by Christopher Bruce

This article was originally published in the April 24, 1998 issue of The Lawyers Weekly. We reproduced it in the Summer/Autumn 2002 issue of the Expert Witness because the topic had arisen in one of our cases.

In a number of recent cases, the courts have been asked to calculate the loss of dependency of orphaned minors. In all of the reported cases, these children have been taken into the care of relatives – aunts, uncles, grandparents, or stepparents. An important issue that is raised by this arrangement is whether the expenditures incurred by the surrogate parents should be set off against the children’s loss of dependency on their natural parent(s).

In the leading Ontario case of Butterfield v. Butterfield Estate (1996) 23 M.V R. (3d) 192 (Ont.C.A.), for example, two children aged six years and six months, respectively, were taken into the care of their aunt and uncle. The defendants argued that the children’s claim of dependancy on their mother’s income should be reduced by the value of the expenditures which their aunt and uncle (who planned to adopt them) would make on them.

Similarly, in the leading British Columbia case of Skelding (Guardian ad litem of) v. Skelding (1994) 118 D.L.R. (4^th) 537 (B.C.C.A.), the defendants argued that the children’s loss of dependency on their mother was extinguished because their father (with whom they lived) had remarried.

The differing approaches that have been employed to resolve this issue provide evidence of a fundamental division in our courts with respect to the purpose of tort law. Two conflicting paradigms can be identified.

In what I will call the ex post approach, the court takes the view that, as the tortious act has already occurred, that act cannot be undone. Rather, the best the court can do is to ensure that the victims are restored, as best as possible, to the position they would have been in had the act not occurred.

In the competing, ex ante approach, the court recognises that any decision that it makes in the current case may influence the behavior of parties in similar, future cases. Hence, what is important in the current case is to set a precedent which will direct future parties to behave in the socially desirable manner.

The appellate court in Skelding clearly adopted the ex post approach. Relying heavily on the Supreme Court decision in Ratych v. Bloomer (1990) 69 D.L.R. (4^th) 25, the majority concluded that the stepmother had replaced the natural mother. Hence, no further claim was necessary to return the children to the position they would have been in had their mother lived.

Notably, this decision is exactly consonant with a well-developed line of cases which have concluded that a widow(er)’s loss of dependency may be extinguished upon marriage to a new spouse whose income is similar to that of the deceased. Particularly important for Skelding was the B.C. case of Ball v. Kraft (1966) 60 D.L.R. (2d) 35 in which both the widow and her children were denied compensation after the date of her remarriage.

In addition to its reliance on Ratych, the B.C.C.A. also defended its decision by reference to B.C.’s Family Relations Act. This Act imposes a legal requirement on parents to provide “reasonable … support and maintenance of the child.” As “parents” are defined in the Act to include stepparents and guardians, the court found that the services which Mr. Skelding’s new wife provided to his children were not “gratuitous.”

Interestingly, in making the latter decision, the court came into direct conflict with its own decision in Grant v. Jackson (1986) 24 D.L.R. (4^th) 598 which it had made only eight years earlier. In Grant the court had held that services provided by a father to his children, following the death of their mother (his wife), were not required by the Family Relations Act.

Despite Skelding’s grounding in Ratych, the vast majority of cases dealing with orphaned children have been careful to distinguish themselves from Skelding. Most of these cases employ what I called the ex ante approach to justify their decisions. In particular, they argue that the precedent established by Skelding may create perverse incentives for the friends and families of orphaned children.

The leading statements of this view appear in Tompkins (Guardian ad litem of) v. Byspalko (1993) 16 C.C.L.T. (2d) 179 and Ratansi v. Abery (1995) 5 B.C.L.R. (3d) 88. In both cases, the trial judges argued that if Skelding was followed, the risk would be created that “… in some cases, family members who would otherwise take orphaned children into their care may decline to do so until or unless an award has been made in the children’s favour.”

And in Tompkins, Spencer, J. went further, arguing that “… a surviving parent may refrain from remarriage, advantageous from the children’s point of view, because the presence of a new spouse who replaces services to the children may reduce their award”.

These cases, therefore, adopt the view that the finances and services provided by family members are in the nature of collateral benefits and should not be deducted from the children’s dependency on the deceased parent.

Most of the cases that adopt this view also respond to the argument in Skelding that the Family Relations Act (or its equivalent in other provinces) imposes a legal requirement that a “parent” provide reasonable support.

In Butterfield (cited above), for example, the Ontario Court of Appeal implied that an aunt and uncle had no legal obligation to provide for the children, even though they intended to adopt the children formally. And in Ratansi (cited above) and Schellenberg v. Houseman (1996) 18 B.C.L.R. (3d) 209, the courts concluded that support provided by family members who had been appointed legal guardians was also to be treated as a collateral benefit.

Yet, B.C.’s Family Relations Act would have defined the family members in all three of these cases to be “parents.” Either these cases were in error or Skelding was.

Finch, J.A., the dissenting judge in Skelding, offered a resolution to this dilemma. He noted, first, that fatal accident legislation generally requires that the damages must have “resulted” from the death of a family member. Conversely, he argued, support received from a third party could not be considered to have offset the plaintiffs’ loss unless that support also resulted from the death in question. He concluded, therefore, that, as the marriage of Mr. Skelding to his second wife could not be considered to have resulted from the death of his first wife, the support provided by the second wife to Mr. Skelding’s children must be considered to be a collateral benefit.

This argument notwithstanding, the minority view in Skelding faces another challenge. Finch, J.A. argued that the income of Mr. Skelding’s new wife should not be offset against the children’s loss of dependency on their natural mother; yet it is well-settled law in Canada that the new wife’s income should be offset against Mr. Skelding’s loss of dependency on his first wife. This discrepancy remains to be “explained.”

