Deborah Azrael earned her doctorate in health policy with a concentration in statistics and evaluative sciences. She is Director of Research at the Harvard Injury Control Research Center (HICRC).
Matthew Miller, a physician with training in internal medicine and medical oncology, is a Professor of Health Sciences and Epidemiology at Northeastern University, Adjunct Professor of Epidemiology at the Harvard School of Public Health, and Co-Director of the Harvard Injury Control Research Center.
In collaboration with their research assistant Wilson Zhang, a 2020 graduate of Northeastern University, Deborah Azrael and Matthew Miller published a study in Annals of Internal Medicine that investigates the 2020 surge in gun purchases. They wanted to find out approximately how many gun buyers were purchasing guns for the first time, versus how many already owned guns. What they discovered through an in-depth analysis of new survey data sheds light on changing gun ownership patterns in the U.S.
First, some background. Starting in March 2020, as Covid-19 swept through the nation, leaving a trail of medical, economic, and political chaos in its wake, Americans flocked to gun stores. The FBI’s National Instant Criminal Background Check System (NICS), which records all firearm transactions that require background checks but provides no information about gun purchasers, recorded an astonishing 33% more background checks in 2020 than in 2019.
The unprecedented surge caused many to wonder what was motivating these buyers, and who exactly they were. But the absence of data on who was purchasing these guns created a “data gap,” and the National Shooting Sports Foundation (NSSF), a trade association that champions unrestricted “gun rights,” attempted to fill it. They surveyed 175 gun store owners on the race and gender composition of their spring 2020 customers. Based on these limited data – dependent on the perception and memory of 175 gun store owners – the NSSF concluded that women constituted 40% and African-Americans 58% of first-time gun buyers from January through April 2020. Given the lack of alternative, more reliable, data on who was buying guns, mainstream media seized hold of the enticing claim that the surge consisted largely of people who had never owned guns before, specifically a significant increase in firearm ownership among women and people of color.
Caroline Light: Your study seems to confirm what the NSSF declared, that a significant portion of buyers in the 2020 “gun surge” are female and non-white. Where does your study diverge from the NSSF findings?
Deborah Azrael: We found that the vast majority (80%) of people who bought guns already owned them, and that these purchasers, like gun owners overall, were mostly white men, Still, because people who bought their first guns over the period we studied were more likely to be women and people of color, the overall demographic composition of gun owners appears to be changing, and is now less white, less male, and less rural than it was, say, 10 years ago. We don’t know when that shift in who is buying guns occurred, but it’s worth noting that, in contrast to the narrative that emerged – understandably – in March 2020, the survey doesn’t seem to be clearly related to Covid, since about the same percentage of gun purchasers bought their first gun in 2019 as in 2020. It’s hard to know for certain what factors have motivated purchasing, but gun industry groups have been saying that Black people and women have been buying guns at high rates for a long time, and maybe that claim itself was an effective marketing tool. It’s hard to know in the absence of historical data.
Matthew Miller: In our study, about 20% of people who purchased a gun in 2019 (pre-Covid) and about 20% of those who purchased a gun in 2020-21 (during Covid) were new gun owners. 20%, not 40%, as the NSSF reported. There were a lot more new gun owners in 2020 compared with 2019, but not because the proportion of new gun owners increased; there were more new gun owners because a lot more people bought guns in 2020 than in 2019.
We didn’t witness an abrupt change in the proportion of people purchasing who were new gun owners, or in the proportion of those new gun owners who were female or people of color. What we saw was a larger number of people in general who were buying guns. NSSF said 40% were new without further characterization of uncertainty around their estimate; we included confidence intervals – margins of error – as is standard practice when using survey data. Journalists were hungry for numbers, but because the U.S. does not have a registry for guns, like we do for cars, and because the U.S. also does not have a federally funded ongoing survey-based surveillance system to keep track of firearm ownership, acquisition and divestment, the choice journalists were faced with was to run with the NSSF estimates, or wait for scientists dependent on privately funded largess to generate accurate data. As a result, we never have anything approaching real-time estimates for changes in firearm ownership, purchasing or use, and that makes it especially hard to understand in real-time how secular shocks change gun purchasing patterns.
CL: What do you mean by “secular shocks”?
MM: Anything like George Floyd’s murder, or Biden’s election, or things that we can’t predict…
DA: Like mass shootings…
MM: …that could influence gun purchasing patterns. If you want to know what happens in response to these sorts of incidents, you need to have an on-going surveillance system that allows you to track and analyze the correlation of behaviors with stimuli. This would allow us to plan ahead, the way we can for other forms of violence and causes of injury and misery. There’s little money invested in on-going surveillance. Deb and I pulled this survey together over the summer with financial support from the Joyce Foundation. We have to rely on private foundations for this kind of maintenance work.
