The fatal toll of driving to drink: the effect of minimum legal drinking age evasion on traffic fatalities.

There is a sizeable literature on the effect of minimum legal drinking age (MLDA) restrictions on teenage drunk driving. This paper adds to the literature by examining the effect of MLDA evasion across states with different alcohol restrictions. Using state-of-the-art GIS software and micro-data on fatal vehicle accidents from 1977 to 2002, we find that in counties within 25 miles of a lower-MLDA jurisdiction, a legal restriction on drinking does not reduce youth involvement in fatal accidents and, for 18 and 19-year-old drivers, fatal accident involvement actually increases. Farther from such a border, we find results consistent with the previous literature that MLDA restrictions are effective in reducing accident fatalities. The estimates imply that, of the total reduction in teenager-involved fatalities due to the equalization of state MLDAs at 21 in the 1970s and 1980s, for 18-year olds between a quarter and a third and for 19-year olds over 15 percent was due to equalization. Furthermore, the effect of changes in the MLDA is quite heterogeneous with respect to the fraction of a state's population that need not travel far to cross a border to evade its MLDA. Our results imply the effect of lowering the MLDA in select states, such as has been proposed in Vermont, could lead to sizeable increases in teenage involvement in fatal accidents due to evasion of local alcohol restrictions.