Crime and Economy

I don't know the literature on this topic well enough to be able to say whether or not this claim that it is difficult to find a link between macroeconomic conditions and crime is accurate:

Crime and Economy, by James Q. Wilson, Commentary, LA Times: ...During the last two decades, scholars have made great progress in explaining why some individuals are more likely than others to commit crimes, but very little in explaining why the crime rate ... rises or falls.

Everyone knows that there is more crime in economically depressed inner-city neighborhoods than in affluent suburbs. That fact leads naturally to the assumption that if a community becomes more prosperous, crime rates will go down, and if income levels decline, crime rates go up. Economists who have checked this view have discovered that it is often true, but not always. ...

A lot of other factors affect the crime rate... It often goes up when the population gets younger, and when drug abuse becomes more common. Murder rates are profoundly influenced, at least in big cities, by gang activity. ...

So can the economy help explain fluctuations in crime? Sometimes yes, sometimes no. It would be difficult to link rising crime during the prosperous 1960s to economics. On the other hand, a declining economy provides a plausible theory to explain increases in crime during the 1990s. Matters become even more complicated if one goes back to the Depression of the 1930s. We had no FBI data on crime rates at that time, but several studies of individual cities suggest that crime rates fell even though one-quarter of all Americans were unemployed. Why? One reasonable hypothesis is that the Depression pulled families together, and this cohesion inhibited crime. ...

The role of the police in reducing crime is often overlooked by those preoccupied with the jobs-crime link. The sharp decline in crime in New York -- and now in Los Angeles -- has a lot to do with how those police departments changed. ...

To try to sort out the combined and complex relations between crime and the economy, the age of the population, imprisonment, police work, neighborhood culture and gang activity, the National Academy of Sciences Committee on Law and Justice (which I chair) has begun an effort to explain something that no one has yet explained: Why do crime rates change? If you have any good ideas, let me know.

Comments

[gordon]

"The sharp decline in crime in New York..."

Perhaps Bernard Madoff can explain it to us.

And as for the poverty/crime connection, J.K.Galbraith suggested that for fraud the connection was exactly the reverse of Wilson's suggestion; the bigger the boom, the bigger the "bezzle".

Back in August - see this comment and subsequent ones - I had a little exchange with Larry (some other commenters contributed too) about what crime is and whether white-collar crime really is crime. You know you're on a middle-class blog (is there another kind?) when there is mass amnesia about fraud, tax evasion, insider trading and so on.

[gordon]

"The sharp decline in crime in New York..."

Perhaps Bernard Madoff can explain it to us.

And as for the poverty/crime connection, J.K.Galbraith suggested that for fraud the connection was exactly the reverse of Wilson's suggestion; the bigger the boom, the bigger the "bezzle".

Back in August - see this comment and subsequent ones - I had a little exchange with Larry (some other commenters contributed too) about what crime is and whether white-collar crime really is crime. You know you're on a middle-class blog (is there another kind?) when there is mass amnesia about fraud, tax evasion, insider trading and so on.

[gordon]

Ooops.

[Beezer]

You want a "crime wave" simply start enforcing laws more energetically. Voila! Crime wave.

[ken melvin]

"If you have any good ideas, let me know.

First, drop the ideology. Only a fool could not see. Only a Pepperdine fool could sit in LA or Oakland and watch crime go down in the 90s and roar back when things got worse in 2001-08 . Any damned fool could have predicted what was going to happen. Anyone who couldn't, didn't want to.

[I, Kahn O'Clast]

I thought it had been long established that a key determinant of crime rates is the rise and fall in the level of young males between 15-25 years of age and that economic conditions, while linked, are secondary.

[Markel]

Wilson published Thinking About Crime in 1975, and has made essentially the same arguments ever since. He has not had much success dislodging the Jean Valjean theory of poverty-driven crime in the minds of either the public or policymakers. He doesn't, as some claim, completely dismiss the connection between unemployment and crime, but says the evidence is weak and the relationship, if it exists, is very complex. He thinks the link is much stronger between crime rates and the swiftness and certainty of punishment (not necessarily the severity). In this sense, he now sounds a bit old-fashioned to folks wowed by the Freakonomics abortion theory.

[StMac]

The idea of a crime rate implies that there is some "amount" of crime committed over a period of time. But how does one even measure crime? Crimes committed? Crimes reported? Laws enforced (i.e. does jaywalking count towards the total)? Arrests? Convictions? Lives lost/# Victims? Dollars lost? Potential dollars lost?

Is a double murder twice the crime of a murder with only one victim? Is murdering a banker a greater quantity of crime than murdering a homeless person? Perhaps less? The same?

