The Problem of Political Authority   —   Part II: Society without Authority

11. Criminal Justice and Dispute Resolution

  1. The integrity of arbitrators
  2. Corporate manipulation
  3. Refusing arbitration
  4. Why obey arbitrators?
  5. The source of law
  6. Punishment and restitution
  7. Uncompensable crimes
  8. Excess restitution
  9. The quality of law and justice
    under a central authority
    1. Wrongful convictions
    2. Oversupply of law
    3. The price of justice
    4. The failure of imprisonment
    5. Reform or anarchy?
  10. Conclusion

Libertarian anarchists envision a system in which disputes between individuals are resolved peacefully through the mediation of wise and fair-minded private arbitrators. Is this wishful thinking? In the present chapter, I review several questions and objections concerning this picture of justice in the anarchist society.

11.1 The integrity of arbitrators

What mechanism will keep arbitrators honest and impartial? We can best address this question by first considering in more detail what makes arbitration a viable dispute resolution mechanism to begin with. If two parties have a dispute that they cannot resolve by direct discussion with one another, they may nevertheless be able to agree upon a general procedure for resolving their dispute. This depends upon a contingent but robust fact about human beings in a wide range of cultures: that appeal to a neutral third party is widely perceived as a fair and reasonable dispute resolution mechanism.

But how is it that two parties who disagree about some practical matter are able to agree upon a third party to resolve the dispute? Why isn't the first dispute simply replaced with a second dispute about whom to appeal to to resolve the first dispute? Again, this depends upon a contingent but robust fact about human normative perceptions: people tend to agree to a large extent on who constitutes an impartial judge.

But why would both parties to a dispute seek an impartial judge instead of each insisting on a judge biased in his own favor, such as a personal friend or family member? The reason is that they are attempting to reach a peaceful resolution of the original dispute. The fundamental idea behind arbitration as a strategy for reaching such a resolution is that the parties seek something that they can agree upon that might be used to generate a solution to the original dispute. Given that goal, it makes sense for both parties to choose an arbitrator who is generally viewed in their society as fair. They should not each propose an arbitrator obviously biased in their own favor, since that would not be a viable strategy for generating the needed point of agreement. Of course, if the two parties do not both desire a peaceful resolution of their dispute, then they may simply fight it out; there is no need to propose a biased or corrupt arbitrator in that case.

Based on this understanding of the logic of arbitration as a solution to conflict, an arbitrator has one critical asset: his reputation for honesty, impartiality, and wisdom. That reputation is the central determinant of the perceived quality of his product, and only if he jealously guards that asset can he expect that contentious parties, frequently unable to agree upon anything else, will be able to agree upon him as the person to resolve their disputes. If an arbitrator acquires a reputation for corruption, bias, or capricious decision making, his business will quickly disintegrate. An arbitration company, therefore, would need to be careful in its choice of arbiters, knowing that a corrupt judge could ruin the business.

In many cases, it may be that no matter how a dispute is resolved, one party or the other will regard the decision as unfair after the fact. The best that an arbitrator can do in such a case is to render a decision that will be perceived as fair by most third party observers. It is the perception of such observers that will determine how well the arbitrator's reputation is maintained and thus how much business he can expect to attract in the future. Admittedly, public perception is an imperfect guide to justice, as the public might misunderstand a case or have incorrect values. Nevertheless, the reputational mechanism provides incentives for arbitrators to uphold justice at least approximately in most cases.

In the present system, by contrast, mechanisms for insuring the integrity of judges are much weaker. Judges' decisions are reviewed only by other judges, with the exception of Supreme Court members, whose decisions are reviewed by no one. If the judicial system acquires a reputation for unfairness, inefficiency, and so on, its members can nevertheless retain their positions without fear of being supplanted by the competition.

11.2 Corporate manipulation

Why won't corporations manipulate the system by requiring employees or customers to sign an agreement to have all disputes settled by an arbitrator biased in the company's favor, such as an arbitrator in the permanent employ of the corporation itself?

Figure 11.1 A standard price theory diagram shows the competitive market price of a good
at the intersection of the supply curve, as determined by marginal costs of production,
and the demand curve, as determined by marginal utility of consumption.

Here is a more fundamental question: why don't businesses make unlimited demands on employees and customers? Why not require customers to give the company all the money they have? Why not require employees to work for free? These arrangements would certainly be more favorable to the company than the sort of arrangements businesses actually offer.

To understand why businesses do not behave in these ways, we should first consider how market prices are determined. For any given business, there is some optimum level at which the business should set its prices to maximize its profit. If it sets prices below that level, the company will lower its total profits due to lower profit per unit. If it sets its prices above that level, the company will lower its total profits due to lower volume of product sold. A precise account of the optimum price level is given in standard price theory, where this price is said to lie at the intersection of supply and demand curves (see Figure 11.1).[1] For our purposes, the important point is simply that market forces determine an optimal price level such that the company does worse for itself if it exceeds that level.

