There’s been a lot of arguing about gun violence (and the scary AR-15) so we’re taking a look at mortality data. Everyone seems to have an opinion about this subject but almost no one seems to review the data, including the people writing legislation. So it’s up to us to look into the numbers, and fortunately the CDC keeps a set of this data and it’s consistent with data from the FBI.
To make this simple we’re gong to look at the data from 2019, which is the most recent. The previous several years have similar numbers so using 2019 will give us a good feel for the numbers without burying ourselves in data.
Let’s start with this introduction from the CDC1. Here’s what they have to say, summarizing all homicides and suicides:
In the United States, more than seven people per hour die a violent death. In 2019, more than 19,100 people were victims of homicide and over 47,500 people died by suicide.
To help find ways to prevent violent deaths, we need to know the facts. The National Violent Death Reporting System (NVDRS) links information about the “who, when, where, and how” from data on violent deaths and provides insights about “why” the deaths occurred.
And already there’s something we need to double check. The folks at the CDC are doing their normal quality of writing, stating that 19,100 people were “victims of homicide.” (The exact number from their data set is 19,141.) But they’re not all “victims.”
They have a page for definitions2 and here are the entries for both homicide and legal intervention (that second one will explain one of the categories later):
Homicide – injuries inflicted by another person with intent to injure or kill, by any means. Excludes injuries due to legal intervention and operations of war. Justifiable homicide is not identified in WISQARS.
Legal Intervention – injuries inflicted by the police or other law-enforcing agents, including military on duty, in the course of arresting or attempting to arrest lawbreakers, suppressing disturbances, maintaining order, and other legal actions. Excludes injuries caused by civil insurrections.
So they don’t distinguish between homicides that are justified and ones that are not – it’s just one data point. From Cornell’s Legal Information Institute, a justifiable homicide is3:
The taking of a human life under circumstances of justification, as a matter of right, such as self-defense, or other causes set out in statute.
Which means there were not 19,100 “victims” because the CDC number includes self defense (and it turns out this is a fairly large fraction of the total).
Let’s look at their breakdown for firearm deaths. To do this we’re going into WISQARS4 (Web-based Injury Statistics Query and Reporting System), and these numbers come from the fatal reports database (https://wisqars.cdc.gov/fatal-reports):
All deaths involving firearms: 39,707
Unintentional: 486
Suicide: 23,941 (60% of the total)
Homicide: 14,414 (36%, including murder and justifiable homicide)
Legal intervention: 520
Undetermined: 346
We can break down the homicides further using FBI data5. The FBI lists 13,927 murders in 2019, of which 10,258 used firearms (of all types).
First, let’s look at all homicides (with and without firearms). Subtracting the FBI’s 13,927 total murders from the CDC’s 19,141 total homicides leaves 5,214 total justifiable homicides (most likely these are all or nearly all self defense).
How about homicides using firearms? Subtracting the FBI’s total of 10,258 firearm murders from the CDC’s total of 14,414 firearm homicides leaves 4,156 firearm justifiable homicides.
This means that of all deaths involving firearms in 2019, 25.8% (10,258 out of 39,707) were murders and 10.5% (4,156 out of 39,707) were justifiable homicide. Accidental deaths are actually rare (1.2%).
For perspective let’s look at some of the other big mortality issues from that same year.
All deaths involving motor vehicles: 39,368
Unintentional: 39,107
Suicide: 171
Homicide: 68
All deaths involving falls: 40,727
Unintentional: 39,443
Suicide: 1,183
Homicide: 12
All deaths involving poisoning: 75,795
Unintentional: 65,773
Suicide: 6,125
Homicide: 159
All deaths involving suffocation: 21,250
Unintentional: 7,076
Suicide: 13,563
Homicide: 455
So for starters homicides aren’t really common in these other categories. Also note that the total numbers in each category are all in the tens of thousands: firearms (39,707), motor vehicles (39,368), falls (40,727), poisoning (75,795), and suffocation (21,250).
These five categories represent 216,847 deaths from a 2019 total of 246,041 injury related deaths (about 88%). By category, they are the following percentages of all injury related deaths: firearms (16.1%), motor vehicles (16.0%), falls (16.6%), poisoning (30.8%), and suffocation (8.6%).
[It’s also interesting that accidental firearm deaths are very rare compared to other types of accidental deaths. There are only 486 out of 173,040 total accidental injury deaths in the CDC database (0.28%).]
But firearms are by a large margin the majority (14,414) of all homicides (19,141), which includes both murder and justifiable homicide.
