiff, notwithstanding his non-payment of the said church rate, was qualified and entitled to vote, and to be a candidate at the said election, as he alleged: and that it was incumbent on the plaintiff to make out that the acts of the defendants complained of were malicious; and that malice might be proved, not only by evidence of personal hostility or spite, but by evidence of any other corrupt or improper motive . . ."(see p. 379).

    In Bourgoin S.A. v. Ministry of Agriculture, Fisheries and Food [1986] Q.B. 716 the issue was debated whether the defendant must intend to injure the plaintiff and it was held that damages could be recovered for misfeasance in public office where the defendant acted deliberately, not with the intent to harm the plaintiff, but with knowledge that he had no power to act as he did and that his action would injure the plaintiff. In that case the Minister of Agriculture, knowing that he had no power to do so, prohibited the importation of French turkeys into the United Kingdom knowing that the prohibition must necessarily injure French turkey producers but acting with the motive, not of injuring them, but of benefiting English producers. French producers claimed damages for misfeasance in public office.

    Mann J. rejected the argument that an intent to injure the plaintiff was an essential ingredient of the tort and stated at page 740D:

"I do not read any of the decisions to which I have been referred as precluding the commission of the tort of misfeasance in public office where the officer actually knew that he had no power to do that which he did, and that his act would injure the plaintiff as subsequently it does. I read the judgment in Dunlop v. Woollahra Municipal Council [1982] AC 158 in the sense that malice and knowledge are alternatives. There is no sensible reason why the common law should not afford a remedy to the injured party in circumstances such as are before me. There is no sensible distinction between the case where an officer performs an act which he has no power to perform with the object of injuring A (which the defendant accepts is actionable at the instance of A) and the case where an officer performs an act which he knows he has no power to perform with the object of conferring a benefit on B but which has the foreseeable and actual consequence of injury to A (which the defendant denies is actionable at the instance of A). In my judgment each case is actionable at the instance of A . . . "

His judgment on this point was upheld by the Court of Appeal and Oliver L.J. stated at p. 777G:

"For my part, I too can see no sensible distinction between the two cases which the judge mentions.

If it be shown that the minister's motive was to further the interests of English turkey producers by keeping out the produce of French turkey producers— an act which must necessarily injure them— it seems to me entirely immaterial that the one purpose was dominant and the second merely a subsidiary purpose for giving effect to the dominant purpose. If an act is done deliberately and with knowledge of its consequences, I do not think that the actor can sensibly say that he did not 'intend' the consequences or that the act was not 'aimed' at the person who, it is known, will suffer them."

    The principal issue which arises for determination on the present appeal is whether in order to succeed the plaintiffs must prove that the public officers of the Bank of England knew that their unlawful acts or omissions would probably injure them or persons in a class of which they were members or that the officers were subjectively reckless as to such likely injury.

    In a learned judgment [1996] 3 All E.R. 558, 632, after a full and detailed consideration of the authorities, Clarke J. summarised his conclusions as to the ingredients of the tort as follows:

"1. Misfeasance in public office

(1) The tort of misfeasance in public office is concerned with a deliberate and dishonest wrongful abuse of the powers given to a public officer. It is not to be equated with torts based on an intention to injure, although, as suggested by the majority in Northern Territory v. Mengel (1995) 69 A.L.J.R. 527, it has some similarities to them.

(2) Malice, in the sense of an intention to injure the plaintiff or a person in a class of which the plaintiff is a member, and knowledge by the officer both that he has no power to do the act complained of and that the act will probably injure the plaintiff or a person in a class of which the plaintiff is a member are alternative, not cumulative, ingredients of the tort. To act with such knowledge is to act in a sufficient sense maliciously: see Mengel 69 A.L.J.R. 527, 554 per Deane J.

(3) For the purposes of the requirement that the officer knows that he has no power to do the act complained of, it is sufficient that the officer has actual knowledge that the act was unlawful or, in circumstances in which he believes or suspects that the act is beyond his powers, that he does not ascertain whether or not that is so or fails to take such steps as would be taken by an honest and reasonable man to ascertain the true position.

(4) For the purposes of the requirement that the officer knows that his act will probably injure the plaintiff or a person in a class of which the plaintiff is a member it is sufficient if the officer has actual knowledge that his act will probably damage the plaintiff or such a person or, in circumstance in which he believes or suspects that his act will probably damage the plaintiff or such a person, if he does not ascertain whether that is so or not or if he fails to make such inquiries as an honest and reasonable man would make as to the probability of such damage.

(5) If the states of mind in (3) and (4) do not amount to actual knowledge, they amount to recklessness which is sufficient to support liability under the second limb of the tort.

(6) Where a plaintiff establishes (i) that the defendant intended to injure the plaintiff or a person in a class of which the plaintiff is a member (limb one) or that the defendant knew that he had no power to do what he did and that the plaintiff or a person in a class of which the plaintiff is a member would probably suffer loss or damage (limb two) and (ii) that the plaintiff has suffered loss as a result, the plaintiff has a sufficient right or interest to maintain an action for misfeasance in public office at common law. The plaintiff must of course also show that the defendant was a public officer or entity and that his loss was caused by the wrongful act."

    In the judgments of the Court of Appeal on appeal by the plaintiffs [2000] 2 W.L.R. 15 there was again a learned and comprehensive review of the authorities. The majority of the court, Hirst and Robert Walker L.JJ., expressed broad agreement with the conclusions of Clarke J. and dismissed the appeal. In his dissenting judgment Auld L.J. held that as the core of misfeasance in public office is abuse of power it is sufficient for a plaintiff to prove that the abuse of power was an effective cause of his loss and that there is no requirement for the plaintiff to prove foresight of probable harm on the part of the public officer. Auld L.J. stated, at p. 167F:

"It follows that, in my view, the test of abuse of power—dishonesty—by a public officer does not necessarily include as an essential ingredient some appreciation by the officer of an injurious consequence. A public officer who dishonestly disregards his plain duty or who does not honestly attempt to do it, acts at his peril, and if injury results he is liable for it.

