Barker, Kit --- "Economic Loss and the Duty of Care: A Study in the Exercise of Legal Justification" [2008] UQLRS 1; (2008) Justifying Remedies in the Law of Obligations Hart

Last Updated: 5 January 2009

THE UNIVERSITY OF QUEENSLAND

LEGAL RESEARCH SERIES

ECONOMIC LOSS AND THE DUTY OF CARE:

A STUDY IN THE EXERCISE OF LEGAL JUSTIFICATION

“Save us from this fog and give us a clear sky, so that we can use our eyes”

(Homer, The Iliad (Trans, Rieu) (1956), 333, cited by Kirby J in Barclays Oyster Pty Limited and Another v Ryan [2002] HCA 54 at [211].

Kit Barker*

This piece was originally published in C Rickett (ed), Justifying Remedies in the Law of Obligations (Hart Publishing, Oxford 2008)(ISBN:978-1-84113-814-5), pp 175-203 and is reproduced with the kind permission of Hart Publishing.

Introduction

Few topics are as vexed in private law as the question of when one party owes a duty of care to avoid harming another’s pure economic interests. The problem is not simply one of identifying the exact boundaries of liability themselves, but in divining the best overall method via which the issue should be approached. In England, at least three nominally distinct approaches are currently taken, all of which bear the authority of the House of Lords - an ‘incremental’ method of reasoning by analogy with existing categories of case;[1] a strategy based on the twin concepts of ‘assumption of responsibility’ by a defendant and ‘reasonable reliance’ by a claimant;[2] and a three-stage approach based on the ‘foreseeability’ of harm, ‘proximity’ and ‘fairness, justice and reasonableness.’[3] The relationship between these approaches is troubled and some judges have sceptically suggested that it does not really matter which one takes, because they all tend to yield the same result, once the facts and policy issues are properly considered.[4] In new types of case, we are now encouraged to apply the second and third in turn (and in that order), with the first acting as a cross-check upon the results, so as to ensure that they remain reasonably consistent with legal precedent.[5]

In Australia, after a period of considerable turmoil,[6] the High Court now seems to have settled (how fleetingly is still unclear) on a single, ‘multi-factoral’ approach.[7] This is intended to be true to the spirit of incrementalism, but eschews both the rigid categorisation of cases and high-level generalisations in favour of lower-level ideas applied across the board, case by case.[8] These ideas include (but are not confined to) the foreseeability of harm, the impact of liability upon a defendant’s commercial autonomy, the prospect of indeterminate liability, a claimant’s vulnerability and the defendant’s knowledge of the risk created by his conduct.[9] In the same way as the English approach has been criticised for being overly abstract and discretionary,[10] the multi-factoral or ‘salient features’ approach has sometimes been said to disclose no general approach at all.[11] Both the criticisms of and the contrasts between the English and Australian approaches may actually turn out to have been exaggerated, as I venture to suggest later in this chapter, but it is clear that neither is seen as ideal.

A single answer to the question when a duty of care is owed is not to be expected, but a stable, principled approach to the way in which it is asked really ought to be an attainable ambition.[12] The purpose of this chapter is to reflect more closely upon the reasons why the task has proven so problematic and to provide some methodological suggestions which are sensitive to the difficulties. The problems for legal reasoning turn out to be of two sorts. They relate in part to the individual justifications (reasons) used by judges in economic loss cases, which are difficult to isolate and much contested. They also stem, however, from a number of more general problems which obstruct the integration of these reasons into coherent legal rules and principles. These distinct, but associated sources of difficulty are considered in Parts 1 and 2 respectively. Part 3 provides some conclusions and critical reflections on method, the thrust of which I hope will be useful to judges in both jurisdictions.

Part 1 - Reasons: Identifying Relevant Justifications

A key to understanding the restrictive approach which courts have taken in cases of negligently inflicted pure economic loss is the realisation that their concerns have not historically related to the nature of the loss itself, but rather to a number of contingent issues which tend to arise in instances in which it happens to take this form.[13] From the point of view of corrective justice, such losses are no less deserving of redress in principle than cases of damage to the person or to property. They represent a form of harm which a wrongdoing defendant has a basic moral responsibility to rectify.[14]

Some writers, it is true, have attempted to set up relevant, moral distinctions between pure economic losses and interferences with proprietary or possessory interests,[15] but their attempts generally run into difficulties of one sort or another. The strongest argument - that such interests simply rank lower in the list of human values[16] - has considerable intuitive appeal, but falls short of explaining why they should be unprotected in principle. Whilst it accurately recites the fact that we tend to regard other interests as more important, it does not explain why economic interests are not important enough to merit protection in their own right against behaviour which is culpably careless.[17] The intuition that intangible financial interests are less valuable than property also seems increasingly questionable, given the central place which they play as investment vehicles in modern market economies. Whereas in the past, much of our wealth has been tied up in tangible form, this is less so in today’s world that ever before. Moreover, whilst property clearly has some personal ‘added value’ over abstract wealth in many instances,[18] it does not always. Landlords frequently choose to buy houses, for example, simply for investment purposes - as instantiations of wealth pure and simple - not because the property has any real meaning to them, or expresses anything particular about who they are. In a modern, fast-moving, mobile and market-driven world, economic interests can no longer be viewed as a peripheral social phenomenon. They are central to modern life.

From this point of view, Lord Oliver’s famous statement in Murphy v Brentwood[19] that causing economic loss does not need to be justified in the same way as damage to property and is not in itself ‘wrongful’ admittedly looks very odd. If one scrutinises more closely the passage in which the statement appears, however, it is clear that his Lordship’s point was not that a moral distinction can be drawn between the different types of loss. Rather, it was that the normal approach to the question of legal duty had to be ‘qualified’ when it came to economic harm, such that something more than its mere foreseeability would be required for liability to arise.[20] It is not, therefore, that carelessly causing economic harm needs no moral justification (is not in itself morally questionable), but that, whatever the basic moral position of a careless actor, we may be justified in restricting her legal responsibility for other reasons. This is certainly the way in which the issue was understood by Lord Denning right from the start[21] and courts in both England and Australia now tend to take that view. Unravelling the various judicial reasons for discriminating between moral and legal responsibility in economic loss cases is, however, a decidedly more problematic business.

Absence of Social Loss – Neutral Wealth Transfers

One possible reason provided by McHugh J in Perre v Apand [22] is that, whilst economic losses invariably harm a particular person, they sometimes entail no ‘social’ harm, because the individual loss gets cancelled out by the benefits it yields to third parties. An example would be where the careless interruption of the electricity supply to a defendant’s business premises detrimentally affects his production and thereby reduces his market share, which is then simply taken over by an enterprising competitor. Negligence in these circumstances results in wealth transfer, it is said, but no harm is done overall. ‘Tis an ill wind that blows nobody any good.

There are two immediate difficulties with this reasoning, even assuming the example to be realistic.[23] For one thing, exactly the same ‘wealth transfer’ phenomenon occurs where the claimant loses production as a result of damage to his property (e.g. his plant) or person (as where he is no longer able to operate his business as a consequence of injury).[24] But at another level entirely, as a system of private rights tort law is generally unconcerned with aggregate social harm. It focuses on damage done to individual interests and indeed exists precisely to protect them against the potential tyranny of collective calculus. The possibility that a claimant’s carelessly inflicted economic ‘minus’ will be cancelled out by ‘pluses’ to other market operators fails to provide a convincing reason to rule out a duty of care. Indeed, if it did, we would live in a very insecure world indeed. Our rights would logically be up for grabs by anyone able to make at least as much profit out of them as we ourselves.

Stifling Legitimate Competition

A second similar, but distinct concern is that liability for pure economic loss could stifle legitimate competitive behaviour.[25] According to McHugh J in Perre v Apand,[26] this argument focuses on the potential impact of liability on the private right of an individual to pursue her commercial interests, not (like the previous one) on broader social effects. It is an argument for the protection of a defendant’s personal autonomy.

Whilst the right to compete is clearly an important aspect of individual freedom, there is actually more than one way in which it can be understood. It can thus be construed narrowly, as a right to engage in legitimate competition with others[27] (in which case it is presumably an issue in only a minority of cases where claimant and defendant are active in the same economic sector, or commercial transaction); or, much more broadly, as a right to run one’s business free of precautionary costs which are so high as to unreasonably undermine its general competitiveness (economic viability).[28]

The narrow version of the argument describes a good reason for ruling out a duty of care in some cases. If claimant and defendant are economic competitors, their position will be analogous to that of parties engaged in the arms-length negotiation of a contract, who tacitly embrace the value of self-interest. The intrusion of ‘neighbourly’ obligations of care would represent an unwelcome violation of an accepted commercial norm, the fruits of which are reaped in the same way as its costs are consensually born.[29] By contrast, the broad interpretation of the right to compete (as a right to be commercially competitive) discloses a far weaker basis for eradicating a defendant’s responsibility. The potential burden of preventive precautions to a defendant’s business may be one consideration in determining whether the risk to which she has exposed a claimant’s interests is unreasonable, but it does not seem a good reason to entirely displace her moral responsibility for considering them. Where the prospective risk is one of damage to property or person, for example, potential precautionary costs are weighed and balanced in the scales of the breach inquiry,[30] but they do not rule out the obligation to take care. The moral justification for taking a different approach simply because the prospective harm is economic rather than physical is very hard to see.

