Research needed to tell if fewer arrests increase crime

Would arresting fewer low-level offenders increase crime? In the wake of Hurricane Rita, researchers have a chance to empirically measure that notion -- I'd love to see somebody in academia take up the project.

Rita's aftermath affords an almost unique chance to research the impact of reducing low-level arrests on crime rates. The Beaumont Enterprise reports that the Orange County jail failed inspection and still cannot re-open ("Orange County jail fails inspection, "Dec. 2). "Since the jail has been closed, law enforcement officers have had few spots to house criminals so arrests are way down," the paper reported. Grits earlier noted that officials in Gregg and Chambers Counties are in the same boat, making fewer arrests, especially of non-violent offenders, because of jail overcrowding.

So, during the period when those jurisdictions have reduced arrests, did the number of crimes reported increase, decrease, or stay about the same? I don't know the answer to that, though I haven't heard of any new crime wave reported in those areas. Local officials don't seem to think crime has worsened as a result: “In no way are the lives and safety of the citizens of Chambers County being put at risk,” [Chambers Sheriff Joe] LaRive said. “We’re still doing our job — we’re still protecting the public.”

To tell if that's really true, researchers would have to examine local records about the number of crimes reported before and after the hurricane for counties that reduced arrests. I'll bet he's right, though -- I've long suspected that a lot of low-level arrests, especially for non-violent offenses, aren't really improving public safety. With a little research it'd be possible to find out for sure.