entry archive

December 25, 2005

Further Comments on Capital Punishment-BECKER

Our discussion last week on capital punishment generated a lot of comments that are worth discussing in more detail. Since capital punishment is so controversial, we decided to continue the same subject this week.

First, let me correct a misunderstanding in some of the comments. I never claimed the evidence is anywhere near conclusive that capital punishment has an important deterrent effect. I stated that the evidence from quantitative studies is decidedly mixed, yet I concluded that "the preponderance of evidence does indicate that capital punishment deters". Although the weight of the positive evidence should not be overstated, the frequently stated claim that these studies prove that capital punishment does not deter is clearly false.

My belief in its deterrent effect is partly based on these limited quantitative studies, but also because I believe that most people have a powerful fear of death. David Hume said in discussing suicide that "no man ever threw away life, while it was worth living. For such is our natural horror of death…". Schopenhauer added also in discussing suicide "…as soon as the terrors of life reach a point at which they outweigh the terrors of death, a man will put a an end to his life. But the terrors of death offer considerable resistance…".

Nevertheless, the main point of my comment last week was not to try to prove that capital punishment deters murders, but rather to argue against the view that it is "immoral" for the State to take lives through capital punishment even if we assume that the deterrent effect on murders is sizeable. Indeed, I believe that deterrence can be the only reasonable basis for capital punishment. Revenge, retribution, and other arguments sometimes made to justify capital punishment are too subject to government abuse, and have been abused.

Some readers interpreted my views as implying that a major goal of government policy should in general be to save lives. That is not my belief. I am against governments interfering, for example, with the rights of people to overeat even when that causes obesity, disease, and possibly early death because overeaters are primarily "harming" themselves. In my view, people should have the right to do that.

Murder, on the other hand, involves taking the lives of others, and any reasonable discussion has to distinguish such behavior from individuals taking actions that affect only their own lives. In economists' language, murder involves the most severe negative externalities. If we assume for the sake of this discussion that there are two fewer murders for each murderer executed, the State would reduce two of these severe externalities for each murderer that it executes. This issue of the effect of capital punishment on innocent victims has to be confronted by even those most opposed to its use. And I frankly do not see how any reasonable and relevant philosophy could oppose the use of capital punishment under the assumptions of this example.

Admittedly, the argument gets less clear-cut as the number of lives saved per execution falls from two to lower values, say, for example, to one life saved per execution. In this case, I compared the qualities of the life saved and the life taken, to the dismay of some readers. In particular, I wrote that "wouldn’t the trade-off still be desirable if the life saved is much better than the life taken, which would usually be the case?" I do not see how to avoid making such a comparison. Consider a person with a long criminal record who holds up and kills a victim who led a decent life and left several children and a spouse behind. Suppose it would be possible to save the life of an innocent victim by executing such a criminal. To me it is obvious that saving the lives of such a victim has to count for more than taking the life of such a criminal. To be sure, not all cases are so clear-cut, but I am just trying to establish the principle that a comparison of the qualities of individual lives has to be part of any reasonable social policy.

This argument helps explain why capital punishment should only be used for some murders, and not for theft, robbery, and other lesser crimes. For then the trade off is between taking lives and reducing property theft, and the case in favor of milder punishments is strong. However, severe assaults, including some gruesome rapes, may approach in severity some murders, and might conceivably at times call for capital punishment, although I do not support its use in these cases.

A powerful argument for reserving capital punishment for murders is related to what is called marginal deterrence in the crime and punishment literature. If say perpetrators of assaults were punished with execution, an assaulter would have an incentive to kill the victims in order to reduce the likelihood that he would be discovered. That is a major reason more generally why the severity of punishments should be matched to the severity of crimes. One complication is that capital punishment may make a murderer fight harder to avoid being captured, which could lead to more deaths. That argument has to be weighed in judging the case for capital punishment. While marginal deterrence is important, I believe the resistance of murderers to being captured, possibly at the expense of their own lives, is really indirect evidence that criminals do fear capital punishment.