An added complication is that, whilst the dominant contemporary interpretation of the competition argument is based on a defendant’s personal autonomy, it has also, on occasion, been understood as a broader, instrumentalist concern about the effects which liabilities might have upon the functioning of the market economy as a whole.[31] In this guise, it has more in common with the first argument about ‘wealth transfer’ than it does with McHugh J’s concern about personal autonomy. It abstracts from the particular interests of the parties to the dispute and is speculative in the sense that the effects of tort liabilities on the health of a market-economy are hard to predict and courts have little empirical data upon which to base their conclusions.

It is not insignificant that whilst the High Court of Australia has often referred to the significance of a claimant’s right to compete, in no case to date has it been used as a justification for ruling out a duty of care. A little reflection discloses good reasons for this. Although the narrower version of the autonomy argument is sound, it is rarely in point: there is no sense in which builders and house purchasers,[32] solicitors and beneficiaries,[33] or potato suppliers and growers[34] can realistically be construed as economic competitors. Equally, the broader (and decidedly more suspect) version is largely irrelevant where the defendant is already under an independent obligation to incur precautionary costs. Such an obligation will arise whenever a defendant’s conduct poses a foreseeable physical risk to the claimant, or consists in the provision of services pursuant to a contract.[35] In these cases, obligations of economic precaution add little extra to the defendant’s bill.

Indeterminate Liability

A third concern, which is most often cited in cases involving negligent advice or ‘relational’ economic loss, relates to indeterminate liability. This argument is notoriously indeterminate in itself. In fact, Cardozo CJ’s infamous statement in Ultramares Corporation v Touche[36] has too often been used as a hold-all for a variety of sub-forms of argument, some of which purport to be concerned with the moral implications of economic loss liabilities and some of which do not. These need to be isolated and addressed individually if they are to be the slightest bit meaningful.

The less common forms of the argument relate to (a) the possibility that actions might be postponed for an indeterminate period of time as a result of financial harm assuming a ‘latent’ and therefore undiscoverable form[37] (b) administrative arguments about the prospects of courts being swamped by vexatious litigation[38] (c) the idea that allowing one type of economic loss claim will open the door to others (the ‘slippery slope’ argument)[39] and (d) the moral impropriety of imposing on a defendant high levels of liability disproportionate to his fault.[40] The spectre of large liabilities is said to stem either from the ease with which inaccurate financial information can be transmitted to a wide audience, or from the close interdependency of financial interests and the capacity of a single act to send foreseeable ‘ripples’ down the financial chain. More popular versions relate to (e) the undesirable effects which such high levels of liability might have upon the availability and/or price of services and the economy as a whole[41] (f) the potential uncertainty for defendants about the likely ‘class’ of affected persons,[42] given the ‘ripple effect’ phenomenon referred to above and (g) the potential uncertainty for defendants about the extent or type of their prospective liabilities.[43]

Many of these arguments have legitimately been questioned. Unacceptable time delays seem logically best precluded via the limitation of actions[44] and ‘administration of justice’ arguments by summary judgement, the striking out of unmeritorious claims, or increased judicial resources. The idea that compensation for economic loss ought to be proportionate to moral wrongdoing is also highly dubious[45] and in any event ruling out a duty of care where loss is large does not achieve proportionate responsibility, it eradicates it completely. The perverse result of succumbing to this strain of reasoning would be that the more harm one did, the less likely one would be to be held responsible not just for some, but for any of it.

Problems of this nature may explain why the High Court of Australia has recently preferred to understand the concept of ‘indeterminacy’ in terms of the last two arguments identified. It is not immediately obvious, however, how ex ante uncertainties about the likely class of potential victims or the type/quantum of economic loss are relevant morally or instrumentally to a defendant’s liability, other than by making it more difficult (and therefore more expensive) for her to arrange appropriate levels of liability insurance. The worries are then not about uncertainty itself, but about the indirect effects which it is likely to have on the provision or pricing of services (sub-argument (e) above), or on the most ‘efficient’ (cheapest, easiest) way of distributing the loss. These concerns are again speculative, in the sense that any assessment of their merits entails detailed knowledge of relevant insurance markets and (even more problematically, perhaps) accurate prediction of the effects which liabilities might have on insurance premiums, market prices and service provision.

An alternative way of understanding the concern about the uncertainties associated with ‘ripple effects,’ is that they are relevant not from the point of view of their indirect social effects on prices, services, or loss-distribution, but because of their potential impact upon a defendant’s autonomy. The ability to plan one’s future is, after all, an important aspect of personal choice and liabilities which are wholly incalculable might represent a sword of Damacles suspended over defendants’ heads, unduly inhibiting their freedom of action. This argument would avoid difficult speculation about the possible social impact which the uninsurability of such losses might have, but the question remains whether the mere incalculability of a foreseeable loss is a good enough moral reason for denying a person’s responsibility to pay for it, when it has been carelessly caused. If the potential size of a liability is insufficient reason from the point of view of a defendant’s autonomy to relieve him of liability (as McHugh J suggests),[46] it is hard to see how uncertainties about the nature or extent of that liability can be, since the indirect costs of such uncertainties for a defendant’s freedom of action are almost certain to be less than the full weight of the liabilities themselves.

In the event, then, not only have arguments relating to indeterminacy until very recently been insufficiently distinguished from one another, but if, as seems to be the current thinking in Australia, they relate to uncertainties about liability for defendants, one still needs to know why such uncertainties are important if one is to formulate anything like a coherent approach to liability. It is hard to see how the incalculability of a prospective liability weighs more heavily upon a defendant’s freedom of action than its actual size, so that if it is really the uncertainty rather than the size of liability that is key, it is tempting to conclude that this must be because of its indirect bearing upon the question of which party is in a better position to insure against the loss. Those further from the ripple’s epicentre will be closer to its potential effects and therefore in a better position to assess and insure against the harm. The snag with this rationalisation is that the High Court of Australia currently bluntly denies that the respective insurance positions of the parties are relevant to liability,[47] which leaves us at something of a loss. A possible reaction is to regard the denial with circumspection, but that brings significant justificatory difficulties of its own. Perhaps the most one can say at the current point in time is that if the relative insurability of economic losses is relevant, this is something that should be openly admitted, rather than buried within ‘indeterminacy’ arguments.[48] Accepting this, it is probably best considered as part of the debate about ways in which a claimant might reasonably have protected himself (considered further below), though even then it is questionable whether it should rule out a duty of care, as we shall see.

Coherence and Conflict

A fourth concern relates to potential conflicts which economic duties of care might yield with other legal systems or norms - to the potential ‘emasculation of other bodies of legal doctrine.’[49] This issue typically attends the expansion of any body of law into fields which are already the subject of legal governance or commercial bargaining, of which this is clearly one. As with the indeterminacy argument, it has a number of distinct manifestations, which include: (a) the worry that duties of care might contradict the way in which risks have been consciously allocated by parties to a dispute in the context of a contractual matrix[50] (b) the concern that they may undermine an existing domestic or international statutory regime regulating parties’ conduct or remedies[51] (c) worries that liability might produce ethical conflicts for professionals by extending legal responsibility beyond their immediate clients[52] (d) fears that liability might undercut the immunities normally provided to directors by principles of corporate personality[53] and (e) concerns that the creation of new negligence liabilities might bring courts into conflict with their own constitutional role, by embroiling them in areas of social policy better suited to reform via the Parliamentary process.[54]

Conflict concerns arise regularly, even where ‘indeterminacy’ is absent. As tort law gradually expands its protection of economic interests, they will feature ever more prominently, since such interests are so regularly (and rightly) the object of commercial risk-allocation and statutory attention. All versions of the argument require detailed consideration of individual facts to determine whether potential conflict is present. They are not unique to cases involving economic loss, or the law of tort.[55] Moreover, whilst ruling out a duty of care will often be the only way of avoiding conflict, it should be born in mind that more sensitive mechanisms may exist for resolving the tension. It may sometimes be more appropriate, for example, to align the standard of care owed in tort with a contractually specified standard,[56] than to reject any tort duty out of hand on the basis that the two obligations might hypothetically clash. Courts should also not be overly ready to assume that statutory remedial regimes are intended to exhaustively define parties’ remedies, particularly when the remedies they do offer are ill-adapted to effectively compensating the harm done.

Alternative Means of Protection

A final, highly topical judicial concern is that a claimant may have an alternative means of protecting himself against economic harm.[57] Whether or not private insurance is regarded as one such means is still a matter of great uncertainty, as we intimated above, but a contractual warranty apparently might be,[58] as might a private law remedy,[59] a public law appeal mechanism,[60] or the possibility of obtaining independent verification of the reliability of information[61] or building work.[62] The High Court of Australia currently targets this concern via the concept of ‘vulnerability.’ Whether a claimant was ‘vulnerable’ to economic harm turns on whether she was unable reasonably to protect herself against it by reason of ‘ignorance, or social, political or economic constraints.’[63]

An immediate uncertainty is whether this is a concern about the (un)reasonableness of the individual claimant’s own failure to protect herself on the individual facts, or embodies a broader judicial strategy for distributing the burden of precaution between different social groups, such as the builders of defective premises and their purchasers. If the former, then it is common to all negligence cases and in that way wholly uncontroversial, but the justification for regarding it as reason to rule out a duty of care is then unclear. An unreasonable failure to protect oneself is a well-established basis for reducing the damages one can claim on grounds of contributory negligence or failure to mitigate loss. Exceptionally, it might be regarded as founding a conclusion that one caused one’s own loss, or that the latter was too remote a consequence of the defendant’s negligence to justify holding him legally accountable for it. It is not usual, however, to regard the fact that one has a reasonable responsibility for protecting one’s own interests as relieving a defendant in principle of his responsibility to avoid harming them. A claimant’s own failings do not make his injurer’s actions right. This suggests either that courts are mistaken in regarding the availability of an alternative means of protection as a factor negating a defendant’s duty of care (though it has a clear role to play at lower levels in the negligence inquiry in relation to causation, remoteness and the mitigation of damage), or that that the second interpretation of the argument better represents courts’ true concerns.

