Help End Prison Gerrymandering Prison gerrymandering funnels political power away from urban communities to legislators who have prisons in their (often white, rural) districts. More than two decades ago, the Prison Policy Initiative put numbers on the problem and sparked the movement to end prison gerrymandering.
Census Bureau should conduct small-scale tests of methods for collecting home addresses of incarcerated people to determine how best to implement a revision of the Residence Criteria and Residence Situations for people who are in correctional facilities on Census Day.
Today, we submitted a joint comment letter from 36 criminal justice, voting rights, and census and data equity advocate organizations in response to the Census Bureau’s recent federal register notice about small-scale tests as part of its planning for the 2030 Census. While the Bureau is only seeking generic clearance — not specifying any particular tests — we took this opportunity to highlight the need for early testing of how best to count incarcerated people at home and end prison gerrymandering .
Prison gerrymandering is a problem created because the Census Bureau incorrectly counts incarcerated people as residents of their prison cells rather than their home communities. As a result, when states use Census data to draw new state or local districts, they inadvertently give residents of districts with prisons greater political clout than all other state residents.
As the comment notes, it has been nearly 20 years since the Census Bureau last considered how it could go about counting incarcerated people at their home addresses. It is past time that that Bureau update that analysis, and the comment letter proposes three specific methods that the Cesnus Bureau should test. If you’d like to read the (rather technical) details, the full letter is available at this link.
The Census Bureau’s current method of counting people in prison and jail is prone to errors with sizable consequences. Counting incarcerated people at home can produce more accurate data.
It is no secret that the Census Bureau counts incarcerated people in the wrong place — a prison or jail cell — instead of in their actual communities. As a result, when states and local governments use census data to draw new political boundaries during their redistricting process, they inadvertently give residents of districts with correctional facilities greater political clout at the expense of everyone else — a problem known as prison gerrymandering.
For two decades, we’ve worked to fix this problem, but it isn’t the only way that the Census Bureau gets it wrong when counting incarcerated people. The Bureau also routinely mistakenly and unknowingly places prisons and jails in the wrong location in its data or dramatically miscounts them, further distorting political representation across the country. This problem exacerbates prison gerrymandering and warrants further scrutiny.
We examined 2020 Census data to better understand the impact of this often overlooked error. We found that it is not a rare mistake; it is perpetuated across the country and across the decades. And the solution is clear: by counting incarcerated people as residents of their homes the Census Bureau will minimize technical inaccuracies while also ending prison gerrymandering.
The Bureau is wrong to count incarcerated people as if they were residents of correctional facilities
Before looking at the impact of the Census Bureau misplacing or miscounting entire prisons and jails, it is worth revisiting its flawed reasoning for counting incarcerated people at those facilities in the first place.
The Bureau says that it determines where to count people using its “usual residence rule” which says it counts them where they live and sleep most of the time. In attempting to justify its decision to count incarcerated people in prisons and jails, it argued, “counting prisoners anywhere other than the facility would be less consistent with the concept of usual residence since the majority of people in prisons live and sleep most of the time at the prison.”
As we’ve noted before, this is simply not true. Short stays in prison are common. Even people incarcerated away from home for a year or longer do not stay in one place; they are usually moved between multiple facilities. While they are being shuffled between facilities, incarcerated people generally maintain a usual residence elsewhere; their pre-incarceration home remains the only actual stable address. There are plenty more reasons — backed up by hard data — that make clear that by counting incarcerated people at prisons, the Census Bureau is choosing the least accurate option.
The Bureau’s flawed way of counting incarcerated people makes its other mistakes even worse
We often think of the census as a gold standard for population data. And while it is, on the whole, fairly accurate, no matter how careful you are, perfection is not possible when attempting to count over 330 million people.
However, when the Bureau counts an entire prison or jail — which can account for hundreds or thousands of people — in the wrong place, it can have a big impact.
The Bureau routinely misplaces correctional facilities
During the 2020 count, the Census Bureau counted at least one prison or jail in every state in the wrong place. Most of the populations counted erroneously were small, but occasionally, the census reported hundreds or thousands of incarcerated people in the wrong place.
Here are just a few examples from a single state, New York.
The Census Bureau didn’t count any people at Adirondack Correction Facility — a place that holds bout 300 incarcerated people. However, roughly the same number of incarcerated people were counted in a state forest about 175 miles away.1 Our best guess is the Bureau accidentally counted this facility in the wrong place.
Woodbourne Correctional Facility appears to have been counted twice — once in its actual location,2 and also in the empty field next door.3
Empty fields are a common place for the Census Bureau to mistakenly report correctional populations. In Wyoming County, N.Y., rather than counting them in the facility where they were confined, the Bureau counted nearly a thousand incarcerated people in a field,4 and another group of over 800 was reported just across the street from their facility’s location — also in empty fields. This is a perennial problem with the census. In the 2000 Census, the Bureau reported nearly 1800 incarcerated people — likely from the Wyoming Correctional facility in a random field halfway across the county.5
These examples are all just from a single state, and these problems are not confined to New York. The scope of the problem becomes clear when considering that these miscounts and misallocations exist in every state.6
Some might be tempted to argue that the Census Bureau might make similar mistakes if it counted incarcerated people in their home communities rather than counting them at the prison or jail. This would almost certainly happen. Despite its best intentions, the Bureau is not perfect and will make mistakes.
The difference is the size and scale of the mistakes. If the Bureau places a prison that holds hundreds of people in the wrong place, it can severely skew the counts. However, if it counts a single or even several incarcerated people in the wrong place, it would have a negligible impact on political representation and the underlying Census data.
The Bureau’s mistakes create incorrect race and ethnicity data
The prison accounts for a third of the Parish’s census population, so any population data reported for the prison will heavily influence the demographic picture of the parish. The 2020 Census counted over 4,000 Black people incarcerated at Angola as if they were white. At that time, white people actually accounted for less than a quarter of Angola’s population. Not only does such an error further distort the demographic data of the surrounding parish, but it blatantly obscures the disproportionate toll that mass incarceration places on Black communities.
Furthermore, by creating a false picture of the population in a facility, it also makes it harder for states ending prison gerrymandering to comply with voting rights laws that require accurate race and ethnicity data for drawing districts.
Misplacing prisons & jails exacerbates prison gerrymandering and diminishes trust in the Census Bureau
The Census Bureau’s current approach — counting incarcerated people as if they were residents of a correctional facility — creates inaccurate pictures of communities across the country and leads to prison gerrymandering. This problem is made worse by the Bureau’s persistent, multi-decennial errors that place correctional facilities in random other locations. A single error (and it’s never limited to single errors) haphazardly injects hundreds or thousands of people into a random community’s redistricting data.
