Home
| Databases
| WorldLII
| Search
| Feedback
Sydney Law Review |
Liability of Educational Providers to Victims of Abuse:
A Comparison and Critique
Anthony Gray[∗]
Abstract
The principle of vicarious liability is, to some extent, incoherent. It is indisputable that the case law has moved well beyond the original confines of the doctrine — the basis of its imposition having, to some extent, undercut by development elsewhere in tort law, and its rationale continuing to be subject to conjecture and disagreement. This article seeks to improve the situation by suggesting that the law of vicarious liability should be reconceptualised as having its basis in the law of agency. It does so in the context of the liability of educational providers to victims of abuse.
The Royal Commission into Institutional Responses to Child Sexual Abuse (‘Royal Commission’) continues to hold hearings into alleged abuse within institutions, primarily educational institutions. The scale of the alleged abuse is staggering. The Royal Commission’s Redress and Civil Litigation Report estimates a possible 60 000 survivors of institutional sexual abuse thus far.[1] The Commission’s hearings and findings will have flow-on effects on many facets of the legal system. It is expected that there may be an increase in civil claims against educational providers for alleged abuse, as well as substantial reform to relevant legislation.[2]
Several cases have recently considered the civil liability of educational providers for abuse, committed by their employees, of victims under their care (usually children), including the recent High Court of Australia decision Prince Alfred College Inc v ADC.[3] This sad situation raises many important and difficult legal issues, primarily: the question of the extent to which the educational provider is vicariously liable for the abuser’s actions; the extent to which the educational provider may be deemed directly negligent for what has occurred; and, in some cases, questions of limitation periods, given that the survivor of the abuse may not seek legal redress until long after the events. It goes without saying that, quite apart from any legal issues, survivors of sexual abuse require society’s deepest compassion and caring for what they have suffered, the symptoms and consequences of which may endure through their lives. As a society, we must do what we can to assist in the healing process, and it is recognised that the legal system can play its part in this, as part of a much broader response.[4] However, in and of itself, the fact a person has suffered abuse does not (and should not) automatically translate to an actionable legal claim against the institution who engaged the abuser.[5]
The extent to which an educational provider may be vicariously liable for the actions of their employee is part of a bigger issue of the future of the principle of vicarious liability in the law. Courts in the modern era continue to struggle to articulate a rational basis for the doctrine, tests of liability remain very difficult to apply in particular cases, and fine distinctions are made that may not reflect the realities of current employment settings. The likely increase in the number of claims in this area will place in the spotlight the appropriateness of current legal principle.
This article surveys recent major developments in the law on vicarious liability. The main focus of the discussion will be the question of the liability of educational providers for abuse. However, these questions cannot be answered in isolation from the more general principles of vicarious liability and other liability issues, including limitation periods.[6] Thus, some cases that do not directly involve questions of educational institutions and abuse will be discussed. Part II of this article summarises the existing law in three jurisdictions: Canada, the United Kingdom (‘UK’) and Australia, including the recent High Court of Australia decision in Prince Alfred College.[7] A comparison among these jurisdictions is appropriate given the common law tradition they share, and because superior courts of each jurisdiction regularly refer to, and often adopt, principles developed from the others in this area of law. Part III critiques existing theories of vicarious liability.
Part IV explores the potential of agency principles to rationalise principle in this area. This is important because, while historically vicarious liability had an agency basis, modern case law tends not to use this concept. The thesis of this article is that vicarious liability is best sourced in principles of agency, and the law should return to it, and its limits. Concededly, the High Court in Prince Alfred College did not utilise agency principles. However, over the long term, the basis of vicarious liability has changed greatly. Furthermore, in tort law more broadly, concepts have undergone rapid transformation and refinement, for example strict liability[8] and liability of public authorities for non-feasance.[9] In this light, it is not too late for reconceptualisation of the law of vicarious liability. In saying this, it is acknowledged that the legal principle of vicarious liability is just one of the legal principles affecting the question of financial compensation for survivors of institutional sexual abuse. Other legal rules, for instance statutes of limitation and the so-called ‘Ellis defence’,[10] have made it difficult in practice for survivors of sexual abuse to obtain legal redress. There has been both successful[11] and unsuccessful[12] legislative reform in this area. An outcome of the Royal Commission may well be further legal reforms, in substantive and procedural law, to facilitate compensation claims by survivors of institutional sexual abuse.[13] Further, the full force of negligence principles should be applied directly against institutions that do not take proper care in the selection or supervision of staff members. While in the past, institutions may have argued ignorance about possible child sexual abuse, or difficulty in obtaining information about a person’s past record, no one today can claim to be unaware of the risk of abuse, or legitimately complain of difficulties in researching an individual’s past.[14]
In other words, there have been legitimate arguments in the past that vicarious liability must be robustly applied to institutions because otherwise survivors of such abuse may be left without a remedy.[15] However, given robust application of negligence principles, developments in accessibility of information, past recent reforms and possible future reform due to the Royal Commission’s work, these concerns may be lessened. The article will now briefly consider the history and rationale of vicarious liability, examine the authorities, and critique existing rationales, before suggesting a reconception of this area of law on the basis of agency principles. It will be argued that given changes in the law of negligence, the rationale for imposing vicarious liability has (largely) evaporated, even if this is little recognised in the cases.
There can at least be agreement about the history of the principle of vicarious liability. It was created in a very different societal context than that pertaining today. It applied typically to a head of household, where most businesses were small, familyrun enterprises.[16] The head would be liable for the actions of members of that household, whether family members or ‘serfs’.
There was a move to limit the head of household’s liability in this context. It was decided that the head of household was only liable for actions of household members that they had expressly consented to or commanded (the ‘consent and command theory’), excepting fire. This theory applied initially to criminal cases, but later extended to civil liability. There is evidence of this ‘consent and command’ theory in statute[17] and case law[18] in medieval and Middle Age England.[19]
This theory was workable when businesses were small, household-centred operations. As the size and complexity of business grew, and as owners of businesses grew remote from day-to-day business operations, the theory became redundant.
A theory emerged that an employer was liable for things they had ‘impliedly commanded’, things analogous or closely related to things they had expressly commanded.[20] It was a short step from this theory of implied command to concepts of ‘scope of employment’: that an employer was liable for what an employee did acting within the scope of what their employment.[21] Relatedly, an employer was not vicariously liable for employee actions committed while on a ‘frolic’.[22]
At one time, an employer’s liability was limited to actions of their employee committed for the employer’s benefit, because this was one of the rationales given for why an employer was liable for their employee’s actions.[23] It continues to be reflected in relevant American law.[24] However, this requirement was abandoned in England in 1912.[25] This development was critical, and it will be returned to below. Limits of the doctrine and its ‘implied command’ basis were tested when it was sought to make an employer liable for actions of an employee that had been specifically prohibited by the employer, an attempt that proved successful.[26] They were also tested when it was sought to make an employer liable for deliberate criminal activity by an employee, as opposed to mere negligence. Possible liability in such cases was also admitted,[27] making the implied command theory no longer tenable.
To some extent, the law of vicarious liability was developed at a time when legal principles such as the doctrine of common employment could otherwise bar recovery against an employer,[28] and prior to general acceptance of fault-based liability under the neighbour principle, which liberalised recognition of duty of care responsibilities.[29] Today it sits precariously within the law of tort;[30] at odds with the explosive growth of fault-based negligence liability, the High Court of Australia’s rejection of longstanding principles of strict liability elsewhere in tort,[31] and the general principle that one person is not liable for actions or omissions of another.[32] This might explain the variety of means, including policy means, used to try to justify the doctrine.[33] It might suggest that some examples traditionally given as applications of vicarious liability are better seen, at least with hindsight, as instances of personal or direct liability, and that it is unhelpful to mix them.[34]
Intuitively, it is important to determine the rationale for particular legal principles. This is useful where there are reasonable differences of opinion regarding application of the principle in difficult cases. Specifically, where the rationale for a tort doctrine would be coherent with a finding of liability in a particular case, a finding of liability is more defensible there. In this light, one must acknowledge strong scepticism that any particular rationale can explain the principle of vicarious liability as it currently stands:
I am not persuaded that there is any reason of principle or policy which can be of substantial guidance in the resolution of the problem of applying the rule in any particular case. Theory may well justify the existence of the concept, but it is hard to find guidance from any underlying principle which will weigh in the decision whether in a particular case a particular wrongful act by the employee would be seen [to be within the scope of employment].[35]
Five members of the High Court of Australia acknowledged earlier cases had not disclosed any ‘clear or stable principle ... underpinning the development of this area of the law’.[36] The reasons noted the policy basis of the modern doctrine of vicarious liability, but acknowledged that what the precise policy actually was had not been properly articulated.[37] These sentiments were recently repeated in Prince Alfred College.[38] Respected scholars are also of the same view.[39]
Despite these misgivings, various rationales have been utilised to justify the doctrine. To some extent, they overlap. They include the agency idea that the employee is acting with the apparent or ostensible authority of the employer,[40] which connects with the command (express and implied) theory traditionally underpinning this area of law, and reflected in some Australian case law,[41] if not Prince Alfred College. Alternatives include that:
• as between two innocent parties, it is better that the loss fall on the employer because they employed the wrongdoer;[42]
• the concept of ‘enterprise risk’;[43]
• the employer will be better able to spread the risk through insurance;[44]
• it is simply a ‘loss-distribution device’;[45]
• its purpose is to find a solvent defendant with means to pay compensation;[46] and/or
• imposition of vicarious liability will deter undesired conduct.[47]
Or, perhaps the rationale of vicarious liability is simply to impose liability on an employer where it is ‘fair, just and reasonable’ to do so.[48] These theories will be considered in more detail in Part III of the article.
It would usually be expected that a rationale for legal doctrine would explain its contours and case outcomes. This does not occur in relation to vicarious liability. For instance, the sharp distinction the law has traditionally made between vicarious liability for employees and (lack of) vicarious liability for independent contractors[49] is difficult to justify on any of the above rationales.[50] This situation is unsatisfactory.
The leading Canadian authority is Bazley.[51] The case involved a defendant which operated residential care facilities for emotionally troubled children. The defendant hired an employee who was, unknown to them, a paedophile. The employee was convicted of 19 counts of sexual abuse against children in the defendant’s care. The question was whether the defendant was vicariously liable for the actions of its employee.
The Supreme Court of Canada noted the oft-cited Salmond test that an employer was vicariously liable for acts of employees that were authorised by the employer, and unauthorised acts so connected with authorised acts that they could be considered modes, albeit improper modes, of doing authorised acts.[52] The Court further noted that it was often difficult to distinguish between improper modes of doing authorised acts (for which the employer would be liable), and independent acts, for which the employer would not be liable.[53]
After considering various categories of case in which vicarious liability had been recognised, the Court suggested a common theme in such cases was so-called ‘enterprise risk’[54] — that the employer’s enterprise had created the risk that led to the tortious act, where the employee’s conduct was ‘closely connected’ with that risk.[55] The Court acknowledged this question may be difficult to answer, and that an employee’s obligations to their employer may be described at various levels of generality, which could influence the answer to the application of the ‘close connection’ test.[56]
The Court also acknowledged the role of policy in developing appropriate principles in this area. It referred with approval to Fleming’s articulation of two fundamental policy concerns in the area of vicarious liability, providing a ‘just and practical’ remedy for the harm, and deterrence of future harm.[57] Vicarious liability was considered fair because the employer placed in the community an enterprise with elements of risk. It was fair that the body which created the risk should bear the loss.[58] This conclusion was supported by the fact that an employer was often in the best position to spread any losses through insurance or higher prices for its services. Further, an employer had tools available to reduce these risks through careful management.[59]
In terms of the ‘close connection’ requirement, the mere fact the employer employed the employee, creating the opportunity for the employee to do the wrong, was insufficient. The wrongful act would have to be sufficiently related to conduct authorised by the employer to justify imposing liability.[60] It would have to be said the employer ‘significantly increased the risk of the harm by putting the employee in his or her position and requiring him [or her] to perform the assigned tasks’.[61] Several factors were of possible relevance.[62]
The question of the closeness of connection explained the fact that, on the same day that Court decided Bazley, it denied vicarious liability in another case, where the defendant did not offer residential care, and the extent of the employee’s responsibilities was much narrower in scope.[63] In Jacobi, a majority found the employer was not liable for the employee’s abuse. In this case some of the limits of the principles set out in Bazley became clear, the majority indicating limits to the economic principles cited in Bazley as rationales for vicarious liability: ‘[m]uch as the Court may wish to take advantage of the deeper pockets of the respondent to see the appellants compensated, we have no jurisdiction ex aequo bono to practise distributive justice.’[64] The majority in Jacobi also appeared to limit the discussion in Bazley regarding the ability of employer defendants to spread the risk through insurance or by raising premiums. In Jacobi, the defendant was a non-profit organisation. The majority found the ability of non-profit organisations to raise prices to users was severely constrained, and this ought to be borne in mind in applying such principles to cases of non-profit defendants.[65]
The leading Australian case of Lepore[66] involved conjoined appeals. One of the cases involved allegations of excessive physical force being applied by a teacher to a pupil under his supervision and care; another involved allegations that a teacher at a one-teacher school had sexually abused his students. The question for the High Court of Australia was whether the plaintiffs could proceed against the defendant school authorities for causes of action based on a non-delegable duty of care that a school was alleged to owe a student or, in addition or in the alternative, based on vicarious liability.
