When it comes to gun violence, it’s tough to beat street gangs. They have made a sport of it.

But the spooky question has been, “How much gun violence is gang violence?” Thanks to a little triangulation on our part, we think it is amplissimum.

Major Take-aways

• Gangs likely constitute 85% of firearm homicides in the US.
• There is a close correlation between metropolitan gang-age homicides and estimates of gang populations by degrees of urbanization.

The Data Dearth

For all the great scorekeeping the FBI does, when it comes to murder, they have understandable gaps.

Every year the FBI reports on the circumstances that led to a homicide, in their Expanded Homicide Tables. There are two rows of data that list what police were absolutely certain were gang related murders.

Combined, those two rows of definitive gang homicides make a measly 4% of homicides. This is because police are only allowed to classify circumstances for which they are certain about the origins. This is why the category “unknown” is always the largest bucket for murders. Likewise, if people near the murder heard an argument before the gunshot, police can classify the circumstance as “argument,” though it could have been an argument between gang members, over turf, over drug distribution territory, et cetera.

The fact is that many of the argument and unknown circumstances are gang related. So, we tried to ballpark gang homicides by triangulation.

Crips and Bloods

A popular documentary details the historic feud between two street gangs, the Crips and the Bloods. The documentary asserts that in the 30-year war between these gangs, they have killed 15,000 of each other. For contrast’s sake, there were “only” 2,401 U.S. military deaths in the 20-year Afghanistan war. That makes the Crips and the Bloods about four times as deadly on a year-in, year-out basis.

[EDITOR NOTE: We reached out to the documentary’s director to learn the source of their estimate, but at the time of this writing they had not responded.]

This brings home the point that any homicide detective in a major metropolitan area will make: most of the homicides there are gang related in some fashion. They question remains, though – how many is that?

Triangulations

What we know from data is that:

• Street gangs are largely based in metropolitan areas.
• The primary age range for street gang activity is between 15 and 30.

Gang Penetration Rates and Firearm Homicide Rates

Thanks to two data sources – the Center for Disease Control mortality databases and the National Youth Gang Survey – we can compare firearm homicide stats with gang penetration stats and see if they align.

They do, at a 96% correlation level.

This alignment is important to the rest of the analysis. If the degree of street gang membership did not agree with the degree of firearm homicides, then there would be no possible cause and effect (confounding variables, though, might apply). But such a strong covariance (0.96 using R²) indicates that the two are not divorced. In other words, if gangs were not a primary source of firearm homicides, then areas where street gangs were rare would also have higher firearm homicide rates, which they don’t. ¹

Gang Membership Rates to Documentary Claim

In the year 2012, there were an estimated 850,000 street gang members in the United States. ² The estimates fluctuate from year to year, but they were at the same level in 1996. Despite a growing population, street gang participation remained fairly constant.

Wikipedia reports that combined, the Crips and the Bloods have about 55,000 members, or about 6.5% of all American gangsters. If the estimate provided by the documentary is anywhere near accurate, and assuming that the Crips and the Bloods are not staggeringly more homicidal than other street gangs, their share of gang homicides should be about the same as their share of gang memberships.

With 15,000 hits over 30 years, this equates to 500 per year for these two gangs (just between them… no telling how many innocent people or members of different gangs were victims). In the most recent year for which data was available, there were 9,063 firearm homicides. The Crips and Bloods count of 500 (average annual) is 5.5% of the firearm homicide rate, which is close to the estimated 6.5% membership rate among gangs.

Now, there are wild variables here. Not every gang member is a murderer, but some members make up for that deficit by killing repeatedly. We’ll accept these variations since we are only trying to ballpark an estimate.

When we take this analysis and apply it to the members of gangs other than the Crips and Bloods, and assume they are roughly as homicidal, the math works out that roughly 85% of firearm homicides are gang related.

It may or may not be fair to say that most gunshot wounds are gang related as well. Security camera footage of drive-by shootings where multiple people are injured would lead one to this belief. But, alas, there is no solid way we at the Gun Facts project have yet discovered to triangulate such, and the CDC’s WISQARS reporting system is not transparent on firearm injuries. But let’s say that the odds are high… as high as Snoop Dogg, who was a Crip… that most inflicted gunshot injuries are gang related.

What All These Numbing Numbers Really Mean

Even if this triangulation is fractionally off, it still shows that in terms of gun violence, the single largest contributing factor is street gangs. This is not news; criminologists have been noting it for years.

So why isn’t this problem the priority for politicians? Because they only care about poor people of color at election time. The rest of their term, while they are hobnobbing with wealthy donors and powerful politicos, they are content to let otherwise good but poor people get caught in gang crossfire.

UPDATE: What if there were no gang homicides

One astute reader asked this basic question. Since America has a much higher gang participation rate than other countries, and our gangs (as demonstrated by the Crips and Bloods saga) are much more homicidal than gangs elsewhere, what would happen if America could magically eliminate gang homicides.

A lot. Using U.N. crime data (because it is normalized across countries, for 2016 (the same year as the most recent CDC data is available for comparison sake), we see the U.S. homicide rate fall to be inline with other first-word, western industrialized nations.

UPDATE 2: 100X homicide rate for gang

One criminology paper ³ states that “gang homicide rates are estimated at up to 100 times that of the broader population.”

UPDATE 3: Gang Density and Homicides

We we alerted to an older study ⁴ that looked at homicide frequencies depending on how many active street gangs were in a 2-square mile area. There is a nearly 20X increase in homicides from having no gangs present to having upwards of 30.