To summarise, the courts’ treatment of claims by orphans for loss of dependency offers insight into a question that goes to the foundations of tort law. Should the courts concentrate strictly on the facts of the case at hand – the ex post approach? Or should they take into account the impact that the decision in the current case will have on the future behaviour of other individuals – the ex ante approach? Although the response to Skelding (and to Ratych) suggests that most courts are leaning towards the ex ante approach, the issue is far from settled.

Christopher Bruce is the President of Economica and a Professor of Economics at the University of Calgary. He is also the author of Assessment of Personal Injury Damages (Butterworths, 2004).

by Christopher Bruce

This article was originally published in the Winter 2001/02 issue of the Expert Witness.

In many cases, the information required to establish negligence remains in the possession of one of the parties. In the absence of any penalties, a party who believes that this evidence may suggest that he or she should be held liable will have an incentive to destroy the evidence.

The purpose of this article is to develop a model of the legal process that will offer insight into the determination of legal remedies for the destruction of evidence by a defendant. I base this model on the assumption that the first role of such remedies must be to discourage the defendant from destroying any information that might reasonably be expected to assist the court in the determination of liability.

Three Questions

I believe that any legal analysis of the destruction of evidence by a defendant must investigate three questions:

• Did the defendant have reason to believe that it was the subject of litigation?
• Did the defendant believe that the information in its possession could assist the court in the determination of negligence or liability?
• Did the defendant intentionally or negligently destroy the evidence, or was that destruction “accidental?”

Was the Defendant the Subject of Litigation?

Businesses and individuals destroy private information every day. That destruction only becomes of concern to the court when it has an impact on the court’s ability to assess liability (and assign damages). For that reason, no legal “remedy” is required if a party had no reason to believe that destruction of information would have a bearing on any legal proceeding.

For example, if a factory has no reason to believe that its emissions have any harmful effects on its neighbours, destruction of information concerning those emissions should not subject it to penalties. To rule otherwise would require that all individuals and all businesses save all information indefinitely. Only if a subjective test concludes that the defendant should have been aware that its actions might be the subject of a legal action should it be held responsible for preserving records of those actions. This test should be stronger, the greater was the likelihood that the actions in question might become the subject of litigation.

Was the Information Determinative of Liability?

Assume that the first question has been answered in the affirmative – the defendant has been found to be aware of the possibility of litigation. Assume also that it could be determined ex post that the defendant knew that information in its possession would prove it to be negligent and liable; and that the defendant has intentionally destroyed that information.

The appropriate legal remedy would be to impose the same level of liability and damages on the defendant as would have been imposed if the information had been preserved. Such a ruling would simultaneously retain the plaintiff’s right to compensation and remove the incentive for the defendant to destroy the information.

Thus, if the court was able to determine ex post that the destroyed evidence would have contributed to the determination of liability, its appropriate response would be to reach the same conclusion that would have been reached had the information been preserved.

Conversely, if the court was able to determine ex post that the destroyed information would not have contributed to the determination of liability, its appropriate response would be to excuse the destruction of that information.

But when evidence has been destroyed, the courts often cannot determine whether that evidence would have been determinative of liability. (If liability could have been determined without that information, the issue of destruction of evidence would not have arisen in the first place.)

It is always in the defendant’s interest to argue that the information that it has destroyed was irrelevant to the case at bar. Once that information has been destroyed, it will be difficult, if not impossible, to lead either objective or subjective evidence to contradict the defendant’s argument.

The court is left with a dilemma. If it knew that the destroyed evidence would have proven the liability of the defendant, it should set damages as if liability had been proven. Whereas if it knew that the destroyed evidence would not have been of assistance to the court, it should ignore that destruction. But the defendant will always argue the latter and the court (and the plaintiff) will not be able to prove otherwise.

The issue then, is how can the court induce the defendant to preserve evidence that might prove to be relevant to the determination of liability? The simplest rule would be to place the onus on the defendant to prove that the destroyed evidence did not bear the importance that the plaintiff has claimed for it. That is, the normal onus of proof would be reversed.

Under this rule, if the defendant was of the honest opinion that the information it proposed to destroy was not relevant to the case, it would be induced to preserve that information until the trial date, if it was inexpensive to store. Or, if the information was expensive to store, it would be induced to offer to obtain the permission of the plaintiff to destroy that information. (If the plaintiff refused, the defendant might be allowed to claim storage costs against the plaintiff if the plaintiff’s case was not successful.)

And if the defendant was of the opinion that the information was relevant to the determination of liability, it would have an incentive to preserve that information. If it preserves the information, there may be some chance that it will be able to convince the court that it was not negligent or liable. Whereas if it destroyed the evidence, the proposed rule would find it liable with certainty.

That is, under all circumstances, the proposed rule would induce the defendant either to preserve potentially damaging evidence or to obtain the plaintiff’s permission to destroy that evidence. As this is the desired outcome, the rule may be said to be efficient.

Did the Defendant Destroy the Evidence “Intentionally?”

In the preceding section, I argued that if the defendant intentionally destroyed evidence, it should be found responsible for the same level of damages that would have been awarded had it been found liable. But what would the efficient rule be if the evidence was destroyed for reasons that were beyond the control of the defendant? Or if the evidence was destroyed as a result of the negligence of the defendant?

In the former of those cases – the destruction was “an act of God” or was, for other reasons, unforeseeable – the imposition of damages could not have the desired effect discussed above, of encouraging the defendant to preserve the information.

