University of Manchester School of Law
Economics of Crime and Criminal Law ARMANDO JIMENEZ SAN VICENTE
University of Manchester School of Law
Course Content 1. Introduction to the economics of crime. 2. The basic model. 3. Extensions of the basic model: marginal deterrence, type I and type II errors, and other justifications for non-maximal sanctions. 4. Economics of criminal law. 5. Economics of criminal procedure, including plea-bargaining, pro-defendant bias procedural rules and prosecution 6. The econometrics of crime.
University of Manchester School of Law
SESSION APRIL 14 Armando JimĂŠnez San Vicente
University of Manchester School of Law Economics of Law Enforcement
Economics
Individual Behavior: Microeconomics Normative Approach: Welfare Economics
Law
Role and Goals of Law Content of Law Structure of Law: Private vs. Public
Enforcement
Certainty vs. Severity Institutional Arrangements
General Model Individual Behavior Behavior Individual -Illegal gains gains -Illegal -Costs of of breaking breaking -Costs the law law the -Rational behavior behavior -Rational -“Born to to be be Bad” Bad” -“Born
Law Enforcement Enforcement Law -Goals -Goals -Sanctions -Sanctions -Resources -Resources -Organization of of -Organization Justice System System Justice
Social Behavior Behavior Social -Social Contract Contract -Social -Political Economy Economy -Political
Law Law -Content -Content -Completeness -Completeness -Institutions -Institutions -Tort vs. vs. Crime Crime -Tort -Administrative -Administrative
Background The Early Days • Cesare Beccaria (1738-1794) [Cesare Bonesana, Marchese di Beccaria] Of Crimes and Punishment, 1764 • Jeremy Bentham (1748-1832) Introduction to the Principles of Morals and Legislation, 1789
Beccaria •
Fundamental Ideas
– There should be as little law as possible – Criminal law should not simply uphold moral values for their own sake, but should furnish the requirements of society – Criminal offenses should be clearly defined – Punishment should be proportional to the harm done to society; it should not exceed that which was necessary to ensure that the criminal would not re-offend; – Punishment should deter in a general sense; – There must be certainty of punishment and it should be imposed promptly – Certain moderate punishment is easier to enforce and therefore is more effective than the mere possibility of a severe one; – Fines should dominate over imprisonment and death penalty (which he disliked); – Corporal punishment is the appropriate way to deal with violent offenders.
Impact of Beccaria’s work • Abolishment of death penalty in Tuscany and, for a while, in Austria. • Declaration of Rights of Man in France. • Revolutionary Code of 1791: the level of punishment was related to the precise nature of the offense to such an extent that the only function for the judge was the determination of guilt or innocence. • American Declaration of Independence.
Impact in America “We perceive... that the severity of our criminal law (that is, the old criminal law inherited from England) is an exotic plan, and not the native growth of Pennsylvania... as soon as the principles of Beccaria were disseminated, they found a soil that was prepared to received them.� William Bradford, 1793, US Attorney-General
Bentham • Fundamental Ideas – Criminal law should be subject to the principle of “the greatest happiness of the greatest number” (utilitarism); – Every human act should be first evaluated in terms of whether it will cause pleasure or pain; – To deter a crime, the amount of pain derived from the forbidden act should be greater than the amount of pleasure; – Harsh punishment should be constrained by the social contract, noone could be taken to have agreed to the possibility of receiving excessively harsh punishment;
Impact of Bentham • Reports for a Criminal Code (1833-1848) in the UK; • Code Napoleon of 1810.
Neoclassical School • Rossi (1787-1848); Garraud (1849-1930); Joly (1839-1925). • Fundamental Ideas: – There are limitations to rational choice; – Punishment should fit the crime rather than the criminal; – People are free not to choose crime.
And Rational Choice Goes Away... • Positivism... The predestined model; [from 1880s to 1960s] • Labelling Theories ... The victimized model; [from the 1960s on to today]
... And Rational Theory comes back • The Modern Rational Theory – – – – –
James Q. Wilson, Thinking About Crime (1975) Clarke and Cornish, introduce limited rationality (1987) Cohen and Felson, Routine Activities Theory (1979) Hirschi, Social control theory (1969) Gottfredson and Hirschi, General theory of crime (1990)
• The Economic Theory – – – –
Becker (1968, JPE) Stigler (1970, JPE) Ehrlich (1973, AER) Work by Polinsky and Shavell in early 1980s
OUTLINE 1.
2.
3.
Basic Model
1. 2. 3.
Becker, 1968 Rational Criminals The Pareto Conflict Theorem: Should Illegal Gains Matter?
1. 2. 3.
Multiplier Principle High-Fine Low-Probability Result Should Sanctions be Based on the Harm to the Victim or the Gain to the Criminal? Inefficiency of Multiplier Principle
Deterrence with Fines
4.
Risk-aversion
BASIC MODEL • • • •
Objective of law enforcement: DETERRENCE. Why?: Externality (social harm or social damage). How?: Internalize externality. Market Approach to Law Enforcement: – Is there a market for offenses? – Is a sanction really a price? • Price approach (Becker, 1968) • Pigou taxation approach (Cooter, 1982)
•
Policy instruments: – Sanction: Fine; – Resources devoted to punishment: Probability;
BASIC MODEL • Individuals compare: – Benefits
• Illegal Gain • Other Gains (Psychological, Biological,…)
– Costs
• Expected Punishment • Social Costs: Stigma • Other Costs (Psychological)
• Decision to offend is the rational outcome of comparing costs and benefits There will be crime if Benefits > Costs
NORMATIVE FRAMEWORK • What are the payoffs? – Criminal : Benefits of crime – Costs of crime Taxes – Victim: Harm from crime – Taxes – Government: Taxes + Benefits from punishment– Enforcement Costs
NORMATIVE FRAMEWORK • Utilitarian Social Welfare: Benefits of Crime [C] + Benefits from Punishment [G] – Costs of Crime [C]– Harm from Crime [V] – Enforcement Costs [G]
NORMATIVE FRAMEWORK • Should Illegal Gains be Considered? – Social Contract – Kaldor-Hicks Efficiency – Pareto Conflict Theorem (Kaplow and Shavell, 2000)
NORMATIVE FRAMEWORK • Example 1 – – – – –
Harm: 100 Gain: 80 Fine without fairness consideration: 1000 Fine with fairness consideration: 500 Probability: 10%
– Without fairness consideration, the crime is not committed (10% of 1000 > 80); – With fairness consideration, the crime is committed (10% of 500 < 80). Victims are worse-off.
NORMATIVE FRAMEWORK •
Example 2 – – – – –
Harm: 5 Gain: 100 Fine: 5 Probability: 100% Indirect taxes on gains: 20%
– –
Illegal Gains Matter: Welfare if the crime is committed is 95 and welfare if the crime is not committed is 0. Crime should take place. Illegal Gains do not matter: Crime should not take place because it generates a loss of 5.
