Posner:
    31: Tort law: "concerned with protecting property rights, including the right to bodily integrity."

Posner
Ch.6 tort law
6.1 the economics of accidents and the learned hand formula of liability for negligence (167-171)

    When, as in this example, the person taking precautions and the person who may be injured if they are not taken are the same, the optimal precautions will be achieved without legal intervention. (P167)
    Hand wrote that a potential injurer is negligent if but only if B < PL, which is what our example implied would be the formula for optimal accident avoidance. (P168)
    See United States v. Carroll Towing Co., 159 E2d 169, 173 (2d Cir. 1947). For contemporary applications of the formula, see Dobson v. Louisiana P~ower & Light Co., 567 So. 2d 569, 574-575 (La. 1990); Cross v. Berg Lumber Co., 7 P. 3d 922,936 n. 3 (Wyo. 2000); Brotherhood Shipping Co. v. St. Paul Fire & Marine Ins. Co., 985 F. 2d 323, 327-329 (7th Cir. 1993); Barnmerlin v. Navistar Intl. Transport. Corp., 30 E3d 898, 902 (7th Cir. 1994); I & M Rail Link, LLC v. Northstar Navigation, Inc., 198 F.3d 1012, 1016 (7th Cir. 2000). (P168n)
    . This example shows that expected accident costs and accident prevention costs must be compared at the margin, by measuring the costs and benefits of small increments in safety and stopping investing in more safety at the point where another dollar spent would yield a dollar or less in added safety. (P168)
    A possible objection to the Hand Formula is that it assumes risk neutrality. We waved this point aside by pointing out that people can buy insurance, and yet we know from Chz/pter 4 that the doctrines of contract law do not on similar grounds assume that the people affected by that law are risk neutral. (P168-169)
    It is difficult to buy insurance against business losses, because a businessman insured against loss will have little incentive to try to prevent loss and because business losses are potentially so open-ended; an unexpected rise in the price of a key input could cause even a large business to go broke. (P169)
    However, there have been well-developed markets in insurance against personal injury and death for a long time, and although this insurance also involves, as we discussed in Chapter 4, a "moral hazard" problem, it is not so severe (why not?), or the losses to be insured against so open-ended, as to prevent the market from working. (P169)
    Although the Hand Formula is relatively recent, the method it capsulizes has been used to determine negligence ever since negligence was first adopted as the standard to govern accident cases. (P169)
    For example, in Blyth v. Birmingham Water Works4 the question was whether a water company had been negligent in failing to bury its water pipes deep enough to prevent them from bursting because of frost and damaging the plaintiff's home. (P169) 11 Exch. 781, 156 Eng. Rep. 1047 (1856).
    In Adams v. Bullock, 5 a 12-year-old boy, while crossing a bridge over the defendant's trolley tracks, swung an 8-foot-long wire over the bridge. (P169) 227 N.Y. 208, 125 N.E. 93 (1919) (Cardozo, J.).
    And here is a case that went the plaintiff's way: Hendricks v. Peabody Coal Co.  A 16-year-old boy was seriously injured while swimming in the defendant's abandoned strip mine, which had become filled with spring water. 115 Ill. App. 2d 35, 253 N.E.2d 56 (1969). (P170)
    A reasonable man would only neglect such a risk if he had some valid reason for doing so, e.g., it would involve considerable expense to eliminate the risk [high B]. (P170)
    Wagon Mound (No. 2), Overseas Tankship (U.K.), Ltd. v. Miller Steamship, [1966] All E.R. 709, 718 (Privy Council). (P170n)

6.2 the reasonable-person standard (171)

    It might seem that if the Hand Formula really created the proper incentives to avoid negligent accidents, no one would be negligent and there would be no negligence cases, at least none won by plaintiffs. (P171)
    But of course there are many negligence cases won by plaintiffs, and one explanation--besides the obvious one that judges and y jurors make mistakes--is that in deciding whether an accident could have been avoided by either party at a cost less than the expected accident cost, the courts do not attempt to measure the actual costs to the parties, taking account of their individual capacities for avoiding accidents. Rather, they estimate the accident-avoidance costs of the average (in legal parlance "reasonable") person in each party's situation. This approach is justified by the costs of individualized measurement. Or suppose the average accident-avoidance cost was only $50 but some people could not avoid the accident at a cost of less than $110. Making them liable would not affect their behavior; it would shift rather than reduce accident costs. (P171)
    Where differences in capacity to avoid accidents are ascertainable at low cost, the courts do recognize exceptions to (or subclasses of) the reasonable-person rule. For example, blind people are not held to as high a standard of care as sighted ones, although within the class of blind people a uniform standard of care is imposed. (P171)
    The discussion in this section is a reminder that administrative costs, and in particular information costs, play an important role in the formulation of efficient legal rules. (P171)
    Some people who are classified as negligent by the legal system cannot in fact avoid the expected cost of a negligence judgment at a lower cost in precautions. So they go ahead and have accidents for which the legal system deems them negligent. They are acting efficiently--and so is the legal system, when administra-tive costs are taken into account as they must be in a complete economic analysis. (P171)

6.3 custom as a defense (171-172)

    Since in these circumstances there is no presumption that the average safety level in the industry is optimum, the law properly rejects compliance with custom as a defense. Where, however, the type of accident is dangerous only to the industry's customers, the level of precau-tions taken by sellers is more likely to be efficient. (P172)
    It is therefore ironic that the classic statement of the principle that compliance with custom is not a defense to a negligence action should have been made--and by Judge Hand!--in a case in which the plaintiff was the defendant's customer. (P172) The T.J. Hooper, 60 F.2d 737 (2d Cir. 1932).
    In one area of negligence, that of medical malpractice, the courts, consistent with the distinction just suggested, have traditionally allowed a defense of custom. (P172)

