Census Bureau should use forms, not administrative records, to count people in prison

by Peter Wagner, June 25, 2008 Link

In May, I sent this letter to the Census Bureau asking it to use Census forms to count people in prison. In a step backwards for accuracy in the prison count, the Bureau is proposing to rely only on administrative records in 2010. –Peter Wagner

May 27, 2008

Diana Hynek
Departmental Paperwork Clearance Officer
Department of Commerce, Room 6625
14th and Constitution Avenue, NW
Washington, DC 20230

Re: The Census Bureau’s plan to rely on administrative records to count incarcerated people in the 2010 Census

Dear Ms. Hynek,

I am writing out of concern that the Census Bureau, as announced in the Federal Register, plans to use administrative records to count people in prison in the 2010 Census. As you may know, the quality of the data the Census collected from prison populations in 2000 was deemed “poor” by the National Research Council and that the poor quality of the data derived in part from the Bureau’s higher than expected reliance on administrative records.[1] I am concerned that the decision to rely solely on administrative records in prisons will further decrease the accuracy of the 2010 and future Censuses.

Accurately counting the population in prison is increasingly important. Currently, there are more than 2 million people in prisons and jails. Since the 1980 Census, the percentage of Americans incarcerated in correctional facilities has increased four-fold, with more than 0.7% of Americans currently incarcerated. Certain demographic groups are disproportionately affected by these trends; for example, 11% of the population of African-American men in their 20s and 30s is currently incarcerated.

Three recent National Research Council reports have urged the direct enumeration of people in prison.[2] All three reports believed that direct enumeration would be more accurate than administrative records, and all three reports encouraged the use of a special form that asked for an alternative address. The first two reports made this recommendation to facilitate de-duplication where data is accidentally processed twice; and the third did so as part of a multi-step proposal to modernize how all populations are counted.

Unfortunately, a decision to rely solely on administrative records to count people in prison is a step in the wrong direction. The Bureau should be seeking to expand its direct enumeration in correctional facilities, not curtailing it.

Despite the concerns expressed by some Census Bureau officials, direct enumeration of prisoners can be both safe and appropriate. Although prison officials often warn of safety problems, social workers, volunteers and other civilians enter prisons quite frequently to no ill effect. Further, the Census Bureau’s own experience in 2000 suggests that direct enumeration of people in prison can produce higher quality data at lower cost.[3]

The National Research Council has repeatedly recommended that the Census Bureau create special forms optimized for each type of special housing. The Census Bureau should create such a form for people in correctional facilities and use it where ever possible.

Given that about 1% of all adults in the United States are expected to be incarcerated during the 2010 Census, increasing the accuracy of the prison count would have a large and positive effect on the overall accuracy of the 2010 Census.

Sincerely,

Peter Wagner
Executive Director

cc:
Steve H. Murdock,
Director, U.S. Census Bureau

Footnotes

Legislators ask Census Bureau to change prison count

by Eric Schneiderman and Tedra Cobb, May 28, 2008 Link

May 27, 2008

Diana Hynek
Departmental Paperwork Clearance Officer
Department of Commerce, Room 6625
14th and Constitution Avenue, NW
Washington, DC 20230

VIA ELECTRONIC MAIL: dHynek@doc.gov

Re: Collecting the addresses of incarcerated people in the 2010 Census

Dear Ms. Hynek,

We are writing to urge the United States Census Bureau to collect the home addresses of all incarcerated persons in the 2010 Census. In October 2007, we sent then director Kincannon a letter co-signed by 30 federal, state and local legislators requesting that the Bureau change how it counts the prison population. Counting people in prison at their pre-incarceration address is essential for compliance with the “One Person, One Vote” rulings of the Supreme Court, which require that legislative districts at every level of government contain equal numbers of residents in order to ensure fair and equal representation for all. The Census is uniquely suited to perform this function, which will help us better serve our constituents, our states, and our country.

Currently, the Census Bureau includes everyone housed in federal, state, and local prisons in its count of the general population of the Census block that contains the prison. State laws, however, often define residence as the place where one voluntarily lives. Many states also have constitutional clauses or election law statutes that explicitly declare that incarceration does not change a residence. Prisoners therefore remain legal residents of their pre-incarceration addresses, and in situations where they retain voting rights, they send absentee ballots to their home districts. Unfortunately, the current census methodology opts to disregard this, instead counting a significant proportion of our national population in the wrong place. Crediting the population of prisoners to the Census block where they are temporarily and involuntarily held creates electoral inequities at all levels of government.

