Causes of Crime

by Russ Roberts on November 27, 2006

in Crime

The Washington Post finds it surprising that crime rates in New York City have dropped while the prison population has declined.

It is one of the least-told stories in American crime-fighting. New
York, the safest big city in the nation, achieved its now-legendary
70-percent drop in homicides even as it locked up fewer and fewer of
its citizens during the past decade. The number of prisoners in the
city has dropped from 21,449 in 1993 to 14,129 this past week. That
runs counter to the national trend, in which prison admissions have
jumped 72 percent during that time.

"If you want to drive down crime, the experience of New York shows
that it’s ridiculous to spend your first dollar building more prison
cells," said Michael Jacobson, who served as New York’s correction
commissioner for former mayor Rudolph W. Giuliani (R) and now is
president of the Vera Institute of Justice, which studies
crime-fighting trends worldwide.

"I can’t tell you exactly why
violent crime in New York declined by twice the national rate," he
said. "But I can tell you this: It wasn’t because we locked up more

It doesn’t seem obvious to either Jacobson or the author of the article, Michael Powell, that crime might be down for reasons unrelated to incarceration, say a change in the demographics of New York and that both crime and incarceration are down because of this third factor. Correlation isn’t causation.

The article goes on to look at other states:

Perhaps as intriguing is the experience in states where officials spent
billions of dollars to build prisons. From 1992 to 2002, Idaho’s prison
population grew by 174 percent. the largest percentage increase in the
nation. Yet violent crime in that state rose by 14 percent. In West
Virginia, the prison population increased by 171 percent, and violent
crime rose 10 percent. In Texas, the prison population jumped by 168
percent, and crime dropped by 11 percent.

I guess I don’t find it so intriguing.

In Idaho, something changed—say an increase in the proportion of young
men relative to the population as a whole—and this in turn increased
crime and—surprise!—led to an increase in the number of people in jail.
And sometime when your criminal population increases, you have to build more prisons. Maybe that’s why prison spending is up rather than because of some theory that by building more prisons you’d deter crime. Without more information, these numbers have no meaning.

What am I missing here? Ah, here it is:

"Crime is down and people
realize, sure, we can lock up more people, but that’s why your kid’s
pre-K class has 35 kids — all the money is going to prisons," Jacobson
says. "There’s a sense of urgency that for the first time in two
decades, we can talk about whether it makes sense to lock up even more

According to this logic, we should let some criminals go free rather
than tie up valuable money taking care of them in prisons or building
new ones. Use that money for schools and the crime rate won’t change. In fact, it will go down, because better reading pre-K reading programs mean fewer criminals!

I don’t know what’s sadder. The claim that we’re spending less on
education (we’re not), the idea that if we spent more money on
education, we’d get more education (not true given the incentives of
the public school system) or the implication that punishment doesn’t
deter crime (it does). 

(HT: James Taranto’s Best of the Web at the WSJ)

Be Sociable, Share!



17 comments    Share Share    Print    Email


matt November 27, 2006 at 3:29 pm

Maybe crime is down in NYC because the criminials have moved down here to Philly?

I see more people from NYC in this area than ever before in my lifetime.

Charlie Quidnunc November 27, 2006 at 4:26 pm

Malcolm Gladwell talked about this in "The Tipping Point". He claims that once it became clear enough to potential criminals that the police would not tolerate any level of crime, the rate dropped precipitously. This is where the "zero tolerance" method of policing got started. They first attacked the graffiti problem on subway cars, and dropped that level of crime below a "tipping point", after which little more effort had very large rewards.

alex November 27, 2006 at 9:01 pm

Crime does not rise or fall due to the disincentives of punishment. Crime tends to follow societal trends on the local level. So, if poverty is down in an area, as well as falling infant mortality rates, hunger, etc., crime will tend to fall as well, since growing economic stability will cause people to go find jobs than join the bloods and crips, for instance. New York City has experienced an economic boom from the eighties to the nienties, and I suspect this has something to do with the falling crime rates their, as does the property bubbles in East Harlem and other deprived neighborhoods (giving local residents much needed cash).

Mr. Econotarian November 27, 2006 at 11:36 pm

One also needs to consider the ratio between entrepreneurial / consentual crime rates (like drug dealing) versus violent crime rates or property crime rates, and the incarceration rates of those crime types as well.

bbartlog November 28, 2006 at 11:11 am

Unemployment rates are strongly (inversely) correlated with crime rates. And unemployment has been low for a long time now.
For some of the rural places like West Virgina and Idaho I wonder whether those huge percentage increases just reflect previously very low crime rates, now increasing to more typical urban levels as things like crystal meth or urban gangs reach the more rural parts of the US.

David November 28, 2006 at 6:16 pm

I suspect that prisoners sentenced to terms longer than one year ended up in state prisons rather than the city jails. It would be helpful if data on the number of New York City residents in state prisons were also examined.

Forbes November 30, 2006 at 2:39 pm

The poster just above, David, is correct in questioning the level of the state prison population, and although I do not have the figures at hand, the State of New York is closing something like 10-20% of its prison beds due to a reduction in prison population.

Now the other observation pertinent in the NYC crime/prison situation is that this is the first notice, of falling prison populations, that has been given to the results–a la the tipping point. Since Guiliani's introduction of various policing strategies in '94, the prison population went up, up, up, and the crime rate went down, down, down, if only because the criminals were behind bars, and could not commit crimes (pace Alex's comments to the contrary).

Also, it's worth noting that the major crime reduction occurred in the most economically impoverished neighborhoods, e.g. Harlem, Wash Hts, East New York, Bed-Stey, South Bronx,, therefore paving way for economic renaissance to occur as such neighborhoods become viable for investment.

Michael December 24, 2006 at 5:05 pm

The article–which I wrote–in no way dismisses Giuliani's war on crime. But it's important to have a clear-eyed view of the phenomenon. To wit:

From the very beginning of NYC's decline in crime (which began dropping noticeably during the last year of the Dinkins administration), New York's jail population fell. The state prison population began falling in about 1994, and today NYC's contribution to that decline is disproportionate.

In other words, the decline in prison populations is not the END result of a get-tough policy. It tracked the decline in crime from the very beginning.

As for declines in poverty accounting for the decline in imprisoment, as suggested above? There's not much empirical evidence to back up that assertion. The poverty rate in New York throughout this period has hovered at between 18 and 22 percent. In fact, it was higher in the early 90s as crime began falling. Similarly, as late at 1988, the unemployment rate in the Bronx and inner city Brooklyn topped 15 percent. Yet crime fell rapidly in these areas.

As for Idaho, there was no striking change in the demographics of the place, save for an influx of Californians of higher income. It's doubtful they contributed to the rise in crime, though.

More likely, the national obsession with metamphetimines has hit Idaho particularly hard, leading to vastly increased rates of imprisonment. It is wise policy to lock up low-level drug users? That's a hard case to make.

Lastly, the blogger asks, well, how do we know the crime decline wasn't driven by demographics? A little research on census websites shows that's not the case, at least during the early and most dramatic years of New York's crime decline. The percentage of people living in poverty, the percentage of blacks, of Latinos, all stayed more or less the same throughout this period. Of late, there HAS been a very large influx of immigrants, and there is intriguing work correlating this influx with a decline in crime.

{ 1 trackback }

Previous post:

Next post: