Census corrections show how Southern communities benefit from prisons

(Graphic from the Prison Policy Initiative.)

In January, the U.S. Census Bureau released its first round of corrections for the 2020 population count. Among these corrections were revised counts for municipalities in Southern states where geographic boundary-related errors occurred, misplacing prisons. 

In Arkansas, Georgia, and Tennessee, these misplaced prisons meant some towns saw an increase to original population counts by thousands of people. In the town of Whiteville, Tennessee, the census count jumped from 2,606 to more than 4,500 when the population of a correctional facility was counted back after previously being attributed to another area in 2020, according to Stateline, an initiative of The Pew Charitable Trusts.

For these corrections, the Census Bureau uses a process called the Count Question Resolution (CQR) operation, in which elected officials from tribal, state, and local governments can file a challenge and request a review of the census count for their jurisdiction if they suspect an error with boundary or housing counts. Annual population estimates can have significant political impacts in the form of redistricting and electoral representation. 

Across the U.S. today, nearly 2 million people are incarcerated, with most losing the right to vote at least temporarily under state felony disenfranchisement laws. When it comes to the census, the way the Bureau counts prisoners — in the places they are imprisoned as opposed to the communities where they lived before their incarceration — is part of a process some deem as "prison gerrymandering" for how it distorts political representation.

A number of states and local governments have taken some action towards addressing prison gerrymandering over the past few decades. While advocates for reform applaud the strides made, they also underscore a need for additional change to achieve what they see as equitable democratic representation. 

What is prison gerrymandering?

Since the first U.S. census in 1790, individuals have been counted at their "usual residence," defined as the place where a person sleeps and lives most of the time. Prisons fall under "group quarters," a term the Census Bureau attributes to places with relatively transient populations like college dorms, nursing homes, and military barracks. 

The issue of prison gerrymandering has existed since the census was first taken, but became more noticeable in the late 1970s and early 1980s alongside the explosion in prison populations, said Mike Wessler, communications director for Prison Policy Initiative (PPI), a nonprofit founded to document the impacts of mass incarceration.

At the local level, prison gerrymandering can result in what Wessler called "paralyzing impacts." He pointed to the infamous example of Anamosa, Iowa, where in 2008 a local man was elected to city council after receiving just two votes — both write-ins — from his wife and a neighbor. The man's ward had around the same population as neighboring areas, but it was home to about 1,300 individuals who were incarcerated within the state's largest penitentiary and thus unable to vote.  
 
"The residents who live closest to those prisons are getting disproportionate clout in political decisions, not just at the voting box," Wessler said. "Obviously prison gerrymandering is a problem that impacts voting, but we think of it as a problem that impacts representation more broadly." 

Critics of prison gerrymandering point to how counting incarcerated individuals where they are imprisoned as opposed to their home communities skews population counts towards the rural, typically whiter communities where state prisons tend to be concentrated. For example, a 2017 paper from the Center for Administrative Records Research and Applications found that a disproportionate share of prisons and inmates are situated in rural areas, while a disproportionate share of inmates hail from urban areas. 

At the same time, people from Black and Brown communities are incarcerated at disproportionately higher rates than their white counterparts. According to a 2021 report from The Sentencing Project, Black Americans are imprisoned in state facilities at almost 5 times the rate of white Americans, while the rate for Latinx individuals is 1.3 times that of white people.  

The rural, predominantly white town of Calico Rock, Arkansas, is another community that saw its population count more than double as a result of the first round of 2020 census corrections, increasing from 888 individuals to 1,815. Those corrections included adding back in the population of the North Central Unit prison, which has a capacity of 800 prisoners. While Black Arkansans make up around 15% of the state's total population, they constitute 41% of the state's overall prison population.

Similarly, the Georgia municipalities of Chester and Glennville had population gains of 161% and 38%, respectively, based on census appeals that moved state prisons back into their population counts. Unlike many communities that benefit from prison gerrymanders, Chester is majority-Black. 

Alison Wright, who heads up the Data Center Division with the Arkansas Economic Development Institute, noted that misplaced prison populations can have a significant impact on smaller municipalities. 

"Depending on what city you're in, they (prisons) do make a big difference if they don't get counted," Wright said. 

