A Critique of Barnum et al. (2012)

A new study by Barnum et al. (2012) on drinking age law enforcement was recently published online this month.  In this first-of-its-kind study, using data from 1975-2006 they find that birth cohorts exposed to tougher enforcement of PAULA (possession of alcohol under the legal age) laws (as measured by underage drinking arrest rates) from ages 15-20 had modestly lower arrest rates for assault and vandalism between the ages of 15-24.  This remained true even when age effects, period effects, relative cohort size, and percentage of nonmarital births were controlled for.   But was it really a causal relationship?

While the authors appear to be convinced that this relationship is causal, we at Twenty-One Debunked note that there are plenty of reasons that this relationship could easily be spurious.  For example:

• Several other potentially important variables were not controlled for, including ones that may not have been captured by age and period effects.
• State-to-state variation was not explored at all.
• Arrest rates may have given biased estimates due to underreporting and changes in reporting rates and police practices over time, which can yield specious inferences.
• The strengths of the underage drinking laws themselves were not explored, only the enforcement of such laws as measured by arrest rates.
• Cohorts exposed to drinking ages of 18, 19, 20, and 21 were all lumped together, with no attempt to distinguish between them.
• No other crimes were explored besides assault and vandalism, and no distinction was made between types of assault (i.e. simple vs. aggravated).
• Allocating more resources towards arresting underage drinkers (and those who commit other victimless crimes) takes away from resources used to fight real crime, including assault and vandalism.  So, an increase in the former could lead to a spurious decrease in arrests for the latter in the absence of any real change in the latter.

Even if we accept the authors' conlcusions at face value, arresting underage drinkers does not appear to be a particularly cost-effective crime reduction strategy from a public safety standpoint.  The study reported that for every full unit increase in underage drinking arrests, there was only a 0.125 unit decrease in assault arrests and a 0.134 unit decrease in vandalism arrests.  Other sources note that on average, for every 1000 or so incidents of underage drinking, only one PAULA arrest occurs, making it a highly inefficent use of resources.  Law enforcement resources would thus be better spent actually targeting real crimes rather than victimless ones like underage drinking.

The example of San Francisco is highly instructive in this regard.  In 1990-1992, they were a crack-infested, gang-ridden hellhole.  By 2000, violent crime had plummeted by half (and even more so for juveniles), and by 2009 they became one of the safest big cities in America.  And how did they manage to do this?  Did they employ a "broken-windows," zero-tolerance approach to the most minor offenses, especially by young people?  Hardly!  In fact, beginning in 1992 (the city's peak year for violent crime), they stopped enforcing their youth curfew law (which was completely abolished in 1995), and since then they have actually cut back on arresting young people for "status" offenses (such as underage drinking) and cannabis possession, making such offenses the lowest priorities.  In other words, "don't sweat the small stuff."   Meanwhile, the police freed up more resources to tackle serious crime, and managed to build better relations with the community.  While the exact reasons for the drop in crime are not entirely clear, and several other cities nationwide saw similar improvements, it certainly casts doubt on the authors' thesis that cracking down on underage drinking reduces crime.

Interestingly, 'Frisco teens also show significantly lower rates of violence, "binge" drinking, driving after drinking, drinking in general, cannabis use, huffing, and crack/cocaine use compared with the national average according to the Youth Risk Behavior Survey.   Furthermore, the rates in San Francisco are comparable to or lower than those in NYC (and dropped at a similar or faster rate) despite the latter city's notoriously heavy-handed police tactics under Mayors Giuliani and Bloomberg.

A Critique of Carpenter and Dobkin (Part Deux)

We recently debunked a new study by economists Carpenter and Dobkin (2011) that has apparently concluded that lowering the drinking age from 21 to 18 will lead to an 8% increase in deaths among 18-20 year olds and an estimated externality cost of $12 million per 100,000 person years (or $1.44 billion per year), for a combined "social cost" of nearly $10 billion per year.  (There are about 12 million people in that age group.)  The cost estimates were (most ironically) based on the increase in death, injury, and crime that occurs upon turning 21, and the authors suggest that the same would happen for 18-20 year olds if they were allowed to drink legally.  While we believe these estimates are grossly overstated (and most likely just represent delayed rather than decreased costs when the legal drinking age is 21), let us take these estimates at face value for a moment for the purpose of a proper Pigouvian analysis.

