THE EFFECT OF RECREATIONAL MARIJUANA LAWS ON TRAFFIC FATALITIES AND CHANGES IN ALCOHOL CONSUMPTION

Abstract

This paper investigates the implementation of recreational marijuana laws on the number of traffic fatalities, the leading cause of death amongst Americans ages 16-25, in the United States by using cross-sectional data from the Fatality Analysis Reporting System (FARS) covering the years 2012 to 2014. To date, four states and the District of Columbia have legalized marijuana for recreational use. However, little is known about the effects of recreational marijuana laws and how the effects may differ from those associated with medical marijuana laws. Using the difference-in-differences method gives estimates of great magnitude and showed that the legalization of marijuana for recreational use is associated with a 22.1 percent decrease in the number of traffic fatalities per 100,000 people, across all age groups. This result is not only large in magnitude but also statistically significant at conventional levels. The impact of legalization on traffic fatalities involving young people (aged 15-19) is larger in magnitude and estimated with more precision than the impact of legalization on traffic fatalities on other age groups. Legalization is associated with a 43.3 percent decrease in the number of fatalities involving 15 to 19 year olds per 100,000 people. Although the results were statistically insignificant at conventional levels, the negative relationship is relatively constant throughout and the largest effect is seen in young people for all regressions and estimated with the greatest precision relative to other age groups. Legalization is also associated with a much larger decrease in traffic accidents involving alcohol than in traffic accidents involving drugs, which suggests that marijuana and alcohol are substitutes. Legalization is associated with a 12.3 percent decrease in the number of traffic fatalities involving alcohol per 100,000 people and a 2.62 percent increase in the number of traffic fatalities involving drugs. These results were statistically insignificant but the magnitudes and direction of the effects signify that the reduction in the number of traffic fatalities is due to a decrease in alcohol consumption and an increase in consumption of marijuana, indicative of the substitution effect. Additionally, data from the Behavioral Risk Factor Surveillance System on alcohol consumption provided further evidence for this conclusion. Legalization is associated with an 11.8 percent decrease in the number of drinks consumed across all age groups. Additionally, the largest decrease in alcohol consumption is again in the youngest age group (those aged 15 to 19 years old): legalization is associated with a 13.9 percent decrease in the number of drink consumed for the youngest age group. Moreover, these results are statistically significant at conventional levels in addition to being substantially large in magnitude. However, alternative mechanisms cannot be ruled out and this paper does not attempt to attribute this negative relationship between legalization and alcohol-related traffic fatalities to the fact that it is safer to drive under the influence of marijuana than under the influence of alcohol.