Check out this Policy Review piece on the 1) ineffectiveness of drug prevention programs, and the 2) wild resistance towards any empirical assessment of such ineffectiveness. When I think of all the laws that have counterproductive effects (see my previous posts here and here), I'm on the verge of believing that if a politician wants to help any particular cause or group of people, the best way to do it might be to vote against any piece of legislation that is explicitly announced as helping that cause or group of people. This principle wouldn't work in all cases -- some laws do accomplish their intended effects, at least in part -- but it might be just as useful a heuristic as its opposite.