The ‘End and Aim Rule’ [1397] and Indeterminate Liability
Recall that in Hercules the Court held that foreseeable reasonable reliance was a sufficient test of proximity to found a prima facie duty of care. It is clear that LaForest did not define reasonable reliance as reliance grounded in what responÂsibilities the defendant had assumed.[1398] He was speaking of the plaintiff's reasonÂably relying on the information having been prepared with due care.
That duty formulation created a potentially indeterminate ambit of liability. To control that practical problem, the Hercules Court adopted the ‘end and aim' rule. Most courts use the ‘end and aim' rule differently, incorporating it into the duty or proximity analysis.[1399] This makes sense. What can it mean, or should it mean, to say liability is limited to losses incurred in ‘precisely the purpose or transaction for which (the information or advice) was prepared', [1400] other than that the scope of liability is defined by the responsibilities the defendant assumed? This is a proximity quesÂtion of principle, not a policy concern about indeterminate liability.[1401]This raises the question of whether the potential for indeterminate liability can ever arise if there has been a proper application of the ‘end and aim' test at the proximity stage of a misrepresentation action. In the Imperial Tobacco case, the Supreme Court of Canada effectively held that the potential for indeterminate liability could survive a proper proximity analysis.101 I disagree.102
In brief, in Imperial Tobacco, tobacco companies were advised by their regulaÂtor, Health Canada, to introduce, promote and market low-tar cigarettes as a harm reduction measure. The companies were sued by smokers, who alleged that low-tar cigarettes did not reduce the harm from smoking, and that negligently encouraging smokers to use low-tar cigarettes caused many of them to smoke more, or to forego other harm-reduction measures, particularly quitting smoking, altogether.
This in turn caused some smokers to suffer unnecessary illness or premature death. The smokers' class action against Health Canada was dismissed for want of proximity. The tobacco companies also sued Health Canada, seeking to recoup some of the losses that they expected to incur when they were held liable in damages to the smokers. This part of the case was framed in negligent misrepresentation.The Supreme Court applied the foreseeable reliance test from Hercules with litÂtle discussion. However, it is evident that the facts of the Imperial Tobacco case would easily have satisfied a proper assumption of responsibility test. There was no doubt that Health Canada intended the companies to rely on its advice about low-tar cigarettes. The Court dismissed the action based on a Step Two Anns conÂcern about public authority policy immunity from negligence liability. This is conÂtroversial in its own right.103 In addition, again with little discussion, the Court held that the prima facie duty had to be negated because it created a potentially indeterminate ambit of liability. As we know from Hercules, the foreseeable reaÂsonable reliance test may actually create potentially indeterminate liability. The present question concerns whether indeterminate liability will arise in misrepreÂsentation if the court uses the assumption of responsibility approach.
Usually when one speaks about an indeterminate risk there are background concerns about a ‘ large risk', possibly a ‘ruinous risk', or a risk disproportionate to fault. Imperial Tobacco illustrates that a defendant may assume an extremely concerns were expressed to suggest that there was an error in the duty analysis if such exposure existed, not expressed for the purpose of limiting the exposure generated by the (improper) duty formulaÂtion. Cardozo CJ presumably would have found Hercules a good example of his point. See J Neyers, ‘Donoghue v Stevenson and the Rescue Doctrine: A Public Justification of Recovery in Situations Involving the Negligent Supply of Dangerous Structures' (1999) 49 University of Toronto Law Journal 475 at n 51, citing P Benson, ‘The Basis for Excluding Liability for Economic Loss in Tort Law' in D Owen (ed), Philosophical Foundations of Tort Law (Oxford, Clarendon Press, 1995) 427, 434; R Bernstein, Economic Loss (London, Longman Group, 1993) 14; Weinrib (n 48) 231.
101Imperial Tobacco (n 4).
102 See also L Klar, ‘Imperial Tobacco Ltd: More Restrictions on Public Authority Tort Liability' (2012) 50 Alberta Law Review 157 [49].
103 Public authority negligence immunity and its application in Imperial Tobacco are criticised by B Feldthusen, ‘Public Authority Immunity from Negligence Liability: Uncertain, Unnecessary, and Unjustified' (2014) 92 Canadian Bar Review 211. large risk, although probably not ruinous in the case of the federal government of Canada. As for disproportionate, I am not sure it would have been. Regardless, that matter would be better dealt with by a judge apportioning liability among the government and the tobacco companies. Large, ruinous or disproportionate loss may be legitimate concerns in their own right. However, our positive law has not embraced openly a policy argument that exculpates defendants simply because the damage it has done is too large. It is as if using the adjective ‘indeterminate' instead renders it unnecessary to do so.
Indeterminacy as a policy matter is not the same as large damage, although the outcome of indeterminacy may be large damage. Indeterminacy is imporÂtant because a prospective defendant cannot rationally assess an unknown and unknowable risk. One would have to question the mental capacity of a defendant who voluntarily assumed open-ended responsibility and the mental capacity of a plaintiff who claimed to have reasonably relied on the defendant having done so. Insurance companies will either not cover open-ended risks, or will do so only at a literally prohibitive price. More fundamentally, is it coherent to speak of a relaÂtionship being both proximate and indeterminate at the same time?
It is doubtful that the risk in Imperial Tobacco was indeterminate in this sense. LiterÂally, the class was limited to the tobacco companies named as plaintiffs. Ultimately, the exposure turns more on the smokers than the companies. Health Canada either knew or could have known the number of smokers in Canada who smoked the plaintiffs' brands.
It would have had an excellent idea about the aggregate costs to individual smokers who continued to smoke, and the aggregate savings, if they quit. These are the outer parameters of the risk. It is inconceivable that Health Canada would have adopted low-tar cigarettes as its signature harm-reduction policy unless it had reliÂable indications about the number of smokers who would not have quit, but who would switch to low-tar cigarettes, and what difference this would make to the risk of illness and death. It is difficult to see the smokers recovering damages from the companies without being able to assemble this information. This information narÂrows the parameters of exposure considerably. Health Canada either had this informaÂtion and more, or could have had it. This is precisely the risk Health Canada intended the tobacco companies to run. The pleadings suggest they miscalculated the benefits of low-tar cigarettes. This has nothing to do with the risk being indeterminate. The assumption of responsibility proximity test defined the risk. Properly applied, there should never be a residual indeterminacy problem.[1402]V.