In U.S., prisoners don’t live in prison
By The NYT Editorial Board
Where do you live? For most people, that’s an easy question to answer when the census comes around. It’s much harder for those locked up in a state or federal prison, often hundreds or even thousands of miles from the place they last called home.
Long-standing Census Bureau policy is to count people as residing wherever they usually eat and sleep, known as the “usual residence” rule. For prisoners, that means being counted in the legislative districts where they are incarcerated.
But that makes no sense, because virtually everyone who goes to prison comes from somewhere else, and almost all will return there after being released. While they are behind bars, they can’t vote, nor do they have any attachment to the local community or its elected officials. They are counted, even though they can’t hold their representatives accountable.
The result is one of the more persistent and pernicious distortions in the redistricting process, known as “prison gerrymandering.” Now that the 2020 census count is over, and the nation begins its decennial struggle over how to draw new congressional and other legislative district lines — and who gains or loses political power as a result — it’s a good time to talk about how we can get rid of prison gerrymandering at last.
A healthy representative democracy needs an accurate picture of who lives where in order to allocate the proper number of lawmakers to represent their interests. It’s why the census is the first job the Constitution gives Congress.
Prison gerrymandering distorts this picture in two ways: by artificially inflating the population of places where prisons are, and artificially decreasing the population of the places where prisoners come from.
Add to that the constitutional principle of one person, one vote, which requires all legislative districts to be roughly the same size. If large groups of people are counted in the wrong place, the result is a government skewed in favor of some and against others.
When it comes to prisoners, the skew follows a clear pattern: Prisoners are disproportionately Black and brown people from urban areas, and prisons are disproportionately built in more rural areas. So counting people where they’re imprisoned takes political power away from racial minorities in cities and transfers it to whites in rural areas.
In one Rhode Island House district, for instance, prisoners account for nearly 16% of the population, even though only 4% of them are from that district.
In Juneau County, Wisconsin, it’s even worse: Prisoners account for 80% of the entire population of one district.
In Connecticut, a redistricting expert calculated that nine of the 151 state House districts are able to meet the required minimum population only thanks to prisons within their borders, and that eight of those nine districts encompass predominantly white communities. If the state stopped counting prisoners where they are locked up, the expert said, 22 districts would need to be redrawn.
For most of American history the distortions caused by prison gerrymandering didn’t make much difference. There weren’t that many people behind bars. That changed with the incarceration boom that began in the 1980s. Today, more than 2 million people are held in state and federal prisons and local jails, with concrete consequences for politics and policy.
In New York, the harsh Rockefeller drug laws, which mandated absurdly long sentences that fell on many young men and women of color, were consistently unpopular with the public.
They survived largely because they were defended by legislators who came from districts that benefited from prison gerrymandering. At the height of the prison-building boom in New York, 28 of 29 new prisons were built in upstate and often rural districts, providing a reliable cash flow for those districts even as the inmates they housed were overwhelmingly from the big cities.
Sometimes lawmakers are forthright about the benefits of exploiting prisoners for political gain. In 2015, state Rep. Janet Adkins, R-Fla., told party activists that the best way to oust a Democratic incumbent was to pack her district full of prisoners. Draw it “in such a fashion so perhaps, a majority, or maybe not a majority, but a number of them will live in the prisons, thereby not being able to vote,” Adkins said.
The best solution to all this is for the Census Bureau, which provides the data for congressional and state districting, to change the “usual residence” rule: that is, stop counting prisoners where they are locked up and start counting them in the place they call home, or at least in the last place they lived before going to prison. That would be consistent with a line of Supreme Court cases holding that a person’s residence isn’t necessarily where she happens to be found at the moment the census occurs, but the place to which she has some “allegiance or enduring tie.”
In 2018, the bureau asked for public comment on the usual residence rule. Of the 77,887 comments it received about prisoners, 77,863 — 99.97% — said they should be counted at their home address. Despite the virtually unanimous consensus, the bureau didn’t change the rule, although it agreed to provide states with access to data that make it easier for those that want to reduce the impact of prison gerrymanders.
Many states have grown tired of waiting, and have begun outlawing prison gerrymandering on their own, by incorporating their own data on their prison populations into the census data.
Ten states now ban the practice outright — including California, New York and Virginia — together representing 30% of the U.S. population. Bills are on the table in four more states, and roughly 200 local jurisdictions have stopped counting prisoners where they are locked up. This is progress, but it’s piecemeal.
Democrats in Congress are trying to force the issue by including it in H.R. 1, the omnibus democracy-reform bill also known as the For the People Act. The bill would require the Census Bureau to count prisoners at their last known address. But even if H.R. 1 becomes law — which is far from assured at this point — it can’t fix the problem for the 2020 redistricting cycle, because the count is already done.
Rather than wait another decade, states without a ban already in place can take action now.
With the help of the newly detailed data from the Census Bureau, they can divide up their inmate populations among multiple districts, essentially diluting the impact of prison gerrymandering by ensuring that no one district gets to claim too many prisoners. It’s not a perfect solution, but it may be the best way to fight this anti-democratic, racially biased practice until it can be eliminated for good.