DA: Especially to the extent that the public health community tries to intervene with gun owners to reduce their risk and their family’s risk, that research is based on the 20 or 30 years in which most gun owners were older white men. And while they’re still disproportionately older white men, there are many new entrants, and we should be nimble enough to consider new exposures and where we might address new gun owners to minimize the risk they pose to themselves and others.
The other question is “what happened?” Gun ownership was relatively stable for two decades and now things are shifting, and we’re not sure exactly how. A new organization, called Project Unloaded, modeled on the effort to prevent young people from starting to smoke, has just started work to try to reduce the chances that young people will choose to become gun owners. Our survey is evidence that we’re starting even further behind than we thought. The public health community is seeking to reduce the harms of gun ownership even as more and more people own them.
CL: We have to rely on private institutions to fund the research, since there’s not a federal effort to mitigate the risk. And this work is taking place as more people view guns as a source of legitimate, essential self-defense and home security. You’re pushing the boulder uphill.
MM: The proportion of people who say “a gun in the home makes the home safer” is growing significantly. In the 1990s, roughly two-thirds of people said guns make the home less safe. Today, two-thirds say a gun makes the home safer. How did that happen? This shift means that we are not only pushing the boulder uphill, it means that the hill is steeper and the boulder heavier.
CL: What main take-aways do you have for readers trying to parse the significance of the 2020 “gun surge,” which appears to be continuing into our present moment?
MM: From an injury and death prevention standpoint, let’s consider what these data mean for households that now have firearms, but didn’t before. By bringing guns into homes that didn’t have them before, the new gun-owners have imposed a substantial mortality risk on their family members, including children. In fact, we estimated five million children live in homes with guns in 2021, each of whom is at several-fold higher risk of suicide and unintentional firearm death than they would be if they lived in homes with no guns. And there’s no evidence whatsoever that having a gun in one’s home confers any protection from injury and death from assaults. Indeed, for women and children in particular, the risk that they will be killed by a gun, most commonly by an intimate partner or family member, is substantially higher when they live in homes with guns, compared to those in a home without guns.
DA: And, when a household has guns, we don’t know enough about how to motivate gun owners to store their guns safely, i.e. locked up and with ammunition stored separately. We just published a study of Child Access Prevention Negligent Storage (CAP-NS) laws, which are in seventeen states and provide criminal sanctions for adults who fail to prevent their children or dependents from accessing firearms in the household. Our study concluded that these laws don’t appear to significantly improve gun-owners’ storage practices. We also discovered that most gun owners surveyed did not even know whether their state had a CAP-NS law. The bottom line is that, as we’re seeing guns spread into more households, we’re not seeing more attention to safe storage or other safety measures to keep family members safe from gun accidents and suicide.
CL: Recent interviews with Black women gun owners indicate that the messages about guns as essential tools of self-defense, that guns make your home safer and yourself safer, are persuasive. There’s not clear evidence that COVID caused the surge, but I wonder about the increased visibility of violent, state-sanctioned, white supremacist and sexist ideals of belonging that particularly exclude Black and Brown women. It seems logical that many would look to a rich history of Black women’s armed self-defense – dating back to abolitionist resistance – to take up firearms as tool of self-defense amidst the rise of an increasingly visible movement of reactionary white supremacy.
DA: I think that’s right. There’s a kind of arms race…it scares me. What I haven’t seen that I’m interested in is, what’s the calculus when being an armed person of color puts you at an enormously increased risk of violence from the state? How does a person of color think about whether the risks of gun ownership outweigh the perceived risks of not having a gun?
CL: I can’t help but think about Amir Locke, a 22-year-old Black musician, a licensed gun owner, shot and killed in a police raid in Minneapolis. Amir was a classic “good guy with a gun” who bought a gun to protect himself as a DoorDash deliverer, and the police shot him before they could discern whether he was an actual threat. This pattern includes Black gun owners like Atatiana Jefferson, killed in her home in Fort Worth, TX in 2019; Philando Castile, killed in his car in Minnesota in 2016. It reminds me too of Kenneth Walker, Breanna Taylor’s fiancé, who tried to protect himself and Breanna from home intruders in March 2020. Those intruders turned out to be police, who shot and killed Taylor, then roughly arrested Walker for endangering a police officer. Where anti-Black violence is so widespread and normalized, the calculus of whether or not to own a gun is indeed complex.
DA: It is painfully complex. But as Matt noted, what we know is that the evidence is overwhelming that bringing a gun into a home increases, not reduces, the chance that someone in that home will die by suicide, and for women and children, homicide. There aren’t any easy solutions, but the evidence is that guns increase rather than decrease risk.