Vicarious liability in the UK Supreme Court and High Court of Australia

Introduction

Vicarious liability imposes strict liability on one defendant (the second defendant) for the torts of the first defendant where:

1. (1) Stage One: the relationship between the second defendant and the first defendant is one which makes it proper for the law to make the one pay for the fault of the other; and

2. (2) Stage Two: the tort committed by the first defendant is sufficiently connected with, or linked to, the parties’ relationship that it can be regarded as committed in the course of the first defendant's employment.¹

Both stages of this test have been addressed by the UK and Australian courts in recent years as they seek to determine the scope of vicarious liability. At Stage One, Australia, unlike the UK,² continues to refuse to extend the relationship test to those ‘akin’ to employees.³ Different approaches have also been taken to the Stage Two test. The UK Supreme Court (UKSC) in Morrison Supermarkets plc v Various Claimants ⁴ held in 2020 that the tort must be so closely connected with acts the employee was authorised to do that it may fairly and properly be regarded as done by him while acting in the ordinary course of his employment. In contrast, in 2016 in Prince Alfred College Inc v ADC,⁵ the High Court of Australia (HCA) rejected the UK ‘close connection’ test. Instead, it favoured a test that enquired whether the apparent performance of any special role that the employer had assigned to the employee may be said to give the ‘occasion’ for the wrongful act, given the position in which the employee was thereby placed vis-à-vis the victim.

This commentary will focus on two vicarious liability decisions delivered in 2023: the UKSC's ruling in Trustees of the Barry Congregation of Jehovah's Witnesses v BXB ⁶ and that of the HCA in CCIG Investments Pty Ltd v Schokman. ⁷ Both are significant judgments which will have impact across the common law world.⁸ They also address a key question: to what extent should cases of sexual abuse give rise to a distinctive test for course of employment? While not a sexual abuse case, Lord Reed argued, obiter, in Morrison that the close connection test had been applied differently in cases concerned with the sexual abuse of children,⁹ and that, in such cases, a ‘tailored’ version of the close connection test is applied.¹⁰ Australia in Prince Alfred College also suggested that in applying the course of employment test ‘in cases of this kind’ (sexual abuse), particular features needed to 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.¹¹

BXB and Schokman both address the extent to which a ‘tailored’ test should be applied in sexual abuse cases, while setting out revised versions of the Stage 2 test. BXB also re-examines the ‘akin to employment’ test. Schokman is particularly significant in providing a fresh appraisal of the Stage 2 test in the contexts of negligence and intentional torts.

1. BXB in the UK Supreme Court

(a) The decision

Mrs B had been raped by one of the elders (Sewell) of the Jehovah's Witnesses congregation she attended with her husband and children. Her religion encouraged her to associate with other witnesses, and the families of Mrs B and Sewell had been good friends. The rape had taken place at the home of Sewell and his wife after the families had returned there with their children, having spent the morning together engaged in religious activity followed by a family lunch. At the time of the rape, Sewell was not engaged in any religious activity. However, Mrs B argued that their shared religion made her vulnerable to his power as an elder. The Court of Appeal accepted that the authority of the elders would extend to all aspects of the lives of congregation members. Mrs B argued that but for this, she and her husband would probably have ended their friendship with Sewell by the time of the rape and the rape would not have occurred.

The Court of Appeal, influenced by Lord Reed's dicta in Morrison, accepted that the Stage Two test for vicarious liability should be ‘tailored’ for cases of sexual abuse, including those involving adults. Nicola Davies LJ, giving the leading judgment, argued that ‘in cases of this kind, the test is more open textured and requires an analysis of all aspects of the relationship between the tort and the abuser's status’.¹² She found that the defendant religious organisation had created and/or significantly enhanced the risk of sexual abuse:

On the facts of this claim, what is relevant for the purpose of the close connection test is the conferral of authority by the Jehovah's Witness organisation upon its elders, coupled with the opportunity for physical proximity as between an elder and [members of] the congregation.¹³

The Court held that the authority conferred on Sewell as an elder encouraged Mrs B to trust him and not break off their friendship and that the power it engendered made it just and reasonable for the defendants to be held vicariously liable for his act in raping Mrs B.

