Grendel / Grabic has challenged me—nay, dared me!—to find him evidence that Europe’s crime rate has risen since the abolition of the death penalty. He even had the nerve to suggest I couldn’t back up my claims. The gall on that German! Here, ladies and gentlemen, I present to you a brief outline of the evidence that the death penalty reduces crime. Ideally, I would have found perfect statistics for just the right years. Unfortunately, that was not available, but I nonetheless found plenty of miscellaneous statistics that support my point.
First: does the death penalty deter crime? I like this comparison in particular.
Homicide Rate/100,000 by Date in US:
1980 – 10.2
2000 – 5.5
Wow, homicides cut in half! That’s quite an accomplishment. Of course, the correlation could be stronger. The death penalty was banned in the US in the late 1960s and was reinstated in the 1970s. (The crime rate skyrocketed during that period, part of what inspired the US Supreme Court to backtrack on its ruling that the death penalty was unconstitutional.)
A more important issue in the US may be that we have imprisoned a lot of people over the last two decades. While our fourth and fifth amendments to the Constitution put Byzantine procedural restrictions on police conduct, those who are caught, even for nonviolent crimes, are often put away for a very, very long time.
Next, comparisons of Europe and the US. I don’t think mere numerical comparisons mean much. Trends are the more important factor.
Interpol: 1995 crime statistics/100,000:
5278 – US
6316 – France
7206 – England & Wales
And six years later:
Interpol: 2001 crime statistics/100,000:
4161 – US
6941 – France
9927 – England and Wales
Raw comparisons don’t mean much: I’d rather look at 1.) trends and 2.) subpopulations. The US is made up of numerous subpopulations that vary wildly in their propensities to commit crime or be the victim of crime. Comparison between countries that have different ratios of different kinds of subpopulations doesn’t really tell you that much unless you look deeper.
I don’t want to get racial about it, but what Michael Moore tried to cover up in Bowling for Columbine was that the homicide rate for the US is basically the same between the US and Canada if you look at the populations of European and Asian origin. The hard truth is that Blacks and Hispanics, who make up 27% of the United States, account for a majority of the violent crime in the United States. One reason for Canada’s relative peace and tranquility is partly because only 5% of the population is Black or Hispanic.
More information from the DOJ, the UN, and the UK Crime Reduction page, or see blogs here, here, and here. I also like this superb AEI report, which points to the arrogance of the European sophisticates who go on and on about the barbarity of America and it’s rarely enforced death penalty all while doing little to tackle the problem of growing crime rates in their own cities.
That capital punishment deters crime perfect sense: those who murder are killed by the state, and (suprise!) this stops them from committing murder again. Additionally, other (potential) criminals see the consquences and some think twice. About the only argument people make against this rather obvious correlation is a vague insinuation that killing criminals somehow encourages a culture of death. Excuse me, but I have to call it like I see it: typical left-wing nonsense. May I also recommend examining another very safe country with a streamlined death penalty: Japan.
Of course, the US doesn’t kill nearly as many people as it should. About 30-50 people are executed a year, most of whom are killed by the states, not the federal government. I’d like to see thousands of violent people executed every year. Alas, the US is caught in a precarious position where our society cannot live with no death penalty or a meaningful deterrent. I highly, highly recommend this Slate article for more.
On a side note: comments to not have to agree with me to be constructive. Thoughtful comments that contribute to the argument are greatly appreciated. All contributions are welcome, and I particularly like those that provide new evidence to challenge what I think. Many comments on this site have changed how I view the issues.

Comments to this entry
Martey
June 25, 2005
8:20 pm
If you are claiming that Canada has less crime because its population is more homogenous than the United States, I will agree with you. However, your post seems to imply that Blacks and Hispanics are somehow inherently more violent than other ethnicities. I assume that this not what you meant, though.
My main problem with the death penalty is that in recent years, too many people on death row have been found to be innocent of the crimes that it was assumed that they had committed. If homicide is important enough that the state decrees that murderers should be punished with death, it should ensure that everyone killed is truly guilty. If the United States did execute thousands of people a year, it could well mean the end of the death penalty if even a small fraction of them were found to be innocent.
bp32
June 25, 2005
11:10 pm
1) Large scale correlations tell us nothing without specifying the causal mechanisms at work. Don't see any of those here. As far as the US crime rate goes there are two initially that I can think of:
First, the death penalty might indirectly influence the crime rate by offering a significant deterrent to potential offenders. Certainly one could argue that the decrease in homicide rates might be attributable to the penalty. However, given the restricted number of crimes that the death penalty is applied to I doubt that this could account for the drop (you mention increased incarceration rates which is a better candidate at the moment) in the overall crime rate, unless we posit another connection between violent and non-violent crimes.
