Thank you, Members of the Panel, for providing the opportunity to submit testimony. My name is Sarah Walker and I am the Chief Operating Officer of 180 Degrees, Inc. and co-founder of the Minnesota Second Chance Coalition.
I am also a member of the Board of the Prison Policy Initiative, a Massachusetts-based non-partisan, non-profit center which for the last decade has been the leading organization studying how the U.S. Census counts people in prison and working to quantify the policy and legal implications flowing from those technical decisions. Prison-based gerrymandering is also on the Second Chance Coalition’s Legislative Policy Agenda, and we supported the passage of S.F. No. 3097, introduced by Senator Higgins, and H.F. 3536 introduced by Rep. Champion during the last legislative session, which would have addressed this issue.
Each decade, Minnesota redraws its legislative districts on the basis of population to ensure that each district contains the same population as other districts. In this way, all residents are given the same access to representation and government, fulfilling the Supreme Court’s “one person, one vote” rule.
However, the Census Bureau’s practice of counting incarcerated people as residents of the prison location, instead of their home communities, results in significant distortions in achieving fair representation. Traditionally, Minnesota uses Census Bureau data for redistricting purposes, engaging in prison-based gerrymandering, even though Article VII, § 2 of the Minnesota Constitution says that a person’s residence does not change upon incarceration.
The problem in Minnesota is wide-spread; currently there are 10 Minnesota House districts where state and federal prison populations were counted as residents during the last redistricting cycle, significantly enhancing the weight of a vote cast in those districts. (See table below.) Treating prison populations as if they were actual constituents of the districts where prisons are located dilutes the votes of everyone who does not have a large prison in their district.
District Census Population Prison population Percent deviation between actual resident population and ideal district size
56A 36,812 1,746 -4.48%
20A 36,641 1,324 -3.80%
52A 36,767 1,124 -2.91%
26B 36,496 1,097 -3.58%
26A 36,780 943 -2.38%
08A 36,721 861 -2.32%
08B 36,998 847 -1.53%
30A 36,890 839 -1.80%
15B 36,573 800 -2.56%
06B 36,697 599 -1.67%
Minnesota could follow the example of other states that have already ensured the end of prison-based gerrymandering within their borders. New York and Maryland are counting incarcerated individuals at their home addresses this redistricting cycle. Similar legislation is awaiting the Governor’s signature in California and already passed in Delaware, allowing both states to take the same approach after the 2020 Census.
Although it is too late for Minnesota to collect the information necessary to count incarcerated people at their residential address, the state need not wait another decade to take action to lessen the harm of prison-based gerrymandering. Arizona and Massachusetts are already currently working on interim solutions for the current redistricting cycle to avoid exacerbating the major distorting effects of prison populations in redistricting.
There are three ways that the Panel can greatly reduce the impact of prison-based gerrymandering in Minnesota during this redistricting cycle. First is to simply remove any prison populations form the redistricting data, thus eliminating large concentrations of phantom constituents from districts with large prisons. The second option is to deliberately overpopulate any district that contains a correctional facility by approximately the population size of the correctional facility, and slightly under-populate the home districts of incarcerated individuals. Third, the Panel could identify prison populations in any redistricting tables it publishes so that the inclusion of such populations would be transparent to all, and use that information to take prison populations into account when drawing majority-minority districts.
The Panel can use the Census Bureau’s Advance Group Quarters Summary File to remove the correctional facilities from the Census counts used in redistricting. Mathematically, counting incarcerated people at the prison location has a larger vote dilutive effect than simply failing to count them at the correct home address. Below is a subset of this file in tabular form showing the 22 census blocks that contain state prisons, federal prisons, private prisons and halfway houses in Minnesota along with the populations counted within those facilities by the Census Bureau. The Prison Policy Initiative has also made point shapefiles of this Minnesota data in both ESRI and Maptitude formats . The shapefiles include direct links to Census data on the race and ethnicity of the incarcerated population within each block. (For more technical information on these files or using the Census Bureau’s group quarters summary file, you can contact Peter Wagner at Aleks Kajstura at the Prison Policy Initiative at (413) 527-0845.)