Now the plan of making unreasonable legal demands is essentially equivalent to a plan to increase the price of one's product. Suppose that Sally's widget business requires all customers to agree that, in case of any dispute arising in connection with the sale of a Sally widget, including complaints regarding product quality or safety, the customer will accept binding arbitration by Sally's nine-year-old daughter, Susan. Sally's Widgets is then in effect adding to the price of Sally's widgets: in addition to the $150 that one must pay for a widget, the customer must also accept the risk of having a dispute with the company resolved by the owner's daughter. Clients might consider this undesirable.[2] They might even take the policy as a signal that the company intends to cheat its customers. For this reason, if $150 was the market price for widgets, then Sally's addition of the unreasonable stipulation in regard to the resolution of disputes with her company will have the effect of placing the real price of her product above the market level and thus lowering Sally's total profit.

What if the market price for widgets is $200 and Sally charges only $150, leaving Sally some leeway to make additional demands on customers? Even in this case, insisting that all disputes should be resolved via Susan is not Sally's best option for taking advantage of that leeway. The reason is that customers are likely to place a greater negative value on Sally's dispute resolution procedure than the positive value that Sally places on it, because customers tend to place negative value on perceived unfairness in addition to the potential monetary costs of unfair procedures. Instead, Sally's best (profit-maximizing) option is simply to raise her price by $50.

The same principles apply to employer-employee relationships. There is an optimal wage for an employer to pay such that, if the employer pays more than that, he lowers his total profits due to increased labor costs, but if he pays less, he lowers his total profits due to difficulty in attracting desirable employees. Any provision in an employment contract that employees regard as unfair or simply disadvantageous amounts to an extra cost of accepting a job with this employer or, equivalently, a decline in the rewards of the job. Aprovision that would naturally be taken as signaling an intention to cheat one's employees would normally lower the attractiveness of any job too much to be worth inserting. If an employer feels that he is giving too much to employees, it would make more sense for him simply to offer lower wages.

Empirically, businesses in free market economies rarely take unreasonable positions in disputes with customers. The following is not an uncommon sort of consumer experience: I buy a product from the local Target, take it home, cut off the packaging, and then decide that I don't like it. I go back to the store and ask for my money back. 'Is there anything wrong with it?' the cashier asks. 'Nope', I say, 'I just decided I don't like it. So take it back.' My position in this dispute, if one can call it that, is utterly capricious. I voluntarily bought the product, I know that they can't resell it after I have opened it, and I have no real complaint about the product. The product is not defective, nor was it misrepresented either by the manufacturer or by the store. I have no argument for why they should take it back. Yet in my experience, the company has never refused a return.

This evidence about the behavior of businesses is of course anecdotal, and certainly others could relate anecdotes of unsatisfactory experiences. Nevertheless, I do not think it an unfair illustration of the overall tendency of the market: consumers are much more likely to take unreasonable positions - and prevail - than the businesses they patronize.

11.3 Refusing arbitration

We have discussed the reasons for accepting arbitration as a mechanism of dispute resolution. But what if, in a particular case, you have strong reason to believe that any reputable arbitrator will find against you? This could be true for any of a number of reasons, including that you have in fact violated someone else's rights and are attempting to get away with it; that you are out of step with the values of the majority of your society, so that what you consider acceptable behavior a typical arbitrator will not; or that there is a large amount of misleading evidence that indicates that you are guilty of some crime of which you are in fact innocent. In any of these cases, it may seem that you would be best advised to reject arbitration.

But even in these cases, you will probably be forced to accept arbitration. If you reject the option of having your dispute arbitrated, your security agency will probably draw the reasonable inference that you are in the wrong according to prevailing norms, since the most likely explanation for your rejection of arbitration is that you expect any reputable arbitrator to decide against you. For the same reasons that protection agencies will not defend criminals (Section 10.4), they will not defend people who reject arbitration as a means of dispute resolution. Security agencies will anticipate this eventuality, writing provisions into their contracts specifying the procedures that customers must accept for resolving disputes and absolving the company of the responsibility to protect clients who violate these procedures.

In some cases, this system would generate unjust or ethically objectionable results, as in the case where strong evidence points to the guilt of someone who is in fact innocent, or where the values of the majority of society are wrong. But the anarchist system nevertheless does as well as could reasonably be asked. In any functioning justice system, whether government-based or market-based, if powerful but misleading evidence points to someone's guilt, then that person will be treated as guilty. Only the unattainable standard of absolutely conclusive proof of guilt could eliminate the possibility of misleading evidence leading to punishment of the innocent.

Likewise, every social system generates unethical outcomes if the people who make decisions in that system have incorrect ethical beliefs and values. Under anarchy, unethical outcomes result if most members of society have incorrect values, which will be reflected in the decisions of arbitrators seeking to cultivate a good reputation with the public. In a government-based system, unethical outcomes result if legislators, judges, or other public officials have incorrect values. This is not less likely to be true than that the majority of society's members have incorrect values.

11.4 Why obey arbitrators?

In the event that you have a dispute with another member of an anarchist society, why should you not agree to try arbitration to resolve the dispute, hoping that the arbitrator will side with you, and then simply ignore the arbitrator's decision if it goes against you?

This sort of behavior, if anything, would be even less tolerated by the rest of society than a refusal to accept arbitration to begin with. For the same reasons that security agencies would not agree to defend criminals, you could expect your security agency to leave you to fend for yourself if you violated an arbitration decision.