Some readers asked whether I also favor public executions of convicted murderers, mangling of their bodies, and other methods used in some countries still, and in most countries in the past? I do not because they seem unnecessarily abusive of convicted murderers without any compensating gains. However, I admit I would reconsider this position if it were demonstrated that such added punishments have a large effect in reducing the number of murders. For those who find such a position "barbaric", I would ask how many innocent victims are they willing to tolerate before they might take a more positive position on these additional punishments?

Of course I am worried about the risk of executing innocent persons for murders committed by others. In any policy toward crime, including capital punishment, one has to compare errors of wrongful conviction with errors of failing to convict guilty persons. My support for capital punishment would weaken greatly if the rate of killing innocent persons were as large as that claimed by many. However, I believe along with Posner that the appeal process offers enormous protection not against wrongful conviction but against wrongful execution. And this process has been strengthened enormously with the development of DNA identification. However, lengthy appeals delay the execution of guilty murderers, and that can only lower the deterrent effect of capital punishment.

So to summarize once again my position on this controversial question, I favor capital punishment because and only because I believe it has "sizeable" deterrent effects. I would join the anti-capital punishment side if this view turns out to be wrong, if it were proven that many innocent persons are wrongly executed, or if it is administered in such a racially biased manner as to wrongly convict many black persons, and to be little used against white murderers. But I do not believe that the available evidence strongly supports any of these arguments against the use of capital punishment.

Posted by Gary Becker at 11:51 PM | Comments (32) | TrackBack (0)

Trackback Pings

TrackBack URL for this entry:


Becker: "I believe it has 'sizeable' deterrent effects."

A panel of the National Academy of Sciences reviewed the available research on the general deterrent effects of criminal sanctions. The panel reported that the empirical studies suffer from methodological weaknesses so severe that no conclusions could be drawn. Among the studies reviewed was Ehrlich's, "The Deterrent Effect of Capital Punishment: A Question of Life and Death," American Economic Review, June 1975.

Posted by Harris at December 26, 2005 02:09 AM | direct link

I wonder what level of evidence it would take to convince you that "[the death penalty] is administered in such a racially biased manner as to wrongly convict many black persons, and to be little used against white murderers." I cannot cite you studies, but perhaps you could cite some studies you know of that dispute the often-cited claim that executions are racially-biased -- some modern ones.

In dealing with such powerful negative externalities (one of the strongest, really, though we can imagine stronger externalities yet), what level of wrongful execution are we willing to tolerate? Wrongful executions are perhaps one of the most fundamental violations of a human's right to life -- though admittedly most economists and consequentialists do not believe in "rights." Still, it seems odd to advocate strong property rights because of its consequence in raising welfare levels over time, but advocate weak life rights because of its possible consequence in reducing the deterrence effect of the criminal justice system.

Posted by Mitch at December 26, 2005 03:04 AM | direct link

"For those who find such a position "barbaric", I would ask how many innocent victims are they willing to tolerate before they might take a more positive position on these additional punishments?"

as a more-or-less determinist, I view extremely violent crime as indicative of a "diseased" mind (in a social, but not necessarily medical, sense ). from that perspective, the primary objective of "punishment" for such crime is to purge society (including that portion in prisons) of those who pose extreme danger to others and to do so in a humane way, including possibly capital punishment. to intentionally inflict suffering suggests holding criminals not only accountable for their antisocial acts but also responsible for being willfully "evil".
by itself, of course, this doesn't answer prof becker's cost-benefit trade question. to make it less abstract, suppose science and technology one day provide the ability to keep a person alive and conscious indefinitely and in continuous agony (sound familiar?). would you do that to one guilty of even an extremely violent crime if it saved scores of innocent lives? only one innocent life? just for vengeance? for lesser offenses if it had a large deterrant effect?

my concerns are that viewing criminal sentences as "punishment" rather than mere accountability is based on philosophically (and biologically?) questionable assumptions and diverts attention from the fundamental objective of protecting society, and that adding torment is way too close to religious inquisition, even if it can be rationalized due to its deterrence value.

