The anti-pot group Project SAM claims drug test data show that marijuana legalization in Washington, approved by voters in that state at the end of 2012, already has made the roads more dangerous. The group notes with alarm that the percentage of people arrested for driving under the influence of a drug (DUID) who tested positive for marijuana rose by a third between 2012 and 2013. "Even before the first marijuana store opens in Washington, normalization and acceptance [have] set in," says Project SAM Chairman Patrick J. Kennedy. "This is a wakeup call for officials and the public about the dangerousness of this drug, especially when driving."
In truth, these numbers do not tell us anything about the dangerousness of marijuana. They do not even necessarily mean that more people are driving while high. Furthermore, other evidence suggests that legalizing marijuana could make the roads safer, reducing traffic fatalities by encouraging the substitution of marijuana for alcohol.
According to State Toxicologist Fiona Couper, the share of DUID arrestees in Washington whose blood tested positive for THC, marijuana's main psychoactive ingredient, rose from 18.6 percent in 2012 to 24.9 percent in 2013. That's an increase of more than 33 percent, as Project SAM emphasizes with a scary-looking bar graph. At the same time, the total number of DUID arrests in Washington rose by just 3 percent, about the same as the increases seen in the previous three years, while DUID arrests by state troopers (see table below) fell 16 percent.
These numbers do not suggest that Washington's highways are awash with dangerously stoned drivers. So why the substantial increase in positive marijuana tests? Lt. Rob Sharpe, commander of the Washington State Patrol's Impaired Driving Section, notes that additional officers were trained to recognize drugged drivers in anticipation of marijuana legalization. So even if the number of stoned drivers remained the same, police may have pulled over more of them as a result of that training.
Another point to keep in mind is that a positive THC test does not necessarily indicate a driver is impaired. The current cutoff for these tests in Washington is two nanograms of THC per milliliter of blood, less than half the state's new per se standard for DUID, which itself is not a very good indicator of impairment, especially for regular users. Sharpe concedes that the increase in positive tests could simply reflect an overall increase in marijuana use. When cannabis consumption becomes more common or more frequent in the general population, the percentage of DUID arrestees who test positive for THC is apt to rise, even if the percentage who are actually impaired does not.
According to the National Survey on Drug Use and Health, marijuana consumption has been rising across the country in recent years. The share of respondents who reported using marijuana in the previous month rose from 6.2 percent in 2002 to 7.3 percent in 2012. As Columbia University researchers Guohua Li and Joanne E. Brady pointed out a few months ago in the American Journal of Epidemiology, this increase in marijuana consumption has been accompanied by an increase in the percentage of drivers killed in car crashes who test positive for cannabinol, a marijuana metabolite.
But as with the increase in DUID arrestees who test positive for THC, this trend does not necessarily mean marijuana is causing more crashes. A test for cannabinol, which is not psychoactive and can be detected in blood for up to a week after use, does not show the driver was under the influence of marijuana at the time of the crash, let alone that he was responsible for it. "Thus," Li and Brady write, "the prevalence of nonalcohol drugs reported in this study should be interpreted as an indicator of drug use, not necessarily a measurement of drug impairment."
Another reason to doubt the premise that more pot smoking means more deadly crashes: Total traffic fatalities have fallen as marijuana consumption has risen; there were about 20 percent fewer in 2012 than in 2002. Perhaps fatalities would have fallen faster if it weren't for all those new pot smokers. But there is reason to believe the opposite may be true, that there would have been more fatalities if marijuana consumption had remained level or declined.
While marijuana can impair driving ability, it has a less dramatic impact than alcohol does. A 1993 report from the National Highway Traffic Safety Administration, for example, concluded: "The impairment [from marijuana] manifests itself mainly in the ability to maintain a lateral position on the road, but its magnitude is not exceptional in comparison with changes produced by many medicinal drugs and alcohol. Drivers under the influence of marijuana retain insight in their performance and will compensate when they can, for example, by slowing down or increasing effort. As a consequence, THC's adverse effects on driving performance appear relatively small." Similarly, a 2000 report commissioned by the British government found that "the severe effects of alcohol on the higher cognitive processes of driving are likely to make this more of a hazard, particularly at higher blood alcohol levels."
