Liability of public authorities and duties of affirmative action.

• Author: Siebrasse, Norman
• Position: Canada

INTRODUCTION

The question of whether a duty of care is owing is perhaps nowhere more important that in the context of liability of public authorities. New duties do, of course, arise in respect of private parties. (1) But the increasing range of state activity and the diversity of governing statutes means that in practice, difficult questions of duty arise on a continuing basis with respect to public authorities. Kamloops v. Nielsen, (2) a modern leading case on the general duty of care, was also a case of public authority liability. The stream has remained steady since then. From public works departments in Just, Brown and Swinamer, (3) to financial regulators in Cooper v. Hobart, (4) to current cases involving SARS, West Nile and BSE ("Mad Cow"), (5) difficult questions of law relating to whether a duty of care is owing arise disproportionately in respect of public authorities.

Canada continues to adhere to a modified version of the approach set out by Lord Wilberforce in Anns v. Merton London Borough (6) in determining whether a duty of care is owing, but the Supreme Court has rightly recognized that Anns alone cannot provide the certainty and predictability required to guide trial judges and litigants. Categories are needed to supplement the Anns approach. In this paper I sketch an argument that what I will call "regulatory liability," which corresponds to duties of affirmative action with respect to private individuals, should be recognized as a distinct category of claims brought against public authorities. While the conclusion that generally no duty should be owing, subject to certain exceptions, is the same for public and private actors, I note some differences in the underlying policy rationale in the two cases. The argument is illustrated by a discussion of Kamloops, which also addresses the relationship between categories of regulatory liability and pure economic loss.

REGULATORY LIABILITY AND DUTIES OF AFFIRMATIVE ACTION

The common law draws a strong distinction between risks caused by the defendant and those caused by a third party. Subject to some exceptions, if a plaintiff is imperilled by a source unconnected with the defendant, the defendant owes no duty to assist, even though it would be easy to do so; there is no duty to be a Good Samaritan. (7) There is no entirely convenient label for this class of cases. Scholarly articles often raise the question as "the duty to rescue", but while rescues are a good example, it is well recognized that the principle applies more broadly than to rescues in a narrow sense. The other commonly used rubric, "duties of affirmative action," is broader, but it, on the other hand, is too broad. It is not really correct to say that the common law does not impose duties of affirmative action generally, as such a duty will normally arise when the defendant itself created the risk. (8) It is more accurate to say that the common law does not impose duties of affirmative action to prevent harm to another when the defendant did not originally cause or increase the harm. (9) I will therefore use the term "regulatory duty" to refer to a duty arising from a duty to prevent harm to caused by third parties, and "regulatory liability" to refer to liability arising from a breach of such a duty. Using a term distinct from that used in the private context also ensures that we do not assume that policy and doctrine will be exactly the same whether a private or public actor is concerned.

In the public authority context, claims for failure to prevent or diminish harm not originally caused by the public authority itself encompass an important subset of government activity, ranging from regulation of air traffic safety, to police prevention of crime, food safety regulation, animal and human disease prevention, and inspection of new homes for building code compliance. My argument is that regulatory duties constitute a distinct category of public authority liability, and that the state should not owe a duty of care in the absence of reasonable reliance on the state activity. In effect, this is an attempt to revitalize distinction between nonfeasance and malfeasance by means of a principled comparison with private duties of affirmative action.

There is a straightforward statutory interpretation argument for recognizing regulatory liability as a distinct sub-category of public authority liability. In Canadian jurisdictions, the statutes waiving the Crown's common law immunity from suit provide that the Crown is subject to liability in tort to the same extent as "if it were a person." (10) Since private individuals are not subject to duties of affirmative action, this statutory wording implies directly that the Crown is similarly not subject to such duties. In most US jurisdictions the statutory waiver of sovereign immunity similarly provides that the state is subject to liability "to the same extent as a private individual under like circumstances." (11) It is well recognized that such statutory provisions mean that the state is not liable as a Good Samaritan except in circumstances in which a private individual would also be liable. (12)

A number of important cases apply this rule in the public authority context, albeit without direct reference to the common law reluctance to impose duties of affirmative action. South v. Maryland (subject to exceptions discussed below) established what is now know as the "public duty" doctrine, when it held that absent a special relationship, when a duty is owed by a public official to the public as a whole, a tort action does not lie for neglect: "the officer is answerable to the public and punishable by indictment only." (13) In Cooper, the Supreme Court of Canada asserted essentially the same doctrine: "The Registrar's duty is rather to the public as a whole. Indeed, a duty to individual investors would potentially conflict with the Registrar's overarching duty to the public." (14) The question, which was answered in neither South v. Maryland nor in Cooper, is to know when a duty is owed to the public and when it is owed to an individual. One answer is that the scope of a public duty is defined by the scope of Good Samaritan activity: when it is acting to prevent harm caused by a third party, the state owes a duty only to the public as a whole. Brennan C.J. of the Australian High Court made much the same point in Pyrenees Shire Council v. Day:

No duty breach of which sounds in damages can be imposed when the power is intended to be exercised for the benefit of the public generally and not for the protection of the person or property of members of a particular class. And I doubt whether a duty breach of which sounds in damages would be held to exist if the power were conferred merely to supervise the discharge by a third party of that party's duty to act to protect a plaintiff from a risk of damage to person or property. (15) This is not to suggest that the leading cases never impose a duty of affirmative action on a public authority. This was done in Kamloops for example, and indeed Brennan C.J. in Pyrenees imposed a duty on the municipality in question to enforce its fire prevention by-laws. Whether these cases fall within recognized exceptions to the rule, or are true counter-examples will be discussed below. The point for the moment is that the "public duty" doctrine generally supports the view that courts are reluctant to impose duties of affirmative action, even on public authorities.

Apart from statutory interpretation argument and this smattering of consistent cases, it is important to consider on its own merits the policy basis for recognizing regulatory liability as a distinct category of public authority liability. If it is a nonsensical category from a policy perspective, to be recognized only because of the demands of the Crown Liability and Proceedings Act, it might be interpreted more narrowly than if it is stands independently on a sound policy basis. Moreover, it is not always obvious when a public regulatory activity is analogous to an act of a private Good Samaritan. An understanding at the policy level of differences and similarities between public and private actors will help us decide when the analogy is appropriate. Because the Good Samaritan duty in the private context has been extensively studied, it is useful to take arguments relevant to private actors and consider the extent to which they are applicable to the state.

The argument in favour of a general duty to rescue is very straightforward. The basic insight of law and economics is that liability should be placed on the person able to avoid the accident at least cost in order to provide an incentive to take care to prevent the loss. If a passer-by can prevent a blind person from walking off a cliff simply by shouting a warning, then a great loss can be prevented at minimal cost. Whether the least-cost accident avoider happens to have caused the loss in the first place is irrelevant to this basic argument. (16)

The more difficult question is why the common law is reluctant to impose a duty to rescue on a private individual. A very significant literature defends the common law rule on the basis of principles requiring respect for the autonomy of the individual. (17) Whatever their merits, these types of arguments do not apply to the state. The state is not a person, and we need not respect its autonomy. (18)

On the other hand, arguments based on incentive effects, that is, the desire to avoid and minimize harm, are potentially applicable to the state. (19) There are two main arguments. One explanation is the problem of salience, or, as Lord Hoffmann has termed it, the "why pick on me?" argument. (20) It turns on the difficulty of identifying an appropriate 'tortfeasor' when many bystanders could potentially have prevented the harm. Apart from the objection of moral arbitrariness, imposing a duty on an indeterminate class of persons may have unintended...