Common sense says that the provision of death penalty would make it costlier for people to commit crimes like murder or rape, thus reducing the expected payoffs and accordingly the occurrence of such crimes. But I have also read articles that say that death penalty is not effective in practice and can even lead to some perverse results, for example, increasing the chances that a rapist murders the victim to reduce possibility of getting caught.

Is there a general consensus or if not consensus, widespread agreement on whether, when and why does capital punishment work or does not work in such cases, and other effects of it? Are there good recent reviews of the general state of research on this topic, especially about empirical papers that try causal inference?

  • 6
    $\begingroup$ The repeat-recidivism rate for murder is greater than zero. It follows that giving someone the death penalty for murder is 100% successful in preventing that individual from committing all future crimes, including that small proportion of murderers who go on to commit more murders after getting out. $\endgroup$
    – Richard
    Jul 10 at 20:18
  • 7
    $\begingroup$ If "death" was a deterrent which was always more powerful that "staying alive, but not living in an acceptable environment", no human would ever commit suicide. $\endgroup$
    – alephzero
    Jul 10 at 21:19
  • 2
    $\begingroup$ you might want to reduce the scope to "premeditated violent crimes" $\endgroup$
    – njzk2
    Jul 10 at 21:54
  • 2
    $\begingroup$ @Richard using the same reasoning the best treatment for any disease is death: 100% success that the individual won't be sick again. And if we generalize killing everybody solves every possible problem. $\endgroup$
    – Erwan
    Jul 12 at 15:26
  • 6
    $\begingroup$ Why is this on topic for economics? $\endgroup$ Jul 12 at 16:07

The literature on this issue flip flops. The first seminal study by Sellin (1959)¹ did not find any significant effect, but early literature was riddled with methodological errors. Ehrlich’s (1975)² study at the time was far more sophisticated, and it showed a strong deterrent effect of capital punishment. However, ironically, this study later too became criticized for lack of empirical rigor (as the statistical method advanced). Cochran and Chamlin (2000)³ and Lamperti (2008)⁴ showed that not only it did not deter violent crimes, but it could also have even increased them via the 'brutalization' effect. However, later Dezhbakhsh, Rubin, and Shepherd (2003)⁵ employing panel data again showed it has a deterrence effect, but that study was later itself challenged by Kovandzic, Vieraitis, and Boots (2009)⁶.

New meta analyses show that evidence is somewhat mixed too. For example, meta analysis of Yang & Lester (2008)⁷ argues there is a significant deterrence on some crimes, but meta analysis of Dölling et al. (2009)⁸ finds that the effect of capital punishment on violent offenses is not significant. It is very hard to fish out the true effect because of endogeneity issues. Places in US with high violent rates often have capital punishment, is that because those places need the deterrent most or because it is ineffective deterrent and other places have better policies? Although it is matter of ongoing discussion most researchers now tend to lean more to the side that capital punishment is not effective deterrent but the research is not fully settled.