But, even if I am wrong about that, I can see no basis, whether as a matter of interpretation of Mann J.'s judgment in the Bourgoin case [1986] Q.B. 716 or otherwise, for requiring a plaintiff in a claim for misfeasance in public office to prove foresight in at least the form of suspicion of probable harm. That would be a much more onerous burden than that imposed on a plaintiff in a negligence claim, where the defendant is not necessarily a public officer and the claimant has not had to go to the lengths of proving dishonesty. Interestingly, it could also be more onerous than that in a tort of intentional injury such as fraudulent misrepresentation."

    Before this House the principal submission of Lord Neill Q.C., on behalf of the appellants, was that Bourgoin established that there was a second limb of the tort, separate and distinct from the limb constituted by abuse of power with the intent to injure, and that the second limb was constituted by a public officer abusing his power by knowingly acting unlawfully. When such an abuse of power is established and where it is also proved that the abuse has caused loss to the plaintiff the law does not require, and there is no reason why it should require, that for liability to arise there should be foresight by the public officer that loss will probably occur to the plaintiff or to a class of persons to whom the plaintiff belongs, or that there should be subjective recklessness as to such likely loss.

    My lords, I am unable to accept this argument for two principal reasons. The first reason is that I consider that the second limb of the tort cannot be viewed in isolation from the first limb and that the concept of targeted malice which is the underlying principle of the first limb exercises a restrictive effect on the ambit of the second limb. This is implicit in the passage of the judgment of Oliver L.J. in Bourgoin [1986] Q.B. 777H which I have quoted above because the learned Lord Justice said:

"If an act is done deliberately and with knowledge of its consequences, I do not think that the actor can sensibly say that he did not 'intend' the consequences or that the act was not 'aimed' at the person who, it is known, will suffer them."

Therefore it was his opinion that if a person acts deliberately, knowing that his action will injure another person, he must be taken to intend the consequences and is equated with the person who acts with the intent to cause injury. This is a view which is inconsistent with a liability which arises where there is an abuse of power without knowledge that it will probably injure the plaintiff but where the abuse is an effective cause of such injury.

    The judgments of the High Court of Australia in Northern Territory of Australia v. Mengel (1995) 69 A.L.J.R. 527 and the judgment of the Court of Appeal of New Zealand in Garrett v. Attorney-General [1997] 2 N.Z.L.R. 332 are the second factor which leads me to reject the wider ambit of the second limb of the tort contended for by the plaintiffs. In those two cases there was a full discussion of the issue now before this House (save that in Mengel the distinction between foresight by the public officer and objective foreseeability was not directly considered) and in both cases it was held that it was insufficient for the plaintiff to show a knowing breach of duty by a public officer coupled with resultant injury.

    In Mengel stock inspectors employed by the defendant, without statutory or other authority, wrongly quarantined the plaintiffs' cattle whereby the plaintiffs suffered loss. Before the High Court of Australia the plaintiffs contended that they were entitled to succeed on a claim for misfeasance in public office, and they argued that the mental element of that tort is made out if the public officer either knows or ought to know that he is acting without authority and the unlawful exercise of the power results in damage. This argument was rejected by the High Court. In their joint judgment Mason C.J., Dawson, Toohey, Gaudron and McHugh JJ. stated, at p. 540:

"The cases do not establish that misfeasance in public office is constituted simply by an act of a public officer which he or she knows is beyond power and which results in damage. Nor is that required by policy or by principle. Policy and principle both suggest that liability should be more closely confined. So far as policy is concerned, it is to be borne in mind that, although the tort is the tort of a public officer, he or she is liable personally and, unless there is de facto authority, there will ordinarily only be personal liability. And principle suggests that misfeasance in public office is a counterpart to, and should be confined in the same way as, those torts which impose liability on private individuals for the intentional infliction of harm. For present purposes, we include in that concept acts which are calculated in the ordinary course to cause harm, as in Wilkinson v. Downton, [[1897] 2 QB 57] or which are done with reckless indifference to the harm that is likely to ensue, as is the case where a person, having recklessly ignored the means of ascertaining the existence of a contract, acts in a way that procures its breach.

It may be that analogy with the torts which impose liability on private individuals for the intentional infliction of harm would dictate the conclusion that, provided there is damage, liability for misfeasance in public office should rest on intentional infliction of harm, in the sense that that is the actuating motive, or on an act which the public officer knows is beyond power and which is calculated in the ordinary course to cause harm. However, it is sufficient for present purposes to proceed on the basis accepted as sufficient in Bourgoin, namely, that liability requires an act which the public officer knows is beyond power and which involves a foreseeable risk of harm."

Brennan J. stated, at p. 546:

"I respectfully agree that the mental element is satisfied either by malice (in the sense stated) or by knowledge. That is to say, the mental element is satisfied when the public officer engages in the impugned conduct with the intention of inflicting injury or with knowledge that there is no power to engage in that conduct and that that conduct is calculated to produce injury. These are states of mind which are inconsistent with an honest attempt by a public officer to perform the functions of the office. Another state of mind which is inconsistent with an honest attempt to perform the functions of a public office is reckless indifference as to the availability of power to support the impugned conduct and as to the injury which the impugned conduct is calculated to produce. The state of mind relates to the character of the conduct in which the public officer is engaged— whether it is within power and whether it is calculated (that is, naturally adapted in the circumstances) to produce injury."

