ATLANTA -- When the U.S. Census is tallied, prison inmates are included in the count for the town nearest to the prison where they are being held -- not the communities where they once lived.
The practice known as prison gerrymandering has gone unchanged since the first census more than 200 years ago but with prison populations increasing, state and local officials are starting to question the federal judgment.
Critics say redrawing voting districts to include prison populations uses residents who can't vote as a way to gain political pull both on state and local levels. Since the first U.S. Census in 1790, the policy has been that prisoners count toward the districts where they are held.
The policy wasn't recognized as a problem until recent decades, when the incarcerated populations in the U.S. climbed to exorbitant levels. In many cases, rural, predominantly white communities benefit from the presence of a predominantly minority prison population.
Some municipalities where prisons are located, including some Georgia cities and counties, have made the decision to exclude the incarcerated population when redistricting. But while policy advocates say districts are trending toward excluding prison populations in their counts, with no rule against it, there's nothing stopping the incarcerated populations from being counted toward communities that are not their own.
"Now that prison populations are so large, what was previously an innocuous quirk in the census," Wanda Bertram, communications strategist at the Prison Policy Initiative, said, "has become a relatively serious threat to fair electoral representation."
A problem of representation
Lobbyists for including prison populations into census counts support the decision with the argument of general "wear and tear" on the community infrastructure by inmates -- creating an extra strain on things such as roads or hospitals. In this thought, inmates are still residents of an area despite their disenfranchisement.
"Many of them see the people that are in prisons still somehow as counting as residents even though they cannot vote and have no ability to participate in civic life," Bertram said, "In all likelihood, they have much more of a stake in politics in their hometown's politics than the prison towns."
While it makes sense, she said, that argument could better be used to lobby the governments for additional funding and grant money, not an argument to miscount a person who can't vote.
The Prison Policy Initiative is a nonprofit, non-partisan advocacy campaign against mass incarceration. Prison gerrymandering is not a party issue, Betram said, even though at the state level it is often portrayed as one.
"Prison gerrymandering is really an issue that impacts everyone regardless of political leaning, regardless of race," she said.
Based on a Prison Policy Initiative, analysis of federal grant programs determined by census counts, Bertram said the algorithms are a little "too smart to be fooled by prison gerrymandering."
The problem then? Skewed political representation.
"Prison gerrymandering is a basic problem of fairness," Bertram said, not just for people in communities that are over policed and send a large portion of the residents to prison but also for people that live in the vicinity of the prison."
For state officials, no answers
Georgia houses more than 52,000 inmates across 34 state prisons and more than 55,000 inmates total as of December -- including those in private facilities and county correctional institutions. Of the total number of incarcerated individuals, more than 60% identify as black or Hispanic.
Like the rest of the nation, Georgia has seen a heavy increase of incarcerated individuals. A 2018 report from the Georgia Council on Criminal Justice Reform stated the state's adult prison population more than doubled from 1990 to 2011. Numbers fluctuated until they reached what it is today.
When prisons close and open, the problem can be exacerbated. The state Department of Corrections said tracking where inmates were moved when facility changes happen would be nearly impossible.
"The fact it is so hard to track is part of why we need the census to make it easier on local governments," Bertram said, "to actually figure out how many people are incarcerated who are in our districts and how do we exclude those people; it's not fair to make ... local government do that on their own."
With no law governing counties and municipalities, state officials can only suggest they consider removing prison populations from their census counts.
Gina Wright, executive director of the state Legislative and Congressional Reapportionment Office, said there's not much they can do. The office can simply point out the prison population size, break down how much of the total district it makes-up and explain how those individuals cannot vote, she said.
"We inform local governments, and we let the local governments make the decision on whether or not to include that population or not," she said.
Including the prison population is not helpful for redistricting purposes, she said, one district will end up significantly smaller in terms of the voting population and even the people who are going to be able to participate in government and activities.
The complicated logistics of reviewing prison populations, county by county, makes filing census materials an individualized task. A note is added to paperwork submitted to the federal government that a prison population was removed -- removing the entire block that houses the prison could exclude people living around it.
