How We Can Reduce Gun Violence

The massacre of 20 young children and six adults at Sandy Hook Elementary School, in Newtown, Conn., has revived interest in gun regulation—a topic that was almost entirely ignored during the presidential campaign. During President Obama’s first term, there were several other mass shootings, but the only Congressional action was a new law to permit tourists to carry guns in national parks. Now Vice President Biden has been asked to develop a plan that includes new regulations.

Four measures that he will surely consider are: a ban on assault weapons; funds to improve the instant-background-check system for screening gun buyers; an expansion of that system to include private sales (rather than only sales through federally licensed dealers); and stronger enforcement to reduce illicit gun carrying and misuse. But are any of those options likely to be effective in preventing the next massacre or in reducing rates of “routine” gun violence?

Direct evidence of the potential effectiveness of such measures is limited. We did have a federal ban on military-style assault weapons from 1994 to 2004. The aspect of the law that was most related to rampage shooting was a ban on large-capacity magazines—clips that hold dozens of rounds of ammunition and allow a shooter to kill many people before reloading. Unfortunately, the ban did not extend to the millions of such magazines already in circulation.

An evaluation by Christopher S. Koper, director of research at the Police Executive Research Forum, concluded that there was little effect on the use of those magazines while the law was in force, and no statistically detectable effect on homicide rates in the United States. If a renewed ban went further and prohibited transfers of any large magazines (new or used), perhaps coupled with a program to buy back and destroy those currently in private hands, the result might be more favorable.

What about background checks? Current law requires all sales by federally licensed gun dealers to be preceded by a background check for a disqualifying criminal record or involuntary commitment to a mental hospital. The “instant” check system relies primarily on state records, which vary widely in quality and completeness. Modest federal programs give financial support to states to improve their record keeping. After the Virginia Tech massacre of 2007, there was particular interest in making mental-health records accessible as part of the instant background check.

A continuing research project headed by Jeffrey W. Swanson, a professor of psychiatry and behavioral sciences at Duke University, has found that since mental-health records have become accessible as part of the instant-check system in Connecticut, violent crimes by those with serious mental illnesses, disqualifying them from owning guns, have declined sharply in that state. That is in comparison with violent crimes by a natural control group—those with serious mental illnesses who were not disqualified from owning guns (because they committed themselves to a mental hospital rather than waited for a judge to order them there).

While serious mental illness constitutes a small part of the gun-violence problem, it figures more prominently in rampage shootings. But federal support for upgrading records has dwindled in recent years. Reviving that program would be relatively uncontroversial and would yield payoffs well beyond the domain of gun violence.

A much more ambitious proposal would be to expand the requirement for background checks as part of gun transfers. Currently the 40 percent or so of all sales that are conducted informally, at gun shows or elsewhere, are not subject to the federal requirement for a background check. That gaping barn door of a loophole may account for the disappointing effects of the instant-check system for sales by dealers.

Jens Ludwig, co-director of the University of Chicago Crime Lab, and I evaluated the Brady Act, the federal law that first required every state to carry out such checks on sales by dealers, and could find little evidence that homicide or suicide rates had been reduced.

A universal background-check requirement is not that radical an idea—in fact, California state law includes just such a mandate. Most transfers between private parties must go through a dealer, who then performs the check and registers the gun. Does that requirement save lives? Direct evidence of its effectiveness is lacking, which does not mean that it is ineffective—only that it is difficult to evaluate.

Strengthening enforcement to combat illicit gun carrying and misuse is less controversial. In fact, here the evidence is quite strong; it indicates that policing focused on guns can be effective in reducing gun violence on the street.

Among the tactics that seem promising are proactive, gun-oriented policing to deter illicit carrying and programs that adopt “retail” deterrence, a strategy developed in 1995 as Boston Ceasefire. In that program, created by the criminal-justice professors David M. Kennedy and Anthony A. Braga (at John Jay College of Criminal Justice and Rutgers University, respectively) and others, gang members are personally informed by law-enforcement authorities that gun misuse by any member will result in punitive consequences for the entire gang.

Braga and others have evaluated several such programs, with promising results. It is “just” a matter of resources, but resources are in short supply in Detroit, Newark, and other cities where gun violence is epidemic. Restoring funds for a Clinton-era program, Community Oriented Policing Services, which offers federal support for local police hiring, would easily pass a cost-benefit test.

Incidentally, it is not surprising that direct evidence of the effects of specific regulations or programs is lacking. The quest for evidence-based policy related to gun control is commendable but sometimes beyond reach. Usually it is not possible to run randomized controlled trials, and in any case, the impact of those policies may be delayed and diffuse—and hence hard to detect.

The available evidence, then, rests on some combination of direct evidence and (more commonly) logical extrapolation from broader findings. One well-established finding is that guns are intrinsically more lethal than other commonly used weapons. It is not just the intent of the assailant that determines whether the victim lives or dies, but also the type of weapon—and a gun is far more likely to kill than is a knife or a club.

Since the debate over gun control is being waged partly in bumper-sticker slogans, it might be appropriate to summarize that evidenced-based conclusion like this: Guns don’t kill people, they just make it real easy. The implication is that programs that are effective in reducing gun use in assaults and robberies will save lives.

Philip J. Cook is senior associate dean for faculty and research at Duke University’s Sanford School of Public Policy. This commentary was originally published in The Chronicle of Higher Education.