Economists approach the analysis of crime with one simple assumption—that criminals are rational. A mugger is a mugger for the same reason I am an economist—because it is the most attractive alternative available to him. The decision to commit a crime, like any other economic decision, can be analyzed as a choice among alternative combinations of costs and benefits.

Consider, as a simple example, a point that sometimes comes up in discussions of gun control. Opponents of private ownership of handguns argue that, in violent contests between criminals and victims, the criminals usually win. A professional criminal, after all, has far more reason to learn how to use a gun than a random potential victim.

The argument is probably true, but the conclusion—that permitting both criminals and victims to have guns will help the criminals—does not follow. To see why, imagine that the result of legal handgun ownership is that one little old lady in ten chooses to carry a pistol in her purse. Further suppose that, of those who do, only one in ten, if mugged, succeeds in killing the mugger; the other nine miss, drop the gun, or shoot themselves in the foot.

On average, the muggers are winning. But also on average, each one hundred muggings of little old ladies produce one dead mugger. Very few little old ladies carry enough money to be worth one chance in a hundred of being killed. Economic theory suggests that the number of muggings will decrease—not because the muggers have all been killed, but because some of them have chosen to switch to safer professions.

If the idea that muggers are rational profit maximizers seems implausible, consider who gets mugged. If a mugger’s objective is to express machismo, to prove what a heman he is, there is very little point in mugging little old ladies. If the objective is to get money at as low a cost as possible, there is much to be said for picking the most defenseless victims they can find. In the real world, little old ladies get mugged a lot more often than football players do. Following out this line of argument, John Lott and David Mustard, in a controversial article, found that laws making it easier to get permission to legally carry a concealed firearm tended to reduce crime rates.

This is one example of a very general implication of the economic analysis of conflict. To stop someone from doing something that injures you, whether robbing your house or polluting your air, it is not necessary to make it impossible for him to do it—merely unprofitable.

Economic analysis can also be used to help understand the nature of organized crime. Newspapers, prosecutors, and the FBI often make organized crime sound almost like General Motors or IBM—a hierarchical organization with a few kingpins controlling thousands of subordinates. What we know about the economics of organizations makes this an unlikely description of real criminal organizations. One major limitation on the size of firms is the problem of control. The more layers of hierarchy there are between the president and the factory worker, the harder it is for management to monitor and control the workers. That is one reason that small firms often are more successful than large ones.

We would expect this problem to be especially severe in criminal markets. Legitimate businesses can and do make extensive use of memos, reports, job evaluations, and the like to pass information from one layer of the hierarchy to another. But the very information that a criminal uses to keep track of what his employees are doing can also be used by a district attorney to keep track of what the criminal is doing. What economists call “informational diseconomies of scale” are therefore a particularly serious problem in criminal firms, implying that such firms should, on average, tend to be smaller, not larger, than firms in other markets.

Criminal enterprises obviously are more difficult to study than ordinary ones. The work that has been done, however, such as that of Peter Reuter and Jonathan B. Rubinstein, seems to confirm what theory suggests. Criminal firms seem to be relatively small and the organization of criminal industries relatively decentralized—precisely the opposite of the pattern described in novels, movies, and the popular press. It may well be that “organized crime” is not so much a corporation as a sort of chamber of commerce for the criminal market—a network of individuals and small firms that routinely do business with each other and occasionally cooperate in their mutual interest.

Economic analysis can also be used to predict the effectiveness of law enforcement measures. Consider the current “War on Drugs.” From an economic standpoint, its objective is to reduce the supply of illegal drugs, thus raising their prices and reducing the amount people wish to consume. One enforcement strategy is to pressure countries such as Colombia to prevent the production of coca, the raw material used to make cocaine.

Such a strategy, if successful, would shift coca production to whatever country is next best at producing it; since coca can be grown in many different places, this shift is not likely to result in a very large increase in cost. Published estimates suggest that the cost of producing drugs abroad and transporting them to the United States represents only about 1 percent of their street price. So, even if we succeed in doubling the cost of coca—which seems unlikely, given experience with elasticity of supply of other crops—the result would be only about a 1 percent increase in the price of cocaine and a correspondingly small decrease in the amount consumed. Thus, economic analysis suggests that pressuring other countries not to produce drugs is probably not a very effective way of reducing their use.

One interesting issue in the economic analysis of crime is the question of which legal rules are economically efficient. Loosely speaking, which rules maximize the total size of the economic pie, the degree to which people get what they want? This is relevant both to broad issues such as whether theft should be illegal and to more detailed questions, such as how to calculate the optimal punishment for a particular crime.

Consider the question of laws against theft. At first glance, it might seem that, however immoral theft may be, it is not inefficient. If I steal ten dollars from you, I am ten dollars richer and you are ten dollars poorer, so the total wealth of society is unchanged. Thus, if we judge laws solely on grounds of economic efficiency, it seems that there is no reason to make theft illegal.

That seems obvious, but it is wrong. Opportunities to make money by stealing, like opportunities to make money in other ways, attract economic resources. If stealing is more profitable than washing dishes or waiting on tables, workers will be attracted out of those activities and into theft. As the number of thieves increases, the returns from theft fall, both because everything easy to steal has already been stolen and because victims defend themselves against the increased level of theft by installing locks, bars, burglar alarms, and guard dogs. The process stops only when the next person who is considering becoming a thief concludes that he will be just about as well off continuing to wash dishes—that the gains from becoming a thief are about equal to the costs.

