I’m what you might call a lukewarmer on criminal-justice reform. I agree that some aspects of the current system have gotten out of control; I think that we have the statistical tools we need to identify and release more low-risk inmates; I am also keenly aware that our prisons are not, in fact, filled to the brim with people who just got caught smoking a joint once. On balance I think the current bill is an improvement, though I wish more of Ted Cruz’s amendment — making certain serious offenders ineligible for “time credits” when they participate in recidivism-reduction programs — had been incorporated. Regarding Tom Cotton’s more aggressive proposal for changing the bill further, I recommend his own piece right here at NRO as well as the argument from FreedomWorks that it’s an overly broad “poison pill.”
But in this post I just wanted to spell out the bill’s math a little bit. The Congressional Budget Office finds that it would reduce the prison population by 53,000 “person-years” over a decade. It adds that this is “roughly equivalent to reducing the federal prison population by 53,000 inmates in one year.”
Unsurprisingly I’ve seen this quoted a few times. What it actually refers to, though, is reducing the prison population by 53,000 for a year and then returning it to its previous level. (If the population stayed at the lower level it would keep adding person-years to the total.) It’s not even slightly helpful as a hypothetical outside the budget context.
Here’s a more useful approach: The current federal prison population is about 180,000. If you just crudely multiply that by ten, you get 1.8 million person-years over a decade. 53,000 is about 3 percent of that. That’s the ballpark we’re talking about, and don’t forget that only 13 percent of U.S. prisoners are in the federal (rather than state) system to begin with.
To be fair, there are wrinkles to this number that cut in the other direction as well. The effects of the legislation grow over time because the score starts in 2019 and some of the policies won’t go into effect immediately, and of course some of the effects are hard to predict because prison authorities will have discretion in their implementation. Bear all of these limitations in mind when you see the figure thrown about.