Sunday, December 16, 2012

An Inconvenient Truth: Homicide Edition

Now that we've had another mass murder, everyone is arguing about gun control law. At the risk of making the discussion more rational, let's study some correlations.  Keep in mind that correlation is NOT causation - just because two factors happen together does not mean one caused the other. Both might be caused by a third, as yet unidentified, factor.

For instance, it is the case that child abuse has been dropping for five straight years. It is also the case that gun ownership has been skyrocketing the last five years. So, which one is the cause, which one is the effect? Have parents become so frightened of their own children that they no longer dare neglect them, and now choose to arm themselves against the little tykes? Or have children gotten a hold of newly acquired household guns and threatened their parents with death if the latter ever again breathe a word about eating spinach?

Well, probably neither. Probably, a third (or fourth) factor is at play - perhaps older parents are less likely to neglect their children AND more likely to buy a gun to protect themselves and their families than young parents are. Or maybe it's the phase of the moon. Who knows?

In fact, we don't know that the two trends are connected at all.
All we know is that both trends are real.

With that in mind, look at this Wikipedia world map of homicide rates, showing the murder rate per 100,000 inhabitants by any cause (not just guns), based on UN Office on Drugs and Crime 2012 data:

 Murder rate per 100,000 inhabitants most recent year.
   2-5
   5-10
   10-20
   >20




Notice that liberals are always comparing the murder rate in the United States to the murder rate in Europe, but never to the murder rate in Cuba or Russia - both nations with strict gun control laws. Why? Well, both Cuba and Russia's murder rates are much higher than the murder rate in most of Europe. If we want to lower homicides, we should look at nations with lower homicide rates, right?

Well, maybe.
Or maybe not.

Let's take a look at another map, based on data from the CIA's 2009 World Factbook, comparing global median age ranges in various countries. The median age is the age which half the country is above, and the other half is below:


Notice how closely median age correlates to homicide rates. In South America and Africa, the correlation is darned near 1:1 - the younger the population, the more likely it is to be homicidal. Within the United States, FBI statistics bear this out - murders are committed primarily by males between the ages of 17 and 35, tailing off rapidly above the age of 50. So, it is not too surprising to find that countries whose median age is between 17 and 35 will have higher murder rates than those which aren't.

We also see from the map, that this correlation between youth and homicide is not perfect. Lithuania has a very old population (40+), but a relatively high homicide rate. The three Scandinavian countries are all old (median age is 40+), but have very different homicide rates. Similarly, Russia, America, China, Ukraine, Poland and Australia have similar median ages, but quite different homicide rates.

But, overall, the median age correlates well enough to the homicide rate for us to make one relatively safe conclusion: if we really want to compare apples to apples, we shouldn't be comparing homicide rates in the United States with homicide rates in the much older nations of Europe, but rather to the more age-equivalent societies of Russia, China, France, Ukraine and Australia.

When we compare these age equivalent countries, we can now see rather a different story. Russia has the highest homicide rate (10.2), followed by the Ukraine (5.2), the United States (4.2), France (1.1), Poland (1.1), Australia (1.0) and China (1.0).

What could account for those differences? It can't be gun laws or gun ownership.  Here is the list of gun ownership rates.

If you click the link, you will notice that countries with a lot of young people, and therefore extremely high homicide rates, also tend to have the lowest number of guns per person. That is, the lack of guns in a young population doesn't appear to inhibit homicide rates at all.

But we care about populations with median ages between 35-40. Below is a short version of the per capita gun ownership list for countries with a median age comparable to the United States. The number of guns per 100 residents is listed:

  • 88.8 - United States (#1 on the list) 
  • 31.2 - France 
  • 15.0 - Australia
  •   8.9 - Russia 
  •   6.6 - Ukraine 
  •   4.9 - China 
  •   1.3 - Poland

Yeah, not much correlation there. Ukraine and Russia have higher homicide rates even though both have only one-tenth the guns the US has. France has three times the guns Russia has, Australia has twice the guns, but Russia easily outpaces the homicide rate in both those countries as well. The Ukraine is in the same situation - fewer guns than similar countries, but a lot more murders.