In our letter, we asked the Census Bureau to permanently eliminate this problem by counting people in prison as residents of their home addresses. Recognizing that time was short, we asked the Bureau to implement, as an interim solution, the recommendation of the National Research Council, and publish a special version of the PL94-171 redistricting data file for the prison population. Publishing detailed block-level counts of prison populations would empower legislators to remove those populations prior to redistricting. New York has relatively good record keeping procedures tracking the home addresses of inmates, but it is ultimately the duty of the U.S. Census Bureau to provide this information in a usable format for State and County Legislators. Furthermore, not all states find themselves with the same level of resources to locally correct this failing on the part of the Federal Census Bureau.

Unfortunately, we did not receive a positive response from then-Director Kincannon. On November 20, 2007 Director Kincannon wrote that the Bureau was not intending to change its method of counting the prison population. His arguments were based on the Census Bureau’s 2006 report, “Tabulating Prisoners at Their Permanent Home of Record Address,” the findings of which were rejected by the Census Bureau’s own commissioned experts at the National Research Council later that year. The Director’s response also failed to address our request for a special version of PL94-171 to facilitate data correction.

While it is true that improving the accuracy, fairness and utility of the census by changing how incarcerated people are counted would create an additional burden on the Bureau, we feel that the Bureau is making the burdens seem more significant than they really are.

For example, the Census Bureau argued that:

If a valid residential address (i.e., a complete address that can be verified to exist) were provided, the Census Bureau would have to verify the validity of tabulating the prisoner at that address which would require a new census operation to interview the current residents of the address.[1]

The solution need not be so difficult. The Bureau could simply count prisoners the same way that people in non-institutional group quarters like migrant farmworkers, people in hostels, and members of religious orders[2]: Count prisoners at the facility only if they do not report a usual and valid address elsewhere.

Likewise, the Census Bureau has argued that collecting home residence information would be difficult because it does not exist in a complete and usable form in administrative records. While the Census Bureau should certainly respond to the National Research Council’s repeated criticism of the Bureau’s over-reliance on administrative records to enumerate group quarters populations[3], the Census Bureau’s response runs to an opposite extreme. Interviewing an estimated 2.6 million incarcerated people in a private and confidential room would indeed be an expensive, time consuming and difficult operation. To date the Bureau has been reluctant to consider solutions based on its successful practice in other parts of the Census: use individual forms in most cases, and conduct in person interviews or indirect means of collecting the data only where necessary.

Given the Bureau’s failure to take concrete steps to change how incarcerated people are counted in the 2010 Census, we concede that the full reform we seek is difficult to realize in the 2010 Census. But time still exists for two very important steps.

First, the Census Bureau should adopt the recommendations of the National Research Council and conduct experiments during the 2010 census as part of a major research and testing program to evaluate assigning incarcerated people to other addresses outside the facility.[4] The 2010 enumeration is an opportunity that should not be missed to test improvements.

Second, we again repeat our request that the Bureau implement as an interim solution, the recommendation of the National Research Council, and publish a special version of the PL94-171 redistricting data file for the prison population. Publishing detailed block-level counts of prison populations would empower legislators to remove those populations prior to redistricting. Again, Director Kincannon did not address this portion of our proposal in his November 2007 letter.

This special dataset would greatly assist counties and states with large prisons to drawing fair districts. The interest in an alternative tabulation is clear. Thirteen counties in New York already manually remove the prison population prior to redistricting[5], as do counties in Alabama, California, Colorado, Georgia, Illinois, Florida, Michigan, Mississippi, Oklahoma, South Carolina, and Virginia. Many of the counties that retain the prison populations do so only because they are unaware that prisoners are included in the Census, because they feel obligated to use the Census, or because they do not know how to adjust the data.[6]

To manually adjust the Census, counties must rely on incompatible Department of Corrections data or on Summary File 1 data which is published several months after PL94-171. Jurisdictions covered by Section 5 of the Voting Rights Act have particular difficulties because they must often start — if not also complete — redistricting prior to the publication of Summary File 1. Further, these jurisdictions find it difficult to reconcile the partially incompatible racial and ethnic classifications in PL94-171 with the data published for correctional populations in Summary File 1. The Census Bureau is uniquely suited to eliminate these problems.

