Defusing The Powder Keg

Or, "It's the Final Countdown!"

The 21 drinking age is an ageist and illiberal abomination that has NO place in a free society.  It's very existence to this day is evidence that we are living in the wrong timeline.  And the sooner it is jettisoned to the trash heap of history, the better.

Twenty-One Debunked has, unwaveringly and  uncompromisingly, since our founding in 2009, sought to speedily lower the legal drinking age (in every sense) to 18, and not a day later.  Period.  That is, the same rights that people over 21 currently enjoy, should be democratized to include 18-20 year olds as well, and yesterday is NOT soon enough!  That is our North Star for the near term, basically.  (And for the very long term?  Don't ask!) The relative purgatory and limbo of setting it at 19, 20, or a split/graduated age limit first has always been best considered a journey, NOT a destination. 

That said, doing so is a LOT easier said than done politically.   The Overton Window is clearly NOT in our favor, to put it mildly (that brief moment in the late 2000s notwithstanding, whose ship has long since sailed).  And while the objections to lowering it to 18 are essentially specious and spurious, they are not entirely pulled out of the ether, either.  So much "referred pain", as it were, is also present as well.  Not least of which is the Law of Eristic Escalation (i.e. Imposition of Order = Escalation of Chaos), which as a corollary, causes a "chaos deficit" to compound over time.  This sets a massive "powder keg" that is best defused carefully and gingerly, to avoid too much chaos unleashed all at once in the short term, and the reactionary backlash that would follow.  Plus, the fear of it would make it harder to get any bill that would lower the drinking age to have a prayer of passing.

So here is a potential "menu of options" to defuse the powder keg and its associated fears, and make the age-lowering bill easier to pass, and all of which should have a sunset clause built in so as not to defeat the purpose:

• Keeping the Zero Tolerance driving BAC age limit at 21 for the time being, or better yet, expanding it to include the first two or three years of having a driver's license, or 21, whichever is longer, largely similar to some other countries with lower drinking ages.  And keep that in place for the first couple of years at least.  (To defuse the fear of DUI.)
• Keeping the purchase age limit for kegs, cases, and other large quantities of alcohol at 21 for the time being, reasonable limits on quantities purchased per transaction, and no more than one transaction per person per store per day for people under 21.  And keep that in place for the first year or couple of years.  (To defuse the fear of high school keggers, as well as high school drinking more generally.)
• Phasing down the drinking age, or at least the purchase age (especially on-premise at bars and restaurants), by one month each day.  That is, 20 years 11 months, 20 years 10 months....until it is 18 years and 0 months.  And that phasedown will take all of....36 days to complete.  (Hat tip to a guy on Facebook many years ago who came up with that idea, which we call "The Final Countdown", like the famous song by Europe.) (To defuse the fear of overall chaos in general, especially at bars and surrounding areas.)
• Best done through explicitly allowing "local option" to do so, and ideally kept only for a limited time, perhaps we could also set only the off-premise purchase age (but NOT drinking age, furnishing age, or on premise purchase age!) to 19 without conditions, and 18 only with a college or military ID, or proof of high school diploma or GED.  (To further defuse any remaining fear of high school drinking and straw purchasing for younger friends.)
• "25 Mile Rule": To alleviate fears of "blood borders", for on-premise service in any bar or restaurant within 25 miles from the border of a neighboring state with a higher drinking age, no out-of-state IDs will be accepted for anyone from that state for anyone under that age, except 1) students within a college town (or a 5 mile radius from campus) showing valid college ID, 2) overnight guests being served in a hotel where they are staying.
• Perhaps temporarily (for the first few months or first year or so) keep the off-premise purchase age at 21 (or 20) for hard liquor, or any beverages exceeding 15% or 20% ABV, similar to what Finland (off-premise) and Norway (both on and off-premise) currently do.  (To further defuse fears.)
• (Holds nose while grudgingly saying this) Continue allowing bars and clubs to set their own age limits higher than the legal drinking age, which many already do. (To defuse the (mostly apocryphal) fear of young people engaging in rowdiness, chaos, "pregaming", and arriving already at least half-drunk and not buying much at the bar.)
• And be sure to include a severability clause!