At a time of increasing distrust in the census, the Bureau should ensure that communities view its data as accurate and feel represented in the process and final results.
The most accurate and equitable solution is to allow incarcerated people to self-enumerate.
Officials at the Census Bureau may worry about how they would implement a change to count incarcerated people in their home communities. The good news for them is they already have a process in place that they use for all other people in the U.S.: simply allow incarcerated people to fill out their census forms themselves — or, as the Bureau calls it, “self-enumeration ” — rather than relying on the facility administrators to provide the date for them. In fact, this is a solution that was already part of the Census Bureau’s plan for the 2020 Census, giving more facility types this option than in 2010.7
States that currently reallocate incarcerated people back home in their redistricting data must now rely on their internal facility or correctional agency records to access home address information. These records are sometimes incomplete. By allowing incarcerated people to self-enumerate, the Bureau would provide states with more accurate and complete data and alleviate state officials of a significant burden.
Not only would this change allow incarcerated people to provide their addresses directly, it would also improve race and ethnicity reporting. Currently, correctional facilities don’t report race and ethnicity data in accordance with census standards. For example, in 2020, correctional facilities failed to report Hispanic status for over a quarter of the people they reported. Moving to self-enumeration would mean a more complete and accurate picture of the demographics of people who are incarcerated.
Of course, despite this, there would still be some non-responses for various reasons, just as people on the outside fail to participate in the census for various reasons. In these circumstances, facility records could still be used to fill in any blanks left after everyone had the opportunity to self-enumerate, ensuring that the 2030 Census is more accurate than prior decades.
If the Bureau chooses not to allow self-enumeration to collect home address data, the Census Bureau can follow the limited-but-tested processes used by the states and rely solely on the data already in the records collected by correctional agencies. Information missing from the records can either be imputed, as the census already does for race and ethnicity data, or, as some states do, continue to count people at the facility if all other efforts fail.
The states have pioneered a variety of paths to counting incarcerated people at home, and the Census Bureau has many more tools, giving it the ability to go far beyond state efforts and create an even more accurate enumeration. At the very least, the Census Bureau can replicate state efforts to count incarcerated people at home.
If the Census Bureau wants to be accurate, it should try something different: count incarcerated people at home
The Bureau’s current way of counting incarcerated people isn’t working. Not only does it dramatically distort political representation, but persistent errors in this data create redistricting chaos far outside the prison walls.
Counting people at home would produce census data that is, on the whole, a more accurate reflection of the population. There are concerns that counting people at home will lead to some inaccuracies because of missing home address data in facility records or non-response from incarcerated people asked to fill out the census themselves. However, the current methods used by the Census Bureau have already resulted in serious inaccuracies in the data. By attempting to count incarcerated people at home, the Census Bureau would be abandoning a seriously flawed status quo and aiming for a more accurate census.
Counting incarcerated people at home is crucial to ensuring equal representation, particularly for communities disproportionately impacted by mass incarceration. In fact, nearly half the US population is now living in a place that has rejected the Bureau’s redistricting data in order to avoid prison gerrymandering. It is time for the Bureau to count incarcerated people at home in order to create a more accurate 2030 Census.
You can browse through the errors noted in our annotations of 2020 Census correctional populations in our “Correctional Facility Locator 2020” tool, searchable by state. Or in the meantime, here are two more examples:
In Michigan, the Kalamzoo County Jail was expected to have a population of at least 400 people; the Office of the Sheriff states that, “The jail regularly supervises over 400 inmates.” Yet, Block 26 077 001801 1000, clearly located at the Sheriff’s office and jail, counts only 27 incarcerated people. The only other block in the vicinity with any incarcerated population has only 12 incarcerated people counted in it.
Conversely, in North Carolina, Block 37 165 010400 1004 is clearly located at Scotland Correctional Institution and counts 2,515 incarcerated people at that facility. North Carolina’s DPS website states that Scotland CI has a maximum capacity of 1,756. According to North Carolina DPS’s custom offender reports, there were 1,654 people in Scotland CI on June 30, 2020. There are no other facilities or blocks nearby to which the missing 1,000 people were attributed. ↩
We’ve identified over 200 cities and counties that have taken action to avoid prison gerrymandering and some local governments that still continue to base representation on flawed Census Bureau data.
The movement to end prison gerrymandering started with local governments. Because of their relatively small size, city and county governments often experience the most distortive impacts of the problem. When drawing districts in these communities, a single prison can easily account for an entire city council or county commission district. This not only makes it impossible for officials to draw fair districts, but it also gives some residents as much as five times1 the political clout of other residents of the community.
Prison gerrymandering is a problem created because the Census Bureau incorrectly counts incarcerated people as residents of their prison cells rather than their home communities. As a result, when states and local governments use Census data to draw new political boundaries, they inadvertently give residents of districts with prisons greater political clout than all other residents.
It can be incredibly difficult to know whether a local government has ended prison gerrymandering, but we’ve done much of that legwork for you. We have two pages covering our prison gerrymandering research on local governments — spanning three decades:
Building these lists takes painstaking research. Very few states have centralized records of local redistricting maps, so we reached out to nearly every city, county, and other local government on those lists to figure out whether they ended prison gerrymandering. We gather the maps, analyze the population data, sift through council meeting notes, and interview local officials to come to our conclusions. Sometimes, regional or state organizations do similar research to supplement our efforts, or individuals working in or living in these areas will add their findings to the collection.
There are two important notes about these lists:
As you can guess, it’s impossible to do this research for all of the nearly 40,000 town, city, and county governments in the US. So, we focus on those that are most likely to yield significant or interesting results. As a result, both of these lists are almost certainly undercounts of the places that do — or don’t — engage in prison gerrymandering.
Many states that have passed prison gerrymandering reform laws also cover local governments under those measures. To make the best use of our time and resources, we didn’t research to confirm whether cities or count in those states ended prison gerrymandering, so they don’t appear on these lists.
Because of these two factors, these lists shouldn’t be considered comprehensive or used for any conclusive comparison between decades. These lists are only a partial reflection of the general decline in the number of local governments engaging in prison gerrymandering, and the general increase in local governments avoiding prison gerrymandering, and don’t show the full story of prison-hosting communities increasingly rejecting prison gerrymandering.
We also encourage you to check out our Pathfinder tool, for more information on the issue, including dozens of state-by-state and district-by-district reports, hundreds of articles and fact sheets, and more spread across our website (and even some notable resources from other organizations, too).
These resources provide important information and context about prison gerrymandering for local officials, journalists, and advocates. It can help to show the scale of the problem in their communities, as well as put a spotlight on those places that have taken on the task of solving this problem.
