May 21, 2008

If the death penalty deters murder, how often must it be applied to have that deterrent effect?

Jac has 2 posts on this subject. In the first post, he catches Steven Levitt (of Freakonomics) in what I think is a real contradiction. And in the second post, he points to something striking about the statistics on deterrence and deals with whether he's contradicted himself about what he said about Levitt's supposed contradiction. I'm not going to try to summarize all that here, but it has to do with how a would-be murderer contemplates the prospect of execution — rationally? — and how quickly and often the state needs to execute convicts in order to maximize the deterrence of murder.

60 comments:

Sloanasaurus said...

I have never accepted the idea that the death penalty deters. I think the death penalty does two things: 1) it is sociatal vengence - it makes us feel good to put the bastards to death, and 2) it represents a finality that the murderer will never again walk the streets.

I don't have any moral ghast against the death penalty, but I wouldn't care much if it was eliminated (via the legislative process). However, there is some truth to worrying that some murderers will eventually get out of prison if not put to death, especially with liberals in charge of the government. The free Mumia movement is a perfect example of that.

Sloanasaurus said...

I think that if liberals would support the island prison concept, where death row innmates can be sent to live out the rest of their days living as cavemen and eating each other as cannibals, we could reach a compromise on ending the death penalty. The island prison would be a cheap and effective way to torture these murders without having to actually torture them.

rhhardin said...

It doesn't matter if it deters murder. It's there because it's justice.

So operations research on maximizing a deterrence effect (presumably with the least executions) is a red herring.

Minimizing executions isn't the point.

I'm Full of Soup said...

Once?

Was this a trick question?

Beldar said...

In states like California, where executions are extremely infrequent, those very few executions that do happen suffer from the same apparent arbitrariness that motivated the effective ban on executions in Furman v. Georgia (and its companion cases) in 1972.

By contrast, in the thirty years since that ban was lifted in Gregg v. Georgia (and its companion cases) in 1976, the State of Texas has made abundantly clear to everyone here that convictions for capital crimes can and do indeed regularly lead to executions. And that's a feature, not a bug.

Freder Frederson said...

This is where economists and statisticians are just so full of it and forget the limitations of their own tools. To conclude that each execution prevents X number of murders through a deterrent effect is just sheer speculation.

Your son's speculations are just as silly. He makes general observations about "human nature" and the criminal mind without apparently even bothering to do the barest research on the demographics or circumstances behind the typical murder, or more specifically, death row inmate. Not that you can blame him, because apparently neither did the authors of the studies.

People kill for lots of reasons. If JAC had bothered to listen to Fresh Air the other night he would have heard an interview with the former Chaplain of Texas' death row who presided over 95 executions. According to him, almost without exception, the people executed by Texas were poor, predominately minority, and many of them retarded. One of them was so mentally impaired he didn't even understand why he was being executed. They also were almost always either high or drunk when they committed their crimes.

Ann Althouse said...

Let's try to stick to the topic of deterrence. With deterrence out of the picture, people have strong opinions, but those who oppose the death penalty may change sides because of the statistics on deterrence. So if you are for the death penalty anyway, you should still care about this, because it has to do with convincing opponents to move to your side. I myself am a death penalty opponent, and it isn't helpful to me to hear people say, once again, that they want to save money or they want vengeance, etc. The statistical material is new and important, so let's focus on that here.

And Freder, the same to you. You are rehearsing the old anti-death penalty arguments. We all know that. We are trying to focus on something specific here. So, seriously, focus.

Beldar said...

freder:

If you want to know about deterrence, wouldn't it make sense to also interview the ones who were deterred?

Oh, yeah -- they're the ones who aren't in prison, because they decided not to kill the cop/hire a contract killer/kidnap and torture their victims etc.

I would agree with you that this is very hard to measure empirically. I don't agree that an inability to get exhaustive and unquestionable empirical measurements is a reason to categorically reject the concept of deterrence. Your argument asks that people ignore common sense, of which Prof. A's son's posts are good examples.

Freder Frederson said...

the State of Texas has made abundantly clear to everyone here that convictions for capital crimes can and do indeed regularly lead to executions. And that's a feature, not a bug.

But the state of Texas does not have a particularly low crime or murder rate. For all its tough on crime attitude, love of the death penalty and gun loving populace, Texas had a 2006 murder rate of 5.9 per 100,000, IL's was 6.1, MA 2.9, NY 4.8, CA 6.8, LA (which both likes to execute people and has more people in jail for life without parole than any other state) 9.9, VA (which is more rural than the other states) 5.2.

