In the society described in part III, law enforcement and law are private goods produced on a private market. Law enforcement is produced by enforcement agencies and sold directly to their customers. Law is produced by arbitration agencies and sold to the enforcement agencies, who resell it to their customers as part of the bundle of services they provide.

Enforcement agencies want to provide their customers with legal rules the customers like; arbitration agencies want to provide legal rules that enforcement agencies want for their customers. The result should be good law. In this chapter I use the concept of economic efficiency to make more precise what that means. Having done so, I will show in what ways the legal rules produced by the market for law may be inefficient, less good than the rules that would be designed by a benevolent and perfectly wise legislator. It is, in the language of the previous chapter, an explanation of market failure on the market for law.

The concept of economic efficiency, although not the term, goes back to Alfred Marshall, who put together modern economics a little more than a century ago. It was his solution to the problem of how to decide whether a benefit to one person does or does not outweigh the loss to someone else. It is an imperfect solution, for reasons briefly described in Chapter 43, but a useful one since, unlike more philosophically satisfactory approaches, it is an objective that economics tells us a good deal about how to achieve.

The idea, as earlier explained, is simple. We measure the benefit to someone of any change by the largest amount he would pay to get it, the cost by the largest amount he would pay to prevent it. The net effect of any change is the sum of all benefits (positive) and costs (negative) to everyone it affects. A change is an improvement if the sum is positive, a worsening if it is negative. We describe an outcome as efficient if there remain no ways of changing it that would be improvements.

Seen from the standpoint of the rights enforcement agencies, the argument I have already offered implies that the market for law will generate efficient law. If a change in the legal rule between two agencies benefits both, it is in the interest of the arbitration agency to make it so as to improve the quality of the product it sells. If it benefits one agency by more than it hurts the other, it is in the interest of the two to agree on the change, with the agency that supports it compensating the other either in cash or by agreeing to accept its preference on some other issue where the preferences go the other way.

It follows that the legal rules the arbitration agency produces should maximize the net benefit to the agencies. Whether it also maximizes net benefit to their customers is a more complicated question, for at least two reasons.

Monopolistic Competition in the Law Enforcement Market

In the perfectly competitive market of the economics textbooks, many producers produce identical products, products that are perfect substitutes for each other. When I buy a bushel of wheat or a barrel of oil I do not care which firm made it, with the result that firms end up selling identical goods at the same price. A firm that charged more would sell nothing.

Contrast that to the market for computers. A Macintosh and a Windows computer are not perfect substitutes; some people prefer one, some another. If they happen to sell for the same price, some customers will buy the Mac, some the PC. If the Mac gets a little more expensive, a few customers, the ones an economist describes as “marginal,” will switch, but most will not. Like an ordinary monopoly, Apple can vary its price over some range, trading off higher prices against fewer sales. But because PCs are substitutes, although imperfect substitutes, how many machines Apple can sell at what price depends in part on the price of the alternative. This is what economists call monopolistic competition. It is a common market pattern, one somewhat more difficult to analyze than the simpler case of perfect competition.

Suppose Apple, which is currently selling its machines at the price that maximizes its profits, is considering some improvement, perhaps a faster processor, that will increase both the cost of making the machine and its value to the customers. The increased value means that Apple can raise its price and still sell as many machines as before. But how much it can raise the price without losing sales depends not on the value to all customers but on the value to the marginal customers. So far as the customers who greatly prefer Mac to Windows are concerned, Apple could have raised its price before—if it knew who they were. But raising the price for everyone would have lost it sales to the marginal customers, which is why it did not do so.

Suppose the increased speed is valuable only to the enthusiasts. In that case, even if the total value to the enthusiasts is greater than the cost to Apple of the improvement, Apple will not make it, since the value to the enthusiasts is irrelevant to how much Apple can sell its computers for. If, on the other hand, Apple makes an improvement that is valuable to the marginal customers, for whom it increases value more than it increases cost, it pays Apple to make the improvement even if it is of no value to the enthusiasts. It follows that it may pay Apple to make some changes whose total value, considering all of the customers, is less than their total cost and fail to make some whose total value is more than their total cost. The design of Apple’s computers has some tendency towards economic efficiency, towards the design that maximizes the value to the customers net of the cost of making them, but an imperfect tendency, since it is based on value only to marginal customers. By offering different models tailored to different segments of its market Apple can reduce the problem but not eliminate it.

