The past decade's research, whether based on aggregate statistics, results from employment programs, or direct observation, tempers hopes for a direct and easily managed relationship between crime and employment. A tempered perspective, however, should not hold that no relationship exists between economic status and crime, but rather that the relationships are complex and indirect as well as direct. Effective policy and programs in this area require identifying both the direct and indirect links between employment opportunities and crime and assessing the quantity and quality of jobs in the labor market, the degree to which entry-level employment is linked to a future career ladder, the social supports afforded employment in local neighborhoods, and the wage levels and duration of jobs available either in programs or after program participation. Mixed results from broadly conceived programs should direct policy to a more narrow targeting of programs toward particular populations influenced by specific social conditions and moving within certain labor markets. Such programs should also have realistic expectations about what they can and cannot do. Some have taken the poor results of some job programs and recommended discarding such efforts in favor of increasing punitive measures that it is hoped will deter crime. Deterrence measures, however, are not likely to succeed where some positive alternative to crime is not offered. Efforts must be directed toward improvement in the labor market prospects of those whose behavior society is trying to control.