Is it Safe to Be High on the Highway?

In case you haven’t heard, California’s Proposition 19 was defeated last week. The initiative, which would have revoked state laws that prohibit the possession and small-scale cultivation of marijuana, drew only 46 percent of the vote. I’m no expert on what this will mean for incarceration rates, California’s budget, drug-related violence or the sales of Phish concert tickets, but what will keeping pot illegal mean for transportation?

A study by the RAND Corporation concluded that passage of Prop 19 would have reduced pretax marijuana prices by something on the order of 80 percent. Chapter one, page one of your microeconomics textbook tells you that consumption of a good rises as its price drops. So it’s probably safe to say that cannabis use would have gone up, perhaps considerably (though RAND could not hazard a guess about the magnitude of the increase). This would certainly have translated into more driving under the influence of pot. And obviously, this would have led to more dangerous driving, more accidents and more carnage on our roads. Maybe.

As a recent review of the evidence by R. Andrew Sewell, James Poling and Mehmet Sofuoglu (all of Yale University’s School of Medicine) demonstrates, the links between cannabis use and dangerous driving are more complex than one might think.

One thing is certain: marijuana does erode your driving skills. Past work shows that “attentiveness, vigilance, perception of time and speed, and use of acquired knowledge” are all impaired by pot. In fact, an analysis of 60 studies (performed by G. Berghaus and B. Guo) found that pot impairs you “in every performance area that can reasonably be connected with safe driving of a vehicle, such as tracking, motor coordination, visual functions, and particularly complex tasks that require divided attention.”

Yet despite the fact that driving while high is self-evidently a terrible idea, the evidence on whether stoned drivers are indeed a menace to society is surprisingly mixed. How can this be?

Surprisingly, most lab experiments show that stoned drivers perform little worse than sober ones, except when pot is combined with alcohol. Many researchers have concluded that this is because drivers who are high are very aware that they are impaired, and deal with their neurological and psychological deficiencies by adopting coping mechanisms to compensate – even overcompensate – for their altered state. This leads to stoned drivers driving at lower speeds, leaving greater following distances between cars and making fewer efforts to pass other vehicles.

Interestingly, it seems that pot impairs “automatic” functions, like reaction time or the ability to stay in one’s lane, while leaving “cognitive” functions (e.g. the choice of speed) comparatively intact. Alcohol, on the other hand, degrades cognitive function, with drunks much more prone to making risky choices behind the wheel.

Of course, lab experiments may be a poor reflection of real-life driving behavior: after all, subjects know they’re being studied. Lab work is a better reflection of what people can do than what they actually do.

So numerous studies have looked at pot’s effects on actual driving behavior. This is primarily done using two methods. One is culpability study, which determines which driver was at fault in an accident and whether the guilty party was more likely to be stoned than the innocent one.? Some of these inquiries have found drivers who are high are more likely to be at fault in crashes, but many conclude that stoned drivers are no more dangerous – and indeed might be less dangerous – than others.

The ambiguity in the literature is due to numerous factors. For example, the mechanics of testing for pot use are tricky. Also, even when elevated accident levels for pot users have been documented, it could be argued that this is because they tend to share certain common characteristics – youth, male gender, a predisposition to risk-taking – that are more responsible for the accidents than the pot itself. A couple of studies suggest that this is the case. In all, the jury’s still out.

The other main method for studying this issue is the case control technique. As the name suggests, stoned drivers are compared with an appropriate control group. The trick is finding that control group, and being sure it matches well with the stoned driver group, something most studies to date have often not done satisfactorily. In all, as with culpability studies, the results from case control studies have been inconclusive.

There are three things, however, that we do know conclusively. First, driving stoned is not “safe;” impairment unquestionably exists. Sewell et al. suggest that “patients who smoke marijuana should be counseled to have a designated driver if possible, to wait at least three hours after smoking before driving if not, that marijuana is particularly likely to impair monotonous or prolonged driving, and that mixing marijuana with alcohol will produce much more impairment than either drug used alone.”

Second, the kind of ambiguous results we get for stoned drivers have never been found with respect to drunk drivers, who have proven conclusively and ad nauseum to be a serious menace, as I’ve written about extensively.

Finally, since the adaptive techniques used by stoned drivers – sticking to the speed limit, leaving plenty of room between you and the car in front of you, foregoing gratuitous passing of other vehicles – seem to be effective, think of what a better place the world would be if we all adopted them – while driving sober.


