Why did violent crime drop by 13% last year?

By Alan Bean

The Bureau of Justice Statistics reports that the rate of violent victimization declined a jaw-dropping 13 percent in 2010.  Nobody seems to know why.  During the prison boom, drops in the crime rate were often associated with mass incarceration–they can’t commit crimes if they’re all locked up.  But we have been sending fewer American citizens to prison in recent years and the drop in crime has only accelerated.  And all of this during the worst recession in 80 years.  What’s going on?

There is little association between violent crime and unemployment statistics.  More often than not, violence is a rage response.  Cost-benefit calculation is only involved when criminals kill to eliminate potential witnesses, but few murders or assaults fit this scenario.  Drug and alcohol abuse can be contributing factors, but people rarely get violent because they are high and the use of some drugs (marijuana, for example) actually inhibit violent behavior. 

The best explanation for shifting rates of violent crime comes from Randolph Roth’s excellent American Homicide.  After analyzing statistics on violent crime from the earliest days of European settlement, Roth shows that America is a far more violent nation than other Western democracies.  For instance, “Th next most homicidal democracy, Canada, has had only a quarter of the homicides per capita that the United States has had since Worth War II.”

Here are a few of Roth’s big-picture observations:

The risk of being killed is highest in the South, moderately high in the Southwest, and lowest in the North.  It is greatest for the poor, but even wealthy and middle-class northerners run a higher risk of being murdered than do people in other affluent democracies.  Their risk is highest when they are in their teens and twenties, living on their own in apartments, in the military, away at college, or spending a night out of town.

 Roth points out that poverty and violence are not usually correlated.

Urban poverty and unemployment may appear at first glance to be responsible for America’s homicide problem, but during the Great Depression of the 1930s, when urban poverty and unemployment were at their worst, homicide rates dropped in most cities, from New York to New Orleans to San Francisco.    

Deterrence is a factor, of course.  During times of general lawlessness (the post Civil War period in the South; frontier towns in the West) homicide rates soared.  But deterrence, by itself, can only do so much.  When waves of homicide strike, Roth says, systems of deterrence are powerless to stop them.

When homicide rates spike it isn’t because jilted lovers or irate acquaintances suddenly become more inclined to resort to violence–violent in-group behavior is more or less constant.  Sharp increases in the murder rate happen Roth says, when we see a sharp increase in the number of people (usually men) who are “predisposed to violence, willing to prey on others or to view them as enemies or rivals.”

So, why are more people predisposed to violence in some times and places than in others?  For instance, why did the murder rate begin to plunge from a high of almost 10 homicides per 100,000 population per year in the mid-1930s, to a low of 5 homicides per 100,000 in in late 1950s and early 1960s?  What made the murder rate rise to 11 per 100,000 in the mid 1980s?  And why were young black males far more inclined to indulge in predatory violence than any other demographic?

“The predisposition to violence is not rooted in objective social conditions,” Roth contends.  Young black males faced far more objective oppression in the 1940s and 50s than was the case in the mid 80s, yet rates of violent crime among this group were far lower during the earlier period. 

Roth’s thesis may surprise you.

Criminologist Gary LaFree confirms the fundamental importance of feelings and beliefs when he points out that of all the variables on which social scientists have collected data in the past fifty years, homicide rates among unrelated adults in the United States have correlated perfectly with only two: the proportion of adults who say they trust their government to do the right thing and the proportion who believe that most public officials are honest.  When those proportions fell, as they did in the late 1960s and early 1970s, the homicide rate among unrelated adults increased.  When those proportions rose, as they did in the 1950s and mid-1990s, the homicide rate fell.

Young black males in the 1940s and 50s faced oppression at every turn, but their life expectations were improving dramatically and there a general expectation that this trend would continue.  The 70s and 80s were a time of deep disillusionment in urban black communities.  Conditions were deteriorating and there was a general sense that the folks making the decisions could not be trusted. 

The election of Bill Clinton reversed this trend even thought the Arkansas Democrat adopted policies that were often detrimental to poor African Americans.  The rhetoric was more inclusive and engaging; Clinton was comfortable in African American settings and was generally popular with the black electorate.

The presidency of Barack Obama is unfolding in an atmosphere of intense ideological conflict that has undermined confidence in government.  But the mere fact that an African American occupies the White House has changed the emotional and social game for young black males.  Objectively, their circumstances are bleak and getting bleaker; but, as in the 40s and 50s, there is a sense of fragile hope.  It’s hard to feel if you live in a predominantly white suburb; but it is very real in poor black neighborhoods and this, more than any other factor, explains the drop in predatory violent behavior.


