Ezra Klein: American Furies

« Metaphors | Main | The Consumers Are Speaking »

June 12, 2007

American Furies

I haven't read Sasha Abramsky's American Furies: Crime, Punishment, and Vengeance in the Age of Mass Imprisonment, but it's next on my list. This country's culture of imprisonment is a critical, and usually under-noticed, contributor to crime, inequality, poverty, the breakdown of the nuclear family in urban areas, depressed earnings among black men, intergenerational transmission of economic status, etc, etc, etc. It's hard to find good reporting and analysis on these issues, as the imprisoned are, by definition, out of sight, and deeply unsympathetic, so their lives and outcomes take on less urgency. But Silja Talvi's review suggests Abramsky's book is more than up the challenge.

Good as Abramsky's book may be on the informational issues, the "what can be done" part of this is tough, as rehabilitative, humane prison reforms are considered politically suicidal. But though I agree with that conventional wisdom, Rhode Island has restored the vote to felons, and in a more amazing shift, Florida's Republican governor did the same thing, so maybe the politics of the issue are changing.

June 12, 2007 in Books | Permalink

Comments

"This country's culture of imprisonment is a critical, and usually under-noticed, contributor to crime, inequality, poverty, the breakdown of the nuclear family in urban areas, depressed earnings among black men, intergenerational transmission of economic status, etc, etc, etc."

Sure is, and it's not just a cause, it's also effect.

"the "what can be done" part of this is tough, as rehabilitative, humane prison reforms are considered politically suicidal."

I don't think so, not if you focus on nonviolent drug-related offenses: there'd be good bipartisian, libertarian-liberal coaltion to reform the country's drug laws.

And because I'm incapable of making a comment without tying it to the Primary,
I really wish Edwards would make withdrawal from the WOD a centerpiece of his campaign. It's not only a critical issue, it could help him with black voters.

Posted by: david mizner | Jun 12, 2007 12:44:07 PM

You won't see that promise from anybody but Ron Paul, and possibly Bill Richardson. Sorry.

Posted by: Mike | Jun 12, 2007 1:09:04 PM

Another good book on the same subject is called A Rage to Punish. I can't remember the author's name, but she was a Pennsylvania judge who resigned out of frustration with mandatory sentencing laws. Her point was that most criminals were handicapped by their lack of basic skills (literacy was the one she focused on), and that punishing them, without addressing the skills deficits, was useless. She was also concerned with how the system discriminates against the poor and minorities.

Posted by: beckya57 | Jun 12, 2007 1:15:44 PM

According to this paper by Steven Levitt, the biggest cause of the dramatic decline in crime rates in the 1990s was the increase in the prison population. The effect applies to both violent crime and property crime.

Posted by: JasonR | Jun 12, 2007 1:51:46 PM

Here are some more sources on this issue:

- The Urban Institute publications
- US Department of Justice reentry website (with links to publications)
- Socialogist Chris Uggen's blog

Posted by: anon | Jun 12, 2007 2:10:31 PM

Ha ha, very funny, and was there a 'dramatic decline' in corporate crime during the 90s? But there's no need to savage the paper, as the author himself admits that it is probably of little predictive value.

The fact remains that the Salem Witch Hunts and the casting our of demons are the models for American imprisonment- that, and, increasingly, the use of prisoners as slave labor, and of guards as a political slush fund.

Just as hospitals are used to make most of us feel 'well' by segregating the 'sick', and schools are used to make most of us feel 'smart' by segregating the 'ignorant', so prisons are used to make most of us feel we are the good guys and the bad guys are in prison.

Even lawyers who witness the most brutal of white collar crimes will simply swallow their gorge, because, as one patiently explained to me in a civil matter, "Everyone in the courtroom knows each other, they all dress alike, they work with each other every day. As an outsider, you haven't got a chance."

It oughta be a crime.

Posted by: serial catowner | Jun 12, 2007 2:30:38 PM

People who don't want violent criminals incarcerated and punished aren't serious. Comparing thugs (who can hurt you) to witches (who can't) is ludicrous.

Crime prevention efforts that focus on social issues as well as law enforcement are worth discussing. [P.S. Crime is itself one of the social issues that hampers troubled communities.] Improving our nation's efforts to rehablitate inmates and support the rehabilitation post-release should be a winner from virtually any perspective. However, a realistic observer will concede that rehabilitation is only a piece of the puzzle and one that won't take for many the first time(s) no matter how stellar the programs.

Posted by: slickdpdx | Jun 12, 2007 2:46:19 PM

Met a guy at an AA meeting yesterday who's in the Montgomery County (MD) Pre-Release Program, a kind of supervised incarceration under which he can hold a job and attend functions like AA. He received a three year sentence for selling two pounds of weed. It's a safe bet that the county could have sent him to the University of Maryland for less money..... Note that Montgomery County is among the most liberal in the nation. Oh, did I mention that he's black?