It does seem that (at least looking at the accidents involving fatalities) alcohol use while driving is simply one of a series of bad choices that the drivers make. I don't know if the standard mantra that alcohol impairs your judgement so that you are more likely to drive more aggressively etc is true, or whether people who get behind the wheel of a car when they can barely see straight are also more likely to make other bad driving choices (which is to say-- they probably drive aggressively when sober).
Another example is seatbelt use. Certainly seatbelts save lives but also the kind of person who is likely to not wear a seatbelt is also likely to make other poor driving choices. So if half of fatal accidents are single vehicle accidents, I would say that one of the most important issues is driver behavior-- choice of speed in particular. I am not seeing the "couldn't stop in time" accidents-- I'm seeing people losing control of their cars and either hitting a tree or an oncoming car.
It is also not lost on me that the "I know I'm impaired so I drive more carefully" argument has been used by drinkers in the past as well...


J Omega T

Hmm. The adaptive techniques described in the last paragraph make sense. But one unintended consequence, based on my personal observations, concurred with by those I've told about them, would seem to be an increase in road rage exhibited by other drivers.

So, if I avoid doing something stupid, but thereby cause someone else to do something stupid, have we really accomplished anything?


Any Cop worth his/her salt will tell you it's the juicers causing the problems out there.
I propose an all electric computer controlled individual transportation system. No insurance needed. No traffic jams. No licence. No pull-overs. No internal combustion engines. And the vehicle travels as fast as the system will allow. Just sit back and leave the driving to IBM.


Maybe chapter one in your economics books tells you that demand decreases with increase in price, but you might consider reading the next few chapters before you come to any conclusions.

Degree of demand, or elasticity, is higher for some consumer goods than others.

A community that craves ice cream, for example, is willing to pay more to satisfy their craving.

So the demand for controlled substances may not be affected much by an increase in price. For those who want something bad enough, the price they're willing to pay is often more than you realize.


very sorry, i had read your article too quickly and made my own error! i do agree with you!

Drill-Baby-Drill Drill Team

Since dope is illegal, no driver will admit to being stoned at an accident scene. There is motive to cover up any drug use.

Police officers should do hair or blood test to drivers to check for THC. However the ACLU complains right now about intrusion of privacy for simple alcoholic breathalyzer tests.

I think any mind altering drugs including alcohol, sleep drugs and antianxiety medication can affect driving. Dope is a mind altering drug.

I do not think the proper controlled studies have been performed to really state that pot makes drivers safer than a blinded control. In fact, the reckless aimless attitude of the stoned, make them poster children for irresponsibility.

" Yes there was an accident, Yes several were killed. But screaming about it doesn't help! Hey Man! chill out.! Puff on a toke and mellow out. This whole thing is kind of funny...."


There's also the small issue that if pot prices went down, many would get high instead of getting drunk. It's not lab-tested or anything, but I'd guess more pot users plan on not driving before getting high than drinkers plan on not driving before getting drunk.

Ian Callum

The question of how legal prohibition impacts drug usage is probably not as clear cut as it may seem. A study of the usage rates in countries where laws were changed could provide more information. My opinion is that people who are predisposed to use drugs generally will not comply with drug laws, so the laws are largely superfluous. I'd guess that only a minor increase in use would follow legalization, but without reliable data there's really no way to know.


Some time back, I was on percocet and hydrocodone due to severe surgical complications. Since I have a young son, I DARED not drive, even though, to be honest, I have no doubt that I would have fared well (maybe, as the article indicates, because I would compensate for my being on the drugs).

However, while handling machinery and other potentially lethal devices may not be a good idea for anyone on ANYTHING (e.g., Tylenol P.M), I can tell you that when I was on these pain killers, I was not "stupid" or in a dumb haze. Yes, I was certainly more gentle, more sweet, and more grateful than usual--that is, I was my "best self," I believe--I was not impaired from writing a letter, thinking logically, researching a matter, etc. The only time my wife noticed was when I told her one night, apparently out of character, I'm ashamed to admit, that I loved her.

"Are you on those pills?"

Indeed I was. But it only brought out what I truly felt. After all, I do love my wife--I just didn't tell her often enough.

