Does Legalizing Marijuana Cause 'Sharp Increases in Murders and Aggravated Assaults'?

The link that Alex Berenson perceives between cannabis and violence is not apparent in careful research on the issue.


The New York Times

Alex Berenson, the former New York Times reporter who has just published an anti-pot polemic that he aptly named after a notoriously hysterical 1936 anti-pot movie, says marijuana legalization "appears to lead to an increase in violent crime." Like his claim that marijuana causes schizophrenia and other "serious mental illnesses," his claim that it causes violence is based on a highly selective reading of the evidence.

"The first four states to legalize—Alaska, Colorado, Oregon and Washington—have seen sharp increases in murders and aggravated assaults since 2014, according to reports from the Federal Bureau of Investigation," Berenson writes in the Times. "Police reports and news articles show a clear link to cannabis in many cases."

As Jesse Singal notes, selecting 2014 as the starting year seems suspect, since two of those states (Colorado and Washington) approved legalization in 2012. But 2014 does coincide with the lowest national violent crime rate since the late 1960s. The national rate rose by 3.5 percent in 2015 and by 3.4 percent 2016, then fell by about 1 percent in 2017, for a total increase of about 6 percent between 2014 and 2017. It's true that the increase in violent crime was sharper in the four states that Berenson mentions. Can the difference be attributed to marijuana legalization?

Probably not. University of Oregon economist Benjamin Hansen finds that "the homicide rates in Colorado and Washington were actually below what the data predicted they would have been given the trends in homicides from 2000-2012." He says "we can't conclude that marijuana legalization increases violence, and perhaps even there could be small negative effects."

Nor is the effect that Berenson perceives apparent in national data. The share of Americans reporting past-month marijuana use in the National Survey on Drug Use and Health rose by 55 percent from 2002 to 2017, a period when the national violent crime rate fell by 23 percent.

How plausible is it that legalizing marijuana would immediately cause "sharp increases in murders and aggravated assaults"? Here is how a bunch of experts at the RAND Drug Policy Research Center summarized the evidence in a 2013 report commissioned by the Office of National Drug Control Policy: "Even though marijuana is commonly used by individuals arrested for crimes, there is little support for a contemporaneous, causal relationship between its use and either violent or property crime. There is evidence supporting a possible intertemporal relationship, but it is not clear to what extent this is unique to marijuana." The authors flatly state that "marijuana use does not induce violent crime," while "the links between marijuana use and property crime are thin."

Free Press

In line with that research, several studies have found that relaxing legal restrictions on marijuana is not associated with an increase in violent crime. A 2016 analysis of data from 11 Western states, published in the Journal of Drug Issues, found "no evidence of negative spillover effects from medical marijuana laws (MMLs) on violent or property crime." To the contrary, the researchers found "significant drops in rates of violent crime associated with state MMLs."

A 2017 study published in Contemporary Drug Problems compared FBI crime data in states with different legal regimes and found that "property and violent crime rates appear to be lower in both decriminalized and medically legalized states, but the difference is not statistically significant." A 2018 study published by Germany's Institute of Labor Economics compared California counties with different policies regarding medical marijuana dispensaries and found "no relationship between county laws that legally permit dispensaries and reported violent crime." Another 2018 study, published in the Journal of Economic Behavior & Organization, found "no causal effects of medical marijuana laws on violent or property crime at the national level" and "no strong effects within individual states, except for in California, where the medical marijuana law reduced both violent and property crime by 20%."

If letting people use marijuana for recreational purposes leads to "sharp increases in murders and aggravated assaults," you would expect to see something similar in jurisdictions that allow medical use, especially when the rules are loose, as they were in California for two decades before full legalization. Yet these studies find nothing of the sort. And if more marijuana use means more "paranoia and psychosis," resulting in "an increase in violent crime," as Berenson claims, you would expect that the national increase in cannabis consumption would have been accompanied by a national increase in violent crime. Yet exactly the opposite happened.

Berenson's book has received respectful reviews in The New Yorker and Mother Jones, along with considerable pushback from people who study these issues. You can expect to see more of the latter.