I have no idea with those statistics how they determined that Texas' or Virginia's murder rate would have been even higher.

Beldar said...

freder:

The question isn't what Texas' crime rate is compared to other states (whose populations vary dramatically from ours), but what our crime rate would be without the death penalty. You seem to have a hard time finding apples-to-apples comparisons today, friend.

Freder Frederson said...

Your argument asks that people ignore common sense, of which Prof. A's son's posts are good examples.

No my argument asks that we don't assume murderers think and act like educated, stable, reasonably intelligent, middle class people who are not either already completely enmeshed in a criminal lifestyle or have serious substance abuse problems.

Freder Frederson said...

The question isn't what Texas' crime rate is compared to other states (whose populations vary dramatically from ours), but what our crime rate would be without the death penalty. You seem to have a hard time finding apples-to-apples comparisons today, friend.

Which is my point exactly. Statistics cannot ever show a deterrent effect because there is no control population available. Anyone who claims otherwise should know better.

Freder Frederson said...

And Freder, the same to you. You are rehearsing the old anti-death penalty arguments. We all know that. We are trying to focus on something specific here. So, seriously, focus.

I am attacking the basis of the statistics as flawed because they assume that murder in general, and capital murder in particular, is an act that is deterrable through punishment. All the dubious statistics in the world cannot prove causation when there is reams of data that shows that most people who commit murder don't think or are incapable of considering the consequences and another good chunk think they are going to die young and violently anyway.

Your son falls into the liberal trap of assuming murderers are just like the girl or boy next door and when confronted with the possibility of punishment, it will change their behavior.

Ken Stalter said...

As I just argued on JAC's posts, I wonder whether the alleged deterrent effect can be explained by other correlations. For example, it seems that the liberal states are least likely to impose the death penalty and conservative states are most likely. We also know that liberal states are the more urban states and conservative states are most rural.

Could it be the case that there is just more homicide in urban areas and less in rural areas, regardless of how the death penalty is applied? I suspect that demographics may be at least a partial explanation of what we're seeing.

Freder Frederson said...

Could it be the case that there is just more homicide in urban areas and less in rural areas

Actually, that's not even necessarily the case. Look at MA, a fairly urban state. It has a murder rate half that of Texas, which is quite an urban state itself.

Wayne said...

While I technically oppose the death penalty on the grounds that you cannot undo an execution if the person is later found innocent, I'm not very concerned with the deterrent effect in principle.

I'm more concerned with: What's the recidivism rate for executed convicts? That's right, it's 0.

Anonymous said...

".... how often must it be applied to have that deterrent effect?"

Once per convict should do it.

Laura Reynolds said...

I don't think the way the death penalty is applied now has a deterrent effect and I doubt, even under a more expeditious system, that one could be proven.

That upon reflection and the prospect of execution a criminal may express regret, is of no use in preventing another person from commiting a crime which reveals a lack of respect for life, theirs or anyone else's.

Anonymous said...

"It doesn't matter if it deters murder. It's there because it's justice."

Thank you! This is precisely the reason that I don't much care whether capital punishment is a deterrent or not -- except that in these discussions, nobody ever mentions the one situation in which it is definitely a deterrent: In prison. Executed inmates cannot murder other inmates.

"Actually, that's not even necessarily the case. Look at MA, a fairly urban state. It has a murder rate half that of Texas, which is quite an urban state itself."

Actually, it is the case. Download the FBI stats on violent crimes and take a look. Urban areas have a much higher murder rate than rural areas. Looking at the stats by state doesn't tell you anything about urban v. rural crime, though I'm sure you know that, and you're just being disingenuous.

Anonymous said...

sloanasaurus said: "...some murderers will eventually get out of prison..."

or be let out for the weekend by some sympathetic liberal governor.

Meanwhile, it strikes me that the effort to show a correlation between inmate executions and crimes not committed could easily be an example of attempting to prove a negative. I also don't see how just the relevant variable(s) can be sufficiently isolated to make the study valid even if it doesn't suffer from the aforementioned fatal flaw.

Ken Stalter said...

"Actually, that's not even necessarily the case. Look at MA, a fairly urban state. It has a murder rate half that of Texas, which is quite an urban state itself."

Actually, it is the case. Download the FBI stats on violent crimes and take a look. Urban areas have a much higher murder rate than rural areas. Looking at the stats by state doesn't tell you anything about urban v. rural crime, though I'm sure you know that, and you're just being disingenuous.


Freder, thanks for your comments. I don't know you so I certainly don't want to accuse you of being disingenuous, but I do agree with rightwingprof about the urban versus rural murder rate.