Indeed, while we are NOT wedded to any of these, and would rather NOT start out with any compromised position lest we have to compromise further, we endorse and strongly encourage at least the first three items on the list in the short term.  And of course, it should go without saying that it should be accompanied by a renewed crackdown on drunk driving, drunk violence, drunk vandalism, and drunk and disorderly conduct, for ALL ages.  For DUI specifically, combining sobriety checkpoints (subject to constitutionally-correct protocols), roving and saturation patrols, and especially "fish in a barrel" aka "Jack in the box" (police waiting patiently outside of bars for would-be drunk drivers to stumble to their cars, and gotcha!) would essentially make obsolete any concerns of increasing drunk driving, even if the drinking age was lowered to 18 overnight in one fell swoop.

Simple.  Elegant.  Surgical. 

So what are we waiting for?  "It's the Final Countdown!"

What Australia Gets Right (Updated)

One thing our movement has a habit of doing is comparing the USA to Europe for the purpose of ascertaining what the effects of a lower drinking age would be like.  While there is some truth to such a comparison, the pro-21 side routinely calls us out on the important differences between here and there.  For example, they have much better public transportation than we do, they are more urbanized, driving licenses are much more difficult to obtain, gas prices are much higher, and thus they are much, much less of a car culture that we are.  All of which would dramatically affect traffic fatalities and skew any comparisons.  As a result, Twenty-One Debunked typically prefers to make comparisons to Canada instead, which is also a car culture that is the most similar to the USA.  And they have seen a similar or faster drop in traffic deaths than the USA despite NOT raising the drinking age to 21, and their traffic death rates have been consistently lower than the USA.  But there is also another major car culture as well with a drinking age of 18--Australia.

In the Land Down Under, they have in fact seen a faster drop in alcohol-related traffic deaths than the USA and even Canada since 1982 despite keeping the drinking age at 18.  For all age groups.  Not only that, a recent study in Australia found essentially no link between being able to drink legally and motor vehicle accidents of any type, at least not in the state that was being studied, New South Wales.  Using a regression discontinuity design similar to the sort that pro-21 researchers have been doing lately, they found NO discontinuous jump in such deaths or injuries in young people upon turning 18.  This stands in stark contrast to the USA, in which various pro-21 researchers have found a significant jump in alcohol-related deaths and injuries among young Americans upon turning 21, an increase that in many cases lingers well beyond one's 21st birthday.

Of course, such a phenomenon is not unique to the USA, as a jump in alcohol-related deaths and injuries has also been observed in Canada at their own respective MLDA (18 or 19, depending on the province), albeit of a shorter duration than in the USA and limited primarily to males who participate in "extreme" binge drinking.  It would seem that a "powder-keg" effect is unfortunately an almost inevitable consequence of the very concept of a drinking age, regardless of what it is.  So what does Australia do right that seems to defuse the powder keg?

Most importantly, Australia has tougher DUI laws, and tougher and more frequent enforcement of such laws.  For example, not only are penalties tougher, but the BAC limit is 0.05 (as opposed to 0.08 in the USA in every state except Utah) and they have random breath testing (RBT), which has been effectively ruled unconstitutional in the USA, and even Canada only began doing it in 2018.  Though one could argue that nowadays the USA effectively practices a form of "de-facto RBT" via a combination of "no-refusal" laws (i.e. the police often have a judge on speed-dial to issue a telewarrant to compel those who refuse to be tested) and often quasi-randomly pull people over for trivial reasons as a pretext and use that as an excuse to test drivers, and the initial effectiveness of RBT in Australia seems to decay over time to converge to the still-significant level of effectiveness demonstrated by American-style sobriety checkpoints (in 38 states) and/or roving/saturation patrols (in all states).   So America can indeed do what Australia does, we just need to be more creative about it.  Additionally, driver's licenses are harder to get and easier to lose over there than in the USA, and the road test there is significantly more difficult as well.  Alcohol excise taxes are also higher in Australia as well.  But truly the biggest and most salient difference is the seriousness with which they take the issue of drunk driving.  You really do NOT want to get busted for DUI in the Land Down Under!

(They also have zero tolerance BAC laws as well, albeit for the first three years of licensed driving regardless of age.)

