Drunk-Driving Deaths Are Way Down, But …

How much do we really value human life? I know, we all believe life is priceless, but we make transportation decisions all the time that involve painful trade-offs that cost lives. Where do we draw the lines?

Last post, I wrote about the repeal of the national speed limit, a transportation policy that has (arguably) caused thousands of deaths without much apparent public concern or debate. And if on deeper reflection it seems trivial to cause carnage so we can go a little bit faster, keep in mind that we also kill simply so we can have fun. Shocking, but true.

Unless you were abducted by aliens and have been held captive on the planet Xandar for the last 100 years, you know that drunk driving is a bad thing. As if they really had to bother, myriad scientists and social scientists have documented its ill effects. (The first study on the topic was published in 1904.)

According to this review of the literature by E. J. D. Ogden and Herbert Moskowitz, alcohol impairs the analysis of sensory information, slows reaction time in complex situations, impedes multitasking, complicates the processing of visual information, slows tracking, reduces vigilance, dulls alertness, and reduces control of intricate movement. Obviously, none of this helps you to drive any better.

As if that wasn’t bad enough, drunk driving not only increases your chances of being in a crash, but it increases the physical harm that befalls you when a crash does take place.

How serious a problem is this for our society? It depends on your point of view.

Let’s start with the good news. In the last 30 years, we have seen a terrific reduction in the number of fatalities due to DUI. In 1982, we experienced 1.64 alcohol-related road fatalities per 100 million miles driven; in 2007, the figure was 0.43. As a society, we really do deserve a pat on the back for this achievement.

What gets the credit? As Peter Howat, David Sleet, Randy Elder, and Bruce Maycock report, some of this reduction in the fatality rate would have happened anyway, as cars and roads have gotten safer in general. But the majority of the improvement is due to stricter law enforcement, harsher penalties and other regulations, publicity campaigns by worthy organizations like Mothers Against Drunk Driving, and organizational policies and economic incentives.

But before we celebrate, things aren’t quite so simple. Obviously, drunk driving is still with us, and over the next couple of posts I’ll look at it in more depth, including the many things we could do to reduce it — if we really have the will.

Correction: Last post, I misspelled the name of one of the authors of the informative report on the speed limit that I cited: my apologies to Jon Bottom (i.e., Jon with no “h”).


From what I remember, the legal limit used to be 0.1% BAL. Nowadays, 0.08% is the standard. Organizations like MADD espouse the philosophy that "impairment begins with the first drink", and I would guess they would support further lowing the legal threshold.

However, I'm wondering if the data shows that an outsized proportion of serious / fatal crashes involve people who are very intoxicated--maybe 0.2% and up. And if so, wouldn't the limited the resources of police, etc. be better utilized by focusing on these people, compared to finding people with BAL's between 0.08% - 0.1%?

bobby g

As economists, you should know the importance of counterfactuals. Please provide comparison data on deaths per 100m miles for the same 25-year window; that will help isolate the effects of drunk driving awareness and prevention vs. other general safety initiatives and improvements in car safety.


We could dig all the roads up, and burn down all the gasoline refineries. That would save at least one child, so its clearly worth it.


"I know, we all believe life is priceless, but we make transportation decisions all the time that involve painful trade-offs that cost lives. Where do we draw the lines?"

Whenever there's a traffic jam to do with a suicidal person or a person who's 'lost it' for whatever reason, the airwaves fill w/ CB calls for the police to just shoot the person and get it over with.

Collectively, people don't value lives at all if it affects their traffic flow, whatever they may profess individually.


I like the idea of a series.

In fact, it could be a part of a larger series (or even a book!) titled, "How to Legislate Against Stupidity."


Lets just require all cars have the devices you have to blow in to start your car. Whatever they cost now, surely it would be much cheaper if we made millions per year. Not foolproof, but I bet it would have a huge effect. They require this of repeat offenders so it must have some effect.

Also, why don't we require this of commercial pilots before every flight?


I'm thinking built-in breathalyzer and mandatory BAC test wired into the starting program of a vehicle.


The solution will inevitably happen when we have robotic cars. There are several technologies leading towards this already, but it will end up with autonomously driven vehicles. I would guess this is 15-20 years away.

