The Only Thing That Stops a Bad Guy With a Gun is a Good Guy With Data

/* That which can be destroyed by the truth, should be. */

Every few months or so, the United States experiences an act of violence perpetrated on innocents by one or or more intensely troubled individuals.  For the next few weeks, there is public outcry from both sides of the political spectrum; the left rallies behind the banner of gun control, and the right reverts to some variant of the slogan “the only thing that stops a bad guy with a gun is a good guy with a gun.”  And then something else happens.  And everyone promptly forgets about gun control.

In this case, I am grateful for the public’s stunted attention span.

It makes absolutely no sense to base gun control policies on rare, emotionally affecting events. Numbers on mass shootings are a bit hard to track down- in no small part because the definition of “mass shooting” is somewhat nebulous- but the consensus seems to be less than 100 deaths per year on average.  If you only count incidents that capture the public’s attention, that number drops precipitously.

Number of deaths attributed to incidents of firearm homicide in which at least 4 people were murdered

These deaths are tragic, and we are right to collectively mourn the senseless death of innocents. But this is not a basis for a policy decision. Our desire to legislate our emotional reaction is a bug, not a feature, of humans.  It is a cognitive bias, not an argument. This question should be answered with data.


In 2010 there were 8,775 murders involving firearms- 6,009 of them (68.5%) from handguns, and another 1,939 (22.1%) with the firearm type unreported.  A reasonable estimate would be around 87% of firearm homicides committed with handguns. Firearm homicides accounted for 66.9% of all homicides in the United States in 2010.  Given a US population of 308 million in 2010, that makes the firearm homicide rate 3.2 per 100,000 in population.  A random US citizen has a 0.0032% chance of dying in a gun related homicide in a given year.

Suicides accounted for 19,392 firearm related deaths in 2010.  There were an additional 606 accidental deaths, along with 73,505 Americans hospitalized with non-fatal gunshot wounds (59,344 intentional, 14,161 accidental).

A relatively small study (56 fatal, 68 nonfatal injuries) found that more than 75% of guns used in suicide attempts and unintentional injuries of 0-19 year-olds were stored in the residence of the victim, a relative, or a friend.

In 2009 there were an estimated 310 million firearms in the United States (not including weapons on military bases)-114 million handguns, 110 million  rifles, and 86 million shotguns.  These numbers appear to be approximate, as there is no central gun registry.

According to the FBI Uniform Crime Statistics, in 2011 firearms were used in 68 percent of the nation’s murders, 41 percent of robberies, and 21 percent of aggravated assaults.


The firearms industry employs about 98,000 people domestically in the US. In 2012 the firearms and ammunition industry was responsible for as much as $31.8 billion in total economic activity in the country, according to a study done for the National Shooting Sports Foundation.

However, gun-related violence also has significant monetary costs, usually in the form of emergency medical care for gunshot wounds.  Some estimates place that cost as high as $100 billion annually.  The psychological cost associated with gun violence is also well-documented, though difficult to quantify.

$70 Billion is a small but non-trivial portion of the entire US GDP.


The US is by far the worst of any industrialized nation in both firearm homicides (3.2/100,000) and total homicides (4.75/100,000).  The only countries worse than the US in firearm homicides were Guatemala, Columbia, Mexico, Paraguay, Zimbabwe, and Costa Rica- let’s just call them PWDLAIC.  The developed countries with some of the strongest gun laws- Germany, England, and Australia- not surprisingly also have the lowest gun homicide rates.  Less obvious, they also have some of the lowest overall homicide rates.

The United States has 3 times as many guns per capita as Germany, 6 times as many as Australia, and 15 times the number of the United Kingdom.

Australia in particular makes an interesting case study for the US.  Australia had a seminal moment- the Port Arthur massacre- in which it’s gun control politics shifted abruptly.  In the space of a few months, the conservative administration had put in sweeping gun reform, extensive gun buy-back programs, and highly restrictive new regulations.  Since the introduction of the legislation in 1996, Australia has seen a 54% reduction in gun related homicides, while total homicides have fallen by 28%.  They also have not had a single mass shooting.  In the same time period, the US has seen a 4% increase in gun related homicides, while seeing a 16% decrease in total homicides. Both countries have seen a decrease in homicides, but Australia has seen nearly twice the decrease of the US. This is suggestive, though not conclusive, that a reduction in firearm homicide had a significant impact on total homicides in Australia- i.e., people did not simply find other ways of killing each other.

