That Rise in U.S. Crime [Edited]

The FBI released preliminary data on crimes known to the police in 2012. The New York Times will let you know only about a portion of the data. Their author Timothy Williams doesn't tell you that property crimes are down by .8%, but presents a story about how violent crime has increased by 1.5%. Then he find an academic who's willing to go into story time:

Joseph Pollini, another John Jay College professor, said that one possibility was that there were fewer police officers on patrol in some metropolitan areas that have cut spending sharply in recent years because of the recession.

“You’re dealing with depleted police resources,” he said of budget cuts that have caused a reduction in the size of nearly every urban police department.

That's a foolish statement to make even if the rise in overall crime were 1.5%, which it is not. That's because 1.5% is very little. It doesn't call for a big explanation. That's not to say that the rise in violent crime is uncaused, but rather, that you'll have a hard time explaining such a small rise. And, to reiterate, property crime is down (calling into question Pollini's police story). The tables I've found won't give you numbers for total index crimes, but given that property crimes known to the police are much more common than violent crimes (e.g., a ratio of about 8:1 in 2011), this means that the overall number of crimes known to the police is down, contrary to the impression you could get from reading the New York Times.

Of course, one might wonder how valid these numbers are in the first place. O'Brien (1996; gated link) concludes that changes in violent crime were measured with high accuracy between 1973 and 1992, and if the convergence between victimization and police data in more recent years (e.g., here, pp. 391-393) is anything to go by, one may think that the accuracy of police data has gone up rather than down. Taken together with other research, this literature leads me to believe that changes reported by the FBI are probably close to the true change rate for overall violent crime, robbery (+.6), burglary (-3.6), and motor vehicle theft (+1.3). Taken together, this still ain't much of a trend.