Posted by CTW at December 26, 2005 12:02 PM | direct link

Add to the hypothetical equations a very real number. For each innocent person convicted for a murder, a real murderer or murderers are free to kill or harm others again. Once the penalty has been carried out the search for the perpetrators will likely never begin again. This is rarely addressed by police, prosecutors or judges that are proponents of the death penalty.

However, an innocent man sentenced to life in prison may shout loud enough and long enough that someone will take notice and the investigation in to what really happened may take place. If this occurs, there is a chance that the real criminal may then be brought to justice at some point in the future. This chance seems to be virtually non-existent in the current system.

Posted by Justin at December 26, 2005 07:19 PM | direct link

Granted what we are arguing here is much like debating politics and the existence of God, but I would still like to throw a few ideas out there that have been somewhat sidestepped. How does one begin to measure a murder that has been deterred? If we have an instance of no-murder, this does not mean a murder has been deterred. I seriously doubt the validity of the deterrent effect of the death penalty given that it is impossible to measure its effects in a sound, valid, and reasonable manner. Do murderers travel over state lines from pro-death penalty states that impose such a punishment to states where there are no such penalties to avoid being sentenced to death? Would that be considered the deterrent effect?

Allow me to play devil?s advocate and ask you run a thought experiment where the federal government forced each state to carry the death penalty and fill a quota (by quota I mean if there is a murder we issue the death penalty to the convicted murderer) of death sentences each year (or whatever time period you like) based on the amount of severe negative externalities we wish to reduce and based on the number of murders committed. The quota would have to be based on the evidence to date of the deterrent effect of the death penalty. The quota would also have to be based on the murder rates idiosyncratic to each state and would of course differ across states based on such information. Then, say we had no murders in a given state in a given year; we would obviously have no death penalty sentence handed out and no quota to be filled. Would you be comfortable saying that the death penalty has in effect done its job and deterred a given number of unknowable murders? What if we run the same experiment in the next time period and in the state where there were no murders, we have 3 (or any number) and in a state where we had X murders we now have none. Where is the deterrent effect in that? I could go on forever with such an experiment and ask you over and over again how you feel about the way I am measuring the deterrent effect of the death penalty and whether or not it has worked, what kind of conclusions do you think we would come to?

Just because have lower murder rates does not in anyway mean that this is because of the death penalty. Just because a town is having a lot of babies and that town also has a high population of storks, does not mean the storks are flying the babies to town in little white handkerchiefs and dropping them off in bedrooms across the town. We have to be careful deeming spurious relationships as causal relationships. The death penalty is in fact a useful tool when it is used as Hammurabi intended but to use it as a deterrent is somewhat questionable as we can neither measure uncommitted murders, murders deterred by such a penalty, etc.

Posted by ChrisJ at December 27, 2005 03:17 PM | direct link

I have a hard time believing 100% in deterrence as the sole reason for capital punishment. We have had capital punishment for a long time, and yet people keep murdering.

It becomes necessary to get into more details on academic studies.

Some people appear to have a death wish and appear to not care if they are executed after murdering other people. This situation seems to be happening a lot recently. It would be interesting to know how common this is and how it has trended over time. Perhaps my perspective is biased by frequent consumption of mass media. It may be possible that deterrence attributable to capital punishment has changed significantly over long periods of time due to changes in other factors (eg, media and information proliferation).

At least two movies may be relevant:

"Minority Report": Maybe science and technology will give us new methods to manage murders in the future (and new government abuses of power).

"Natural Born Killers": Some people may kill or murder for media attention and posterity. This motive may be getting worse in modern time - not sure. Murderers know that murderous events will be reported over cable news and the internet instantly.

P.S. What about "crimes of passion" where people murder spontaneously? Do such people grasp the consequences at the time? And how common are unpremeditated murders and is this even relevant to this discussion?

