Monday, January 7, 2019

Some Economics of Gun Regulation

When it comes to regulate gun ownership, the ratio of confident predictions to actual research evidence can be distressingly high. A substantial report from the RAND think tank, which I wrote about last spring, spells out this theme in some detail ("The Distressingly Weak Lessons of Research on Gun Control", March 12, 2018).  The Regulatory Review, A Publication of the Penn Program on Regulation, has entered the fray with a series of nine short essays on "Bringing Expertise to the Gun Debate." which ran from November 5-15, 2018. Here are some points that caught my eye:

If the public policy goal is reducing gun violence and saving lives, it may be that direct regulations on guns are not the most cost-effective method to success. From "Gun Regulation Is Costly—and Not the Only Option," by Jennifer Doleac:
But, in general, the effect of gun regulations on public safety is less clear than many advocates on either side think ... It is difficult to disentangle the effects of gun laws from the effects of a community’s feelings about guns, from a community’s motivation to reduce gun violence, or from an increase in gun purchases that often comes before the laws take effect. ...   The significant time and money required to pass gun regulations—not to mention the time and money needed to enforce such laws through policing and incarceration—could be spent advocating for and implementing other programs. Are there other life-saving programs more deserving of these resources?
Several programs are at least worthy of consideration. Summer jobs programs for teens reduce mortality by 18 to 20 percent among participants. This effect is driven by a reduction in young men killed by homicide or suicide. Cognitive behavioral therapy for at-risk young men lowers violent crime arrests by 45 to 50 percent for participants. Access to Medicaid in early childhood decreases suicide by 10 to 15 percent later in life. Mandating that health insurance cover mental health benefits at parity reduces the suicide rate by 5 percent. Access to antidepressants also reduces suicide rates: An increase in antidepressant sales equivalent to one pill per capita reduced suicide by 5 percent.
In addition, repealing duty-to-warn laws for mental health providers—which require that they report a patient’s violent threats, perhaps causing patients to be less honest—could reduce teen suicides by 8 percent and decrease homicides by 5 percent. Repealing juvenile curfews could lower urban gunfire by two-thirds. And if the goal is to reduce mortality in general—not just gun deaths—then there are many more options policymakers should consider.
Some people will argue that policymakers can and should pursue all of these policy options—that pursuing gun control does not mean advocates cannot also lobby for summer jobs and mental health care—which may be true. But ... [i]n the war over gun deaths, vast armies have gathered to contest gun regulations, a territory of uncertain value. Meanwhile, other zones of clear value are available and virtually unguarded.
Guns are in some ways a magnifier of intentions. Thus, if you try to commit suicide, a gun means your attempt will be more likely to succeed. If you are involved in a violent altercation, a gun (and a larger-caliber gun) makes it more likely that the violence will cause death. From  "Reducing Information Asymmetry in the American Gun Market," by Amanda LeSavage:
Suicides constitute two-thirds of annual gun deaths in the United States. Individuals who live in homes with guns are approximately five times more likely to commit suicide by any means and approximately 17 times more likely to commit suicide with a gun than individuals who do not have guns in their homes. Despite these enhanced risks, gun owners and their family members are not more likely to engage in suicidal ideation or planning. This finding indicates that individuals who commit suicide with guns typically do not purchase their guns with the intention of committing suicide.
Suicide usually results from an impulsive decision that can come as a surprise even to the victim. If an individual has access to a gun in a moment of such crisis and attempts suicide with that gun, there is an 85 percent chance of death. But less than 10 percent of people who attempt suicide by any other means actually die. That statistic is why the United States, which possesses almost half of the civilian-owned guns that exist worldwide, suffers from an alarmingly high suicide rate.
From "Guns Do Kill People," Anthony A. Braga and Philip J. Cook:
Fifty years ago, law professor Franklin Zimring demonstrated that serious knife assaults are similar to shootings in many respects, including apparent determination to kill or injure the victim, yet the gun assaults had a much higher “case fatality rate.” In 1972, Zimring followed with an analysis comparing attacks with different types of guns. Once again, he demonstrated that nonfatal and fatal shootings were similar with respect to the circumstances and observed characteristics of the victims and assailants. He further found that the likelihood of death increased sharply with the caliber of the shooter’s firearm, as would be expected if the intrinsic power and lethality of the weapon mattered. Zimring concluded that there was a large random component to the outcome of gun assaults and that the firearm caliber was a systematic factor that influenced whether the victim lived or died. ... 
[W]e replicated the 1972 Zimring analysis but with better data and more sophisticated statistical techniques. We collected detailed Boston Police Department data on fatal and nonfatal shootings in Boston between 2010 and 2014. The working sample included all 221 gun homicides and 300 randomly selected nonfatal cases drawn from the 1,012 non-fatal gun assaults where the victim suffered a gunshot wound that occurred during the five-year period. Boston Police investigations determined firearm caliber in about 63 percent of nonfatal cases and about 83 percent of fatal cases. ... Comprehensive statistical analysis revealed that firearm caliber had no systematic association with the number of wounds, the general location of wounds, the circumstances of the assault, or victim characteristics. This lack of systematic association is what would be expected if caliber were “assigned” at random in these criminal assaults, akin to a natural experiment. Just as with Zimring’s findings, we found a strong positive association between death rate and caliber.
A lack of information about guns and gun ownership hinders research and sensible public discussion. From "A Call to Arms Research," by David S. Abrams:
But gun regulations are an area where even basic facts, such as annual gun sales, are not well known, which is a massive impediment to progress in the field. The lack of data has forced researchers to devise creative proxies for gun sales and ownership, because the true numbers are not generally available. For example, one of the best-known studies on the relationship between guns and crime, by Mark Duggan, uses sales of the magazine Guns & Ammo to proxy for the sales of guns. Another influential article on the topic, by Philip Cook and Jens Ludwig, uses the share of suicides committed with handguns as the proxy. Both are excellent papers that provide compelling evidence on the significant social cost of firearms. But serious policy-making in this area is still hamstrung by a lack of access to data.  ...  This lack of information could be remedied simply by a government effort to collect data on this important public health topic, as former Surgeon General Vivek Murthy and others have suggested.
From "Defining “Assault Weapons," by  David B. Kopel
Handgun control has always been about guns that are genuinely distinctive. Compared to rifles or shotguns, handguns are easier to conceal, faster to deploy, and more maneuverable, especially indoors. For this reason, handguns are the most preferred guns for lawful defense—as Justice Antonin Scalia pointed out in District of Columbia v. Heller—and are also “overwhelmingly” preferred by criminals—as Justice Stephen Breyer pointed out in his Heller dissent. ... [W]hen we discuss regulations for “handguns,” everyone understands the type of gun at issue. In contrast, “assault weapon” has no fixed meaning. ...  “Assault weapon” is just an epithet to stigmatize the largest possible number of guns and gun owners—the breadth of the definition of the moment depending on the politics of the moment.
From "Effective Gun Regulation Can Be Compatible with Gun Rights." by Robert J. Spitzer
The number of ATF agents charged with inspecting gun dealers has remained roughly the same since 1973. Thanks to a provision passed by Congress in the Firearms Owners Protection Act of 1986, agents are limited to one unannounced on-site gun dealer inspection every year, regardless of the dealer’s past record. The ATF is also barred by law from computerizing its records. Background checks are still conducted by hand, laboriously, by employees who must sift through paper records housed at its tracing center in Martinsburg, West Virginia. A routine trace takes about five days—a process that would take seconds by computer.