Summary of Findings

While overall homicide levels in the United States have fluctuated minimally in recent years, those involving young victims and perpetrators—particularly young black males—have surged.

From 2002 to 2007, the number of homicides involving black male juveniles as victims rose by 31% and as perpetrators by 43%. In terms of gun killings involving this same population subgroup, the increases were even more pronounced: 54% for young black male victims and 47% for young black male perpetrators.

The increase in homicide among black youth, coupled with a smaller increase or even decrease among their white counterparts, was consistently true for every region of the country and nearly all population groupings of cities. The pattern also held individually for a majority of states and major cities.

After some decline during the 1990s, the percentage of homicides that involve a gun has increased since 2000, both among young white offenders and black offenders of all age ranges. The percentage of gun homicides for young black offenders has reached nearly 85%. These trends are concomitant with various legislative initiatives at the federal level that have lessened the extent of surveillance on illegal gun markets.

Time-of-day patterns of violent crime victimization for youngsters, ages 6-17, reveal clear differences between school days and out-of-school periods. On school days, the risk spikes during the after-school hours—the primetime for juvenile crime—while the late evening hours are most problematic on non-school days, particularly summertime weekends.

Future demographics suggest that the concern for at-risk youth should increase over the next decade. The number of black and Hispanic children should continue to expand, contrasting with the rather limited increase expected among Caucasian children. There is a significant need for reinvestment in children and families—in essence an at-risk youth bailout during these difficult economic times.

Federal support for policing and youth violence prevention has declined sharply in recent years, perhaps precipitated by complacency brought about by the significant 1990s decline in crime. The resurgence in homicide, especially among minority youth, signals the importance of restoring federal funds for crime prevention and crime control.

At least on the surface, the news from the crime front has seemed encouraging. The FBI, in its recent release of crime figures for the nation, reported that violent crime in 2007 was down across the board compared to 2006, including a 1.3% decline in murder. Although welcomed and widely applauded, the news contrasts sharply with the experience of countless Americans living (and some dying) in violence-infested neighborhoods—those for whom the frightening sound of gunfire is a far too frequent occurrence.

It is not that the FBI figures tell an inaccurate story about crime trends in America. Rather, they obscure the divergent tale of two communities—one prosperous and safe, the other poor and crime-ridden. The truth behind the fears and concerns of the nation’s under-classes about crime and violence lies deep beneath the surface of the FBI statistical report.

Recent Homicide Trends

Over the past few years for the nation as a whole, rates of violence, and homicide in particular, have been relatively stable, with rather modest fluctuations since the beginning of the decade. Exploring deeper, this is generally the case as well for whites, black females, and adult black males over the age of 25. But the picture for young black males, especially teenagers, is radically different.

Figures 1-2 display the percentage change over the past five years in the estimated number of homicide victims and offenders (see technical notes on p. 8) among selected groups (males, male juveniles, black male juveniles, and black male juveniles involving a gun). As shown, between 2002 and 2007, the number of homicides involving black male juveniles as victims grew by 31% and as perpetrators by 43%. In terms of gun killings involving this same population subgroup, the increases were even more pronounced: 54% for young black male victims and 47% for young black male perpetrators.

Greater detail pertaining to the number of victims and offenders by age, sex and race—separately and in combination—is provided in Tables 1-3 for each year since 2000. In order to lessen the effects of the volatility from year-to-year in measuring trend, the percentage change rates included in these and other tables compare pooled counts for 2006 and 2007 against a baseline of pooled counts for 2000 and 2001. By this gauge, the number of males committing homicide has increased, particularly for young black males (14-17 and 18-24) and especially involving a gun.

Moving below the national level, the increase in homicides by black youthful offenders is consistent for all nine geographic regions and nearly all population subgroups, as shown in Tables 4-5, suggesting that the problem is not limited to only certain parts of the country. Moreover, these increases contrast with smaller increases and even decreases among white youth.

Finally, Tables 6-7 provide similar percentage change calculations, separately for white and black youthful offenders, for states as well as cities with populations over 500,000 and more than 25 overall homicides annually. Given the smaller base figures at the local level, the change rates are more volatile. Overall, however, a majority of states and a majority of cities have experienced increases in homicides committed by young black offenders compared with smaller increases or even decreases among their white counterparts.

Long Term Homicide Trends

While recent increases in homicides involving young offenders, particularly black males, are of significant concern, when compared to the longer term trends of the past few decades, a different perspective emerges. Tables 8-9 and Figures 3-4 contain the rates of victimization and offending per 100,000 population for males by age and race from 1976 to 2007. The recent surge in homicide among young black males clearly falls far short of the extraordinarily high levels witnessed during the crack-related street gang wars of the late 1980s and early 1990s.

