Despite Marijuana Legalization, Traffic Fatalities in Colorado Continue to Fall


Last month I noted that marijuana legalization in Colorado does not seem to have interfered with a downward trend in traffic fatalities, which have been falling there since 2004. That period includes the commercialization of medical marijuana (which started to take off in 2009) and the legalization of recreational use (which took effect at the end of 2012). My former Reason colleague Radley Balko, who blogs about civil liberties and criminal justice at The Washington Post, points out that the downward trend in deadly crashes has continued since the beginning of this year, when state-licensed pot stores began serving recreational customers. According to data from the Colorado Department of Transportation, the total number of fatal crashes in the first seven months of this year was 258, down from 263 during the same period last year. Here are the monthly totals for 2014, compared to the same months in 2013:

Colorado Department of Transportation

The trend in Colorado is broadly similar to the national trend during the last decade. Balko notes that Colorado's drop since 2004 looks more dramatic when compared to miles traveled, which have continued to rise in Colorado while falling nationwide.

Does the continuing decline in fatal crashes mean that legalization is reducing fatalities by encouraging the substitution of marijuana for alcohol (which has a more dramatic impact on driving ability)? Not necessarily. In fact, from January through July the total number of alcohol-related fatalities was 91, exactly the same as the total for the first seven months of 2013:

Colorado Department of Transportation

Still, these numbers clearly are not consistent with warnings from prohibitionist groups such as Project SAM, which predicted that legalizing marijuana would mean more blood on the highways. Perhaps that grim prophecy will be realized one day, but so far it looks wrong. 

NEXT: John Stossel on Border Patrol Agents Without Borders

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Bullshit! We need to build a giant fence around CO before the reefer madness escapes into the other states. For the children!

    1. It’s to late for fences because of the second hand smoke. What we need is a Giant airvac that can rid the atmosphere of the dangerous skunk weed.

      1. Hmm, I wonder if that thing could also suck CO2 out of the air? We could combat this warming and save the children at the same time!

  2. Despite Marijuana Legalization, Traffic Fatalities in Colorado Continue to Fall

    Only because Obama. Could you imagine if BOOSH were president!?


    1. I am completely confident that vehicular fatalities will continue to fall since nobody will have to drive clear across town to find their dealer — they can have their bud delivered,right to their home 🙂

  3. In fact, from January through July the total number of alcohol-related fatalities was 91, exactly the same as the total for the first seven months of 2013

    I don’t put much trust in those numbers. If any party was in an accident was drinking, then it is ruled as alcohol related. So if a sober person runs a red light and kills someone who was drinking, it will be ruled as alcohol related. Even though the drunk would be alive had the sober person not run a red light.

    1. If any party was in an accident was drinking, then it is ruled as alcohol related.

      Not only that, but they don’t even consider BAC when declaring that it’s “alcohol related.” Yeah, that .01 really made me crash my car officer!

      1. Yeah, that .01 really made me crash my car officer!

        And this is why we need zero tolerance.

    2. I agree that the definition is problematic, but year-to-year comparisons are still meaningful as long as the definition remains the same. Some subset of “alcohol-related fatalities” really were caused by drunk drivers.

      1. Some subset of “alcohol-related fatalities” really were caused by drunk drivers.

        True. I’m just saying that whenever alcohol or drugs are involved, they always get the full blame. Even if the cause was a sober but negligent driver.

  4. So people don’t really drive in Colorado in the winter? That’s the only conclusion I can actually draw. These charts prove absolutely nothing, and it’s ridiculous to try to draw any conclusions about whether legalization is good or bad from them. This is why we lose the argument on legalization: people put out meaningless statistics/information instead of making an argument.

    1. The graph compares the numbers for the first seven months of 2014 with the numbers for the same period in 2013, so seasonal variations in weather have nothing to do with it.

      1. I know. But the only thing I can tell is that there are fewer fatalities in the winter in general. What the hell does that have to do with pot?

        1. It may have nothing to do with pot one way or the other. But the total number of fatalities so far this year is lower than the total number for the same period of 2013, which suggests that legal sales are not leading to more traffic deaths, contrary to what prohibitionists predicted.