6.4 victim: contributory and comparative neglgence, assumption of risk, and duties to trespassers (172-177)

    Since, as every pedestrian knows, many accidents can be prevented by victims at lower cost than by injurers, the law must be careful not to impair the incentives of potential accident victims to take efficient precautions. (P172)
    Suppose that an expected accident cost of $1,000 could have been avoided by the defendant at a cost of $100 but by the plaintiff at a cost of only $50. The efficient solution is to make the plaintiff "liable" by refusing to allow him to recover damages from the defendant. (P172)
    The traditional common law approach, which goes by the name of "contributory negligence," was, after asking whether the defendant had been negligent and concluding that he had (if he had not, that would be the end of the case), to turn around and ask whether the plaintiff had been negligent. If the answer was "yes," the plaintiff lost. (P173)
    However, the appearance is misleading, provided that the law defines due care--as it does--as the care that is optimal if the other party is exercising due care. (P173)
    Why won't the court listen to the defendant if he says that if the plaintiff were exercising due care, the defendant's optimal care would be zero, so he should be let off the hook? ...
    In the above examples the implicit assumption was that the case was one of "alternative care," that is, that the efficient solution was for one party, but not both, to take care. (P173)
    In a "joint care" case, in contrast, we want each party to take some care rather than one to take care and the other to do nothing. (P173)
    But will a negligence/contributory-negligence rule induce the parties to adopt it? It will. (P173)

    It will be destroyed, but since the railroad will be held negligent and the farmer will not be held contributorily negligent, the farmer won't care. (P174)
    Under a negligence regime, if the injurer is not negligent the victim will bear the whole cost of the accident whether negligent or not. The defense of contributory negligence comes into play only when the injurer is negligent as well as the victim. And if the injurer is negligent, why should he get off scot-free and the victim be left to bear the whole cost of the accident? The economic answer is that shifting the cost from the victim to the injurer will not do any good as far as creating incentives to take due care in the future is concerned, but will be costly. (P174)
    The injurer has an incentive to take care to avoid having to pay damages if he is careless, an accident occurs, and the victim was not careless; the victim has an incentive to take care to avoid the cost of the accident if it occurs though the injurer was careful. Since efficiency is not enhanced by making the negligent injurer pay damages to the negligent victim, the common law traditionally allowed the cost of the accident to lie where it fell, in order to minimize the costs of administering the legal system. (P174)
    Most states have, however, replaced contributory negligence with comparative negligence, whereby if both parties (injurer and victim) are negligent the plaintiff's damages are reduced, but not to zero. Surprisingly, comparative negligence has the same effects on safety as contributory negligence. (P174)
    It does not follow that there is no economic difference between contributory and comparative negligence. Comparative negligence entails a transfer payment that generates no allocative gain and transfer payments involve administrative costs. (P174-5)
    This requires the expenditure of additional resources by the parties and the courts, and by making it harder to predict the extent of liability may increase the rate of litigation. (P175)
    Yet it is unclear at the level of theory which rule--contributory negligence or comparative negligence--is more productive of uncertainty, and likewise unclear is the effect of uncertainty on the amount of care taken by injurers and victims respectively. (P175)
    Comparative negligence makes most economic sense when society wants to use the tort system to provide insurance to accident victims, because it gives the careless victim of a careless injurer something; contributory negligence gives him nothing. It is, therefore, no surprise that comparative negligence got its first foothold in admiralty law. The rule in collision cases when both vessels were at fault was that each party was liable for one-half of the total damage to both ships. (P175)
    But why in an age of much more widely available market insurance than when contributory negligence held sway in tort law there should be a desire to provide insurance through the tort system is a mystery to the positive economic theorist of the common law. (P175)
    Another important doctrine of victim responsibility is assumption of risk. It is like contributory negligence in being a complete bar to recovery of damages but differs in important respects that economics can illuminate. (P176)
    Another rule of victim responsibilits; although one in decline and subject to many exceptions, is that a landowner is not liable for negligent injuries to trespassers. (P176)
    [Ploof v. Putnam] But the plaintiff would probably have been liable to the defendant for any damage caused by his boat. Such liability is appropriate to assure that the rescue is really cost-justified, to encourage dock owners to cooperate with boats in distress, to get the right amount of investment in docks ... (P176n)
    But in cases of public necessity, as where the fire department pulls down a house to make a firebreak, compensation is not required. This illustrates a common technique of common law regulation, that of encouraging the provision of external benefits (saving the rest of the city from the fire) by allowing costs (the cost to the person whose house is pulled down) to be externalized. (P177) Incidentally, why might the cost to the owners of the houses be a good deal less than the pre-fire market value? (P177n)
    Ploof v. Putnam is a special application of the last clear chance doctrine. A man is using the railroad track as a path. Since he is a trespasser, the railroad has no duty to keep a careful lookout for him.... But if the crew happens to see him (and realizes he is oblivious to the train's approach), it must blow the train's whistle and take any other feasible precautions to avoid running him down. (P177)
    It might be thought that, were there no doctrine of last clear chance, there would be fewer trespassers and maybe, therefore, fewer accidents than with the doctrine. But this ignores the probabilistic character of taking care. To stray across the center line in a two-way highway is negligent, but everyone does it occasionally because it would be too costly to adopt a driving strategy that reduced the probability of straying to (or very close to) zero. (P177)

6.5 strict liability (177-182)

