When the counterintuitive becomes the norm, arguments get twisted out of shape

I was bothered by a recent post on the sister blog. The post was by political scientist David Fortunato and it was called, Would “concealed carry” have stopped Dylann Roof’s church shooting spree?.

What bugged me in particular was this sentence:

On its face, the claim that increasing the number of gun carriers would reduce crime seems logical (at least to an economist). If more people carry guns, then criminals would understand that the likelihood of their victims defending themselves with a gun is higher and would therefore be less likely to commit crime. In simple economic terms, easing concealed carry seeks to increase the cost assailants pay to commit a crime, so they choose not to, we hope.

On its face, I think the above argument is ridiculous in its naive division of the population into “criminals” and “their victims,” with no mention of the idea that the presence of the gun could lead to escalation. Also the silly idea that “a criminal” decides whether to “commit a crime,” with no sense that events develop in unexpected ways. The above argument is “simple” indeed, but I think it does no service to economists to consider it as logical.

This sort of thing bothers me about a lot of op-ed style writing, that the author has a certain flow in mind (in this case, it seems that Fortunato wanted to say that evidence on gun control is weak, and so he decided to lead up to it in this way). Also a problem with social science that everything has to be a “puzzle.” Fortunato writes:

But scientific research on the ability of concealed carry to reduce crime has yielded mixed results. A few studies suggest these policies are effective, but even more suggest that making concealed carry easier does not reduce crime and may even increase instances of firearm injury. Why is this the case?

Why indeed? It hardly seems a mystery that if more people are carrying guns, then they might use them accidentally or on purpose to shoot people.

Again I see a problem with Fortunato’ theoretical framework when he writes: “More people choose to carry firearms . . . Criminals observe (or infer) that more people are carrying firearms,” as if there is some division between “people” and “criminals.”

I don’t mean this as a slam on Fortunato’s research, and I’m not saying that the obvious answers to his question tells the whole story. Of course I have a huge respect for research on topics that laypeople consider to be obvious. But his post really bothers me as representing some of the unfortunate tendencies of journalism and social science to take a counterintuitive idea (giving out guns is a way to reduce gun crime!) and normalizing it so much as to make it the default explanation.

From the Monkey Cage perspective, I don’t think much can be done here—in the old days I would’ve done a follow-up post but I know we don’t really do these anymore—but maybe we can all be aware of the pitfalls of taking a counterintuitive framework as the norm. I think the content of Fortunato’s post is excellent but I’m disturbed by how it’s framed, as if it’s some sort of surprise that a wacky counterintuitive bank-shot theory of the world doesn’t actually seem to be borne out by the data.

P.S. I elaborate here in comments:

When someone focuses on an indirect effect instead of the first-order effect, that’s what I call a counterintuitive argument. Indeed, this might be the core of many or even all counterintuitive arguments: The first-order effect is obviously in one direction (increase the number of guns, make it more normal to carry guns as a way of life, and people like Dylann Roof are more likely to go shoot people), but there’s this second-order effect (Dylann Roof might not shoot if he thinks the people in the church are armed too).

The counterintuitive claim is that the second-order effect outweighs the first-order effect. It’s counterintuitive for the same general reason that an echo is typically not as loud as the original sound, or for the same general reason that elasticities are typically less than 1.

Some counterintuitive claims are true. But it hardly seems puzzling when a bank-shot, counterintuitive claim is not supported by the data.

66 thoughts on “When the counterintuitive becomes the norm, arguments get twisted out of shape

  1. If you want to give yourself a real headache, look at the gross statistical manipulations occurring around the gun control topic (on both sides). On the anti-2A side of things we see use of overly coarse data, gross oversimplifications of context/environments, and bizarre apples-to-oranges comparisons (for example, if you look at the FBI’s openly published crime data, you see that a) the vast majority of US gun-related crime ties directly to gangs and drugs, and b) the vast majority of these crimes are with handguns, not the much-maligned and mythical “assault weapon” – remove the gangs+drugs activities and you find the crime rates rival or beat those in other countries – and then ask the question: “Is it access to the guns or the war on drugs causing/escalating the problem?”).

