Policy Review Banner

Arresting Ideas

Friday, September 1, 1995

Serious crime is declining in many big cities across America. That's the good news. Meanwhile, the country's largest and most violent cohort of young males will soon reach its crime-prone years. That's the bad news. But demography is not fate. Smarter law enforcement and tougher sentencing policies explain much of the recent drop in crime, and can minimize the damage from the next crime wave.

Between 1993 and 1994, the violent crime rate declined by 10 percent or more in eight of the 10 cities with the highest violent-crime rates (Miami, New York City, Los Angeles, Tallahassee, Baton Rouge, Little Rock, Jacksonville, and Pueblo, Colorado). In many cities, a sizable reduction in homicides accounts for much of the fall in these rates. For example, the number of murders in Atlanta, Chicago, and New Orleans together plummeted by 17 percent during the first half of 1995 compared with the same period a year ago.

New York City and Houston have enjoyed truly phenomenal drops in serious crimes, including murder. In 1992 and again in 1993, more than 1,900 homicides were committed in the Big Apple. But in 1994 New York City's murder count fell to 1,581. Through July 1995, it suffered fewer than 700 murders, and it continued to show declines of 10 percent or more in robberies, burglaries, and most other serious crimes. Likewise, the number of people murdered in Houston declined by 32 percent during the first half of 1995 compared with same period a year ago. Rapes in Houston decreased by 21 percent, robberies by 15 percent, and the overall violent crime rate by 7 percent.

While New York City and Houston are leading the pack, other cities are catching up. During the first half of 1995, for example, the overall crime rate was down by more than 16 percent in San Francisco, 10 percent in San Antonio, and 6 percent in both Los Angeles and Philadelphia. And the number of murders declined by more than 6 percent in Philadelphia and Los Angeles, 9 percent in Detroit, and 10 percent in Boston and St. Louis.

What is going on here? Some criminologists dismiss the recent improvement in the crime rate as a mere statistical fluke. But it is hard to imagine that these downward trends, occurring in consecutive years in given jurisdictions, could have happened by chance. Others insist that the slide in crime rates is greased by a dwindling population of teenage boys. There is something to this claim, but it ignores the inconvenient fact that Houston and some other places with growing populations of at-risk youth have nonetheless experienced sharp reductions in crime.

Finally, a few criminologists have rushed to relate the dive in crime rates to everything from a sudden surge in the efficacy of gun control laws (which is patently absurd) to changes in the patterns of drug use (for example, the decline in crack-cocaine use which, they insist, has had nothing to do with anti-drug law enforcement). One much-quoted criminologist has even declared, "What goes up must come down."

Bratton's Law: Enforcement Counts

In many cities, the decline in crime rates can be explained at least in part by law-enforcement efforts that capitalize on community crime-fighting initiatives and take bad guys off the streets. I call this explanation Bratton's Law in honor of New York City's police commissioner, William Bratton. Like most veteran professionals in the justice system, Bratton understands perfectly well that crime rates are not determined solely by what cops, courts, and corrections agencies do. But his impatience with criminological cant about the inefficacy of policing practices and sentencing policies on crime rates is both ennobling and enlightening. Three brief examples illustrate Bratton's Law in action.

  • Jacksonville. In July 1991, Harry L. Shorenstein became state attorney for the Fourth Judicial Circuit in Jacksonville, Florida. At that time Jacksonville was besieged by violent crime, much of it committed by juvenile offenders. In the year before Shorenstein arrived, juvenile arrests had risen by 27 percent, but most young habitual criminals were released quickly. Jacksonville's finest were doing their best to remove serious young criminals from the streets, but the rest of the system was not following suit.

    Then, in March 1992, Shorenstein instituted an unprecedented program to prosecute and incarcerate dangerous juvenile offenders as adults. In most parts of the country, juvenile criminals for whom the law mandates adult treatment are not actually eligible for state prison sentences and are routinely placed on probation without serving any jail time. But Shorenstein's program was for real. He assigned 10 veteran attorneys to a new juvenile-prosecutions unit. Another attorney, funded by the Jacksonville Sheriffs Office, was assigned to prosecute repeat juvenile auto thieves.