Beyond that, arbitration companies could maintain lists of individuals who had violated an arbitration agreement. There might be criminal-record-reporting agencies, functioning analogously to credit-reporting agencies, providing reports of past criminal activity for a nominal fee. Given knowledge of your past violation of an arbitration agreement, it would not be rational for others in the future to enter into business relationships in which you might attempt to cheat them and then refuse to pay compensation. It might therefore become very difficult to find a job, get a credit card, take a bank loan, rent an apartment, and so on.

11.5 The source of law

In the status quo, the decisions of judges and juries are based largely on laws written by legislators or bureaucrats working for regulatory agencies. Since the anarcho-capitalist society contains neither legislators nor regulators, on what basis could arbitrators make their decisions?

There would be two sources of law in the anarchic society. First, property owners or local associations of property owners could specify the body of law to govern interactions occurring on their property. Provided that all who entered the property were given fair notice of the legal code in effect there, arbitrators would most likely honor the owner's choice of law. Legal scholars might develop suggested standardized legal codes, with business owners, landlords, or homeowners' associations choosing which of several widely used legal codes should hold sway on their land. Consumers with strong objections to a particular legal code would avoid patronizing businesses that adopted that code. In choosing a home, individuals would weigh the advantages of the legal code subscribed to by the local homeowners' association.

The other major source of law would be the arbitrators themselves. When the solution to a particular dispute was not determined by any law of the kind described in the foregoing paragraph, the judge would look to similar cases for guidance, attempting to apply the same principles in the present case that had generally been used to decide cases of this kind in the past. If the case before him had novel features, the judge would exercise his own judgment to devise a resolution that seemed fair and in keeping with the generally accepted values of his society. He would then write an explanation for his decision, which would be added to the body of precedent for other judges to consult in future cases. It makes sense for arbitrators to follow this tradition, since it usually results in decisions that most observers regard as fair, and it preserves the sort of consistency that most observers value in a legal system.

This bottom-up approach to generating law has three key advantages over the top-down approach of law created by a legislature. First, judge-made law is more closely tied to the problems that ordinary people encounter and their actual circumstances, because it is made by individuals with regular experience in resolving interpersonal disputes - the problems that give rise to the need for law in the first place - and it is made only in the context of deciding such disputes. Second, judge- made law is more flexible than statutory law. No rule of conduct that human beings devise can foresee all possible future contingencies. In a common-law system, when a court encounters a case of a kind not previously considered, it can decide that case in the manner that seems most fair rather than being forced by earlier oversights to render unjust rulings. Third, the common-law system makes far smaller cognitive demands on the individual lawmaker. A legislature faces the nearly impossible task of anticipating every issue that might occur in every area of human conduct and writing rules valid for all circumstances. Ajudge in a common-law system faces at any given time only the task of understanding the case now before him and deciding how that case should be resolved; at no time need a judge or anyone else attempt to anticipate every possible type of problem.

We know that this is a viable way of developing an extremely sophisticated and subtle system of law, because this is in actual fact the source of the common law that now holds sway (alongside statutory and regulatory law) in Great Britain and several other countries influenced by Great Britain, such as the United States, Canada, Australia, and New Zealand. In these countries, most of contract law and tort law is judge-made common law. Most of the criminal law was also common law prior to the twentieth century. In the anarchist society, given the absence of statutory and regulatory law, common law would play an even greater role than it does in these countries today.[3]

11.6 Punishment and restitution

Existing government-based criminal justice systems rely on imprisonment of criminals as a response to crime. It is thought that society as a whole benefits from this practice because it keeps criminals off the streets for a time and deters others from entering a life of crime. The victims of a particular crime, however, generally receive nothing in the way of compensation, and the rest of society is forced to pay for criminals' upkeep during their terms of imprisonment.

The anarcho-capitalist justice system would most likely focus on restitution rather than punishment. That is, criminals would be forced to pay compensation to their victims. This system would be preferred over punishment-based systems because it is better for the crime victims and it does not require anyone to pay for the criminals' upkeep. The required compensation would most likely include compensation for the inconvenience and lost time suffered by the victim in attempting to secure justice, as well as reasonable costs incurred by the victim's protection agency in identifying, apprehending, and prosecuting the criminal. As a result, a thief, for example, would have to pay back significantly more than the value of what he stole. This would provide a deterrent to crime.

What if the victim of a crime was dead (whether killed by the criminal or killed by other causes after the crime) and thus unable to collect compensation? In this case, the victim's family or friends might collect the owed compensation. Alternately, individuals might, in advance, authorize their protection agencies to collect compensation on their behalf in the event that they were unable to receive compensation for a crime. The compensation that a crime victim is owed can be thought of as property of the victim, which he therefore has the right to give, sell, or bequeath to someone else. Granting one's protection agency the right to collect compensation in the event that one is murdered might serve to deter potential murderers.

11.7 Uncompensable crimes

What would happen if a criminal lacked the funds to compensate his victims? One possibility is that the criminal would be remanded to a private prison where he would be required to work off his debt.

But what if the criminal could not work off his debt? Imagine, say, a criminal con artist who has defrauded his victims of $20 million, almost all of which has been spent. The criminal has no realistic hope of ever paying his victims back. What would be done with this criminal? One possibility is that the criminal might be housed indefinitely in a prison-labor facility, to pay as much of his debt as possible. Or the victims might settle for some partial repayment, such as the criminal could realistically make within his lifetime. It would be up to the arbitrator in the case, in consultation with the victims, to decide upon the most appropriate remedy. In any case, the information as to what the criminal had done would most likely be made publicly available and possibly sent to criminal-record-reporting agencies so that future landlords, employers, and so on could be on guard.

