MILWAUKEE — Until a worker representing the U.S. Census Bureau showed up at his door last April, Sylvester Jackson felt like a ghost in his own community.
Jackson, 54, lives in Milwaukee. The last time census workers had come around, a decade earlier, he was 200 miles from home, serving a prison sentence in Black River Falls, Wisconsin. That’s where the census counted him as a resident the next 10 years.
It wasn’t a clerical error. Since its first count in 1790, the Census Bureau has counted incarcerated people as residents of wherever they’re incarcerated at the time of the census — as opposed to where they call home — even though in 48 of 50 states, they cannot vote in that community, or anywhere. In effect, this transfers each prisoners' vote to another voter in the district, undermining the "one person, one vote" principle of American democracy.
Census data is used to redraw congressional, state legislative and local voting district, as well as inform federal research, policymaking and funding allocations.
Each district is supposed to contain roughly the same number of voters for equal representation.
Jackson was convicted of sexual assault of a minor in 2007 and spent 10 years in the Wisconsin Correctional System. Because Jackson was in Jackson Correctional Institution on April 1, 2010, he counted among those "represented" by officials elected in Jackson County. However, in effect, his power — and the power of all the prisoners there — transferred to non-imprisoned people in the district. At the same time, the places where the prisoners came from and will return to live after serving their sentences lost representation.
Now on extended supervision and still unable to vote, Jackson has become an activist to stop counting prisoners where they are incarcerated. In the company of several legislators, he's spoken about the issue at the state Capitol and is an organizer with EXPO, which works on prisoner and ex-prisoner concerns.
“It’s kind of hard to say you’re being represented by anyone, because representation means someone representing your interests," he said. "The representation that should be allocated to you is given elsewhere.”
The result: Census data is used to give inflated political power to people who live in communities near prisons — often places that are predominantly rural and white.
“It’s using people who have no rights to give greater power to those people who are oppressing or incarcerating those people,” said Oscar Blayton, a civil-rights attorney and former combat pilot who lives in Virginia.
How prisons disperse people in Wisconsin
Use this map to see how prisoners in Milwaukee and Dane counties are moved related to their home ZIP codes.
Jared Knowles of Civilytics Consulting
Blayton learned about counting African-American population in third grade at the segregated school he attended: How as the new nation was forming, white Southerners proposed counting people they enslaved as part of their population, to pad their numbers and increase their political power.
Getting to count enslaved people as even a part of their population meant the South got more seats in Congress. Blayton said the same dynamic is at play in how incarcerated people are apportioned today.
He and other experts and policy groups refer to the policy as “prison-based gerrymandering.”
The ripple effects of how the census counts prisoners may have been subtler only a few decades ago, before new “tough on crime” and "truth in sentencing" policies in the 1980s and 1990s sent to prison tens of thousands of people — many of them Black and brown — and held them there longer.
Today, Black people are incarcerated at five times the rate of white people. Prisons remain disproportionately located in majority-white areas. As of the 2010 census there were 161 counties in the U.S. where the incarcerated Black population outnumbered the Black general population eligible to vote.
In recent years, prison-based gerrymandering has gotten more attention. But despite the 77,000 letters the Census Bureau received leading up to the 2020 count asking that prisoners be counted at their home address, about 2.3 million people were once again counted as residents of where they’re in jail or prison.
Some states have taken up calls to count prisoners differently. Since the 2010 census, 11 states — including New York, California and Illinois — have passed laws to count prisoners at their pre-incarceration address when they factor them into redistricting.
Other states require counties to omit prison populations altogether from their count, in an effort to adhere to the principle of one person, one vote. And more than 100 counties and municipalities in the U.S. — even though not required by law — have opted to do so when drawing their local districts.
The people drawing the lines in Wisconsin have largely ignored or refused to act on the question — even though the combined prison and local jail population is equivalent to about 80% of the adult population of a typical Assembly district. This means the state offers up some of the starkest examples of how political representation gets distorted when prisoners are counted as residents of where they’re incarcerated, instead of where they call home.
1,000 new residents — in prison
On a midsummer evening in 2011, John Wenum drove from his house to the Mauston County Courthouse. Wenum, a longtime county supervisor, was leading the redrawing of the county’s new maps. All around the state and the country, lawmakers and redistricting officials were doing what could sound like a straightforward job: break down a population into a certain number of districts, each with roughly the same amount of people in them.
Over in Madison, Wisconsin, the new Republican majority was drawing what would be one of the most controversial maps in state political history. But in Juneau County, Wenum’s biggest problem was a county supervisor or two hoping to lasso a block or two into their district. The process in past decades had typically gone off without a hitch.
