What’s the impact of raising the drinking age to 21?

I’m distracted by other things today, but couldn’t resist the PM’s call for evidence on the costs and benefits of raising the Australian minimum drinking age from 18 to 21. Here are 3 possibly relevant economics papers.

Does the Minimum Legal Drinking Age Save Lives?
Jeffrey A. Miron, Elina Tetelbaum
The minimum legal drinking age (MLDA) is widely believed to save lives by reducing traffic fatalities among underage drivers. Further, the Federal Uniform Drinking Age Act, which pressured all states to adopt an MLDA of 21, is regarded as having contributed enormously to this life saving effect. This paper challenges both claims. State-level panel data for the past 30 years show that any nationwide impact of the MLDA is driven by states that increased their MLDA prior to any inducement from the federal government. Even in early adopting states, the impact of the MLDA did not persist much past the year of adoption. The MLDA appears to have only a minor impact on teen drinking.

Long Term Effects of Minimum Legal Drinking Age Laws on Adult Alcohol Use and Driving Fatalities
Robert Kaestner, Benjamin Yarnoff
We examine whether adult alcohol consumption and traffic fatalities are associated with the legal drinking environment when a person was between the ages of 18 and 20. We find that moving from an environment in which a person was never allowed to drink legally to one in which a person could always drink legally was associated with a 20 to 30 percent increase in alcohol consumption and a ten percent increase in fatal accidents for adult males. There were no statistically significant or practically important associations between the legal drinking environment when young and adult female alcohol consumption and driving fatalities.

Alcohol and Marijuana Use Among College Students: Economic Complements or Substitutes?
Jenny Williams, Rosalie Liccardo Pacula, Frank J. Chaloupka, Henry Wechsler
College campuses have been cracking down on underage and binge drinking in light of recent highly publicized student deaths. Although there is evidence showing that stricter college alcohol policies have been effective at discouraging both drinking in general and frequent binge drinking on college campuses, recent evidence from the Harvard School Of Public Health College Alcohol Study (CAS) shows that marijuana use among college students rose 22 percent between 1993 and 1999. Are current policies aimed at reducing alcohol consumption inadvertently encouraging marijuana use? This paper begins to address this question by investigating the relationship between the demands for alcohol and marijuana for college students using data from the 1993, 1997 and 1999 CAS. We find that alcohol and marijuana are economic complements and that policies that increase the full price of alcohol decrease participation in marijuana use.

Of course, the issue that we also need to consider are the magnitude of the benefits that young people gain from drinking (if you find this hard to swallow, pretend the proposal was to reduce road deaths by banning all drinking). I haven’t seen any good empirical evidence on this point, but it’s a critical one.

In the Australian context, Harry Clarke or Jenny Williams would be my go-to people if I was a journalist writing a story on this topic. Any other papers or experts that readers can suggest?

Update: Here’s Harry Clarke’s view on the issue.

(xposted @ andrewleigh.com)

14 thoughts on “What’s the impact of raising the drinking age to 21?”

  1. I’m sure I’ve asked this question on here before (but don’t think I got a response): 

    – Has there been much research into the effect of legal driving  (or licensing) and driving ages being aligned versus staggered?

    What I’m getting at is might an alternative solution be lowering the Australian minimum driving age to 16 (and extending P-plate – i.e. zero tolerance of alcohol in driver’s blood – out to 4 years)? 

    Like

  2. The second paper finds an impact among young males aged 18-20, legal drinking, and road fatalities. I bet this relationship still holds in this at risk group when they are aged 21!    

    Like

  3. It seems a little odd to me that we are contemplating introducing a new law (to ban drinking) simply to enforce an existing law (that bans drink-driving).  Wouldn’t a more direct approach to be to seek better ways to enforce the existing law.  Or, conversely, if the existing law cannot be enforced, who is to say that the new one will be.
     
    Andrew is right to suggest that there would be outrage if it was suggested that drinking is banned for everyone, simply to deal with a drink-driving minority.  And yet, this is what seems to be proposed for young adults.
     
     

    Like

  4. I don’t think it’s just the drink-driving angle behind a push for a higher drinking age. I think there’s also the issue of binge drinking and alcohol related violence / anti-social behaviour. Well – thats what we were being told when the push for an alocpops tax was on wasn’t it?