In some cases, however, a criminal's behavior is so heinous that not only is it impossible for him to compensate his victims but the criminal can never be safely released. Imagine, for example, that a protection agency has taken Ted Bundy into custody. Bundy protests his innocence, but the arbitration company finds him responsible for at least thirty murders. Bundy will never compensate his victims, and if he is ever released, he will kill again. There would seem to be two options: he can be imprisoned indefinitely (probably in a forced-labor facility), or he can be executed. Again, it would be up to the arbitrator in the case to determine the best course of action. As in the case of the real Ted Bundy, execution appears a likely possibility.

11.8 Excess restitution

The victim of a crime is justly entitled to collect full compensation for the crime; that is, sufficient compensation to return him to the welfare level he would have enjoyed if the crime had not occurred. But what if a particular court regularly awarded excess compensation - say, twice what the victim was justly entitled to and twice what other courts generally awarded for a given crime? Wouldn't the excess compensation court be favored by victims? And since almost everyone considers himself more likely to become a crime victim than to become a criminal, almost everyone would want any future disputes of theirs to be resolved by such a court. Taking account of this, protection agencies would agree to use courts that provide excess compensation. Soon, almost all criminal cases would be tried in courts of this kind. Criminals could protest at the injustice, but their voices would be little heeded, since protection agencies and arbitration firms would be more keen to satisfy the overwhelming majority of law-abiding customers than to satisfy the criminals.

What is problematic about this result? The obvious problem is that this situation is an injustice, albeit one over which we may find it difficult to rouse much indignation - it is an injustice to the criminals. But Paul Birch argues that the problem would go deeper than this, undermining the entire anarcho-capitalist system.[4] Once the practice of awarding excess compensation started, firms would compete to offer higher and higher awards to victims, perhaps ten times, twenty times, or even fifty times the amount to which the crime victim was justly entitled. These excessive awards would create powerful deterrents to crime, resulting in a dramatic drop in the crime rate. While this may sound like a happy result, it would put increasing financial pressure on arbitration firms. As the crime rate dropped, arbitration firms would continue to raise their compensation awards in the effort to collect a larger share of the dwindling market. This would only cause the market to shrink further. Eventually, either all firms would go out of business, in which case society would devolve into a state of chaos, or the last firm able to hold out would acquire a monopoly on the industry, whereupon it would evolve into a state.

There are several reasons why the foregoing scenario is unlikely to transpire:

  1. The argument unrealistically assumes that actual and potential crime victims favor unlimited compensation. This assumption may be driven by a conception of human beings as homo economicus, pure profit maximizers: since higher compensation equals higher profit, crime victims will favor unlimited increases in compensation. Normal human beings, however, do not see criminal victimization as an opportunity to get rich; that sort of thinking is generally reserved for scam artists. Most normal people wish to avoid being crime victims, if possible, and to secure justice in the event that they are victimized.

    A more plausible concern is that crime victims will be motivated by vengefulness, rather than profit seeking, to push for excessive sanctions on their malefactors. Surprisingly, this concern is undermined by empirical evidence: surveys of attitudes toward criminal sentencing have found that crime victims in fact harbor attitudes no more punitive than those of the average member of the population.[5]

  2. Birch imagines arbitration companies advertising that they award excessive compensation - announcing, for example, that they award each victim compensation equal to ten times the loss the victim has suffered. This is very close to a court's explicitly announcing that it is unjust. It is difficult to imagine this occurring. For reasons discussed earlier, arbitration companies would carefully select their judges and guard their reputations for fairness, impartiality, and wisdom. The sort of people who would wind up as judges would be unlikely to explicitly and intentionally promote injustice for the sake of profit maximization.

    A more realistic concern is that arbitration companies would be biased in favor of victims rather than explicitly embracing injustice. They would almost certainly claim to be administering justice, but their perceptions of what justice demands might be slanted in favor of victims; for instance, they might tend to perceive most crimes as more damaging than they really are. It is plausible that arbitration companies could hire judges with such slanted perceptions without unduly tainting their reputation for integrity. I therefore think it plausible that in an anarcho-capitalist society, criminals would often suffer somewhat more than they deserved.

    This is a possible problem with the system, but it is not a terrible problem. Moreover, it is plausible that overpunishment occurs also in governmental systems, and it is not obvious that governmental systems deliver more just punishments than those that would emerge from an anarcho-capitalist system.