    On the pro-2A side we see arguments like the one Fortunato dissects about whether or not increased concealed carry would stop attacks like those in Charleston, not to mention secondary negative impacts (there are numerous anecdotal examples of concealed carrier stopping mass shootings, but we also know that having firearms in the house correlates to increased likelihood of suicide, though this is also likely a simplification). These arguments oftentimes falter because you can’t prove the negative. We can look for correlations, but those are a far cry from anything meaningfully causal.

    If nothing else, these issues highlight that none of these topics are easily simplified or resolved. Stats can help us understand some things, but it’s no panacea. Interesting times. :/

  2. Hmmm… My training as an economist obviously underlies my notion that there is nothing counterintutive here (to me!). This is no different, it seems to me, than the commonplace notion that nuclear proliferation may have (weakly) reduced the incentive for the use of nuclear weapons, with the weakness stemming at least partially from your point that the escalation scenario goes in the other direction. As to the division between “people” and “criminals” the division happens along various margins. As an economist, I would simply say that we are all people or criminals depending on the milieu in which we find ourselves. While there are certainly some who would never resort to criminality even if everyone kept their doors open 24/7, and there were no police or courts, there would almost certainly be *more* observed criminals under those circumstances. Why wouldn’t you expect, at the margin, to find a few individuals who, confronted with a heightened probability of being shot during their current occupation, choose a different metier? We can certainly discuss how many such people there are, and whether the net effect once you add in mistaken shootings and various accidents are positive or negative (and whether there are some who might be emboldened to become criminals at the margin because they have access to firearms), but I guess I just don’t see what’s counterintuitive. but that’s the great thing (and the bad thing) about intuition, right?

    • Jonathan:

      We’re talking about two things going on when weapons proliferate: (a) that they can get used; (b) they can deter.

      Nuclear weapons have not been used (testing aside) for 70 years so it makes sense to focus on the deterrence. Guns get used all the time. My problem with Fortunato’s argument is that he seems to be focusing on (b), even though the consequences of (a) are so apparent.

      When someone focuses on an indirect effect instead of the first-order effect, that’s what I call a counterintuitive argument. Indeed, this might be the core of many or even all counterintuitive arguments: The first-order effect is obviously in one direction (increase the number of guns, make it more normal to carry guns as a way of life, and people like Dylann Roof are more likely to go shoot people), but there’s this second-order effect (Dylann Roof might not shoot if he thinks the people in the church are armed too).

      The counterintuitive claim is that the second-order effect outweighs the first-order effect. It’s counterintuitive for the same general reason that an echo is typically not as loud as the original sound, or for the same general reason that elasticities are typically less than 1.

      Some counterintuitive claims are true. But it hardly seems puzzling when a bank-shot, counterintuitive claim is not supported by the data.

      • I don’t disagree with any of that, but the problem is that the categorization of first- and second-order is part of the intuition. Start with the status quo and some horrifying level of violence. Now assume that there is some unique publicized incident that induces 1,000,000 people to purchase guns in self-defense. The use of these guns is clearly second-order, so net incremental effect is the difference between two second-order effects. (The weak part of this argument, of course, is the increase in lethal arguments, accidents and suicides, but I’m assuming we don’t count those.) The indirect effects of the single precipitating incident could be either positive or negative. And could easily, just through magnification, swamp whatever the initial direct effect might be, which could have been a single death.

      • To Jonathan’s response I would add that economists seem to think that incentive effects _are_ first order in their framing of the world.

        To paraphrase Fletch, “It’s so simple maybe you need a refresher course. [Leans arm on hot gun barrel, then jumps away] Heyya! It’s all incentives nowadays.”

        To an economist, I can see how the first order question is not about the gun per se, but about the decision making process of the shooter.

        • First/Second order effects are about real, underlying causal mechanisms. The Garden is about spurious findings posing as real, underlying causal mechanisms.

