The Mischief in Hedley Byrne

Though I have in a sense done so, it is wrong to say that Hedley Byrne created liability for negligent misstatement. Rather, it expanded such liability beyond the bounds identified with Derry v Peek,^[552] which had been felt to be excessively con­straining in a number of cases, the most important of which was Candler v Crane, Christmas and Co.^[553] It was, of course, perfectly possible at the time of Derry v Peek that making an incorrect statement could lead to legal liability.

In Candler, the claimant invested £2,000 (now circa £75,000) in a mine in reli­ance on a statement of the mine's accounts by its owner's accountants, the defend­ant. The investment proved disastrous and the £2,000, as well as other subsequent investments of money and labour by the claimant, were lost. There was no question that the accounts were prepared negligently and there was no question of proxim­ity, for an employee of the defendant, at the mine owner's request, discussed the matter with the claimant and the mine owner.^[558] The case focused on whether the defendant had been fraudulent, and so liable under Derry v Peek, or negligent, and so not liable, on the same authority as interpreted in Le Lievre v Gould.^[559]

The problem in Candler was caused by the mine owner, who was a scoundrel who both harried and misled his accountants (their negligence in the prepara­tion of the accounts was nevertheless unquestionable) and badly mismanaged the mine, including misusing the £2,000.

Now I come to the great question in the case: did the accountants owe a duty of care to the plaintiff? If the matter were free from authority, I should have said that they clearly did owe a duty of care to him. They were professional accountants who prepared and put before him these accounts, knowing that he was going to be guided by them in making an investment in the company. On the faith of those accounts he did make the investment, whereas if the accounts had been carefully prepared, he would not have made the invest­ment at all. The result is that he has lost his money. In the circumstances, had he not every right to rely on the accounts being prepared with proper care; and is he not entitled to redress from the accountants on whom he relied? I say that he is, and I would apply to this case the words of Knight Bruce LJ in an analogous case ninety years ago: ‘A country whose administration of justice did not afford redress in a case of the present description would not be in a state of civilisation'.^[562]

Before turning to Hedley Byrne, I wish to stress what I would call the tendentious­ness of Denning LJ's approach.

I will be extremely brief about the facts of Hedley Byrne, which will be thor­oughly well known to any reader of this chapter. When considering whether to undertake advertising work for a manufacturer which put it at a particular finan­cial risk of up to £9,000 (now circa £175,000), the claimant advertising agency sought a credit reference from its bank. The bank made inquiry of the defendants, the manufacturer's bank, and was given what it understood to be a satisfactory reference,^[565] the substance of which it communicated to the claimant. When the manufacturer shortly thereafter went into liquidation, leaving the claimant with the loss it feared, the claimant brought the action in what was to become negligent misstatement against the defendant.

These facts are, of course, on a first look similar to those of Candler, but they were significantly different, especially as regards two important points. First, though it was found to be so,^[566] the balanced way in which it gave the credit refer­ence raises, in my opinion, a serious question whether the defendant was negligent at all, and, however this is, it certainly was less culpable than the defendant in Candler.

In regard of the issue I want to address, the reason these pains were taken emerges most clearly from the speech of Lord Devlin, who thought:

[T]hat the law, if settled as [the majority in Candler settled it], would be defective. As well as being defective in the sense that it would leave a man without a remedy where he ought to have one and where it is well within the scope of the law to give him one, it would also be profoundly illogical. The [defendants] in this case cannot deny that they were performing a service. Their sheet anchor is that they were performing it gratuitously and therefore no liability for its performance can arise. My Lords, in my opinion this is not the law. A promise given without consideration to perform a service cannot be enforced as a contract by the promisee; but if the service is in fact performed and done negligently, the promisee can recover in an action in tort... it is true that this principle of law has not yet been clearly applied to a case where the service which the defendant undertakes to perform is or includes the obtaining and imparting of information. But I cannot see why it should not be; there is ample authority to justify.

I want to add only the observation I made about Candler: all this is tendentiously addressed just to the duty of the defendant and nothing is said about anything the claimant has to do to obtain the benefit of that duty.