Past High Court decisions suggested schools might owe students a nondelegable duty, so that a school would be liable if students were injured independently of evidence the school was negligent, regardless whether it had purported to delegate its responsibilities to others.[67] Members of the High Court in Lepore indicated both disagreement with the concept of a non-delegable duty in such cases,[68] and agreement.[69] There was clear ambivalence about applying nondelegable duties to intentional, rather than merely negligent, wrongdoing.[70]
Members of the High Court cited the Salmond test approvingly, noting that while it was relatively easy to categorise a negligent act as an unauthorised mode of performing an authorised act, it was more difficult to do so for intentional acts.[71] Chief Justice Gleeson noted a difficulty with the ‘improper mode of performing an authorised act’ was that the answer was not always clear-cut, but depended on the level of generality at which one classified the range of acts that the employee was authorised to conduct.[72] As applied to school authorities, Gleeson CJ suggested the answer may depend on the nature and seriousness of the criminal act.[73] In the past sexual misconduct by a teacher would have been regarded as so far outside of what a teacher was employed to do as to be regarded as an independent act.[74] However, His Honour acknowledged that elsewhere successful claims had been made in such circumstances, and that such a claim could succeed in Australia, depending on the nature of the connection between what the employee concerned was employed to do, and the misconduct.[75] Chief Justice Gleeson referred to the Canadian decisions, and their use of ‘enterprise risk’, before noting that enterprise risk had not been adopted in Australia for determining whether conduct was within the scope of employment.[76] Physical discipline could be within the scope of a teacher’s responsibilities.[77]
Justice Gaudron preferred to couch the test in terms of principles of agency and estoppel. On this approach, an organisation could be liable both for actions of employees and independent contractors,[78] provided the defendant was (or should be) estopped from denying the person whose acts are in question was acting as their agent or representative.[79] Her Honour linked the estoppel approach with the ‘close connection’ test of Bazley, concluding that a person would not be estopped from denying a person was acting as their agent or representative unless there was close connection between what was done and what the person was engaged to do.[80]
Justices Gummow and Hayne disagreed with the Bazley approach. It did not give sufficient significance to the fact that the abuser’s conduct was intentional, that the employee’s conduct was a breach of the employment agreement, and that the offender had clearly not been deterred by the criminal law.[81] This undermined the argument in Bazley that vicarious liability should be imposed because it deterred undesirable behaviour.
Their Honours indicated the concept of ‘enterprise risk’ may be too broad, not focused sufficiently on risks linked with furtherance of the venture, but could quickly develop into a principle that an employer was effectively liable for everything its employees did, because it provided the opportunity for wrongdoing to occur.[82] Justices Gummow and Hayne also suggested an agency-type approach, stating an employer would be vicariously liable in either of two cases: (a) where the act complained of was done in intended pursuit of the employer’s interests or the contract of employment; or (b) where the employee’s conduct was done in ostensible pursuit of the employer’s business or apparent execution of the authority the employer held out the employee as having.[83] In so doing, they were applying a version of the test articulated by Dixon J in an earlier decision.[84] Their Honours lamented the law of vicarious liability had become difficult because of over-reliance on policy at the expense of principle. It was not enough to create liability that the defendant had ‘deep-pocket[s]’.[85]
Justice Kirby agreed with the approach in Bazley, including the ‘enterprise risk’ theory and rationales for imposition of vicarious liability posited there; namely, fair and efficient compensation for wrongful conduct, and deterrence of future harm.[86] His Honour favoured broad application of the ‘close connection’ test, [87] asking when it was ‘just and reasonable’ to hold the enterprise liable for actions of its employee.[88] His Honour agreed the fact the employment provided the occasion for the employee’s conduct was insufficient.[89] Justice Kirby concluded, in the context of an educational establishment employer, the risk of an employee abusing students was inherent in, and characteristic of (in a small number of cases), the enterprise.[90]
Justice Callinan found deliberate criminal conduct would usually lie well outside the scope of a teacher’s responsibilities, and it was unreasonable to impose liability for such conduct on an educational authority.[91] Deliberate criminal conduct was unconnected with employment. His Honour rejected use of the ‘fair and just’ test because it would lead to uncertainty.[92]
Principles of vicarious liability were considered in the recent High Court decision in Prince Alfred College.[93] The case involved abuse committed by a boarding master at the appellant school many years before the legal action. The victim sought compensation from the appellant on the basis of direct negligence. It was alleged the appellant was negligent in employing the abuser, given his past criminal history of sexual abuse, and in choosing him for a boarding master role when he was a young and inexperienced teacher. It was suggested his activities around students were insufficiently supervised and, more generally, issue was taken with use of only one teacher where students were vulnerable to predatory behaviour. It was also alleged that the appellant was vicariously liable for the abuse, because the offender committed the abuse while a boarding master at the school, such that his activities occurred within the course of his employment. Since the victim brought the legal claim many years after the abuse occurred, he required an exercise of the Court’s discretion to grant an extension of time under the relevant legislation.
The High Court overturned the decision of the state court that favoured the victim. The lower court had granted an extension of time, and had found the appellant vicariously liable for the actions of the boarding master. Only one (state) judge found the appellant directly liable in negligence.[94]
The High Court decided that too much time had elapsed between an apparent resolution of the dispute, and these subsequent proceedings. This delay was, in the Court’s view, not justified and precluded the possibility of a fair trial.[95] The Court expressly declined to deal with the substantive legal issues surrounding vicarious liability and/or direct liability of the appellant.[96] Though the Court declined to do so, it provided some parameters around the law of vicarious liability; obviously, these comments are obiter dicta. However, they are designed to provide guidance for lower courts, and will be influential. Having said that, given the past differences of opinion among the High Court and other courts on the scope of the principle, it would be a brave person to assume that what the High Court said in Prince Alfred College about vicarious liability presents the ‘last word’ on the doctrine’s scope in Australia.
Members of the High Court recognised the notion of ‘scope of employment’ had been utilised to frame the parameters of vicarious liability, and this was defensible because it provided some objective, rational basis for liability,[97] though it acknowledged that the question of ‘connection’ with employment did not add much to an understanding of an employer’s liability.[98] Prince Alfred College heralded use of a new concept to explain vicarious liability principle. This was the notion of ‘occasion’. The Court agreed with earlier case law that the mere fact employment provided an opportunity for the commission of the wrongful act was not sufficient in itself to attract vicarious liability. However, if the employment situation not only provided the opportunity, but was also the occasion for the commission of the wrongful act, vicarious liability could arise:
[I]t may be sufficient to hold an employer vicariously liable for a criminal act committed by an employee where, in the commission of that act, the employee used or took advantage of the position in which the employment placed the employee vis-à-vis the victim. Consequently, ... the relevant approach is to consider any special role that the employer has assigned to the employee and the position in which the employee is thereby placed vis-à-vis the victim. In determining whether the apparent performance of such a role may be said to give the “occasion” for the wrongful act, particular features may be taken into account. They include authority, power, trust, control and the ability to achieve intimacy with the victim. The latter feature may be especially important. Where, in such circumstances, the employee takes advantage of his or her position with respect to the victim, that may suffice to determine that the wrongful act should be regarded as committed in the course or scope of employment and as such render the employer vicariously liable.[99]
The High Court considered whether the abuser’s role as boarding master placed him in a position of power and intimacy with respect to the victim, such that the abuser’s role as boarding master ‘gave the occasion’ for the wrongful acts.[100]
It is considered somewhat difficult to distinguish, as the Court did in Prince Alfred College, between cases where the employment provided the opportunity for the wrongful act/s to occur, and cases where the employment provided the occasion for same. The distinction is important, because it has long been accepted the fact the employment provided the opportunity for the wrongful act to occur is insufficient to attract vicarious liability, and in Prince Alfred College the Court said the ‘occasion’ principle explained and justified imposition of vicarious liability in particular cases. It is not entirely clear how the concept of ‘occasion’ adds to ‘scope of employment’ notions. If ‘occasion’ is informed by issues like power, control, trust and opportunity for intimacy with victims, as the High Court said, one assumes the existence of these elements would already be relevant in considering the scope of the employee’s employment. As such, it is not clear what, if anything, the concept of ‘occasion’ adds. This observation is supported in the decision itself, where the joint reasons conclude, in discussing the recent UK decision in Mohamud, that
the role assigned to the employee ... did not provide the occasion for the wrongful acts which the employee committed outside the kiosk ... What occurred after the victim left the kiosk was relevantly unconnected with the employee’s employment.[101]
In other words, ‘occasion’ and ‘connection with employment’ tend to run together. Somewhat paradoxically though, in the judgment of Dixon J in Deatons,[102] the judgment from which the High Court derived this ‘occasion’ principle, Dixon J himself disaggregated the concept of ‘occasion’ from scope questions:
The occasion for administering [the throwing of the glass] and the form it took may have arisen from the fact that she was a barmaid but retribution was not within the course of her employment as a barmaid.[103]
The fact that, of the Court in Deatons, only Dixon J referred to the concept of ‘occasion’, and that His Honour found no vicarious liability on the facts although admitting the ‘occasion’ for the barmaid’s acts arose from her employment position, suggests the need for respectful scepticism as to the utility of the ‘occasion’ concept in clarifying the law here.
If this concept is to continue to be applied, much will turn on its articulation and development in future cases. It may be, for example, that the High Court was distinguishing a situation where a school cleaner abuses a student from a case where a boarding master does so. The Court might utilise the ‘occasion’ principle to justify imposition of vicarious liability on the school authority with respect to the abuse of the latter, not the former. However, this can lead to fine and arguably arbitrary, incoherent distinctions. Say the abuse occurs while the school cleaner is cleaning the showers in the dormitories. Can it justifiably be argued the employment was not the ‘occasion’ for the abuse, because pastoral care was not part of the cleaner’s responsibilities, but if the same abuse were committed by the boarding master in the same place, the school authority would be liable? And again, what does ‘occasion’ add here that considerations of employment scope did not? While the difficulty in framing appropriate principle in this area is fully acknowledged, I am doubtful the concept of ‘occasion’ can provide the desired certainty and workability in this area.
The House of Lords considered the liability of a school authority for sexual abuse committed by their employee boarding master in Lister.[104] It accepted and applied the ‘close connection’ test espoused in Bazley.[105] It found the abuse was ‘inextricably interwoven’ with the carrying out by the warden of their responsibilities.[106] The House acknowledged the two-stage Salmond test struggled with cases of intentional wrongdoing.[107] Lord Clyde was critical of the theoretical basis of the doctrine of vicarious liability.[108] Lord Millett thought it was best understood as a ‘loss-distribution device’.[109] It was not enough that the employer gave the employee the opportunity to commit the abuse by employing them.[110] Lord Hobhouse rejected the Bazley approach, particularly its articulation of the policy basis of vicarious liability. His Lordship preferred to phrase the inquiry as whether the defendant had ‘assumed a relationship to the plaintiff’,[111] and entrusted performance of aspects of that relationship to others, with consideration of the scope of the employee’s employment critical. Specifically, His Lordship found that if the abuse had been committed by a groundsperson, the school authority may not have been liable, but because it was committed by a boarding master whose responsibilities extended to the provision of pastoral care to boarding students, the school was liable.[112]
In the course of finding an Institute (Brothers of the Christian Schools) vicariously liable for the actions of one of its brothers in abusing children, all members of the UK Supreme Court noted:
The policy objective underlying vicarious liability is to ensure, in so far as it is fair, just and reasonable, that liability for tortious wrong is borne by a defendant with the means to compensate the victim. Such defendants can usually be expected to insure against the risk of such liability, so that this risk is more widely spread.[113]
The Court found it was usually ‘fair, just and reasonable’ to impose liability on an employer in cases where the following criteria were satisfied:
(a) The employer was more likely to have means to compensate the victim than the employee and can be expected to have insured against that liability;
(b) The tort will have been committed as a result of activity by the employee on behalf of the employer;
(c) The employee’s activity is part of the employer’s business activity;
(d) The employer, by employing the employee, has created the risk the employee will commit the tort; and
(e) The employee has to some extent been under the employer’s control.[114]
The Supreme Court noted the House of Lords had applied the ‘enterprise risk’ approach of Bazley in Dubai Aluminium, before concluding the Institute was liable here because of the close connection between the nature of the brothers’ employment at the school and the abuse committed.[115] Subsequent recent decisions have endorsed the enterprise risk approach, ‘fair and just’ criterion, and the close connection test.[116] Recently, it meant the owner of a service station was liable for the actions of their kiosk attendant in leaving the kiosk, and then punching and kicking a customer, causing them serious injury.[117]
Having summarised the existing position in the three jurisdictions, this article will now critique the various principles said to underpin these decisions.