    • I think the deterrent effect of nuclear weapons is dubious at best too. You can tell by the very expensive efforts we go through to keep them out of the hands of countries that aren’t considered stable. Throw in chemical weapons and you have a very good counter example where countries have used chemical weapons despite any deterrent effect. World War I, for example.

      • What explains the unusually peaceful decades we have enjoyed post-WW-2? I had always assumed the nuclear deterrent played a big role in preventing any new full blown wars between the major nations.

        But is there another explanation?

  3. I talk to students about this a lot, which is that just because someone has a story that has some kind of internal logic that makes sense, that doesn’t mean that it is actually true. “Scared Straight,” “Abstinence only sex education” and “Project DARE” all have internally coherent logics, but they also have built into them unstated and untested and undifferentiated ideas about how 16-22 year olds think and act and the same is true about a lot of the argumentation about guns.

  4. As an economist I find it puzzling that the NFL can’t stop people from fighting at football games. They get drunk and are passionate and antagonize each other until fists fly, but only fists. If the NFL simply let, or better yet mandated, people to bring Guns to the game no one would get hurt.

    (-sarcasm and typed on an IPhone)

  5. A bit of comparative analysis internationally seems to suggest the problem isn’t guns in the US, thus arguments around guns and their availability will never solve the issues you guys face. But then again this is coming from a guy who has spent most of his life in countries where neither people or police typically carry weapons, and gun deaths are rare (here in NZ we at least give the army water-pistols and nerf-guns so they won’t cry <>). Hmm though I have also spent time in Switzerland where everyone has guns, and gun deaths are also rare (via rate based comparison).

    Perhaps all this comes down to the illusion of appeasing false dichotomies in culture. Half the population are adamant x, half are adamant not-x, all discussion/compromise tends to do is cement ones prior belief and push behavioural within-group norms toward the extremes. Oh and that nutty anachronistic constitution can’t help either.

    You’re a weird lot :p

  6. Two more general concepts needed: (multiple) Nash Equilibria and Conditional Probability.

    1) I would suggest that when most people discuss economics they hone in on the story supporting one preferred Nash Equilibrium and pretend that there are not many other points of stability in the interactions of various collections of people. The problem isn’t focusing on the indirect effect per se, it’s using an oversimplified single equilibrium mental model that neatly conforms to a story they like (affect heuristic) that ignores the complexity of reality (and most always does not offer useful predictions)

    2) Whether gun control ‘works’ or not may not be an intelligent question. I would suggest that whether gun control cuts murders, suicides, etc. perhaps depends whether you condition the existence of gun control on being in a country that has very few guns in circulation and a history of strict gun control (e.g. Britain) or perhaps condition gun control’s efficacy instead on imposing it on a country that has had freely flowing firearms for decades – leading to more guns than people and no centralized / decentralized bodies having much of an idea as to where those guns currently are. I suspect it ‘works’ in the former case but I have a significant amount of skepticism for the latter case, which approximately describes the United States.

  7. If statistics were outlawed, only outlaws would practice statistics. Think about that before you strap on your 38 before your daily bike ride to Columbia.

  8. Andrew,

    “no mention of the idea that the presence of the gun could lead to escalation.”

    What is the mechanism whereby presence leads to escalation? Why is there no escalation at gun shows? Or shooting ranges? Or hunting in the outdoors?

    I am NEVER afraid of escalation at any of these places, and I frequent them regularly. In fact, the thought never occurred to me. In all my gun discussions over the years, I am pretty sure that no one has ever mentioned this idea.

    If people who are familiar with firearms and use them regularly give the notion of escalation a probability less than epsilon, is it really worth considering seriously?