Posted by: Anon | Jun 12, 2007 3:15:34 PM

catowner,

Ha ha, very funny, and was there a 'dramatic decline' in corporate crime during the 90s?

I don't know. According to the paper, there was a dramatic decline in both violent crime and property crime, as measured by both the Uniform Crime Report and NCVS crime victimization survey.

But there's no need to savage the paper, as the author himself admits that it is probably of little predictive value.

There's a need to savage the paper if you dispute its finding that the biggest factor in reducing crime in recent years has been the increase in the prison population.

Posted by: JasonR | Jun 12, 2007 3:16:35 PM

Well, I doubt things have gotten better since the mid-70's--and in the mid-70's, G. Gordon Libby thought his fellow prisoners were more admirable than the people running the jails.

As far as the War on (some) Drugs, it's worth noting that there'd be a substantial anti-anti-drug constituency in "God and Guns" culture too.

Posted by: SamChevre | Jun 12, 2007 4:16:51 PM

First of all, a very large number of our prisoners are simply victims of prosecutors in the drug war seeking the longest possible sentences. People who are not violent criminals and never should have been prosecuted, much less imprisoned.

Second, most of the remainder are just early-20s screwups who could probably be made into valuable members of society if we wanted to.

Give them training, employ them at a real job during their imprisonment, pay them a minimum wage that goes into a post-release trust fund, and furnish real contacts for real employment after release.

Do you see any of that happening? If so, put down your crack pipe and back slowly away from it- you've had enough already.

This largest prison system in the world is also the largest criminal manufacturing system in the world- fittingly run by other criminals who pay the inmates 70 cents an hour, and then take that money and charge the inmates for their "room and board".

It's nothing other than legalized slavery, and the comment by slickdpdx illustrates perfectly how a main purpose of the system is to make the people outside the prisons feel that they are better than the people inside.

If JasonR would brush up his reading skills and then re-read the paper, he might notice the part where the author says it has little predictive value. I did not dispute the papers finding, mainly because JasonR doesn't yet seem to understand that, in reality, a contractor killing an employee by ignoring safety standards is a crime, even though it doesn't count as such in our present legal system.

That's the difference between law and morality, and in our society, it's a big difference.

Posted by: serial catowner | Jun 12, 2007 7:44:27 PM

If JasonR would brush up his reading skills and then re-read the paper, he might notice the part where the author says it has little predictive value.

JasonR already did read that part. If serial catowner would brush up his thinking skills, he might realize that the purpose of the paper was not to predict future crime trends, but to identify the causes of past ones. And the author's finding is that the primary cause of the dramatic decline in crime during the 1990s, both violent crime and property crime, was putting more people in jail.

Posted by: JasonR | Jun 12, 2007 8:41:21 PM

A couple of observations on Jason's link. First, the author acknowledges that the scale of incarceration has significant social costs in poor communities that are not related to crime and questions whether the reduction in crime is worth the cost. Second, the author also points out that an increased police presence produced a significant chunk of the crime reduction much more cheaply. So it's accurate but not quite the point to say that prison was the "biggest factor" causing the decline in crime rates -- it was also by far the costliest per unit of crime reduction in financial terms and human terms. Police presence was the most efficient and least controversial factor. Indeed, the author notes that greater reductions could be achieved at less social and economic costs by "substituting" expanded police forces for prisons. Third, I have never seen a study that touts the crime reducing benefits of prison compare them to the increased crime that occurs in prison in absolute terms because the population is larger there. I am not willing to say that shifting the victim of crimes to those who have done something bad is worth the social and financial costs. Fourth, and relatedly, remember that expansion of prison carries an implicit utilitarian costs. If the evidence isn't clear, you are just engaged in punishment for it's own sake.
Finally, the paper relies on other researcher's studies to estimate the relationship between prison and crime. Its own contribution is limited, and I didn't find its descriptions of the methodology particularly transparent. That's fine, the paper had its own goals and these may have been to estimate the different contributors to the crime rate reduction in that historical period using what it perceived to be the consensus in the area.
But, obviously the studies it cites hardly exhaust the field (if anyone is more familiar with them, please share). Here is a more exhaustive "meta-analysis" that attempted to aggregate date from 50 studies of the issue, published by the Canadian Solicitor General. It finds slight increases in recidivism associated with the use of prison as a penal sanction. Here is the Canuck Study This only tests the deterrence and school of crime theories, and does not test incapacitation. It persuasively critiques ecological studies of the relationship between prison population and crime rates, however, as producing unrealistically large correlations that suggest the presence of confounding variables.
A 2004 study by Raphael and Stoll Here is the prison release study
employs the "prison release" methodology referenced in the study JasonR provided. This study does show a relationship between prison releases and increased violent crime. It also has a number of caveats: 1) it notes that the relationship between prison releases and crime rates is less in the 1990's, suggsting that the potential for increased crime was being suppressed by other social variables at work during that time period, and 2) more critically, it notes that the increase in crime associated with the prison releases is suppressed where the releases were managed by strong parole boards, and argues that reductions in crime could be achieved by changing who is incarcerated rather than the scale of incarceration. If follows, also, that some net crime reduction could be achieved with a reduction in the scale of incarceration. I would also add that a judicial prison release (when was the last time we had one of those? surely no later than 92 or so) may be a poor device to measure the effect of reducing the number of inmates committed to prison in the first place; the increase in crime associated with a prison release is consistent with school of crime theories. It might even be a poor device to measure the effect of shorter sentences, if you believe in the efficacy of prison rehabilitation programs that would be disrupted by an early release.