Is there a distinction between a bank robbery where $10,000 are stolen versus $50,000...does one have more criminality than the other?

Maybe someone who has a better background in criminology can easily answer these questions (I'm sure it's been disussed)...but I doubt it. It seems like a highly subjective concept that will greatly color any conclusion derived from it.

[malcolm]

If I remember correctly, this is related to Steven Levitt's work on crime. There should be a Journal of Economic
Perspectives piece by Levitt reviewing his research.

Whenever I teach the costs of unemployment, students a re surprised when I tell them that economy wide downturns don't
have a statistically significant impact on crime.

[Patrick]

Violent crime seems to be associated with populations of young, unmarried, unemployed males.

[Christian B]

I know the data better for Japan than the United States, but statistically factors such as GDP growth, inflation, unemployment do not track changes in the amount of crime. Also, wealthier countries tend to have more crime than poorer countries. Changes in income inequality do track changes over, at least in Japan, and cross-nationally the levels in crime.

There are obviously other problems with getting an accurate measure on the Dependent variable.

[Russell]

I do consulting work with Internal Audit departments, and they are all convinced that fraud (broadly defined) at their organizations increases when there is a downturn in the economy. I remain highly skeptical of this view and none of them has been able to provide any empirical support. It's just a hunch.

I find it interesting that Madoff was uncovered post-downturn (in the midst of recession), but almost all of his bad behavior came during the boom.

Really interesting questions Wilson is proposing to study, but there are too many methodological challenges to be overcome. The committee will likely find whatever fits their general view of the world. That's a shame.

[Sarah]

I would echo Christian's comment. I have been most impressed by the correlation between income inequality and crime. It mirrors what I have seen in other countries where poverty itself doesn't seem to be an issue-- provided most people are equally poor. But the existence of a visible rich elite seems to provoke anger, which often spills over into all populations. Economists need to start looking more at psychological research. We humans seem to be hard-wired, at some level, for fairness.

[Rusty]

Anecdotally, in this region, armed robberies appear to be increasing sharply. No hard data though.

[Denis Drew]

WHEN THE ECONOMY WAS RUNNING AT ITS RECENT BEST A COUPLE OF YEARS BACK THE MINIMUM WAGE WAS AT A 1939 LEVEL -- $5.15/HR IN 2007 V. $4.65/HR W/NO TAXES IN 1939?

COULD THIS GREAT WAGE DEPRESSION POSSIBLY EXPLAIN OUR HOOVERVILLES (GHETTOS) POSSIBLY EXPLAIN THE EXISTENCE OF STREET GANGS FIGHTING OVER THE (PROHIBITED) DRUG BUSINESS?

AS CASEY STENGLE ONCE ASKED: DOES ANYBODY HERE KNOW HOW TO PLAY THIS GAME? ??????????????????????????????

[George Kalogridis]

I believe it was Woody Guthrie that wrote the line; "Some men rob you with a six gun, others with a pen." The real question is why the thief with a gun does long hard time and the "White Collar" criminal does a few years in a Federal Boarding School.

I have been wondering if "the troubles" will return to Northern Ireland now that their economy has tanked. This might be a good indicator of the poverty - crime thesis.

[Alex Tolley]

Interesting how little distinction is made between economic crime - burglaries and violence used as part of theft and non-economically motivated crimes, e.g. rape.

Then there are classes of economic crime that are not even seriously reported - e.g. retail store pilferage by employees.

Then there are crimes created by local laws - e.g. using or dealing in drugs - particularly harshly enforced in the US.

So factors influencing crimes depends at least partly on what behaviors the society regards as crimes.

What factors raise or lower the rate of "criticizing the state or leader", a crime in many countries?

[Donald A. Coffin]

Suppose we have some data (time-series or cross-section or pooled time-series/cross-section). Suppose we think
Y = f(X; V),
where X is a variable of primary interest and V is a vector of other control variables. And suppose we estimate the following relationship:
Y = a + bX
and we find a weak or insignificant relationship. So?

I can buy Wilson's conclusion if we properly estimate a full model, which may be a fairly complex, multi-equation, structural model. But looking at a simple bivariate relationship is unlikely to be terribly revealing.

[Bruce Wilder]

Patrick: "Violent crime seems to be associated with populations of young, unmarried, unemployed males."