  3. Apart from their concern for the rights of criminals, which admittedly is limited, there is another reason for ordinary individuals to oppose absurdly excessive compensation for crimes: in any realistic criminal justice system, innocent people are sometimes convicted. Most people find this prospect troubling even in the abstract and perhaps more so when they reflect that they themselves or someone they are close to may one day be among the wrongly convicted. The problem cannot be eliminated without entirely dispensing with the criminal justice system; however, most people would find the problem much less troubling if the penalties for crimes were reasonable than if they were absurdly excessive. This would lead most people in the anarchist system, just as in the present system, to support some degree of restraint on the part of judges in the process of assigning compensation awards.
  4. Excessive compensation awards tend to be more difficult or expensive to collect. If, for example, the compensation for the theft of a video-game cartridge was $100,000, this might prove difficult to enforce. If a shoplifter could expect to be imprisoned for life in a compulsory labor facility if caught, then shoplifters might be willing to kill to escape or to fight to the death rather than surrendering. Knowing this, security agencies would have a reason to favor reasonable compensation awards.
  5. A criminal who is wronged by a clearly excessive compensation award would seem to have a valid complaint against the court that made the unjust award. There is no obvious reason why he could not file a lawsuit against that court in a different court.

    If all the courts had the same excessive standards for compensation, then the criminal's suit would fail. But if the courts generally started out with approximately just standards and one court decided to seek a larger market share by offering excessive compensation awards, then that court would suffer for its indiscretion as other courts found its judgments unjust and awarded compensation to those who had been wronged by the court. Thus, if the system starts out in a generally just position, it will be stable.

  6. Even extreme increases in the penalties for crime would not eliminate all crime. This is because some criminals, unfortunately, are highly resistant to deterrence. They recklessly ignore the future or blithely assume that they won't get caught.[6] Thus, a market for private courts would continue to exist even in a regime of absurdly high compensation awards.
  7. Even if excessive compensation awards resulted in a dramatic drop in crime rates, this would not cause all or nearly all arbitration firms to go bankrupt. However much crime might drop, honest disputes among ordinary people would continue to arise, and they would still need to be adjudicated by arbitration firms. If crime suffered a precipitous drop, arbitration companies would experience a decline in revenues and would need to scale back operations to the point that the market would support. But this would not lead all of them to go bankrupt, nor would it cause the industry to be monopolized.

    Consider an analogy. As automobiles became more practical in the early twentieth century, the demand for horses suffered a drastic drop. But the entire industry did not collapse, nor was it monopolized - there remains more than one horse breeder in the world today. The industry simply shrank to the size that could be supported by the new levels of demand. Likewise, if we should be so blessed as to find ourselves worrying about unduly low crime levels, the arbitration industry will shrink so that it includes only the number of courts needed to satisfy however much demand remains.

11.9 The quality of law and justice under a central authority

To better assess the merits of a nongovernmental justice system, we must first consider some of the flaws of the present system.

11.9.1 Wrongful convictions

One disturbing aspect of the present system is the rate at which the innocent are punished. Michigan law professor Samuel Gross studied cases in which convicts were exonerated in the United States between 1989 and 2003.[7] He found 340 such cases, including 205 murder cases, 121 rape cases, and 14 cases involving other crimes. Prosecutors and police often refuse to accept that they arrested and prosecuted an innocent person, even after proof of the person's innocence has been uncovered.[8] On average, these defendants suffered eleven years of wrongful imprisonment before finally being officially exonerated.

Why were murder and rape so overrepresented among the crimes of which defendants were exonerated? The main reason for the dramatic overrepresentation of rape cases lies in the development of DNAtesting in the late 1980s and thereafter, which led to the reexamination of a number of rape cases in which semen samples had fortunately been preserved. Application of the new techniques revealed that many convictions prior to the advent of reliable DNAtesting were erroneous. The main reason for the overrepresentation of murder cases seems to lie in the much greater scrutiny that such cases receive as compared to less serious cases, especially when the death penalty is involved.[9]

Omitted from Gross's statistics are cases of mass exonerations due to the exposure of large-scale police corruption. One such case involved the CRASH ('Community Resources against Street Hoodlums') program of the Los Angeles Police Department. In 1999, Officer Rafael Perez revealed that he and other officers in the program had routinely lied in arrest reports, shot unarmed suspects and innocent bystanders, planted guns on suspects after shooting them, fabricated evidence, and framed innocent defendants. In the wake of these revelations, over 100 defendants had their convictions vacated in 1999 and 2000.[10]

Why were the defendants in Gross's sample wrongly convicted? Most cases involved witness misidentification. Many involved perjury by police, forensic scientists testifying for the government, the real criminal, jailhouse snitches, or others who stood to gain by providing false testimony. In 15 percent of the cases, the defendants, under the stress of high-pressure police interrogations, actually confessed to crimes they had not committed. Most of those 15 percent were under the age of 18, mentally retarded, or mentally ill.

Since the defendants in these cases were ultimately exonerated, may we rest easy that the system works and that justice is served? There are two reasons for rejecting such complacency: first, there are the eleven years that these defendants, on average, were forced to spend in what may be the worst conditions that any significant segment of society must live under. Second, and more importantly, there are the implications for the number of people who continue to be unjustly imprisoned.

There are no reliable estimates of the frequency of wrongful convictions, due to the inherent elusiveness of such cases. Though it is reasonably clear that all or nearly all of Gross's 340 cases were indeed wrongful convictions, we have no way of knowing how many additional erroneous convictions went undiscovered during the same time period. The 74 death row inmates who were exonerated constituted about 2 percent of the death row population.[11] This suggests that if we applied the same level of scrutiny to all cases that we apply to death penalty cases, we might find a 2 percent false positive rate in these other cases as well.

