We wish to know what the laws of a society—statist or anarchist—ought to be. The obvious way to find out is to start with general principles of justice and see what laws are necessary to implement them. In an earlier chapter, I argued that that cannot be done; libertarian principles of justice cannot, at least as they now exist, answer enough of the relevant questions. They provide no way of deciding what ought to be included in property rights, how they may legitimately be defended or how violations ought to be punished.
When I say that libertarian principles cannot answer the questions, I do not merely mean that answering them is hard. That would be true wherever we started; these are hard questions. I mean that I cannot see how to even start answering these questions—what facts I need, what calculations I should do. It is as if I were faced with an engineering problem and had no way of finding out how to start setting up the relevant equations.
Perhaps someone else does know how to do it—but someone else is not writing this book. My solution is to find a different starting point from which to solve the problem. That starting point is utilitarianism. As a moral philosopher I am a libertarian, insofar as I am anything. As an economist I am a utilitarian.
One could describe most of this book as a utilitarian approach to libertarianism, but only by using “utilitarian” in a very general sense. My approach would be more accurately described as consequentialist. I have tried to show that libertarian institutions produce attractive results but I have not defined “attractive” as anything so specific as tending to maximize the sum total of human happiness. In this chapter, however, I am trying to answer much more specific questions, not merely “should we have property rights?” but “exactly what sort of property rights should we have?” To do so I require a more precise definition of the objective I am trying to achieve. When I am finished, your conclusion, if you agree with everything I say, should not be “we should have property rights X, Y, and Z” but rather “If we wanted to maximize total utility we would want property rights X, Y, and Z.”
Even if I can demonstrate that, why should I bother? By adopting a philosophical position that I believe is false merely because it makes it easier to answer a particular set of questions, am I not making the same error as the drunk who, having lost his wallet in the middle of the block, looked for it under the streetlamp at the corner because the light was better there?
I think not. Even if utilitarianism is not true it may still be useful. There seems to be a close correlation between rules that make people free and rules that make them happy; that is why it was the East Germans and not the West Germans who erected barbed wire fences and guard towers on their common border. Perhaps that correlation comes from some deep connection between freedom and happiness, perhaps it is merely an accident. In any case, it is there. I conclude that by figuring out what legal rules would best make people happy I may learn something about what legal rules are suitable for a free society.
A second reason utilitarian arguments may be useful is that even if they cannot tell us what the legal rules should be they may, under some circumstances, tell us what they will be. In Chapter 31 I tried to show that the institutions of anarcho-capitalism tend to produce economically efficient law. By figuring out what legal rules would be economically efficient we can learn something about what rules would be generated in such a society. Richard Posner, one of the leading writers on the economic analysis of law, has made the same claim for the existing body of common law. If he is right, then economic efficiency is useful for understanding what the law is as well as what it ought to be. Economic efficiency and total happiness are, as you will shortly see, closely related; the former is best understood as an approximate measure of the latter.
A third reason was suggested at the end of the previous chapter. Most people, myself included, are at least partly utilitarians. While a demonstration that a particular legal rule tends to increase the total of human happiness does not prove that the rule is a good one, it is a strong argument for it. Since I have no very good way to settle disagreements about values, it makes sense to base my argument on values that most people share.
The final reason is that, whether or not people care about the sum total of human happiness, most of us care a good deal about our own happiness. If a particular legal rule increases the average level of happiness there is at least a presumption that it will, on average and in the long run, make me better off. That is a reason, although not necessarily a compelling reason, why I should favor it.
For all of these reasons it makes sense to ask what legal rules tend to maximize human happiness. The rest of this chapter is devoted to trying to answer that question. My tool for doing so is the economic analysis of law. The first steps are to explain what economic efficiency means, how it can be used to choose legal rules and why it may be a useful measure of total happiness.
