Jeremy Travis

September 10, 2008

TESTIMONY OF JEREMY TRAVIS
PRESIDENT OF JOHN JAY COLLEGE OF CRIMINAL JUSTICE
HEARING BEFORE THE SENATE JUDICIARY COMMITTEEE ON
"New Strategies for Combating Violent Crime: Drawing Lessons from Recent Experience"

September 10, 2008

Chairman Leahy and members of the Senate Judiciary Committee:

It is indeed an honor to be invited by this Committee to offer testimony on the issue of violent crime in America. I am doubly honored to be sharing the panel with these distinguished experts from academia, law enforcement, and communities hard hit by violence.

My testimony this morning consists of two parts. First, I propose to offer some perspectives on the phenomenon of violent crime in America, with a particular focus on ways to place our current rates of violence into appropriate and useful contexts. Second, I would like to make some recommendations about the appropriate federal response to the phenomenon of violent crime, specifically an agenda for the new Administration and new Congress that will take office in January 2009.

Perspectives on the Level of Violence in America.
As this Committee is well aware, over the past twenty years our nation has experienced a dramatic rise and fall in the levels of violence in our communities. In his presentation, Professor Blumstein, who has chronicled changes in crime rates in America for many years, documented that the rates of robbery and murder, as measured by the Uniform Crime Reports (UCR), experienced a sharp increase beginning in 1985, then peaked in 1993, and dropped dramatically until 2000, when the rates of both crimes basically leveled off. Setting aside for a moment some year-to-year fluctuations since 2000, we can confidently say that we now experience the lowest levels of violence in a generation.

This new reality obviously constitutes very good news for the nation. We need only remember the very scary atmosphere of the late 1980s -- when violent crime rates were rising rapidly, the introduction of crack cocaine to urban America was destabilizing inner city communities, and commentators announced the emergence of a generation of "super-predators" and warned of a "coming blood bath" - to place the current level of safety and security in proper perspective.
The good news of the unprecedented drop in violence has led to a predictable search for explanations - Why did this happen? What factors contributed to this turn-around? A number of academics, most prominently Dr. Blumstein, have tested various hypotheses, including the strong economy of the late 1990s, the expansion of our prison population, the emergence of crack markets, gun control policies, new policing strategies, demographic shifts, etc. Certainly we need to understand the factors that led to the decline in violence to craft policies to reduce those rates even further.

Rather than enter into the debate over which factors contributed to the decline or speculate as to the changing nature of violence in America, I would prefer to focus the Committee's attention on a question often overlooked in today's discussions namely how should we view the current rates of violence?

In my view, we should not be complacent, for one minute, about the current rates of violence. Yes, we are justifiably proud that our nation no longer experiences the high rates of violence seen in the early 1990s. We should celebrate the fact that homicide and robbery rates are below their 1970 levels. Yet, three different perspectives on these national data should give us reason to set our sights much higher. We have no reason to be complacent, and every reason to implement policies that will bring our rates of violence much, much lower.

International Perspective.

While the United States no longer leads the developed world in all forms of violence and property crime, it still has the highest levels of lethal violence. Even after U.S. homicide rates fell by more than 40% during the 1990's, they remained four-to-ten times higher than those of other developed nations. For example, the latest available data on homicide from 2006 show that the homicide rates in the United States (5.7 per 100,000) are more than four times the homicide rates of England and Wales (1.4 per 100,000). The distinguishing characteristic of violence in America is the widespread availability of illegal firearms that are used in the commission of crimes. If we aspire to bring our homicide rates lower, and to provide a level of safety approaching that seen in other countries in the developed world, we need for focus on strategies that reduce the illegal use of firearms.

Sub-national Perspective. We typically measure crime rates at the national level and ask whether property crime and violent crimes are up or down across the country. For many years, these national trends in turn reflected sub-national trends. In other words, if crime went up - or down -- nationally, it likely went up--or down-- in all cities. The increase or decrease may have been sharper or flatter in any given city, but the trends were mostly in the same direction.