The problem that, of course, immediately arises from this, the discussion of which surely would now form a substantial library, is that, by going beyond con­tract and consideration one goes beyond privity, and having got rid of contractual boundaries to liability, one has to set tortious boundaries. I will simply say without argument that setting those boundaries on the basis of proximity understood as reasonable foreseeability, however this is in its turn understood, manifestly can­not work in respect of a tort which will have its main application in cases of pure economic loss.^[572] Other than to disavow any impression I may just have given that I believe that reasonable foreseeability is generally workable in respect of other forms of negligently caused loss, I have nothing more to say about the negligence model of liability for misstatement, and I turn to the additional ‘special relation­ship’ that has almost always been thought necessary to limit the reach of such liability,^[573] sometimes to the point it entirely displaces it, and specifically to the assumption of responsibility model of that relationship.

How can a party legitimately come to rely on another party’s assumption of legal responsibility for a negligently made statement? It should contract.^[574] If a party contractually agrees to provide professional advice or the like, the justification for then holding it responsible in this way is that it actually has voluntarily undertaken the responsibility.

Now, even in a contractual context the categories of fiduciary relationships are not closed, and indeed their proliferation has on occasions proven hard to keep in check, but when they arise from contracts, fiduciary relationships do (or should) partake of the essential quality of being paid for. But the whole point of Hedley Byrne was to move beyond contractual liability. One way in which this was bound to happen was that the understanding of ‘ fiduciary' as it related to negligence would be expanded so that its connection with contract was broken.^[576] But, more importantly, this was merely a subordinate part of what was essential to the assumption of responsibility model, which is that it is used to describe situa­tions about which the last thing that could be said was that there actually was an assumption of responsibility.

Lord Devlin's identification of a category of ‘relationships which... are “equivalent to contract”, that is, where there is an assumption of responsibility in circumstances in which, but for the absence of consideration, there would be a contract',^[577] was intended ‘to settle the law so that the presence or absence of consideration makes no difference'. ^[578] But a contract without consideration is not a contract, and by ‘equivalent to contract' Lord Devlin meant something entirely different to contract. This, with the greatest respect but one has to say, very unhelp­ful equivocation was at the heart of the assumption of responsibility model from the outset.

Summarising the speeches of his brethren in Hedley Byrne, Lord Devlin said: ‘ I do not understand any of your Lordships to hold that it is a responsibility imposed by law upon certain types of persons or in certain sorts of situations. It is a responsibility that is voluntarily accepted or undertaken'.^[579] I can neither interpret this as a flat inability to understand the issue, which I find hard to attribute to this most intellectually gifted of judges, nor as ratiocination aimed at unscrupulously getting one's way, which I find almost as hard to attribute to this most decent of legal professional men. I will try to explain how Lord Devlin came to see things this way below. But, however this is, he avoided the basic issue. One can voluntarily make a statement and indeed intend it be relied upon, but whether one voluntar­ily assumes the indemnification of the reliance is a quite separate matter. This can be determined by assessing what the parties intended, which is a matter of contractual interpretation, or be determined by imposing the indemnification and then reading this back into the ‘voluntary' assumption of responsibility. The more that the assumption of responsibility model is based on the Henderson v Merrett ‘objective' assumption of responsibility,^[580] the more it departs from a voluntary assumption of responsibility, as was latent in Hedley Byrne from the outset.^[581] It is not going far enough to say that, if one decouples ‘voluntary' and ‘assumption', one decouples ‘assumption' from its legitimate meaning in a way of which only Humpty Dumpty could approve. One is, in fact, calling an imposition an assump­tion in a Newspeak fashion of which only Minitrue could approve.^[582]

I would hope to be allowed to claim, on the basis of previous work, that I am not insensitive to the shortcomings of the law of contract in general or the doctrine of consideration in particular. But I nevertheless can clearly recall the incredulity with which I read the following 1998 dicta of Lord Steyn in Williams v Natural Life Health Foods Ltd:

[T]he restricted conception of contract in English law, resulting from the combined effect of the principles of consideration and privity of contract, was the backcloth against which Hedley Byrne was decided and the principle developed in Henderson’s case... while the present structure of English contract law remains intact the law of tort, as the general law, has to fulfil an essential gap-filling role. In these circumstances there was, and is, no better rationalisation for the relevant head of tort liability than assumption of responsibility.^[583]

