“If you think mass shootings are becoming more frequent and worse – you’re right,” declared one newspaper headline in 2017 following a spate of horrific and highly publicized mass shootings. Meanwhile, a brief glance at the National Rifle Association’s (NRA) website yields a different spiel of statistics – the first bullet point of its “Get the Facts” page on crime states that “as gun ownership has risen to an all-time high, the nation’s total violent crime rate has fallen to a 44-year low and the murder rate has fallen to an all-time low.” Statistics can be construed to support any story depending on where the author’s sympathies lie; therefore it is more important than ever for us to be discerning in our judgment and base our reasoning in sound statistical and economic evidence.

Conflicting accounts of gun violence and the effects of gun policy proliferate, but despite these discrepancies – which beg for greater clarity over the nature of gun violence in the US – there has been a dearth of concrete, reliable research on the topic. This can be traced back to the passage of the 1996 “Dickey Amendment,” which stated that the Center for Disease Control and Prevention (CDC) could no longer use money to “advocate or promote gun control.” When federal research dollars stopped, so did the flow of funding into private research, resulting in an alarming lack of data collection and analysis surrounding gun research. The recent passage of a bill that included language giving the CDC the authority to resume gun-related studies (though only on the causes of gun violence) and a commitment by Kaiser Permanente to invest $2 million in gun violence research gives hope that more conclusive studies on guns can be conducted in the near future.

MORE GUNS MORE CRIME, OR MORE GUNS LESS CRIME? EXAMINING THE WEAPONS INSTRUMENTALITY HYPOTHESIS

Perhaps the battleground on which the debate is most centered – and where the evidence is most disparate – is that of the weapon instrumentality hypothesis, which purports that the increased accessibility of guns contributes to the escalation and lethality of violence. Misleading assertions abound on both sides: it is commonly cited that the U.S. has by far the most firearm homicides in any developed nation at 29.7 homicides per million people (versus 1.4 in Australia and 5.1 in Canada) and the highest per-capita civilian-owned guns around the world, suggesting that more guns leads to more homicides. Graphs illustrating the relationship between gun ownership and gun deaths on the state and country level are commonly cited in pro-gun control articles. Yet, as any statistics course will tell you, just because there appears to be a positive relationship between two variables doesn’t mean that one has a causal effect on the other. Many other factors within a state or a country can influence the disparities: the average level of education and socioeconomic environment, the severity and deterrence effects of criminal punishments and the nature and degree of gun laws, among others. Similarly fallible reasoning is implicit on the NRA-ILA website, as points like “the nation’s violent crime rate and its murder rate have decreased by more than half [since 1991], as Americans have acquired over 170 million new guns” suggest. In order to filter out misleading statistics from substantiated evidence, systematic reviews of academic literature on gun violence are necessary.

I. GUN OWNERSHIP PROXIES & GUN-RELATED DEATHS

Due to the scarcity of longitudinal data on gun prevalence, many academic studies use the proportion of suicides committed with a firearm (FS/S) as a proxy of gun prevalence. FS/S is purported to be highly correlated with state-level estimates of gun ownership, and therefore an appropriate proxy to use. Whether or not this proxy is valid and appropriate is contested, especially since the correlation with gun ownership estimates can only be shown on a state-level – more disaggregated data of gun ownership (e.g. on the county level) is at worst non-existent, and at best unreliable. But this is the reality of what economists have to work with; if sufficiently disaggregated reliable gun ownership data existed, there wouldn’t be a need for a proxy in the first place.

Most of these studies conclude that firearm prevalence is positively associated with total homicides, with the exception of Kovandzic et al., which concludes the opposite in some of its instruments. The study finds that increased subscriptions to the Field & Stream, Outdoor Life, and Sports Afield magazine and an increased percentage of people in the county voting for Bush in the 1988 election significantly predicts less firearm-related homicides. Whether or not these are relevant proxies for gun ownership is debatable, but the study does provide some support for the view that higher gun ownership potentially leads to more deterrence and better self-defense, and therefore lower violent crime.

