Your gun control ideas won’t work. This one will.

[FrankH]

Interesting. Note: If you want to see the graphs you will need to open the article.

https://zachmortensen.net/2018/02/20/your-gun-control-ideas-wont-work-this-one-will/

Five years ago in the aftermath of the Sandy Hook shooting I was frustrated by the simplistic narratives being trotted out to explain why these tragedies happen. Gun violence is a multi-factorial problem: Yes, other countries with fewer guns have fewer gun deaths. Yes, military-style rifles have been used in many recent massacres of innocents. Yes, we have the 2nd Amendment. Yes, the NRA and the Brady Campaign and other organizations influence elections. Yes, many who commit mass murder are mentally ill or have extreme ideologies. Yes, we have a system of federal background checks with known loopholes in some states. Yes, we are actually talking about legislating the behavior of people who seem to have no regard for the law in general. Yes, in some cases local and federal authorities have failed to do their job.

control-all-the-guns
Image credit: http://hyperboleandahalf.blogspot.com
I’m skeptical of any “simple, common-sense” solution to a multi-factorial problem. For decades I have dealt with this class of problems daily in my professional life. Simple, common-sense solutions — however well-intended — are almost always wrong and often make things worse.

So in January of 2013 I assembled data from public sources to study the relationship between firearm ownership and homicides across the United States and worldwide. That analysis found the variance in homicide rates was inversely correlated with the household firearm ownership rate: Both the lowest and the highest homicide rates in the US and worldwide occurred in the places with the fewest guns. The homicide rate was also inversely correlated with the firearm ownership rate: States and countries with more guns had relatively fewer homicides on average.

I published my analysis and (rather verbose) commentary here. The article resurfaces on Facebook, Reddit, and Twitter after every mass-shooting event, and I get a push notification about the increase in website traffic. These notifications now bring feelings of dread; each signals a new tragedy.

A new, deeper look at the problem

Following the Pulse shooting in Orlando in July of 2016 I received an email from Jeff Dyer, one of my professors at Wharton. He pointed me to an article that he had published in Forbes calling for legislative and innovative solutions. I responded with a link to my 2013 article, and we engaged in a thoughtful dialogue that prompted me to revisit my original work and test some new hypotheses. The conclusions were nothing short of mind-blowing, but I was busy with other projects, so this work sat unpublished until now.

I sought to test whether variance in violent crime (homicide, rape, robbery, aggravated assault) and property crime (burglary, larceny/theft, motor vehicle theft) can be explained by any of the following variables at the state level:

Firearm ownership rate
CCW status (shall-issue, may-issue, no-issue, unrestricted)
CCW permits per capita
GDP per capita
Income inequality (as measured by the Gini coefficient)
Racial diversity index
I’ll spare you the details of the statistical analysis. For those who are interested, the raw data are available here. Visually:

predicted-actual-homicide-rate

Here is a brief summary of key findings:

State CCW status and CCW permits per capita are not statistically significant predictors of any violent crime or property crime. There is no evidence of any relationship, positive or negative, between the number of legally-armed citizens and the number of violent crimes or property crimes.
No firearm-specific variables in the data are statistically significant predictors of any kind of property crime whatsoever.
Firearm ownership rate, racial diversity, GDP, and income inequality are statistically significant predictors for violent crime at the state level, especially for robbery (F=85, R-squared=0.88) and homicide (F=31, R-squared=0.73).
Factors that predict violent crime

Firearm ownership is a statistically significant predictor of homicide rate (t=4.43), but the effect of each firearm is vanishingly small (and that of each homicide disproportionately large) at the margin. For example, the model suggests that a population of 1.2 million must reduce its firearm ownership rate by 1 per 100 (i.e. destroy up to 12,000 firearms) to avoid one homicide per year. Alternatively, each additional homicide per year would prompt the purchase of up to 12,000 additional firearms.

Racial diversity. This analysis was colorblind. I used publicly available data from the Kaiser Family Foundation for the racial composition of each state (White, Black, Hispanic, Asian, Native, and of two or more races). The diversity index represents the probability of a random pairing of individuals being of different racial groups. The analysis found that more-diverse populations have higher rates of homicide (t=4.75) and robbery (t=3.41). This statistical finding might seem disturbing, but the magnitude of the effect is rather small: If we were to make our hypothetical population of 1.2 million of any single race, the model predicts that we would avoid only three homicides per year.

GDP per capita. The analysis found an inverse relationship between GDP and both homicide (t=-5.86) and robbery (t=-5.29). In other words, as a population gets wealthier, homicides and robberies both decrease. The model implies that our hypothetical population of 1.2 million could avoid one homicide per year by increasing GDP by $1,700 per person.

Income inequality. The analysis found an interaction between the Gini coefficient and the GDP per capita that was a strong predictor of both homicide (t=6.80) and robbery (t=7.06). In other words, the wealthier the population and the bigger the gap between the highest and lowest income earners, the more homicides and robberies. The model suggests that our hypothetical population of 1.2 million, assuming the current US GDP per capita of $57,466 and Gini coefficient of 0.41, would avoid 60 homicides per year if it had Canada’s Gini coefficient of 0.34 while holding all other variables constant.

Factor effects and implications

Let’s compare homicide rates and their predictors between the US and Canada and illustrate what the US homicide rate might look like under Canadian levels of firearm ownership, racial diversity, GDP, and income inequality:

us-canada-comparison-1

Canada has more racial diversity and lower per-capita GDP than the United States, and in our model these are associated with more homicides. However, Canada’s lower firearm ownership rate and lower Gini coefficient more than offset this increase: The model predicts the net effect of the US adopting these Canadian attributes would be a reduction of 2.1 homicides per 100,000 people, or 6,898 homicides per year.