Second, and this is a long shot, the death penalty has reduced the crime rate by taking the most probable and frequent offenders out of the population, therefore reducing the occurrence of crime. Again, this hardly seems plausible given the small number of executions per year (note: however one could make a different argument based on the number of prisoners sitting on death row who are not on the streets, as this number is certainly larger than those executed).
Before we say that the existence of the death penalty reduces crime we need to specify what the causal mechanisms are that get us from DP to less crime and then test those mechanisms to see if they are actually having the hypothesized effect.
2) State by state data should be used to determine whether there is a correlation between the death penalty and crime rates and whether there is actually anything causal going on in the relationship. Since different states have readopted the death penalty (and with different restrictions) and not all capital crimes would be covered by the feds we cannot say for certain that crime varies with the death penalty. A longitudinal, state-by-state comparison would probably give us a better feel for what relationship, if any, exists.
Finally, is the argument about crime rates overall or violent crimes, because the statistics you cited for Europe/US seem to include all crime. I am not sure how one connects the supposed deterrent value of the death penalty to non-violent crimes (i.e. crimes which do not carry the possibility of the death penalty).
3) That capital punishment deters crime perfect sense: those who murder are killed by the state, and (suprise!) this stops them from committing murder again. Additionally, other (potential) criminals see the consquences and some think twice. About the only argument people make against this rather obvious correlation is a vague insinuation that killing criminals somehow encourages a culture of death. Excuse me, but I have to call it like I see it: typical left-wing nonsense. May I also recommend examining another very safe country with a streamlined death penalty: Japan.
Actually, the first statement is logically impossible"”?to deter means to prevent someone from taking certain actions by threatening to punish them in the future should they alter the status quo"”?if you have already killed them then, by definition, you are not threatening to punish them in the future. As for the deterrent value of the death penalty, I have yet to see strong empirical proof that criminals are in fact deterred by the death penalty (meaning, the specter of the penalty actually weighs in their decision whether to commit a capital crime)"”?I would love to see it if such a report exists (I am honestly curious). As for Japan, there are any number of reasons why their crime rate is lower"”?better policing, streamlined legal proceedings that do not involve the death penalty, etc. Again, we need to tease out causal mechanisms and test them against other factors before we can say that the correlation is causal.
Daniel Nexon
June 26, 2005
12:02 am
You compare 1980 to 2004, but you don't look at trend lines within that period. For example, if you look at 1989-1993, the homicide rate increases by approximately 8%, while the number of executions increased from 16 to 23. The next year, the execution rate drops to 14, while the homicide rate increases 4%. Even the "eyeball" of the data for the period from 1980-2004 (let alone something that might be useful here, like multivariate regression), suggests very little relationship between the death penalty - either in magnitude shifts in its implementation or in existential deterrence terms - and homicide rates. Most of the drop in the homicide rate corresponds to the low unemployment, high growth, new policing techniques of the Clinton years (I'm not suggesting causality here, just observing that the falloff in homicide rates is a very recent phenomena).
I'll discuss how your US-Canada and US-Europe comparison involves completely ad hoc, and contradictory, reasoning if I get around to posting a full response.
Daniel Nexon
June 26, 2005
12:04 am
Dan
June 26, 2005
12:29 am
As to your detractors...
My main problem with the death penalty is that in recent years, too many people on death row have been found to be innocent of the crimes that it was assumed that they had committed
"Too-many" is what percentage?
If homicide is important enough that the state decrees that murderers should be punished with death, it should ensure that everyone killed is truly guilty.
Ah, a zero defects mentality.
http://www.google.com/search?q=zero+defects+mentality
If the United States did execute thousands of people a year, it could well mean the end of the death penalty if even a small fraction of them were found to be innocent.
Thousands of Americans are raped -- not a lawful punishment for any crime -- in prison every year. That has hardly meant the end of incarceration!
Now, to bp32...
First, the death penalty might indirectly influence the crime rate by offering a significant deterrent to potential offenders.