Of course, the picture down under is not entirely rosy.  Australia's drinking culture is quite extreme even by American, Canadian, and British standards (though tame by New Zealand standards), and binge drinking is quite the art form over there.  Indeed, even the aforementioned study found that while traffic deaths and injuries do not increase discontinuously at 18 when they become legal to drink, there is still a discontinuous increase in hospital visits and admissions for alcohol poisoning and injuries from assault at 18.  But the fact that, even in a country with a more Anglo-Celtic, drink-to-get-drunk culture than the USA, it is nonetheless possible to break the link between drinking and drunk driving casualties, really speaks volumes indeed.

In other words, lowering the drinking age in the USA should really not be something to fear.  But we also need to get tougher on drunk driving if we wish to continue the progress of decades past.

UPDATE:  We should clarify that we do NOT support Australia's zero tolerance per se policy for driving with cannabis.  Since THC is fat-soluble, the pharmacokinetics of cannabis are far too complex, and thus do NOT lend themselves to any sort of per se limits, let alone zero tolerance.  If there must be any official limit, is should be "permissible inference" or "prima facie" instead, and of course significantly greater than zero.  After all, merely testing positive at a too-low cutoff days or longer after last use should NOT be seen as evidence of current impairment.

Would A Graduated Drinking Age Be Better? (Part Deux)

In our previous post, we discussed the idea of possibly having a split (graduated) drinking age for the first few years of lowering the drinking age, whether for quantity limits, beverage type, or both.  And we concluded that we would be open to that, but ONLY if that would eventually sunset to make it 18 across the board within a few years at most.  Remember, our goal at Twenty-One Debunked remains to fully lower it to 18, and not a day later.

Another idea could be to lower it completely to 18 for on-premise purchases (bars and restaurants), while requiring anyone 18-20 buying alcohol at stores for off-premise consumption to simply have someone 21 or older with them for the purchase, except if those 18-20 have and present a college or military ID.  And they could not have anyone under 18 with them (except their own children) either.  That would be more than enough to discourage 18-20 year olds from buying for their younger friends, which of course would still remain just as illegal as it is now.  

And of course, 18-20 year olds should be fully allowed to drink, possess, and be furnished alcohol in private as well, just like people over 21 currently are.  That would be head and shoulders better than ONLY allowing them to possess and consume alcohol on premise, a truly half-baked idea which would thus create a perverse incentive to drive under the influence, especially in rural areas where such a policy would be tantamount to legalized entrapment in practice.

What about the fear of "blood borders" between states that lower their drinking ages while adjacent states remain 21?  Well, first of all, that is really no worse than the problem of dry counties adjacent to wet counties in terms of perverse incentives to drive under the influence.  (And we sure as hell don't blame it on the wet counties!)  Secondly, they could always have it be where on premise establishments less than X number of miles from the state line of such adjacent states would not be allowed to accept out of state IDs for people under 21 from those specific states, except in college towns for those who also show a college ID.  And the same "temporary residence" exception should also apply to hotels serving their overnight guests as well.  There is a precedent for Wisconsin (a state that has experimented with just about every possible alcohol policy under the sun) doing something similar before 1977 when it was rescinded.

Another "novel" idea:  simply enforce existing laws against actual drunk driving better.  Checkpoints and/or roving patrols can really go a long way yo reduce the problem.

But should we even bother to start out with a compromised position at all?  That remains an open question.

The pro-21 crowd really lives in an alternate reality, it seems.  Just recently, I was debating someone on Reddit that kept on fruitlessly picking apart the methodology of Miron and Tetelbaum's groundbreaking 2009 study that thoroughly debunked the idea that the 21 drinking age saved lives.  The person I debated with kept on straining at proverbial gnats and swallowing proverbial camels, tossing red herrings, using the word "weird" repeatedly for anything they disagreed with, and kept failing to actually refute it.  The pro-21 crowd really cannot to see the forest for the trees, it seems.  They are apparently way too left-brained to do so (and perhaps smooth-brained as well).

The point being:  trying to reason with them is fundamentally an exercise in futility.  And compromising with them would also likely be just as futile as well.

Would A Graduated Drinking Age Be Better?

Twenty-One Debunked has long called for the drinking age to be lowered to 18, completely, and yesterday is not soon enough.  And that hasn't changed one iota since our founding in 2009, and it never will.  That's our North Star.