Consider some change that affects only two people. For each, one may ask how much the change is worth to him, how many dollars he would if necessary pay in order to get it (positive value) or prevent it (negative value). One could then sum the answers to get a dollar value for the effect of the change. If one person was willing to pay four dollars to get the change and the other two dollars to prevent it, the change could be described as increasing total value by two dollars. A similar calculation could be made with any number of people, summing the positive values of those who favor the change and the negative values of those opposed to it. If the net is positive we describe the change as an economic improvement or an increase in efficiency, if negative as an economic worsening or decrease in efficiency.
Although we are measuring values in dollars, no money need actually be involved. The change might be the transfer of an apple from you to me. The apple is worth two dollars to you and four to me. You would pay up to two dollars to keep the apple, so the change has a value to you of minus two dollars. I would pay up to four to get the apple, so the change has a value to me of plus four dollars. The change produces an economic gain of two dollars.
How would we find out whether a particular change produced a net gain or a net loss? The best way would be to observe people’s values as reflected in their actions. Suppose I offer you three dollars for the apple and you accept. The fact that I make the offer implies that the apple is worth more than three dollars to me. The fact that you accept implies it is worth less than three dollars to you. Assuming that we are the only people affected, the transfer must result in a net gain. Generalizing the argument, we conclude that any voluntary transaction that has no effect on third parties must result in an economic improvement.
Voluntary transactions are improvements, but improvements are not necessarily voluntary transactions. Suppose I am lost in the woods and starving. I stumble upon your locked cabin, break in, and use the telephone to summon help. Being both grateful and responsible, I leave you an envelope containing enough money to pay for the damage several times over. The exchange is not voluntary; you did not give me permission to break into your cabin. But, just as with a voluntary transaction, we have both ended up better off (assuming my calculation of how much to leave was correct), so there was a net improvement.
In both cases—selling the apple and breaking into the cabin—the cash payment provided evidence that the change was a net gain, but the gain was produced by the change not the payment. The same two-dollar gain would have occurred if you had accidentally lost the apple and I had found it, although in that case it would have been the sum of a four dollar gain and a two-dollar loss instead of the sum of two one-dollar gains (you lose an apple valued at two dollars, get three dollars; I gain an apple valued at four dollars, pay three dollars).
So far I have been talking about changes, not about rules. The next step is to ask what legal rule will result in only efficient changes, changes that produce a net economic benefit. In the case of the apple, we want a rule that will result in the apple being transferred to me if and only if it is worth more to me than to you, since only then is the transfer an economic improvement. The obvious solution is to allow the transfer if and only if both of us agree to it. If the apple is worth more to me than to you I will make you an offer for it that you will accept; if it is not I will not. In this case, the solution is simply property rights, enforced by a punishment for anyone who steals an apple.
What about the case of the cabin? Property rights will not solve that problem, since the owner of the cabin is not available to rent out the use of his phone. This time the solution is a damage rule. If I break into the cabin and turn myself in for doing so, I owe the owner a payment equal to the amount of damage I have done to his property. If the use of his phone is not worth that price, I will keep wandering; if it is, I will break in. That is, in each case, the economically efficient outcome.
I have now gotten far enough so that you can see how, in principle, economic analysis can be used to figure out what laws ought to be. Before I go on to discuss these two examples in more detail and to apply the analysis to some of the problems mentioned in Chapter 41, I should first fill in a missing step in the argument. I have talked about maximizing total happiness and about economic improvement, but have not shown that the two have anything to do with each other. I have not shown when or why the fact that some change is an economic improvement implies that it increases total utility.
There are two important differences between the economist’s criterion and the philosopher’s. The first involves the measurement of utility for an individual, the second the comparison of the utility of different people.
In defining value, the economist accepts the individual’s own evaluation of whether something does or does not make him better off. If I prefer gaining an apple and losing four dollars to doing neither, that shows that the apple is worth at least four dollars to me. That definition of value is what economists refer to as the principle of revealed preference. The possibility that I am wrong in judging my own interest, that I am willing to pay for apples even though they are bad for me, is assumed away.