Posted by nate at December 27, 2005 05:07 PM | direct link

The singlest largest problem with the concept of deterrence I see, from studies of criminal behavior, is the questionable assumption that engagers of crime take the risk of punishment into consideration.
Deterrence would only work if there was a reasonable surety of consequences, and a rational understanding of those consequences.
However, as forensic psychologists have long known, few criminals believe that they will get caught, or care if they do (and that includes murderers). Not to mention the fact that a large percentage of the criminal population has rather low intelligences, the mean is considerable below that of the average -- their ability to make the kind of decisions that the judicial idea of deterrence implies is inherently handicapped.
The only viable use of punishment, whether execution or imprisonment, is not deterrence or revenge, but simply to withdraw those individuals who have commited crimes from the general population; preventing them from further actions.

Posted by HamletsMill at December 27, 2005 07:30 PM | direct link

Professor Becker said: "Some readers asked whether I also favor public executions of convicted murderers, mangling of their bodies, and other methods used in some countries still, and in most countries in the past? I do not because they seem unnecessarily abusive of convicted murderers without any compensating gains. However, I admit I would reconsider this position if it were demonstrated that such added punishments have a large effect in reducing the number of murders."
If it were demonstrated that such added punishments have a large effect in reducing the number of murders, it might be better to outsource such executions to certain Middle East countries. The information asymmetry introduced by such an outsourcing or rendition policy in the minds of criminals will act as an additional deterrent.

Posted by Arun Khanna at December 27, 2005 10:54 PM | direct link

Nate said.."P.S. What about "crimes of passion" where people murder spontaneously? Do such people grasp the consequences at the time? And how common are unpremeditated murders and is this even relevant to this discussion?"

-----I believe premeditation is required for the death penalty. I'm no expert though, and this is an interesting question.

Posted by CH at December 28, 2005 10:10 AM | direct link

I am not an expert either.

HamletsMill makes interesting points. Murderers may not weigh the consequences prior to the crime - premeditated or unpremeditated.

And even though murders continue despite capital punishment and even though not all murderers weigh consequences, there may be to this day a small subset of potential murderers out there who will be deterred by the threat of capital punishment. Maybe this deterrence alone justifies capital punishment. Maybe the idea of removal from society and justice (or "revenge" or "retribution") is superfluos or dangerously subject to abuse.

Posted by nate at December 28, 2005 10:25 AM | direct link

"superfluous" not "superfluos"

Posted by nate at December 28, 2005 10:27 AM | direct link

Here in Florida we kill with great abandon, officially and privately. We are one of the top states when it comes to executions, yet we have a high rate of murders. If the death penalty deters murder why do states like Florida and Texas which have high execution rates have high murder rates?

Frankly, this is a topic which needs to be illuminated by rigorous mathematical analysis. I think that when the government kills it sends a message that killing is good. The death penalty is the opposite of a deterrent, it encourages people to kill.

Texas leads the country in executions, but, like Florida it has a high murder rate. Why?

Posted by Cogliostro Demon at December 28, 2005 10:46 AM | direct link


When the state arrests people and locks them up for murder, does that send the message that kidnapping is good? (And let's not get into tax collection....)

Similarly, Chicago has a lot more snowplows and road salting trucks than Miami, yet also has more snow and more icy roads in the winter. Why?

Posted by albatross at December 28, 2005 11:29 AM | direct link

While economists have no authority nor competence to discuss the morality of public policy, and it is best that they abstain to enter into such discussions, it is a whole different issue to argue that morality should not matter at all.

Economists are needed to caution against possible unintended consequences of measures taken with lofty moral objectives.

But to say that morality does not matter at all? Does the end justify the means? Why do humans, generally, get repulsed and rebel against fully efficient governments with no morals? Maybe it is because morals enter our utility functions? Is it part of our biological survival programming? I don't know, but I would never say that morals are not important to evaluate public policy.

Posted by Bob K at December 28, 2005 11:40 AM | direct link

Cogliostro has a point. If I murder someone else, then I know that if caught I will get the electric chair. After that first murder, the marginal cost of killing a cop, or another person is zero. I can't get the death penalty twice.
So, for criminals and serial killers, the death penalty makes murder cheaper, not more expensive as Becker says. That's why Florida and Texas might have a higher murder rate.

Posted by Bob K at December 28, 2005 11:47 AM | direct link

In other words, the death penalty has a deterrent effect on the number of murderers per capita, but, in principle, a positive effect on the number of murders per murderer. The net effect is uncertain.

