The Cost of Accidents Two principle goals of accident law: Justice Two separate notions of justice: the communitys

and the critics. The critics goal is susceptible purely to reason, where the communitys understanding may be tied to the past or the status quo in a way thats difficult to shake. This goal stays in the background until the end of the book. Reduction of Accident Costs Apart from the requirements of justice, I take it as axiomatic that the principal function of accident law is to reduce the sum of the costs of accidents and the costs of avoiding accidents. This has three sub goals: 1) Primary Accident Cost Avoidance A reduction in the number and severity of accidents. There are two means of accomplishing this:
a) We can forbid specific acts or activities thought to cause

accidents (Specific Deterrence)

b) We can make these activities more expensive and thereby

less attractive to the extent of the accident costs they cause (General Deterrence)

2) Secondary Cost Reduction Concentrates on reducing the societal costs resulting from accidents OR their avoidance. Do the expense of accidents bankrupt victims or tortfeasors or third parties? The most obvious question here is how we provide for victims after accidents, but secondary cost reduction includes other things. For example, if we decide to ban motorcycles this may cause dislocation and unemployment. Secondary Cost Reduction calls for spreading: a) Spreading Instead of leaving the losses on the victim or the injurer we spread the losses around. This includes social insurance, private insurance, and enterprise liability. Guido gives reasons why private insurance may give inadequate loss spreading. He also points out that spreading may eliminate

market (general deterrence) pressures to make activities safer (achieve primary accident cost reduction).

3) Lowering Administrative Costs Just what it says.

After describing the goals, Guido goes into more depth on the means of achieving primary accident cost reduction (preventing accidents). General Deterrence This market-based approach puts accident costs on the activities that cause the accidents. This may mean something like the current fault system or a special charge to actors engaging in dangerous activities whether those individuals actually have accidents or not (e.g. insurance or some kind of liability-related tax such as workers compensation funds). This reduces accident costs in two ways:
1) It encourages people to engage in safer activities (i.e. they are

cheaper than dangerous ones)

2) It encourages us to make activities safer (i.e. if you have to pay your

accident costs you will install any safety equipment that is cheaper than the accidents that it would prevent). General deterrence cannot be our sole approach to accident law, however, because in a world of perfect general deterrence any activity that could pay for the accidents it caused would be considered worthwhile. (Imagine Donald Trump throwing lighted sticks of dynamite off of his penthouse balcony.) Specific Deterrence (aka Collective Deterrence) Suggests that decisions regarding accident costs should be made collectively through the political process. In practice, this takes two forms: First we can bar certain activities altogether, or, alternately, we can penalize undesirable activities and subsidize desirable ones. Guido outlines three limits on specific deterrence: 1) it is impossible to make political judgments with regard to every decision involving choices for or against accidents, 2) the fact that individuals cannot control each individual act (e.g. we can ban drunken

driving but we cant really ban driving absentmindedly, even if that causes the same number of accidents, because people dont always know when they are driving absentmindedly and such a prohibition would be unenforceable), (3) limits imposed by our other goals. Neither General nor Specific Deterrence alone will allow us to meet all of our accident reduction goals, so in discussing the ideal means of accident cost reduction we are talking about a mixed system for example, one that prohibits certain activities outright while leaving the market to control others. Guido describes four systems that combine parts of General and Specific Deterrence. The most relevant of these in terms of the course is Penalties on an Involvement Basis: Activities Defined after an Accident (p 123-8). This section discusses what a happens when we prohibit a broad set of conduct with a vague prohibition (like careless conduct) that is given concrete application only After the accident. This is of course what we do in torts, and it is noteworthy that Guido points out four theoretical problems with this method of deterrence: 1) It must be assumed that the act or activity can be controlled by the individual at the moment it becomes apparent to him that it is the kind of conduct the collective deciders have said they would penalize. This is a big theoretical stumbling block because people frequently dont realize they are behaving negligently until after the accident so society has prohibited conduct that individuals cannot control completely.
2) Because of this vagueness of the prohibition, our sense of justice

prohibits certain penalties. We wont guillotine people if their absentminded driving causes an accident.
3) The vagueness of the prohibition especially if combined with a

serious penalty may make people abstain from broad activities that might put them in a position of having to make this difficult guess. (e.g. Afraid of the penalty for driving absentmindedly, I just stop driving altogether.)
4) Following from the third problem - its difficult for society to prohibit

only a narrow range of conduct and not affect related activities which may be desirable. The conclusion of all of this is that penalty imposed on the basis of vague prohibitions and applied ex poste after an accident what Guido calls an

uninsurable tort fine must be very carefully tailored to the situation. It should be serious enough to limit the undesirable activity but not so big that it affects related but desirable areas of conduct.