The following link goes to a DOJ report on urban versus rural victimization.

http://www.ojp.usdoj.gov/bjs/pub/pdf/usrv98.pdf

The report seems to suggest that all violent crimes, including homicides, occur at a much higher rate in urban areas than in rural areas.

Freder Frederson said...

Actually, it is the case. Download the FBI stats on violent crimes and take a look. Urban areas have a much higher murder rate than rural areas.

I am not being disingenuous at all. All states with low murder rates are not entirely rural (e.g. MA). I would imagine that as far as urban/rural breakdown MA and TX are probably pretty close demographically. If anything, MA is probably more urban (and liberal and has tougher gun control laws and no death penalty) than TX, yet has a murder rate less than half that of TX. IL has nearly the same murder rate of TX with about a 50/50 urban/rural-small city split.

Freder Frederson said...

Freder, thanks for your comments. I don't know you so I certainly don't want to accuse you of being disingenuous, but I do agree with rightwingprof about the urban versus rural murder rate.

I never meant to imply it was, but the argument was made that conservative, rural states have lower murder rates. For the most part this is true. But Texas, which carries out the most execution, has a very large urban population, probably comparable to IL or MA. It is more than fair to use statewide figures for those states.

Original Mike said...

Don't you have homework to do, Freder?

Ken Stalter said...

I never meant to imply it was, but the argument was made that conservative, rural states have lower murder rates. For the most part this is true. But Texas, which carries out the most execution, has a very large urban population, probably comparable to IL or MA. It is more than fair to use statewide figures for those states.

My argument boils down to this:

Urban tends to imply both liberal politics and higher murder rates. Liberal politics tends to imply less frequent death penalty.

Rural tends to imply both conservative politics and lower murder rates. Conservative politics tends to imply more frequent death penalty.

Now for other demographic reasons, Texas may be a counterexample to the general trends I'm pointing to. I'm not sure that we're in actual disagreement.

My point is that I because of the differences between rural and urban areas, we can't necessarily say that more frequent death penalty is the cause of fewer murders. I suspect and believe that the deterrent effect may be real, but I have just not seen a fully persuasive argument that it is.

KCFleming said...

It seems the deterrent effect of killing the murderer saves at least one (or greater) murders from occurring because the murderer is now dead.

Since one cannot assume a "life sentence" is in any way a meaningful concept any longer, deterring at least one murder by killing the murderer is sufficient for me.

The fact that it also serves as a more complete justice than 20 years-to-life is a better reason, however. Anti-death penalty advocates routinely neglect this more important issue.

Freder Frederson said...

Urban tends to imply both liberal politics and higher murder rates. Liberal politics tends to imply less frequent death penalty.

If you want to look at those statistics it really doesn't help you either. Dallas (17.3/100,000), Houston (17) and Chicago (15.6) all have similar murder rates. Boston (12.9), Seattle (4.3), and NYC (6.6) are much lower. (2005 numbers)

The real trend that the crime statistics show is that urban and rural (comparing like to like), the upper and left and right hand corners of the country tend to be the least violent. The south and west are the most violent--even though they are the most obsessive about "law and order" and like to execute people the most, and the middle of the country is just that--in the middle.

N

Ken Stalter said...

If you want to look at those statistics it really doesn't help you either. Dallas (17.3/100,000), Houston (17) and Chicago (15.6) all have similar murder rates. Boston (12.9), Seattle (4.3), and NYC (6.6) are much lower. (2005 numbers)

Comparing a number of urban areas to one another doesn't tell us anything about the overall difference between urban and rural.

Methadras said...

The death penalty should be applied as often as possible in cases of murder. If distinctions of the degrees of murder were removed, no more murder in the 2nd degree, and we just said murder, the nature of it would change. Also, if upon conviction you are given 1 appeal and given a year to do it that would also be deterrent effect. As of right now the death penalty is a toothless old man who waddles his way on a long lazy walk. It takes so long to execute anyone with endless appeals and legalistic manueverings, that it renders the death penalty ineffective as a real deterrent.

If people thinking about committing a murder thought to themselves that a conviction and a year before appeal and execution occurred would make them think twice about doing it and possibly made them stop, then it was worth it. Right now, you commit a murder, get convicted of murder 1, get the death penalty, and you go away for 30 years, while you learn to become a lawyer in prison to keep yourself from getting executed. Where is the justice in that?

Richard Dolan said...

Ann: "Let's try to stick to the topic of deterrence."