This leads me to think that while, indeed, care must be shown when there is true potential for danger or damage, in hundreds of other day-to-day cases, most people, if they don't abuse the drugs, can operate perfectly well as receptionists, customer service reps, accountants, etc.

It does not affect your logical thinking (unless you take too much and go hazy), but it does affect your emotions. At least that's the story for me.


Bobby G

@DBDDT (#8), I agree with you most of the time, but not today. I really get the sense based on what you say that you are bashing something you've never tried. I don't see the example you propose at the end of your response ever really happening, not with marijuana at least, maybe with some other substances.

I've heard *cough* marijuana can increase paranoia which creates that offsetting behavior mentioned in the original post... paranoia not only about being caught, as DBDDT mentioned, but about danger in general, including being in a car accident.

Given that, it's hard to talk to Eric's final point about everyone driving with an increased sense of "paranoia"... it kind of doesn't make sense. Drivers use what they think are the appropriate amounts of "paranoia" already, and they can (for example) take a drug that increases paranoia. Increasing the base amount of care taken when driving will just mean the increased "paranoia" (I keep using this term for lack of a better one) makes the user even MORE careful, right? There's no maximum for carefulness besides maybe abstaining from the activity altogether (being too cautious to even drive), in which case one could even argue that consuming marijuana makes the roads safer! After all, one less driver on the road is safer than a driver that is 99.999% accident-averse.

To tie this all back in, I'm not saying driving while stoned should be legal, but just that I'm with Eric in that whether or not it's dangerous is still up in the air (no pun intended... "high"? nevermind). I do believe marijuana, specificially the mind-altering aspects of it, is not as dangerous as many people make it out to be, so I felt compelled to step in and defend it against such a confusing attack.



I've seen these compensation mechanisms in action: My stoner freind once had to move his car to avoid a parking ticket. He drove two blocks at about 10mph with a white-knuckled death grip on the steering wheel.

That "awareness of impairment" (or just paranoia) probably makes stoners a lot less likely to drive impaired in the first place.

I welcome everyone to come to a "Town Meeting" On a State level. is a private Citizen's attempt to offer a Neutral venue for further discussion on Legalizing Cannabis in California.

Legalization is a complex issue involving Billions of dollars in International Trade so it's no wonder the Federal Government wants to keep it's Cannabis Trade strategy in place an threatened California this election cycle.
On the issue of Stoned Driving.

It comes down to two things:
Alcohol has been rated the #1 most dangerous drug. I agree.
Alcohol impairment causes a driver to speed up trying to stimulate a neurologically depressed brain.
Driver drunk on alcohol don't think twice about driving under the influence for the most part and we allow it!
If we are that concerned why do we allow Bars to have Parking lots?
Cannabis imp[aired drivers slow down because the know they are impaired. They for the most part drive more cautiously.
Now it is not good to drive impaired on anything but people have and people will.
If I have to choose to ride with a drunk or a Stoned driver i would pick the Stoned driver. For one the Stoned driver will be more likely to be reasonable to deal with while a drunk driver is more likely to argue and fight since their brain is being attacked by a poison.
Oh and by the way, Google it to be sure, Cannabis helps protect the brain cells from the damage of alcohol. Also smoking some cannabis before you drink cuts down on the amount you drink while protecting your brain cells.
But Don't take my word.. Test out there theories for yourself.
Get drunk and see if you want to drive and on another day get stoned and see if you want to drive.
myself? I know when I am too stoned but I don't seem to know when I am too drunk.. Must be something about Alcohol damaging the brains cells while Cannabis doesn't
If you reader are thinking there was Proof Cannabis destroys brain cells well we know how that data was generated now after many freedom of information requests were denied. basically they starved monkeys for oxygen while providing cannabis smoke to them and then they died!
So has the war on Cannabis been fair? Nope it has been racial and economic. Now it is all about prison profits and law enforcement industries income at the Citizens expense.


T Charles

I am especially intrigued by this sentence:

"Some of these inquiries have found drivers who are high are more likely to be at fault in crashes, but many conclude that stoned drivers are no more dangerous - and indeed might be less dangerous - than others."

I am guessing that this can only be the case because driving, to begin with, is already an incredibly hazardous activity. Even in 2009, the safest year on record, nearly 100 individuals died each *day* in motor vehicle accidents in the US. This is death on the scale of 9/11 every month. Where is the outrage?