Deane J. stated, at p. 554:

"In the context of misfeasance in public office, the focus of the requisite element of malice is injury to the plaintiff or injury to some other person through an act which injuriously affects the plaintiff. Such malice will exist if the act was done with an actual intention to cause such injury. The requirement of malice will also be satisfied if the act was done with knowledge of invalidity or lack of power and with knowledge that it would cause or be likely to cause such injury. Finally, malice will exist if the act is done with reckless indifference or deliberate blindness to that invalidity or lack of power and that likely injury. Absent such an intention, such knowledge and such reckless indifference or deliberate blindness, the requirement of malice will not be satisfied."

The judgment of Deane J. is important because, as in the judgment of Oliver L.J. in Bourgoin, it emphasises that the second limb of the tort is a species of malice, and that the requirement for malice is satisfied where the public officer knows that the abuse of power will cause injury, or is recklessly indifferent or deliberately blind to the likely injury.

    Lord Neill relied on a passage in the judgment of Brennan J., at p. 547, where he said:

"It is the absence of an honest attempt to perform the functions of the office that constitutes the abuse of the office. Misfeasance in public office consists of a purported exercise of some power or authority by a public officer otherwise than in an honest attempt to perform the functions of his or her office whereby loss is caused to a plaintiff. Malice, knowledge and reckless indifference are states of mind that stamp on a purported but invalid exercise of power the character of abuse of or misfeasance in public office. If the impugned conduct then causes injury, the cause of action is complete."

    But in my opinion this passage cannot be read in isolation and must be read together with the earlier passage at p. 546 which I have quoted.

    In Garrett v. Attorney-General [1997] 2 N.Z.L.R. 332 the plaintiff claimed that there had been an abuse of power by a police sergeant who had failed to investigate properly her complaint that she had been raped by a police constable in a police station. The Court of Appeal held that to succeed in a claim for misfeasance in public office the plaintiff had to prove that the public officer knew that his disregard of his duty would injure the plaintiff or that the officer was recklessly indifferent to the consequences for the plaintiff.

    In its judgment the Court of Appeal set out Clarke J.'s summary of his conclusions which I have set out earlier in this judgment and stated their agreement with them, at pp. 349-350:

"We are in respectful agreement with Clarke J. that it is insufficient to show foreseeability of damage caused by a knowing breach of duty by a public officer. The plaintiff, in our view, must prove that the official had an actual appreciation of the consequences for the plaintiff, or people in the general position of the plaintiff, of the disregard of duty or that the official was recklessly indifferent to the consequences and can thus be taken to have been content for them to happen as they would. The tort has at its base conscious disregard for the interests of those who will be affected by official decision making. There must be an actual or, in the case of recklessness, presumed intent to transgress the limits of power even though it will follow that a person or persons will be likely to be harmed. The tort is not restricted to a case of deliberately wanting to cause harm to anyone; it also covers a situation in which the official's act or failure to act is not directed at the injured party but the official sees the consequences as naturally flowing for that person when exercising power. In effect this is no more than saying the tort is an intentional tort. In this context, a person intends to bring about the known consequences of his or her actions or omissions, even if other consequences form the primary motive. Bourgoin is an example. The concept of attributing intention by necessary inference in this way is well established."

The court also stated the reasons why it considered that the tort should not be extended to cover a wrongful act done without a realisation of the consequences for the plaintiff and stated, at pp. 350-351:

"The purpose behind the imposition of this form of tortious liability is to prevent the deliberate injuring of members of the public by deliberate disregard of official duty. It is unnecessary, to attain this objective, to extend the tort to catch an act which, though known to be wrongful, is done without a realisation of the consequences for the plaintiff. The law may still provide a remedy in negligence if the situation is one of those in which it is appropriate to impose a duty of care or, if the plaintiff is someone intended by statute to have the particular benefit or protection of an Act of Parliament or subordinate legislation, the plaintiff may have a remedy in the form of an action for breach of statutory duty; or the circumstances may give rise to another of the traditional tort actions, for instance, for false imprisonment or assault . . . .

In our view this intentional tort should not be allowed to overflow its banks and cover the unintentional infliction of damage. In many cases the consequences of breaking the law will be obvious enough to officials, who can then be taken to have intended the damage they caused. But where at the time they do not realise the consequences they will probably not be deterred from exceeding their powers by any enlargement of the tort. As Clarke J. observes, they may well think that they are acting in the best interests of those persons whom they actually have in mind. In any modern society administration of central or local government is complex. Overly punitive civil laws may oftentimes deter a commonsense approach by officials to the use or enforcement of rules and regulations. We prefer to err on the side of caution and not to extend the potential liability of officials for causing unforeseen damage. To do so may have a stultifying effect on governance without commensurate benefit to the public. . .

The common law has long set its face against any general principle that invalid administrative action by itself gives rise to a cause of action in damages by those who have suffered loss as a consequence of that action. There must be something more. And in the case of misfeasance of public office that something more, it seems to us, must be related to the individual who is bringing the action. While the cases have made it clear that the malice need not be targeted there must, as we have said, be a conscious disregard for the interests of those who will be affected by the making of the particular decision."

The opinion of the New Zealand Court of Appeal, in agreement with Clarke J., that it is insufficient for the plaintiff to show objective foreseeability that the breach of duty will probably cause damage and that it must be proved that the public officer himself foresaw the probability of damage, or was reckless as to the harm which is likely to ensue, is the same as the view taken by Deane J. in Mengel. I also consider that the opinion of Brennan J. is not inconsistent with it as, at p 546 he referred to "conduct … calculated to produce injury." In Bourgoin [1986] Q.B. 716 the Minister knew that the prohibition would harm French turkey producers and therefore the issue as between foresight and foreseeability did not have to be determined by Mann J., but as Clarke J. pointed out in his judgment, at p. 568, Mann J. had previously referred more than once to the public officer knowing that his act would injure the plaintiff, and I agree with Clarke J. that Mann J. must have meant "foreseen" at page 740F in the passage of his judgment which I have quoted above.