    Strict tort liability means that someone who causes an accident is liable for the victim's damages even if the injury could not have been avoided by the exercise of due care. (P177)
    this is like liability for breach of contract where the breach is efficient (P177n)
    Only if the benefits of the activity are obviously very slight, as where a man runs into a burning building to retrieve an old hat and does so as carefully as he can in the circumstances but is seriously burned nonetheless, will the court find that engaging in the activity was itself negligence, even though once the decision to engage in the activity was made, the actor (plaintiff or defendant) conducted himself with all possible skill and cir-cumspection. (P178)
    Accident costs unavoidable by due care and borne entirely by strangers to the industry (hence external to the industry) are assumed to be proportional to the industry's output. (P178)
    Under strict liability, MCs becomes the industry's private marginal cost curve, inducing it to reduce output from qo to q*, which results in eliminating socially wasteful accident costs (the shaded area in the diagram). (P179)
    The problem with using this analysis to support a general rule of strict liability is that changes in activity level by victims are also a method of accident avoidance, and one that is encouraged by negligence liability but discouraged by strict liability. (P179)
    So if a class of activities can be identified in which activity-level changes by po-tential injurers are the most efficient method of accident prevention, there is a strong argument for imposing strict liability on the people engaged in those activities. Conversely, if there is a class of activities in which activity-level changes by potential victims are the most efficient method of accident prevention, there is a strong argument for no liability, as by applying the doctrine of assumption of risk to participation in dangerous sports. (P179)
    Through the concept of ultrahazardous activ-ities, tort law imposes strict liability on activities that involve a high degree of danger that cannot feasibly be prevented by the actor's being careful or potential victims' altering their behavior. (P179-180)
    The category of ultrahazardous activities is not fixed; the tendency is to apply the label to new activities (often called nonnatural), such as reservoirs in England or ballooning in early nineteenth-century America. (P180) Guilie v. Swan, 19Johns. Ch. 381, 10 Am. Dec. 234 (1822).
    The trial of a strict liability case is simpler than that of a negligence case because there is one less issue, negligence; and the fewer the issues, the easier it should be to settle the case without a trial. On both counts we can expect litigation costs to be lower under strict liability than under negligence--for the same number of claims. But the number may not be the same. In principle, under strict liability, every accident to which there is more than one party gives rise to a claim, not just every accident in which the defendant may have been negligent. (P180)
    But if most of the accidents that occur in some activity are unavoidable in an economic sense either by taking greater care or by reducing the amount of the activity (because the costs of greater care, or less activity, exceed any savings in reduced accident costs), the main effect of switching from negligence to strict liability will be to increase the number of damages claims. (P181)
    Another difference is that strict liability operates to insure victims of unavoidable accidents. It is a gain only if the cost of insurance through the tort system is less than the cost to potential victims of buying accident insurance policies in the insurance market; almost certainly it is greater. (P181)
    Courts make mistakes; which regime--strict liability or negligence--is more robust against mistakes? On the one hand, an erroneous finding that an injurer is not negligent cannot have misallocative consequences under strict liability, because the injurer's negligence is not an issue. On the other hand, the consequences of a mistaken ascription of causation, or an overestimation of damages, are worse under strict liability. Under negligence, a person is sanctioned only for inefficient conduct; under strict liability, he may be sanctioned for efficient conduct, and if the actual costs of that conduct are exaggerated, the conduct may be deterred. (P181)
    Because of these differences between negligence and strict liability, we would not expect the tort system to opt all for one or all for the other. (P181)
    Negligence has a strict liability component (and in the next section we shall see that strict liability has a negligence component). This is a result in part of the reasonable-person rule, which makes persons having above-average costs of taking care strictly liable for their accidents, and in part of the doctrine of respondeat superior.... (P181)
    How, finally, to explain the greater role of strict liability in contract law than in tort law? (P181)
    The difference may reflect the greater availability of market insurance in tort cases (and hence the lesser value of providing insurance through the legal system) and the fact that contract cases are less likely than tort cases to involve an interactive mishap that either party could have prevented, although possibly at very different costs. Ordinarily one of the contracting parties is performer and the other payor. The former has complete control over performance, the latter complete control over payment. (P182)
    In contrast, most tort situations are collisions between two activities, such as driving and walking, and there is no basis for a general presumption, such as would warrant a general rule of strict liability, that the injurer was in a better position than the victim to have prevented the collision. (P182)

6.6 products liability (182-184)

    But the term strict liability is something of a misnomer here, because in deciding whether a product is defective or unreasonably dangerous in design or manufacture the courts generally use a Hand Formula approach, balancing expected accident costs against the costs of making the product safer. (P182)
    The manufacturer is liable only if the design or construction of the car was defective, which in most cases means just about the same thing as negligent. (P182)
    But a slight cost to each consumer, when aggregated over millions of consumers, may be substantial. (P182)
    But it will lead him to increase his price (why?); and although the increase will be small (again by hypothesis), it will lead some consumers to substitute other, and probably safer (why?), products. (P182-183)
    This is the doctrine of foreseeable misuse and is related to the point just made about the obvious hazard. A manufacturer sells a machine whose moving parts are not shielded, and a worker is injured when he sticks his hand in them. He was careless in doing so, the danger being apparent, and yet the manufacturer could have shielded the moving parts, and thus prevented the accident, at a trivial cost. In many states, he would be held liable to the worker. Is this an efficient result? Can you see an analogy to the doctrine of last clear chance? Should the doctrine of foreseeable misuse allow every negli-gent user of a product to recover damages, on the ground that it is well known that many consumers are negligent? (P183)

6.7 causation and foreseeability (184-188)