But we have no idea how many death penalty cases there were in which erroneous convictions went undiscovered. The wrongful convictions in Gross' sample were due mainly to witness error, perjury, and false confessions. But when a witness misidentifies a suspect, a witness lies on the stand, or the police extort a false confession, in how many of such cases can we assume that proof of the defendant's innocence, elusive at the time of trial, will later luckily appear and rescue him from prison? Proof of innocence is not generally very easy to come by, and the authorities, having closed the case, will not be looking for any such evidence. The defendant himself will have difficulty uncovering such evidence from his position in prison. For these reasons, it would seem overly optimistic to assume that in the majority of wrongful convictions (even in death penalty cases), proof of innocence is later discovered. It therefore seems probable that the actual false conviction rate is much greater than the 2 percent exoneration rate that Gross found among death penalty cases.

Could anything be done to improve the system, or are these mistaken convictions simply the price of criminal justice? Several measures have been suggested to improve the reliability of the system: reducing the use of high-pressure interrogation techniques, particularly for underage or mentally disabled suspects; having witnesses questioned by officers who do not know the details of the investigation and therefore cannot influence the witnesses; showing witnesses one suspect at a time rather than a group of suspects all at once; and instructing juries on the limitations of eyewitness evidence. Despite studies indicating that these measures would reduce the risk of wrongful convictions, American police and courts have generally not adopted them.[12]

11.9.2 Oversupply of law

Under a legal system based on a central authority with legislative powers, a great deal more law is provided than under a pure common-law system. Some see that as an advantage - perhaps we need a strong network of regulations to protect us against the failures of laissez-faire capitalism. Nevertheless, it is worth considering whether a governmental system might provide too much law.

As an exercise, try to imagine an ideal legal system. Before reading on, try to estimate how many pages worth of laws that system would contain. There are many difficulties with making such an estimate; nevertheless, attempting at least a vague, order-of-magnitude estimate before finding out how much law actually exists may help to forestall the tendency to rationalize the status quo.

Most citizens in modern states, whether they would describe themselves as supporting a strong regulatory regime or not, have little idea of how much regulation they actually have. In the United States, the rules promulgated by regulatory agencies of the national government are recorded in the Code of Federal Regulations (this does not include statutes passed directly by Congress, nor does it include state or local laws). Over the last half century, the quantity of these regulations has ballooned from about 23,000 pages to about 150,000 pages (see table below).

Year Length of CFR
in pages[13]
1960 22,877
1970 54,834
1980 102,195
1998 134,723
2010 152.456

These statistics cannot capture qualitative information about the content of these regulations, and of course there is no prospect of reviewing any significant fraction of these regulations here (or anywhere). Nevertheless, I suggest that these numbers might prompt even the strongest ideological supporter of regulation to consider whether dedicated lawmaking bodies might have a tendency to provide a greater than optimal quantity of regulation. The reader unfamiliar with regulation is invited to peruse the CFR at random to obtain a qualitative sense of the regulatory regime. One may, for example, chance upon a paragraph describing the spacing of spark plug gaps, another prescribing the use of the expression 'all day protection' in antiperspirant labels, another describing the signing of documents related to excise taxes on structured settlement factoring transactions, and so on.[14]

What is objectionable about such overprovision of law? The first objection is that it represents an excessive reliance on coercion. Each of these regulations is a command backed up by a threat of force issued by the state against its citizens. While some of these threats may be justified, those that are not constitute a violation of the rights of all those who are thereby coerced.

Second, a surplus of laws can have large economic costs. Ronald Coase, Nobel laureate and former editor of the Journal of Law and Economics, reports that his journal published a series of empirical studies of the effects of a wide variety of regulations, in which it turned out that every regulation studied had overall negative effects on society.[15] The Small Business Administration of the U.S. government has estimated the annual cost of federal regulations to the U.S. economy at $1.75 trillion, a burden that they find falls disproportionately on small businesses.[16]

Third, an excessive quantity of law, as well as an excessively complex and technical body of law, renders it unreasonable to demand that citizens know, understand, and follow all laws. To threaten to punish citizens for violation of rules that, in the light of the extreme cognitive burdens, they could not reasonably be expected to know or understand, is a form of injustice. These cognitive burdens at some point defeat the primary purpose of establishing written laws to begin with - namely, that the law should be accessible to all who are expected to follow it.

One solution to the last problem is for citizens to hire experts to advise them in any area in which the law is complex and difficult to follow. This, however, leads us to the next problem with the currently accepted system of justice.

11.9.3 The price of justice

For most citizens of modern states, the costs of government justice in both time and money are prohibitive. The typical civil dispute requires anywhere from several months to a few years to resolve through governmental channels.[17] In 2009, the average American law firm billed $284 per hour, with a typical divorce costing between $15,000 and $30,000. To the average American, with an annual income of $39,000, any use of the government's justice system represents an overwhelming financial burden.[18]

Why are legal services so expensive? One reason is the oversupply of law mentioned above. The complexity, technicality, and sheer length of the laws and legal procedures forces individuals to pay experts to handle any legal procedure, and it forces those experts to expend a great deal of labor on each case. Another reason can be found in the restrictions on the supply of legal services, which by law may only be purchased from government-approved sources (lawyers who have been admitted to the bar, generally after a lengthy and very expensive law school education).[19]

These costs are troubling for at least three salient reasons. First, the high cost of legal services means that only the wealthy can afford justice. Middle-and low-income individuals cannot afford to seek justice or must take justice into their own hands when they believe they have been wronged. In criminal cases, low-income defendants may receive inadequate legal representation due to heavy case loads on public defenders.