But what about a phase-in period to make the age lowering more likely to get passed at all?  We have already proposed for the first few years of the new law, to have it remain at 21 for kegs, cases, handles (large bottles) of liquor, and other bulk quantities, while otherwise lowering it to 18 yesterday, and having the higher age limit automatically sunset after X number of years (likely three to five years).  And also, keep the zero tolerance age at 21 for drinking and driving as well for the first few years.  The question remains, however:  what if that still is not enough to realistically get it passed?  We face an extremely uphill battle these days, after all.

Thus, we at Twenty-One Debunked would now grudgingly support, and ONLY for the first few years at most, also keeping the purchase age at 21 for the first few years for hard liquor, or really anything with more than 18% alcohol by volume.  Everything else in non-bulk quantities would be lowered to 18 right away, while the higher age limit would automatically sunset to 18 within a few years.  The higher age limit would not apply to drinking the liquor, of course, only for the specific act of buying it.

Alternatively, the higher age limit can be phased down gradually to 20, then 19, then 18 as well.

The higher age limit could also be kept for all internet, phone, or delivery app orders of alcohol for the first few years as well, and also perhaps have shorter trading hours allowed for off-premise sales for those below that age.

All of this would pour cold water over any real or imagined fears of a short-term increase alcohol-related problems and casualties among teens and young adults, particularly those involving high school students and keggers.  (Longer-term problems and casualties resulting from a lower drinking age have already been thoroughly debunked by Miron and Tetelbaum (2009) and several other studies, of course.)

Hey, if it actually gets us to our real goal sooner than being absolute purists about it would, why not?  Especially in light of the fact that there are more and more places (coffee shops, movie theaters, etc.) these days that serve beer, wine, cider, and/or alcopops, but not hard liquor.  That said, we will NOT support any further compromises beyond that, as that would be a compromise OF a compromise, and thus that would ultimately vitiate our goal of lowering to 18.

So what are we waiting for?

Latest Regression Discontinuity Study Only Confirms Powder Keg Theory (Again)

As we have noted before back in 2016, the latest type of drinking age studies, i.e. "regression discontinuity" (RD) studies, superficially appear to support the 21 drinking age, but upon closer examination actually kind of imply the opposite, namely 1) there is really nothing magical about 21, and 2) delaying legal access to alcohol to 21 compared with 18 only creates an even larger ticking time-bomb and powder keg that goes off at 21, with no real net benefit.

In 2016, pro-21 researchers Kitt Carpenter and Carlos Dobkin were at it once again.  As you may recall, we at Twenty-One Debunked have critiqued much of their previous work, particularly their use of the "regression discontinuity" approach.  The researchers found a significant jump in statistical death rates, arrests, and stuff like that immediately after young people turn 21 compared with before.  And their latest study seems to be more of the same, this time looking at non-fatal injuries as measured by both ER visits and inpatient hospital admissions.

Oddly, these researchers actually (and without even a hint of irony) claim that these studies show that the 21 drinking age is effective in saving lives and reducing alcohol-related harm!  But we at Twenty-One Debunked see it rather differently--if anything, it shows that there is nothing at all magical about turning 21 that makes one invulnerable to the deleterious effects of excessive alcohol consumption.  And setting the drinking age at such an arbitrarily high age only sets a powder keg (pun intended) that goes off when young people reach that age.  The higher the drinking age, the larger the powder keg, it seems.  And it also shows that the Law of Eristic Escalation (i.e. imposition of order leads to escalation of chaos) is correct, as well as Fenderson's Amendment (the tighter the order is maintained, the longer it takes for the chaos to escalate, but the more it does when it does).  Hardly a ringing endorsement for the 21 drinking age!

Of course, Carpenter and Dobkin also find evidence of a jump in both drinking and alcohol-related deaths among Canadians upon reaching their MLDA (18 or 19, depending on the province), particularly among males, and the increase in mortality seems to be due to a sudden jump in "extreme" binge drinking.  It seems there is always a risk of increased alcohol-related harm in the short-run after suddenly turning legal, regardless of age.  But as much other research shows, there is good reason to believe that such an effect is worse and longer-lasting when the the legal drinking age is higher rather than lower.  And furthermore, a recent study in Australia (where the drinking age is 18 and DUI laws are tougher) found essentially no link between being able to drink legally and motor vehicle accidents of any type in the state of New South Wales.  Food for thought indeed.