Posted by Bob K at December 28, 2005 11:51 AM | direct link

Professor Becker writes: "I believe that deterrence can be the only reasonable basis for capital punishment. Revenge, retribution, and other arguments sometimes made to justify capital punishment are too subject to government abuse, and have been abused."

What makes the case for capital punishment as a deterrent any less subject to abuses from government? In fact, give the age-old axiom of "lies, damn lies and statistics", I would think that "deterrence" as the sole policy motivator would be even more subject to abuse, especially given the decidedly mixed perponderance of the evidence. For example, unquestionably capital punishment if performed very quickly after the crime would have a great deterrent effect; maybe 10 lives saved per execution if all executions were performed within six months after the murder.

Economically speaking, such a policy would make tremendous sence, even if a few "innocents" were executed the greater good would have been served.

I would suspect that injecting some moral or social standard of abstract justice, such as revenge or retribution would in fact have the opposite effect. Revenge is an emotional good only obtained when the right person is put to death, the same with retribution. Sanitizing the policy considerations for capital punishment to merely "deterrence", in my opinion, increases the likelihood of government abuse, by reducing the considerations to a variable completely unconnected with the person to be killed.

Posted by Josh at December 28, 2005 01:15 PM | direct link

Both deterrence and the satisfaction of revenge are based on the people experiencing these feelings/effects belief that the right person gets punished most of the time. If I'm a prospective murderer, and I think the police will, upon finding the body, round up the usual suspect and beat a confession out of one of them selected at random, then there's not much deterrent effect on me. If I'm an angry family member of a murder victim, and I think the police have beaten a confession out of some random street criminal and executed him, I'm not going to feel very satisfied.

These effects are independent of the truth. I'll be deterred by the threat that the police will catch and punish me if I believe it, even if the police are really too incompetent to catch any real criminals. I'll fail to be deterred if I think the police are incompetent, even if they're really very likely to catch me.

Posted by albatross at December 28, 2005 01:58 PM | direct link

Bob K

My only discomfort with the claim that the marginal cost of a second murder is zero under the death penalty is that the chance of capture or of a death penalty verict can still increase by murdering more. The second murder is therefore still has costs for the criminal.

Btw, the second-murder argument doesn't require the death penalty to work: if a state's maximum sentence - whatever it is - is imposed for the first murder then a second murder will be similarly undeterred, notwithstanding my previous point.

Posted by ben at December 28, 2005 02:54 PM | direct link


A kidnapping is not an arrest. An execution, be it gangland or Bush style, is still an execution.

Killing a woman strapped to a hospital gurney and killing a man strapped to an electric chair are both killing. Why is murder OK if the government does it?

Posted by Cogliostro Demon at December 28, 2005 02:56 PM | direct link

Couple points to add:

The stubborn persistance of wrongful convictions in criminal cases, usually based on mistaken eyewitness accounts.

It does not deter. Too many studies say so and the ones that say it does upon closer scrutiny turn out to be gamed. To say that it does because men fear death is to move into the realm of faith and belief. Not that these fields have nothing to add to the debate but to frame a rational defence of capital punishment around them is darkly comic.

The company we keep. When you look at the list of countries that use capital punishment as a form of justice and those that don't one can only ask the United States. Why? Why must you do this?

The historical experience. The institution of African chattle slavery in the old confederacy and capital punishment are closely linked. It was a tool of social control and dominance in the Old South. To use it in the modern era without hesitation is to cheerlfully ignore America's darker impulses, as if the past had no possible hold on the present.

Which brings me to a final comment. To believe in Capital Punishment you must believe in a from of human perfection and human judgement that in my opinion surpasses the ability of real men to achieve in the persuit of justice. All the countries that execute have this in common, the arrogance to think that they can be perfect. It is a conceit that has led America to do evil in the world of men in many fields and I look forward to the day when those who shape public opinion in the Republic approach the world and its problems with a little more humility.