    Regards,

    Bruce

    • Chris Kyle was shot and killed at a shooting range near Chalk Mountain, Texas, with his friend, Chad Littlefield–Wikipedia

      I would say epsilon is positive from this occurrence. Incidentally, inner city areas are where guns are the most prevalent and they also have a very high murder rate. “Packing” was very common in the antebellum South and there was also a large number of firearm murders (Strom Thurmond’s father, for example, shot and killed a man–I realize this is post-bellum but it’s a recognizable name even to current readers–Andrew Jackson was in multiple duels and carried a lead shot ball in him the rest of his life after one of these–no doubt contributing to his ill-temper through lead poisoning).

      • Numeric,

        What incident “escalated” at the range when Kyle was shot? By all accounts there was no fight, and Kyle’s gun was still in his holster when he was shot. Apparently it was an ambush, not a fight that “escalated” to the use of guns.

        Regards,

        Bruce

        • Well, technically it was an escalation–a nearly infinite one, going from zero to death in a split second (escalation means rapid increase and this was truly rapid). And in actual fact Kyle would have had a better chance away from the gun range as he could have (possibly) seen the shooter was armed outside of a situation where one would normally expect firearms (though with open carry, I guess you expect them everywhere). I think the central point is whether we would all be better off it no one was armed, some were armed (the current situation), or everyone was (open carry nirvana–though I should point out an early Congress passed a bill requiring every (free) adult male to own a firearm and George Washington signed it–I think of that whenever someone claims the individual health care mandate is unconstitutional). I tend to lean towards we’d all be better off without firearms, looking at the experience of European countries and their low homicide rate.

    • > “Why is there no escalation at gun shows? Or shooting ranges?”

      The word “escalation” implies that there was something–a heated argument or dispute–to escalate. Why weren’t you afraid of escalation in these places? Because there wasn’t a heated dispute to escalate.

      My guess would be that if there was a heated dispute between armed people at a shooting range, that you would in fact worry about escalation.

      • I’ve seen arguments at gun shows and shooting ranges. Never worried about escalation. Just because someone is mad and has a gun, it does not follow that he will use the gun. Are you projecting here?

        • It’s an interesting and unaddressed empirical question (also a separate one from whether or not one person worries about escalation). From what I was able to find, there hasn’t really been any research on it.

          Though it does look like the biggest danger at a gun show isn’t escalation, but accidental firing.

    • Anon:

      From what I’ve read, many deaths involved in crime are accidental in that the killer did not come into the situation intending to kill, but the gun enabled the action. An example would be when George Zimmerman killed Trayvon Martin. I don’t think anyone thinks Zimmerman started out with an intent to kill, but the gun enabled it. From a statistical point of view, the point is that if you condition on the fact that someone is being killed, this is conditioning on an intermediate outcome and it interferes with any estimates of causal effects.

      Also, it is a crime to shoot someone by accident, no? I mean, sure, they didn’t put Dick Cheney in jail. But if my neighbor was playing around with a gun and shot me by accident, yeah, I’d want them arrested! I wouldn’t just say, hey, it’s just an accident so no crime.

    • Crystal clear? I’m trying not to be too snarky, but, on the face of it, we have both the most guns per capita and the most people in prison on any country. So in the cross-country sense, that is not crystal clear. Do you mean: given the current level of gun ownership, new laws marginally changing the percentage of gun owners leads to a reduction in crime? Do you also mean “legal gun ownership” reduces crime?

      It also looks to me on a quick glance that gun ownership rates have been unchanged in the USA or slightly trending down over the years, while violent crime has been dropping rapidly.

      What is the evidence you have in mind?

    • Your “accident” may well be my “negligent homicide” or “depraved indifference to human life” not to mention how many “accidents” are really suicides. If you are going to argue the case argue i as a serious policy case that the net effects are what you claim, just like a drug has to argue benefit net of adverse events.

    • The evidence that more guns reduce crimes is clear crystal

      Actually it’s not. Lott suggested it did but were were few problems with his analysis. Other analyses with the datasets did not find a difference but then for some technical reason which I totally forget it became apparent that the FBI data could not not analyzed to answer the questions being asked.

