Pondering the link between sentencing policy and crime rates

Sequential posts at CrimProf Blog this morning — titled “Use of Death Penalty Dropped in ’04 for Fifth Year in a Row” and “Crime Rates Decline in Early ’04, FBI says” — has me thinking about the links (or lack of links) between sentencing policy and crime rates. 

There has long been, of course, a robust debate over whether and how the criminal law deters, but social scientists still struggle with more refined questions of whether and how marginal changes in sentence lengths and types may impact crime rates.  I find fairly compelling the conclusions of Professors Paul Robinson and John Darley from this recent paper’s abstract:

There seems little doubt that having a criminal justice system that punishes violators, as every organized society has, does have an effect in influencing conduct. Having a punishment system does deter. But the evidence increasingly accumulates that there is little added deterrent effect that can be derived from the manipulation of criminal law rules for the distribution of criminal liability and punishment within that system.

I flag these issues in part because 2004 ought to be a uniquely rich and interesting time period for examining the relationship between sentencing policies and crime rates.  First, on the death penalty front, as noted here, 2004 has given us the first execution-free month in a decade and in the last few years we have seen nationwide declines the total number of death sentences and in the total number of executions.  And yet it appears that murder rates continue to decline even though we apparently are making less frequent use of capital punishment.

Second, in the non-capital sentencing arena, though the Blakely earthquake has had a profound impact on sentencing law and policy, I am inclined to doubt Blakely is having any real impact on crime rates.  Under rational deterrence theory, I think we should expect crimes rates in the second half of 2004 to rise in at least some jurisdictions: rational criminals doing cost-benefit analysis ought to realize that Blakely means it will be harder for those jurisdictions with constitutionally problematic guideline systems to impose long sentences.  But, especially for the most serious crimes, I do not think there are really any “rational criminals doing cost-benefit analysis.”

Because I do not have any economics or social science training, I can do no more than spotlight these issues and hope that other folks much smarter than me start giving these matters serious attention.