Professor Alfred Blumstein

May 13, 2003

Senator Hatch, Senator Leahy, and Members of the Committee:

Thank you for inviting me to this session on the issue of gun violence and the efforts to deal with it through Project Safe Neighborhoods. I am honored by the opportunity to appear before you as you consider one of the continuing problems confronting our nation.

As background to my own involvement in these issues, I have engaged in a variety of criminological research since my involvement as Director of Science and Technology for the President's Commission on Law Enforcement and Administration of Justice in 1966. I have been involved in practical policy matters as the chairman for over eleven years of the Pennsylvania Commission on Crime and Delinquency (PCCD), the state's criminal justice planning agency, which manages Federal criminal justice funds in Pennsylvania. I also served as a member of the Pennsylvania Sentencing Commission for ten years between 1987 and 1997. Most recently, since 1996, I have been the Director of the National Consortium on Violence Research (NCOVR), a research consortium that has been particularly concerned with issues of gun violence. Attached to my testimony is a short biographical statement for your information.

Some Background on the Gun Violence Problem in the U.S.

As this Committee knows too well, the United States is by far the leading nation in suffering from gun violence. In 2001, we suffered 15,980 murders, not counting those of September 11, 2001. Of these, 63.4 percent or over 10,100 were by firearms . In the last few years, we have been elated that our homicide rate has finally dropped over 40 percent to a level below 6 per 100,000 from a high of just under 10 as a result of the steady decline of the 1990s . In contrast, however, most of the countries we compare ourselves with are much lower: The average of the EU member states is 1.7 and Norway, Switzerland, and Belgium are around 1.0. But, we can be comfortable that Russia, a country whose prison incarceration rate we recently exceeded, is still well ahead of us with a rate over 20 .

However advanced we are in murder, it is well known that we are especially the leader in gun murders. Frank Zimring developed a table that contrasts our murder rate with that of England and Wales, our closest comparison country. Our rate is 8.5 times theirs for all murders, only 3.7 times for non-gun murders, 63 times for gun murders, and 175 times for handgun murders . Thus, the U.S. has a powerful incentive to reduce the rate of gun violence, and especially that associated with handguns.

Considerations in Reducing Handgun Violence

In contrast to the great complexity of dealing with the nation's drug problem, where so much of what we do can be seen to be counter-productive, one would think that there should be some reasonably straightforward approaches to the problem of gun violence based on what we have learned over the past decade of research:

? We know that young people carrying handguns hands can be very dangerous
? We know that a growing presence of illegal guns stimulates more gun carrying as a defensive response, and that a declining presence of such guns can stimulate less gun carrying as a result of the reduction of the threat5
? We know that most crime guns were obtained illegally, either through their own straw purchaser or through an illicit dealer, and not largely through burglary
? We know that in any jurisdiction a small number of dealers can account for a large share of the crime guns
? We know that many crime guns are relatively new6.
? We know that a strong commitment to tracing of crime guns captured by the police in any community can be very helpful in understanding the flow of guns in local secondary markets and can help identify licensed dealers and straw purchasers who are disproportionately involved with the marketing of crime guns6
? We know that aggressive police patrols in high gun-use areas - as reflected in "shots fired" calls to 911 - can be effective in suppressing gun carrying and use. Jacqueline Cohen has analyzed such operations by the Pittsburgh police and shown striking suppression effects .
? We know from deterrence theory that more certain and more severe sanctions should reduce crime, but we are still uncertain about the magnitude of those effects for different types of crimes and different types of offenders. Also, we know that improvement in the certainty is more important than improvement in the severity - that is, increasing the risk of punishment through more effective policing and prosecution is more effective than increasing the sentence length .

All of this suggests that there are a few basic approaches that any community should consider:

? Crime-gun tracing and follow-up to identify the illicit secondary markets and to prosecute those involved
? Aggressive police pursuit of illegal gun carrying, especially in those neighborhoods with high rates of gun crimes and shots fired
? Deterrence of illegal gun carrying and especially use by enhancing the sanctions, with an emphasis on increasing the certainty that the sanctions will be imposed

When we have this limited range of options, then it becomes critical that we develop some coherent research program to evaluate the various versions of these approaches to come to an understanding of how well they work, how they interact with each other, and how that effectiveness varies with the crime and cultural context of different communities. This calls for a coordinated national program of research and evaluation to develop such knowledge.

My understanding of Project Safe Neighborhoods (PSN) is that no such coordinated program has yet been organized. Rather, each U.S. Attorney's office is mandated to link up with a "research partner," to evaluate its local effort, but that those individual program and evaluation efforts are largely independent and unconnected. That may result in some important local knowledge, but the size and the effect of each of those efforts is likely to be too small to generate much measurable effect. That is likely to be the case in all but the handful of jurisdictions with large gun-violence problems, and so we will miss the opportunity to generate the level of knowledge that is needed to provide the guidance for the future. I believe that the efforts to combat gun violence would be much better served if could devote at least a significant part (at least 5-10 percent) of the annual $300 million committed to PSN to the organization of such a program. Only with that kind of coherent program of evaluation will the nation be able to move forward with increasing effectiveness in combating the gun-violence problem.

That will require a central research planning and analysis office that will get agreement of the program actors (usually the U. S. Attorneys offices or the local agencies to which they allocate action funds) on a coherent multi-site action plan. That research coordination office will work with the action agencies in developing consistent research designs, measurements to be taken, and development of analysis protocols to ensure rich and completes assessment of the major approaches being taken. With PSN funding, NIJ could be the contracting agency to fund and oversee the operation of such an office.

Some Other Observations about PSN

This need for a coherent research program overlaying the PSN activities seems quite central in using the efforts to become more effective. A number of other considerations would also seem relevant:

? The funds for PSN are allocated to U.S. Attorneys' offices on the basis of their population. This seems to miss the point of the need based on the level of gun violence there. Some measure of gun violence (e.g., gun homicides, gun robberies, gun suicides, or some weighted sum of these per capita) would seem so much more appropriate.
? The resources for crime-gun tracing at both the local level and at ATF must be increased to accommodate the increased load that is likely to result.
? There are important technical issues that must be pursued nationally in support of the PSN efforts. For example, one of the continuing questions that arose during the Washington-area sniper shootings was the feasibility and utility of a national ballistics identification system. That assessment must be done nationally, by the National Academy of Sciences, for example.
? There will have to be regional and national coordination of the information derived from the crime-gun tracing information to identify the prime and consistent violators.
? We are still woefully ignorant of the mix of factors contributing to gun violence and how that mix varies across locality. The violent-injury reporting system being developed by CDC will be most important in developing such knowledge. We are tracking SARS around the world very carefully, but are doing so little in tracking violent injury, which kills so many more people in the United States.

Summary

Let me summarize the major points of my testimony. First, I agree that gun violence is an important issue to be addressed. While I accept some of the benefits of the decentralized strategy, I think this is an area where coordination to gain improved knowledge will be very useful, and I don't think we are pursuing that effort sufficiently. Also, while intensified prosecution after the crime has occurred is useful, I think the program could pay much more attention to deterring the carrying of guns on the street and pursuit of the markets supplying the illicit guns on the street should become more prominent aspects of the program.

Thank you for the opportunity to testify on this important issue. I will be happy to answer any questions.