The Latest on Homicide Rates


Nothing grabs headlines like dire warnings about homicide trends. And there is no criminologist better at garnering headlines than James Alan Fox, whom you might remember from Freakonomics for the ominous reports he produced about juvenile homicide for Attorney General Janet Reno in the 1990’s, even as crime began to plunge.

James Alan Fox is baaaack with a new report that is garnering headlines. The New York Times writes: “Homicides by Black Teenagers Rise, Bucking a Trend.”

The Wall Street Journal is even more precise: “Murders of Black Teens Are Up 39 Percent Since 2000-01,” although actually murders by black teens are up more than murders of black teens.

So is it time to panic, or at least to take dramatic action? If you look at The New York Times graphic accompanying the story, it sure seems that way.

When the data are displayed differently, you likely come to a different conclusion. Here is Figure 4 from the James Alan Fox report:


This figure presents homicide rates by age for blacks from 1976 to 2007. The dominant pattern in this picture is the huge spike in black youth homicides in the early 1990’s. The phenomenon captured in the scary New York Times graphic above corresponds to the barely perceptible rise in the black circles at the far right of the figure.

Why do things look so different in The New York Times graphic versus the figure from Fox’s own report?

1) Compared to the early 1990’s, what is happening now is much smaller in scale.

2) When put side-by-side with no trend in the homicide rates of blacks aged 18 to 24 (the gray circles in the graph), the blip by 14- to 17-year-olds doesn’t seem so frightening.

And the most important (and I would say devious) difference between the two figures:

3) The numbers in The New York Times graphic and most of the James Alan Fox report fail to control for the change in the population of young black males over this time period.

According to U.S. Census data, the number of blacks aged 15 to 19 rose by about 15 percent between 2000 and 2007.

So even if any individual black teen’s propensity for crime was unchanged over this time period, the aggregate amount of black-teen crime would have risen by 15 percent. In other words, in that New York Times graphic on perpetrators, just based on changes in population, the number of perpetrators would have been expected to rise from a little over 800 to nearly 1,000. Knowing that, the actual rise to roughly 1,150 doesn’t seem that noteworthy.

In his report, James Alan Fox argues for the “importance of restoring federal funds for crime prevention and crime control.” While I suspect that directing federal money toward crime control would be a better use of funds than continued bailouts, I would argue that it is time to experiment with something more radical that would actually save the government an enormous amount of money: ending the war on drugs.

More on that soon.


very nice piece. It's always good to show that newspapers tend to write what people want to read.

I coudn't agree more on your last statement!


So there should be a graph with the figures adjusted to constant population.



Indeed is another example of how some social researchers manipulates the available data in order to gain more than news spreading.


Newspaper, more than tending to write what people want to read, actually write things they (the journalist) know very little about.


Really looking forward to more on ending the war on drugs. Should cocaine and heroine remain illegal because they are "hard" addictive drugs and marijuana legalized? Either way, while I agree that some legalization should happen I won't hold my breath because there are MAJOR cultural forces arrayed against such an outcome.


Be careful...your willingness to uncover the truth about this situation may lead to your being called racist, since your findings may lead to less concern about a situation affecting black people. O.K. I am just going to go ahead and put it out on the table so it will sting less when others say it: you are a racist. Please get back to your bell curve studies, you racist you.


Sure, but don't you think that if the "barely perceptible rise" in black youth homicides was noticed in 1983-5 and action was taken to curb that rise, the spike in the early 90s could have been averted or at least not so severe?


I was glad to see that you responded to this report. I had been intrigued by your explanation about legalized abortion leading to lower crime rates in your book and thought here was another alarum that required debunking. I agree with your statement about the need to end the spurious war on drug, however, do you think all of the thousands upon thousands of DEA, FBI, ATF and local police personnel who are engaged in this game of whack the mole can be re-trained as drug counsellors? Because for sure, they will be out of a job.


I bet if you opened a factory in the inner city (near public transit), and offered 200 jobs starting at $20/hr. with benefits, you'd get 2000 applicants. And I bet those workers (targeting young black men) would work hard to hold onto those jobs. And so even if this factory built TVs in America, and sold them at a loss to compete with Asian pricing, we'd still come out ahead: people would be able to get out of the cycle of subsistance welfare, learn the social and job skills required to succeed, and be able to support families.