One implication of that assumption is that the value of heroin to a heroin addict is just as real as the value of insulin to a diabetic. If you are unwilling to accept such implications you will conclude that an economic improvement is not inevitably an increase in total human happiness; some of the values gained may represent mistakes by individuals about what is in their own interest. You may still agree that, for most people most of the time, revealed preference is the best available way of measuring value and that economic efficiency is therefore a good, although not a perfect, measure of total happiness.
The second divergence between economic improvement and increased utility involves comparisons between people. In summing individual values in order to decide whether some change is an improvement or a worsening, we count a one-dollar gain to one person as just cancelling a one-dollar loss to another. We act as if a dollar (or what a dollar can buy) were worth the same amount of happiness to everyone. Pretty clearly, that isn’t true.
If the rule that the economist uses for making interpersonal comparisons is wrong, why should we use it and how can it tell us anything about what legal rules maximize total happiness? The answer to the first question is that we use the rule because my value for an apple is much easier to observe than my utility for an apple. We can observe my value for an apple by how much I am willing to pay for one and we can, as I have just demonstrated, set up legal rules (property rights) that give me the apple if and only if its value to me is greater than its value to anyone else.
A system of rules that gave me the apple only if I got more utility from it than anyone else would be very much harder to construct. My actions show my utility for an apple relative to my utility for some other good that I am offering to exchange for it (dollars in this case), not relative to someone else’s utility for the same apple. In order to give the apple to the person who got the highest utility for it, someone would have to judge how much happier an apple made each of us. Observing other people’s utility may not be impossible, but it is much harder than observing our own. It follows that it is much easier to design institutions that maximize value, that produce changes if and only if they are economic improvements, than to design institutions that maximize total utility.
It is easier to figure out what increases value than what increases utility, but is the answer of any use? Am I not again searching where the light is best instead of where I dropped my wallet? I think not. In many situations, although not in all, the fact that a change is an economic improvement, increases total value, is strong evidence that it also increases total utility. Since changes in economic value are much easier to measure than changes in utility, we may use the former as a proxy for the latter.
Consider, for example, the abolition of a tariff on U.S. imports. Suppose we could show (as in many cases we can) that, in addition to benefitting our trading partners abroad, it is an economic improvement for residents of the U.S., that the gain to Americans who are better off as a result of abolishing the tariff (workers and stockholders in U.S. export industries and American consumers of imported goods or goods that compete with imports), measured in dollars, is greater than the loss to those who are worse off (workers and stockholders in industries that compete with imports). Individual gainers and losers may have greatly varying values for a dollar; a change that benefits one of them by six dollars and hurts another by five is not necessarily an improvement in total utility. But both gainers and losers are large and diverse groups and there is no obvious reason to expect the one group, on average, to value dollars more or less than the other. If the average is about the same for both groups, then a change that produces a gain in value probably produces a gain in utility as well. That was the argument used by Alfred Marshall, who invented the idea of economic improvement, to justify using it as an approximate way of identifying changes that increase total utility.
The approximation should be a good one as long as we are considering situations where there is no reason to expect gainers and losers to have, on average, different utilities for a dollar, different relations between value measured in dollars and utility measured in some absolute units of happiness. In many cases that is a reasonable assumption. Buyers and sellers of apples, lost hunters and owners of locked cabins in the woods, are likely to be similar people, even the same people at different times.
There is one obvious exception. We expect, as a general rule, that the more money you have the less an additional dollar is worth to you and therefore that, on average, a dollar represents more happiness to someone with very little money than to someone with a lot of money. That is why we rarely give charity to millionaires. We therefore expect that, if gainers and losers have very different incomes, the net change in value will be a poor measure of the net change in happiness.