Rights enforcement agencies are also in a monopolistically competitive market. Just as with computer companies, the amount an agency can get its customers to pay is an imperfect measure of the value of its services to them since it depends only on the value to the marginal customers. Like Apple, the agency can try to tailor its services and prices to the different requirements of different customers and so come closer to collecting from each the highest price he is willing to pay. But unless it can do so perfectly it will not collect all the value, so there may be changes in what it provides, including the legal rules it agrees to, that increase or decrease its revenue by more or less than they increase or decrease the value of its services to its customers.

It follows that although the market has some tendency to produce efficient law, law that maximizes the benefit to those who live under it, the tendency is imperfect; the desires of marginal customers are weighted more heavily than those of customers who strongly prefer their agency to any of its competitors.

Externalities: Market Failure in the Market for Law

We can expect the legal rules between Agencies A and B to maximize their joint welfare. We can expect the rules to maximize the joint welfare of all of their clients, subject to the problem I have just pointed out. But we cannot expect the rules applying between A and B to take into account the effects of those rules on customers of other enforcement agencies. So we can expect the result to be optimal only when the legal rule between A and B produces no net effect on customers of other agencies. In many cases this is plausible; the rule that determines what happens if A breaks his contract with B or breaks into B’s house or breaks B’s arm should have little effect on C.

But consider intellectual property law. Imagine that Anne, a computer programmer, is bargaining with Bill, a consumer, for Bill’s agreement to pay Anne $10 for each copy of any of her computer programs that Bill makes. The benefit to Anne of receiving $10/copy is exactly balanced by the cost to Bill of paying $10/copy; if these were the only costs and benefits, agreement and disagreement would be equally efficient.

But there are at least two other costs. One is that Bill will make fewer copies of the program than if copying were free. Perhaps he will put a copy on his desktop machine but not on his laptop, perhaps he will buy copies of two of Anne’s programs but not a third, since that one is worth only $5 to him. This is the familiar deadweight cost of copyright, the inefficiency due to the difference between the (positive) price of making an additional copy to the user and the (zero) marginal cost to the copyright owner of permitting an additional copy to be made, resulting in an inefficiently low number of copies. A second cost is the cost of enforcing the agreement. Keeping track of what copies Bill has made will be costly, perhaps impossible, and any resulting dispute may lead to expensive litigation.

To balance these costs there is a benefit, the programs that Anne will write if she expects to be paid for them and won’t if she does not. If we were considering the effect of requiring all consumers of Anne’s intellectual property to pay for it, that benefit might well outweigh the costs we have described, making copyright protection for programs economically efficient. If Anne does not write any programs there will be nothing for Bill to copy.

But we are considering the question not with regard to the whole world but only with regard to Bill. The additional revenue Anne will receive as a result of Bill being obliged to respect her copyrights will produce only a very small increase in output. That increase will benefit everyone who uses Anne’s programs, but only the part of that benefit that goes to Bill will be relevant to the negotiation between them. He is, however, bearing all of the cost of using fewer programs than he would if he had not agreed to respect Anne’s copyrights and she is bearing all of the costs of enforcing that agreement against him. It follows that the net benefit to them of an agreement between them is almost certainly negative.

That may still be true from the standpoint of their enforcement agencies, which are the ones actually negotiating. Bill’s agency will take into account not merely the benefit to him from the increased output due to his being bound by Anne’s copyright but the benefit to all of its customers due to the increased output from all of them being bound by Anne’s copyright. The result is still only a small fraction of the total benefit from the agreement, assuming that there are many enforcement agencies each serving only a small part of the population. The fraction becomes larger if we allow for the possibility of copyright negotiations among groups of enforcement agencies, with each agreeing to recognize the copyrights of the customers of all of the others if they will all agree similarly. Such negotiations would be analogous to the negotiations among nations by which international intellectual property rights are now established. Even allowing for such multiparty negotiations, our result, although weaker, still remains; we would expect an inefficiently low level of protection for intellectual property. We might well get no protection at all.

This raises two questions, both outside the range of this chapter. One is whether intellectual property protection is desirable at all—readers interested in arguments that it is not will find them in Against Intellectual Monopoly by Michele Boldrin and David Levine. The other, whether the equivalent of intellectual property protection could be provided in other ways, for instance by contract, is discussed in the context of digital intellectual property and encryption in my Future Imperfect.

A similar situation will exist for pollution law, where Anne’s right to sue Bill for polluting her air results in a reduction of Bill’s emissions and thus an external benefit for Anne’s neighbor Carl. It may exist in other important contexts as well. In all of these cases, we would expect the legal rules generated by the private market to be less than perfectly efficient. They should still be better than the rules generated by political mechanisms under our current system, since we have little theoretical reason to expect those legal rules to be efficient at all. From that standpoint, this chapter is offering a special case of the argument of the previous chapter. Market failure sometimes exists on the private market. On the political market it is the rule.