Given these differences, it stands to reason that if more pot smoking is accompanied by less drinking, the upshot could be fewer traffic fatalities. Consistent with that hypothesis, a study published last year in the Journal of Law and Economics found that legalization of medical marijuana is associated with an 8-to-11-percent drop in traffic fatalities, beyond what would be expected based on national trends. Montana State University economist D. Mark Anderson and his colleagues found that the reduction in alcohol-related accidents was especially clear, as you would expect if loosening restrictions on marijuana led to less drinking. They also cite evidence that alcohol consumption declined in states with medical marijuana laws.
Anderson et al. caution that the drop in deadly crashes might be due to differences in the settings where marijuana and alcohol are consumed. If people are more likely to consume marijuana at home, that could mean less driving under the influence. Hence "the negative relationship between legalization and alcohol-related fatalities does not necessarily imply that driving under the influence of marijuana is safer than driving under the influence of alcohol," although that is what experiments with both drugs indicate.
Arrest data from Washington are consistent with the idea that marijuana legalization could result in less drunk driving. Last year drunk driving arrests by state troopers fell 11 percent. By comparison, the number of drunk driving arrests fell by 2 percent between 2009 and 2010, stayed about the same between 2010 and 2011, and fell by 6 percent between 2011 and 2012. The drop in drunk driving arrests after marijuana legalization looks unusually large, although it should be interpreted with caution, since the number of arrests is partly a function of enforcement levels, which depend on funding and staffing.
Two authors of the Journal of Law and Economics study, Anderson and University of Colorado at Denver economist Daniel Rees, broadened their analysis in a 2013 article published by the Journal of Policy Analysis and Management. Anderson and Rees argue that marijuana legalization is apt, on balance, to produce "public health benefits," mainly because of a reduction in alcohol consumption. Their projection hinges on the premise that marijuana and alcohol are substitutes. If marijuana and alcohol are instead complements, meaning that more pot smoking is accompanied by more drinking, the benefits they predict would not materialize. Anderson and Rees say "studies based on clearly defined natural experiments generally support the hypothesis that marijuana and alcohol are substitutes." But in the same issue of the Journal of Policy Analysis and Management, Rosalie Liccardo Pacula, co-director of the RAND Corporation's Drug Policy Research Center, and University of South Carolina criminologist Eric Sevigny conclude that the evidence on this point "remains mixed."
A study published last month by the online journal PLOS One suggests that the substitution of marijuana for alcohol, assuming it happens, could affect crime rates as well as car crashes. Robert G. Morris and three other University of Texas at Dallas criminologists looked at trends in homicide, rape, robbery, assault, burglary, larceny, and auto theft in the 11 states that legalized marijuana for medical use between 1990 and 2006. While crime fell nationwide during this period, it fell more sharply in the medical marijuana states, even after the researchers adjusted for various other differences between states. Morris and his colleagues conclude that legalization of medical marijuana "may be related to reductions in rates of homicide and assault," possibly because of a decline in drinking, although they caution that the extra drop in crime could be due to a variable they did not consider.
How relevant is research on medical marijuana laws to the debate about broader forms of legalization? Highly relevant, if you take the view that medical marijuana is mostly a cover for recreational use, as prohibitionists tend to argue. In truth, the legal regimes governing the medical use of marijuana range from very strict (such as New Jersey's) to very loose (such as California's). But it is fair to say that a lot of people with doctor's recommendations in the looser states are recreational users in disguise. It therefore makes sense that legalizing medical marijuana would be accompanied by a decline in drinking and that such an effect would be magnified by full legalization.
This article originally appeared at Forbes.