« The Case for Big Government | Main | Eating local »
Lowering the drinking age
Here is another reader request:
There's been recent talk about what would happen if the legal drinking age were lowered to 18. Would there be a net increase or decrease in risky binge drinking, accidents, etc?
New Zealand lowered its drinking age to 18 in 1999 and bad consequences followed, including a higher rate of drinking-related car crashes. Illegality, even when it can be circumvented, really does raise the price of an activity in many instances.
Nonetheless I still think that 20-year-olds -- legal adults in just about every other way -- have the right to drink alcohol. Sometimes I call myself a "two-thirds utilitarian." I am a pluralist who thinks that utility is often but not always the primary consideration behind policy choice.
There's always another paternalist intervention to save children's lives but no one is for all of them. We could ban swimming pools and buckets for instance. We could ban high school football. We could raise the drinking age to 25. How about a drinking age of 50? How about a driving age of 21?
I see at least two major analytical questions. First, how much normative force should "extra death" have in a policy argument? Second, what is special about the number 18? Consistent with the latter question, I think that 15-year-olds should be able to drink in a restaurant when clear parental permission is present.
Posted by Tyler Cowen on August 28, 2008 at 06:59 AM in Food and Drink, Law | Permalink
Comments
i wonder what the drinking related deaths for europe are, where theres much lower drinking age, controlling for car use.
Posted by: c8to at Aug 28, 2008 8:00:32 AM
right on!
Posted by: mario at Aug 28, 2008 8:20:28 AM
I think it's reasonable that drunk-driving, etc. would increase after the lowering of the drinking age, if for no other reason then that current law bottles up (and grows) demand.
If a lower drinking age were held long enough for the culture to normalize it (France, for example, doesn't have a binge drinking problem), I think the long run is lower alcohol abuse.
What's exceptionally dumb now is that the drinking age now is set precisely so that the rebellion against it is exactly when young people are out on their own for the first time. Set the drinking age to 16, and have kids learn to drink responsibly when they're still in their parent's house.
Posted by: Dave at Aug 28, 2008 8:30:27 AM
Contrary to popular myth, France and Italy have the highest alcoholism rates in Europe.
If you were making decisions based purely on statistics, the way insurance companies do, the age limit for most activities, from driving a car to drinking, would be much higher - 25.
Posted by: Ted Craig at Aug 28, 2008 8:40:48 AM
Interesting idea from McArdle yesterday, that if you lowered the drinking age and added a zero-tolerance rule for driving, that you might get a better result. Similar to a "nudge", if you will.
Posted by: Larry at Aug 28, 2008 8:51:01 AM
A recent study conducted of 100 Presidents of leading universities contend somewhat otherwise that making the object of drinking illegal is making it additionally more unsafe than it already is.
Full article here: http://www.cnn.com/2008/HEALTH/08/18/college.drinking.age.ap/index.html?eref=rss_topstories
As an individual 19 days away from his 21st birthday, of course young people will drink more or less no matter what; at least when popular culture tells us that it's tacitly OK, when most of our parents supply alcohol, when establishments prefer to let a poorly designed fake-ID slide because they want the business of younger college students. All of these factors drive an illegal function to cultural legitimacy.
The idea of raising the driving age is ok, but given the disparity in geography, particularly when you depart from the east and west coast's megalopolis, many young people absolutely require the ability to drive - to work, school, to overcome t he simple distances of our heartland. Making the drinking age in city's to be 19 and the driving age to be 20 would be a good first step, but of course enforcement would be a nightmare.
No ready solution presents itself, except to allow the status quo to proceed.
Posted by: Buck Bobbin at Aug 28, 2008 8:58:04 AM
It's the large population effect.
With a large enough population, everything results in death.
But the fallacy is that risk is a personal thing, not a population thing. Something does not become more risky to do if lots of people are doing it. It's safe for you if the risk of your own death is low enough. So you can forget about death by tornado!
Always divide by population size to get risk.
Posted by: Ron Hardin at Aug 28, 2008 8:58:41 AM
How about a maximum drinking age of 65? Don't old people have enough health problems that we should protect them from themselves and keep them from being a danger on the road by not allowing them to drink alcohol?
Posted by: AZ at Aug 28, 2008 9:01:20 AM
Since the problem is one of traffic accidents, which are more likely for drivers under 20 regardless of whether or not alcohol is involved (Source: http://www.ncbi.nlm.nih.gov/pubmed/7718080), maybe the answer is to lower the drinking age, and raise the driving age.
Posted by: Jennifer at Aug 28, 2008 9:03:05 AM
"Second, what is special about the number 18?"
Surely if you are liable to fight to your death for your country via a gov't issued draft then you should be able to enjoy a drink of alcohol.
Posted by: Richard at Aug 28, 2008 9:10:54 AM
I learned to drive in my twenties. Although I never got into an accident, in the first couple of years there were a couple of times that came close. The issue isn't young drivers, it's inexperienced drivers. Raising the driving age wouldn't help.
You could argue instead that there should be a maximum driving age, given all the recent news stories about disoriented elderly drivers plowing through crowds and killing a bunch of people, but of course this is politically impossible.
Posted by: at Aug 28, 2008 9:13:15 AM
The argument comes down to this: should a legal adult be able to buy a drink? Of course.
Would it lead to bad results (more alcoholism, more traffic fatalities, etc?). Of course.
My vote: leave the drinking age at 21.
Posted by: Dan Tarrant at Aug 28, 2008 9:16:37 AM
"if you lowered the drinking age and added a zero-tolerance rule for driving, that you might get a better result. Similar to a "nudge", if you will."
