Untapped Potential: An Empirical Analysis of the Effects of Adolescent Alcohol Consumption on Educational Attainment

Abstract

This paper examines the relationship between adolescent alcohol consumption and educational attainment. There is a division in prior research over the effects of adolescent alcohol consumption on education outcomes; for example, Dee and Evans (2003) find that alcohol consumption has no effect on the likelihood of graduation whereas Mullahy and Sindelar (1989) conclude that alcoholic behaviors reduce education attained by 1.5 years. We use OLS estimations to consider years of schooling and logit models observe likelihood of achieving education milestones, specifically graduation from high school and a 4-year college. Further, we contribute to the existing literature by decomposing alcohol use into various types of consumption habits to analyze which drinking patterns drive the relationship between alcohol use and education. Using The National Longitudinal Study of Adolescent to Adult Health, we find that frequency of drinking has the strongest negative influence on education, specifically those that consume alcohol three times per week or more complete roughly a half-year less of formal schooling, are 42% less likely to graduate from high school, and are 64% less likely to graduate from college. We also find that those who begin drinking at the youngest age (ages 5 to 12) are 26.8% less likely to graduate from college. Lastly, we find that those who consume eight or more drinks, 3 or more times per week face a 77.1% reduction in the likelihood of college graduation. Lastly, we consider a set of plausible assumptions surrounding omitted variable bias and conclude that our findings create the upper bound of the effect of alcohol consumption on educational attainment. Therefore, we conclude that the effects on total years of education are quite small, but the effects on the likelihood of college graduation likely remains substantial.