Study: Higher Murder Rates in States With Restrictive Carry Laws

Smith & Wesson Range Jacket courtesy

A study was released recently in the Applied Economics Letters journal that comes to some interesting conclusions. It states that “… states with restrictions on the carrying of concealed weapons had higher gun-related murder rates than other states. It was also found that assault weapons bans did not significantly affect murder rates at the state level. These results suggest that restrictive concealed weapons laws may cause an increase in gun-related murders at the state level.” That’s from the abstract, and on the surface it sounds pretty good. But there’s an issue.

This is the same problem that we run into with gun control advocates: they mistake correlation for causation. Point in case is the old chestnut about being more likely to kill yourself with a gun if there’s a gun in the home. It’s like saying you’re more likely to be killed by an alligator if there’s an alligator in the home. It doesn’t say anything about whether the gun makes you more likely to die, just that one specific cause of death is more common. Gun control advocates mistake that for proof that guns cause death, and we’re in danger of doing the same thing here.

The study in question is being paraded around as proof that restrictive gun control measures increase the crime rate, but that’s not really a conclusion we can draw. The measures may have been reactive instead of causative, or maybe there were much more important factors that were not examined. We probably will never know. So while it’s a solid report, the conclusions being drawn from the results aren’t necessarily valid.

What we can do, however, is disprove the opposition’s position. Gun control advocates claim that restrictive gun laws prevent crime and make us safer, but the study pretty conclusively shows that restrictive gun control measures do not reduce crime. In fact, they might increase the crime rate instead. We can’t necessarily accept the hypothesis presented by the report, but we can definitely reject the hypothesis presented by the opposition.


  1. avatar GS650G says:

    Spell check the headline

    1. avatar Accur81 says:

      And it’s fixed. RF is on it.

  2. avatar MarcusAurelius says:

    It still bothers me to see a focus on “gun related” violence but maybe reading the damned thing will shut me up. Still good to have more ammo for the “honest,” national discussion on this topic.

  3. avatar Dirk Diggler says:

    Did you send a complimentary copy of the study to Shannon Watts aka Bloomy’s street walker?

    1. avatar Vhyrus says:

      I have a feeling that if she were, in fact, ‘for sale’ (or rent), your ammo budget would be significantly impacted.

  4. avatar BillC says:

    Seems like every progressive forgets this premise (correlation/causation) from Stats101. Thanks, TTAG, for keeping with the truth (and better writing. I can’t count how many times I catch very poorly written articles in the gun rags and blogs, in terms of grammar and spelling).

  5. avatar ajc111 says:

    Good catch – while we gotta give them a mulligan on the spelling the other side won’t.

  6. avatar anonymous says:

    See “Brady Grades vs. Crime Rates” for a comparison of violent crime rates, homicide rates, and Brady Grades per state. If anyone can see a correlation in the scatter-shot graphs, let me know.

    via , which compares murder rates in the U.S. to the gun-free zones known as prisons.

  7. avatar PeterK says:

    Aw yeah, nice to see the intelligentsia gettin’ it right. 🙂

    It vexes me that so few understand that causation and correlation are not the same.

    1. avatar KCK says:

      But sometimes related.

      1. avatar BillC says:

        At least warrants further studies. Don’t even get me started about Z scores though. It’s amazing that so few understand stats, especially in how it’s used to lie to us about everything.

  8. avatar KCK says:

    Well take Illinois (Chicago) for example.
    For the most part, thugs kill thugs and they are both often carrying concealed, just not legally.
    In this case restrictive laws for carry are irrelavent.

    All states have very strict law against murder, but illinois (Chicago) has a high murder rate, obviously not causation.
    The problem with Illinois is that they pass laws to attempt “control” murder by guns which also have no effect.
    I think that the restrictive carry environement is because politicions “need to do something” and create neither causation nor solution.

    1. avatar GS650G says:

      Guns don’t vote but thugs do. That includes relatives and friends of thugs. Coming down hard on thugs means fewer votes for law and order candidates and more votes for community organizers who,,,,,,,,,
      blame guns not people. Look at who is talking about guns from other places, about “how none of this would have happened if _______ didn’t get a gun”, and the all time greatest line about getting guns off the street.
      This all is a problem without a real solution that works. Only nature taking it’s course will solve it eventually unfortunately with many innocent victims. Sad that it comes down to it.

      1. avatar KCK says:

        Did you kinda mean, take the thug off the street and there is no one to fire the gun?

        We could use a quasi-Islamic approach and remove the finger they used to pull the trigger.
        The shooting may stop before they have only their pinkie.
        (not really advocating that, only for comedic effect. No one wants a Muslim sht storm.)

