Census count hurts prison towns right now

by Peter Wagner, March 28, 2006

This is a letter I sent last week to the editor of the Tribune-Star in Terre Haute, Indiana responding to Mark Bennett’s column If a coalition of congressmen get their way, Vigo County and Congressional District 8 could be facing a serious population decrease.

Dear editor,

I share Mark Bennett’s concern that how prisoners are counted in the Census could become a “tug-of-war between cities and rural communities.” If that happens, it will obscure how the current Census policy hurts rural communities right now.

Bennett is right that counting prisoners at home would cost Vigo County some clout in the state legislature and in Congress. But while the 8th Congressional District has more prisoners than most — the district is 1% prisoners — that small amount of extra clout comes at a big price for local democracy.

The city of Terre Haute is divided in to 6 council districts, each of which has roughly the same Census population. The Supreme Court’s “One Person One Vote” rule requires districts to be drawn to contain the same number of people. In theory, this gives every resident of the city the same access to government regardless of where she or he lives.

I say “in theory” because the Census Bureau counted 1,764 prisoners at the United States Penitentiary as if they were residents of District 1 in southwest Terre Haute. The prisoners cannot vote, but the votes of the remaining 7,778 residents of District 1 have as much clout as the approximately 10,000 residents in each other district. Using prisoners as population gives each group of 8 residents in District 1 the same clout as 10 residents in other districts.

Left uncorrected, this problem is about to get worse. The federal prison complex has grown to contain 2,910 prisoners. If the next Census was taken today, the resulting District 1 would be 30% prisoners, giving each group of 7 District 1 residents the clout of 10 elsewhere in the city.

If the Census Bureau started counting prisoners at their home addresses, Terre Haute’s residents would have a more equal City Council, and there would be little impact on the city’s federal funding.

For example, many grant programs shun “per-capita income” statistics to use “household income” or “poverty status” because those statistics do not include prisoners and other group quarters populations. Still other grant programs require that institutional populations be deducted from the population figures. Simply put, most government programs are designed with sufficient precision that they are not affected by where prisoners are counted.

Ironically, the place that prisoner counts most frequently affect funding is in the distribution of funds within rural communities. Like our analysis of the impact on democracy, the biggest victims of this census policy are the rural residents who live near — but not immediately next to — large prisons.

Peter Wagner
Executive Director
Prison Policy Initiative