    By the end of 1994, the program had sent hundreds of juvenile offenders to Jacksonville's jails and scores more to serve a year or more in Florida's prisons. Jacksonville's would-be juvenile street predators got the message, and the effect of deterrence soon appeared in the arrest statistics. From 1992 to 1994, total arrests of juveniles dropped from 7,184 to 5,475. From 1993 to 1994, juvenile arrests increased nationwide and by over 20 percent in Florida. But Jacksonville had a 30 percent decrease in all juvenile arrests, including a 41 percent decrease in juveniles arrested for weapons offenses, a 45 percent decrease for auto theft, and a 50 percent decrease for residential burglary. Although Jacksonville still has a serious violent crime problem, the number of people murdered there during the first half of this year declined by 25 percent compared with the same period a year ago.

  • Houston. Almost a thousand officers have been added to the city's police force since 1991. Led by Police Chief Sam Nuchia, Houston has a cost-effective police overtime program that puts more cops on the street when and where they are most needed. Residents of Washington, D.C., which fields the highest number of police officers per capita of any major city, know that more police manpower does not necessarily produce less crime or better police performance. But in Houston, Nuchia has used the additional manpower to jump-start community anti-crime activities.

    To cite just one example, Houston's Citizen Patrol Program has operated in more than a hundred of the city's neighborhoods. Among other things, thousands of citizen patrollers have observed and reported suspicious or criminal behavior, from assaults to narcotics dealing to vandalism. Many once-troubled neighborhoods have gone as long as three consecutive months without needing to call for police service. Indeed, two recent studies found that Nuchia's enforcement efforts not only contributed to Houston's falling crime rates, but also improved police emergency response times, raised police productivity, and reduced citizens' fear of crime.

  • New York City. Like Houston, New York City has greatly expanded its police force. Since 1990, the NYPD has grown by 7,000 officers. Under Bratton, police have been directed to crack down on public drinking, graffiti, vandalism, and other public disorders. The NYPD has beefed up action against street gangs and drug traffickers, returned to a policy of frisking suspects for guns and other weapons, and redoubled precinct-level efforts on a wide range of community-policing projects.

    In the process, Bratton has promoted a new breed of precinct commanders and made them responsible for finding innovative, cost-effective ways of serving citizens and cutting crime in their neighborhoods. Despite recent corruption scandals, the precinct-based management system is working, NYPD morale is high, and New Yorkers are getting results that range from fewer aggressive panhandlers to fewer shootings and murders.

Criminologically Correct

Why, then, are many criminologists so unwilling to admit that law enforcement can cut crime? Part of the answer is that more than a dozen major empirical studies over the last two decades have failed to demonstrate either that police manpower and crime rates vary inversely or that particular types of community-oriented policing practices prevent crime. The most famous of these studies is the Kansas City, Missouri, "preventive-patrol" experiment.

For a year in the early 1970s, Kansas City was divided into three areas, each of which received a different level of auto patrol. The 1974 report on the experiment found that criminal activity, reported crime, rates of victimization (as measured in a follow-up survey), citizen fear, and satisfaction with the police were about the same in all three areas. Active auto patrol -- beats where cars cruised the streets conspicuously two to three times more frequently than in the control areas -- no difference at all.

But academic experts who treat such negative findings as the final words on the subject are badly mistaken. George L. Kelling of Northeastern University, the father of the Kansas City research and many other major studies, recently cautioned his colleagues that "generalizing from a study about a specific tactic to other tactics or uses of police is inappropriate." As Kelling observed, "random preventive patrol by automobile for the purpose of creating a feeling of police omnipresence" is a relic of "mid-century policing tactics."

Kellling has scolded those "academic ideologists" in criminology who "do not let research interfere with their conclusions." He keenly characterizes as defeatist dogma their views that "crime stems from basic structural features of society, and until problems like homelessness, social injustice, economic inequalities, and racism are addressed, police impact on crime will be negligible."

As a matter of ideology, denying that law enforcement counts in cutting crime may be trendy, but the policy science of the subject remains far from settled. Following an exhaustive review of the empirical literature on policing, David H. Bayley of SUNY–Albany recently concluded that there has never been "a rigorous, clear-cut test of the association between the visible presence of the police and crime rates." Bayley is now in the early stages of a quasi-experimental study designed to test this relationship while controlling for demographic and other variables related to the incidence of crime. Such fine-tuned research on patrol presence, policing strategies, crime rates, and other key variables has become possible only in the past few years with the development of computer-assisted information systems for police dispatching and management.

Bayley's cutting-edge research will help to identify the general conditions under which tactics and increases in police manpower can curb public disorders and cut crime. For now, there is no solid evidence to dismiss, and every practical reason to uphold, Bratton's Law.

