Testimony calls for New York to adjust Census counts of people in prison as part of larger redistricting reforms

by Peter Wagner, November 3, 2006  

New York State is considering a number of reforms to its redistricting practices that would transfer control over the process from the legislature to an independent commission. The State Assembly is holding a series of hearings on the issue around the state. At the invitation of Citizen’s Union, Prison Policy Initiative Executive Director Peter Wagner testified at the October 17 hearing held in New York City. His testimony, reprinted below, discusses why relying on the Census Bureau’s count of people in prison as residents of the prison town violates the New York State constitution. His testimony offers three ways that the state could remedy the situation.

Many of the other democracy and good government groups who testified at the hearing agreed that where people in prison are counted is an important issue. Citizen’s Union and Common Cause NY called for the state to address the issue of prisoner counting as part of its redistricting reform and Demos, the Brennan Center for Justice and the New York Voting Rights Consortium explicitly supported changing how people in prison are counted.

Testimony of
Peter Wagner, Executive Director,
Prison Policy Initiative
Before the
Assembly Standing Committee on Governmental Operations
and the
Assembly Legislative Task Force on Demographic Research and Reapportionment
October 17, 2006

Good morning Madam Chairwoman, Mr. Co-Chair and Assemblymembers. I thank you for the opportunity to testify today. My name is Peter Wagner and I am an attorney and Executive Director of the Prison Policy Initiative. For the last 5 years I have been conducting legal and demographic research on how the Census Bureau’s method of counting people in prison negatively impacts the political redistricting process. In 2002, I wrote a report called Importing Constituents: Prisoners and Political Clout in New York which was the first report to quantify the unintentional harm to democracy caused by New York State’s practice of drawing legislative district lines based on Census Bureau counts of prisoners as residents of the prison town.

Citizens Union asked me to appear here today because the issue of prisoner counting has been raised at your two previous public hearings. I would like to address the relevant requirements of the 14th Amendment and the New York State Constitution and suggest how New York can fix the problem created by the Census Bureau’s outdated method of counting the population.

Counting prisoners for redistricting purposes as residents of the prison town has profound implications for minority voting strength because of three simultaneous trends in New York State’s prison policies: a growing incarceration rate; a large and increasing racial disparity in the prison population; and a large and growing trend to locate prisons in majority White rural areas.The percentage of New Yorkers who found themselves incarcerated on census day in 2000 was 5 times higher than 3 decades prior. While the state is 62% White, the prison population is overwhelmingly people of color. Eighty-one percent of the prisoners in New York State are Black or Latino. Alongside this racial disparity has been a growing geographic disparity: New York City is home for about two-thirds of the state’s prisoners, but all of the 43 new prisons built in New York since 1976 have been built upstate. Crediting 71,466 state prisoners to the wrong place seriously distorts the demographic face of New York State. I brought with me a map that shows that in 10 New York counties more than half of the Black adult population is not local residents but state prisoners. As this map vividly demonstrates, the data that New York relies on to draw its legislative districts is not an accurate reflection of the state.

Federal requirements

Since Reynolds v. Sims in 1963, the Supreme Court has required state legislative districts to be drawn on an equal population basis.[1] As the court remarked, “legislators represent people, not trees or acres. Legislators are elected by voters, not farms or cities or economic interests.”[2] The principle is that “the weight of a citizen’s vote cannot be made to depend on where he lives.”[3] Using Census counts of prisoners to draw legislative districts violates this principle by increasing the weight of a citizen’s vote if a large prison is nearby.