Footnotes
For example, Juneau County, WI contains one of the most prison gerrymandered county commission districts in the nation, where incarcerated people account for roughly 83% of the district’s population. As a result, 17 actual residents of this district have as much political clout as 100 residents of any other district in the county. ↩
Census Bureau should stop its one-size-fits-all approach to privacy protections and refrain from adding “noise” when it reports the number of incarcerated people; states need accurate counts to end prison gerrymandering
Last week the Prison Policy Initiative and Common Cause sent a letter to the Census Bureau urging it to publish data in a way that makes it easier for states to end prison gerrymandering.
Prison gerrymandering is a problem created because the Census Bureau incorrectly counts incarcerated people as residents of their prison cells rather than their home communities. As a result, when states and local governments use Census data to draw new political boundaries, they inadvertently give residents of districts with prisons greater political clout than all other residents.
After the 2020 count, the Census Bureau added another wrinkle to prison gerrymandering reform by taking a new approach to its data privacy protections that unintentionally made it harder for states to adjust their redistricting data to count incarcerated people at home. Every decade, the Bureau’s privacy protections ensure that no single person can be identified in the census data, and that everyone’s answers on the census forms remain private. This privacy guarantee is a statutory requirement, but the Census Bureau is free to achieve this privacy through whichever method it sees fit.
For the 2020 Census, the Bureau picked a new privacy method called differential privacy, which injects “noise” into the population data published by the census. This “noise” essentially takes the correctional facility populations and adds or subtracts a statistically-determined number of people in the population data published for the census. This had the consequence of undermining states’ efforts to address prison gerrymandering, as the letter explains:
There are 19 states that have formally addressed prison gerrymandering, these states account for nearly half (49.6) of the US population.
Although the approaches to population adjustments differ among the 19 states, they all rely on accurate counts of correctional facility populations to adjust redistricting data to avoid or limit prison gerrymandering.
In each case, the Bureau’s application of privacy protections added an unnecessary impediment to their data adjustment efforts. Facility populations are public. In most states, they are published – along with other demographic statistics – on state websites.
The letter concludes by asking the Bureau to stop undermining state efforts to end prison gerrymandering:
The Bureau should update its residence criteria to count incarcerated people at home. But if the Bureau continues to count all incarcerated people as if they were residents of the correctional facility they happen to be held at on census day then the correctional facility populations should be held invariant (or kept unchanged) to decrease the burden on the states that are correcting their redistricting data to count incarcerated people at home or last known address.
Because of their relatively small size, city and county governments often experience the most distortive impacts of prison gerrymandering. When drawing districts in these communities, a single prison can easily account for an entire city council or county commission district, making it impossible for officials to draw fair districts. So it should come as no surprise that, while state government efforts to address the practice receive most of the attention, the movement to end prison gerrymandering started and has been most successful at the local government level, in mostly small, rural, conservative cities and counties. The 2020 redistricting cycle continued this pattern.
Prison gerrymandering is a problem created because the Census Bureau incorrectly counts incarcerated people as residents of their prison cells rather than their home communities. As a result, when states and local governments use Census data to draw new political boundaries, they inadvertently give residents of districts with prisons greater political clout than all other residents. Progress in addressing this problem has been so swift in recent years that the National Conference of State Legislatures recently called the movement to end prison gerrymandering “the fastest-growing trend in redistricting.”
The distortions caused by prison gerrymandering have real-world implications in these communities. It gives some residents a louder voice in the political process, which means local government often prioritizes the issues they want addressed — whether it be deciding which potholes get filled, where to expand sewer services, or which parks to improve — over the needs of residents who live in all other parts of the town or county.
Hundreds of local governments ended prison gerrymandering on their own, with still new ones joining their ranks this decade
In addition to the hundreds of cities and counties that avoided prison gerrymandering after the 2000 and 2010 Censuses (decades when zero and two states, respectively, tackled the issue), we identified an additional 21 local governments1 scattered across 10 states that addressed the problem for the first time after the 2020 Census.2
Local governments that ended prison gerrymandering for the first time after the 2020 Census
State
Local Government
Arkansas
Izard County
Arkansas
Jackson County
Delaware
City of Willmington
Florida
Columbia County
Florida
Jefferson County
Georgia
Coffee County
Georgia
Emanuel County
Georgia
Long County
Georgia
Monroe County
Georgia
Montgomery County
Georgia
Wayne County
Louisiana
Allen Parish
Louisiana
Catahoula Parish
Oklahoma
Atoka County
Oklahoma
Woodward
Rhode Island
City of Central Falls
North Carolina
Granville County
South Carolina
Williamsburg County
Tennessee
Hardeman County
Tennessee
Johnson County
Tennessee
Morgan County
It is important to note that this does not represent the complete list of communities that addressed prison gerrymandering; it is a list of places we identified that have done so for the first time this decade — and an incomplete list at that. That’s because many of the laws passed in recent years ending prison gerrymandering in state legislative districts also apply to city and county districts. For example, California’s law, which ended the practice, applies to local governments, too, meaning the communities containing the 34 state prisons and 114 local jails also avoided the practice. As a result, our analysis certainly underrepresents the number of local governments that no longer take part in prison gerrymandering.
Local governments got creative to avoid prison gerrymandering
To avoid prison gerrymandering, communities found creative solutions to overcome the Census Bureau’s flawed counting methods. Regardless of which method they employ, they all led to the same result: districts more representative of the community’s actual population.
There are several common ways local governments avoided prison gerrymandering:
Adjusting their redistricting data to remove the out-of-town population counted at the prison;
“Overpopulating” the district with the prison so that it contains the same number of true residents as every other district in the city/county;
Cutting holes in the maps around the prison to exclude the prison population altogether;
Trying to split prison facilities across multiple districts to even out the impact.
In addition, a handful of communities avoid prison gerrymandering by forgoing redistricting altogether, a particularly common practice when a new prison has been built in the area since it last drew new district lines. Communities can maintain equal representation without redistricting if their population outside of the prison has not changed. We’ve discovered a number of places that have chosen not to redistrict because the only significant change in the total Census population is the correctional population.
It is important to remember, though, even when local governments can solve their own prison gerrymandering problems, their jurisdiction ends at their city or county line. While they can ensure that prison isn’t included in their redistricting data, they can’t actually count incarcerated people in their home communities, meaning those people often aren’t included anywhere for redistricting purposes.
Some local governments face unnecessary obstacles
Even as more governments take on the task of ending prison gerrymandering in their local districts, some face arcane state rules that turn these simple reforms into complex mazes.
In Wisconsin, for example, leaders in Juneau County recognized prison gerrymandering was a problem and wanted to fix it, so they sought guidance from the state on how to do that but were told their hands were tied. They learned an outdated and confusing 1981 attorney general’s opinion directed local governments to draw new district lines without excluding incarcerated people. It requires local governments to take part in prison gerrymandering, even though they didn’t want to. Wisconsin’s attorney general can and should issue a new opinion that gives local governments the freedom to finally fix this problem, but until he does, their hands are tied.