Wenum and his colleagues had calculated that each county supervisory district needed to have 1,200 people in it, give or take.
But when Wenum took his seat in Mauston to begin drafting the new map, he and his team faced a big question: what to do about the 1,000 new residents who’d come to Juneau County since the last census count.
“They weren’t residents in the normal sense, they weren’t voting citizens in the normal sense. And it actually skewed things in some very significant ways,” Wenum said.
All these new residents had arrived at the same time and lived under the same roof: the prison that had opened that year, New Lisbon Correctional Institution in New Lisbon, Wisconsin.
Wenum said he looked at his options. He’d heard from a Massachusetts-based legal group, the Prison Policy Initiative, which urged him and his committee to consider excluding the 1,000 incarcerated people from the map. But Wenum said the only legal guidance he could find from the state was an opinion written by the state’s attorney general 30 years earlier, instructing him to do just the opposite.
“We had the attorney general’s opinion which was the only binding legal force in this state and we responded to it by saying we’ve got to count them in,” he said.
The opinion piece, however confusing, remains the clearest guidance on the matter In Wisconsin. So it’s what Wenum leaned on.
“The attitude here was ‘We’re not going to rock the boat, we’ll go with the flow,’ ” he said.
It felt wrong to hand 1,000 non-voting constituents to any one supervisor who’d been elected to represent people who lived in the county, Wenum said. But in the end, Wenum and his team felt that placing those 1,000 people into one district was their only choice.
Roy Granger grew up in New Lisbon, and got elected to the Juneau County Board of Supervisors two decades ago. When he took his oath of office, he pictured himself representing his neighbors, people he’d gone to high school with, folks he ran into at the hardware store or the holiday craft fair.
“It’s a small town, 1,500 people — now it's 2,500 people because of the prison,” Granger said. “But it's just small-town living and it’s neat.”
When the new state prison opened, the community celebrated it as a boon for the local economy. Granger went to the ribbon-cutting ceremony.
But it was only after Granger was sworn in as county supervisor that he learned who his constituents would be.
“So when it comes time for elections, I have like 250 eligible voters, or people … 250 residents that can vote if they’re of age,” Granger said. “And I’ve got a thousand residents that are locked up in that prison that can’t vote, but they still count in my district.”
Population demographics of Juneau County
Race breakdowns of Juneau County including freed and jailed adult populations. Explore the data .
Jared Knowles of Civilytics Consulting
That means that someone in Granger’s district has five times the political influence of someone in the next district over, or any district in the county without a prison. In other words, each eligible voter in Granger’s district has outsized influence because four other residents — inmates — can't vote, a distortion of one-person, one-vote.
The way the census counts those men, and the way Wenum and the county redistricting committee chose to include them in their district map, creates a kind of fiction about what the community there looks like.
Leonard Collins was incarcerated for homicide beginning in 1976 and got home to Milwaukee this past February. He took notice of something when he was in prison at Waupun Correctional Institution in 2000: It was a census year, but inmates weren't being handed census forms. The prison filled them out.
“When I came to prison, there were unanswered questions that I just kept in the back of my mind and I started gradually finding out about different things. So I'm not ignorant of these facts.”
Neither is his younger sister, Sharyl MacFarland.
“We're undercounted in this community. Because they've been counted somewhere else,” she said.
From her front porch, MacFarland can count down her block house after house where someone has been sent to prison. She includes her own house. All three of her sons have spent time behind bars; two of her brothers have been in prison most of their lives.
This transferring of political clout out of her community to districts where it is essentially erased has become an obsession for MacFarland.
Over 21 years of being separated, MacFarland and her brother Leonard Collins remained connected by the same question: How is change going to happen as long as the people in power are the ones who write the rules?
And as long, Collins said, as people like him, who are most invested in reforming the policies of mass incarceration, are locked out of the conversation.
“It's a domino effect. People are powerless, and that's because incarceration makes you powerless,” MacFarland said. “But if you keep all the power in one place, around one area, this is where it's going to remain and will continue to take the power.”
It’s a situation, she said, that makes the possibility for democratic reform harder and harder to imagine.
The Census Bureau counts people according to what it calls “usual residence” —essentially, where people eat and sleep most of the time. College students are counted where they’re in school; deployed military personnel are counted at their base, if they’re stateside.
What makes prisoners unique is that they have no contact with the community they’re being counted in, and they are there against their will. Ruling on the question of where a prisoner calls home, one U.S. Court of Appeals judge wrote in 1973, “It makes eminent good sense to say as a matter of law that one who is in a place solely by virtue of superior force exerted by another should not be held to have abandoned his former domicile.”