Crediting prison communities with large external population of prisoners — who are local residents in no sense other than in the Census — turns the “One Person One Vote” principle on its head. There can be no doubt that rural prison towns do not share a “community of interest” with urban prisoners or their loved ones. As three residents of Franklin County explained in a comment letter to the Census Bureau’s Redistricting Data Program:

“We do not consider the prisoners to be residents of our community as they originate outside of our community, have no interaction with it and immediately leave the district when their sentences expire or the Department of Corrections chooses to transfer them elsewhere.”[4]

Crediting thousands of non-resident prisoners to the prison location serves to unconstitutionally enhance the weight of each vote cast by rural residents. The size of this distortion can even be large enough to violate the maximum population deviations allowed by White v. Regester. One legislative district in New York is 7% prisoners;[5] a legislative district in Texas is 12% prisoners[6]; and 15% of one Montana district is prisoners imported from other parts of the state.[7]

Given the racial dynamic in New York where tens of thousands of disenfranchised Blacks and Latinos from New York City are credited to White legislative districts, the Voting Rights Act may also be violated by drawing legislative districts based on Census counts of prisoners. On May 4, the U.S. Court of Appeals for the Second Circuit signaled that it was interested in hearing a case which would argue that New York State’s practice of crediting prisoners to upstate districts constitutes vote dilution in violation of Section 2 of the Voting Rights Act.[8]

Critically, although states are required to redraw state legislative districts each decade to assure compliance with the federal Constitution’s one-person, one-vote requirements, they are not required to use federal Census data in doing so.[9] The Third Circuit has explained:

Although a state is entitled to the number of representatives in the House of Representatives as determined by the federal census, it is not required to use these census figures as a basis for apportioning its own legislature.[10]

State constitutional requirements

While federal law gives New York State the option of not using the federal census data to draw its districts, the state constitution requires the state to seek another source of data. The New York State Constitution defines residence in a way that precludes using Census counts of prisoners: “for purposes of voting, no person shall be deemed to have gained or lost a residence, by reason of his presence or absence … while confined in any public prison.”[11]

The case of People v. Cady, 143 NY 100, 37 N.E. 673 (1894), vividly illustrates the reality that a prison is not a home. As described in Importing Constituents:[12]

Michael Cady would repeatedly confess to vagrancy and have himself committed for six months at a time to the Tombs. He had been doing so for about seven years, and intended to do so indefinitely. He was even allowed outside on occasion to do paid errands. As he was only committed to prison for vagrancy, Cady was allowed to vote, and he registered using his Tombs address.

Cady was prosecuted for illegal registration — not for registering to vote — but for registering to vote as a resident of the Tombs. The prosecution’s theory was that under the Constitution and common sense, a prison cannot be a residence, and Cady must have lived somewhere else before he went to prison. The New York Court of Appeals upheld the conviction, citing to the New York Constitution prohibition on gaining or losing a residence from imprisonment, and further defined residence:

The domicile or home requisite as a qualification for voting purposes means a residence which the voter voluntarily chooses and has a right to take as such, and which he is at liberty to leave, as interest or caprice may dictate, but without any present intention to change it.

The Tombs is not a place of residence. It is not constructed or maintained for that purpose. It is a place of confinement for all except the keeper and his family, and a person cannot under the guise of a commitment, or even without any commitment, go there as a prisoner, having a right to be there only as a prisoner, and gain a residence there.

The New York Constitution requires that the state use the Census data for redistricting only “in so far as such census and the tabulation thereof purport to give the information necessary therefore.”[13] The data produced by the Census Bureau is not the “information necessary” for New York’s redistricting because it does not accurately reflect where in New York its citizens live.

Unlike many other states, the framers of the New York State Constitution foresaw a time when the federal census data would not meet the state’s needs and left open the possibility of using other sources of data. I will next address three ways that timely action by New York State could eliminate or greatly reduce the vote dilution caused by the Census Bureau’s method of counting people in prison.

The remedies

The best place to change where prisoners are counted in the Census is at the Census Bureau, but active intervention by New York is necessary. Given the rapid approach of the next census and the stubbornness of the Census bureaucracy, a major change in where the Census counts incarcerated people is unlikely for 2010; but the prospects for change in 2020 are much better if action is taken soon. Last month, the National Research Council of the National Academies released a report commissioned by the Census Bureau which called for a major test within the 2010 Census to study alternative ways of counting prisoners in future Censuses.