For murders, firearms were 73.7% (10,258 out of 13,927)
For justifiable homicide, firearms were 79.7% (4,156 out of 5,214)
Most firearms deaths were actually suicides (23,941), accidents were rare (486) and legal interventions – the police - (520) were also not large in 2019.
Among all suicides, firearms (23,941) represent 50.4% of the total (47,511).
Now let’s get into the details of those 10,258 firearm murders. The FBI has data on the type of firearm for 6,977 of them (the categories are handguns, rifles, shotguns, other guns), and 364 of those (5.2%) used rifles (of all types combined). They don’t show more specific detail on the types within those categories.
[If we use the 5.2% number to estimate the total rifles for the cases where the type is not stated, we get another 171 rifles used in murders for an estimated total of 535 rifles, of all types, used in murders in 2019. Note that this is just my attempt to not understate the total number for rifles by ignoring the cases for which the FBI gives no type - this number could be anything between 171 and zero.]
Here’s an interesting comparison (again, for 2019). And at the big picture level, of all 13,927 murders:
Rifles were 2.6% (364) but maybe as much as 3.8% (535 by my estimate)
Blunt objects were 2.9% (397)
Hands, feet, etc. were 4.3% (600)
Knives were 10.6% (1,476)
And rifles were 0.15% (364 out of 246,041) of all injury related deaths.
Bare hands and blunt objects were used for about as many murders as rifles, and knives were used about four times as often as rifles. Also, recall from earlier that there were 455 homicides via suffocation.
So very few people are murdered by rifles, of all types, both as a percentage and as a total number. You are just as likely to be murdered via suffocation, blunt object, or bare hands - and much more likely to be murdered with a knife - as a rifle.
Okay, but what prompted us to look into this?
There has been a lot of talk lately (for a long time, really) about banning “assault weapons” and Senator Feinstein (D-CA) has a bill right now (S.736 - Assault Weapons Ban of 2021)6 to ban a long list of something like 200 specific models, including the often-mentioned AR-15. The weapons on this list are only a subset of all rifles, so could represent no more than a very small percentage of all murder weapons (some fraction of that 2.6%).
There’s always a very spirited discussion on this topic, and I don’t intend jumping into the debate over the impact of specific pieces of legislation on crime statistics. But even if Senator Feinstein’s bill became law, the impact on murders (up, down, or no change at all) would be indistinguishable from normal year-to-year variations:
2015 (9,143)
2016 (10,398)
2017 (11,014)
2018 (10,445)
2019 (10,258)
And quite honestly those few rifle murders could likely have been committed using a handgun or shotgun instead. Maybe Senator Feinstein’s legislation will make a handful of professional hit-men work a little harder, but they’re not going to surrender their long range sniper rifles just because of this bill (that’s a joke – I don’t think Agent 47 is bumping up the numbers here).
Anyone looking at the data, and interested in reducing the number of deaths caused by firearms, would not be focused on rifles in general, “assault weapons” as a subset, or the AR-15 specifically. There’s no reason to think the proposed ban on “assault weapons” will have any impact on the total number of homicides, or on injury related deaths in general.
So why all the sound and fury about an item that’s large, difficult to conceal, and represents only a small percentage of the issue these politicians are supposedly trying to address?
Original:
https://www.cdc.gov/violenceprevention/datasources/nvdrs/index.html
Archived copy:
https://web.archive.org/web/20211119010120/https://www.cdc.gov/violenceprevention/datasources/nvdrs/index.html
Original:
https://www.cdc.gov/injury/wisqars/fatal_help/definitions_fatal.html
Archived copy:
https://web.archive.org/web/20211002235054/https://www.cdc.gov/injury/wisqars/fatal_help/definitions_fatal.html
Original:
https://www.law.cornell.edu/wex/justifiable_homicide
Archived copy:
https://web.archive.org/web/20211118214121/https://www.law.cornell.edu/wex/justifiable_homicide
https://www.cdc.gov/injury/wisqars/index.html
Original:
https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-2019/tables/expanded-homicide-data-table-8.xls
Archived Copy:
https://web.archive.org/web/20211120184907/https://ucr.fbi.gov/crime-in-the-u.s/2019/crime-in-the-u.s.-2019/tables/expanded-homicide-data-table-8.xls
Original:
https://www.congress.gov/bill/117th-congress/senate-bill/736/text
Archived copy:
https://web.archive.org/web/20211122045039/https://www.congress.gov/bill/117th-congress/senate-bill/736/text
You're too logical for this discussion when it comes to brain dead leftist zombies. A stronger argument to those still capable of rational thought on the left is; "once guns are banned, only cops and criminals will have guns and it is a fine line separating both of them."
Because it is a big bad scary monster that politicians can create FEAR around to push a bad policy based on emotion instead on facts.