Private Enforcement Of Law Essay, Research Paper
4. Private Enforcement of Law4.1 Octracism and boycottUnlike governments, private bodies do not normally have jails, police, and soldiers to enforce their rulings. This is why many people assume that only governments can enforce private agreements, and, indeed, private arbitration. However, the whole thrust of the economic analysis of crime is that a criminal sanction is much like a price; what precludes private bodies from exacting the same price that the government does? While the usual criminal sanction is imprisonment, it is likely that time in prison has some implicit price. It should therefore be possible to secure compliance by threatening to reduce a person’s income. And private bodies do have this power.The classic non-violent sanctions are ostracism and boycott. A boycott is essentially a refusal to trade with an offender. Ostracism is a more extreme form of boycott; the community completely cuts off an ostracized individual. We frequently hear about sanctions like this in primitive and ancient societies. Cut off from their tribe or nation, ostracized individuals had to either survive autarchically or find another community. The latter was naturally difficult because an individual ostracized by one group might be a bad member of any group.A common assumption is that boycott and ostracism can only work in face-to-face societies. There are two reasons why people make this assumption. First, the usefulness of boycott and ostracism depends crucially on information. In a small tribe, everyone knows who has been ostracized; it is not feasible for an offender to assume some new identity or move to a new part of town. The complaint is that modern societies are too anonymous for this to work; it is easy to cast off a stigma by finding a new group of friends, trading partners, and community when there are five billion humans on earth.The second complaint is that boycott and ostracism suffer from a serious free rider problem; hence, if they work at all, they would have to be enforced by the government. The sanction isn’t really private at all. Imagine, for example, individuals who find the working conditions of grape workers offensive. They could try a consumer boycott of grapes to punish the employers. Yet there is a serious free rider problem here: all people offended by the grape- workers’ conditions benefit if the employers get punished via reduced sales (assuming that this is a good strategy in the first place). But the costs of enforcement are borne solely by the individuals who have to forego their favorite fruit. We might expect cheating and chiselling unless the boycotters are ferociously principled. Like most free rider problems, a boycott could only work if the state enforced it. That seems self-defeating if we want a private means of punishment.While these two doubts about the feasibility of boycott and ostracism are plausible, they simply aren’t true in many cases. First of all, while imodern societies lack the face-to-face character of simpler times, information storage and transmission have advanced in step with modern technology. Credit ratings, for example, or rental histories, have become ever more feasible and inescapable with the development of the computer. We could easily imagine comparable data banks for information about employment history. More to the point, in the modern world it would be easy to register people who violated arbitration agreements or defied legitimate rulings. There is actually a strong case that the requisite information for boycott and ostracism is more available in modern societies than in earlier ones. Before the advent of modern information storage and retrieval technology, one could simply move to another city and leave one’s past behind. As the world’s economy globalizes, it grows ever more difficult to get rid of a bad commercial reputation.The second doubt about boycott and ostracism misses the whole point. Of course, if the benefit produced by the boycott/ostracism really is collective (as in the grape worker example), then it is a public good. But boycotting and ostracizing dishonest trading partners, far from being selfless service to the public good, is in one’s immediate interest. Just as landlords would voluntarily refrain from renting to people with bad rental or credit histories because they are bad risks, so too is it in the interest of merchants and employers to refuse to trade with people who break arbitration agreements. This is why the damage to one’s reputation is so harmful to merchants — they won’t get customers if they can’t win their trust. (Or at the least they would have to pay a price premium to compensate trading partners for the extra risk.)Ostracism and boycott work well in the merchant community precisely because the common good — sanctioning dishonest businesspeople — is perfectly consistent with the private interest in avoiding bad risks. If people want to boycott grapes to pressure employers to give grape workers better pay, then the cost is private while the benefit is public. But if merchants want to boycott a cheater