Wattenberg's Law: Prison Works

By the same token, there is tremendous empirical support for another proposition that many criminologists reflexively reject: Sentencing policies that keep violent and repeat criminals behind bars contribute mightily to reductions in crime.

I call this proposition Wattenberg's Law in honor of Ben Wattenberg, a scholar at the American Enterprise Institute and a nationally syndicated columnist. As Wattenberg has quipped, "A thug in prison can't shoot your sister." Whatever else incarceration buys us in the way of criminal deterrence, rehabilitation, or retribution, it most definitely pays dividends by preventing crimes that prisoners would commit if they were free.

The U.S. Bureau of Justice Statistics has reported that fully 94 percent of state prisoners have committed one or more violent crimes or served a previous sentence in jail or on probation. Between 1980 and 1993, violent offenders were the greatest contributors to state prison population growth.

Even so, today more convicted violent offenders are serving time on probation and parole than in prison. About a third of all violent crime arrestees are on probation, parole, or pretrial release at the time of their arrest. Recent studies by me and others estimate that most prisoners commit between 12 and 21 serious crimes a year when on the loose.

From 1980 to 1992, the aggregate violent-crime rate in the 10 states where incarceration climbed the most decreased by 8 percent. In the 10 states with the lowest increases in incarceration, violent crime soared, in aggregate, 51 percent. A study published in Science calculated that in 1989 alone, the increased use of imprisonment spared Americans an estimated 66,000 rapes, 323,000 robberies, 380,000 assaults, and 3.3 million burglaries. In a research report targeted against California's three-strikes law (life without parole for thrice-convicted felons), the RAND Corp. conservatively estimated that the measure would spare Californians about 340,000 serious crimes a year.

Nationally, state prisoners convicted of violent crimes who were released in 1988 had served an average of only 43 percent of their sentences in confinement. Violent convicts released in 1992 had served an average of 48 percent of their time behind bars. And violent offenders released from state prisons this year will have served, on average, 50 to 52 percent of their time in confinement. This slow but steady increase in incarceration is the result of a nationwide trend toward tougher sentencing policies, and has already spared millions of Americans from serious crimes.

It is not yet possible to calculate precisely how much tougher incarceration policies have contributed to falling crime rates in particular cities. But in explaining New York City's falling crime rate, consider the fact that roughly half of New York's state prison population, and an even larger fraction of its violent offender population, comes from New York City. Over the last decade, the Empire State's prison rolls have more than doubled, and the amount of time served behind bars by violent and repeat criminals has increased by as much as 50 percent.

DiIulio's Law: Be Prepared

Apparently, it takes a Ph.D. in criminology to doubt that keeping dangerous criminals incarcerated cuts crime and to wonder whether releasing any significant fraction of the nation's 1 million prisoners tonight would result in more serious crime tomorrow.

But criminologists are right about one thing: Americans are sitting on a demographic crime bomb. Most predatory street crimes are committed by men under 25. Today there are about 7.5 million males aged 14 to 17. By the year 2000, we will have an additional 500,000. About 6 percent of young males are responsible for half the serious crimes committed by their age group. Thus, in five years we can expect at least 30,000 more young murderers, rapists, and muggers on the streets than we have today. Worse, since the 1950s each generational cohort of young male criminals has committed about three times more crime than the one before. Despite the recent decline in murder rates, homicides committed by 14- to 17-year-olds between 1985 and 1993 increased by 165 percent (more for minority males). The next wave of homicidal and near-homicidal violence among urban youth is bound to reach adjacent neighborhoods, inner-ring suburbs, and even the rural heartland.

This crime bomb probably cannot be defused. The large population of seven- to 10-year-old boys now growing up fatherless, Godless, and jobless -- and surrounded by deviant, delinquent, and criminal adults -- will give rise to a new and more vicious group of predatory street criminals than the nation has ever known. We must therefore be prepared to contain the explosion's force and limit its damage.

While there is some room for reasonable disagreements about policy tactics -- for example, whether the federal role in crime control should be expanded, or whether we should invest more in drug treatment or drug interdiction -- any effective anti-crime policy must advance one or more of three interlocking anti-crime strategies: hardening targets, targeting the hardened, and targeting resources.

