Criminal deterrence research at the outset of the twenty-first century

Evidence for a substantial deterrent effect is much firmer than it was two decades ago. However, large gaps in knowledge on the links between policy a ctions and behavior make it difficult to assess the effectiveness of policy options for deterring crime. There are four major impediments. First, anal yses must estimate not only short-term consequences but also calibrate long -term effects. Some policies that are effective in preventing crime in the short term may be ineffective or even criminogenic in the long run because they may erode the foundation of the deterrent effect-fear of stigmatizatio n. Second, knowledge about the relationship of sanction risk perceptions to policy is virtually nonexistent; such knowledge would be invaluable in des igning effective crime-deterrent policies. Third, estimates of deterrent ef fects based on data from multiple governmental units measure a policy's ave rage effectiveness across unit. It is important to understand better the so urces of variation in response across place and time. Fourth, research on t he links between intended and actual policy is fragmentary; a more complete understanding of the technology of sanction generation is necessary for id entifying the boundaries of feasible policy.