    A and B, out hunting, carelessly mistake C for a deer and shoot. Both hit him, and each shot is fatal. This means that, viewed separately, neither A nor B caused C's death; he would have died anyway. Yet it would be an economic mistake to let both off scot-free (why?). C's estate should be allowed to get damages from A and B--in what proportions we shall discuss later. Oddly to anyone who thinks that causal principles ought to play an independent role, un-related to economic considerations, in tort liability, the analysis is the same as in the case just put even if only one bullet hits C and we do not know whether A or B fired it, provided that both A and B were careless. There is growing judicial support for liability in this situation (P184) Sindell v. Abbott Laboratories, 26 Cal. 3d 588, 607 P. 2d 924 (1980), a DES case. (P184n)
    If a 10 percent increment in the probability of death is deemed causation, then each of the 110 will be able to recover his full damages, and the owner of the nuclear reactor will be forced to pay 11 times the actual damage it caused. But if 10 percent is deemed too little for causation, the owner will pay nothing, and there will be no tort sanction for its negligence. (P184)
    tort law might be well advised to consider defining the injury as the increased risk of death from cancer, rather than as the cancer itself. Then the whole population exposed to the burst of radiation would be able to sue when the accident occurs, using the class action device (discussed in Chapter 21) to economize on the cost of litigating many small claims. (P184-5)
    A train breaks down as a result of the railroad's negligence and a passenger is put up at a hotel and is injured when the hotel burns down.... To hold the railroad liable would thus be to impose (in effect) punitive damages for its negligence,just as in the superficially very different case with which we began this section. (P185)
    In Rickards v. Sun Oil Co., (23 NJ. Misc. 89, 41 A.2d 267 (1945)) the defendant's negligence put the only bridge between an island and the mainland out of commission. Merchants on the island who saw their business dry up as a result of the collision sued the defen-dant but lost, on the "theory" that pure economic loss is not recoverable in an accident case. The theory doesn't make much sense, but the result may. Although the island merchants did lose money, their loss was a gain to mainland merchants who picked up business from the island merchants when customers could no longer reach the island. Since the defendant could not seek from the mainland merchants restitution of the gains he had conferred upon them, it would have been punitive to make him pay the losses of the island merchants. The damage to the bridge was a net social loss, but one recoverable in a suit by the owner of the bridge. But the analysis is incomplete. It leaves out of account the customers. (P185)
    Here is another example. In White v. Jones, ([1995] All E.R. 691 (H.L.)) a man of 78, who had cut his daugh-ters out of his will, decided to reinstate them and instructed his solicitors to draw up a new will. The solicitors dilly-dallied without excuse and as a result when the man died two months later the new will had not yet been prepared. So the daughters got nothing, and they sued the solicitors--and won. (P186)
    The difference between this case and Rickards is that in this case the econom-ically apt solution--solicitors' liability coupled with restitution by the recipients of the windfall--may be feasible, as it was not in Rickards because of the large num-ber of those recipients. In White, the will could have been reformed4 so that the original legatees were made to transfer their legacies (or so much of those legacies as necessary to carry out the decedent's intentions) to the intended legatees. If the transfer caused any cost (other than the loss of the legacies) to the original legatees--they may have altered their position in reliance on having received the legacies--they would have a claim against the solicitors for that loss. (P186)
    By an application of, or analogy to, the doctrine of fraudulent conveyance. For example, if a charity innocently received a donation from someone who had obtained the money for the donation through securities fraud, the charity would have to return the money because it had not given consideration for it, any more than the original beneficiaries in White v. Jones gave consideration for their legacies. See Scholes v. Lehmann, 56 E3d 750 (7th Cir. 1995). It would be a case of what the law calls "constructive fraud." (P186n)
    One is that there is considerable doubt whether the defendant's negligence was a cause of the injury. A man is told that his pet canary has died through the negligence of a veterinarian, and upon hearing the news drops dead of a heart attack; such a consequence would, be deemed unforeseeable, but what this seems to mean is that either we can't believe the shock was enough to kill him or we think that if he was in such a vulnerable state he couldn't have had long to live. (P187)
    The other meaning of unforeseeability in the 1aw of torts is that high costs of information prevented a party from taking precautions against the particular accident that occurred; put differently, B in the Hand Formula was prohibitive once information about risk is recognized to be a cost of avoiding risk. A trespasser beds down for the night in a newly constructed but unoccupied house in a real estate development and is asphyxiated because the developer had accidentally spliced a gas main and a water pipe to the house. (P187)
    The analogy is imperfect, however, because the developer was merely negligent, and could not have communicated a risk of which he was unaware. But there is another reason for liability in the trespass case: that the trespass did not in fact increase the probability of injury. (P187)
    An apparent anomaly in the tort treatment of foreseeability is the eggshell skull principle.... A reason for nevertheless imposing liability in such cases is that, in order for total tort damages awarded to equal tort victims' total harm, there must J be liability in the eggshell skull case to balance nonliability in the "rock skull" case (i.e., where the victim has above-average resistance to damage). (P187)
    But why are these considerations not equally relevant to Hadley v. Baxendale? (P188)

6.8 joint torts, contribution, indemnity: herein of respondent superior and sexual harassment (188-190)