Beginning in 2000, this relationship between national and sub-national trends began to weaken. We need only look at some recent examples to illustrate the point. As Dr.
Blumstein pointed out, according to the Uniform Crime Report, homicide rates increased slightly in 2005 (1.8%) and 2006 (1.8%), and robbery rates increased in both years as well (3.0% and 6.1%). Yet these national statistics mask important local variations. Between 2004 and 2006, homicides decreased by 25% in Dallas and 31% in Portland, and increased by 23% in Philadelphia and 25% in Seattle. Robbery rates were essentially flat over those two years in New York and Los Angeles, but increased 44% and 63%, respectively, in Milwaukee and Oakland.
We do not yet have a good understanding of the reasons for these very different crime trends at the sub-national level. But the fact that we are seeing these divergent trends underscores two points. First, in those communities experiencing upward trends in violence, the fact that the national trends are showing only slight increases present little comfort. Second, any national strategy adopted by Congress and the new Administration must include a robust analytical capability to diagnose these local trends, and must target resources to communities where the rates of violence are highest.

Inner City Perspective. A third perspective is perhaps the most important as we consider future directions for policy. We know that crime does not affect all Americans equally. Crime is concentrated in urban America, and particularly in the poorest urban neighborhoods, which are typically communities of color. Furthermore, violent crime is most often committed by, and committed against, young men. So, within this demographic group, of young men living in America's urban neighborhoods, violence is a daily fact of life. Allow me to cite two studies that illustrate this point:

Rochester, NY, has one of the highest homicide rates among the cities in New York State. Beginning in 2001, a team of local and federal law enforcement agencies, working with academics and community groups, conducted an analysis of homicides in Rochester as part of the Strategic Approaches to Community Safety Initiative (SACSI) of the Department of Justice. This analysis, carried out by Professor John Klofas of the Rochester Institute of Technology, found that violent crime was concentrated in a core urban area he called the "high crime crescent."

Professor Klofas calculated the homicide rate using a simple methodology that we should replicate in every city across the country. At the time of his research, the homicide rate for the nation as a whole was 8 per 100,000. Among those aged 15-19, it was nearly triple that: 22.4 per 100,000. Among males in that age group, it was more than quadruple the national rate, or 36.3 per 100,000. For black males in that age group, the national rate was 147 per 100,000, yet for black males aged 15-19 in Rochester, it was 264 per 100,000. And for black males aged 15-19 in the high-crime crescent, the homicide rate was 520 per 100,000, or 65 times the national rate. This means, nearly incredibly, that one in 200 young black men was killed in the "high crime crescent" every year.

Dr. Klofas then calculated the ripple effects of homicides in the "high crime crescent." Assuming that for each homicide victim, five friends were affected by that murder (a conservative assumption), Klofas calculated that 6.2% of the young African-American men in those neighborhoods lost a friend to homicide each year. For the rest of Rochester, homicides affected only .1% of the population.
Cincinnati, OH, provides a second illustration of the importance of looking below the national data. This city has long been plagued by high levels of violence. Last year, a group of police officials, public health officials, civic leaders and business representatives came together to launch CIRV, the Cincinnati Initiative to Reduce Violence. Prof. David Kennedy, Director of the Center for Crime Prevention and Control at John Jay College, with colleagues from the University of Cincinnati, the research partner for CIRV, conducted an analysis of the patterns of homicide in Cincinnati. According to their analysis, 48 high-rate offending groups - drug crews, "gangs," and the like - with around 1100 members total were involved as offenders and/or victims in nearly three-quarters of the homicides in Cincinnati.

The studies from Rochester and Cincinnati underscore three important points that are relevant to the deliberations of this Committee: (1) the phenomenon of violence in America is concentrated in a small number of neighborhoods; (2) a significant share of the violence is committed by, and against, a small number of young men living in those neighborhoods; (3) within these communities and subpopulations, the levels of violence are dramatically higher than the national experience - in Rochester, by a factor of 65. The national data about violence in America do not tell this story, but I believe this is the central story. If we want to produce a safer nation, advance an urban development agenda, and provide equal opportunities for Americans from minority groups, then we must bring these levels of violence down.