Yes, and James Q. Wilson, in other work, has emphasized the "testosterone" factor. In the U.S., violent crime is also cultural: the "honor" culture of the South and the "Wild West" is associated with much higher rates of violent crime than is the conscience culture of greater New England or the vaguely Nordic cultures of the upper Midwest. There's a nasty dispute over whether Southern "honor" culture or some genetic basis in race, or a racist bias in the criminal justice system, or economic conditions, is responsible for the disproportionate participation of young, black males in violent crime in the U.S. 2% of the population, young, black males, commit close to half the violent crime (excepting rape) -- according to FBI statistics; that's roughly half the murders, robberies and assaults. Burglary, by contrast, is more of a white crime.

James Q. Wilson, it ought to be noted, is very conservative politically, and has been associated with "constitutional learning" (aka "high testosterone") explanations of crime that put a lot of weight on individual predispositions. He co-wrote a book on Crime and Human Nature, with the late psychologist Herrnstein, who also did a lot of controversial work on race and intelligence (Herrnstein was a co-author, too, of the infamous Bell Curve).

The crime rate did rise during the 1960's and early 1970's, and spectacularly. It fell again in the 1990's, though it has never returned to the levels of the 1950's. The rise in the crime rate in the 1960's had profound political effects, and contributed mightily to the rise, imho, of conservatives and the Republican Party. Liberals tend to have little memory of this profound social disturbance, when they review the history of Nixon and the "Southern Strategy" and all that. But, the rise in crime rates accompanied a rise in fear and anger and resentment.

One component explaining the rise in crime in the 1960's, which is relatively uncontroversial, is the baby boom, and the general increase in the numbers, and proportion, of teens and young adults in the population.

Another component explanation, which oddly doesn't get a lot of attention, is the incarceration rate, which fell, and by a lot, in the course of the 1960's, as the rates of violent and property crime were rising. Perhaps not coincidentally, the crime rate fell in the 1990's as the prison building boom took hold. Improved policing methods get generally favorable attention, but the U.S. imprisons a lot of people today.

[bp]

the curse of ceteris paribus
lifted by a depression

all other things equal
crime pays more in depressions

worseoff robs betteroff
trickle down turns to take down

all other things equal
crime pays more in bubbles

madoff robs betteroff
financial parasites rob the rest

more guns, less crime, ceteris paribus
those without concealed carry robbed the most

even a blind economist
finds an acorn once in a while

[Patricia ShannonP]

As others have said, there are are several factors, including ones of definition, involved in crime rates.

But I, who tries to keep an open mind and usually avoids non-equivical statements, have no doubt that rising unemployment does cause an increase in property crimes, other things being equal.

I volunteered for several years around the time of the Reagan recession. I visited the women in the county jail every Saturday morning. Before the recession, there were usually only one or two women. The largest at one time was four. They were each in a separate cell.

When unemployment rose, so did the prison population. The number of women rose to about a dozen. They were crowded into a single cell, with mattresses on the floor, because the number of male prisoners had also increased so much, they didn't have spare cells.

Almost all the women then were in for passing bad checks. When they had jobs, they weren't passing bad checks. Almost all of them had children.

[Patricia ShannonP]

Also, from personal experience, I know that when you are out of work and desperate, you will consider doing things you wouldn't even consider at other times.

[ken melvin]

First, one must lie to oneself. It follows, that in order cease lying to others, one must first cease lying to oneself. If my children were hungry and in need of clothing, and I could not get a job that payed enough to feed and clothe them, or get assistance; I would steal to get them food. If I were a young black man with no hope of getting a job being constantly bombarded with images of what I'm supposed to do and wear; I would probably steal, which brings me back to Wilson's. Wilson's laws are meant to protect property, specifically the property of his. To Wilson, this is paramount, and so he would have crime control based on fear of punishment rather address the distribution of opportunity. property or wealth. The Wilsons of So Cal have hastened our becoming a third world nation.

[gordon]

StMac raises an interesting point, which also came up in previous discussions (see my earlier comment). I wonder if anybody has studied the harmfulness of crimes from an economic perspective. Is white collar crime more economically harmful than blue collar crime? Are police/legal resources deployed according to the harmfulness of crimes? How would you measure the economic harmfulness of crimes - would you look, for example, at resource misallocation and opportunity cost?

[Patricia ShannonP]

gordon

Like history, the "crime" is defined by those in power.

[Patricia ShannonP]

That should have been

Like history, the "crime" is defined by those in power.

[kthomas]

@ Ken Melvin "Only a Pepperdine fool....."

Very funny. They do graduate some oddballs, but the view of the Pacific from thier campus is spectacular.

[Aesthete]

Another explaination of lower crime rates after the 70's supposedly was the legality of abortions at that point.

[kthomas]

Aesthete, great comment. We definitely did not consider that earlier.