    We have already given some examples of joint torts, and here are two others. (1) Two landowners negligently set fires on their properties. The fires spread, join, and roar down on the plaintiff's house, which is consumed. Either fire would have done all the damage. Both landowners are liable. (2) An employee is negligent and injures someone. The employer is liable, too, under the doctrine of respondeat superior, which makes an employer liable for the torts of his employees, committed within the scope of their employment, even if he was not negligent in hiring, training, supervising, or failing to fire them. (P188)
    The rationale for the second is that most employees lack the resources to pay a judgment if they injure someone seriously. They therefore are not very responsive to the threat of tort liability. The employer, however, can induce them to be careful, as by firing or otherwise penalizing them for their carelessness (P188)
    Notice that his liability is strict; this may be because activity-level changes by the employer (e.g., substituting capital for labor inputs or reducing the scale of the enterprise) are potentially efficient methods of reducing torts by employees. (P188)
    There is an exception to the exception for .. the case where the independent contractor's work is highly dangerous. (P189)
    Sexual harassment in the workplace is now widely recognized as a tort; and one might suppose therefore that if employee A harasses employee B, their employer C will be liable to B. But under the prevailing view of this tort, C will be liable only if A was a supervisor of B (why should that matter?) or if C had reason to know that sexual harassment was a problem in its workplace and failed to do anything about it. Thus in the case of coworker as distinct from supervisor harassment, the employer is liable only if negligent; respondeat superior is not applied. An economic explanation is the virtual impossibility of an employer's preventing all sexual harassment by low-level employees, which implies that the imposition of strict liability via respondeat superior would have no beneficial allocative effects (P189)
    The examples at the beginning of this section--the two landowners' liability for fire and the employer's liability for his employees' torts--illustrate the two funda-mental types of joint tort (discussed earlier in the context of victim fault, for an accident in which both injurer and victim are at fault is functionally a joint tort). (P189)
    The two types are the joint-care joint tort and the alternative-care joint tort. At common law, the rule in joint-care joint-tort cases was "no contribution among tortfeasors," and the rule in alternative-care cases indemnity. Does this pattern make economic sense? (P189)
    No contribution among tortfeasors means that if the plaintiff gets a judgment against one of several tortfeasors, the defendant cannot compel the others to contribute any share of the judgment; and if the plaintiff gets a judgment against a group of tortfeasors he can collect it from them in any proportions he wants even if as a result one or more of them get off scot-free. (P189)
    Although the rule of no contribution is efficient, it should be apparent from the analogy to contributory-comparative negligence that a rule of contribution, which allows a joint tortfeasor made to pay more than his "fair" share of the plaintiff's damages to require contribution from the other joint tortfeasors, will also create the right safety incentives for all joint torffeasors--and this regardless of how the contribution shares are determined (pro rata, relative fault, etc.). But contribution is more costly to administer than no contribution because it requires the courts to decide another issue and supervise another set of transfer payments. (P189-90)
    Under the rule in alternative-care joint-tort cases--indemnity--the joint tortgeasor who would incur the higher cost of preventing the accident can get the other tortfeasor to reimburse him for the whole damages bill. Thus, if Firm X is held liable to the plaintiff for $100,000 in damages as a result of employee Y's negligence, X is entitled (although in practice will rarely be able to obtain) indemnification from Y for the whole $100,000. (P190)
    The economic explanation for the complete shifting of liability from one joint torffeasor to another that is brought about by indemnity is straight-forward. In an alternative-care case we do not want both torffeasors to take precau-tions; we want the lower-cost accident avoider to do so. The liability of the other is a backstop in case insolvency prevents the threat of tort liability from deterring the primary accident avoider. (P190)

6.9 rescue: liability versus restitution (190-192)

    Causation defines the pool of potential defendants: those who in some sense caused the plaintiff's injury. Since the universe of those who might have prevented the injury is not so circumscribed, there would be practical difficulties in limiting good Samaritan liability to those who really could have prevented the injury at reasonable cost. (P190)
    Another economic objection to good Samaritan liability is that it would make it more costly to be in a situation where one might be called upon to attempt a rescue, and the added cost would presumably reduce the number of potential rescuers-- the strong swimmer would avoid the crowded beach. (P191)
    It might seem that liability would impose a cost only on one who, but for the liability, would not attempt a rescue gthat altruists would be unaffected. But this is doubtful for two reasons. First, even the altruist wants to have a choice at the moment of crisis as to whether or not to attempt a rescue that may be dangerous to him; he doesn't want to be coerced by the law. Second, one of the benefits that a person receives from being an altruist is public recognition. (P191)
   A more serious objection to not imposing liability is that it assumes that potential rescuers are not also potential rescuees. (P191)
    If, however, there are not more rescuers, it is unclear what is gained by imposing liability. (P191)
    An alternative to liability and altruism as spurs to rescue is restitution, an approach followed, as we have seen, in the case of physicians and other professionals who render assistance in an emergency, and also widely utilized in admiralty law, where it is known as the law of salvage, which we have also met before. (P191)
    But a right of restitution creates a legal claim whenever a benefit is conferred; and the costs of processing the claims can be horrendous. Allowing the benefactor to externalize some of his costs is a cruder, but also a cheaper, method of encouraging the provision of external benefits. (P191)

6.10 the function of tort damages (192)

     There are two reasons why compensatory damages must be paid to the victim and not the state. The first is to give the victim an incentive to sue, which is essential to the maintenance of the tort system as an effective, credible deterrent to negligence. The second is to prevent victims from taking too many precautions. (P192)
    If the farmer will not be compensated for the railroad's negligence, the railroad, even if it would be punished by the state for damaging the farmer's crops, will go ahead and be negligent, knowing that the farmer will have an incentive to incur a $110 cost to avoid expected damages of $150 and that if he does so the accident will be prevented and the railroad will be off the hook. (P192) [an important point--we want only the lowest cost avoider of damages to adjust. This scenario involves strategic behavior though. What if the farmer calls the RRs bluff?]