If the latter is the case, the idea is that whatever an individual claimant’s actual circumstances, the fact that claimants in the same class normally have the capacity to protect themselves (for example by virtue of their economic power or commercial sophistication) is a good enough reason in itself to allocate the responsibility for economic precaution to her, rather than the defendant. The debate about ‘vulnerability’ and ‘alternative means of protection’ is then pertinent in the sense that it raises the question of which of two social groups ought to bear the onus of precaution, but is overtly distributive in nature. From the point of view of those who think that broad, distributive criteria are irrelevant to the outcome of negligence claims (McHugh J seems to be one),[64] the first interpretation will be preferable, though it will require detailed investigation in each case into the individual circumstances of the particular claimant. The majority in Woolcock seem to endorse this type of approach. Callinan J’s blanket suggestion, however, that all purchasers of real estate (as a group) have ways of protecting themselves[65] seems closer to the second view, since it makes no apparent attempt to assess whether or not such means were available to an individual claimant on the facts.

If we adopt the former view, as I suggest here that we should, a key question which courts may need to reconsider is the extent to which the availability of private insurance against economic harms is something which individual claimants should now consider as a matter of course. Courts’ formal insistence on the irrelevance of such insurance seems inconsistent with the suggestion in Woolcock that a claimant may sometimes be expected to protect herself by obtaining a contractual warranty from a vendor, which is, after all, simply another way of insuring herself against the risk. The relevance of insurance to tort liability is supported to some degree by recent comments of Lord Hoffmann,[66] (albeit in the context of a debate about liability for property damage under the rule in Rylands v Fletcher[67]) and Lord Mance in the House of Lords’ most recent economic loss decision, Commissioners of Customs and Excise v Barclays Bank.[68] In the latter case, his Lordship clearly thought that the difficulties for a claimant in obtaining first party protection were relevant, though they were ultimately outweighed by the insurance costs which non-commercial defendants might have to bear, if economic duties of care were extended in the manner proposed.[69]

Summary

The justifications currently used by judges for restricting legal responsibility for pure economic loss are therefore very ill-defined at best and in no sense logically unique to cases involving losses of this type. Two of the justifications we have examined (legitimate competition, indeterminate liability) can be interpreted either as autonomy-based arguments focusing on the moral effects of liabilities on defendants, or as broader, instrumentalist arguments about the social or economic repercussions which such liabilities might have. The argument about alternative means of protection can likewise be understood either as a way of re-focusing our attention on familiar issues about a claimant’s reasonable responsibility for protecting herself which are highly individualised and fact-specific, or as relating to broader issues about the distribution of precautionary costs between different social groups. The most clearly defined justifications relate to conflict and wealth transfer, the former of which seems a perfectly sound and increasingly relevant concern in a system seeking to navigate its way through other systems and norms, but the latter of which is, with respect, incompatible with the idea of tort law as a system for the protection of private rights and should probably be discarded altogether.

Part 2: Concepts: Integrating Justifications into Legal Reasoning: Problems and Ideals

(a) Problems

The preceding section discloses a number of good explanations for the general incoherence of courts’ approach to the duty of care question in economic loss cases. Courts’ justifications are almost universally ill-defined and some are normatively suspect. Aside from their individual failings, these justifications also exhibit a number of collective characteristics which make them difficult to integrate coherently into legal principles and rules. These features have implications for the style of approach that should be taken to the duty of care question, which we address shortly.

Positive Principles, Negative Justifications

Firstly, all the justifications are negative, acting as they do as reasons to restrict the scope of liability for harm caused by fault. This makes them difficult to express in terms of positive duty principles without some risk of obscurity and inaccuracy. Positive justifications for liability convert relatively precisely into positive liability criteria. The reasonable foreseeability of harm, for example, features universally as a prerequisite for duty in all common law approaches and is the best - perhaps the only really well - understood duty ‘principle’. By contrast, it is hard to express negative concerns such as those relating to indeterminate liability or economic competition in anything other than their own terms. Attempting to address them through positive liability principles such as ‘proximity of relationship’ ‘reliance,’ ‘assumption of responsibility,’ ‘vulnerability,’ or ‘knowledge of risk’ tends to obscure them and can draw judicial aim away from its mark. This may indeed explain why Lord Wilberforce chose in Anns[70] to break the duty of care inquiry down into two, distinct stages - one (Lord Atkin’s neighbour principle) reflecting primarily positive reasons for liability based upon the reasonable contemplation of likely harm, and one consisting simply in ‘negativing considerations.’ Whatever the difficulties associated with this approach,[71] it had the obvious merit of avoiding difficult translations from negative reason to positive principle and drove the various concerns out into the open, enabling them to be better identified and targeted.

By contrast with Lord Wilberforce, courts in both England and Australia remain hesitant about expressing legal requirements in economic loss cases directly in terms of negativing policy concerns.[72] This is not to say that the latter are regarded as unimportant - they are actually now identified and discussed as central even in England - but simply that their translation into liability principles is thought to be an important component of the process of legal justification. This may be because courts are still uncomfortable being seen as direct arbiters of ‘policy,’ but these days it is actually at least as likely to be that legal rules and principles are simply thought to offer a level of certainty which policies inherently lack.[73] Anyone familiar with the confusion currently attaching to duty of care concepts in both jurisdictions may be forgiven for finding this more than a little ironic, but the theory is sound enough.

Coherent Principles, Multiple Justifications

Secondly, even if we eradicate some of the more dubious concerns we explored in part one, they are clearly multiple. This makes their distillation into a singular principle of liability extremely difficult. The greater the number of reasons we try to accommodate within a rule, the more impossible it becomes to formulate it in a way which coheres with them all.

There are only a limited number of ways of dealing with this problem. One is to abandon the search for general principles as a will-o-the wisp and deploy something akin to a structured judicial discretion instead. This enables a wide variety of reasons to be deployed across a very broad range of cases, but it can markedly reduce the predictability of outcomes. The second option is to break up the case law into a number of smaller, factual categories or ‘pockets’ in the hope that a narrower range of concerns can be identified with each category.[74] That ought in theory to give the development of coherent liability rules within each category a more realistic chance. The difficulties with this, as Professor Stapleton has explained,[75] are that the various judicial concerns we have identified tend to spill over between categories; that any increase in the coherence of liability principles within the categories can be at the expense of coherence between them; and that it leaves one with no obvious reasoning apparatus to deploy outside categories, in entirely new types of case.

English and Australian courts now grapple with the problem of multiple policy concerns in subtly different ways. England has persisted with general principles, whilst also recognising that the law must respect and develop incrementally from existing categories. By contrast, the Australian multifactoral approach purports to firmly reject both generalisation[76] and categorisation.[77] In some respects, it resembles the operation of a structured discretion, though decision- making is much more tightly constrained than that expression implies, because it entails the use of only those factors recognised in the cases and at a cautious, incremental pace.[78] Also, although the approach formally rejects categorisation (in the sense that the failure of a case to fit within an existing pocket of case law is not regarded as determinative), it is only used when a case falls outside a space in which liability rules are clearly settled.[79] In this respect, the strategy is much closer to that prevailing in England than may at first appear to be the case. Both approaches purport to leave existing rules and categories intact and to work outwards from them. The key difference is that in deciding new cases, they deal with the wide variety of policy arguments through concepts formulated at rather different levels of abstraction. The English approach works through a superstructure of higher-level principles, whereas the multi-factoral approach deploys concepts which are closer to ground level in the sense they relate more directly and individually to the various normative concerns.

Legal Principles and Social Justifications

A third feature is that some of the justifications cited by courts relate to the social effects of economic loss liabilities, rather than to concerns which are identifiably particular to the parties on the facts. This is difficult in so far as it opens up the possibility of a gap between the ends of private law and its means, in ways that have deeply troubled Professors Weinrib[80] and Cane.[81] Even if it is impossible (and inappropriate) for judges to ignore all of the effects which liabilities might have when setting the scope of tort rights, private litigation is clearly not the best mechanism for proactively implementing broad social policies such as loss-distribution, because of the informational constraints to which it is subject and the constitutional limits of the judicial role.

This is not necessarily to say that all arguments which are framed in terms of fears about the effects of tort liabilities are inconsistent with private law’s constraints and therefore inappropriate.[82] Some arguments which look ‘social’ in orientation may simply be generalised ways of stating the implications of tort liabilities for individual rights. We have seen, for example, that concerns about the effect which liabilities might have on economic competition can be understood in more than one way - one of which centres on the health of the market economy as a whole (which is largely imponderable), and the other of which expresses a legitimate concern for the rights of individual defendants to make their way freely in the world. The same ambiguity exists in relation to arguments about indeterminacy and alternative means of protection, some versions of which are more compatible with private law’s bi-partite systemic constraints than others. Unless these constraints are to be significantly relaxed, however, their necessary implication is that courts should generally avoid reference to justifications which are based on the achievement of social ends, rather than on the effects which liability rules might have for private rights.

