Ron Unz caused quite a stir in the blogosphere when he published an article in The American Conservative—yes, Pat Buchanan’s former magazine—arguing that Hispanic crime rates are not higher than white crime rates. Unz called Hispanic criminality a “myth” that immigration restrictionists should stop repeating if they want to be taken seriously.
Already familiar with much of the relevant data, I wrote a detailed critique of Unz’s article for the new webzine Alternative Right, and Unz then responded to my critique. I won’t rehash all of the points of contention here, as interested readers can follow the links to decide for themselves who has the better of the argument. In my opinion, much of Unz’s response was already anticipated and answered in my original critique.
But one part of his response did bring some new data from California to bear on the question, so let me tackle that here. First, some quick background. Unz calculated the Hispanic incarceration rate divided by the white rate (call the result “HDW”) using state-level data published by the Bureau of Justice Statistics. He then crudely controlled for age and gender, and declared that HDW in California is only about 1.1. (I’m inferring that number from one of his bar graphs.)
I found his approach problematic for several reasons. I preferred to use American Community Survey data on institutionalization. (Note: Institutionalization does not include people in college dorms or military barracks, as has been alleged.) Using regression analysis to adjust for age, my HDW calculation for California comes to 1.53, which is substantially higher than Unz’s figure.
Though this is only one of many contentious issues, Unz suggested that a 2006 report from the Public Policy Institute of California (PPIC) could help resolve it. The PPIC report uses incarceration data from California’s government that Unz says could be among the most reliable available. The report gives an HDW of 1.48, which Unz claims as a victory. Once we control for age, he says, the HDW comes down to just about his 1.1 estimate.
But Unz missed something important in the report. The 1.48 number is already fully controlled for age. I exchanged emails with one of the PPIC report’s coauthors to confirm this fact.
To reiterate, Unz’s estimate for California was 1.1, mine was 1.53, and the estimate of his suggested referee gives us 1.48. To the extent that the PPIC report can corroborate anyone’s analysis, it is mine—not his—that is confirmed.