Should the Minimum Drinking Age Be Lowered?

Lots of articles coming out this week because of the release of a report of sorts urging for a lowering of the mandatory drinking age from 21 to 18. It was signed by around 100 college presidents, who believe that doing so would reduce binge drinking on college campuses. You can read the the Amethyst Initiative's statement yourself here.

I'm fairly skeptical that reducing the drinking age would do anything to reduce binge drinking, to be honest. In his 2007 book, Paying the Tab: the Costs and Benefits of Alcohol Control, Duke economist Philip Cook who spent part of his career studying alcohol control policies writes the following.

"Some of the strongest evidence we have on the effects of alcohol-control measures arises from the `laboratory' of the states. The Twenty-first Amendment ensures that each state establishes its own alcohol-control policy. ... The quasi-experimental method has been used most prominently and persuasively in the case of the minimum drinking age. As of 1970, most states had adopted twenty-one as the minimum age for legal purchase of any sort of alcoholic beverage ... by 1988, every state had established twenty-one as the minimum age. The frequent changes in law generated a great deal of evidence on the consequences of the minimum legal drinking age (MLDA). ... The most important potential effect of these changes was with respect to traffic fatalities involving youthful drivers. A decrease in the MLDA from twenty-one to eighteen could logically increase traffic fatalities by increasing the amount of drinking that was coupled with driving. ... States that lowered their MLDA experienced an increase in youthful highway fatalities relative to states that did not.

The entire book is a good one to read for those interested in the Amethyst Initiative's statement, but it's particularly relevant for this debate because it summarizes all of the research done on MLDA. In essence, the evidence that we have suggests the following: a reduction in MLDA from 21 to 18 would likely increase traffic fatalities, and possibly other risky behaviors associated with alcohol like risky sex. Chesson, Harrison and Kassler found that MLDA was negatively correlated with syphilis and gonorrhea incidence among youth, and more recently, Angela Fertig and Tara Watson find evidence linking MLDA to poor child health outcomes. Lower minimum age drinking laws were associated with higher incidence of low birth-weight babies, premature births, and unplanned pregnancies.

As Cook shows in his book, there has been a mountain of studies done on this, and the overwhelming conclusions from them suggests that lowering the MLDA from 21 to 18 would have significant social costs. Something we libertarian-sorts are reluctant to face, I think, is that alcohol and illicit drugs have externalities. They are associated with more driving fatalities, for instance, as well as bad outcomes for babies. In other words, the full costs of drinking are not absorbed by the drinker - they "spill over" (lol!) onto others not involved in that calculus. Alcohol control should be focused, imo, on greater restrictions, not fewer, making the Amethyst Initiative's statement a troubling one.