A change that makes a rich man ten dollars worse off and a poor man nine dollars better off is an economic worsening but it may well increase the amount of happiness in the world. The same is true for a change that harms a large group of rich people by a total of ten million dollars and benefits a large group of poor people by a total of nine million. The obvious conclusion, and one that many utilitarians have drawn, is that income redistribution is a good thing. Taxing the rich and giving the money to the poor may be an economic worsening, due to collection costs and disincentives, and yet a utilitarian improvement.
My reasons for disagreeing with that conclusion are two. The first is that since the poor are, as a rule, politically weak, they are at least as likely to be the victims of governmental income transfers as they are to be the beneficiaries. That is the point that I made in Chapter 4. The second is that the struggle among groups trying to make themselves beneficiaries rather than victims is likely to be an expensive one, making practically all of us, rich and poor, worse off in a society that permits such redistribution than in one that does not. That is the point that I made in Chapter 38. Those two chapters were a utilitarian attack on one of the chief doctrines that divides utilitarians from libertarians.
Some pages back I abandoned the subject of specific rules in order to show the connection between economic improvement and increases in total happiness, to show why designing rules to maximize economic efficiency makes sense as a way of increasing human happiness. I have now done so. I have not shown that economic improvement and increases in total utility are the same; they are not. I have shown why the former is an approximate measure of the latter and may, for practical purposes, be the best measure available. Readers who are not convinced may want to look at Marshall’s original argument or at the more detailed discussion of economic efficiency in one or another of my other books, the former listed in Appendix II, the latter in Appendix I. Readers who are economics students should be warned that those are almost the only places to look. Modern economics texts other than mine use a different, although for most purposes equivalent, definition of improvement.
It is now time to go back to discussing specific rules. The question I shall be investigating is how one would design legal rules to maximize economic efficiency, to permit changes that are economic improvements and prevent changes that are economic worsenings.
Consider again the solution to the apple problem. If we do not enforce property rights in apples at least two kinds of inefficient change may occur. First, apples may be transferred from owners who value them more to thieves who value them less. Second, thieves may spend time and money stealing apples instead of buying them.
Suppose the apple is worth two dollars to you and four dollars to me. Instead of buying it for three dollars I sneak into your orchard at night and steal it, at a cost of a dollar’s worth of time and effort. You are worse off by two dollars (the value of the apple to you) and I am better off by three dollars (the value of the apple to me minus the cost to me of getting it), so there is a net gain of one dollar; my stealing the apple is an economic improvement over my not getting it at all. But not getting the apple is not the only alternative; I could have bought it instead. Stealing the apple is worse than buying the apple, since that would have produced a net gain of two dollars. An efficient legal system will include some way of making it in the interest of people who want apples to buy them instead of stealing them. That is why we punish thieves.
How much should we punish them? If all thieves were caught, a fine equal to the value of what is stolen would be sufficient; since stealing things is more trouble than buying them, theft would be the less attractive of the two alternatives. If only a fraction of thieves are caught, say one in ten, the same argument suggests that the punishment should be scaled up accordingly. If the fine for stealing an apple is ten times the price of buying one, then stealing costs the thief, on average, as much money as buying and more trouble.
We now have the same rule for apples and for cabins. The rule I suggested for someone who broke into a cabin was that he should pay a fine equal to the damage done—provided he turned himself in. I included that condition in order to make it a case where the probability of being caught was one.
In order to eliminate inefficient transactions, the amount of the fine (or the probability times the amount, if only a fraction of thieves are caught) must be at least the value of what is taken. The case of the cabin in the woods is an argument against making the fine any higher than that. While we could have one legal rule for apples and a different one for cabins, it may be easier to have a single set of rules defining what property rights are and what happens if you violate them. Such a set of rules should take account of the possibility that some violations of property rights, such as the lost hunter breaking into the cabin, are desirable changes that for some reason cannot be arranged via a voluntary exchange. A punishment lower than the damage done permits some inefficient changes; a punishment higher than the damage done prevents some efficient ones. So the ideal punishment equals the damage done, appropriately adjusted for the probability of catching and convicting the criminal.