Rural municipal and county districts still suffer from the largest impact
Despite this progress, there are still some communities that face huge impacts of prison gerrymandering.
Local governments’ efforts to end prison gerrymandering are sometimes stymied by state governments, such as in Wisconsin. But even when state governments reverse course to support avoiding prison gerrymandering, some counties stay adrift on their previous heading. For example, since 2016, Tennessee has explicitly allowed (Tenn. Code Ann. S 5-1-111) counties to deviate from the Census population to avoid prison gerrymandering. Since then three counties (Hardeman County, Johnson County, Morgan County) that had engaged in prison gerrymandering have stopped doing so. Unfortunately, most counties impacted by prison gerrymandering in the state continue to use unadjusted Census data for redistricting and, as a result, draw some of the most unequal districts in the country.
As a result, Tennessee, along with Wisconsin — which, as we explained above, prohibits local governments from ending prison gerrymandering — account for six out of the eight cities or counties in the country where correctional facilities make up at least half of local government district
Cities or counties where incarcerated people account for more than 50% of the population
County or City
State
Percent of the population in a City Council or County Commissioner district counted at a correctional facility
Lake County
Tennessee
86%
Jackson County
Wisconsin
83%
Wayne County
Tennessee
79%
Juneau County
Wisconsin
79%
Mansfield City
Ohio
66%
Adams County
Wisconsin
64%
Marion City
Ohio
61%
Lauderdale County
Tennessee
56%
Rural residents would benefit from the Census Bureau solving the problem
Despite the progress made in recent years to address the problem, residents of rural communities continue to be disproportionately harmed by prison gerrymandering and stand to uniquely benefit if the Census Bureau were to solve the problem.
There are three reasons for this:
As we explained in this briefing, because of their small size, rural cities and towns that contain prisons often have their local government districts extremely distorted by prison populations.
Rural state governments have been slower to adopt statewide reforms to address prison gerrymandering. As a result, our analysis earlier this year found most of the worst state legislative prison gerrymanders are located in rural areas. This analysis also shows the issue defies many preconceived notions about the partisan impact of the problem, with Democrats controlling six of the most gerrymandered districts, while Republicans control four.
Mass incarceration is no longer just a problem in urban areas. Our analysis of where incarcerated people come from found that rural areas often have the highest incarceration rates in their state. These increasing incarceration rates mean that while some rural communities see their political boundaries dramatically distorted by prison gerrymandering, others lose political clout because of the problem.
If, in 2030, the Census Bureau chose to count incarcerated people in their home communities instead of in prison and jail cells, all three of these problems would be solved for rural communities. This change would save these communities time and money, ensure they didn’t lose political clout to prison gerrymandering, and help communities that contain prisons create local government boundaries that accurately reflect their true population.
Looking toward 2030
Roughly half of all U.S. residents now live in a city, county, or state that has ended prison gerrymandering. Even as more local governments and states address the issue, they can only do so much. Prison gerrymandering is a national problem that crosses jurisdictional boundaries. No city, county, or state can completely fix this issue on its own. Only the Census Bureau can fully solve this problem nationwide. As it begins to plan the 2030 count, it should finally listen to the growing consensus among state and local governments and end prison gerrymandering across the country.
Keep reading for an explanation of why this number undoubtedly underrepresents that number of communities that addressed the problem for the first time after the 2020 census. ↩
Rhode Island’s adjusted data did not account for Wyatt Correctional Facility, located in Central Falls. So the state’s partial data adjustment was not usable for their local redistricting. ↩
2030 Census Advisory Committee's inaugural meeting ended in recommendations that future agenda include discussions of the Census Bureau's approach to counting incarcerated people in the 2030 Census.
On Friday the 2030 Census Advisory Committee held their inaugural meeting and voted to put prison gerrymandering on its agenda. (That’s actually more exciting than it sounds.) The recommendation to discuss the topic in depth was both in reaction to a public comment submitted to the Committee by the Leadership Conference on Civil and Human Rights on behalf of its prison gerrymandering working group, as well as Committee members’ own expertise identifying prison gerrymandering as an issue that needs to be addressed for the 2030 Census.
The topics of discussion for the Committee’s first meeting were largely (and procedurally) limited to on hard-to-count and historically undercounted populations, but committee members Jeri Green (National Urban League’s 2020 Census Black Roundtable) and Ben Williams (National Conference of State Legislatures) requested that the committee address two issues related to prison gerrymandering at a future meeting:
The Census Bureau’s residence criteria for incarcerated people, which is based on an outdated interpretation of the Bureau’s “residence rule”, and means that people are counted as if they were residents of a correctional facility.
The 2030 Census Advisory Committee is charged with “advis[ing] the Census Bureau in conducting an accurate decennial census.” And it is in that role that it recommended that the “a future committee meeting include a discussion on residence criteria for those who are incarcerated. … [And] that there be a future discussion regarding the impact of disclosure avoidance policies on the 2030 Census.” These issues accounted for two out of the three topics recommended for discussion by the Committee.
The 2030 Census Advisory Committee operates in an advisory capacity to the Census Bureau, and as such their focus on prison gerrymandering sends a strong message to the Census Bureau that it must consider how to count incarcerated people at home in the 2030 Census.
Last week, The Leadership Conference on Civil and Human Rights — a coalition of more than 240 national organizations to promote and protect the rights of all persons in the United States — sent a letter to the Census Bureau urging it to end prison gerrymandering. Specifically, the group called on the Bureau to “update the current residence rules to maintain the accuracy of the census, ensure fair representation of incarcerated individuals within their home communities, and to prioritize equity among all residents regardless of their incarceration status.”
Prison gerrymandering is a problem created because the Census Bureau incorrectly counts incarcerated people as residents of their prison cells rather than their home communities. As a result, when states use Census data to draw new state or local districts, they inadvertently give residents of districts with prisons greater political clout than all other state residents.
The letter follows up on a meeting with staff from the Census Bureau’s Population Division, which oversees how the “residence rule” — is applied to people in different living situations at Census time. To end prison gerrymandering, the Bureau needs to update how it applies its residence rule for incarcerated people to ensure an accurate Census. While is a plethoraof reasonsit should do so, the letter focuses on 5 key points:
The residence rules must reflect the real-life circumstances of incarcerated persons and the needs of the states and communities in which they live.
Representation and, therefore, democracy, are distorted by counting incarcerated persons where they are housed on Census Day rather than in the home communities to which most will return.
Incarcerated people are not part of the social and economic fabric of the communities surrounding the facilities where they are serving their sentences.