This ‘tailored’ test may be contrasted with the general position taken in Morrison that a discussion of policy factors such as enterprise risk¹⁴ were not to the point when applying the close connection test.¹⁵ Rather, advised the Court in Morrison, in applying the close connection test it was necessary to have regard to the assistance provided by previous court decisions to ensure that cases can be decided on a basis which is principled and consistent. The tailored test also conflicts, as Silink and Ryan¹⁶ point out, with statements in Lister v Hesley Hall Ltd ¹⁷ that sexual abuse cases should be approached in the same way as any other case where questions of vicarious liability arise. It further presupposes that a clear line can be drawn between sexual abuse and other tortious actions. Natasha Armes in Armes v Nottinghamshire CC,¹⁸ for example, suffered appalling physical and sexual abuse. The Court quite rightly did not distinguish between these different forms of abuse. Physical abuse will at times involve abuse of authority and power. Can it logically be subjected to a different test? A distinct ‘tailored’ test, in the manner interpreted by the Court of Appeal, raises more questions than answers.

The UK Supreme Court in April 2023 agreed. Lord Burrows, giving judgment in BXB on behalf of the Court (which included Lord Reed), denied the existence of a distinct test for sexual abuse cases. In his view, Lord Reed in Morrison had simply acknowledged the existence of criteria that were particularly relevant in sexual abuse cases, such as the fact that abusers had been placed in a position where they were able to exercise authority over the victims.¹⁹ Lord Reed, further, in Cox v Ministry of Justice ²⁰ had indicated that sexual abuse of children was not a special category of case. On this basis, Lord Burrows argued:

The same two stages, and the same two tests, apply to cases of sexual abuse as they do to other cases on vicarious liability … The idea that the law still needs tailoring to deal with sexual abuse cases is misleading. The necessary tailoring is already reflected in, and embraced by, the modern tests.²¹

While the Morrison test, therefore, is the sole Stage Two test, different factors may be relevant in each case.

Lord Burrows concluded that the Court of Appeal had been in error in constructing its own ‘tailored’ test and this had led it to consider factors irrelevant to the Morrison test.²² His Lordship gave six reasons why the close connection test was not satisfied on the facts. These included Sewell not carrying out religious activities at the time of the rape and that he had not used his position to groom the victim. It was a shocking one-off attack. Fundamentally Sewell had not abused his position as an elder, but his position as a close friend of Mrs B.

The case also involved the Stage One question – was an elder ‘akin’ to an employee of the religious organisation? This was more straightforward and can be dealt with briefly. The Supreme Court approved the Court of Appeal's ruling that the elders could not be said to be carrying out a business on their own account. They were integral to the organisation's activities.²³ Sewell had carried out work on behalf of the religious organisation and in the furtherance of its aims and objectives.²⁴ He was appointed by a designated process and there was a hierarchical structure into which his role as elder fitted.

(b) Reflections

This is a deliberately narrow reading of the law, drawing on the tests set out in Barclays Bank and Morrison in 2020 and focusing on the facts of the case. The obiter suggestion of Lord Reed in Morrison of a ‘tailored’ test for sexual abuse claims is rejected by reference to the words of Lord Reed in Cox. Discussion of the controversial 2016 case of Mohamud ²⁵ should also be noted. Lord Burrows described the decision as an application of the close connection test, as stated in Lister and Dubai Aluminium. ²⁶ This replicates the exposition of the test in Morrison which sought to correct any ‘misunderstandings’ of Lord Toulson's judgment in Mohamud.²⁷ Thus corrected, we are now advised that Mohamud's main relevance is in indicating that courts might find it helpful to focus on the functions or ‘field of activities’ entrusted to the employee. Lord Burrows did not comment further. While, therefore, the UKSC seems resolute not to overturn Mohamud, its precedential value is now premised on a very narrow reading of the case.

The key paragraph of BXB is, however, [58] where Lord Burrows set down his summary of the modern law of vicarious liability: ((i)–(v)). This bears re-reading and will set the agenda for future cases. Paragraph [58](ii) addresses the ‘akin to employment’ test. His Lordship emphasised that the traditional position – that there is no vicarious liability for a ‘true’ independent contractor– stands. The relevant factors he listed for determining whether a relationship is ‘akin to employment’ resemble those of Lord Phillips in CCWS:

• • whether the work is being paid for in money or in kind;

• • how integral to the organisation is the work carried out by the tortfeasor;

• • the extent of the defendant's control over the tortfeasor in carrying out the work;

• • whether the work is being carried out for the defendant's benefit or in furtherance of the aims of the organisation;

• • what the situation is with regard to appointment and termination; and

• • whether there is a hierarchy of seniority into which the relevant role fits.

