Tanvi Misra is a staff writer for CityLab covering immigrant communities, housing, economic inequality, and culture. She also authors Navigator, a weekly newsletter for urban explorers (subscribe here). Her work also appears in The Atlantic, NPR, and BBC.
A new project visualizes what the science says—and doesn’t say—about the effects of firearm regulations.
Americans are suddenly—and seriously—talking about guns. Since the shooting at Stoneman Douglas High School in Parkland, Florida, in which 17 were killed, the public conversation about firearm regulations has acquired a new urgency. The teens who survived the tragedy have spearheaded nationwide rallies for more protective gun laws, and they have shown no signs of letting up. Businesses are withdrawing support from the National Rifle Association (NRA). Several retailers are changing their policies on selling weapons to buyers under 21. And President Donald Trump is whipsawing between demanding armed teachers and stricter gun control laws. (And then, according to the NRA, re-reversing himself.)
In many ways, Trump embodies in one person this nation’s intense polarization and deep-seated confusion surrounding firearm regulations. But which position is backed up by the best evidence? What does science say about what laws save lives and what don’t? That’s where the Gun Policy in America project comes in.
Over the last two years, researchers at the RAND Corporation have synthesized all the available research on the effect of various gun laws on a range of outcomes. To that, they’ve added a survey of policy experts both for and against gun control. The results of this self-funded project have been visualized in interactive maps and charts on their website. Think of this as your one-shop-stop to understand what we know about gun policy—and importantly, what we don’t.
There are two big takeaways. One, some gun control policies clearly seem to decrease certain types of violence. Two, while there is more overlap than one would have expected among experts on both sides of the issue, the huge gaps in research make it impossible to reach any kind of consensus.
“We haven’t invested, as a country, as much as we should in building a data infrastructure,” said Andrew R. Morral, the lead researcher on the project. “The second thing is: There’s not very much investment in research to use the data that is available or to collect more data.”
Let’s dive into the first part of the project. Clicking on the little magnifying glass in the “policy analysis” section pulls up the table below, which gives a snapshot of the literature review. In the vertical column on the left are the gun policies the RAND researchers have examined—both restrictive ones such as bans on assault weapons and minimum age requirements, as well as permissive ones such as concealed-carry and “stand your ground” laws. In the first row, you can see the spectrum of outcomes the researchers were interested in. Among the negative ones are mass shootings, suicide, accidental deaths, and violent crime. But they also looked for research on links between these policies and gun sales, incidents of defensive gun use, and hunting and recreation.
The table is populated with the results of the analysis: The darker the purple, the stronger the evidence supporting the effect of a policy on the outcome. The gray cells, on the other hand, show where there’s no research at all. (On the website, you can click through to get a detailed explanation of how the researchers came to these conclusions.) Here, science staunchly backs up the claim that laws that prevent children from getting their hands on guns help reduce unintentional injuries and suicide (deepest purple boxes).
For four outcomes that are central to the gun debate going on in the country—how gun control laws hinder people’s ability to defend themselves, whether they discourage hunting for sport, and whether they increase or decrease the likelihood of mass shootings and officer-involved shootings—the researchers found very little authoritative evidence. (See: the many boxes marked “inconclusive.”) Via the report accompanying the project:
This does not mean that these policies are ineffective; they might well be quite effective. Instead, it partly reflects shortcomings in the contributions that science has made to policy debates. It also partly reflects the policies we chose to investigate, all of which have been implemented in some U.S. states and so have proven to be politically and legally feasible (at least in some jurisdictions).
Here’s another way of visualizing the results, available on the website. Below are seven policies (on the left) that research has found to have substantive effects on at least four outcomes (listed on the right). The brown lines show a positive effect, whereas the green ones indicate a negative effect. The thicker the lines, the stronger the science. Below, you can see moderate evidence that background checks and ownership restrictions for individuals with mental illness decrease suicide and violent crime. In other words, the review demonstrates that some gun control policies, at least, do work.
Given the lack of scientific research regarding many of the policies, the RAND researchers went ahead and surveyed experts and advocates on both sides of the battle. Their sample included those who supported looser restrictions on guns (NRA, Gun Owners of America, etc.) and those who supported tighter laws (Brady Campaign to Prevent Gun Violence, Mayors Against Illegal Guns etc.). On their website, the researchers have put up a nifty interactive map to see the results of the survey.
If you toggle a particular law “on,” you can see what both camps of experts think will happen to a particular outcome if that law is enacted across the country. So let’s say I want to see the effect of a blanket ban on assault weapons and high-capacity ammunitions—a key demand in the post-Parkland gun conversation—on mass shootings. When I select those options, the results show that pro-gun experts believe the policy will produce no change (first map below). The anti-gun experts believe it would lead to a 11 percent decrease in mass shootings, nationwide (second map below).
Similarly, this is what the experts think will happen to firearm homicides if “stand your ground” laws pop up across the country. The first group thinks firearm-related murders will go down 4 percent; the second thinks they’ll go up 2 percent. (A small caveat: If more policies are selected via this tool, more assumptions come into play, and the resulting maps get less reliable.)
Across all outcomes, the researchers found that the disagreements weren’t as widespread as they seem. They were particularly stark when it came to three laws in particular: concealed-carry, elimination of gun-free zones, and stand your ground. Four other laws—prohibitions against gun ownership for people with mental illnesses, required reporting of lost and stolen guns, awareness campaigns to prevent children from accessing and misusing guns, and the surrender of firearms by individuals prohibited from owning them—elicited a bit more overlap in opinion.
A key hopeful finding of this survey: The two groups do, in fact, have shared goals. They just disagree on how to achieve them. “We find some pretty good evidence that there’s not a values disagreement here,” Morral said. “It's not that one side favors community violence reduction and the other side favors Second Amendment rights more. Everyone wanted to reduce homicides, suicides, mass shootings, and accidents—the other things were more secondary. … They appeared to evaluate policies in the same way, but they were working from different assumptions about what the policies could achieve.”
That’s important, because more rigorous research around some of the most contentious gun policies could actually help bridge the gap in opinion, the researchers concluded. But getting that data has long been a challenge, for nakedly political reasons: a 1996 amendment to a spending bill—lobbied for by the NRA—that forbade the Centers for Disease Control and Prevention (CDC) from using government funds for research that “may be used to advocate or promote gun control.”
For more than two decades, that legislation has hindered researchers from treating gun violence as the public health problem that it is. While there are national datasets that contain information about the circumstances of each traffic fatality, no such database exists for gun deaths. (Traffic deaths slightly outnumber gun deaths overall, but are less numerous in 21 states. Along with suicides, injuries, accidents, and defensive shooting, all gun incidents came to over 60,000 in 2017, according to one count.) If decreasing gun violence is indeed a shared value in America, both camps will have to start with the numbers, so that we can build a shared understanding of the scale of the problem.