« High costs of sex offender civil confinement driving reform talk in Minnesota | Main | Ohio asks SCOTUS to vacate stay of state execution »

January 20, 2012

"The Law and Economics of Fluctuating Criminal Tendencies"

The title of this post is the title of this interesting-looking new paper up on SSRN from Murat Mungan. Here is the abstract:

Law and economics is one of the most successful legal disciplines, and it is used frequently to study various legal issues, doctrines, and policies.  Its application to criminal law has received great attention by economists as well as lawyers.  But, economic analyses of criminal law are often criticized for employing unrealistic assumptions and performing poorly in providing rationales for various criminal law doctrines.  In light of these criticisms, one may be tempted to jump to the conclusion that law and economics in particular, and consequentialist approaches in general, are inappropriate for studying criminal law and procedure.

In this article, I demonstrate that the poor performance of economic analyses in providing rationales for a number of criminal law doctrines is not the result of some inherent problem with consequentialism, but its narrow application.  Specifically, I show that the two main criticisms mentioned above are interrelated, and that addressing the first — by employing more realistic assumptions — at least partially resolves the second.

In particular, economic analyses of criminal law can achieve greater explanatory power by incorporating two simple facts: (i) the degree of control one can exert over himself to abstain from committing a wrongful act fluctuates over time, and (ii) the incapacitative function of imprisonment contributes to reductions in crime.  Despite the simplicity of observing these two facts, almost all normative economic analyses of criminal law ignore them. Instead, they assume that criminals have constant criminal tendencies, and that the only benefit of imprisonment is deterrence.

This article demonstrates that an economic analysis incorporating the two simple facts mentioned above provides convincing rationales for (i) why repeat offenders are punished more severely, (ii) why there is a de facto tendency to punish offenders, who are believed to be remorseful, less severely, and (iii) why voluntary manslaughter is punished less severely than murder.  Because previous economic analyses employing narrower approaches face difficulties in providing similar rationales, I conclude that the economic approach gains significant explanatory power by applying consequentialism more broadly.

January 20, 2012 at 05:47 PM | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d83451574769e20168e5e064f1970c

Listed below are links to weblogs that reference "The Law and Economics of Fluctuating Criminal Tendencies":

Comments

From the abstract: "[T]he incapacitative function of imprisonment contributes to reductions in crime."

Heresy!

Posted by: Bill Otis | Jan 20, 2012 8:52:41 PM

Good Evening Bill: I am fine, I hope you are too.

Leave it to you to search through a haystack of BS and find the needlle you were looking for.

Have a good night!

Posted by: albeed | Jan 21, 2012 12:07:03 AM

albeed --

I think it's a compliment to say that a person can find a needle in a haystack. I have to confess that I have practice. When you review as many defendants' statements as I have about how those ten bricks of cocaine happened to wind up in the trunk of their car without their having anything to do with it, you get used to sorting through a lot of, ummmmm, hay.

I live in Hawaii in the winters, so I'm just now off to dinner. Sleep well!

Posted by: Bill Otis | Jan 21, 2012 12:54:42 AM

Aggregate some consequentialism and you get utilitarianism.

And as Bill points out, incapacitation is the sole benefit of the criminal law.

Very good start.

Next we need an admission is that the criminal law is to protect the public, and not provide make work jobs for lawyers.

Posted by: Supremacy Claus | Jan 21, 2012 1:00:26 AM

Bill:

I have the means but don't live in Hawaii. In my mind, the Big Island is the one MUST See. What do you believe?

Posted by: albeed | Jan 21, 2012 11:10:16 PM

albeed --

I couldn't agree more. The Big Island is a miracle of nature -- you can sit on the beach on the Kohala Coast and look up at snow-capped Mauna Kea. The rain forests, tropical gardens, black sand beaches, on and on.

It doesn't hurt that today, statistically the coldest day of the year in the Northern Hemisphere, was 83 degrees and (as usual on the west side of the Island), cloudless. The climate here simply must be the best in the world.

We live here until March, about three miles north of the tiny town of Kawaihae on what is, for a variety of reasons, called the Gold Coast. Drop on by!

Posted by: Bill Otis | Jan 22, 2012 1:57:15 AM

nice place to live bill going by the photo's ive seen. that is one of the few states i have not been to!

Posted by: rodsmith | Jan 22, 2012 3:43:14 PM

rodsmith --

You should come on down too. You're one of the good guys of this forum. Not that I agree with all you say, but you never have this superior, snarky tone that's getting all too popular around here.

Incidentally, the place looks even better than in the travel books. It's amazing. And you're never cold.

Posted by: Bill Otis | Jan 22, 2012 8:52:35 PM

Bill:

Thank you for your description and comment. One of these days we'll talk over coffee at Georgetown.