As indicated above, one of the rationales[118] currently accepted by courts in the UK and Canada for imposition of vicarious liability on an employer is that it is, or may be in given cases, fair, just and reasonable to do so.[119] This reference to imposition of (vicarious) liability where it is fair, just and reasonable has unmistakable parallels with the approach to imposition of a duty of care in those jurisdictions.[120] Specifically, in the UK a three-stage test applies to determine whether a duty of care is owed; one stage asks whether it would be ‘fair, just and reasonable’ to recognise a duty of care in the given situation.[121] A similar approach applies in Canada, which uses a modified two-stage approach from Anns v Morton London Borough Council,[122] where policy considerations including fairness are relevant at the second stage.[123] In Australia, only Kirby J adopted the three-stage approach in determining whether a duty of care was owed, involving consideration of policy factors such as fairness.[124] In noting this apparent overlap between the use of concepts of fairness and justness in establishing that a duty of care is owed, and establishing that vicarious liability applies, it is acknowledged that the contexts differ: the former involves questions of direct liability; the latter relates to liability for actions or omissions of another.
A unanimous High Court in Sullivan v Moody expressed difficulties with the ‘fair, just and reasonable’ approach to the question of the existence of a duty of care:
There is ... a danger that the matter of foreseeability ... having been determined, the succeeding questions will be reduced to a discretionary judgment based upon a sense of what is fair, and just and reasonable as an outcome in the particular case ... The question as to what is fair, and just and reasonable is capable of being misunderstood as an invitation to formulate policy rather than to search for principle. The concept of policy, in this context, is often ill-defined. There are policies at work in the law which can be identified and applied to novel problems, but the law of tort develops by reference to principles, which must be capable of general application, not discretionary decision-making in individual cases.[125]
Although it is conceded that these remarks were made in the context of application of concepts of fairness, justice and reasonableness to the question of a duty of care, they are relevant to use of similar concepts in rationalising imposition of vicarious liability in a particular case. They are subjective contested concepts unlikely to provide clarity of principle in the vexed area of vicarious liability.[126] Recently in Prince Alfred College, members of the High Court again rejected use of concepts of ‘fair and just’ in determining the ambit of vicarious liability.[127]
Included within the concept of ‘fair, just and reasonable’ that is rejected as a principled basis for vicarious liability are comments indicating it is ‘better’ that the loss should fall on the employer than the innocent third party, or that there ‘ought’ to be a remedy for the wrong done to the victim.[128] Expression of this sentiment is usually traced to Holt CJ in Hern,[129] and then reflected in subsequent English[130] and Australian authority.[131] It is reiterated that assessments of where it is ‘better’ that the loss lie were made at a time when the generalised duty of care concept did not exist. Had it been otherwise, the Court might have found a remedy for the wronged third party on the basis the employer had directly breached a duty of care owed to the third party. Now that such a duty does exist, the rationale for vicarious liability on the basis of where it is ‘better’ or more fair and just that the loss lies has largely evaporated. With respect, that observation has not been specifically recognised in the case law as it should have been. Such recognition should prompt greater willingness to reconceptualise the law in this area.
Others have pragmatically observed that several of the vicarious liability cases[132] appear to be based on a desire to find a defendant with pockets sufficiently deep to pay compensation to a wronged individual.[133] I disagree, at the level of principle, that any cases can or should be decided based on whether a defendant has ‘deep’ pockets. Decisions create precedents and refine principles that will be used in subsequent cases. If a precedent establishing liability is made, no-one can be sure the context in which it will be used in future. Perhaps it might be used against a defendant with ‘shallower’ pockets. Should the original decision be applied? How deep must the defendant’s pockets be? This shows the folly of deciding cases based on whether a defendant has deep pockets. Courts develop and apply legal principle; they are not social welfare agencies concerned with distribution or redistribution of wealth.[134] I author agree with the High Court in Soblusky v Egan that, while deep pockets considerations might explain past cases and development of vicarious liability principle, they do not justify it, if the cases depart from sound principle.[135] Williams, canvassing possible rationales for vicarious liability, considered the deep pockets theory:
There is the purely cynical theory that the master is liable because he has a purse worth opening. The master is frequently rich, and he is usually insured — two arguments that might be used by any burglar, if he ever troubled to justify his thefts. The strange thing is to find them put forward by judges of eminence [citing Lord Lyndhurst and Willes J] ... Whatever ... can have put this extraordinary idea into judges’ heads, that the mere possession of wealth is enough to justify the imposition of legal liability for a wrong ...[136]
Clearly, this rationale does not explain or justify the traditional distinction between employees and independent contractors, to provide one example.
Further, I do not agree with the creation and application of legal principle based on who has, or who has access to, insurance. This is an orthodox position,[137] although some judges have apparently challenged it, expressly taking into account the availability or otherwise of insurance in determining whether vicarious liability should be imposed.[138] This type of reasoning is not employed expressly elsewhere in tort law; it should not be permitted here.[139] As Atiyah succinctly puts it, the extent of insurance is (should be) determined by the law, not the other way round.[140] The principle of vicarious liability was established before insurance was available;[141] there is an incongruity in using a product developed after the creation of a principle to justify and rationalise that principle.[142]
One rationale the Supreme Court of Canada gave for its interpretation of principles of vicarious liability in Bazley was that of ‘enterprise risk’ that the employer had introduced the risk of employee wrongdoing, and damage caused by that employee was basically a cost of doing business. This theory has to some extent been accepted by courts in the UK[143] and Australia.[144]
While the High Court of Australia did not expressly embrace the concept of ‘enterprise risk’ in Prince Alfred College, there is an argument that the test espoused there of whether the employment provided the ‘occasion’ for the wrongful act is somewhat similar to notions of enterprise risk.[145] The idea is that just as the fact that an enterprise conducts certain activities carries with it inherent risks, the costs of which are allocated to the enterprise (according to the enterprise risk theory), so too when an enterprise engages individuals to do particular tasks, it might be thought to have taken on liability risk associated with performance of such tasks. This is concededly speculative; the High Court did not in Prince Alfred College expressly make this link.
Notions of enterprise risk were refined among law and economics scholars in the United States (‘US’), and there are some early traces in English cases.[146] The basic idea is that the full costs of an activity should be borne by the body responsible for that activity, rather than shifted elsewhere. This will cause optimal allocation of resources.[147] As has been noted, this concept of ‘enterprise risk’ is inherently difficult to apply to non-profit organisations such as educational authorities.[148] Many would find it inappropriate and/or offensive to view sexual abuse as a ‘business risk’, and compensation to be paid by an employer as a ‘cost of doing business’.
Even if economic analysis of law is accepted as being of possible relevance in this context, and legal principles should be governed or influenced by ‘efficiency’, leading law and economics scholars are dubious about the imposition of vicarious liability. For example, Sykes posits a situation involving ‘passive victims’, where victims will not take ‘cost effective measures’ to protect themselves from harm by adopting precautions or avoiding activities that might subject them to it.[149] A student in a school is considered a ‘passive victim’ in this context.
In such a situation, Sykes raises the possibility that some employees who produce the employer’s goods (or services, in the case of an educational authority) commit assaults creating civil liability:
By hypothesis, neither the probability that an employee will commit an assault nor the magnitude of the resulting liability depends upon the employee’s wealth, upon whether he is employed or unemployed, or upon any characteristic of his employment. In addition, by hypothesis, no incentive mechanism exists that would enable the employer to reduce the incidence of employee assaults. To hold the employer vicariously liable for such employee assaults, regardless of when or where they occur or the identity of the assault victim, is almost certainly inefficient. The employer cannot affect the number of assaults or the total amount of damages. Thus [where] vicarious liability imposes additional costs on the enterprise, it may drive the enterprise out of business or at the very least cause it to shrink as the marginal costs of production increase. Society then will lose at least some of the economic surplus [not to mention the social benefit] from the production of the enterprise.[150]
Sykes concludes vicarious liability is ‘probably inefficient’ if two conditions exist — the enterprise does not cause the wrong, and the enterprise cannot reduce the probability of it occurring through incentive contracts with employees.[151]
He finds that in circumstances where the causal relationship between the activities of the employer and the prospective wrong is weak, direct negligence is more efficient than vicarious liability.[152] Regarding this first requirement, the courts have stressed the mere fact that employment of the abuser gave them the opportunity to commit the abuse is insufficient to find the employer liable, so it cannot be said the employer ‘caused’ the wrong, in relation to one of their employees abusing students. Indeed, if they could be said to have caused the wrong, they would be liable directly, without the need for recourse to vicarious liability. Regarding the second, it is hardly practical to reduce the likelihood of abuse by including incentives in employment contracts. If the prospect of jail term and instant dismissal from employment for abusing children does not deter wrongdoers, there is nothing else the employer can do to dissuade a would-be abuser. In summary, of the two conditions articulated by a leading law and economics scholar for vicarious liability being a more ‘efficient’ means of considering liability, neither applies in the current context.
Ironically then, according to one leading law and economics scholar, it is not ‘efficient’ to make an educational authority liable for abuse committed by an employee, yet efficiency underlies the basis of the Canadian Supreme Court’s ‘enterprise risk’ doctrine in Bazley, as subsequently applied by courts in the UK. Further, tort law is incoherent if it accepts concepts of enterprise risk in the context of vicarious liability when it does not, for instance, in the context of product liability or car accidents.[153] Further, notions of enterprise risk do not explain or justify why traditionally vicarious liability distinguishes actions of employees and independent contractors, nor why an employee remains liable for the tort, even if it occurred pursuant to furtherance of the enterprise (and, logically, its costs might be borne solely by it).[154] It is difficult to apply it to non-profit organisations, as the UK Supreme Court concluded in Jacobi.
Even if enterprise risk justifies, to some extent, the imposition of vicarious liability, its rationale requires limited application of the doctrine. As Smith noted:
To make the entrepreneur responsible for acts of his employees in no way connected with the enterprise would be undesirable because it would result in including in the cost of production an item which economically does not belong there.[155]
I therefore agree with the reticence expressed by members of the High Court of Australia in Lepore towards adoption of ‘enterprise risk’ theory to rationalise the imposition of vicarious liability,[156] at least in the current context of the liability of educational providers to victims of abuse. The mere fact the employment provided the ‘occasion’ for the wrongful act/s to occur would, on this basis, not be sufficient either to impose liability, if I am correct in at least broadly equating notions of ‘occasion’ with ‘enterprise’. Sykes concluded vicarious liability was not efficient where an employer did not cause the wrong, and could not reduce its probability through contract provision. The fact an employment context provided the occasion for the wrong does not mean that the employer caused the wrong.
Another justification given for the imposition of vicarious liability is the policy role it can arguably play in providing deterrence to socially undesirable behaviour. In Bazley v Curry, McLachlin J for the Court cited Fleming as identifying deterrence, together with what was a ‘just and practical remedy’, as two key rationales of vicarious liability.[157] Her Honour said that making an employer liable for the wrongful acts of employees could have a deterrent effect, given employers could often reduce risks by ‘efficient organisation and supervision’.[158]
However, as Gummow and Hayne JJ explained in Lepore in terms of deterrence, if the possibility of criminal conviction and punishment does not prevent the abuser from committing the abuse, it seems unlikely that the possibility of the imposition of civil liability will do so.[159] And employers can be held directly liable for negligence in the selection of employees, or bad workplace practices including failure to supervise, quite independently of the imposition of vicarious liability. As such, deterrence is not accepted as a theoretical justification for imposition of vicarious liability. McIvor rejects deterrence reasoning in Bazley as ‘naïve and simplistic’.[160]
In summary, none of the concepts of fair, just and reasonable; deep pockets; availability of insurance; deterrence or enterprise risk adequately explain and justify imposition of vicarious liability in particular instances. Part V of this article will now consider the potential of agency doctrine to provide the theoretical basis for vicarious liability, acknowledging that not every precedent can be fitted within its bounds.