    Like

  5. If the drinking age were to be raised, the law would be flouted to a greater degree than it is now. I wonder what effect that has on regard and respect for laws in general. Anyone know of any research along those lines?

    Like

  6. @HeathG: that is a fair point, but not one brought out in the post.  Of the papers referred to by Andrew, 2 are about drink driving and 1 is about Marijuana use.  The latter is another example of creating a new law to better enforce an existing one.
    @Toby: agreed. Look at the second abstract.  Legalising drinking led only to a 20-30% increase in alcohol consumption.  So, around 75% of drinkers are prepared to break the law.

    Like

  7. I have long argued for increasing the drinking age on the grounds of health benefits. A blog post I did on the topic is <a href=”http://www.harryrclarke.com/2008/03/24/a-case-for-increasing-the-minimum-age-for-legal-drinking/”>here</a> and there are a couple of academic papers of mine available in Agenda and Melbourne Review. A recent contributer to the area is  Melbourne University’s  John Freebairn.

    A lot of interesting research centres on research into brain development. The main insight is that the brain is still developing up to around age 25 and that alcohol (like nicotine) is a neurotoxin which impedes this.

    Note my arguments are based on heath benefits not on an assessment of overall costs and benefits.  You need to account for enforcement costs and effectiveness. But the general evidence I have seen supports the second reference above not the first.

    Youth whose parents are alcoholic should probably never drink if they wish to avoid alcoholism themselves for genetic predisposition reasons. If any young person thinks about initiating drinking then it is wisest to delay initiation to around age 25 since that is when the really severe neurotoxic effects of alcohol on the brain start to diminish and when the risk of alcoholism recedes markedly.

    Hormonal youth are often not smart (or patient!) enough to follow such a rule so I am happy with the idea of involuntary restrictions. The idea that the 5 kids who recently killed themselves in Bundoora rationally elected to drive in a car with an intoxicated yound driver is too silly to contemplate. 

    The argument is often raised that driving under the influence should be penalised directly rather than penalising under age drinking.  This presupposes that the major cost of under age drinking takes the form of traffic accidents. My understanding is that costs in terms of brain development are also significant. There are a lost of other anti-social externalities as well.

    Like

  8. Harry are there any other bad habits you’d like adults to have no right to pursue?  Smoking? Masturbation? Wearing jeans that expose their underwear? Voting?
    18 year olds are adults in this country and seriously impeding their choice set on the basis of the possible outcomes of extreme consumption (“stop it or you will go blind”) is draconian and, I suspect, unenforceable.  As was noted above, all that this woud do is encourage law-breaking, and more dangerous (i.e. surreptitious) forms of drinking.
    How about accepting that we can’t fix everything with laws and let people learn from their mistakes (and the mistakes of others)? Personal (and parental) responsibility will typically outperform a nanny state.
     
     
     
     
     

    Like

  9. I’d like to end smoking in one generation  – it is by far the most significant public health issue in the world today and, on current estimates, will kill 1 billion over the coming centur. Masturbation has always seemed to me to outperform valium and to be a harmless recreation. Jeans that expose underwear convey positive externalities to some but I think the transaction costs would make a subsidy non-viable.

    As a parent I feel real sadness for those 5 kids who died near Bundoora a few weeks ago.  What a waste for them and what an ongoing tragedy for their parents. I am happy to seek to restrict individual foolishness on this scale. It isn’t nanny-statism. 

    Young males with lots of testosterone have high discount rates and behave in ways at age 18-24 that they themselves regret a few years later.  They should know that drinking during this period permanently reduces their brain development and be discouraged from consuming such poisons until it is at least much safer to do so.

    Like

  10. Daniel Todd you are a moron! This is a group of civilised adults trying to have a serious argument about an important issue. Take your immaturity elsewhere.

    Like

  11. Jack Jeffery you are the moron!  dont talk shit about total!! i can tell by your level of maturity that your head is too far up your rectum. get a life FAGGOT!!

    Like

  12. Oh great, trolls are on the prowl.

    My point is not original, but we have no problem with young males aged 18-21 yo to fight -and possibly die – for the nation. How far should the (nanny) state go to  remove inate risk-taking behaviour from young males, or anybody else for that matter? Very specific laws aimed at very specific behaviour, i.e. drinking and driving in high-risk age groups, is fundamentally social engineering. The implication seems to be that drinking oneself into long-term health problems is OK if you are past a certain age constraint.

    Like

Comments are closed.