Until there are respectable jobs that pay a good wages, there will be a disproportionate criminal economy, gang activity, and all the violence and loss associated with it. Right now, it seems like young black men are on a tightrope: be targeted by the gangs if they try to do well in school, trying to make a living on minimum-wage, low interest jobs; or getting sucked into gang life in the hope of fast money (or need for protection).

We've been trying to help people on the cheap--how's that working? What we need is a "surge"--an influx of life-changing opportunties. And if the jobs were competitive (based on achievement and productivity) those most likely to benefit from the opportunity would rise to the top.

As a matter of fact, every American, regardless of race or gender, should be able to work for a decent wage if willing and able. Just think how many of these "factories" the $700 billion bailouts could have built. And suppose the factories didn't build TVs--but green cars, laptops for schools, exercise equipment for parks, wheelchairs for the disabled....



Quite a few years ago, I was James A. Fox's student at Northeastern. Even then, when I was young and impressionable, he had the air of a very (self) Important Person. Anything to get on CNN (at the time, who cares about CNN these days?).

His scholarship needs work, but his ambition is boundless.


In a bad economic climate, crime rates tend to increase. please check out my article on associatedcontent which explorers job grwoth this century as compared to the prior century:


Just another example of how things went to hell during the Reagan administration. He was such a successful con man that people still don't realize they've been duped.

Jonathan Katz

Maybe we should stop paying unmarried girls to have babies?
Almost all street criminals are illegitimate, with no father to guide them growing up.

Despite the reform of 1996, unmarried girls can still get welfare for their first child for five years, which looks like forever when you are 15 (these aren't people who plan for the long-term in any case: those go to college and don't get pregnant until they marry). The sum is small compared to a middle class cost of living, but it is a lot of spending money for a teenager living at home (probably with her unmarried mother).


You say there is a rise from an expected value of 1000 to 1150. Put another way as a 15% increase in murders. Since when is 150 murders not noteworthy? That this is not noteworthy to you illuminates the limits of economics.

More importantly, each of those crimes ends not just life of the victim. The countof course begins with the victim. But it should also include any family that they have, the productive life of the person who committed the crime, the family of the person who committed the crime, and on and on.

Thought I would disagree with the premise that a 15% increase is a small increase, the multiplier effect of such crimes magnify the impact any increase.

Zanger Zuniga

What dubious data.

If you could control for household income levels, education, employment outlook, etc., and re-run the numbers, they might actually offer some information.


Ending the war on drugs might save the government an enormous amount of money, but it would adversely affect two things: GDP and interest groups. The interest groups, besides gangsters and terrorists, are corporate prisons, defense attorneys, corrupt government officials, corrections officers unions, police departments (both honest and corrupt), DEA officials, threat assessors, pharmaceutical companies (which want to hook you on their chock-full-of-side-effects chemicals), and rehab clinics that don't have to show effectiveness because they ply their trade on prisoners and are paid by the government. GDP, of course, is morally and ethically neutral, so all of the money spent making the groups mentioned above happy would no longer be spent, and the politicians would cry that our economy has become stagnant. Marijuana, cocaine and opium make money only when criminalized.


So, let us if I got this right.......

It is not too bad if more people are killed, so long as the proportion (as the percent of that sub group), of those killed remains level, or is lower that the peak level ! Somehow it implies that the previous level was acceptable.

Is this a new standard by which we value certain human life? If so, somebody please get word out to that poor mother in the ghetto to look at the bright side to all this - her child's death represents an improvement in our ability to protect our citizens.

Tell me what is wrong with this picture?


Excellent! When I read this article in the WSJ yesterday, I had the exact same reaction. The author gives raw numbers but doesn't provide the meaningful data. I'm so glad you've called him out on this.

Rich Accetta-Evans

I think your analysis is probably correct, but I'm a little confused about how a rise in the young black population would lead to a spike in this graph. The graph doesn't appear to be a graph of the absolute number of murders but of the "rate per 100,000"

Could you provide a corrected graph that clarifies this?


It's comforting to know that if you are killed by a black teen, still the relative risk is not as great as it would appear.