Every time New York City sends a convicted criminal to an upstate prison, it also gives up money and power.

That’s because the Census Bureau counts prisoners as residents of the towns they are incarcerated in, instead of the neighborhoods they lived in before being locked up. The result: Every calculation that uses Census data–from federal funding formulas to drawing state legislative districts–gives more money and representation to communities with prisons than they would get if reckoning were based solely on their non-incarcerated populations. That makes prisoners a hot commodity that everyone wants to claim.

Take Franklin County, a bucolic getaway on the Canadian border. Of the county’s 51,000 residents, more than 5,000 are behind bars. An estimated 3,600, many black and Latino, come from New York City to the overwhelmingly white area. This influx of big-city convicts brings extra anti-poverty funds to the little towns of Franklin County. It also inflates the population count for the legislative district, even though convicted felons cannot vote. As a result, in one Franklin County district, one state senator represents 287,000 free individuals; in Queens, each state senator represents 318,000 people. That gives each free person in Franklin County a bigger voice in Albany than someone in, say, South Jamaica.

That disparity in political power has researchers and advocates, both in New York and nationally, taking a closer look at the system–and pushing to change it. One option is to convince the Census Bureau to count differently; another is to allow state legislatures to conduct counts themselves, instead of relying on the Census. Proponents of these changes say the current situation has economic and political ramifications. They note that the current method bolsters Republican domination of the State Senate, helping to perpetuate drug sentencing laws that discriminate racially. Furthermore, proponents note, the current counting system siphons resources from downstate, so ex-cons returning to impoverished city neighborhoods get fewer services to help them reintegrate into the community.

Commentators like Harvard Law School professor Lani Guinier have remarked on the injustice of the arrangement. “The strategic placement of prisons in predominantly white rural districts often means that these districts gain more political representation based on the disenfranchised people in the prison,” she wrote in The American Prospect in 2001, “while the inner-city communities they come from suffer a proportionate loss of political power and representation.” The situation, according to Guinier, is comparable to the three-fifths rule that allowed states to count slaves to bolster the South’s representation in Congress–which helped the South preserve slavery.

While many agree the injustice exists, it’s harder to pin down the best solution. Advocates admit that repatriating money and power to urban areas might not have an immediate, significant impact. Nationally, each prisoner brings in only about $100 in local and federal aid–and figures for New York state might be lower, according to rough estimates by Eric Lotke, a Soros Justice Senior Fellow. Currently, 43,740 people from New York City are incarcerated elsewhere, and Lotke’s research suggests that these prisoners export, at most, $4.37 million annually from the city–peanuts compared to budget gaps in the billions. Further, prisoners from New York City collectively represent only about one-seventh of a New York City state Senate seat. That’s not enough to tip the 38-to-24 Republican advantage in Albany, or to compel significant change in the Assembly, where Democrats already outnumber Republicans.

Short-term calculations aside, advocates believe that the count should be brought in line with demographic reality and social justice principles, so the chips can fall–even if they fall slowly.

The issue of counting incarcerated convicts could have its biggest impact on the larger prison reform movement. “This is a complicated issue that’s been below the radar,” says Peter Wagner, a Soros Fellow and assistant director of the Prison Policy Initiative in Springfield, Massachusetts. “Interest has been growing since the 2000 Census showed that there are two million people in prison in this country, and we are beginning to see some of the effects” of that ballooning population.

Wagner’s research has focused on the state of New York, where several senate districts straddle a statistical border that may eventually push them into court challenges. When prisoners are subtracted from five of New York’s state Senate districts, the disparity between their size and that of the most populated districts in the state is larger than federal law allows. If prisoners were excluded from those districts’ counts, all five would have to be expanded geographically. It’s anybody’s guess what political ramifications that would have. But Wagner speculates it could shift the political balance of the districts by bringing in new voters.


The political obstacles to changing the way the Census counts, however, are daunting. Advocates would have to win over a lot of people with a lot of interests at stake–including the president, the U.S. Congress and the state Legislature. Defenders of current practice have tradition on their side–the Census has been counting prisoners the same way since 1790. Some officials do not even pretend the method is rational; they’re simply pleased to see their communities benefit from it. “From a selfish point of view, hey, whatever works,” Henry Rausch, a mayor in upstate New York, told the Boston Globe three years ago. Rausch’s town, Coxsackie, counts a third of its residents in local prisons. “I’m not about to set out and change it if it helps us,” Rausch added.