Was Lord Steyn really saying that defects in the law of contract meant that we should turn to negligence? I thought it significant that, if this was his position, he seemed unaware that it repeated the argument made by Professor Markesinis a decade earlier that an ‘expanding tort law' was ‘the price of a rigid contract law'.^[584] This argument was an encomium to the House of Lords' decision in Junior Books Ltd v Veitchi Co Ltd,^[585] of which our appellate judges had in that intervening decade become so ashamed that they tried to airbrush it from legal history in a manner most unusual if not unique, effectively overruling it in the Court of Appeal.^[586] One immediately thinks that Lord Steyn proposed to jump from the frying pan into the fire, but this is not enough. He proposed to jump from a sometimes uncomfort­ably hot bath into a ladle of molten metal. Though Lord Steyn simply discounted it,^[587]9 I have nothing to add to Professor Hepple’s conclusion, based on a compelling comparative assessment of the ways the law has been advanced in contract and in negligence, that ‘ The way to create liability in situations “equivalent to contract” was surely to broaden the conception of contract’.^[588]

What I might add to Hepple, however, is to say that Lord Steyn did not really have specific defects in the law of contract in mind. He had it in mind that con­tract was defective tout court.^[589] He found it unacceptable that contract could place limits on the indemnification of a claimant when the defendant objectively had assumed responsibility. This position is tenable only if one has a complete lack of sympathy with the core value of liberal economics, which is the autonomy of economic actors institutionalised in freedom of choice or contract.

It overwhelmingly is the case that such consideration as economics occasionally receives from those whose strength is in law is framed in terms of evaluating the efficiency with which, as it is very tellingly put, ‘society’s resources’ are utilised. It is undeniable that the law of negligent misstatement, indeed of negligence and much other rights discourse, has proceeded in blithe ignorance or disregard of the fact that the creation of rights imposes costs as well as confers benefits. But even when it is realised that taking a decision to indemnify reliance on a statement imposes costs, the issue is viewed from the theoretical perspective of a planner who is able to determine whether the contractual outcome or the tortious out­come is superior. This implies that the planner, speaking on behalf of ‘ society’, is cognisant of an optimal use of society's resources and can decide which outcome best approximates to it.^[590]

The criticism I have so far made on Hepple's authority is one of lack of com­putational competence. How can the thinking Lord Steyn expresses dare to claim that it can identify an optimal use and assess an outcome's proximity to it? It can't. But this is not the main point, which is that it shouldn't. When an economic actor chooses to rely on a statement, then, as the statement could be wrong, the actor creates and bears a risk. Whether and to what extent the actor nevertheless chooses to rely is, in the first instance, a personal matter. But it is no longer a personal mat­ter if reliance includes indemnification of the risk that the statement is wrong, for this requires the actor to obtain the good of indemnification from another. Putting aside direct indemnification by the government, indemnification may be bought from the party making the statement (or from an insurer, a situation I also shall put to one side). How much, if any, indemnification should an actor buy? If the economy is one which respects the actor's choice, then no one can know other than the actor herself. The choice is a function of both the situation the actor faces and her capacity to decide, and that capacity to decide is, precisely, hers, expressing her status as an autonomous actor. In advance of the buyer ‘revealing' her prefer­ences by her choice, it is senseless to talk of having the information by which the optimum can be determined and so approximations to that optimum assessed, for the optimum is derived from the choice, and if an optimum is chosen by someone other than the actor, it is not an optimum.

Liberal democratic society's basic claim to legitimacy rests not on the moral value of particular social goals set by that society, but on the extent of the freedom of its citizens to set their own goals. One may say that the goal of liberal demo­cratic society should be to eschew the pursuit of particular social goals and to maintain neutrality between the goals set by its citizens. In particular, the claim that the market economy is efficient is not a claim that that economy efficiently produces a particular set of morally valued goods but that goods are produced according to the choices of economic actors.

In the rather formal, but I think helpful to those who I anticipate might read this chapter, terms of neo-classical economics, the first theorem of welfare eco­nomics, Pareto optimality, identifies a perfectly efficient distribution of goods as an equilibrium established under conditions of general competition in which mutually beneficial exchanges have taken place in complete accordance with the voluntary choices of economic actors. Under general competition, goods will be exchanged up to the point where the increase in one actor's utilities achieved by further exchange would be more than offset by the diminution in the sum of another actor's. At this point, the market is in equilibrium because there are no further mutually beneficial exchange opportunities and, vitally importantly, it has been brought there by the uncoordinated working out of voluntary exchanges, which automatically identify the point of Pareto optimality by reaching equilib­rium. The beautiful symmetry of the model lies in its being driven by voluntary exchange and working only because it is so driven. This is the source of the power of the rejection of ‘patterned principles' of distribution in favour of the ‘pure pro­cedure' of the market in liberal political philosophy, for any state imposition of a ‘fair' distribution of goods must prevent the perfectly efficient distribution which would be voluntarily reached at general competitive equilibrium. Efficiency, in sum, is not a matter of the achievement of an objectively determined optimum but the actualisation of a procedure in which the optimum is identified by the voluntary choices of economic actors.