It only needs to be a marginal deterrent to potential offenders. Pretty basic economics, really.
However, given the restricted number of crimes that the death penalty is applied to I doubt that this could account for the drop (you mention increased incarceration rates which is a better candidate at the moment) in the overall crime rate, unless we posit another connection between violent and non-violent crimes.
Of course there are many connections.
Actually, the first statement is logically impossible"”?to deter means to prevent someone from taking certain actions by threatening to punish them in the future should they alter the status quo"”?if you have already killed them then, by definition, you are not threatening to punish them in the future.
Those who have committed murder are much more likely to commit murder in the future than those who have not. Therefore killing them reduces the number of murders.
As for the deterrent value of the death penalty, I have yet to see strong empirical proof that criminals are in fact deterred by the death penalty (meaning, the specter of the penalty actually weighs in their decision whether to commit a capital crime)"”?I would love to see it if such a report exists (I am honestly curious).
Please see the recent interview by the author of "Freakonomics" on NPR. I would assume that if he discussed his findings in the radio interview, he also mentioned them in the book.
As for Japan, there are any number of reasons why their crime rate is lower"”?better policing, streamlined legal proceedings that do not involve the death penalty, etc.
Irrelevent. Of course there are many factors. "Saying X implies Z" doesn't mean that "Y doesn't imply Z."
Daniel Nexon
June 26, 2005
12:40 am
Daniel Nexon
June 26, 2005
1:03 am
"Irrelevent. Of course there are many factors. "Saying X implies Z"Â? doesn't mean that "Y doesn't imply Z."Â?"
You know what a spurious correlation is, right? Bill's asserting that we're looking at one right now, called the "death penalty" and "lower homicide/crime rate." Thus, the presence of other potential explanatory factors is extremely important, since it might disprove the grounds for believing that there is any causal relationship (or correlative relationship) of the type Curzon posits.
Dan
June 26, 2005
1:20 am
Thank you for the correction. Levitt was surprised to find that "three-strikes" laws had a significent impact. I am in your debt.
A better criticism of the death penalty numbers might be that the occurance of the death penalty is now so low it could not have an effect in the current regime. One way around this is to have the federal government encourage some states to use the death penalty much more, and see how the experiment turns out.
Your final point about correlations misdirects. It doesn't address my criticism. You start with a vague statement about "other factors," and then proceed with an argument that relies on precise coeffecients for a certain number of factors.
Of coures, the best argument for the death penalty is that it protects the criminal's basic human rights more than long prison terms, but that is a discussion for another time...
- tdaxp
bp32
June 26, 2005
1:42 am
It only needs to be a marginal deterrent to potential offenders. Pretty basic economics, really.
Ok, marginal, significant--what does this have to do with the overall point I was making about testing causal mechanisms?
Of course there are many connections.
OK, what are they? What do most studies say about this connection and is the nature of those connections such that the potential deterrent effect of the death penalty will trickle down to non-capital crimes? I was merely responding to a hole in Curzon's analysis and suggesting that before we close the book on the subject this is a question we ought to ask and evaluate"”?you seem to think I was asserting that there are no connections"¦
Those who have committed murder are much more likely to commit murder in the future than those who have not. Therefore killing them reduces the number of murders.
You seem to have missed the entire point I was making--yes, by killing those who commit murders we prevent those specific people from committing murder in the future--but that is not deterrence, not by any acceptable definition of the term, which is what I was specifying above. You can't deter someone by physically preventing them (in this case by killing them) from doing something--to assert as much is to partake in conceptual stretching to a degree which makes the term meaningless (unfortunately, many IR scholars seem to have done as much which makes discussions of deterrence difficult in that field as well). To deter is to prevent action indirectly, by threatening to punish them in the future if they commit a certain act. If you kill them that is prevention by physical force, not by threats of punishment.
Irrelevent. Of course there are many factors. "Saying X implies Z"Â? doesn't mean that "Y doesn't imply Z."Â?
As for this one, see Dan Nexon's response"¦
Dan
June 26, 2005
3:14 am
Exactly as much as your original claim that it was a "significant deterrent." That seemed to be an argument from reason, while now you are resting your argument on empiricism.
You seem to think I was asserting that there are no connections"¦
I'm not sure how I was, considering I agreed with you that there were many conections. Reality is a complex system.