If New York State wants to use Census Bureau data for redistricting without violating the state constitution, the legislature should formally ask the Census Bureau to change how it counts prisoners. The Census Bureau considers redistricting to be the second most important use of the Census, so the opinion of New York State would carry significant weight with the Bureau.

A second option that would lessen the harm from miscounting prisoners in the 2010 redistricting cycle would be for New York to take the prison populations out of the Census data prior to redistricting. While ignoring any part of the population is not ideal, ignoring the incarcerated population is a better solution than crediting tens of thousands of disenfranchised minority men to entirely different communities of interest. Removing the prison populations is not a new idea. In fact, at least twelve upstate counties including Franklin and Greene counties already take prisoners out of the data used to draw their local county legislative districts.[14] As David Sommerstein of North Country Public Radio explained about the decision in Franklin County:

The legislature there voted last month to continue to exclude prisoners. Legislative Chairman Earl LaVoie says their stance was a no-brainer, because one district near Malone has more prisoners than residents. But he adds, prisoners shouldn’t be counted among North Country residents on principle.[15]

New York would find it even easier to remove prison populations from the redistricting data if these populations were specially marked in the redistricting data. (Currently, correctional facility data is only available in Summary File 1 which is published separately 3-5 months after the PL94-171 redistricting data.) In their recent report, the National Research Council recommended that the Census Bureau assess the interest of the states in having prisoner counts separately identified in the PL94-171 Redistricting Data.[16] If adopted, this recommendation could eliminate a great deal of the electoral inequities at negligible expense, but it must be acted upon before the Census Bureau finalizes the format of its data products. Therefore, New York should immediately notify the Census Bureau’s Redistricting Data Program that it would like the correctional facility populations identified within the PL94-171 Redistricting Data.

Third, New York could adopt the bill advanced by Senator Schneiderman (S2754) and Assemblymember Espaillat (A9688) which would collect the home addresses of people in prison and adjust the Census data prior to redistricting. The bill is loosely modeled after how Kansas adjusts Census counts of soldiers and students to comply with how the Kansas state constitution defines residence. The Schneiderman/Espaillat bill requires correctional facilities to submit home address information for each incarcerated person to the Board of Elections which will then modify the PL94-171 redistricting data to remove incarcerated people from the census block where the prison is located and add them to the census block containing their last address prior to incarceration. The resulting dataset would reflect where all people in New York — including people in prison — legally reside and would be required for use in drawing congressional, senate, assembly and county legislative districts.

The Schneiderman/Espaillat bill is an appropriate solution for New York in 2010. The National Academies thought that 2010 was too soon to change how people in prison were counted for the national census. In their report, they called for a major research project to determine the state of electronic record keeping in prisons and jails in every state and to determine the best way to gather the data where these records do not exist. A state based solution would have little of this complexity or risk, particularly in New York State where the Department of Correctional Services already has a largely complete database of home residence information and where a strong Commission of Correction has been actively guiding the development of county jail systems. With enough notice, it will not be difficult for the operators of state prisons and county jails to send the necessary data to the Board of Elections.


The Census Bureau’s outdated method of assigning residence to people in prison creates a large problem for democracy in New York. The problem was left uncorrected in the last redistricting cycle because the importance of the issue was discovered only shortly before the Census began. Today, we have the benefit of advance notice. If New York State begins lobbying the Census Bureau and planning to adjust federal Census data today, the state will be able to draw districts that give residents near prisons and residents far from prisons the same access to government.

Thank you for the opportunity to participate in your reexamination of the redistricting process in New York. I would be happy to answer any questions that you may have.

Selected references:

Prison Policy Initiative and National Voting Rights Institute, Brief Amici Curiae In Support Of Plaintiff-Appellant Jalil Abdul Muntaqim, a/k/a/ Anthony Bottom, Urging Reversal Of The District Court, January 28, 2005.