in order to avoid bad risks, the benefit of sanctioning is just as private as the cost. There is no public goods problem here. Each individual need merely attend to his private interest, and the public good of punishment springs up automatically.Here, then, lies a simple tool with which private arbitration might be enforced. It is non-violent and wholely private; but it works in many cases. Private enforcement of law is not only possible but real.One mustn’t forget, however, that boycott and ostracism depend on the free flow of information. Information is hardly costless even in competitive markets; and it is likely that individuals would only check out the history of their trading partners if the value of their deal were fairly high. Landlords, for example, usually run a credit check on potential tenants, but motel owners do not. But, as mentioned earlier, it isn’t necessary for every link in a contractual deal to investigate the history of every other link. The traders can get guarantees from some third organization that specializes in information. Consider the motel case. Quite possibly, the owner might bill the cost of damages (e.g., stolen towels) straight to the customers’ credit card company. Even though the motel owner finds it too costly to research the history of the tenant, the owner does have information about the trustworthiness of the credit card company. The credit card company researches its customer’s credit history because they have a long-term relationship. So it is possible to economize on information costs by contracting with guarantors. The guarantor need be the only body that knows the client’s history, and the client’s trading partners need only know the guarantor’s reputation in order for the client and the trading partner to trust each other. The problem of information costs in such situations actually becomes trivial. Observe how stores don’t care about the credit history of customers who use credit cards. They don’t care because they know that the credit card company pays no matter what; the credit card company bears the risk that the store’s patron will default. Since the credit card company has a big stake in the trustworthiness of its clients, it will likely spend the money to check their credit history. But the parties must incur this expense only once before they can enjoy the benefits of mutual trust.4.2 Aside: the moral argument against credit ratingsThis system of enforcement is workable without government help. But it does have trouble if there is government hindrance. Alas, the latter is far more common. Many people find it offensive if a bureau records individuals’ wrongful acts and warns people who might be harmed by them. They therefore try to get the government to restrict or even illegalize information storage. A common prohibition placed upon credit bureaus is that they must erase information after X years. Others want to declare that credit ratings are an invasion of privacy.Such regulation would indeed make private enforcement impractical in the modern world. Since private enforcement is non-violent, it has to rely on subtle methods to punish those who break faith. Without the free flow of information, word of mouth would be the only alternative. In narrow business associations this works; but broader privatization rests on free access to information. Critics might object that the right to privacy trumps lesser values like wealth. Some things are more important than social utility, and individual liberty is one of them. I happen to agree with this view, but its application to this question is senseless. For what opponents of credit ratings want to do is suppress the liberty of individuals to announce to the world that they have been wronged, and to suppress the liberty of individuals who would like to protect themselves from dishonest trading partners. Why doesn’t the liberty of these individuals count? Moreover, since individuals get bad credit ratings because they did not live up to their own agreements, how is their liberty violated anyway? They consented to the deal in the first place, so they consented to be reported if they broke faith. Since they consented to everything that happened to them, their liberty isn’t violated.Even more shocking is that governments’ invasions of privacy and punishments are much more objectionable from a moral point of view. Tax forms, for example, must be filled out by everyone, willy-nilly; why isn’t this an invasion of privacy? Boycott and ostracism are unpleasant sanctions. But if private bodies don’t punish fraud, the government will have to. It will probably use its traditional sanction: prison. Surely this is a worse violation of the right to privacy than boycott and ostracism. Advocates of the right to privacy should, on their own grounds, support a wider role for non-violent sanctions. Despite the moral high ground that opponents of ratings assume, their case is weak and hypocritical.4.3 The Becker-Stigler modelIn an important article, Gary Becker and George Stigler suggest another method of private enforcement. Suppose that a firm wants to deter employee