Search About Newsletters Donate
In California, sentencing reforms have reduced incarceration rates in state prisons, but not so much in county jails. Above, Madera County Jail in Madera, Calif.

Measuring Incarceration

Don’t overlook local jails.

As the national spotlight burns bright on criminal justice reform, policymakers and the public are closely watching the data for signs that we’ve turned the page on mass incarceration. Indeed, a recent report from the Brennan Center for Justice noted that 28 states have decreased imprisonment over the last ten years. However, this analysis of state incarceration trends, like every similar one before it, is subject to an important limitation: It excludes individuals in local jails.

One-third of the 2.2 million incarcerated Americans are in jails and, if jail and state trends diverge, the prison numbers alone can give a false picture of a state’s trajectory and make it hard to compare one state with another. As we recently learned, they do diverge—in 16 states.

State jail data is left out of most analyses for a practical reason: it has not been readily available. Although the U.S. government annually collects data from every state prison system, it collects jail data from individual facilities and does not regularly report this data by state. This leaves a severe data gap for those looking to track the nation’s progress in reducing mass incarceration, especially given that some recent sentencing reforms, intentionally or not, have shifted incarcerated populations from prisons to jails. The most notable example is California, where a 2001 court-ordered realignment shifted punishment of certain nonviolent felonies to the jurisdiction of local counties, and a 2014 voter initiative reclassified certain property and drug offenses from felonies to misdemeanors. So while the state prison incarceration rate (the number incarcerated per 100,000 residents) has plummeted by an astonishing 27 percent from 2006 to 2014, jail incarceration rates fell less than five percent. (California still had the greatest percentage drop overall of any state.)

To draw attention to the importance of jails, the Vera Institute of Justice has published comprehensive and comparable state-by-state incarceration data for the period from 1978 to 2014, merging federally-collected prison population data with the jail population data we released in December through the Incarceration Trends Project data tool. This new dataset allows us to examine whether the frequently-cited state prison population trends accurately depict the trajectory of incarceration.

As it turns out, jail and prison incarceration rates can interact in a number of different ways. Between 2006 and 2014, the direction of jail and prison trends mirror one another in 28 states—either both jail and prison incarceration rates are falling (14 states) or both jail and prison incarceration rates are rising (14 states). So in these states, the prison incarceration rate provides a pretty good picture of the overall incarceration trend. In six states no analysis is possible because the state runs both prisons and pretrial incarceration, and the data is commingled.

But in seven states, jail populations have continued to grow while prison populations have declined, and in two of these states (Iowa and South Dakota) the growth of jail populations has exceeded the small decline in prison populations, which means that incarceration has actually risen in these states. On the other hand, in nine states, jail populations have fallen while prison populations have risen. In four of these states (Florida, Idaho, Minnesota, and Oregon) the falling jail population has exceeded the rising prison population to the extent that the state incarceration rate has actually decreased, contrary to what the prison incarceration rate tells us.

The data from these four states suggest where we might look for leadership and ideas in the effort to reduce incarceration. Although prison incarceration has increased slightly in these states, and in small counties jail populations are on the rise, many of the large counties—including Miami-Dade, Broward, and Palm Beach Counties in Florida, Ada County in Idaho, Hennepin, Ramsey, and Anoka Counties in Minnesota, and Multnomah County in Oregon—have made sizable reductions in their jail population. As the saying goes, you manage what you measure. Ending mass incarceration is one of the rare reform efforts where success should be easy to measure. But the metric matters. If success is defined as falling prison populations, there is a real risk of ignoring a key factor—and missing an opportunity to focus on places that are doing something right.

Christian Henrichson is research director of the Center on Sentencing and Corrections at the Vera Institute of Justice.