The Drinking Fallacy

I would say that that I don't mean to quibble, but that would be false, because I precisely do mean to quibble.Will Wilkinson has weighed in against there being a drinking age at all.

Here's an example of his argumentation:

"UCLA professor of public policy Mark Kleiman, an ex-advocate of age restrictions, told PBS that he came around to the no-limits position when he saw a billboard that said, 'If you're not 21, it's not Miller Time--yet.' Age limits make drinking a badge of adulthood and build in the minds of teens a romantic sense of the transgressive danger of alcohol. That's what so often leads to the abuse of alcohol as a ritual of release from the authority of parents. And that's what has the college presidents worried. They see it."

Smells like a fallacy of false cause. It might be true that restricting legal drinking to 21 lends a weird romanticism to the activity (really though, who knows), and the abuse of alcohol is indeed a problem, but the idea that kids abuse alcohol because it is romantic is specious inference based on some pretty sketchy folk sociology. What is probably true is that some aura of romanticism encourages some extra amount of drinking, but drinking abuse is undoubtedly caused by many other factors, very few of which have to do with any sort of aura, and that together dwarf this supposed romanticism effect. Kids drink because its fun. In part it's fun because it's rebellious, but its fun for an whole lot of other reasons as well (inebriation feels good, individuals feel more sociable, you're more likely to get laid, worries are easy to forget, it's a social activity with the all the benefits of group membership, for some people the stuff just tastes good, etc.)
Will goes on:

"There's certainly evidence that if we got rid of age limits, teens would drink more. But drinking more is a drinking problem only in the minds of neoprohibitionists. In a 2003 survey 22% of American tenth graders said they'd had five or more consecutive drinks in the last 30 days. But in Denmark, where there's no legal minimum to drink (though you have to be 18 to buy), 60% of 15- and 16-year-olds said they'd thrown back five or more in a row within the last couple of fortnights. Maybe you think that's too much. But the European champion of heavy teen drinking ranks as the world's happiest country and scores third in the United Nation's 2007 ranking of child welfare. In the UN listing the U.S. came in 20th out of 21 wealthy countries."

Um, maybe Danes are so happy because they drink so much. But regardless, it's not unimportant that Denmark is a wealthy, relatively homogenous and very well educated nation. I've spent a fair amount of time in Denmark. There is a lot of conspicuous drunkenness. Drunkenness is a problem in Denmark, as most Danes would admit on those occasions when they're not drunk. But being wealthy, well-educated, and committed to a generous social welfare state, they can afford a level of alcoholism that there's very little reason to think that the United States could afford. We have here a fallacy of false analogy. In any case, I don't think I'm going out too far on a limb to again assert that, even if alcohol policy has some effect on metrics like happiness and child welfare, the effect is going to be very, very small, to the point where overall social happiness and child welfare are completely unrelated and so can't support any inference either way.

Will also suggests that drinking-age and drunk-driving traffic accidents may not be positively correlated. I don't know any of the research, and so won't comment on that angle.

But he continues:

"Salt makes things taste better. If you eat too much, it can kill you. But we don't need laws regulating salt."

Again, false analogy, in this case so obvious that there's hardly need for comment. Crack makes you feel better too, but if you smoke too much, it can kill you. A-bombs give you a sense of security, but if you let one off, it can kill lots of people. Point is: just about everything has some sort of benefit, and just about anything can be dangerous. We need to decide which are too dangerous to allow to be legal. He concludes:

"In an America without a minimum drinking age, we would shift our focus from demon rum and car crash statistics to creating an environment where parents are expected to supervise their children and alcohol would become for teens just another thing, like bicycles or swimming pools, that can either make your day or take your life."

I'm pretty sure that this is perfectly fine anyway in most states. If parents want to ease their kid into a responsible drinking habits starting at an early age, I'm rather certain that there's no legal obstacle to this, and that even if there were, no one bothers to enforce it. I've at least never heard of a 15 year-old kid get into trouble with the law for enjoying a glass of red wine with his parents. Final point: kids drink and party too much for the same reason many have sex too early and too often: it's fun, and there's not much that legislation either way is going to affect that fact.

Personally, I'm agnostic on the issue. I do remember what a bummer it was not being able to drink legally as a 19 year old in college. But I drank anyway, and if it had been legal, that wouldn't have been different. Point being: the two are not all that related. I'm sure that arguments for Will's position are out there, but they need to be on principled grounds, not on utility effects. The arguments ought to be of the sort: 18 year olds should be able to legally drink, period, and if that entails some net costs, such is the price of freedom. We allow them to join the military. We allow them to vote. We allow them to have children and to marry. It seems a little arbitrary to prohibit them from drinking. If we are not going to argue the issue on these grounds, then if someone is goint to persuade me that lowering the drinking age would be better, they'd have to convince me that there's not after all anything wrong with the following inference: we will lower alcoholism among kids by making it easier for them to get it. That said, the arguments I've made suggest that there would be little effect either way if the drinking age were lowered. This is why I remain agnostic on the issue--I don't think it matters all that much.