Was the prison boom driven by locking up people sentenced for nonviolent crimes? Politifact.com, the Tampa Bay Times’ fact checker website, recently checked a statement by Sen. Cory Booker, who wrote on Instagram that, “We built a new prison every 10 days between 1990 and 2005 to keep up with our mass incarceration explosion of nonviolent offenders”. The site said of Booker’s statement that, “On balance, we rate his claim Mostly True”. Which raises the question of what exactly does it mean to say something is “Mostly True”? In this case, it means that the first part is correct, but the second part is only partly so. So, while the facts might (mostly) add up, the meaning of the statement is misleading.

It turns out that the statement is true up to, but not including, the last three words; “of nonviolent offenders”. In the posting on Politifact.com, reporter Keely Herring does a good job in parsing out the statement. Herring cites the Congressional Research Service, which reported in 2010 that the number prisons in the U.S. had increased from 1,277 in 1990 to 1,821 in 2005. The increase of 544 new prisons over the 15-year period works out to one every 10 days.

But, writes Herring, “The second part of Booker’s claim is somewhat less clear”. He cites the work of John Pfaff, a professor at Fordham Law School, who has written extensively about what has driven incarceration rates in recent years (hint: its prosecutors). In a paper he wrote for the Harvard Journal on Legislation, Pfaff noted that violent offenders contributed 60 percent to state prison growth between 1990 and 2009, while nonviolent offenders (drug and property offenses combined) contributed only 27 percent. There is some truth to the point that “nonviolent” offenders once drove the lion’s share of the increase; 55 percent of state inmate growth came from nonviolent offenders between 1980 and 1990, (the period before the building boom Booker mentions). Concludes Herring, “So Booker is wrong to focus on the period 1990 to 2005 when discussing a “mass incarceration explosion of nonviolent offenders.” A big rise did occur, but it started earlier than Booker’s post indicated.”

So the statement was “Mostly True” as far as it goes. It does, however, leave aside that the original statement referred to data that is fifteen years old and to a period when, in fact, it was violent offenders who were driving incarceration rates. But isn’t a statement that’s only mostly true, really and actually not true? This is the problem with factoids; they sound, and are, mostly true, but they leave an impression that is misleading. I don’t believe Sen. Booker is trying to mislead people. He’s just picking up a factoid and fitting it into a narrative that fits with what has become a common assumption about too many nonviolent offenders in custody. In this view, we just went a bit overboard on putting in prison who shouldn’t be there (ergo, all we have to do is cut that out and everything will be fine). In previous posts on low risk drug offenders and minimalist policy proposals, I explain why I’m not sanguine about how well this will work.