– – –
Criminal: gets 75 (80% of 100 – fine) if crime takes place, 0 otherwise; Victim: gets 15 (20% of 100 – harm) if crime takes place, 0 otherwise; Government gets 5 (fine) if crime takes place, 0 otherwise.
DETERRENCE WITH FINES • Compliance is achieved if Benefit from Crime = Cost of Crime • Benefit = Illegal Gain • Cost of Crime = Probability x Fine
DETERRENCE WITH FINES • Compliance is Efficient if... Illegal Gain < Harm
• Compliance is NOT Efficient if... Illegal Gain > Harm
• Therefore, we should have... Probability x Fine = Harm
THE MULTIPLIER PRINCIPLE Fine=Harm/Probability Harm = 100
LIMITATIONS OF MULTIPLIER PRINCIPLE
• Resistance to Inflating Sanctions – Punishment should fit the crime
• Difficulty in estimating accurately the probability of punishment • Individuals are not risk neutral • Does not say which policy is efficient
BECKER’S FUNDAMENTAL RESULT Fines are costless transfers Resources devoted to punishment are socially costly
High Fine – Low Probability Fine= Entire Wealth Probability should go to Zero
BASIC PROPERTIES OF LAW ENFORCEMENT • Acts that have a benefit higher than the external cost they cause SHOULD NOT be deterred and SHOULD be punished. • Acts that have a benefit lower than the external cost they cause SHOULD be deterred.
BASIC PROPERTIES OF LAW ENFORCEMENT â&#x20AC;˘ Punishment should be based on harm to the victim and not on the gain to the criminal. â&#x20AC;˘ A sanction based on the gain to the criminal would deter by removing the illegal gain. However, if the gain is above the level of harm, the deterrence of these acts would be inefficient.
OPTIMAL LAW ENFORCEMENT • The Multiplier Principle is not Efficient because it does not care about enforcement costs. • Efficient probability and sanction should maximize social welfare • Fine equals entire wealth (Becker) • Probability equals Harm/Fine – Marginal Cost of Enforcement
• Expected Fine is less than Harm • Some under-deterrence is efficient (Polinsky and Shavell) • Complete deterrence is not efficient (Polinsky and Shavell)
OPTIMAL LAW ENFORCEMENT • Social welfare: Utilitarian Approach Criminals: Illegal Gains – Expected Punishment -Taxes Victims: - Social Harm- Taxes Government: Taxes+ Expected Punishment – Enforcement Costs Total: IG – H – Enforcement Costs
OPTIMAL LAW ENFORCEMENT • Efficient Deterrence is not… … Complete Deterrence. … achieved by the Multiplier Principle.
• Efficient Deterrence is… … maximization of Social Welfare. … about criminals being able to compensate victims.
QUICK OVERVIEW OF RISK AVERSION • Risk averse individuals are not indifferent between any probability and fine equal to social harm; • Risk premium is strictly increasing in the fine, but not in the probability; • Efficient fine must balance… … risk generates deterrence; … disutility of risk is a social cost! • Therefore, the efficient fine is usually less than the entire wealth.
QUICK OVERVIEW OF RISK-AVERSION • Example 1:
– Policy 1: Probability is 50% and Fine is 200. – Policy 2: Probability is 100% and Fine is 100. – Illegal Gain: 75 – Utility of Criminal under Policy 1: 75- 100- disutility of risk – Utility of Criminal under Policy 2: 75-100 – Policy 2 could be better if: Social Cost from Disutility of Risk > Difference in Enforcement Costs from Higher Probability
OUTLINE 1. 2.
3.
Previous Results Deterrence with Non-Monetary Sanctions
1.
Imprisonment
1. 2.
Polinsky and Shavell, 1984. Alternative approaches.
2. 3. 4.
Parole Death Penalty Shaming Penalties
1. 2. 3.
Stigler, 1970 Attempts Repeat Offenders
Marginal Deterrence
IMPORTANT RESULTS • • •
Multiplier Principle… fine=harm/probability High-Fine Low-Probability Result (Becker, 1968) Efficient Law Enforcement: – Complete deterrence is not efficient: • … because illegal gains matter, • … because enforcement is costly.
– Some under-deterrence is efficient: • … because enforcement is costly, • … multiplier principle is not efficient.
DETERRENCE WITH NONMONETARY SANCTIONS • Nonmonetary sanctions are different because: – They are socially costly to impose (NOT costless) – They create disutility (NOT transfer)
DETERRENCE WITH NONMONETARY SANCTIONS • Social welfare: Criminals: Illegal Gains – Expected Punishment -Taxes Victims: - Social Harm-Taxes Government: Taxes– Enforcement Costs Total: IG – EP - H – Enforcement Costs
DETERRENCE WITH NONMONETARY SANCTIONS • Optimal Nonmonetary Sanctions (Polinsky and Shavell, 1984): • Should not be maximal… • Should be used infrequently… • Should be implemented once fines are exhausted… » Less wealthy people
• Should only be applied for more harmful acts…
DETERRENCE WITH NONMONETARY SANCTIONS • Why not? • Wealth is private information: imprisonment as a device for information revelation (Levitt) • Disutility of imprisonment (Polinsky and Shavell) – Risk-aversion with respect to fines – Risk-aversion with respect to imprisonment
• Corruption or other means of punishment avoidance (Garoupa and Klerman)
DETERRENCE WITH NONMONETARY SANCTIONS • Death Penalty… – It can be effective without being efficient!!
• Example 1: – Each crime deterred saves 100… – Death penalty deters 10 and costs 500: 1,000-500= 500 expected gain – Imprisonment deters 6 and costs 50: 600-50= 550 expected gain
DETERRENCE WITH NONMONETARY SANCTIONS • Parole or substitution of imprisonment by electronic tagging: – We need to balance the benefits from a «good behavior» effect against the reduction in deterrence.
DETERRENCE WITH NONMONETARY SANCTIONS • Shaming Penalties: – Disutility of humiliation is high... – Cheap for the government... – But involves loss of reputation... • ... If extralegal sanctions are not taken into account ... Over-deterrence!!
MARGINAL DETERRENCE • If sanctions are maximal for all acts, the marginal expected cost for a second crime is zero... It will not be deterred! • Suppose there are two possible criminal acts: – Objective 1: Deter act A and act B. – Objective 2: If act A has been committed, deter act B. – Objective 3: If act B has been committed, deter act A.
MARGINAL DETERRENCE • If we apply maximal fines objectives 2 and 3 cannot be achieved!! • Therefore, in order to satisfy objective 2 and 3, we need nonmaximal fines. • Objective 1 is deterrence in the general sense whereas objectives 2 and 3 are marginal deterrence.