NEXT: Kamala Harris' New Book Tries to Massage Her Record as a Prosecutor, But the Facts Aren't Pretty

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. Even if you could find a correlation, that would be meaningless without an explanation of the causality. In fact, looking for the correlation at all seems to be a waste of time. Assaults and murders are generally well documented in the court system. If you think pot might be causing people to be violent, the way to confirm that is to go study cases of violence and the people known to have committed it. If it is the case that significant numbers of people convicted of assault or murder say that committed their crime while high on pot and that in their view being high contributed to their doing it such that had they not been high they wouldn’t have done it, I will at least be willing to listen to the argument that legalized pot causes violence. Yes, the post facto claims of criminals would hardly settle the issue but it at least would create a good reason to think it might be true and justify looking at the issue closer. But if all you have is some raw causlaity numbers, I really don’t give a shit what you have to say.

    1. Not really, because they may have committed the crimes not while high, but after coming down, or after a period of abstinence when they were craving more. There are some notorious cases of violence while high, but overall I’d hypothesize that pot availability and decriminalization reduces overall violence and crime.

      1. That is a different issue. I would argue someone committing violence because they were too lazy or poor to obtain whatever pot fix they wanted is not quite the same thing as pot leading to violence. Indeed, as you point out, that argues for legaizing it not against it.

  2. All the potheads I know are too lazy and unfocused to murder anyone.

    1. You and everyone else. To say that legalizing pot increases violence, you have to be able to show that people committed acts of violence while using it that they would not have commtted had they not used pot. I have never heard of a single example of such a case personally and have never once seen a scientific study that examined the existence of and frequency of such cases.

      Everyone knows alcohol increases violence. Doctors can show how the biology of alcohol consumption lowers people’s inhibitions. And anyone over the age of 20 who isn’t Amish has seen examples of drinking making otherwise peaceful peole violent. I have yet to see anything like that with pot. There is no biological explanation for why it would make people violent and everyone I have ever met who has used it or been around it claims that it doesn’t. Yet, somehow we are supposed to believe that it does because of some raw causality numbers.

        1. That is an interesting case. It might be an example. It is absolutely the kind of case that should be investigated. This guy says he went berserk because of the pot, but we don’t know if that is true. Maybe he is lying, maybe he had a reaction to something else and just thinks it was the pot. Or maybe it was the pot and he is some kind of special case that reacted violently to a large dose of the stuff. All of those questions should be answered. And investigating cases like the one you link is how this issue should be examined instead of wasting time looking for raw correlations.

          1. Or maybe it was the Twinky and he is some kind of special case that reacted violently to a large dose of the sugary stuff.

            It could happen.

            What about the many studies showing ” a strong association between schizophrenia and tobacco smoking, whereby people with schizophrenia are much more likely to smoke than those without the disease.[1] For example, in the United States, 80% or more of people with schizophrenia smoke, compared to 20% of the general population in 2006.” -wiki

            So, the self-medicating weed users with a prior propensity for violence or mental health issues are supposed to be more proof of the inextricable connections and causation of the evil weed. Just tell me it’s a scientific consensus so I can get on board.

          2. Or, more likely, it’s a completely bullshit excuse.

  3. All I can say to Alex Berenson is, amen sir. But?.

    ?did you know that alcohol also leads to violence?

    According to the US Dept of Justice. ” nearly 4 in 10 violent victimizations involve use of alcohol,”

    This is much worse than people smoking the marijuana, much worse.

    I’d say the problem with prohibition is that the government repealed this law.

    Think of all the lives that would have been saved if prohibition wasn’t repealed.

    I’m sure Alex Berenson would agree that it’s time to bring back prohibition.

    Beer is a gateway drug to whiskey and rum, some of the most addictive and destructive drugs of all time.

    1. That is the thing, the public health case against alcohol is at least as stong or maybe stonger than it is pot. Whatever the effects of pot, you are very unlikely to kill yourself from using it and the correlation between alcohol and violence is much better establsihed than between pot and violence. Yet, rightfuly, no one would argue for prohibiiting alcohol again but somehow ending pot prohibition is the end of civilization.

  4. I agree the danger of marijuana is greatly exaggerated. However at the same time we shouldn’t downplay the risks either. The fact is, it’s a powerful and dangerous drug, even if you can’t o.d. on it. As a first time user, you should take only a small amount (half a dose) and then wait an hour before taking more (which you probably shouldn’t). If you take too much or if you are not in a safe space, then you are far more likely to have an adverse and frightening experience. As I have learned the hard way. I don’t like it and don’t use it, though I have used it as a tool in the past for cognitive exploration. It is not particularly pleasurable, and I am honestly baffled by the people who enjoy it.

