Pennsylvania voting rules explain prisoner residence

by Peter Wagner, April 18, 2005 Link

In previous articles, I’ve explained how the two states that currently allow prisoners to vote, Maine and Vermont, require prisoners to vote via absentee ballot back home. Prisoners are not allowed to vote in local elections because they are considered residents of their home communities, not the prison towns.

But this rule applies in other states as well. In contrast to people in state prison for felonies, most states allow jail inmates (generally those awaiting trial or serving a short misdemeanor sentences) to vote. A voting rights pamphlet [PDF] issued by the Pennsylvania Secretary of State and distributed to incarcerated people shows how that state interprets its own residence rules for incarcerated people.

The rules state that a prison cannot be used as a voting address, and that the person should vote using either their last registered address or register at their last pre-incarceration address. But the rules also contain a common sense provision that keep the rule fair and accurate: prisoners may also establish a new residence outside of the correctional facility, “for example, if the inmate’s spouse establishes a new residence in which the inmates intends to reside upon his/her release from confinement.”

Pennsylvania, like most states, defines residence and structures its democracy around the idea that people are residents of the part of the state they choose to be in. Incarceration does not qualify as a residence because it is involuntary. That makes sense, but this carefully constructed concept breaks down when the state relies on federal census data to draw its legislative districts because prisoners and prisons are unevenly distributed in the state. Forty percent of the state’s prisoners are from urban Philadelphia, and all are incarcerated outside of the city in often very remote locations.

In order to comply with their own rules on residence, states like Pennsylvania need the Census Bureau to change how they count incarcerated people. While it may have made sense to count prisoners as residents of the facility during the first census in 1790, both high incarceration and modern uses of the census for redistricting require a new approach.

Thanks to Jon E. Yount from bringing this pamphlet and its coverage in the February 2005 issue of Graterfriends, published by the Pennsylvania Prison Society to my attention.

Zoe Gottlieb takes prisoner-census analysis to southern states

by Peter Wagner, April 11, 2005 Link

Zoe Gottlieb, a law student at the New York University School of Law, has made available on Prisoners of the Census the first review in southern states of the impact of how the U.S. Census counts prisoners. The paper, Prisoner enumeration and the “Usual Residence” rule in southern states, [PDF] looks at the county of origin and county of incarceration for prisoners in Georgia and North Carolina.

In the states I have previously studied, the urban areas often have large Black and Latino populations while the rural areas house the majority of the prisons but are largely white. In New York, I found that although New York State’s prisoners are 82% Black or Latino, 98% of New York’s prisoners are incarcerated in Senate districts that were disproportionately White for the state as a whole.

Ms. Gottlieb’s study is the first attempt to quantify at the county level in the South whether prisons are likely to be built in white rural areas. Given the very different demographics of the South and the fact that that region has the highest incarceration rate in the country, Gottlieb’s review is well timed to the growing national interest in the accuracy and utility of Census data.

Although prisoners are come from large portions of both Georgia and North Caolina rather than a single large city, Gottlieb confirmed that the majority of prisoners are residents of what the Census Bureau labels “metropolitan” counties and are incarcerated in “non-metropolitan” counties. However, unlike in the North and despite disproportionate Black incarceration, Gottlieb found no evidence of a transfer of prisoners from largely Black counties to majority White counties.

This latter finding was not unexpected because although prisons in the South are likely to be in non-metropolitan counties, Southern non-metropolitan counties have large Black populations. (A review of employment patterns I conducted last year with Rose Heyer found that the South was the only place besides the urban areas of the Northeast where any prisons had a racial parity between prisoners and staff.)

However, Ms. Gottlieb was only able to extend this analysis down to the county level. It is possible that when this analysis is extended down to the legislative district level, additional detail may become apparent. But regardless, the high concentration of prisons in rural Georgia identified by Gottlieb has disturbing implications for the ability of all states to rely on Census data for an accurate representation of where their population actually resides.

Ms. Gottlieb also included some detailed tables about Alabama, but errors in how the Census Bureau classified correctional facilities by type (state, federal or local) frustrated a complete analysis of that state at the county level. However, her data in all 3 states provides an insightful look in to prisoner origin and prison construction patterns at the county level, and offers some valuable insights to guide future research in to how the redistricting process is skewed in the South by the Census Bureau’s method of counting our large incarcerated population.

Pace Law Review: Prisoners of the Census: Electoral and Financial Consequences

by Peter Wagner, April 5, 2005 Link

The U.S. Census Bureau counts people in prison where their bodies are confined — in prison — not the communities they come from and where they are genuine members. This would be an item of statistical trivia, but the new numbers give it new meaning. More people now live in prison and jail than in our three least populous states combined. Organized differently, they would have six votes in the United States Senate. It is not trivia anymore.

A new article in Pace Law Review traces the unanticipated effects of this counting method. Prisoners of the Census: Electoral and Financial Consequences of Counting Prisoners Where They Go, Not Where They Come From [PDF] by Eric Lotke and Peter Wagner is the first academic article to quantify the impact of prison population counts on both the political process and on state and federal funding streams.