MARGINAL DETERRENCE • Implications (Stigler and others): – ... More harmful acts have higher fines, but less harmful act should have lower fines... – ... The fine for act A should increase with the level of harm generated by act A and decrease with the level of harm generated by act B... – ... There is a tension between marginal deterrence objectives and general deterrence objectives...
MARGINAL DETERRENCE • Attempt: – A potentially harmful act that does not happen to result in harm. – There are three types of attempts to consider: • It fails due to lack of effort by the criminal; • It fails due to pure chance; • It fails but the criminal thinks otherwise (DF’s rational voodoo).
MARGINAL DETERRENCE • First case: – Apply marginal deterrence theory. • Sanction for attempt should be less than sanction for full crime; • Sanction for attempt should be increasing on the level of harm generated by it, but decreasing on the level of harm caused by the full crime.
MARGINAL DETERRENCE • Second case: – Should be treated as full crime because... • ... It improves deterrence. • ... There is no incentive problem here...
– Practical restriction: it could be difficult for a government to distinguish an attempt that resulted in no harm from an act that does not create harm. • Probability of detection of an attempt can be extremely low.
MARGINAL DETERRENCE • Third case: – Punishing attempts is irrelevant... Because the potential criminal is contemplating a crime and not an attempt...
MARGINAL DETERRENCE • Repeat offenders: – By the marginal deterrence theory, the sanction for the first crime should be less than the sanction for the second crime if more harmful; – Furthermore, the sanction for the first crime should be increasing on the level of harm it produces, but should be decreasing on the level of harm produced by the second crime.
MARGINAL DETERRENCE • First problem: if the first crime is deterred, so will be the second crime. – Tension between marginal deterrence and general deterrence.
• Second problem: by the multiplier principle, it should be irrelevant... – Escalating sanctions would be explained by wealth constraint!!
REPEAT OFFENDERS • Information-based explanations: – The likelihood that a second crime is a mistake is lower than the first time... – A second crime signals a higher likelihood of future crimes...
OUTLINE 1.
Victims
1. 2. 3.
Precaution
1.
Should we have comparative negligence rules in criminal law?
Compensation Hate Crimes
VICTIMS • Victims are not passive... – Precaution – Compensation for reporting – Hate Crimes
VICTIMS â&#x20AC;˘ Private Precaution is not SOCIALLY optimal because there is a fundamental difference between the social and private motives for precaution. â&#x20AC;˘ Victims do not care about the illegal gain for the criminal, losses for 3rd parties, and general deterrence.
VICTIMS • Precaution could be above or below socially efficient level: – Above:
• Because victims do not care about the gain to the criminal (Ben-Shahar and Harel) • Because precautions activities just divert crime (Png) – The case of Lojack (Ayres and Levitt)
– Below:
• Because victims rely on the government for general deterrence (Hylton): moral hazard
– Need for negligence rules in criminal law to achieve efficient precaution (Harel).
VICTIMS • Reporting is important for effective deterrence... – ... But compensation might create moral hazard... (Garoupa).
• Other problems with compensation for reporting: – Fabrication of offenses... just like for insurance purposes...
VICTIMS • Hate Crime Laws (HCLs): – Interest group driven (Posner); – Could be efficient (Dharmapala and Garoupa)... • ... HCLs might reduce victim costly and wasteful avoidance activities...
University of Manchester School of Law
SESSION APRIL 16 Armando JimĂŠnez San Vicente
OUTLINE 1.
2.
Deterrence with a Reduction in Punishment Costs
1. 2. 3. 4.
Plea-Bargaining Corruption Avoidance activities including defense expenditure Legal Aid
1.
Precaution
Victims
2. 3.
1.
Should we have comparative negligence rules in criminal law?
Compensation Hate Crimes
PLEA-BARGAINING • Advantages – 1) Reduces enforcement costs; – 2) Reduces risk disutility for criminals and prosecutors.
PLEA-BARGAINING •
Disadvantages – 1) Reduces expected sanction, hence dilutes deterrence • Increase the underlying sanction (problem if it is already maximal).
– 2) Infrequent use of trials • Harm is private information: A trial outcome could be better to get an approximation to the level of harm than a settlement between the criminal and the prosecutor. • Plea bargaining can also hinder the amplification of evidence and development of law enforcement through more investigation and discovery. • A criminal who has committed several offenses may be willing to agree on pleading guilty to one of them before others are discovered and investigated.
•
Similar approach is useful for self-reporting behavior… (Kaplow and Shavell, Innes)
PLEA-BARGAINING • Information-based theories: – Plea-bargaining as signal… • … From Criminals… • … From Prosecutors…
– Even with a semi-pooling equilibria, signals convey useful information!!
CORRUPTION • What can be corruption in this context? – Punishment avoidance (bribes, collusion, compounding). – Fabrication of offenses. – Entrapment.
• Corruption distorts efficient enforcement by … diluting deterrence; … conviction of innocent people; … misallocating enforcement effort.
CORRUPTION • Bribing… – … Increases enforcement costs to offset effect on expected sanctions; – … Increases sanctions for the underlying offense if still not maximal; – … By the marginal deterrence theory, the sanction for the underlying offense should be less than otherwise; – … It should be increasing on the harm it generates, but decreasing on the harm generated by corruption; – … Should have in mind that there is only corruption if individuals are not deterred from the underlying offense.
CORRUPTION • Becker and Stigler’s collusion-proof equilibrium… • Fines should equal reward to enforcers… • Hence, fines should not maximal… – … lower fines cause under-enforcement!!
• And rewards should not be too low… – … the possibility of collusion induces over-enforcement!!
• Ambiguous effect on deterrence… – … certainly more expensive…
AVOIDANCE ACTIVITIES • Criminal avoidance activities include… – Defense expenditure; – Obstruction of justice; – Hiding evidence; … any activity that reduces the probability of conviction…
AVOIDANCE ACTIVITIES • The existence of avoidance activities generally suggests that the optimal fine should not be as high as otherwise. – The potential for costly avoidance activities makes the fine ultimately costly at the margin (a higher fine implies more waste of resources in avoidance activities); – Setting a fine ignoring the existence of these costs eventually generates over-deterrence.
• Imposing a lower fine reduces the waste of resources in avoiding punishment and makes enforcement more effective.
AVOIDANCE ACTIVITIES • Information-based theories: – Level of avoidance activities as signal… • … For Guilty Behavior… • … For Disutility of Punishment…
– Even with a semi-pooling equilibria, signals convey useful information!!
OUTLINE 1.
Extensions of the Basic Model
1. 2.
2.
Errors
1. 2.
Deterrence Miscarriage of justice
Monitoring and Investigation
Other Goals of Law Enforcement
1. 2. 3.
Incapacitation Rehabilitation Preference Shaping
ERRORS • Two types – False positives (acquitting the guilty); – False negatives (convicting the innocent).
