skip to Main Content
Police Office In Street

Robinson v Chief Constable of West Yorkshire [2018] UKSC 4 and Poole BC v GN [2019] UKSC 25: public authority negligence liability today

Nimrah Ashraf

Nimrah Ashraf

Nimrah read law at Cambridge University and is currently working with legal NGOs. Her interests lie in Tort Law, Employment Law and Public Law.

Before Robinson v Chief Constable of West Yorkshire, the law on public authority negligence liability was confused. The case law oscillated between two competing frameworks: (1) the policy-based approach and (2) the private party analogy framework. On the policy-based approach, a duty of care would only be found if it was fair, just and reasonable to impose liability. In contrast, the private party analogy framework sees public authorities and private individuals being subject to the same rules. Robinson settled the law in favour of the latter approach, with Poole BC v GN following suit.

This case note will analyse these two cases in detail and assess whether the approach they represent is sound.

Robinson Supreme Court judgement: [2018] UKSC 4
Poole Court of Appeal judgement: [2017] EWCA Civ 2185
Poole Supreme Court judgement: [2019] UKSC 25

Robinson v Chief Constable of West Yorkshire [2018] UKSC 4

The facts

Police officers approached a suspected drug dealer in a busy shopping street. The drug dealer resisted arrest, which led to a struggle. The police officers and the suspect collided with Mrs Robinson, knocked her over and fell on top of her. Mrs Robinson was injured and instituted proceedings against the Chief Constable on the grounds of the police officers’ negligence.

Crucially, Mrs Robinson’s case did not concern an omission – she did not allege that the police officers failed to protect her from the drug dealer. Rather, she argued that the police officers’ positive acts resulted in her injury. The key question for the Supreme Court was whether the police owed Mrs Robinson a duty of care in such circumstances.

The Supreme Court

The Supreme Court unanimously held that the police did owe Mrs Robinson a duty of care, which had been breached. Lord Reed (with whom Lady Hale and Lord Hodge agreed) reached this conclusion through an application of the “established principles of the law of negligence” ([30]). Lord Hughes and Lord Mance agreed with the majority about the outcome, though their reasoning differed.

The Majority Reasoning of Lord Reed

Private party analogy framework

The premise of Lord Reed’s approach is that “at common law, public authorities are generally subject to the same liability in tort as private individuals and bodies” ([32]). In other words, the ordinary principles of negligence apply. This means a duty of care will generally be owed in respect of positive acts. However, there will be no liability for a pure omission, unless the facts fall within an established exception ([34]). These exceptions are where:

  1. The defendant creates a source of danger with which a third party interferes to cause harm (Haynes v Harwood [1936] 1 KB 146 per Lord Goff in Smith v Littlewoods [1987] 2 AC 241);
  2. The defendant prevents an alternative means of rescue, making the situation worse (Barrett v Ministry of Defence [1995] 1 WLR 1217);
  3. D has control over the third party who causes the damage (Dorset Yacht Co v Home Office [1970] AC 1004);
  4. D assumes responsibility for the claimant’s safety (Hedley Byrne v Heller) having made an express assurance to this effect on which the claimant has relied (Michael v Chief Constable of South Wales Police [2015] UKSC 2), or;
  5. The status of D creates an obligation to protect the claimant (Barnes v Hampshire [1969] 1 WLR 1563).

Limiting the role of policy

Policy considerations are not relevant to the question of whether a public authority owes a duty of care. Therefore, the approach taken by some earlier cases, in which the non-liability of public authorities was justified through reliance on policy arguments, is wrong. Lord Reed gave Hill v Chief Constable of West Yorkshire [1989] AC 53 as an example of this incorrect approach ([40]).

Instead, recourse to policy considerations is only permissible in two circumstances. Firstly, “a novel type of case, where established principles do not provide an answer” ([27]). Secondly, when a higher court is invited to determine “whether the precedents should be departed from” ([29]). If the facts of a case do not fall within either of these two categories, then it is “unnecessary and inappropriate” to consider whether the existence of a duty of care is ‘fair, just and reasonable’ ([26]).

Application to the facts

Firstly, Lord Reed established that the case concerned a positive act, not an omission. In other words, the police officers had not failed to act, but instead played an “active part in the critical events” ([72]). He drew a distinction with the facts of Hill. In that case, police officers failed to arrest a murderer before he killed a victim. On these facts, it was the police officers’ own actions which resulted in Mrs Robinson being injured ([73]).

Secondly, Lord Reed held at [74]:

It was reasonably foreseeable that if the arrest was attempted at a time when pedestrians – especially physically vulnerable pedestrians… – were close to [the suspected drug dealer], they might be knocked into in the course of his attempting to escape. That reasonably foreseeable risk of injury was sufficient to impose on the officers a duty of care towards the pedestrians in the immediate vicinity… including Mrs Robinson.

In so doing, Lord Reed treated physical loss which reasonably foreseeably followed from positive acts as an established category. Therefore, the facts of this case constituted a situation where ‘established principles’ did provide an answer. Thus, a duty of care could, without recourse to policy, be found.

The Minority Reasoning of Lord Hughes and Lord Mance

Lord Hughes and Lord Mance agreed that this case concerned a positive act and a duty of care was owed. However, they both disagreed with the limited role that Lord Reed ascribed to policy considerations.

Lord Hughes argued that “analysis in terms of omissions cannot be the only, or sufficient, reason why … duties of care are imposed” ([114]). In addition to applying the ordinary principles of negligence, regard must be had to “vital policy considerations,” whose treatment in “principal cases are … simply too considered, too powerful and too authoritative to be consigned to history” ([113]).