If gun ownership correlated to intentional homicide rates, the United States should top the list of homicidal countries. Instead, it is only about half-way down.

If we were to drill down into American statistics, it gets even more disturbing. Men make up only 50% of the population, but account for 90% of the murders. Blacks males make up less than 6% of the population, but they account for 53% of the homicides in the United States and 50% of the homicide victims. Most murder victims and offenders know each other, either through friendship or family relationship. And the majority of murders happen during or after an argument.

In short, if young black men who knew each other were to stop getting into fights, America's homicide rate could be cut by half.

So, what should we do? While we arguably have a problem with homicidal violence, the availability of guns doesn't appear to have much to do with it.

Apparently, if we really wish to stop homicidal violence in America, we would have to criminalize being male, especially being a young black male. We would also have to criminalize the establishment of interpersonal relationships, either via friendship or family. And having personal arguments would have to be categorically outlawed.

If we passed just those simple laws - jail young men, especially young black men, outlaw families and friends, and incarcerate anyone having an argument - homicidal violence should disappear. Additional gun laws don't even have to be considered.

See how easy it is?



6 comments:

  1. Wonderful comparison. I would have liked you to have included the statistics on the number and break down of people incarcerated on a per country basis. This is an area of personal interest as in the course of my employ I am required to visit all of these facilites in my state.
    I do not have those statistics or I would provide them, but I have in the past been told some of these indicators on a percentile basis of the population and it was astounding for the good ole USA.
    By rout of observance it is my estimate that the lack of fathers in the household and the devaluation of life that are the causes that have percipitated our current problems.
    Of course we can not point out the obvious drastic differences based on race of the prison population or that would be..... the truth!

    ReplyDelete
  2. The incarceration rate by country, with map, can be found here.

    We do have the highest incarceration rate in the world, but incarceration rate doesn't correspond to homicide rate. Africa and South America both have much higher homicide rates, but much lower incarceration rates.

    It is interesting to note that something like 40% of prisoners are black men. The rate of incarceration has skyrocketed since 1980.

    More young black men go to prison then get employed. But the majority of these jail terms are for non-violent drug offenses.

    So, even though we do incarcerate young black men preferentially, our homicide rate still places us halfway up the list of most homicidal countries.

    ReplyDelete
  3. In the U.S. this type of mass killing did not happen often before 1970... counted on single hand. Since then we have had 61 of them.

    I do not think that our gun laws account for the increase. Like you say, correlation does not equal causation but I have a hunch that legalizing the murder of children in the womb has not made our country a safer more pleasant place to live.

    Doug

    ReplyDelete
  4. There's a pretty complete listing of school shootings in the United States since the 1700s here.

    A worldwide list can be found here.

    Prior to about 1950, these kinds of things were one-off killings of teachers by spurned would-be lovers or killings of administrators by disgruntled employees.

    As you say, Doug, the idea of mass slaughter on a grand scale seems to be a pretty recent phenomenon. The perps are almost NEVER young black men. Instead, they are generally young white or young Asian men, the two groups of men LEAST prone to interpersonal violence.

    I DID try to correlate it to worldwide stats on alcohol consumption and injectable drug use. I couldn't see any correlations between those two things and homicide.

    Given the typical early deaths of men in Russia and the Ukraine, and the lack of drug correlation, I strongly suspect that the increase in gun violence among young men can be attributed to the loss of fathers in the household.

    Unfortunately, I can't find any stats on world-wide family breakup, so I can't support that thesis with any statistics.

    ReplyDelete
  5. My wife asked a good question, how many people will now start homeschooling their children now that we know public / private schools are not safe?
    What are the statistics in the areas where school shootings have happend in the past? Have parents started homeschooling, is there an increase?

    ReplyDelete
  6. If they chose to homeschool based on this event, they would not be paying close attention to the evidence.

    According to the National Center for Education Statistics, the rates of ALL crimes, including homicide, at schools has been dropping for the last 20 years.

    ReplyDelete