ON 18 October 2017, the UK Supreme Court decided Armes v Nottinghamshire County Council [2017] UKSC 60. The Court ruled that a local authority could be vicariously liable for intentional torts committed by foster parents against a child whom the authority had placed in their care. The outcome was not entirely unexpected. Less than two decades ago it would have been inconceivable. After all, isn't it the case that the common law does not recognise a general principle of liability in tort for the acts of third parties? And that in so far as it does, it holds an employer vicariously liable for a tort committed by an employee in the course of their employment? This is a very long way from the facts of Armes.
Armes was made possible by a line of cases which have transformed the law of vicarious liability since Lister v Hesley Hall [2001] UKHL 22, [2002] 1 A.C. 215 was decided by the House of Lords in May 2001. In Lister the century-old “Salmond test” for determining when a wrong arose in the course of employment was abandoned. In place of the distinction between “authorised acts” and “unauthorised modes” came the more modern-sounding “sufficient connection” test.
Now an employer could be vicariously liable if the tort had a sufficient connection to the job or role which the employee was undertaking. Around the same time the Canadian Supreme Court was using, in Bazley v Curry (1999) 17 D.L.R. (4th.) 45, the slightly more functional language of the “enterprise risk” test: “the question in each case is whether there is a connection or nexus between the employment enterprise and that wrong that justifies imposition of vicarious liability on the employer for the wrong, in terms of the fair allocation of the consequences of the risk and/or deterrence” (17 D.L.R. (4th.) 45, at [72]). The result was the same: an organisation responsible for the care of children could be liable for assaults committed by an employee to whom that task had been assigned.
We are still some way from Armes: foster parents are not employees of the council. But in the Christian Brothers case (Catholic Child Welfare Society and others v Various Claimants and The Institute of the Brothers of the Christian Schools and others [2012] UKSC 56, [2013] 2 A.C. 1) the Supreme Court held that a relationship “akin to that between and employer and employee” (Lord Phillips at [47]) could establish vicarious liability. Tort law, being mostly judge-made, has turned out to be more flexible than labour law, which, confined by statutory wording, still requires a contract for there to be an employment relationship. It is not then very hard to see that if a prison inmate negligently injures a member of the prison staff who is effectively in the position of being a co-worker, the latter may have a cause of action against the organisation running the prison, as the Supreme Court decided in Cox v Minister of Justice [2016] UKSC 10, [2016] A.C. 660.
Assigning tortious liability to organisations is not straightforward. Organisations acquire the legal capacity to manage their affairs by being constituted as corporations or companies. Corporations, statutory or otherwise, can only act through the agency (using that term broadly) of human persons. In tort law there are two routes by which an act or omission of a human being can be attributed to a corporate person: vicarious liability is one; the concept of a non-delegable duty is another.
Juridically, the two routes are very different. In the case of vicarious liability, the human person must commit a tort. The liability of the corporate person is strict once a sufficient connection between the tort and the organisation is established. In the case of the non-delegable duty, there is no need to find an individual tort. The tort is personal to the corporate entity, and requires fault. This tends to mean, in practice, that it relates to a managerial or organisational failure of some kind. In employer's liability this can take the form of a failure, for example, to provide a safe working environment. Following the decision of the Supreme Court in Woodland v Essex County Council [2013] UKSC 66, [2014] 1 A.C. 537, the concept of the non-delegable duty extends beyond employment; in that case it was applied to a school's duty to ensure a safe environment for its students.
In Armes, the Supreme Court found against a non-delegable duty of care, in part because of the statutory framing of the local authority's responsibilities. Thus “the duty of the local authority is not to perform the function in the course of which the claimant was abused (namely, the provision of daily care), but rather to arrange for, and then monitor, its performance” (at [47], per Lord Reed). As there was no evidence to suggest that the local authority had been negligent in placing the claimant with her foster parents, or in subsequently monitoring the placement, the claim for breach of the personal duty failed.
However, the vicarious liability claim succeeded. The Supreme Court found that “the foster parents provided care to the child as an integral part of the local authority's organisation of its child care services” (at [60], per Lord Reed). Moreover, “considering next the issue of risk creation, the local authority's placement of children in their care with foster parents creates a relationship of authority and trust between the foster parents and the children, in circumstances where close control cannot be exercised by the local authority, and so renders the children particularly vulnerable to abuse” (at [61]). Although there was no day to day control, the local authority retained certain powers: “the local authority exercised powers of approval, inspection, supervision and removal without any parallel in ordinary family life” (at [62]; this is the point over which Lord Hughes's dissent mostly turned). And the local authority had the means to pay damages: while “foster parents have insufficient means to be able to meet a substantial award of damages, and are unlikely to have (or to be able to obtain) insurance against their own propensity to criminal behaviour”, local authorities, “can more easily compensate the victims of injuries which are often serious and long-lasting” (at [62], per Lord Reed). Many of the same policy considerations which were considered under the heading of the non-delegable duty came up again under the heading of vicarious liability. Thus the majority found it relevant, for the purposes of vicarious liability, that the placement of the child put her at risk, and that the local authority exercised a degree of control over the foster parents. This is the language of “enterprise risk”, as is the suggestion that the foster parents were an integral part of the “organisation” of the local authority's child care services. It is all the more striking then that the court arrived at different outcomes via the two routes.
Whatever the outcome in Armes, the route chosen was going to be important (S. Tofaris, “Vicarious Liability and Non-Delegable Duty for Child Abuse in Foster Care: A Step Too Far?” (2016) 79 M.L.R. 884). In opting for vicarious liability, the Supreme Court chose a route which is in many ways more familiar to tort lawyers than the emerging alternative of the non-delegable duty. But vicarious liability, in its current form, is something of a historical accident and, when coupled with the insurer's right of subrogation, can produce some peculiar outcomes (T. Weir, “Subrogation and Indemnity” [2012] C.L.J. 1, first published privately in 1973). We may wonder if it is really fit for purpose.
In vicarious liability, the corporate person is strictly liable, but a personal tort must have been committed. There has to be a wrong of some kind, but, because of the way the vicarious liability tests are phrased, the issue of whether there was a wrong is detached from the issue of managerial or organisational fault. With the non-delegable duty, it is precisely this issue of managerial failure which comes back into focus. In employer's liability, the non-delegable duty is not strict; it may be towards the stricter end of negligence, but managerial fault is still required, a “defect in the structure of the enterprise” as German law puts it (BGH NJW 1956, 1106; BGH NJW 1971, 1313.)
The concept of the non-delegable duty has its critics. Perhaps it is just as much a fiction as vicarious liability, and just as doctrinally contorted (J. Morgan, “Vicarious Liability for Independent Contractors” (2016) 31 Professional Negligence 235). As with any developing area of liability, many issues remain to be clarified. However, the need to show organisational fault could provide courts with a control device which may not be available if they go down the vicarious liability route.
From the point of view of enterprise liability, the right question to ask in Armes was whether the risk of harm to the claimant was one for which the defendant council retained organisational responsibility even after it discharged to the foster parents the task of caring for the claimant. If the answer to that question is “yes”, the council should have been liable for a breach of a non-delegable duty of care. Did Armes reach the right result by the wrong route?