      For international comparisons I remember, dimly, looking at homicides and gun ownership. The US was much higher about 10-15 years and the differences very noticeable.The USA seems to have more of a culture of violence and lots more guns than most countries in something Western Europe, Australia Canada, New Zealand and France.

      Gun violence was higher in Canada than the other countries but even in Canada the rates were lower by, at least, an order of magnitude. I couldn’t find much on Switzerland and my languages skills were not good enough for other countries

      It also looked as if the UK had more petty crime but less violent crime than the US. It could have just been different reporting rates but there was no way to tell. I think Canada was about midway between the UK and US but different crime legislation and reporting rates make it difficult to know for sure. Comparing stats across countries an be a real pain

      It also it appeared from a couple of studies that having a gun in the house greatly increased the chance of getting shot by one of your nearest and dearest however this was more dubious especially as one study compared Vancouver Canada and IIRC Seattle Washington and with the drastically different base rates it was hard to really have a lot of confidence in it.

      IIRC, Switzerland had almost no gun violence, yet most adult males had army weapon. A correspondent from Switzerland told me that it was probably partly due to the fact all weapons owners were trained solder who would never think of using their military weapons unless on training.

      I remember, dimly, looking at violence and gun ownership and the US was much higher about 10-15 years and the differences very noticeable. The USA seems to have more of a culture of violence and lots more guns than most countries in something Western Europe, Australia Canada, New Zealand and France. Gun violence was higher in Canada than the other countries but even in Canada the rates were lower by, at least, an order of magnitude. I couldn’t find much on Switzerland and my languages skills were not good enough for other countries

      It also it appeared from a couple of studies that having a gun in the house greatly increased the chance of getting shot by someone, quite possibly one of your nearest-and dearests however this was more dubious especially as one study compared Vancouver Canada and IIRC Seattle Washington and with the drastically different base rates it was hard to really have a lot of confidence in the study

      IIRC, Switzerland had almost no gun violence, yet most adult males had army weapon. A correspondent from Switzerland told me that it was probably partly due to the fact all weapons owners were trained solder who would never think of using their military weapons unless on training.

      Australia, Canada and the UK had quite strict handgun laws and I think were just bringing in longun requirements.

      I suspect that there are so many guns in the US that gun control wouldn’t work but someone mentioned at that time, it’s difficult to do a drive-by with a axe. As someone mentioned at the time, it’s difficult to do a drive-by with a axe

  9. The National Academy of Sciences did an expert report on the numerous econometrics studies on this question (whether concealed-carry laws increase or decrease violent crime). The panel concluded that the evidence didn’t support a confident inference 1 way or other b/c of the fragility of the regression models (w/ 1 dissenting vote by James Wilson, who concluded that evidence supported the “reduces crime” conclusion).

    The Panel makes a very convincing case that multivariate regression family of observational studies won’t work in this context — or in death penalty, where another panel reached same conclusion on deterrent effect: who the hell knows? — given inherent difficulties in model specification.

    But I think the very fact NAS commissioned the study supports the view that the “reduce crime” hypothesis is one plausible enough to warrant empirical examination. I believe the hypothesis is plausible but hold priors that the impact is inconsequential either way.

    This is also a classic issue, of course, where people tend to form perceptions, both of evidence & of what experts believe, that fit their ideological predispositions.

    And no, in case anyone wonders, <a href="http://www.culturalcognition.net/blog/2015/3/25/you-talking-to-me-are-you-talking-to-me-actually-no-im-not-t.html"I wouldn't say that such studies support any inferences *about* Andrew's reasoning given Andrew's own belief that the “reduce crime” thesis is implausible.

    • Dan:

      It’s not that I’m saying the “reduce crime” hypothesis is implausible. (I didn’t think that’s what I was saying, and I did a quick search on the above post and discussion and didn’t find myself saying that either.)