Anyway, there certainly are plenty of people who study the issue rigrously who believe that prison reduces crime and is worth it even at large scale, so I would be interest to see any other data.

Prediction -- my links will fail because of my html idiocy.

Posted by: RW | Jun 13, 2007 2:41:37 AM

Hey the links worked! I did it!

Hey look everybody I...My very first... They go to the studies and everyth...

Excuse me.

Posted by: RW | Jun 13, 2007 2:43:55 AM

This all reminded me of a story I usually try to forget.

A 19-year old in my county was sentenced to life in prison for "three strikes". The first strike, brandishing a knife. The second strike, stealing a jacket while brandishing a knife. The third strike, brandishing a knife.

It seems that, age the age of 13, this Native American was stolen from his family by white social workers using bogus charges of "child abuse". A good student up to that time, he became sullen and angry after that occurred.

During the first five years I lived in this county, I became aware of two people murdered by these social agencies, who systematically deny clients the assistance they are entitled to by law, and reward employees who will lie about the clients by promoting them.

Reading the news stories, especially those about the lawsuits on behalf of abused clients, makes it plain that this abuse is widespread in "liberal" Washington state. Don't bother thinking it is easy for abused clients to find a lawyer who will represent them, it is almost impossible.

I don't think putting people in prison during the 90s led to any fall in crime at all, and the author of the paper cited by JasonR quite rightly restricts conclusions to the data he studied and refuses to say it is predictive.

Naturally, all of this is very depressing, and none of it tells us what we want to hear- that we are the kindest most generous people in the world, that our prisons serve a valuable social function, and that our justice system puts the bad guys in jail but lets the good guys roam free.

Well, go ahead and believe what you want. But don't blame the judge, who in this case put the blame squarely where it belonged- on a society that, by law, forced him to impose a sentence that was clearly not fitted to the crime. That's us, folks.

Posted by: serial catowner | Jun 13, 2007 8:22:07 AM

RW,

I'm glad you acknowledge Levitt's finding that increased imprisonment was by far the biggest factor in reducing crime in the 1990s, but your other comments misrepresent Levitt's statements and findings in important ways.

First, the author acknowledges that the scale of incarceration has significant social costs in poor communities that are not related to crime and questions whether the reduction in crime is worth the cost.

This is an exaggeration of Levitt's statement. He says merely that "it seems quite plausible that substantial indirect costs are associated with the current scale of imprisonment, such as the adverse societal implications of imprisoning such a large fraction of young African American males." I see no statement questioning whether the reduction in crime is worth the cost.

Second, the author also points out that an increased police presence produced a significant chunk of the crime reduction much more cheaply. So it's accurate but not quite the point to say that prison was the "biggest factor" causing the decline in crime rates -- it was also by far the costliest per unit of crime reduction in financial terms and human terms. Police presence was the most efficient and least controversial factor. Indeed, the author notes that greater reductions could be achieved at less social and economic costs by "substituting" expanded police forces for prisons.

Again, this statement is a significant misrepresentation of what the author actually wrote. Levitt found that "a dollar
spent on prisons yields an estimated crime reduction that is 20 percent less than a dollar spent on police." I don't think a 20% difference qualifies as "much more cheaply." But more importantly, the amount of crime reduction produced by increased policing was found to be only about half the amount of crime reduction produced through more imprisonment, and the confidence level for the policing finding is only "medium," whereas the confidence level for the imprisonment finding is "high." Hence, increased policing may be a more cost-effective substitute for increased imprisonment only "at the margins" rather than as general policy.

Though Levitt doesn't mention them, there are also other costs associated with increased policing, most obviously the loss of privacy and the civil liberties concerns associated with increased government surveillance of the civilian population. Britain has pursued an aggressive anti-crime policy of increased police video surveillance of public spaces. This "Big Brother is Watching You" approach may be an effective way of reducing crime, but it is not without its own social costs.