          1. Did they predict more fatalities or more accidents related to marijuana consumption. You’re just proving that Colorado is a little safer for drivers. That could be the result of demographic changes, construction, laws pertaining to driving, etc. Honestly, the only meaningful data would be the ratio of marijuana related accidents to accidents where marijuana was not a factor.

            1. I think measuring by fatalities is problematic. Cars are much safer these days, so you have to be pretty reckless to kill yourself. And do stoned drivers tend to drive too fast? I thought the old joke was that they’d drive too slowly.

              I’d like to see a chart of total accidents.

            2. The point of the article is not to speculate on the cause, or absence of causes, of traffic fatalities. The point is to say, “there’s no spike in traffic fatalities since marijuana legalization, meaning that anti-legalization hysteria about how many more traffic fatalities there’d be if marijuana were legalized is not justified by the data.”

              1. I know what the point of the article is. The problem is that if marijuana related accidents increased (I have no idea if they did or did not because that vital piece of data is missing) it wouldn’t matter if the total number went up or down.

                Likewise, if the total number tripled, it wouldn’t matter if the percent related to pot decreased. All in all these charts are meaningless with regards to the marijuana debate.

                1. I wouldn’t say the charts are totally meaningless, because they do disprove the extreme case (fatalities). But they do nothing to disprove the more reasonable objection that legal pot will increase pot-related traffic accidents.

                  1. I just want some data that shows what happens in cases where individuals were high on marijuana while driving. I honestly don’t think it’s a good argument against banning pot either way (if it were we’d ban fast food…), but if that’s your argument, then that’s the necessary data.

                    1. There will never be any accurate data of impaired driving from marijuana until they start using a test that measures impairment instead of the current test that only shows marijuana use at some point in the last 30 days.

                2. Your mindless speculation is meaningless.

    2. Well, it does sort of refute the predictions that legalization would increase traffic deaths. I agree it isn’t relevant to any argument for legalization, but it does take some away from arguments against it. Which is good as the “OMG everyone is going to drive stoned” is a pretty common one since most people have stopped believing the other bullshit they used to use to scare people about pot.

      1. That is true. There has not been an increase in deaths. The more relevant statistic though would be accidents specifically related to marijuana use. Since that information isn’t here, these charts don’t support anything in my view, and it takes away from the argument when they are used to support the legalization agenda or even to attempt to refute the anti legalization agenda.

        1. I agree it is not a big strong argument, but still worth mentioning.
          Hell, I think that most of the arguments commonly put forward for legalization are weak and silly. The only argument that should really matter is that it is no one else’s business and it is completely immoral to criminally punish someone for doing something that harms no one else.

          1. That’s my biggest problem with arguments like this one. That they detract from real arguments.

            It’s very similar to the pro life movement. They constantly get screwed because they don’t just say “it’s a person, we have laws against killing people.” Instead they say the facilities aren’t clean enough or whatever.

            Never make a weak argument when you have a strong one because you are giving your opponent something to argue against.

  5. if it saves even one child…

  6. The madness! The charts all spike during the post-summer weed-binging month known as ‘total’?

    1. Yeah, that graph is extremely misleading.

      1. Poorly laid out, anyway.

        1. Which is to say the same thing.

          The entire purpose of making charts out of numbers is to make a point *clearer* and highlight something about their directional trend/relative proportions/rates of change/correlations with other info etc.

          If instead it fails to convey information as clearly as it might *in a table* (which in this case, where numbers are compared in part then in aggrigate)… then the table is actually better.

          But then we are talking about government work. We should be glad they didn’t make things any worse. (like change the definitions of the data mid-stream as they often do)

  7. Facts don’t matter to people with an agenda. Just the other day in my court-mandated DUI class, one of the instructors told our group that since legalization marijuana-related traffic fatalities in Colorado have DOUBLED. I knew it was bullshit (and so did a few others) but we had no way to prove otherwise. And I’m sure most people there believed it.

