In volume 30, issue 4 of this journal Bariş Yörük and Ceren Yörük (Y&EY) used data from the National Longitudinal Study of Youth, 1997 (NLSY97) and a regression discontinuity design to estimate the effect of the minimum legal drinking age on a variety of substances including marijuana.
They obtained evidence that the probability of marijuana use increased sharply at the age of 21, consistent with the hypothesis that alcohol and marijuana are complements, but inadvertently conditioned on having used marijuana at least once since the last survey.
Applying the Y&EY research design to all NLSY97 respondents ages 19 through 22, we find no evidence that alcohol and marijuana are complements.
Read Full Abstract
Request Reprint E-Mail: ben.crost@gmail.com