• Effect of False positives – Reduces expected sanction -> More Crime
• Effect of False negatives – Reduces the opportunity cost of crime -> More Crime
ERRORS • Legal error implies that the government may have to increase the probability of detection and the fine to offset the dilution of deterrence. • However, a higher fine might produce a substantial cost of miscarriage of justice… – … it is better to acquit 10 criminals by mistake than convict an innocent…
• It seems a choice has to be made:
– Policy 1: Severe punishment, high level of deterrence, more innocents, high social cost of legal error; – Policy 2: Mild punishment, low level of deterrence, fewer innocents, lower social cost of legal error.
ERRORS • Probability of apprehension and conviction: – More Apprehension
• More innocents are apprehended • More criminals are apprehended
– Accuracy
• Fewer innocents are convicted • More criminals are convicted
– Optimal choice between apprehension and accuracy • Trade-off between deterrence and legal error
– Examples of improving accuracy (?) • Better prosecution • Better defense counseling • Raising standard of proof
MONITORING vs. INVESTIGATION • Enforcement: – General: several offenses are detected at the same time; [MONITORING] – Specific: a certain offense is investigated. [INVESTIGATION]
• In Investigation… – The probability is chosen independently for each type of harmful act.
• In Monitoring… – The probability is the same for all harmful acts.
MONITORING vs. INVESTIGATION • The multiplier principle can be applied to – … investigation: p=h/f – … monitoring: f=h/p
• Therefore in the case of monitoring – … Less harmful crimes pay low fines… – … More harmful crimes pay high fines…
• Should the government lower the probability and raise fines for low harm acts?
MONITORING vs. INVESTIGATION • Example 1: – Maximal fine: 10,000 – Probability is 10% – Expected sanction is 1,000 – Offenses imposing harm above 1,000 should be punished with maximal fines… – Offenses imposing harm below 1,000 should be punished with fines given by f=10xharm
MONITORING vs. INVESTIGATION • Suppose we consider imposing a lower probability, say 5%. – Offenses below 500, keep the same principle: f=20h [pay more now], but same expected sanction (equals h by multiplier principle). – Offenses between 500 and 1,000 pay maximal fine now, but lower expected sanction (from h dropped to 500). – Offenses above 1,000 still pay maximal fine, but lower expected sanction (from 1,000 dropped to 500).
MONITORING vs. INVESTIGATION • Conclusion from Example 1: – Reducing the probability saves on enforcement costs… – And more crimes are punished with maximal fine… – But expected sanction is reduced for more harmful crimes…
OTHER GOALS OF ENFORCEMENT • • • •
Incapacitation Rehabilitation Retribution Preference Shaping – Stigma – Social Norms
OTHER GOALS OF ENFORCEMENT • Incapacitation – Benefit of incapacitation is the harm s/he would commit otherwise; – Imprisonment is more effective than fines; – However, in some particular contexts, other sanctions can be equally effective… • … losing driver license, suspending professional rights,…
– Probability of detection and punishment is irrelevant.
OTHER GOALS OF ENFORCEMENT • Incapacitation – Deterrence is about dissuading an individual from committing a criminal act; – Incapacitation is about preventing an individual from engaging in a criminal act and remove parties from positions in which they could cause harm; – What an individual knows is irrelevant for incapacitation.
OTHER GOALS OF ENFORCEMENT • Incapacitation – The marginal benefit from incapacitation should equal the marginal cost; – Only those people who are highly prone to harm society should be incapacitated. – Empirical question: can we distinguish incapacitation from deterrence? • See Kessler and Levitt on using sentencing enhancements to distinguish INC and DET.
OTHER GOALS OF ENFORCEMENT • Rehabilitation – Marginal benefit from changing criminogenic preferences should equal the marginal cost of rehabilitation programs. – Empirical literature is not exciting about rehabilitation…
• Retribution – Sanctions will be higher than otherwise…
OTHER GOALS OF ENFORCEMENT • Preference Shaping – Stigma • Deterrence • Intertemporal inconsistency: entry vs. exit barriers.
– Social Norms • • • •
Norms management Relationship between extralegal and legal norms Criminal law as a signal Criminal sanctions as part of optimal morality rules
– Expressive Law and Economics and Criminal Law as a focal point
OUTLINE 1.
The Choice of Private versus Public Enforcement of the Law
1.
2.
Theory
1. 2. 3. 4.
Becker and Stigler, 1974 Landes and Posner, 1975 David Friedman, 1984 Enforcement Problems: 1. 2. 3.
Under-enforcement; Polinsky, 1980 Error; Garoupa, 1997 Rent-seeking; Garoupa and Klerman, 2002
Historical Cases
1. 2. 3. 4.
Roman Law England France Iceland
THEORY •
Becker and Stigler, 1974 JLS • •
A system where the enforcer receives the fine paid by the offender eliminates corruption; We need private enforcement to get efficient deterrence.
THEORY •
Landes and Posner, 1975 JLS •
•
A fine is the price for criminal market and price for the enforcement activities – it cannot be set at a level which will optimize both criminal and enforcement activities. Private enforcers will not internalize the positive externality from raising the probability, hence crime rates will be higher.
THEORY •
David Friedman, 1984 JLS • •
•
Government enforcement is also subject to inefficiencies, therefore the question is which system is superior Empirical answer… Enforcement contracts are prior to detection, and therefore the deterrent effect of catching criminals can be internalized by “advance contracting” as in any other market. Allocation of the right to catch criminals: private property of the victim who should be able to sell it to the highest bidder.
THEORY •
Private Enforcement Problems: 1. Under-enforcement; Polinsky, 1980 JLS • Monopoly • Competitive Market 2. Error; Garoupa, 1997 IRLE • More errors than efficient; • Sanction enforcer when accused is found not guilty pay for performance • Make sure performance measure cannot be manipulated: evidence,… 3. Rent-seeking; Garoupa and Klerman, 2002 ALER • Enforcers and Government Bureaucrats compete for rents.
HISTORY AND INSTITUTIONS •
Roman Law – Offenses (crimina) punishable by criminal Courts, following a State-controlled procedure, were either offenses directly against the political community (like treason, for instance), or offenses which were directed at individuals but produced insecurity in the entire community (some instances of murder, for example). – These Courts, together with their rules and procedures, covered only the offenses understood to affect the state or the society as a whole.
HISTORY AND INSTITUTIONS • Roman Law – Other actions that nowadays we consider as punishable crimes, like simple (furtum) and aggravated (rapina) theft, and assault, were considered not to be crimina, but delicta (the ancestor of torts), and left to private Law courts and private prosecution. They were thought to have only an individualized effect upon a particular victim. – However, forgery and counterfeiting (falsum) was a public quaestio because it was seen as making one’s property more insecure.