Of course, the two approaches are complementary, and both increased policing and increased imprisonment seem to be important ways of reducing crime.

Posted by: JasonR | Jun 13, 2007 1:35:49 PM

RW,

Here is a more exhaustive "meta-analysis" that attempted to aggregate date from 50 studies of the issue, published by the Canadian Solicitor General. It finds slight increases in recidivism associated with the use of prison as a penal sanction. Here is the Canuck Study This only tests the deterrence and school of crime theories, and does not test incapacitation.

As far as I can tell it addresses only recidivism. It doesn't test deterrence (other than deterrence of future crime by the offender himself as measured by recidivism) or incapacitation. Its finding seems to be consistent with the findings of the Levitt study, which addresses overall crime rather than just crime resulting from recidivism.

Posted by: JasonR | Jun 13, 2007 1:55:06 PM

WARNING --PEDANTRY FOLLOWS

So. The author approximates that increases in prison population caused “about one third of the observed decline in crime.” He also says that “Extrapolating the conservative estimates of Donohue and Levitt (2001) to cover the period 1991–2000, legalized abortion is associated with a 10 percent reduction in homicide, violent crime and property crime rates, which would account for 25–30 percent of the observed crime decline in the 1990s.”

1/3-25% = 8%
1/3-30%= 3%

You then say that Levitt called the increased prison population “the biggest cause” of the “dramatic drop” in crime, “the primary cause of the dramatic decline” in crime, and “by far the biggest factor in reducing crime in the 1990s” (actually you attribute the last characterization to me, which is not what I said).

Well, OK, I am not going to call you a liar. A reasonable person could think that a 3%-8% difference between two causes made the greater cause “by far” the biggest cause. Why not?

On the other hand, I look at the same author and note that he says “a dollar spent on prisons yields an estimated crime reduction that is 20 percent less than a dollar spent on police.” I then paraphrase this finding by saying that “an increased police presence produced a significant chunk of the crime reduction much more cheaply…”

Your response is to call my conclusion that 20% more cheaply = “much more cheaply” a “misrepresentation" (indeed, a "significant misrepresentation") because you "don't think a 20% difference qualifies as "much more cheaply."

If saying that 20% more cheaply = “much more cheaply” is so wildly implausible that the only possible explanation is dishonesty, then what is the explanation for your statement that a 3%-8% difference between two crime reducing factors renders the larger factor “by far the biggest factor”?

I guess that means you are a 2.5-6.66 times bigger liar than I am.

But I tend to think that neither one of us is lying. Maybe we are reading the same study and certain factors about it are salient to us, and we are relating them the best we can, and shouldn’t accuse each other of dishonesty.

I have built my professional life on the belief that less prison is better than more. But I am genuinely willing to reexamine that belief in light of credible research. I can recognize that my ideological lens provides insights that should be listened to, but also distorts some of the picture I am seeing. Are you generally willing to consider possible limitations in the data on which you rely? I hope so, because you find interesting articles, and sometimes make some pretty intelligent points.

Posted by: RW | Jun 15, 2007 2:01:10 AM

The main issue is not the elimination of prisons but in determining what in fact is a crime and where "punishment" should be considered.

This a complex issue entailing morality, racial, socio-economic, social justice, victim restoration, cost, and societal cohesion consideration.

If we look at our results we find the US has more people behind bars than any other nation, most are poor, minority, African males. Many of these are non-violent offenders (to start) who with skyrocketing recivitism rates. And all at a cost that is substantially per capita greater than anywhere else.

Most non-violent crimes can be kept out of the state's legal system through restorative justice. This holds the offender accountable to the victim and community rather than the state. Results have created a drastic reduction in recivitism (as low as 6%), the victim and community meet the accused directly and work through atonement in the form of community service and where necessary financial retribution. All around the community is restored with a cost of hundreds of dollars rather than hundreds of thousands of dollars to lock up and offender.

When the offense is repeated or violent than it is turned over to the state. In the case of this book, what is called for is the decriminolizaton of drug addicts and the addication is treated as a public health problem rather than a crime. This again is at a fraction of the cost of incarceration, and allows the individual to move away for his/her addiction and to a contributing member of society.

Statistics showing increased rates of incarceration correlate to decreases in crime are based on how one defines a crime. So, if we insist on calling addiction to narcotics a crime and throw people in jail, the rate of that "crime" may go down, but so many other critical factors worsen.

Posted by: Maxfield | Aug 18, 2007 9:01:08 PM

托盘
 托盘
 钢托盘
 钢制托盘
 塑料托盘
 木托盘
 木制托盘
 纸托盘
 木塑托盘