For example, if the defendant had stored information on a type of video tape that would disintegrate over time, the threat of damages could not induce the defendant to alter that behaviour if it had no reason to suspect that the tape had that characteristic. Similarly, the threat of damages could not induce it to protect itself against unforeseen floods or acts of terrorism.

In such cases, therefore, the courts’ rulings could not have the desired effect that I discussed above; namely, that of encouraging defendants to preserve valuable information. Rather, the only effect that imposition of damages could have in such cases would be to make the defendant the “insurer” of the plaintiff – a result that the courts have often rejected. Therefore, in cases of unforeseen destruction of evidence, the defendant should be excused from any liability to the plaintiff.

The more complex cases are those in which defendants recognised that their actions (or inactions) might lead to the destruction of evidence, but failed to alter their behaviour accordingly. For example, the defendant might have recognised that information would be lost if certain documents became wet but failed to provide storage facilities that were protected against moisture.

The efficient rule in this situation is to place the onus on the defendant to prove that the evidence does not have the import claimed by the plaintiff if it can be shown that the defendant’s actions (or inactions) were negligent. As with the rule concerning the intentional destruction of evidence, discussed above, this rule would leave the decision concerning the destruction of information in the hands of the party that is able to determine the probability that that information will be relevant to the assessment of liability.

As long as the defendant is aware that the information in its possession may be relevant to the determination of liability, this rule will normally induce the defendant to take all those precautions necessary to avoid a finding of negligence. That is, rather than face the prospect of being found liable for the plaintiff’s damages, the defendant will normally prefer to meet its standard of care. As this is the desired outcome – the court would never ask the defendant to take more precautions than necessary to meet its standard – this is the efficient rule.

Spoliation

In some jurisdictions the courts have been asked to treat the destruction of evidence as a tortious act, independently of whether that destruction affected the determination of liability. Under this tort, often called spoliation, the plaintiff asks that the defendant be punished for the harm it has caused to the legal system.

Implicitly, the argument in this article has been that no independent tort need be established. The harm caused by the destruction of evidence is that both the compensatory and deterrent effects of tort law are impaired. If the underlying function of the law is to ensure that innocent victims are compensated, for example, the destruction of evidence may prevent that function from being performed.

If a set of rules can be designed that induces defendants to take all reasonable steps to preserve relevant information, the basic function of tort law will also be preserved. But that is precisely what the rules described in this article can be expected to do. Hence, no additional rules – such as punishing the defendant for the destruction of evidence that could not reasonably have been expected to assist the court in its deliberations – are necessary.

Summary

My argument in this article has been that the goal of rules concerning the preservation of evidence must be to induce defendants to avoid the intentional or negligent destruction of any information that they believe may be useful in the determination of liability (or damages). I have argued in this paper that a sufficient rule is that a defendant that has intentionally or negligently destroyed evidence be treated as if liability had been found against him or her.

Furthermore, no distinction should be made between those cases in which the defendant argues that the evidence would not have been of value to the court and those in which it admits that the evidence would have been relevant. The only defences available to the defendant should be (a) that the destruction of the evidence was unforeseeable; or (b) that the destruction of the evidence had occurred even though the defendant had met its standard of care.

Christopher Bruce is the President of Economica and a Professor of Economics at the University of Calgary. He is also the author of Assessment of Personal Injury Damages (Butterworths, 2004).

by Christopher Bruce

This article was originally published in the winter 1998 issue of the Expert Witness.

This is the third in a series of articles in which I examine the application of economic reasoning to questions of liability in torts. In the previous two articles, I argued that the principles of tort liability can best be understood if it is assumed that the goal of the courts has been to deter future inefficient behaviour – rather than to compensate victims for past harms.

One corollary of that analysis is that if the threat of tort damages is not the most effective means of preventing a particular type of harm, the courts should refuse to treat the case under the rubric of tort law. In short, one would expect that there would be a “gatekeeper” doctrine in law that would allow the courts to divide the cases appearing before them into two streams: “tort cases” and “not tort cases.”

Economists argue that “duty of care” rules act as this doctrine. That is, to say that a party owes a duty of care is tantamount to saying (i) that any (potentially) negligent behaviour in which the party engages could be deterred by threat of tort damages; and (ii) that tort law is the most efficient technique for deterring any such behaviour. The advantage of viewing duty of care as having this gatekeeping function is that it provides a relatively simple framework in which to understand one of the most complex and misunderstood areas of tort.

Foreseeability

In some cases, it is clear, even before the court has heard evidence concerning the actions of a party, that the threat of tort damages could not have induced that party to change his or her behaviour. The clearest instance of this situation is that in which the party in question could not have foreseen that its actions had the potential to cause an injury.

In the classic Canadian case of Nova Mink v. Trans-Canada Airlines [1951] 2 D.L.R. 241, a low-flying airplane so scared the animals in a commercial mink farm that they ate their young, causing the owner considerable harm. The airline was held to owe no duty of care to the mink farm and, therefore, was not required to pay damages.

This decision is consistent with the view that tort actions are to be allowed only when they can deter harmful behaviour. (And it is strongly inconsistent with the view that the function of tort law is to compensate “deserving” plaintiffs.)

To have ruled in favour of Nova Mink would have established a precedent to the effect that injurers owe a duty even when they cannot foresee the consequences of their actions. Yet when those consequences could not (reasonably) have been foreseen no precautions against such consequences could have been taken. Therefore, any court action in such a situation could have produced no change in the behaviour of the parties. It could only have resulted in a transfer of income from the defendant to the plaintiff, at a great cost (in terms of judicial expenses) to society.

