What Gun Laws Work?
The research is not as robust as you'd think.
In what’s surely a futile effort, Andrew R. Morral, head of RAND’s Gun Policy in America initiative, tries to cut through the emotional and ideological arguments about guns laws with evidence. Alas, the effort is underwhelming.
We have reviewed thousands of scientific articles to identify those that credibly estimate the effects of 18 different gun laws that are commonly debated in state legislatures. In particular, we have looked for evidence that these laws caused changes in one of the outcomes of interest — not just that they are correlated with these outcomes (because mere correlation is poor evidence of causation). Several policies, we find, do have substantial support in the scholarly literature — with child access prevention legislation, also known as CAP laws, or safe storage laws, boasting some of the most potent evidence about effectiveness. Studies make clear that CAP laws decrease self-injuries and suicides among youths in states that adopt them, and also decrease unintentional injuries and deaths. Yet only 19 states have such laws.
Policies requiring firearms owners to keep guns in safes or other places where children can’t access them would not have changed events in Uvalde, but they deserve strong consideration by any legislators concerned about gun deaths. It’s an unfortunate fact that mass shootings are sufficiently rare that it is hard to establish with scientific rigor whether policies affect them — although some laws that reduce gun violence in general may also reduce mass shootings.
We had firearms around the house when I was growing up. While they were not locked in a safe, they were certainly kept out of reach.
Yet, while this seems like a no-brainer solution—it has no impact on one’s ability to go hunting or otherwise engage in shooting sports—the obvious rejoinder is “Well, what if a criminal breaks into the house in the middle of the night and I need my gun to protect my family?!” Rationally, the odds of this happening are pretty slim in comparison to those of killing someone in the family by accident. The idea that you’re going to be a hero when startled awake, successfully engaging the perpetrator in the dark while flushed with adrenaline—rather asinine. But it’s a widely-held fantasy that ostensibly motivates keeping a loaded firearm at the ready.
Of course, we should not expect to implement laws only for which we have strong scientific evidence. Often — usually — no such evidence is available. Even when it is, failure to find a statistically significant effect does not mean the law has no effect: It usually means the study wasn’t strong enough to say what the law’s effects might be. When no rigorous evidence of law effects is available, policymakers and the public instead must rely on logical considerations (for example, is it plausible that restrictions on magazine capacity might reduce the carnage mass shooters cause?) and weaker evidence (such as correlations).
You’re writing a column for a national newspaper about what research shows is effective and can’t get past one policy change before moving on to conjecture? That’s . . . really not helpful.
The results of Rand’s regularly updated and ongoing review of evidence might be, at first glance, underwhelming: Most of the gun policy effects for which we sought evidence have not been evaluated or haven’t been evaluated well enough to draw strong conclusions from. The two problems — weak studies or no studies — are related: For decades, we in the United States have underfunded research and data collection efforts that could help us establish the effects of gun laws and other firearm violence prevention interventions. (In fact, for nearly a quarter-century, almost no federal funding was available for research in this area, for political reasons.)
Well, okay. But, for example, RAND is a federally-funded research center that has had a Gun Policy in America initiative headed by you for the last six years. What have you been doing if not researching gun policies?
As mentioned, where positive effects are concerned, child access prevention laws have the strongest evidence in their favor. (According to the Centers for Disease Control and Prevention, 842 children younger than 18 died in firearms accidents or firearms suicides in 2020, and a larger number were injured, though some of these casualties were at the hands of older people.) Meanwhile, the research is similarly persuasive that “Stand your ground” laws are associated with an increase in firearm homicides, and there is moderate evidence suggesting they also drive up total homicides after their passage. “Stand your ground” laws remove the traditional obligation to avoid using deadly force in a conflict if retreat is a safe option. Such laws have been popular in recent years, and by now more than half of all states have implemented them.
Okay, how we’re getting somewhere. We have a second policy. And, like the first one, it’s one that I intuitively support. While I think people should have a right to defend themselves when being attacked, I agree that they shouldn’t just be able to haul off and shoot people simply because they’re scared. It makes sense to expect people to flee to safety if that’s an available alternative. As with the requirement to lock up firearms, it wouldn’t solve the problem of spree shootings. But, while those are what get the headlines, they’re a pretty small chunk of firearms deaths.
There is less robust, but still notable, evidence for the effects of other laws. Since the mid-1990s, all states have required background checks for firearms purchased from a licensed dealer. There is mounting evidence that these checks decrease homicides. Although almost half of U.S. states have some form of “universal” background check laws, which also require checks for sales between private parties, the effects of these laws are not yet well established. There is also moderately strong evidence that waiting period laws (found in fewer than half of all states) decrease firearm suicides and homicides, and that laws prohibiting firearms possession by people with domestic-violence restraining orders — variants of which are present in most state laws — decrease intimate partner homicides.
These policies all seem reasonable. Since the overwhelming number of people who purchase firearms through legal channels will not use them in a homicide, waiting periods and background checks are a nuisance. But, presuming they’re legitimately being used to screen for those with a history of violence or mental illness or to allow someone who is simply angry or depressed to cool off, they’re a pretty small burden. And, yes, if your behavior is so erratic that a magistrate has issued a restraining order, you probably shouldn’t be in possession of firearms.
Other policies — such as assault weapons bans of the sort the United States had in place for the 10 years after 1994 — don’t yet provide enough scientific evidence to indicate what their effects might be. That is not to say that these laws do not have effects, only that they have not been rigorously demonstrated. By some definitions, for example, mass shootings declined in the United States during the period of the federal ban, but because mass shootings remain, at least in a statistical sense, relatively rare, and because rates of mass shootings highly variable from year to year, there are methodological challenges to reliably detecting even fairly strong effects for these laws.
There is at least anecdotal evidence that the kind of individual who is going to engage in mass shootings is attracted to military-style weapons. Certainly, a lot of the highest-profile shooters have used AR-15 variants. Considering that they’re actually fairly lousy weapons for sporting use, I’m not opposed to banning or at least seriously restricting their use. But, at the end of the day, they’re just regular, semi-automatic rifles that look scary.
Some people who argue about guns are simply never going to be swayed by scientific evidence regarding firearms policies, but many citizens and lawmakers are willing to take sound evidence into consideration. We have extraordinary levels of firearms violence in this country — but also research-proven ways to reduce it.
Alas, the column doesn’t provide much evidence. How many deaths would gun safes, registration requirements, cooling-off periods, and the abolition of “stand your ground” laws prevent? Presuming we could pass them into law (I know, we can’t) I’m willing to find out. But I’m guessing the impact would be relatively small.