Expanding our Knowledge on Local Incarceration Trends

Christian Henrichson Former Director, Vera Insights // Jacob Kang-Brown Senior Research Fellow // Oliver Hinds Former Senior Data Scientist
Dec 13, 2018

Below are two perspectives today on the public release of the Incarceration Trends dataset, one from Vera Research Director Christian Henrichson and one from Senior Research Associate Jacob Kang Brown and Senior Data Scientist Oliver Hinds.

Photo by Jack Norton

Take a Look at Mass Incarceration in Your Own Backyard

By Christian Henrichson

The state of incarceration in America—2.2 million people behind bars—is such a massive problem that it can feel abstract and distant.

But it’s actually quite tangible, proximate, and solvable. And as Vera’s Incarceration Trends tool shows, it’s a problem that’s actually in all of our backyards—the place where the solutions also lie because it is city and county officials—such as police, prosecutors, and judges—who decide who and how to arrest, prosecute, and sentence. All people that are directly accountable to their local constituents.

To show how the tool can be used, let’s look at Broome County, NY, which lies in the state’s Southern Tier, bordering Pennsylvania, and west of the Catskill Mountains. The county is the birthplace of IBM and was a hub of defense industry during the Cold War. It’s also home to Binghamton University, a prominent state university that is now at the center of the region’s economic and cultural life in the wake of the declining industrial sector.

About 200,000 people live in Broome County. In 2015, 722 of the 50,000 people in the New York state prison system were sentenced from Broome County. 722 hardly seems like a lot when compared with the 22,301 people in the prison system who were sentenced from New York City. But a comparison by volume fails to capture the way that incarceration affects communities of different sizes—the right way to think about incarceration is using the rate per capita. The prison incarceration rate in Broome County is now 45 percent higher than in New York City (551 per 100,000 residents aged 15-64 in Broome, versus 380 per 100,000 residents of the same age in NYC).

The same trend can be observed in the local jail. In 2015, the 467 people in the Broome County jail were a mere fraction of the 9,531 in New York City. But once again, that’s not the right comparison. The 467 people in the Broome County jail equated to an incarceration rate that was 120 percent higher than in New York City (357 per 100,000 residents aged 15-64 in Broome, versus 162 per 100,000 residents of the same age in NYC).

The trends in Broome County are not dissimilar from other mid-sized upstate counties. Jail and prison incarceration rates have taken nearly the same path in Onondaga County (Syracuse) and Monroe County (Rochester).

Jail: Incarceration’s “front door”

The first phase of Vera’s Incarceration Trends project focused on jail incarceration precisely because unwinding a place’s use of confinement needs to start with incarceration’s “front door”—the local jail. In Broome County, an average of 80 percent of people in the jail on any given day are being held pretrial, meaning that they are behind bars while still legally presumed innocent, awaiting the resolution of their case. At the same time, the reach of the jail is grossly understated by these average daily population counts. There are nearly 6,000 jail bookings each year in Broome County—more than four times as many as New York City on a per capita basis. Another way to contextualize the scale of jail admissions in Broome County is to compare the number booked into the jail with the number sent to prison. In 2015, Broome County sentenced 423 people to prison during the same year that nearly 6,000 bookings into the jail were recorded. This begs the question: if the county meted out 423 prison sentences, why was the jail used more than 5,500 other times? Nationally, jail admissions are 17 times higher than prison admissions, revealing the scale of impact that jails can have on communities.

Unpacking Racial Disparity

Prison incarceration throws the racially disparate impact of the Broome County justice system into sharp relief. In 2015, 347 black people and 301 white people from Broome County were incarcerated in New York state prisons. From these numbers alone, it’s evident the incarceration rates are highly skewed: there are fewer white people from Broome County in prison than black people—even through the majority of the people in the county are white. In fact, the prison incarceration rate of black people from Broome County is more than ten times the rate of the white population—3,986 per 100,000 black residents aged 15-64, versus 312 per 100,000 white residents of the same age. Nationwide, the rate of black jail incarceration is 3.6 times the rate of white incarceration, whereas the rate of black prison incarceration is more than five times higher. So the racial disparity in Broome County is not just high, it’s an outlier. Put another way, the number of black people in prison from Broome County is equivalent to 3.9 percent of the county’s total black population, whereas only 0.3 percent of the county’s total white population is incarcerated in the state prison system.