The Misfeasance/Nonfeasance Distinction

Even if the defendant has foreseen the harmful event, he/she will often not be found to owe a duty of care if his/her failure to act is one of nonfeasance rather than misfeasance. If it is the actions of the defendant which create the circumstances in which a third party may be harmed, failure to take precautions to avert that harm is called misfeasance. In that circumstance, the defendant will be held to owe a duty of care. If, however, the defendant has merely observed that a third party may be harmed if a certain precaution is not taken, and has not taken that precaution, that failure to act is termed nonfeasance. In that circumstance, the defendant may be found to owe no duty of care (assuming that he/she did not create the circumstances – i.e. that he/she was not also a misfeasor).

For example, if A knocks down a stop sign and lack of that sign subsequently contributes to the injury of B at that intersection, A may be found to have owed a duty of care to B – and may be found negligent for having failed to report the initial accident. On the other hand, if, after A has knocked over the stop sign, C notes the absence of the sign and fails to report that fact, C will not be found to have owed a duty of care to B.

On economic grounds this distinction initially appears arbitrary. If it is efficient for a person who knocks over a stop sign to report that fact to the authorities, it must also be efficient for an individual who observes that a stop sign has been knocked over to report that fact. How, then, can the difference in legal duty between these two situations be reconciled?

The answer may lie in the relative difficulty of identifying potential defendants. When A has knocked over the stop sign it will be much simpler to identify him as the defendant, ex post, than it will be to so-identify “innocent” passerby C. Whereas there will be only one individual like A (or at least a very limited number of such individuals), who will generally leave evidence of their involvement; there may be a very large number of individuals like C. Furthermore, very few individuals like C will leave any evidence of their presence at the scene. And most, if identified as being present, will be able to deny plausibly any knowledge of the potential harm, or may be able to argue that they thought someone else was attending to the matter. Thus, whereas the pursuit of efficiency may require that the individual whose actions initiate a harmful situation owe a duty to those who are (potentially) harmed, that pursuit may require that some alternative mechanism be employed to induce third parties to offer their assistance.

One such alternative would be to offer third parties incentives to induce involvement, rather than deterrents to prevent non-involvement. That is, the common law might provide a means by which those who performed “good deeds” – benefactors – would be able to force those who benefited from those deeds – beneficiaries – to provide them with rewards. The advantage of this approach, in terms of the analysis of the preceding paragraph, is that the problem of identifying potential benefactors, ex post, would be avoided. Those who observe a (potentially) harmful situation and feel that the benefit of their actions will exceed the costs will present themselves as “rescuers”, that is, they will become involved in the attempt to rectify the harmful situation.

In fact, we observe that if the costs of identifying benefactors are low relative to the benefits of the rescue, the law does operate in this manner. Awards are provided to those who rescue salvage at sea; doctors can charge fees to individuals whom they have rescued from imminent danger; and individuals who have stored lost property can claim for their expenses.¹

The preceding analysis also helps to explain why a duty of care is found in one class of (apparent) nonfeasance – that in which it is inexpensive to identify the potential benefactor, ex post. In particular, a duty is often owed in situations in which the nonfeasance occurs on the property of the potential benefactor and in cases in which the benefactor has a pre-existing legal and/or contractual relationship with the beneficiary. For example, a homeowner has a duty to visitors to keep his sidewalks clear of ice; a municipality may have a duty to the users of its roads to ensure that stop signs are erected (and maintained) at dangerous intersections²; and shopkeepers may have a duty to their customers to ensure the safety of their premises. In each case, there is a party who is clearly-identifiable, ex post, who could have acted to protect the plaintiff.

There is also an efficient exception to this exception. Various altruistic groups – usually governments and charities – offer free services that may be interpreted as the provision of warnings concerning potential “harms”. For example, a local government may offer to send out its engineers to check earthen banks to ensure that there is no danger of them becoming unstable.³ Failure to respond to a request to provide these services clearly constitutes a nonfeasance. Yet, although the nonfeasor is easily identified, no duty of care is found. The reason that this ruling may be considered to be efficient is that if the “altruist” is found liable for failing to provide a service, the response of the altruist can be expected to be to withdraw that service. Such an outcome cannot generally be considered to be in society’s best interest.

Conclusion

The economic analysis of torts leads to the suggestion that the function of duty of care rules is to act as a “gatekeeper,” separating tort cases from non-tort cases. If the harmful behaviour of either party could have been deterred through the threat of tort sanctions, (and if such sanctions are the most efficient method for altering that party’s behaviour), the case should be considered to fall within the rubric of tort law. Otherwise, the gatekeeper should redirect the case away from the tort system.

Footnotes

1. For an economic analysis of the laws concerning “rescue”, see W. Landes and R. Posner, “Salvors, Finders, Good Samaritans, and Other Rescuers….,” 7 Journal of Legal Studies (January 1978), pp. 83-128. [Back to text]

2. See Anderson v. County of Ponoka (1980) 12 A.L.R. 320. [Back to text]

3. See Windsor Building Supplies v. Art Harrison Ltd. (1980) 14 C.C.L.T. 129. [Back to text]

Christopher Bruce is the President of Economica and a Professor of Economics at the University of Calgary. He is also the author of Assessment of Personal Injury Damages (Butterworths, 2004).

by Christopher Bruce

This article was originally published in the autumn 1998 issue of the Expert Witness.

In the summer 1998 issue of this newsletter, I introduced the academic discipline of “the economic analysis of tort law.” I argued in that issue that the traditional legal analysis of torts employs an ex post orientation. That is, legal scholars traditionally assume that the function of tort law is to compensate victims for harms that have already occurred. Economists, on the other hand, argue that the evolution of torts can better be understood if one uses an ex ante paradigm. In this paradigm, the courts behave as if they are less concerned with the resolution of the cases before them than they are with the establishment of precedents that will affect future behavior. In particular, economic analysis “predicts” that the courts will prefer those rules that encourage parties to select cost-minimising behaviors.