My current joy is living in trout country like John Travis, author (and Michigan Supreme Court Justice) of "Anatomy of a Murder".

Posted by: albeed | Jan 22, 2012 10:53:57 PM

thanks for the invite bill. IF i ever manage to pick the right 6 numbers and hit the florida lotto i'll take you up on it!

Posted by: rodsmith | Jan 23, 2012 2:30:57 AM

Regarding the two key insights described, they don't seem as new or original as the author seems to believe, and it's untrue that "almost all normative economic analyses of criminal law ignore them. Instead, they assume that criminals have constant criminal tendencies, and that the only benefit of imprisonment is deterrence."

The literature of "deterrence," though, discusses both general and specific deterrence, the latter of which is a synonym for incapacitation. So like Bill, the author fancies he found a needle in a haystack but the discovery hinges on ignoring the pile of nails sitting alongside it and pretending the needle is some great wonder.

Changing criminogenic risk over a person's lifetime are a well-known phenomenon, not some new idea among anybody but the rarified set of L&E number crunchers whose opinion the author values. As Bill and SC's comments show, the incapacitation aspect of incarceration is not ignored but in fact frequently overstated and fetishized. The best econometric studies of incarceration (e.g, Levitt, Spelman) estimate locking people up is responsible for about a quarter of the crime reduction over the last two decades.

Finally, changing criminogenic risk over time also argues AGAINST super-long sentences in most cases, since the cost-benefit ratio goes up as the risk of recidivism goes down with age. If you're going to embrace a concept, embrace it all, not just the slender "needle" that catches your fancy.

Posted by: Gritsforbreakfast | Jan 23, 2012 5:36:34 AM

Gritsforbreakfast:

you seem to have two main criticisms; (i) that the author's claim regarding the incorporation of points (1) and (2) are not novel, and the author is unaware of the effects of aging on a person's tendency to commit crime.
(i) It is clear that you have not read this interesting looking paper. The author states clearly that the incorporation of (1) and (2) are novel in the NORMATIVE L&E analysis. Name one paper which simultaneously is a normative L&E paper and incorporates points 1&2. Can't? Because it doesn't exist. He gives a detailed lit. rev. to back up his claim. And his lit. rev. is to a great extent VERY accurate.
[He does not but should have a section on whether fluctuating tendencies have been incorporated in the norm. L&E analysis (because Cooter (1991) includes what the author calls fluctuating crim. tendencies). I've informed the author of this. But, this is not harmful, because for the authors claim to work, both (1) and (2) need to be in the model...]
The author does, of course, provide information concerning Levitt's studies. Just search for Levitt, and youll see. I suspect you do not know the difference between normative and descriptive studies? If so, you must understand that Levitt's studies are descriptive, and the author does not claim that such studies do not exist...

(ii) It is obvious that the author is aware of the "ageing effect" see p. 19 ("However, if an offender’s dangerousness declines with age..."). Granted, he does not consider the effect of ageing in his analysis, but must or even should he? Afteralll, his claims are already consistent with what you claim (i.e. "changing criminogenic risk over time also argues AGAINST super-long sentences in most cases"). Why over-complicate the analysis with an issue that helps make his claim even stronger?

Again, it is obvious that you haven't read the paper, the author specifically notes this, see p. 11 "A second goal of this article is to re-iterate the fact that economic models of crime do not necessarily lead to policy implications endorsing extremely harsh penalties".

This is the problem with criticisms over blogs. People get extra-brave, because there are no consequences to making false claims…

Posted by: Golyadnik | Jan 23, 2012 7:20:57 AM

Golyadnik (if that is your real name), Levitt and Spelman included calculations for costs averted from future crime from both specific and general deterrence, so if what you say is accurate it's only by defining away those who don't fit into the small category of L&E normative theorists. My point was precisely that just because L&E theorists haven't considered a point doesn't mean it's been unconsidered, only that this narrow group of self-avowed experts has ignored a small mountain of social science research in their Olin-funded exploration of the allegory between prices and criminal sentencing.

Interesting that you speak of having "no consequences" but use a pseudonym with no identifying info. You may disagree with what I say, but I don't hide behind anonymity on the web.

Posted by: Gritsforbreakfast | Jan 23, 2012 8:03:01 AM

Grits --

"As Bill and SC's comments show, the incapacitation aspect of incarceration is not ignored but in fact frequently overstated and fetishized."

I assume you're not including this blog, where the incapacitation effects of incarceration are ignored, shuffled off into the corner or point-blank denied by the great majority of commenters.

"The best econometric studies of incarceration (e.g, Levitt, Spelman) estimate locking people up is responsible for about a quarter of the crime reduction over the last two decades."

