By Jeremy J. Collins
Though most experts have long dismissed any measurable deterrent effect from the use of the death penalty, a recent AP story helped spark new discussion on the topic.
According to the report, some academic studies have purported to find such a deterrent impact. The story cited a 2003 Emory University study which concluded that each execution deters an average of 18 murders. To read the story, one might believe that new life has been pumped into what had largely been a settled argument.
A closer look at the facts, however, reveals that there was very little to the story. The truth is that leading academics have roundly rejected these studies. A rigorous 2006 study conducted by John Donohue of Yale Law School and the National Bureau of Economic Research (NBER) and Justin Wolfers of the Wharton School of Business and NBER analyzed the same data used in the Emory study and like studies and debunked their conclusions in striking terms: “The view that the death penalty deters is still the product of belief, not evidence.†In fact, the researchers found that, if anything, “the evidence suggests that the death penalty may increase the murder rate.â€
These dueling findings have a deja vu quality. The studies purporting to find a deterrent effect all build upon the foundation of a 1975 article in which Prof. Isaac Ehrlich claimed that each execution averted eight murders. Economists and social scientists attempted to replicate his findings by using different data and improving on his methodology. The overwhelming majority of such studies found no evidence that the death penalty deters murderers. Indeed, a 1978 panel of experts appointed by the National Academy of Sciences strongly criticized Ehrlich’s work and methodology.
Jeffrey Fagan, a professor at Columbia Law School and an expert on statistics, testified before Congress that the Emory study and similar Ehrlich-inspired studies finding a deterrent effect are “fraught with numerous technical and conceptual errors.†Important among these problems is that the studies “avoid any direct tests of deterrence.†That road would likely not lead to deterrence findings: numerous studies “show the limits of the assumptions or rationality that underlie deterrence†while others show the cognitive, organic and neuropsychological impairments which characterize violent offenders.
Instead of attempting a direct test of deterrence, the Ehrlich-inspired studies acknowledge that the factors leading a person to murder (or not murder) are complex and numerous — including socioeconomic variables, crime rates and the efficacy of the criminal justice system in catching, convicting and punishing criminals. The studies purport to isolate every other factor but the availability of the death penalty as punishment for murder. But common sense and respected scientists such as Profs. Donahue and Wolfers tell us that the number of homicides that executions “can plausibly have caused or deterred cannot be reliably disentangled from the large year-to-year changes in the homicide rate caused by other factors.â€
When touting superficially powerful arguments in favor of the state executing our fellow human beings, the media and academics have a duty to acknowledge the facts contradicting their claims. These facts include not only that respected academics have rejected claims that the death penalty deters, but also the following:
- Â Murder rates are lower in states without the death penalty. This holds true even when comparing neighboring states.
-  While Southern states account for over 80 percent of the executions in this country, they have consistently had the highest murder rate of the nation’s four regions.
- Â Since 1972, homicide rates in Canada and the United States have moved in lockstep, yet in that period, Canada has not executed a single person and the United States has executed over 1,000 people. When homicides go down in the United States, they go down in Canada, even though Canada does not use capital punishment.
-  One of the authors of the Emory study (Joanna Shepherd) found in a separate study that while the death penalty deterred murder in six states, it actually increased murder in 13 states and had no effect on the murder rate in eight states. Other studies have found that the death penalty has a “brutalization effect,†increasing the number of murders.
Danger lurks whenever we look at statistical claims without a skeptical eye, never more so than when the issue is life and death. We need straight information before making an informed decision on the death penalty. Statistical claims that wilt under the mildest scrutiny woefully fail to meet that test.
I (as have others) suggest that it may not be accurate to claim that the death penalty offers no deterrence. Death sentences may provide some level of deterrence, but probably not at the levels claimed by certain economists (thank you for a well written review of this subject!). Anyone interested in this topic is encouraged to review the excellent work of Ruth Peterson and William Bailey. The valid research has suggested that there is no statistically significant difference between capital punishment and life in prison in terms of deterrence. The latter is also reversible and costs far less.
nicely put mr. blankenship…
This article is quite good! The rebuttal to the “academics” assertion that statistics “prove” that the death penalty is a deterrent certainly validates Mark Twain’s assertion about the efficacy of statistics. In fact, a strong case can be made that the existence of the death penalty can actually put more people at risk (particularly police officers) because of the “life or death” decision to eliminate witnesses. The National Association of Police Chiefs certainly agree that the existence of the death penalty creates a greater risk to police officers in those states which use the death penalty as opposed to those states which do not have the death penalty.
I find this article very interesting because I have always questioned whether capital punishment was a deterrence to murder or any other crime. I am glad that the article points out the fact that statistically it can be proven that states such as Texas have higher capital punishment rates but their murder rates are also higher. It is also a good point to make that the neighboring states may also be affected by the rates of capital punishment of their neighbors. In my opinion I think that if capital punishment does deter crime as some experts would argue, it deters in very small amount. More so though I believe that in those states that the death penalty is not so common there is a lower murder rate, that would make me believe that the deterrence is not significant enough to keep allowing for the death penalty in any state.
Recently, a colleague produced a 2003 study concluding that the death penalty had deterrent effects. Of course, recent studies suggest otherwise, for reasons that turn on not only the justice question but also the brutalization effects, both before, during, and after the commission of the murder. Indeed, people charged with murder, by and large, have background experiences that predispose them to a sort of caloused approach to questions of life. In quite a few cases, they live brutalized existences prior to the tragic events. This is consistent with what researchers have uncovered when they look at street crimes. In the corporate suites, a different dynamic may be operating. I wonder how we assess the murders committed by corporate thugs, for instance, via toxis dumping and other such environmentally (human) destructive processes? Any deterrence here? In any event, supporters of the death penalty have no problem cottoning onto illogical and contradictory stances. they suffer no pangs of guilt while thousands of people are brutally murdered by, sometimes, their own governments. Otherwise how does one account for the shame that is Abu Ghraib, Guantanamo, renditions, water boarding, etc.?
The Donahue and Wolfers criticism has been strongly rebutted.
16 recent US studies, inclusive of their defenses, find a deterrent effect of the death penalty.
Â
All the studies which have not found a deterrent effect of the death penalty have refused to say that it does not deter some. The studies finding for deterrence state such. Confusion arises when people think that a simple comparison of murder rates and executions, or the lack thereof, can tell the tale of deterrence. It cannot.Â
Â
Both high and low murder rates are found within death penalty and non death penalty jurisdictions, be it Singapore, South Africa, Sweden or Japan, or the US states of Michigan and Delaware. Many factors are involved in such evaluations. Reason and common sense tell us that it would be remarkable to find that the most severe criminal sanction — execution — deterred none. No one is foolish enough to suggest that the potential for negative consequences does not deter the behavior of some. Therefore, regardless of jurisdiction, having the death penalty will always be an added deterrent to murders, over and above any lesser punishments.
Â