First, let me begin with the usual caveat that correlation does not prove causation. Actual proof is a difficult, if not impossible, thing to do with crime statistics, and I do not claim that these simple comparisons prove the case that Realignment and Proposition 47 have caused the results noted here. I will return to this point at the end of the post.

The violent crime graph shows California and United States violent crimes rates falling in tandem prior to Realignment. That bill took effect in October 2011, so 2011 is mostly a pre-Realignment year. We would expect effects to show up in later years. There is a bump in 2012, while the national number is flat, followed by a drop the year after that. Beginning in 2015, California's violent crime rates have been above the national rate to a larger extent than previously.

Overall, the California rate averaged 9% above the national rate before Realignment and 12% above since Proposition 47.

A report by the Public Policy Institute of California notes two statistical problems in violent crime rates -- the FBI's change in the definition of rape to be more inclusive and underreporting by LAPD in 2008-2014. The first problem would not affect the comparison of California to the country as a whole because it applied nationwide. The second problem I will examine more closely in a later post by looking at the state minus the City of Los Angeles.

As noted in my prior post, property crimes are where we would expect to see the largest effect if there is one. Before Realignment, California's property crime rate declined steadily with the nation's at about 9% below. The year after Realignment, California's rate had a spike and came close to the national rate. In the next two years both national and state rates decline, but California never made it back to the 9% below.

After Proposition 47, California's property crime rate jumped to 5% above the national rate, and it has maintained that differential for two years now.

There may be a statistics problem here that makes the real differential even greater. The FBI numbers only track crimes reported to the police. A major reason for not reporting is that the victim feels the police won't do anything. For theft, which makes up nearly three-quarters of the property crime index, that feeling is largely true if the crime is a misdemeanor rather than a felony. By changing a large number of thefts to misdemeanors, Proposition 47 may well have reduced the reporting rate. The official statistics are bad, but reality may well be worse.

Some studies have attempted to compare California with a selected number of states rather than the country as a whole, claiming that the states were selected to form a "synthetic control." I am skeptical of this approach, as it seems to be readily subject to manipulation. I hope to be able to examine these studies in more detail and will report the results here when I do.

So what degree of proof would we say the simple comparison above establishes that California's soft-on-crime legislation has increased crime? In terms familiar to lawyers, is it "proof beyond a reasonable doubt"? Certainly not. "Preponderance of the evidence." No, I wouldn't claim that. "Probable cause"? Arguably. "Reasonable suspicion"? Certainly.

Update:  This post is noted at Sentencing Law and Policy, which I appreciate, but the excerpt there includes only the paragraph on violent crime rates followed by the concluding paragraph, as if the conclusion is based on the violent crime rates.  Once more, with feeling, given the nature of the changes made by Realignment and Proposition 47, we would expect the stronger statistical connection in the property crime rates, and that is exactly where we find it.