Meaning, specifically, that there are over one million fewer serious, non-drug crimes PER YEAR now than there were 20 years ago, because of increased incarceration. Yet more specifically, it means in excess of 2000 fewer murders.

Is that something we should sneeze at?

"Finally, changing criminogenic risk over time also argues AGAINST super-long sentences in most cases..."

Super-long sentences are not given in "most cases." Surely you must know this.

"...since the cost-benefit ratio goes up as the risk of recidivism goes down with age."

I do not disagree with that, but would note the related and universal statistical phenomenon of diminishing returns to scale. The fact that there are diminishing returns to the scale of imprisonment does not mean, however, that there are zero returns. It means you're getting less per dollar spent, not that you're getting nothing.

Posted by: Bill Otis | Jan 23, 2012 4:27:20 PM

Bill -
I hesitate do dive into this one, but I've seen something in your comments on other posts (as well as this one) that I've wanted to point out for a while.

The study by Levitt, et al., point to reasons besides incarceration-as-incapacitation too. Consistently, these numbers are thrown out and you will agree thus:

"Meaning, specifically, that there are over one million fewer serious, non-drug crimes PER YEAR now than there were 20 years ago, because of increased incarceration. Yet more specifically, it means in excess of 2000 fewer murders." (Jan 23, 2012 4:27:20 PM)

My question has two parts, if you so choose to answer. (1) Are these numbers indicative of 25% of the crime decrease (per Levitt), or the full amount of crime decrease as observed from 1993-present. (2) If those are the full numbers, why does it seem you attribute the full amount of the crime decrease over these 18 years to increased incarceration when the econometric studies cited only attribute a quarter of this decrease to it?

OK, one more. (2)(a) If 75% of the crime decrease observed over the time period of the Levitt study are attributable to factors other than increased incarceration, why don't these factors get more press or legislative attention?

Posted by: Eric | Jan 23, 2012 5:15:08 PM

Eric --

Thank you for your questions.

1. In 1991, there were 14,872,900 serious, non-drug crimes. By 2010, there were 10,329,135. That is a total drop of over 4.5 million. Twenty five percent of that -- the percentage caused by increased incarceration -- is 1,125,000. Thus my statement was correct: Because of increased incarceration, there are over one million fewer serious, non-drug crimes now than twenty years ago.

To me, that is a mind-blowing success, not the "broken system" we constantly hear about here.

1a. In 1991, there were 24,700 murders. By 2010, the number was 14,478 -- a total drop of almost 10,000. Twenty five percent of that is about 2500 fewers murders. Thus I slightly understated my case by saying that there were 2400 fewer murders attributable to increased incarceration (which is far more common as the sentence for murder than the DP).

Source: http://www.disastercenter.com/crime/uscrime.htm

2. The other factors leading to the massive crime decrease (e.g., more targeted and aggressive policing and the federal government's long but successful efforts to stifle the crack wars) in fact do get mention by the press and academia -- which, however, mostly pretends to be "baffled" by the crime decrease. Incarceration will grudgingly get mentioned, but only grudgingly.

Posted by: Bill Otis | Jan 23, 2012 5:47:24 PM

Bill, no one said the 1/4 of crime reduction attributable to mass incarceration is anything to "sneeze at," only that folks like you who want to attribute ALL crime decline to incarceration dramatically overstate its benefits and ignore other factors that reduce crime a lot more.

As for diminishing returns on incarceration over time, I'm glad you agree, and I agree with you that as inmates age, "you're getting less per dollar spent, not that you're getting nothing." But in a cost-benefit analysis, when costs exceed benefits, especially by a significant margin, that truth doesn't justify continued long-term incarceration past the point where marginal cost equals the marginal benefit from crime reduction, at least not according to any econometric model. Where we disagree is that you support incarceration far past that point and I think criminal justice spending should be subject to cost-benefit analysis just like any other taxpayer funded policy.

Posted by: Gritsforbreakfast | Jan 23, 2012 6:35:12 PM

Grits --

"Bill, no one said the 1/4 of crime reduction attributable to mass incarceration is anything to 'sneeze at,' only that folks like you who want to attribute ALL crime decline to incarceration..."

At no point, here or elsewhere, by statement or implication, have I "attribute[d] ALL crime decline to incarceration...," nor do I want to. Instead, I have explicitly stated the opposite -- that there are other factors which also contributed to the decline. The study upon which I principally rely -- Spelmand and Levitt -- says as much.

When we have 1,125,000 fewer serious crimes now than 20 years ago because of the increase in incarceration, however, that is a fact of enormous significance. Indeed it's staggering. And it should be constantly borne in mind in the debate about "incarceration nation."