On the plus side, one should also note that another recent study using a regression-discontinuity approach found that being able to drink legally reduced the consumption and initiation of hard drugs (cocaine, heroin, meth, etc.) among young people, with no effect on tobacco use.  And still another RD study finds that when alcohol retreats, cannabis advances (and vice-versa).  These findings blow yet another big hole in the junk science that is the roundly-debunked "gateway" theory of drugs.

Fast forward to 2018, when another researcher, UW-Madison Professor Jason Fletcher, conducted yet another regression discontinuity study in the USA, this time looking at previously unexplored risk behaviors and consequences relating to drinking.  In addition to replicating the results of the aforementioned previous studies, these novel measures included self-reports of drunk driving, boozy and risky sex, interpersonal troubles, violence, and of course, hangovers.  And not surprisingly, these effects also jumped at 21 as well, with larger effects on males than females.  And again, this is simply further confirmation of the powder keg theory.

So what do we call it when you do the same thing over and over again and expect different results?

UPDATE: A later study in Austria, where the drinking age is 16, also dovetails with this same theory as well.  Similarly, so does another study in Germany (also 16).

And another Canadian study, looking at sexual assault perpetration, finds mixed results.  That is, for provinces with a legal drinking age of 19, they find such a significant discontinuous increase (for male perpetrators only), but in provinces with a drinking age of 18, in contrast they find no significant jump at all.  A second Canadian study, looking at sexual assault victimization among women, found mixed results that varied by both location of occurrence as well as by legal drinking age, and had very wide confidence intervals.  It appears that in provinces with a drinking age of 18, it was primarily a shifting of the location where the assaults took place rather than an overall increase, while in the provinces with a drinking age of 19, it was more of an increase (though modest) for the "any location" category.

And finally, a later Dutch study also dovetails with this general theory as well in terms of crime victimization.  And in stark contrast to a recent American study, no effect was seen for sexual assault victimization in the Netherlands.

Put together, at least for sexual assault, it seems that drinking age laws merely shift the timing and location, and not really change overall rates in the long run.  And forcing drinking underground only makes it more dangerous than it has to be.   It's almost like rape is, in fact, actually caused by rapists, not by alcohol, victims, differences in drinking age laws, or anything else.  But hey, what do I know?  I'm just the 800 pound guerrilla in the room.

What Australia Gets Right

One thing our movement has a habit of doing is comparing the USA to Europe for the purpose of ascertaining what the effects of a lower drinking age would be like.  While there is some truth to such a comparison, the pro-21 side routinely calls us out on the important differences between here and there.  For example, they have much better public transportation than we do, they are more urbanized, driving licenses are much more difficult to obtain, gas prices are much higher, and thus they are much, much less of a car culture that we are.  All of which would dramatically affect traffic fatalities and skew any comparisons.  As a result, Twenty-One Debunked typically prefers to make comparisons to Canada instead, which is also a car culture that is the most similar to the USA.  And they have seen a similar or faster drop in traffic deaths than the USA despite NOT raising the drinking age to 21, and their traffic death rates have been consistently lower than the USA.  But there is also another major car culture as well with a drinking age of 18--Australia.

In the Land Down Under, they have in fact seen a faster drop in alcohol-related traffic deaths than the USA and even Canada since 1982 despite keeping the drinking age at 18.  Not only that, a recent study in Australia found essentially no link between being able to drink legally and motor vehicle accidents of any type, at least not in the state that was being studied, New South Wales.  Using a regression discontinuity design similar to the sort that pro-21 researchers have been doing lately, they found no discontinuous jump in such deaths or injuries in young people upon turning 18.  This stands in stark contrast to the USA, in which various pro-21 researchers have found a significant jump in alcohol-related deaths and injuries among young Americans upon turning 21, an increase that in many cases lingers well beyond one's 21st birthday.

Of course, such a phenomenon is not unique to the USA, as a jump in alcohol-related deaths and injuries has also been observed in Canada at their own respective MLDA (18 or 19, depending on the province), albeit of a shorter duration than in the USA and limited primarily to males who participate in "extreme" binge drinking.  It would seem that a powder-keg effect is unfortunately an almost inevitable consequence of the very concept of a drinking age, regardless of what it is.  So what does Australia do right that seems to defuse the powder keg?