HISTORY AND INSTITUTIONS • Roman Law – The victim suffering from a given behavior could go to civil courts or before the authorities to lay charges. – In the vast majority of the cases, the authorities had no reason to intervene since reparation was paid. – Certain classes of thieves were identified as very dangerous to society (usually due to large scale operations) and aggravated penalties were imposed by criminal courts. In these cases, the state intervened and it is not clear if a choice of prosecuting these classes of thief in civil courts was given to the victim.
HISTORY AND INSTITUTIONS •
Roman Law – Roman criminal or standing jury Courts (iudicia publica) are somehow different from the institutions with that name nowadays. – The criminal inquiry (quaestiones) was heavily adversarial. – The defendant, represented by one or more counsels, was faced by a private person (singular or plural) who acted as prosecutor. – Neither side was typically a legal professional, though these were usually consulted. – The state could accept or reject (by the praetor or a less important officer) the case, and would arrange for the best prosecutor in case more than one claim was made.
HISTORY AND INSTITUTIONS • Roman Law
– The penalties were fixed by law, and damages should be assessed by the same jury in a separate proceeding (litis aestimatio). – The court was presided by the praetor or his subordinate (quaesitor). – No clear guidelines existed to determine the burden of proof.
HISTORY AND INSTITUTIONS • Roman Law – Most of the offenses in Roman Law gave the injured party a right to civil actions. These offenses were designated as delicta, including larceny and theft (furtum), robbery with violence (rapina), damage or losses to property (damnum iniuria datum), and personal injury (iniuria). – Some other acts were described as quasi delicts. These quasi delicts bear, to some extent, a resemblance to modern vicarious liability principles, when one person is liable for acts of other people (e.g., employees) even though he is unaware of what they are doing.
HISTORY AND INSTITUTIONS •
Roman Law
– The delict of theft could be manifest (furtum manifestum) or nonmanifest (furtum nec manifestum). – Manifest theft took place when the thief was caught in the act (or before the thing was transferred to the place the thief had designated for its storage). The penalty would be four times the value of the stolen good, whereas for non-manifest theft the penalty was just the double. – A ritual search (lance et licio) for stolen property was established. If the stolen property was found, the theft was treated as manifest. When the stolen property was found without the formal search, the penalty was reduced to three times the value of the stolen goods.
HISTORY AND INSTITUTIONS • ROMAN LAW
– The incentive for the victim not to kill the criminal was reinforced by the existence of noxal liability since it created a fair chance of reparation. – When a delictual action was brought against the father or the master (paterfamilias) for the wrong committed by the son or by the slave, the former had the choice of either paying the amount due, or of handing over his dependent. However, if the master himself was directly involved in the wrongful action, he usually lost his right to noxal surrender. – Violent robbery (rapina) was initially subject to the same penal remedy (in quadruplum) than manifest theft, and even later, during the Justinian period, the penal remedy was reduced to three times the value of the stolen goods, although the victim could claim restitution of the property through rei vindicatio or condictio.
HISTORY AND INSTITUTIONS •
• •
Was Private Enforcement under Roman Law Efficient?
There are some efficiency-based explanations of certain aspects of Roman Law (Posner, Parisi, and others). The literature suggests that these rules could have been efficient in the technological and social context of the time.
HISTORY AND INSTITUTIONS • –
England • • •
– • • • •
–
Private Prosecution of Crime
System of rewards for individual prosecution that was largely abandoned by 1750 due to entrapment and negative effect on juries. Reimbursement of prosecution costs introduced by 1752. Only after 1778, prosecutors could be reimbursed for failed prosecution. Creation of London metropolitan police in 1829.
Typology of Crime
Minor offenses, usually punished with whipping and other shaming penalties; Clergyable felonies, not subject to capital punishment; Non-clergyable felonies, subject to capital punishment; Definition of clergy had been broadened to include anyone who could read and, by 1706, almost any defendant, although defendants who were not actually clergymen could plead clergy only once.
Extensive use of pardons
HISTORY AND INSTITUTIONS • England – Was private enforcement efficient? • Yes (David Friedman, 1996 UCRT) – Deterrente as a Private Good » Development of private associations of prosecutors that internalize the costs without relying on government’s money. – Extreme punishment (either minor or capital punishment). – Compounding Felonies was efficient to guarantee “punishment”. – Extensive Use of Pardons » Avoid unnecessary punishment costs correcting for too severe sanctions imposed by juries; » Internalizes the externality of severe punishment; » Allows a choice between a fine and execution: it creates a market for punishment.
HISTORY AND INSTITUTIONS • France – Public enforcement due to centralized government since the 1550s. – Inefficient system of punishment, including an ineffective use of the galleys. – More imprisonment than capital punishment contrary to the English preference for extreme punishment.
HISTORY AND INSTITUTIONS • Problem: – According to Bruce Smith (UIUC), private enforcement in Britain is a myth... By Radzinowicz (1956) and Langbein (1976).
HISTORY AND INSTITUTIONS • If efficient, why public enforcement after the 1800s? (David Friedman, 1996, BULR) – Urbanization • Externalities are easier to internalize in rural societies.
– Centralization of power: rent-seeking at the expense of victims of crime. – Discontentment with private enforcement was misunderstood!!
LEGAL ORIGINS AND ENFORCEMENT • Paper by Glaeser and Shleifer , QJE, 2002 – Explain why England had local judges and juries whereas France had royal judges; – Explain Code Napoleon.
• Basic Theory: Control of Enforcers and Enforcement.
LEGAL ORIGINS AND ENFORCEMENT • England – Strong King (Henry II) – Weak local magnates
• France – Weak King – Strong local magnates
LEGAL ORIGINS AND ENFORCEMENT • Local judges/juries – Maximize social welfare Preferences of the people – Subject to contrains imposed by local magnates.
• Royal judges – Maximize the welfare of the king – Subject to protection of the king.
• Basic Theory: Civil law and law enforcement is the consequence of... – Need to protect enforcers – High ability of local magnates to subvert the law • ... And nevertheless they accepted a royal judge...
LEGAL ORIGINS AND ENFORCEMENT • Codification... – Need to control enforcers, – They do not aim at make adjudication less complex, but at restraining judges, • ... Very inefficient given that they are educated and appointed by the Government...
– Politicization of the judiciary • Undermines the security of property rights... • Less able to perform well as a transplant...
HISTORY AND INSTITUTIONS • Iceland (David Friedman, 1979) – Unification of Norway by the end of 800s led to massive imigration to Iceland; – A new political system was created: the King was replaced by an Assembly; – No feudalism was developed. – Relationships were contractual, not territorial.
HISTORY AND INSTITUTIONS • Courts deliver verdicts; • If against the defendant, he should pay a fine to the victim or the prosecutor who bought the case from the victim; • If he did not, case goes to court again and he will be declared an outlaw; • The killer of an outlaw could not be prosecuted.