  1. Sellin, Thorsten. 1959. The death penalty. Philadelphia, PA: American Law Institute.
  2. Ehrlich, Isaac. 1975. The deterrent effect of capital punishment: A question of life and death. American Economic Review, 65: 397–417.
  3. Cochran, John K. and Mitchell B. Chamlin. 2000. Deterrence and brutalization: The dual effects of executions. Justice Quarterly, 17: 685–706
  4. Lamperti, John. 2008. Does capital punishment deter murder? A brief look at the evidence. Retrieved October 8, 2009 from math.dartmouth.edu/~lamperti/capitalpunishment.pdf
  5. Dezhbakhsh, Hashem, Paul H. Rubin, and Joanna M. Shepherd. 2003. Does capital punishment have a deterrent effect? New evidence from postmoratorium panel data. American Law and Economics Review, 5: 344–376
  6. Kovandzic, T. V., Vieraitis, L. M., & Boots, D. P. (2009). Does the death penalty save lives? New evidence from state panel data, 1977 to 2006. Criminology & Public Policy, 8(4), 803-843
  7. Yang, B., & Lester, D. (2008). The deterrent effect of executions: A meta-analysis thirty years after Ehrlich. Journal of Criminal Justice, 36(5), 453-460.
  8. Dölling, D., Entorf, H., Hermann, D., & Rupp, T. (2009). Is deterrence effective? Results of a meta-analysis of punishment. European Journal on Criminal Policy and Research, 15(1), 201-224.
  • 7
    $\begingroup$ Interesting bit about the correlation between high crime prevalence and harsh laws. A conjecture coming to my mind is that such communities focus on the wrong things, namely ineffectual punishment instead of "soft" interventions that strengthen the social fabric and provide alternatives to a criminal career for young men. The Freaconomics drug dealer chapter anecdotally shows how being a low-earning, high-risk criminal is a rational decision when no other opportunities exist. The dealers directly risk their lives -- their death rate is higher than on death row -- but that does not deter them. $\endgroup$ Jul 11 at 11:07
  • $\begingroup$ All these studies seem focused on the USA. I'm curious if there are any studies related to the execution of drug smugglers & the low crime rate in Singapore. $\endgroup$
    – axsvl77
    Jul 12 at 4:02
  • 6
    $\begingroup$ @axsvl77 If you look at international comparisions here you should also notice that every single European country does not have the death penalty and every single one of them has a much lower murder rate than the US. Singapore also has lower crime rates than the US and has the death penalty. $\endgroup$
    – quarague
    Jul 12 at 13:41
  • 3
    $\begingroup$ @axsvl77 It seemed like you wanted to argue that the death penalty works by citing the example of Singapore. I think this would be a highly selective view as there are plenty of other countries with low crime rates without the death penalty. So it is not death penalty may or may not work but rather low crime rate can be achieved with or without death penalty. $\endgroup$
    – quarague
    Jul 12 at 14:30
  • 1
    $\begingroup$ @quarague History is complicated. I don't want to speculate about what studies related to Singapore's prolific executions might show. I just do know that this answer, despite the upvotes, does not answer the question because it is focused on only the USA. I was, and am suggesting that this answer should be expanded to include more information. $\endgroup$
    – axsvl77
    Jul 12 at 16:02

The death penalty as a deterrent to crime is ineffective is the prevailing view amongst researchers and experts in the field of criminology. Some studies would indicate that there is no correlation between the death penalty and the reduction of violent crime. However, like any topic, you can find contrarian views that say there is no conclusive evidence to say that it does not. The increased use of econometric models is inconclusive aswell.

"The data alone cannot reveal what the homicide rate in a state without (with) a capital punishment regime would have been had the state (not) had such a regime."

"In their book, Professors Hood and Hoyle say almost all the academic studies available for review are concerned with the deterrent effect of capital punishment on the rate of murder in the United States.

The authors say theoretical and methodological issues have 'dogged the attempts to prove or disprove the existence of the deterrent effect of executions in the United States' and 'a fierce controversy continues' in the United States over attempts to use econometric models to address the question.

After reviewing the literature they conclude that 'it is not prudent to accept the hypothesis that capital punishment, as practised in the United States, deters murder to a marginally greater extent than does the threat and application of the supposedly lesser punishment of life imprisonment'."

"His position is shared by the majority of criminologists in relation to homicide, according to a 2009 survey of members of the American Criminology Society, who were asked to limit their answers to their understanding of the empirical research and to exclude their personal opinions. That study found that over 88 per cent of the criminologists did not believe the death penalty deterred murderers."

"A comprehensive review of the research in this area over 34 years was conducted in 2012 by a committee of the American National Academy of Sciences National Review Council. The committee concluded that "research to date on the effect of capital punishment on homicide is not informative about whether capital punishment decreases, increases, or has no effect on homicide rates"

Furthermore, as a penalty itself, the death penalty is morally repugnant and an anti-thesis to any government that would claim moral authority over its citizens.

Additionally, it is extremely costly as a form of punishment, some within the UN argue it can constitute torture and for its finality as a punishment, there is far too much human error involved in its delivery.