A more precise analysis would qualify this conclusion in many ways to take account of complications such as the cost of enforcing law (preventing inefficient crimes may sometimes cost more than it is worth) and the possibility of error in determining guilt. Readers interested in such an analysis will find it in my Law’s Order.
So far I have treated the probability of catching a thief as if it were simply a fact of nature. It is not. By hiring more policemen or offering higher rewards we can increase the probability that thieves will be caught. In setting up a system of legal rules, one of the decisions to be made is whether to catch half the thieves and fine each of them twice what he stole, catch a tenth of the thieves and fine each ten times what he stole, or catch one thief in a thousand and shoot him.
In choosing the proper combination of punishment and probability, we are trading off two kinds of costs. Enforcement cost is the cost of catching criminals: paying policemen, distributing pictures of wanted criminals, or whatever. Punishment cost is the cost of punishing criminals once we have caught them. As we move from a combination of high probability and small punishment to a combination of low probability and large punishment, enforcement costs decline, since we only have to catch one criminal in a hundred instead of one in two. Punishment costs, however, tend to rise with the size of the punishment. So we maximize total value by choosing the combination of probability and punishment that produces the appropriate level of deterrence—probability times punishment equal to damage done by the crime—at the lowest cost.
What is punishment cost and why does it increase with the size of the punishment? Consider first a fine. The cost to the criminal is the amount of money he has to pay; having to pay a ten-dollar fine makes me worse off by exactly ten dollars. That cost is balanced, however, by the benefit to whoever receives the fine: the victim under a system of civil law, where the fine is called a damage payment, or the state under a system of criminal law. The net cost of the fine is only the administrative expense of collecting it.
As the size of the punishment becomes larger it becomes less likely that the criminal can pay it as a fine and more likely that it must take some other form, such as imprisonment or execution. Imprisonment and execution serve at least as well as fines to discourage people from violating other people’s property rights, but the cost to the criminal is no longer a benefit to someone else. When the criminal loses his life, nobody else gets an extra life in exchange. When you are imprisoned, nobody gets the freedom you lose and someone must pay the additional cost of maintaining the prison.
The recognition that punishment is costly provides part of the answer to another problem mentioned in Chapter 41, how sure we have to be that someone is guilty before convicting him. Punishing the innocent results in the same sorts of costs as punishing the guilty without providing the benefit of deterrence. In designing the optimal system of legal rules, we must balance the punishment cost of convicting innocent defendants against the costs of a higher standard of proof: hiring more policemen and acquitting more guilty defendants.
One conclusion is that we will want a higher standard of proof for an offense that results in a costly punishment, such as execution, than for an offense that results in an inexpensive punishment, such as a fine. That is, in fact, the way our present legal system works. A higher standard of proof is required in criminal cases (“beyond a reasonable doubt”) than in civil cases (“the preponderance of the evidence”). This is not simply a matter of taking more care in more important cases; a million dollar damage payment is a bigger punishment than a two-week jail sentence, but the standard of proof required to impose it is lower.
We have now seen, at least in a general way, how and why property rights should be enforced. There is one feature of the analysis that I find interesting and some readers may find shocking. In calculating the costs and benefits whose sum we try to maximize, costs and benefits to the thief have the same weight as costs and benefits to the victim. In judging whether a change was inefficient and should therefore be prevented, gains to the criminal were balanced against costs to the victim. In choosing a combination of probability and punishment we included the cost of the punishment to the criminal along with costs of enforcement and costs paid (or benefits received) by the court system in the total to be minimized.
What is interesting about this is that we are deriving libertarian results rather than assuming them. We start with an assumption, utilitarianism, that says nothing at all about the relative virtue of thieves and victims. We end with a legal system in which thieves are punished.
