Fixing the unfairness of prison gerrymandering

The South is home to several states with high rates of incarceration and criminal justice systems rooted in slavery and Jim Crow. People of color are disproportionately incarcerated across the region, as they are across the United States, and harsh sentencing laws have inflated the prison population exponentially over the past several decades. Though efforts to address mass incarceration have helped to decrease the U.S. prison population since it peaked in 2009, the U.S. still incarcerates 2.27 million people, and has one of the highest incarceration rates in the world.

That has had profound effects on U.S. democracy, from limiting access to the ballot box to shaping voting districts for federal, state and local elections. The 2020 census will once again count incarcerated people, most of whom do not have the right to vote, as residents of the communities in which they are imprisoned rather than as residents of their home communities — a practice known as "prison gerrymandering." This gives an advantage in representation to communities where prisons are located.

"Only in the last four decades have you built these concentrated prisons, where you can really upset a district line" by using prison populations in the count, said Bruce Reilly, the deputy director of Voice of the Experienced (VOTE), a grassroots organization that advocates against mass incarceration in Louisiana. "With redistricting generally, and with things like the Electoral College, [you get] questions of whether politicians represent land masses or humans. I think our democracy depends on some of these basics about what does 'one person, one vote' mean?"

Prison gerrymandering has important implications for representative democracy. Counting prisoners as residents of prisons, rather than as residents of their home communities, tilts population counts in favor of the whiter, more rural areas where prisons tend to be located, and away from communities of color, often in more urban areas where incarceration rates have historically been much higher.

"It unjustly awards extra political power to the regions that are the host of these prisons," said Jeri Green, a senior advisor for the census at the National Urban League and the Census Bureau's former senior advisor for civic engagement. "And brown and black people are the pawns in this whole situation — doubly the pawns."

Prison gerrymandering is yet another way the system is stacked against people in the criminal justice system as well as against their families, neighbors, and communities. Though prison gerrymandering doesn't affect the distribution of state and federal funding, according to the Prison Policy Initiative (PPI), it can have a deep impact on representative institutions.

Take for example Allen Parish in Louisiana, the state with the highest incarceration rate. Because the parish did not exclude incarcerated people from its redistricting maps following the 2010 census, two of its county government districts included large prisons, with prisoners making up 66 percent of the population of District 1 and 39 percent of District 6, according to PPI. Incarcerated people are counted in the district's population, and each district must have approximately the same population. But the incarcerated cannot vote — and so voters in District 1 make up just 44 percent of the district. This gives each voter in these districts disproportionate sway relative to voters in districts that do not house incarcerated people; a voter in District 1 has as much weight as three voters in any other district except District 6.

The effects of prison gerrymandering are more apparent in local elections because of how small the districts can be on a local level. "The presence of a thousand people [in prison] impacts the clout of a district that's smaller way more than if it's larger," Wanda Bertram, a spokesperson for the Prison Policy Initiative, which advocates against prison gerrymandering, said. “State legislative districts tend to be larger than, for instance, the school board."

Some states have recognized this disparity and are taking steps to ameliorate it. Two Southern states, Tennessee and Virginia, have passed laws explicitly allowing counties and cities to discount prison populations when drawing district lines; these laws were prompted by the need to avoid creating state legislative or local city council districts that would be composed entirely of a prison and no eligible voters. In other states, including Arkansas, Florida, Georgia, Kentucky, Louisiana, and Texas, bills to address prison-based gerrymandering have been filed in current or previous sessions but not passed.

But the laws that have been enacted in Tennessee and Virginia are piecemeal in nature and do not require counties to take action to avoid prison gerrymandering unless they want to. Localities that choose not to take prison population distortion into account when drawing local election districts generally don't have to. And that is a blow to the political power of black, Native, and Latinx Southerners in particular — populations whose voting rights are already under attack.

"You've got prison-based gerrymandering that removes, and then you've got the good old boy gerrymandering that plagues the South in particular," said Green.  

How prison gerrymandering persists

Prisoners are counted differently than other transient populations, including long-haul truck drivers, deployed military personnel, and boarding school students, all of whom are counted in their home communities. Meanwhile, the county commissioner or state legislator who represents a district with a prison likely won't think of the people incarcerated there as constituents. And because those prisoners usually can't vote, they have no say in who those representatives are.

"In a prison district, when you have an artificial number of people who are just kind of phantom, it's kind of like the three-fifths slave clause, right?" said Reilly. "'My slave counts for three-fifths humans,' and bulks up my representation. Really? Are you really representing those people? You're holding them in bondage."

Though prisons aren't built with gerrymandering in mind, Bertram said, representatives who are elected in distorted districts are loathe to give up the bump they get from counting prisoners. "They often don't want to let that go, for the same reason they don't like any kind of redistricting that impacts their chances of re-election," she said.

Democracy advocates have been pushing the Census Bureau to change its policies on counting incarcerated people for nearly two decades. They point to the fact that in many states, the definition of "permanent residence" does not include a prison or jail facility.

Some counties around the South removed prisons from their redistricting efforts after the 2010 count to avoid having districts composed mostly or entirely of incarcerated people. A 2002 opinion from the Mississippi attorney general directed that inmates should not be used for redistricting purposes, but some counties in the state still draw districts by counting prisoners. 

While states can pass laws changing the way incarcerated people are counted for state and local redistricting purposes, they cannot change the way census data is used at the federal level, including for congressional redistricting. That's up to the Census Bureau and the federal government. And it's controversial even there: During redistricting after the 2010 census, for example, Rep. Corinne Brown, a Florida Democrat, contended that by adding 7,000 prisoners to her newly-drawn district, the state legislature's proposed map diluted the votes of her district's black residents.

There have been attempts to remedy prison gerrymandering at the federal level. H.R. 1, the For the People Act that was introduced in Congress this year, included a provision to end prison-based gerrymandering alongside a slew of other pro-democracy reforms. The Democrat-controlled House passed the bill in March, but the Republican-controlled Senate is unlikely to bring it to a vote.

The Census Bureau itself considered changing its residence policy between the 2010 and 2020 censuses. After advocacy groups like PPI released detailed reports showing the unfair impact of prison gerrymandering, the Bureau seemed to understand the issue, according to Green, who worked there until 2017.

When the Bureau opened the comment period for the Proposed 2020 Census Residence Criteria and Residence Situations, it received 77,877 public comments addressing how the census counts prisoners. Just four of those were in favor of the current system of counting prisoners as residents of the communities where their correctional facilities are. The rest were in favor of counting the incarcerated as residents of their home communities. Yet the Bureau did not budge. Citing previous practice and its definition of "usual residence" as "the place where a person lives and sleeps most of the time," the Bureau said in 2018 that it would continue to count prisoners as residents of the district where they're incarcerated.

But because some states and localities are moving away from the practice of counting prisoners as residents of their prisons, the Census Bureau will produce a population tally that counts prisoners at their home addresses for states that request it. Democracy advocates say that's a step in the right direction, especially for states and municipalities that already take prison gerrymandering seriously, but doesn't go far enough.

"The census really is about power, representation, and money," Green said. And without a standardized method of counting the incarcerated, who are some of the most vulnerable U.S. citizens, power and money tend to win.