First, lets examine the other side of the ledger.  For one, there is the potential 3% increase in wages for each year of education in which alcohol is consumed (versus not consumed), presumably due to increased human and social capital formation.  One study even found that men who go to bars once a month or more make 7% more in wages compared to those who do not--which is significant since under the current laws, most 18-20 year old drinkers do their drinking in locations other than bars.  Since Carpenter and Dobkin estimate that the proportion of drinkers among young adults increases by 6.1 percentage points (or a 10% relative increase) upon being able to drink legally, one could say that, based on per-capita personal income figures for 2009, an economic benefit of roughly $1-2 billion per year in the long run would result from lowering the drinking age to 18.  Tax revenues that fill government coffers (to the tune of $5.6 billion per year for the general population) are another benefit of alcohol, and foregone drinks translate to foregone taxes.  Americans consume 8.4 liters of alcohol per capita, or about 117 billion standard drinks per year in total, making that a tax benefit of about $0.05 per drink.  Since Carpenter and Dobkin estimate that an additional 547 million drinks total will be consumed per year if the drinking age was reduced to 18 (which we strongly doubt, but let's go with it anyway), that would be a benefit of $27 million per year.  However, that only includes excise taxes.  When all taxes are taken into account, beer sales alone bring in about $30 billion per year, so we estimate that total alcohol brings in at least $40-45 billion, or $0.40 per drink.  Using that figure, lowering the drinking age to 18 would yield $219 million to the economy.  The fact that more drinking would occur in bars would only boost that number even higher.  Due to the uncertainty over health benefits of moderate alcohol use for young adults, we exclude them from our calculations.

We also need to examine the externalities of the 21 drinking age itself.  There is the cost of enforcement, of course, which while we don't know the exact number, we do know that states receive $25 million each year from the feds for the purposes of underage drinking enforcement.  Since states most likely spend their own money as well on this effort, this is a lower-bound estimate of the real number, which we believe is probably double that.  But some of these funds would still be spent if the drinking age was 18, so we'll stick with the $25 million figure for now.  We also know that due to the laws' ineffectiveness, numerous other ancillary laws (dram shop, social host, use and lose, keg registration, etc.) are often used to prop up the 21 drinking age.  There are also significant opportunity costs to enforcing all of these laws, so that fewer resources are devoted to real crimes, including violence and drunk driving.  Also, the estimates of the costs of legal alcohol use by 18-20 year olds should be adjusted downward, since forcing alcohol use underground makes it more dangerous than it has to be.  The penalties for those who do get busted can be quite harsh in some states, and when you criminalize normative behavior you also create costs such as an increased number of individuals with criminal records, which limits their job opportunities in the future and hinders their ability to become productive members of society.  There are also increased court costs and increased lawsuits as well.  Other costs, such as loss of social cohesion and loss of civil liberties are probably very high (and many would even say priceless), but exceedingly difficult to measure.  But for now let's estimate all of these total externalities conservatively at, say, $200 million per year.

Looking at both sides of the ledger, lowering the drinking age to 18 would now give a net social cost of $7.5 billion, including a net external cost of $1 billion per year.  However, if we take into account that the increased wages would be taxed as well, the estimate of the net external cost (which is really the only cost that matters for public policy) drops to about $800 million.  And if we use a somewhat smaller value for the statistical value of a life than Carpenter and Dobkin used (there are many legitimate values given in the economic literature), the estimate shrinks even further to as little as $500-750 million.  And if a less conservative estimate of the externalities of the 21 drinking age law, say, $200-500 million (remember the measurement difficulty) and hedonic effects of alcohol are factored in, you're now talking chump change, at least relative to our society's $14 trillion GDP. 