I do not accept the position of some courts and academics that a search for a coherent theoretical underpinning or justification of the law of vicarious liability will be in vain, and that there are multiple rationales, none of which is entirely satisfactory of itself.[161] I believe that if legal principles are to be applied, they must have theoretical underpinning because this assists the objective resolution of novel cases, and encourages the resolution of disputes via principle, not deep pockets, insurance or subjective assessments of what is ‘fair and just’.
The most promising theoretical basis upon which to justify the principle of vicarious liability is considered that of agency. It is conceded that some existing vicarious liability precedents cannot be justified as instances of agency doctrine. If so, they should no longer be considered good law. It is time the law made an effort to reconceptualise principle in this area into a coherent and functional whole. It will be impossible to do this if we remain wedded to all of the past case law. Some see the purpose of academic writing as an attempt to provide a theoretical underpinning for past case law, explaining what has occurred in the past in terms of a coherent, all-inclusive theory. In the area of vicarious liability, we must accept such an attempt is actually not possible. We must be prepared to jettison some of the old, to move forward with a coherent approach for the future.
Arguments that vicarious liability is an example of agency theory are not new. However, those arguments are becoming less easy to discern in the modern case law. They justified the initial conception of vicarious liability as relating to something the employer had expressly or impliedly commanded or authorised, as Holmes discusses in his treatment of vicarious liability in an article entitled ‘Agency’.[162] The agency nature of vicarious liability was acknowledged in Colonial Mutual Life Assurance Society Ltd v Producers and Citizens Co-Operative Assurance Co of Australia Ltd. [163] Justice Dixon, with whom Rich J agreed, held ‘the rule which imposes liability upon a master for the wrongs of his servant committed in the course of his employment is commonly regarded as part of the law of agency’.[164] Chief Justice Gavan Duffy and Justice Starke discussed vicarious liability in terms of what the employer had expressly authorised the employee to do or was within the scope of the employee’s (agent’s) authority.[165] It was referred to expressly by the Canadian Supreme Court,[166] and by Lord Wilberforce.[167] The judgments of Latham CJ and Dixon J in Deatons include some indirect support for the agency basis of vicarious liability, as will be noted presently.
The basis of vicarious liability in the law of agency was discussed at length by McHugh J in Scott,[168] and applied by him in Hollis.[169] Justice Kirby accepted and applied agency theory of vicarious liability in Sweeney.[170] In Lepore, Gaudron J accepted agency as providing theoretical justification for vicarious liability:
If vicarious liability is to be imposed so that a person is to be held liable in damages for injury suffered without fault on [their] part, it ought to be imposed only in circumstances where it can be justified by reference to legal principle ... [T]o hold an employer liable for the authorised acts of an employee or acts done in the course of his or her employment, is simply to apply the ordinary law of agency.[171]
In the same case, Gummow and Hayne JJ used words associated with agency in clarifying the extent of vicarious liability.[172] Agency principles were not discussed in Prince Alfred College.
Acknowledging that the concept of ‘agency’ can mean many different things to different people, and be used in different contexts, it is used here to mean the situation where one person (A) is acting on behalf of another (B), such that it can be said that A’s actions are really those of B (the ‘broad view’). Obviously this can include express, apparent and ostensible authority. This is a wider meaning of the word agency than that of someone who has the capacity to make legal relations between a principal and a third party (the ‘narrow view’).[173] This distinction is important. Support for the broad view of the concept of agency appears (implicitly) in Scott, Hollis, Sweeney and Lepore. These cases did not involve situations where the worker had the ability to enter into legal contracts on behalf of another, yet they were rationalised by some as involving ‘agency’ principles. Implicit is acceptance of a broad view of agency.
In the US, § 7.07 of the Restatement (Third) of Agency[174] confirms links between vicarious liability and agency. Stating an employer is liable for an employee’s actions occurring within the scope of employment,[175] the section states an employee is an agent whose principal controls or has the right to control the manner and means of the agent’s performance of work.[176] It further clarifies that actions are not within the scope of employment when they occur ‘within an independent course of conduct not intended by the employee to serve any purpose of the employer’.[177] This mirrors comments by both Latham CJ and Dixon J in Deatons, a vicarious liability case where both considered whether the employee’s actions were done ‘in furtherance of the interests of the employer’.[178] Focus on whether the act complained of was intended to serve the employer’s purpose has been a consistent theme of agency restatements, and is consistent with questions regarding whether the employer is on a ‘frolic’, and past concern in English case law with whether the employee’s actions were, or intended to be, for the employer’s benefit. An employer may also be liable on the basis of apparent authority.[179]
Given Beuermann’s important work in this area,[180] this article should explain the extent to which my theory relates to Beuermann’s theory. I will focus primarily here on Beuermann’s theory of conferred authority strict liability. Fundamentally, our work is not inconsistent, but has different objectives. Neither of us apparently think vicarious liability is an appropriate vehicle for resolution of institutional child sex abuse claims.[181] Neither of us think vicarious liability is appropriate when dealing with intentional wrongful conduct. We agree vicarious liability should be confined to what an employer institution has expressly or impliedly authorised.[182] While my concern here is to clarify the contours of the law of vicarious liability, Beuermann has a different concern. She advocates for recognition of ‘conferred authority strict liability’.[183] She expressly distinguishes this from vicarious liability.[184] At first, it might seem our work overlaps, because Beuermann uses concepts such as ‘authority’ and ‘apparent authority’, terms sometimes associated with the law of agency, which is the principle I utilise. However, the learned scholar does not use the word ‘agency’ in explaining her theory, and my reading of her work suggests she does not mean these words in the ‘agency’ sense. As such, it is a different theory than the one posited here.
Her theory is that an institution is (should be) strictly liable for child sexual abuse committed by a person on whom authority has been conferred, by an institution which itself has been conferred with authority.[185] School teachers for example are conferred with authority over students by the educational institution, which derives such authority through legislation. This gives teachers power to direct students. Beuermann says this power is the source and rationale for the imposition of strict liability. This is her ‘conferred authority strict liability’, similar to the previously discarded non-delegable duty principle in the context of intentional wrongdoing. Her suggestion is largely supported by the recommendation in the Redress Report to reintroduce non-delegable duties in this context.[186] I hope to consider her interesting theory of ‘conferred liability strict liability’ and the Report’s recommendations in relation to non-delegable duties in future work.
However, for the purposes of this article, Beuermann’s work on conferred authority strict liability is not of central importance, because the purpose of this article is to outline the confines of the vicarious liability doctrine, while the purpose of Beuermann’s work is effectively to seek the reintroduction of a non-delegable duty of care in the current context. She does not argue for change to the law of vicarious liability; while I do. She advocates for the reintroduction of a nondelegable duty in this context; I do not.
In a book chapter, Beuermann does discuss vicarious liability in the context of ‘agency’.[187] However, importantly, she is using the ‘narrow view’ of agency to mean someone with the authority to effect (transactional) legal relations on behalf of another,[188] though she admits that in some cases a broader conception of agency has been utilised.[189] I indicated above the broad view of agency is adopted here, and has support in Australian case law. Beuermann’s comments about vicarious liability in relation to an ‘agent’ in that chapter must be seen in the context of her narrow view of agency, because that affects her position on the extent to which an organisation should be vicariously liable for conduct of an ‘agent’, in the sense she uses the term.[190] As indicated above, the ‘broad view’ of agency is used here, in line with Australian authority, and this difference in definition is critical. In adopting the narrow view of an ‘agent’, she does not suggest this concept be used to reconceptualise the law of vicarious liability, which is the argument made here. In this way, our work differs.
While no doubt imperfect, the idea that vicarious liability is based on principles of agency has several practical advantages. First, it removes the existing strict dichotomy between employees (for whom the employer will be vicariously liable for actions within scope of employment), and independent contractors (for whom the employer is not vicariously liable). As indicated by McHugh J and others,[191] the continued merit of this distinction is questionable given changes in the nature of employment over time. It is incoherent with the enterprise risk basis of liability or the fair, just and reasonable theory, since loss caused by an independent contractor may be considered just as much ‘a cost of doing business’ as what an employee does, and it might be fair, just and reasonable for an employer to be vicariously liable for what an independent contractor did, yet the traditional distinction precludes it. To be clear, neither the enterprise risk approach nor the fair, just and reasonable approach is defended here; rather the point is that the distinction between employees and independent contractors is not supported by any of the current suggested rationales for the doctrine. Its discard should be unlamented.
The second advantage is to prevent situations where an employer is liable for actions of an employee that have been specifically prohibited by the employer.
I cannot support precedents where an employer is liable for injuries caused when the employee has defied specific employer instructions in causing loss to a third party.[192] These are not really acts of the employer, and the employer should not be vicariously liable for them. Nor can it be accepted that an employer should be (vicariously) liable for actions they never would have approved of or authorised, had they been consulted. This means that in most, if not all, cases in which an employee is guilty of intentional, and in some cases criminal, wrongdoing, the employer would not be vicariously liable for this behaviour.[193] If the rationales of enterprise risk, deep pockets and fair, just and reasonable are not accepted (as has been suggested above), there is little left to conceptually justify why an employer should be liable for actions they either specifically prohibited, or would never have authorised, if asked. Thus,
I must respectfully disagree with Beuermann’s acceptance of liability in cases where an employer in this context would never have authorised the particular acts of wrongdoing and/or cases of intentional wrongdoing,[194] though I concede her acceptance of this outcome is via use of the doctrine of ‘conferred authority strict liability’, which she distinguishes from vicarious liability.[195]
In this regard, the focus of the US Third Restatement of Agency on whether the employee’s actions were intended to serve the purpose of the employer is welcome. It was the position of the common law in the UK until Lloyd in 1912.[196] In 1873, for example, the Court found in Barwick that the ‘general rule is, that the master is answerable for every such wrong of the servant or agent as is committed in the course of the service and for the master’s benefit’.[197] In 1919 in Australia, a member of the High Court asked whether the employee in question ‘assumed to act within the scope of their employment’.[198]
However, the House of Lords in Lloyd interpreted the 1873 decision as not requiring the act of the wrongdoer be for the master’s benefit; merely that this was the factual scenario in Barwick. With respect, some reasoning of the judges in Lloyd is open to serious question. In explaining why the employer was liable, Earl Halsbury claimed the employee had breached a duty to apply diligence and honesty in carrying out business within their delegated powers.[199] The Earl cited with approval comments by Holt CJ in an earlier case that a merchant was liable for the deceit of his factor because someone had to be a ‘loser’, and there was more reason that the employer who employs the deceiver and places their trust and confidence in them should be the loser than the customer.[200] Lord Macnaghten took a similar view, asking ‘who is to suffer for this man’s fraud’, the innocent customer or the employer ‘who put this rogue in his place and clothed him with his authority’.[201] His Lordship added an employer could insure against the dishonesty of employees or make them sign a ‘fidelity’ policy.[202]
With respect, these comments are findings of fault against the employer for employing the dishonest employee. In 1912, a generalised duty of care did not exist. Findings of negligence were dependent on bringing relations between the parties into a pre-existing recognised category in which duties of care were owed. This might explain the apparent intrusion of fault concepts into consideration of a doctrine of strict liability in vicarious liability. However, since 1932, when a generalised fault-based duty of care was recognised,[203] it surely became illegitimate and incoherent to base the imposition of strict liability on notions of fault around the employment of a dishonest individual. Fault in the employment of a dishonest employee is relevant to a direct claim of negligence against the employer; it is not relevant to consideration of a vicarious liability claim against them. So, while Lloyd is explicable according to other aspects of tort law as they stood in 1912, its reasoning has been significantly undercut by broad acceptance of fault-based liability in negligence. This has not been recognised in the case law. And its other bases in deep pockets and insurance are not accepted, for reasons given earlier.