Before leaving the question of enforcing property rights and going on to discuss how those rights should be defined, there are a few more things worth noting. As I pointed out in the previous chapter, there are two ways of measuring utility. One is from the outside, by trying to estimate how much someone else values something; I argued that doing so is not impossible but that it is difficult to do it very well. The other is from the inside; each of us knows quite a lot about what he values and his actions reflect that knowledge.
The legal rules I have suggested use both methods. Apples are allocated by revealed preference; if I think the apple is worth more to me than you think it is worth to you, I buy it from you. Locked cabins in the woods are allocated by a combination of revealed preference and outside observation. The hunter decides whether to break in according to how much he values access to a telephone, but the court decides the damages he must pay according to how much it thinks that the owner values not having his door broken down. This is a point I made earlier, when I suggested that the existence of courts making damage awards is evidence that we believe it is possible to know something about other people’s values.
If revealed preference is a better way of measuring values, why not construct a legal system that depends entirely on revealed preference and never tries to measure someone else’s value for something? The answer is suggested by the example of the cabin in the woods. Since the owner is not present when the lost hunter shows up, there is no way to negotiate a price for the use of the owner’s telephone.
Are there ways to solve this problem without having a court measure value? Perhaps. The owner might decide for himself how much he objected to people breaking into his cabin and post a price list on the door: 50 dollars for breaking the lock and another ten for using the phone. The problem is that there are many different situations in which one person might very much want to use someone else’s property and not have an opportunity to get his permission first; the price list would have to be a long one and it might be necessary to post it not only on the door but on every tree. It would have to cover not only breaking down the door to use the telephone but also trespassing onto the property while running away from a bear, cutting dead wood to make a fire to keep from freezing, and perhaps even bulldozing down the cabin to stop the spread of a forest fire. All things considered, using a court to estimate damages seems a more practical solution.
Another alternative would be to arrange a contract in advance between the hunter and the property owner defining the circumstances and conditions under which the former could use the latter’s property. Here again, there are practical difficulties, due to the variety of possible problems and the large number of people involved. Each individual hunter has a very low probability of being lost and having to break into a cabin and an even lower probability of having to break into any particular cabin. Negotiating terms in advance for an event that has only one chance in ten million of happening is unlikely to be worth the trouble. If we try to draw up advance contracts covering every possible contingency we shall have no time to do anything else.
What these examples suggest is that it is not practical to set up a legal system in which outcomes are entirely determined by revealed preference and voluntary transactions. At the same time, because the market provides a less expensive and more accurate way of measuring values, we would like a system that uses courts only when markets are not a viable alternative. If, for example, there is some class of cases where we are sure that market transactions are always practical and the efficient level of crime is therefore zero, we might make the punishment much more than the court’s estimate of damage done (‘punitive damages’) in order to make it less likely that mistakes by the court system will encourage inefficient crimes. A full discussion of such issues would again carry us beyond what can be done in a chapter.
I have now finished sketching the answer to one of the problems raised in an earlier chapter, the appropriate punishment for a thief. In doing so, I have laid the groundwork for answering two other questions raised in that chapter: the proper restrictions on risky activity and the proper definition of property rights.
The case of risky activity, as exemplified by the pilot with a small chance of crashing anywhere within a thousand miles of his starting point, is similar to the case of the starving hunter. The pilot, unlike the hunter, does not actually decide to break into someone’s house. He does decide how often to fly, how often to have his plane checked and what kind of safety equipment to buy. By making those decisions he controls the probability that he will end up entering someone’s house through the roof. Similarly, someone who drives a car or uses dynamite to remove stumps from his land does not choose to have an accident that results in injury to someone else’s person or property. He does, however, choose how much to drive or blast and how carefully. In each case, the proper legal rule is one that forces him to pay for any damage produced by his actions. Under such a rule, he will take an action if and only if its value to him is great enough to make up for the probabilistic damage he causes.