    1. Having said that – I have definitely had some mind bending trips. But again, this can only happen if you plan it carefully in a relaxed and safe environment. As more people become familiar with pot, acceptance of it will grow. So I’m not too concerned by this fearmongering. I am however a bit perplexed why Tucker Carlson is pushing it. He’s a modern day prudish teetotler. Why would the ‘deplorables’ be attracted to that?

    2. “…if you are not in a safe space…”
      FFS As the official self-appointed Reason Comment Moderator I’m suspending your posting privileges. Failure to abide by my official determination will be registered as a failure to respect my athoritah and may lead to an unenforceable permanent ban. You’ve been warned.

      “…though I have used it as a tool in the past for cognitive exploration.” To boldly go where no man has gone before? Timothy Leary just called and said you were a puss.

  5. I don’t know if MJ increases violence or not, but speaking for myself, I usually get fucked up as polio when I use it.

    1. Speaking for myself, I usually Burn it till it’s burned out.
      Turn it till it’s turned out. Lose control, of body and soul. let’s get retarded in here.

  6. Right — let’s criminalize pot (or booze) so that, *by definition*, the crime rate rises. Then decriminalize it, and claim the crime rate also rises.

    This doesn’t pass the smell test any more than drug dogs.

  7. University of Oregon economist Benjamin Hansen finds that “the homicide rates in Colorado and Washington were actually below what the data predicted they would have been given the trends in homicides from 2000-2012.”

    Economics: The science of explaining why your predictions were wrong.

  8. Does Legalizing Marijuana Cause ‘Sharp Increases in Murders and Aggravated Assaults’? Yes, Yes,Yes
    Of couse, We knew it in the 930’s:
    Refeer Madness, it could get you too!
    We knew it in the 1970’s:
    We knew in in the late 80’s and 90’s:
    And we know it now! LOL

  9. Have they given up on the “gateway drug” argument? That was all the rage when I was a teen.

  10. Epidemiology is a very nonspecific tool. You can find all kinds of correlations that are meaningless. The winner of the Super Bowl, old NFL vs AFL teams predicts whether the stock market will go up or down. That is or was true in a statistical sense but it means nothing.

  11. We know bath salts cause cannibalism, so perhaps marijuana makes you mistake other people for Cheetos?

  12. I essentially started three weeks past and that i makes $385 benefit $135 to $a hundred and fifty consistently simply by working at the internet from domestic. I made ina long term! “a great deal obliged to you for giving American explicit this remarkable opportunity to earn more money from domestic. This in addition coins has adjusted my lifestyles in such quite a few manners by which, supply you!”. go to this website online domestic media tech tab for extra element thank you……

  13. Six hours later, the cops showed up to make sure Sarah wasn’t being abused or worked to death.

    “If you’d hired my daughter, she could’ve ridden her bike there in 15-20 min. and let you know whether it was an emergency situation that might require you to intervene within the next six hours.”

    I’m tickled by the fact that, apparently, the police can’t imagine someone being worked to death, especially a kid, in less than 6 hours

  14. Those who believe in limited government, personal responsibility, free markets, and individual liberty should embrace the ending of this irrational, un-American cannabis prohibition. It should be the cornerstone of current GOP policy.

    Federal studies show about half of the U.S. population has tried cannabis, at least 15% use it regularly, over 80% of high school seniors have reported cannabis “easy to get” for decades. This prohibition, like alcohol prohibition has had little of its intended effect. In many cases cannabis prohibition makes cannabis usage problematic where it would not have been otherwise, be it light, moderate, or heavy usage. For the most part, cannabis prohibition only successfully prohibits effective regulation.

    A few issues created by prohibition: there are no quality controls to reduce contaminants (harmful pesticides, molds, fungus, other drugs), there is no practical way to prevent regular underage sales, billions in tax revenue are lost which can be used for all substance abuse treatment, underground markets for all drugs are empowered as a far more popular substance is placed within them expanding their reach and increasing their profits, criminal records make pursuing many decent careers difficult, police and court resources are unnecessarily tied up by pursuing and prosecuting victimless ‘crimes’, public mistrust and disrespect for our legal system, police, and government is increased, which is devastating our country.

