The high cost of the exclusionary rule
...I think the better policy is to let the evidence in as well as evidence of how it was obtained. the latter will discourage police from obtaining evidence by abuse and coercion. Juries would be unlikely to convict in a case where the police beat a confession out of a witness or defendant. Police would be reluctant to engage in egregious conduct if they knew it would be exposed at trail.
How did we get here? When Mapp v. Ohio (the Supreme Court decision applying the exclusionary rule to the states) was adopted in 1961, exactly half of the states (24) already had an exclusionary rule in place, and half did not. This set up an ideal situation for a statistical experiment to determine what the effects of the rule had been on crime rates. In a 2003 paper published in the Journal of Law and Economics, Raymond Atkins and I looked at just this issue. We compared crime rates in states that had adopted their own exclusionary rule with states that had the rule forced on them by the Supreme Court. We statistically controlled for other factors, such as the demographic characteristics of the population.
The results were striking. In our basic model, we found significant increases in crime in jurisdictions forced by the Supreme Court's ruling to exclude evidence. Those increases were 3.9% for larceny, 4.4% for auto theft, 6.3% for burglary, 7.7% for robbery, and 18% for assault. For murder, the increase was small and statistically insignificant.
To put these numbers in perspective, for the entire U.S. in 2007 (the most recent year for which FBI data is available), these percentage increases would translate into 246,000 additional cases of larceny, 46,000 more auto thefts, 129,000 additional burglaries, 32,000 more robberies, and 130,000 more aggravated assaults.