OK. In this context, "deterrence" is just the term used to describe the response of the impacted population (here, would-be murders) to the costs of the proposed action. We know (at least everyone who is alive and sentient knows) that people respond to incentives and disincentives, and that behaviour can and is influenced by such costs/benefits. (The level of rationality that this model of human behaviour assumes is quite minimal.) And whether a particular outcome counts as a "cost" for these purposes depends on the ease with which the outcome can be avoided -- i.e, whether engaging in the particular behaviour regularly results in having to pay the "cost." (That is why, for example, that "pay what you want" entrance fees to museums don't work well if the object is to cover operating expenses.) This is all just basic economics and describes the functioning of any pricing system.

There is no reason to expect that would-be murderers as a class are a different breed of humanity, and thus immune from responding to incentives/disincentives. Obviously some are so mentally impaired that they may not respond in the way that normal people would to such costs/benefits. But murders who are so impaired may also be poor candidates for the death penalty (it would be a mitigating factor that could be argued at the penalty phase). Murders occur most frequently between people who already know each other -- family feuds and the like being more the norm than the outlier here. I can't see any reason to assume that such a population would not respond to costs/benefits in the expected pattern. If anyone knows of such a reason, I'd be interested to hear it. In all events, the key point is that those who contend that "deterrence" doesn't work have the burden of proof here, since they are basically claiming that the behaviour of the subject population on an overall basis (again, would-be murderers) does not respond to normal costs/benefit considerations.

Designing a study to test the impact of the death penalty as a deterrent is thus, in its general approach, a standard-issue study in economics. The results of the studies JAC cites, to the effect that the frequency with which the death penalty is imposed increases its deterrent effect, is the result one would expect to see. And JAC is surely on solid ground in contending that a properly conducted study cannot brush aside the jurisdictions where the subject population has the best reason to take into account the relevant "cost." If you want to see how a pricing system impacts on museum attendance, for example, it makes no sense to focus only on museums that use a "pay what you want" system. You'd certainly want to include museums where there was no option to avoid the entrance fee. The studies of capital punishment that ignore Texas are making that kind of mistake.

Nor does it make sense to suppose that the authors of those studies are entirely ignorant of what they're doing. By that gets away from the "deterrence" topic, and more into the limitations of agenda-driven studies.

Trumpit said...

If society decides that I should spend a year in jail because I was caught smoking a joint, then that's harsh, but who's to say it's unjust? Does the punishment fit the crime is always somewhat (or a lot) arbitrary and reflects the values and judgment of society. But to tell me that my sentence is now 5 years for the same crime and the reason is to deter others is highly questionable from a moral standpoint. I, the marijuana-smoking culprit, am now being used by society to further some objective other than my own debt to society. I'm paying a price to stop others from doing the same thing. I think that is morally wrong. Why I'm I responsible from what others may do in the future? It doesn't seem fair to me. Why not put me to death for the same crime? That would probably stop 99% of pot smoking. Then kill the next guy or gal on tv and you'll stop 99.9% of pot smoking. Kill a few more and the maybe you'll get to 99.99%. There's probably a diminishing returns effect to deterrence, but you see my point. If you can repeal the "cruel and unusual" part of the constitution then you can add torture to the punishment. I'm sure the threat of sadistic torture, a la Nazis, would lower the incidence of pot smoking even further. Do we really want to go that route?

KCFleming said...

Do we really want to go that route?
No, but that's not the issue here.
Deterrence works even if only for the murderer put to death, who is unable to kill again, being completely dead.

And that is reason enough.

Does it deter others?
"You may rely on it" is about the best answer you'll get from these "studies", and is the same answer you get from a Magic 8-Ball. Social science studies can never prove or disprove the "causes" of human behaviors or predict their results.

To expect any certainty before agreeing to the death penalty is disingenuous, because the answer will always contain a high degree of uncertainty. I think therefore the burden is on anti-death penalty advocates. The tradition for millenia has been that justice is served by killing murderers. Opponents need to provide compelling reasons why they disagree, rather than demanding "proof" that the death penalty "deters" others.

Hoosier Daddy said...

I never believed it was a deterrent. The average Joe Blow murderer doesn't think about the death penalty because none of them think they're ever going to get caught in the first place. Hell that goes with crime in general. Few think in terms of well this is a misdemeanor which will land me 6 months in the pokey whereas Plan B is a felony and I'm looking at 5-10or life or whatever.

Trying to justify executing on deterrence is pointless. I would much prefer the public be provided the details as to what the murderer did and whether or not $30,000 per year in taxpayer money should be spent feeding, clothing, providing 24/7 medical attention, a law library and cable TV keeping the human waste alive.