In a similar vein, Lord Mance argued “it would be unrealistic to suggest that, when recognising and developing an established category, the courts are not influenced by policy considerations” ([84]). In examining the previous case law on police and CPS liability, Lord Mance pointed to a tendency for judges to ‘buttress’ their conclusions with policy considerations ([88]-[94]). Lord Mance’s approach reflected such an approach. He concluded that although there was a “genuine policy choice” to be made, “that policy choice should … be made unequivocally in the sense indicated by Lord Reed” ([95]).

Poole Borough Council v GN [2019] UKSC 25

Robinson brought clarity to a muddled area of law. However, the Court of Appeal’s earlier judgement in Poole threatened to complicate this through endorsing an approach to public authority liability incompatible with that of Lord Reed. The Supreme Court therefore had two options: maintaining the coherence to the case law brought by Robinson or disturbing the new orthodoxy. They choose the former option.

The facts

As children, the claimants were abused by a family in their housing estate. Notwithstanding the youngest child’s attempted suicide and the mother’s complaints to the defendant local authority, the claimants were rehoused five years after the abuse began. They sought damages for physical and psychiatric harm, claiming the defendant had been negligent in (1) failing to rehouse them and (2) failing to take them into care.

The Court of Appeal [2017] EWCA Civ 2185

The Court of Appeal held that a duty of care did not arise. There were two strands of reasoning to the leading judgement of Irwin LJ. Firstly, this was an omissions case and none of the exceptions to the principle of non-liability applied. Secondly, policy considerations pointed against finding a duty of care. Specifically, the imposition of negligence liability would “complicate decision-making in a difficult and sensitive field, and potentially divert [officials] into defensive decision-making” ([94]). The difficulty is that, on Lord Reed’s approach, such recourse to policy considerations is impermissible.

The Supreme Court

Lord Reed followed his approach in Robinson to the letter. Thus, he considered that Irwin LJ had erred in his reliance on policy considerations. Instead, the question of liability was to be resolved solely through an application of the ordinary principles of negligence ([75]).

On the facts, Lord Reed held none of the exceptions to the principle that there is generally no liability for an omission applied. The defendant had not created a source of danger in allocating the claimants accommodation neighbouring the abusive family ([77]), nor had they assumed responsibility for the claimants’ safety through the provision of services or taking the children into care ([80]-[81]). Therefore, a duty of care did not arise.


Robinson as a step forward

Lord Reed’s approach in Robinson and Poole can be praised on two grounds. Firstly, it brought stability to a previously tumultuous area of case law. Secondly, and most importantly, it curtails reliance on flawed policy arguments which are unsupported by empirical evidence.

Prior to Robinson, judges relied on four policy arguments to find no duty arose. The first was that, as best endeavours are always used, negligence liability would not result in improved standards. However, this cannot be reconciled with examples of institutional misconduct (for example, as outlined in the Lammy Report (2016)). Secondly, negligence claims touch on matters of ‘policy and discretion,’ unsuitable to judicial determination. While this a legitimate concern, it is relevant to justiciability and not the duty of care.

The third consideration is that liability would divert resources away from public authorities’ functions. Yet this might result in the greatest wrongs (for which the most compensation is owed) going without a remedy. Finally, it was argued liability lead to a defensive exercise of functions. This would inhibit authorities from acting purely in the public interest. However, there is a lack of empirical evidence in support of this, or indeed any of the above arguments.

It should be noted that policy cannot be as tidily confined as Lord Reed suggests. For instance, policy is relevant to determining the extent of binding precedents and the scope of the omissions exceptions. However, the sharp reduction in the role of deficient policy-based reasoning is to be commended.

The limitations of Robinson

The difficulty with Lord Reed’s insistence on treating public authorities and private individuals the same is that they are not. However, his framework results in the acts/omissions distinction being determinative of liability, notwithstanding that it was designed primarily with private individuals in mind. Two problems follow from this.

Firstly, the rationale of the omission principle does not convince when applied to public authorities. Take, for instance, the erosion of individual autonomy rationale. This posits that requiring a person to act is more restrictive than imposing an obligation not to do something. Therefore, in general, it would be too great an inroad into individual autonomy to impose liability for omissions. However, public authorities are already under a public law duty to act and, in any case, their freedom of action is arguably only of instrumental significance. Similarly, in the case of private individuals, it may be difficult to justify why one person is singled out for their inaction over another (the ‘why pick me?’ argument). However, public authorities have a statutory duty or power which differentiates them, such as Poole Borough Council’s powers under the anti-social behaviour legislation.

Secondly, there are positive reasons for imposing liability for omissions more readily on public authorities. It could be argued their special status coupled with public dependence on such bodies should result in more onerous obligations being owed. Poole exemplifies this: while the claimants could take some steps to protect themselves from their neighbours, beyond that their only option was to seek assistance from Poole Borough Council.

The counterargument to the above view is that, irrespective of its logical faults, the omissions principle shields public authorities from excessive and speculative litigation. Public authorities are an easy target: traceable, guaranteed to have the necessary financial resources, and most failures in society can be attributed to them. However, all this suggests is that there is a need for adequate control devices on public authority liability. These can be found elsewhere in the negligence framework, such as when determining whether there has been a breach of duty.


The Robinson approach to public authority liability leaves much to be desired. While it has the benefit of side-lining policy-based reasoning, it is by no means the only framework capable of doing so. Approaches which may have struck a better balance include a public authority specific regime and developing further exceptions to the omissions principle specifically for public authorities. That being said, stability has only just been achieved in a controversial and chaotic area of law; it may be ill-advised to disrupt this so soon.


You can read more about Robinson v CC West Yorkshire‘s impact on the theory behind the duty of care in the tort of negligence here: Robinson v CC West Yorkshire

Back To Top