HISTORY AND INSTITUTIONS • Killing was made up for by a fine; • For a murder a men could be outlawed even if willing to pay a fine; • Murder was a killing followed by an intent to conceal the body.
HISTORY AND INSTITUTIONS • It was private enforcement... – Because no public prosecution; – Because no punishment but only compensation to victims.
HISTORY AND INSTITUTIONS • Was it efficient? – Yes • Poor could sell prosecution to wealthy market for prosecution existed; • Nobody was impune; • Economically efficient punishment – Fines – A person unable to pay could be reduced to a state of temporary slavery until he had worked off his debt – Severe punishment only when probability of detection is quite low.
HISTORY AND INSTITUTIONS • Why collapsed after 1250s? – Concentration of wealth; – Kingship imposed by Norway.
OUTLINE 1. 2.
Determinants of the Structure of Law Enforcement. Why Criminal Law?
1. 2.
Criminal and Tort. Criminal, Tort and Administrative.
GENERAL STRUCTURE OF LAW ENFORCEMENT (from Shavell, 2004)
DETERMINANTS OF STRUCTURE OF THE LAW (from Shavell, 2004)
• Determinants of Optimal Stage of Intervention – Magnitude of Possible Sanctions for Deterrence – Prevention – Information about Acts: » When » Who … State… Parties
– Enforcement Costs
DETERMINANTS OF STRUCTURE OF THE LAW (from Shavell, 2004)
• Determinants of Optimal Form of Sanctions – Level of Wealth – Illegal Gain – Harm – Resources Devoted to Detection and Punishment » Enforcement Costs » Distribution of Costs
DETERMINANTS OF STRUCTURE OF THE LAW (from Shavell, 2004)
• Determinants of Public vs. Private Enforcement – Information is primarily private » Reporting » Compensation, Retribution
– Effort must be Expended to Identify Injurers » Information systems may constitute natural monopolies
CRIMINAL VS. TORT •
Economic Explanations for Crime or Tort:
– Harm caused by crimes is different from that of torts because… » » » »
… victimless crimes; … too many victims (collective action problem); … victims do not care about deterrence but compensation; … other external aspects of enforcement (shaping preferences);
– Political economy or rent-seeking theories:
• Groups shift the cost of deterring and punishing a certain act from private (tort) to public (crime). • The Government takes over the benefits from prosecution at the expense of victims (David Friedman, 1996, BULR).
– Criminals should be stigmatized, but not tortfeasors. • Why are punitive damages used in tort? • Are parking tickets imposing any kind of stigma?
– Criminals are uninsured, thus public intervention to guarantee compensation.
CRIMINAL VS. TORT •
Posner’s Typology of Crime (1985, CLR):
– Criminal Law deters pure coercive transfers of wealth, pure implying it is not an accident of productive activities. – Crimes are acts that aim at “market bypassing” • Acquisitive crimes bypass explicit markets • Crimes of passion bypass implicit markets
•
– Allowing coercion would create incentives for potential victims to spend heavily on selfprotection.
Problems:
– What is the difference between intentional torts and crime in a “market bypass approach”? – Why are some conducts crimes at common law (counterfeiting) and others made criminal by statute (tax evasion or price-fixing)? – Some crimes are voluntary exchanges, prostitution, gambling…
•
• Effect on third parties.
Criminal law is designed primarily for when tort law does not work… – For the poor… – The affluent are kept in line by tort law…
CRIMINAL VS. TORT •
Economic Explanations for Crime, Tort and Regulatory Violation: – Regulatory Enforcement is cheaper but less effective; – Forces production of information by victims and enforcers; – Generates “competitive enforcement” » Duplication of Costs » Deters Corruption » Avoids Error
University of Manchester School of Law
SESSION APRIL 17 Armando JimĂŠnez San Vicente
OUTLINE 1.
Advanced Topics on Law Enforcement
1. 2. 3.
Corporations. The Political Economy of Business Crime. Organized Crime.
THE CASE OF CORPORATIONS • Firms and Contracts as Nexus of Contracts: – “Cooperative Crime” – Allocation of Liability b/w Employee and Employer: » Irrelevant [Coase Theorem] » Relevant due to Transaction Costs [Agency Costs]
• Corporate Liability: Civil vs. Criminal • Social harm of Corporate Crime • Political Economy of Business Crime
THE CASE OF CORPORATIONS • Corporate Criminal Liability – Agency Costs and Moral Hazard: • Distortion of production.
– Reputation Costs; – Internal Punishment and Wealth Constraints: • Lower salaries.
– Internal Control: • Perverse incentives to collusion.
THE CASE OF CORPORATIONS • The Political Economy of Corporate Crime – Lobbying; – Corruption; – Easier access to legal experts; – Control information over activities; – Technology: • Free movement of capitals; • Globalization.
ORGANIZED CRIME • Buchanan’s idea: monopoly of bads. • Why not? – Barriers to entry and to exit; – Conglomerate effects: • Dumping • Legal business
– Path dependence
OUTLINE 1. 2. 3.
Economics of Criminal Law Principles Some Aspects of Criminal Procedure
1. 2.
Criminal Procedure as a Market System. Removal of Illegal Gains.
Competitive Federalism in Law Enforcement.
CRIMINAL LAW • Substantive Principles of Criminal Law – Deterrence and Crime Prevention – Criminal Intent – The defense of insanity – The defense of necessity
CRIMINAL LAW • Deterrence: – Multiple Offender Laws – Attempt – Conspiracy • More dangerous (Posner, 1985) • Cooperative behavior – Trust – Commitment
CRIMINAL LAW • Deterrence: – Aiding and Abetting • Knowledge of the purpose: reduction of information asymmetry • Costs of acquiring information
– Entrapment as defense to criminal prosecution.
CRIMINAL LAW • Criminal Intent – Distinguish intent from awareness. – Unless the criminal defendant confesses, his state of mind has to be inferred from actions… – Deterrence theory requires premeditation…
CRIMINAL LAW • Defense of Insanity – Cannot be deterred… – But what about incapacitation?
CRIMINAL LAW • Defense of Necessity – No question of incapacitation, – Lowers transaction costs… – Widespread use could bypass market transactions. – Example: Dudley and Stephens (1884). • One of them was near dead in any case, and that killing and eating save the other three men.