Criteria used to apply the residence rule to incarcerated people should be applied in the same manner as for other people away from home on Census Day.
Equity requires counting incarcerated people at home.
For details on all 5 points, check out the letter (or PDF).
We explore state redistricting reports to show that, despite facing immense challenges, states were remarkably successful ending prison gerrymandering after the 2020 Census.
During the 2020 redistricting cycle, more than a dozen states took it upon themselves to do what the Census Bureau has refused: end prison gerrymandering. Through their redistricting process, they worked to unwind the flawed way the Bureau counts incarcerated people — as residents of a prison cell — and instead count them where they actually live, creating legislative districts that more accurately reflect actual resident populations.
Prison gerrymandering is a problem created because the Census Bureau counts incarcerated people as residents of a prison cell rather than at their true communities. When states then use Census data during their redistricting process, they artificially inflate the population of prison communities and give them a larger voice in government.
It is an important question with huge implications; 19 states are already poised to adjust their redistricting data to address prison gerrymandering in the 2030 redistricting cycle. These questions can help states evaluate how they can do better during the 2030 redistricting cycle. It can guide states that haven’t yet adopted anti-prison gerrymandering laws but are considering them. Finally, it can illustrate how states are limited in their ability to solve this problem in ways the Census Bureau is not.
To answer these questions, we reviewed redistricting reports and data from 13 states that addressed prison gerrymandering after the 2020 census to understand how many people in state prisons they attempted to count in their home communities and how many they were able to reallocate successfully.
Perfection isn’t possible, so it’s not the goal
Before diving into our analysis, it is important to remember that it is impossible and unrealistic for states to successfully reallocate everyone in their prisons to their home communities.
There are three primary reasons why a state can’t count an incarcerated person at their home address:
The person’s last address is in a different state: States don’t have the power to reallocate people across state lines, so they cannot count people from out of state in their home community.
The state has decided not to reallocate an incarcerated person for a policy-based reason: Unfortunately, some states have implemented arbitrary limitations on their reforms by excluding certain incarcerated people from being counted in their home communities. For example, Rhode Island did the mapping work for all people incarcerated in the state correctional facilities. However, it chose to count incarcerated people with more than two years left on their sentence at the prison location in the redistricting data. While we and other organizations think this is misguided, it is the choice policymakers have made, so redistricting staff could not count them in their home community.
Erroneous or missing data: State redistricting officials often rely on their Department of Corrections (DOC) to get address data for incarcerated people. Generally, that data contains everything they need to implement their reforms. However, sometimes data is missing or erroneous. Apart from typos and other record-keeping errors in home address data, there are a few situations where home addresses are missing altogether. For example, if an incarcerated person indicated they were unhoused before coming to prison, the DOC will likely leave the address field blank or note “homeless.” This may be an appropriate response for prison staff, but it creates problems for redistricting officials.1While there are some steps redistricting officials can take to reduce this number, it often takes time, which is something they don’t have much of.
These limitations make it impossible for a state to reallocate 100% of people in its prisons to their home communities, so when they evaluate their success, they shouldn’t grade themselves against perfection. Instead, they should ask themselves: “Of the people it was possible to reallocate, how many did we reallocate?”
How did states do?
For our analysis, we looked at two key numbers:
The state’s success rate for reallocating people that it was possible to count at home.2
The total percentage of people in state prisons counted at their home address.
Our analysis 3 shows that most states did remarkably well. This is noteworthy because 2020 was a notoriously difficult redistricting cycle, in which Census data was delayed, and many states adopted prison gerrymandering reform laws late in the decade.
Of the 13 states we analyzed, 10 published their methodology. Nine out of 10 were able to reallocate at least 80% of people with addresses that met the states’ criteria, with Nevada being the lone exception. Impressively, California successfully mapped every address that it attempted.
2020 redistricting state reallocation success rates
State
Success rate for reallocating people who’s address met the state’s criteria
Percentage of people in state prisons successfully counted at their home address
California
100.0%
99.7%
Colorado
88.5%
82.1%
Connecticut
96.9%
87.6%
Delaware
85.5%
79.2%
Maryland
84.8%
77.0%
Montana
96.3%
45.3%
Nevada
68.9%
64.1%
New Jersey
95.3%
89.5%
New York
91.9%
91.9%
Pennsylvania
State did not publish methodology
60.6%
Rhode Island
83.1%
41.0%
Virginia
State did not publish methodology
88.2%
Washington
State did not publish methodology
91.4%
States were so successful that even in two of the three states where we could not fully analyze the success rate because they didn’t publish their methodology — Virginia and Washington — we can tell that roughly 90% of their total state prison population was successfully reallocated to their home communities. Importantly, like Rhode Island (which we explained above), Pennsylvania also made a policy choice to exclude certain incarcerated people from these reforms, instantly lowering the portion of its prison population that could be counted at home.
Getting to 100%
While states have done a remarkable job implementing these reforms, every incarcerated person should be counted in their home communities for redistricting purposes. What can government officials do to accomplish this?
Start collecting, assessing, and cleaning data early: Getting good data from the Department of Corrections is critical to successfully implementing reforms. That’s why state redistricting officials should work with that agency now to ensure it is collecting the data they’ll need during the next redistricting cycle. Similarly, states that haven’t yet addressed prison gerrymandering should adopt and implement policies to do so now. This will ensure prisons collect and maintain high-quality address data for people they incarcerate.
Don’t adopt arbitrary limitations on reforms: When adopting anti-prison gerrymandering reforms, states should include all incarcerated people, regardless of the length of their sentence. Even if a person is likely to spend decades behind bars, they’re still not a resident of the legislative district that contains the prison4 — they don’t have ties to that community, they don’t call it home, and if they have a problem that they want lawmakers to fix they’re not likely to call the legislator who represents the prison but rather the elected officials from their home community — so the town containing the prison shouldn’t get an outsized voice in government simply because the facility happens to be located there.
The Census Bureau must change its policies: Even if states do everything in their power to address prison gerrymandering, there will still be some incarcerated people that they can’t count in their home districts, most frequently those who come from different states. Similarly, but not covered in this analysis, if a state contains a federal prison, redistricting officials cannot reallocate those incarcerated people to their home communities. The Census Bureau is the only government entity with the power to address this problem completely and across state lines.
This analysis shows that states have been incredibly successful in implementing reforms to address prison gerrymandering. History suggests that if the Census Bureau fails to resolve this problem, states will do even better during the 2030 redistricting cycle. This should give confidence to places considering similar reforms and motivate them to take action now.
We measured the states’ success in two different ways:
Success rate for reallocating people who met the state’s criteria for reallocation
Total percentage of people in state prisons successfully counted at their home address
It is worth noting that we only provided partial analyses for three states that have ended prison gerrymandering for the following reasons: Pennsylvania and Virginia have not made their methodologies publicly available; Washington has some documentation available, but it does not provide the number or percentage of people successfully reallocated, or the total prison population the state began with.