Even if we accept the notion of ‘enterprise risk’ is at least an arguable justification for imposing vicarious liability, on the basis of properly allocating the costs of an activity to ensure a more optimal allocation of resources, we must be aware of ‘over-allocation’ of costs. Smith warned against including costs that ‘economically do not belong’. It is argued the cost of actions of an employee not acting pursuant to the employer’s interests does not ‘belong’ with the employer. In other words, there is a link between a theory of enterprise risk and the idea that an employer should only be liable for actions of an employer designed to benefit the employer. This link remains in the US jurisprudence, evidenced by the Third Restatement of Agency, but it was severed in 1912 when the House of Lords decided Lloyd, and incoherence in the UK continues when the courts apply the principles of Bazley, but do not accept the enterprise risk methodology underlying it.
The practical effect of applying vicarious liability principles through the doctrine of agency will be a reduction in claims for which the employer will be liable. This is acknowledged by Oxford. Writing of two cases involving alleged vicarious liability, one involving a police officer involved in sexual activity with a store employee while on duty, another involving a pastor sexually abusing a child, he states:
In neither case did the agent act on express or implied directions of the respective principal. In addition, neither party acted with intent to serve his principal, but acted out of his own self-interest. Accordingly, both sexual assaults were outside the scope of employment. If both [complainants] understood that their attacker committed torts without acting on behalf of the respective principal — lacking the requisite reasonable belief of authority — then neither principal should be vicariously liable for their agent’s intentional physical torts.[204]
Acceptance of this approach would mean that precedent such as Lister in the UK and Bazley in Canada would no longer be regarded as good law. Nor does it fit with the Australian High Court’s articulation of the ‘occasion’ principle in recent times.[205] The recent Mohamud decision, where an employer was liable for a retail employee who left his work station and violently assaulted a customer, would not be accepted.[206] There are numerous cases in the US where vicarious liability for sexual assaults of employees has been denied because the employee was not acting in furtherance of the employer’s interests.[207] Counter-intuitively, it appears easier to make an employer liable for wilful acts of their employees in the UK than the US.
Reduction in the ambit of vicarious liability claims against employers is independent of the extent to which an employer is directly liable for breaching their duty of care. Specifically, that an educational institution is not vicariously liable for sexual abuse committed by one of their employees on students within the institution’s care is not the end of the (liability) story. That institution may be directly liable in negligence, for employing a person with a past criminal history of sexual abuse, engaging an employee who was otherwise inappropriate for the position, not supervising that employee sufficiently, or inadequate systems (for example, a policy requiring employees not be alone with students at ‘vulnerable’ times). This article says nothing about the educational institution’s direct liability in negligence for these types of failures. It simply takes issue with the unprincipled use of the vicarious liability concept in such a situation. Suffice to say that a narrowing of vicarious liability in the way suggested in this article is not intended to mean, and does not necessarily mean, that a person subject to sexual abuse within an educational institution will be left without a legal remedy.
More victims of institutional sexual abuse are coming forward to acknowledge the pain and suffering to which they have been subject, and to seek redress for past atrocities. One aspect of the societal response to such events is the legal response, and, in particular, the question of liability of the institution, particularly here educational institutions, either directly, or through the doctrine of vicarious liability. These events cause us to consider the theoretical basis of these doctrines and their place in tort law in more detail.
The doctrine of vicarious liability has moved well beyond its historical roots, undermining its traditional rationale. Some of the reasoning of previous cases has been undermined by developments elsewhere in the law of tort. Various rationales continue to be suggested, but none of these in themselves can adequately explain and justify the state of the existing authorities. This article has found that none of the theories of what is fair, just and reasonable, deep pockets, insurance, enterprise risk or deterrence provide the needed theoretical underpinning for the doctrine. Though some simply lament this unsatisfactory state of affairs, this article has assumed that such a state of affairs is unacceptable, and some reconceptualisation of the law in this area necessary.
Though no resolution is perfect, this article has suggested the doctrine of agency could be utilised in this reconceptualisation. It has indicated that in some of the past vicarious liability cases, notions of agency were evident. The focus in terms of vicarious liability should be on the extent to which the employee was acting in pursuit of the employer’s interests. This means that where an employee is on a frolic of their own, when they are engaged in activities which the employer did not and would never have authorised, and/or where they are engaged in criminal activity, vicarious liability should not apply. This approach should also be taken in the context of institutional sexual abuse. This relates only to the doctrine of vicarious liability. It does not, and is not designed to, preclude victims of institutional child sexual abuse from obtaining legal redress through other avenues, for example by arguing that the institution at which the abuse occurred is directly liable, rather than vicariously through actions of an employee of which they were not aware, and would never have authorised. Victims of wrongdoing are entitled to justice, but legal principle must be coherent.
[∗] Professor, School of Law and Justice, University of Southern Queensland, Springfield, Queensland, Australia. Thanks to the two anonymous referees for helpful comments on an earlier draft of this article.
[1] Commonwealth, Royal Commission into Institutional Responses to Child Sexual Abuse, Redress and Civil Litigation Report (2015) 8 (‘Redress Report’).
[2] The Redress Report recommended implementation of a redress scheme, as an alternative to civil litigation as a means of compensating survivors: ibid, 322, 390. It would have a maximum payout cap of $200 000 (at 252); and would only be available to those who agreed not to pursue a civil claim (at 390). Given that a successful civil damages claim would often be for a substantially higher figure, it is not known whether, even if implemented, this redress scheme will substantially reduce the number of civil claimants. And at the time of writing, it is not clear whether such a scheme will be implemented, particularly given the cost involved. The Commission recommended substantial reforms to legislation in relation to limitation periods (at 52–3); introduction of a non-delegable duty of care owed by institutional care providers (at 77–8); a reverse onus of proof provision (at 56); and reforms to make it easier for survivors of abuse to identify an appropriate body to sue (at 58–9). Again, it is not currently known whether these suggested reforms will be implemented. See also Commonwealth, Royal Commission into Institutional Responses to Child Sexual Abuse, Consultation Paper: Criminal Justice (2016).
[3] [2016] HCA 37; (2016) 335 ALR 1 (‘Prince Alfred College’).
[4] On restorative justice in this context, see Anne-Marie McAlinden and Bronwyn Naylor, ‘Reframing Public Inquiries as “Procedural Justice” for Victims of Institutional Child Abuse: Towards a Hybrid Model of Justice’ [2016] SydLawRw 14; (2016) 38(3) Sydney Law Review 277; Simone Degeling and Kit Barker, ‘Private Law and Grave Historical Injustice: The Role of the Common Law’ [2015] MonashULawRw 14; (2015) 41(2) Monash University Law Review 377; Redress Report, above n 1. The Royal Commission canvassed other remedies such as apologies (at 140–56); counselling and psychological care for survivors (at 177–218); and statutory and other schemes for compensation (at 219–62). It also recommended introduction of a national redress scheme (at 26) and noted deficiencies with civil litigation responses (at 92–3).
[5] Richard Townshend-Smith, ‘Vicarious Liability for Sexual (and Other) Assaults’ (2000) 8(2) Tort Law Review 108, 122: ‘That there may be particular sympathy for the victims of sexual assaults cannot itself provide an adequate justification for the imposition of vicarious liability.’
[6] Recently, the High Court of Australia found an action against an educational provider for abuse committed many years earlier by one of its employees was barred due to the passing of time: Prince Alfred College [2016] HCA 37; (2016) 335 ALR 1.
[7] Ibid.
[8] The so-called doctrine in Rylands v Fletcher (1886) LR l Ex 265 was subsumed into the general law of negligence in Burnie Port Authority v General Jones Pty Ltd [1994] HCA 13; (1994) 179 CLR 520 (‘Burnie’).
[9] Brodie v Singleton Shire Council [2001] HCA 29; (2001) 206 CLR 512.
[10] The Ellis defence provided that church leaders may use limited liability as a shield against claims for abuse committed by others if a church was an unincorporated association: see Trustees of the Roman Catholic Church for the Archdiocese of Sydney v Ellis [2007] NSWCA 117; (2007) 70 NSWLR 565. See also Redress Report, above n 1, 58–9.
[11] Regarding the removal of limitation periods for alleged child sexual abuse, see Limitation Act 1969 (NSW) s 6A; Limitation of Actions Act 1958 (Vic) s 27P; Limitation of Actions Act 1974 (Qld) s 11A. The Redress Report recommended the removal of the limitation period with respect to child sexual abuse claims: above n 1, 434–59.
[12] Roman Catholic Church Trust Property Amendment (Justice for Victims) Bill 2014 (NSW).
[13] The Redress Report recommended substantial legal reforms to overcome difficulties in survivors identifying the appropriate body to sue, issues often described in shorthand form as the ‘Ellis defence’: Redress Report, above n 1, 496–511. It also recommended a reverse onus be applied to such claims: at 56–7.
[14] The Royal Commission noted that while abuse continued, past years probably reflected higher rates of abuse, explained by different social attitudes towards children, and past unquestioning acceptance of authority: Redress Report, above n 1, 5.
[15] Redress Report, above n 1, 92–3, 461–4.
[16] Justice O W Holmes Jr, ‘Agency’ (1891) 4(8) Harvard Law Review 345, 351–2.
[17] Staple, Merchant Strangers, Money Act, 27 Edw 3, c 19: ‘no merchant nor other ... shall lose or forfeit his goods nor merchandizes for the trespass and forefeiture of his servant unless he do it by the command or procurement of his master’.
[18] Waltham v Mulgar [1688] EngR 87; (1606) Moo KB 776; 72 ER 899; John H Wigmore, ‘Responsibility for Tortious Acts: It’s History’ (1894) 7(6) Harvard Law Review 315, 333–4.
[19] John H Wigmore, ‘Responsibility for Tortious Acts: Its History — II’ (1894) 7(7) Harvard Law Review 384.
[20] Middleton v Fowler [1795] EngR 220; (1699) 1 Salk 282; 91 ER 247 (Holt CJ) (‘Middleton’); Sir William Blackstone, Commentaries on the Laws of England (1765–9) vol 1, 429.
[21] Stone v Cartwright [1795] EngR 4211; (1796) 6 TR 411; 101 ER 622; Barwick v English Joint Stock Bank [1867] UKLawRpExch 30; (1867) LR 2 Ex 259, 265 (Willes J) (‘Barwick’).
[22] Joel v Morison (1834) 6 Car & P 501, 503; 172 ER 1338, 1339 (Parke B).
[23] ‘[T]he reason that I am liable is this, by employing him I set the whole thing in motion; and what he does, being done for my benefit and under my direction, I am responsible for the consequences of doing it’: Duncan v Findlater [1839] EngR 1005; (1839) 6 Cl & F 894, 910; [1839] EngR 1005; 7 ER 934, 940 (Lord Brougham) (emphasis added) (citations omitted) (‘Duncan’). In Turberville v Stampe, Holt CJ, the judge credited with the initial recognition of vicarious liability, explained that an employer could be liable for actions of an employee not specifically authorised where ‘it shall be intended, that the servant had authority from his master, it being for his master’s benefit’: [1792] EngR 2684; (1697) 1 Ld Raym 264, 265; [1792] EngR 2684; 91 ER 1072, 1073. For critique, see William O Douglas, ‘Vicarious Liability and Administration of Risk I’ (1928) 38(5) Yale Law Journal 584, 584. Baty asks how obiter dicta of Holt CJ in various cases of the late 17th and early 18th centuries ‘acquired the force of law’: T Baty, Vicarious Liability (Claredon Press, 1916) 28.
[24] American Law Institute, Restatement (Third) of Agency (2006) § 7.07 discusses the ‘scope of employment’, excluding from it an employee’s independent course of conduct not intended by the employee to serve any purpose of the employer.
[25] Lloyd v Grace, Smith & Co [1912] AC 716 (‘Lloyd’).
[26] Limpus v London General Omnibus Co [1862] EngR 839; (1862) 1 Hurl & C 526; 158 ER 993 (‘Limpus’).
[27] Lloyd [1912] AC 716.
[28] Priestley v Fowler [1837] EngR 202; (1837) 3 M & W 1; 150 ER 1030.
[29] Donoghue v Stevenson [1932] AC 562. As an example, Laski wrote in 1916: ‘public policy obviously requires a means of forcing masters to keep continual watch over the conduct of their servants, and it is difficult to see how that end would otherwise be attained [other than through the principle of vicarious liability]’: Harold J Laski, ‘The Basis of Vicarious Liability’ (1916) 26(2) Yale Law Journal 105, 116. Obviously, failure by an employer to keep watch over their employees would today be remedied by negligence law, and a finding that the employer was directly liable for negligently failing to supervise. However, this remedy was not generally available in 1916.
