November's elections will be the first time voters in Wilmington, Delaware, cast their ballots in city council districts that aren't distorted by the location of prisons.

That's because the city decided after the last census to abolish a long-standing practice sometimes known as "prison gerrymandering," in which people in prison are counted as residents of the places where they're incarcerated rather than where they're actually from.

"It allows for one district to have way more people than what it really should have," said Wilmington Councilmember Shané N. Darby, who championed the city's new approach. "I want to have a redistricting process that is equitable."

Under the new law, people incarcerated at state correctional facilities located in Wilmington were counted based on their last known address in the city, rather than as residents of the city council district where the facilities are located. Those who last lived outside of the city were not counted at all.

More than a dozen states and scores of municipalities like Wilmington have made moves on their own in recent years to end prison gerrymandering. But the U.S. Census Bureau has resisted pressure to do the same. So activists and lawmakers are pushing to ensure that more state and local governments make the change themselves in time for the next redistricting process, just six years away.

"The longer that the Census Bureau drags its feet holding onto its misguided and outdated way of counting incarcerated people, the more difficulty they're creating for cities and states who recognize this is a problem," Prison Policy Initiative Communications Director Mike Wessler told Law360. "And those cities and counties have to clean up the mess."


"Unnervingly Reminiscent"

Prison gerrymandering is a result of the Census Bureau's "usual residence" rule, under which the government counts people as living in the place where they primarily eat and sleep on the day the once-per-decade census is conducted. As a result, people in prison are considered residents of the district in which they are incarcerated, rather than where they lived before being locked up.

Since political representation is based on population, the practice siphons political power away from the largely urban and minority communities that most incarcerated people come from, and inflates the political power of the mainly white, rural areas where prisons tend to be located, according to activists fighting to end the practice.

"Black and brown communities are unquestionably some of the most harmed communities by prison gerrymandering, the communities that are losing the most political clout," Wessler said.

That political clout winds up in places like Jackson County, Wisconsin, home to the Jackson Correctional Institution. Redistricting that followed the 2010 census left Jackson County with a county commission district where people in prison accounted for 51% of the population — a majority that can't vote and the members of which aren't really considered constituents by the district's representatives.

"Most lawmakers who represent prison communities say, 'I don't think of those people as my constituents. I've never held a town hall meeting there. I've never asked them what their concerns are. I've never met most of them,'" Wessler said.

As a result, the remaining 49% of district residents who aren't incarcerated wind up with the same political power as 100% of the people in districts without prisons, according to Wessler.

"It reduces everyone's political clout a little bit and gives certain people, just because of where they live, a lot greater [of a] voice in government," he said.

And Jackson County isn't unique. After the 2010 census, people incarcerated in Adams County, Wisconsin, accounted for 59% of the population of one Board of Commissioners district and 63% of another, according to the Prison Policy Initiative. Meanwhile, a prison accounted for 87% of the population of a county commissioner district in Lake County, Tennessee, as of 2021.


That distortion has consequences for voting and political representation, according to Ismar Volić, chair of the math department at Wellesley College and director of the Institute for Mathematics and Democracy.

Given the prison system's racial disparities, Volić called the practice of prison gerrymandering "unnervingly reminiscent" of the three-fifths compromise, which allowed early states to count slaves as three-fifths of a person for census purposes.

If Black people who lived in Philadelphia before being incarcerated had been counted as residents of the city rather than of the places where they were imprisoned during 2011's redistricting, for instance, Pennsylvania would have had two more majority-minority state House districts, Volić explained.

That lost representation is then cemented for a decade until the next census, even though most incarcerated people will be released from prison long before that, advocates point out. Then, when people are released from prison, they almost invariably return to their home communities, which now lack political representation, exacerbating already existing racial inequalities.

"The discriminatory practice thus continues to be in effect long past the period of incarceration," Volić said.

Gaining Traction

Activists and lawmakers are working to change how states count people in prison.

New York and Maryland were the first to implement that change when they drew new maps in 2011. Since then, more than a dozen other states have followed suit, according to Ben Williams, associate director of the elections and redistricting program at the National Conference of State Legislatures.

In May, Minnesota passed a bill making it the latest state to end prison gerrymandering, though the new policy won't affect redistricting until after the 2030 census, according to Democratic State Rep. Esther Agbaje, the legislation's primary author.

"We wanted to make sure that we're treating our prisoners with respect, that they're not being discounted or not counted or just counted in the wrong places," Agbaje said.

At least 200 cities and counties have addressed the issue in their local redistricting efforts as well, according to the Prison Policy Initiative.

That gathering momentum is the result of a renewed public focus on gerrymandering in general, combined with recent years' emphasis on criminal justice reform, experts say.

"This is one aspect of redistricting that has historically not gotten a lot of attention. There are advocacy groups that are trying to push this as a policy, and they clearly have had some traction," Williams said.

"I'm not going to pretend like it's the biggest topic in gerrymandering," Williams acknowledged, "but it's something that has risen in public salience."

Education, Data and Opposition

But that success doesn't mean there aren't obstacles to changing how people in prison are counted.

For starters, activists say, legislators and the public need to be educated about the impacts of prison gerrymandering.

"There's a lot of misconceptions around the whole census process," said Kim Murphy-Kovalick, programs director at Voters Not Politicians, which is working to end prison gerrymandering in Michigan.