You seem to have missed the entire point I was making
I apologize. I assumed you weren't making an argument more substantial than semantic. Again, apologies.
bp32
June 26, 2005
3:37 am
Exactly as much as your original claim that it was a "significant deterrent."Â? That seemed to be an argument from reason, while now you are resting your argument on empiricism.
It wasn't a claim but merely a hypothesis about the causal mechanisms that might help Curzon make his case...
I'm not sure how I was, considering I agreed with you that there were many conections. Reality is a complex system.
Actually I wasn't saying that there were many connections but that in order to make the case about the death penalty and non-death penalty crime rates there would need to be these connections. Realty is a complex system but complexity theory doesn't necessitate that everything is connected or that they are connected in linear or meaningful ways--we need to investigate what these connections (if any) are and how they function--again, I would like to see some studies on this for my own curiousity...
Mutantfrog
June 26, 2005
4:47 am
The claim that executing murderers lowers the murder rate by keeping them off the street. This is total nonsense because in cases where the death penalty is an option the alternative is life imprisonment, often without the possibility of parole -either way they are probably never getting out again.
Look at New York- They only reinstated the death penalty in 1995, but the murder rate had already been declining before then. NYC has the lowest murder rate of any American large city, but have used the death penalty quite rarely. The high murder rate of the late 80s early 90s, and its subsequent drop, seems for more likely to be linked to the rise and fall of the crack epidemic than to capital punishment.
New York State has had no death penalty since early 2004 when the statute was ruled unconstitutional and the legislature has declined to issue a new statute, and yet the murder rate continues to drop as steadily as before.
The death penalty is completely irrelevant to New York's success, it is due to the improvement of poor neighborhoods and more efficient policing.
Some links:
http://www.safestreetsdc.com/subpages/murdercap.html
http://66.102.7.104/search?q=cache:NG8-6r_RS48J:www.philly.com/mld/philly/news/11190868.htm+new+york+city+murder+rate+2005&hl=zh-TW&client=firefox-a
http://www.nynewsday.com/news/local/crime/nyc-murd0425,0,5501545.story?coll=nyc-homepage-breaking2
Mutantfrog
June 26, 2005
4:50 am
Joe
June 26, 2005
6:49 am
Personally, I agree with Bill O'Reilly's notion (for once) that we should have imprisonment at hard labor in Alaska instead of a death penalty. It's probably just as scary to most people (and I would imagine a particularly good deterrent for those types who don't care if they live or die) but it removes the problems of irreversibility and God-playing that cause many people to oppose the death penalty. But then again, I don't really believe in speaking Victorian or thinking pagan: I'm more in the school of speaking ghetto and thinking economic.
Michael
June 26, 2005
8:37 am
It might not hurt to clarify that the crime rate among blacks and hispanics is correlated more or less to their typically low economic status.
GraBlog
June 26, 2005
10:19 am
On some mornings you know what you're going to do over the day, on some morning you wake up and see what the day keeps ready for you. Then there are morning like this one when you think you know what's going to happen, but then you read your name on ...
Chief Wiggum
June 26, 2005
2:32 pm
It is common to read of people in prison being exonerated by DNA evidence. Former witnesses sometimes recant their testimonies or new exculpatory evidence is found. I think it is better to err on the side of life.
Most death row inmates are indigent. In California, it costs the state from 1-3 million dollars per inmate to move the legal appeals process forward. It just doesn't make economic sense to spend so much money to kill someone who is locked up and no longer a threat to the public.
People convicted of murders in some instances are more likely to be represented by overworked public defenders or court-appointed lawyers of questionable diligence than those who are acquitted. Do you think O.J. Simpson would have walked if he had been represented by a public defender?
In some instances, the only available "evidence" used to convict is provided by other prisoners, criminals or others trying to create leverage with the police in their own legal situations. This "evidence" is suspect, and there is no reliable way to confirm or deny it.
Finally, isn't it more of a punishment to locked someone up for a lifetime in the company of fellow sociopaths?
Chirol
June 26, 2005
2:57 pm
First of all, I'd like to add something with regard to deterrence. Considering the amount of privately owned guns (legal and illegal) and the willingness of the police in the US to use deadly force (compared to countries in Europe who allow people to kill others and still DONT shoot them) a criminal has a MUCH higher chance of being killed while comitting the crime than years afterwards. Thus, I think the deterrent argument is fairly worthless. It has already been proven that people have a very difficult time weighing the consequences of actions if they are far off in the future (eg: smoking) and thus possibly being executed in 10 or 15 years isn't going to convince anyone to not do something.
