A new study published by the Institute for the Study of Labor, a German think tank, finds that medical marijuana laws in the United States are associated with a 9 percent decline in traffic fatalities. That result is based on data from the Fatality Analysis Reporting System for 13 states that legalized medical use of cannabis between 1990 and 2009. The study's authors, Montana State University economist D. Mark Anderson and University of Colorado at Denver economist Daniel Rees, argue that the most plausible explanation is the substitution of marijuana for alcohol, since laboratory research suggests that smoking pot impairs driving ability substantially less than drinking does (citations omitted):

Laboratory studies have shown that cannabis use impairs driving-related functions such as distance perception, reaction time, and hand-eye coordination. However, neither simulator nor driving-course studies provide consistent evidence that these impairments to driving-related functions lead to an increased risk of collision. Drivers under the influence of marijuana reduce their velocity, avoid risky maneuvers, and increase their "following distances," suggesting compensatory behavior. In addition, there appears to be an important learning-by-doing component to driving under the influence of marijuana: experienced users show substantially less functional impairment than infrequent users.

Like marijuana, alcohol impairs driving-related functions such as reaction time and hand-eye coordination. Moreover, there is unequivocal evidence from simulator and driving-course studies that alcohol consumption leads to an increased risk of collision. Even at low doses, drivers under the influence of alcohol tend to underestimate the degree to which they are impaired, drive at faster speeds, and take more risks.

In support of their hypothesis that these differences explain why states with medical marijuana laws have seen especially big drops in fatal car crashes, Anderson and Rees cite data from the National Survey on Drug Use and Health for three states that enacted such laws during the last decade: Montana, Vermont, and Rhode Island. Marijuana use by adults rose more in those states after they adopted their laws than it did in neighboring states, a difference that was statistically significant in Montana and Rhode Island. (At the same time, the survey data do not indicate that medical marijuana laws boosted consumption among minors—a charge commonly made by drug warriors.) Anderson and Rees also note that the drop in fatal car accidents associated with medical marijuana laws was statistically significant for crashes involving alcohol but not for other crashes, and they present some evidence that alcohol consumption has declined in medical marijuana states.

Assuming that people are substituting marijuana for alcohol, Anderson and Rees note, the impact on traffic fatalities might be due to shifting locations of consumption rather than the difference between the two drugs' effects on driving ability:

Alcohol is often consumed in restaurants and bars, while many states prohibit the use of medical marijuana in public. Even where it is not explicitly prohibited, anecdotal evidence suggests that the public use of medical marijuana can be controversial. If marijuana consumption typically takes place at home, then designating a driver for the trip back from a restaurant or bar becomes unnecessary, and legalization could reduce traffic fatalities even if driving under the influence of marijuana is every bit as dangerous as driving under the influence of alcohol.  

Either way, this benign effect of medical marijuana laws hinges on a tradeoff between a recreational drug and a purportedly medicinal one:

Because it is prohibitively expensive for the government to ensure that all marijuana ostensibly grown for the medicinal market ends up in the hands of registered patients (especially in states that permit home cultivation), diversion to the illegal market likely occurs. Moreover, the majority of MMLs allow patients to register based on medical conditions that cannot be objectively confirmed (e.g. chronic pain and nausea).

So while the study undermines one complaint about medical marijuana laws—that they encourage minors to smoke pot—it reinforces another: that they allow people to get high under the cover of treating unverifiable illnesses. Then again, those who believe adults should not need an officially recognized excuse to consume marijuana might view that aspect of the laws as a feature rather than a bug. 

You can read the whole report here (PDF). I discuss stoned vs. drunk drivers here and here. I discuss pot-smoking malingerers here.

[Thanks to Eric Sterling for the tip.]