    I further consider that the judgment in Garrett provides an answer to the submission that if there is a deliberate or reckless abuse of power which causes harm the injured party is entitled to recover damages whether or not the officer foresaw the harm. This view of the tort separates the abuse of power from the resultant harm and regards the unlawful exercise of the power, viewed in isolation from its consequences, as the essence of the tort, whereas the New Zealand Court of Appeal, rightly in my opinion, does not detach the unlawful conduct from its consequences but regards the abuse of power as consisting in the unlawful exercise of a power by a public officer with knowledge that it is likely to harm another citizen, when the power is given to be exercised for the benefit of other citizens. As the court stated, at p. 349:

"The tort has at its base conscious disregard for the interests of those who will be affected by official decision making."

    In the present case Clarke J. and the Court of Appeal were of opinion that to constitute the tort of misfeasance there must be a dishonest abuse of power by the public officer. In the course of his submissions that the tort is established under the second limb in Bourgoin when a public officer knowingly acts unlawfully in the purported exercise of his power and the unlawful exercise of the power causes loss to the plaintiff, Lord Neill accepted that an improper motive is an essential ingredient and said that improper motive was alleged by the plaintiffs. On behalf of the Bank Mr. Stadlen Q.C. submitted that it was a requirement of the tort in every case that the public officer should have acted dishonestly. In considering this point it is necessary to recognise, as Mr. Stadlen did in his submissions, that in most cases, as a matter of reality, a finding by the tribunal of fact that a public officer knew that he was acting unlawfully and that his actions would probably injure the plaintiff would lead to a finding that he was acting dishonestly. Mr. Stadlen submitted that in most cases the element of dishonesty would permeate a finding of knowledge of unlawfulness and probability of harm, but he contended that dishonesty was an element which was always necessary and that there might be exceptional cases where, notwithstanding knowledge of unlawfulness and probability of harm, the plaintiff would fail in his action because he failed to prove dishonesty.

    My Lords, I consider that dishonesty is a necessary ingredient of the tort, and it is clear from the authorities that in this context dishonesty means acting in bad faith. In some cases the term "dishonesty" is not used and the term "in bad faith" or acting from "a corrupt motive" or "an improper motive" is used, or the term "in bad faith" is used together with the term "dishonesty". In Cullen v. Morris (1819) 2 Stark 577, 589, 171 E.R. 741, 745 Lord Abbott C.J. in directing the jury said:

"The question for your consideration is, whether the refusal of the vote in this instance, was founded on an improper motive on the part of the defendant, it is for you to pronounce your opinion, whether the defendant's conduct proceeded from an improper motive, or from an honest intention to discharge his duty acting under professional advice."

The judgment of Buxton L.J. in Barnard v. Restormel Borough Council [1998] 3 P.L.R. 27 is a modern example of how the terms "dishonesty" and "bad faith" are used interchangeably. In that case where misfeasance in public office was alleged the learned Lord Justice in delivering his judgment said, at p. 33: "We bear in mind the need to establish dishonesty" and later, at p. 37 he referred to the strong emphasis placed "in the tort of misfeasance on the requirement of subjective bad faith." However, as the term "dishonesty" in some contexts implies a financial motive, I consider that the term "in bad faith" is a preferable term to use and, as I have stated, I consider that it is an essential ingredient in the tort.

    I agree with the opinion of Clarke J., at p. 583A of his judgement that the tort can be constituted by an omission by a public officer as well as by acts on his part. As Brennen J. stated in Mengel at p. 545: "Any act or omission done or made by a public official in purported performance of the functions of the office can found an action for misfeasance in public office." But whether the public officer is sued in respect of an act or an omission, it must be a deliberate one involving an actual decision and liability will not arise from injury suffered by mere inadvertence or oversight. I also agree with the opinion of Clarke J., at p. 583B and D that it is sufficient for the plaintiff to prove that the public officer foresaw that his action would probably injure the plaintiff; to require foresight of certainty of harm would be unrealistic and, being very difficult to prove, would give inadequate protection against abuse of power.

    I further consider that if the public officer knows that his unlawful conduct will probably injure another person, or is reckless as to that consequence, the plaintiff does not need to show, before liability can arise, some other link or relationship between him and the officer. The requirement of foresight of probable harm to the plaintiff, or recklessness as to such harm, is sufficient to ensure that the tort is confined within reasonable bounds.

    I have had the advantage of reading in draft the speeches which have been prepared by my noble and learned friends Lord Hope of Craighead and Lord Millett, and for the reasons which they give and with which I am in agreement, I would dismiss the appeal on the Community law claim and I would make the same order as that proposed by my noble and learned friend Lord Steyn.

LORD HOBHOUSE OF WOODBOROUGH

My Lords,

This appeal has raised two questions of law. The first is that relating to of the alleged breach of the First Council Banking Co-ordination Directive (77/780/E.E.C.) of 1977. I agree that the claim under this head cannot succeed in law for the reasons given by my noble and learned friend Lord Hope of Craighead and that the appeal under this head should be dismissed.

    The second is that relating to the definition of the tort of misfeasance in public office. On this, subject to what I will say in this speech, I am in substantial agreement with the views expressed by my noble and learned friends Lord Steyn and Lord Hutton. None of your Lordships are willing to accept the main submissions of the appellants on this aspect but it will be necessary, as your Lordships propose and I agree, to hear further submissions upon the actual pleaded allegations of the appellants before deciding what order to make. On the hearing of the appeal your Lordships have heard full argument upon what are the constituents of the tort. Their correct identification is of an importance which extends beyond the present case. Your Lordships have been greatly assisted by the judgments in this case of Clarke J. and the Court of Appeal which contain a careful examination of the authorities as have the written and oral submissions of counsel to which tribute should be paid. I am also indebted to the re-examination of the authorities in the speeches of my noble and learned friends Lord Steyn and Lord Hutton.