theft. Rather than rely on the law, it offers a higher than normal wage. If an employee gets caught, he gets fired — and presumably must accept a lower wage at the next job. (The argument would be stronger if the employer were to “blacklist” dishonest employees, just as landlords report bad tenants to a credit rator.) Becker and Stigler point out that one could spend very little on enforcement if the salary differential were high enough. As they explain, “Trust calls for a salary premium not necessarily because better quality persons are thereby attracted, but because higher salaries impose a cost on violations of trust.”53Becker and Stigler also point out that there is actually no need for overall pay to be higher. An employer might pay employees, say, $100 a month extra. Since workers’ expected earnings exceed their opportunity cost, employers could easily charge employees an “entrance fee” equal to $100 times the expected number of months worked discounted by the interest rate. Alternately, the would-be employee might post a bond for good behavior. If they are honest, their bond earns interest until they quit, at which point they get the bond back. But if they break faith, the employer could fire them and keep the bond. Perhaps a neutral holding company (or arbitrator!) might hold the bond to control any moral hazard problem. A third alternative might be to put the salary premium into a pension fund which an employee must forfeit if the employer gets caught cheating.While Becker and Stigler main goal is to prevent corruption among public police, judges, and the like, they recognize that its applicability is wider. “Our analysis of malfeasance is applicable not only to enforcers but to all public and private employees who must be ‘trusted.’”54 If ostracism and boycott is too costly, individuals in long-term contracts might use Becker-Stigler incentives to win mutual trust. Labor contracts are the clearest possibility. While it is quite rare for employers to exact explicit deposits, they frequently require employees to pay for their own uniforms and equipment. A system of this kind permeates the apartment and housing markets. Typically, a landlord charges a “security deposit.” The point of this deposit is to protect the landlord if the tenant damages the premises or refuses to pay rent . Other rental systems combine deposits with boycott sanctions. For example, when you rent a videotape, you consent for your credit card company to pay any rental fees that you renege on. In other words, you put down a sort of deposit, and your credit card company collects from you by threatening a credit boycott if you decline to pay your monthly statement.We can also imagine a system of mutual deposits. Suppose you and I draft a long-run contract with each other. We might require each of us to put $10,000 in the hands of an arbitrator. If cooperation breaks down and we arbitrate the conflict, it would be easy to enforce a judgment. Since the arbitrator already holds our two deposits, he could simply return the winner’s half of the deposit plus damages out of the other’s person’s half. (The remainder would go to the losing party.) By providing our arbitrator with the tools of enforcement before cooperation breaks down, we can assure each other that all arbitration award will get enforced.There are some problems with the Becker-Stigler system.55 Sometimes employees face liquidity constraints, making it difficult to charge an admission fee. The result might be underemployment in industries which use the Becker-Stigler method of enforcement. It is also possible that employee resentment against such a system might undermine its effectiveness. Employers must often rely on willing worker participation to get good performance; if workers resent deposit systems, employers may be worse off than if they used a less intrusive sort of enforcement. There is also the danger that employers would issue false charges to steal a worker’s bond. While these complaints are not all equally convincing, they limit the effectiveness of Becker-Stigler incentives.Nevertheless, the use of deposits is one effective way to enforce agreements. It could both substitute for and complement other sorts of private enforcement such as ostracism and boycott. In rental markets, deposits are more efficient; in credit markets, boycotts are. Other markets, like videotape rentals, use both in concert.Like most of the policy suggestions that I have made throughout this paper, Becker-Stigler enforcement requires no positive government help. Parties use it because it is simple and cost-effective. But for this sanction to be maximally effective, the state must not hinder it. And a number of legal restrictions on deposits exist. In some cities, the government caps the maximum possible rental deposit. The result is that tenant breaches of trust increase and government courts handle more landlord-tenant disputes. The public courts also restrict parties’ use of liquidated damages clauses. They routinely refuse to enforce allegedly “punitive” damages in contracts.56 One can imagine that the public courts might, in