The Sandy Hook shootings were brutal; the gun control debate has reignited. I tried to make some sense of the debate by looking at the numbers involved, but it didn’t work out well. The numbers simply don’t provide a compelling case for gun control. If we want to argue for gun control, we need a different angle.
Let’s look at the numbers. In any given year, the chance of being murdered by firearm is the chance that you meet a murderer times the chance that the person actually has a gun to kill you with with (and manages to shoot you successfully). In the United States, the yearly rate of deaths by firearm due to homicide is 3.7 out of every 100,000 . The average rate of gun ownership is 88.8% (wow!) . With these numbers we can back-calculate a rough idea of the “craziness rate”: the average chance of encountering a crazy person who want to shoot you (and does so successfully). The resulting craziness rate is about 4.2 in 100,000.
(Yes, I accept that this is a really simple calculation that leaves out plenty of stuff including the facts that many murders aren’t totally random, that some people own many guns and that some murderers kill multiple people. It still gives us a rough idea of the size of the problem.)
For comparison, in Canada, a similar country with somewhat similar gun laws, the ownership rate is 30.8% , the death rate is 0.76 per 100,000 , and the resulting craziness rate is 2.5 out of every 100,000 – roughly two-thirds of the craziness rate in the U.S. Keeping the craziness rate in the U.S. unchanged, we would need to reduce guns to 20% of the current level to reduce the death rate to Canada’s level. Or we could work on the craziness rate and make it less likely that somebody out there wants to kill us.
Opponents of gun control will argue that more guns will actually reduce the craziness rate so much so to counteract the fact that there are more guns in the first place. Let’s test this. If we assume that all Americans are armed (100% gun ownership), we would need the craziness rate to drop to 0.76 per 100,000 to reach the mortality rate of Canada. That would be over an 80% reduction! It seems hard to believe that more guns can lower the rate by that much; and if they did lower it by that much, we need to ask if that world of fear is a world we’d like to live in.
So far, this is sounding like a good case for gun control, but a bit more context shatters the argument. It turns out that gun deaths just aren’t that common. For instance, driving fatalities in the U.S. are much higher at 12.3 per 100,000 . There are no big news stories about driving fatalities; it’s a number we seem willing to live with. The top five causes of death are heart disease, malignant neoplasms, chronic lower respiratory disease, cerebrovascular disease, and accidents. The first two have death rates over 280 per 100,000 and they alone account for just under half of all deaths . If all we care about is reducing the mortality rate, there are many causes of death that should frighten us much more than death by firearm.
But, alas, our brains work strangely – illogically – at times. In this case, we’re succumbing to the availability heuristic, our tendency to think that just because we hear about something a lot, it is happening a lot. It especially happens with dramatic events like airplane crashes or mass shootings, and our sensationalist news media has a lot to do with it. Compare the news coverage about Sandy Hook with the death of ten afghan girls who stepped on a mine just a few days later. Both stories are terrible tragedies for the people and communities. One story dominated the headlines for days; the other just flew by without much notice. The death rates for the two are quite similar  but we certainly don’t feel that way in our gut.
Speaking personally, the trouble is that I still do support significant tightening of gun regulations in the US. I just can’t find a rational, data-based argument for it. More precisely, I can’t find an argument for gun control consistent with utilitarianism (which roughly speaking seeks to minimize suffering and maximize happiness). Gun deaths just don’t affect the overall rates of suffering all that much.
If we can’t use cold, numerical calculus to argue for gun control, how can we argue for it? That’s another topic for another day, but we need to, because this it makes zero sense to me that this gun, a mildly handicapped, military-grade weapon, is in the hands of my neighbor… and in the hands of psychos who want to shoot up schools.
- http://www.cdc.gov/nchs/data/nvsr/nvsr61/nvsr61_07.pdf (Table 1)