Even liberal redistricting experts admit that the most obvious alteration–rearranging prisoners in Census counts based on where they come from–might not be constitutional. David Bositis, a senior political analyst at the Joint Center for Political and Economic Studies in Washington, D.C., predicts that approach would fail. “They would have to get the Supreme Court to come up with a decision contrary to the last 40 years of jurisprudence,” Bositis said. “Do I think that’s particularly likely? No.”

State-managed counts that depart from Census methods are not unprecedented, however. For example, Kansas subtracts out-of-state students and military personnel for state legislative districts. In 2001, a Texas state representative from Houston, Democrat Harold Dutton Jr., proposed a bill to count prisoners in their hometowns. It got out of committee but languished on the floor.

As for New York, Wagner argues that legal precedent and the state constitution are on his side. New York’s constitution does require using Census data to draw legislative districts, but it also allows alternate methods when the Census “fails to give the information as to any civil or territorial divisions which is required to be known.” Narrowly interpreted, this caveat suggests that if the Census did not, for example, break out Brooklyn’s population from the whole of New York City, the state could do its own count. Wagner, however, says it gives the state some discretion in figuring how the numbers should best be used to craft legislative districts.

For legal precedent, Wagner points to an 1894 Court of Appeals decision involving a vagrant, one Michael Cady, who was in the habit of getting jailed in order to get free room and board. Before an election, Cady attempted to list the Tombs as his voter registration address. The court said no. Key to the decision was a part of the state constitution that stipulates, “for the purpose of voting, no person shall be deemed to have gained or lost a residence, by reason of his or her presence or absence…while confined in any public prison.” So if jailed people awaiting trial must vote where they were last registered, why shouldn’t they be counted at home after being convicted?

For now, Wagner’s arguments are moot–the legislature approved new districts in 2002, and the conversation won’t start up again for seven or eight years. So he and others have turned to efforts to influence federal Census officials, who are already discussing the 2010 count. Last October, a team from the Brennan Center for Justice at New York University School of Law presented its case before two panels that advise the Census Bureau: the Decennial Census Committee and the Census African American Advisory Committee. The latter has recommended that the bureau count prisoners based on where they come from.

Former African American Advisory Committee Chairman Robert Hill acknowledged that his group would likely favor reapportionment arguments before the Census Bureau staff or members of Congress would. Still, says Hill, “Our recommendation…gives a basis of support” to sympathizers in Congress. In 2005, the bureau will begin outlining its plans for the next decennial count.

Advocates are also trying to show that the Census has always had changes and exceptions. For example, it counts people at their “usual place of residence,” commonly interpreted as wherever they spent the night of April 1. Wagner points out that boarding-school students are counted with their families, and travelers based on where they usually sleep. The Census Bureau has also modified its procedures in the past. In 1950, for instance, college students became residents of places where their dorms were located, instead of their parents’ homes. Though that change seems similar to counting inmates in prisons, the deeper point, says Wagner, “is that the Census Bureau has changed with the changing times.”

A 1992 U.S. Supreme Court case could help persuade the Census that a home is more than just a place to hang your hat, says Patricia Allard, an associate counsel at the Brennan Center. There, the court upheld a new methodology counting soldiers abroad as residents of states where they maintain their “home of record.” The court held that a “usual place of residence can mean more than mere physical presence, and has been used broadly enough to include some element of allegiance or enduring tie to a place.”

As for figuring out a prisoner’s home address, advocates suggest using the last legal residence listed in court papers, and if that is not available, the county where he or she was convicted. Census enumerators likewise approximate when counting a homeless shelter as the residence of homeless people, says Soros Fellow Lotke: “Even if they sleep in a different shelter each night, they are still more or less in that area and in that community of interest.”

Adds Allard, “The idea of a home being a prison is absurd. People who go to prison do not all of a sudden lose their hometown.” When they return and find social service resources shortchanged because of the current Census count, “the individual and the community are being cheated out of what is rightfully theirs.”

Matthew Schuerman is a Brooklyn-based freelance writer.