After this discussion of general and specific deterrence Guido moves on to what he calls the Two Major Problems in Reducing Accident Costs: Any system of accident law that seeks to achieve primary cost reduction [fewer accidents] must provide a basis for making two very complicated decisions: (1) the extent to which each of the activities combining to cause an accident is responsible for the accident costs... [Guido calls this what-isa-cost-of-what] and (2)the proper valuation of the cost of accidents [what-is-the-cost]. What-is-a-cost-of-what What causes the accident cost and the Market Approach (General Deterrence) He explains that a general deterrence (purely market-based) approach to primary accident cost avoidance (fewer accidents) would require allocation of accident costs to those acts or activities (or combinations of them) which could avoid the accident costs most cheaply. This is the same as saying that the system would allocate the costs to those acts or activities that an arbitrary initial bearer of accident costs would (in the absence of transaction and information costs) find it most worthwhile to bribe in order to obtain that modification of behavior which would lessen accident costs most. (We discussed this in class, but see 136 if you want an example of how it works.) He then outlines some guidelines by which we are to determine what party in a given situation can avoid accidents most cheaply: 1) an initial rough guess, 2) by remembering the relationship between Avoidance and Administrative Costs, and 3) by avoiding externalization, 4) by choosing the best briber (e.g. a corporation may be able to bribe everyone who uses its products, but it would be much more difficult for the product users to unite and bribe the corporation), 5) weighing all of these factors to choose the person we think is the cheapest cost avoider, and 6) using controlled experiments or statistics to test our guess of who is the cheapest cost avoider. Even under a market model, we as a society must chose a forum and method which can best make the decision of who is the cheapest cost

avoider following these guidelines. Guido says that if our goal is the marketbased goal of finding the cheapest cost avoider, case-by-case decisions are undesirable and thus a body like the jury is not a good forum. He elaborates on this in section V. Here Guido also discusses Vincent v. Lake Erie Transp. Co. (p 162) and the assumption that it wouldnt matter which side - Dock Owners or Ship Owners you put the cost on because, if the cost was more cheaply allocated to the other side, that side would just bargain with the other side to change its behavior. (If you put the cost on the ship owners but it was actually the dock owners who were in a better position to insure or prevent accidents, then the ship owners would just bargain with the dock owners to do whatever is necessary to minimize the cost of accidents and their prevention.) Guido says that this assumed symmetrical behavior looks good in theory but often isnt possible in practice. The market can often correct errors and find the cheapest cost avoider efficiently. It does not mean that it can always do so. In practice, the situations where it is most likely to be able to do so are those in which the bargainers are of approximately equal size, number, expertise, and wealth. And in these situations, conscious allocation of losses among bargainers may be quite unnecessary and wasteful. What causes the accident cost and the Specific Deterrence Approach While the market approach looks for the cheapest avoider of accident costs, Specific Deterrence seeks to find the best avoider. This best takes into account the desirability or undesirability of the various possible causes of accidents. I am not going to summarize his treatment of this because it doesnt look as relevant for the exam as the market-related questions. (Or at least I didnt underline anything.) What-Is-the-Cost? This question runs parallel to the discussion of damages in torts, and most of the points in the book were discussed in class. He notes that there are essentially two answers to the question of what is the cost of an accident under a market approach. You can either make case by case determinations (as we do with a jury) or you can establish a schedule of values. The first is probably accurate but expensive; the second, inaccurate but cheap.

In this section he also discusses Pain and Suffering and Sentimental Damages (p 215) and Administrative Expenses read Lawyers Fees (p 225). Again, almost all of his points were made in class.

In Part IV, Guido moves on to treat the Fault System. This is really the meat as far as the exam is concerned. Even if its the only time you open the book, you should read pages 239-243. He says that in terms of the preceding discussion the three most important things to realize about the fault system are (1) that it is essentially a mixed system that is, as system that is not committed to just one goal or subgoal but attempts instead to achieve all the goals to some degree through mixed methods; (2) that it begins by viewing each accident as being the exclusive concern of the parties immediately involved and tries to allocate costs according; and (3) that it allows insurance of both victim and injurer. The fault system makes ex poste decisions about liability (as discussed above) and does this in a way that includes consideration of moral questions. Thus the fault system is not strictly market-based. Guido admits that the fault system is buttressed by a gamut of specific deterrence rules and regulations, such as safety legislation with criminal penalties but he excludes these from his discussion of the system. Leaving aside the specific rules and regulations, the question is whether the particular characteristic of the fault system, i.e. the ex post allocation of responsibility on a case-by-case basis despite the knowledge that the costs will be spread to actuarial groups, gives us enough in specific and general deterrence to justify the secondary and tertiary costs of this particular mixture. He will conclude that fault is too expensive, cannot be made to spread costs adequately, and has serious failings in terms of the prevention of future accidents. If you have time I would recommend skimming section IV. On the other hand, sections V and VI are probably more worthwhile, if you have limited time. Section four is about criticizing the fault system from market (general deterrence) and collective (specific deterrence) perspectives. I will pull out the main criticisms and put page numbers so that you can look them up as you see fit. I will put an asterisk next to those that seemed novel to me (either because we didnt discuss them in class, or because I was reading the newspaper at the time).