We may note greater reference here to the characteristics of a contract of employment (eg remuneration; appointment/termination provisions). There is, unlike CCWS, no reference to the deeper pocket argument nor any mention of enterprise risk. These factors self-evidently favour practical resemblances to the contract of employment over policy.²⁸

Paragraph [58](iii) articulates the close connection test using the Morrison definition with minor adjustments to include ‘quasi/akin to employment’ and to remove the word ‘ordinary’ before course of employment in order to include abuse cases. The revised test thus reads:

whether the wrongful conduct was so closely connected with acts that the tortfeasor was authorised to do that it can fairly and properly be regarded as done by the tortfeasor while acting in the course of the tortfeasor's employment or quasi-employment.

What it does not permit is a causal ‘but for’ test.

More generally, Lord Burrows reminds us (in paragraph [58](iv)–(v)) that, in the vast majority of cases, both the relationship and connection tests should be applied without considering the underlying policy justifications for vicarious liability.²⁹ When facing doubtful or difficult cases, Lord Burrows advised that it may nevertheless be a useful final check on the justice of the outcome to stand back and consider whether the outcome indicated is consistent with underlying policy.³⁰ And that policy?

At root the core idea … appears to be that the employer or quasi-employer, who is taking the benefit of the activities carried on by a person integrated into its organisation, should bear the cost (or, one might say, should bear the risk) of the wrong committed by that person in the course of those activities.³¹

This, he described as a policy of enterprise liability or risk, drawing on Lord Reed's statements in Cox and Armes. However, it is clearly meant as an afterthought. The formulation is also more reminiscent of the work of Atiyah³² than the law and economics-based analysis of Bazley v Curry. ³³

BXB amounts to a further attempt by the UKSC to slow down the momentum of vicarious liability and reinstate guiding principles to provide greater certainty and predictability to the law. Policy is confined to a final check after the two-stage test has been applied. The close connection test is refined yet again with the Supreme Court favouring a dissection of Sewell's functions and activities as an elder over broad notions of risk creation. Morrisons and Barclays Bank in 2020 sought the same goals. Will BXB be more successful?

2. Schokman in the High Court of Australia

(a) The decision

The facts of Schokman are memorable, if not for the right reasons. The plaintiff and another employee (Hewett) were employed to work at a restaurant on the defendant's island resort and spa. They were required under their contract of employment to live in shared staff accommodation provided by the defendant company. In the early hours of the morning, Schokman was awoken by a drunken Hewett standing over him and urinating on his face. Schokman suffered a cataplectic attack as a result. Was this negligent act in the course of Hewett's employment? The Queensland Court of Appeal (QCA) focused on the terms of the employment contract. It was a term of the contract that the parties reside in shared accommodation, and this required employees to take reasonable care that their acts did not adversely affect the health and safety of other persons. On this basis, Hewett ‘was not occupying the room as a stranger, but instead as an employee, pursuant to and under the obligations of his employment contract. There was in this case the requisite connection between his employment and the employee's actions.’³⁴

The High Court overturned this very generous decision and argued that the QCA's focus should not have been on the terms of the contract of employment but on whether there was a connection between what Hewett was employed to do and his negligent act, looking to decided cases for guidance.

In other words, the QCA should have examined:

whether the tortious act in question has a sufficiently strong connection with the employment, and what is entailed in it, so as to be said to have been done in the course of that employment.³⁵

Reference to ‘connection’ might be regarded as surprising given that seven years earlier, the HCA in Prince Alfred College had argued that a ‘test of connection does not seem to add much to an understanding of the basis for an employer's liability’,³⁶ favouring a test of ‘occasion’.³⁷ The majority in Schokman ³⁸ did not, however, regard the two tests as necessarily distinct:

where an employee is placed in a special position … so that the act in question may be seen as one to which the ostensible performance of the employer's work by the employee ‘gives occasion’ … the requisite connection would be present.³⁹

The connection, however, would be too tenuous where the employment only provided an opportunity for the tort to take place.⁴⁰

The majority did accept, however, that in cases of sexual abuse something more than the sufficiency of the connection between the wrongful act and the employment might be necessary to better explain the basis for an employer's liability. Here reference to the special role assigned to the employee who was the abuser might well be helpful.⁴¹ Features such as authority, power, trust, control, and the ability to achieve intimacy would indicate a strong connection between the employment and the wrongful act. In common then with BXB above, sexual abuse cases are seen as an application of the general Stage Two test where specific additional factors are relevant.