Second, even defendants who win their cases, whether the cases be civil or criminal, may be financially ruined. This acts as a kind of unjust punishment imposed on all defendants, whether they are guilty of wrongdoing or not.

Third, large businesses may be able to afford the legal fees necessary to ensure compliance with complex bureaucratic regulations, while the same costs may prove prohibitive for small businesses. As a result, the present legal regime tends to promote concentration of industries in the hands of large corporations, even if those corporations are in themselves less efficient than smaller firms.

11.9.4 The failure of imprisonment

Today's governments rely on imprisonment as a response to serious crimes. Imprisonment serves two main functions: first, it protects society from convicted criminals for a limited time by separating the criminals from the rest of society. Second, it punishes the criminals by forcing them to live in highly undesirable conditions. The suffering on the part of the criminals may be valued intrinsically as a form of retributive justice or it may be valued instrumentally as a means of deterring future criminal behavior.

Existing jails and prisons, however, suffer from a number of very serious problems. In the United States, these facilities are regularly extremely overcrowded, and inmates live in danger of gang violence, rape by other prisoners, beatings by guards and other prisoners, and other forms of abuse. The rate of such violence and abuse is unknown, but anecdotal reports are numerous.[20] In recent years, the use of solitary confinement has become increasingly common, a practice that leads to mental deterioration on the part of the prisoner and higher rates of recidivism once the convict is released.[21]

Under these conditions, incarceration could hardly be expected to rehabilitate criminals. Accordingly, two-thirds of criminals are rearrested within three years of being released from prison.[22] This statistic must be assumed to underestimate the true rate of recidivism, given the low rate at which law enforcement solves crimes (Section 10.7); thus, the great majority of criminals return to a life of crime shortly after their release. Some observers have argued that incarceration not only fails to rehabilitate criminals but actually renders them more dangerous when released than they were when they entered. This may be true, for example, because inmates make new criminal contacts and learn new criminal skills and ideas from other inmates while in prison, because they absorb antisocial values from the other inmates, and because inmates become more angry and resentful as a result of the abuse they suffer while in prison. Some have gone so far as to suggest that incarceration may cause more crime than it prevents.[23]

These problems are not inevitable in a criminal justice system; critics have offered numerous potential reforms that would seem likely to significantly reduce these problems. Some rehabilitation programs have been found to reduce recidivism rates by up to 30 percent. Policymakers simply have not chosen to adopt these reforms.[24]

11.9.5 Reform or anarchy?

The problems listed in the preceding subsections are only the most prominent of those afflicting the present government-based system of justice. Asanguine observer, however, while acknowledging the seriousness of the problems, might take them to show merely that the justice system ought to be substantially reformed while still remaining in the hands of government.

Indeed, there are a number of measures that would greatly mitigate the problems listed above, and we cannot rule out the possibility that government officials will one day begin a serious reform of the prison and court systems. Nevertheless, it is no mere accident that problems of the sort we have been discussing are found to persist in government-based justice systems. Coercive monopolies have a systematic tendency to foster a variety of problems, and they tend to be slow to recognize and address their own shortcomings.

The reasons are familiar. Because government collects its revenues in the form of taxes which citizens have no choice but to pay, government programs can survive financially even with extreme levels of consumer dissatisfaction. More importantly, because government is monopolistic, citizens have nowhere else to turn if they find its services inefficient, of low quality, or abusive. Most of the problems with America's justice system are obvious and have been well known for a very long time. National and state governments have done little to address these problems, not because the problems are difficult or impossible to address, but because the government suffers no negative consequences as a result of its failure to address them.

Consider the problem of wrongful convictions. In a competitive system, a local homeowner's association could choose from among many protection agencies, arbitration firms, and bodies of law to apply to its neighborhood and could alter its choice if and when it became dissatisfied with the security and justice arrangements. Furthermore, residents dissatisfied with their HOA's decisions could relocate at relatively little cost. Since no one wants to be wrongly convicted, a protection agency that used unreliable methods of investigation or an arbitration firm that used unreliable methods of assessing guilt or innocence would have to worry about being supplanted by competitors who offered services with less risk of wrongful convictions. Similar points apply to the problem of oversupply of law and excessive costs of legal services.

What about the problems associated with incarceration of criminals? These would be greatly reduced by a justice system that focused on restitution rather than punishment. In such a system, the hundreds of thousands of people presently incarcerated for victimless crimes, chiefly drug-related offenses, would be free. Only individuals who had harmed another person and were otherwise unwilling or unable to pay the required compensation to the victim would be held captive in prison-labor facilities. These facilities' focus on productive work would diminish the risk of in-prison violence as well as recidivism. It is theoretically possible for a government to reform itself - to eliminate all victimless crime statutes, shift its focus from punishment to restitution, and so on. But when we look around and see that no government has in fact done so and when we notice that this kind of unresponsiveness to problems has a systemic explanation rooted in the basic incentive structure of government, conversion to an alternative system begins to seem like a more rational and less utopian solution than that of reforming the present system. There will always be room for improvement in any justice system. In governmental systems, reform will tend to be slow and difficult to implement. By contrast, businesses in a competitive industry tend to move quickly to improve their products or reduce their costs when the opportunity presents itself.

