This is a continuation of my previous post on a discussion on guns and violence in the United States. Read Part 1 here. As I mentioned there, I used to be for gun control legislation. A lot of this is the data that convinced me it was not a viable solution to the problem of violence.
This second part will be more about crime. Because they have no stake in influencing public policy (making them as unbiased as possible), I will be using primarily (but not only) data from the FBI’s Universal Crime Reporting system, the CDC, and the Bureau of Justice Statistics.
Because both organizations do have a stake in policy, I will not be including anything from either the NRA or the Brady Campaign. There will be no sarcasm, scare tactics, or hyperbole. These are things I believe are having a negative effect on the ability of Americans to have an honest political discussion in general.
I will also not be comparing violent crime in the United States to that of any other country. There are too many factors involved in the causes of violent crime to be able to provide an honest comparison.
As the final piece to this boilerplate, what’s actually being discussed has to be determined. Using the sources I listed above, we run into a sort of reporting bias. It isn’t strictly so, but the data on crime is aggregated from crimes that were reported. Historically, certain types of crimes (non-fatal and sexual assaults, for example), aren’t always reported and so there isn’t much data on those. The BJS tries to capture that information using the National Crime Victimization Survey, designed to find out about crimes never reported to police. That said, the CDC, FBI, and BJS are still the most complete sources available for unbiased crime and mortality information.
So what are we talking about here? Well, most of the time, people are focusing on violent crime deaths involving firearms, so that’s what I’ll focus this discussion on. Murders and non-negligent manslaughter are a good place to start because they generally aren’t subject to quite the same reporting bias as other assaults. I’m going to set aside negligent deaths because those aren’t statistically significant in either direction and suicide because that merits its own discussion entirely.
The first thing to look at here is the FBI overview of crime between 1991-2010, found here.
1991 happened to be the year of the highest volume (not just rate, but total volume) of murders and non-negligent manslaughter (referred to just as murder for the rest of the post for ease) in the history of the United States. For the breakdown, here is the breakdown of firearm death by intent (1991-2001) from the BJS. As mentioned, we’ll set aside the suicide and undetermined categories for right now since those categories aren’t counted in the FBI numbers.
Of those 24,703 murders in 1991, we’ve got 19,063 of them as a result of people intending to do harm with a firearm that resulted in fatality.
Now start looking at those other years from the FBI table. The total volume of murder continues a downward trend, as does the overall volume of violent crime. Note that I’m not taking mention of the rate, (number of crimes per 100,000 people) as the population isn’t static. Yes, the murder rate dropped by just over half, from 9.8 per 100,000 to 4.8 per 100,000 but relying solely on that isn’t quite accurate since the population grew so, even if the volume didn’t change, population changes would affect the rate.
I feel like looking at the raw numbers is more useful and intellectually honest.
In 1994, President Bill Clinton signed into law what became known as the Assault Weapons Ban, as I mentioned in the last post. By this point, as we can see, the incidence of murder had already begun to decline.
Here are some more tables to consider. First is the breakdown of murder by weapon type for 2000-2004, 2002-2006 and from 2007-2011 from the UCR. There isn’t a table that goes more than five years in a row, so there are two here. Data older than 2000 isn’t in a linkable table format.
I went back to 2000 in order to illustrate three things.
- Overall, murder rates have been declining steadily since 1991.
- There were periods during the Assault Weapons Ban where the incidence of murders rose slightly.
- Rifles make up a small portion of murders, generally somewhere between 3-5%. Assault weapons are a subcategory of rifles.
Looking at this data, it becomes obvious that more people killed with “personal weapons” (fists, feet, etc.), blunt objects, and cutting weapons than with any rifle. This led me to one of the big questions in changing my mind on gun control: why is it that semi-automatic rifles that only look like military weapons are being particularly demonized, despite being a subset of a group of weapons used relatively infrequently in murders?
So we see that murders have been decreasing. Not just the rate because, as mentioned before, the rate is affected by a non-static population. We’re looking at the total volume. This must mean that there are fewer guns, right? You’d be surprised.
Since the United States has no national register, there is no reliable way to tell how many firearms are privately owned. What we can look at to get an idea is the data from the FBI on background checks called in by licensed firearms dealers. Here is a table that contains NICS checks from Nov 30, 1998 to April 30, 2011. This is the most current information I could find. Here is a link to everything you could ever want to know about the NICS.
It seems like there are about 1.1-1.2 million NICS denials annually, on average. Not every NICS approval means a purchase is being made and not every purchase requires a background check, but we can get a pretty good idea of how many sales happen annually. What we can infer from this is that the number of legally owned private firearms is increasing while the overall instance of murders is decreasing. Of course, this isn’t to say that there is a causal relationship. What it does say is that the notion that more guns necessarily equals more crime is not true.
A study conducted for the Department of Justice after the sunset of the Assault Weapons Ban in 2004 finds that the results of the ban were inconclusive. The study, in its entirety, can be read here. The study acknowledges that what were defined as assault weapons by the law were not heavily used in violent crimes before the ban and that the impact of low-capacity magazines could not be adequately measured.
There’s a lot of time, money, and effort being thrown at this, but why? Is it actually a matter of public safety, or is it just that certain people really don’t like guns?
I’ll be looking into that a little in the next post. Read part 3.
