Locked Up, But Still Counted: How Prison Populations Distort Democracy

by Peter Wagner, September 5, 2008

This article originally appeared in Open Society News.

A 200-year-old glitch in the U.S. census is causing an increasingly serious problem for American democracy. Members of the booming U.S. prison population continue to be counted as residents of the city or town where the prison is located, even though they don’t vote, pay taxes, or do anything else that would make them active citizens in the community outside the prison’s walls.

This arrangement may have made sense to the founding fathers more than two centuries ago when there were few prisons and few uses for the census. By 2008, however, that has all changed. Almost 1 percent of the American adult population — and 3.6 percent of the black adult population — is temporarily behind bars, and the census has become a crucial tool for making important decisions ranging from where to build schools and hospitals to how to draw voting districts. The impact of counting people in prison in the wrong place is now simply too big to ignore. Increasingly, the practice of counting the prison population as residents in communities with prisons is distorting democracy at the state and municipal levels and fueling prison expansion projects at the expense of other forms of economic development.

According to a Prison Policy Initiative analysis of 2000 U.S. census data, there are 21 counties where at least 21 percent of the reported census population is actually composed of incarcerated people from outside the county. In 173 counties, more than half of the black population reported in the census is incarcerated. Fifty-six counties that were actually losing residents and in serious distress were recorded by the census as growing because they had opened prisons during the previous decade.

Along with distorting government priorities, crediting large portions of a population to the wrong communities makes drawing fair legislative districts impossible. Because people in prison cannot vote and remain legal residents where they lived prior to incarceration, counting them as residents in the district where they are imprisoned creates artificial populations that increase the voting strength of the prison district’s real residents and dilute the voting strength of residents in other districts. In New York State, seven rural state senate districts with large prisons would not meet the U.S. Supreme Court’s minimum population requirements without counting the prison population as local residents. Four of those prison-district senators sit on the powerful Codes Committee and oppose reforming the state’s draconian Rockefeller drug laws, which boost the state’s prison population by mandating harsh penalties. The inflated populations of these senators’ districts give them political power and little incentive to consider or pursue policies that might reduce the numbers of people sent to prison and the length of time they spend there. Republican New York State Senator Dale Volker, boasts that he is glad that the almost 9,000 people confined in his district cannot vote because “they would never vote for me.”

In New York, the state legislature relied on 2000 census data to update district boundaries and wound up drawing one legislative district in which 7 percent of people were in prison. A similar situation in Texas resulted in one district where 1 out of every 8 people was in prison. In Montana, a district padded its population by 15 percent with a prison population imported from other parts of the state. This means that every group of 85 residents who live near the prison is given the same say over state affairs as 100 residents elsewhere in the state.

The most visible support for changing the way the Census Bureau counts people in prison is coming from urban areas like New York City and Chicago that see large percentages of their population sent off to remote prisons. Yet some of the strongest support for reform is coming from rural communities that have prisons and need census data to draw legislative districts. Because county legislative districts tend to be small, a large prison can dominate the district and dilute the votes of residents elsewhere in the county. For example, the residents of St. Lawrence County, New York, who happen to live near a state prison have 25 percent more say over the future of their county than people who do not live next to the prison. Tired of how a prison distorted the county’s districts, people in St. Lawrence County signed a petition to stop the inclusion of prison populations in local districts. After the petition effort failed, county voters used local elections to throw out most of the incumbents responsible for the prison population district gerrymandering.

Senator Eric Schneiderman, a Democrat from New York City, is leading the effort in the New York State Senate for reform. “We need to deal with this, not just as a political issue, but as a moral issue,” he recently told City Limits. He went on to explain why this issue matters in his district: “Working in the legislature, we look at poor communities, like the one I represent in Washington Heights: people tend to have fewer resources for their schools, fewer resources for transportation, less access to health care. Our constituents are told to get into the political process and fight for change, and yet the same state actors dilute their vote.”

A relatively simple solution would be for the Census Bureau to collect home addresses rather than institutional addresses for people in prison in the next census. In the past, the bureau has updated its method of counting students, missionaries, overseas Americans, and other groups as changing demographics and changing needs for data demanded it. Yet the bureau moves slowly, and even though the 2010 census is two years away, time for reform is limited because the bureau has already begun planning the count. A number of states are taking steps to resolve the problem on their own. A bill is currently pending in New York that would collect the home addresses of incarcerated people and then adjust the federal census data when it arrives. This is not an ideal solution, but it is a viable option for some states and could alleviate the political distortions that come from counting transplanted prison populations as local residents.



Stay Informed


Get the latest updates:



Share on 𝕏 Donate