Criticism of the Fault System from a Market Perspective: It pays little attention to which of the possible categories of cost bearers is most likely to be aware of the risk involved (244). It never asks who really pays. E.g. if you find the poor man liable for his injury, it is really social services that picks up the tab. (This is externalization due to transfer.) (246) *It does not ask which types of actors are susceptible to better subcategorization for insurance purposes (246-250) The system does a terrible job of achieving an optimal balance between avoidance costs and administrative costs. This is partly because it fails to ask which party is the cheapest spreader or insurance buyer. (251) The forum and method of the fault system The Jury and the limitations the fault system imposes on itself make the selection of the cheapest cost avoider far less likely than need be. (255) The self-imposed limitations are significant. They are: (1) little or no quantitative division of damages, (2) little or no qualitative division of damages, and (3) allocation of damages to the victim where no fault is found. (259) *Guidos discussion of the third limitation (and of Justice Holmes) starts on 261 and is worth looking at. The Fault System from a Moral Perspective (Specific Deterrence) If the fault system exists to impose societys moral will with regard to faulty or wrongful behavior, this purpose is undercut by permitting people to insure against the costs of accidents. (269) The fault system is a much more expensive enforcer of societys opinions about behavior than simple non-insurable penalties. (272) Faults failures as a mixed moral-market system The conclusion is that the fault system can only be justified if what we wish to minimize is neither the sum of the costs of accidents and their avoidance (the market goal), nor faulty or wrongful behavior (the collective goal), nor both of these (the mixed goal), but rather only those accidents in which faulty behavior is a but for cause. The more on examines this proposition the more absurd it becomes. (277)

Possible modifications to help the fault system spread accident costs better Guido treats and rejects the following ideas as means of solving the faults systems high secondary costs (recall that secondary cost concerns the societal cost resulting from accidents, like the victims medical bills, bankrupted businesses, etc.): Comparative Negligence - Its compatible with the fault system but doesnt necessarily help spreading (279) Compulsory Insurance It might achieve spreading but it would eliminate the fault systems successes in general deterrence (ie. It would leave no economic incentive to behave more safely) (281) Considering Spreading Ability When Determining Fault While it happens occasionally under the current fault system, as a broad solution it would be erratic and expensive (283) Combine Social Insurance with Fault Same problem as Compulsory Insurance (284)

Section V Justice and the Fault System Again if you had infinite time, you would probably want to read this section at least from the bottom of 299 on. It isnt imperative, however. The main ideas are below: Justice has two aspects: the communitys sense of it, which may include a thoughtless adherence to tradition, and the critics, which is based purely on practicalities. (291-2) There is some evidence that the moral imperatives that the public attaches to systems of accident law are directly related to cost avoidance and hence subject to change as what brings about cost avoidance changes. (296) One of the confusions in current discussions of accident law is the belief that there must be individualized relationships between injures and injured (297).

The moment one accepts the notion that justice does not require that an individual injurer compensate his individual victim and the allowance of insurance for faulty parties is a clear indication that this notion is accepted and the moment on realizes that wrongdoers can be punished for wrongful acts quite apart from whether they must compensate victims, it becomes very hard to see how the fault system can be supported on grounds of justice. (302-3) In the end, justice will support the fault system only if there is no sensible alternative system presented, only if the choice is solely between crushing on relatively wrongful and one relatively innocent party. It will not support the fault system in a world where faulty or undesirable acts, activities, and actors (whether victims or injurers) can be penalized according to their undesirability, and injured parties can be compensated according to their injury. (307) The fault system may have arisen in a world where one injurer and one victim were the most that society could handle adequately, and in such a world it probably was a fairly good mixed system. It did a good job of meeting our combination of goals: general and specific deterrence, spreading, justice, and even efficiency. But even assuming that such was the world in which the fault system grew, it is not todays world. Today accidents must be viewed not as incidental events linking one victim with one injurer, but as a more general social problem. That is why the fault system has become totally inadequate for any of our mixed goals, even justice. It has become so inadequate, in fact, that other mixed systems can improve our record as far as each of these goals is concerned, even though at their extremes some of the goals are inconsistent with one another. (307). Toward a New System of Accident Law Again this might be worth reading, though it doesnt say a great deal. I will give you two good quotes: We can do better, that is, with regard to every goal of accident law except the goal of retaining the status quo. In such a situation, change is virtually inevitable, whether it is made consciously or not. (316) A look at most of the systems proposed for the reform of accident law suggests that the real issue of the coming decades will not be the show battle over the fault system, but rather the quiet war between

those who, by pushing various systems of so called enterprise liability or first or third-party nonfault insurance, seek to give primacy to primary cost avoidance [fewer accidents] and some market control over accidents and those who, by urging great and greater social insurance paid from general taxes, seek to give primacy to secondary cost avoidance [social costs], with perhaps some collective primary cost control attached. (317)