In fact, the recent increases may say more about the success of crime prevention and crime control efforts of the past decade than about contemporary failures. In essence, the recent apparent spike in violence indicates that the nation is victim of its earlier success. Were it not for the 1990s downturn, recent figures would hardly stand out as cause for alarm.

Actually, there are naturally-occurring cycles to crime rates. Although not as firm and deterministic as Newton’s law of gravity, when it comes to the crime rate, what goes up, generally comes down, and what goes down generally rebounds. While no level of victimization can be termed “acceptable,” compared to the early 1990s when a deadly mix of gangs, guns and emerging crack markets fueled an unprecedented surge in violence, the current state is not out of control. Reinvestment in the programs and strategies that worked successfully in the past, along with restored funding levels for policing and prevention, can reverse the current spate of street and gang violence.

The Role of Firearms

The role of firearms in the recent increase in youth killings, shown in Figure 5 (and Table 10) is particularly significant and noteworthy. The percentage of homicides involving a gun has risen to nearly 85% among young black offenders, matching the high-point reached during the early 1990s.

The percentage of gun homicides for young white offenders has also grown in recent years, though not quite equaling the level seen during the early 1990s. While the role of guns in homicides committed by older white offenders has continued its steady decline of the past few decades, gun use among black offenders over age 25 has rebounded in recent years.

The especially prominent upturn since 2000 in gun homicide coincides with legislative restrictions upon ATF regarding the dissemination of gun tracing information and other pro-gun legislation that passed through Congress early in this decade. Congress has passed amendments in recent years making it more difficult to identify illegal market sources of crime guns through ATF data.

It is noteworthy that increases since 2000 in gun killings by young offenders have occurred as levels of non-gun homicide remained relatively flat or even decreased. This divergence suggests the need to rethink our nation’s approach to reducing availability of firearms to young offenders—those who are more apt to pull the trigger, even over trivial matters, without fully considering the consequences for themselves, much less for others.

Primetime for Juvenile Crime

Regardless of trend, be it upward, downward or stable, the concern for the safety of children is genuine and critical. With parents spending less time supervising their children—some out of choice, others out of necessity for the sake of managing expenses, and a few out of sheer indifference or negligence— an increasing number of youngsters are unsupervised during out-of-school hours. Poor supervision, combined with idleness and boredom, is a recipe for trouble. Far too many youngsters, therefore, are especially at-risk during the after-school hours for a range of problems, such as violence, as well as drinking, drug use, and teen pregnancy.

Figures 6-7 display the time-of-day patterns of violent victimization for 2006 among juveniles, ages 6 through 17, separately for the months between September and June when school is in session and the two summer months of school vacation. Clearly, the incidence of victimization peaks in the after school hours—the primetime for juvenile crime—when many parents are working and kids are often unsupervised, and then begins to tail off in the evening hours when parents typically are home to monitor their children.

Weekend days during the school year reflect a very different pattern in which the evening hours are more problematic. The summer months reveal patterns that are close to that for weekend days of the school year. However, the pronounced peak in the late evening hours of weekend days in the summertime warrants special attention in terms of providing constructive programs and alternative forms of supervision.

Attraction of Gangs

Notwithstanding the tale of official crime statistics, it hardly takes a rocket scientist—or a research criminologist—to recognize that there are increasing numbers of wayward and poorly-supervised youngsters with guns in their hands and gangs in their plans. Regrettably, as the nation celebrated the successful fight against violent crime back in the 1990s, we grew complacent and eased up on our crime-fighting efforts.

Unfortunately, the crime problem and the gang problem do not disappear, and rebounded once we shifted priorities elsewhere. Unless we restore the sense of urgency, some day we may look back and call these the “good old days.”

Even while targeting gangs for intensive enforcement, we need also understand their special appeal. Gangs offer youngsters many desirable advantages—status, excitement, power, praise, profit, protection, mentoring, and opportunity for advancement—healthy goals fulfilled in unhealthy ways. Today’s youngsters who are drawn to gang membership are too young to have witnessed the gang wars of the early 1990s when joining a gang could mean an early grave.

Our challenge, therefore, is to identify and promote healthier means for youngsters to achieve the same need-fulfillment, constructive ways to feel good about themselves and their prospects for the future, while at the same time having fun. This, of course, is where programs like the Boys and Girls Clubs and other youth enrichment initiatives play a significant role, and a role that, given ongoing trends, needs to be expanded.

While many Americans rail on about underage, under-prepared, and under-motivated parents “who just need to do a better job of raising and supervising their children,” we recognize that these families cannot do it on their own. We must assist families, not assail them, when they become overwhelmed with the day-to-day struggles of raising children, particularly during an economic downturn. The alternative forms of supervision and mentoring are extraordinarily critical.