A remaining externality gap of at most a few hundred million dollars could easily be made up by raising the alcohol taxes, especially since this has also been shown in several studies to reduce traffic fatalities and other alcohol-related social ills, especially among young people.  And at least one economist, Donald Kenkel (1993), considers higher alcohol taxes combined with a drinking age of 18 to be more socially efficient than a drinking age of 21.  Raising the tax rates to a level that would increase the price of alcohol by as little as 12% could theoretically be enough.  In another paper published the same year, he also notes that tougher penalties for drunk drivers would be more efficient than a 21 drinking age as well.  And we at Twenty-One Debunked believe that Pigouvian taxation (for all ages) makes far more sense than prohibition of drinking for 18-20 year old young adults.

Remember, as we have noted in previous posts, that we do not believe that Carpenter and Dobkin's estimates are accurate, and that several studies have debunked the claim that the 21 drinking age saves lives on balance, and even the claim that it leads to large reductions in alcohol consumption.  Any reductions that do occur are really just delays.  Thus, since our analysis assumes Carpenter and Dobkin's estimates to be reasonably correct, we still understate the net benefits to lowering the drinking age by a great deal.

A proper Pigouvian analysis of young adult drinking would also consider the marginal social cost (MSC) versus the marginal private benefit (MPB) of alcohol consumption, in this case for 18-20 year olds.  While this is not easy since we have yet to encounter any serious quantitative analysis of these parameters, we at Twenty-One Debunked believe that, as for those over 21, there does exist some level of consumption (which we do not attempt to quantify here) above which MSC exceeds MPB and below which MPB exceeds MSC.  This can be logically inferred from the fact that a heavy-drinking minority cause the majority of the problems associated with alcohol consumption, while most drinkers are non-problem drinkers.  Thus, the goal of public policy should be to reduce alcohol consumption to a socially efficient level via taxation and/or other cost-effective means.

Of course, there are some folks who believe that for people under 21, the most socially efficient level of consumption is zero (i.e. the marginal social cost exceeds the marginal private benefit at all nonzero levels of consumption), making outright prohibition of drinking for that age group the only solution.  We clearly disagree with that statement, but what if it somehow was true?  We know that underage drinking cannot be practically eliminated even under the strictest current regimes, but we could do a thought experiment.  What if, in a relatively Orwellian version of the future USA, our nation decided to fit everyone between the ages of 13-20, inclusive, with a SCRAM bracelet that monitored the alcohol in their sweat 24/7, and then had the bracelet removed upon turning 21?  Costs for all those bracelets, which are currently $300/month/person, would total up to a whopping $90 billion per year.  Even neglecting all other costs to the economy (i.e. lost revenue), transition costs, loss of liberty, and negative hedonic effects, this value would exceed even the most liberal estimate of the total social costs of underage drinking ($60 billion) by $30 billion--a massive net deadweight loss.  Clearly, this is not a viable option by any stretch of the imagination!  And the only other way to guarantee that no one drinks a drop before turning 21 is if no one has kids--which would ultimately lead to our extinction. 

Besides, the cost of "overage drinking", at least in terms of lives lost, is FAR worse than the cost of underage drinking.  That's the pink elephant in the room that the pro-21 crowd doesn't want to talk about.

Let America be America again, and lower the drinking age to 18.  If you're old enough to go to war, you're old enough to go to the bar.  'Nuff said.

A Critique of Carpenter and Dobkin (2011)

Recently, a new study by economists Carpenter and Dobkin (2011) has apparently concluded that lowering the drinking age from 21 to 18 will lead to an 8% increase in deaths among 18-20 year olds.  The authors argue that those who are proposing lowering the drinking age would therefore face "a very high burden of proof" for their position.  However, there is less to this study than meets the eye, and we at Twenty-One Debunked do not agree with their conclusions.  In fact, much of what the study adds actually can be said to fuel our side of the debate rather than the pro-21 side, and the burden of proof actually falls on them, not us.