Different Orders of Justification

Finally, the justifications we have identified are of different normative orders, in the sense that they bear in different ways and with differing degrees of force upon a defendant’s legal responsibility. Some, such as the concern to avoid undercutting express contractual allocations of risk, statutory policy and the freedoms of commercial competitors, disclose reasons which are strong enough entirely to negate the positive case for a precautionary duty. Others, however, such as the concern that duties of care might lead to ‘disproportionate’ liability, or obligations which are unduly burdensome to individuals in terms of precautionary cost, or that they might detract from a claimant’s responsibility for taking reasonable measures to protect himself, seem to describe reasons only for limiting the extent of the precautions which a defendant should be required to take, or the amount of liability which ought to be imposed on a case-by case basis.

This mixing of concerns of different orders within the duty inquiry further complicates the search for a principled approach and raises the question whether arguments of the latter sort might not be accommodated better elsewhere within the negligence inquiry, when courts consider issues of breach, causation and remoteness of damage. For example, an alternative response to the ‘ripple’ phenomenon which apparently creates such difficulties for the size and determinacy (certainty) of claims, could be to develop distinct remoteness of damage rules in cases of pure economic loss, which require something more stringent than simply that the loss be ‘foreseeable’ in a general way if a defendant is to be liable for it. This is effectively what Lord Denning proposed in Parsons (Livestock) Ltd v Uttley Ingham & Co Ltd.[83] The idea was never taken up, but the distinctions beginning to creep into Australian law between ‘first line’ economic loss claimants and others[84] (or between defendants who are merely negligent and those who are ‘reckless’ as to risk because they have greater ‘knowledge’[85] about its prospective effects) are not far removed from his idea that economic harms should have to be foreseen at higher levels of probability to in order to satisfy the remoteness inquiry.[86] Which approach one takes to the ‘indeterminacy’ issue (duty of care or remoteness of damage?) turns upon the relative importance accorded to certainty and particular justice respectively. It is generally assumed that certainty requires a response to the problem which rules out a duty of care, on the basis that this promotes more open consideration of the relevant concern and sends a very clear social message about who bears the relevant precautionary burden. The effect of this strategy, however, can be to wield a sledgehammer to crack a nut. The legal response (no duty on the part of the defendant to consider the claimant’s interests at all) becomes disproportionate to its underlying normative justification (uncertain or excessive amounts of liability). This, as we indicate further below, is something that should generally be avoided, if the law is to claim to be in touch with its underlying reasons in a sensitive and coherent way.

Accommodating concerns about ripple effects via the remoteness inquiry would ironically do no more than return us to a mode of analysis deployed more than a century and a half ago in Cattle v Stockton Waterworks.[87] In a similar vein, there seems to be space to consider the relative burdens of precaution as between claimant and defendant (i.e. both the broader version of the personal autonomy argument and the narrower interpretation of the argument about alternative means of protection) through fact-sensitive inquiries into breach, causation, contributory negligence and mitigation.[88]

(b) Methodological Ideals

Given the nature of these difficulties, how, then, should we approach the task of legal reasoning in economic loss cases? What are the ideals? Without attempting to be exhaustive, I believe there are five.

Sensitivity to the Force of Justification

Firstly, courts should accommodate the various justificatory arguments at a place within the overall negligence inquiry which accurately reflects their normative force. Otherwise, justice is only bluntly done. Whilst blunt justice is often to be found in the early stages of a legal system’s development and has the advantage of sending very powerful social messages, a mature, sophisticated system of law should constantly be increasing the sensitivity of its rules to their underlying justifications. This alignment between rules and their reasons is a vital precondition of the law’s normative coherence.

The implication of this in the current context is that probably only a small, powerful minority of the justifications we have examined ought to entirely negate a duty to take care of another’s economic interests. This minority may still apply in a good number of cases, but its membership should be less. It should comprise the concern that liabilities will yield some direct conflict with other systems via which economic dealings have been structured or regulated; and the analogous concern that tort duties may contradict commercial norms operating between economic competitors.

Respect Structural Constraints

Secondly, the justifications used by courts must respect the structural and constitutional limits of private litigation.[89] Both forms of constraint raise legitimate question marks against the use of arguments about the distribution of loss between different social groups, as opposed to the claimant and defendants’ respective individual responsibility for the harm suffered. For this reason, I have suggested, it may be preferable to understand concerns about the availability of alternative means of protection as individualistic arguments about the reasonableness of particular claimants’ behaviour on the facts, rather than as arguments about the way in which loss should be distributed between different interest-groups. Whilst ‘commercial’ purchasers of defective real estate do generally have ways of protecting themselves which consumers do not, this should not, therefore, preclude the High Court of Australia from allowing a claim in instances in which such means were in fact lacking, any more than it should necessarily mean that consumers should be entitled to claim in full, where such means were readily and reasonably available to them. The informational constraints of private law also suggest that McHugh J is right to cast the concern about legitimate competition primarily in terms of individual autonomy, rather than in broader, utilitarian terms.

Occam’s razor

Thirdly, courts should avoid unnecessary conceptual multiplication. This is the principle of Occam’s razor.[90] In the current context, it means that we should avoid approaching the duty of care question in a host of different ways. The multiplication of conceptual approaches, all of which are aimed at addressing the same issue is likely to yield duplication, inconsistency and incoherence as between cases raising similar concerns. This has proven an obvious problem in the past and remains a particular difficulty in England, where several nominally distinct approaches are taken to the duty issue, the interrelationship of which still remains unclear. If it is true that they can all be manipulated to yield the same result, [91] this clearly begs the question as to why we need more than one.

Optimal Transparency between Reasons and Concepts – The avoidance of fiction

Fourthly, courts should aim for optimal transparency in their reasoning, by which I mean that whatever approach is taken, it should leave the justifications for decisions as visible as possible. We have already touched on the difficulty of this, given that the relevant justifications are both negative and multiple, but, assuming the legitimacy of the judicial desire to translate arguments into legal concepts and principles, there is one obvious pitfall which can usefully be avoided. This is the use of fictional analogies between the situations in which duties of care are imposed by courts and relationships of contract or trust. In particular, it is unhelpful to continue to mediate the various negative justifications for restricting economic loss liability in negligence through the question of whether a claimant has ‘assumed responsibility’ for his conduct.[92] There are, of course, good explanations as to why courts use fictions, which have been explored in depth by Professor Fuller.[93] Such devices help to overcome inevitable gaps in the fabric of legal justification, which judges do not necessarily have the time to fill, as new cases arise.[94] More generally, analogies may actually be an integral part of the way in which we think. Psychologically, we tend to digest new experiences (here, new legal questions) by converting them into terms which are familiar to us.[95]

Both these observations about the attraction of fictions help to explain why a judge faced in 1963 with the unfamiliar question of when duties of care should be owed in respect of economic loss caused by negligent misstatement might choose to answer in terms of relationships ‘equivalent to contract.’[96] Such duties were, after all, owed prior to that date in relationships in which a contract or relationship of trust was present and Hedley Byrne was, it is often pointed out, very close to being a contract case. Nonetheless, even Fuller accepts that there comes a stage at which fictions must be allowed to die a dignified death.[97] Like scaffolding, they serve as a valuable temporary instrument in the construction of new structures of thinking, but can messily obstruct the view, once the building is complete.[98] There is a particular danger, he notes, that having initially used a fiction as a tool for understanding things that are unfamiliar to us, we become so fixated on it, that it becomes in our minds not merely a means for better understanding the issue, but the very issue itself.[99]

To my mind, this is what has happened in cases in England, where the ‘assumption of responsibility’ concept remains current as one determinant of duty,[100] despite several laudable judicial attempts to kill it off.[101] Now that our imposition of economic duties of care has advanced beyond the point at which anything like the contractual analogy holds water, we should, I suggest, dispense with it altogether. It is surprising that the House has declined to do so[102] now that it has another technique to hand (the Caparo approach), not least because the chief saviour of the modern assumption of responsibility approach - Lord Goff - has shown himself in the past openly hostile to misleading, ‘quasi-contractual’ analogies.[103] As Gummow J has wisely observed in the current context, [104] the spirit our times is hostile to such fictions. Aside from proving hopelessly indeterminate,[105] the assumption of responsibility concept is yielding damaging anomalies of its own. Most strikingly, it has led to the not uncommon, but clearly mistaken view that, where an assumption of responsibility is present, a duty of economic care should follow without further inquiry.[106] This was never likely to be the case, since there may be negativing justificatory concerns present on the facts of a case which call for a duty to be rejected, even where an ‘assumption of responsibility’ is present, as the old cases on advocates’ immunities demonstrate.[107] The analogy has given rise to a significant misconception: a concept, the design of which is apparently to accommodate a number of negative justificatory concerns (in particular about indeterminate liability and conflict) is now thought to provide a sufficient positive justification for a duty of care. Fuller’s warning has gone unheeded. The fiction of contract, once a means to an end, has become an end in itself with a beguiling, disruptive force of its own.

Interestingly, whilst Australian judges have expressed equal scepticism about the Caparo approach,[108] it has not been on the basis that the concepts it uses are ‘fictional.’[109] This seems right. On Fuller’s view, even the elusive concept of ‘proximity’ is unlikely to qualify as a fiction, since it expresses no proposition which is obviously counter-factual. It acts more as a shorthand, abstract way of expressing a complex idea about legal relationship.[110] In fact, the real substance of Australia’s antipathy to the Caparo method, aside from the accurate but slightly bizarre accusation that it engages courts in discussing policy issues,[111] seems to lie in the view that general concepts are unhelpful in determining concrete outcomes in particular cases, because they are too far removed from the facts and too prone to discretionary manipulation.[112] That is a different, difficult, connected matter to which we now turn.

Balance the Particular and the General - Reasons, Dispositive Rules and Abstract Principles

The fifth methodological ideal is that courts strike an appropriate balance between the development of lower-level, dispositive rules governing outcomes in particular categories of economic loss case; and higher level principles. We must not focus on either the general or the particular, but on both and creating a constructive and reflexive relationship between them, which affords to courts a degree of both certainty and flexibility.