Posted by Northern Observer at December 28, 2005 03:22 PM | direct link


A dozen guys with guns show up at your house, and by threat of violence carry you off to some heavly secured house, where they hold you for several years against your will. How is this not kidnapping? If it were done by the Mafia or Al Qaida, we'd call it that without hesitation.

The point is, we normally don't accept private citizens applying punishments for crimes. Instead, we want the police and courts doing that, since it seems more likely that they'll be somewhat neutral about it, will make some effort to imprison or hang the right person, etc. That means we have the police/courts doing things we don't accept that private citizens can do to punish a crime, whether that's imprisoning people, flogging them, hanging them, torturing them, whatever. Which of those the state ought to be doing is a different moral question.

Posted by albatross at December 28, 2005 04:19 PM | direct link

Northern Observer:

The use of slavery and capitol punishment is closely linked? That is no different than saying cp is closely linked to agriculture. They all go back thousands of years. The US and its 'dark history' have nothing to do with either. They were invented elsewhere, and would continue, with or without us.

Posted by Me at December 28, 2005 04:48 PM | direct link

I did not post on this blog on capital punishment during week 1 because:

-I am not a lawyer or judge
-it is easy to be misunderstood on a blog
-sometimes it is nice not to think about capital punishment

The book "Crime and Punishment" comes to mind:


Posted by nate at December 28, 2005 08:41 PM | direct link

Capital punishment is a deterrent. Studies have shown otherwise, but studies show a lot of things. Correlating murder rates and the death penalty is especially tricky, because of all the possible factors that could affect crime. If we could randomly introduce the death penalty in half of the world's cities, and ban it in the rest, we might be able to get some useful data.

Until then, we should conclude that capital punishment deters based on some very common-sense assumptions.

1) Criminals prefer life to death

2) Criminals are aware of the penalties they face for the crimes they commit

3) Criminals behave rationally

Most arguments against the deterrent effect of capital punishment focus on criminals' lack of ability to reason properly. ("Criminals don't think that they will get caught" or "Criminals don't think about what they're doing.")

Shooting rival gang members in the leg is very common, since the shooter's offence for that is "assault with a deadly weapon" instead of "attempted murder." Sounds pretty rational to me.

But all of this is so obvious that I can't believe it even has to be said. Living proof of the deterrent effect exists in my rommate, who tests my self-restraint with silly arguments like that all the time, without reprecussion.

- C

Posted by Sane Canadian at December 28, 2005 09:47 PM | direct link

One issue that comes to mind is that of the "utility distribution of murderers". What I mean by this is the fact that not all murderers are getting caught, but those who are, are punished. If capital punishment is introduced, then this in-equality in utilities between caught and not-caught murderers rises. This is especially so, if the introduction of capital punishment is complemented with decreasing probability of being caught (to hold constant the expected punishment), as fewer murderers were faced with increased punishment. The fact that in-equality rises holds true for every rise in punishment and/or fall in probability of getting caught.

If public has preference for equality in general -as at least in Europe it seem to have-, then capital punishment clearly decreases social utility in this respect and should be taken into account when determining the total utility gain/loss of capital punishment. To assess the scale of the problem, it would be beneficial to know what proportion of murderers are not getting caught and go unpunished.

Posted by twestlin at December 29, 2005 05:34 AM | direct link

I frankly do not see how any reasonable and relevant philosophy could oppose the use of capital punishment under the assumptions of this example

I think that's rather myopic. You're taking as a given a utilitarian philosphy (like Mills, Bentham), but there's certainly other reasonable, relevant philosophies that would disagree with this.

You ought to consider Locke, in particular, for his natural law philosophy. I emphasize this one, because it seems to me that much of the protections in the Constitution's Amendments 4, 5, and 6 reflect this very philosophy.

According to our Constitution, we can't convict someone based on a forced confession, e.g., even though it might clearly save lives to put away this serial killer now.

Posted by ChrisW at December 29, 2005 11:00 AM | direct link

Although it was not the focus of the post, i have a comment on the question of whether the deterent effect exists, which is the premise of the rest of the argument.

