Interpreting the Juvenile Incarceration Drop
by Jeffrey A. Butts, March 20, 2013
Juvenile Justice Information Exchange
On March 17, Nate Balis and Tom Woods from the Annie E. Casey Foundation responded to my JJIE opinion column from March 7 in which I cautioned that it was too soon to claim intentional reform as the cause of recent declines in juvenile incarceration. Nate and Tom argue that policy and practice reforms around the country have at least contributed to the decline. They cite specific examples of states where substantive reforms in juvenile justice were followed by marked shifts in juvenile incarceration.
Nate and Tom make some valid points, and I’m willing to concede two of them. First, they chide me for writing that “incarceration numbers follow the crime rate,” and they are right to do so. I did not intend to characterize the relationship between crime and incarceration as consistent and predictable. I was describing the statistical trends portrayed in my John Jay College data brief. The trends from 1995 to 2010 show a percentage decline in juvenile incarceration that closely tracks the percentage change in juvenile arrests and court referrals.
My observation was accurate, but the sentence was misleading. I would never argue that the relationship between crime and incarceration is always direct and proportionate. In fact, I’ve made the opposite point in previous publications. In our 2011 report on juvenile justice realignment, for example, Doug Evans and I wrote: “The scale of incarceration is not simply a reaction to crime. It is a policy choice.”
Second, Nate and Tom correctly note that my argument generalizes from national data that might obscure important variations among states. This is certainly true, but while the state-level patterns they describe are encouraging, they are not sufficient to establish a causal connection between reform and lower incarceration. Reliable evidence requires more than convenient correlations. Even if we observe a number of instances when state reforms are followed by lower incarceration, we have to test whether the causal hypothesis holds up in the absence of reform. If we lined up all the states according to whether they had enacted meaningful reforms in their juvenile justice systems, would their incarceration trends line up in the same way, with high reform states showing more decline and low reform states showing less? Moreover, does the relationship persist over time and under varying circumstances?
If we had reliable data with which to track system reforms across jurisdictions and in a standard metric, we might be able to disentangle all of the competing influences on the use of incarceration, and we might be able to identify the extent to which recent changes in juvenile incarceration were due to intentional reform or to larger social forces, including crime trends. Of course, it is far easier to imagine such an analysis than to create one. Measuring juvenile justice reform is very difficult because policies and practices vary so much from place to place, and because the organizational components of juvenile justice systems are themselves dependent on local legal culture.
Until we have such an analysis, we are left to speculate about why some juvenile justice systems are using incarceration much less than they did 15 to 20 years ago. Is it largely the result of falling crime, or is it due to the efforts of advocates and reformers? When we know the answer to this important question, we will know whether we should rely on existing strategies for future improvements in juvenile justice policy and practice.