Australia has rejected the idea of categorisation as determinative and this is clearly right in the sense that the arbitrary factual classification of a case should not dictate its outcome. The multi-factoral approach also eschews high-level abstractions (generalisations) in favour of flexible, lower-level ideas spanning the various case-groupings. These ideas are undoubtedly located closer to the operative policy concerns driving judicial decisions and in this sense much more transparent in the sense discussed in the preceding section. At the same time, however, the approach makes no attempt to express what Lord Atkin would have termed a ‘general conception’[113] of the types of relations giving rise to a duty to take care. The list of factors is self-confessedly non-exhaustive, which leaves no broader reasoning strategy available in new cases, other than the mandate that courts should proceed analogically, case by case, like ancient mariners hugging an unknown coastline,[114] or that they should address the duty of care question bluntly in its own terms in the light of a case’s ‘salient features.’

At the end of the day, perhaps this is right. It is unlikely that the method will in itself produce results which are strikingly different to those under the various English approaches. Both approaches are avowedly incremental in outlook and such (admittedly very significant) differences as exist are more likely to be the product of the different emphasis which judges in England and Australia place upon the various underlying policy concerns, than on the precise conceptual vehicle chosen to organise and express them. Nonetheless, the multi-factoral approach, more transparent as it is, tends to disregard the organising and synthesising function of higher-level principles, in just the same way that the English approach can be criticised for failing to develop concepts which are sufficiently close to judicial reasons to enable more precise predictions to be made about likely outcomes.

The ideal, in my respectful view, lies somewhere between the two approaches and if I were rashly to make a prediction, that is where I think we shall all end up, if we are not actually there already. Although factual categories should not prove normatively determinative, the practical likelihood is that lower-level liability criteria will develop within each type of economic loss case. Although factual categories mean nothing in themselves, they are initial filters for our understanding of normative concerns. Even under the Australian approach, they are accorded respect.[115] At the same time, in developing these loose factual categories and in confronting new facts, courts will inevitably reach for higher-level concepts and principles to give structure and expression to the common reasons underpinning existing categories and rules. There are already signs of a reversion to this higher-level approach in the High Court of Australia, where McHugh J, pioneer of the lower-level ‘multi-factoral’ approach, has subsequently located these factors within a broader investigation into the existence of a sufficiently ‘close and direct relationship’ of ‘neighbourhood’ between the parties.[116] With respect, it is hard to see how this differs from the formally disapproved concept of ‘proximity’ and gives credence to Kirby J’s accusation[117] that the law in Australia has now come full circle to something approximating its Atkinian point of departure. Conversely, in England, there are clear signs in Customs and Excise Commissioners, that their Lordships recognise the perils of over-abstraction and the need to supplement abstract principles with detailed discussion of lower, operative concerns. Lord Mance has thus noted that what really matters is how and by reference to what low-level factors the various abstract concepts are interpreted in practice.[118]

The approach hinted at by Lord Mance deploys higher-level principles only as a basis for organising, changing and developing lower-level rules, not as a substitute for them, which has significant advantages for reasons explained by Prof. Raz.[119] An approach which relies solely on such principles will necessarily be uncertain; whereas one which conversely relies solely on particular rules in individual cases will run into developmental difficulties when familiar territory is left behind. It will have only weak dynamic force. An approach which combines lower-level norms with higher-level principles has a reasonable chance - the best chance I believe - of accommodating the judicial need to keep an eye on both the particular and the general; and of the law to be both stable and reasonably dynamic.

According to this view, abstract principles like foreseeability of harm and proximity of relationship have a continuing and important role to play in synthesising and developing the various different factual categories and precedents - in allowing the justifications in the individual cases to be abstracted upward and then deployed reflectively downward into other cases, both old and new. This method of reasoning (illustrated crudely in figure 1 below) is not actually greatly different from the process of analogical reasoning between one type of case and another,[120] except in so far as principles enable the law’s collective reasons to be deployed more dynamically - further from existing case law - than analogical reasoning alone might allow. They also provide a common analytical framework for thinking - as Kirby J would say, a means for ‘steering the judicial mind through the task at hand,’[121] which is vital, if an ordered, consistent approach to the consideration of cases is to prevail across the field.

NEGLIGENCE

Duty of Care

Principles Foreseeability Proximity/Legal Relationship ‘Justice/Policy’

(Reason 1) (Reason 2, Reason 3, Reason 4)

upward

Categories Category 1 Category 2 Category 3 Category 4 abstraction

(Facts) to principles



Rules Rule 1 Rule 2 Rule 3 Rule 4 downward

reflection

Reasons Reason 1 Reason 1 Reason 1 Reason 1 to rules

Reason 2 Reason 3 Reason 2

Reason 4

Fig 1. Abstraction to and Normative Reflection from Higher-Level Principles

The use of higher-level abstract principles only really causes problems when we try to set them to ends for which they were not designed. Naturally enough, if we try to use them not as tools to assist in organising, creating or developing rules, but as determinate rules or ‘tests’ of liability, we will come away with an acute sense of disappointment. Students tend, I find, either to treat the Caparo approach as if it were such a test (at which point they become understandably frustrated at its lack of dispositive power), or, at the other extreme, as a fictional sham, somehow designed to pull the wool over our eyes and disguise the policy-based decisions which judges are making. It is then as if judges were engaged in some conspiracy to conceal their legislative role. Neither interpretation is true to the way the approach is used. The former misconstrues the purely organisational and developmental function of its abstract concepts, whilst the latter is generally unfair to judges. There is also an obvious tendency to assume that abstract concepts necessarily invite subjective discretions and appeal to open, moral standards.[122] That is again a misconception in the current context, as least as far as the Caparo approach goes. This is because the concepts of ‘proximity’ and ‘justice’ deployed within its framework look downward to - and derive their meaning from - existing precedents, not upward towards an uncertain, star-filled, discretionary sky. Legal principles, unlike their moral counterparts, have a grounding in the fabric of the law.

Part 3 - Conclusions and Reflections

A stable method for approaching the issue of when a duty of care is owed in respect of another’s intangible economic interests is a vital precondition to the long-term coherence of the law and it is hoped that, at the very least, this chapter has helped to develop a clearer vision of what it is that makes the task so difficult. Both England and Australia have struggled with the difficulties of defining relevant reasons for constraint in economic loss cases and in working these reasons into positive rules and principles which are sensitive to the normative force of the various individual justifications, the structural constraints of private litigation, the need for transparency in legal reasoning and the need to accommodate the inevitable ‘new case’ lying awkwardly outside existing categories and precedents.

England and Australia are currently set on their own methodological paths and it would be an overly simplistic assessment which indicated that one was right and the other wrong. The best hope for a coherent approach seems, in fact, to lie somewhere between them; in the adoption of a single reasoning strategy deploying both legal rules which are sufficiently close to the ground to provide reasonable certainty and predictability in particular cases; and higher-level abstract principles which enable courts to take a structured overall approach to the inquiry and to synthesise their experience of a wide range of individual cases in developing the law dynamically. Within this strategy, the coherence of courts’ approach can also be increased by eliminating as far as possible the use of fictions from the language of legal justification; and by unloading the duty inquiry of some its current justificatory burden.

This lightening of the justificatory load can be achieved, I have suggested, firstly by eradicating from courts’ consideration some concerns which are simply not appropriate; and secondly, by relocating others to lower levels of the negligence inquiry, where more sensitive weighting can be given to their normative force. Most clearly overdue for eradication is the argument that economic losses are not socially harmful. With respect, this is unlikely to be true and cannot in any case be in point in a system of private rights that takes itself at all seriously. Broad arguments about loss-distribution, or the effects of economic duties on the health or the market economy are probably also suspect, if for no other reason than that they exceed the informational constraints of the private litigation system. By contrast, a number of justifications are up for eviction from the duty of care inquiry and relocation elsewhere. These are: concerns about the possible precautionary burdens to defendants associated with economic duties of care (to be dealt with via breach); concerns about the moral limits of responsibility for losses caused by ‘ripple effects’ (remoteness); arguments about delay and latent loss (limitation), fears of floods of claims causing administrative chaos (strike-out, summary judgment, settlement) and arguments about alternative means of protection open to a claimant (causation, remoteness, contributory negligence, mitigation). This strategy of relocation would leave the duty inquiry free to accommodate two of the most powerful arguments - the importance of an individual’s freedom to compete (narrowly interpreted) and, connectedly, the importance of ensuring that economic duties do not conflict with other legal mechanisms or norms for dealing with economic interests. Other, equally powerful concerns may need to be added to the list in particular cases, but this reorganisation of the various justificatory arguments would, in my view, still lead to a more coherent approach to the duty question, as well, vitally, to the more sensitive dispensation of justice.

No doubt there will be some who strongly object to this ‘downgrading’ of the majority of the arguments, on the basis that most disputes about pure economic loss will then dissolve into detailed ‘factual’ inquiries. That, it might be argued, will simply lead to more uncertainty not less, because relevant policy arguments will get buried out of sight instead of being openly aired. It will also reduce the strength of the social message, evident in Lord Oliver’s bold statement cited at outset, that causing pure financial loss is not really ‘wrong.’ There is no reason, however, to think that any other stage of the negligence inquiry is any less charged with ‘policy’ issues, or that they cannot openly be discussed as such. Moreover, if one focuses on any of the recent cases on pure economic loss in Australia, it soon becomes evident that courts’ approach to the duty issue is already very particularised, entailing detailed consideration of the circumstances in every case. This in itself says something about the appropriateness of the traditional message about the careless causing of economic loss not being ‘wrongful’ in today’s world. It is not so much, these days, that carelessly doing harm to economic interests fails to ignite our moral interest, but that we need to assume a balanced, co-ordinated approach to their protection, which takes account both of a claimant’s own capacity to protect these interests; and of other ways in which economic relationships are (for very good reason) regulated and protected.