    The tort, concerning as it does the acts of those vested with governmental authority and the exercise of executive powers, has developed over the centuries as circumstances have changed. Terminology still tends to be used which is of little assistance to anyone not familiar with the legal history. The use of the word malice also causes confusion both as to its meaning in relation to this tort and the role it has in the analysis of the tort. The particular elements emphasised as being of the essence of the tort have varied from time to time. There has been little consistency of language. It is therefore right to take the opportunity to attempt to draw together the threads and assist a more definitive view to be taken. It is not necessary for this purpose to repeat the review of the authorities.

    I will start by putting the tort in its legal context. Typically, a tort involves the invasion by the defendant of some legally protected right of the plaintiff, for example, trespass to property or trespass to the person. Conversion is another example. Such conduct on the part of the defendant is actionable as such and the belief of the defendant as to the legality of what he did is irrelevant. It is no defence for the defendant to say that he believed that he had statutory or other legal authority if he did not. The legal justification must actually exist otherwise he is liable in tort. (Northern Territory v. Mengel 69 A.J.L.R. 527, 547)

    On the other hand, where the plaintiff is not entitled to complain of the invasion of such a right but bases his claim on some loss which he has suffered consequentially upon some act of the defendant which the defendant mistakenly believed was authorised by the law, the defendant's honest belief provides him with an answer to the plaintiff's claim notwithstanding any actual illegality. Thus the holder of a public office who acts honestly will not be liable to a third party indirectly affected by something which the official has done even if it turns out to have been unlawful. Illegality without more does not give a cause of action. (Lonrho Ltd. v. Shell Petroleum Co. Ltd. (No. 2) [1982] A.C. 173, 189; Dunlop v. Woollahra Municipal Council [1982] AC 158, 172; Mengel at p. 546) There is no principle in English law that an official is the guarantor of the legality of everything he does; but he is liable if he injures another by an act which is itself tortious if not justified and he is unable to justify it, however honestly he may have acted.

    The subject matter of the tort of misfeasance in public office operates in the area left unoccupied by these limits. It does not, and does not need to, apply where the defendant has invaded a legally protected right of the plaintiff. It applies to the holder of public office who does not honestly believe that what he is doing is lawful, hence the statements that bad faith or abuse of power is at the heart of this tort. Similarly, it covers the situation where the plaintiff has suffered some financial or economic loss and therefore raises the question what relationship between the plaintiff's loss and the defendant's bad faith is required: hence the use of such expressions as "targeted malice". It is necessary to locate these words and concepts in the right places in the analysis of the constituents of the tort otherwise confusion can and does occur.

    My Lords, features of this tort have to be found both in the origin and in the consequence. The official must have dishonestly exceeded his powers and he must thereby have caused loss to the plaintiff which has the requisite connection with his dishonest state of mind. The correct formulation of this nexus is one of the points of difficulty coupled with the formulation of what state of mind of the official has to be shown.

    It is not necessary to discuss further who comes within the description 'holder of a public office'. It is a broad concept (Calveley v. The Chief Constable of the Merseyside Police [1989] AC 1228, Henly v. The Mayor of Lyme 5 Bing 91, 107-8) and has been further extended by recognising that there may be a vicarious liability of the relevant governmental authority for the acts of the public official. (Racz v. Home Office [1994] 2 A.C. 45)

    The argument before your Lordships has touched upon the question whether omissions as well as acts can give rise to the liability in this tort. I would answer this question by saying that the position is the same as in the law of judicial review. If there is an actual decision to act or not to act, the decision is amenable to judicial review and capable of providing the basis for the commission of the tort. If there is a legal duty to act and the decision not to act amounts to an unlawful breach of that legal duty, the omission can amount to misfeasance for the purpose of the tort. (Reg. v. Dytham [1979] 1 Q.B. 722; Henly at p. 107) What is not covered is a mere failure, oversight or accident. Neglect, unless there is a relevant duty of care, does not suffice and the applicable tort would then be negligence not misfeasance and different criteria would apply. (X (Minors) v. Bedfordshire County Council [1995] 2 AC 633, Mengel at p. 547)

    The relevant act (or omission, in the sense described) must be unlawful. This may arise from a straightforward breach of the relevant statutory provisions or from acting in excess of the powers granted or for an improper purpose. Here again the test is the same as or similar to that used in judicial review.

    The official concerned must be shown not to have had an honest belief that he was acting lawfully; this is sometimes referred to as not having acted in good faith. In Mengel, at p. 546, the expression honest attempt is used. Another way of putting it is that he must be shown either to have known that he was acting unlawfully or to have wilfully disregarded the risk that his act was unlawful. This requirement is therefore one which applies to the state of mind of the official concerning the lawfulness of his act and covers both a conscious and a subjectively reckless state of mind, either of which could be described as bad faith or dishonest.

    The next requirement also relates to the official's state of mind but with regard to the effect of his act upon other people. It has three limbs which are alternatives and any one of which suffices.

    First, there is what has been called 'targeted malice'. Here the official does the act intentionally with the purpose of causing loss to the plaintiff, being a person who is at the time identified or identifiable. This limb does not call for explanation. The specific purpose of causing loss to a particular person is extremely likely to be consistent only with the official not having an honest belief that he was exercising the relevant power lawfully. If the loss is inflicted intentionally, there is no problem in allowing a remedy to the person so injured.

