A growing number of state and local governments are using a new set of data produced by the U.S. Census Bureau to recalculate the residences of hundreds of thousands of prison inmates who are serving their sentences far away from their homes.
Legislatures in Connecticut and New Jersey have passed new laws this year that require state redistricting panels to count those who are incarcerated as living at their home addresses, joining California, Nevada, Colorado, Delaware and Virginia in adopting the policy for the first time this decade.
This week, Pennsylvania’s redistricting commission said it would follow the same practice.
During the last census, only two states, Maryland and New York, counted prisoners at their home addresses.
Those new adopters have been aided by the Census Bureau itself, which for the first time this year included a separate data set in official figures that states will use for redistricting purposes that includes counts of prisoners in their physical location and counts allocating them to their home districts.
Supporters of the reallocation say it is necessary to even out representation among districts. Prisons in many states are located in rural communities far from urban centers; counting prisoners in their place of incarceration can give uneven influence to those areas, to the detriment of larger population centers that, because of their size, send a disproportionate number of inmates.
“You end up with all these folks who aren’t counted in their home communities where that’s their legal residence, where that’s their representative who still represents them,” said Aleks Kajstura, legal director at the Prison Policy Initiative. “You can have a small county out there somewhere with a couple districts, and just a single state prison can account for most of the district side.”
In Pennsylvania, the 2,389 inmates at the State Correctional Institution in Marienville account for a little under a third of the 7,716 residents of Forest County, north of Pittsburgh. About a quarter of the population in Pershing County, Nevada, is made up of the 1,680 residents of the Lovelock Correctional Center. And about 10 percent of the residents of Fremont County, Colorado, live in one of seven state-run prisons in and around Cañon City.
Almost 2.3 million people are serving time in thousands of jails, prisons and correctional facilities across the country, or about seven-tenths of the nation’s total population.
“In certain states, the prison population comes from cities and is essentially reallocated to rural areas with predictably partisan and racial consequences,” said Nathaniel Persily, a Stanford Law professor and redistricting expert. “There’s the moral argument that it’s wrong to essentially pad the representation of a community that contains a prison merely because it has a disenfranchised population.”
The question of where to count prison residents may only impact the political power of a handful of places across the country, and it may only marginally diminish the representation in most big cities.
But in an era when the redistricting process has taken on a heightened partisan significance, and when every seat counts toward the narrowest of majorities, Democrats tend to favor counting prisoners at home, while Republicans tend to prefer to count them in their physical location.
“Reallocating prison populations is a tool Democratic political operatives like to use to move generally non-voting populations into urban areas so they can adjust those districts to elect more representatives with fewer actual voters,” said Jason Torchinsky, general counsel to the National Republican Redistricting Trust (NRRT). “When prison population is reallocated away from where those individuals actually reside, it negatively impacts the ability of the elected representatives of those areas to seek the various federal and state level support that other areas have.”
“Prisons generally rely on local communities for food, water, power, health care, support services and all of the other things that go towards maintaining those facilities and taking care of those individuals,” Torchinsky said.
The National Democratic Redistricting Committee, the NRRT’s Democratic counterpart, applauded the Pennsylvania commission’s decision to count prisoners at home. The committee said 200,000 Pennsylvania voters are underrepresented because of the way prison populations were counted a decade ago.
“Fair districts should accurately reflect the will of the voters, and should be drawn in a way to provide effective representation to the communities that call those districts home,” said Kelly Burton, the NDRC’s president. “Counting individuals where they are temporarily incarcerated is simply unfair. It artificially inflates the population of rural areas where most prisons are located and dilutes the power of voters in their home districts in other parts of the state.”
The decennial census has wide-ranging applications beyond redistricting: Population data is used to allocate hundreds of billions in federal, state and local programs every year.
In some instances, counting residents at their physical location, even in prison, is smart planning, Persily said.
“Sometimes we actually want to know that people are physically in prison or not. It is relevant to other questions like funding or emergency planning or other purposes for which we use the census,” he said. “If you’re trying to plan for flood emergencies or something, you actually need to know how many people are in each location, where they actually are, not where they may have been before they were incarcerated.”
Kajstura, of the Prison Policy Initiative, said modern funding mechanisms mean few federal or state programs allocate money today based on formulae as simple as a straight-up headcount. Instead, most programs take in complicated calculations that weigh a population’s demographics or wealth, rather than its absolute size.
“There’s very little out there that’s dollars per head,” Kajstura said. “So incarcerated populations are removed from the funding formulas.”