11.10 Conclusion

There are two main systems by which a society may provide for the resolution of disputes and the remedying of rights violations. The first is the coercive, monopolistic system, in which a single organization assumes exclusive authority for making laws, resolving disputes, and punishing criminals. Large and well-known problems tend to occur in systems of this kind, including frequent erroneous convictions, excessive and excessively complex legal restrictions, high monetary costs, long delays, overcrowded prisons, abuse of prisoners, and high rates of recidivism. Governments on the whole have done little to address these problems, despite the identification by social scientists and other experts of numerous steps that could be taken to greatly improve the system. This neglectfulness on the part of government can be traced to the defining characteristics of this approach to justice; namely, its coercive and monopolistic character. Because the governmental system is funded through compulsory taxation, courts, prisons, and other elements of the justice system can continue to collect as much revenue as the government wishes to allocate, regardless of consumer satisfaction. Because the government holds an effective monopoly on the provision of justice, these organizations need not fear replacement by competitors, regardless of their performance.

The alternative is a market-based system of justice, in which arbitration companies compete with one another in the resolution of disputes. When one individual violated the rights of another, an arbitrator would decide upon the compensation to be paid by the criminal to the victim. In cases in which a criminal had no other means of making payment, the criminal would be housed in a private prison, where he would be required to work off his debt. Individual property owners or associations of property owners, such as homeowners' associations, would choose the body of law to apply to interactions occurring on their property. Any issues not resolved by such laws would be dealt with through a form of law devised by the arbitrators, similar to the British common law in the actual world.

To attract customers, arbitrators in the free market justice system would seek to maintain a reputation for fairness, consistency, impartiality, and wisdom. Security companies would most likely require their customers to resolve any disputes through reputable third-party arbitrators and would refuse to defend customers who either rejected arbitration or violated the arbitrator's decision after submitting a dispute to arbitration.

In this system, arbitrators might evince a bias in favor of crime victims and against criminals, so that criminals might be forced to pay somewhat higher amounts in compensation for their crimes than justice truly demanded. However, it is far from clear that this problem would be more severe than the overpunishment that occurs in government-based systems, which focus on incarceration of criminals in oppressive and dangerous conditions. The problem of excess compensation awards would likely be a relatively modest and tolerable problem in comparison with the problems of the status quo.


1 See Friedman 1990 for an accessible account of the standard theory.

2 For a different perspective, see Caplan (2010), who suggests that most customers would be unconcerned by such a clause, since they do not expect ever to sue the company.

3 See Barnett 1998 for a more thorough account of nongovernmental legal systems. 4 Birch 1998.

5 Walker and Hough 1988, 10; Hough and Moxon 1988, 137, 143-6.

6 Banfield 1977.

7 Gross et al. 2005.

8 Gross et al. 2005, 525-6.

9 Gross et al. 2005, 531-2, 535-6. Gross et al. (532-3) point out that there may also be more pressure to convict someone in capital cases, leading to more mistakes. However, there may also be greater care exerted by defense attorneys, judges, and juries in cases where extremely severe punishments are at issue.

10 Gross et al. 2005, 533-4.

11 Gross et al. 2005, 532, n. 21.

12 Duke 2006.

13 Figures for 1970 and 1998 are from Longley n.d. Figures for 1960 and 1980 are from Crews 2011, 15. The figure for 2010 is computed from the edition of the CFR available from the Government Printing Office,; I have omitted the 'Finding Aids' at the end of each volume from the total page count for 2010.

14 40 CFR, Appendix I to subpart V of part 85 (H)(1)(b); 21 CFR 350.50(b)(3); 26 CFR 157.6061.

15 Hazlett 1997, 43.

16 Crain and Crain 2010.

17 In the United States, delays vary from about six months to about three years, with an average of eleven months (Dakolias 1999, 18).

18 On average lawyer fees, see California Attorney's Fees 2011, quoting a survey by Incisive Legal Intelligence. On the price of divorce, see Hoffman 2006. On average incomes, see U.S. Census Bureau 2011b, 443, table 678. 19 The price of law school often exceeds $100,000. Only seven U.S. states allow individuals to take the bar exam without attending law school (Macdonald 2003).

20 Commission on Safety and Abuse in America's Prisons 2006, 11-12, 24. AU.S. Department of Justice survey found that 4.4 percent of prison inmates and 3.1 percent of jail inmates suffered some form of sexual abuse in the preceding year (Beck et al. 2010); however, such incidents may be underreported.

21 Commission on Safety and Abuse in America's Prisons 2006, 14-15.

22 Commission on Safety and Abuse in America's Prisons 2006, 106.

23 Pritikin 2008.

24 Pritikin 2008, 1092; Commission on Safety and Abuse in America's Prisons 2006, 12, 28, 108.

Previous Chapter       Next Chapter

The Problem of Political Authority

An Examination of the Right to Coerce and the Duty to Obey

Michael Huemer




Ancap FAQ

Library of Liberty