Test scores and drinking age - a legitimate link?

Radley Balko thinks, and Glenn Reynolds agrees, that the US national drinking age should be lowered. One of the points raised was this:

If the research on brain development is true, the U.S. seems to be the only country to have caught on to it.

Oddly enough, high school students in much of the rest of the developed world — where lower drinking ages and laxer enforcement reign — do considerably better than U.S. students on standardized tests.

I'm not going to debate the drinking age right now, but I do think that there are plenty of reasons to be skeptical about the test score comparisons, skeptical enough that it smacks of dishonesty to include that in the debate.