Peter Wagner, Importing Constituents: Prisoners and Political Clout in New York, April 22, 2002.

Peter Wagner, Eric Lotke and Andrew Beveridge, Why the Census Bureau Can and Must Start Collecting the Home Addresses of Incarcerated People, Submitted by to the U.S. Census Bureau on February 10, 2006 in advance of the Bureau’s report to the Appropriations Committee on using prisoners’ homes of record in the Census.

Daniel Jenkins, et. al, comment letter to the Redistricting Data Program, September 6, 2004. (Letter from three residents of Franklin County New York documents the harm to rural county democracy from how the Census Bureau currently counts people in prison)

National Research Council, Once, Only Once, and in the Right Place: Residence Rules in the Decennial Census, September 2006.

David Hamsher, Counted Out Twice – Power, Representation & the “Usual Residence Rule” in the Enumeration of Prisoners: A State-Based Approach to Correcting Flawed Census Data, 96 J. Crim. L. & Criminology 299 (Fall 2005).


[1] Reynolds v. Sims 377 U.S. 533 (1963) The equal protection clause “requires that the seats in both houses of a bicameral state legislature must be apportioned on a population basis… An individual’s right to vote for state legislators is unconstitutionally impaired when its weight is in a substantial fashion diluted when compared with the votes of citizens in other parts of the State.”

[2] Reynolds at 562.

[3] Reynolds at 567.

[4] Daniel Jenkins, et. al, comment letter to the Redistricting Data Program, September 6, 2004. (Letter from three residents of Franklin County New York documents the harm to rural county democracy from how the Census Bureau currently counts people in prison.)

[5] Peter Wagner, Importing Constituents: Prisoners and Political Clout in New York, Prison Policy Initiative, April 2002.

[6] Peter Wagner and Rose Heyer, Importing Constituents: Prisoners and Political Clout in Texas, Prison Policy Initiative, November 2004.

[7] Peter Wagner, Importing Constituents: Prisoners and Political Clout in Montana, Prison Policy Initiative, December 2004

[8] Hayden v. Pataki [PDF], (04-3886) May 4, 2006.at In January 2005, the Prison Policy Initiative and the National Voting Rights Institute submitted a brief amici curiae in a related felon disenfranchisement case arguing that New York legislature’s use of Census Bureau counts of prisoners when drawing legislative districts should be considered as one Gingles factor when evaluating New York’s felon disenfranchisement law under the Voting Rights Act. The Court rejected the argument that the Voting Rights Act gives people in prison the right to vote, but it seized upon our argument and went even further, questioning whether New York State’s use of Census counts of prisoners by itself amounts to impermissible vote dilution.

[9] See Mahan v. Howell, 410 U.S. 315, 330-332 (1973) (rejecting Virginia’s argument that it was compelled to use Census Bureau assignments of residences of military personnel in its state legislative redistricting).

[10] Borough of Bethel Park v. Stans, 449 F.2d 575, 583 n.4 (3rd Cir. 1971).

[11] N.Y. Const. art. II, § 4.

[12] Importing Constituents (Main Report, Part III) (quoting People v. Cady, 143 NY at 106).

[13] New York State Const. Article 3, Section 4

[14] This is a simpler operation than it may appear. In the 2000 Census, New York State was divided in to 298,506 census blocks, but only 154 of these contained correctional facilities. These 154 blocks are the only ones that would have to be adjusted to remove the correctional populations.

[15] David Sommerstein, Prisoners: North County Residents?, North County Public Radio (New York), March 5, 2004.

Tweet this page Follow @PrisonPolicy on Twitter Get our newsletter Donate Contact Us


  • September 21, 2016:
    Executive Director Peter Wagner will be speaking at Claremont McKenna College in California. Details TBA.

Not near you?
Invite us to your city, college or organization.