Philip J. Cook of Duke University and Jens Ludwig of the Brookings Institution draw on the work of UCLA’s Mark Kleiman and authors to frame a more cost-effective anti-crime strategy, which relies less on long prison sentences and more on “swifter, surer, but more moderate punishment.” If you are very unlikely to be caught for committing a crime, you might be willing to chance it if the return is sufficiently high, even if the punishment is extreme. This is particularly true if your other employment opportunities aren’t so hot. But if you are very likely to be caught, the calculation changes, even if the punishment isn’t quite so terrifying. For a number of reasons, we’ve gotten the balance wrong. One of the obvious reasons is that an expanding prison system has generated very visible employment opportunities for the kind of people who vote, e.g., organized public workers, employees of private prisons, etc. Reducing crime through less punitive strategies would produce vastly more private wealth, but the resulting job creation would be widely dispersed across the economy, not concentrated in a fairly discrete political constituency.

Consider the following brief excerpt:

Much of the reluctance to reduce the prison population reflects a belief that the extraordinary reduction in crime that occurred in the 1990s was caused by a surge in imprisonment. But even a casual look at the actual statistics challenges the view that prison trends get all or most of the credit for the crime drop.
Looking at three periods from recent history, we see that the crime drop of the 1990s did coincide with a large increase in the prison population. But the large crime increase during the prior period was also associated with a jump in imprisonment – and so was the relatively static crime pattern since 2000. If the prison surge of the 1990s gets credit for the crime drop, then fairness requires that the prison surge of the 1980s gets the blame for the crime increase of that period, while the prison increase of the 2000s was largely irrelevant. This type of armchair analysis supports almost any conclusion.
Studies suggest that increased use of imprisonment indeed should receive part of the credit for the crime drop of the 1990s, in the sense that crime was lower than it would have been had we taken all the funds devoted to prison increases and spent it for purposes other than crime control. But is that the right counterfactual? If the vast increase in prison expenditures came at the expense of alternative crime-control efforts that might be even more effective, then the net effect of the imprisonment boom is not so clear, even qualitatively.

The authors also offer thoughts on “private co-production” of crime prevention:

Most of the research on crime control strategies focuses on the role played by government and non-profit interventions. But private citizens and businesses account for a surprisingly large share of resources devoted to preventing crime. State and local governments can help reduce crime indirectly by encouraging private actions that make law enforcement more productive. Two examples for which benefits exceed costs by an order of magnitude are building the police-tracking infrastructure for Lojack, and creating the legal framework for Business Improvement Districts (where local businesses are subject to tax payments that go in part toward making the neighborhood clean and safe).

The paper is well worth your time, and of particular interest to conservatives interested in reducing the burden of wasteful spending.