The first part of their study separately examines daytime and nighttime traffic fatality rates (from FARS) from 1975-1993 for four age groups:  15-17, 18-20, 21-24, and 25-29.  (Nighttime deaths should be affected much more than daytime ones since the former are much more likely to involve alcohol.) They use a fixed-effects panel regression that, while it controls for state and year fixed effects, state-specific trends, and population changes, still fails to control for any other variables that would not be subsumed under these (in contrast to Miron and Tetelbaum (2009) and Dee and Evans (2001)).  The drinking age was expressed as the proportion of 18-20 year olds who are legal to drink in a given state-year, hence the expected sign is positive.  The results are summarized below as percent changes, with statistically significant values (p < 0.05) in bold:

Age  Group	% Change (Night)	% Change (Day)
15-17	7.9	8.3
18-20	16.9	4.7
21-24	11.3	6.9
25-29	9.7	1.7

As one can see, the apparent effect occurred among all age groups rather than just the 18-20 year olds who were targeted by the changes in the legal drinking age.  For 15-17 year olds (the spillover group), whose effects were statistically insignificant, the day-night counterfactual does not appear to work would be predicted if banning 18-20 year olds from drinking really saved the lives of the former.  However, it does appear to work for 21-24 year olds and 25-29 year olds, both of whom should not have been affected by the change in the drinking age since the age groups were evaluated contemporaneously rather than as cohorts.  In fact, the effects on 18-20 year olds and 21-24 year olds are rather comparable, as opposed to a mere spillover which would be much smaller in magnitude.  Thus, it is very likely that the drinking age was a proxy for something else, i.e. one or more of the many possible variables that was not controlled for such as anti-drunk driving campaigns or tougher DUI laws.

We should compare this to other studies that looked at the effects on various age groups.  Miron and Tetelbaum (2009), who thoroughly debunked the idea that the 21 drinking age saves lives (at least in the long run) for 18-20 year olds, found that a legal drinking age higher than 18 has no effect either way on 21-23 year olds but actually increases under-18 driver fatalities.  Dee and Evans (2001) found that a drinking age of 18 or 19 (relative to 21) increases traffic fatalities among 18-19 year olds but decreases fatalities among cohorts of 22-24 year olds who were allowed to drink at 18 or 19, with no consistent effect on 16-17 year olds.  This echoes Asch and Levy (1987 and 1990) and Males (1986), who both found that raising the drinking age to 21 likely just shifts traffic deaths from 18-20 year olds to 21-24 year olds and possibly even increases the net probability of dying by age 25.  In addition, a new, award-winning paper by Dirscherl (2011) finds that raising the drinking age to 21 not only shifts deaths from 18-20 year olds to 21-24 year olds, but actually leads to a net increase in deaths among 18-24 year olds, a phenomenon we like to call "White Noise Syndrome".

Next, the authors examined the effects of the drinking age on the mortality rates of various causes of those same four age groups.  These death rates were gleaned from vital statistics from the National Center for Health Statistics, and the panel estimates of the effects were obtained from a similar model to the one discussed above.  Again, the expected sign is positive.  The results are summarized below, with statistically significant values (p < 0.05) in bold:

Age Group	% Change All-Cause	% Change Traffic	% Change Suicide	%Change Homicide	%Change Alcohol	% Change Other External
15-17	5.5	8.4	9.3	6.3	-30	-4.0
18-20	6.9	9.1	10.1	-4.6	-10	9.0
21-24	5.5	10.5	3.7	-6.5	2.5	11.6
25-29	-0.8	4.4	0.0	-1.9	-18	5.2

We see that only for suicide is the effect statistically significant for 18-20 year olds and at the same time insignificant (and smaller) for the other age groups.  That is, only for suicide can one actually infer a potentially significant lifesaving effect of the 21 drinking age.  For traffic fatalities and other external causes of death, there now seems to be a greater effect for 21-24 year olds than for 18-20 year olds, which casts doubt on whether these effects were actually due to the drinking age.  Interestingly, homicide and alcohol-related deaths (e.g. alcohol poisoning, etc.) were not only statisically insignificant in all cases but even had the "wrong" sign for most of the age groups.  Finally, for all-cause mortality, we see that none of the estimates are significant, not even at the 10% level, which means that they are likely due to chance (and thus spurious).  However, the above results are nonetheless taken by the authors of the study to indicate an overall lifesaving effect.