In the summer 1998 article, I provided examples of a number of situations in which the courts appeared to have made explicit use of ex ante reasoning. In this article, I will argue that the economic approach can also be used to cast light on the development of the tort rules of negligence.

Legal Versus Economic Analysis

If one adopts the proposition that the function of tort law is to compensate accident victims, it is difficult to rationalise the rules of negligence. Those rules require that the victim show that the defendant was negligent before compensation will be ordered. But as many defendants are not negligent, many victims are not compensated. Why would a body of law whose purpose was to compensate victims contain a major “escape clause” which would deny compensation to a large percentage of victims?

One could begin to answer this question by modifying the traditional argument. Perhaps tort law is not designed to compensate all accident victims, just those who are deemed “worthy.” But this begs the questions of who is worthy and why it is that the “worth” of the victim should be defined by the behavior of the defendant.

I will argue in this article, that negligence rules can better be understood if we view their purpose as being cost-minimisation. Under this approach, a party will only be found to have been negligent (and therefore potentially liable to pay damages) if he or she had failed to take some precaution for which the cost was less than the benefit (measured in terms of accident costs avoided). That is, I will argue that the function of the tort rules of negligence is to send signals to potential (future) “injurers” that if they fail to take appropriate precautions, they will be made to bear the costs that result.

Negligence Rules: an Economic Exposition

The economic model can best be understood using a numerical example. Assume the following “facts,” (based, loosely, on Anderson v. County of Ponoka [1980] 12 ALR 320):

• One of the stop signs at the intersection of two country roads is knocked over sometime on Saturday evening.
• The County responsible for those roads becomes aware of this on Sunday morning but decides to wait until Monday morning to replace the sign, in order to save $100 overtime pay to its road crew.
• Sunday evening, Mr. A, unaware that the sign is missing, assumes that he has the right-of-way, enters the intersection without slowing, and collides with Ms. B’s car.
• The two cars and their occupants suffer damages which total $25,000.
• At trial, the court accepts the evidence of a traffic expert that the probability, per day, that such an accident will occur is 3/1,000 if there is no stop sign in place and 1/1,000 if there is a stop sign.

Was the county negligent? Economic analysis predicts that the court will say “no.” Why? Assume that rural stop signs are frequently knocked down on Saturday evenings. If the relevant highway departments wait until Monday to replace the stop signs, there will be three accidents every 1,000 times a sign is knocked down. Hence, there will be $75,000 damages for every 1,000 such occurrences. ($75,000 = 3 x $25,000.) If the counties replace the stop signs immediately, the number of accidents will fall to one in every 1,000 occurrences, reducing the accident costs to $25,000, a saving of $50,000. But, in order to obtain that “saving,” counties will have to send out 1,000 repair crews at an overtime cost of $100 each, or $100,000 in total. The $50,000 “saving” will have cost $100,000.

Put another way, the average cost of precautions, per event, will be $100 and the average benefit of those precautions (measured in terms of accident costs saved) will be (2/1,000) x $25,000, or $50. As the economic model predicts that the court will only encourage behaviour whose cost is less than the benefit, the economic prediction is that the court will not find the county to be negligent in this case.

It can be seen from this case that three factors are predicted to enter the court’s calculations:

• the cost to the defendant of taking a precaution to avoid the accident, (C);
• the probability that a precaution which could have been taken by the defendant would have prevented the accident, (P); and
• the expected cost of the accident, (A).

In particular, it is predicted that the defendant will be found to have been negligent if there was some precaution, not taken by the defendant, whose cost was less than the cost of the accident multiplied by the decrease in the probability of an accident which would have occurred had that precaution been taken. In algebraic terms, the party is found negligent if C < (A x P).

Is the Law Consistent With the Economic Model?

In the U.S., this prediction was confirmed in one of the leading cases on negligence, U.S. v. Carroll Towing. In that case, Justice Learned Hand concluded that negligence was to be found only if the burden (cost) of precautions was less than the probability of the accident multiplied by the gravity (cost) of the accident — precisely the formulation which I derived above from the economic model.

In British/Canadian jurisprudence, confirmation of the prediction is less direct, but persuasive nevertheless — sufficiently persuasive that in recent editions of Canadian Tort Law Allen Linden has organised his discussion of the rules of negligence around the “Learned Hand rule.” In Wagon Mound No. 2, for example, the court concluded that a party could be found negligent even if the probability of an accident was low as long as the cost of the accident was high. Arguably, it was the court’s view that the cost of the accident multiplied by the probability that it could be avoided should be weighed against the cost of avoidance in order to determine negligence — again, precisely the prediction made by economic reasoning. Other leading cases which are consistent with the economic model include Bolton v. Stone, Priestman v. Colangelo, and Reibl v. Hughes.

Applying Economic Analysis to the Law

Many “day-to-day” cases also employ reasoning which is consistent with the economic approach to the determination of negligence. For example, in Hewson v. City of Red Deer (1977) 146 DLR (3d) 32 (Alta. CA), a City employee left the keys in the ignition of a bulldozer. Subsequently, the bulldozer was stolen and driven into the side of Hewson’s house. The City was found not to be negligent largely because (a) the bulldozer was left two blocks from Hewson’s house; (b) it was left at midnight; and (c) the operator was absent for only 25 minutes. All three of these factors suggest that the probability of an accident was quite low. And the first factor suggests that the average damages which might occur if the bulldozer was stolen were low (because the bulldozer would have to be driven a long distance before causing any harm.)

