The effectiveness of alcohol interlocks in reducing repeat drink driving and improving road safety
By Sara Rahman
AIM To evaluate the impact of the first phase of the Mandatory Alcohol Interlock Program (MAIP), introduced in NSW in February 2015, on repeat drink-driving, driving while disqualified, traffic infringements, and crash outcomes. METHOD We use a dataset of 98,501 observations involving offenders with a proven ‘exceed the prescribed concentration of alcohol’ (PCA) offence or a ‘refuse to provide a breath sample’ offence finalised in a NSW court between 1 June 2012 and 30 April 2018. We identify the impact of taking up an alcohol interlock on reoffending and road crashes using a regression discontinuity design. This analysis compares outcomes for first-time PCA offenders in a small bandwidth on either side of the high range blood alcohol concentration (BAC) threshold of .15. We also estimate the overall impact of the introduction of MAIP on drink driving and road crashes using a difference-in-differences approach. This compares outcomes for eligible and ineligible offenders before and after the introduction of the program. We implement a range of robustness checks and analyse outcomes for various groups of offenders, including those receiving different interlock periods, and those with different demographic characteristics and criminal histories. RESULTS MAIP reduced the likelihood of drink driving during the interlock period for first-time high range PCA offenders who start the program by 11 percentage points (p.p.; a reduction of 86%) compared to mid range offenders just below the high range PCA threshold. We also observe reductions in PCA offending within 36 months of finalisation (of 3.4 p.p.; 43%) and within 60 months of finalisation (of 6.0 p.p.; 43%) among all eligible offenders compared to offenders committing eligible offences before the program’s introduction. The program reduces traffic infringements committed after court finalisation, but these effects are concentrated among repeat low range PCA drink drivers. We do not find significant effects of the program on reducing the likelihood of involvement in an alcohol-related crash nor on crashes resulting in injuries and fatalities. The reductions in PCA offending are particularly large for offenders convicted of the most severe offences (i.e., repeat high range drink driving and repeat refuse to provide a breath sample offenders), those residing in disadvantaged areas, and to a lesser extent, those residing outside major cities. CONCLUSION Alcohol interlocks significantly reduce drink driving while interlocks are active and (to a modest extent) following their removal.
(Crime and Justice Bulletin No. 251), Sydney: NSW Bureau of Crime Statistics and Research. 2022. 35p.