Correlation Between Crime and Punishment

Property crime reduces investment, because individuals resist investment, if criminals reap the returns. The time of criminals, which could otherwise be spent productively, is also a social loss.
These significant social costs imply that research on crime is among the most important areas in economics. Furthermore, as economics addresses issues beyond its conventional topics, work on economics and crime has indeed been among the most productive areas for research. Over the past 30 years since Becker (1968), there has been a surge of activity in trying to understand the economics of crime. Economists have brought tools such as hedonic real estimate work, using the differences in real estate costs between high crime and low crime areas to assess individual's willingness to avoid crime, and thereby to measure the economic costs of crime. But this research extends far beyond just estimating the costs of crimes. Economists have used sophisticated statistical techniques to estimate the extent to which increases in punishment deter crime. Economists have also investigated the link between crime and a rich variety of social factors.
In this brief essay, I review some of the recent work on the economics of crime and punishment. I specifically suggest ways in which the World Bank's exciting new initiative addressing crime can be connected with the most pressing issues in new research. The World Bank's focus on crime is important both because crime represents a major social problem, but also because most research on crime has heretofore been centered on developed countries. Developing countries present an even more pressing area for future research. Indeed, poorer countries often have the worst crime problems.
A new focus on crime research in developing countries offers the promise of both significant policy impact and sizable scientific relevance.
I begin this essay by reviewing the basic Becker (1968) model of crime and punishment.
While this model is well known to almost all economists, it may be useful to present the model to non-economists and policy makers to ensure that all readers are on the same page. A common set of terms and intellectual framework is valuable in understanding the determinants of crime.
In the next section of this essay, I review what I consider to be the three most important current questions in economic research on crime. The first section reviews recent advances in the efficacy of police action and incentives. Traditional economic research tended to look for a correlation between policing and crime. Unfortunately, causal links between crime and deterrence are almost always two sided. Often, researchers would find a positive link between the level of crime and the amount of deterrence, which are explained by the fact that localities often increase the amount of policing when the levels of crime rise. Recent research, particularly that of Steven
Levitt, has explicitly addressed this causality problem. Using a variety of clever statistical techniques,
Levitt has been able to identify the purest estimates of the causal link between deterrence and crime.
Ideally, similar estimates of the power of deterrence should be able to be done in developing countries as well.