Let's Spark Up a Fatty and Discuss the Early Results of Marijuana Legalization
Remember to stay hydrated!
Our national dialogue about marijuana has not been very dank. For years, the focus was on the obvious proxy issue of medical marijuana. When Last Week Tonight did a story on marijuana in 2017, we included a gut-wrenching clip of a Vietnam veteran arguing that marijuana helped him overcome his nightmares about the war. The other primary writer on the piece and I argued that: 1) Anecdotal evidence isn’t how you establish the efficacy of a medicine, and 2) Grisly war flashbacks, in which the soldier tearfully lists the names of his fallen comrades, might be a tad heavy for a comedy piece about reefer. In the end, we couldn’t get the clip removed, but we managed to get the list of fallen soldiers edited down, and we also removed the guy’s grisly description of a dead VC soldier. Because, again, this was supposed to be a comedy show.
I think I can safely say this now: One of the main arguments for legalizing weed is that weed can be fun. It makes bad movies good, makes good music great, and -- pro tip -- makes Mario Kart absolutely fucking hilarious. In a perfect world, a responsible adult would be free to relax with a joint at the end of a long week the same way my Dad used to relax with a few cans of Natural Light (no, Dad wasn’t a college student -- he just had the beer-buying habits of one).
But marijuana is also a controlled substance, and we shouldn’t lose sight of that. I feel like decades of Reefer Madness-type fear-mongering have led to an over-correction in which people talk about marijuana as if it’s benign as Wild Cherry Luden’s Throat Drops, which I can’t believe are marketed as “medicine”. So, while it’s true that no overdose deaths from marijuana have ever been reported, and the fact that the government puts cannabis in the same class of drugs as heroin and LSD makes about as much sense as creating a definition of “weapons” that includes both Predator drones and rubber band guns, we shouldn’t assume that marijuana is totally benign. Strangely, one of the benefits of legalization would be that we could learn more about marijuana’s possible negative effects (more on that in a bit).
We’ve had a few years of marijuana being completely legal in several states (18 states and DC have fully legalized, it’s illegal in only five states, and 27 states do some in-between “you have to pretend to have anxiety” thing). It’s worth asking: In places where weed is legal, how’s that going? What’s worked, and what hasn’t? What changes might states with legal weed want to make? I picked three issues that seem worth addressing: 1) States are creating problems by being tight-fisted with sales licenses; 2) It looks like legalization might be causing a slight increase in traffic accidents; and 3) We don’t know enough about marijuana’s effects, especially in relation to psychosis.
States are probably too stingy with sales licenses
Here’s an incredible stat: More people were arrested for cannabis in 2019 than for all violent crimes put together. That year saw 545,602 people arrested for cannabis, which is about the population of Sacramento, and I like to imagine that every man, woman, and baby in Sacramento was busted for weed and nobody else was.
It’s hard to think of a dumber reason for sending someone to jail.1 I don’t feel at all protected by these arrests, because the worst crime someone is likely to commit while stoned is to write a Rotten Tomatoes review about how Aqua Teen Hunger Force is an allegory for capitalism. It’s also hard to think of a worse use of law enforcement resources than busting people for marijuana possession; a police officer’s time should be split 80/20 between stopping serious crime and getting shredded for cop hunk-of-the-month calendars.
Of course, there’s also a racial element to these arrests: Black people are about 3.64 times more likely to be arrested for marijuana possession than white people, even though they report similar rates of use. It’s not hard to deduce some major drivers behind that stat: Black people are more likely to live in heavily-policed areas, and they’re also more likely to be stopped for minor offenses (even in the absence of formal stop-and-frisk policies, there’s evidence that this happens). So, what you have is an unequal application of the law that correlates with -- and is almost certainly influenced by -- race. That’s real bad.
As you’d expect, arrest rates are lower in states that have legalized weed. Which should be one of the major benefits of legalization: Fewer pointless arrests. But some states have been so stingy in handing out licenses to sell legal weed that there’s still a large illegal market.
By “some states”, I mostly mean “California”. California has managed to screw everything up through an obsession with local control and an ignorance of markets. I’m starting to think that “an obsession with local control and ignorance of markets” should probably be California’s state motto.
California’s first mistake was to give cities and counties a a lot of control over what level of legalization is permitted. As a result, it was reported that as of 2019, less than a quarter of California’s cities allow legal sales. Maybe I would have more patience for local control if I didn’t know that California has engineered a housing crisis and botched a gigantic rail project in part by giving every locality a say at every point in every project. California: You can’t give everyone a veto over everything and expect anything to get done. Much like how at some point, the election for governor really needs to be over. Things have gotten so bad that New Yorkers like me feel comfortable lecturing you about efficient government and sound gubernatorial politics.