    1. Prohibition is also very expensive, though, a cash cow for a number of powerful groups such as those related to law enforcement and the prison industry. These organizations have powerful lobbies and influence that perpetuate a failed drug policy through ignorance, fear, disinformation and misinformation. This ensures an endless supply of lucrative contracts, grants and subsidies from the government and its taxpayers to support their salaries, tools of the trade, ‘correctional services’, and other expenses. Cash, property and other assets from civil forfeiture laws also significantly fatten their coffers while often violating civil rights.

      America was built on the principles of freedom and liberty. In some cases there are extreme circumstances that warrant intervention with criminal law. In the case of mind-altering drugs we have already set this precedent with alcohol. Cannabis is less harmful than alcohol to the consumer and especially to others. If we are to have justice, then the penalties for using, possessing and selling cannabis should be no worse than those of alcohol.

  15. While, like alcohol and tobacco, there are associations with cannabis use and psychosis, causation has not been established. However, even IF we did assume that cannabis is an independent cause of psychosis, then it so rarely does that you would have to stop thousands from using it to prevent just one case:

    “IF we assume that cannabis use plays a causal role in psychosis, it will be difficult to reduce psychosis incidence by preventing cannabis uptake in the whole population: an estimated 4,700 young men in the United Kingdom aged 20?24 years would have to be dissuaded from smoking cannabis to prevent one case of schizophrenia” [Hall 2014]

    If cannabis was a significant cause of psychosis, then varying rates of cannabis usage over time in the U.S. and other countries should have a corresponding change in rates of psychosis in those countries, but they have not despite decades of increased use by millions:

    “The most parsimonious explanation of the results reported here are that the schizophrenia/psychoses data presented here are valid and the causal models linking cannabis with schizophrenia/psychoses are not supported by this study.” [Frisher et al. 2009]

    1. “There was a steep rise in the prevalence of cannabis use in Australia over the past 30 years and a corresponding decrease in the age of initiation of cannabis use. There was no evidence of a significant increase in the incidence of schizophrenia over the past 30 years”
      “Cannabis use does not appear to be causally related to the incidence of schizophrenia…” [Degenhardt et al. 2003]

      “The current data do not support low to moderate lifetime cannabis use to be a major contributor to psychosis or poor social and role functioning in high-risk youth” [Auther et al. 2012]

      One of the most important confounders is the fact that cannabis helps some people with schizophrenia, especially varieties high in cannabidiol (CBD), a cannabinoid which can treat and prevent psychosis:

      “Additionally, CBD prevented human experimental psychosis and was effective in open case reports and clinical trials in patients with schizophrenia with a remarkable safety profile…These results support the idea that CBD may be a future therapeutic option in psychosis, in general and in schizophrenia, in particular.” [Zuardi et al. 2012]

      Correlation does not imply causality. There are many factors involved, including the fact that many with undiagnosed psychosis are self-medicating with cannabis. It is especially difficult to tease out this confounder when trying to establish causation.

  16. The following from the CBI web-site puzzles me:

    “(September 27, 2018?CBI?Lakewood, CO)?There’s a new name and look for the state’s annual crime statistics published by the Colorado Bureau of Investigation (CBI). The Colorado Crime Stats system (formerly the Crime in Colorado report) includes crime statistics reported by 245 law enforcement agencies across Colorado.

    The new format transitions the state of Colorado to the National Incident-Based Reporting System (NIBRS), which offers extensive information about each crime. This differs from the Summary Reporting System (SRS) that was part of previous Crime in Colorado reports. In the former Summary Reporting System, each criminal incident was summarized to the single most serious offense, whereas NIBRS counts each crime committed in a given incident. As a result, NIBRS statistics generally provide counts greater than the Summary Reporting System. It’s important to know that this is not necessarily an indication of more crime, only greater detail regarding the crimes committed.

    Because NIBRS data is collected differently from the older reporting method used for Crime in Colorado, the CBI will be removing Crime in Colorado links and allowing users to use the Colorado Crime Stats system to compare NIBRS data collected for each year dating back to 2013.”

    It’s an interesting coincidence that the Colorado change happens to coincide largely with the change in marijuana laws.

Please to post comments

Comments are closed.