Other NCPA Websites

AboutSupportA National Center for Policy Analysis Project

An Economic Analysis of the Conflict Between the Patriot Act and Civil Liberty

Hugo M. Mialon and Paul H. Rubin *
Emory University

The 2005/2006 high school debate topic is:
Resolved: The United States federal government should substantially decrease its authority either to detain without charge or to search without probable cause.

The topic relates directly to the Fourth Amendment of the U.S. Constitution which states:

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

It may seem surprising to readers to learn that economics has anything interesting to say at all about this topic. However, the term “unreasonable” means that the issue of allowed searches requires some balancing of costs and benefits. Since economics is the science that studies decisions involving tradeoffs between costs and benefits, the issue being debated is actually an economic issue.

The particular tradeoff here is between what has come to be called “privacy” and “security.” In general, the benefit of allowing easier searches is increased security and protection of citizens from crime and terrorism. The cost is reduced privacy. Therefore, to examine the issue, we must compare the value of security to the value of privacy.

In normal times, the downside of reduced security is increased crime. The social decision that has to be made is a decision regarding the proper tradeoff: what level of increased crime is worth some level of reduced privacy? The results of such a decision then reflects the implicit tradeoff. (Some of the decisions are statutory, and the result of laws passed by legislatures. Some are the result of judicial interpretations of these statutes and constitutional provisions.) But we do not now live in “normal” times. Laws balancing levels of privacy and security must now consider terrorism as well as crime. Thus, while the value of privacy is presumably not affected by the possibility of terrorism, the value of security has increased because of this possibility. For example, Ted Bundy, a notorious serial killer, confessed to 28 murders. The events of September 11 led to 3000 deaths. Routine homicide bombings in Israel and Iraq often kill 20-50 people. Thus, risks of any one act of terrorism are much greater than the risks of any one act of crime. This means that society must rationally put more value on security now than previously (D. Mueller, 2004).

In its war on terrorism and crime in general, the U.S. government has weakened Fourth Amendment protections against invasions of privacy by the police. The USA Patriot Act, which was passed by Congress shortly after the September 11, 2001 terrorist attacks, allows the government to obtain warrants to monitor and search suspects without meeting previous standards of probable cause, in any criminal investigation, whether related to terrorism or not. The Act also allows greater information sharing between police and counter-terrorism officials. As a result, some searches that would have been unreasonable before September 11 are now “reasonable” and some information will be shared by various law enforcement agencies that otherwise would not have been shared. By weakening Fourth Amendment rights, the Patriot Act could increase security against crime and terrorism, but could also reduce individual liberty and privacy.

Whether or not we should uphold the Patriot Act is affected crucially by the answers to the following four basic questions:

  • Does reducing Fourth Amendment rights actually reduce harmful activities (crime and terrorism) and if so, by what amount?
  • Does it actually reduce privacy, and to what extent?
  • What are the benefits of privacy?
  • And what are the costs of crime and terrorism?

Effect on Crime

In the United States, protection against “unreasonable” searches is enforced through the “exclusionary rule.” If a search is held to be unreasonable, then whatever evidence is found in that search cannot be used in court and the evidence is excluded, or “thrown out.” The Patriot Act implies a weakening of the exclusionary rule, in the sense that evidence uncovered by searches that would have been deemed unreasonable by previous standards is no longer thrown out. In theory, a weakening of the exclusionary rule should reduce crime.