Mitchell Brown, voting rights senior counsel with the nonprofit Southern Coalition for Social Justice (SCSJ), sees education and advocacy as key to addressing prison gerrymandering. For example, SCSJ partnered with Democracy North Carolina and PPI on a 2021 letter-writing campaign encouraging lawmakers across the South to exclude prisoners from redistricting data. 

Under prison gerrymandering, incarcerated populations are treated "as pawns in an electoral game," Brown observed. 

"You have a group of people that are invisible on one hand but then used as a pawn on the other," he said. "That is all due to prison gerrymandering. And so the more people know about that, I think the more reform we can make."

Changes ahead

Across the country, Wessler says 16 states and over 200 local governments have taken some action on prison gerrymandering reform.

At the state level, Virginia passed legislation in 2020 that ensured prisoners would be counted at their last-known addresses as new legislative districts are drawn. The following year the Virginia Supreme Court rejected a challenge to that law brought by individuals including Virginia state Sen. Travis Hackworth (R), who argued that inmates should be counted as residents of the prisons where they are housed. Hackworth's district, population 184,289, is home to five state prisons: Bland Correctional Center (population 572 as of December 2022), Keen Mountain Correctional Center (910), Pocahontas State Correctional Center (947), Red Onion State Prison (728), and Wallens Ridge State Prison (1,004).

In Tennessee, lawmakers have passed legislation allowing counties to exclude prison populations in drawing district lines. And a number of local governments in Southern states including Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, and Texas have taken measures to avoid prison gerrymandering practices in some way, whether by "ignoring the prison population, cutting a hole in their maps around the prison, overpopulating the district with the prison by the exact amount of the prison population, or splitting the prison population between all districts equally," according to PPI.

"It's kind of a pivot point for the country and inflection point where the choices that state and local leaders, in particular, make will really decide whether or not we turn the page on the failed experiment of mass incarceration or continue down this path that harms communities of color, harms urban communities, and harms rural communities," Wessler said.

But Wessler sees the ultimate solution to prison gerrymandering as residing with the Census Bureau, saying the government agency should change how it counts incarcerated people. 

"Anything short of that is going to be kind of an imperfect solution," he said. "States are working hard, jumping through lots of hoops to make sure that their legislative lines align with true representation. But there's certain gaps that they can't fill."

David Ayala is the executive director of Formerly Incarcerated, Convicted People and Families Movement (FICPFM), a national network of more than 50 human and civil rights groups led by individuals with conviction histories and their families. Among the issues that FICPFM advocates for is the enfranchisement of convicted peoples and prison gerrymandering reform.

Ayala, who has served time in correctional facilities, said it's critical to prioritize prison gerrymandering as a criminal justice issue because it impacts re-entry for imprisoned people. "To count a person in an area that they will not be living in, it affects the resources for them that they can receive when they are released back into their communities," he said.

The CQR process does not mean the Census Bureau will collect new data or that the official 2020 census results will be updated with revised counts. However, the process does allow the Bureau to redistribute corrected population counts to government entities impacted by errors. 

Ahead of the 2020 count, the Census Bureau solicited feedback regarding its residence criteria and received close to 78,000 public comments, the majority of which suggested imprisoned people should be counted at their home addresses. In response, the Bureau stated in 2018 that counting prisoners anywhere else apart from their correctional facility wouldn't be consistent with the notion of usual residence, "since the majority of people in prisons live and sleep most of the time at the prison." 

But the Bureau also noted that some states may "move" incarcerated populations back to pre-incarceration addresses in legislative redistricting, and instituted a one-time process called the 2020 Post-Census Group Quarters Review (PCGQR) for governmental units to request reviews of 2020 population counts for group quarters. The deadline for submitting a challenge is June 30, 2023. 

The Census Bureau has stated that if the inconsistencies in 2020 population counts are found for a group quarters facility through the PCGQR, certified updates to the count will be supplied to the Population Estimates Program. These population estimates are used in critical statistical surveys like the American Community Survey, which various agencies and policymakers use for community planning purposes. 

Ultimately, when counting populations, it's crucial to consider the social ties a person has to a community, Wessler said. Once released, imprisoned individuals, though they may be used to pad population counts under the system of prison gerrymandering, aren't guaranteed to stay in or necessarily feel connected to a community where they were incarcerated.

"I don't think anyone would argue that somebody who's locked behind a prison wall is a member of that community that the prison happens to be in," Wessler said. "That's not where they consider home."