U.S. Census and Voter Suppression? (1 of 2)

A recent University of Memphis Law Review article fleshes out how routine U.S. Census undercounting or possible miscounting facilitates voter suppression and gerrymandering. The article, by Molly Danahy and Danielle Lang, provides context to the recent controversy over the proposed addition of a citizenship question to the 2020 U.S. Census.

The authors beleive the Census undercounts and distorts the representation of minority communities and has done so throughout its history. Some of the reasons for this concern the ongoing problem of counting hard-to-count populations and how the Census deals with prison populations. This post focuses on “prison gerrymandering.” The next post will cover census undercounting.

Prison gerrymandering

The authors state that “prison gerrymandering, or the practice of counting incarcerated persons as permanent residents of prisons for redistricting purposes, is a problem of miscounting a particular population, rather than undercounting a particular population.”

The goal of redistricting is to create districts with nearly equal population so that each person has the same representational power in the district. The primary goal of the decennial census is to provide an accurate snapshot of the number of people in the population and population geography.

Prison gerrymandering substantially harms minority communities disproportionately. Counting incarcerated persons as residing in the prisons that house them rather than in their home political jurisdictions inflates the population of voting districts with prisons relative to jurisdictions that have no prisons.

Danahy and Lang provide an example from Lake County, Tennessee. The county after the 2000 census “drew a district ‘where 88% of the population was not local residents, but incarcerated people’ and therefore ‘every group of 3 residents [had] as much say in county affairs as 25 residents in other districts.'”

Another example comes in 2016 from Florida. Jefferson County divided the county into five commission districts each having about 3,000 voting-age people. One district contained the Jefferson Correctional Institution, which had 1,000 prisoners who were unable to vote. Thus the voters in this district received disproportionate representation, ruled the federal judge Mark Walker.

Judged Walker remarked, “The inmates at JCI, unlike aliens, children, etc. living in Jefferson County, are not meaningfully affected by the decisions of the Boards [The Jefferson County Board of Commissioners and the Jefferson County School Board]. To say they are ‘constituents’ of the Board representatives from District 3 is to diminish the term constituent. To treat the inmates the same as actual constituents makes no sense under any theory of one person, one vote, and indeed under any theory of representative democracy.”

He concluded that the county’s districting scheme “violates the Equal Protection Clause.” He enjoined the county boards from using their current districting plan and required them to submit a revised plan that complies with his Order.

Mass incarceration and representative equality

Peter Wagner notes, “‘many politicians admit they do not view the prisoners inside the walls of their districts’ prisons as constituents.'” Consequently, it is more difficult for localities “most affected by mass incarceration to obtain the political power to necessary to redress its effects.” Also, a disconnect likely exists between the prisoners’ political interests and the interests of the prison districts’ representatives.

Prison gerrymandering existed as a theoretical issue until the 1990 census. Significant incarceration did not begin until Nixon’s emphasis on the War on Drugs and the Burger Supreme Court’s emphasis on being tough on crime. In June 1971, Nixon declared a “War on Drugs” saying that drug abuse was “public enemy number one.” In 1973, he created the Drug Enforcement Administration. During a 1994 interview, Nixon’s domestic policy chief explained that the War on Drugs had two enemies, “the antiwar left and black people.”

After 1973, the War on Drugs “took a slight hiatus” until the 1980s. Reagan expanded many of Nixon’s drug policies. His refocus on the drug war increased incarcerations for nonviolent drug crimes. In 1986, Congress enacted the Anti-Drug Abuse Act. The act established mandatory minimum sentences for certain drug abuses. During this time many states also enacted harsher sentencing laws.

According to the Sentencing Project, total prison population steadily increased from about 316,000 in 1980 to about 1.56 million in 2009. At that time, the rate of increase slowed and began to reverse in 2014. The estimated prison population for 2017 is about 1.44 million. The United States has the highest incarceration rate in the world, 655 per 100,000 people. Only El Salvador is relatively close with an incarceration rate of 618.

Racial differentiation exists in the prison population. The Sentencing Project estimates in 2017 the white incarceration rate was 275 per 100,000 compared with the Black rate of 1,408 and the Hispanic rate of 378. Further, the disenfranchisement rate in 2016 is estimated at 6.1 million, about 2.5% of the U.S. population. This compares to an estimated Black disenfranchised population of 2.2 million, about 7.44% of the total population.

Prison building

The most recent estimates show that the American criminal justice system holds about 2.3 million people. These people are held in 109 federal prisons, 1,719 state prisons, 1,772 juvenile correctional facilities,3,163 local jails, and 80 Indian Country jails as well as in military prisons, immigration detention facilities, civil commitment centers, state psychiatric hospitals, and prisons in the U.S. territories.

Between 1995 and 2005, when the drug war was at its peak, a new rural prison opened on average every 15 days. As the Washington Post noted, “we have slightly more jails and prisons in the U.S. – 5,000 plus – than we do degree-granting colleges and universities. In many parts of the country, particularly in the South, there are more people living in prisons than on college campuses.”