However, a study conducted by Stanford professor Mark Duggan using the Guns & Ammo magazine sales proxy determined that changes in gun ownership are significantly positively related to changes in the homicide rate. He concludes that the positive relationship between magazine sales and the homicide rate is driven by the impact of gun ownership on murders in which a gun is used – that is, guns are more likely to be used by perpetrators in murders if ownership is high. Criticism of the study has largely been focused on the use of Guns & Ammo as a proxy for gun ownership. From the differing conclusions of these studies, then, it seems that the conclusions drawn about the relationship between gun prevalence and violence depend on the proxy used for ownership.

II. RIGHT-TO-CARRY (RTC) LAWS

Given the lack of direct measures of gun prevalence, it could be helpful to look at the effect of specific gun policies on crime, rather than look at gun ownership as a whole. Professor Duggan does exactly that: he examines the impact of concealed weapons legislation on crime, but ultimately rejects that these laws led to reduced crime. Other research studies are unsurprisingly divided on the topic. John R. Lott, Jr. and David B. Mustard’s seminal 1997 work – which coined the “More Guns, Less Crime” phrase and influenced many states to relax their concealed-carry laws – found that murder rates dropped by 7.65 percent in 10 states after they passed laws making it easier for their citizens to carry concealed weapons. Many economists and experts, including a majority of members on the National Research Council, have since denounced the methodology used in that paper. A more recent study by Stanford Law School Professor John Donohue concluded that states that adopted RTC laws – granting citizens the presumptive right to carry concealed handguns in public – experienced a 13-15% increase in aggravated assault in the decade after enacting those laws. There seems to be an increasing number of studies corroborating the positive relationship between RTC laws and increased gun violence, but this is unlikely to be the final word on the topic; it just happens to be the most recent in the decades-long academic exchange between the authors.

III. THE ASSAULT WEAPONS BAN

Much of the argument on the pro-gun-control side of the debate not only targets RTC but also advocates the wholesale ban of assault weapons. In considering the potential effects of an assault weapons ban, those supportive of a ban often point to one clearly successful case – Australia. Following a horrific shooting in Port Arthur in 1996 involving an AR-15 and a military-style rifle resulting in 35 deaths, elected officials in all of Australia’s states and territories banned semi-automatic and self-loading rifles and shotguns. A national buyback program for newly prohibited weapons was instated in 1996-7, and more than 700,000 weapons were surrendered. Two decades later, the results of the firearms regulation are overwhelmingly positive: the percentage of victims of firearm-perpetrated homicide, as a proportion of all homicide, declined between 1989–90 and 2009–10 from 24% to 12%. There hasn’t been a mass shooting in 22 years, with the exception of a terrifying incident in Osmington in May where a father murdered his entire family of seven in their sleep.

It may not be suitable, however, to apply the Australian model to the American context. First, Australia’s assault weapons ban constituted only one of the country’s many legislative reforms under the National Firearm Agreement. Other policies, like requiring standard licensing and permit criteria, storage requirements and inspections, a 28-day wait time between permit approval and weapon purchase and greater restrictions on the sale of firearms and ammunition also contributed to the reduction of firearm-perpetrated homicide, and there is little evidence that an assault weapons ban in the US would have a similar positive effect in the absence of other reforms. Moreover, efforts to reduce the number of guns in circulation have not proven to be effective in the US: as the NRA website points out, the congressionally-mandated study of the federal assault weapon ban in place from 1994-2004 found that the ban had no conclusive impact on crime. That being said, the ban had only applied to weapons manufactured after the date of its enactment and thus had not reduced the number of guns in circulation (in fact, there was a surge in assault weapons production and purchases in the months before the ban took effect). The results of this study, therefore, have little bearing on whether a complete ban on assault weapons – accompanied by a comprehensive buyback program – would have positive effects.

Past buyback programs in the US, like those initiated in the wake of the Sandy Hook Elementary School massacre in Connecticut, often end up with more hunting rifles or old revolvers than semi-automatic weapons that can cause the most damage. Those most likely to use their weapons to commit crimes are also the least likely to participate in a buyback program – Australia’s gun buybacks were accompanied by the threat of jail time, requiring all citizens to surrender their assault weapons. A buyback in isolation will not be effective; rather, a ban on assault weapons – that, to some, will essentially amount to the confiscation of private property – will be necessary to achieve a similar effect.