Interestingly, 99% of the reduction in homicides in the scenario above was due to the change in the Gini coefficient.

What if the US were to adopt only Canada’s Gini coefficient and keep all other attributes the same?

us-canada-comparison-2

The model predicts a drastic reduction in homicides, from the FBI reported total of 17,250 homicides in 2016 to a hypothetical 655; a difference of 16,595.

Let that sink in for a minute: Reducing income inequality to match that of our nearest neighbor — by helping the poorest among us to be less poor — would do 100x more to reduce homicide than would adopting Canada’s more-aggressive gun control measures. And we’re talking about reducing all homicides, not just those related to guns.

Getting the same effect from gun control alone, according to the model, would require disarming twice as many US households as actually own guns, an impossibility.

Getting the same effect from GDP growth alone would require a $102,000 increase in per-capita GDP, or growth of about 177% net of price inflation. That’s not going to happen.

The model is telling us that homicide is not a race issue. It’s not a gun issue. It’s an economic issue, specifically one of prosperity and equality. Few seem to understand this, and if they do they aren’t talking about it. Until we have a national conversation about what’s actually at the root of the problem, we have no hope of solving it.

A call to action

I’m calling on you — each of you, regardless of your persuasion — to change the way you engage on this issue. What you’ve done in the past hasn’t worked. At all. You need to change your approach in order to get different results, and I don’t mean that you keep saying the same simple and misguided things, only more loudly and obnoxiously than before.

To my friends who feel the need to do something about this plague of senseless violence: I feel it, too. These data show that most of your ideas about how to solve the problem aren’t going to work very well. Ironically, many of you also feel passionately the need to address economic inequality, but you’ve allowed yourselves to put that on the back burner while you get worked up over the latest tragedy or whatever the Russians may have done to influence the last election. Please stay focused on what matters most.

To my friends who resist gun control because they know that the guns are not the problem: The data show that you are partly right. But most of you are rather stupidly ignoring that a serious problem still exists, and you offer no solutions. Many of you also resist any intrusion by the state to address economic inequality, and I empathize with that sentiment. But today you must choose: Do you value economic inequality more than you value the 2nd Amendment? Today you have an opportunity to reframe the issue of violence around economic inequality, find common ground with the other side, and earn a seat at the table to solve this defining problem of our generation. If you fail to do so, you are complicit in allowing the statists to frame violence exclusively as a gun issue. You will ultimately be marginalized, disarmed, and denied a voice in the conversation about the larger economic issue that is literally tearing our nation apart.

To my friends who find themselves in the middle, neither passionately for nor against gun control: Your instincts are right, gun control will not solve the problem. And you don’t need to buy into the false dichotomy of us-vs.-them thinking on this or any other issue. But you owe it to yourself, your friends, your family, and your community to join the conversation. It will affect you at some point. Share ideas that resonate with you, even if you think they might be lost in the cacophony of the loudest voices in the room.

Please, all of you — I’m begging you — pull your heads out of the sand and lead.

[Pirate]

Thanks Frank. I bet if one runs this model on SA the results would (1) be very interesting, and (2) show that gun control (as predicated by the FCA and planned ammendments) won't make a difference.

It may even suggest an interesting tack for gun-rights organisations to take...

Quoting directly from the article:

"Let that sink in for a minute: Reducing income inequality to match that of our nearest neighbor — by helping the poorest among us to be less poor — would do 100x more to reduce homicide than would adopting Canada’s more-aggressive gun control measures. And we’re talking about reducing all homicides, not just those related to guns."

and

"To my friends who resist gun control because they know that the guns are not the problem: The data show that you are partly right. But most of you are rather stupidly ignoring that a serious problem still exists, and you offer no solutions."

[WCC]

“But most of you are rather stupidly ignoring that a serious problem still exists, and you offer no solutions."

But we are spending all our time fighting the evil that is gun control - just imagine the solutions we will come up with if we could rather spend our time finding solutions for crime instead. Double that up with the Government spending time and resources on combatting crime rather than the abysmal failure and cluster f’up that is gun control and we will have a winner for sure.

[ARM505]

Actually a surprising step forward in the debate, but still some obvious problems that'll encounter resistance.

- Most pro-firearm people are all about personal responsibility, maybe to a greater or lesser degree, but making your own choices always features. Hence the argument 'don't blame me when somebody else does something stupid/illegal with a gun - that wasn't on me, and shouldn't affect my rights'. This still applies.

- Likewise, with economic well being. With an ethos of personal responsibility, those people are often loathe to be viewed as welfare cases and make real effort not to be a burden. Efforts to 'help the poor be less poor' is basically taken to mean 'tax the wealthier more', and creates a handout mentality amongst some. That being said, I completely agree with the wealth-inequality problem, it *is* real, and has been painfully obvious to most - South Africa is a case in point, made even more critical by wealth often being aligned along racial lines.

I also take (very mild) exception to the last section, with the false dichotomy of 'But today you must choose' (ie. you must climb on this bus which I've defined or you'll be left behind, where in actual fact no such thrust currently exists).

Nonetheless, it makes for interesting talking points, many of which *have* actually been brought up by pro-firearms groups before, contrary to what he says.

[Pirate]

I also take (very mild) exception to the last section, with the false dichotomy of 'But today you must choose' (ie. you must climb on this bus which I've defined or you'll be left behind, where in actual fact no such thrust currently exists).

Nonetheless, it makes for interesting talking points, many of which *have* actually been brought up by pro-firearms groups before, contrary to what he says.

I couldn't agree more!