According to the economic theory of crime and punishment (Becker, 1968), rational individuals commit a crime only if their expected benefit from committing the crime is greater than their expected cost. Their expected cost of committing a crime is the probability of being caught multiplied by the punishment that they would face if they were caught. Whether or not this same rationale applies to terrorists will be discussed shortly. Where common criminals are concerned, Mialon and Mialon (2005) explain the mechanism by which a weakening of the exclusionary rule would affect a rational individual’s expected punishment for committing a crime. If the individual commits the crime, then the evidence against him might rightfully indicate that he is guilty. If the police were to stumble upon this evidence, they would search him and he would likely be convicted. But the evidence against him might also wrongfully indicate that he is innocent. In this case, the police would not have probable cause to search him. If the police searched him anyway, thereby finding evidence of his guilt, then he would likely escape conviction only if this evidence were excluded from trial; that is, only if the exclusionary rule were enforced. The Patriot Act, by weakening the exclusionary rule, should therefore increase a rational individual’s expectation of punishment for committing a crime, and hence should reduce crime.

Allowing the police to search with fewer constraints also would allow them to stop more crimes before they actually occur, thereby reducing crime even further. Additionally, if criminals know that they are subject to search they might change their behavior in various ways that lead to a reduction in crime. For example, criminals might be less willing to use tools aimed at crime (lock picks, illegal guns) if these tools can be found in a search and used against them as evidence at trial. Terrorists might be less willing to conceal explosives. Similarly, if it is easier for the security officials to perform electronic searches (say, of email or of computers), this might disrupt communication between members of (criminal or terrorist) gangs, and again reduce the value of harmful activities to potential criminals, and so perhaps the amount of crime as well. For criminals, easier searches would reduce the incentive to retain unique stolen goods which could be found and identified, and so would create incentives not to steal such goods or sell (“fence”) them sooner, again reducing the value of crime to criminals. Of course, easier searches are particularly disruptive of the drug business, since illegal drugs once found are themselves evidence of a crime, and these drugs comprise the inventory of the criminals.

As can be seen from reading the text of the Fourth Amendment, the exclusionary rule is not explicitly stated in the Amendment. It is the result of judicial interpretation of this Amendment. Alternatives are to allow searches and then find the police liable (perhaps for damages) if the search is unsuccessful. The exclusionary rule is the result of two Constitutional rulings. In 1914 in a case called Weeks v. United States, the Supreme Court held that illegally obtained evidence could not be used in prosecutions for Federal Crimes. In a case called Mapp v. Ohio in 1961, the Court ruled that this rule also applied to state crimes. (In the United States, most crimes are state crimes, so Mapp had a larger impact than did Weeks.) Interestingly, even before Mapp, exactly one half of the states had adopted a version of the exclusionary rule on their own. This created an ideal setting for a statistical examination of the effect of the exclusionary rule on crime rates, discussed below.

Many studies have found that very few cases are thrown out because of violations of the requirements for search. For example, see Oaks (1970), Cannon (1979), Davies (1983) and Crocker (1993). These findings led many scholars and policy-makers to erroneously conclude that the exclusionary rule had little impact on crime rates. However, once the courts enunciate the rule and the police learn the details of the rule, we would expect them not to violate the rule because they know that violation will lead to losing the case. Instead, they will use other methods to find criminals, rather than searches. For example, police might rely more heavily on informants or surveillance and less heavily on searches. But if searches are in some circumstances more efficient, having to turn to substitute alternative methods will lead to lower arrest and conviction rates, and this will in turn lead to higher crime rates.

In fact, this is exactly what has been found. Employing statistical or econometric techniques, Atkins and Rubin (2002) find that crime increased after 1961, when the exclusionary rule was put in place for all states by the U.S. Supreme Court, and that the increase cannot be attributed to other factors that might also affect crime, such as demographic characteristics and economic conditions. The 1961 ruling caused an increase in most types of crime: 3.9 percent for larceny, 4.4 percent for auto theft, 6.3 percent for burglary, 7.7 percent for robbery, and 18 percent for assault. Thus, since a broadening of Fourth Amendment protections caused an increase in crime, restricting those protections through the Patriot Act should have the effect of reducing crime by some amount. This is of course a benefit of the Patriot Act, although not the intended benefit (which is reducing terrorism).