WASHINGTON – During 2010, U.S. residents age 12 or older experienced a double-digit drop (down 13 percent) in the rate of violent victimization, the Bureau of Justice Statistics (BJS) announced today. Violent crime includes rape or sexual assault, robbery, aggravated assault and simple assault. The rate of property victimization, which includes burglary, motor vehicle theft and household theft, also declined by six percent during the year.

The drop in violent victimization, from about 17 victimizations per 1,000 residents in 2009 to 15 per 1,000 in 2010, was three times the average annual rate of decline experienced over the last nine years. The property victimization rate dropped from 127 victimizations per 1,000 households in 2009 to 120 per 1,000 in 2010, which was about two times the average annual rate of decline from 2001 to 2009. During the 10-year period from 2001 to 2010, the overall violent victimization rate decreased by 40 percent and the property victimization rate fell by 28 percent.

These declines in violent and property victimizations continued a larger trend of decreasing criminal victimization in the United States. In 2010, violent and property victimization rates fell to their lowest levels since the early 1990s. From 1993 to 2010, the violent crime victimization rate decreased 70 percent, dropping steadily from about 50 victimizations per 1,000 persons age 12 or older in 1993 to about 15 per 1,000 in 2010. The property crime victimization rate fell 62 percent, from about 319 victimizations per 1,000 households in 1993 to 120 per 1,000 in 2010.

Overall, U.S. residents age 12 or older experienced an estimated 18.7 million violent and property crime victimizations during 2010, down from 20.1 million in 2009. This included 3.8 million violent victimizations, 1.4 million serious violent victimizations (rape or sexual assault, robbery and aggravated assault), 14.8 million property victimizations and 138,000 personal thefts (picked pockets and snatched purses).

Almost two-thirds of violent crime victimizations occurring during 2010 were simple assaults (2.4 million), in which the victim did not suffer an injury. The decline in simple assaults (15 percent) between 2009 and 2010 accounted for 83 percent of the total decrease in violent victimizations; there was no measurable change in the number of serious violent victimizations during that time.

The nature and severity of violent victimization changed during the 10-year period from 2001 and 2010. The percentage of violent victimizations involving weapons declined slightly from 26 percent in 2001 to 22 percent in 2010. Strangers perpetrated 39 percent of violent victimizations in 2010, down from 44 percent in 2001. The percentage of males who were victimized by a stranger declined from 55 percent to 48 percent over the 10-year period, while the percentage for females remained relatively stable. After a slight decline from 2001 to 2008, the percentage of victims of violent crimes who suffered an injury during their victimization increased from 24 percent in 2008 to 29 percent in 2010.

In 2010, about 50 percent of all violent victimizations and nearly 40 percent of property victimizations were reported to the police. These percentages have remained stable over the past 10 years.

For the first time, males (15.7 per 1,000) and females (14.2 per 1,000) had similar rates of violent crime victimization in 2010. Historically, males have had higher rates of violent victimization compared to females. Males and females were equally likely to report violent victimizations to the police.

As in other years, persons of two or more races (52.6 per 1,000) had higher violent victimization rates in 2010 than white non-Hispanics (13.6 per 1,000), black non-Hispanics (20.8 per 1,000), or Hispanics (15.6 per 1,000). Asians and Pacific Islanders had the lowest rates of overall violent victimization (6.3 per 1,000).

BJS’s National Crime Victimization Survey (NCVS) collects self-reported information from victims on their experiences of criminal victimization. The survey provides the largest data collection on criminal victimization independent of crimes documented by law enforcement. Estimates from the NCVS, which includes offenses both reported and not reported to police, complement those from the Federal Bureau of Investigation’s (FBI) Uniform Crime Reporting Program (UCR), which measures crimes reported to law enforcement agencies across the nation.

During 2010, 40,974 households and 73,283 individuals were interviewed twice for the NCVS. The NCVS, unlike the UCR, is a self-reporting survey and does not collect data on murder or homicides.

The report, Criminal Victimization, 2010 (NCJ 235508), was written by BJS statistician Jennifer Truman. Following publication, the report can be found at http://www.bjs.gov.

For additional information about the Bureau of Justice Statistics’ statistical reports and programs, please visit the BJS website at http://www.bjs.gov/.

# # #