A. The New Deterrence
The modern debate on deterrence and capital punishment, now in its fourth decade, was launched by two closely timed events. The first was the 1976 United States Supreme Court decision in Gregg v. Georgia,1 which restored capital punishment after its brief constitutional ban following Furman v. Georgia2 in 1972.3 In 1975, Professor Isaac Ehrlich published an influential article saying that during the 1950s and 1960s, each execution averted eight murders.4 Although Ehrlich's article was a highly technical study prepared for an audience of economists, its influence went well beyond the economics profession. Ehrlich's work was cited favorably in Gregg and later was cited in an amicus brief filed by the U.S. Solicitor General in Fowler v. North Carolina.5 No matter how carefully Ehrlich qualified his conclusions, his article had the popular and political appeal of a headline, a sound bite, and a bumper sticker all rolled into one: "every execution deters eight killings."
Reaction was immediate: Ehrlich's findings were sharply disputed in academic forums such as the Yale Law Journal,6 launching an era of contentious arguments in the press and in professional journals.7 In 1978, an expert panel appointed by the National Academy of Sciences issued strong criticisms of Ehrlich's work.8 Over the next two decades, economists and other social scientists attempted (mostly without success) to replicate Ehrlich's results using different data, alternative statistical methods, and other design modifications that tried to address glaring errors in Ehrlich's techniques and data. The accumulated scientific evidence from the NAS report and these later studies weighed heavily against the claim that executions deter murders.9
The debate both revived and shifted within the past decade. Since 1996, more than a dozen studies have been published claiming that each execution can prevent anywhere from three to thirty-two homicides.10 The new deterrence studies analyze data that span a twenty-year period since the resumption of executions following the United States Supreme Court's decisions in Furman11 and Gregg.12 The claims of these new studies are far bolder than the original wave of studies by Professor Ehrlich and his students.13 Some claim that pardons, commutations, and exonerations cause murders to increase.14 One says that even murders of passion, among the most irrational of lethal acts, can be deterred.15 In short, these studies suggest that the deterrent effects of capital punishment apparently are limitless, leading some proponents to offer execution as a cure-all for all types of murder.16
Both legal scholars and social scientists have transformed this new social science evidence into calls for more executions that they claim will save lives,17 and new rules that will remove procedural roadblocks and hasten executions.18 Others challenge the scientific credibility of these new studies,19 and warn about the moral hazards and practical risks of capital punishment.20
Obviously, the stakes are high in this latest round of the recurring debate on deterrence. We think the new results are wrong, for a simple reason. The measures of homicide used in the new deterrence studies are overly broad: by studying whether punishments affect all homicides, these studies fail to identify a more plausible target of deterrence-namely, those homicides that are punishable by death. By broadening the target of the search for deterrent effects, these studies have overestimated not just the number of lives saved by deterrence, but whether any murders are averted by the threat of execution.21 In this study, we find no evidence of deterrence when the effects of execution are estimated for the subset of homicides that are most directly affected by execution.
B. Errors in Aggregation
The question of whether the threat or actuality of execution adds to the deterrent effect on homicide produced by lengthy imprisonment alone has been the subject of statistical debate for more than a century. …