It is more difficult to make predictions about the impact on terrorism, because sufficient data on terrorist acts in the United States is not available. One might expect terrorism to be less responsive to changes in expected punishments, as terrorist acts are often driven by strong hatred of enemies or love of country. Moreover, terrorism is harder to prevent, since most terrorists are part of highly organized groups that provide them with the training and means to evade government authorities. On the other hand, if terrorists must communicate with each other or if they use explosives or other equipment, then searches might be more disruptive to terrorists than to garden variety criminals. For example, Landes (1978) finds that the use of metal detectors at airports has led to vastly fewer airline hijackings. According to his estimates, if mandatory screening had not been instituted, between 41 and 67 additional hijackings would have taken place between 1973 and 1976.

This evidence tells us that levels of security are directly responsive to rules regarding searches: if we impose more restrictions on the police in searching, law enforcement becomes less efficient and levels of undesirable behavior increase. Relaxing these restrictions should therefore lead to reductions in crime, and probably to reduced terrorism as well.

Effect on Privacy

A reduction in Fourth Amendment protections results in a greater invasion of privacy only to the extent that it increases police searches of suspects without probable cause. Theoretically, as explained in Mialon and Mialon (2005), a reduction in Fourth Amendment protections should increase police searches without probable cause directly by reducing the probability that these searches will be thrown out at or before trial. But it should also reduce police searches without probable cause indirectly by reducing crime. As explained above, a reduction in Fourth Amendment protections reduces crime, so that for any given search intensity by the police, the police would be searching more innocent citizens. The police should then respond by reducing their intensity of search without probable cause. Hence, the effect of the Patriot Act on police searches is theoretically ambiguous, as is its effects on police searches of innocent suspects. If the Patriot Act strongly reduces crime, it may only weakly increase police searches, causing only a minor threat to individual privacy. However, if the Act only weakly reduces crime, it might strongly increase police searches, causing a major threat to individual privacy.

Oaks (1970) finds that the Fourth Amendment’s exclusionary rule had no significant effect whatsoever on arrests by the police in Cincinnati. Cannon (1973) replicated Oaks's research in thirteen other cities and showed that the effect of the exclusionary rule in Cincinnati was not typical. In Baltimore, for example, the number of arrests for stolen property and narcotics fell by almost 45 percent from 1960 (one year before the exclusionary rule) to 1962 (one year after), and remained at the lower level for several years after that. In Buffalo, stolen property and narcotics arrests fell by about 30 percent. On the other hand, there were no significant changes in these types of arrests in Atlanta or New York (as well as Cincinnati). However, Cannon (as well as Oaks) did not control for other factors that may also have been responsible for changes in arrests.

More recently several researchers have used interviews with individual officers (Orfield, 1987, Canon, 1991), and others have used field observation (Skolnick, 1994, Gould and Mastrofski, 2005), to analyze the impact of the exclusionary rule on police behavior. The results of these studies are also mixed.

Orfield (1987) finds that the exclusionary rule has caused police officers from the Narcotics Section of the Chicago police department to use warrants more often and to exercise more care when conducting warrantless searches. Gould and Mastrofski (2004) reviewed reports from trained field observers who accompanied police officers from a major metropolitan police department on 115 searches. They find that 30 percent of the searches were in clear violation of Fourth Amendment prohibitions, and most of these illegal searches did not result in arrest and were therefore invisible to the courts. Police officers searching for drugs were most likely to conduct illegal searches. One explanation is the possession of drugs is highly correlated with that of illegal weapons. If the police search vehicles without probable cause and find drugs and guns, then even though their search may not result in a conviction for drug possession, since it violated the Fourth Amendment, they can nevertheless seize the guns.

The extant empirical research neither proves nor disproves the inhibitory effect of the exclusionary rule on police searches, although as indicated above Atkins and Rubin (2002) find that the rule unambiguously increases crime. It is not clear, therefore, that the Patriot Act, by weakening Fourth Amendment protections, would increase police searches and hence reduce individual privacy. Instead, it seems much more likely that, by strengthening police authority to search and seize, the Act would lead to less crime, which in turn would result in fewer searches and therefore more privacy.