Under general competition, transacting is costless and so economic actors' choices are made with perfect information and are perfectly well expressed in the actors' negotiations. Such general competition is, of course, purely theoretical, and the concept of Pareto optimality is akin to what Kant called a ‘ regulative' idea or principle. It states an end which is of great value in ordering our thought and action, though it is not vouchsafed to us to realise that end. Pareto optimality allows us to understand the nature of autonomous economic action, but it is of no direct value at all to the analysis of any empirical process of Pareto optimisation, of which the complete satisfaction of voluntary choices is the unrealisable but nevertheless essential goal. Under all empirical conditions of choice, with infor­mation and negotiation costly and imperfect, risk is universal, and the way one handles this risk is a most important part of one's qualities as an autonomous economic actor.

It makes no sense whatsoever to speak of handling only the upside of risk. Unless there is a possibility of suffering the consequences of one's mistakes, one cannot make an economic choice. There is a very compelling argument that the possibilities both of making a gain and of suffering a loss provide the best incen­tives to rational economic action. But it is much more important to acknowledge that autonomous economic choice cannot exist without the possibility of success and failure. If the government intervenes to prevent or limit either or both by means backed by its monopoly of violence, this is a coercive overriding of auton­omy, and the political justification of such intervention is possible only if this is acknowledged at the outset. As Kant himself somewhat imperfectly understood, autonomy is by no means the only value which a system of (social) justice, as opposed to a system of (personal) morality, must respect, and such justification is entirely possible. Practical welfare economics have to contemplate many second, not first, best choices which are given effect by coercive transfers, the principle of which runs entirely counter to Pareto optimality. Nevertheless, the technicalities of neo-classical economics are expressive of fundamental moral rights and duties which emerge from acknowledging the autonomy, and therefore responsibility, of economic actors.

It is, as a normal case, economically irrational and, what is the same thing, mor­ally unjust that one's choice to rely on a statement is legally indemnified without one having to pay for the benefit of the indemnification. Exactly the wrong posi­tion about this was established by Hedley Byrne and then adopted, and of course extended, in the cases which have found liability on the basis of it. The process, one trusts, reached its doctrinal culmination in Henderson v Merrett, which, in its opening of the possibility of indemnification, effectively by everyone who ulti­mately pays the costs of insurance, of certain claimants who categorically should in private law have suffered the consequences of their contractual position,^[591] is the second most unjustifiable modern case I can recall in almost 40 years of study of English private law.^[592] That the finding of tort liability extended the limitation period to the advantage of the claimants seems to have been treated as the justi­fication of that finding,^[593] a deplorable p etitio principii wholly expressive of the tendentiousness of Hedley Byrne reasoning.

I hope I am not putting words in his mouth if I say that, in the casenote on Hedley Byrne I have mentioned, Weir said all that I would like to say about what was done in that case, and I will quote him by way of conclusion of this section of the chapter:

Now a law journal is no place for considerations of justice, but a glance at the plaintiffs in [the] line of cases [culminating in Hedley Byrne] reveals that their claims to redress are not indisputably high. They made bad business deals, having taken only a free opinion before hazarding their wealth in the hope of profit, no part of which, had it eventuated, would they have transferred to the honest person whom they now seek to saddle with their loss. The defectiveness of a system which refuses in such a case to sever the risk of loss from the chance of profit is not obvious. It would admittedly be defective if it were made impossible for the investor to share the risk, but... the plaintiffs here could have found a credit investigation agency; had they done so, the system would have afforded them a remedy, through the appropriately commercial institution of contract, against such of their advisers as were careless; the risk on the adviser would be justified by the fee. One can hope, perhaps, that in most cases it will continue to be ‘reasonable’ to rely only on a word one has bought. A free tip is relevantly distinguishable from a remunerated opinion.^[594]

III.