* Associate Professor, University of Queensland. I am greatly indebted to Jenny Steele and Mark Lunney for their helpful insights on an earlier draft. All views and defects remain entirely my own.

1 Caparo Industries v Dickman (HL) [1990] UKHL 2; [1990] 2 AC 605 (‘Caparo’).

[2] This approach was affirmed most recently in Commissioners of Customs and Excise v Barclays Bank plc [2006] UKHL 28 (‘Barclays’), but has been formulated in a number of very different ways historically. Two models now dominate: a ‘narrow’ one (‘relationship equivalent to contract’ entailing direct communication or dealing between the parties) Hedley Byrne & Co Ltd v Heller & Partners Ltd [1963] UKHL 4; [1964] AC 465 529, per Lord Devlin; Williams v Natural Life Health Foods Ltd (HL) [1998] UKHL 17; [1998] 1 WLR 830; and a ‘wide’ one (‘undertaking of a task’ in circumstances in which another is generally reliant on the defendant for protection) Henderson v Merrett Syndicates Ltd (HL) [1994] UKHL 5; [1995] 2 AC 145; White v Jones [1995] UKHL 5; [1995] 2 AC 207, 273, per Lord Browne-Wilkinson.

[3] Caparo [1990] UKHL 2; [1990] 2 AC 605.

[4] Bank of Credit and Commerce International (Overseas) Ltd. (In Liquidation) v Price Waterhouse (No 2) [1998] EWHC 333; [1998] PNLR. 564, 586, per Sir Brian Neill; Dean v Allin & Watts [2001] EWCA Civ 758 at [33], [2001] EWCA Civ 758; [2001] 2 Lloyd's Rep 249, 259, per Lightman J; Barclays [2006] UKHL 28, per Lord Mance. In Barclays, both Lords Hoffmann and Mance suggest that the second approach displaces the third in cases involving the negligent provision of advice.

[5] Barclays [2006] UKHL 28 at [4] per Lord Bingham, [93] per Lord Mance; [2004] EWCA Civ 1555 at [23, 40] per Longmore LJ.

[6] Various approaches have been tried, including: (i) knowledge or means of knowledge that the plaintiff individually, and not merely as a member of an unascertained class, will suffer loss - Caltex [1976] HCA 65; (1976) 136 CLR 529, per Gibbs J (ii) proximity: Sutherland Shire Council v Heyman [1985] 157 CLR 549 (‘Heyman’), per Deane J; Bryan v Maloney (1995) 182 CLR 609 (iii) general reliance: Heyman [1985] 157 CLR 549, per Mason J.

[7] Perre v Apand [1999] HCA 3 (‘Perre’), per McHugh J; Woolcock Street Investments Pty Ltd v CDG Pty Ltd [2004] HCA 16 (‘Woolcock’), per McHugh J. This approach has since been deployed in a case involving physical harm and omissions in Crimmins v Stevedoring Industry Finance Committee [1999] HCA 59, but recent cases on psychiatric harm revert to the concept of ‘reasonable foreseeability’ as the determinant of duty in cases of that type: Annetts v Australian Stations Pty Ltd [2002] HCA 35; (2002) 211 CLR 317; Gifford v Strang Patrick Stevedoring Pty Ltd [2003] HCA 33. Note that the multi-factoral approach is confined to new cases and is not intended to undermine existing precedents.

[8] Ibid.

[9] Perre [1999] HCA 3 at [105] per Mc Hugh J.

[10] Ibid at [79-82]; Sullivan v Moody [2001] HCA 59 at [49] per Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ.

[11] Witting, ‘The three Stage Test Abandoned in Australia- or Not’ (2002) 118 LQR 214, 218. It is assumed for these purposes that the multifactoral and salient features approaches are the same, which is probably an oversimplification. They are nonetheless generally regarded as consistent.

[12] On the need for a stable analytical method, see Graham Barclay Oysters Pty Ltd v Bryan and Others [2002] HCA 54 at [211] per Kirby J.

[13] Murphy v Brentwood [1991] UKHL 2; [1991] 1 AC 398, 485, per Lord Oliver; Gordley, ‘The rule against recovery in negligence for pure economic loss: an historical accident ?’ ch 2 in Bussani, Palmer (eds) Pure Economic Loss in Europe (Cambridge, 2003), 25; Cane, ‘The blight of economic loss: is there life after Perre v Apand ?’ (2000) Torts Law Journal 246. For a seminal analysis of some of these contingent ‘policy’ issues, see Stapleton, ‘Duty of Care and Economic Loss: A Wider Agenda’ (1991) 107 LQR 249.

[14] Caltex Oil (Australia) Pty Ltd v The Dredge Willemstad [1976] HCA 65; [1976] 136 CLR 529 (‘Caltex’), 570, per Stephen J (there is no difference in ‘morality’ or ‘justice’ between pure and consequential economic loss); Perre [1999] HCA 36 at [123] per McHugh J (the law of negligence protects economic interests ‘in its quest for corrective justice’); Arthur JS Hall & Co v Simons [2002] 1 AC 615, 681, per Lord Steyn (an advocate’s liability for pure financial loss follows the ‘basic premise that there should be a remedy for a wrong’); Woolcock [2004] HCA 16 at [106] per McHugh J (liability of builders for defective premises ‘would advance the cause of corrective justice, one of the rationales of the law of negligence’).

[15] Weir, ‘Abstraction in the Law of Torts - Economic Loss’ (1974) City of London L Rev 15; Witting, ‘Distinguishing between property damage and pure economic loss in negligence: a personality thesis’ (2001) LS 481; Benson, ‘The Basis for Excluding Liability for Economic Loss in Tort’ in Owen, Philosophical Foundations of Tort Law (Oxford, 1996). Benson’s argument (carelessly causing economic loss is legally wrongful only where it violates some prior exclusive right) finds some resonance in the speech of Gaudron J in Hill v Van Erp (1997) 188 CLR 159 (‘Hill’) 198 (who indicates that it is easier to establish liability where there is interference with a claimant’s ‘precise legal right’) but appears, with respect, to beg the question which the law of negligence has to decide. As Witting indicates (at 501), it also runs into difficulties in explaining misstatement cases.

[16] Weir, above n 15; Cane, above, n 13.

[17] Bussani and Palmer, above, n 13, 21-23 indicate that arguments about ‘relative priorities’ work on the basis of a ‘silent premise’ that legal resources are insufficient to accommodate them all. The argument then effectively dissolves into a pragmatic concern about the effect of liability on judicial resources, akin to that sometimes encountered in the context of arguments about ‘indeterminacy,’ below.

[18] Weir, Witting, above n 15.

[19] [1991] UKHL 2; [1991] 1 AC 398, 487.

[20] Ibid. His Lordship rejected any moral distinction at 485.

[21] SCM Ltd v W J Whittall [1971] 1 QB 337, 344; Spartan Steel & Alloys v Martin [1972] EWCA Civ 3; [1973] QB 27, 36.

[22] [1999] HCA 36 at [72]; Feldthusen, ‘Pure Economic Loss in the High Court of Australia: Reinventing the Square Wheel?’ (2000) 8 Tort Law Review 33, 50-51; Feldthusen and Palmer, ‘Economic Loss and the Supreme Court of Canada: An Economic Critique of Norsk Steamship and Bird Construction,’ (1995) 74 Can Bar Rev 427.

[23]The ‘zero social loss’ hypothesis is questionable. See Rizzo, ‘The Economic Loss Problem: A comment on Bishop’ (1982) 2 OJLS 197; Witting, above n 15, 495-6.

[24] In either case, the loss of production is readily recoverable, as ‘parasitic’ upon injury to property or person: Spartan Steel & Alloys v Martin [1972] EWCA Civ 3; [1973] QB 27.

[25] Heyman [1985] HCA 41; (1985) 157 CLR 424, 503, per Deane J; Bryan v Maloney, (1995) 182 CLR 609, 618 per Mason CH, Deane, Gaudron JJ, 632 per Brennan J; Hill (1997) 188 CLR 159, 175 per Dawson J, 193 per Gaudron J; 211 per McHugh J; Esanda Finance Corporation Ltd v Peat Marwick Hungerfords (1997) 188 CLR 241, 254 per Dawson J; Perre [1999] HCA 36 at [32-33] per Gaudron J, [114] per McHugh J, [182] per Gummow J; Dovuro Pty Ltd v Wilkins and Others [2000] FCA 1902 at [30-31] per Brennan J; Woolcock [2004] HCA 16 at [21] per Gleeson CJ, Gummow , Hayne and Heydon JJ, at [78] per McHugh J.

[26] Ibid. The same interpretation appears to be taken by Gaudron J at [32-33]. See also Woolcock [2004] HCA 16 at [78]; Dovuro Pty Ltd v Wilkins and Others [2000] FCA 1902, at [30-31] per Brennan J (rev’d on other grounds [2003] HCA 51).

[27] This appears to be Cane’s interpretation, above, n 13.

[28] In Perre [1999] HCA 36, McHugh J shifts between interpretations, but his reference at [101] to the need to ‘avoid unreasonable burdens’ on defendants clearly encompasses the broader argument. He affirms this in Woolcock [2004] HCA 16 at [78].

