Question: 'The Caparo three-stage test has failed to provide a coherent and predictable framework for determining the existence of a duty of care.' Discuss.
---
Introduction
The Caparo three-stage test—foreseeability, proximity, and fair/just/reasonable—was intended to discipline the duty inquiry after the expansive and ultimately unsustainable approach in Anns v Merton LBC. Yet it has attracted sustained criticism for indeterminacy, policy-laden discretion, and inconsistent application. This essay evaluates the coherence and predictability of the Caparo framework, examining doctrinal practice, academic critique, and the Supreme Court's recent clarifications in Robinson v Chief Constable of West Yorkshire Police.
I. The structure of the Caparo test
The test comprises three ostensibly distinct requirements. Foreseeability is objective and fact-sensitive, but largely uncontroversial: it screens out remote or unforeseeable claimants. Proximity is more elusive: it is not spatial or temporal but normative, encompassing assumption of responsibility, reliance, directness, control, and the nature of the harm. The fair/just/reasonable limb invites explicit policy reasoning: indeterminacy, floodgates, alternative remedies, coherence, and public functions.
In practice, proximity and policy overlap extensively. Many factors labelled as proximity (e.g. generality of the claimant class, remoteness of harm) are equally plausible as fairness concerns. This conflation undermines the analytical clarity supposedly gained by distinguishing three stages.
II. The incremental method and the problem of novelty
Lord Bridge in Caparo emphasised that the three stages are not mechanical tests but aids to incremental reasoning by analogy to established categories. Robinson reinforced this: in established duty categories (e.g. road accidents, medical treatment), courts do not rehearse Caparo; duty is simply assumed. Caparo operates only in novel cases.
This is doctrinally coherent but raises two difficulties. First, it is not always clear whether a case is novel or falls within an established category. Michael v Chief Constable of South Wales Police illustrates the problem: was the claim a straightforward omissions case (no duty per Hill) or sufficiently distinct (identifiable victim, express assumption of responsibility) to warrant separate analysis? The Supreme Court treated it as established, but the result was controversial.
Second, in genuinely novel cases, the third limb becomes decisive, and it offers little constraint. In Customs and Excise Commissioners v Barclays Bank, the House of Lords split on whether it was fair, just, and reasonable to impose a duty on a bank to freeze an account pursuant to a freezing order. The disagreement turned on competing policy intuitions, not doctrinal principle.
III. Incoherence across harm types
The framework produces divergent outcomes depending on the type of harm. Physical injury caused by positive acts: duty readily assumed. Pure economic loss from negligent acts: presumptively no duty (Spartan Steel, Murphy). Pure economic loss from negligent misstatement: duty if assumption of responsibility (Hedley Byrne, Caparo). Psychiatric harm to secondary victims: restrictive control mechanisms (Alcock).
This typology is doctrinally entrenched but arguably lacks principled foundation. Why should psychiatric harm be treated more restrictively than physical injury? The conventional answer is policy: fear of indeterminate liability and malingering. But the Caparo framework obscures this choice: the Alcock criteria are presented as defining proximity, yet they are plainly policy-driven limitations.
Stapleton and others argue that the framework's failure to distinguish openly between principle (does the defendant owe an obligation?) and policy (should we impose liability?) generates confusion and unpredictability.
IV. Public authority cases: justiciability and accountability
Public authority liability illustrates the indeterminacy problem acutely. Hill established a broad immunity for police in relation to investigation of crime, grounded in both proximity and policy. Yet the boundaries are unclear. Robinson held that where police cause physical injury by positive act, Hill does not apply. Michael extended Hill to cover failure to respond to emergency calls. Poole clarified that assumption of responsibility requires a specific undertaking, not merely the performance of a statutory function.
These distinctions are defensible but fact-sensitive and unpredictable. The outcomes turn on fine distinctions between acts/omissions, general/specific duties, and justiciable/non-justiciable decisions. Critics argue that the third limb permits the courts to immunise public bodies on policy grounds while denying accountability.
V. Academic responses
Rights theorists (Stevens, McBride) argue that the third limb is misconceived: liability should depend on infringement of the claimant's primary rights, not policy balancing. On this view, proximity is superfluous if we identify the relevant rights correctly.
Others defend the framework as appropriately flexible and policy-sensitive. Nolan argues that Caparo should be understood not as a test but as a framework for reasoning, encapsulating the multi-factorial nature of the duty inquiry. Lord Reed in Robinson emphasised the common law method: incremental development guided by analogy, coherence, and principle.
Conclusion
The Caparo framework has not provided a fully coherent or predictable structure. The three stages overlap; the incremental method defers hard questions; the third limb invites unstructured policy discretion. Yet Robinson and Poole suggest a more disciplined approach: Caparo applies only in novel cases; established categories are respected; assumption of responsibility is narrowly construed.
Whether this represents genuine clarification or post-hoc rationalisation remains contested. The duty inquiry is irreducibly normative and fact-sensitive. Caparo may be the best framework available, not because it is determinate, but because it makes explicit the policy choices that any duty framework must address. Predictability may be an unrealistic aspiration in a system that privileges flexibility and incremental development over rigid rules.
---
(Word count: ~950)