County Tract Block Correctional Population Facility Name(s) Facility Type(s)
Anoka County 050229 1020 1,305 Minnesota Correctional Facility-Lino Lakes State
Carlton County 070500 5021 1,128 MCF Willow River/Moose Lake State
Chisago County 110200 1049 980 MCF Rush City State
Hennepin County 008400 1000 25 Volunteers of America Private
Hennepin County 021602 1014 16 Damascus Way Private
Hennepin County 105600 3000 21 180 Degrees Halfway House
Olmsted County 002300 5012 954 Federal Medical Center, Rochester Federal
Pine County 950400 2132 1,315 FCI Sandstone Federal
Polk County 020700 1013 8 Red River Valley Juvenile Center Halfway House
Polk County 020700 1031 139 Tri-County Community Corrections Halfway House
Ramsey County 033200 1015 12 RS Eden (Women’s) Halfway House
Ramsey County 035500 2007 34 RS Eden (Men’s) Halfway House
Ramsey County 041602 2008 46 Volunteers of America- Female Private
Rice County 070700 3013 2,058 MCF-Faribault State
Scott County 080500 2009 588 MCF Shakopee State
Sherburne County 031500 4008 1,000 MCF St. Cloud State
St. Louis County 000300 1043 772 Federal Prison Camp Duluth Federal
St. Louis County 002000 2004 45 Bethel Work Release Program Halfway House
St. Louis County 011100 2003 150 Northeast Regional Correction Center Private
Waseca County 790500 2104 1,067 FCI Waseca Federal
Washington County 070801 1001 1,587 MCF Stillwater State
Washington County 070802 1001 448 MCF Oak Park Heights State
The second option is to deliberately overpopulate any district that contains a correctional facility by approximately the population size of the correctional facility. Conveniently for Minnesota, no individual prison is larger than the allowable population deviation. As long as multiple large prisons are not clustered together in the same district, it should be possible keep the actual population of each district within the traditional maximum population variance of 5%.
The Panel can do this through a careful use of the allowed population deviation between districts. The ideal size of House districts to be drawn this year is 39,582, with each district ranging in size from 37,603 to 41,561 (allowing for the 5% deviation in each direction). The Panel can deliberately overpopulate each district that contains a correctional facility by about the same population as the correctional facility.
For example, if a prison has a population of 1,000 people, try to draw the district that contains the prison to contain about 40,582 people, instead of the ideal size of 39,582:
2012 District: Hypothetical District with prison
Ideal District Size in 2012: 39,582
Allowable District Size in 2012: 37,603 to 41,561
(assuming ± 5 % deviation)
Census population: 40,582
Prison population: 1,000
Actual population: 39,582
Deliberately overpopulating this district by the prison population counted within it will result in a district with the ideal number of actual residents.
Third, the Panel should at the very least identify prison populations in any redistricting tables it publishes so that the inclusion of such populations could be taken into account by anyone reviewing the maps. Disclosing which districts contain prison populations as part of the Panel’s demographic analysis would be useful to the public. A healthy discussion about redistricting needs to include information about the number of people incarcerated in each district.
Critically, being aware of prison populations while drawing districts will also help the Panel minimize the vote dilutive effect of prison-based gerrymandering because it will make it less likely that several large prisons are concentrated in the same district, and it will lessen the odds that a large prison will be placed within a district that is already underpopulated.
The Panel should also take prison populations into account when drawing majority-minority districts, Arizona’s Redistricting Commission, for example, is currently excluding prison populations when analyzing districts for Voting Rights Act purposes. The Arizona Redistricting Commission states that it would exclude prisons from its Voting Rights Act Section 5 analysis in order to avoid creating “artificial majority-minority districts” comprised largely of non-voting incarcerated populations. As the Commission’s redistricting expert Bruce Adelson emphasized at a September 8 hearing: “The election analysis in determining what are effective majority-minority districts where minorities have the opportunity to elect, as we’ve talked about, cannot include felons who are incarcerated because they can’t vote.”
Crediting all of Minnesota’s incarcerated people to a few locations enhances the political clout of the people who live near prisons, while diluting voting power of all other Minnesotans. I urge the Panel to consider the above solutions to ensure that the weight of a person’s vote is not based on the happenstance of where they live.
- More information is available:
Preventing Prison-Based Gerrymandering in Redistricting: What to Watch For is a guide for advocates who want to minimize the effects of prison-based gerrymandering in their state or community: http://www.demos.org/pubs/Preventing_pbg.pdf
States are Authorized to Adjust Census Data to End Prison-Based Gerrymandering, and Many Already Do is a fact sheet summarizing the discretion given under federal law to adjust the Census for redistricting purposes: http://www.prisonersofthecensus.org/factsheets/adjusting.pdf
Importing Constituents: Prisoners and Political Clout in Minnesota, is a district-by-district analysis of prison-based gerrymandering in Connecticut state legislative districts: http://www.prisonersofthecensus.org/mn/report.html