Pew recently reported on how the homicide rate in the US has been cut in half over the past twenty years, even as gun ownership has greatly expanded in the US. The US public, meanwhile, thinks the murder rate is increasing.
What Pew didn’t mention is is how remarkably low the homicide rate has fallen in some US states. Pew reported FBI murder data up through 2013, but in more recent FBI homicide data, total homicides fell again in 2014, while the homicide rates in some states reached historic lows.
In New Hampshire, the homicide rate in 2014 was 0.9 homicides per 100,000, making New Hampshire in 2014 one of the safest places (in terms of homicide) on planet earth.
Several other states came in with remarkably low homicide rates, as well, with Minnesota reporting a rate of 1.6, Idaho at 2.0, and Iowa at 1.9 (rates measured in events per 100,000).
US States vs. Canadian Provinces
Clearly, these places rank among the US states with the lowest homicide rates, but for a broader context, let’s look at homicide rates compared among both US states and Canadian provinces:
Indeed, the northern United States in general tends to have quite low homicide rates in a global context. (There are problems with comparing across national boundaries. For more on that see below.)
Within North America, the jurisdictions with the lowest homicide rates include all of New England, the northern plains states of the US, and the Pacific Northwest. Most of Canada reports low rates as well, with the exception of the rural north, where Nunavut territory has the worst homicide rate in both the US and Canada.
For a more detailed look, let’s see each state, province or territory compared:
The blue states are USA states, and the red ones are Canadian provinces or territories. Most Canadians live in either Ontario or Quebec or British Columbia, so we can see right away what’s driving the overall homicide rate in Canada.
And yet, we should also note that nine US states that have homicide rates that are either lower than or the same as (statistically speaking) British Columbia. Manitoba has a murder rate that is statistically the same as, or higher than, 23 states.
So, does the United States have a high homicide rate? Well, some parts of the country, including Maryland, South Carolina, Louisiana and a few other states certainly report relatively high homicide rates when compared to many high- and middle-income countries. And some of those states also have lax gun laws. This should lead us to evaluate what is it that leads to these regional issues with homicide. Given that these are regional issues, and clearly not national ones, however, its difficult to see why any national mandates would be appropriate.
Additionally, If one can point to Canada or the UK or Luxembourg and immediately conclude that gun control in those jurisdictions produced their low homicide rates, why can one not also point to New Hampshire or Oregon or Utah and conclude that their lack of gun control is the reason for their low homicide rates?
This has yet to be explained by gun-control advocates.
Fewer Guns, Fewer Homicides?
When we look at the similarities in homicide rates between some American states and some Canadian provinces, the question we quickly find ourselves asking is “what is it about these areas that make for such low homicide rates?
Well, as the anti-gun Brady Campaign inadvertently makes clear, there is little apparent connection between homicide rates and state-level gun laws. States with more “gun freedom” get lower scores. For example, New Hampshire, Vermont, Idaho, and Oregon, all of which have among the lowest homicide rates in North America, get Brady Campaign scores of D-, F, F, and D+, respectively. Almost comically, Maryland and Illinois, which have homicide rates that are much higher, get scores of A- and B, respectively. But to be fair, there are other examples that show a lack of correlation altogether. New York, which has a relatively low homicide rate, gets a score of A-, while South Carolina, which has a very high homicide rate, is scored at F.
Canada, meanwhile, does not have the same level of diversity in gun laws that the USA has among states.
Canadian gun laws are dominated by national law which mandates restrictions on certain weapons nationwide, and there is a national gun registry centralized under the RCMP. All handguns are subject to registration. Nevertheless, at the provincial level, we might note that Quebec is traditionally among the most anti-gun in Canada while New Hampshire is among the least restrictive states when it comes to guns. In 2014, the two jurisdictions had identical homicide rates.
Yet, given the significant legal differences between US and Canadian law, we can’t simply point to gun laws as the reason for similar homicide rates between Canadian provinces and US states.
Other issues might include urbanization, ethnic conflict, median age, and weather.
Not surprisingly, many of the lowest-homicide states and provinces have high median ages. However, in the US, low-homicide states like Idaho, Utah, Colorado, and others have relatively low median ages. And median age is often affected by foreign-born populations.
But maybe if we could just control for all those factors, we could know the effect of guns, right?
The problem is, there’s no agreement as to whether or not a variety of these other factors drive homicide rates up or down. Urbanization, for example, is assumed by many to be a driver of more homicide. But, according to numerous studies, urbanization does not, in fact, drive homicide rates up. See here and here.
More Guns, Fewer Robberies
Other assumptions cause problems as well. For example, in a recent Vox article, author German Lopez cited research claiming that if we control for the robbery rate, gun control can be shown to reduce homicides. The reasoning being that robbery is a given, but fewer robberies will turn lethal if there is gun control. The problem here is this theory assumes that robbery is an independent variable when it really is a dependent variable (dependent on gun ownership by non-criminals).
It makes just as must sense to assume that the robbery rate is dependent on the potential cost of attempting a robbery. In states where gun ownership is very common, robbery is riskier to the robber, and thus — ceteris paribus — the robbery rate is lower. With fewer robberies, there are — naturally — fewer lethal robberies.
Lopez’s theory theory builds on research (also cited in the Vox article) from the late 1990s that further attempted to claim that Americans are more lethal than other countries because of guns. The study claims to show that there are not notable differences between crime levels among countries, but that crime is more lethal in the US because there are more guns. The book-length study, based on this journal article, treats the US as a single legal and demographic entity.