Future Possibilities

The fact that the problem of youth violence, especially among minorities, has emerged and persisted for several years suggests that it is hardly an aberration or statistical blip. Moreover, it could worsen in the years ahead as the population of at-risk youth (blacks and Hispanics) grows as a result of both demographic patterns and immigration. Figure 8 shows projected trends in the numbers of young children—infants and toddlers under age 5—over the next decade, using the 2008 counts as a baseline. While the number of white children should change minimally, the pattern is remarkably different among race and ethnic minorities.

The number of black children is projected to grow in the years ahead. Growth in the number of Hispanic children, partially tied to immigration patterns, is especially pronounced. Given the social and economic strains that unevenly impact minority communities, growth in the population of at-risk youth signals the clear potential for increased problems of homicide, violence and other social ills associated with an expanding population of underclass youngsters.

Whether these demographic trends translate into increasing crime problems ahead largely depends on our willingness to be proactive. The urgency is clear: we must reinvest in children—not just for the purpose of crime prevention, but for the wide range of benefits associated with promoting their healthy and successful development.

Untimely Budget Cuts

Lulled into complacency by the sharp decline in crime during the 1990s, our nation’s priorities appear to have moved away from fighting street crime. A triple whammy at the federal level—related to cops, guns and kids—has hampered proven strategies for crime control.

Federal appropriations in support of law enforcement have been slashed since the early part of the decade, as reflected in Figure 9 in relation to funds for the Community Oriented Police Services (COPS) initiative as well as the Byrne Justice Assistance Grants (JAG) program. In addition, federal support for juvenile justice and prevention programs (Juv Just) has been reduced by half, now a shadow of the former investment.

Drastic funding cuts have led to a significant reduction in police resources among large cities, those with populations of more than 250,000. Specifically, as shown in Figure 10, the number of police officers per 1,000 population for large cities has dropped 8.5% since the year 2000, with most of the decline occurring in the first few years of the decade. By contrast, the level of police protection in cities with populations under 250,000 has remained virtually constant.

Of course, much of the decline can be traced to the changing priorities following the September 11, 2001 terrorist attack on America. Much of the federal support for law enforcement shifted from hometown security in patrolling high crime neighborhoods to homeland security in protecting the nation’s transportation, government and financial centers. Regardless of the level of terrorist threat, however, many more Americans are murdered each year by gunfire than were killed on 9/11. While the focus on combating terrorism is undeniably important, we cannot lose sight of the carnage taking place on our city streets.

In these difficult economic times, the banking and automobile industries have looked to the federal government for assistance. Where is the voice to appeal on behalf of the needs of at-risk youth, as various support programs for children and families are being slashed? We need an at-risk youth bailout.

Principles of Prevention

Unfortunately, not all Americans are convinced about the value of prevention—especially early childhood and youth enrichment efforts. As a result, prevention initiatives are too often funded and implemented on a shoestring, and a rather short shoe-string with a brief window of opportunity to show results. This is a recipe for failure and provides additional fodder for skeptics.

Smart crime fighting involves a balanced blend of enforcement (from community policing to identifying illegal gun markets), treatment modalities (from drug rehab on demand to community corrections and post-incarceration services) as well as general and targeted crime prevention (from family support to summer jobs for high-risk youth).

Regrettably, the prevention approach has at times been disparaged as “worthless” and as “soft of crime.” Yet, this cynical perspective reflects gross misunderstanding of the process and goals of prevention, and a selective examination of outcomes. Simply put: Prevention programs can work; good prevention programs that are well implement do work.

Besides the matter of funding adequacy, five fundamental principles of crime and violence prevention are critical for effective investment:

1. No program is successful all the time or for all individuals. Regardless of the initiative, there will be failures—those who commit crimes or recidivate despite best efforts to prevent it. Rather than focusing on the failures, the goal should be a reasonable reduction in offending rates. In light of the enormous social and administrative costs associated with each criminal act, even modest gains are worthwhile.

2. Prevention should have an emphasis on the prefix “pre.” While it is unwise and inappropriate to “give up” on even a seemingly hardened offender, the greatest opportunity for positive impact comes with a focus on children— those who are young and impressionable and will be impressed with what a teacher, preacher or some other authority figure has to say. It is well-known that early prevention—during grade school if not earlier—can carry the greatest and lasting impact, before a youngster is seduced by gangs, drugs and crime.

3. Patience is more than a virtue, it is a requirement. Prevention is not a short-term strategy. Rather, it involves a continued effort, undaunted by setbacks. Unfortunately, many prevention programs are given short window periods in which to show progress, and are often terminated before the final results are seen.

4. Prevention should take a multi-faceted approach. Understandably, there is much temptation to target gang activity as perhaps the most visible and immediate threat to public safety. While the focus on anti-gang initiatives is laudable and should be strengthened, there are many other points of intervention for successful crime reduction programming.