CRIMINAL PROCEDURE AS A MARKET SYSTEM • Ideal Criminal Procedure Market System: – Optimal pricing: internalize social harm; – Efficient trade-off between fines and severity; – Optimal allocation of resources given constrained budgets: • Price set by negotiations; • Feedback effect
CRIMINAL PROCEDURE AS A MARKET SYSTEM • Criminal Procedure Market System: – Prosecutorial Discretion: • European compulsory prosecution model is inevitably ineffective; • Rules produce behavior to escape clauses: outcomes are arbitrary and unexpected. • Market Failure objections: – Prosecutor does not seek to maximize social welfare; » Agency costs. – Results are inequitable. » The Role of ex post equality in defining fair treatment
CRIMINAL PROCEDURE AS A MARKET SYSTEM • Criminal Procedure Market System: – Efficient Prosecution: • • • •
Decision not to prosecute; Selective prosecution; Selection of charge; Upping the Ante.
– Domination of the Grand Jury • Limit private prosecution • Discovery
CRIMINAL PROCEDURE AS A MARKET SYSTEM • Criminal Procedure Market System: – Plea-Bargaining: • Price-establishing function at low cost • Market failure arguments: – Defendants’ lawyers will use it to undermine the interests of the accused. – Discrimination against poor defendants. – No voluntary sales by defendants of their rights. » Reward for pleading guilty or a penalty for insisting on trial?
CRIMINAL PROCEDURE AS A MARKET SYSTEM • Criminal Procedure Market System: – Efficient rules of plea-bargaining: • Rule 11 as a Statute of Fraud – Written contract.
• Plea-Bargains are Binding • Unconscionable Pleas • Broken Bargains – Deal is off.
• Involvement of the Judge in Bargaining – Should be limited because has less access to evidence than prosecutor.
CRIMINAL PROCEDURE AS A MARKET SYSTEM • Criminal Procedure Market System: – Sentencing Discretion: • Price discrimination; • Reduction of information costs; • Market Failure Objections: – Inequitable sentencing: » Ex ante expected penalties are the same; » Elimination of discretion does not reduce ex post inequality; » Why should criminals have a moral claim to equal treatment when the costs fall on the law abiding?
REMOVAL OF ILLEGAL GAINS • Puzzle: – It should have been done more frequently, – Not enough to deter.
• Should the penalty be based on the harm to the victim or the gain to the criminal? – Kaldor-Hicks approach: harm…
JURISDICTIONAL COMPETITION IN CRIMINAL JUSTICE • Competitive federalism. • Problems: – Displacing crime; – Displacing criminals.
• Regulating the Market for Criminal Justice – Race to Bottom: displace. – Race to Top: raise probability of detection. – Local solutions vs. central planners (federal criminal law).
University of Manchester School of Law
SESSION MAY 4 Armando JimĂŠnez San Vicente
OUTLINE The successfulness of a theory depends crucially on the empirical adherence. 1.
Empirical Analysis
1. 2.
Understanding Why Crime Fell in the 1990s (Levittâ&#x20AC;&#x2122;s paper) The Controversies
1. 2. 3. 4.
The Case of the Death Penalty; Effects of Concealed Weapons Laws; The Impact of Legalized Abortion on Crime; The Impact of Race on Policing and Arrests.
EMPIRICAL ANALYSIS • The decline of crime in the 1990s – The magnitude – The broad range – The universality – The persistent, continuous nature of the crime decline – Unexpected drop
(Levitt, 2004)
(Levitt, 2004)
(Levitt, 2004)
(Levitt, 2004)
(Levitt, 2004)
(Levitt, 2004) US 1995-1999
EUROPEAN UNION 1995-1999
HOMICIDE RATES
-28%
-4%
VIOLENT CRIME
-20%
+11%
BURGLARY
-19%
-14%
MOTOR VEHICLE
-22%
+7%
(Levitt, 2004)
(Levitt, 2004) • Strong Economy – Growth – Unemployment – Inequality
(Levitt, 2004) • Changing Demographics – An increasing share of the population is Black; – The baby boom is leading to a temporary increase in the number of teenagers and young adults.
(Levitt, 2004)
(Levitt, 2004) • Better Policing Strategies
– Examples: Guiliani’s NYC; Community policing; Boston’s multi-agency collaboration for gang violence. – Why not?
• Crime decline started in 1991, while Guiliani was elected in 1993; there is no obvious break in the trend. • Enormous growth of police force (45% between 1991 and 2001, 3 times more than national average). • NYC had abortion rates among the highest and was legalized (1970) before Roe v. Wade (1973).
(Levitt, 2004) • Gun Control Laws – Brady Act (1993): no statistical effect; – Gun buy-programs • Surrendered guns are those less likely to be used in criminal activities; • Replacement guns are easily obtained. • More stringent polices (DC 1976 ban on handgun acquisition, Chicago 1982 ban on handgun ownership): minor impact even in short-run.
(Levitt, 2004) • Concealed Weapons Laws – Argument: in face-to-face contact and for particular crimes, armed victims raise cost faced by a potential offender. – DEBATE: • Lott and Mustard (1996) YES • Black and Naggin (1998), Duggan (2001), Ayres and Donohue (2001) NO
(Levitt, 2004) • Increased Use of Capital Punishment – DEBATE • Ehrlich (1973, 1975): seven murders per execution • Levitt: believes if there is any effect, it is small.
(Levitt, 2004) • Crime Prevention Programs – No evidence of success; – Rehabilitation: ??
(Levitt, 2004) • Increase in the Number of Police – 14% per-capita increase in the 1990s; – Evidence: • Police is endogeneous variable; • Need for appropriate instrumental variables (Levitt, 1998): – Timing of elections for police hiring – Firefighters for changes in number of police
– DEBATE: The Impact of Race and Sex on Police Arrests and Crime • Lott (2000): Very important • Donohue and Levitt (2001): Not so much...
(Levitt, 2004) • The Rising Prison Population – 4 times more in 2000 than 1972; – Extremely expensive ($50 billion year)
(Levitt, 2004)
(Levitt, 2004) • The Receding Crack Epidemic – Emergence of crack cocaine: sales, low price, intense effect. – Also explains why crime did not fell in the 1980s: • Decline of adult crime rate were masked by sharply rising youth crime – Crack epidemic – Falling punishment in the juvenile justice system
(Levitt, 2004) • The Legalization of Abortion (Roe v. Wade) – Theory • Unwanted children are at greater risk for crime; • Legalized abortion leads to a reduction in the number of unwanted births.
– DEBATE • Donohue and Levitt (2001): Very important • Lott: Not so much...
(Levitt, 2004) • Predictions:
– Crime will decline further (on a lesser scale) • Effect of more police officers • Effect of legalized abortion
– However...
• Prison system has peaked due to state financial constraints; • Crack-related violence has reached steady-state.
– New factors:
• Coming of age of “crack babies”
CAPITAL PUNISHMENT •
DEBATE
• Ehrlich (1973, 1975): seven murders per execution. • A number of critics demonstrated the sensitivity of Ehrlich’s findings to minor changes in specification. • Similar debate in Britain (Wolpin, Pyle, Cameroon). • Leamer and McManus in the late 1980s argue that the evidence of deterrence is essentially a product of the researcher’s prior beliefs. • Ehrlich and Liu (1999): focus on 1940 and 1950, there is a deterrent effect. • A series of more recent studies that incorporate data from the 1990s find yet again a deterrent effect. But others do not. Measurement problems and econometric specification are at the heart of the debate. • Levitt: believes if there is any effect, it is small.