'I represent everybody in the district'
Federal dollars aren’t affected by where prisoners are counted at the state and local level, because the overall state population usually stays the same. Most federal funding comes to states in block grants or is determined by more complicated formulas.
The real benefit of “prison-based gerrymandering” is power — who’s representing whose interests. And prison gerrymandering is one way that locking up so many people — about 2.3 million nationwide, a larger population than 15 individual states — has reshaped our democracy.
In a district with a prison, like Granger’s, a small number of voters gets all the attention of the person they put into office. Conversely, in a community that loses a lot of people to prison, the district lines must widen, potentially watering down its priorities by lumping together people with different needs.
As for districts where the boundaries are tighter because a concentrated bloc of people are incarcerated, the rest have exaggerated power.
“If you don’t have to answer to half of your people or a quarter of your people or however many it is depending on your community — if 50% or more of your constituents can’t vote — it means the other 50% is almost getting more than one vote,” state Sen. Lena Taylor said.
Past attempts at reform have failed
The state Assembly and Senate — the people elected to represent people in these various districts — are where policy begins in the state. They’re also where policies like mandatory minimums and truth-in-sentencing that led to the state’s mass incarceration rates were born.
Rep. Michael Schraa, a Republican, heads up the Legislature’s Corrections Committee — the committee that drafts bills on how prisons should be run.
Taylor said Wisconsin's way of accounting for incarcerated people has also limited the ability to elect people of color to office. Taylor is one of two Black senators in the state Legislature. She said this was a direct result of the maps that were drawn in 2010.
“If you go back and look at the maps that were drawn in 2010," Taylor said, "and then you look at the numbers when those maps were first used in 2012, we lost about half of our Black representatives in our Assembly” compared with the maps drawn after the 2000 census.
In the 2003-04 Legislature, there were seven representatives of color, compared with the 2013-14 session, when there were four.
Before those maps were drawn in 2010, a group of legislators in the statehouse did try to do things differently. Then-state Rep. Fred Kessler, a Democrat, led a group of colleagues in arguing for a constitutional amendment requiring the state to just exclude prisoners from the redistricting data.
This was in June 2009, during a rare moment when the Democratic Party held control of the Assembly, Senate and Governor’s office.
But after a series of public hearings and approval from the Committee on Elections and Campaign Reform, the amendment failed in the Assembly. Kessler retired in 2019.
There was another moment when a couple of Wisconsin lawmakers sounded the alarm about how prisoners were factored into redistricting — but their motivations were different.
In 1999, U.S. Reps. Paul Ryan and Mark Green, both Republicans, convened a hearing to talk about the census and how incarcerated people were factored into redistricting. They were worried about losing their seats in Congress.
At the time, Wisconsin was shipping many prisoners to private prisons out of state. If those prisoners were counted as residents of those other states, the two feared, it could amount to Wisconsin’s nine seats in the U.S. House of Representatives dropping to eight.
Nothing came out of Ryan and Green’s hearing. Wisconsin did lose a seat, but not because of how inmates were counted.
The body best positioned to change where people in prison are counted, of course, is the U.S. Census Bureau. And while it hasn’t changed its rules on how and where prisoners are counted, for the first time this year the bureau is taking steps to make it easier for states to try something different as they draw their new districts: It has promised to share more detailed data on prison populations, and sooner than usual, and to help states figure out how to adjust the numbers if they choose.
While other states around the country are recognizing the way prisoners are factored in redistricting as a problem, and enacting big changes to solve it, Wisconsin’s leadershave signaled little interest in change. The odds are that Wisconsin will remain a striking example of prison gerrymandering.
About this story
Freelance journalist Natasha Haverty spent the 2018-19 academic year as an O'Brien Fellow in Public Service Journalism at Marquette University examining the impact of the way prisoners are counted in the census — as residents of the place they are incarcerated, not where they call home — and how the data affects political power and how districts are drawn. She was assisted by student researchers Lucie Sullivan, Robyn Di Giacinto and Claire Hyman.
Initial work was aired on the NPR program Reveal from the Center for Investigative Reporting. This story, focused on Wisconsin, is an outgrowth of that effort. Additional support for Haverty's reporting came from the Fund for Investigative Journalism and the Rockefeller Brothers Fund.
Work on the story was done under the guidance of Milwaukee Journal Sentinel editors. The various funding organizations played no role in the reporting, editing or presentation of this project.
This article originally appeared on Milwaukee Journal Sentinel: Wisconsin voting districts skew power to small prison towns