[30] P S Atiyah, Vicarious Liability in the Law of Torts (Butterworths, 1967) 12.
[31] Burnie [1994] HCA 13; (1994) 179 CLR 520.
[32] Smith v Leurs [1945] HCA 27; (1945) 70 CLR 256, 262 (Dixon J), cited approvingly in Modbury Triangle Shopping Centre Pty Ltd v Anzil [2000] HCA 61; (2000) 205 CLR 254, 264 [20] (Gleeson CJ), 299 [140] (Callinan J).
[33] [T]ort law generally is deeply fault-based. It reflects our deep-seated psychological and moral convictions that responsibility should be attributed to someone on the basis of fault and that other forms of responsibility have less validity. The issue of vicarious liability is thus a difficult one in a fault-based system. ... [I]n the cases on vicarious liability ... courts [are] struggling to justify the imposition of liability on a person who has done nothing wrong. It is this struggle which compels the courts to rely heavily on policy arguments in an attempt to justify the outcome.
Prue Vines, ‘New South Wales v Lepore; Samin v Queensland; Rich v Queensland: Schools’ Responsibility for Teachers’ Sexual Assault: Non-Delegable Duty and Vicarious Liability’ [2003] MelbULawRw 22; (2003) 27(2) Melbourne University Law Review 612, 625. See also Paula Giliker, ‘Making the Right Connection: Vicarious Liability and Institutional Responsibility’ (2009) 17(1) Torts Law Journal 35, 35.
[34] Giliker criticises the House of Lords decision in Lister for this: Paula Giliker, ‘Rough Justice in an Unjust World’ (2002) 65 (March) Modern Law Review 269, 276. See also Nick McBride, Vicarious Liability Cases Mcbridesguides, 2 <https://mcbridesguides.files.wordpress.com/2014/09/vicarious-liability-cases.pdf> (emphasis altered): ‘Now — it seems that the House of Lords in Lister was completely incapable of grasping this point; that ... liability in ... situations [involving intentional misconduct by employees are] examples of personal liability rather than vicarious liability.’ McIvor also documents confusion in decisions between direct liability and vicarious liability: Claire McIvor, ‘The Use and Abuse of the Doctrine of Vicarious Liability’ (2006) 35(4) Common Law World Review 268, 288–96.
[35] Lister v Hesley Hall Ltd [2002] 1 AC 215, 232 [35] (Lord Clyde) (‘Lister’). See also E v English Province of Our Lady of Charity [2012] EWCA Civ 938; [2013] QB 722, where Ward LJ, after considering various suggested rationales for the doctrine and explaining deficiencies in each as explanations of current principle, concluded ‘no single rationale provides a complete answer for the imposition of vicarious liability’: at 761 [53]. His Lordship agreed with Fleming that the doctrine could only be seen as based on a combination of policy considerations: 761 [53] citing Carolyn Sappideen and Prue Vines (eds), Fleming’s The Law of Torts (Lawbook, 10th ed 2011). See also Justice Oliver Wendell Holmes Jr, ‘Agency II’ (1891) 5(1) Harvard Law Review 1, 22.
[36] Sweeney v Boylan Nominees Pty Ltd [2006] HCA 19; (2006) 226 CLR 161, 166–7 [11] (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ) (‘Sweeney’).
[37] Ibid.
[38] ‘[V]icarious liability has been regarded as an unstable principle, one for which a “fully satisfactory rationale for the imposition of vicarious liability” has been “slow to appear in the case law”’: Prince Alfred College [2016] HCA 37; (2016) 335 ALR 1, 8 [39] (French CJ, Kiefel, Bell, Keane and Nettle JJ), quoting Hollis v Vabu Pty Ltd [2001] HCA 44; (2001) 207 CLR 21, 37 [35] (Gleeson CJ, Gaudron, Gummow, Kirby and Hayne JJ) (‘Hollis’).
[39] See, eg, Glanville Williams, ‘Vicarious Liability and the Master’s Indemnity’ (1957) 20(3) Modern Law Review 220, 231: ‘[V]icarious liability is the creation of many judges who have had different ideas of its justification or social policy, or no idea at all. Some judges may have extended the rule more widely, or confined it more narrowly than its true rationale would allow.’ Justice Oliver Wendell Holmes Jr called vicarious liability an irrational fiction: Holmes, above n 35. 22–3; Baty dismissed attempts to find a rationale for vicarious liability as ‘hopeless groping’: Baty, above n 23, 148.
[40] Middleton [1795] EngR 220; (1699) 1 Salk 282; 91 ER 247, 248 (Holt CJ); Barwick [1867] UKLawRpExch 30; (1867) LR 2 Ex 259, 265–6 (Willes J); Kay v I.T.W. Ltd [1968] 1 QB 140; Branwhite v Worcester Works Finance Ltd [1969]
1 AC 552, 587 (Lord Wilberforce) (‘Branwhite’); New South Wales v Lepore [2003] HCA 4; (2003) 212 CLR 511, 554–5 [108]–[111] (Gaudron J) (‘Lepore’). Gaudron J also used the concept of estoppel in her agency analysis. Agency and its links with vicarious liability are explored in Gerard McMeel, ‘Philosophical Foundations of the Law of Agency’ (2000) 116(3) Law Quarterly Review 387.
[41] For example, in Deatons Pty Ltd v Flew [1949] HCA 60; (1949) 79 CLR 370 (‘Deatons’), Dixon J considered whether the employee’s actions were done under the employer’s ‘express or implied authority’ (at 381–2); Latham CJ referred to the ‘authority’ the employment conferred on the employee (at 378); McTiernan J referred to ‘implied authority’ (at 382); and Webb J referred to the employee’s ‘express or implied authority’ (at 388). Contrasting views appeared in Scott v Davis (2000) 204 CLR 333 (‘Scott’), where Gleeson CJ asked whether the employee was the defendant’s agent (at 341 [13]–[16]) and Gummow J denied vicarious liability was based on agency concepts (at 413 [239]). Justice McHugh used the concept of agency to clarify vicarious liability in Hollis [2001] HCA 44; (2001) 207 CLR 21,
51 [74]. Members of the High Court in Lepore also referred to and applied concepts of agency in applying vicarious liability: Lepore [2003] HCA 4; (2003) 212 CLR 511, 560 [128] (Gaudron J), 593–4 [235]–[239] (Gummow and Hayne JJ).
[42] Hern v Nichols (1709) 1 Salk 289, 289; [1795] EngR 1995; 91 ER 256, 256 (Holt CJ) (‘Hern’). Obviously, this argument is only sensible if the employer is not directly liable for negligence — for example, by employing the wrongdoer, by giving them inappropriate duties given their experience, or by not supervising appropriately. If the employer were directly liable to the victim/survivor in negligence, a rationale for making the institution liable vicariously (namely, because it is better that they should bear the loss than the victim) becomes redundant.
[43] Bazley v Curry 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 548–9 [22] (McLachlin J, for the Court) (‘Bazley’); Dubai Aluminium Co Ltd v Salaam [2003] 2 AC 366, 377 [21] (Lord Nicholls), 386 [65] (Lord Slynn agreeing), 386 [66] (Lord Hutton agreeing) (‘Dubai Aluminium’). In Lepore, some members of the Court apparently accepted enterprise risk as a satisfactory rationale for vicarious liability: Lepore [2003] HCA 4; (2003) 212 CLR 511, 612–13 [303] (Kirby J); while others did not: at 543 [65] (Gleeson CJ),
589 [223] (Gummow and Hayne JJ).
[44] Bazley 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 554 [31] (McLachlin J, for the Court).
[45] Lister [2002] 1 AC 215, 243 [65] (Lord Millett); Lepore [2003] HCA 4; (2003) 212 CLR 511, 614 [306] (Kirby J). See also Atiyah, above n 30, 27: ‘[L]oss distribution is the most rational justification that can be offered for vicarious liability today’.
[46] ‘The policy objective underlying vicarious liability is to ensure ... liability for tortious wrong is borne by a defendant with the means to compensate the victim’ where it is ‘fair, just and reasonable’: Vicarious Claimants v Catholic Child Welfare Society [2013] 2 AC 1, 15 [34] (Lord Phillips, for the Court) (‘Various Claimants’). Willes J refers in Limpus to the ‘deep pockets’ of the defendant: Limpus [1862] EngR 839; (1862) 1 Hurl & C 526, 539; [1862] EngR 839; 158 ER 993, 998. Glofcheski states that ‘introduction of the concept of vicarious liability into the law of tort was largely driven by the pragmatic consideration of finding a “deep pocket” to ensure that compensation is paid’: Rick Glofcheski, ‘A Frolic in the Law of Tort: Expanding the Scope of Employers’ Vicarious Liability’ (2004) 12(1) Tort Law Review 18, 20. See also Townshend-Smith: ‘the quality of these justifications [for vicarious liability is] ... sufficiently weak that the real practical justification for the doctrine is little more than that of finding a deep pocket ... by which to ensure that an innocent victim does not go uncompensated’: above n 5, 108.
[47] Bazley 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 554–5 [32]–[33] (McLachlin J, for the Court).
[48] Ibid 552–4 [29]–[31] (McLachlin J, for the Court); Various Claimants [2013] 2 AC 1, 27 [94] (Lord Phillips, for the Court); Cox v Ministry of Justice [2016] UKSC 10; [2016] AC 660, 675 [42] (Lord Reed, for the Court) (‘Cox’); Mohamud v Wm Morrison Supermarkets plc [2016] AC 677, 694 [47] (Lord Toulson, with whom Lord Neuberger, Baroness Hale, Lord Dyson MR and Lord Reed agreed) (‘Mohamud’); Lepore [2003] HCA 4; (2003) 212 CLR 511, 619 [320] (Kirby J), contra 626 [345] (Callinan J). Beuermann suggests a new rationale based on the employer’s power to direct the conduct of an employee, and the possibility of a conflict between the employee’s obligation to obey their employer, and the employee’s obligations under the general law: Christine Beuermann, ‘Dissociating the Two Forms of So-Called “Vicarious Liability”’ in Stephen GA Pitel, Jason W Neyers and Erika Chamberlain (eds), Tort Law: Challenging Orthodoxy (Hart Publishing, 2013) 463, 474. Though interesting, this rationale has not been accepted in any case law of which I am aware, so I do not discuss it in detail here.
[49] Sweeney [2006] HCA 19; (2006) 226 CLR 161, 167 [12] (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).
[50] In fact, that distinction was (at the very least) substantially minimised in Various Claimants [2013] 2 AC 1, 18 [47] and by McHugh J in Hollis [2001] HCA 44; (2001) 207 CLR 21, 57 [93].
[51] 1999 Can LII 692 (SCC); [1999] 2 SCR 534. See also G H L Fridman, ‘“The Course of Employment”: Policy or Principle?’ [2002] NewcLawRw 4; (2002) 6(1) Newcastle Law Review 61.
[52] 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 543.
[53] Ibid 544.
[54] Townshend-Smith has pointed out three difficulties with the concept of enterprise risk in terms of a rationale for vicarious liability: (1) as an economic concept it forces us to assume a value attached to avoidance of damage; (2) the concept is ill-suited to the public sector where the ‘costs’ cannot as easily be passed on to the ‘market’; and (3) economic analysis implies ability to calculate the tangible economic effects of a change in legal principle and put an economic value on the activities thereby affected: above n 5, 121–2.
[55] Bazley 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 567 [57] (McLachlin J, for the Court). See also: at 548–9 [22].
[56] Ibid 550 [24] (McLachlin J, for the Court). The ‘close connection’ test has been criticised for providing insufficient guidance on when liability should be imposed: Dubai Aluminium [2003] 2 AC 366, 377–8 [25]–[26] (Lord Nicholls), 386 [65] (Lord Slynn agreeing), 386 [66] (Lord Hutton agreeing); Giliker, above n 33, 39–41; and for ‘nudging open’ the ‘floodgates’ of liability: Glofcheski, above n 46, 39.
[57] Bazley 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 552–3 [29] (McLachlin J, for the Court) citing John G Fleming, The Law of Torts (LBC Information Services, 9th ed, 1998) 410.
[58] Ibid 554 [30] (McLachlin J, for the Court).
[59] Ibid 554–5 [32]–[33] (McLachlin J, for the Court).
[60] Ibid 558–9 [40] (McLachlin J, for the Court).
[61] Ibid 560 [42] (McLachlin J, for the Court) (emphasis altered).