a parallel fashion, overrule an arbitrator who awarded damages out of the parties’ deposits. The stricter the legal restrictions on the use of deposits, the less effective Becker-Stigler enforcement will be.A more appropriate policy, and one which would encourage more extensive private resolution of disputes, would be to leave deposit enforcement systems unregulated. The technique is imperfect, but Becker-Stigler enforcement can be a cheap and effective way to enforce private arbitration rulings — and private enforcement in general. It is non-violent and doesn’t undermine the government’s monopoly on coercion. It just cleverly designs incentives that make coercive enforcement superfluous.4.4 Posting rewards and the Posner criticismBecker and Stigler, again in “Law Enforcement, Malfeasance, and Compensation of Enforcers,” suggest a novel alternative to government supply of police. Why not simply offer rewards to firms that successfully catch and prosecute criminals? The accused would still get a trial in the government’s courts. If found innocent, the enforcers would have to offer compensation to the wrongfully accused. But if found guilty, the enforcers would exact a fine from the criminal. If that is impossible because the criminal is too poor, the government might pay the enforcers a bounty and then imprison the criminal. Though Becker and Stigler do not suggest this, we could auction off criminals as indentured servants to profit-making jails. The government would then merely order that criminals get released once they pay off their debt.As Becker and Stigler admit, “Free competition among enforcement firms may seem strange, even terrifying, and much more radical than the method of compensation proposed earlier to eliminate malfeasance by salaried enforcers.”57 But they find that if we look at their system less emotionally, it has many virtues. First, it would entirely eliminate malfeasance. It is only possible for criminals to bribe enforcers now because the private costs to the criminal of getting punished exceed the private benefits to the enforcer of carrying out the punishment. Police officers, for example, get a fixed salary; it makes little monetary difference to them how many criminals wind up in jail. But it matters greatly to the criminal. Profitable trades can and do spring up between criminals and enforcers, since the criminal’s welfare falls if he goes to jail, but the officer’s income is invariant with his number of arrests. But if enforcers were paid out of the pockets of convicted criminals, the situation would change. If the fine for a crime is $1000, there is no way that a guilty individual could bribe an enforcer not to prosecute, because the only sufficient “bribe” would exactly equal the fine. As Becker and Stigler design their system with a perfect link between costs to criminals and benefits to enforcers.Second, they argue, their system would leave the selection of means to the market. Just as market-like schemes for pollution abatement mandate nothing but outcomes, and leave the choice of methods to the market, the bounty system would set a reward for apprehending criminals but let the market determine the most effective way to apprehend them. As Becker and Stigler put it, “rules usually provide neither the slightest hint of where to look for violations nor the incentive to convict violators.”58 Both of these functions, in their view, might be left to the market: the bounties would provide both the incentive to figure out the least-cost enforcement methods and the incentive to use them.Third, Stigler and Becker criticize existing compensation systems for giving victims bad incentives. If there is no compensation for victims, they grow excessively cautious. A system that offers a bounty if criminals get caught would, they say, lead to optimal victim care. If it is easier for victims to take care than extract compensation from criminals, they take care; if the opposite is true, the victim take the risk and relies on compensation. As Becker and Stigler explain, “the right amount of self-protection by potential victims is encouraged, not the excessive (wasteful) self-protection that results when victims are not compensated, or the inadequate self-protection that results when they are automatically compensated.”59While the Becker-Stigler proposal is open to criticism, they are sophisticated enough to anticipate and adjust for the obvious objections. To prevent overzealous (or indeed criminal) enforcement, the enforcer would have to compensate any innocent person they apprehend. (Incidentally, this might make it possible for defense attorneys to serve on a contingency basis just as plaintiffs’ attorneys often do today.) If criminals has no money, the government would pay the bounty and then imprison the criminal. Since firms might not have enough money to compensate the wrongfully accused, they suggest mandatory bonding or malpractice insurance.In fact, Stigler and Becker anticipate a much more sophisticated complaint later popularized by Judge Posner. Posner begins with some simple claims in the economics of crime. There are