        1. avatar GS650G says:

          We call them thugs, others call them earners. Plenty of people depend on their activity, from family to businesses where they spend the proceeds. The dirty side of street business is the result. Disarming them and legalizing drugs simultaneously will not convert them into law abiding tax payers any time soon. And I am not thrilled with paying for their arrest, trial, and incarceration. While having them walk among us is not preferred, this also presents a problem since there are so many complaints about the quantity and makeup of the prison population on one side and a real need to protect the rest of us on the other.
          In the middle are ordinary people who want to be left alone to make it in this country.

    2. avatar KCK says:

      I also meant to say that there is probably more causation for restrictive laws to be passed where there is a high murder rate where reversing the two, correlation holds but not cause.

  9. avatar Greg in Allston says:

    Nick, just curious, is this gleanings from the abstract or have you read the whole study? I’m not so curious that I’d drop the $39.00 to buy a copy of it myself and I’ve yet to see someone who has gotten the paper/study to parse out the good professor’s analysis.

    1. avatar ChuckN says:

      I agree. We need to see the full paper. For all we know they
      might have factored in crime rates before and after gun-
      control legislation and many other factors. Regardless,
      the study should make for an easy platform as a baseline
      for other correlating evidence.

  10. avatar Ralph says:

    the study pretty conclusively shows that restrictive gun control measures do not reduce crime.

    No, it doesn’t do that either. For all we know, crime rates would be ten times worse without the restrictions. And in less-restrictive states, crime rates might also be worse now than they would if there were restrictions.

    Not that I believe any of that, but there it is.

    1. avatar Accur81 says:

      The observation, in and of itself, is true. States with restrictive carry laws have higher murder rates. Therefore, states which allow concealed carry have lower murder rates. The stated goal of the antis is to pass laws against gun ownership because they will reduce murder rates. However, the presence of lower murder rates in states without restrictive gun laws is objective evidence which refutes their claims.

      When WI went to “shall issue,” one of the objections to the measure was the standard “blood in the streets” claim – shootings over parking spaces, toy shopping, petty disputes and such. It hasn’t happened. Once again, the objective support that gun control reduces crime is left wanting by the observation that increasing gun rights does not lead to an increase in crime. In fact, it usually leads to a decrease.

      Obviously the 2A supersedes social utility, but it is there nonetheless, and should be thrown loudly in the face of the antis at every opportunity.

      1. avatar Ralph says:

        States with restrictive carry laws have higher murder rates. Therefore, states which allow concealed carry have lower murder rates.

        That statement only establishes correlation, not causation.

        Establishing causation will require a lot more than statistics. We cannot prove that more guns causes less crime, only that there is a correlation between the two. Let the statement stand on its own — states with permissive gun laws suffer less murders.

        1. avatar Accur81 says:

          Works for me. I think we’re saying the same thing. I’m pretty sleep – deprived at this point.

        2. avatar Daniel Silverman says:

          I agree. I think it is a solid start. On the surface the conclusion stands as valid. So I say we run with it.
          Also the study, and I think this is where our beef is, does not deal with the social dynamic. Urban or rural. Rich, poor, religious affiliation?
          There are a lot of things not being looked at here, and should. But then again MAIG, MDA, and CSGV could care less about these things, they just want action.
          We tend to forget that this is a PR war. Logic is pretty much lost, so we will take any support where we can get it.

  11. Good report, thanks.

    1. avatar Myshkin says:

      *Off-topic: Paul, I just wanted to let you know that you taught me how to get the slide stop back in my 1911! I’ve also (mostly) avoided the idiot scratch, I just have the faint, luckily barely-visible beginnings of one from my first attempt before finding your video. I had to also have your video on standby for my second attempt but can now do that step pretty quickly!

  12. avatar Michael G Marriam says:

    I say go ahead and trumpet the illogical false conclusion and dare the anti-gunners to call us on it. Their whole position is built on rhetoric and factual misrepresentation. See how they like a dose of their own medicine.

  13. avatar Jus Bill says:

    I feel the report abstract is correct. That should be good enough for the libs…

  14. avatar BluesMike says:

    I have the report. The report is only four pages, so, it’s roughly $10 per page. I had a brief email conversation with a person who does a lot of study in this area and we agreed that the regressions that were run were simple. There is too much variation within a state to really draw conclusions from state to state comparisons. I highly encourage anybody who is interested in this area to get John Lott’s “More Guns, Less Crime.” You can get a Kindle version (on your iPad, iPhone, Kindle or any of many other devices) for a reasonable price. If we would all arm ourselves with the knowledge from that book alone, it would make us better at arguing the “needs” side of the argument, where the other side wants to stay. Note that the latest edition has John’s answers to many challenges by the other side as well as a compelling story of how hard he worked to make sure there was peer review from all corners of opinion. I don’t know what is inside another person but I believe after reading it, that the response he got from the control side really changed his whole world view. We should of course always be improving our ability to argue the inalienable rights side of the argument.

