Vera’s Incarceration Trends State Fact Sheets

An Important New Resource
Christian Henrichson Director, Vera Insights
Dec 03, 2019

This week we released Incarceration Trends in Local Jails and State Prisons fact sheets for each state.

For several years, Vera researchers have been working to advance the availability of county- and state-level incarceration data through projects such as the Incarceration Trends data tool and the People in Prisons report series. These fact sheets are a new resource that capture the best available jail and prison data in a single document for each state.

In addition to placing a state’s jail and prison data literally on the same page, we designed these documents so that people can easily see how their state compares to others. Mass incarceration manifests differently in each state, so the reform priorities in one state will likely not be the same as in another. We also present data broken down by race, gender, and geography to feature three aspects of incarceration that are at the core of Vera’s research and advocacy: racism as a central feature of mass incarceration, rapid growth in the number of women behind bars, and rising incarceration rates in smaller cities and rural counties.

To distill all this information into a four-page document, we made a number of decisions to aid readability and interpretation. Some of these decisions are not conventional but, in our view, many of the old conventions merit rethinking.

  1. We present historical trends for the longest period possible with the available data. Analysis of incarceration trends, particularly in the media, often use data from a recent period, such as the recent year-over-year change—or the change since a recently enacted reform. Although there is obvious value in examining recent changes, it is also important to reflect on the long-term trends, which provide context to the recent change and reveal the enormous growth of incarceration over time. Trends in these fact sheets look back to 1970 or 1983, depending on data availability.
  2. We present historical trends using the number of people incarcerated rather than incarceration rates. Statistics on historical incarceration trends typically feature the rate of incarceration. The thinking is that—just as you cannot meaningfully compare the total number of people in prison in California to the number of people in prison in Oregon, for example—using incarceration rates can account for differences in the size of the U.S. population as it has grown over time. We feel, however, that there are good reasons to take a different approach. First, the daily number of people in prison and jail is commonly used in contemporary analyses of jail and prison populations—and we believe there is a benefit to creating a document that aligns with the public’s understanding of the number of people in prison and jail today. Second, using the rate can lead to the false impression that a state with a growing resident population but stagnant incarcerated population, such as Texas, is decarcerating.
  3. We present incarceration metrics relative to each state’s neighbors. State rankings are generally used to create 50-state lists. This has intuitive appeal, but we decided to instead rank states among their own neighbors. (See this whitepaper for complete details of the methodology.) States vary widely on economic, social, and political dimensions, so the most relevant comparisons are usually with neighboring states. To be sure, policy goals should never be limited by one’s closest neighbors—but we find that when you look at the data from a more local perspective, it reveals differences that may go underappreciated in a 50-state framework.
  4. We present long-term trends for people in the custody of the state prison system alongside the local jail trends. Statistics on prison populations nearly always center on the number of people in the “jurisdiction” of the state prison authority, which includes both people in the “custody” of the state prison system (held in a state prison), as well as people held in private prisons and local jails because there is no room for them in a state prison. When presenting the long-term prison population trend on page one of the fact sheets, we use the number of people in the custody of the state prison system (plus those in private prisons). We do this in order not to double-count people in local jails who are being held for the state prisons, which is common in states such as Kentucky, Louisiana, Tennessee, and Utah, among several others. Furthermore, where people are serving their sentences is significant because sentencing reform is not true reform if it simply moves people from prison to jail.
  5. We prominently feature county-level jail and prison admission rates. We display these county-level rates to illustrate to the reader that, generally, it is the smallest counties that have the highest rates of jail and prison incarceration. This means that incarceration is now most prevalent, on a per capita basis, in smaller cities and rural counties. (Up until the 1990s, incarceration rates had been the highest in America’s largest cities.) Presenting county-level jail admission rates, in particular, might be subject to controversy because some will argue that jail admissions, even on a per capita basis, are not comparable because some jails hold people for the state prison system or U.S. Immigration and Customs Enforcement. The available data does not disentangle “local,” “state,” and “federal” bookings—so making such a distinction is not possible. But there is also a risk to parsing out this distinction, because a person booked into a jail is a person booked into a jail—and, if we are to make headway at addressing the 10.6 million annual jail bookings, then it is critical that we have clarity on where these booking actually occur.
  6. We parse out the incarceration population that is “jail-like” in the states with unified corrections systems. Six states—Alaska, Connecticut, Delaware, Hawaii, Rhode Island, and Vermont—do not have local (county- or city-operated) jails and instead operate “unified” corrections systems in which the state administers facilities that hold people with both pretrial and sentenced statuses. These states are more difficult to compare with others and are thus sometimes omitted from the national conversation on jails and excluded from state comparisons. Fortunately, the Bureau of Justice Statistics’ National Prisoner Statistics data series disaggregates people by pretrial and sentenced status and provides information on the number of people with sentences of one year or less. We use this data to identify people in unified state systems with sentences of one year of less who would generally be sentenced to a jail in another state, as well as people with pretrial status, and we designate these as people in “jail.”

The data we used to create these fact sheets, along with detailed methodology whitepapers and a codebook, are maintained on Vera’s Github page. These fact sheets were created by directly linking Vera’s datasets to the Adobe Illustrator file that produced the designed reports, so any errors are due to mistakes state and local agencies made when submitting their data to the Bureau of Justice Statistics, which is the source of the data in these fact sheets.

We hope that you find these to be useful resources for your advocacy, research, and policymaking. If you are interested in the data for a particular county, rather than the state as a whole, you can find this detail on Vera’s Incarceration Trends data tool.