    Secondly, there is what is sometimes called 'untargeted malice'. Here the official does the act intentionally being aware that it will in the ordinary course directly cause loss to the plaintiff or an identifiable class to which the plaintiff belongs. The element of knowledge is an actual awareness but is not the knowledge of an existing fact or an inevitable certainty. It relates to a result which has yet to occur. It is the awareness that a certain consequence will follow as a result of the act unless something out of the ordinary intervenes. The act is not done with the intention or purpose of causing such a loss but is an unlawful act which is intentionally done for a different purpose notwithstanding that the official is aware that such injury will, in the ordinary course, be one of the consequences. (Garrett v. Attorney-General [1997] 2 N.Z.L.R. 332, 349-350)

    Thirdly there is reckless untargeted malice. The official does the act intentionally being aware that it risks directly causing loss to the plaintiff or an identifiable class to which the plaintiff belongs and the official wilfully disregards that risk. What the official is here aware of is that there is a risk of loss involved in the intended act. His recklessness arises because he choses wilfully to disregard that risk.

    It is necessary to add some footnotes. I have used the word 'intentionally'. This relates to the doing of the act and covers a similar point to that referred to earlier in relation to acts and omissions. It indicates that the mind must go with the act. It does not require any specific intent (except in so far as having a specific purpose under the first limb imports an intent).

    The tort is historically an action on the case. It is not generally actionable by any member of the public. The plaintiff must have suffered special damage in the sense of loss or injury which is specific to him and which is not being suffered in common with the public in general. The three alternative limbs reflect this. The plaintiff has to be complaining of some loss or damage to him which completes the special connection between him and the official's act.

    The use of the word 'directly' has a similar connotation. The act of the official may have a widespread economic effect, indirectly affecting to some extent a wide range of diverse persons. This does not suffice to give any of them a cause of action. The relevant plaintiff must be able to bring himself within one of the three alternative limbs.

    Subjective recklessness comes into the formulation at the first and last stage because it is in law tantamount to knowledge and therefore gives rise to the same liability. (See Mengel at p. 554.) The word reckless is not normally used in relation to this tort; other words are used including 'blind disregard'. At the first stage the phrase 'without an honest belief' in the lawfulness of his conduct best conveys the requisite state of mind covering both actual knowledge and dishonest disregard. At the last stage, the phrase 'wilful disregard' best describes the element of subjective recklessness in the third limb and the word 'risk' is the appropriate word to use in conjunction with it.

    The use of the words foreseen or foreseeable is to be avoided. They are concepts borrowed from the law of negligence. This tort concerns deliberate acts. Thus in the first limb the criterion is having a specific purpose, a different concept from foresight. In the second limb the concept is acting intentionally with the knowledge that the act will have a particular consequence in the ordinary course. This is like foresight but represents rather a state of mind which colours the intentional or deliberate act.

LORD MILLETT

My Lords,

1. The Community Law Claim.

    I have had the advantage of reading in draft the speech of my noble and learned friend Lord Hope of Craighead. I propose to set out my own reasons for agreeing with him that this claim must fail.

    In my opinion it is plain that the Directive does not have as one of its objects the protection of depositors. That is the function of the various national banking laws of the member states. In order to achieve their purpose these place restrictions on the freedom to provide banking services. Such restrictions are inherently anti-competitive and thus contrary to article 59 of the E.E.C. Treaty (now article 49 E.C.), but (as the recitals to the Directive itself expressly recognise) some restrictions are necessary if the interests of depositors are to be properly safeguarded. As they are imposed by national legislatures such restrictions vary from one member state to another, and this inevitably has a distorting effect on the operation of the single market. The object of the Directive was to begin the process of co-ordination of the national laws in order to remove anti-competitive barriers to the free provision of banking services throughout the single market so far as this could be achieved without weakening or impairing the protection of depositors. Thus the adequate protection of depositors was not an object of the Directive but rather operated as a constraint upon the extent to which its object could be achieved. That this is the case is in my opinion confirmed by the judgment of the Court of Justice in Société Civile Immobilière Parodi v. Banque H. Albert de Bary et Cie (Case C-222/95) [1997] E.C.R. 1-3899.

    The identification of the object of the Directive, however, is only of marginal relevance in the present case, where the question is whether the Directive confers rights on individuals. That question must be determined by reference to the particular provisions of the Directive which are relied on. What is decisive is not the interest which a Directive as a whole seeks to protect, but the interest sought to be protected by the provisions which are alleged to have been breached: see Sacha Prechal Directives in European Community Law (1995), p. 138. Attempting to identify "the underlying object" or "main object" of a Directive and to distinguish it from other subsidiary objects is in my opinion as unprofitable as it is illusory.

    There is a further consideration. The question whether a particular provision of a Directive confers rights on individuals is a secondary question. It does not arise until a breach of the provision in question has first been established. In the present case the plaintiffs allege breaches of articles 3, 6, 7 and 8 of the Directive. Unless the plaintiffs can establish a breach of any of these articles, the question whether it confers rights on individuals does not arise.

    As every banking supervisor would recognise, there is a critical difference between (i) withholding authorisation from a credit institution which is about to commence business and (ii) withholding authorisation from or revoking the authorisation of a credit institution which is already carrying on business. The former cannot injure anyone, not even the credit institution itself; it merely deprives it of an opportunity for profit. The latter damages and may destroy an established business and cause irreparable injury to the institution and its existing depositors. In the case of an existing business, there is a perennial conflict between the need to protect potential depositors and the interests of existing depositors, and difficult questions of judgment are inevitably involved.

    The Directive is careful to draw this distinction. It lays down precise requirements which must be met by a credit institution before it commences business; but it recognises that it would be inappropriate and potentially harmful to the interests of existing depositors to fetter the discretion necessary to deal with an institution already carrying on business.