Anonymous said...

Richard Dolan said: There is no reason to expect that would-be murderers as a class are a different breed of humanity, and thus immune from responding to incentives/disincentives.

That's an unproven assumption on which your arguments rest. No more need be said.

trumpit said: ...the threat of sadistic torture, a la Nazis, would lower the incidence of pot smoking even further. Do we really want to go that route?

Sure, why not? More discipline, less bondage I say.

An Edjamikated Redneck said...

One other point, as to Texas and Mass. and their murder rates:

Texas is #32 in per capita income at 19k a year; MA is #2 at 42k a year.

Is it possible that being urban isn't the issue, but being poor and urban?

Does the death penalty deter anyone other than the convicted? My guess is no, for several reasons. The murderer usually does not have any thought of the consequences of his immediate act, either the fact that they will be caught or will be punished.

There is also a disconnect between the crime and the punishment in the public's eye that tends to lessen Capital Punishment as a deterent.

There is also, IMO, a general lack of worth of human life, either others or even their own, in certain segments of the country that makes any punishment ineffective, even the ultimate one.

That does not stop me form supporting the death penalty; it always stops at least one convicted murderer.

John Althouse Cohen said...

A couple things are worth noting for some of those who've criticized my posts:

(1) I've repeatedly said on my blog that I am not qualified to really judge the statistical methods, and that I'm just working with the small amount of stuff I do understand from the studies.

(2) Joanna Shepherd claims that she and other researchers have made a thorough attempt to control for variables. In contrast, she seems to say that that law review articles, which are typically not peer-reviewed, have tended to find no deterrence but haven't controlled for the relevant variables. I plan to post on this point in the future.

(3) It may indeed be possible to see what Texas would have been like without the death penalty, since Texas actually didn't have the death penalty during our four-year nationwide ban. So be careful about sweeping assertions that the data cannot possibly exist to show such-and-such.

(4) I was arguing that I -- an educated, middle-class person -- am irrational some of the time. Of course it's speculation to draw conclusions about other people who may be less privileged than I am. But it's not like I'm saying: "I'm rational -- they should be too." I'm saying: "I'm irrational -- they're probably at least as irrational." If they're even more irrational, that may open the door even further for the possibility that they'd overestimate the risk of getting the death penalty.

Thanks to everyone for the comments -- I haven't read through all of them yet, but I will, and I might eventually post an update to take some of your points into account.

Freder Frederson said...

Murders occur most frequently between people who already know each other -- family feuds and the like being more the norm than the outlier here. I can't see any reason to assume that such a population would not respond to costs/benefits in the expected pattern.

Actually you've answered your own question here. People are not thinking about consequences when they get into arguments that escalate into violence and then murder. Deterrence assumes that perpetrators are thinking about the consequences (whether or not their perception of those consequences is rational or not). The vast majority of murders are not plotted, reasoned acts, but unplanned, spontaneous acts where the perpetrator is not weighing the costs and benefits and is often impaired by drugs or alcohol.

Applying an economic analysis to an irrational act is just nonsensical.

Methadras said...

John Althouse Cohen said...

(4) I was arguing that I -- an educated, middle-class person


Why don't you refer to yourself as an educated, middle-class man instead of just a neutral personage via labeling yourself a person? There is no shame in saying you are a middle-class man or woman if that is your gender of course. Hate to get off topic, but when I see this type of self-reference, it is irritatingly politically correct and that bothers me.

Freder Frederson said...

3) It may indeed be possible to see what Texas would have been like without the death penalty, since Texas actually didn't have the death penalty during our four-year nationwide ban. So be careful about sweeping assertions that the data cannot possibly exist to show such-and-such.

Well have a look for yourself. I really wish people would bother to do the least bit of research before they started blogging on stuff, especially since you seem convinced that Texas murder rates must show some deterrence effect. As you can see, the murder rate climbed pretty steadily in Texas from 1960 through the mid-1980's (there was actually a slight dip in the mid-'70s during the death penalty hiatus), when it began to drop (I'm sure John Lott would claim this was because of concealed carry, not the death penalty). By the mid-90's it had actually dropped below the 1960 rate, when I daresay Texas was probably a lot more likely to execute people than they are even today.

As for the lack of peer reviewed articles. I think the authors, in their eagerness to show that they published a peer reviewed article, ignored the stack of criminology and psychology peer reviewed data that shows that murderers are generally not thinking about much else (if they are not too high or drunk to think at all) other than how pissed off they are at the person they are killing at the time of their crime. The consequences don't enter into it.