On the facts Hewett had been assigned no role in relation to his roommate – shared physical proximity merely provided the opportunity for his drunken actions, which were not in the course of his employment: ‘In truth, [they] had no real connection to it.’⁴²

The judgment of Edelman and Steward JJ, while agreeing with the result, offers a more nuanced exploration of the relationship between agency, vicarious liability, and non-delegable duties. The reason why vicarious liability is confusing, they argue, is because it conflates different areas of law. Agency cases (which, of course, involve primary liability) should therefore be distinguished from vicarious liability (secondary liability) where the employee's wrongful acts have to be sufficiently or closely connected to the employee's duties or powers of employment so that they could be said to have been performed in the course of their employment.⁴³ Drawing on the history of vicarious liability, they argue that while based upon the same principles as agency (which makes the principal liable through the attribution of acts), vicarious liability has developed as a distinct form of liability based upon an attribution of liability to the employer.⁴⁴

More radically, Edelman and Steward JJ questioned whether institutional sexual abuse cases such as Prince Alfred College might better be regarded as non-delegable duties cases.⁴⁵ This would involve overturning authority to the contrary (NSW v Lepore)⁴⁶ and rewriting the approach taken in Prince Alfred College itself. While worthy of greater academic consideration than I have space for here,⁴⁷ this suggestion was not adopted by the rest of the court.

(b) Reflections

The judgments in Schokman are united, then, in finding that Hewett's act of negligent urination was not sufficiently connected to his employment. The word ‘occasion’ which featured so prominently in Prince Alfred College is used only nine times; five of these feature in Gleeson J's judgment, which expressed concern that the difference between ‘occasion’ and the mere ‘opportunity’ to commit the act (which does not give rise to vicarious liability) is not ‘easy to identify’.⁴⁸ BXB is cited seven times and we might infer that its more restrictive view of the law has reassured the HCA that the term ‘connection’ will not necessarily permit uncontrolled liability on a ‘but for’ basis (indeed, as we saw above, the UKSC expressly denied this). Does Schokman then mark a budding convergence between the approach of the UKSC and HCA as both embrace ‘connection’ as a means of determining the course of employment? Before we get too excited, Prince Alfred College is still regarded as authoritative in relation to sexual abuse, raising specific factors to be weighed in the balance. Yet the bizarre facts of Schokman have given rise to a thoughtful decision that draws liberally on the case law of the UKSC and seeks to rehabilitate the ‘connection’ test – drawn, it may be recalled, originally from the wording of Salmond on Torts ⁴⁹ – as a means of determining the scope of employment and hence the reach of the vicarious liability doctrine.

Conclusions

BXB and Schokman represent leading judgments that will dictate the future shape of the law. They seek to provide guidance for future courts, notably advising them to examine the duties and powers of the wrongdoing employee with reference to previous case law. Both advise caution. Questions remain, however. How will courts treat sexual abuse cases given that both cases suggest specific concerns that need to be addressed in such cases?⁵⁰ To what extent will these ‘special features’ be relevant to other intentional torts such as physical abuse? Will BXB finally be successful in reining in vicarious liability by reducing the role of policy to a final check in ‘difficult cases’ or will the impulse to do justice prove hard to resist? While the messaging is clear, given that the doctrine of vicarious liability ‘should be frankly recognised as having its basis in a combination of policy considerations’,⁵¹ it will be hard to relegate policy to an afterthought in cases where there is no clear precedent (and precedents can be distinguished …). The challenge for Australian law is more fundamental. The ‘occasion’ test of Prince Alfred College is now seen as part of the ‘connection’ test for course of employment. How will Australian courts respond to this new formulation?