Does she mean zero tolerance for those under 21 or for everybody? If the latter, that would kill off the restaurant and bar industry everywhere but in large metro areas that have public transportation and/or taxi service. Most suburban, small town and rural areas have neither, and are not dense enough to survive on a walk-in crowd.
Posted by: liberalarts at Aug 28, 2008 9:17:55 AM
You could perhaps look to Canada for data. The majority of Canadian provinces have a drinking age of 19, some have 18. Quebec for instance has had a drinking age of 18 all along, and there has been little or no clamor to raise it.
Posted by: at Aug 28, 2008 9:20:07 AM
In the future electronics could lead to much safer cars, if that happens we could lower the drinking age.
Posted by: floccina at Aug 28, 2008 9:30:44 AM
what is special about the number 18? Consistent with the latter question, I think that 15-year-olds should be able to drink in a restaurant when clear parental permission is present.
What is special about the number 15?
Posted by: mk at Aug 28, 2008 9:33:48 AM
Lower the drinking age but increase Pigouvian alcohol taxes substantially.
Taxes on cigarettes have been found to disproportionately discourage young people. That's likely to be true for alcohol taxes as well.
Posted by: a student of economics at Aug 28, 2008 9:40:05 AM
I can't quite formulate the argument precisely, but I suspect underage drinking is subject to the Alchian-Allen Theorem. Since there is fixed, relatively high legal cost to drinking both moderately and excessively when underage, I believe the Theorem states that people who drink when underage will more often drink to excess.
I recently graduated college, but when I was attending, I noticed this effect in myself and others. Tyler, is my economic reasoning correct?
If my reasoning is correct, I think there is a strong case to be made for a lower drinking age. The primary goal of controlling the drinking age is preventing harm, and harm most often comes from drinking lots, not from having one or two beers with friends.
Posted by: Matt at Aug 28, 2008 9:54:33 AM
In England they are working toward raising the drinking age to 21, in an effort to curb some of their culture's emphasis on drinking. English drinking culture, especially in the Northeast (where I'm studying) is different from American drinking culture in that habitual drinking (2+ times each week) is expected from teen years until death. Also, drinking until vomiting isn't stigmatized so much here.
How do you know you're spending the evening in Newcastle?
There is a layer of puke on the sidewalk.
Posted by: grant at Aug 28, 2008 10:02:07 AM
How about for people age 16-21 the DMV issue "Driving Licenses" and "Drinking Licenses" and let the teens choose (you can only pick one)?
Posted by: zacharij at Aug 28, 2008 10:02:27 AM
Philip Cook's Paying the Tab: Costs and Benefits of Alcohol Control has a large section devoted to the economic studies of the effect minimum drinking age laws on traffic fatalities among other things. I think nearly all of the papers found that the drinking age laws were correlated with traffic fatalities. The lower the minimum drinking age, the more traffic fatalities. There's also papers finding that younger drinking ages causes increases in syphilis and gonorrhea (see Harrell Chesson's late 1990s, early 2000s JLE paper on that: "Sex Under the Influence" is the name, I think). And more recently papers have found that higher minimum drinking age laws causes improvements in birth outcomes.
I've not seen anyone reference this pretty large literature when debating this topic. It'd be helpful if advocates of lowering the law would respond to this work. That a lower law would decrease binge drinking at college seems speculative, but even if that's true, it's not clear why that would swamp the plausible damages found in other studies.
Posted by: jason voorhees at Aug 28, 2008 10:26:03 AM
How about not having a specific age but instead focus on the maturity and situation of each individual, including parental guidance? Maybe guidelines are better than laws and maybe some people are better off never drinking at all while others are ok even at very young ages. I know that it's a lot easier to just have a simple age limit, because everyone understands it, it's easy to use etc, at least on paper. But what I think would be much better and less arbitrarily rigid is a change in drug use culture. Sure, young people are often rash and tend to do stupid things like overuse certain substances, but I do believe that a better way to modify that kind of behaviour is by way of example and mutual respect rather than stern rules that are neither understood nor respected by adolescents. They are prone to rebel, but much more so when there are actually things to rebel against.
Posted by: Alexander at Aug 28, 2008 10:28:09 AM
"I learned to drive in my twenties. Although I never got into an accident, in the first couple of years there were a couple of times that came close. The issue isn't young drivers, it's inexperienced drivers. Raising the driving age wouldn't help."
I think the issue for, say, a 16 year old driver isn't basic competence so much as an inclination towards risky behavior that is much different for most 22 year olds. I would take a less skilled driver who is careful than a "natural" driver who takes unnecessary risks any day. Combine the risky behavior of a 16 year old with alcohol use and you get a recipe for disaster. I frankly think the best policy might be to lower the drinking age to 15 and raise the driving age to 21, but that will never happen in this country with the present state of its public transportation system.
Posted by: wugong at Aug 28, 2008 10:29:47 AM
What happens in countries where the drinking age is lower than the driving age? I had a friend from Germany and his opinion was that a drinking age of 16 and driving age of 18 worked better because it allowed everyone to learn from a couple years' experience how they handle alcohol before they were at risk of killing someone while drunk.
Posted by: Jacqueline at Aug 28, 2008 10:50:08 AM
In response to Dave:
Lowering the drinking age to 16 would make it incredibly easy for not only high school students but middle school students as well to get alcohol. I agree that under parental supervision, someone as young as 16 should be allowed to drink, but the distribution of alcohol to people of middle school age should be parents' responsibility, not 16 year olds'.
Posted by: Tim at Aug 28, 2008 10:52:41 AM