It is important to note, however, that there are major differences between the United States and these countries.  In particular, the income inequality in the United States can be argued to have a real and measurable affect on crime rates.  A larger class of people who struggle to survive generally leads to more crime.  Additionally, some academics have challenged the notion that restricting gun ownership is responsible for Australia’s decrease in murders, citing meta-trends of reduced homicide that extend to before the 1996 gun reform.  Australia also had a much lower initial gun ownership rate than the US.  There are also ample counter-factual examples of countries with high gun ownership and low murder rates, such as Switzerland, and low gun ownership but high murder rates, such as Honduras.  Based on data from The Guardian, there is actually a slight negative correlation between guns per capita and overall homicide rate:

Homicide rates in all countries with data available

Homicide rates in all countries with data available

A ready explanation comes to mind: underdeveloped countries, where rule of law tends to be a foreign concept, cannot afford firearms.  There are relatively few places economically stable enough to afford lots of firearms, but simultaneously lacking the political stability to effectively enforce rule of law. It would be reasonable to expect these countries to have much higher murder rates than developed countries. And in fact, there is a clear correlation between gun ownership and murders in developed countries, excluding South Africa (accounting for about 950 million people):

Homicide rates in countries meeting the CIA Factbook definition of "Developed," excluding South Africa

Homicide rates in countries meeting the CIA Factbook definition of “Developed,” excluding South Africa

…But look again.  If you exclude USA, you’re back to basically no correlation:

Homicide rates in developed countries excluding South Africa and USA.

Homicide rates in developed countries excluding South Africa and USA.

Now obviously we can’t just go around excluding data points willy-nilly.  But if one data point  is the difference between strong correlation and no correlation, your brain should be screaming “sample size!”

So far, every nation has been given equal weight in this regression- so Iceland, with it’s 300,000 people, counts for as much as Japan, with it’s 120,000,000.  However, even if we get rid of all countries with populations less than 10 million, we see basically no change:

Homicide rates in developed countries, excluding countries with populations less than 10,000,000

Homicide rates in developed countries, excluding countries with populations less than 10,000,000

This last chart contains data on Australia, Belgium, Canada, England, France, Germany, Greece, Italy, Japan, Netherlands, Portugal, Spain, and Turkey- a fairly good approximation for “USA-like” countries.

The available data does not suggest a strong correlation between gun ownership and homicide rate.


In 2010, 880 total homicide victims were reported killed during the commission of a robbery, burglary, or larceny.  Adjusting for the number of unknown felony types, roughly 15% of homicides happened in the traditional self-defense type scenario. Of those, 663 (75%) involved firearms, 499 of which were handguns, and 122 of which were unknown gun types.  Adjusting for unknown, roughly 92% of robbery, burglary, and larceny homicides involving firearms used a handgun.

The same year, there were 278 justifiable homicides. 232 (83%) of those incidents involved firearms, 170 of which were handguns.  For every person that killed an assailant in self defense, 38 were killed in the commission of a crime.

There is highly conflicting data about the effectiveness of firearms as a deterrent to crime, both directly and indirectly.  In particular, accusations of over-reporting defensive gun usage and data messaging have called into question early studies that claimed to suggest a negative correlation between gun ownership and crime.


The Second Amendment to the United States Constitution states:

A well regulated militia being necessary to the security of a free state, the right of the people to keep and bear arms shall not be infringed.

The original intention of the right to keep and bear arms is clearly stated as being for use in a well regulated militia.  This seems rather silly today, given that the US spends almost as much as the next 14 nations combined on defense.  We have no need of a militia, nor would a militia stand any reasonable chance should the military decide to turn on us.  This rule made a lot of sense in 18th century America.  It makes very little sense in 2013 America.

However, in District of Columbia v. Heller, the Supreme Court has held (5-4) that:

The Second Amendment protects an individual right to possess a firearm unconnected with service in a militia, and to use that arm for traditionally lawful purposes, such as self-defense within the home.

The Amendment’s prefatory clause announces a purpose, but does not limit or expand the scope of the second part, the operative clause. The operative clause’s text and history demonstrate that it connotes an individual right to keep and bear arms

The handgun ban and the trigger-lock requirement (as applied to self-defense) violate the Second Amendment. The District’s total ban on handgun possession in the home amounts to a prohibition on an entire class of “arms” that Americans overwhelmingly choose for the lawful purpose of self-defense. Under any of the standards of scrutiny the Court has applied to enumerated constitutional rights, this prohibition – in the place where the importance of the lawful defense of self, family, and property is most acute – would fail constitutional muster. Similarly, the requirement that any lawful firearm in the home be disassembled or bound by a trigger lock makes it impossible for citizens to use arms for the core lawful purpose of self-defense and is hence unconstitutional

Basically, they said that the second amendment protects an individual’s right to bear arms, even if the original reason given for it no longer remains relevant, and that handguns cannot reasonably be banned from sale to law abiding citizens intending to use them for lawful purposes.