[62] Ibid 560 [41] (McLachlin J, for the Court). The relevant factors included: (a) the opportunity the enterprise afforded the employee to abuse their power; (b) the extent to which the wrongful act may have furthered the employer’s aims (and, thus, the employee was more likely to commit them);
(c) the extent to which the wrongful act related to friction, confrontation or intimacy inherent in the employer’s business; (d) the extent of power conferred on the employee in relation to the victim/s; and (e) the vulnerability of the victim/s to the wrongful exercise of the employee’s power. Specifically, where the employer places the employee in a position of respect and trust, suggesting the child should emulate and/or obey the employee, the risk is enhanced and the more likely that abuse could be sheeted home to the employer: at 562 [44] (McLachlin J, for the Court).
[63] Jacobi v Griffiths [1999] 2 SCR 570 (Cory, Iacobucci, Major and Binnie JJ; McLachlin, L’Heureux-Dubé and Bastarache JJ dissenting) (‘Jacobi’).
[64] Ibid 589 [29] (Binnie J, for Cory, Iacobucci and Major JJ).
[65] Ibid 613 [71] (Binnie J, for Cory, Iacobucci and Major JJ). See also EB v Order of the Oblates of Mary Immaculate in the Province of British Columbia [2005] 3 SCR 45. Subsequently, the Supreme Court of Canada confirmed Jacobi does not mean non-profit organisations are never vicariously liable for their employees’ sexual assaults, and the old charitable immunity doctrine was no longer good law: John Doe v Bennett [2004] 1 SCR 436, 447–8 [24]. Indeed, generally there has been a trend towards no special rules for non-profits in terms of accountability: Elizabeth AdjinTettey, ‘Accountability of Public Authorities Through Contextualised Determinations of Vicarious Liability and Non-Delegable Duties’ (2007) 57 University of New Brunswick Law Journal 46; Jason W Neyers and David Stevens, ‘Vicarious Liability in the Charity Sector: An Examination of Bazley v Curry and Re Christian Brothers of Ireland in Canada’ (2005) 42(3) Canadian Business Law Journal 371; William I Innes, C Michael Kray and Brian J Burke, ‘Selected New Developments in the Liability of Directors and Officers of Charitable and Nonprofit Corporations’ (2004) 18(3) The Philanthropist 199. However, the Australian Royal Commission, in its Redress Report in recommending the introduction of a non-delegable duty of care on certain kinds of institutions regarding child sexual abuse, stated that the reform should not apply to non-profit providers: Redress Report, above n 1, 491.
[66] [2003] HCA 4; (2003) 212 CLR 511; see Vines, above n 33; Jane Wangmann, ‘Liability for Institutional Child Sexual Assault: Where Does Lepore Leave Australia?’ [2004] MelbULawRw 5; (2004) 28(1) Melbourne University Law Review 169.
[67] Commonwealth v Introvigne [1982] HCA 40; (1982) 150 CLR 258.
[68] [2003] HCA 4; (2003) 212 CLR 511, 533 [34] (Gleeson CJ), 598–603 [254]–[270] (Gummow and Hayne JJ), 624 [339]–[340] (Callinan J). Justice Kirby stated where vicarious liability was relevant, it, rather than principles of non-delegable duty, should be applied: at 609–10 [293]–[296].
[69] Ibid 559 [124] (Gaudron J), 565–6 [144]–[145] (McHugh J).
[70] Ibid 532 [31] (Gleeson CJ), 624 [339] (Callinan J agreeing on the point), 599 [256] (Gummow and Hayne JJ). Justice Gaudron’s discussion of non-delegable duties referred only to negligence claims: at 551–3 [99]–[105]. Justice McHugh found non-delegable duties could apply to intentional conduct (at 562). Justice Kirby left the matter open, but seemed to indicate agreement with the position of McHugh J on that point: at 609–10 [293]–[296], at least in cases where vicarious liability was not applicable, though he clearly disagreed with Justice McHugh on other points (610). See also Vines, above n 33, 614–16. The Redress Report recommended reintroduction of a non-delegable duty of care on institutions for deliberate criminal activity in the context of child sexual abuse: Redress Report, above n 1, 495.
[71] Lepore [2003] HCA 4; (2003) 212 CLR 511, 537 (Gleeson CJ).
[72] Ibid 539 [51].
[73] Ibid 540 [54] (Gleeson CJ).
[74] Ibid.
[75] Ibid 546 [72]–[74] (Gleeson CJ). Relevant factors would include the age of the students, their vulnerability (if any), the tasks allocated to teachers, the number of adults who were responsible for the students, and the nature and circumstances of the offending.
[76] Ibid 543 [65] (Gleeson CJ).
[77] Ibid 547 [78] (Gleeson CJ).
[78] This was also the position of McHugh J, because His Honour applied the non-delegable duty principle, for which questions of employee versus independent contractor were irrelevant: Ibid 566 [146].
[79] Ibid 561 [130]–[131] (Gaudron J).
[80] Ibid 561 [131] (Gaudron J).
[81] Ibid 587 [218] (Gummow and Hayne JJ).
[82] Ibid 589 [223] (Gummow and Hayne JJ).
[83] Ibid 594 [239] (Gummow and Hayne JJ).
[84] Deatons [1949] HCA 60; (1949) 79 CLR 370, 381.
[85] Lepore [2003] HCA 4; (2003) 212 CLR 511, 594 [242] (Gummow and Hayne JJ).
[86] Ibid 612–13 [302]–[303].
[87] Ibid 618 [320] (Kirby J).
[88] Ibid 619 [320] (Kirby J).
[89] Ibid 621 [326] (Kirby J).
[90] Ibid 621 [327]–[328] (Kirby J).
[91] Ibid 625 [342].
[92] Ibid 626 [345] (Callinan J).
[93] [2016] HCA 37; (2016) 335 ALR 1.
[94] A, DC v Prince Alfred College Inc [2015] SASC 12 (4 February 2015).
[95] Prince Alfred College [2016] HCA 37; (2016) 335 ALR 1, 3–4 [8] (French CJ, Kiefel, Bell, Keane and Nettle JJ),
25 [125] (Gageler and Gordon JJ).
[96] Ibid 4 [10] (French CJ, Kiefel, Bell, Keane and Nettle JJ), 26 [128] (Gageler and Gordon JJ).
[97] Ibid 8 [40] (French CJ, Kiefel, Bell, Keane and Nettle JJ).
[98] Ibid 14 [68] (French CJ, Kiefel, Bell, Keane and Nettle JJ).
[99] Ibid 17 [80]–[81] (French CJ, Kiefel, Bell, Keane and Nettle JJ), 26 [130] (Gageler and Gordon JJ agreeing).
[100] Ibid 18 [84] (French CJ, Kiefel, Bell, Keane and Nettle JJ), 26 [130] (Gageler and Gordon JJ agreeing).
[101] Ibid 17 [83] (French CJ, Kiefel, Bell, Keane and Nettle JJ) discussing Mohamud [2016] AC 677.
[102] [1949] HCA 60; (1949) 79 CLR 370.
[103] Ibid 382.
[104] [2002] 1 AC 215.
[105] Ibid 229 [23] (Lord Steyn), 238 [52] (Lord Hutton agreeing), 237 [48] (Lord Clyde), 245 [70]
(Lord Millett).
[106] Ibid 230 [28] (Lord Steyn), 238 [52] (Lord Hutton agreeing).
[107] Ibid 226 [20] (Lord Steyn), 238 [52] (Lord Hutton agreeing), 244 [66] (Lord Millett); this was also noted in Mohamud [2016] AC 677, 687 [26] (Lord Toulson, with whom Lord Neuberger, Baroness Hale, Lord Dyson MR and Lord Reed agreed).
[108] Lister [2002] 1 AC 215, 232 [35]: ‘I am not persuaded that there is any reason of principle or policy which can be of substantial guidance in the resolution of the problem of applying the rule
[of vicarious liability] in any particular case.’
[109] Ibid 243 [65].
[110] Ibid 235 [44]–[45] (Lord Clyde), 241 [59] (Lord Hobhouse) and 244 [65] (Lord Millett).
[111] Ibid 238 [54] (Lord Hobhouse).
[112] Ibid 243 [62] (Lord Hobhouse).
[113] Various Claimants [2013] 2 AC 1, 15 [34] (Lord Phillips, for the Court). The school managers had been found vicariously liable for the abuse; this was not challenged on appeal.
[114] Ibid 15 [35] (Lord Phillips, for the Court). Subsequently in Cox [2016] UKSC 10; [2016] AC 660, Lord Reed (for the Court) indicated the second, third and fourth of these criteria were of most relevance: at 669–70 [20]–[23]. In terms of discussion below, the concept of ‘fair just and reasonable’ is critiqued in Part III(A), the first of Lord Phillips’ criteria is considered in Part III(B), and the second, third and fourth considered under Part III C. The final criterion is a standard traditional test for distinguishing between an employee and an independent contractor, a distinction also questioned in Part III(C).
[115] Various Claimants [2013] 2 AC 1, 27 (Lord Phillips, for the Court).
[116] Cox [2016] UKSC 10; [2016] AC 660; Mohamud [2016] AC 677.
[117] Mohamud [2016] AC 677.
[118] It is conceded that, to some extent, the rationales overlap.
[119] Bazley 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 552–3 [23] (McLachlin J, for the Court); Cox [2016] UKSC 10; [2016] AC 660, 675 [42] (Lord Reed, for the Court); Lepore [2003] HCA 4; (2003) 212 CLR 511, 619 [320] (Kirby J).
[120] ‘The imprecise concepts of fairness, justice and reasonableness are central to the law of negligence. The test for the existence of a duty of care is whether it is fair, just and reasonable to impose such a duty’: Mohamud [2016] AC 677, 695 [54] (Lord Dyson MR).
[121] Caparo Industries plc v Dickman [1990] UKHL 2; [1990] 2 AC 605, 617–18 (Lord Bridge).
[122] [1977] UKHL 4; [1978] AC 728, 751–2 (Lord Wilberforce).
[123] Cooper v Hobart [2001] 3 SCR 537, [14].
[124] Pyrenees Shire Council v Day [1998] HCA 3; (1998) 192 CLR 330, 419. Justice Kirby abandoned the three-stage approach in Graham Barclay Oysters Pty Ltd v Ryan [2002] HCA 54; (2002) 211 CLR 540, 626 [238].
[125] (2001) 207 CLR 562, 579 (the Court). Similarly, in the context of vicarious liability, see Callinan J in Lepore [2003] HCA 4; (2003) 212 CLR 511, 626 [345]: ‘cases would, as a practical matter, be decided according to whether the judge or jury thought it “fair and just” to hold the employer liable. Perceptions of fairness vary greatly. The law in consequence would be thrown into a state of uncertainty.’
[126] McIvor, above n 34, 279.
[127] [2016] HCA 37; (2016) 335 ALR 1, 9 [45] (French CJ, Kiefel, Bell, Keane and Nettle JJ). The majority observed that:
if a general principle provides that liability is to depend upon a primary judge’s assessment of what is fair and just, the determination of liability may be rendered easier, even predictable. But principles of that kind depend upon policy choices and the allocation of risk, which are matters upon which minds may differ. They do not reflect the current state of the law in Australia and the balance sought to be achieved by it in the imposition of vicarious liability.
[128] Limpus [1862] EngR 839; (1862) 1 Hurl & C 526, 539; [1862] EngR 839; 158 ER 993, 998 (Willes J):
there is virtually no remedy against the driver of an omnibus, and therefore it is necessary that, for injury resulting from an act done by him in the course of his master’s service, the master should be responsible; for there ought to be a remedy against some person capable of paying damages to those injured by improper driving.
[129] (1709) 1 Salk 289, 289; [1795] EngR 1995; 91 ER 256, 256: ‘seeing somebody must be a loser by this deceit, it is more reason that he that employs and puts a trust and confidence in the deceiver should be a loser, than a stranger ...’.
[130] Lloyd [1912] AC 716, 727 (Earl of Halsbury) citing Hern with evident approval, 738 (Lord Macnaghten), 739 (Lord Atkinson agreeing).
[131] Bugge v Brown [1919] HCA 5; (1919) 26 CLR 110, 117 (Isaacs J) (‘Bugge’): ‘it is more just to make the person who has entrusted his servant with the power of acting in his business responsible for injury occasioned to another in the course of so acting, than that the other and entirely innocent party should be left to bear the loss’.