      I didn’t object to Fortunato being interested in studying the hypothesis that concealed-carry laws reduce violent crime, or even the hypothesis that concealed-carry laws reduce the number of violent deaths (which is a slightly different thing). What I objected to was Fortunato’s unidirectional statement of the question (seemingly not considering the possibility of the other direction) and also that he (a) shifted the question from “violent death” to “crime” (which to me isn’t quite right, given that the concern was not that Roof committed a crime but rather that he killed a bunch of people), (b) set up a “criminals” and “victims” frame which has serious endogeneity problems (as we say in econometrics), and (c) set up the counterintuitive position as the default.

      I’d really have had no problem with Fortunato’s post at all, had he just said that it’s very difficult to use available data to estimate the effects of gun laws. What I was bothered by was his implication that it’s somehow a surprise that the second-order effect (other people have guns, maybe I won’t shoot) does not, in the data, outweigh the first order effect (I have a gun and I can shoot it).

      As commenters have noted above, reasonable people can differ on what is the first-order and what is the second-order effect, but I think Fortunato is (unintentionally) stacking the deck by, essentially, conditioning on the intermediate outcome and by framing the potential effects in a one-sided way.

    • James Q Wilson was also the leading proponent of the “broken windows” theory of policing, which has lead to such results as a sixth of the Baltimore population being arrested every year, the creation of the type of oppressive governmental policing as we say in Ferguson, Mo., and an incredibly high rate of incarceration of the poor and minorities. The point is that he is hardly an impartial scientist examining data on gun laws.

    • There’s a well-defined definition in mathematics and in many physics applications, in which one is performing a series expansion, but I think it’s an ill-defined term in the context we have here…although still a useful concept, I think.

      Suppose a company sells X widgets per year, at $D per widget, so they take in a total of X*D dollars per year. Their marketing department decides to increase the cost by $d per widget. How will this affect their revenue? One guess would be, their new revenue will be X*(D+d) dollars per year. We might call that the first-order estimate: we have directly changed one parameter — the price — and we hold everything else constant. But this would be a lousy estimate in this case, because if the price goes up the number of sales will go down, from X to X-x.

      The terms “first-order” and “second-order” might make sense in this example: the first-order effect is what you get if you change just the one parameter, the second-order effect is what you get if you take into account how that first parameter affects the other parameters.

      The number of gunshot victims is given by (number of people who carry guns) * (average number of people who get shot by each person who carries a gun). If you directly change the first of those quantities, that’s what Andrew means by the first-order effect. You have a product, and you are increasing the first term. By changing that first term, you expect the other term to change too, and it’s certainly possible that it could change even more than the first term does (and in the opposite direction).

      • This is actually wrong. If $d rise in price leads to x drop in sales (with x~d) then the first order effect on revenue is d*X-D*x and the second order is -d*x (and, of course, if it is a real series expansion you will have x^2 and d^2 as well).

    • I guess that Prof. Gelman means that the probability of violent (or gun-violent) altercation is ~N*n where N is total population and n is gun-bearing population (n<<N) and probability of deterrence is ~n^2. Thus change in n by dn should lead to ~N*dn more violent altercations (0 order in n) and the probability of deterrence is changed by ~n dn (first order in n).

  10. I live in the San Fernando Valley on the fringes of the entertainment industry, which is an interesting natural experiment because the area is both liberal and gun-crazy:

    http://www.unz.com/isteve/hollywood-liberal-gun-nuts/

    My impression is that gun possession among homeowners went way up after the 1992 Rodney King riots and the TV news footage of Korean shopkeepers blasting away in self-defense at rioters when the LAPD wouldn’t defend them.

    My impression is that home invasions are way, way down, and so is burglary and even graffiti.

    • Do you really think that the nationwide decrease in violent crime is attributable in any identifiable or measurable way to changes in gun ownership?

      I mean, in the case of LA, everything has been getting better since 1992 – crime rates, air quality, Hollywood Blvd, the LA River… even Dr. Dre managed to up his game another level.