    Article 3(1) imposes an obligation on member states to require credit institutions to obtain authorisation before commencing their activities. The Directive is not retrospective. B.C.C.I. had commenced its activities in the United Kingdom before the Directive came into force, and accordingly article 3(1) has no application to it. Article 3(2), which limits the power of the member state to grant authorisation, deals with the same subject-matter, viz. the authorisation of credit institutions proposing to carry on business in a member state. It requires the member state to withhold authorisation from a credit institution which does not comply with the relevant criteria. In the context of article 3 this does not include an institution like B.C.C.I. which is already carrying on business in the member state concerned. If the article applied to credit institutions already carrying on business in the member state, it would be inconsistent with the provisions of article 10(4).

    Credit institutions which commenced business in a member state before the Directive came into force are dealt with by article 10. Such institutions are deemed to be authorised and are subject to modified requirements. Whereas article 3(1) obliges member states to require credit institutions to obtain authorisation before commencing their activities, article 10(4) permits them to require institutions which are legitimately carrying on business without authorisation to obtain authorisation for the future. The United Kingdom could have exempted B.C.C.I. and other credit institutions which had commenced business in the United Kingdom before the Directive came into force from any requirement to obtain authorisation. It did not do so, but went further than the Directive required (though not further than it permitted) by obliging such institutions to obtain authorisation. In these circumstances an unlawful grant of authorisation in breach of the provisions of the Banking Act 1979 will amount to a breach of United Kingdom law but not of Community law.

    The appellants rely on Wagner Miret v. Fondo de Garantía Salarial (Case C-334/92) [1993] E.C.R. 1-6911 for the proposition that where a member state voluntarily assumes obligations laid down by a Directive it cannot escape liability by reason of the absence of any Community requirement to assume them. But that case dealt with the converse situation. There the Directive applied, but permitted member states to exclude certain categories of case from its scope. Spain did not exercise this power, and accordingly the Directive applied. The Court of Justice held that the fact that Spain could have excluded the operation of the Directive was no defence to a claim under Community law when it had not done so. The fact that the legal obligation had been voluntarily assumed did not alter its source, which was Community law. But in the present case there is no applicable Community law: article 3(1) of the Directive does not apply. The United Kingdom was not obliged by Community law to require B.C.C.I. to obtain authorisation or to withhold authorisation from it. The United Kingdom chose to impose its own requirements. Any obligation to enforce those requirements arises from national law, not Community law.

    I cannot see that these articles impose any such obligation. They impose obligations of co-operation and the establishment of appropriate ratios with a view to eventual standardisation. They take the existence of supervisory regimes in the member states for granted.

    But it is not necessary to decide this, because it is obvious that an obligation to provide "continuous and effective supervision" is far too imprecise and leaves far too great a margin of appreciation in member states to be capable of conferring rights on individuals in accordance with the jurisprudence of the Court of Justice.

    Lastly the appellants allege a breach of an obligation said to be imposed by article 8 of the Directive to revoke an authorisation in certain defined circumstances. But the article imposes no such obligation. It provides that the competent authorities "may withdraw the authorisation . . . only" in the circumstances prescribed. The word "only" is critical to the meaning. "May" sometimes means "must"; but "may only . . . if" means "must not . . . unless." The article is not permissive, let alone obligatory. It is prohibitory. It forbids the member state to withdraw authorisation from a credit institution, with the attendant risk of loss to existing depositors, except in the circumstances prescribed. Where those circumstances exist, the article does not interfere with the power of the member state to withdraw authorisation at its discretion. But it does not oblige it to do so without regard to any of the countervailing considerations to which I have referred.

    When properly understood, the scheme of the Directive makes sound regulatory sense. A member state must withhold authorisation from an institution wishing to commence business in the member state unless the criteria prescribed by article 3(2) are met. A member state is, however, free to select its own criteria for allowing an institution to continue to carry on an existing business. A member state may not withdraw authorisation or prohibit an institution from carrying on an existing business except in prescribed circumstances. Where those circumstances exist it is left to the member state to have regard to any countervailing considerations in exercising its own discretion whether to withdraw authorisation or close down an existing business.

    It follows that the appellants are unable to allege any breach of article 3 (which does not apply to B.C.C.I.) or article 8 (which does not impose a relevant obligation), while neither article 6 nor 7 confers rights on individuals.

2. Misfeasance in public office

    I have had the advantage of reading in draft the speeches of my noble and learned friends Lord Steyn and Lord Hutton with which I am in full agreement. It may, however, be helpful if I set out in my own words what I consider to be the elements of the tort of misfeasance in public office.

    The tort is an intentional tort which can be committed only by a public official. From this two things follow. First, the tort cannot be committed negligently or inadvertently. Secondly, the core concept is abuse of power. This in turn involves other concepts, such as dishonesty, bad faith, and improper purpose. These expressions are often used interchangeably; in some contexts one will be more appropriate, in other contexts another. They are all subjective states of mind.

    It is important to bear in mind that excess of power is not the same as abuse of power. Nor is breach of duty the same as abuse of power. The two must be kept distinct if the tort is to be kept separate from breach of statutory duty, which does not necessarily found a cause of action. Even a deliberate excess of power is not necessarily an abuse of power. Just as a deliberate breach of trust is not dishonest if it is committed by the trustee in good faith and in the honest belief that it is for the benefit of those in whose interests he is bound to act, so a conscious excess of official power is not necessarily dishonest. The analogy is closer than may appear because many of the old cases emphasise that the tort is concerned with the abuse of a power granted for the benefit of and therefore held in trust for the general public.