The impact of including a prison population is completely dependent on the size of the prison and the size of the district, she said.
State officials and advocates agree, counties have been leaning toward excluding prison populations from their counts.
"We're letting local governments know this is a concern that you need to consider," Wright said. "Usually once you show the why, they've always been pretty easy and understanding. I don't think there's been too many that once they really understood and say how large of a number it would be -- especially in rural counties -- I don't remember there being much debate about it."
As awareness of the issue rises, Bertram said, more cities and counties are jumping on board to exclude prison populations during redistricting.
Several Georgia cities and counties avoided prison-based gerrymandering after the 2000 census by excluding prison populations from their counts when drawing city or county lines.
The cities of Garden City and Milledgeville and the counties of Calhoun, Dooly, Macon, Tattnall, Telfair and Wilcox were among those, according to the Prison Policy Institute. Even more excluded prison populations from redistricting legislation after 2010.
"For all the Census Bureau has disappointed us in its refusal to change how it counts incarcerated people," Bertram said. "We've been very encouraged by the people who are in local government, who recognize this as a problem and are taking action on their own."
'One person, one vote'
Nancy Abudu, deputy legal director of the Southern Poverty Law Center, litigated a 2016 case in Florida where prison-based gerrymandering was deemed unconstitutional by a federal judge.
The case in Jefferson County was ruled on the "one person, one vote" principle. Abudu, who argued the case on behalf of the American Civil Liberties Union, said the county purposefully included the prison population which then made up 42% of the voting age population in the district.
According to the ACLU, of the 1,200 inmates in the correctional center involved, only nine were convicted in Jefferson County.
Elected officials did not do anything to gain the political support of the inmates, Abudu told CNHI, let alone their votes.
"With the racial disparities in the criminal justice system, coupled with the fact that folks can't vote for the most part, except in a couple of states while they are incarcerated," she said. "It all makes for what has been called a ghost population."
Often rural, low-income areas where elected officials have benefited from prison-based gerrymandering, benefit from the district looking more racially diverse than it really is.
"Residents are just seeing these dollar signs coming in but not really understanding the full social and economic impact," Abudu said. "Especially when you have predominantly white communities that are taking advantage of the racial disparities in the prison system to make it look like they have much more racially diverse communities than they really do."
But when doling out dollars for minority development programs, she said, it wouldn't take federal officials long to recognize the numbers are actually "fudged."
Rule in standstill
In 2015 and again in 2016, the U.S. Census Bureau sought public comment on how prison populations should be counted in the census.
Of the 77,887 comments relating to the question in the 2016 responses, more than 99% suggested "prisoners should be counted at their home or pre-incarceration address."
But the bureau ruled out the change.
"Counting prisoners anywhere other than the facility would be less consistent with the concept of usual residence, since the majority of people in prisons live and sleep most of the time at the prison," the bureau said.
The policy standstill is not because people aren't speaking up, Betram said; it concerns a lot of people.
"I think that to a large degree it comes down to the fact that the census has been chronically underfunded -- as so many agencies right now -- suffers from a lack of stable staffing and leadership," she said. "Compared to other issues that the census has to deal with, frankly, I think it doesn't consider the counting of people in prison as very important."
A handful of states have enacted legislation to eliminate prison-gerrymandering -- Nevada and Washington the most recent to make the change requiring prisoner counts from where they're incarcerated to their last known home address for redistricting.
In 2011, a group of Georgia Democrats pushed legislation that would have excluded certain prison populations from redistricting counts, but the bill did not see a committee hearing.
Last month, the governor of New Jersey signed a bill ending prison gerrymandering -- now counting incarcerated inmates at their home addresses.
Change at the top level would take a federal mandate, which is why at the federal level, there has been so little movement, Abudu said, and yet at the local level lobbying has seen some success.
"The goal is to count everyone, it's not like they disappear off the face of the planet," Abudu said. "So, either they're not considered at all and those lines are drawn based on a population that doesn't include the prisoners, or they're counted where they're from which shouldn't be too difficult."
Riley Bunch covers the Georgia Statehouse for CNHI's newspapers and websites.