The dangers of the drinking age (updated)

For the past 20 years, the U.S. has maintained a Minimum Legal Drinking Age of 21 (MLDA21). Recently more than 100 college and university presidents signed the Amethyst Initiative, a public statement calling for "an informed and dispassionate public debate over the effects of the 21-year-old drinking age." The reaction of many of those who support MLDA21 was to decry the statement for not recognizing that the MLDA21 saves lives by preventing traffic deaths among 18 to 20-year-olds. And who can argue with that?

Actually economists Jeffrey A. Miron and Elina Tetelbaum can. In a recent article at Fobes.com Miron and Tetelbaum write on The Dangers Of The Drinking Age. They note that the federal government in the US pressured states to raise the drinking age to 21 and then ask, So why didn't the move save lives? To understand why lives have not been saved, Miron and Tetelbaum start by explaining that a bit of history is useful. They write,

When the U.S. repealed the prohibition of alcohol in 1933, states were free to legalize, regulate or prohibit access to it as they saw fit. Most legalized but regulated it. In particular, 32 states adopted an MLDA of 21, while 16 chose an MLDA between 18 and 20. With few exceptions, these disparities persisted through the late 1960s.

Between 1970 and 1976, 30 states lowered their MLDA from 21 to 18. These changes coincided with other national efforts to enfranchise youth, exemplified by the 26th Amendment, which granted those 18+ the right to vote.

In 1984, however, Congress passed the Federal Underage Drinking Act (FUDAA), which withholds transportation funding from states that do not have an MLDA21. The justification given for the act was that higher MLDAs would result in fewer traffic fatalities.

By the end of 1988, after passage of the FUDAA, all states adopted an MLDA21. Several states had adopted an MLDA21 before the FUDAA, but the other states were less eager to change. Several passed MLDA21 legislation but set it up for repeal if the FUDAA were held unconstitutional. Others enacted "sunset provisions" in case federal sanctions expired. But when the Supreme Court upheld the FUDAA, states faced a strong incentive to maintain an MLDA21.

In recent research Miron and Tetelbaum compare traffic fatality rates in states before and after they changed their MLDA from 18 to 21. But in contrast to all earlier work they examined separately the impact in states that adopted an MLDA21 on their own and those that were coerced by the FUDAA. The Miron and Tetelbaum research shows that states that raised the drinking age to 21 since 1984, in response to FUDAA, enjoyed no statistically significant decrease in traffic fatalities for 18- to 20-year-olds. They point to the decades-long, steady decline in the rate of traffic fatalities (deaths per billion passenger miles), a decline due in large part to safer cars, improved driver education and better medical technology. Raising the drinking age did little or nothing. Miron and Tetelbaum explain that,

The results are striking. Virtually all the life-saving impact of the MLDA21 comes from the few early-adopting states, not from the larger number that resulted from federal pressure. Further, any life-saving effect in those states that first raised the drinking age was only temporary, occurring largely in the first year or two after switching to the MLDA21.

Our results thus challenge both the value of the MLDA21 and the value of coercive federalism. While we find limited evidence that the MLDA21 saves lives when states adopted it of their own volition, we find no evidence it saves lives when the federal government compels this policy.

This makes sense if a higher MLDA works only when state governments can set a drinking age that responds to local attitudes and concerns--and when states are energized to enforce such laws. A policy imposed from on high, especially one that is readily evaded and opposed by a large fraction of the citizenry, is virtually guaranteed to fail.

The major implication of these results is that the drinking age does not produce its main claimed benefit. Moreover, it plausibly generates side effects, like binge drinking and disrespect for the law--the very behavior that events planned for this month's alcohol awareness theme are designed to deter.

Offsetting behaviour.

(HT: Greg Mankiw)

Update: Matt Nolan discusses the article at TVHE here.