    1. avatar tdiinva says:

      I believe Lott used a technique called seemingly unrelated regression. This method uses both cross sectional and time series data. Each state forms a separate time series and the use of all states is the cross section. It is a method that captures unique characteristics of each unit over time.

      1. avatar BluesMike says:

        Another key to Lott’s studies is that he learned that states have enough variation that is it possible that any one state encompass the variation of most of them. Or, a city on a border of one state could impact the state next door. So, he concentrated more on county and city data, and where he was able to get information on the likelihood of getting a permit/license (may issue for example), he was able to really narrow down to more specific effects. Combining this with time series allowed him to get closer to being able to show cause versus after-because. After he expanded his studies with so many more variables, he was running into the limits of the software available at the time. Luckily, the software and hardware available today allows for regressions with many more variables.

  15. avatar ValleyForge77 says:

    Very sober analysis, Nick. I can dig it man. Can’t become like those we so distrust. Gotta keep it real.

    Now that I got that out 😉 …the other part of me wants to energetically agree with Michael above – and trumpet it up like a damn brass section, just to get the soundbite out there and beat them at their own damn game.

    Heck, we all know it’s highly likely that it’s true. Why do we have to hold ourselves to a higher standard of truth-telling when all they do is lie and distort and propagandize?

    (I know, I know…. I sure do feel like a pouty kid asking his parent a rhetorical question though 😉 )

  16. avatar RockOnHellChild says:

    Let’s say it was an incontrovertible fact, completely beyond reproach, that less restrictive state had way more crime and more restrictive states had virtually none, would anyone on here be willing to give up their gun rights?

    I, for one, would not… So, in my mind, it’s a moot point, regardless.

  17. avatar MAC][ says:

    Great article. Thanks for “keeping it real”.

  18. avatar WI Patriot says:

    Really, go figure…

  19. avatar GS650G says:

    A major contribution to lower crime rates is technology. Video cameras, databases, cell phones, DNA, and general knowledge about criminal activity makes it harder for people to get away with crimes. It also infringes to some degree on people’s rights. If all that fails, there is the gun as the last deterrent.

  20. avatar Javier says:

    Correlation, causation, in the end the conclusion is the same: better to live elsewhere.

  21. avatar peirsonb says:

    Huh, wonder how this one made it by the censors…..

  22. avatar Hannibal says:

    Good article (here). The issue is probably too complicated with confounding variables to draw conclusions from statistics alone.

    But when there’s someone breaking in your house, would you rather have a phone or a phone + a gun? I don’t tell other people they need to play the odds a certain way with their life, I’d appreciate if they’d let me run my own numbers.

    1. avatar uncommon_sense says:

      ^ This.

      Society is for all intents and purposes infinitely complex. And the variables that affect crime rates are almost infinitely complex. The only way anyone can claim that statistics prove something is if they are omniscient (all knowing — in other words God himself). Last time I checked, no one was omniscient.

      While statistics give us a clear sense of what is happening, they can only tell us what causes something to happen in an exceedingly simple environment and society is NOT exceedingly simple. Anyone who tells you otherwise is proud and arrogant.

      In the end every citizen has the Natural right to honestly procure and possess any personal property that he/she desires, period. That right is not subject to anyone’s approval, period. We need to stop arguing with the gun grabbers about “data”, “needs”, or “consensus”. Simply state your rights and stand behind them.

  23. avatar AJ says:

    Here’s what sets us “fanatics” apart from the grabbers. Here’s a study which clearly disproves a huge argument put forth as “fact” by the state committee for disarmament. But instead of running with the ball and drawing all sorts of wild conclusions and then publicizing them until our eyes bleed, we take an objective approach and actually ask what, exactly, has been proven.

    Not like the “90% of gun owners support gun control” trope that continues to be trotted out day after day after day by the opposition. A meaningless factoid borne out of a flawed “study” on its best day.

  24. avatar ShaunL. says:

    First: Duh.

    Second: Though correlation may not be the same as causation I’ll take what I can get in this battle. After all, how often do facts, stats, real numbers or the truth have anything whatsoever to do with the anti-gun agenda being foisted upon law abiding citizens????

    I say we use it. Write it in the sky so bright it burns into their retinas and forces them to have an actual debate on the subject.

    Sometimes there ARE three paths to choose from… The low road, the high road and the STFU road. I believe each has it’s merits from time to time. Why should we be honorable against an enemy that doesn’t know the meaning of the word.

  25. avatar Grumpy in Kali says:

    “You’re more likely to be killed by an alligator if there’s an alligator in the home.”

    Dang it!

  26. avatar DL says:

    The full article is now freely available online here so you can view the full piece

Write a Comment

Your email address will not be published. Required fields are marked *

button to share on facebook
button to tweet
button to share via email