An assault weapons ban will be exponentially more difficult for the US than Australia. Since 1996, Australia has collected around 1 million assault weapons, drastically reducing the number of assault weapons in circulation. The US, on the other hand, has approximately 310 million firearms in circulation, according to a 2012 Congressional Research Service report. It is unclear how many of these are assault weapons, although NRA research coordinator Mark Overstreet remarked in the Heller v. District of Columbia case that, from 1986 to 2007, at least 1.6 million AR-15-style semi-automatic rifles were produced and not exported from the United States. It is clear that an assault weapons buyback program in the U.S. would have to be of a much larger scale and would come at a greater cost to the state and federal government.

IV. WEAPONS OF CHOICE

At any rate, even if an assault weapons ban and a mass buyback were possible, it isn’t clear how much of an impact they would have on reducing the incidence of firearms-related homicides. A study by the Firearm and Injury Center at Penn (FICAP) in 2011 showed that handguns are the homicide weapons of choice – 71% of the weapons used in homicides the previous year were handguns, not semiautomatic rifles. (It is telling that FICAP has not released such a study in the years since due to lack of funding, highlighting the acute need for research in this area.) That being said, a recent piece by the Washington Post in the aftermath of the Santa Fe shooting reveals that 9mm semiautomatic handguns showed up more than any other weapon in the analysis of 152 mass shootings since 1966. Semiautomatic guns are undoubtedly more lethal and have more potential to kill than their non-automatic counterparts, but whether or not banning them will reduce gun deaths remains a topic for debate.

V. BACKGROUND CHECKS AND MENTAL HEALTH

Unlike in Australia, the demographic of violent gunmen in the US demonstrates a clear pattern. In Australia’s case, over half of mass shooting victims in the two decades before the legislative reforms were passed were shot by licensed gun owners, and most shooters had no previous history of mental illness. In this context, there was little Australia could do to target individual agents of harm: in order to decrease the risk of another mass shooting tragedy occurring, it was necessary to reduce the means of mass destruction by banning and destroying semi-automatic firearms. It is not clear that the same holds true in the US, as there is certainly more room to target the types of individuals that are more likely to commit gun crimes. The pro-gun didactic of “Guns don’t kill people, people kill people” may have more substance to it than its opponents might like to believe, at least in the context of the US. And there is evidence to support the potential impact of individual-discerning gun laws – a systematic analysis of thousands of gun policy studies by the non-profit think tank RAND corporation revealed there was supportive evidence that child-access prevention laws may decrease suicide, unintentional injuries and deaths, and moderate evidence that background checks decrease suicide and violent crime.

It would be hard to argue a case for why individuals seeing psychiatrists for serious mental illnesses should retain a fundamental right to own a gun, given that they are not always treated as agents capable of conscious choice. But under federal law, a person can only be barred from purchasing or possessing a firearm due to mental illness under two conditions: (i) if they are involuntarily committed to a mental hospital, or (ii), if a court or government body declares them mentally incompetent. Many people who are diagnosed by psychiatrists to be mentally ill or known to be “experiencing and enduring mental illness [their entire lives]” like 19-year-old Parkland school shooter Nikolas Cruz can, at present, purchase semiautomatic rifles perfectly legally. Regardless of whether a general ban of assault weapons is passed, gun access should be further restricted for those most prone to violence.

