The most incisive guide to issues facing the American family today . . . An invaluable resource for anyone wishing to stay on the cutting edge of research on family trends.

-W. Bradford Wilcox
Associate Professor of Sociology, University of Virginia 

A Risky Proposition

Bryce J. Christensen and Robert W. Patterson

When thirty-nine states lowered the drinking age during the 1970s, the country experienced an increase in automobile fatalities. Congress responded with the National Minimum Drinking Age Act of 1984, which prompted those states, using the carrot of federal highway funding, to reset the age back to 21.

Even as at least half of all undergraduates are not legally permitted to imbibe, many institutions of higher education act as if the drinking age doesn’t apply to their campus. Not only is there very little effort to curb chug-a-lugging among their students, but also more than a few college presidents have called for lowering the drinking age to 18, claiming the age change would result in “more responsible” drinking among students. These are the same administrators that years ago pushed for coed dorms, seemingly unaware that such living arrangements foster an “Animal House” atmosphere that not only promotes drinking but also the related promiscuous “hookup” culture, each of which places young coeds (and their future marriages) at risk.

Yet a study by economists Christopher Carpenter and Carlos Dobkin of the University of California (at Irvine and Santa Cruz, respectively) and the National Bureau of Economic Research exposes the folly of high-minded attempts to “reassess” (meaning lower) the drinking age. Among their findings, they quantify the extent to which a lowered drinking age would boost alcohol consumption and all its attendant consequences. Using a “panel fixed-effects” methodology with data on the drinking habits of high-school seniors as reported by the Monitoring the Future study (1976–2003), they calculated “that allowing 18–20 year-olds to drink increases drinking participation [or having a drink in the past month] by 6.1 percentage points, heavy episodic drinking by 3.4 percentage points, and instances of past month drinking by 17.4 percent.” The coefficient in each of these categories was statistically significant (p<.01).

A “regression-discontinuity” approach using data from the National Health Interview Survey (1997–2005), measuring the effect of turning 21 on alcohol consumption, yielded similar results. This methodology found that becoming legally eligible to drink increases the probability of drinking participation by 6.1 percentage points (p<.05), heavy episodic drinking by 4.9 percentage points (p<.10), and instances of past-month drinking by 19.6 percent (p<.05).

As might be expected, lowering the drinking age would exact a huge social toll. After reviewing what is already known about the relationship between drinking and fatalities, injuries, and crime, the two economists again use both panel fixed-effects and regression-discontinuity methodologies to estimate the anticipated negative outcomes. Their results are sobering: “If the drinking age were lowered to 18, there would be an additional 8 deaths for 100,000 person-years for the 18–20 age group.” But the effects are not limited to this age group. “Our best estimate is that the typical young adult killed while driving drunk kills another person 21 percent of the time,” meaning that lowering the drinking age will kill at least an additional 0.77 people annually for every 100,000 young adults, ages 18–20, that were allowed to drink.

Additional external costs involve the increased risk that a drinker will commit a crime or generate health-care costs. Here, the researchers cite their early study finding 63 additional arrests and 8 additional robberies for every 100,000 newly minted drinkers, as well as 408 additional emergency-room visits and 77 additional hospital stays per 100,000 person-years that would occur with a lower drinking age. Then after calculating the financial costs of all these harms (and deaths) on a per-drink basis of nearly $18 (in addition to the cost of the drink), Carpenter and Dobkin believe that most 18–20 year-olds would not be aware of the personal costs imposed by their drinking behavior nor the huge costs their drinking imposed on others, if the legal age were lowered.

While the study does not specifically address the impact of a lowered drinking age on the well-being and sexual health of young women, its findings suggest that the impact would also be negative. So the full import of this study offers more reasons than one to keep the legal age at 21.

(Christopher Carpenter and Carlos Dobkin, “The Minimum Legal Drinking Age and Public Health,” Journal of Economic Perspectives 25.2 [Spring 2011]: 133–56.)