Rural jobs

One facet of the prison gerrymandering issue deals with the location of prisons. Most prisons are in rural districts with predominantly white populations. For example, the 2000 U.S. Census found that Franklin County NY (an upstate rural county bordering Canada with a 2000 population of 51,000) more than doubled its Black population from 1990. Few in the county noticed this because 91% of Blacks in Franklin County are incarcerated in the county’s five state prisons.

According to Mapping Census 2000, areas like upstate Maine, upper Michigan, central Pennsylvania, southern West Virginia, and southeastern Oregon saw relatively large growth in their Black populations between 1990 and 2000. Danahy and Lang point out that “prison gerrymandering allows white, rural districts to meet minimum population requirements by ‘padding’ their population counts with prisoners of color…”

John Eason, author of Big House on the Prairie, remarks that “the number of prisons in the U.S. swelled between 1970 and 2000, from 511 to nearly 1,663…The prison boom is a massive public works program that has taken place virtually unnoticed because roughly 70 percent of prisons were built in rural communities. Most of this prison building has occurred in conservative southern states like Florida, Georgia, Oklahoma, and Texas…Many may be surprised to learn that residents of these often distressed rural communities view local prisons in a positive light.”

Eason also says, “Many feel that prison building is the end product of racist policies and practices, but my research turned up a more complicated relationship. People of color have undoubtedly suffered from the expansion of prisons, where they are disproportionately locked up, but they have also benefited. Blacks and Latinos are overrepresented among the nation’s 450,000 correctional officers. Prisons are also more likely to be built in towns with higher Black and Latino populations.”

Rural communities with prisons, or those who would like to get prisons, see prisons as an economic development/job growth measure. This includes rural minority communities. As Eason notes, “communities that have prisons are opposed to legislation like sentencing reform that would reduce the number of prisoners locked up in America.

“Reforms like repealing three strikes are vital to reducing the number of people imprisoned. The Sentencing Project estimates that without sweeping sentencing reform it would take almost 90 years to return the prison population to its 1980 level.

“Supporting a large number of prisons housing millions of inmates is expensive. In 2014, states spent $55 billion on corrections, meaning the economic benefits to towns come at a high cost to taxpayers. [The estimate for 2017 is $60 billion.]
“It does not look like the footprint of prisons will be shrinking any time soon. Given our current political climate, it’s more likely we will see more prisons built.”

Political barriers

In addition to a rural jobs benefit barrier, a political barrier also exists. A recent Pew Trust report elaborates on the political aspect. “More states plan to count prisoners as residents of their home communities, rather than residents of the places where they are incarcerated – a change that would shift political power away from conservative rural areas to more liberal cities during legislative redistricting.

“Many inmates hail from neighborhoods in or near cities, but most are incarcerated in small towns and rural areas. Counting prisoners as residents of their hometowns would, for the most part, boost the legislative representation of Democratic-leaning urban areas with large minority populations while diminishing the power of Republican, mostly white rural areas.

“New York and Maryland made the change after the 2010 census, and California and Delaware will start with the next redistricting cycle after the 2020 count. Connecticut, Rhode Island, and New Jersey could follow suit.”

Outgoing New Jersey Governor Chris Christie in 2017 vetoed a bill that would have shifted prisoners to their home communities for legislative districting purposes. He said the bill “‘smacks of political opportunism.'”

The Pew report provides an example from New York. Before the state began counting inmates as residents of their home communities, state legislators from upstate areas helped block measures designed to reduce incarceration rates. In 2012 New York counted nearly 60,000 state prisoners in their home districts. “Nine Republican Senators and nine other plaintiffs sued to stop the change, claiming it would shift political power ‘from upstate Republican districts to downstate New York City Democratic districts which constitutes political gerrymandering.'” “A state judge upheld the law saying that counting prisoners in districts where they are incarcerated ‘tends to dilute minority voting strength in violation of the Voting Rights Act of 1965.'”

State Initiatives

In addition to the state actions identified above, other states have enacted measures to deal with prison gerrymandering. These include as of early 2019:

• Colorado – affecting county commissions and school districts
• Massachusetts – requiring all districts to put into place one person, one vote schemes
• Nevada – applying to legislative, congressional, and districts of the Board of Regents
• Tennessee – allowing counties to avoid prison gerrymandering
• Virginia – allowing local governments to exclude prison populations from district calculations
• Washington – ending the prison gerrymandering of state legislative districts.

Census 2020

The Census Bureau addressed again its position on counting prisoners as it prepared for the 2020 census. Again, the Bureau proposed a rule to maintain counting prisoners where they are located. In response to the Bureau’s 2016 proposed rule to maintain its past practices, the Bureau received 77,887 comments from the public. Only four comments supported the proposal to continue to count prisoners at the facilities where they are located. Twenty comments were neutral on the proposal.

