Social Control Through Deterrence: Drinking-and-Driving Laws

The anti-drunk-driving movement has resulted in a profusion of deterrence-based legislation, providing the opportunity to test the deterrence proposition via quasi-experiments employing high-quality data series. Considerable evidence bearing on this proposition has accumulated to date. Efforts directed mainly at increasing potential drunk drivers’ perceived certainty of punishment frequently have a deterrent effect in the short run. In the long run, however, i.e. over several months or a few years, indexes of drunk driving return to prior levels. This phenomenon may be explained by the very low actual probability of punishment. Efforts directed principally at increasing the perceived severity of punishment have not had the desired deterrent effects, perhaps because of the low levels of certainty that these punishments will be applied. The utility of deterrence-based laws appears to be limited owing to the cost of raising the actual likelihood of punishment to a point that would support a perception of reasonable certainty. Alternative policies based on changing institutions in the fields of recreation and transportation are more promising, though politically problematical. Most promising, in terms of feasibility and effectiveness, are technological measures that prevent driver errors from resulting in crashes or that reduce the forces experienced in those crashes that do occur.