Most importantly, Australia has tougher DUI laws, and tougher and more frequent enforcement of such laws.  For example, not only are penalties tougher, but the BAC limit is 0.05 (as opposed to 0.08 in the USA) and they have random breath testing (RBT), which is unconstitutional in both the USA and Canada.  Though one could argue that nowadays the USA effectively practices a form of "de-facto RBT" via a combination of "no-refusal" laws (i.e. the police often have a judge on speed-dial to issue a telewarrant to compel those who refuse to be tested) and often pull people over for trivial reasons and use that as an excuse to test drivers, and the initial effectiveness of RBT in Australia seems to decay over time to the level of effectiveness demonstrated by American-style sobriety checkpoints and/or roving patrols.  Additionally, driver's licenses are harder to get and easier to lose over there than in the USA, and the road test there is significantly more difficult as well.  Alcohol excise taxes are also higher in Australia as well.  But the biggest and most salient difference is the seriousness with which they take the issue of drunk driving.  You really do NOT want to get busted for DUI in the Land Down Under!

Of course, the picture down under is not entirely rosy.  Australia's drinking culture is quite extreme even by American, Canadian, and British standards (though tame by New Zealand standards), and binge drinking is quite the art form over there.  Indeed, even the aforementioned study found that while traffic deaths and injuries do not increase discontinuously at 18 when they become legal to drink, there is still a discontinuous increase in hospital visits and admissions for alcohol poisoning and injuries from assault at 18.  But the fact that, even in a country with a more Anglo-Celtic, drink-to-get-drunk culture than the USA, it is nonetheless possible to break the link between drinking and drunk driving casualties, really speaks volumes indeed.

In other words, lowering the drinking age in the USA should really not be something to fear.  But we also need to get tougher on drunk driving if we wish to continue the progress of decades past.

Latest Regression Discontinuity Studies Only Confirm Powder Keg Theory

This year, pro-21 researchers Kitt Carpenter and Carlos Dobkin are at it again.  As you may recall, we at Twenty-One Debunked have critiqued much of their past work, particularly their use of the "regression discontinuity" approach.  The researchers found a significant jump in statistical death rates, arrests, and stuff like that immediately after young people turn 21 compared with before.  And their latest study seems to be more of the same, this time looking at non-fatal injuries as measured by both ER visits and inpatient hospital admissions.

Oddly, these researchers actually (and without even a hint of irony) claim that these studies show that the 21 drinking age is effective in saving lives and reducing alcohol-related harm!  But we at Twenty-One Debunked see it rather differently--if anything, it shows that there is nothing at all magical about turning 21 that makes one invulnerable to the deleterious effects of excessive alcohol consumption.  And setting the drinking age at such an arbitrarily high age only sets a powder keg (pun intended) that goes off when young people reach that age.  The higher the drinking age, the larger the powder keg, it seems.  And it also shows that the Law of Eristic Escalation (i.e. imposition of order leads to escalation of chaos) is correct, as well as Fenderson's Amendment (the tighter the order is maintained, the longer it takes for the chaos to escalate, but the more it does when it does).  Hardly a ringing endorsement for the 21 drinking age!

Of course, Carpenter and Dobkin also find evidence of a jump in both drinking and alcohol-related deaths among Canadians upon reaching their MLDA (18 or 19, depending on the province), particularly among males, and the increase in mortality seems to be due to a sudden jump in "extreme" binge drinking.  It seems there is always a risk of increased alcohol-related harm in the short-run after suddenly turning legal, regardless of age.  But as much other research shows, there is good reason to believe that such an effect is worse and longer-lasting when the the legal drinking age is higher rather than lower.  And furthermore, a recent study in Australia (where the drinking age is 18 and DUI laws are tougher) found essentially no link between being able to drink legally and motor vehicle accidents of any type in the state of New South Wales.  Food for thought indeed.

On the plus side, one should also note that another recent study using a regression-discontinuity approach found that being able to drink legally reduced the consumption and initiation of hard drugs (cocaine, heroin, meth, etc.) among young people.  That blows yet another hole in the junk science that is the "gateway" theory.