TAKING A STEP BACK: THE AGE-OLD UTILITARIANISM VS. INALIENABLE RIGHTS DEBATE

The confusing mélange of different policies and conflicting economic research renders it unsurprising that the American public is so conflicted about gun policy.  Yet the division can hardly be attributed to the disparate statistical evidence – very few look to the evidence to determine their stance on gun control; rather, the evidence serves to corroborate their beliefs. And those beliefs find their roots not in statistics and economic analysis, but in a much older jurisprudential debate: what role should legal rules play in regulating society? The general, albeit simplified, outline of the conflict places maximizing overall social welfare (under the assumption that abolishing guns will make society as a whole better off) in opposition to the institution of inalienable rights. Of course, it is much more nuanced than the aforementioned oversimplified binary choice – after all, the statistical debate surrounding the weapons hypothesis contends with the former argument that banning guns will lead to less crime and therefore greater total welfare. As for the latter, “the institution of rights” can perhaps be most contemporaneously conveyed through American philosopher Ronald Dworkin’s theory of rights. He asserts that the importance of taking rights seriously is based on two important ideas: firstly, human dignity in the Kantian sense, “recognising [an individual]as a full member of the human community,” and secondly, political equality that “supposes weaker members of a political community are entitled to the same concern and respect of their government as the powerful members.” This speaks to the interpretation of the Second Amendment that suggests it protects “the right of people to keep and bear arms for defense of life and liberty.”

Second Amendment

A well regulated Militia, being necessary to the security of a free State, the right of the people to keep and bear Arms, shall not be infringed.

Dworkin’s jurisprudential framework is relevant because he believes that “jurisprudential issues are at their core issues or moral principle, not legal fact.” We should not just look to legal fact when reviewing gun policy – in this case, the specific wording of the Second Amendment – because, as the debate surrounding the District of Columbia v. Heller decision suggests, the issue of gun rights is far more nuanced than the sweeping single line of the amendment. Even if we fully recognize the right of the American people to “keep and bear Arms,” we have to decide what type of arms are permissible and under what situations. Regardless of where our opinions fall, we are inherently concerned with moral arguments of “should” or “should not.”

For Dworkin, there are only three grounds to consistently limit the definition of an individual’s rights: (i) if the values protected by the original right are not really at stake, (ii) if there is an abridgment of a competing right, and (iii) if the cost to society is far beyond the cost paid for granting the original right, a degree great enough to justify any assault on dignity or equality that may be involved. The first is hotly debated in the interpretation of the original rights that the Second Amendment affords: some argue that the presumptive right to guns should not even exist; it should be a conditional privilege rather than an untouchable right. As for the second, pro-gun-control supporters often cite the “right to safe schools” (and implicitly, safety in general) as a competing inalienable right to “the right to guns”: the question of which right is more important depends on which side of the debate one falls on. The third line of reasoning is one often cited by public health practitioners: to them, the cost to society of gun violence far outweighs the benefits of granting the original inalienable right. The public health perspective is being increasingly adopted for other current criminal issues, such as drug abuse – because it affects all of us, not just the perpetrator. As an agent of harm, public health practitioners argue, the gun should be amenable to standard injury prevention procedures. But all these arguments find their basis in moral rights – even if the relationship between guns and violence were affirmatively ascertained, such a utilitarian view does not necessarily address the issue of the “inalienable right”.

On closer inspection, the arguments on both sides – though fraught with statistical errors of fact – aren’t diametrically opposed. Rather, their disagreements boil down to a presupposition of what constitutes a fundamentally inalienable right and whether utilitarian arguments of overall social welfare justify the abridgment of such rights. Neither side seems to be able to penetrate this fundamental presupposition with their statistical arguments because it finds its roots in beliefs of a moral nature that no amount of statistical evidence can displace. That isn’t to say that there is no hope for consensus on issues of gun policy – a survey by Pew Research Center reveals that large majorities of both Republicans and Democrats find common ground in favoring the prevention of people with mental illnesses from buying guns, barring gun purchases by people on federal no-fly or watch lists and background checks for private gun sales and sales at gun shows. Such policies also happen to be the ones with moderate supportive evidence: they can, and should be, implemented. As for broader, more contentious policies like RTC laws and assault weapons bans, there is hope that an increase in research funding can help shed more conclusive light on the potential effects of such policies. Conclusivity may not breed agreement – as the utilitarian vs. inalienable right debate suggests – but increased research and significant evidence can galvanize both sides to reexamine their beliefs, regardless of which direction the evidence weighs in favor. A paradigm shift is needed – policy-makers should look to well-grounded statistical research, not partisan ideology, as the foundation of their policies.  It is then, and perhaps only then, that we can begin a substantiated conversation on how to address broader gun policy in the nation.


Ruru Hoong, a senior studying economics, is a staff writer for Stanford Politics.