Eric M. Jones

Hey wow...The Brits (in The Lancet) have this figured out--alcohol is the most dangerous drug. MJ is rated MUCH LESS dangerous than tobacco. See:


It strikes me as queer that California voters turned down MJ. Maybe the stoners were unmotivated, or maybe the freely accessible MJ marketers didn't want taxes. Like Far-Out weird. Man....

My observation is that MJ causes virtually no problems compared with alcohol. And if you drive stoned, you are probably drunk anyway.


It is all about experience. Sans alcohol it is easy to drive when stoned. It is called maintaining. It is possible to work, study, test, create, and love when maintaining on pot.

I don't recommend computer file maintenance while stoned though.


Assuming that cannabis usage would increase following legalization and the subsequent price drop does not necessarily follow. As was mentioned earlier, those who consume cannabis will do so regardless of prohibition. This is true for any prohibited substance. Studies tracking the rates of drug consumption in Portugal, which decriminalized all narcotics in 2001, show an initial spike in usage followed by an overall decline in use of all substances. Prohibition does lend itself to black markets, which results in a net loss to the economy as a whole, as the revenue from taxation and sales is now diverted into the hands of criminals and out of the open market.

The main point to this post was regarding cannabis us and driving, however, and I would like to address that issue. Even if we agree with the assumption that legalization leads to increase in use, can we still assume that this leads to increased intoxicated driving? We would have to have the statistics for current number of people who drive while under the influence of cannabis as a starting point for any conversation. Reasonably we can assume that most people accept that while under the influence of substances that inhibit driving ability, they should not drive. Knowing, as you mention in the article, that cannabis does not interfere with cognitive processes like alcohol, we should assume that most people under the influence would not operate a vehicle. However, as you mentioned as well, there are other factors to consider when evaluating accident potential, notably age and sex, as many insurance company actuaries have calculated for decades. Also, we should consider that people who engage in risky or thrill-seeking behavior will continue to do so, regardless of prohibition of a substance. Can we assume that those accidents would happen regardless of the legalization of cannabis? I think it warrants investigation.

It is true that legalization would bring with it new problems that would necessitate new solutions. Just as we do for alcohol and other substances, education is key to informing people of the dangers of operating a vehicle while under the influence of any mind-altering substance.

What we must realize, though, is that the economic and social costs of continued cannabis prohibition far outweigh any benefit that may arise from retaining the status quo.



One thing that I think could be a boon from marijuana legalization, is abolish the concept of a "limit" on consumption beyond which you can't drive. Keep it simple - if you've smoked, you basically don't drive for until you are fully sober again. "Zero" tolerance, as it were. I think that the BAC limit we currently use leads to increased cases of drunk-driving, from drivers who calculate that perhaps they are below the limit when they aren't, or drivers who try to reach exactly the limit before driving (I for one remember, in my youth, when drunk and at a party and time to leave, that i'd frequently drink a water and wait half an hour before driving, thinking that was sufficient to drive sober). Having a similar policy for alcohol (ie, any level, not legal to drive) seems like it would be appropriate and not require drinkers/drivers to do any calculations.


Given the choice of the other drivers on the road having consumed alcohol or marijuana, I'd choose marijuana. In general, I think our society would be better off if we had more smoking and less drinking.


I know when I am too stoned but I don't seem to know when I am too drunk.. Must be something about Alcohol damaging the brains cells while Cannabis doesn't

Also smoking some cannabis before you drink cuts down on the amount you drink while protecting your brain cells.

quotes from -

I personally tried to quit smoking tobacco and hash but continued to drink and because alcohol impairs your judgment and dose so fast, I could not tell that I was getting to drunk (ending up doing stupid things).
So if you are trying to quit tobacco and/or hash do not compensate with alcohol.
I believe that if cannabis was made legal it would make people drink less and since cannabis does not impair your judgment as badly as alcohol dose, we would end up with a much safer society.


Legalization will of course give an immediate increase in use of cannabis but that will only be temporary due to the lowering in price when it becomes legal. Recent studies in Amsterdam shows a decrees in cannabis use among adults. One of the problems with banning smoking tobacco in pubs in Europe is that it made many elderly people antisocial as they no longer could go out to the pub. This being the case for cannabis connoisseurs today. They have no place to meet. Sad really.