Excessive alcohol consumption costs the U.S. one-quarter of a trillion dollars each year, with alcohol-attributable crime accounting for one-third of these costs.  Problem drinking is particularly common among criminal justice populations with one-third of probationers and  nearly one-half of sentenced jail inmates meeting clinical criteria for alcohol abuse or dependence in the previous year (for the latter it was in the year before incarceration).  A range of scholarly studies suggest that alcohol-abusing populations have self-control problems in part due to strongly present-biased preferences, but an important challenge to monitoring consumption amongst this group is that alcohol passes through the body quickly.  For example, a 160-pound man who exceeds the legal drinking limit for driving after consuming five drinks in two hours will likely register a 0.00 in a breathalyzer test seven hours after drinking. 

There are a growing number of strategies targeted at deterring offenders from using alcohol, but rigorous empirical evidence on their effectiveness in reducing problem drinking and consequent economic costs is rare.  One approach is to monitor offenders’ alcohol consumption by ordering them to wear a Secure Continuous Remote Alcohol Monitor (SCRAM) bracelet, which monitors their alcohol intake up to every 30 minutes and submits the data to criminal justice oversight.  Roughly 175,000 people in 48 states have been ordered to this type of supervision since 2003.  Another approach is to combine frequent monitoring with swift, certain but brief sanctions to address the present-bias preferences among alcohol-abusing populations.

Between March 2006 and July 2011, 3,960 individuals in South Dakota were assigned to SCRAM either as part of a traditional oversight and sanctioning regime or as part of South Dakota’s innovative 24/7 Sobriety Project.  The 24/7 program utilizes the swift, certain, and moderate sanctions described above to deter offenders from consuming alcohol—if you skip or fail, you go to jail.  By increasing the immediacy and certainty of punishment, 24/7 seeks to achieve more comprehensive deterrence than the traditional enforcement model which typically delays sanctions.

In this paper, we test the hypothesis that swift and certain sanctions for a violation are more effective at creating a deterrent threat than traditional sanctioning regimes which prioritize sanction severity. Using a unique database from Alcohol Monitoring Systems, the company that invented SCRAM, this paper presents a survival analysis of drinking violations for the nearly 4,000 individuals ordered to wear a continuous monitoring bracelet. Of these individuals, approximately half were assigned to 24/7 and half were not.  In addition to the treatment assignment (24/7 vs. traditional sanctions), the analyses control for a range of individual and community factors including age, gender, prior criminal record, and the type of criminal offense resulting in SCRAM assignment. Community-level controls include the density of alcohol sellers, weather patterns, and socioeconomic characteristics.