While mistakes are possible, I'm very much for the death penalty provided it is used on in very clear cases (which I would personally extend not only to 1st degree murder, but to rape and child abuse as well). Lawyers would of course have to set down very clear rules for when the death penalty is applicable. DNA evidence, multiple witnesses, admission of guilt, video etc. In all other cases, even when it seems to be 100%, I'd have to opt for the hard labor for life in Alaska. If further evidence comes to light later on that backs up their guilt, then I'd move to execute them, and I think an excellent point was made that life long hard labor is probably more of a deterrent than death which is too easy for those who truly don't care.
This has been an excellent debate thus far. I only wish I had internet at home so I could follow it!
Grendel
June 26, 2005
11:08 pm
I find it very interesting no one has come out with moral arguments for or against the death penalty and instead all practical ones.
I think we owe this to
These capital punishment arguments have been around forever.
...the moral ones are of relative value, never mind the death penalty's effectivity, one can still believe it to be right or wrong, justified or not, etc.
Indeed, on that level we could go on forever. Fortunately, it was just about refuting the claim that capital punishment decreases crime.
regards,
Grendel (T. Grabic)
Dan
June 27, 2005
1:59 am
which I would personally extend not only to 1st degree murder, but to rape and child abuse as well
So we should manage crime by encouraging rational rapists and child abusers should kill their victims?
snow
June 27, 2005
8:09 am
It's just too bad it can't be all simple and clear-cut like those old Hollywood movies, where the criminals knew very clearly that they if they got caught, they'd get the chair. I'm thinking of the Lana Turner movie "The Postman Always Rings Twice" and other film-noir type movies (great stuff, I might add).
Curzon
June 27, 2005
3:51 pm
In the US, 40,000 (mostly innocent) people died in car accidents in 2003. We basically accept that as a cost of transportation. I think we should also accept that some innocent people -- perhaps as many as a few hundred every year -- would be killed in a more robust system of capital punishment. That's my cost benefit analysis that I think is wholly moral.
Jing
June 27, 2005
7:23 pm
Curzon
June 27, 2005
9:15 pm
Of course my tune would change if I was falsely accused, just like many who are against the death penalty forget their moral stance when a love one is assaulted, raped, and murdered.
snow
June 28, 2005
7:15 am
ComingAnarchy.com » Blog Archive » This is what justice looks like
June 29, 2005
4:29 am
bp32
July 1, 2005
2:15 am
Grendel
July 3, 2005
11:24 am
Youndhusband "mentions":http://grabic.blogspot.com/2005/06/death-penalty-in-japan.html#111785176302880104 that a system going into that direction is currently used in Iran. I wasn't able to find detailed information about it though. There's an important difference between revenge (as in the case of homicide being penalized by capital punishment) and ... sorry, I wasn't able to translate "Genugtuung" (German) into English properly. I found "gratification", "amends" and "satisfaction". Abolitionist countries don't allow revenge as an aspect of jurisdiction, which indeed has its roots in history (in ancient times, in Egypt i.e. it was the death penalty or nothing. If the crime was serious enough, you got the death penalty, if it was not, you were free to go). Over the centuries punishment diversified quite a lot. In this respect, Ghandi said something very smart... "an eye for an eye and we'll all be blind". Metaphorically speaking, supporters of the death penalty are blind at least on one eye and shortsighted on the other.
Families who lost someone of their own are absolutely in no condition psychologically *and* physically or otherwise to decide something as you suggested, bp32 - beside that normally a human is never supposed to decide about life and death of another person due to extreme stress (doctors and police men i.e. struggle enough with such decisions). Take the case of "Marc Hoffmann":http://www.spiegel.de/panorama/0,1518,362910,00.html for example. On the day of the rendition, one of the mothers who lost their child was inable to even come close to the court room and stayed in a hotel nearby where she waited for closure.
GraBlog » Europe’s Crime Rate and Capital Punishment Part I
September 2, 2005
8:36 pm
7oob
December 21, 2006
1:09 pm
I Googled for something completely different, but found your page...and have to say thanks. nice read.