Lots of people think the War on Drugs is useless, claiming that it's unjust and doesn't do much to reduce drug use. It may very well be unjust, but Victor Morton has some information on the effectiveness of Prohibition in the 1920s and how it reduced alcohol consumption, and I think there's a strong parallel. I asked him for hard evidence, and here's his response (which he kindly allowed me to post here):
Well, a lot of guesswork is involved for any number of reasons. First of all, figures for any illegal activity are necessarily unreliable. There were also great gaps in the availability of alcohol between urban areas, especially in the Midwest and Northeast (where the evidence is good that Prohibition did fail to restrain alcohol consumption), and more rural areas, especially in the South and West (where Prohibition was obviously effective and remained in place in some ways after 1933). Anyway, these figures are for U.S. per-capita consumption of alcohol in gallons (http://www.drugtext.org/library/articles/craig102.htm)
As I say, these figures are obviously imperfect at precisely the point they're most needed. So, it's a bit more reliable to measure alcohol consumption through proxy figures, such as the death rate from cirrhosis and mental-home commitments for alcoholism. They tend to suggest Prohibition was at least somewhat successful.
For example, the average death rate from cirrhosis of the liver was 7.3 per 100,000 in the years 1920-1933; the average rate for the rest of the 20th century was 11.5. According to a Mark Moore New York Times op-ed column "Actually, Prohibition Was a Success," in 1989, cirrhosis death rates for men went from 29.5 per 100,000 in 1911, to 10.7 per 100,000 in 1929. Admissions to state mental hospitals for alcohol psychosis also fell from 10.1 per 100,000 in 1919 to 4.7 per 100,000 in 1928. There are problems with figures this crude, but they tend to cut both ways -- i.e. the time lag that often happens between alcohol consumption and cirrhosis on the one hand, but the effects of the Temperance movement that resulted in Prohibition in the years prior to Constitutional prohibition and the lingering effect of formed habitual behavior in the years after (more on that anon).
There's two other issues to consider after all these numbers. The first is a weakness of all social science, in that, mimicking natural science and mathematics, it seeks to isolate variables. The problem is that laws and human conduct simply never occur in a vacuum or a scientist's germ-free lab. To cite a point relevant to this case and to which I've already alluded above, the people who argue against Prohibition's effect on alcohol consumption point out that the cirrhosis death rate dropped faster during the 1910s, before Prohibition, than in the 1920s. And that the repeal of Prohibition didn't noticeably increase cirrhosis rates in the late 1930s. The numbers plainly support that (if you read Moore carefully with that in mind, you can see the residue). So the social scientist is satisfied. The problem is that in the real world, the country that passed Prohibition had to first become a country that *would* pass Prohibition (and that also meant passing mini-prohibitions in states and counties). Or to use the pro-life movement's formulation, we seek a culture where every unborn child is welcomed into life *and* protected in law. Changes in culture and the habits of the heart are logically distinct from the law, but not practically. We express our notions of good through the law, and the law reinforces those notions of good. Or to use Aristotle's formulation, we becomes virtuous by doing the virtuous things and vicious by doing the vicious things.
Secondly, and this might sound a little a priori and anti-intellectual, there's a certain level at which I simply refuse to listen to something so counterintuitive as "Prohibition didn't reduce alcohol consumption" (as opposed to "Prohibition had excessive countervailing costs"). I mean if making something illegal doesn't raise its cost (in the broadest possible sense), if raising the cost of something doesn't decrease its sales, and if being less a part of unthinking visible routine doesn't make a thing less popular -- if these things aren't true, ceteris paribus (and keep in mind what a stiff demand that is of history and the social "sciences"), then every thing we think about man is false and we've learned nothing about him in the history of civilization.