Argument: 21 drinking-age just transfers deaths to 21-24 age group
National Youth Rights Association. FAQ - as researchers Peter Asch and David Levy put it, the "minimum legal drinking age is not a significant-or even a perceptible-factor in the fatality experience of all drivers or of young drivers." In an in-depth and unrefuted study Asch and Levy prove that raising the drinking age merely transferred lost lives from the 18-20 bracket to the 21-24 age group. The problem with the 20,000 lives saved statistic is that it looks only at deaths for people aged 18-20. This is like rating the safety of a car by looking only at the seat belt and ignoring the fact that the car frequently tips over while driving. Raising the drinking age may have reduced deaths 18-20 but resulted in more deaths among people 21-24.
John J. Miller. "The Case Against 21. Lower the drinking age." National Review Online. April 19, 2007 - McCardell suggests that one effect of raising the drinking age was not to prevent deaths but merely to delay them. “The most common age for drinking-related deaths is now 21, followed by 22 and 23,” he says. “It seems that the minimum drinking age is as likely to have postponed fatalities as to have reduced them.”