two central variables : the probability and the severity of punishment. Since there are two variables, it is necessary to choose some “mix” of the two. Does economic analysis have anything to say on this matter? Posner suggests that the costs of increasing the probability of punishment are high and the costs of increasing the severity are low (indeed, virtually zero). But if this is true, then we could always get the same level of criminal activity at a lower cost if we reduced our spending on enforcement and increased the punishment. The logical conclusion is that we should couple horrific punishments with negligible enforcement. As Posner puts it, “every increase in the size of the fine is costless, while each corresponding decrease in the probability of apprehension and conviction, designed to offset the increase in the fine and so maintain a constant expected punishment cost, reduces the costs of enforcement.”60The Becker-Stigler bounty system is imperfect, says Posner, because it cannot reach this optimal mix of probability and severity of punishment. Why? Because the level of the fine automatically determines the probability, too. A $1000 fine, for example, would lead, in equilibrium, to marginal enforcement expenditures (neglecting the interest rate) of $1000. This in turn should yield some probability of punishment — suppose .2. Now how would the government reach the optiminal probability-severity mix. If it lowers the fine, it lowers the probability of conviction (say to .1); if it raises the fine, it raises the probability of conviction (e.g., to .25). There isn’t any way to increase the severity and lower the probability at one and the same time. To do so, the government would have to use a second policy tool, such as an excise tax on enforcers.Becker and Stigler anticipate Posner’s point (and, indeed, his tax-solution) in their original article. While conceding the logic of the point, they suggested that Posner’s super-high punishments would just lead to more malfeasance. Raising punishments, then, might not be effective since it increases the gains of corruption. If the state taxes bounties to induce optimal enforcement expenditures, the incentive for malfeasance re-emerges. All three systems, whether bounties without taxes, bounties with taxes, or public enforcement, have some problems. But the surprising conclusion that arises out of their debate is that the public sector has no clear advantage over the private sector for the enforcement of crime. On top of this, the debate between these three economists sheds light on the problems of restiution-based systems of criminal justice, which I investigate in the next section.4.5 Torts and crimes: incentives for enforcement”[M]ost cases, both civil and criminal, in the public courts are settled out of court rather than litigated to judgment, and most of the inputs into the litigation of such systems are private,”64 explain Landes and Posner. Private inputs are especially prominent in civil cases — a wealth trasnfer from loser to winner functions as both an incentive (to try cases) and a deterrent (against wrongful activity). And both sides in a civil suit pay their own legal costs. Evidently a sizable chunk of civil law enforcement relies on private resources. All that the government does is decree some “bounties” that convicted losers have to pay to winners, and then enforce the judgment. After the courts create these incentives, the system largely runs itself: there are incentives to prosecute and dis-incentives to commit civil wrongs. Here is a startlingly close approximation of the Becker-Stigler enforcement system suggested in section 4.4.The role of non-governmental parties in civil suits is indeed impressive, and testifies for the importance of the private sector in law enforcement. This section will take tort principles a step further and consider applying the same model to crimes. Crime victims would have a legal right to collect restitution from criminals convicted of harming them; crimes would, like torts, be offenses against individuals rather than society. (We might solve the problem of inchoate crimes by requiring restitution for them, too.) Since criminals frequently have liquidity constraints, the likely corollary of a restitution-based system for crimes would be a system of indentured servitude for criminals. If a court rules, for example, that a penniless criminal owes a victim $10,000 for an assault, the victim could accept bids from free-market jails. The victim would presumably get his money upfront, and then the business/prison would be allowed to “exploit” $10,000 worth of value out of the convicted criminal.Such a system would get rid of government prosecutors and government jails; instead of taxing the general public for such functions, the system would instead set up incentives for private individuals to prosecute and deter criminals. Why don’t we do this already? The usual answer is that deterring crimes is a public good, while deterring torts is a private good. This is the economic rationale for private enforcement of civil wrongs and government enforcement