The thief who is just on the margin of being a thief pays, with his time and effort, the price of what he steals. Thus, the victim’s loss is a net social loss—the thief has no equal gain to balance it. So the existence of theft makes society as a whole poorer, not because money has been transferred from one person to another, but because productive resources have been diverted out of the business of producing and into the business of stealing.

A full analysis of the cost of theft would be more complicated than this sketch, and the social cost of theft would no longer be exactly equal to the amount stolen. It would be less to the extent that people who are particularly skillful at theft earn more in that profession than they could in any other, giving them a net gain to partly balance the loss to their victims. It would be higher to the extent that theft results in additional costs, such as the cost of defensive precautions taken by potential victims. The central conclusion would, however, remain—that we will, on net, be better off if theft is illegal.

This conclusion must be qualified by the observation that, to reduce theft, we must spend resources on catching and punishing thieves. Theft is inefficient—but spending a hundred dollars to prevent a ten-dollar theft is still more inefficient. Reducing theft to zero would almost certainly cost more than it would be worth. What we want, from the standpoint of economic efficiency, is the optimal level of theft. We want to increase our expenditures on law enforcement only as long as one more dollar spent catching and punishing thieves reduces the net cost of theft by more than a dollar. Beyond that point, additional reductions in theft cost more than they are worth.

This raises a number of issues, both empirical and theoretical. The empirical issues involve an ongoing dispute about whether punishment deters crime and, if so, by how much. While economic theory predicts that there should be some deterrent effect, it does not tell us how large it should be. Isaac Ehrlich, in a widely quoted (and extensively criticized) study of the deterrent effect of capital punishment, concluded that each execution deters several murders. Other researchers have gotten very different results.

One interesting theoretical point is the question of how to choose the best combination of probability of apprehension and amount of punishment. One could imagine punishing theft by catching half the thieves and fining them a hundred dollars each, by catching a quarter and fining them two hundred each, or by catching one thief in a hundred and hanging him. How do you decide which alternative is best?

At first glance, it might seem efficient always to impose the highest possible punishment. The worse the punishment, the fewer criminals you have to catch in order to maintain a given level of deterrence—and catching criminals is costly. One reason this is wrong is that punishing criminals is also costly. A low punishment can take the form of a fine; what the criminal loses the court gains, so the net cost of the punishment is zero. Criminals generally cannot pay large fines, so large punishments take the form of imprisonment or execution, which is less efficient— nobody gets what the criminal loses and someone has to pay for the jail.

A second reason we do not want maximum punishments for all offenses is that we want to give criminals an incentive to limit their crimes. If the punishments for armed robbery and murder are the same, then the robber who is caught in the act has an incentive to kill the witness. He may get away, and, at worst, they can hang him only once.

One final interesting question is why we have criminal law at all. In our legal system, some offenses are called civil and are prosecuted by the victim, while others are called criminal and are prosecuted by the state. Why not have a pure civil system, in which robbery would be treated like trespass or breach of contract, with the victim suing the robber?

Such institutions have existed in some past societies. In fact, our present system of having the state hire professionals to pursue criminals is actually a relatively recent development in the Anglo-American legal tradition, dating back only about two hundred years. Several writers, starting with Gary Becker and George Stigler, have suggested that a movement toward a pure civil system would be desirable, whereas others, most notably William Landes and Richard Posner, have argued for the efficiency of the present division between civil and criminal law (see also law and economics).

About the Author

David D. Friedman is a professor of law and economics at Santa Clara University. His most recent book is Law’s Order: What Economics Has to Do with Law and Why It Matters. His Web site is

Further Reading


Ayres, Ian, and John J. Donohue III. “Shooting Down the ‘More Guns, Less Crime’ Hypothesis.” Stanford Law Review 55 (2003): 1193. Online at:
Becker, Gary S. “Crime and Punishment: An Economic Approach.” Journal of Political Economy 76 (1968): 169–217.
Becker, Gary S., and George J. Stigler. “Law Enforcement, Malfeasance, and Compensation of Enforcers.” Journal of Legal Studies 3 (1974): 1–18.
Benson, Bruce. The Enterprise of Law: Justice Without the State. San Francisco: Pacific Research Institute for Public Policy, 1990.
Ehrlich, Isaac. “The Deterrent Effect of Criminal Law Enforcement.” Journal of Legal Studies 1 (1972): 259–276.
Friedman, David. “Efficient Institutions for the Private Enforcement of Law.” Journal of Legal Studies 13 (1984): 379–397. Online at:
Friedman, David. Law’s Order: What Economics Has to Do with Law and Why It Matters. Princeton: Princeton University Press, 2000. Available online at:
Friedman, David. “Private Creation and Enforcement of Law—a Historical Case.” Journal of Legal Studies 8 (1979): 399–415. Online at:
Landes, William M., and Richard A. Posner. “The Private Enforcement of Law.” Journal of Legal Studies 4 (1975): 1–46.
Lott, John, and David Mustard. “Crime, Deterrence, and Right-to-Carry Concealed Handguns.” 1997. Abstract online at
Reuter, Peter, and Jonathan B. Rubinstein. “Fact, Fancy, and Organized Crime.” Public Interest 53 (Fall 1978): 45–67.