I would be interested in a comparative analysis of drunk driving with other crimes where there is no harm committed. Most things defined as crimes involve someone being harmed or the perp having the intent to cause harm, neither of which occur in drunk driving (in cases where no accident occurred). What other laws act this way and does this make drunk driving a victimless crime?


Of course, cracking down on drunk-driving could have some nasty side-effects in the job market.

Bartenders, liquor store clerks, beer truck drivers, police officers, tow truck drivers, impound lot attendants, bail bond agents, EMTs, doctors, nurses, coroners, casket makers, grave diggers, grief counselors, prosecutors, defense attorneys, trial lawyers, claims adjusters, jail guards, and probation officers will all suffer if drunk-driving is eliminated.

In addition, the people who would have been killed in drunk-driving incidents won't need replacements.

In this economy, we need pro-job policies, but cracking down on drunk-driving only benefits one special interest group, taxi drivers, while infringing on the sacred right of Americans to drink, drive, and kill with relative impunity.

Nicholas Ferrante

A more robust network of public transit options would deal with the drunk driving and driving while texting problem simultaneously.


"[A]s cars and roads have gotten safer in general", are there relevant statistics on how many accidents total or accidents requiring police assistance per million miles driven we now have? That is to say, do we have any proof that or ability to quantify how much drunk driving-related accidents have declined, let alone that drunk driving itself has declined? I would imagine both have, but the statistics cited do not seem to tell a particularly deep story about drunk driving, only about the fatality aspect.


"How much do we really value human life?" I have no respect for people who use death statistics to issue some sort of self-righteous crusade against this or that. Why not just require helmets for all automobile drivers and passengers? That would SUBSTANTIALLY increase accident survivability. More so than decreasing speed limits. More so than adding every sort of airbag. Don't agree we should impose helmet laws? Well, I guess you just don't have much value on human life, right?


Lets just require all cars to have the devices that govern the speed at 5 mph (or lower) That way no one dies!


I accept your premise that higher speed limits leads to more automobile fatalities, but I must dispute that it has caused more deaths on net. If higher speeds make transportation costs lower, then our society is richer as a result in ways that make us healthier.

As a thought experiment, consider what would happen if the speed limits were lowered to 5 mph. There would be no more deaths on the highway (even the drunk drivers would be ok), but most of us would starve to death within 6 months.

Alex J

Channeling my inner MADD:
We should just ban alcohol. If no one can get drunk in the first place we will have no drunk driving occurring. Problem solved

Mark Wolfinger

How much will does it take to pass a serious law?

1) One drunk driving conviction gets a severe warning and a shot-term license suspension

2) No more chances. A 2nd conviction during an entire lifetime should result in mandatory loss of license - in all 50 states (this must be a federal law with teeth). No exceptions.

3) Anyone caught driving when the license has been lost via #2, faces big trouble

a) If intoxicated, mandatory 20 year prison sentence. Plus loss of car, a la RICO. No possibility of early parole. Keep dangerous people off the streets

b If not intoxicated, Mandatory one year prison sentence plus loss of car.

4) After being convicted via #3, next conviction is mandatory life imprisonment, confiscation of all assets. no possibility of parole. Does not matter whether intoxicated.

You want zero tolerance. This is it.

rube goldberg

why don't we round up all those dangerous octogenarians that are out there behind the wheel?

clearly old age impairs the analysis of sensory information, slows reaction time in complex situations, impedes multitasking, complicates the processing of visual information, slows tracking, reduces vigilance, dulls alertness, and reduces control of intricate movement.

incarcerate and impoverish them for our collective safety.

rube goldberg

can I count on an offshoot of MADD emerging, to be known as...

Mothers Against Doddering Drivers?


MADD is an organization that isn't "worthy" of anything but criticism. They have evolved into a powerful neoprohibitionist lobbying organization, which is why their founder left the organization.

Furthermore "alcohol-related" and "alcohol-caused" are NOT the same thing and virtually every piece of literature published by the NHTSA flagrantly conflates the two.

Sobriety checkpoints are already in use in several parts of the country (http://tucsoncitizen.com/hot-off-the-press-release/2010/04/12/dui-checkpoint-hits-tucson-streets-thursday/), which the courts have upheld as legal in the face of the fourth ammendment.

You can not and will not EVER stop the .25+ drunks from driving, and the .12 and below are not dangerous.

"They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety."


Due to economic conditions:
1. People drink @ home now
2. Their cars were reposessed.