But the biggest problem is that, again, it treats violent crime (such as robbery) as an independent variable when it is really a dependent variable that is shaped, among other things, by the overall rate of gun ownership.
Moreover, claims such as these can’t explain how gun ownership and conceal-carry have expanded in the US while homicide rates have fallen. Lopez writes:
The prevalence of guns can cause petty arguments and conflicts to escalate into deadly encounters. People of every country get into arguments and fights with friends, family, and peers. But in the US, it’s much more likely that someone will get angry at an argument, pull out a gun, and kill someone.
And yet guns are more prevalent now than they were twenty years ago. And there is less homicide now than before.
Lopez would likely answer that this can explained by less alcohol consumption, lower poverty rates, or other factors.
In that case then, we’re left asking why the insistence on gun control? If less alcohol consumption and poverty are the driving issues here, why is it assumed that gun control is the most pressing “solution”? If gun-control advocates were really concerned about reducing mortality, they’d take a closer look at prescription drugs.
The answer, of course, is that it’s easier to simply impose gun control and call it a day (whether it works or not) than it is to address more central issues like mental illness, addiction, ethnic strife, and poverty. If the homicide rate goes up after gun control is imposed simply impose more gun control and then declare victory (as they do in the UK).
A Note on Data Collection and Statistics
First of all, its important to note, yet again, the manufactured and arbitrary distinction between “developed” countries and everyone else. Lopez invokes this distinction when he says:
Vox’s charts look at the correlation between gun ownership and gun violence in developed countries: It helps weed out the many, many social and economic factors involved if you compare the US with, for instance, Honduras — a nation mired by poverty and weak government institutions.
Note the bizarre and dishonest false dichotomy. You can compare the US to it’s “peers” or you can compare it to countries like Honduras.
I’ve already shown here why this is a nonsensical position: it cherry picks which countries are peers and ignores that fact that the US in many ways has more in common with other countries in the Americas than with the countries of Europe. There are numerous countries with “high” development indices (according to the UN) with restrictive gun control and high homicide rates. Lopez insists on restricting analysis to a handful of Western European countries, but no objective standard for this is ever given.
Secondly, there are issues with the actual collection of data. Generally, there are two ways of collecting homicide data. We can collect it from the law enforcement agencies, and we can collect the “mortality” data from medical examiners and other medical personnel.
Much of the time, however, what counts as a homicide, can be rather subjective. In fact, as The Los Angeles Times noted back in 2007, Japanese officials are often biased against autopsies and tend to minimize the number of deaths that are declared homicides for purposes of padding the percentages of “solved” cases:
Police discourage autopsies that might reveal a higher homicide rate in their jurisdiction, and pressure doctors to attribute unnatural deaths to health reasons, usually heart failure, the group alleges. Odds are, it says, that people are getting away with murder in Japan, a country that officially claims one of the lowest per capita homicide rates in the world.
“You can commit a perfect murder in Japan because the body is not likely to be examined,” says Hiromasa Saikawa, a former member of the Tokyo Metropolitan Police security and intelligence division. He says senior police officers are “obsessed with statistics because that’s how you get promotions,” and strive to reduce the number of criminal cases as much as possible to keep their almost perfect solution rate. Japan’s annual police report says its officers made arrests in 96.6% of the country’s 1,392 homicides in 2005.
But Saikawa, who says he became disillusioned by “fishy” police practices and in 1997 left the force in disgust after 30 years, claims that police try to avoid adding homicides to their caseload unless the identity of the killer is obvious.
“All the police care about is how they look to people; it’s all PR to show that their capabilities are high,” Saikawa says. “Without autopsies they can keep their percentage [of solved cases] high. It’s all about numbers.”
(This article also details some of the corruption behind Japan’s highly suspect 99% conviction rate.)
Moreover, one is likely to find suspiciously low homicide rates in many authoritarian countries in Asia, such as China which is clearly manipulating official data on a variety of issues.
Even in Canada, there are issues around reporting and compiling homicide data (although there is no indication of deliberate deception.) This document notes that there are persistent differences between the mortality data and the data from law enforcement agencies.
In the US as well, it is important to remember that the mortality data put out by the Centers for Disease Control includes justifiable homicide simply as “homicide.” In other words, if a violent maniac breaks into your house, and you shoot him dead, that will be included in the CDC’s official “homicide rate.”
In practice, it is likely that the US is far more liberal in its counting of homicides than numerous countries. For example, infanticide in the US is generally counted as homicide, although that is not the case in some other jurisdictions. The UNODC reports:
Nonetheless, the challenges of cross-national comparability are considerable. National legal systems may have different thresholds for categorising a death as intentional homicide.Whilst intentional homicide usually requires that the perpetrator purposefully intends to cause the death or serious injury of a victim, in some countries a death that occursin the act or attempted act of another serious crime may also qualify as ‘intentional’ homicide or murder. Infanticide, assault leading to death and killings carried out by law enforcement officers (acting legitimately in the line of duty or not) all may or may not be included in police recorded statistics.
In addition,differences in police recording practices such as differences in counting units (offences, suspects or cases), whether or not attempted homicide or non intentional homicides are included in published figures,and the point in the investigation at which a suspicious death is classified as homicide all vary as between countries.
Map and graph by Ryan McMaken