[29] Note however that the law already makes incursions into the competitive domain via the ‘economic torts.’ Parties to contractual negotiations also owe duties of care in respect of information given, both at common law and (commonly now) under statute. The key question is thus what constitutes ‘legitimate’ competition according to the commercial norm. In Hill (1997) 188 CLR 159, 198, Gaudron J suggests that the legality of behaviour is key (it is never ‘legitimate’ pursuit of self interest to interfere with the (‘precise’) ‘legal rights’ of others). In Perre [1999] HCA 36 at [116] Mc Hugh J regards legality as relevant but inconclusive, the ultimate question being whether behaviour constitutes ‘sharp or ruthless practice’ going beyond community tolerance.

[30] Bolton v Stone (HL) [1951] UKHL 2; [1951] AC 850.

[31] Bryan v Maloney (1995) 182 CLR 609, 632, per Brennan J. See also Holmes J, ‘The Path of the Law’ 1897 10 Harvard law Review, 457, 466 (‘the public good [my emphasis] is supposed to be best preserved by free competition’).

[32] Bryan v Maloney (1995) 182 CLR 609; Woolcock [2004] HCA 16.

[33] Hill (1997) 188 CLR 159; White v Jones (HL) [1995] UKHL 5; [1995] 2 AC 207.

[34] Perre [1999] HCA 36

[35] Hill (1997) 188 CLR 159 (contractual duty of care to testator); Perre [1999] HCA 36 (contractual duty of care to potato purchasers); Woolcock [2004] HCA 16 (tort duty already owed to purchasers regarding person and property); Dovuro Pty Ltd v Wilkins and Others [2000] FCA 1902 (rev’d on other grounds [2003] HCA 51) (tort duty regarding seed user’s property interests); Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [962] (duties owed to third party employees regarding physical safety).

[36] (1931) 174 NE 441, 444.

[37] Bryan v Maloney (1995) 182 CLR 609, 626, per Mason CJ, Deane and Gaudron JJ; Woolcock [2004] HCA 16 at [107] per McHugh J, [166], [180-181] per Kirby J, [219] per Callinan J. The concern is most often voiced in defective premises cases, but was also mentioned by McHugh J in Hill (1997) 188 CLR 159, 216, on the basis that it may be a long time before a solicitor’s advisory errors in respect of a will are discovered.

[38] Esanda Finance Corporation Ltd v Peat Marwick Hungerfords (1997) 188 CLR 241, 288, per McHugh J; Perre [1999] HCA 3 at [169] per Gummow J (obiter); Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [778], [950-955] per Gillard J (obiter). Woolcock [2004] HCA 16 at [97] per McHugh J (obiter).

[39] Leigh and Sillivan Shipping v The Aliakmon (HL) [1985] UKHL 10; [1986] AC 785 (‘The Aliakmon’), 816-7, per Lord Brandon; Barclays [2006] UKHL 28, per Lord Walker (liability of banks would also entail liability of other non-commercial parties with notice of freezing injunctions).

[40] Caltex [1976] HCA 65; [1976] 136 CLR 529, 551 per Gibbs J, 591 per Mason J; San Sebastian Pty Ltd v Minister Administering the Environmental Planning Act 1979 (1986) 162 CLR 349, 354 per Gibbs, Mason, Wilson, Dawson JJ; Perre [1999] HCA 3 at [169] per Gummow J; McFarlane v Tayside Health Board (HL) [1999] UKHL 50; [2000] 2 AC 59, 91 per Lord Hope, 106 per Lord Clyde; Barclays [2006] UKHL 28 at [111] per Lord Mance; Rabin, ‘Tort recovery for Negligently Inflicted Economic Loss: A Reassessment’ (1985) 37 Stanford Law Review 1513,1534. For criticism, see Stapleton, ‘Duty of Care Factors: A Selection from the Judicial Menus’ in Cane and Stapleton (eds) The Law of Obligations: Essays in Celebration of John Fleming (Oxford, 1998), 59, 65-6.

[41] This argument is often voiced in cases involving advisory services (eg Hill (1997) 188 CLR 159, 214, per McHugh J (dissenting), but is not so confined: Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [777] per Gillard J. See further Bishop, ‘Negligent Misrepresentation through Economists’ Eyes’ (1980) 96 LQR 360.

[42] McMullin and Another v ICI Australia Operations Pty Ltd and Others [1997] FCA 541; [1997] 72 FCR 1, 76 per Wilcox J; Hill (1997) 188 CLR 159, 180 per Dawson J, 193 per Gaudron J, 215 per McHugh J (dissenting); Perre [1999] HCA 3 at [50], [107-108] per McHugh J; Dovuro Pty Ltd v Wilkins and Others [2000] FCA 1902 at [29], per Branson J; Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [906] per Gillard J; Goodwill v. British Pregnancy Advisory Service [1996] 1 WLR 1397.

[43] Hill (1997) 188 CLR 159, 180 per Dawson J (damage must not be ‘at large’), 193 per Gaudron J, [215-216] per McHugh J (dissenting). Esanda Finance Corporation Ltd v Peat Marwick Hungerfords (1997) 188 CLR 241, 285, per McHugh J; Perre [1999] HCA 3 at [107-108] per McHugh J; Woolcock [2004] HCA 16 at [219] per Callinan J. See contra, Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [906], [920] per Gillard J. For an approach focusing on uncertainty about claimant class and type (but not exact amount) of loss, see Beach, ‘Indeterminacy: The Uncertainty Principle of Negligence’ (2005) 13 Torts Law Journal 1.

[44] Woolcock [2004] HCA 16 at [180-181] per Kirby J (dissenting); Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27, at [906] per Gillard J.

[45] It does not apply in cases of property damage. Introducing it here therefore reintroduces the same moral distinction between property interests and economic interests which judges have generally rejected.

[46] Perre [1999] HCA 3 at [107-108].

[47] Caltex [1976] HCA 65; [1976] 136 CLR 529, 580-81 per Stephen J; Perre [1999] HCA 3 at [130] per McHugh J. See, by contrast, Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [1093-1110] per Gillard J (insurability of a claimant’s loss is relevant to the question of whether she had an alternative means of protection).

[48] Interestingly, English courts have recently begun to make open reference to arguments of this type at the highest appellate level: see eg Barclays [2006] UKHL 28, per Lord Mance, discussed further below. For the view that the role of insurance has been drastically underplayed by judges and academics alike (albeit in the context of personal injury litigation), see Lewis, ‘How important are Insurers in Compensating Claims for Personal Injury in the UK’ (2006) 31 The Geneva Papers 323.

[49] Hill (1997) 188 CLR 159, 184, per Dawson J; Perre [1999] HCA 3 at [5] per Gleeson CJ, Woolcock [2004] HCA 16 at [102], per McHugh J. For the same concern expressed outside the economic loss context, see Sullivan v Moody [2001] HCA 59, at [50] per Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ.

[50] Fleming, ‘Tort in a Contractual Matrix,’ 1995 (3) Tort Law Review 12; Bryan v Maloney, (1995) 182 CLR 609, 622 per Mason CJ, Deane and Gaudron JJ (obiter); Johnson Tiles Pty Ltd v Esso Australia Ltd [2003] VSC 27 at [1144-1145]; Simaan v Pilkington Glass [1988] UKHL 4; [1988] QB 758; Henderson v Merrett Syndicates Ltd (HL) [1994] UKHL 5; [1995] 2 AC 145.

[51] The Aliakmon [1985] UKHL 10; [1986] AC 785 (Hague Convention giving rise to the Hague Rules); Caparo [1990] UKHL 2; [1990] 2 AC 605 (Companies Act); Woolcock [2004] HCA 16 at [102-105] per McHugh J (Statutes of Limitation); Barclays [2006] UKHL 28 (Civil Procedure Rules and Contempt Proceedings). This concern is most obviously relevant where the defendant is a public body imbued with statutory powers or duties: Sullivan v Moody (2001) 207 CLR 562.

[52] Clarke v Bruce Lance [1988] 1 WLR 881.

[53] Williams v Natural Life Health Foods Ltd (HL) [1998] UKHL 17; [1998] 1 WLR 830.

[54] D & F Estates v Church Commissioners (HL) [1988] UKHL 4; 1989 AC 177, 210 per Lord Bridge; Murphy v Brentwood (HL) [1991] UKHL 2; [1991] 1 AC 398, 457 per Lord Mackay, 472 per Lord Keith, 492 per Lord Oliver, 498 per Lord Jauncey; Bryan v Maloney (1995) 182 CLR 609, 644 per Brennan J (dissenting).

[55] The same constraints operate upon the recent, rapid expansion of the law of restitution, which constitutes another device for the protection of economic interests.

[56] The Aliakmon [1985] UKHL 10; [1985] QB 350, 397-8 per Lord Goff (although note that this approach was rejected in the House of Lords [1985] UKHL 10; [1986] AC 785, because of the complexity of the relevant contractual provisions, which made synthesis of the duties apparently very difficult); Henderson v Merrett Syndicates Ltd (HL) [1994] UKHL 5; [1995] 2 AC 145, 206 per Lord Browne-Wilkinson.

[57] Perre [1999] HCA 3 at [120-121] per McHugh J; Woolcock [2004] HCA 16 at [85] per McHugh J.

[58] Ibid.