Data drives change

The incarceration statistics in Broome County may be a surprise to those outside of the region, but they are not surprising to the people that live there. Sociologists at Binghamton University have been writing about upstate incarceration for years. And there are well organized local efforts led by groups such as Justice and Unity for the Southern Tier (JUST) and Progressive Leaders of Tomorrow (PLOT) that have been broadening awareness about upstate incarceration and developing partnerships to drive reform.

Whenever we talk about the high, and rising, rate of incarceration outside of big cities—we are almost always asked “what’s driving this trend?” There are many reasons, and they vary by place; our research has profiled several, including the impact of criminal justice fines and fees in Montgomery County, NY, the devolution of prison sentences to local jails in Indiana, or immigrant detention in Glades County, FL and Baker County, FL.

But I think that there is another important factor. Policymakers and the public, especially in smaller cities and towns, simply don’t know how much their use of incarceration has increased, or how outsized it now is relative to their population. In a sense, they are driving without a speedometer. It is our hope that better access to data will spark more conversations like the ones happening around Binghamton and countless other places across the country.

So we’ve tried to create the tools to spark those conversations. In addition to an expanded set of metrics that includes the number of people from each county that are in state prison, Incarceration Trends now has functionality to download an image file of the chart you create, or even an Excel file that includes the data behind the chart (just click the “menu” in the upper-right corner of the chart). There is a lot of data to be explored, comparisons to be made, and questions to ask and answer. And all of that work is best done by the people that know their communities the best.

Take a look at mass incarceration in your backyard, and let everyone know what you see.

Huerfano County Correctional Facility, Colorado (now closed). Photo by Jack Norton

Opening up Local Incarceration Data

Incarceration Trends Dataset: County-level jail data (1970-2015) and prison data (1983-2015)

By Jacob Kang-Brown and Oliver Hinds

Basic criminal justice data—such as the number of people in U.S. jails and prisons—is hard to come by. Given the decades-long, intense growth of incarceration across the United States—a phenomenon that has had long-lasting and negative implications for families, communities, and the nation—it is urgent to get better information in order to reverse this trend. To address this problem, we have created a public dataset on local incarceration to encourage research on mass incarceration with detailed, county-level prison and jail data.

The Bureau of Justice Statistics (BJS) collects an estimate on the number of people who are incarcerated in jails and prisons every year. However, these statistics have not previously been combined to tell us how many people are being sent to jail and prison from each county in the United States. Further, research on incarceration has traditionally ignored the jail data and centered on state-level prison population data.

To close this gap, Vera researchers are making public the Incarceration Trends Dataset—including county-level jail data from 1970 to 2015 as well as the recently constructed county-level prison admissions and incarceration data from 1983 to 2015. We will continue to maintain and update this dataset as future data becomes available.

When we first began linking jail data together we realized that the archives were richer than expected, and included detailed information dating back to 1970. But the data were messy, and sometimes included clear errors. We worked to clean up these issues and develop estimated numbers for counties that did not report regularly to BJS, in order to provide the most complete picture possible of jail incarceration across the country. When examining prison data, we found similar errors, and also noticed that many states have public data on prison incarceration by county that can strengthen BJS data.


Using the state as the unit of analysis is sufficient for understanding the broad contours of incarceration in the United States, but it does not provide the level of detail necessary to unpack its causes and consequences. This is because it is largely county officials—judges, prosecutors, people who manage jails—that decide how communities use incarceration (i.e., who is sent to jail and prison, and for how long). Therefore, county-level variability makes for more robust, theoretically-grounded studies of the high rates of incarceration seen across the United States.

The Incarceration Trends project, our web data tool available at http://trends.vera.org, maps incarceration across the United States and lets you dig in to the details of historical jail and prison populations in particular counties and states. By making the Incarceration Trends Dataset used to build this tool available, our hope is that researchers, advocates, activists, county officials, journalists, and the public can use the raw dataset and the trends tool to understand how incarceration is used in their community. Without this understanding, we will never be able to unwind the hugely destructive social problem called mass incarceration.

In October 2018, we hosted a symposium to bring researchers working on these questions together with early access to the data files. Stay tuned for another call for papers in Spring 2019.

Download data from Github here.