lated by most states. However, at least eight state legislatures have prohibited random testing of employees in positions that are not safety-sensitive, and a few state supreme courts have embraced a similar approach in the absence of legislative action. As indicated, however, the law in most states leaves employers free to adopt a random testing program for all employees, and it appears that many employers have done so.
The question of interest is whether the expanding practice of workplace drug testing deters drug use by employees.3 In 1994, the National Academies’ Committee on Drug Use in the Workplace concluded that the preventive effects of drug testing have never been adequately demonstrated and that there existed no conclusive “scientific evidence from properly controlled studies” that employment drug testing programs widely discourage drug use or encourage rehabilitation. Subsequent research has provided some additional evidence bearing on the deterrent effect of drug testing in the military, but this evidence does not cast much light on the effect of testing in civilian employment.
In the U.S. military services, random testing is required of the entire workforce and, since 1995, all services have automatically imposed a severe sanction (discharge) on all violators. Two recent studies yield some evidence that the military’s zero-tolerance drug policies, including routine drug testing (adopted in 1980), have deterred illegal drug use among enlistees. One study, by Bachman et al. (1999), based on the longitudinal panel data from the Monitoring the Future survey, tracked cohorts of seniors (classes of 1976 to 1995) for two years after graduation, and compared active-duty recruits with nonmilitary classmates who entered college and civilian employment.
The study found that the prevalence of marijuana or cocaine use before graduation was about the same among seniors who chose to enlist in the military as it was among those who chose to go to college or enter the civilian workforce. However, during the follow-up interview two years
There are other goals that could be served by employee drug testing, including screening employees to identify applicants or employees at high risk for impaired performance or reduced productivity. However, a 1994 National Academies report found that “clear evidence of the deleterious effects of drugs other than alcohol on safety and other job performance indicators” is lacking (National Research Council, 1994:107). Random testing is a costly and imperfect method of identifying poorly performing employees (Rothstein, 1991). Direct testing of performance, for-cause testing of apparently intoxicated workers, or— possibly—random testing of employees in safety-sensitive positions would be much more cost-effective.