ASAN Opposes Citizenship Questions For 2020 Census

A person writing in a notebook with a blue ballpoint pen.

This document is also available as a PDF here.

August 7, 2018

Jennifer Jessup

Departmental Paperwork Clearance Officer
Department of Commerce Room 6616
14th and Constitution Avenue NW
Washington, DC 20230

Re: Docket ID:  USBC–2018–0005, Proposed Information Collection; Comment Request; 2020 Census

Dear Ms. Jessup,

The Autistic Self Advocacy Network (ASAN)[1] appreciates the opportunity to offer comments on the proposed information collection for the United States 2020 Census.

ASAN represents a community of thousands of autistic people of diverse racial and ethnic groups, religions, genders, and socioeconomic statuses. Our members and allies also have differing co-occurring disabilities and support needs. We have long advocated for the collection of clear, accurate demographic data, particularly with respect to people with disabilities and autistic people.

The 2020 Census, because it is designed to obtain an accurate count of all people living in the United States for the purposes of apportioning seats in the House of Representatives, is a civil rights issue which impacts people with disabilities, including autistic people.[2] Additionally, both the federal government and state and local governments utilize Census data to make important decisions on the apportionment of funds for a myriad of public services, many of which are vital to the continued health and independence of people with disabilities.[3]

In particular, ASAN strongly opposes the addition of a citizenship question to the 2020 Census. Secretary of Commerce Wilbur Ross announced that the citizenship question had been added to the decennial census on March 26, 2018.[4] We believe that the inclusion of this question would have significant adverse effects and could potentially harm the civil and political rights of Americans with disabilities. We outline our concerns with this question in further detail below.

The inclusion of the proposed citizenship question will likely reduce household participation in the 2020 Census, creating an inaccurate count of U.S. residents that undermines the Census’ constitutional purpose.

Secretary Ross’ 2018 announcement stated that, while there was “widespread support” for the belief that the inclusion of the citizenship question would lower response rates, “the Census Bureau’s analysis did not provide definitive, empirical support for that belief.”[5]

However, many stakeholders, including the U.S. Census Bureau itself, have predicted that the addition of the citizenship question to the 2020 Census would decrease participation in the Census and the accuracy of responses. On January 19, 2018 the Census Bureau’s Associate Director for Research and Methodology, Dr. John Abowd, concluded that the inclusion of the citizenship question would have an “adverse effect on self-response, and, as a result, on the accuracy and quality of the 2020 Census.”[6] Dr. Abowd concluded that a reduced number of self-responses would lead to reduced accuracy and quality because the Census Bureau must conduct non-response follow-up (NRFU) when someone fails to respond to the Census questionnaire, which tends to produce less accurate data.[7] Secretary Ross’ announcement appears to ignore these warnings.

Immigrants and their families were reluctant to respond to the 2020 Census before the introduction of the citizenship question. The Census Bureau’s Center for Survey Measurement reported in a memorandum that respondents to Census Bureau interviews conducted in 2017 “spontaneously” expressed concerns that the Census Bureau would not keep their citizenship status, or the citizenship status of their family members, confidential.[8] Many of the respondents’ concerns were related directly to recent changes in immigration policy and the tense political climate around immigration, including repeated references to the elimination of the Deferred Action for Childhood Arrivals (DACA) program and fear of persecution by Immigrations and Customs Enforcement (ICE).[9] The inclusion of a question that directly asks U.S. residents their citizenship status in this political and social climate is therefore likely to increase the number of non-responding households to some extent, particularly households with noncitizen members.

Additionally, a significant number of stakeholders with relevant experience, from all sides of the political spectrum — including six former census bureau directors,[10] more than 160 Republican and Democratic mayors,[11] and more than 60 members of Congress[12] — have all opposed the addition of the citizenship question on precisely these grounds. More than 170 civil and human rights groups, who have heard directly from the constituents most likely to be impacted by an inaccurate census count, also oppose the legislation.[13]

Given the significant concerns that have been expressed, it is premature for the Census Bureau to conclude that the citizenship question will not increase non-response rates. It cannot merely rely on the absence of studies directly addressing this question. Rather than take that risk and damage the Census’ constitutional goal of ensuring that all residents, including immigrants, are adequately represented in Congress, ASAN urges that the citizenship question be removed.

The predicted omissions and falsification of Census data will directly harm populations whose survival depends upon the data the Census provides policymakers, such as people with disabilities.

 Census Bureau data, collected from both the 2020 Census and the related American Community Survey (ACS), affects the allocation of federal funds to hundreds of programs. A September 2017 working paper by the Census Bureau found that 132 programs utilized Census data to distribute more than 675 billion dollars during fiscal year 2015.[14] George Washington University’s Institute of Public Policy estimated even higher numbers. Its preliminary analysis indicated that approximately 300 financial assistance programs utilize data derived from the decennial Census to allocate over $800 billion in funds.[15]

Several of the federal programs that utilize decennial census data are vital to the continued health, welfare, and independence of people with disabilities. For instance, the Federal Medical Assistance Percentage (FMAP) is in part based on the decennial census’ population count.[16] Medicaid and the Children’s Health Insurance Program (CHIP), both vital programs that provide people with disabilities with funding for the healthcare and supports and services we need to survive, utilize FMAP to allocate funds.[17]

As a result, a statistically significant undercounting of the number of people in a state in the decennial census is likely to have a ripple effect on the lives of thousands of unrelated people with disabilities who need Medicaid services to live. ASAN urges the Census Bureau to reconsider the citizenship question in light of the potential impact of an undercount on the lives of people with disabilities.

 The proposed citizenship question will significantly increase the cost of the 2020 Census.

 According to Dr. Abowd, adding the citizenship question to the 2020 Census would likely cost the federal government an additional 27.5 million dollars.[18] Dr. Abowd stated that this was because federal government administrators must conduct NRFU for each non-response to the census, which costs the federal government considerably more money.[19]

Dr. Abowd claimed this was a “lower-bound,” “conservative” estimate of the cost, based on statistics available from previous censuses.[20] He noted that other evidence available to him indicated that the rate of non-response during the 2020 Census would be “amplified … compared to historical levels.”[21] Dr. Abowd estimated that each percentage point increase in the number of households who do not respond would cost the federal government $55 million. In a situation where many more households fail to respond due to fear of government persecution over their citizenship status — the exact situation many stakeholders suggest is likely to occur during the 2020 Census —  the cost of the 2020 Census will dramatically increase.

These dramatic cost increases are particularly unnecessary given the availability of alternatives for collecting data on citizenship. Dr. Abowd, when discussing the possibility that the Census Bureau could utilize data from administrative records to obtain information on citizenship status, found that doing so would only cost the government between $500,000 and $2 million— a fraction of the $27.5 million cost of including the citizenship question.[22]

The increased costs of the 2020 Census would ultimately be borne on taxpayers and people who rely on tax-funded federal programs to survive, including people with disabilities. We therefore urge the Census Bureau to pursue less-costly alternatives.

ASAN believes in the value of an accurate, effective count of all who live in the United States, which will help ensure fair representation in Congress and the proper allocation of limited federal dollars to those who need them most. We therefore urge the Department of Commerce and the Census Bureau to remove the citizenship question. For more information on ASAN’s positions with respect to a fair and accurate 2020 Census, please contact Sam Crane, our Director of Legal and Public Policy, at scrane@autisticadvocacy.org.