For instance, many legislators believe that removing incarcerated people from districts' population counts will lead to those districts getting fewer federal dollars, according to advocates like Murphy-Kovalick. But that isn't the case, since the allocation of federal money to the states is a lot more nuanced and is based on far more than just population numbers, advocates say.

The census data itself presents another challenge.

The Census Bureau isn't planning to alter how it counts people in prison anytime soon, though federal legislation to require it to make that change has been introduced in Congress.

"The Census Bureau has historically counted prisoners at the correctional facility. There have been no recent changes to where the incarcerated population is counted," a spokesperson for the Census Bureau told Law360 in a statement.

That means states, which rely on the bureau's data to draw legislative maps, have to adjust that data themselves using records from state corrections and prisons departments if they want to draw districts that count incarcerated people as part of their home communities. But the information provided by those state agencies is often inconsistent or incomplete, according to Williams.

Files for incarcerated people can contain incorrect or fragmentary last-known-address information, particularly if a person was intoxicated, unhoused or didn't speak English when arrested, Williams explained.

In fact, 12 of the 13 states that ended prison gerrymandering in time for the 2020 redistricting cycle reported that problems with last-known-address information was a major issue when the National Conference of State Legislatures later surveyed the states for a report on their efforts to reallocate that data.

Address information is also handled differently at private prisons than it is at state-run facilities — and some corrections departments don't track inmate address information at all, according to the NCSL's report.

The time needed to address those data issues is why states like Illinois, which passed a law ending prison gerrymandering in 2021, decided not to implement the new policy until after the 2030 census, Williams explained.

"All of that needs to be developed, and it needs to be implemented, and that could take a few years," Murphy-Kovalick said about the legislation she's supporting, "which is why we're pretty keen on getting this passed as soon as possible, so that there is an implementation deadline that is well before the census."

But that effort in Michigan has run into another roadblock to change: political opposition.

Despite what Wessler calls "a lot of bipartisan support" for ending prison gerrymandering in some states, efforts in other states have faced partisan pushback, according to lawmakers.

Before succeeding this year, Minnesota Democrats had attempted to pass legislation changing how the state counts people in prison before both the 2010 and 2020 censuses. Those efforts failed, largely due to Republican control of part of state government at the time, according to Agbaje. Only after Democrats gained control of the governor's mansion and both houses of the state Legislature were they able to enact the law, which was passed along strict party lines and received only Democratic support, Agbaje said.

Democratic Michigan State Sen. Sylvia Santana, who is currently sponsoring the legislation in that state, first introduced her bill in 2017. That effort likewise failed due to Republican opposition, she said.

"A lot of our Republican colleagues do live in those rural areas where the prisons are located, and so they directly benefit from that," Santana explained.

Murphy-Kovalick was slightly more blunt.

"There's not a lot of representatives or senators out there who are saying, 'I want my district to have less power, less influence,'" she said.

One Person, One Vote

So some advocates looking to end prison gerrymandering have turned to the courts.

After multiple bills abolishing prison gerrymandering failed to pass in the Connecticut Legislature, the NAACP, which had supported those bills, sued the state in 2018.

The state's maps, drawn using prison gerrymandering, violated the "one person, one vote" principle of the 14th Amendment, said Yale Law School professor Michael J. Wishnie, who represented the plaintiffs in the case.

The lawsuit, NAACP v. Merrill, was the first legal challenge to a state's prison gerrymandering, according to Wishnie, who said there had previously been a small number of lawsuits over how local governments counted people in prison.

Connecticut cited one of those previous lawsuits when seeking to dismiss the NAACP's case: In Davidson v. City of Cranston, Rhode Island, a federal appeals court ruled that decisions about how to count incarcerated people in the population base are inherently political judgments in which courts shouldn't interfere.

"Plaintiffs' analysis invites federal courts to engage in what have long been recognized as paradigmatically political decisions, best left to local officials, about the inclusion of various categories of residents in the apportionment process," a three-judge First Circuit panel wrote in 2016 when reversing a district court's grant of summary judgment to the plaintiffs in Cranston.

"We decline that invitation," the panel added.

But the Connecticut case was different from the Cranston case because it challenged a state's use of prison gerrymandering, rather than a city's, Wishnie explained.

"The relative power under state law of municipalities and states made these very different cases," Wishnie said. "We also think the First Circuit got it wrong in Cranston."

The Connecticut lawsuit was eventually dismissed without reaching its merits after delays caused by the COVID-19 pandemic made it impossible to resolve the case in time for the 2020 election, according to Wishnie.

But that doesn't mean the legal challenge was unsuccessful. In 2021, Connecticut passed legislation ending prison gerrymandering in the state. And according to Wishnie, the Merrill lawsuit played a substantial role in the passage of the bill.

"I have no doubt that the possibility of renewed litigation was a meaningful factor in the Legislature finally after five or six previous efforts adopting legislation that did end prison gerrymandering," he said.

Whether continued progress in changing how people in prison are counted comes from litigation or legislation, it's likely that even more states and cities will end prison gerrymandering in time for the 2030 census, according to experts.

"We think the Census Bureau should change how it counts incarcerated people," Wessler said. "However, states increasingly aren't waiting for the Census Bureau to take action."

--Editing by Alanna Weissman. Graphics by Ben Jay.

Have a story idea for Access to Justice? Reach us at accesstojustice@law360.com.