Benefits of Privacy

Posner (1978, 1981, 1983), Stigler (1980), and Hirshleifer (1980) discuss several reasons why people value their privacy. There also are several relevant possible definitions of privacy.

Privacy may be interpreted as the ability to achieve seclusion. Reclusive individuals may require seclusion for peace of mind. Seclusion is also essential for sleep, concentration, and creativity, which are productive activities. If, for example, the police barge into citizens' houses in the middle of the night, tearing through their possessions in search of compromising evidence, their peace of mind, whether it is productive or reclusive, is certainly compromised. However, it is not clear that Fourth Amendment protections are aimed at this form of privacy.

Privacy may also be interpreted as the ability to conceal personal information that others might use to one's disadvantage. Concealment protects reputation, which is often a valuable asset in relationships. If citizens are searched by the police and the details of the search are subsequently made public, they may suffer a loss of reputation, which might result in the loss of a job or the loss of a spouse. Society may therefore want to limit the government’s ability to obtain, retain and disseminate discrediting personal information. For these reasons, if the Patriot Act does indeed increase police searches by weakening Fourth Amendment rights, then the resulting invasion of individual privacy may be socially costly.

On the other hand, when information is concealed, people may be interacting with others on a fraudulent basis, and such interactions could be costly to society as a whole.While those who have performed harmful acts may wish to keep information about such acts private, it is not clear that such concealment is socially desirable. An additional reason people may desire privacy is fear of identity theft. In that sense, privacy is a form of security. But easier searches might actually reduce the danger of this form of crime. For example, if police can more easily search the computers of suspected identity thieves, they can perhaps reduce the incidences of identity theft, and so actually increase the security component of privacy for innocent citizens.

An increase in privacy generally means a reduction in the use of information.  But information itself is valuable, and if the use of information is increased, social value is often created (Rubin and Lenard, 2002; Hahn and Layne-Farrar, 2002).  For example, easy and quick credit checks are now possible; in many circumstances a buyer of a car, for example, is able to obtain credit on the spot. Additionally, privately held information can be used to provide security, as when child molesters are identified through use of information and denied jobs involving children, or when those with criminal records are identified and denied jobs in security.

Thus, the effects of privacy on security are themselves ambiguous.

Costs of Crime

Becker (1968) provides a general discussion of the various costs of crime. More detailed analyses are available in Cohen (2000) and Anderson (1999). In general, the costs of crime can be classified into four basic categories:

  • The direct costs to victims,
  • The costs of countering crime through law enforcement,
  • The opportunity cost of the time of incarcerated individuals; and
  • The costs of crime avoidance by potential victims.

The direct costs of crime include injury, loss of life and destruction of property. The costs of countering crime include government spending on police and the court system. The time of incarcerated individuals is an opportunity cost since these individuals could be engaged in socially productive activities if they were not incarcerated. The costs of crime avoidance by potential victims include the costs that potential victims incur to protect themselves against crime.

Generally speaking, the amount of money that society spends on fighting crime is a function of how much crime there is. The greater the incidence of crime, the more resources potential victims will spend on self-protection, resources that could otherwise be invested in more productive activities. Moreover, the greater the incidence of crime, the less are potential victims inclined to invest in productive activities because the rewards are more likely to be stolen or destroyed. Crime reduces investment. Individuals may also avoid certain consumption activities because of a fear of harm. For example, few walk in Central Park at night. Following the September 11 attacks many individuals refused to fly, even though driving (even after adjusting for the probability of a terrorist attack) is much more dangerous.