In Weaver v. Buckle (1982) 35 AR 97 (Alta. QB), Weaver (a child) ran out in front of Buckle’s car and was injured. The court implied that it would not normally have found Buckle to be negligent for causing this accident. However, as the road was narrow, it found that he should have been driving more slowly, to take account of the general conditions of that road. The court concluded that if he had been driving more slowly, the probability that this accident would have occurred would have been reduced substantially. Buckle was found 60 percent liable. In economic terms, this finding suggests that it is not simply this accident which determines the “costs of an accident” (A, in my terminology). Rather, it is all accidents which might have been prevented had the defendant taken additional precautions.

In Jordan v. Schofield (1996) 148 NSR (2d) 104 (NSSC), Schofield’s 7 year-old son played with a lighter and caused a fire in an apartment building belonging to Jordan. The court concluded that Schofield was not negligent. Although parents are responsible for taking “reasonable” precautions to watch their children and to put harmful things out of their way, at some point the costs of additional precautions become prohibitive. Parents will not be found negligent for failing to take precautions beyond that point. For example, parents will not be negligent for leaving children unattended around “ordinary” dangers (such as knives or scissors) for a few minutes. They may, however, be negligent for leaving their children unattended around such dangers for longer periods of time, or for leaving them for only a few minutes around more dangerous items (such as fires burning in fireplaces). In economic terms, the cost of taking additional precautions is to be weighed against the probability that an accident will occur if those precautions are not taken.

To conclude, given that no Canadian court has formally adopted the “Learned Hand rule” it would be difficult to base an argument before our courts on economic reasoning alone. Nevertheless, if one accepts that the three components of the “Learned Hand Rule” (C, P, and A) play an implicit role in the determination of negligence, an understanding of their function may help to clarify the legal analysis in difficult cases.

Christopher Bruce is the President of Economica and a Professor of Economics at the University of Calgary. He is also the author of Assessment of Personal Injury Damages (Butterworths, 2004).

by Christopher Bruce

This article was originally published in the summer 1998 issue of the Expert Witness.

Economists have made important contributions to the analysis of many areas of the law – particularly competition law, labour law, regulation, and international trade – during the last 200 years. It is only in the last quarter century, however, that economists – and legal scholars using economics – have turned their attention in a systematic way to the analysis of torts, contracts, and criminal law. In spite of the youth of the sub-discipline that investigates these branches of the law, it has quickly become a major force within U.S. and, to a lesser extent, Canadian and European law schools. All of the major U.S. law schools – Harvard, Yale, Princeton, Chicago, Stanford, and Berkeley among them – now offer courses in the Economic Analysis of Law and have professors with Ph.D.s in Economics on faculty. (Only the University of Toronto, among Canadian schools, has followed suit.)

Those reading this newsletter will be familiar with some elements of the economic analysis of personal injury damages – for example, through my textbook Assessment of Personal Injury Damages (Butterworths, 1992). But economics, being the imperialistic science that it is, has not stopped there. Economic models have been applied to virtually every aspect of tort law – negligence rules, causation, onus of proof, intentional torts, informed consent, volenti, to name only a few. I do not wish to claim that the economic analysis of these doctrines should supplant the traditional legal analysis. However, I do think that there may be situations in which practitioners may find it useful to consider some of these issues from a different angle.

In this article, and a number of others to follow it, I would like to provide some insight into one such angle – the economic analysis of torts. In this introductory article, I begin by identifying the primary characteristics that distinguish the economic approach from more traditional methods of legal analysis. I then use this approach to discuss collateral benefits and negligence rules.

Characterising the Economic Analysis of Torts

Two fundamental characteristics distinguish the “economic” analysis of torts from other approaches to the study of tort law.

First, economists take a “positive,” or “scientific” approach to the identification of legal doctrines. Instead of trying to determine what the law “should be,” economic analysts attempt to determine what the law “is.” That is, they use the deductive approach to derive hypotheses about the principles which underlie judges’ reasoning and then test those hypotheses by comparing their predictions against the decisions which judges have made. As a simple example, economic analysis can be used to “predict” that the courts will, under most circumstances, reject the defence of “custom.” That prediction can be “tested” by observing whether the courts do or do not accept that defence.

Second, all economic analysis of tort law begins from the working hypothesis that judges behave as if they were attempting to devise legal rules which would encourage individuals to maximise social benefits net of social costs. (For example, if there is some accident-avoiding behaviour whose cost is less than the resulting saving in accident costs, the courts are predicted to adopt rules which will encourage adoption of that behaviour.) It is not argued that judges consciously act in this way; simply that the doctrines that have been selected by the common law courts have developed as though this was the goal of the courts.

This view of the functioning of the courts suggests that the courts will behave as if they were employing an ex ante (or “forward-looking”) approach to decision-making. In this approach, the courts recognise that any decision they make in the current case may influence the behavior of parties in similar, future cases. Hence, it becomes important to set a precedent which will direct future parties to behave in the socially desirable manner.

This approach can be contrasted with the traditional view of the court’s decision-making process, which I call the ex post (or “backward-looking”) approach. In this approach the court is assumed to take the position that, as the tortious act has already occurred, that act cannot be undone. Rather, all the legal system can do is to ensure that the victims are restored, as well as possible, to the position they would have been in had the act not occurred. Contrary to the economic assumption, no thought is given to the impact which decisions will have on future behaviour.

The ex post view is common to most textbooks and was given its most famous expression by the Supreme Court of Canada in Ratych v. Bloomer. There, Justice McLachlan concluded that the function of damages in tort law was to “restore the plaintiff to his pre-accident position.” Further, she emphasised that

[t]he law of tort is intended to restore the individual to the position he enjoyed prior to the injury rather than to punish the tortfeasor whose only wrong may have been a moment of inadvertence. [Emphasis added]

That is, the Court has said that tort damages are intended strictly to compensate harms that occurred in the past, not to deter negligent behavior that might occur in the future.