A full discussion of the complications associated with problems of risk would again take us far beyond the constraints of even a very long chapter. One of the points we would have to deal with is the possibility that someone whose airplane destroys my house may not have enough money to pay for the damage, even assuming he is alive to do so. If so, we might want legal rules that allow potential victims to forbid my taking off unless I can show that I have suitable insurance. A second point is that accidents are frequently a product of decisions made by both parties concerned. Your car would not have collided with my bicycle if I had not been riding in dark clothes at night, but my carelessness would have produced only a close call if your brakes had been functioning correctly. This makes it harder to design efficient rules to control accidents. If I know that you will be liable for all the costs of the accident, I have no incentive to take precautions; if you know I will be liable, you have no incentive; if the liability is divided between us, both of us have an inefficiently low incentive.
The final question to be dealt with is how property rights should be defined, the question implicit in my discussion of trespass by single photons and single molecules of carbon dioxide. We start by noting that what we call a property right, the ownership of a piece of land for instance, is actually a complicated bundle of such rights. Under current American law it includes the right to forbid trespass but not, under most circumstances, the right to shoot trespassers or to plant land mines where you expect them to step. It does not include the right to forbid overflights by airplanes nor trespass by small numbers of carbon dioxide molecules or photons. The questions I raised in Chapter 41 are questions about what belongs in the bundle.
It seems at first that the answer is obvious; when I acquire land I acquire all of the rights associated with it. The problem is that some rights are associated with more than one piece of land. The right to decide whether a light beam crosses the border from your land to mine is associated with both my property and yours. It is useful to me because if I control the right I can keep you from firing laser beams at my front door. I can even keep you from shining a flashlight at my darkroom window. It is useful to you because if you do not own that right you cannot do anything on your property that can be seen from mine.
In this particular case, there is an obvious commonsense solution; you have the right to make any light whose intensity can be seen but not felt. The line is drawn somewhere between the brightest light likely to be produced by your normal activities and the weakest likely to do damage to my property. Unless your normal activities include outdoor testing of high powered lasers or nuclear weapons, there should be no problem finding a suitable dividing line.
The problem arises, however, in a great variety of different forms, for many of which there is no easy answer. One can get some idea of the ambiguity about what right belongs in what bundle by reading a good casebook on tort law. Real-world law cases have included questions such as whether my building can block your sunlight, whether I am allowed to make an addition to my house that prevents your chimney from drawing properly, and whether a candy factory is allowed to produce vibrations in the ground that only become a problem when a neighboring physician builds a consulting room on his own property adjacent to the factory.
The first step in dealing with such problems is to realize that the problem is not simply one person injuring another; if it were, we could prohibit the injury or charge damages. It is rather a case of two people engaged in inconsistent activities. My candy factory would be no problem if you had built your consulting room somewhere else on your lot; your building your consulting room where you did would be no problem if I were not running a candy factory. This is a different way of saying that the relevant right—in this case, the right to decide whether I can run machinery that produces vibrations on your land—seems to belong in two different bundles of rights, my ownership of my land and your ownership of yours.
The second step is to realize that in many cases it does not much matter how the initial bundles of rights are defined, at least from the standpoint of economic efficiency. If a right is valuable to two people and belongs to the one who values it less, his neighbor can always offer to buy it from him. If you have the right to order me to shut down my candy factory I can offer instead to pay the cost of tearing down your consulting room and rebuilding it on the other side of the lot. If the right is more valuable to me than to you, I should be able to make some offer that you will accept.
This insight leads us to the Coase Theorem, named after Ronald Coase, the economist whose ideas are largely responsible for this part of the chapter. The Coase Theorem states that any initial definition of property rights will lead to an efficient outcome, provided that transaction costs are zero.
The condition—zero transaction costs—is as important as the theorem. Suppose we start with a definition of property rights that forbids trespassing photons; anyone may forbid me from making a light that he can see. The right to decide whether or not I turn on the lights in my house is worth more to me than to my neighbors, so in principle I should be able to buy their permission. The problem is that there are a lot of people living within sight of my house. Buying permission from most of them does no good, since I need permission from all. The result is likely to be a difficult bargaining game, with at least some of my neighbors trying to extort from me a sizable fraction of the value of my land in exchange for their permission to use it.
