Prison releases are at an all-time high in the United States, and many of those leaving prison are looking for jobs just as the country is recovering from the worst economic crisis since the Great Depression. What to do? Some on the left call for public job creation programs, while voices on the right suggest mandatory work programs for parolees and men with unpaid child support. Beyond the usual partisan divides, work remains an appealing possibility for reducing social ills such as repeat crimes and drug use.

But do job creation programs work? Past projects have sometimes been labeled failures because they did not magically catapult poor people with problematic histories into sobriety and middle-class success. That is an unrealistic expectation, but perhaps jobs programs can make some measurable difference, such as lowering crime.

To learn more, we have turned to data from the National Supported Work Demonstration Project, a large program conducted between 1975 and 1978, in which former drug users with a history of incarceration were randomly assigned to publicly supported jobs or a control group. Using statistical techniques, we have been able to tease out the specific effects of supported work. In addition, we have interviewed people leaving chemical dependency treatment programs in Minnesota. Our guiding questions are straightforward: Does supported work reduce serious crime and drug use? If so, is it the money that matters or something else?

Money Legally Obtained Reduces Crime

About two-thirds of people released from U.S. prisons commit new crimes within just three years. Former prisoners are not attractive to private employers, yet when they cannot earn wages, they still need money. Nine of every ten crimes tracked by the Federal Bureau of Investigation involve some kind of financial payoff. If people leaving prison were assigned to basic jobs that paid a regular income, perhaps they would not commit new crimes as readily. The importance of regular work may go beyond the paycheck, because employment offers a person the opportunity to interact regularly with law-abiding fellow employees.

Our analysis of the experimental data from the 1970s shows that providing a basic job opportunity does, in fact, reduce the likelihood that heavy substance users will commit serious crimes to steal money from others.

• People assigned to a group given supported jobs were 39% less likely than those not assigned work to be arrested for robberies or burglaries.
• Much of the reduced chance of arrest for these crimes was because of the income people got from their regular jobs—or from some other legal source (such as welfare payments).

Does Work Reduce Drug Use, Too?

Drug use and alcoholism are common among prisoners, affecting more than half. Does work reduce drug use as well as crime? Previous studies have revealed that legally earned income can cut both ways. Work may draw a former substance abuser into sober social networks, and thus make it easier to stick it out in drug treatment programs. But income from a job could also provide the means to support an expensive drug habit. In turn, drug use may make someone a less effective worker, lead to lower or interrupted wages, and thus make a turn to crime more likely.

Our findings using the 1970s experimental data reveal that providing a supported work job did not lead to reduced drug use for the workers—but work was not associated with increased drug use, either. To understand why, we turned to interviews with nearly 30 young adults leaving chemical dependency treatment in Minnesota during the summer and fall of 2007. Talking to people struggling right now with how to work, avoid crime, and escape drug problems provided us with new insights into the complex processes at work:

• Income does affect crime and drug use. When someone has less legitimate income, he or she may very well turn to crimes like robbery and burglary to support the cost of drug use. However, higher regular income can also increase drug use if wages are used to buy drugs.
• For some of those we spoke with, the social bonds and regular daily schedule provided by work inhibited drug use and the temptation to engage in crimes. This meshes with what we found in close probes of the 1970s data, where income from work reduced drug use more than income from welfare. The social ties involved in regular work seem to matter even beyond the money earned.
• A few people claimed to be “productive addicts,” able to hold down quality jobs while also using drugs like heroin and methamphetamines. But if drug use escalates, it soon interferes with effective work.

The Bottom Line: Jobs Programs Make a Difference

Our results pin down real benefits from providing jobs to citizens grappling with multiple barriers to employment. Supported employment programs for heavy substance users represent a promising model for reducing serious economic crimes such as robbery and burglary.

Broader supported jobs programs might be a good idea, too, in this era of high unemployment. If publicly supported jobs programs were available to needy unemployed people beyond the ranks of former prisoners and drug users, the programs would carry lower stigma. A mix of jobs could be offered, creating ladders from minimum-wage posts to better opportunities for all participants.