The response which those who rely on “positive” analysis of the law make to this argument is that the most reliable way to determine what someone thinks is to observe what they do, not what they say. In short, the best way to identify the underlying principles of tort law is to review the courts’ decisions, not their arguments. What I propose to argue in the following sections is that the courts’ decisions can often be more easily understood if it is assumed that they are trying to influence future behavior than if it is assumed they are attempting to “right past wrongs.”

The Collateral Benefits Rule

A clear example of the courts saying one thing and doing another arises in their interpretation of the “collateral benefits rule.” On the one hand, the Supreme Court has made it clear that it prefers the ex post approach. On the other hand, the trial courts have consistently adopted the ex ante approach.

1. Orphaned Children: Consider for example the situation in which orphaned children have been taken into the care of relatives.* Although Ratych would appear to suggest that the children’s claim for loss of dependency was thereby extinguished, most of the decided cases have rejected this view. The trial courts have recognised that the plaintiffs would be “double compensated” but have argued that to deny compensation would be to establish a dangerous precedent for future cases.

The leading statements of the latter view appear in Tompkins (Guardian ad litem of) v. Byspalko (1993) 16 C.C.L.T. (2d) 179 and Ratansi v. Abery [1995] 5 B.C.L.R. (3d) 88. In both cases, the trial judges argued that if Ratych was followed, the risk would be created that

… in some cases, family members who would otherwise take orphaned children into their care may decline to do so until or unless an award has been made in the children’s favour.

And in Tompkins, Spencer, J. went further, arguing that “… a surviving parent may refrain from remarriage, advantageous from the children’s point of view, because the presence of a new spouse who replaces services to the children may reduce their award”

2. Charitable Donations: Similarly, the rationale that is commonly given for the “charity exception” is that to deny a plaintiff compensatory damages because he or she had received a charitable donation would discourage individuals from making those donations. A clear example of this principle was stated in the Northern Ireland case of Redpath v. County Down Railway [1947] N.I.L.R. 167 where Andrews L.C.J. noted that if

the proposition contended by the defendants is sound the inevitable consequence in the case of future disasters of similar character would be that the springs of private charity would be found to be largely if not entirely dried up.

Surprisingly, further confirmation of this view comes from the author of Ratych, Madam Justice McLachlan. In her dissent in Cunningham v. Wheeler (1994) 113 D.L.R. (4^th) 1 she argued that “… people should not be discouraged from aiding those in trouble.”

3. Implications: The common thread running through all of these decisions, I would argue, is that the courts will often consider the impact that their current rulings can be expected to have on individuals’ future behavior. In this view, the function of torts is not merely to compensate particular plaintiffs for past wrongs, but is also to protect potential plaintiffs from future harmful behavior. Children whose parents have been killed are to be protected against the possibility that their relatives may delay the adoption process; and victims of catastrophic events are to be protected against the possibility that donors may be discouraged from providing assistance.

This view opens a number of interesting possibilities for argument in similar future cases. For example, the practice has been to assume that a widow(er)’s loss of dependency comes to an end once she (he) remarries (assuming that the new spouse has a similar income to the first spouse). It could be argued, however, that this rule may encourage widow(er)s to postpone any relationships with the opposite sex until after the fatal accident case has been settled. As it cannot be in the public interest to discourage dating and marriage, a legal rule which has the effect of providing that discouragement may well be contrary to public policy.

In each case, economic analysts of the law would argue that the courts were behaving as if their goals were to encourage (socially) desirable behaviour and to discourage (socially) undesirable behaviour. In the next section, I will argue that it is rules of negligence which distinguish desirable from undesirable behaviour.

Negligence Rules

Assume the following facts:

• One of the stop signs at the intersection of two country roads is knocked over sometime on Saturday evening.
• The County responsible for those roads becomes aware of this on Sunday morning but decides to wait until Monday morning to replace the sign, in order to save $1,500 overtime pay to its road crew.
• Sunday evening, Mr. A, unaware that the sign is missing, assumes that he has the right-of-way, enters the intersection without slowing, and collides with Ms. B’s car.
• The two cars and their occupants suffer damages which total $5,000.
• At trial, the court accepts the evidence of a traffic expert that the probability, per day, that such an accident will occur is 3/10 if there is no stop sign in place and 1/10 if there is a stop sign.

Was the county negligent? Traditional, ex post legal analysis has difficulty answering this question definitively. On the one hand, ex post analysis holds that the function of tort law is to compensate “worthy” victims, creating a presumption that the county should be found responsible. On the other hand, that analysis also argues that a defendant should only be found liable if he or she failed to take those actions that would have been taken by a “reasonable” person. But what actions would have been reasonable in this case? I will argue that the answer the courts will usually give to this question is consistent with the ex ante, or economic, analysis of the law.

In particular, if the function of the law is to encourage behaviour that maximises social benefits minus social costs (the economic prediction), a “reasonable” action will be one for which the benefits exceed the costs. That is, economic analysis predicts that the county will be found negligent only if the cost of ensuring that the stop sign was re-erected exceeded the benefit of doing so. In this section, I will show that the factors that enter the determination of those costs and benefits are the same as those that the courts usually take into account when determining negligence.