Revenant said...

That's an interesting point about the threshold of executions. I suppose it could be a question of whether or not you've ever heard of people being executed for the crime you are contemplating.

Revenant said...

Murders occur most frequently between people who already know each other -- family feuds and the like being more the norm than the outlier here.

Actually, "people who already know each other" usually means "fellow criminals". Most murder victims in America have criminal records themselves, if I recall correctly.

Freder Frederson said...

I was arguing that I -- an educated, middle-class person -- am irrational some of the time.

You were talkinb about irrational beliefs about the consequences of actions, which is different than committing an irrational act without considering the consequences.

summer anne burton said...

Uh,

Why don't you refer to yourself as an educated, middle-class man instead of just a neutral personage via labeling yourself a person? There is no shame in saying you are a middle-class man or woman if that is your gender of course. Hate to get off topic, but when I see this type of self-reference, it is irritatingly politically correct and that bothers me.

This is just silly. It's not political correctness for John to refer to himself as a person. Do you really think that his gender influences his read on these statistics or his conclusions? We all know that Ann has two sons, that John is a masculine name, and if you actually visited his blog you would see that the most recent post is partially on the subject of gender. You're suggesting that he add unnecessary biographical detail to every opinion he states? What about his age, neighborhood, race, sexual orientation, SAT score, or favorite ice cream flavor? Why didn't he include those details, too? Come on!

Fen said...

Criminals are risk adverse. If you want evidence that the death penalty deters murder, simply look at all the statements by criminals that they skipped over a house because the beleived the homeowner was armed and they didn't want to get shot.

/disclaimer: I'm against the death penalty because I do not trust the State.

Larry J said...

The name is capital punishment, not capital deterrence. Some people seem uncomfortable with the idea of punishment and especially with the idea that the punishment should fit the crime. I'm not.

I do believe that where the possibility of DNA evidence may exist, it should be mandated. I also believe that taking 15 years or more after the commission of a crime to carry out the punishment almost certainly isn't a deterrent (if that matters) but is cruel as well.

There are cases where there is reasonable doubt. There are other cases where there is absolutely no doubt. Back in 1988, I was in Sunnyvale, California on a business trip. That day, a man named Richard Farley walked into a building across the street from me and opened fire, killing 7 people and wounding 4 others. He was convicted in 1991 and sentenced to death. Today, 20 years after his crime and 17 years after his conviction, he is still sitting on death row. There is absolutely no doubt that he did it.

John Althouse Cohen said...

methadras:

(1) What Summer said.

(2) I use my real name on my blog and in the comments here. My first name is John. I think that makes it clear what my gender is.

(3) I was responding to someone who specifically criticized my post for glossing over the difference between murderers and me, since I'm better-educated and more affluent than most of them. (That's actually skewed since it's not counting the people who are actually deterred and thus aren't criminals, but that's what I was responding to.) I mentioned education and class because that person was making an issue of the education and class disparities between murderers and ivory-tower bloggers/writers/academics.

Come to think of it, there's probably a good reason they didn't mention gender, isn't there? They wouldn't have had a very convincing point if they had said I'm a different gender from most murderers!

John Althouse Cohen said...
This comment has been removed by the author.
John Althouse Cohen said...

Well have a look for yourself. I really wish people would bother to do the least bit of research before they started blogging on stuff, especially since you seem convinced that Texas murder rates must show some deterrence effect.

I'm not "convinced" that Texas "must" have a deterrent effect. I blogged that Texas seems to have a deterrent effect because Joanna Shepherd did a controlled study showing that the death penalty is particularly effective at deterring murder in Texas and other high-death-penalty states. I think she has extra credibility since she's personally opposed to the death penalty.

As I've been blogging all along, I continue to be skeptical about all the data and assertions on both sides. I'm going to keep looking for more of it and keep blogging what I see. But I don't see how you can fault me for giving more weight to what Joanna Shepherd says about the data -- based on her research that apparently uses sophisticated new statistical methodologies -- than what anyone in the comments on this blog are speculating about what the data show.

former law student said...

But the state of Texas does not have a particularly low crime or murder rate. For all its tough on crime attitude, love of the death penalty and gun loving populace, Texas had a 2006 murder rate of 5.9 per 100,000, IL's was 6.1, MA 2.9, NY 4.8, CA 6.8, LA 9.9, VA 5.2.

I have no idea with those statistics how they determined that Texas'... would be even higher.


Why not compare Texas to its death-penalty-free neighbor Mexico? Despite strict gun control, the Mexican murder rate is more than twice as high, 13 per 100,000. Although some may object that Mexico is a separate country, Texas is full of Mexican Americans, from those who could trace their roots back before New Spain to those who crossed the Rio the night before last.

