Cross-Faded: An Econometric Analysis of Marijuana Legalization, Marijuana Consumption, and Binge Drinking

Abstract

This study analyzes the relationship between marijuana-use and alcohol-consumption by examining changes in the expected probability that individuals engage in binge drinking following the legalization of both medical-use and recreational-use marijuana in their respective states. The study observes changes in marijuana consumption among the full sample and a sub-sample of marijuana users, by observing general-use and frequency of use, following medical and recreational legalization at the state level, while also accounting for respondents’ preferred method of marijuana-consumption.

The analysis utilizes data collected from the Behavioral Risk Factor Surveillance System (BRFSS) over two time periods: 2006-2016 and 2016-2017. The models developed herein were constructed using OLS estimation strategies with State and Time Fixed-Effect and a Two-Staged Least-Squares Model. These models were used to measure changes in the expected probability that individuals engage in binge drinking and marijuana consumption as well as a continuous measure of the frequency of marijuana consumption.

These models show that recreational-use legalization is associated with an 8.6 percentage point (pp) increase in the probability of being a marijuana-user among the full sample and a 3.083-day increase in expected use among marijuana-users. However, medical-use legalization showed no significant changes to the general consumption of marijuana or frequency of use among the full sample and marijuana user sub-sample. Using a measure of the number of days of marijuana consumption in the past 30-days, preference of edibles is associated with a 7-day reduction in expected-use among marijuana users, and preference of dabs was associated with a 7-day increase in expected consumption among marijuana users.

When looking at binge-drinking behavior, we find that marijuana users have a 28.6pp increase in the probability of engaging in binge-drinking. Additionally, medical-use legalization is associated with a 3.67pp increase in the likelihood a respondent engages in binge-drinking among the full sample and an 8.5pp increase among a sub-sample of marijuana users. Recreational-use legalization is associated with a 5.23pp reduction in the likelihood a respondent engages in binge-drinking among the full sample and an 8.05pp decrease among a sub-sample of marijuana users.

The results of the paper find evidence to support the claim that heavy marijuana users treat marijuana and alcohol as substitutes; whereas light-to-medium users treat alcohol and marijuana as complements.