My initial Overlawyered guest post calling for higher excise taxes on alcohol in the US motivated a particularly thoughtful and lengthy reply from Radley Balko over at The Agitator, and his post has been followed by a fair number of comments. While I agree with many of the arguments that Radley and his commentators raise, there are a few points of contention. I will make a couple of remarks here, and then move any further discussion on my part to Vice Squad. If you are already tired of this, do not click on the ?Continue reading?? link.

It seems as if many people have a hard time believing that a small increase in price will actually dissuade alcohol consumption, especially by heavy drinkers or teenagers. And, to take it a step further, that the small price increase will result in any noticeable decline in bad outcomes, such as auto crashes. But there is lots of research on these points, and what it generally (though not universally) finds is that price increases do lead to reduced alcohol consumption, and to, for instance, fewer auto crashes. In fact, taxes seem to have more of an influence on fatal crashes than they do on overall alcohol consumption. One web-available review of these points (that I, for what it is worth, regard as pretty balanced) can be found here (pdf, 14 pages).

Radley doesn?t believe that tax policy should be used for the purpose of manipulating personal behavior. Now one can argue over the extent to which drunk driving is an “external” cost of drinking, and surely what constitutes an externality is not fixed in time or place. (It would be great if we could always rely on perfectly targeted policies that only affect the bad actors and no one else, but often, to the extent that such policies are feasible, they are inadequate. At the extreme, you could even argue that drunk driving should not be illegal, because some people do it pretty well, without causing much of a risk to others. I have some hopes, incidentally, for technological advances that can do a better job of keeping troublesome drinkers from alcohol, without impinging upon other drinkers or nondrinkers. The better these devices are, the fewer the “externalities” for taxes to correct.) But to the extent that some of the associated costs of drinking are real externalities, then internalizing them through a legislated higher price isn?t really “taxation” in the usual sense of the word. Rather, the excises eliminate an existing implicit subsidy. Done correctly (no small feat, of course), they enhance, rather than detract from, the working of the markets.

Can a case be made for taxing alcohol even beyond what would be called for by internalizing external costs? I think so, for a couple of reasons. First, alcohol often involves costs to the drinker that do not seem to be fully accounted for ? “internalities,” as the saying goes. But second, well, here I will turn to Vice Squad hero John Stuart Mill, from Chapter V of On Liberty. (I should also mention that most studies find that higher alcohol taxes in the US would cause an increase in revenue.) After first suggesting that consumption choices should be left to personal judgment, Mill continues:

“But it must be remembered that taxation for fiscal purposes is absolutely inevitable; that in most countries it is necessary that a considerable part of that taxation should be indirect; that the State, therefore, cannot help imposing penalties, which to some persons may be prohibitory, on the use of some articles of consumption. It is hence the duty of the State to consider, in the imposition of taxes, what commodities the consumers can best spare; and ? fortiori, to select in preference those of which it deems the use, beyond a very moderate quantity, to be positively injurious. Taxation, therefore, of stimulants, up to the point which produces the largest amount of revenue (supposing that the State needs all the revenue which it yields) is not only admissible, but to be approved of.”