      Here’s another interesting “natural experiment”: in 1993 the Brady Bill was passed. Rates of homicide by handgun, which had been increasing since 1985, began to drop and (eyeballing it) fell 40% or so in the next 5 years.

      https://en.wikipedia.org/wiki/Gun_violence_in_the_United_States#/media/File:Ushomicidesbyweapon.svg

        • Well… It depends on what you value, I guess. NWA was pretty great, and they probably generated the most pearl-clutching among the news cycle clutching set*, which is cool. I understand why some people think that was the Golden Era of West Coast Hip Hop.

          But the Chronic and Doggystyle sold more albums, and they were better than anything NWA ever did. It was a whole new kind of hip hop: smoother, catchier, better rhymes, groovy as f***. That was like right at 92/93. But that whole G-Funk era played out from like 1991-1996, and it changed the sound, style, and status of hip hop forever. So I think that was his peak.

          Of course then he started Aftermath, invented a 3rd style (¿post G-funk?), signed Eminem, got filthy rich peddling the best beats around to the best rappers and singers in the game and launched a succession of proteges: Em, 50 Cent, The Game, Kendrick Lamar. And then he got involved in Beats by Dre and made $500 Million off that. $500 Million. Some people would call that “upping his game a level.”

          *Because the news cycle pearl clutching set care a lot about threats of violence towards cops, but very little about threats of violence towards women or young black men.

  11. Andrew treats the shooting as binary:
    “An example of the presence of the gun leading to escalation is what happened in that church, the event that prompted the blog post being discussed here.”

    The (binary) incident is that he shot people. Dylan wanted to kill those people. He went and bought a gun legally. He could have illegally acquired the gun as well. He didn’t have to hide that he was carrying it. Concealed carry has nothing to do with this aspect of the incident.

    Concealed carry impacts the severity of the crime, as is alluded to here:
    “…(a) shifted the question from “violent death” to “crime” (which to me isn’t quite right, given that the concern was not that Roof committed a crime but rather that he killed a bunch of people)”

    He would have killed fewer people if someone armed in the congregation had stopped him. Still a tragedy but no longer mass murder. Concealed carry (or more guns in general) does increase the probability that a gun will be used in an incident but it also decreases the probability of mass shootings like this event. Less guns (period, illegal or legal) means less Chicago shootings (which don’t make headlines) but increases the chance of (headline grabbing) mass shootings.

    Andrew is critiquing the argument style, which is fair. There are certainly first- and second-order (binary) effects, but I don’t see any comments regarding the severity of the incident.

  12. Here’s an argument that fits your ‘counterintuitive’ pattern that I found myself propounding recently:

    Mandatory bike helmets do not necessarily save lives, because they may put people off from cycling who would benefit by improving their health.

    I believe this makes a bit more sense than the guns-in-church-would-save-lives argument, because mandating laws that put barriers in the way of everyone who wants to cycle could well outweigh a much smaller number of bike accidents. Although my prejudices (pro-cycling, anti-gun) could well be getting in the way.

    In general many ‘moral hazard’ arguments fit into this pattern too. I wonder if there are any common features where we could say second-order effects were actually likely to dominate.

    • The evidence that more guns reduce crimes is clear crystal

      Actually it’s not. Lott suggested it did but were were few problems with his analysis. Other analyses with the datasets did not find a difference but then for some technical reason which I totally forget it became apparent that the FBI data could not not analyzed to answer the questions being asked.

      For international comparisons I remember, dimly, looking at homicides and gun ownership. The US was much higher about 10-15 years and the differences very noticeable.The USA seems to have more of a culture of violence and lots more guns than most countries in something Western Europe, Australia Canada, New Zealand and France.

      Gun violence was higher in Canada than the other countries but even in Canada the rates were lower by, at least, an order of magnitude. I couldn’t find much on Switzerland and my languages skills were not good enough for other countries

      It also looked as if the UK had more petty crime but less violent crime than the US. It could have just been different reporting rates but there was no way to tell. I think Canada was about midway between the UK and US but different crime legislation and reporting rates make it difficult to know for sure. Comparing stats across countries an be a real pain

      It also it appeared from a couple of studies that having a gun in the house greatly increased the chance of getting shot by one of your nearest and dearest however this was more dubious especially as one study compared Vancouver Canada and IIRC Seattle Washington and with the drastically different base rates it was hard to really have a lot of confidence in it.