I think that there is a general overestimation of the rationality of potential murderers. Killers are not usually engaging in some sort of cost-benefit analysis which the threat of execution could somehow sway in the direction of not killing. Murder is clearly an irrational act for anyone who believes that they may be caught and values their life at all, and this is true with or without capital punishment.

This, combined with the existence of an overwhelming number of confounding variables, severely compromises the validity of any study on capital punishment's deterent effects. And if there is not conclusive evidence of deterence then i think it is clear that executions should not proceed. The default position in the face of inconclusive evidence should not be to kill more people.

Posted by Owen Thompson at December 29, 2005 06:14 PM | direct link

That's very true. I think there's a need to study the issue more closely.

Posted by hannah at December 30, 2005 03:58 AM | direct link

Though Professor Becker's argument stands if capital punishment has any sizable partial/marginal deterrent effect, one must also consider whether any other option is open. I would say, it is hypocritical indeed to oppose capital punishment if it saves lives (yes, even if only more valuable ones) that cannot be saved in any other way. My point is that this institution is not a first-best solution for the problem for sure, and in our culture we should agree that it is very close to being "last-best".

Two related points:

  • I doubt whether after controlling for many other possible deterrent factors or disincentives, there is a well-identifiable marginal/partial effect left, and even less if we also try to consider related social programs and institutions like gun-control and education.
  • And even if we find evidence that there have been some "net gains in lives" (in quantity or quality) from killing murderers (a positive result), I would not rush to the conclusion that the same effect cannot be achieved in any other way, therefore its implementation is recommended (a normative result).

  • Let me note a similar point in another highly-debated issue: however gainful Levitt and Donohue find abortion in the past, whether or not to prescribe the policy still depends on whether the problems of unwanted babies and single mothers could be solved in more acceptable and/or more gainful ways. (Which were not in the data therefore not controlled for while measuring the partial effect of abortion.)

    People and politicians should always compare viable options and choose, that is what economics tells us. So economists should also look for data whether people find capital punishment, a stronger police force, a more cohesive society, or, say, some intrusion into their private lives (like their "inalienable right" to carry a gun) etc. more worthwhile -- comparing marginal benefits and marginal costs.

    Posted by Skeptical Humanist at December 30, 2005 03:48 PM | direct link

    Only in passing is the issue of wrongly convicted death row inmates addressed. The inference is there is some proportion of the two Mr. Becker would find acceptable. In the abstract, it seems perfectly reasonable; however, what is the proportion that would cause Mr. Becker no loss of sleep, and what of the poor devil who while insisting on his innocence is on the cusp of the equation? Do we keep a tally and then execute with impunity based on this proportional threshold? Further, are you suggesting Governors Ryan and Glendening were wrong to declare moratoriums on the death penalty in their respective states? My heart does not bleed for convicts, however, when we allow the state to murder then we provide the conditions under which corruption can cause an innocent man his life. I think the government should be in the business of eliminating opportunities for corruption not providing them. In my opinion, the deterrent aspect of the pro-death-penalty argument is a red-herring; the real question is, as a matter of policy, does the government have any business putting any one to death for any reason. I wonder how history might have played out differently if the German people, however heinous the individual was perceived to be, would have had an underlining cultural aversion to having the government execute its citizens in say, 1933-45.

    Posted by Ignacio J. Couce at December 30, 2005 06:35 PM | direct link

    "This argument helps explain why capital punishment should only be used for some murders, and not for theft, robbery, and other lesser crimes. For then the trade off is between taking lives and reducing property theft, and the case in favor of milder punishments is strong."

    Not really. Suppose we imposed the death penalty for embezzlement of sums over $10,000. That might well deter enough embezzlement to compensate for the loss of a few lives from those who did embezzle and were caught. Or consider the creation of computer viruses. These can have far higher costs than the two million dollars or so at which human life is conventionally valued for cost-benefit purposes.

    A philosophical issue that arises is why we care about the lives of murderers-- should they enter the social utility function?

    Posted by Eric Rasmusen at December 31, 2005 09:02 PM | direct link

    Post a comment

    Remember Me?

    Are you a spam bot? If not, type "human" here:

    (you may use HTML tags for style)