In a previous post, we at Twenty-One Debunked ran several difference-in-differences analyses on mortality rates of 15-19 and 20-24 year olds for all of the external causes listed above.  Those were the only age groups publicly available through CDC's WONDER database, and while not ideal, taken together they are still useful for generating estimates of the net effects in the long run, which we did.  The control group was the 11 states (excluding Utah) that did not change their drinking ages (i.e. they remained 21 throughout since the 1930s and 1940s), while the treatment group was the states that had a drinking age of 18 in 1979 and later raised it to 21.  Comparing 1998 to 1979, we found that the net difference-in-differences between the groups had the "wrong" sign for nearly every cause of death, especially suicide for 15-19 year olds.  Only for homicide was there an apparent lifesaving effect, but removing New York from the data attenuated this effect to almost null.  The pattern for suicide vs. homicide appears to be the reverse of what Carpenter and Dobkin found.  Overall, we found no net lifesaving effect in the long run, echoing what Miron and Tetelbaum found for traffic deaths, and thus perhaps Carpenter and Dobkin's results are primarily capturing short-term effects due to the study design.

Next, the study's authors discuss their previous work on regression discontinuity estimates using more recent data, which we critique here.  This analysis shows a discrete and significant jump in mortality at exactly age 21.  The effect is true only for external causes of death, including motor vehicle accidents, suicides, deaths labled as "alcohol related," and those labeled as "other external," but not homicides or drug-related deaths.  Another similar study they did concerning various types of crime gave similar results overall.  But unfortunately, they also make the specious claim that such an effect is not merely a delay in deaths, but rather constitutes a true lifesaving effect of the policy (which is dubious).  In any case, it certainly shows once and for all that there is nothing at all about turning 21 that magically makes one a safe responsible drinker.

The authors then tie together all of their analyses thus far, and assert that despite all of these limitations, the similarity of the effect size (8-10%) between the regression discontinuity analyses and the panel estimates implies that the effects of the 21 drinking age are likely to be truly causal rather than a proxy for something else.  We find that argument to be puzzling at best, especially since Miron and Tetelbaum also found a similar effect of MLDA-21 (8-11%) in their initial 50-state model relative to MLDA-18, but it nonetheless dropped well below statistical and practical significance when the states were disaggregated and when the persistence of the effect was analyzed.  For some states, it apparently even made things worse.  Thus, an effect of this size may very well be a mirage rather than a truly causal relationship.

But the most tenuous aspect of the author's latest study is their analysis of the "social costs" in dollars per drink consumed by people under 21 if such drinking was legalized.  Not only do they presume that the effects they observed in the aforementioned analyses are causal and represent a net lifesaving effect of the 21 drinking age, which we doubt, they also leave out much of the other side of the ledger with respect to alcohol consumption by ignoring or dismissing several potential economic benefits associated with it.  They also ignore the likely adverse effects of a high drinking age on social cohesion, as well as the fact that forcing alcohol use underground makes it far more dangerous than it has to be, to say nothing of the value of individual liberty.  And the by the same measures, the social cost per drink would likely be at least as high for people over 21 on balance, especially 21-24 year olds who are the most likely of any age group to drive drunk (both in the USA as well as countries with lower drinking ages).  Using their logic, even bringing back Prohibition could potentially be justified, and we all know how well that worked out. 

On balance, the 21 drinking age is an EPIC FAIL.  And even more so are the tired, old attempts to justify it.

QED