of crimes. [Quote Cooter] Yet this rationale is hardly adequate. Imagine that damages in tort cases went to the government rather than the plaintiff. Deterring risky behavior would become a public good! Deterrence is never inherently a public good; if damages went to the plaintiff instead of the state, it would become a private good. Deterring crime is only a public good if there are no incentives to prosecute and jail criminals. One might reply that a crime victim’s choice not to prosecute injures us all. True, but so does a tort victim’s choice not to sue. In both cases, society benefits if crime and negligent behavior get deterred; and if there were no incentives for either, there would indeed be underdeterrence. But just as award of damages gives private parties incentives to supply the public good of safety from torts, so would restitution give private parties incentives to supply the public good of safety from crimes.One important conclusion of the economics of crime is that fines could potentially deter just as well as prison. And a major conclusion of the externalities literature is that from an efficiency viewpoint, it doesn’t matter who gets the fine, so long as the creator of the externality pays it. From these two facts alone, we can conclude that a system which required criminals to pay fines to victims could in principle internalize the social costs of crime.Notice that crime victims suffer a private bad. If you doubt this, observe how people try to avoid becoming another crime statistic: they choose safe routes, don’t travel alone, try to live in nice neighborhoods, and so on. If security were really a public good, there would be no incentives at all for private parties to protect themselves. Thus, there are no effective steps I individually could take to provide national defense for myself. But there are many ways that I individually can take to protect myself against crime. Where is the public good? As far as I can tell, the only public good is prosecution of crimes. We all benefit if someone prosecutes, but the prosecutor bears all of the costs. (Since he is unlikely to be repeatedly victimized by the same criminal.) Yet imagine that the property rules were different — in particular, imagine that crime victims got restitution. Why couldn’t this internalize the social benefits of prosecution?It might be objected that people prefer to live in a society with a low crime rate, even if they are not themselves crime victims. That is quite, but so what? The level of the restitution could incorporate this, too. Tortfeasors don’t need to compensate every person who they put at (some minuscule) risk, only the people who suffered harm. But by requiring compensation for the victims of torts, we reduce the average risk of suffering a civil wrong — and with much lower transaction costs. Similarly, criminals don’t need to compensate everyone who ever heard of their dirty deeds, only their actual victims. But by requiring compensation for the victims of crimes, we reduce the average risk of suffering a criminal wrong — again with much lower transaction costs.One sensible complaint about criminal restitution is that the compensation is imperfect. The probability of apprehension and conviction is less than one,; even if they were, there are some crimes that money can’t undo. True, but we have the same problem in torts. There is no price that can compensate for wrongful death, or probably even the loss of a limb. But isn’t it preferable to do our best to make victims whole rather than do nothing? No system is perfect; but compensation to victims for torts helps somewhat. Is imperfect compensation for crimes really more common than for torts? Many crimes, such as theft and fraud, could probably be compensated easily; many torts, like wrongful death, cannot be. So perhaps the imperfect compensation problem is no worse for crimes than for torts. In any case, we can adjust for imperfect enforcement by multiplying the damages. On top of this, one might argue that compensation systems are better for victims on average, because the high burden of proof and procedural safeguards of criminal trials let many guilty criminals get off. Perhaps the severity of criminal sanctions leads to such low probabilities of conviction that victims in general are worse off than if the sanction were milder. In short, correcting for imperfect compensation with criminal sanctions may backfire by reducing the total level of deterrence. (More on this later.)The lack of restitution for crime victims creates a tragedy of the commons. While the upkeep of common areas is in a sense a public good, it is only a public good so long as the area stays common. If we divide up the commons and create property rights, the area’s preservation turns into a private good. There is an especially simple way to divide up the commons in the case of crimes: let each victim have a property right to some amount of restitution from the criminal who wronged him. Of course, privatization is only one possible way to handle commons problems. But since this paper investigates the prospects of non-state law