This suggests that, in deciding how property rights ought to be bundled, there are two important considerations. The first is that, so far as possible, rights should go in the bundle where they are most valuable. The right to control the air a foot over a piece of land is worth more to the owner than to anyone else, so ownership of land usually includes ownership of the space immediately above it. The second is that, since the proper composition of bundles of rights will often be uncertain and may change over time, they should be defined in a way that makes it as easy as possible to trade rights. Property rights should be defined in a way that minimizes the transaction costs of likely transactions.
One of the questions to be decided is how to bundle the rights; another and closely related question is what the rights are that we are bundling. Does my right to forbid intense lights and sounds from my property mean that I can forbid my neighbor from testing lasers and nuclear weapons—and holding loud parties—or only that I can collect damages afterwards?
The answer has been suggested in an earlier discussion. Where transactions between the two parties are easily arranged, as in the case where only two neighbors are involved, there is much to be said for an absolute right to forbid backed up by punitive damages. That way the court does not have to engage in the difficult task of measuring the cost to me of being blown up or kept awake. If what my neighbor wants to do is sufficiently important to him he can offer to buy my permission—or my land.
But where transactions are impractical, a damage rule may be the best solution. It is not practical to buy the right to emit unpleasant fumes from all of the three thousand people who can occasionally smell what comes out of my smokestack. Even if it is worth much more to me to be able to run my factory than it is to them not to smell it, I will not be able to buy the permission of all of them. I face the same sort of bargaining problem as in the previous case of trespassing photons; one holdout can prevent the entire deal. Efficient legal rules might allocate the relevant right to my neighbors instead of to me but make it a right to collect damages rather than a right to close down the factory.
I believe I have now justified the title of this chapter. I have shown that economic analysis can answer questions about what the law ought to be that I cannot answer—that I believe cannot be answered—on the basis of libertarian principles.
That claim must be qualified in several ways. I have shown what the law should be only in the sense in which an engineering textbook shows how a bridge should be built. The engineering textbook shows how general physical principles can be applied to specific information, such as the strength of available materials and the width of the river to be bridged, to figure out how to build a specific bridge. I have shown how economic principles can be applied to specific information, such as the value of one right to the owner of another or the costs associated with arranging different sorts of transactions, to figure out what legal rules maximize human happiness in a particular society. Economics is a newer field than engineering and more is known about the strength of materials than about the cost of transactions, so the engineering textbook does its job better than I can do mine.
A second qualification is to point out that what I have given in this chapter is a very sketchy description of one part of a large field. A full analysis of what legal rules are implied by economic efficiency requires several volumes, not all of which have yet been written. Furthermore, the question of what rules are economically efficient is not the only question that the economic analysis of law deals with, merely the question that seems to me most relevant to this book. Much of the existing economics and law literature is devoted to the very different objectives of understanding why particular legal rules exist and what their consequences are.
Most of the ideas I have been explaining were invented within the past fifty years; they are part of a field that is still being developed and much of which is still controversial. Readers who are interested in a much more detailed account of the economic analysis of law, including much of my own work in the field, will find it in my Law’s Order. They may also find the next chapter of interest. It is based on one of my published articles and describes a society in which all laws, including the law against murder, were privately enforced.
Before ending the chapter, there is one final qualification to be made. Economic efficiency is only an approximate measure of total utility and total utility is only a very partial description of what I and, I think, other people value. Even if we can prove that certain legal rules are economically efficient, it does not necessarily follow that we should be in favor of them.
What I find interesting and useful about the economic analysis of law is not that it tells me for certain what the law should be but that it starts with objectives based on what most of us want and apparently unrelated to questions of right and wrong and ends with answers—conclusions about what the law should be—not all of which are obvious.