Our state-specific methodologies are below.
California
The state documentation reports a total of 122,730 people in state prisons across the state (MacDonald presentation, p. 4). Ultimately, the adjusted dataset includes 122,393 people in state prisons who were successfully reallocated home in California (Statewide Database). Based on the 122,393 successfully reallocated people, we can assume that the remaining 337 people had a last known place of residence outside of California or an address that “cannot be determined” (Final Maps Report, p. 20). These people were counted in an “unknown geographical location” and excluded “from the population count for any district, ward, or precinct” (Final Maps Report, p. 20).
State documentation:
Final Maps Report from the California Citizens Redistricting Commission
The Colorado state documentation reports 17,506 people in prisons across the state. This does not include any people in federal prisons.5
1,270 people in state prisons reported a known previous address outside of the state and, therefore, were counted at the facility (p. 3). 6
That leaves a remaining 16,236 people in prison with Colorado addresses. 1,872 of those people in state prisons with in-state addresses had unusable addresses and were subsequently counted at the facility (p. 3). The state defined “unusable” as addresses with “no or incomplete street address information” or people who reported being homeless before incarceration (p. 3).
The remaining 14,364 people in state prisons could be reallocated home in Colorado (p.3).
The Connecticut state documentation states the Department of Corrections reported a total of 11,853 people in state prisons across the state (p. 4). An additional 1,057 people were in federal prisons and subsequently counted at large (p.4).
392 records had incomplete or unusable addresses and were subsequently counted at large (p. 5). The reasons for these 392 unusable addresses were listed as: 301 records did not include a street address, 75 records did not have a street number, four records listed a PO box in the street address field, eight records listed “an apartment complex where the precise address location and [Census] block could not be determined,” and four records were for “temporary locations (e.g., hotel, shelter, or campground) where the precise census block location could not be determined” (p. 5). 687 records had an out-of-state address and were subsequently counted at large (p. 4). 7 10 records were for people serving life-without-parole sentences and subsequently counted at the facility. 49 records were for people under 18 years old and were “excluded” (p.3).
Of the remaining 10,715 records, 333 “could not be geocoded in cleaned dataset” (p. 7). 8 The remaining 10,382 people in state prisons were successfully reallocated home in Connecticut (p.6).
The Delaware state documentation reports that a list of records from the Department of Corrections included 4,748 records (p.3). 350 of these records were for out-of-state addresses and were excluded from the counts of the facility and the population totals (p. 3). There were 4,398 remaining in-state addresses. 637 of these records were “without a good geocoded home address” and were counted at the facility (p. 3). The remaining 3,761 people in state prisons could be reallocated home in Delaware (p.3).
The Maryland state documentation reports a total of 19,802 people in prisons across the state (p. 1). This does not include any people in federal prisons. 1,821 of those people had out-of-state addresses and were not counted. That leaves a remaining 17,981 people with in-state residence. Out of those 17,981 people, 2,793 people had “no address” – meaning they did not have “a valid prior address or they did not provide a previous residential address” (p. 1) – and were subsequently counted at the facility. The remaining 15,242 people in state prisons could be reallocated home in Maryland.
The Montana state documentation reports 2,840 people in state prisons (p. 5). 1,505 were deemed “not appropriate for reallocation,” which included: 1,389 records with no addresses, 56 with out-of-state addresses, 28 with addresses in jails or other similar facilities, 20 indicating “transience,” and 12 with P.O. boxes as addresses (p. 5). The remaining 1,335 people in state prisons had addresses deemed “appropriate for geocoding” (p. 2). 1,286 were successfully geocoded, and 49 were not geocoded, therefore excluded from reallocation (p. 2).
The Nevada state documentation reports 12,214 records of people in Nevada state prisons (Killian presentation, p. 3). 860 of these addresses were for out-of-state residences and were not counted (Killian presentation, p. 9). This leaves 11,354 records with in-state addresses. Ultimately, the adjusted dataset includes 7,826 people in state prisons who were successfully reallocated home (Nevada Reapportionment and Redistricting website). Based on the 7,826 successfully reallocated people, we can assume that the remaining 3,528 people had addresses that could not be reallocated. The available documentation does not explain why some addresses could not be reallocated.
State documentation:
October 2021 presentation from Asher Killian, Haley Proehl, and Matthew Lawton on the interim status of the Nevada reallocation process. (Note: This documentation is from October 2021, and subsequently, further work was done to reallocate people in prisons to their addresses. The numbers in this presentation are not the final reallocation counts that are in the adjusted dataset published at the completion of the reallocation process. Those numbers are linked below.)
The New Jersey state documentation reports a total of 18,109 people in state prisons (p. 3). An additional 4,067 people are in federal prisons and were subsequently counted at large (p. 10). 9
Out of the 18,109 people in state prison, 1,114 people had out-of-state addresses. 10
Of the remaining 16,995 people in state prisons with in-state addresses, 797 had in-state addresses that were unusable (p.5). The state defines “unusable” addresses as: residential addresses that were unknown/homeless, incomplete, PO boxes, rural route addresses, or contained an invalid/nonexistent address or street name.
The remaining 16,198 people in state prisons could be reallocated home in New Jersey (p.5).
The New York state documentation (p.2) reports a total of 46,418 people in prisons across the state. 3,926 of those people are in federal prisons and were subsequently counted at large. That leaves a remaining 42,492 people in New York state prisons. The three-step geocoding and address verification process used in New York included the following:
Utilization of the Google Maps Geocoding API 12 to extract the latitude and longitude coordinates from addresses. At this point, geocoding results were accepted when the program confirmed street address precision. These results were then converted to Census blocks.
Reverse geocoding the information generated by the Google Maps Geocoding API latitude and longitude points to a mapped address, and then these results were manually compared to the original addresses. Discrepancies at this stage were investigated using Google StreetView, Bing Maps, and street name lists.
At this point, any remaining “invalid addresses” (i.e., addresses that were not successfully mapped to an address) were investigated manually using Google StreetView, Bing Maps, and street name lists. The state reports that “many of these initially invalid addresses were effectively geocoded” at this point by visually confirming house and building numbers or rectifying street name misspellings. Google StreetView and Bing Maps had poor coverage or limited visibility for some apartment complexes and trailer parks. In these instances, a third source was used, the Cornell University-based New York Block Browser LUCA Evaluation System (NYBBLES), allowing for rooftop geocoding of many individual structures in mobile home parks and apartment complexes.