CAPITAL PUNISHMENT • Levitt’s argument against a substancial deterrent effect of capital punishment: – Rarity of executions and long delays in doing so...the probability is still below 1 in 200 (2% annual execution rate for those on death row, just twice the death rate from accidents and violence among all American men). – Even if Ehrlich’s optimistic empirical estimate would only eliminate 364 homicides per years, less than 1/20 of the observed decline.
CONCEALED WEAPONS LAWS • Concealed Weapons Laws – Argument: in face-to-face contact and for particular crimes, armed victims raise cost faced by a potential offender. – DEBATE over the enactment of Shall Issue Laws • Lott and Mustard (1996) YES • Black and Naggin (1998), Duggan (2001), Ayres and Donohue (2001) NO
CONCEALED WEAPONS LAWS • Under Shall Issue Laws: – A law that allows a citizen to carry a concealed handgun if h/she can demonstrate a need to a governmental official is a discretionary or “may issue” law; – A “shall issue” law is designated to eliminate discretion on the part of officials because permits have to be issued unless specific and easily verifiable factors say otherwise. – Permit cannot be issued for: • Individuals below 18 or 21; • Those who have a criminal record or a history of mental illness.
CONCEALED WEAPONS LAWS
CONCEALED WEAPONS LAWS
CONCEALED WEAPONS LAWS •
Chronology –
1997, John Lott and David Mustard, “Crime, Deterrence, and Right-to-Carry Concealed Handguns,” Journal of Legal Studies. •
–
1998, Dan Black and Daniel Nagin, “Do Right-to-Carry Laws Deter Violent Crime?” Journal of Legal Studies. •
–
Argued that if crime is increasing before and decreasing after the passage of the law, then the sum of the before vs after coefficients will be the same, “inverted V.” Re-estimated model using splines.Shall-issue laws reduce crime.
2000, John Lott, More Guns, Less Crime (AEI). •
–
No significant effect. Criticized Lott and Mustard for aggregating across time and across states. Used dummy variables for 5 years before and 5 years after the law date and compared the sum of the coefficients. No significant difference.
1998, John Lott, “The Concealed-Handgun Debate.” Journal of Legal Studies. •
–
Shall Issue Laws reduce crime. Based on a pooled time series and cross section data set for all 3000 counties in the U.S. from 1977 to1992.
Extended the sample to 1995. Estimated spline models to capture the inverted V. Shall issue laws reduce crime.
2002, John Lott, More Guns, Less Crime, Second Edition (AEI). •
Extended the sample to 1997. Re-estimated spline models. Shall issue laws reduce crime.
CONCEALED WEAPONS LAWS •
Chronology –
2003, Ian Ayres and John Donohue, “Shooting Down the ‘More Guns, Less Crime’ Hypothesis,” Stanford Law Review. •
–
2003 Florenz Plassmann and John Whitley, “Confirming ‘More Guns, Less Crime,” Stanford Law Review. •
–
Selection bias: original sample included only states that adopted in the 1980’s when crime peaked because of the crack epidemic. The post-crack decline in crime is captured by shall-issue variable. They estimated a ‘hybrid model’ with both a dummy and a spline, found that more states had significant increases than decreases. Shall issue laws increase crime (more often than they decrease crime).
Ayres and Donohue misread their own results. When the coefficients in their Table 10 are properly analyzed, they show that shall issue laws reduce crime. They also extend the data to 2000 and find that shall issue laws reduce crime.
2003 Ayres and Donohue, “The Latest Misfires in Support of the ‘More Guns, Less Crime’ Hypothesis,” Stanford Law Review. •
Forget Table 10, it was merely there to show how wrong it is to aggregate across states. Table 13 is the crucial table. Plassman and Whitley’s latest analysis is “fatally flawed” by coding errors. “The bottom line is that the best evidence suggests overall small increases in crime associated with adoption of concealed-carry laws.”
ABORTION AND CRIME •
The Legalization of Abortion (Roe v. Wade) – Theory
• Unwanted children are at greater risk for crime; • Legalized abortion leads to a reduction in the number of unwanted births.
– Evidence
• The five states (NY,W, Alaska, Hawaii and California after SC of California ruling of 1969) that allowed abortion in 1970 experience declines in crime earlier than the rest of US. • States with high and low abortion rates in the 1970s experienced similar crime trends for decades until the first cohort exposed to legalized abortion reached the high-crime ages around 1990. At that point, the high abortion states saw dramatic declines in crime relative to low abortion states over the next decade. • Panel data estimates confirm the strong relationship between lagged abortion and crime.
ABORTION AND CRIME
ABORTION AND CRIME
ABORTION AND CRIME
ABORTION AND CRIME •
Debate: – –
Evidence by Donohue and Levitt (2001) Lott and Whitley criticize: • • • • • •
–
Abortion may prevent the birth of "unwanted" children, who would have relatively small investments in human capital and a higher probability of crime. On the other hand, some research suggests that legalizing abortion increases out-of-wedlock births and single parent families, which implies the opposite impact on investments in human capital and thus crime. The question is: what is the net impact? They find evidence that legalizing abortion increased murder rates by around about 0.5 to 7 percent. Previous estimates are shown to suffer from not directly linking the cohorts who are committing crime with whether they had been born before or after abortion was legal. Though abortion was indeed legalized in New York, California, Alaska, Hawaii, and Washington prior to the 1973 Roe vs. Wade decision, there were also a sizeable number of abortions being performed in other states where abortion was legal for the life or "health" of the mother before 1973. Indeed, several of these states had abortion rates as high or higher than those states where abortion was legalized.
Joyce (2003): •
Demographers have concluded that most legal abortions in the early 1970's replaced illegal abortions. Since no one has data on illegal abortion rates, the results in Donohue and Levitt (2001) may be spurious.
ABORTION AND CRIME
ABORTION AND CRIME
ABORTION AND CRIME
THE IMPACT OF RACE AND SEX ON POLICING AND CRIME •
The Impact of Race and Sex on Policing and Crime:
– Lott (2000): female and minority police are less effective at reducing crime on average than white male police. – Donohue and Levitt (2001):
• Increases in the number of minority police are associated with significant increases in arrests of whites but have little impact on arrests on nonwhites; • More white police increase the number of arrests of nonwhites but do not systematically affect the number of white arrests. • Office race does matter. The normative implication is unclear:
– Restore greater equality in the likelihood of arrest conditional on the race or on unlawful conduct; – Greater harassment is being perpetrated with cross-race policing; – Crime is higher.