Cohen (2000) estimates that the total cost of crime is about $450 billion annually. Anderson (1999) includes costs of the criminal justice system itself and additional costs such as the opportunity cost of time of jailed offenders, and estimates that the total cost of crime (net of transfers) is about $1 trillion per year. This consists of $397 billion in “crime induced production” (resources spent preventing crime, including police, courts etc.), “opportunity costs” (of time in jail) of $130 billion, and “risks to life and health” of $574 billion. Note that this estimate includes a value of $6.1 million for each life lost to crime. This estimate is based on economic research on the “willingness to pay” to prevent death, based on wage premiums paid to workers in risky occupations. ( Anderson also includes $603 billion in “transfers” but since this income is transferred from legitimate owners to criminals, he nets out this figure.)

We can compare this to the cost of the September 11 attack and subsequent spending on prevention of terrorism. Bram et al. (2002) estimate the costs of the attacks at between $33 and $36 billion. This includes lost life and lost property. The cost of lost property (including clean-up and repairs) is $21.6 billion. The cost of lost wage and salary income in New York is between $3.6 and $6.4 billion. The estimate of the cost of the 3000 deaths is $7.8 billion. However, in estimating the economic cost of the 3,000 lives lost, Bram et al. use the “lost earnings” measure of value of life. This estimate is about one-half of the willingness to pay measure used by Anderson (which is theoretically preferable). If we double this amount (to adjust for the willingness to pay method, to be consistent with Anderson), the total cost rises to between $41 and $44 billion. This figure does not include the cost of conducting military operations in the War on Terror.

In other words, the events of September 11 imposed on U.S. citizens a cost equal to about 4% of the total cost of crime per year – for a single terrorist event.

Hobjin (2002) estimates the cost of Homeland Security per year. His estimate is that the cost is about 1% of GDP, or about $70 billion. He argues that this is a “small part of total resources.” This estimate includes both direct and indirect effects, such as additional time spent waiting in airports. Anderson’s estimate of the amount spent on crime prevention is $400 billion. Thus, the September 11 attacks have led to the equivalent of a 17% increase in spending on crime prevention. This may seem large, given that the estimate is that the costs were only 4% of the costs of all annual crime. However, it is possible that a second terrorist attack could be much more costly than the September 11 attack, so this increase may be justified.

On the other hand, J. Mueller, (2004) argues that we are spending too much on anti-terrorism efforts. He argues that the costs of terrorism are actually rather low, and that much anti-terrorist policy is an overreaction. He does not specifically address issues related to search and the Fourth Amendment.

Conclusion

Whether the Patriot Act should be upheld depends most importantly on the Act’s relative costs and benefits. By weakening Fourth Amendment rights, the Act will probably reduce crime and terrorism, which will benefit society since the costs of crime are large and the costs of terrorism are potentially large as well. The Act may or may not lead to more general invasion of privacy, depending on whether its effect on crime and terrorism is small or large. But if it does lead to greater invasions of privacy, this will hurt society since there are benefits of privacy.

Since it is difficult to quantify the benefits of privacy, it is difficult to determine if the reductions in privacy are justified by the benefits. But we can say something about the direction. Since September 11, the cost and the danger of lack of security have increased in America. If society were making the optimal tradeoff between privacy and security on September 10, then that tradeoff is no longer optimal today. If we hold constant the benefits of privacy but increase the risks of lack of security, then reduced protections from the Fourth Amendment is an efficient policy (D. Mueller, 2004). We cannot say that the exact amount of this reduction is correct, but the direction makes sense.