The second thing California seems to have done wrong is to go a bit nuts with taxes and regulations. Dale Gieringer, Director of California NORML -- a group that works for marijuana law reform -- says that state and local taxes can raise costs by up to 25 percent. When you combine that with the fact that California already had a robust illegal weed network -- they all but remind you to add the Weedmaps app to your phone when you land at LAX -- many people have decided to stick with the illegal system. So, California’s “legal” system retains traits of the illegal system, namely arrests, support for cartels, and having to invite your weed guy in to play Xbox even though you’d never hang out with him if he didn’t have weed.
For a legal system to work, it needs to be workable. California (and other states) are still working out the kinks.
Legal weed might be causing slightly more traffic accidents
Incredibly, people have felt the need to commission studies to determine whether marijuana negatively affects your ability to drive. This despite the fact that getting high can make it damn near impossible to operate a remote control. I will also refer you to my afore-mentioned Mario Kart research to conclude: Yes, weed affects your ability to drive (and also dodge turtle shells) big-time, and you should absolutely never drive while stoned.
This is one of the real concerns I have about legalization; when a stoned person gets behind the wheel, their drug use ceases to be a personal choice. If legalization leads to more stoned driving, that’s bad. Unfortunately, the research on the relationship between marijuana legalization and traffic accidents is muddled.
The first problem is that trying to suss out marijuana DUIs by measuring incidents in which a driver tests positive for THC is misleading. This method will miss drivers who were under the influence but weren’t tested, and will count many drivers as “impaired” who actually weren’t, because THC levels affect people differently, and THC can be detectable in the body for up to a month. The second problem is that simply measuring incidents in which THC was a factor doesn’t tell us much about legalization, because -- I hope you’re sitting down -- sometimes, people get high even when it’s not legal!
I found two studies I consider halfway decent.2 The first found a six percent increase in accident reports in states that legalized weed, and the second found no significant difference between legal-weed states and non-legal weed states. Unfortunately, both are a few years old, so they don’t contain much data.
We have more data now, so I re-crunched the numbers. Using the US Fatality Analysis Reporting System, I looked at the eight states (I count DC as a state) that had at least two years of legal marijuana sales as of 2019, the last year for which there’s data. First, I made a five-year average for deaths per 100 million vehicle miles traveled for each state and set this as the baseline. Next, I charted the percent change relative to baseline for each year since marijuana sales became legal.3 Here’s what I found:
For what it’s worth: Oregon looks really bad, but if I had chosen a ten-year baseline instead of five, their numbers would look quite different (they had some unusually-good traffic years in the early 2010s). Massachusetts’ traffic deaths have actually gone down, and I have a theory there: Boston is -- hands down -- the worst driving city in the country. It’s a Hobbesian state of nature on Beantown streets; Boston driving is a depraved, primal exercise that plays to the advantage of rage-fueled nihilists with contempt for human life. Anything that gets those people to mellow the fuck out is all to the good.
I think these numbers are troubling, though not conclusive -- we need more data. But the yearly average, weighted for population, is an 8.73% increase over baseline; if annual traffic deaths were to increase by 8.73% nation-wide, that’s 3,377 people. Not good. The strong social norm against drunk driving -- which any Mad Men episode will tell you didn’t exist a few decades ago -- needs to extend to stoned driving. We also need better DUI tests for marijuana; right now, the best available test is for police to put on some Phish, and if you don’t say “God, this sucks” within 20 seconds, you’re high. If we’re going to be lenient on marijuana, we should also be tough on driving stoned.
We need to know more about marijuana’s effects, especially as they relate to psychosis
Marijuana is still illegal at the federal level. That makes it tough to conduct research; federal grants are out of the question, and most studies are heavily-regulated. In fact, there’s only one federal contract for marijuana for federal studies, and it’s with the University of Mississippi, which grows more than 4,000 pounds of weed a year. Which, from Ole Miss’ perspective, sort of makes sense: When your school is known for confederate iconography and excluding James Meredith, you jump at any re-brand available to you. And if that’s as the spliff-riffic source of the nation’s most primo bud, so be it.
Adults have the right to make choices about their own health. But informed choices require good information, and that’s lacking. Probably the most comprehensive look at the effects of marijuana/cannabinoids comes from a research review by the National Academies of Sciences, Engineering, and Medicine in 2017. It’s not outstanding that our best research is four years old, and it’s even more not outstanding that the research’s top finding is “we need more research”. From the report:
Recommendation 1: Address Research Gaps
To develop a comprehensive evidence base on the short- and long-term health effects of cannabis use (both beneficial and harmful effects), public agencies, philanthropic and professional organizations, private companies, and clinical and public health research groups should provide funding and support for a national cannabis research agenda that addresses key gaps in the evidence base.