In defending its decision, the Bureau remarked that the concept of “usual residence” which it uses is grounded in the 1790 law enacting the first census. This law specified that persons be enumerated at their “usual place of abode.” The Bureau usual residence “is defined as the place where a person lives and sleeps most of their time, which is not always the same as their legal residence, voting residence, or where they prefer to be counted. Therefore, 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.”

The Census Bureau, however, announced two changes that could help states who want to count prisoners at their home residence rather than at their incarceration facilities. For the first time, the Census Bureau will publish their correctional facility population data at the same time as the main redistricting data files they send to the states. In the past, the census published these data too late for use in redistricting. This data will give state officials census counts of correctional facilities’ inmates. States could then subtract incarcerated people from the prison location and using their home address data reallocate them back home for a state’s redistricting.

Also, the Bureau said it would “offer a product that states can request in order to assist then in their goals of reallocating their own prisoner population counts. Any state that requests this product will be required to submit a data file (indicating where each prisoner was incarcerated on Census Day, as well as their pre-incarceration address) in a specified format. The Census Bureau will review the submitted file and, if it includes the necessary data, provide a product that contains supplemental information the state can use to construct alternative within-state tabulations for its own purposes. However, the Census Bureau will not use the state-provided data in this product to make any changes to the official census counts.”

Comment

The issue of prison gerrymandering was new to me, so I found it interesting to review the literature on the topic. Notwithstanding the Bureau’s comments above defending its decision, it seems the counting-prisoners-at-their-residence is the better argument for several reasons.

First, the gerrymandering argument per se possesses enough weight for me to move away from facility-based counting. Such counting is double-sided. One, it overweights the population in the facility communities/localities/congressional districts/states and, two, underweights the places where prison inmates would otherwise reside. It adds, also, to the continued problem of undercounting difficult populations. Overall, facility-counting moves away from one person, one vote.
Second, although the Census Bureau claims fidelity to the 1790 statute, the Bureau has often altered how it counts citizens not residing at their home. As an illustration of this, I provide a lengthy quote from the Bureau’s Technical Paper 62, Americans Overseas in U.S. Censuses, dated November 1993:

“Table 1 lists the residence rules given to census takers for Americans living abroad for an extended period and for crews of U.S. Merchant Marine vessels at sea. As this table illustrates, the residence rules for these groups changed considerably from census to census.”

“A major observation that emerged from reviewing these historic materials was the lack of a single conceptual thread running through the censuses concerning how Americans abroad fit into the overall decennial enumeration. It was partly this absence that led to the inconsistencies – evident in this report – in census treatment of Americans overseas….Legal and political considerations also played an important role in formulating the Census Bureau’s concept and policy of who to count overseas and where to count them. Cross-cutting all these issues has been the fact that decisions relating to overseas enumeration for a specific census have reflected the overall census procedures and prevailing societal conditions (for example, the presence of military conflicts or economic downturns).”

For the 2020 census, a change again has been made. For the 2010 census, all military personnel were counted at the address they provided at enlistment. For 2020, however, deployed service members will be counted as residents of the bases or ports they were temporarily assigned away from.

National Public Radio reported that “For many of the communities surrounding military bases around the country, the bureau’s decision is seen as a long-awaited victory after years for a change in policy.”

NPR noted that “When many of the troops at the U.S. Army base were overseas during the 2010 census, an estimated 10,000 service members were not counted in Christian County, according to Sen Rand Paul, R-KY., who sent a letter to the Census Bureau in support of the policy change. Instead, their numbers were directed to the population counts for the home addresses they provided when they first enlisted.”

The Census Bureau indicates its 2020 guidance “makes a distinction between personnel who are deployed overseas and those who are stationed or assigned overseas. Deployments are typically short in duration, and the deployed personnel will be returning home to their usual residence where they are stationed or assigned in the United States after their temporary deployment ends. Personnel stationed or assigned overseas for longer periods of time and often do not return to the previous stateside location from which they left.”

Assigning for census counting temporarily deployed military personnel at the base or station from which they leave and will return makes sense. Military personnel and their families use area roads, public facilities (such as schools), private establishments, etc. all of which are a cost to local and state governments. And deployments are relatively short term, 5 to 10 months depending often on the branch of military service. Overseas assignments or stations usually run three years.

But compare this with prisons. First, public facilities incur almost no cost from inmates, who cannot move about in the area. And if they have spouses and children, their spouses and children almost invariably stay in the home jurisdictions.

Second, the Department of Justice indicates that in 2016 the median time served for all serious offenses before first release was 1.3 years. For all violent crimes the median time served was 2.4 years; for all property crimes, 13 months, for all drug crimes, 14 months; for all public order crimes, 13 months, and for other/unspecified crimes, 13 months.

It seems inappropriate for a decennial census to count prison inmates as residing in a prison facility when most will be released with 18 months of the census count (the exceptions are for murder, 13.4 years; negligent homicide, 4 years; and rape/sexual assault, 4.2 years). Additionally, many prisoners who serve two or more years in prison serve that time in two or more facility locations.