In the year since Adam Lanza used a Bushmaster rifle to gun down 20 children in Newtown, Conn., the discourse on gun control has focused on mass shootings and homicides. That’s not surprising: Terrible events dominate the cable news cycle, and murders get reported every day in our nation’s newspapers. But if we want to talk about the effects of guns, we should remember this: In a typical year, suicides outnumber homicides by 3 to 1 and a majority of suicides are by firearm. Suicides come in ones and twos, here and there; they rarely make the national news, and when they are reported at all they are veiled in euphemism (“He died suddenly”). Suicides go so underreported that Slate’s Gun Deaths Project, which collects data from news articles and other online sources, categorizes only roughly one-tenth of the reported deaths as suicides.
The national conversation about guns, and about gun control, should include the relationship between guns and suicide. We recently analyzed a decade’s worth of data on guns and suicides in the United States and we found that the relationship is clear: more guns, more suicides.
Suicide is neglected compared with the large quantity of research on the relationship between homicides and guns, a relationship which remains controversial because it’s difficult to demonstrate causality. Places with lots of guns may have high homicide rates, but is this because guns cause homicide or because homicides cause people to buy guns? Or could a third factor—say, a general lack of social trust or high violence in a region—be causing both homicides and gun possession? The relationship between suicides and guns is much easier to tackle because it’s unlikely that an increase in the number of suicides in a community would cause an increase in local gun ownership.
So in a new paper published in the International Review of Law and Economics, we studied the relationship between guns and suicide in the U.S. from 2000 to 2009. Using five measures of gun ownership and controlling for other factors associated with suicide, such as mental illness, we consistently found that each 1 percentage-point increase in household gun ownership rates leads to between 0.5 and 0.9 percent more suicides. Or, to put it the other way, a percentage-point decrease in household gun ownership leads to between 0.5 and 0.9 percent fewer suicides.
Are the people not killing themselves with guns simply committing suicide by other means? Some are—but not all. While reduced household gun ownership did lead to more suicides by other means, suicides went down overall. That’s because contrary to the “folk wisdom” that people who want to commit suicide will always find a way to get the job done, suicides are not inevitable. Suicides are often impulsive decisions, and guns require less forethought than other means of suicide—and they’re also deadlier.
Our research had to overcome the fact that no one knows with great precision how many guns there are in America, how many households own a gun, how gun ownership varies demographically and geographically, what types of guns there are, or how guns are used. In part that’s because in 1996, Congress banned the CDC from funding any research to "advocate or promote gun control." That’s not a ban on gun research, technically, but after Congress extended the wording and expanded the ban to other agencies, it had enough of a chilling effect to reduce CDC funding for gun violence research from $2.5 million per year in the early 1990s to just $100,000 in recent years.
But don’t take our critique of bans on government-funded research as a plea for more funding for our research. We would be satisfied if the CDC and other government agencies such as the Bureau of Justice Statistics were simply allowed to collect more and better data on guns, homicide and suicide. The CDC’s National Violent Death Reporting System, for example, still covers only 18 states. President Obama lifted some of the restrictions on the CDC earlier this year, so we are cautiously optimistic about the prospects for future research. (Gun advocates and supporters of Second Amendment rights shouldn’t assume that more research simply means more arguments against guns. We are eager, for example, to see more studies on the defensive use of guns, a phenomenon about which there is currently very little trustworthy data. Moreover, better research might find ways of reducing gun violence without violating Second Amendment rights. Indeed, reducing gun violence could be one of the best ways of reducing the demand for gun control.)
Despite challenges presented by the data, our findings appear robust and are consistent with a series of “natural experiments” from around the world. For example, following the 1996 killing of 35 people in Port Arthur, Australia, a strong movement for gun control developed in Australia. States and territories made uniform and more stringent regulations for the possession of firearms, and instituted a buy-back of the newly illegal guns, most of which were rifles and shotguns. As Andrew Leigh and Christine Neill determined in a paper published in the American Law and Economics Review, these changes resulted in a reduction of the country’s firearm stock by 20 percent, or more than 650,000 firearms, and evidence suggests that it nearly halved the share of Australian households with one or more firearms. The effect of this reduction was an 80 percent fall in suicides by firearm, concentrated in regions with the biggest drop in firearms. Meanwhile there was little sign of any lasting rise in non-firearm suicides.