Assume that rural stop signs are frequently knocked down on Saturday evenings. If the relevant counties wait until Monday to replace their stop signs, there will be three accidents every 10 times a sign is knocked down. Hence, there will be $15,000 damages for every 10 such occurrences. ($15,000 = 3 x $5,000.) If the counties replace the stop signs immediately, the number of accidents will fall to one in every 10 occurrences, reducing the accident costs to $5,000, a saving of $10,000. But, in order to obtain that “saving,” counties will have to send out 10 repair crews at an overtime cost of $1,500 each, or $15,000 in total. The $10,000 “saving” will have cost $15,000. Put another way, the average cost of precautions per event (knocked over stop sign) will be $1,500 and the average benefit of those precautions (measured in terms of accident costs saved) will be (2/10) x $5,000, or $1,000. (Note: the reduction in the probability of an accident, when the county sends out a repair crew, is only 2/10 because the crew does not reduce the probability of an accident to zero.) As the economic model predicts that the court will only encourage behaviour whose cost is less than the benefit, the economic prediction is that the court will not find the county to be negligent in this case.

It can be seen from this example that three factors were predicted to enter the court’s calculations:

• the cost to the defendant of taking an additional precaution to avoid the accident, (here, $1,500);
• the probability that an additional precaution would have prevented the accident, (here 2/10); and
• the expected cost of the accident, (here, $5,000).

In the U.S., this prediction was confirmed in one of the leading cases on negligence, U.S. v. Carroll Towing. In that case, Justice Learned Hand concluded that negligence was to be found only if the burden (cost) of precautions was less than the probability of the accident multiplied by the gravity (cost) of the accident – precisely the formulation which was derived from the economic model.

In British/Canadian jurisprudence, confirmation of the prediction is less direct, but persuasive nevertheless – sufficiently persuasive that in recent editions of Canadian Tort Law Allen Linden has organised his discussion of the rules of negligence around the “Learned Hand rule.” In Wagon Mound No. 2, for example, the court concluded that a party could be found negligent even if the probability of an accident was low as long as the cost of the accident was high. Arguably, it was the court’s view that the cost of the accident multiplied by the probability that it could be avoided should be weighed against the cost of avoidance in order to determine negligence – again, precisely the prediction made by economic reasoning. Other leading cases which are consistent with the economic model include Bolton v. Stone, Priestman v. Colangelo, and Reibl v. Hughes.

Conclusion

Economists often define their discipline to be the analysis of “the allocation of scarce resources among competing ends.” When this approach is applied to common law, it suggests that one of the functions of torts might be to establish rules that encourage individuals to use resources effectively. Common law precedents should not discourage relatives from adopting orphans, for example. Nor should they find defendants to be liable if they have taken all precautions for which the benefits (of those precautions) exceed the costs.

In this article, I have argued that this “law and economics” method of analysing the common law predicts that

• the courts will employ the ex ante approach when resolving tort disputes; and
• they will base the determination of negligence on: the probability of an accident occuring, the costs of the accident, and the costs of avoiding the accident.

I have also provided evidence that, at least in some cases, the Canadian courts have followed this approach.

I do not wish to conclude from this that the courts should follow the economic approach, nor that they will always adopt it. I merely offer it as another tool for those who are looking for an underlying rationale to the courts’ behaviour. Perhaps in some cases, the advocate and the court will find it of value to think explicitly in terms of the signal which the decision in the current case will send to others in situations similar to those in which the plaintiff and defendant found themselves. In future articles in this series I will discuss the economic analysis of such doctrines as custom, causation, duty of care, volenti, and restitutio in integrum.

Footnotes

* The foregoing analysis is based on an article that I wrote for The Lawyers Weekly (April 24, 1998). This article is also available, as “Duty to Care for Orphaned Minors,” on this website. [back to text of article]

Christopher Bruce is the President of Economica and a Professor of Economics at the University of Calgary. He is also the author of Assessment of Personal Injury Damages (Butterworths, 2004).

by Christopher Bruce

This article first appeared in the spring 1996 issue of the Expert Witness.

In Muir v. Alberta damages were awarded to the plaintiff on two grounds: first, that she was wrongfully confined, at the age of 10, in a home for the mentally defective; and, second, that while so confined, she was wrongfully sterilized. On the first of these claims, she was awarded $250,000 plus $115,500 interest for pain and suffering but was denied both aggravated damages and damages for loss of income. On the second claim, she was awarded $250,280 for pain and suffering and $125,000 for aggravated damages but was denied punitive damages.

Madam Justice Veit denied the claim for loss of earnings primarily on the ground that Ms. Muir had come from a dysfunctional family, leading her to suffer from severe emotional problems prior to her wrongful confinement. The confinement itself was found not to have exacerbated these problems.

Does this imply that all individuals in Ms. Muir’s situation will be denied damages for loss of earnings? We think not. Three sources of claims for lost earnings appear to have survived the decision in Muir.

• First, if the plaintiff did not come from a dysfunctional family, a claim for loss of earnings could arise from the wrongful confinement.
• Second, it might be argued that, had the plaintiff been placed in a foster home or group home for the care of emotionally disturbed children (possibilities which were canvassed by Madam Justice Veit), she would have overcome the effects of her dysfunctional upbringing. Hence, a loss of earnings would have arisen from the government’s failure to take advantage of one of these alternatives.
• Finally, it is possible that a claim for loss of earnings could arise from the action for wrongful sterilization. Madam Justice Veit concluded that the “…sterilization had a catastrophic impact on Ms. Muir.” (Emphasis added, p. 59) She also accepted a psychologist’s opinion that sterilization can have “profound detrimental effects on … education…” (p. 46) and a psychiatrist’s testimony that the impact of sterilization on “…a young woman … would be hard to over-estimate.” (p. 38) On these bases, it could be argued that a wrongful sterilization had impaired the capacity to earn income.

Christopher Bruce is the President of Economica and a Professor of Economics at the University of Calgary. He is also the author of Assessment of Personal Injury Damages (Butterworths, 2004).