Where Criminal Justice and Democracy Intersect: The Strange Case of Prison Gerrymandering

A New York Times editorial last weekend lifted up an important criminal justice and voting rights issue: how incarcerated people will be counted for the 2020 Census.  Although it’s hard to hear anything over the din of what’s happening in Cleveland right now, this issue lies at a critical intersection between democracy and criminal justice, neither of which is served by the Census Bureau’s recently released proposal to continue counting incarcerated persons as “residents” of the prison where they are temporarily and involuntarily confined. 

The Times correctly calls this proposal “a rejection of common sense and fairness.”  The Census Bureau is now inviting comments on its proposal to be submitted by August 1, and the Times’ assessment needs to be repeated and amplified in public comments to the Bureau.  Resources to assist in drafting a public comment are gathered on the website of our partner, the Prison Policy Initiative, a long-time leader on this issue.

Data users, advocates, political scientists, state legislatures, and many others have been calling on the Census Bureau for years to change this practice – commonly called “prison gerrymandering” – because it results in serious distortions in how our nation’s population is tabulated for redistricting purposes, and fails to reflect accurately the demographics of numerous communities throughout our country.  Because of this outdated rule, some 2 million incarcerated people are being counted in the wrong place for purposes of redistricting, undermining the equal representation principle of the 14th Amendment to the U.S. Constitution. 

In particular, using this flawed data to draw local and state districts grants the people who happen to live near large prisons extra representation in government, at the expense of voters everywhere else in the jurisdiction.  By the same token, it denies fair representation to the home communities of incarcerated persons, even though for almost all legal purposes the home community remains the legal residence of a person who is in prison or jail.  For these reasons, Dēmos along with many others has long urged the Bureau to revise its Residence Rule to tabulate incarcerated people at their home address, rather than at the particular facility where they happen to be present on Census Day.

Prison gerrymandering is not just bad policy; it is also increasingly running afoul of the 14th Amendment’s requirement of one person, one vote.  One example is a recent federal district court ruling striking down Cranston, Rhode Island’s districts for City Council and School Board elections (full disclosure – Demos along with the ACLU and the Prison Policy Initiative is counsel in the case).  The City of Cranston houses Rhode Island’s only state prison complex, the Adult Correctional Institutions (“ACI”).   The ACI contains an incarcerated population of 3,433 –all of which is counted as “residing” in city ward – Ward Six – where the entire complex is located. 

Since each ward should include 13,000-14,000 residents in order to achieve population equality, this means that almost a quarter of Ward 6 is made up of persons involuntarily incarcerated in the ACI—who are in no sense true “residents” of Ward 6.  This results in an actual maximum population deviation among all Cranston wards of approximately 28% — which, as the district court concluded, is far outside acceptable deviations from population equality.   Put differently, because Ward 6 has significantly fewer actual residents than any of the other five wards, three Ward 6 constituents enjoy more representation and political power in City government than four similar people across the district line.   

Although the City counts the ACI population to fill out Ward 6, in no way does the City actually treat the incarcerated persons there as actual constituents.  The evidence gathered in the case confirms that no elected official in Cranston has ever made campaign visits to the ACI to seek the electoral support of persons incarcerated there or to identify their needs and views about city governance. 

Indeed, even when persons at the ACI remain eligible to vote, they are generally not permitted to claim the prison as a residence for voting purposes but instead must vote absentee from their pre-incarceration home address – which is outside of Ward 6 for all but a tiny handful of the ACI population.   The absurd result is that Ward 6 claims all these individuals as residents to pad out its population count, even while state law requires the incarcerated persons to vote elsewhere.  As the district court noted, the “ACI inmate population does not participate in the civic life of Cranston or Ward Six” and the City provides “minimal services to the ACI.”  Indeed, the median length of stay for persons assigned to the ACI is only 99 days – yet they are counted as “residents” of the prison location for a decade if they happened to reside there on Census Day.

As a result, the district court concluded that the distortions produced by counting the entire ACI population in Ward 6 make Cranston’s districting plan unconstitutional under the 14th Amendment.  The City has appealed, but it’s worth noting that a similar case in Jefferson County, Florida earlier this year resulted in the same conclusion:  prison gerrymandering that produces severe distortions in population equality violates the Constitution.  

What’s clear from the ongoing controversies over prison gerrymandering is that constitutional challenges will continue so long as the Census Bureau fails to update its policies and recognize that incarcerated persons should be counted at home, not in the prison where they are involuntarily and temporarily confined.