"From 1973 through December 2014, 150 innocent people were exonerated from death row"- https://www.aclu.org/issues/human-rights/human-rights-and-death-penalty

Then there are the issues within the legal institutions that would apply it, and the biases they hold.

• Jurors in Washington state are three times more likely to recommend a death sentence for a black defendant than for a white defendant in a similar case. (Prof. K. Beckett, Univ. of Washington, 2014). • In Louisiana, the odds of a death sentence were 97% higher for those whose victim was white than for those whose victim was black. (Pierce & Radelet, Louisiana Law Review, 2011). • A study in California found that those convicted of killing whites were more than 3 times as likely to be sentenced to death as those convicted of killing blacks and more than 4 times more likely as those convicted of killing Latinos. (Pierce & Radelet, Santa Clara Law Review, 2005).

In terms of economic analysis, I don't know if this is what you had in mind, but it is related, with cited research by economists who've endeavored to study the issue through that lens. Albeit from 1996


A small snippet from the above link - "In 1973, economist Isaac Ehrlich studied data from robberies that occurred in the 1940s,1950s, and 1960s. He found that the higher the probability of conviction for robbery, the lower the robbery rate, ceteris paribus (Cooter and Ulen 1988). However, he found that the severity of the punishment did not affect deterrence in 1940 and 1960. Interestingly, he found a deterrent effect based on the severity of punishment when studying robbery reports from 1950. More recently the Capital Punishment Research Project and the New York Times compared capital and non-capital states to assess deterrence effects (Economist 1994a).The investigation examined the number of murders that occurred in New Jersey before and after the imposition of a death penalty statute in 1982. No statistically significant decrease was found in the number of murders that occurred. The study also compared the number of murders per 100,000 residents in both Massachusetts (a non-capital state) and New Jersey (presently a capital state). No significant difference was found in the number of murders. In the same study, the murder rate in New York (a non-capital state when the study was conducted) was compared to the rate in Texas (a capital state); and there was no statistical difference between the two states with respect to the number of murders per 100,000 residents. Interestingly, while most southern states have the death penalty, they also have higher murder rates. In fact, Louisiana, a capital state, has the highest murder rate in the nation. Among southern states, only Florida has a murder rate below the national average"

In terms of overall views on the subject, it differs highly in which country you are referring to, as it is not just a scientific issue, but a political one. There is also unfortunately a lot of ignorance and emotionally distorted reasoning when dealing with the topic with those unfamiliar with it.

  • 16
    $\begingroup$ this answer just argues against the use of death penalty instead of answering the request for empirical literature review on its efficacy $\endgroup$
    – WilliamT
    Jul 10 at 10:46
  • 8
    $\begingroup$ Your answer has valuable information, it could be pruned a bit to make it more focused and brief. $\endgroup$
    – Giskard
    Jul 10 at 12:02
  • 3
    $\begingroup$ I'm sitting here with a mere 101 rep and can't downvote but this answer should be junked. $\endgroup$
    – Joshua
    Jul 12 at 2:51
  • 1
    $\begingroup$ @Joshua Therein lies the problem with HNQ on controversial topics. Regardless of whether one agrees with this particular answer, drawing in a bunch of 101 rep users from other SE sites on controversial questions leads to massively skewed voting, since the 101 rep users can upvote, but not downvote. This happens all the time on sites like Politics and Skeptics that get a lot of controversial topics. Personally, I think the upvote rep threshold should also increase to 125 while a question is on HNQ, at least on sites that frequently get controversial questions. $\endgroup$
    – reirab
    Jul 12 at 22:34
  • 1
    $\begingroup$ I should also note that if one is assessing this purely economically and trying to isolate 'moral' questions from away from the topic (boo, but economics has a bad habit of getting in trouble when it tries to encroach on moral philosophy) then we also need to factor in the utility of innocent prisoners getting whacked, and while the estimates vary, most fall into the "way too many" camp, and thus we have to evaluate the suffering caused by executing an innocent man against whatever the deterent effect might be. It doesnt bode well for capital punishments defenders. $\endgroup$
    – Shayne
    Jul 13 at 7:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.