    The tort is generally regarded as having two limbs. The first limb, traditionally described as "targeted malice," covers the case where the official acts with intent to harm the plaintiff or a class of which the plaintiff is a member. The second is said to cover the case where the official acts without such intention but in the knowledge that his conduct will harm the plaintiff or such a class. I do not agree with this formulation. In my view the two limbs are merely different ways in which the necessary element of intention is established. In the first limb it is established by evidence; in the second by inference.

    The rationale underlying the first limb is straightforward. Every power granted to a public official is granted for a public purpose. For him to exercise it for his own private purposes, whether out of spite, malice, revenge, or merely self-advancement, is an abuse of the power. It is immaterial in such a case whether the official exceeds his powers or acts according to the letter of the power: see Jones v. Swansea City Council [1990] 1 W.L.R. 1453 C.A. His deliberate use of the power of his office to injure the plaintiff takes his conduct outside the power, constitutes an abuse of the power, and satisfies any possible requirements of proximity and causation.

    The rationale of the second limb is not so transparent. The element of knowledge which it involves is, in my opinion, a means of establishing the necessary intention, not a substitute for it. But intention does not have to be proved by positive evidence. It can be inferred. Proof that the official concerned knew that he had no power to act as he did and that his conduct would injure the plaintiff is only the first step in establishing the tort. But it may and will usually be enough for the necessary intention, and therefore of the requisite state of mind, to be inferred. The question is: why did the official act as he did if he knew or suspected that he had no power to do so and that his conduct would injure the plaintiff? As Oliver L.J. said in Bourgoin S.A. v. Ministry of Agriculture, Fisheries and Food [1986] Q.B. 716, 777M:

"If an act is done deliberately and with knowledge of its consequences, I do not think that the actor can sensibly say that he did not 'intend' the consequences or that the act was not 'aimed' at the person who, it is known, will suffer loss."

As that case demonstrates, the inference cannot be rebutted by showing that the official acted not for his own personal purposes but for the benefit of other members of the public. An official must not knowingly exceed his powers in order to promote some public benefit at the expense of the plaintiff.

    It will be seen from this that the real difference between the two limbs lies in the starting point. If the plaintiff can establish the official's subjective intention to exercise the power of his office in order to cause him injury, he does not need to establish that the official exceeded the terms of the powers conferred upon him. If, on the other hand, the plaintiff can establish that the official appreciated that he was acting in excess of the powers conferred upon him and that his conduct would cause injury to the plaintiff, the inference that he acted dishonestly or for an improper purpose will be exceedingly difficult and usually impossible to rebut. Moreover, as Blanchard J. pointed out in Garrett v Attorney-General. [1997] 2 N.Z.L.R. 332, 350, the consequences of his actions will usually be obvious enough to the official concerned, who can then be taken to have intended the damage he caused. I also agree with him that intention includes subjective recklessness, that is to say (to adopt his words at p. 349) "a conscious disregard for the interests of those who will be affected by" the exercise of the power.

    It is not, of course, necessary that the official should foresee that his conduct will certainly harm the plaintiff. Nothing in life is certain. Equally, however, I do not think that it is sufficient that he should foresee that it will probably do so. The principle in play is that a man is presumed to intend the natural and probable consequences of his actions. This is the test laid down by Mason C.J. writing for the majority of the Hugh Court of Australia and Brennan J. in Northern Territory v. Mengel (1995) 69 A.J.L.R. 527, viz. that it should be calculated (in the sense of likely) in the ordinary course of events to cause injury. But the inference cannot be drawn unless the official did foresee the consequences. It is not enough that he ought to have foreseen them if he did not do so in fact.

    In the present case most (and perhaps all) of the complaints made by the appellants are of the Bank's failure to act. Even their complaint that the Bank of England ought not to have granted B.C.C.I. initial authorisation is essentially a complaint that officials of the Bank failed to exercise an independent judgment and to apply the relevant criteria.

    The parties are agreed that there is no conceptual difference between sins of omission and sins of commission. This may be so; but factually there is a great difference between them. It is no accident that the tort is misfeasance in public office, not nonfeasance in public office. The failure to exercise a power is not in itself wrongful. It cannot be equated with acting in excess of power. The tort is concerned with preventing public officials from acting beyond their powers to the injury of the citizen, not with compelling them to exercise the powers they do have, particularly when they have a discretion whether to exercise them or not. There seems to be only one case in the books where a failure to exercise a power gave rise to the tort: Reg. v. Dytham [1979] 1 Q.B. 722, 727G C.A., where Lord Widgery C.J. said in terms that the neglect must be "wilful and not merely inadvertent." Ferguson v. Earl of Kinnoull (1842) 9 Cl. & Fin. 251 and the cases there cited were all cases of wilful breach of duty. Henly v. Mayor & Burgesses of Lyme (1828) 5 Bing. 91 was in my opinion a case of breach of statutory duty, not of misfeasance in public office.

    In conformity with the character of the tort, the failure to act must be deliberate, not negligent or inadvertent or arising from a misunderstanding of the legal position. In my opinion, a failure to act can amount to misfeasance in public office only where (i) the circumstances are such that the discretion whether to act can only be exercised in one way so that there is effectively a duty to act; (ii) the official appreciates this but nevertheless makes a conscious decision not to act; and (iii) he does so with intent to injure the plaintiff or in the knowledge that such injury will be the natural and probable consequence of his failure to act.

    Although we heard argument directed to the requirement of proximity and in particular the suggested need for the plaintiff to establish "an antecedent legal interest" or, as I would prefer to put it, "a legally protected interest," I cannot see that this presents a problem in the present case. The statutory powers in question were conferred on the Bank of England for the protection of actual and potential depositors, and any member of either class can satisfy the requirement.

     I agree with the order which your Lordships propose.