6.11 damages for loss of earning capacity (192-196)

    A system of periodic disability payments, in contrast, would be the equivalent of a 100 percent tax on earned income. A potential offset, however, is that the incentive to exaggerate one's injury at the trial is less under the periodic-payments approach. (P192-3)
    Courts have had difficulty determining damages in cases involving disabled housewives.... Although it may have been greater, just as the value (discounted lifetime earnings) of an opera singer may exceed her value in an alternative occupation, the valuation of a housewife's services is difficult because of the absence of an explicit market in housewives. Courts do, however, allow testi-mony about the quality of the housewife's household services.  (P193)
    Foreseeable sources of wage change include probability of layoff based on past employment experience in the industry, rising labor productivity, and inflation. (P194)
    Although the probabilities of death, unemployment, etc. were taken into account in estimating the lost earnings stream, it is not double counting to use the existence of those probabilities to reject use of the riskless interest rate to discount future earnings to present value. (P196)
    Here is another complication: The theory of household production discussed in Chapter 5 implies that an individual's real earnings are not limited to the market income that he earns in 40 of the 168 hours in a week....  Wage rates for second jobs (moonlighting) might be used to estimate the opportunity costs of being disabled from productive use of hours not devoted to earning market income. But if this is done, then the portion of a person's moonlighting wages that represents compensation for the costs of working (including the forgone nonpecuniary income from leisure, and any haz-ards or disamenities involved in the job) must be subtracted to determine the worker's net loss from disability. (P196)

6.12 damages for pain and suffering, the problem of valuing human life, and the risk of overcompensation (196-200)

    Damages awards for pain and suffering, even when apparently generous, may v well undercompensate victims seriously crippled by accidents. Since the loss of vision or limbs reduces the amount of pleasure that can be purchased with a dollar, a very large amount of money will frequently be necessary to place the victim in the same position of relative satisfaction that he occupied before the accident. The problem is most acute in a death case.  (P196)
    The courts have resolved the vexing problem of the proper valuation of life largely by ignoring it. Damages in a death case are generally limited to compensat-ing the pecuniary loss to survivors (in some states, to the deceased victim's estate), plus medical expenses and any pain and suffering experienced by the victim before death. The pecuniary loss to survivors is the victim's lost earnings minus his living expenses....The implicit assumption is that the person who has been killed obtained no utility from living! (P196-7)
    Some foreign courts have implicitly recognized this in awarding damages to the employer for an injury to an employee. Common law courts used to award such damages but no longer do--erroneously regarding such awards as implying that the employer "owns" the employee. (P197)
    [valuing children:] Wycko v. Gnodtke, 361 Mich. 331,339, 105 N.W. 2d 118, 122 (1960); Breckon v. Franklin Fuel Co., 383 Mich. 251,268, 174 N.W. 2d 836, 842 (1970). (P197n)
    These studies could be used to estimate the costs of my dangerous driving, for which I could be made liable whether or not my car actually hit anyone. (P198)
    Notice that the sum of the ex ante damages thus computed will not equal the common law damages of someone who is actually injured. The fact that my conduct may have subjected each of 100 people to a 1 percent risk of loss of a life that, in a tort case utilizing conventional methods of damage assessment, would have been valued at $500,000 does not imply that each of the 100 would have demanded only $5,000 from me to undergo the risk. Risk aversion aside, since most people obtain nonpecuniary as well as pecuniary income from life they will demand a higher price to assume a risk of death than the purely pecuniary loss from dying, which is generally all the common law system tries to compensate for. (P198)
    Suppose we know that the average person demands $100 in order to assume a .0001 risk of death. Can we infer that he values his life at $1 million? We can -- at least for the purpose of calculating tort damages for low-probability injuries (i.e., accidents) at the correct level, which is our purpose here. (P198)
    But this approach will not work when the probability of death is high. The fact that someone demands only $100 to incur a .0001 risk of death does not imply that he will demand only $100,000 to incur a 10 percent risk of death--or $1 million to incur a certainty of death. (P198)
    A rational individual wants the most bang for his buck and therefore would want to reallocate income from the post-accident period, when his marginal utility of income is low, to the pre-accident period, when it is high. This could be done by the taxation of damages awards in cases of death or severe permanent disability, since the proceeds of the tax would go to increase (perhaps by enabling reductions in other taxes) the disposable income of the public at large, most of whom of course are not (yet) dead or disabled.... Why would it not impair the incentives of potential accident victims to take care--and might in fact strengthen them? (P199)
    But this discussion assumes, perhaps unjustifiably, that the proper baseline for determining marginal utility of income is the pre-accident self. If the perspective of the future contingent disabled self is used instead, it becomes apparent that unless that future self is destined to be irreversibly comatose, it may well ascribe a higher marginal utility of income in the disabled state than would the pre-accident self, and thus attach a lower value to expenditures on consumption by the pre-accident self. (P199-200)

6.13 the collateral benefits (collateral source) rule (200-201)

    If an accident insurance policy entitles me to receive $10,000 for a certain kind of accidental injury and I sustain that injury in an accident in which the injurer is negligent, I can both claim the $10,000 from the insurance company and obtain full damages (which, let us assume, are $10,000) from the injurer, provided I did not agree to assign my tort rights to the insurer (subrogation). To permit the defendant to set up my insurance policy as a bar to the action would result in under-deterrence. (P200)
    Less obviously, the double recovery is not a windfall to me. I bought the insurance policy at a price presumably equal to the expected cost of my injury plus the cost of writing the policy. The company could if it wished have excepted from coverage accidents in which the injurer was liable to me for the cost of the injury, or it could have required me to assign to it any legal rights that I might have arising from an accident. In either case my premium would have been lower. (P200)
    Some courts have had trouble when the collateral benefit was not rendered pursuant to a contract but was "gratuitous." ... If an employer gives his injured employees medical treatment free of charge, this means only that the employer pays for their labor partly in money and partly in kind, so that the money wage would be higher if the "gratuitous" benefits were lower. (P200)
    To the extent that unemployment insurance is financed by the government, there is an argument for deducting the benefits from the em-ployee's award of damages and permitting the government to sue to recover them. (P200)