J said...

"Texas had a 2006 murder rate of 5.9 per 100,000, IL's was 6.1, MA 2.9, NY 4.8, CA 6.8, LA (which both likes to execute people and has more people in jail for life without parole than any other state) 9.9, VA (which is more rural than the other states) 5.2"

I'm curious if there are stats published that correct for different circumstances of murder. I'm pretty sure a very high percentage of murders in TX and CA, to name two states, involve participation in drug trafficking at that industry's most dangerous chokepoint, something states like MA or WA have negligible exposure to, and a phenomenon that is a result of geography rather than demographics or legal climate. Is it possible to correct for murders in which perp and victim were engaged in criminal activity at the time of the crime?

I also believe this argues in favor of the death penalty being a deterrent, as criminals in this category are pretty likely to return to a life of crime upon release (or even possibly while behind bars). The recidivism rate for convicts who have been executed is very, very low.

Freder Frederson said...

Okay, so I finished reading the two most prominently mentioned papers in the NYT. Both were written by people who supposedly opposed the death penalty. But they must love their preconceived economic theories more than they hate the death penalty, because their methods and suppositions leave a lot to be desired.

What really got me is one of the papers actually tried to tie individual executions and releases from death row to the murder rate. So apparently, potential murderers track every single execution and release from death row and when someone receives clemency, suddenly the criminals feel like it is a good time to commit a murder.

Also one paper found a strong correlation between lesser crimes and the murder rate, the other found none. What's up with that?

I was half joking when I said John Lott probably attributed the drop to concealed carry laws, but apparently his buddy Kleck did complain about it so the Shepherd paper accounted for NRA membership in their analysis.

Unfortunately for you gun nuts, this is what they found:
"Finally, the NRA membership variable has positive and significant estimated
coefficients in all cases, suggesting a higher murder rate in counties with a strong NRA
presence."

Revenant said...

So apparently, potential murderers track every single execution and release from death row and when someone receives clemency, suddenly the criminals feel like it is a good time to commit a murder.

Publicized cases have a notable impact on people's perceptions of how likely they are to be punished. That's one of the reasons, for example, why the federal government focused so hard on getting convictions in civil rights cases during the 60s -- so that the average redneck might realize "hey, if I go out an beat the Jesus out of some black guy, I might actually go to jail for it!".

Conversely, when some gangbanger hears about another gangbanger escaping the death penalty for murder, he worries about the consequences of doing a drive-by that much less.

There are two traits possessed by most criminals: low intelligence and poor impulse control. These are not people who think rationally about the crimes they commit; if they did, they generally wouldn't commit them. But hearing that a guy just got fried for the very thing you're thinking of doing is a pretty good wake-up call no matter HOW dumb you are.

Finally, the NRA membership variable has positive and significant estimated coefficients in all cases, suggesting a higher murder rate in counties with a strong NRA presence

People who live in areas with a lot of crime tend to be more interested in self-defense then people who live in areas with little crime. It is unsurprising that NRA membership would be higher in such areas.

paul a'barge said...

No criminal after being executed has ever committed another crime.

100% effective.

Stack them like cord wood.

Hoosier Daddy said...

Unfortunately for you gun nuts, this is what they found:
"Finally, the NRA membership variable has positive and significant estimated
coefficients in all cases, suggesting a higher murder rate in counties with a strong NRA
presence."


I always love it when studies say 'suggest'.

Freder, when I was 17, home alone watching tv, some guy was breaking into our house through my parent's bedroom window. I caught him halfway through the window and when he looked up and saw me pointing my 12 guage in his face, he promptly departed from whence he came.

So for the record, I am proud to be a gun nut.

Fen said...

/echo

Criminals are risk adverse. If you want evidence that the death penalty deters murder, simply look at all the statements by criminals that they skipped over a house because the believed the homeowner was armed and they didn't want to get shot.

Anonymous said...

"I am not being disingenuous at all."

Then you're being uneducated. When the data are aggregated by state, the differences between urban and rural areas are masked. Therefore, comparing states tells you nothing about urban/rural crime. That's week 1 of freshman stats 101.

Methadras said...

Summer Anne said...

This is just silly. It's not political correctness for John to refer to himself as a person.


Neutralizing his gender to that of an entity known as a person may not be politically correct in the strictest sense of the idea of political correctness, but in my opinion when I see a man or a woman refer to themselves as a person instead of their gender, it bothers me.

