Firearms in the United States: A Hoax?

Dec. 2, 2015: 14 people were killed and 21 were injured in a shooting in San Bernardino, California. It’s the deadliest shooting since 2012, when 20 children and eight teachers were killed in an elementary school in Connecticut. After this new shooting, Barack Obama declared that it was necessary to strengthen arms control to reduce the chances that a similar event happens again.

The United States is, in effect, one of the developed countries with the most arms per inhabitant. It also has the most elevated rate of homicide of the OECD, excluding Russia, Mexico and Brazil. It’s easy to link the two given that it is much easier to kill someone with a firearm, which is on average 13 times more effective than a knife to kill.

So, can reducing access to guns reduce the number of homicides? The answer seems evident and yet the research on this question is inconclusive!

Thus, homicides are not consistently more numerous in the United States where there are the most guns. And the decline in the number of guns in a country does not automatically reduce the number of homicides.

The case of gun shows in the United States is very interesting. These shows, similar to auto shows, allow vendors to show state of the art guns to dumbfounded clients and, of course, to sell them. Even if this varies depending on the state, it is generally easier to buy a firearm in such an event that blocks full control of police records. If the sale of guns facilitates homicides, we would expect an increase of homicides in proximity to these shows during the following weeks. However, Duggan, Hjalmarsson and Jacob (2010) show that there isn’t one. Therefore, there is not a direct link between easy access to guns and homicides.

Another instructive example is the Australian case. After the death of 35 people in a shooting in 1996, Canberra prohibited possession of certain types of guns in 1997, reducing the number of arms in Australia by about 20 percent. Leigh and Neill (2010) compare the Australian states where many guns were removed from circulation to those in which numbers only changed a little. In the five years that followed this new law, the results were essentially the same in all the states. The Australian example once again shows that reducing access to guns does not necessarily reduce the number of homicides.

How to explain these counterintuitive results? To better understand the role of weapons in homicides, it’s necessary to know how criminals get them themselves. Recent work from my colleagues Jens Ludwig and Harold Pollack, with Cook and Harris, examines the origin of guns confiscated from criminals in Chicago. They show that the large majority of weapons used by criminals had been bought, or sold, on the secondhand market. The average age of a firearm seized from a murderer is 10 years.

Thus, further limiting access to guns in stores [where it is] legal to buy new weapons will not have an immediate effect on criminals who have the tendency to use secondhand material. This is why it is so difficult to show a systematic link between access to legal guns and homicides. If such a link exists, it goes through the secondhand market and can take 10 years or more to come into being.

In conclusion, it isn’t demonstrated that reducing access to guns in the United States will have a palpable effect on the number of homicides in the short term. However, it is also not necessary to stop controlling weapons. They remain the most effective instrument to kill: The limitation of the quantity of weapons in circulation can curtail the secondhand market, reducing criminals’ access to weapons and ultimately, the number of homicides in the long term.