    1. When you need more people to attend your crony funded class, then everything always leads to more fatalities.

      1. These people seem to think that somehow more laws & mandated classes lead to less fatalities. Never once have any of them considered that perhaps safer cars = less fatalities, and their laws only serve to generate revenue.

    2. Maybe the number is 0. Double 0 and you get 0.

      Marijuana related traffic fatalities have gone up by 1,000,000,000%!!!

      1. I assumed it went from one to two, or some other insignificant number. Plus, now that’s it’s legal the cops probably characterize ALL accidents where anyone was stoned (even just a passenger) as marijuana-related–just as they’ve done for years with alcohol.

    3. I am aware of that, indeed, the number of marijuana fatalities in Colorado DID indeed double. From 1 to 2. 🙂

      Actually, since we shouldn’t see police cars in high-speed pursuit of the “devil pot smokers,” it might actually drop 🙂

  8. Also worth noting that driving drunk is fun, while driving high is not. Generally speaking at least.

    1. The last thing I want to be doing when I’m high is driving. I’d much rather be watching Aqua Teen Hunger Force.

      1. I used to think rape was funny. Not anymore.

        1. Even when it’s a lady rapist?

          1. I once knew a guy who, back in the ’70s, was essentially gang-raped by a small group of women, all of whom he knew. To naive and horny me, it sounded like a fantasy come true, but he described it as “not fun.” He didn’t seem traumatized by the experience, though.

            1. How did they get him aroused?

              1. A gentle breeze?

                1. Yeah, when you’re a guy that age, it doesn’t take much. And the situation didn’t involve a knife at his throat in an alley, it was some sort of party that got out of hand. So I can imagine him getting aroused and yet not feeling it was a fun experience out of Penthouse Letters.

                  1. I’m shuttering at the idea of a forced erection.

      2. The rare occasion that I might be high and driving I’m thinking, “I just need to get home.”

        Whereas drunk and driving, “I wonder how fast I can take that corner?”

        It’s patently obvious to me which is more likely to cause a fatal accident (or fist fight, or unwise hookup, or…)

        1. or wise hookup!

    2. I think that depends a lot on the driver.

      But it is probably true generally speaking.
      Perhaps more relevant to the safety question is that people tend to think they are driving better than they are when they are drunk and think that they are driving worse than they are when stoned. Which means that a lot of people drive more cautiously when stoned than when sober.

      I honestly have zero concern about stoned drivers on the road.

      1. That really slow guy disrupting traffic and causing accidents? Yeah don’t worry, he’s just stoned.

        Good luck with that. I’d rather have assholes not drive, no matter what they’re doing.

  9. I seem to recall someone saying that drug related accidents were up (not that I necessarily believe that finding drugs at the site of an accident proves that drugs caused the accident).

    I’d be curious to see some statistics on total accidents vs drug related accidents.

    1. Yes. That data would actually be meaningful. But instead, the writer got lazy and decided to provide charts that don’t prove or disprove anything about legalization.

      1. Actually, they prove that anti-legalization speculation that traffic fatalities would rise was bullshit.

        1. But that’s a bad argument in the first place. The argument that weed related fatalities would rise makes sense, but the total number isn’t important. Let’s say, for example, 2% of those fatalities in 2013 involved pot and 4% in 2014 did. Even if the total number went down, the marijuana related number double proportionally.

    2. I put a dime bag on my driver’s seat and left it overnight. The car didn’t even move! This proves that drugs cannot cause an accident. Only people can.

  10. DespiteBecause of Marijuana Legalization, Traffic Fatalities in Colorado Continue to Fall

    Come on Sullum, stop with the anti-pot slanting. Are you on the project SAM payroll?

  11. That’s less than a 2% decrease? I don’t think that a statistically significant decrease.

    1. Well, so that makes you think that it was a statistically significant increase?

      Why do some people not grasp that the only point in presenting these statistics is to demonstrate that the hysterical predictions from the sycophants of prohibition were wrong, wrong, wrong. These numbers are strong supporting evidence of that assertion.

      It’s amazing how many people want to let the enemies of freedom weasel out of being exposed as bald faced lying idiots.

Please to post comments

Comments are closed.