      IIRC, Switzerland had almost no gun violence, yet most adult males had army weapon. A correspondent from Switzerland told me that it was probably partly due to the fact all weapons owners were trained solder who would never think of using their military weapons unless on training.

      I remember, dimly, looking at violence and gun ownership and the US was much higher about 10-15 years and the differences very noticeable. The USA seems to have more of a culture of violence and lots more guns than most countries in something Western Europe, Australia Canada, New Zealand and France. Gun violence was higher in Canada than the other countries but even in Canada the rates were lower by, at least, an order of magnitude. I couldn’t find much on Switzerland and my languages skills were not good enough for other countries

      It also it appeared from a couple of studies that having a gun in the house greatly increased the chance of getting shot by someone, quite possibly one of your nearest-and dearests however this was more dubious especially as one study compared Vancouver Canada and IIRC Seattle Washington and with the drastically different base rates it was hard to really have a lot of confidence in the study

      IIRC, Switzerland had almost no gun violence, yet most adult males had army weapon. A correspondent from Switzerland told me that it was probably partly due to the fact all weapons owners were trained solder who would never think of using their military weapons unless on training.

      Australia, Canada and the UK had quite strict handgun laws and I think were just bringing in longun requirements.

      I suspect that there are so many guns in the US that gun control wouldn’t work but someone mentioned at that time, it’s difficult to do a drive-by with a axe. As someone mentioned at the time, it’s difficult to do a drive-by with a axe

      • Right, the high rate of gun ownership contributes to the relatively lower rates of things like home invasion, burglary, and drunken beatings among American working class whites relative to British working class whites. On the other hand, it boosts homicides and probably suicides.

        To give an example of an English crime that almost doesn’t exist in America, in part due to gun ownership, recall in “A Clockwork Orange” when Alex and his droogs drive out from the slums to a rural cottage and terrorize the residents. When Anthony Burgess wrote that in 1962, that kind of urban->rural home invasion almost didn’t exist in England. He based the scene on the victimization of his wife’s family by American GIs during WWII. But by the 1990s, that horrifying crime had become fairly common in Britain.

        In contrast, in America it was almost unknown. In the 1990s nobody drove out from South Central to Ventura County to terrorize residents, in part because guns are legal for homeowners and common in Southern California, in part because of the effectiveness of cops racial profiling suspicious cars full of youths in America, and some other reasons.

        Since then the British have been fighting crime with a variety of technocratic innovations, such as security cameras everywhere, that aren’t exactly the same as in Clockwork Orange, but Burgess did also anticipate the technocratic spirit of the response.

  13. Andrew,

    I have to say that I find this argument really rather revealing of your own bias here.

    I myself don’t think there’s a particularly good way of designating one potential effect or the other as the default, or “first order” effect, because it depends on one’s priors, and those are mostly ideologically driven these days.

    To see how one might think that the default might conform to Fortunato’s choice, you really have to occupy his frame of mind, and that of many gun owners.

    It’s probably fair to say that people who support robust gun rights really do think that the explicit reason for owning handguns is for self protection. If that is the default position for those who own handguns and those who support doing so, then the most natural position to take is that the more guns that are around for self protection, the less violence there should be. It becomes then counterintuitive that more such gun ownership should engender more violence, rather than less. It is, for them, a second order effect that introducing more guns should engender more violence, because, for them, that would require the greater use of guns for reasons other than the explicit reason for which they were introduced.

    Now of course one can look at this with the very different priors of a gun control advocate and come up with very different default assumptions. But I should think it pretty obvious that that depends on a peculiar ideological perspective as well.

    One would think this overarching point would be obvious particularly to a Bayesian.

Leave a Reply to Steve Sailer Cancel reply

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