Recommendations for the new Administration and new Congress.
I am humbled by the opportunity to present my thoughts on new crime policies to be adopted by the incoming Administration and Congress, and I applaud this Committee for taking the initiative in paving the way. My recommendations fall into three categories: understanding the problem of violence in America; supporting proven interventions; and testing new ideas.
Understanding the Problem. Compared to virtually any other area of high policy interest in America, we have a very limited ability to track, analyze, and describe the phenomenon of violence. Our data from the Uniform Crime Reports are released months after the close of the year. Our National Crime Victimization Survey (NCVS) is conducted annually, but only at a national level - statistically, it cannot capture the realities of crime at the local level -- and is always struggling for adequate appropriations from Congress. The Arrestee Drug Abuse Monitoring (ADAM) program, which provides for quarterly interviews with individuals arrested and charged with crimes, has been cut back to ten cities from thirty-five, still far short of the goal of 75 established under the Clinton Administration. At the local level, police departments are making enormous strides to bring their reporting systems into the modern era, posting crime data on public websites, conducting geo-spatial analysis of crime reports, and using the internet to encourage crime reports, but at the national level we are still operating in a pre-internet, pre-GIS mindset.

The federal government should take the lead in designing and implementing a robust national crime data system that allows police executives, policy makers, elected officials, academics and other researchers, and community groups, to have a data-informed policy discussion about crime trends and effective responses.

Although the exact contours of such a program would necessarily depend on a process of design specification and consultation, I would suggest that such a program include, at a minimum:

?Rapid collection and dissemination of standardized police reporting data on crime, so that every month we would know whether crime rates were increasing or decreasing in every major jurisdiction across the country.

?Funding for annual local victimization surveys, using standardized survey designs, so that we could also track citizens' experiences of crime, independent of the police data. These victimization surveys should also include questions on citizen-police interactions, perceptions of fear, and attitudes toward the justice system, so that we know whether the agencies of our justice system are meeting citizens' expectations.

?Full funding of the ADAM system, expanding from the current 10 cities to at least 75 major cities, so that we can track changes in offender behavior, drug markets, illegal gun distribution, and gang dynamics.

?Funding of an analysis of gang dynamics, similar to that undertaken in Cincinnati, in those jurisdictions that are equipped to use that analysis to carry out the violence reduction strategies pioneered by Prof. Kennedy (see below).
Our goal should be to create a robust crime analysis capability at the national level, just as we have a national capability to understand fluctuations in unemployment rates, housing starts, or business cycles. As this statistical capacity is brought to scale, the federal government should significantly increase its investment in research to analyze the changing nature of crime in America, at the national, regional and local level. This robust analytical infrastructure would then provide the platform for the development of targeted violence-reduction strategies that focus federal, state and localattention and resources on the communities in America that are experiencing high rates, and increasing rates, of violence.

Supporting Proven Interventions. Over the past fourteen years, I have been particularly impressed by the violence reduction strategies pioneered by Prof. David Kennedy, formerly at Harvard's Kennedy School of Government and now at John Jay College as Director of our Center on Crime Prevention and Control. When I was Director of the National Institute of Justice, we funded Prof. Kennedy's work developing a strategy called Operation Ceasefire, that led to the "Boston Miracle," a stunning two-thirds decline in youth homicide. By bringing together local, state and federal law enforcement with community leaders, clergy and service providers, Operation Ceasefire directly engaged the young people who were engaged in the violence, offered them a way out of their anti-social behavior, engaged the positive forces of the community in establishing new community norms, and promised and delivered formal law-enforcement sanctions where violence continued.
This strategy has since been replicated in dozens of jurisdictions across the country, with similar results. In Indianapolis, homicide was reduced by more than a third city-wide. In Chicago, homicide was reduced by 37% in some of the most violent neighborhoods in the city. Most recently, in Cincinnati, the CIRV initiative, previously mentioned, reduced homicide associated with violent groups by about half.

These strategies have earned national acclaim. The Boston Ceasefire model was awarded the prestigious Innovations Award by the Kennedy School of Government and the Ford Foundation. Under Attorney General Janet Reno, the Boston strategies were replicated in ten jurisdictions under the name of the Strategic Approaches to Community Safety Initiative (SACSI). The national evaluation of SACSI, recently published by the Department of Justice, concluded that SACSI was "associated with reductions in targeted violent crimes, sometimes by as much as 50 percent." Under the Bush Administration, these approaches were embraced by Project Safe Neighborhoods, a national anti-crime initiative.