After these three steps, 3,465 of those people in state prisons still had “invalid” addresses and were subsequently counted at large. Invalid addresses were loosely defined in the documentation (p. 2-3) as addresses that were not able to be geocoded successfully via New York’s three-step process. 39,027 people in state prisons were able to reallocated home in New York.
The Rhode Island redistricting consultant’s documentation reports 2,618 people in Rhode Island Department of Corrections custody across the state (Nov. 15 presentation, slide 13). 691 addresses had “obstacles to geocoding,” which included: 271 records with no permanent addresses, 154 with out-of-state addresses, 146 serving home confinement,11 44 incarcerated out of state, 23 addresses listed at the facility, 10 addresses with no street, and 4 with P.O. boxes as addresses (Nov. 15 presentation, slide 13). The remaining 1,927 people in custody had addresses that were deemed geocoded (Jan. 12 presentation, slide 3).
While 1,927 addresses were successfully geocoded, the Rhode Island redistricting process only reallocated people who were not yet sentenced or had less than two years until their expected release date. Out of all 1,354 people (Jan. 12 presentation, slide 3) in Rhode Island Department of Corrections custody who met that criteria, only 1,074 people were reallocated to their home address.
See our methodology for a full explanation of how we conducted our analysis. ↩
There are obviously some people who actually do live in the same legislative district where the prison they are incarcerated is located. However, our analysis shows this is the exception, not the rule. ↩
The Colorado legislation requires the state to request this data from the federal Bureau of Prisons, but that federal agency refused to share it. ↩
The Colorado documentation states, “Nonpartisan staff will not reallocate incarcerated persons with a previous known address outside Colorado and will instead leave these persons in the state correctional facility census block they are listed under in the DOC report. Section 2-2-902 (4), C.R.S. requires only an incarcerated person with an in-state address (in-state incarcerated persons) to be counted at their last known address. Per the DOC report, there were 1,270 incarcerated persons in a Colorado correctional facility on April 1, 2020 with a last known address outside Colorado” (p. 3). ↩
There were 688 records with out-of-state addresses, but one was also under 18 years old, so that individual is counted in the total count of people excluded because they were younger than 18 years old. ↩
These 333 records could not be “geocoded in the cleaned dataset,” but there is no clear definition of what that means. ↩
Of note, the Federal BOP did not provide the population count of federal prisons in NJ: “As required under N.J.S.A. 52:4-1.2 and N.J.S.A. 52:4-1.3, the Secretary of State requested a report from the NJ Department of Corrections and from agencies that operate a federal corrections facility in New Jersey. The Department of Corrections provided such report, and the Federal Bureau of Prisons advised that it would not provide the requested data because the state’s request is not for law enforcement purposes and thus, did not meet the criteria for permissible disclosure of the data sought.” To address this, New Jersey relied on Census data for the population of the two federal facilities in the state, which is where the federal population count of 4,067 originates. The Census data included an additional 35 people who were not accounted for in the “group facilities count” and New Jersey counted these 35 individuals at large (p. 8). ↩
As the state documentation explains, “for each individual included in the report received from the NJ Department of Corrections pursuant to N.J.S.A. 52:4-1.2 where the individual’s address was unknown or not located within the State, and for all individuals reported in the census as residing in a federal incarceration facility for whom a report was not provided under N.J.S.A. 52:4- 1.3, the relevant population counts reported in the census have been reallocated to reflect that the individual resided at “an unknown address within the State” on the day the census was completed, and the individual is not included in any applicable population counts reported in the census for the geographic units where the individual’s prison facility was located on the day the census was completed.” (p. 2-3) ↩
The state documentation (p. 2) explains this program as “an interface that gives programmers a method to actively send requests to the Google geocoding server through customized scripts and functions in order to obtain geographic coordinates from valid addresses.” (p. 2) ↩
In calculating the success of the geocoding process, we removed the following addresses from the denominator: out-of-state addresses, which are by definition not mappable to a location within Rhode Island, and those listed as home confinement because people in that situation would have had the opportunity to fill out the Census at home and be counted there directly. ↩
With Governor Tim Walz's signature, the state is the latest to reject the Census Bureau’s flawed and outdated way of counting incarcerated people.
May 20, 2024
On Friday, Minnesota Governor Tim Walz signed HF 4772 — an omnibus elections policy bill — into law, officially ending prison gerrymandering in the state. With this action, Minnesota joins the rapidly growing list of states that have taken action on this issue. The measure requires state and local governments to count incarcerated people at their home addresses when drawing new political districts during their redistricting process.
Prison gerrymandering is a problem created because the Census Bureau incorrectly counts incarcerated people as residents of their prison cells rather than their home communities. As a result, when states use Census data to draw new state or local districts, they inadvertently give residents of districts with prisons greater political clout than all other state residents.
“With this law, yet another state has rejected the Census Bureau’s flawed way of counting incarcerated people to ensure its residents have an equal voice in their government,” said Aleks Kajstura, Legal Director of the Prison Policy Initiative and the head of the organization’s campaign to end prison gerrymandering. “Roughly half the country now lives in a place that has ended prison gerrymandering. With so many places taking action on this issue, it raises the question, ‘Will the Census Bureau listen to the growing consensus on this issue, or will it cling to its outdated and misguided way of counting people in prisons and jails in 2030?'”
The provisions ending prison gerrymandering in the state were initially introduced as standalone measures and were rolled into this omnibus election bill.
Avoiding carveouts in prison gerrymandering reforms
Carveouts that exclude certain incarcerated people from bills to end prison gerrymandering undermine their impact and continue to distort democracy.
During the House debate on the measure, a representative introduced an amendment to limit the impact of the reforms by continuing to count people who have served ten consecutive years in prison or have over a decade remaining on their sentence at the prison. Fortunately, this poison-pill amendment was rejected, but it explains why other states considering similar reforms should also avoid this type of carveout.
Misguided amendments like this ignore the reality that, regardless of an incarcerated person’s sentence length, they are still not a member of the prison community. Additionally, nationally 75% of people in prison serve time in more than one facility, while 12% of people serve time in at least five facilities before returning home. These amendments also assume that a person will serve their full sentence and that sentencing policies or release mechanisms in a given state will remain static over a ten-year period, two things that are far from certain.
States shouldn’t undermine their own progress on this issue. When they end prison gerrymandering, they should ensure those reforms include all people incarcerated in the state’s prisons.
Most states have clauses in their constitutions or statutes that explicitly say that prisons are not a residence — whether someone is incarcerated away from home for a few months or a few decades — yet the Census Bureau continues to count people as if they are. The Census Bureau has the authority to change how it counts incarcerated people and officially end prison gerrymandering at the national level, but inaction has forced state and local officials to pass reforms and shoulder the burden of correcting flawed Census redistricting data to count incarcerated people at home.