6.14 negligence with liability insurance and accident insurance; no-fault automobile compensation (201-204)

    Studies show that administrative costs, mainly legal expenses, are a high fraction of the total amounts paid to victims in settlements and judgments and that many people injured in automobile accidents receive little or no compensation msometimes because the victim himself was neg-ligent, sometimes because the defendant was uninsured and insolvent or was a hit-and-run driver and unknown. (P201)
    If compensation is the only purpose of the negligence system, it is indeed a poor system, being both costly and incomplete. ... If the system yields sub-stantial savings in accident costs, its heavy administrative costs, which relate pri-marily to the determination of liability--the determination whether the accident was uneconomical--may be justified. As for coverage, the deficiencies of the system could be remedied by wider purchase of accident insurance. (P201)
    The deterrent impact of automobile damage awards is impaired by liability in-surance, although the implications of this point for legal policy are less clear than one might have thought.... because courts make mistakes and because negligence contains as we saw a strict liability component, there is a risk of being adjudged negligent and hence a de-mand for insurance against liability for negligence. (P201)
    With insurance, the cost of an accident to the negligent injurer is no longer the victim's loss; it is the present value of any premium increase that the injurer may experience as a result of being found negligent. (P201)
    If the liability insurance market were not regulated, insurance companies might charge different premiums to their customers keyed more closely to differences in the probability that a customer would, through his negligence, injure someone in an accident. (P202)
    A surprising feature of these laws from an economic standpoint is that they are not concerned with creating better incentives for accident avoidance, but in-stead seek to increase the coverage of the system and to reduce the cost of insur-ance. These goals are inconsistent with each other as well as with the goal of reducing the number of accidents. (P203)
    The Keeton-O'Connell plan, the model for these statutes, illustrates the di-lemma. Under the plan, every motorist is required to carry basic protection that entitles him in the event of an accident to recover his medical expenses plus lost earnings, regardless of the injurer's negligence or his own freedom from negli-gence. Pain and suffering are not compensated, and any collateral benefits are deducted. The victim may waive basic protection and sue in tort in the usual way if he sustains more than $10,000 in damages other than for pain and suffering. Basic protection is first party (accident) rather than third party (liability) insurance. (P203)
    Because the plan compensates victims of faultless drivers and victims themselves at fault, its coverage is broader than that of the tort system. Hence if the average claim were no smaller under the plan than under the existing tort system, the total amount paid out in claims, and therefore insurance premium costs, would probably be greater (even assuming lower administrative costs) than under the present system. (P203)
    Proponents of no fault argue that deterrence is the province of the criminal law. Since it is unlawful to insure against criminal penalties, the effect of liability insurance in sapping the deterrent efficacy of negligence liability is eliminated. (P204)
    One study has found that states whose no-fault statutes place severe restrictions on tort liability can expect 10-15 percent more automobile accident deaths. This result may seem weird, since no-fault statutes leave tort liability intact in death cases. Bear in mind, however, the probabilistic character of care (P204)

6.15 intentional torts (204-208)