That’s recommendation ONE. And maybe I shouldn’t be shocked that a bunch of academics’ top recommendation is “fund us, goddamn it!”. But it still seemed notable.
The report (which is summarized by German Lopez here) finds different amounts of evidence for different effects, both good and bad (e.g. “substantial evidence” that marijuana lessens chronic pain, “moderate evidence” that acute use impairs learning, memory, and attention). Some of the report’s most impactful findings might be those related to psychosis, which they summarize this way:
The evidence reviewed by the committee suggests that cannabis use is likely to increase the risk of developing schizophrenia, other psychoses, and social anxiety disorders, and to a lesser extent depression.
This is not “case closed”. The report notes that the relationship between marijuana and psychoses “may be multidirectional and complex”. If there’s a connection, it might not be “marijuana causes psychosis”; it could be “psychosis causes a person to seek marijuana”, or “factor X leads to both psychosis and marijuana use”. Also, how much use might lead to how much increased risk matters a lot.
Still, people are wondering about the relationship, especially in light of a high-profile book by Alex Berenson that makes the case for a link. I haven’t read the book, but many people say it misuses data, and given Berenson’s record on other issues, I’m not inclined to give him the benefit of the doubt. The NASEM report does essentially the same work as Berenson, except they’re a much larger team doing a much more comprehensive literature review, and they reach a more nuanced conclusion.
But we need to know this. Even if the relationship is only that people who are already at risk of psychosis increase their risk through heavy marijuana use, that could influence people’s decisions. Some people with alcoholism in their family choose not to drink; people at risk for schizophrenia or other psychosis might make a similar choice. There’s some evidence of a relationship, but we need to know for sure if there is and what it might be; we need to keep digging.
Two of the three issues above boil down to “we need more data”. The Senate bill to legalize marijuana at the federal level would help; it would remove some restrictions on research and make us less dependent on the University of Spliff-assippi’s dank stash of chronic. The conversation on this topic is getting more honest; people have stopped pretending that they’re super-interested in glaucoma treatments and making homemade rope. Hopefully, more honesty and more data will help us craft smarter, better -- some would say “danker” -- public policy.
***For Friday’s column***
I’m skipping the poll today, because the idea about “what building a news media thing looks like when you’re doing it” has gotten second place several times, so I’m going to write that up (unless something timely happens between now and Friday).
Also, a few people have asked if they can recommend topics. I haven’t yet figured out how to make that part of the poll, but if you’re subscribed to the e-mail list and hit “reply”, that goes to me. Suggestions are welcome; I want to know what people want to read about, and I also want people to give me a bunch of good ideas for free.
So thanks!
Obviously, far from everyone who is arrested for marijuana possession does time; many won’t even spend the night in jail. But even in a best case scenario, an arrest is a gigantic pain in the ass.
The first study is from the Highway Loss Data Institute in 2018, which compared collision claim frequency in three legal marijuana states (Washington, Oregon, and Colorado) compared to control states. This study is okay, but I’m not nuts about the “compared to control” aspect (I think a better method is to compare the states to themselves before the law change, since there’s enough fluctuation in these numbers that the fluctuations of the control group can add a bit of noise). Also, it seems that changes in how and why people file claims might cause those numbers to wobble. This study found a six percent increase in collision claim frequency.
The second study is from the NIH in 2017, which also compared legal-weed states to non-legal-weed states, but used fatalities instead of accident reports, which seems like a more objective measure. A good thing this study does is that it controls for state-specific population, economic, and traffic characteristics, which I wasn’t able to do. This study found no significant difference between the weed states and the non-weed states.
For states in which marijuana sales became legal mid-year, I used the following year as the “first year”. Which I think makes sense; it takes a little while for retailers to get up and running. Also, if there are a few legal weed months mixed into the baseline, they should be mostly washed out, because there are 60 months in that measure.
Also, sales are not technically legal in DC, but they do this weird thing where you’re allowed to give weed as a gift. It’s very stupid. At any rate, for the purposes of this measure, I decided that counts as “fully legal”.
My wife is a psychiatrist, and marijuana is a constant frustration because her patients think it doesn't have negative effects. Like me, she could give a fuck if people smoke weed for fun, and she wants it to be fully legalized. She also believes it's a beneficial treatment for chronic pain. But as a psych medication, she's not impressed. It makes some psych symptoms worse, and interferes with some medications, in addition to increasing the risk of psychosis as you describe here.
I hope that after legalization, we can talk about the negative effects without people freaking out -- like we can for alcohol. Plus, people who just want to smoke weed can leave doctors out of it.
For a while, my wife worked in a state where the only psych condition that qualified for medical marijuana was PTSD. So, she had all of these patients who were determined to get a PTSD diagnosis. Arguing with college students who clearly just want to smoke weed is.... not what she pictured when she went to medical school.
538 but with better writing!