[59] White v Jones (HL) [1995] UKHL 5; [1995] 2 AC 207, 262, per Lord Goff (rectification of inter vivos transactions). Note, however, that in Walker v Geo Medlicott (Eng CA) [1998] EWCA Civ 1806; [1999] 1 WLR 727 Mummery LJ accepted a duty as having been correctly conceded (obiter) even though rectification might have been available. Its availability was regarded as relevant instead to the mitigation of loss. Note that generally there is no obligation to mitigate by engaging in speculative litigation: London and SE Building Society v Stone [1983] 1 WLR 1242.

[60] X (minors) v Bedfordshire County Council (HL) [1995] UKHL 9; [1995] 2 AC 633. This is questionable, since such remedies ordinarily provide no compensation for harm done: Phelps v Hillingdon LBC (HL) [2000] UKHL 47; [2001] 2 AC 619, 672, per Lord Clyde. Such remedies may however be intended to constitute an ‘exclusive regime’ in some cases, giving rise to the conflict argument, above.

[61] Esanda Finance Corporation Ltd v Peat Marwick Hungerfords (1997) 188 CLR 241.

[62] Woolcock [2004] HCA 16 at [32].

[63] Ibid at [80] per McHugh J.

[64] Perre [1999] HCA 3 at [130]

[65] Ibid at [212],[224]

[66] Transco plc v Stockport Metropolitan Borough Council [2003] UKHL 61 at [46], [49].

[67] (1868) LR 3 HL 330.

[68] [2006] UKHL 28.

[69] Ibid at [102].

[70] Anns v Merton London Borough Council [1977] UKHL 4; [1978] AC 728, 751-52.

[71] The main problem stemmed from the idea of ‘prima facie duty’ arising at stage one, which seems (perhaps mistakenly) to have been interpreted as reversing the onus in new categories of case, thereby leading to an overly expansive approach.

[72] See eg Caltex [1976] HCA 65; (1976) 136 CLR 529, 567, per Steven J. In Perre [1999] HCA 3 McHugh J suggests at [95] that the new multifactoral approach actually deploys the relevant policy reasons as legal principles, but whilst the various factors clearly come much closer to expressing judicial reasons, some are still opaque translations of them: ‘knowledge of risk’ and ‘vulnerability’ stand out in this regard. Note that the view that the various policy considerations are legal principles is rejected by Kirby J at [284].

[73] Ibid.

[74] Feldthusen, Economic Negligence (3rd ed, Carswell, Toronto, 1994), ch 1; ‘Pure economic loss in the High Court of Australia: Reinventing the Square Wheel ?’ (2000) Tort Law Review 33. This approach has the approval of the Supreme Court of Canada.

[75] Above, n 13.

[76] Perre [1999] HCA 3, at [81] per McHugh J.

[77] Ibid at [75]. This must be carefully qualified. All that appears to be meant is that the duty issue will not be determined purely by asking whether a case falls within an existing category or can be seen as a viable extension of one. It does not discard categories completely. See further the text, below.

[78] Ibid at [404], per Callinan J

[79] Ibid at [94], per McHugh J.

[80] For Weinrib, tort law as an instantiation of corrective justice can only coherently sustain arguments based on private, Kantian rights. All instrumentalist justifications are suspect in so far as they fail to respect the correlativity of the parties’ legal relationship in private law disputes. For a classic statement, see Weinrib, The Idea of Private Law (Harvard, 1995), esp ch 5.

[81] Cane, ‘Distributive Justice and Tort Law’ (2001) 4 NZLR 401, 418-9; ‘Corrective Justice and Correlativity in Private Law’ (1996) 16 OJLS 471. Cane’s concerns are different to Weinrib’s. His objection is not to the use of distributive criteria in defining tort rights per se, but to the use of those which are inconsistent with the structural limitations of private law.

[82] In ‘The Disintegration of Duty’ in M. Stuart Madden (ed) Exploring Tort Law (2005) 143, 176-86, Weinrib distinguishes between two different types of ‘policy’ reasoning, only one of which is inconsistent with his view of corrective justice. This is reasoning which refers to ‘independently desirable goals’ external to the parties’ correlative relationship.

[83] [1977] EWCA Civ 13; [1978] QB 791, 802-803.

[84] Perre [1999] HCA 3 at [112] per Mc Hugh J; Fortuna Seafoods Pty Ltd v The Ship ‘The Eternal Wind’ [2005] QSC 4.

[85] Perre [1999] HCA 3 at [10], [13] per Gaudron J, [47], [67], [104], [111], [131-32] per McHugh J, [325] per Hayne J.

[86] Cane, above, n 13 identifies the problems of ‘indeterminacy’ with a similar idea of ‘causal remoteness,’ but relates this idea to remote claimants, not damage. This appears to be his justification for tackling the issue via the duty of care inquiry. There is clear support for this strategy in psychiatric damage cases (where similar issues arise), but it may be logically questionable. It is losses, not claimants that are ‘caused’ by negligent conduct, so that if causal remoteness is the key to cases involving ‘ripple effects,’ it seems logical to assume that it is the remoteness of the effects themselves, not the remote nature of the relationship with any given claimant that is the key concern.

[87] (1874-75) LR 10 QB 453. Note that in many ‘liberal’ European systems, these problems are accommodated via the ‘ordinary requirements’ of conduct, fault, damage and causation, with particular emphasis on the last of these. See Bussani and Palmer, above n 13, pp. 205-6.

[88] On alternative legal remedies and mitigation, see Walker v Geo Medlicott (Eng CA) [1998] EWCA Civ 1806; [1999] 1 WLR 727.

[89] On structural constraints, see Cane, above, n 81.

[90] Russell, History of Western Philosophy, 2nd ed, (London 1961) pp 462-463; Barker, ‘Wielding Occam’s Razor: Pruning Strategies for Economic Loss’ (2006) 26 OJLS 289.

[91] Above, n 4.

[92] Barker, ‘Unreliable Assumptions in the Modern Law of Negligence’ (1993) 109 LQR 461. See, contra, Feldthusen, ‘Liability for Pure Economic Loss: yes, but Why ?’ [1999] UWALawRw 4; (1999) 28 UWALR 84 (advocating the concept as a ‘core justification’ in some categories of case). In more recent defences of this concept, judges have claimed that the concept entails no fiction, because it refers only to the voluntary doing of a task. But it is then hard to see how the idea can usefully discriminate between cases of positive misfeasance in which recovery should be allowed and cases in which it shouldn’t.

[93] Fuller, Legal Fictions (Stanford, 1967), esp ch 2.

[94] Ibid, 65.

[95] Ibid, ch 3.

[96] Hedley Byrne & Co Ltd v Heller & Partners Ltd [1963] UKHL 4; [1964] AC 465, 529 per Lord Devlin.

[97] Above n 93, 14-23. Whether they live or die is ultimately based on whether they can be usefully redefined or become overly strained (22).

[98] Ibid 70, citing Gray, Nature and Sources of the Law (2nd ed, 1921), 35.

[99] Ibid 118-121.This is the ‘original sin’ of human reasoning- a failure to drop fictions out of the ‘final reckoning’ such that they become masters of reasoning, not its servants.

[100] Above, n 2.

[101] Smith v Eric S Bush, Harris v Wyre Forest District Council (HL) [1990] UKHL 1; [1990] 1 AC 831, 862 per Lord Griffiths; Caparo [1990] UKHL 2; [1990] 2 AC 605, 628 per Lord Roskill, 637 per Lord Oliver.

[102] See, most recently, Barclays [2006] UKHL 28. The House unanimously affirmed the approach, at least in cases involving the negligent provision of advice, seeking to strengthen and narrow its meaning to its original form.

[103] Goff and Jones, The Law of Restitution (5th ed, London, 1998) 5-11.

[104] Pyrenees Shire Council v Day [1998] HCA 3 at [163].

[105] Barker, above, n 92.

[106] Henderson v. Merrett Syndicates Ltd. (No 1) [1994] UKHL 5; [1995] 2 AC 145, 181 per Lord Goff. See also Barclays [2006] UKHL 28, per Lord Hoffman.

[107] See Rondel v Worsley [1969] 1 AC 191 (since overruled in England in Arthur JS Hall & Co v Simons [2002] 1 AC 615). Similarly, a lawyer ‘undertaking’ an advisory task may also owe no duty by virtue of the possibility of a conflict between that duty and his duty to his client: Clarke v Bruce Lance [1988] 1 WLR 881.

[108] Above n 10.

[109] By contrast, this accusation is made about ‘general reliance’ in Pyrenees Shire Council v Day [1998] HCA 3 at [163] per Gummow J.

[110] Fuller, above n 93, 28-9.

[111] Sullivan v Moody [2001] HCA 59 at [49], per Gleeson CJ, Gaudron, McHugh, Hayne and Callinan JJ

[112] Ibid.

[113] Donoghue v Stevenson [1932] AC 562, 580.

[114] Perre [1999] HCA 3 [at 93] per McHugh J.

[115] This stems from the fact that the multifactoral approach is primarily designed to assist in dealing with new cases, not to undermine existing categories and rules.

[116] Graham Barclay Oysters P/L v Ryan [2002] HCA 54 at [99].

[117] Ibid at [244].

[118] Barclays [2006] UKHL 28, at [83].

[119] Raz, ‘Legal Principles and the Limits of Law,’ (1972) 81 Yale Law Journal 823, 842.

[120] Sherwin ‘Restitution and Equity: An Analysis of the Principle of Unjust Enrichment’ 79 Tex L Rev 2083 at 2110 (2001). Sherwin expresses a preference for analogical reasoning precisely because it is likely to lead to a slower pace of legal development, but this may be at the expense of guidance in completely new case types, located far from existing precedents.

[121] Perre [1999] HCA 3 at [283].

[122] Ibid at [82] per McHugh J.