For example, several proven and promising strategies are directed at at-risk families with young children. Rather than criticizing struggling underage mothers for their lack of parenting effectiveness, many programs support them in raising children who are less likely to become juvenile offenders. In addition, many school-based initiatives effectively and efficiently enhance the well-being of large number of children.

Behavioral skills training at the elementary school level (such as the modules developed by Boston’s Lesson One Foundation), anti-bulling curricula for middle school students (such as the Olweus bullying prevention program) that recognize the link between bullying and later offending, peer-mediation and mentoring program in high school, after-school activities targeted at the “prime time for juvenile crime” (such as the Boys and Girls Clubs) all have payoffs far greater than the investment.

5. Prevention is significantly cost-effective. Virtually all assessments of crime prevention confirm the adage that an ounce of prevention is worth a pound of prison time. It is, however, a political reality that sound investments in crime prevention can take years to reap the benefits. For example, the Perry preschool program experiment implemented in Ypsilanti, Michigan translated into a 17-to-1 rate of return on investment, yet it wasn’t until years later when the preschoolers matured that their significantly lower involvement in crime, alcohol and drug abuse was observed. It takes a bold leader to earmark funds today for tomorrow’s success that his/her successor will derive.


The current surge in youth violence was anticipated years ago. Even while rates of crime were falling in the 1990s, criminologists warned about the potential for another wave of youth and gang violence ahead, a not-so-perfect storm combining an upward trend in the at-risk youth population with a downward trend in spending on social and educational programs to support youth.

Furthermore, we should not be surprised if the concomitant increase in the number of at-risk youth, especially black and Hispanic children with less than adequate supervision, combined with recent budget cuts for youth programs and crime control initiatives, translates into further increases in gang and gun violence. We’re already seeing the early signs.

The good news–or at least the encouraging word–is that the crime problem is not out of control, at least by contrast to the early 1990s when the nation’s murder rate was almost twice what it is today. It is not surprising that a small bounce back would occur after the glory years of the late 1990s. But let this small upturn serve as a thunderous wake-up call that crime prevention needs to be a priority once again.

At this juncture, we must, of course, look toward immediate solutions for controlling gang activity and easy access to illegal firearms—approaches that depend heavily on police personnel, intelligence, and deployment. At the same time, however, we must maintain a long-range view toward the future as the population of young children—especially race and ethnic minorities—grows. The choice is ours: pay for the programs now or pray for the victims later.

Data Sources and Technical Notes

Several data sources were used in completion of this report. Most prominently, the data on homicide victimization and offending come from a multiply-imputed cumulative file of the Supplementary Homicide Reports for the years 1976-2007, created by the authors. Compiled as part of the FBI’s Uniform Crime Reporting (UCR) Program, the Supplementary Homicide Reports (SHR) include incident-level data on the month and year of the offense; on the reporting agency and its residential population, county, Metropolitan Statistical Area (MSA) codes, geographic division, and population group; on the age, sex, and race of all victims and offenders; and on the victim–offender relationship, weapon use, and circumstances of the crime.

The SHR records are incomplete on a small percentage (2.5%) of victims, yet a substantial percentage (32.5%) of perpetrators largely as a result of unsolved offenses. However, a process of multiple-imputation was employed to fill in the gaps (based on available information about the incomplete reports) so that characteristics of victims and offenders can be reliably and accurately estimated.

In addition, approximately eight percent of homicides are not covered in the SHR data files. To adjust for under-reporting, adjustment weights were applied based on comparisons to mortality data from coroners’ reports and to the aggregated homicide counts reported by the FBI in the Crime in the United States series. Aided by this imputation and weighting process, all counts and rates, particularly for offender data, are estimates, although reliable ones.

Calculations of the time-of-day distribution for violent crime victimization for school-age youngsters (ages 6 through 17) were based on National Incident- Based Reporting Data (NIBRS) for 2006 archived at the University of Michigan. Although NIBRS coverage is not nationally representative, there is little reason to believe that the time patterns are adversely affected by gaps in NIBRS coverage. In fact, state-by-state analyses of these time patterns reveals general consistency across various parts of the country, providing support for the assumed representativeness of the sample data with regard to time-of-day distributions.

Homicide rate calculations and demographic projections relied on U.S. Census Bureau annual estimates of resident population by age, sex and race. Available race-bridged estimates were used to enable a smooth transition between the multiple-race classifications of the 2000 Census counts and the singular-race designations of earlier Census counts.

Finally, data on police personnel were drawn from figures published annually by the FBI in Crime in the United States. Information on federal appropriations for justice-related programs was drawn from figures compiled and reported by the National Criminal Justice Association, a Washington, D.C.-based justice policy organization.