Following these successes in reducing violence, Prof. Kennedy then applied a variant on these strategies to the issue of overt community drug markets, with similar successes. In High Point, NC, which was the first test site, and is represented here by Rev. Reverend James Summey of the English Road Baptist Church, the strategy shut down the worst drug market in the city virtually overnight more than four years ago, with a sustained neighborhood reduction in serious crime of more than 40%. As important, the African-American community in High Point, and other sites that replicated the High Point model, including Providence, RI, represented today by Colonel Esserman, has witnessed a more open, trusting and collaborative relationship between the African-American community and the police. The ABC news program "Primetime" recently highlighted a parallel intervention in a drug market in Hempstead, Long Island, which resulted in a 75% drop in serious crime; I have submitted a copy of that program with my written testimony.
These proven innovations should be brought to national scale, with national leadership. The Boston Ceasefire and High Point strategies represent important breakthroughs because they focus directly on the most pressing manifestations of violence in our country in the communities that are most directly affected. Not surprisingly, there is enormous demand across the country for technical assistance and training in these strategies. I am pleased to note that Kennedy's drug market strategy has recently been embraced by the Justice Department under the Bureau of Justice Assistance's Drug Market Elimination Program (DMIP), and during the first week of July 2008, the Providence Police Department served as host for a BJA-sponsored training conference for 9 jurisdictions. Yet the demand for assistance far outstrips our capacity to meet the demand. And, more importantly, an approach that simply relies on a technical assistance model - working only with a small number of jurisdictions as expert consultants - fails to realize the enormous potential of these new approaches to violent crime.

Prof. Kennedy and I have developed a proposal for a "The National Safety Network" that would achieve four ambitious goals.We believe it is possible to simultaneously reduce violence, abate drug markets, reduce our reliance on incarceration, and promote better relationships between the police and minority communities. Whether through this proposal or a variant, we should build upon this record of federally-supported innovation, with its strong evaluation results, and bring down rates of violence in communities that are suffering. Police agencies around the country are facing enormous pressures to respond to the levels of violence highlighted at this hearing. In my view, the federal government has an obligation to provide leadership in this area, as it has in the past, through targeted allocation of scarce federal dollars. Our highest priority should be to provide effective assistance to those communities facing the highest rates of violence.
Testing New Ideas. When I was Director of the National Institute of Justice, I invited Dr. James Q. Wilson to deliver a lecture on crime policy issues to a large, broadly representative audience of policy makers, researchers and practitioners. He chose as his topic, "What, If Anything, Can the Federal Government Do About Crime?" His answer was instructive. The federal government's role in the arena of crime policy is necessarily limited, he argued, because law enforcement and criminal justice policy is so much the province of state and local government. But, he argued, the federal government should test new ideas, and help jurisdictions embrace those ideas with proven success. The federal government, he posited, should support the creation of a robust "Research and Development" capability for the nation.

We have many examples of successful federal leadership along these lines. The 1994 Crime Act developed with the leadership of this committee supported innovations in policing through the community policing initiative. It also promoted drug courts, new multi-sector responses to violence against women, advances in the use of DNA technology and other forensic science investigative techniques, crime mapping, and responses to sex offenders. At its best, the federal government tests new responses to critical and emerging problems facing the criminal justice system, evaluates those new interventions rigorously, and then disseminates successful models for use by state and local agencies.

The crime and justice challenges facing the country today are enormous. In this statement, I have outlined an approach to a federal strategy for promoting public safety in communities facing unacceptable levels of violence. I also believe strongly that the federal government should show leadership by testing new approaches on a variety of other pressing topics. How can we reduce the recidivism rate and promote the successful reintegration of the 700,000 individuals leaving prison, and the 12 million people leaving local jails, each year? How can we reduce the incidence of identity theft, which strikes millions of Americans each year? How can we reduce our reliance on incarceration, without sacrificing public safety, so that those resources can be redirected to communities experiencing high rates of crime? How can we improve our response to crime victims, so that they can rebuild their lives after the devastation of crime? How can we reduce the levels of violence against women, and the tragedy of abuse and neglect of children? How can we improve the level of trust and confidence in the justice system and the rule of law, particularly in communities of color that suffer the triple impact of high crime, high incarceration, and high rates of prisoner reentry? Bringing down rates of violence is clearly the top priority for the nation, but these other challenges are compelling, and also require national leadership.

I thank the Committee for the invitation to present these thoughts, and would be eager to provide further assistance if called upon.