Do you really think that his gender influences his read on these statistics or his conclusions?

Yes. Because men and women are inherently different. They see things differently, they perceive things differently, they speak and communicat differently, they act differently, their approaches and outlooks on life tend to be different and exclusive to their genders. He referred to himself as an educated middle-class person. I know he is a male, but what perspective can I glean from his opinion if he has already neutralized his gender in the way he characterizes himself. This detail does matter. At least to me. He can call himself anything he wants. I really don't care, but I was just questioning why he did it, not insisting that he not do it.

We all know that Ann has two sons, that John is a masculine name, and if you actually visited his blog you would see that the most recent post is partially on the subject of gender.

Actually, I only recently understood Ann to have 2 sons and I didn't know that he had a blog up until now that you just mentioned it. Me knowing that his name is masculine isn't what I was questioning John about. It was me questioning why he didn't refer to himself as a man in his characterization of himself. I think you are applying a little to much emotionality and seem to be overly invested in convincing me that I shouldn't care how someone refers to themselves.

You're suggesting that he add unnecessary biographical detail to every opinion he states?

Really? And how did you arrive at this conclusion that I 'suggest' that he add unnecessary biographical detail about himself in every opinion he states? Where did you ever see me suggest such a thing. Where in what I said did you derive this suggestion or conclusion? I simply asked him why he didn't refer to himself as a man in his own characterization. You on the other hand have now bloomed this simple question into something much more and I'm suspecting that this is coming from something within yourself. Please stop looking for things that aren't there. You are making this much much more than what it is.

What about his age, neighborhood, race, sexual orientation, SAT score, or favorite ice cream flavor? Why didn't he include those details, too? Come on!

Because he didn't need to include any of those in the face of my simple question of why he didn't refer to himself as a man. If I wanted to know anything else I would have asked him. It's really that simple. You've made a mountain out of mole hill. I suspect you do this often.

Methadras said...

John Althouse Cohen said...

methadras:

(1) What Summer said.


Is nonsense. I hope you don't ascribe to her nonsense.

(2) I use my real name on my blog and in the comments here. My first name is John. I think that makes it clear what my gender is.

It wasn't your name I was concerned with, since I already understood and assumed you to be a male. I was simply interested as to why you referred to yourself as an educated, middle-class person and not a man. Yes, your name makes it clear what your gender it, but what isn't clear is why you referred to yourself in a gender neutral way.

(3) I was responding to someone who specifically criticized my post for glossing over the difference between murderers and me, since I'm better-educated and more affluent than most of them. (That's actually skewed since it's not counting the people who are actually deterred and thus aren't criminals, but that's what I was responding to.) I mentioned education and class because that person was making an issue of the education and class disparities between murderers and ivory-tower bloggers/writers/academics.

Yeah, I understood that to be the case, but that's not what I cared about.

Come to think of it, there's probably a good reason they didn't mention gender, isn't there? They wouldn't have had a very convincing point if they had said I'm a different gender from most murderers!

Well, only if they accused you of being a female I suppose, but since most murderers are male, then I'm still evermore curious as to why you neutralized your gender at all. I'm almost suspecting that it's a habitual reference to yourself. For example most people in conversation would say something like this, "I consider myself to be a nice person." or "I saw that person over there steal something from that other person." Instead of substituting gender in either sentence, e.g. "I consider myself to be a nice man." Maybe it's the sound of it that sounds strange. I'm not sure. But nevertheless I was struck curious by the characterization. No harm or malice was intended on my end.

John Althouse Cohen said...

methadras: Summer and I have dealt with this satisfactorily. I can't fathom why you think it's a problem that I referred to myself as a "person" instead of a "man," in a context that had nothing to do with gender differences.

Methadras said...

John Althouse Cohen said...

methadras: Summer and I have dealt with this satisfactorily. I can't fathom why you think it's a problem that I referred to myself as a "person" instead of a "man," in a context that had nothing to do with gender differences.


John, no offense, but Summers histrionic outburst really added nothing to my initial query to you. If this is how Summer deals with the simplest of issues or questions, then I'd really hate to see how she deals with a legitimate problem. She simply wanted to see and make more out of it then there really was and basically it was a weird attempt to defend some perceived slight. She overreached, overcompensated, and failed.

I know you can't fathom why it's a problem because you don't think it is one since your reference to yourself as a person is at the core of my question. That is why I asked for a clarification because I saw it as a problem for me in that I was curious about you referencing yourself as a person and not as a man in that sentence. I just wondered why you would refer to yourself that way or if you do it out of a natural habituation. No harm no foul.