Harold Pollack posts the following image from “The effect of alcohol consumption on mortality: Regression discontinuity evidence from the minimum drinking age,” a new paper by Christopher Carpenter and Carlos Dopkin:
The red line represents days drinking per month. The black line represents the fatality rate. The horizontal axis is age. There’s a discontinuity at 21, suggesting that legalized drinking immediately increases deaths. I don’t have access to the paper, but here’s Pollack’s interpretation:
Young people’s alcohol consumption increases by over 20 percent as they hit their 21st birthday. Meanwhile, death rates increase by 9 percent exactly at age 21. Carpenter and Dobkin traced this further, finding that the mortality jump was largest for motor vehicle accidents, suicides, and other causes plausibly linked with alcohol use. The correlation isn’t a slam dunk, but it is close. The authors estimate that reducing the minimum drinking age by one year–as some propose–would cause 408 additional deaths every year among 20-year-olds.
That might be true, though I question the accuracy of self-reported drinking days among the under-21s. More importantly though, this graph strongly suggests that our current neoprohibitionist policies leave 21-year-olds woefully ill-equipped to enjoy alcohol safely. It’s likely the case there will be a increase in fatalities at any drinking age. That’s not surprising given that alcohol is a dangerous product when it’s abused. The question is how we teach young people to drink responsibly.
I don’t know that there’s an easy answer to that. A combination of lowering the drinking age so that teens can learn to imbibe with adult supervision and a greater emphasis on driving enforcement is one appealing alternative. That’s not going to change American drinking culture over night, but absent those changes it’s hard to envision better practices developing.
For more, here’s Radley Balko on the Amethyst Initiative and the case for lowering the drinking age.