Hardening the Target. Over the last decade or so, most Americans have taken steps to make the places where they live, work, go to school, or recreate impervious to crime. People have moved out of high-crime neighborhoods, installed anti-burglary devices, made crime-sensitive investment decisions, and lectured their children to be mindful of dangers. Businesses and the 32 million Americans who now live in privately governed residential communities have erected security gates and employed more than a million private security guards.

Neighborhoods of every socioeconomic description have formed "town watch" associations and citizen patrol groups. And the poorest of the inner-city poor have battled the ACLU for the right to target-harden their homes, schools, and parks: They erect concrete barriers on streets frequented by drug dealers and prostitutes, evict convicted street thugs from public housing, install metal detectors, and institute random locker searches in public high schools, and more.

Though to a degree hard to quantify, such initiatives have contributed to recent decreases in crime rates and, over time, insulated us somewhat from the failures of our system of justice. But as the next crime wave approaches, we may well be nearing the limit of what private target-hardening measures can do to foster public safety.

Government at all levels, therefore, should do whatever can be done to bolster these protective measures. To offer just two of many possible examples, urban zoning decisions should begin to take into account the criminal consequences of permitting liquor outlets to be so heavily concentrated in high-crime, inner-city neighborhoods. Likewise, urban enterprise zones make sense as ways of giving de facto tax credits to businesses willing to locate in high-crime places.

Targeting the Hardened. At the same time, we must redouble our efforts to keep violent and repeat criminals behind bars. To consolidate and expand recent gains, we must be vigilant not only in pushing for truth-in-sentencing and three-strikes measures, but seeing to it that these laws are followed both in letter and in spirit.

Make no mistake: The counter-offensive against tougher sentencing policies is well underway. Aided and abetted by activist federal judges, prisoners' rights activists, journalists, and academic "experts," efforts are already being made to depict these laws as failures and to deny or disparage any suggestion that they have helped account for recent drops in crime rates.

But just ask the criminals. In California this year, within several months after the three-strikes law went into effect, an increasing number of parolees began to request inter-state transfers. Likewise, even before Washington State's three-strikes law hit the books in 1993, dozens of sex offenders called the Seattle Police Department with questions about what crimes might count as "strikes." As one career criminal told the detective in charge of the department's sex-offender unit, "It wasn't until [the three-strikes law] passed that I had to say to myself, 'Damn, these people are serious now.'"

As the next crime wave draws near, we must remain deadly serious about targeting hardened adult and juvenile criminals for arrest, prosecution, and incarceration. Anti-incarceration propagandists can be counted on to work overtime with much-publicized tales like the one about the California man whose "third strike" was stealing a slice of pizza from a child in a mall. They failed to note, however, that this criminal had four prior convictions.

Likewise, the anti-incarcerationists are sure to repeat canards about how tougher sentencing policies will bankrupt the country and result in massive prison overcrowding. In truth, we now spend less than half a penny of every tax dollar on prisons, and the costs of prisons can be reduced greatly by cutting back on inmate amenities and services that account for more than half of the prison budget in many states.

As for "overcrowding," despite the growth in the prison population, fewer prisons today are operating over their rated capacity (the number of inmates they were designed to hold) than in 1990. And contrary to popular assertions, there is no systematic empirical evidence to show that double-celling raises the risk of prison disorders, inmate illness, or other serious problems.

Congress is taking steps to target the hardened, as it considers Title III of the Violent Criminal Incarceration Act adopted by the House last February. Known as the Stop Turning Out Prisoners or STOP law, this measure would prohibit activist federal judges from arbitrarily imposing prison caps that result each year in the early release of tens of thousands of dangerous criminals whose return to the streets results in murder and mayhem.

Targeting Resources. Whatever government does henceforth to combat crime should be done in a targeted fashion. It makes no sense, for example, to heed President Clinton's call for spending $8.8 billion in federal dollars for "100,000 cops." As I and many other analysts have proven, not only will that amount not pay for or even seed the funding of anything near 100,000 police officers, but the Justice Department's grant process is rigged to deliver lots of money to small cities that have enough cops and little crime. By the same token, however, it makes even less sense to follow the Republican alternative of dumping more than $10 billion on the states in open-ended anticrime block grants.

So here is DiIulio's Law: Over the next five years, as our euphoria over good news about crime fades and a public panic to "do something" about youth crime begins, let us prepare to honor the overarching conservative principle that government should never spend money it does not have for purposes it has not clearly articulated in order to generate results that cannot easily be evaluated.

About the Author

More from Policy Review