References

Anderson, David A. (1999),
“The Aggregate Burden of Crime,” Journal of Law and Economics Vol. 42, No. 2, 611-642.
Atkins, Raymond A. and Rubin, Paul H. (2003).
“Effects of Criminal Procedure on Crime Rates: Mapping Out the Consequences of the Exclusionary Rule,” Journal of Law and Economics, Vol. 46, No. 1, 157-180.
Becker, Gary S. (1968).
“Crime and Punishment: An Economic Approach,” Journal of Political Economy, Vol. 76, No. 2, 169-217
Bram, Jason, James Orr and Carol Rapaport,
“Measuring the Effects of the September 11 Attack on New York City,” Federal Reserve Bank of New York (FRBNY) Policy Review, November, 5-19.
Cannon, Bradley C. (1973).
“Is the Exclusionary Rule in Failing Health? Some New Data and a Plea Against a Precipitous Conclusion,” 62 Kentucky Law Journal 681.
Cannon, Bradley C. (1991).
“Courts and Policy: Compliance, Implementation and Impact,” in John B. Gates and Charles A. Johnson, eds., The American Courts: A Critical Assessment, Washington, D.C.: CQ Press.
Cohen, Mark A. (2000).
“Measuring the Costs and Benefits of Crime and Justice,” Criminal Justice, Vol. 4, 263-315.
Crocker, L. (1993).
“Can the Exclusionary Rule be Saved?” 84 The Journal of Criminal Law & Criminology 310.
Davies, T.Y. (1983).
“A Hard Look at What We Know (and still need to know) about the ‘Costs’ of the Exclusionary Rule: The NIJ Study and Other Studies of ‘Lost’ Arrests,” 3 American Bar Foundation Research Journal 611.
Gould, Jon B. and Stephan D. Mastrofski (2004).
“Suspect Searches: Assessing Police Behavior Under the U.S. Constitution.” Criminology and Public Policy, Vol. 3, No. 3, 315-362.
Hahn, Robert W. and Layne-Farrar, Anne (2002).
“The Benefits and Costs of Online Privacy Legislation” (October 2001). AEI-Brookings Joint Center Working Paper No. 01-14. http://ssrn.com/abstract=292649
Hirshleifer, Jack (1980).
“Privacy, Its Origin, Function, and Future,” Journal of Legal Studies, Vol. 9, 649–66.
Hobijn, Bart (2002).
“What Will Homeland Security Cost,” Federal Reserve Bank of New York working paper, May 17.
Landes, William M. (1978).
“An Economic Study of U.S. Aircraft Hijacking, 1961-1976,” Journal of Law and Economics, Vol. 21, No. 1, 1-31.
Mialon, Hugo M. and Mialon, Sue H. (2005).
“The Economics of the Fourth Amendment: Crime, Search, and Anti-Utopia,” Department of Economics, Emory University Working Paper, http://ssrn.com/abstract=591667.
Mueller, Dennis C. (2004).
“Rights and Citizenship in a World of Global Terrorism,” European Journal of Political Economy, Vol. 20, 335-348.
Mueller, John (2004).
“A False Sense of Insecurity,” Regulation, Fall, 42-46.
Oaks, Dallin H. (1970).
“Studying the Exclusionary Rule in Search and Seizure,” 37 University of Chicago Law Review 665.
Orfield Jr., Myron W. (1987).
“The Exclusionary Rule and Deterrence: An Empirical Study of Chicago Narcotics Officers,” 54 University of Chicago Law Review 1016.
Posner, Richard H. (1978).
“An economic theory of privacy,” Regulation, 19-26.
Posner, Richard A. (1981).
“The economics of privacy,” American Economic Review, Vol. 71, No.2, 405- 409.
Posner, Richard A. (1983).
The Economics of Justice. Harvard University Press: Cambridge, Massachusetts.
Rubin, Paul H. and Thomas M. Lenard (2002).
Privacy and the Commercial Use of Personal Information, Progress and Freedom Foundation and Kluwer Academic Publishers.
Skolnick, Jerome H. (1994).
Justice Without Trial: Law Enforcement in a Democratic Society, 3rd ed., New York: Macmillan.
Stigler, George J. (1980).
“An introduction to privacy in economics and politics,” Journal of Legal Studies, Vol. 9, 623-644.

* Mialon and Rubin, Department of Economics, Emory University, Atlanta, GA 30322-2240 (E-mails: hmialon@emory.edu and prubin@emory.edu).