The key conclusion to be drawn from this research is that increases in the full price of alcohol—whether they result from increases in monetary price, reduced availability, or increases in the expected legal costs of drinking and driving (i.e., more severe drunk-driving laws)—can reduce drinking and driving and its consequences among all age groups. As is the case with the effects of beer taxes on consumption, however, the estimated magnitude of the beer tax effects on motor vehicle mortality depends somewhat on whether State-fixed effects are included in the statistical models used. For example, Saffer and Grossman (1987a) reported much greater tax effects when using a fixed-effects model. Similarly, significant tax effects in the studies by Ruhm (1996), Dee (1999), Mast and colleagues (1999), and Dee and Evans (2001) also were based on this specification.
