This paper, from the National Bureau of Economic Research (NBER), discusses the economics of crime and punishment, focusing on the optimal enforcement of laws. The author, George J. Stigler, argues that enforcement is costly and that society must balance the costs of enforcement against the benefits of compliance. He suggests that the goal of enforcement is to achieve a level of compliance that society can afford, given the costs involved. He also notes that punishments should be set to deter crime, but that the social value of the gain to offenders from the offense is a factor that is not well understood. Stigler discusses the supply of offenses, noting that they can be either production or consumption offenses. He argues that the structure of penalties and the probability of detection are important factors in determining the supply of offenses. He also notes that the enforcement of laws is not done by society, but by an agency that must be incentivized to enforce the law efficiently. Stigler also discusses the design of enforcement agencies, noting that they often fail to consider the costs they impose on the activity or persons regulated. He argues that the use of inappropriate methods of determining the extent of enforcement is another deficiency in the design of enforcement. The paper concludes that the widespread failure to adopt rational criteria of enforcement is often due to a lack of understanding of the need for and nature of rational enforcement. However, there is also a second reason for the use of inappropriate sanctions and appropriations: the desire of the public not to enforce the laws. The paper suggests that variation in enforcement provides desirable flexibility in public policy.This paper, from the National Bureau of Economic Research (NBER), discusses the economics of crime and punishment, focusing on the optimal enforcement of laws. The author, George J. Stigler, argues that enforcement is costly and that society must balance the costs of enforcement against the benefits of compliance. He suggests that the goal of enforcement is to achieve a level of compliance that society can afford, given the costs involved. He also notes that punishments should be set to deter crime, but that the social value of the gain to offenders from the offense is a factor that is not well understood. Stigler discusses the supply of offenses, noting that they can be either production or consumption offenses. He argues that the structure of penalties and the probability of detection are important factors in determining the supply of offenses. He also notes that the enforcement of laws is not done by society, but by an agency that must be incentivized to enforce the law efficiently. Stigler also discusses the design of enforcement agencies, noting that they often fail to consider the costs they impose on the activity or persons regulated. He argues that the use of inappropriate methods of determining the extent of enforcement is another deficiency in the design of enforcement. The paper concludes that the widespread failure to adopt rational criteria of enforcement is often due to a lack of understanding of the need for and nature of rational enforcement. However, there is also a second reason for the use of inappropriate sanctions and appropriations: the desire of the public not to enforce the laws. The paper suggests that variation in enforcement provides desirable flexibility in public policy.