In addition to Minnesota, eighteen other states — including “red” states like Montana, “blue” states like New York, and “purple” states like Maine — have recognized the impact of prison gerrymandering on political representation and have taken action to provide more equal representation to their residents. Progress on this issue has been so swift that the staunchly bipartisan National Conference of State Legislatures recently called the movement to end prison gerrymandering “the fastest-growing trend in redistricting.”
With Minnesota adding itself to this rapidly growing trend, there is yet another reason for the Census Bureau to finally change how it counts incarcerated people and end prison gerrymandering nationwide.
Download our one-pager for a quick summary of the evidence that shows incarcerated people return home after their release.
Incarcerated people return to their home communities after release. That’s a fact that is obvious to anyone who lives in a community hard hit by mass incarceration. Yet when states seek to end prison gerrymandering by counting incarcerated people at home for redistricting purposes, some people ask why we should use someone’s home address — concerned that they might not return to that exact place after release or even might stay in the prison town.
Unfortunately, because we can’t predict the future and because prisons and jails generally don’t track where people go after their release, this has been a hard question to answer. However, we collected several unique datasets to fill this gap and provide the best evidence possible of where people go after being released from prison. We found:
Incarcerated people don’t generally stay in the area of the prison after their release.
Incarcerated people almost always go back to the communities they came from — if not the exact address they lived at before their arrest.
Counting incarcerated people in prisons is likely the least accurate place to count them.
In theory, every legislative district within a state is supposed to have the same population to ensure everyone has equal representation from elected officials. However, the Census frustrates this goal by counting nearly 2 million incarcerated people as residents of the places in which they are detained instead of at their home addresses. The resulting Census data leads to “prison gerrymandering” — legislative districts that skew representation in favor of people who live near prisons and other correctional facilities.
To avoid prison gerrymandering, states have taken it upon themselves to count incarcerated people at home in their redistricting data. In practical terms, this means counting people at the pre-incarceration address on file with the Department of Corrections.
People don’t stay near the prison after release
There are two widely-accepted truths about mass incarceration:
With this in mind — along with the fact that more than half a million people enter and leave prison every year — you can conclude that if even a small portion of people stayed near the prison after release, those communities would, over time, begin to look similar to those on the inside, but that’s not the case.
In 2015, we released a report that explored the racial geography of mass incarceration. It showed that in 208 counties with prisons, the portion of the free population that was Black was at least 10 times smaller than the portion of the prison population that was Black. These trends were so strong, in fact, that we found 161 counties where the number of incarcerated Black people was actually bigger — by raw number — than the number of free Black people.
If people in prison stayed in those communities after their release, these dramatic disparities would not exist.
People go home after release
Unfortunately, while prison systems generally keep people’s home addresses on file, they don’t track whether they actually return to that address upon release. However, addresses for people on probation and parole can be a reasonably accurate proxy to fill this gap. That’s because a significant number of incarcerated people are on probation or parole after their release.
Comparing these two sets of addresses, we can get a sense of whether people return to where they came from after incarceration. If the portion of incarcerated people from an area is similar to the portion of people on probation or parole from that area, you can reasonably conclude that people are returning to their home communities after incarceration.
Rhode Island serves as a useful example for this analysis because all of its correctional facilities, including pre-trial facilities, are located in one correctional complex (in the city of Cranston). In most states, people are held pre-trial in jails, usually close to home, but in Rhode Island, they are incarcerated equally far from home, whether they are held for a few days or a few years.
Percentage of people incarcerated or on probation/parole in Rhode Island’s five largest cities
Municipality
Percentage of incarcerated people from the city
Percentage of people on probation or parole from the city
Providence
38.2%
33.0%
Pawtucket
11.9%
11.2%
Woonsocket
8.1%
8.2%
Cranston
7.6%
6.2%
Warwick
4.6%
5.4%
Examining this data for the five largest cities in the state, you see that the numbers align quite closely, reinforcing the fact that incarcerated people almost always return to their home communities after their release.
It is worth noting that Providence, the state’s biggest city, represented a larger portion of the incarcerated population than the probation/parole population. While some may be tempted to say this disproves our point, it is important to remember the context of this time period. The probation/parole data was from December 2020, at the pandemic’s peak, when many urban areas saw their populations dramatically decrease due to health and economic concerns and the desire for more space. Based on the patterns in the other communities, it is reasonable to conclude this data represents pandemic-era fluctuations and should be treated as such.
Importantly, you can also see more incarcerated people come from Cranston than are there while on probation and parole. This is another sign that most people don’t stay near the prison after incarceration unless they are from there originally.
These numbers aren’t a perfect match, which means that the exact address someone ends up at is sometimes different than the one they had on file while incarcerated. However, it clearly shows that people almost always return to their home communities after incarceration.
Prison is the least accurate place for the Census Bureau to count incarcerated people
Even if someone is unconvinced by the data showing that incarcerated people are likely to return to their home communities after their release, there are plenty of reasons — backed up by hard data — that make clear that by counting incarcerated people at prisons, the Census Bureau is choosing the least accurate option.
People are incarcerated further from home than most people choose to live
When someone is incarcerated, they are moved far from their daily surroundings and usually locked up somewhere they never visit on their own — let alone live. They’ll likely never know anyone outside of the prison walls, nor rent a house, visit a business, or attend a community event there. So why would the Census Bureau decide this was their home?
People counted at the prison on Census Day aren’t likely there for long
There is a common misconception that a person in a prison will be in that facility for a long time — this is the mistaken belief that is the foundation of the Census Bureau’s choice to count incarcerated people there. But this simply isn’t true.
The average time served by people in state prison is 2.7 years, with this “average” pulled up by the few people serving very long sentences; the median time served is 15 months. Short stays in prison are common. For example, in Rhode Island, the median length of stay for people serving a sentence in the state’s correctional facilities is only 99 days. This means that most people counted at a state prison on Census Day will spend the vast majority of the next decade outside the prison walls.
The home address someone has on file with the Department of Corrections may not be the exact place they return to after incarceration. However, it is a close approximation of where they reside through and after incarceration. While some of these address records may vary in precision, the one address we know is wrong is the facility address.
States that adjust their redistricting data to count incarcerated people at home use the home addresses contained in their Departments of Corrections records. This practice creates redistricting data that better reflects the populations of communities hardest hit by mass incarceration and communities that contain large prison populations. The Census Bureau should count incarcerated people at home in the 2030 Census using home address data as the states have done.
The Probation and Parole data is from December 2020 and the redistricting data is from April 2020 — these are the closest dates available. While the distribution of people in prison in April 2020 geocoded to their hometowns is not a perfect match to the hometowns of people on probation and parole in December 2020, it is close enough to draw some conclusions. It clearly shows that the towns where incarcerated people come from are the same towns where incarcerated people go after release. ↩