Publishing an alternative version of PL94-171 dataset that contained just the prison population at their prison addresses is well within the Bureau’s capabilities. There is no impact on data collection procedures and it requires only the publication of a small amount of data in a different format several months earlier than the Bureau would have done otherwise. Of the 8 million Census blocks in Census 2000, less than 6,000 contained a correctional facility.[7] We are asking that the four tables in the PL94-171 redistricting data set be recalculated and published separately to show only the correctional populations in those 6,000 blocks. With this special tabulation, data users would be empowered to decide whether they want to draw districts based on the prison populations without wrestling with the technical questions. By subtracting one data set from another, data users can have a complete and accurate count of their population without the prisoners skewing the data.

This is not an ideal solution, but it is one that is well within the means of the Census Bureau to implement at this late date. We respectfully urge the Census Bureau to take all steps possible to help reduce and eventually eliminate the political inequities and unnecessary data complications that result from the flawed practice of counting people in prison as residents of the census block that contains the prison.

Sincerely,

Eric Schneiderman
New York State Senator, 31st District
Ranking Minority Member,
Senate Codes Committee
80 Bennett Avenue, LA
New York, NY 10033
Phone: 212-928-5578
Fax: 212-928-0396

Tedra L. Cobb
Vice-Chair
St. Lawrence County Legislature
District 8
365 Townline Road
Hermon, NY 13652
(315)386-4928 Phone and Fax

cc:
Steve H. Murdock, Director, U.S. Census Bureau
Frank Vitrano, U.S. Census Bureau
Catherine M. McCully, U.S. Census Bureau

Footnotes

Prisons Warp Vote

by Brett Blank, January 8, 2008 Link

Cross-posted from TomPaine.com.

There once was a time when in the state of New Hampshire, wealth bought votes. It wasn’t illicit or even unusual. In fact it was the law of the state. The amount of power each Senate district had was directly proportional to its taxes paid. This method of redistricting was struck down in Reynolds v. Sims less than 45 years ago.

New Hampshire was not alone. Other states had fundamentally unfair apportionment schemes. For example, Alabama gave every county the same number of state senators. As a result, sparsely populated Lowndes County had the same number of state senators as densely populated Jefferson County. This too was struck down by Reynolds v. Sims.

In that case, the Court declared its now-famous “one person, one vote”  principle, holding that the Constitution prohibits redistricting plans giving voters in one area power beyond their numbers. The Court explicitly held that “legislators represent people, not trees or acres. Legislators are elected by voters, not farms or cities or economic interests.”

In other words, apportionment plans must be based on population and where voters reside should not affect the power of their vote. As a result, an industry or other economic interest can only exert political influence in proportion to the number of people that support it. The Reynolds case intended to permanently the settle the question of whether other interests besides population can be the basis of representative democracy.

Except it didn’t.

The Census Bureau counts people in prison where their bodies are located on census day, not where they come from and where they will return, on average, 34 months later. Forty-eight states bar prisoners from voting, and most states have constitutional clauses or election law statutes which explicitly declare that prisoner remains legal residents of their home addresses. When states draw districts based on the Census Bureau’s padded counts of prison locales, they give those districts extra representation just because the prison industry has a facility there.

Where prisoners are counted matters not only in big-prison states like Texas, it matters everywhere. New Hampshire has one of the lowest incarceration rates in the nation, yet relying on the prisoner miscount to draw districts creates a serious vote dilution problem. Of the 12,594 people in the district containing the New Hampshire State Prison, 11 percent are incarcerated. As a result, every group of 89 residents in this prison district has the same political power as 100 residents elsewhere in the state.

When the Reynolds Court required legislatures to apportion political power on the basis of population they could not have foreseen that a glitch in Census Bureau methodology would later undermine their groundbreaking decision. Prisons are the only industry that has its raw material counted in the decennial census. For example, tourists are counted at their home addresses, and Detroit receives no additional boost from the automobile industry’s unsold inventory.

The intent of Reynolds was to ensure that industries could not exert political influence. To do so, the Court crafted a rule stating that political power must be distributed equally on the basis of population. The Court could not have foreseen that two million people associated with one particular industry would be counted at the wrong location.

To fulfill the promise of representative democracy embodied in Reynolds, the Census Bureau and the state legislatures should develop a plan to count prisoners properly. There is no excuse that people who are easy to count by virtue of their confinement should present such difficulty.

Brett Blank is a law student and a law clerk at the Prison Policy Initiative.