Overturning this doctrine would require an ideological shift on the court (e.g. a new liberal justice), or a new constitutional amendment nullifying or modifying the second amendment.  It is unlikely that the court’s ruling will be overturned, as the American judicial system highly values precedent.  It is typically only in the most extreme of civil rights cases- e.g. Brown V. Board of Education- that the court will overturn precedent, and even then it typically requires a shift in the public zeitgeist.


Even supposing handguns were banned, it’s not clear how to implement a disarmament on the scale of 300 million weapons.  Again we may point to Australia and it’s compulsory buyback program in the wake of Port Arthur. This program was successful, but also very expensive.  A buyback program to reclaim just one third of US guns, for example, might cost upwards of $110 Billion.

America also already has a large illegal gun ownership problem. Aggressive enforcement of gun control laws could begin to pare this number down, but this would be a gradual solution. We could try something tricky, however: increasing the buyback amount for firearms would incentivize more people to sell back their guns. It would also artificially raise the price of illegal black market firearms. If a pistol sells for $500 on the black market, but you can turn around and sell it to the government (no questions asked) for $1000, then that’s an arbitrage opportunity. This would drive up the price of illegal firearms to a minimum price of whatever the government was willing to pay. This in turn would price a lot of people out of the illegal gun market, reducing demand. Less demand means less illegal gun smuggling.  Of course, it’s all moot, since the Supreme Court has ruled that we can’t ban handguns, even if we were to decide it was the right thing to do.


We are mainly fighting against intentional gun violence, not accidents.  Accidents make up a relatively small fraction of gun violence.  Gun violence, however, makes up a huge proportion of the total violence in the United States- and handguns in particular, being the easiest to conceal and the easiest to wield, account for a huge proportion of all gun violence.  We have excellent data to suggest that gun-related crimes will decrease with stricter gun controls.

Based purely on the available statistics on other countries, we do not have conclusive evidence that the overall number of violent crimes is correlated with stricter gun control.  However, America has by far the highest number of guns per capita of any nation, and the highest murder rate of any developed nation.  Given how large a role guns play in the violence prevalent in America, it seems likely that in our case,  stricter gun controls would lead to a reduction in violence.  It remains possible that what America has is fundamentally a cultural problem or an economic problem, rather than a gun problem- but even if so, our obsession with guns that outstrips everyone else by a hefty multiplier is a best a manifestation of this cultural and economic problem, and at worst a direct cause.

Guns make it easier to kill people.  The question is whether you want to live in a world where it’s easier to kill people, or harder.  If it’s easier for everyone to kill each other, criminals may be incentivized to avoid confrontation (though we do not have good evidence to suggest this either).  If it’s harder for everyone to kill each other, then when conflicts do happen, fewer people are going to die.  Moreover, the data suggests that when guns are used, criminals use them a whole lot more effectively that non-criminals.

Incentives work.  We don’t need to outlaw guns to stop people from buying them- If you make it harder to obtain guns, fewer people will have them.  There are a myriad ways to accomplish this, from taxing firearm purchases to closing loopholes and better enforcement of existing laws to more stringent background and psychological screenings.  However, the Supreme Court has made clear that denying access to firearms- including making it difficult enough to be prohibitive- is unconstitutional.

The data is frustratingly inconclusive.  However, based on all the above, it seems likely that stricter gun controls would reduce the overall homicide rate in the US.  It also seems likely that stricter gun controls would be a significant net positive on our economy.

Tagged , ,

3 thoughts on “The Only Thing That Stops a Bad Guy With a Gun is a Good Guy With Data

  1. Paul Stewart II says:

    It is hilarious to see you parsing data to suit the intended objective. And it is further so to see you ignoring good data. America has far more gun deaths per capita by a margin of 10 times to infinite times compared to all developed countries. Why are you mixing in the third world? Oh, and your complaint about frustrating inconclusive date… get some data. I am sure you know that Congress can’t collect data. It is in part outlawed and more so politically a death sentence. You collect data; speak against guns or in any way reduce this frustrating inconclusiveness of date? Then you get primaried and turfed out of politics. Or whatever job you are in be it even magazine writer…

  2. Brotherwayne says:

    > In the same time period, the US has seen a 4% increase in gun related homicides, while seeing a 16% decrease in total homicides.

    You really should source this. I’d like to see where that data is coming from.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

%d bloggers like this: