Prison Populations, The Census, and Prison Gerrymandering

Prison gerrymandering manipulates the boundaries of electoral districts by inflating the districts with incarcerated individuals who have no real relationship to the districts where the prisons are located. Incarcerated individuals in New York are not allowed to vote. As a result, in districts with prisons and large prison populations the votes of the voters who can vote become more powerful compared to the vote of the voters in districts without a prison population. At the same time, the districts which include the actual residences and homes of the incarcerated individuals have less political representation. This dilutes their political influence as those districts become underrepresented compared with districts with incarcerated persons.

New York State in 2010 ended prison gerrymandering for its State Senate, State Assembly, and local legislative districts. Under the 2010 law, New York reallocates incarcerated persons to their homes of record instead of counting them as residents of the district in which the prison is located. Congressional districts are still redistricted under the old system. New York had proposed a constitutional amendment to include congressional districts in the reallocation of prison populations, but the amendment failed at the polls in the November 2021 election.

The Problem of Prison Gerrymandering

The United States Census Bureau, in 1790, adopted residency rules to determine where people would be counted on Census Day. The Census Bureau defined residency as the place where a person lives or sleeps most of the time. Following that rule, the U.S. Census Bureau counts incarcerated persons at their prison locations where they are incarcerated.

Approximately three-quarters of New York’s prisons are located more than 100 miles from New York City and more than 60 percent are located over 200 miles from the city. The majority of the State’s incarcerated population comes from New York City. The state’s prison population in 2000 was 77 percent Black or Latinx, while 98 percent of the state’s prison cells were located in disproportionately white State Senate districts.

As of 2019, the incarcerated population in New York was 49.5 percent Black and 24.1 percent Hispanic. 48.5 percent of the State’s incarcerated population comes from New York City.

Several upstate State Senate and Assembly district populations were inflated by incarcerated individuals which resulted in more legislative representation and political capital allocated to predominantly rural communities. Seven New York State senate districts drawn after the 2000 census met minimum population requirements only by including incarcerated people who were residents of their communities. Each of these districts was a rural district. In the districts drawn after the 2000 Census, for example, Senate District 45 gained extra political influence by inflating its population with 13,000 incarcerated individuals. In Rome, New York’s second ward, half of the population counted in the census was incarcerated.

New York’s 2010 Law

In 2010, the New York State legislature ended prison gerrymandering, the second state in the nation to do so. The legislation passed as a technical amendment to part XX of the revenue budget bill A.9710-D. Both this amendment and the full bill passed the Assembly on July 1, 2010 and the Senate on August 3, 2010. Governor David Paterson signed the legislation on August 11, 2010.

The 2010 law adjusts census data to reflect the last known address of incarcerated persons. It directs the State’s redistricting commission, the Legislative Task Force on Demographic Research and Reapportionment (LATFOR), to adjust the Census data by allocating federal and State incarcerated individuals from where they were incarcerated on April 1, 2020 back to their home communities for purposes of drawing state and local districts.

To accomplish the reallocation the law directs the New York State Department of Corrections and Community Supervision to deliver to LATFOR by September 1 of the census year, the following information for each person in its custody on Census Day: (1) unique identifier, not including the name; (2) the address of the correctional facility in which the person was incarcerated; (3) the residential address of the person prior to incarceration; and (4) any additional information specified by LATFOR.

Upon receiving the information, the 2010 law requires LATFOR to determine the census block corresponding to the street address of each incarcerated person’s residential address prior to incarceration and the census block corresponding to the address of the correctional facility. State incarcerated individuals are reassigned to their residence on record prior to their incarceration. Federal incarcerated individuals are not reassigned and are subtracted from the state total population. The law also requires LATFOR to maintain the dataset of the amended population and use it to draw state senate and assembly districts.

The 2010 law also amended the Municipal Home Rule law to clarify population base requirements for local redistricting plans. LATFOR is responsible for making the adjusted dataset available to local governments.  Incarcerated individuals from out-of-state or who had unknown pre-incarceration addresses and all federally incarcerated individuals are moved from the redistricting dataset. They are simply subtracted from the prison district and not reallocated.

Since the 2010 law was implemented before Shelby County v. Holder was decided, New York had to submit the law for preclearance to the U.S. Department of Justice. Bronx, Kings and New York counties, because of past discrimination against language minorities, were “covered jurisdictions” under Section 5 and the State was required to seek the Department of Justice’s approval before implementing any changes to their voting laws or procedures.

The law was submitted for preclearance on March 8, 2011. State officials explained that the new law would directly benefit minority voters because incarcerated individuals originate predominantly from urban districts that were subject to section 5 and are incarcerated in non-covered jurisdictions. The Department of Justice granted the preclearance on May 9, 2011, finding that the law satisfied its burden of being free from discriminatory intent or effect. The law was in place and able to be implemented for the 2011 redistricting cycle.

The 2010 Census

New York State’s redistricting cycle following the 2010 Census came with challenges. On August 26, 2010, LATFOR sent a letter to the Department of Corrections requesting information for incarcerated individuals under the department’s jurisdiction as of April 1, 2010. The data was provided by September 2010 and showed that the State had 58,237 incarcerated individuals. and included addresses prior to incarceration. LATFOR was able to reallocate 46,003 incarcerated individuals to their home addresses.

Many Republicans criticized the reallocation of the incarcerated population as an unconstitutional power play by Democrats who had taken control of the Senate the year prior. Senator Betty Little filed a lawsuit in April 2011 against LATFOR aiming to strike down the 2010 law. Senator Little’s upstate district contained eleven state correctional institutions, one federal prison, and an estimated 12,000 inmates. Her district was one of the seven districts that at the time met federal population requirements by counting inmates as residents.

In July of 2011, LATFOR’s Senate leadership announced that it would ignore the new law due to the pending lawsuit seeking to revert to the old method of counting incarcerated persons where they were imprisoned. A joint letter was subsequently submitted by six civil rights advocacy organizations to demand LATFOR comply with the law.

The Supreme Court in Albany County upheld the legislation in Little et al. v. LATFOR.

The opponents of the law had argued that the state constitution was required to count incarcerated persons where they are incarcerated as residents of that district because the Census counts them there. In support of this, the opponents referred to a report issued by the Census Bureau in 2006 that found that it would be highly difficult to determine the precise number of persons in each state for apportionment purposes unless prisoners were counted in their place of incarceration. Judge Eugene Devine stated that the report merely highlighted the difficulties in attempting to collect incarcerated person data and did not preclude the 2010 law.

The opponents also argued that the law violated the one person, one vote principle. The opponents alleged that the new law “exacerbates the weight of the vote differential between upstate and downstate counties that already exists because even with the total population being counted there remains the disparate presence in downstate counties of ineligible voters and traditionally lower voter turnout rates.” The court, however, found that the purpose of the law was to rectify the electoral inequities resulting from the practice of prison gerrymandering and ruled that these policy rationales provided the basis of enacting the law.

2020 Implementation

The 2010 law was in effect for the 2020 Census and applied to incarcerated persons under Department of Corrections jurisdiction as of April 1, 2020. As required, LATFOR undertook the adjustment process in the following stages: (1) the subtraction of all state and federal incarcerated persons, (2) reallocation or the addition of incarcerated persons whose addresses were successfully geocoded, and (3) the production of data files and reports using the adjusted data.

According to the State Department of Corrections, as of April 1, 2020, there were 52 Corrections facilities statewide located on 64 census blocks, and four federal facilities statewide located on four census blocks. According to Corrections, there were 42,292 State incarcerated individuals on April 1, 2020. Based on LATFOR’s research, there were 3,926 federally incarcerated individuals on April 1, 2020.

The geocoded addresses of incarcerated individuals’ addresses was provided by Corrections through a program that generated latitude and longitude coordinates from valid address locations. Verification of the coordinate information was done through a reverse geocoding process that either adjusted the addresses for successful geocoded or rejected them as invalid. All invalid addresses were investigated manually by visual inspection or rectifying misspelling.

LATFOR successfully reallocated for post-2020 census redistricting, 39,027 of the State’s 42,492 incarcerated persons (91.8 percent). The residences of the incarcerated persons were reallocated to 23,195 unique 2020 census blocks distributed among all 62 counties statewide. This represents 8.03 percent of all of the 288,819 unique 2020 census blocks statewide.

Other States Make Strides to End Prison Gerrymandering

Maryland passed a law to end prison gerrymandering for its 2011 redistricting cycle. The law, the No Representation without Population Act, became the law in April 2010. The Act required that the population count used to draw legislative districts not include incarcerated individuals housed in state or federal correctional facilities or those who were not residents of the state of Maryland before their incarceration. Incarcerated individuals were to be allocated to their last known residence before incarceration. Maryland’s law was broader than New York’s as it applied to both state and federal prisons and applied to local, state, and federal legislative districts. This law was upheld by the United States Supreme Court in 2012.

California, Colorado, Connecticut, Delaware, Nevada, New Jersey, and Washington also passed legislation to end prison-gerrymandering. These state laws will apply to the post-2020 redistricting cycle.

California’s law requires county and city governments to count incarcerated persons in their home communities. At the state level, the legislation recommends that the Citizen Redistricting Commission count incarcerated persons in their home community. In 2021, the Commission adopted the legislatures recommendation for the state level.

Colorado’s legislation passed in 2020 and ends prison gerrymandering in state legislative districts.

Delaware passed a very similar law to New York in 2010 to end prison gerrymandering but it was not implemented in the 2010 redistricting cycle. Incarcerated persons will be counted as residents of their home addresses when new state and local legislative districts are drawn.

Legislation in Nevada was signed into law in 2019 to ensure that people in state prisons will be counted as residents of their home addresses when new legislative districts are drawn.

New Jersey’s governor signed a bill ending prison gerrymandering in the state in 2020.

Virginia also passed its legislation to end prison gerrymandering in the state in 2020. In Virginia, persons in state prisons will be counted as residents of their home addresses when new legislative districts are drawn.

Washington passed its legislation in 2019 ensuring that people in state prisons will be counted as residents of their home addresses when new legislative districts are drawn.

Illinois also passed legislation to end gerrymandering in the state, but the implementation of the law will commence in 2030. People in state prisons will then be counted as residents of their home address.

Connecticut became the 11th state to end prison gerrymandering with a law passed in 2021. Incarcerated persons in state prisons will be counted as residents of their home addresses for the purposes of drawing legislative districts.

In August of 2021, Pennsylvania ended prison gerrymandering but not through legislation. The independent Legislative Reapportionment Commission decided to end the practice on its own. Pennsylvania became the first state to end prison gerrymandering this way.

By: Marissa Zanfardino (Marissa is a third year law student at New York Law School and is a Student Fellow at NYLS’s N.Y. Census & Redistricting Institute.)


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.