[132] For example, in Various Claimants [2013] 2 AC 1, Lord Phillips noted for the Court that ‘the policy objective underlying vicarious liability is to ensure [so far as was just] that liability for tortious wrongs is borne by a defendant with the means to compensate the victim’: at 15; and in earlier times ‘there ought to be a remedy against some person capable of paying damages to those injured by improper driving’: Limpus [1862] EngR 839; (1862) 1 Hurl & C 526, 539; [1862] EngR 839; 158 ER 993, 998 (Willes J).
[133] Baty, above n 23, 154.
[134] See above n 64 and accompanying text.
[135] [1960] HCA 9; (1960) 103 CLR 215, 229 (Dixon CJ, Kitto and Windeyer JJ).
[136] Williams, above n 39, 232 (citations omitted).
[137] ‘[A]s a general proposition it has not, I think, been questioned for nearly 200 years that in determining the rights inter se of A and B the fact that one or other of them is insured is to be disregarded’: Lister v Romford Ice & Cold Storage Co Ltd [1956] UKHL 6; [1957] AC 555, 576–7 (Viscount Simonds). See also Hollis [2001] HCA 44; (2001) 207 CLR 21, 68 [115]–[117] (Callinan J).
[138] Scott (2000) 204 CLR 333, 419 [254] (Gummow J), 439 [310] (Hayne J) and 454 [347] (Callinan J).
[139] ‘[T]hat there is no broad formulation of where and why liability insurance or insurability should be a factor relevant to the imposition and shape of liability, suggests that commentators deploy it selectively, and therefore unconvincingly, in the normative debate just as they do in the descriptive debate’: Jane Stapleton, ‘Tort, Insurance and Ideology’ (1995) 58(6) Modern Law Review 820, 829.
[140] Atiyah, above n 30, 27.
[141] Douglas, above n 23, 591.
[142] Ibid. It is conceded that this sometimes, in fact, happens. However, it is logically difficult to accept, and when efforts are made to reconceptualise legal principle, illogical premises are best avoided if possible.
[143] Dubai Aluminium [2003] 2 AC 366, 377 [21] (Lord Nicholls), 386 [65] (Lord Slynn agreeing),
386 [66] (Lord Hutton agreeing); Cox [2016] UKSC 10; [2016] AC 660, 672 [29] (Lord Reed, for the Court).
[144] Lepore [2003] HCA 4; (2003) 212 CLR 511, 612–13 [302]–[303] (Kirby J).
[145] The High Court in obiter dicta in Prince Alfred College [2016] HCA 37; (2016) 335 ALR 1 based vicarious liability on whether the employment not only provided an opportunity, but also, the ‘occasion for the commission of the wrongful act’: at 17 [80] (French CJ, Kiefel, Bell, Keane and Nettle JJ), 26 [130] (Gageler and Gordon JJ agreeing). Gageler and Gordon JJ, who specifically adopted the factors identified in the Canadian and UK case law as relevant to vicarious liability in the context of child sexual abuse, denied the tests and principles said to underpin those decisions (including enterprise risk): at 26 [130]. Some might quibble with ‘cherry picking’ factors identified in case law in comparable jurisdictions, but not the theoretical basis said to underpin them.
[146] Duncan v Findlater [1839] EngR 1005; (1839) 6 Cl & Fin 894, 910; [1839] EngR 1005; 7 ER 934, 940 (Lord Brougham); Broom v Morgan [1953] 1 QB 597, 608 (Denning LJ); Atiyah, above n 30, 17–18. Baty referred to ‘profit’ as one of nine possible rationales for imposing vicarious liability: Baty, above n 23, 148. Others were control, revenge, carefulness and choice, identification, evidence, indulgence, danger and satisfaction.
[147] Guido Calabresi, ‘Some Thoughts on Risk Distribution and the Law of Torts’ (1961) 70(4) Yale Law Journal 499.
[148] Jacobi [1999] 2 SCR 570, 615–16 [75] (Binnie J, for Cory, Iacobucci and Major JJ); Townshend-Smith, above n 5, 121–2.
[149] Alan O Sykes, ‘The Boundaries of Vicarious Liability: An Economic Analysis of the Scope of Employment Rule and Related Legal Doctrines’ (1988) 101(3) Harvard Law Review 563, 573.
[150] Ibid 574.
[151] Ibid 575.
[152] Ibid 579.
[153] Lewis Klar, ‘Vicarious Liability’ in Carolyn Sappideen and Prue Vines (eds), Fleming’s The Law of Torts (Thomson Reuters, 10th ed, 2011) 459.
[154] J W Neyers, ‘A Theory of Vicarious Liability’ (2005) 43(2) Alberta Law Review 287, 298–9; Calabresi, above n 147, 543–7.
[155] Young B Smith, ‘Frolic and Detour’ (1923) 23(5) Columbia Law Review 444, 461.
[156] [2003] HCA 4; (2003) 212 CLR 511, 586–9 [214]–[223] (Gummow and Hayne JJ).
[157] 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 551–2 [29] citing John G Fleming, The Law of Torts (LBC Information Services, 9th ed, 1998) 410.
[158] Ibid 554 [32] (McLachlin J, for the Court).
[159] [2003] HCA 4; (2003) 212 CLR 511, 587 [219].
[160] McIvor, above n 34, 275.
[161] Holmes, above n 35, 22–3. The High Court of Australia recently suggested the way forward was on a case-by-case basis, rather than elucidation of general principle: Prince Alfred College [2016] HCA 37; (2016) 335 ALR 1, 9–10 [46] (French CJ, Kiefel, Bell, Keane and Nettle JJ), 26 [128] (Gageler and Gordon JJ).
[162] Holmes, above n 16. See also Paula J Dalley, ‘All in a Day’s Work: Employers’ Vicarious Liability for Sexual Harassment’ (2002) 104(3) West Virginia Law Review 517, 527–8:
[T]he law of torts fails to provide an adequate explanation for vicarious liability because it is not a tort law doctrine ... Vicarious liability ... is based in the agency relationship itself and is not dependent on tort principles such as fault or on tort policies such as accident prevention.
[163] [1931] HCA 53; (1931) 46 CLR 41.
[164] Ibid 49 (Dixon J), 47 (Rich J).
[165] Ibid 46.
[166] Bazley 1999 Can LII 692 (SCC); [1999] 2 SCR 534, 546 (McLachlin J, for the Court) (citations omitted) (emphasis altered):
The cases confirming vicarious liability on the basis that the employee was acting in furtherance of the employer’s aims rely on the agency rationale implicit in the Salmond test. Because the employee was acting in furtherance of the employer’s aims, he or she is said to have ‘ostensible’ or ‘implied’ authority to do the unauthorized act. This rationale works well enough for torts of negligent accident.
[167] This occurred (indirectly) in Branwhite [1969] 1 AC 552, 587 where he wondered whether ‘some wider conception of vicarious responsibility other than that of agency ... may have to be recognised’.
[168] (2000) 204 CLR 333, 345–70 [30]–[110] (dissenting in the result). Gummow J specifically found that although there was a relationship between tort and agency, agency did not provide the theoretical basis of vicarious liability: at 413 [239].
[169] [2001] HCA 44; (2001) 207 CLR 21, 57 [93].
[170] [2006] HCA 19; (2006) 226 CLR 161, 186–8 (dissenting in the result).
[171] [2003] HCA 4; (2003) 212 CLR 511, 560 [128].
[172] Ibid 594 [239] (emphasis added), concluding that an employer would be vicariously liable for intentional torts of an employee in two circumstances, one of which was where ‘the conduct of which complaint is made was done in the ostensible pursuit of the employer’s business or the apparent execution of the authority which the employer held out the employee as having’. Of course, both the words ostensible and apparent describe types of agency.
[173] International Harvester Co of Australia Pty Ltd v Carrigan’s Hazeldene Pastoral Co [1958] HCA 16; (1958) 100 CLR 644, 652 (the Court).
[174] American Law Institute, Restatement (Third) of Agency (2006).
[175] Ibid § 7.07(1).
[176] Ibid § 7.07(2).
[177] Ibid.
[178] [1949] HCA 60; (1949) 79 CLR 370, 379 (Latham CJ). See also: at 381 (Dixon J).
[179] American Law Institute, Restatement (Third) of Agency (2006) § 7.08.
[180] Christine Beuermann, ‘Conferred Authority Strict Liability and Institutional Child Sexual Abuse’ [2015] SydLawRw 5; (2015) 37(1) Sydney Law Review 113.
[181] Ibid 128.
[182] Ibid 128–9.
[183] Ibid 128.
[184] Ibid; Christine Beuermann, ‘Vicarious Liability and Conferred Authority Strict Liability’ (2013) 20(3) Torts Law Journal 265.
[185] Beuermann, above n 180, 124–6.
[186] Redress Report, above n 1, 495.
[187] Beuermann, above n 48.
[188] Ibid 468.
[189] Ibid 470–71.
[190] Ibid 475.
[191] Richard Kidner, ‘Vicarious Liability: For Whom Should the “Employer” be Liable?’ (1995) 15(1) Legal Studies 47; Ewan McKendrick, ‘Vicarious Liability and Independent Contractors — A ReExamination’ (1990) 53(6) Modern Law Review 770.
[192] See, eg, Century Insurance Co Ltd v Northern Ireland Road Transport Board [1942] UKHL 2; [1942] AC 509; Rose v Plenty [1975] EWCA Civ 5; [1976] 1 WLR 141.
[193] Paula J Dalley, ‘A Theory of Agency Law’ (2011) 72(3) University of Pittsburgh Law Review 495, 528: ‘Under the [Second Restatement’s] test [requiring] that conduct be “actuated, at least in part, by a purpose to serve the master,” masters will not be liable for most intentional torts committed by their servants.’
[194] Beuermann, above n 180, 130.
[195] Ibid 129.
[196] [1912] AC 716.
[197] [1867] UKLawRpExch 30; (1867) LR 2 Ex 259, 265 (Willes J) (emphasis added). See also Cheshire v Bailey [1904] UKLawRpKQB 203; [1905] 1 KB 237, 241 (Collins MR). See Laski, above n 29, 120, agreeing that a master should not be liable for acts of servants involving ‘objects demonstrably and entirely his [sic] own’.
[198] Bugge [1919] HCA 5; (1919) 26 CLR 110, 118 (Isaacs J) (citations omitted).
[199] Lloyd [1912] AC 716, 724. In fact, it may be better to see the case as a breach of contract case, rather than an exemplar of vicarious liability. Neyers reads the case as one where the employer was found to be personally liable for a breach of contract, rather than vicarious liability: Neyers, above n 154, 315.
[200] Lloyd [1912] AC 716, 727 (Earl of Halsbury).
[201] Ibid 738.
[202] Ibid 739 (Lord Macnaghten).
[203] Donoghue v Stevenson [1932] AC 562.
[204] Alan J Oxford II, ‘When Agents Attack: Judicial Misinterpretation of Vicarious Liability Under “Aided in Accomplishing the Tort by the Existence of the Agency Relation” and Restatement 3rd’s Failure to Properly “Restate” the Ill-Fated Section 219(2)(d) Provision’ (2012) 37(2) Oklahoma City University Law Review 157, 199–200 (citations omitted).
[205] Prince Alfred College [2016] HCA 37; (2016) 335 ALR 1, 16–17 [80]–[81] (French CJ, Kiefel, Bell, Keane and Nettle JJ), 26 [130] (Gageler and Gordon JJ agreeing).
[206] [2016] AC 677. It is unclear whether the High Court would apply Mohamud given its decision in Deatons [1949] HCA 60; (1949) 79 CLR 370. The UK Supreme Court in Mohamud indicated disagreement with Deatons: [2016] AC 677, 688 [30] (Lord Toulson, with whom Lord Neuberger, Baroness Hale, Lord Dyson MR and Lord Reed agreed). Five members of the High Court indicated disagreement with Mohamud in Prince Alfred College: [2016] HCA 37; (2016) 335 ALR 1, 15 [72]–[73] (French CJ, Kiefel, Bell, Keane and Nettle JJ).
[207] Byrd v Faber, 565 NE 2d 584 (Ohio, 1991); Iglesia Cristiana La Casa Del Señor Inc v LM, 783 So 2d 354 (3D Fla, 2001); Primeaux v United States, [1999] USCA8 502; 181 F 3d 876 (8th Cir, 1999); Thompson v Everett Clinic, 860 P 2d 1054 (Ct App Wash, 1993); Cooke v Stefani Management Services Inc, [2001] USCA7 257; 250 F 3d 564, 569–70 (7th Cir, 2001).
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/SydLawRw/2017/8.html