Posted by: Bill Otis | Jan 23, 2012 7:45:03 PM

hey bill how are you doing there in hawaii. saw the article about the 5.0 earthquake on the big island yesterday.

hope you and yours are fine.

Posted by: rodsmith | Jan 24, 2012 12:25:00 AM

Bill and Grits -

This was precisely the reason I asked the questions I did above to Bill. Thank you, Bill, for answering them. The underpinning of my questions were not to discredit that mass-incarceration has reduced serious, violent crimes.

I did, however, seek to point out (in a Socratic way) that 3,375,000 serious crimes are not being committed because of other factors beside increased incarceration. The returns from the other contributing factors are three times as high.

While I strongly advocate for long incapacitation periods for violent criminals, especially violent recidivists, the idea of longer sentences caught on and spilled over into non-violent crimes as well. This seems to happen a lot: a good idea gets construed ad absurdum (see 18 U.S.C. §922(g)(3) and SORNA, to name two examples)

I do not, however, advocate for incarcerating non-violent criminals for obscene amounts of time. I know that "obscene" is subjective and not qualitative, but I digress. Does this belief give me a label of left-wing apologist "incarceration nationer"? If it does, I think I can live with that.

Posted by: Eric | Jan 24, 2012 12:25:37 AM

Bill, murderers have pretty low recidivism rates, so the idea that mass incarceration caused the murder decline makes little sense. There are some murderers who are repeat or "serial" killers, but they're a decided minority and their incarceration doesn't account for the lower number of murders. (Incapacitation gets more bang for the buck when it's focused on recidivists, and most murderers will never kill again.) Instead, economic and cultural factors, not to mention improved hospital trauma care, had much more to do with declining homicides than incarceration.

As Eric points out, 3/4 of the recent crime reduction had nothing to do with incarceration. (Levitt thinks legalization of abortion was a major causal factor - perhaps you agree since you say you rely on his work - and that spending on police reduces crime much more than spending on incarceration. You claim "At no point ... have I "attribute[d] ALL crime decline to incarceration." But you never mention any of these other causes and advocate incarceration even when a cost-benefit analysis no longer justifies it. ("It means you're getting less per dollar spent, not that you're getting nothing.")

Dr. Spelman concluded based on a cost-benefit analysis (with the "benefit" being crime reduction), that incarceration levels are more than double what would be required to maximize public safety, so if you're relying on their analyses (are you sure you've read their work? they wrote separate papers, it's not the same study) then your boosterism for expanding incarceration makes little sense. Based on the Levitt and Spelman studies, if that's really what you're relying on, it's hard to understand why you cherrypick the one aspect you like ("incarceration reduces crime") but ignore the larger causes and more profound conclusions the authors draw from the same data regarding policies that reduce crime even more.

Posted by: Gritsforbreakfast | Jan 24, 2012 10:23:21 AM

Grits --

"As Eric points out, 3/4 of the recent crime reduction had nothing to do with incarceration."

Not only did I never deny that, I implicitly but unmistakeably affirmed it by noting that incarceration accounts for 1/4 of the decrease. No sensate person could fail to understand that that means other factors accounted for the rest. And I noted, on this thread, two of the other crime reducing factors Levitt and Spelman point to.

(On earlier threads, I have mentioned the finding of greater availability of abortion as well. I have also repeatedly provided the link to the Levitt paper for those who wish to see for themselves.)

"You claim 'At no point ... have I "attribute[d] ALL crime decline to incarceration.'"

That claim is true, as I would be happy to testify under oath. Your contrary claim is false.

"But you never mention any of these other causes..."

That claim is also false. I mentioned two of them on this very thread, Bill Otis | Jan 23, 2012 5:47:24 PM.

"...and advocate incarceration even when a cost-benefit analysis no longer justifies it. ('It means you're getting less per dollar spent, not that you're getting nothing.')"

That's because amorphous but central concepts like just dessert cannot be quantified and thus cannot be captured by cost-benefit analysis.

Example of the primacy of just dessert: The defense costs alone for the Timmy McVeigh capital case were over $13 million. We could have saved most of that by bargaining it down to LWOP. Had we done so, it would have been an outrage beyond words.

"...it's hard to understand why you cherrypick the one aspect you like ('incarceration reduces crime')..."

Because the great majority of the board cherrypicks everything and anything else.

"...but ignore the larger causes and more profound conclusions the authors draw from the same data regarding policies that reduce crime even more."

As noted, I do not ignore the other causes, which are, contrary to your apparent implication, cumulative to rather than exclusive of incarcertation.

Posted by: Bill Otis | Jan 24, 2012 9:39:17 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB