Although the Supreme Court banned capital punishment for child rape last week, the justices have made it clear that for homicide, states may inflict the ultimate penalty. Last month, capital punishment resumed after a seven-month moratorium. Rapid scheduling of executions followed the Supreme Court's ruling in Baze v. Rees, reaffirming the constitutionality of the death penalty in general and lethal injection in particular.
To support their competing conclusions on the legal issue, different members of the court invoked work by each of us on the deterrent effects of the death penalty. Unfortunately, they misread the evidence.
Justice John Paul Stevens cited recent research by Wolfers (with co-author John Donohue) to justify the claim that "there remains no reliable statistical evidence that capital punishment in fact deters potential offenders." Justice Antonin Scalia cited a suggestion by Sunstein (with co-author Adrian Vermeule) that "a significant body of recent evidence" shows "that capital punishment may well have a deterrent effect, possibly a quite powerful one."
What does the evidence actually say?
One approach notes that in states with the death penalty, the average murder rate is about 40 percent higher than in states without the death penalty. Yet such comparisons are surely confounded by other influences, as those states that impose the death penalty also have a historic culture of violence, including lynching.
If we compare countries, the United States has higher execution and higher homicide rates than nearly all other industrialized countries. Here, too, many alternative explanations remain, making it hazardous to conclude that the death penalty does not deter murder.
Other studies have evaluated changes in homicide rates over time. In the 1960s, as the death penalty fell into disuse, homicide rates rose sharply, leading some studies to infer a deterrent effect. Moreover, a large-scale decline in homicide in the past two decades coincided with renewed use of the death penalty. Countering this, homicide and execution rates rose together in the 1920s and early 1930s, then fell together through the 1940s and 1950s. Because conclusions are so sensitive to the time period evaluated, these studies fail to provide much help.
More sophisticated studies compare the evolution of homicide rates across jurisdictions. Over the past six decades, the homicide rate in Canada has tracked that in the United States even as the countries' punishment policies have diverged sharply. Similarly, the 12 states that have not executed a prisoner since 1960 comprise a useful comparison group; murder rates in these states have largely tracked those in states that subsequently adopted or rejected the death penalty.
One might like to conclude that these latter studies demonstrate that the death penalty does not deter. But this is asking too much of the data. The number of homicides is so large, and varies so much year to year, that it is impossible to disentangle the effects of execution policy from other changes affecting murder rates. Moreover, execution policy doesn't change often or much. Just as a laboratory scientist with too few experimental subjects cannot draw strong conclusions, the best we can say is that homicide rates are not closely associated with capital punishment. On the basis of existing evidence, it is especially hard to justify claims about causality.
Justice Stevens argues, "In the absence of such evidence, deterrence cannot serve as a sufficient penological justification for this uniquely severe and irrevocable punishment." Perhaps. But the absence of evidence of deterrence should not be confused with evidence of absence.
Justice Scalia relies on the suggestion by Sunstein and Vermeule that some evidence suggests a possible deterrent effect. But that suggestion actually catalyzed Donohue and Wolfers's study of available empirical evidence. Existing studies contain significant statistical errors, and slightly different approaches yield widely varying findings, a problem exacerbated by researchers' tendency to report only those results supporting their conclusions. This led Sunstein and Vermeule to acknowledge: "We do not know whether deterrence has been shown. . . . Nor do we conclude that the evidence of deterrence has reached some threshold of reliability that permits or requires government action."
In short, the best reading of the accumulated data is that they do not establish a deterrent effect of the death penalty.
Why is the Supreme Court debating deterrence? A prominent line of reasoning, endorsed by several justices, holds that if capital punishment fails to deter crime, it serves no useful purpose and hence is cruel and unusual, violating the Eighth Amendment. This reasoning tracks public debate as well. While some favor the death penalty on retributive grounds, many others (including President Bush) argue that the only sound reason for capital punishment is to deter murder.
We concur with Scalia that if a strong deterrent effect could be demonstrated, a plausible argument could be made on behalf of executions. But what if the evidence is inconclusive?
We are not sure how to answer that question. But as executions resume, the debates over the death penalty should not be distorted by a misunderstanding of what the evidence actually shows.
Cass R. Sunstein, the Felix Frankfurter Professor of Law at Harvard Law School and Justin Wolfers, assistant professor of business and public policy at the University of Pennsylvania's Wharton School.