    Most accidental injuries are intentional in the sense that the injurer knew that he could have reduced the probability of the accident by taking additional precautions.... Conversely, in many intentional torts the element of intention is severely attenu-ated, as when a surgeon who unwittingly exceeds the limits of the patient's express or implied consent to a surgical procedure is held to have committed a battery. In the usual medical battery case, the issue is whether there was a sufficient emergency to justify a procedure to which the patient's consent had not been obtained in advance. This in turn depends on whether the costs of delay (such as the risk that the patient's condition might deteriorate, and the added danger of again subjecting him to a general anesthetic) exceeded the value to the patient of an opportunity to consider whether to undergo the procedure; if so, implied consent to the procedure will be found (P204)
    The defendant in Bird v. Holbrook (4 Bing. 628, 130 Eng. Rep. 911 (C.P. 1828) owned a valuable tulip garden about a mile from his home. (P204-5)
    since spring guns do not discriminate between the thief and the innocent trespasser, they deter owners of domestic animals from pursuing their animals onto other people's property and so increase the costs (enclosure costs or straying losses) of keeping animals. The court in the Bird case implied an ingenious accommodation: One who sets a spring gun must post notices that he has done so.... The analysis thus turns out to be the same as in a negligence case--the archetypal unintentional tort case. (P205)
    The set of intentional torts that are economically distinct from unintentional torts consists of such torts as trespass, assault, simple battery (for example, mugging, as distinct from medical battery or the technical battery that is committed in an illegal prizefight), fraud, and conversion (the tort counterpart of theft)--torts that resemble such common law crimes as rape, murder, robbery, burglary, and larceny. (P205)
    Consider a good worth $100 to both the owner and the thief, and suppose that if the owner spends nothing on protection the thief could steal the good by expending $20 in time and burglar tools (P205)
    I want a car, and I decide to save time by stealing your car. B is not only lower than in an accident case; it is actually a negative number, because rather than saving resources by injuring the victim (implying a positive B) I would save resources by not injuring the victim (implying a negative B), since it must cost me something to steal the car....  P, furthermore, is very high--much higher than in an accident case--because wanting to do someone an injury makes it much more likely that an injury will occur than if the injury if it does occur will simply be an undesired by-product of another activity, such as carrying freight from one point to another. (P206)
    This has two important implications for legal policy. 1. We would expect, and find, the law to be much more willing to award punitive damages in "real" intentional tort cases.... We know:that in a strict liability case punitive damages would lead to overdeterrence. Less obviously, the same thing is true in a simple negligence case. Because of judicial mistake and the strict liability component in negligence, negligence cannot be completely avoided by spending B on care.... But since the gap between B and PL is so much larger in the "real" intentional tort case, the danger of deterring socially valuable conduct by making the damages award greater than L is minimized and other policies come to the fore, such as making sure that the damages award is an effective deterrent by resolving all doubts as to the plaintiff's actual damages in his favor; this can be done by adding a dollop of punitive damages to the estimate of his actual damages. (P206)
    Moreover, because we want to channel resource allocation through the market as much as possible, we want to make sure that I am not allowed to be indifferent between stealing and buying my neighbor's car. (P206) [link this to F with his ex of making burglary a tort]... Another way, also common in intentional tort cases, is to make the tortfeasor pay the victim what the thing taken was worth to the tortfeasor. (P206)
    2. There is no reason to allow a defense of contributory negligence in what we are calling a "real" intentional tort case (pure coercive transfer), since the cost of avoidance is plainly lower to the injurer than to the victimmis indeed negative to the injurer and positive to the victim. The victim cannot be the lowermost avoider. Stated otherwise, the victim's optimal care is always zero. (P207)
    Thus far in our discussion of intentional torts the emphasis has been on acquis-itive torts--conversion, robbing (conversion plus assault, in tort terms), killing for money, etc. (P207)
    The spitting example suggests a further economic reason for awarding punitive damages in some tort cases: to relieve pressure on the criminal justice system di-rectly and indirectly by providing a substitute for violent self-help (itself criminal).... Even for major crimes that are also torts (wrongful death, for example, the tort counter-part of homicide), punitive damages assist the criminal justice system by increasing the expected punishment of wealthy criminals, whose wealth gives them a consid-erable advantage in the criminal process.  (P207)
    There is a movement to cap punitive damages at a fixed multiple (say 3) of compensatory damages. Would it make more sense to make the amount of punitive damages awarded inverse to the amount of compensatory damages? See Kemezy v. Peters, 79 E3d 33 (7th Cir. 1996). (P207)
    "deliberate indifference" is a necessary rather than, as the judicial formula might be taken to imply, a suffi-cient condition of liability as an intentional tortfeasor. The higher that PL is (es-pecially P), the likelier is it that the potential injurer actually knows that his conduct is dangerous; so knowledge becomes a proxy for a high PL, and we know that, other things being equal, the higher it is the greater the defendant's fault.... Furthermore, the information-costs component of B is smaller if the defendant was actually aware of the risk and decided to do nothing about it. So deliberate indifference increases the likelihood of a big spread between PL. and B. But it is essential, if the defendant is to be pronounced an intentional tortfeasor (more precisely, a reckless torffeasor where recklessness blends into deliberateness), not only that PL be very high but also that B be very low. (P208)

6.16 defamation (208-210)

    The tort of defamation mixes intent, negligence, and strict liability. Usually classi-fied as an intentional tort, because Writing or speaking critically about a person is a deliberate act, it has a strong flavor of strict liability, as in the rule that it is no defense that the defendant may have made a reasonable effort not to defame the plaintiff. In Jones v. E. Hulton & Co., [[1909] 2 K.B. 444, aff'd, [1910] A.C. 20.] the author of a fictitious newspaper story accidentally gave a character in the story the name of a real person, Artemus Jones, who sued for libel, and won by showing that his neighbors thought the article was about him. (P208)
    The law of defamation has two puzzling gaps that economics may be able to explain. The first is that group libels (e.g., "All doctors are quacks") are not ac-tionable. The libel is not likely to hurt members of the group because substitution away from an entire trade is much more costly than substitution away from an individual. ... The second, and oddly related, rule is that you cannot defame the dead.... But if the libel were that the deceased had had an inheritable disease, the adverse consequences would not terminate with death, and the law recognizes an exception for such cases. (P209)
    The law treats written defamations (libel) more harshly than oral (slander). (P209)
    There is no external benefit conferred in such a case (why not?), so neither is there any reason to permit a cost to be externalized. But truth is an absolute defense because there is an external benefit even if the defendant thought he was lying. (P210)
    With the original publisher and the republishers all liable for any defamation in the original article, there is no longer an incentive for a publisher to hang back; as a result, news gets out faster. (P210)

6.17 successor liability (210-211)

    The growth of scientific knowledge has increased the awareness of long-delayed adverse consequences of activities such as the sale of asbestos or DES. The longer the delay between an activity and its consequences, the greater the likelihood that the actor will no longer be around when suits for damages are brought. (P210)
    The tradeoff is a difficult one, but a couple of relevant factors can be identified. One is whether the corporation's tort liability was foreseen at the time of dissolution. If it was, successor responsibility will not be as costly as it would be if tort liability had not been foreseen, because it should be possible to estimate the cost of that liability and adjust the purchase price accordingly. (P210-211)
    Successor responsibility will also be less costly if the corporation is sold in a lump rather than its assets' being dispersed among many purchasers, for then the number of successorship lawsuits will be reduced from many to one. ... With delayed consequences more common today, the rule is eroding, as eco-nomic analysis predicts it would. (P211)
    Consistent with this analysis, it has been found that risky industrial undertakings are disproportionately concentrated in small firms, because their potential tort liability is truncated as in the example. (P211)